RECORD: 1833. Report of the first and second meetings, at York in 1831 and at Oxford in 1832, including its proceedings, recommendations, and transactions. London: John Murray.

REVISION HISTORY: Transcribed (single key) by AEL Data 7.2013. RN1

NOTE: This work formed part of the Beagle library. The Beagle Library project has been generously supported by a Singapore Ministry of Education Academic Research Fund Tier 1 grant and Charles Darwin University and the Charles Darwin University Foundation, Northern Territory, Australia.


[page i]

REPORT

OF THE

FIRST AND SECOND MEETINGS

OF THE

BRITISH ASSOCIATION

FOR THE

ADVANCEMENT OF SCIENCE;

AT YORK IN 1831, AND AT OXFORD IN 1832:

INCLUDING ITS

PROCEEDINGS, RECOMMENDATIONS, AND TRANSACTIONS.

LONDON:

JOHN MURRAY, ALBEMARLE STREET.

1833.

[page ii]

PRINTED BY RICHARD TAYLOR,
RED LION COURT, FLEET STREET.

PREFACE

[page iii]

THE plan of the British Association for the Advancement of Science, and the principles on which it was founded, are given in the first pages of this volume, which contain a reprint of the Report of the Meeting at York in September 1831: the second part of the volume presents a specimen of the results furnished by the Meeting at Oxford in June 1832.

The contents of the present publication will show distinctly the path which the Association is pursuing, and the difference between its objects and those proposed by any other scientific Societies or Meetings at home or abroad.

It will be observed that the Papers here printed in detail consist chiefly of reviews of the progress of various branches of science, drawn up expressly at the request of the Association and by the recommendation of its Committees. The want of better information respecting the recent advances and actual state of our knowledge has long been felt in every department of inquiry, and the influence which the Association has been able to exercise, in procuring the supply of this desideratum, may be judged of from the declaration of the Professor of Astronomy at Cambridge, who stated at the late Meeting that no inducement but that of such a solicitation as he had received, could have impelled him to undertake the task which, in the following pages, he has fulfilled. The ability and industry which have thus been enlisted in rendering a laborious and responsible service to science, prove the efficacy of a system of public invitation in giving incitement and direction to the energies of individuals, and show the existence of a public spirit entirely in accordance with the designs of the institution.

In publishing these reviews or reports, the responsibility which the Association takes upon itself must be understood

A 2

[page] iv

to be limited to the selection of the subject and of the reporter, leaving the author accountable for his own opinions, and exercising only that general censorship which an Editor is entitled to claim.

The Association also holds itself liable to print in detail those researches on particular points of inquiry which it has requested individuals or Societies to undertake. A few experiments on the magnetic intensity of the earth, and on the quantity of rain which falls at different heights in the atmosphere, are all that will be found under this head in the present volume. It is to be hoped that hereafter they may hold a more prominent place in these Transactions, and that that part of the designs of the Association may be diligently prosecuted, which aims at promoting in a direct manner the investigation of such questions as, in the existing state of science, especially require to be solved in order to open the way to the application of abstract reasoning and the deduction of general laws.

The rest of the Transactions printed in this Report consist of notices or abstracts of the miscellaneous papers which were read at the Meeting, arranged under general heads. It may probably be found necessary in future, that these contributions should not only be printed, but delivered in and read to the Meetings in the same abbreviated form; the business of the session would thus be brought within compass, and the subsequent trouble and delay which it costs to collect the abstracts for publication would be saved. To prevent delay however, and to enable the Officers of the Association to publish the Transactions of the Meeting soon after it has been held, it is of still greater moment that those who draw up the REPORTS on the state of science, which are printed at length, should put the finishing hand to their labours previous to the Meeting, and bring them ready prepared for the press.

The discussions on questions of science which occurred in the Sectional Committees were not the least interesting of the proceedings: but of these it has scarcely been attempted to offer any account; they are a part of the spirit

[page] v

of the Meetings which it is not possible to embody in a Report. The speeches also, which were delivered on various occasions, have only been given so far as they were materially connected with the order of the proceedings, or might serve to exemplify the principles on which the Meeting was conducted. It was a gratification indeed, of no common kind, to listen to the sentiments of so many men of varied talents and high reputation collected together from every part of the United Kingdom; but the record of those sentiments which could have been here presented would have been cold and imperfect, and in such a publication as this would have appeared also redundant and misplaced.

A supply of information not less copious and valuable than that which is now laid before the public, is in preparation to enrich the next volume of these Reports. The printing of an Account of the recent Additions to our Knowledge of the Phœnomena of Sound, which was delivered at Oxford by the REV. MR. WILLIS, has been deferred, to allow the author leisure to prepare it for publication. The postponed Report on the Advances which have been recently made in the Integral and Differential Calculus, by the REV. MR. PEACOCK; that on the principal Questions debated in the Philosophy of Botany, by PROFESSOR LINDLEY; and that on the Question of the Permanence of the relative Level of the Sea and Land, by MR. STEVENSON, are promised for the next Meeting; and in addition, Reports have been undertaken on the following subjects:—

On the State of our Knowledge respecting the Magnetism of the Earth, by MR. CHRISTIE;

On the present State of the Analytical Theory of Hydrostatics and Hydrodynamics, by the REV. MR. CHALLIS;

On the State of our Knowledge of Hydraulics considered as a Branch of Engineering, by MR. GEORGE RENNIE;

On the State of our Knowledge of the Strength of Materials, by MR. BARLOW;

On the State of our Knowledge respecting Mineral Veins, by MR. JOHN TAYLOR;

[page] vi

On the State of Physiological Knowledge, by the REV. PROFESSOR CLARK;

On the State of Zoological Knowledge, by MR. VIGORS.

There is also reason to hope that the ensuing Meeting will be favoured with the communication of some results of researches which have been undertaken in compliance with the Recommendations contained in this volume. It is highly desirable that the attention of the Members of the Association should be particularly directed to these Recommendations of its Committees, and that the suggestions offered by them in 1831 as well as in 1832 should be attended to (See page 48, &c. and page 115, &c.)

The time fixed for the Association to assemble at Cambridge, is Monday the 24th of June 1833.

POSTSCRIPT.

BY the direction of the First Meeting of the Association, a request was made to the chief Secretary of the Government in India, Mr. Swinton, to form a Corresponding Committee at Calcutta, with the aid of Sir Edward Ryan, Major Benson, Mr. Herbert, Mr. Prinsep, and Dr. Christie. An answer has recently arrived from Mr. Swinton, announcing that he has received the Report of the Association, and will have the greatest pleasure in becoming a Member of the Calcutta Committee, in concert with the gentlemen whose names had been conjoined with his, to such of whom as are resident in Calcutta he had communicated the invitation. Sir Edward Ryan, President of the Asiatic Society, has accepted the office of President of the Committee. A further communication is promised respecting "the means which the Committee possess of following up the objects to which their attention has been particularly directed, and the steps which have been taken for inviting the cooperation of the lovers of science in the other Indian Presidencies."

YORK, April 9, 1833.

[page vii]

CONTENTS.

Page
Objects and Rules of the Association ix

FIRST REPORT—1831.

Preface 9
Proceedings of the General Meeting 15
Proceedings of the General Committee 45
Recommendations of the Committees 48
Transactions 56

SECOND REPORT—1832.

Proceedings of the General Meeting 95
Proceedings of the General Committee 111
Recommendations of the Committees 115

TRANSACTIONS:—

Report on the Progress of Astronomy during the present Century. By G. B. AIRY, M.A., F.R. Ast. Soc., F.G.S., Fellow of the American Academy of Arts and Sciences; late Fellow of Trinity College, Cambridge; and Plumian Professor of Astronomy and Experimental Philosophy in the University of Cambridge. 125
Report on the Tides. By J. W. LUBBOCK, V.P. & Treas. R.S. 189
Report upon the Recent Progress and Present State of Meteorology. By JAMES D. FORBES, Esq. F.R.S.L. & E. F.G.S. Member of the Royal Geographical Society, of the Society of Arts for Scotland, and Honorary Member of the Yorkshire Philosophical Society.
Introduction, Discoveries on Heat; Systematic Works on Meteorology. Constitution of the Atmosphere. Temperature, Thermometers; Atmospheric Temperature; Climatology; Decrease with Height; Proper Temperature of the Globe. Atmospheric Pressure, Barometers; Periodical Variations; Accidental Variations; Variation with Height. Humidity, Hygrometers; Distribution of Vapour in the Atmosphere. Atmospheric Phœnomena and Precipitations, Winds; Rain; Atmospherical Electricity, Hail; Aurora Borealis, its Influence on the Magnetic Needle. 196

[page] viii

Report on the present State of our Knowledge of the Science of Radiant Heat. By the Rev. BADEN POWELL, M.A. F.R.S., Savilian Professor of Geometry in the University of Oxford. 259
Report on Thermo-electricity. By the Rev. JAMES CUMMING, F.R.S., Professor of Chemistry in the University of Cambridge 301
Report on the recent Progress of Optics. By Sir DAVID BREWSTER, LL.D. F.R.S. &c. 308
Report on the Recent Progress and Present State of Mineralogy. By W. WHEWELL, M.A., Fellow and Tutor of Trinity College, and late Professor of Mineralogy in the University of Cambridge. 322
Report on the Progress, Actual State, and Ulterior Prospects of Geological Science. By the Rev. W. D. CONYBEARE, F.R.S. V.P.G.S. Corr. Memb. Institute of France, &c. &c. &c. 365
Report on the Recent Progress and Present State of Chemical Science. By JAMES F. W. JOHNSTON, A.M. &c. 414
Remarks on the Application of Philological and Physical Researches to the History of the Human Species. By J. C. PRICHARD, M.D. F.R.S. 530

TRANSACTIONS OF THE SECTIONS:—

Mathematics 545
Optics 547
Acoustics 556
Magnetism.—Electricity 557
Chemistry 566
Meteorology 574
Geography.—Geology 576
Zoology.—Anatomy.—Physiology 587
Botany 595
Arts 597
Miscellaneous 602
Index 603
List of Members. 607

[page] ix

OBJECTS OF THE ASSOCIATION.

THE ASSOCIATION contemplates no interference with the ground occupied by other Institutions. Its objects are,—To give a stronger impulse and a more systematic direction to scientific inquiry,—to promote the intercourse of those who cultivate Science in different parts of the British Empire, with one another, and with foreign philosophers,—to obtain a more general attention to the objects of Science, and a removal of any disadvantages of a public kind, which impede its progress.

RULES.

MEMBERS.

All Persons who have attended the first Meeting shall be entitled to become Members of the Association, upon subscribing an obligation to conform to its Rules.

The Fellows and Members of Chartered Societies in the British Empire shall be entitled, in like manner, to become Members of the Association.

The Office-Bearers, and Members of the Councils or Managing Committees, of Philosophical Institutions shall be entitled, in like manner, to become Members of the Association.

All Members of a Philosophical Institution, recommended by its Council or Managing Committee, shall be entitled, in like manner, to become Members of the Association.

Persons not belonging to such Institutions shall be eligible, upon recommendation of the General Committee, to become Members of the Association.

SUBSCRIPTIONS.

The amount of the Annual Subscription shall be One Pound, to be paid in advance upon admission; and the amount of the composition in lieu thereof, Five Pounds.

MEETINGS.

The Association shall meet annually, for one week, or longer. The place of each Meeting shall be appointed by the General Committee at the previous Meeting; and the Arrangements for it shall be entrusted to the Officers of the Association.

GENERAL COMMITTEE.

The General Committee shall sit during the time of the Meeting, or longer, to transact the Business of the Association. It shall consist of all Members present, who have communicated any scientific Paper to a Philosophical Society, which Paper has been printed in its Transactions, or with its concurrence.

Members of Philosophical Institutions, being Members of this Association, who may be sent as Deputies to any Meeting of the Association, shall be members of the General Committee for that Meeting.

[page] x

COMMITTEES OF SCIENCES.

The General Committee shall appoint, at each Meeting, Committees, consisting severally of the Members most conversant with the several branches of Science, to advise together for the advancement thereof.

The Committees shall report what subjects of investigation they would particularly recommend to be prosecuted during the ensuing year, and brought under consideration at the next Meeting. They shall engage their own Members, or others, to undertake such investigations; and where the object admits of being assisted by the exertions of scientific bodies, they shall state the particulars in which it might be desirable for the General Committee to solicit the co-operation of such bodies.

The Committees shall procure Reports on the state and progress of particular Sciences, to be drawn up from time to time by competent persons, for the information of the Annual Meetings.

LOCAL COMMITTEES.

Local Committees shall be appointed, where necessary, by the General Committee, or by the Officers of the Association, to assist in promoting its objects.

Committees shall have the power of adding to their numbers those Members of the Association whose assistance they may desire.

OFFICERS. A President, two Vice-Presidents, two or more Secretaries, and a Treasurer, shall be annually appointed by the General Committee.

COUNCIL.

In the intervals of the Meetings the affairs of the Association shall be managed by a Council, appointed by the General Committee.

PAPERS AND COMMUNICATIONS.

The General Committee shall appoint, at each Meeting, a Sub-Committee, to examine the papers which have been read, and the register of communications; to report what ought to be published, and to recommend the manner of publication. The Author of any paper or communication shall be at liberty to reserve his right of property therein.

ACCOUNTS.

The Accounts of the Association shall be audited, annually, by Auditors appointed by the Meeting.

TREASURER.

JOHN TAYLOR, Esq. 14 Chatham Place, London.

LOCAL TREASURERS.

DR. DAUBENY, Oxford.

PROP. FORBES, Edinburgh.

JONATHAN GRAY, Esq. York.

PROF. HENSLOW, Cambridge.

WILLIAM HUTTON, Esq. New-castle-on-Tyne.

REV. THOMAS LUBY, Dublin.

DR. PRICHARD, Bristol.

GEORGE PARSONS, Esq Birming ham.

REV. J. J. TAYLER, Manchester.

H. WOOLCOMBE, Esq. Plymouth.

[page 7]

FIRST REPORT

OF

THE BRITISH ASSOCIATION

FOR

THE ADVANCEMENT OF SCIENCE.

[page 8]

[page 9]

PREFACE TO THE FIRST REPORT.

IN giving to the public a Report of the Proceedings of the BRITISH ASSOCIATION FOR THE ADVANCEMENT OF SCIENCE, it has been considered an important object to add to the account of the past Meeting a distinct view of what is to be expected from the next, and to announce the result of the applications which have been made to individuals, requesting them, in the name of the Association, to undertake the reports and researches recommended by its Committees in different branches of science.

The success of these applications will appear from the following statement.

[It has not been thought necessary to reprint this statement, as the Recommendations are contained in the First Report, and the effect of the applications now appears in the Second.]

It will be observed that the object to which the Committees have in general paid the first attention has been, to procure Reports on the state and desiderata of the several branches of science, preliminary to measures which may be hereafter adopted to advance them. To the investigation, however, of a few points of prominent interest and importance they have at once proceeded to invite attention; and of these there are some which it is highly desirable should receive the consideration of experimenters and observers who cannot be individually solicited to take a share in them. Such is the examination of those first data of chemistry, (Recommendations, p. 53), which, lying at the very foundation of the science, are proposed to be settled by the common consent of experienced chemists, and to which it is hoped that every one possessing the necessary means

[page] 10

and habits of accurate experiment will lend his assistance; such, also, are those meteorological and botanical researches, (Recommendations, p. 50, 55), which, belonging to a lower order of facts, are open to a much wider class of observers, and are capable of being extended through all parts of the country by the exertions of individuals, and still more effectually by those of Societies.

The nature and value of the aid which Provincial Societies might render to science through the system of the British Association, and the advantages which they may themselves derive from it, have been lately adverted to by the Council of the Yorkshire Philosophical Society in the following manner*.

"The object of this system is not only to give connection to the efforts of insulated inquirers, but to link Societies themselves together in unity of purpose, and in a common participation and division of labour. There are many important questions in philosophy, and some whole departments of science, the data of which are geographically distributed, and require to be collected by local observations extended over a whole country; and this is true not only of those facts on which single sciences are founded, but of many which are of more enlarged application. Thus, for instance, were the elevation above the sea of all the low levels, and chief heights and eminences, of a country ascertained so generally, that every observer of nature might have a station within his reach from which he could fix the relative position in this respect of whatever might be the object of his research,—of how many questions, in how many sciences, would these facts contribute to the solution? Again, supposing it to be ascertained also, at these stations, what is the temperature of the air, and of the water,—as it falls from the sky, and as it is held in the reservoirs of the earth,—these are data of the same kind, interesting not only to meteorological science, but to the philosophy of organized and animated existence. Yet, extensive as might be the importance of such facts, and simple as are the processes for ascertaining them, and numerous as are the individuals capable of contributing to their in-

* Report of the Council of the Yorkshire Philosophical Society for 1831—32.

[page] 11

vestigation, how little, nevertheless, even of this elementary work has yet been accomplished, either by insulated observers, or by those who are associated together for the express purpose of advancing the sciences to which it is of such essential interest.

None of our Societies has ever pretended to collect observations of this kind on a regular system, nor to form a national catalogue of the scattered particulars of any one science, accurately detailed; and yet the great value which would attach to such collections of facts, when reduced and analysed, must often have occurred to the enlightened conductors of such institutions; but that which has prevented any single Society from venturing on the undertaking, has been the impracticability of carrying it on over so extensive a territory as an entire kingdom. There is a method, however, by which these important objects might be achieved. Were there in every county one or more provincial Societies, having some members competent to superintend, and others ready to execute, the observations within definite limits, and were these Societies willing to work together on a common plan, the natural history of the country, and all the geographical data of philosophy included within it, might easily be collected in a manner far more perfect than has ever yet been attempted.

With a just sense, therefore, of the consequence to science of combining the Philosophical Societies dispersed through the provinces of the empire in a general cooperative union, the British Association has not only invited them to join its Meetings, but has given to those whom they may specially depute to represent them, the privilege of becoming members of the Committee by which its affairs are conducted.

It appears to the Council that in availing themselves of the bond of connection thus offered, Societies will not only contribute most essentially to the success of this extensive plan, but will add greatly to their own efficiency. When individuals meet for scientific objects, the effect of the general effort, emulation, and example, is to produce a spirit of exertion which gives to such meetings their principal value. And if Societies shall concur in thus meeting each other, in proposing certain

[page] 12

common objects, in communicating from year to year the means which they are employing and the progress which they are making,—it seems impossible that this should be done in the presence of an assembly concentrating a great part of the scientific talent of the nation, without kindling an increased ardour of emulous activity; it seems impossible that the deputies of any Society should attend such meetings without bringing back into its bosom an enlargement of views, and communicating to its members new lights of knowledge, new motives for inquiry, and new encouragement to perseverance.

The actual assembling of one of the meetings at the place in which any Society is established, has a tendency to produce the same effect in a still more powerful degree, and the Council does not hesitate to state that this institution has received a sensible impulse in all these respects, from the visit with which it has recently been honoured. The plan indeed on which it was first founded, and on which it has been since conducted, was in the spirit of the design which may now be contemplated for the whole kingdom. Its especial aim has been to collect information respecting its own County, and the end to which it aspires has been described in a former Report to be the execution of such a History of Yorkshire as the Natural Philosopher and the Antiquary may be contented to possess. But how greatly will the importance of this object be heightened when it is incorporated into a national system, and when all the results of our inquiries become part of the materials of a far more extensive analysis. It could not but be felt before by a provincial Society, that, in executing the task which it had undertaken, advice and consultation were wanted. With how much more confidence may it proceed when it has the advantage of consulting with the Committee of this great national Association. In comparing the views which it entertains, and the methods which it employs, with those that may be offered to its consideration, how largely may it profit by such a commerce, without sacrificing any portion of its real dignity or independence."

Should views like those which are here expressed be generally adopted, should the Societies established in different

[page] 13

districts be disposed to combine their exertions through the medium of this Association, for the purpose of carrying a general system of observations into effect, each Society would then become a centre of instruction to its own neighbourhood, from which correct means and methods of investigation might be derived. Thus, for instance, a large proportion of the philosophical instruments at present in use are so imperfectly constructed, and so discordant in their indications, as to be of little service to science; but if Societies will send to the next Meeting of the Association the Thermometer or portable Barometer which they employ, in order that they may be examined, and that any error which may be found in them may be rectified or estimated, the instruments will thenceforward not only speak the same language among themselves, but will become standards with which in every part of the kingdom those of insulated observers may be compared.

The principles which have been already noticed as having regulated the choice of some of the subjects of investigation recommended in the present Report, are important to be borne in mind, at the ensuing Meeting, by those who may take a share in proposing matter of inquiry or discussion. To come to a common understanding on unsettled questions of general interest, to fix the data on which important points of theory hinge, to collect and connect extensive series of observations; these appear to be the objects which peculiarly belong to the Association, and which should therefore be chiefly, if not exclusively, contemplated. It is also very material that those who propose any subject of inquiry should have considered it well in a practical point of view. It is not enough to put forth general recommendations of inquiries without making specific arrangements for their being actually undertaken. The Committee which met for the first time at York laboured under a disadvantage in this respect, from not knowing on what auxiliaries to reckon. Much was in consequence left to subsequent correspondence with the members of the different Sub-committees, which, had it been possible, ought to have been settled at the Meeting itself.

These deficiencies, however, have been so far surmounted,

b

[page] 14

that a highly valuable store of appropriate scientific communications, as has been seen, is already provided for the approaching Meeting; and in this respect also it will possess a great advantage over the last. The Transactions, of which an account is given in the Report, were miscellaneous contributions not expressly designed for the use of this Institution, and in consequence they occupy but a small space in the present publication. It is a principle of the Association to claim no right of property in the papers which it receives; and, with the exception of one Essay, which, by leave of the accomplished writer, has been printed at length, the remainder of this part of the Report consists of abstracts or notices of memoirs which will be communicated to the public through other channels. A few interspersed memoranda of the occasional discussions which followed the reading of the papers, have been inserted, chiefly to illustrate the plan of proceeding which was pursued at the meeting.

It only remains to be added, that the time which has been fixed upon as that on which it will be most convenient for the Association to assemble at Oxford, is the 18th day of June, 1832.

YORK, February, 1832.

[page 15]

FIRST REPORT.

PROCEEDINGS

OF

THE GENERAL MEETING.

1831.

ON the morning of September the 27th, 1831, the Theatre of the Yorkshire Museum was filled by an assemblage of more than three hundred persons*, including many distinguished members of learned and scientific bodies in different parts of the united kingdoms, who were collected together in consequence of a general invitation to the friends of science, which had been issued by the Yorkshire Philosophical Society. At half past twelve o'clock, on the motion of Dr. Brewster, Viscount Milton, the President of the Society, was called to the Chair, and addressed the Meeting nearly in the following words:

"Gentlemen,

You have been kind enough to call me to the Chair of this Meeting, which is indeed one of the most important description; and I only regret, that you have not turned your eyes towards a person, whose acquirements would render him more qualified to fulfil the duties imposed on him. But I trust that, although I may be in some respects deficient, at least I am not deficient in an anxious desire to promote those objects which have been in the view of the authors of the Philosophical Society established in this city, and also those which will be brought under the consideration of the Meeting now assembled. It must undoubtedly be highly satisfactory to the Members of the Society who have taken an active part in making the arrangements for the purpose, to see that we are honoured with the attendance of persons from all parts of the kingdom, who testify,

* The number of Tickets issued, was three hundred and fifty-three.

[page] 16

by coming from so great a distance, their desire to cooperate with the movers of this Meeting, and to carry its objects into effect. Similar Meetings, it is well known, have taken place on the continent of Europe, which have been attended by the most beneficial effects, and I trust that the same effects will attend those that we are now commencing here. In our insular and insulated country, we have few opportunities of communicating with the cultivators of science in other parts of the world. It is the more necessary, therefore, to adopt means for opening new channels of communication with them, and at the same time of promoting a greater degree of scientific intercourse among ourselves. Nor do I see any reason to doubt the successful issue of this undertaking. When I consider what the Yorkshire Philosophical Society has accomplished,—when I view the establishments it has founded, and when I recollect, that it has not existed for more than eight or ten years; having owed its origin, I believe, to the curious discoveries which were made at Kirkdale,—when, I say, we can trace the progress of a body now so considerable, to so inconsiderable a source,— may we not entertain a confident hope, that the Meetings thus auspiciously begun, will rapidly advance to still greater importance, and become the source of incalculable advantage to science hereafter? In addition to other more direct benefits, I hope they will be the means of impressing on the Government of this country the conviction, that the love of scientific pursuits, and the means of pursuing them, are not confined to the metropolis; and I hope that when the Government is fully impressed with the knowledge of the great desire entertained to promote science in every part of the empire, they will see the necessity of affording it due encouragement, and of giving every proper stimulus to its advancement. Perhaps the most effectual method of promoting science is by removing the obstacles which oppose its progress; though I am aware of the fact, that there are some investigations which require to be carried on upon so great a scale, as to be beyond the reach of individual enterprise: and to these, undoubtedly, the energies of Government should be directed. We all know, that the laws of this country,—I mean in particular the fiscal laws of this country,—offer numerous obstacles to scientific improvements. I will name only one instance. In the science of optics very serious obstacles are found to result from the regulations relative to the manufacture of glass. I mention only this; but it must occur to many of the persons present, that there are various other instances, in which the laws interfere materially with the progress of science. With regard to the more direct advantages which we have a right to anticipate from these Meetings, I have no

[page] 17

doubt, that, if they shall be extended to different parts of the country, and held in well-selected places, this result will be obtained: the men of science, now scattered over the empire, will be enabled to meet each other, and mutually communicate their ideas; they will state the advances which have been made in their own respective spheres of action, and also what the deficiencies may be. Thus not only will an extraordinary impulse be given, but the individuals and the Societies taking part in the Meetings will learn what parts of science they can cultivate with the greatest utility, and will give their researches the most advantageous direction. Such, Gentlemen, are a few of the benefits which, it appears to me, will be derived from Meetings of this description; and if they shall be extensively held, and shall be found thus pregnant with important consequences, sure I am that it will redound to the honour of this Society to have been the first to set the example."

Lord Milton concluded by expressing the sense which he entertained of the services which, his friend and predecessor in the office of President, Mr. W. V. Harcourt, had rendered to the Institution, within whose walls they were assembled.

The Rev. William Vernon Harcourt, Vice-President of the Yorkshire Philosophical Society, and Chairman of the Committee of Management, then addressed the Meeting:

"Gentlemen,

I am desired by the Council of the Yorkshire Philosophical Society to submit to your consideration a plan, which they beg leave to propose for the conduct of this Meeting, and for the establishment of a system, on which similar Meetings may continue to be conducted hereafter.

The Meeting, Gentlemen, owes its origin to some distinguished cultivators of science* here present, who were of opinion that great advantage might be expected from an Association for scientific intercourse in these kingdoms, formed upon the model of that which has subsisted in Germany for several years,—an Association which appears to have answered the hopes of its founders, as well in approximating men of science to each other, and promoting among them friendly feelings and an instructive interchange of ideas, as in giving to their union a collective efficacy, and bringing their aims and views more prominently into public notice.

* The Meeting was proposed by Dr. Brewster to the Yorkshire Philosophical Society in a letter to one of the Secretaries (Mr. Phillips). The proposal was approved and encouraged by the Society, and it received the most zealous and effective support from Mr. Robison, Mr. Forbes and Mr. Johnston in Edinburgh, and from Mr. Murchison in London.

B

[page] 18

Fully concurring in the utility of such objects, our Society cordially embraced the proposal which was made to us, that the first Meeting should be held in these apartments—happy if the accommodation which we have to offer could be made serviceable to a purpose of so much public interest, and not insensible, Gentlemen, to the honour and advantage which the presence of so distinguished an assembly would confer upon our own Institution.

In conformity also with the express desire of the promoters of the Meeting, we undertook to make all the arrangements for it, and to prepare the plan of a permanent Association. I will request the Secretary of the Committee of Management to state, in the first place, what arrangements were made, and will afterwards proceed to give an account of the plan which I have to offer to your consideration."

Mr. Phillips, Secretary of the Society and of the Committee of Management, made the following statement:

"The Committee, Gentlemen, being of opinion that the invitations to this preliminary. Meeting should be co-extensive with whatever desire there might be in the country to promote the objects of Science, drew up in the first instance a circular letter inviting the attendance of all persons interested in scientific pursuits, which, in case any one who is here present should not have received it, it may be proper for me to read:—

Sir,

The Council of the Yorkshire Philosophical Society having received intimation from men of scientific eminence in various parts of the kingdom, of a general wish that a Meeting of the Friends of Science should be held at York during the last week in September next, we are directed to announce that the Society has offered the use of its apartments for the accommodation of the Meeting, and that arrangements will be made for the personal convenience of those who may attend it. It will greatly facilitate these arrangements, if all who intend to come to the Meeting, would signify their intention as early as possible to the Secretaries.

The apartments, which the Yorkshire Philosophical Society has to offer for the use of the Meeting, consist of a Theatre, which affords seats for about three hundred persons, five rooms containing the Museum of Natural History, a Library, Laboratory, and Council Room.

All persons interested in scientific pursuits are admissible to the Meeting.

WILLIAM VERNON HARCOURT, VICE-PRESIDENT.
JOHN PHILLIPS, SECRETARY.

Yorkshire Museum, York,

July 12, 1831.

[page] 19

Copies of this circular letter were sent to the Presidents and Secretaries of all the Scientific Institutions in England, metropolitan and provincial, which were known to the Committee, with a request that the invitation might be communicated to any members of those Institutions who might be disposed to accept it. The number of Societies in London thus addressed was THIRTEEN; the number in other parts of England was TWENTY-SIX, NINE of these being in the County of York.

The letter was sent individually to the more distant Members of our own Society, and to persons, whether belonging to any Society or not, who were known to be active cultivators and promoters of science. One hundred and eighty-nine copies were issued on the latter account. In this list, and even in the list of Societies, it is more than possible that the Committee may have been guilty of some omissions, which they hope, however, will be pardoned, when the number of letters sent out is considered, amounting, in the whole, to more than four hundred. One hundred copies were also transmitted for similar distribution to Societies and individuals, by the correspondents of the Committee in Scotland and Ireland; and two or three eminent foreigners were in like manner individually invited, though the Committee did not deem it prudent to extend invitations abroad, till it should be seen what reception the plan of the Association might meet with at home. Lastly, to ensure more general publicity, advertisements of the Meeting were inserted in the Philosophical Magazine for the months of August and September, an announcement of it having before appeared in the Edinburgh Journal of Science."

Mr. Phillips then proceeded to read the answers which had been received to these invitations from persons who had been prevented, by unavoidable engagements, from being present at the Meeting:—answers, which, whilst they excited a deep regret for the absence of the distinguished writers, showed what valuable support the Association might justly count upon receiving from them hereafter. He stated, "that in several instances deputations had been appointed by provincial Institutions to attend the Meeting, and that gentlemen were present, who had come for that purpose from London, Edinburgh, and Dublin, from Newcastle, Manchester, Liverpool, and Birmingham, and even from Bath and Bristol. The great distance of the Plymouth Institution had prevented any of its members from being present; but the official letter received from that body was a gratifying proof of the general interest felt in these proceedings, and of the benefit to be expected from a migratory Association, which might another year be as conveniently at-

B 2

[page] 20

ended by the Southern, as on this occasion by the Northern Societies.

"When the time appointed for the Meeting drew nearer, the Committee of Management put into circulation another notice, specifying more particularly the nature of the regulations which they proposed to adopt. The second circular notice was as follows:—

GENERAL SCIENTIFIC MEETING AT YORK.

It is requested that persons proposing to attend the Meeting will give notice of their intention to the Secretaries of the Yorkshire Philosophical Society.

Models of Inventions, Specimens of Natural and Artificial Products, to be exhibited at the Meeting, Instruments or Drawings to illustrate any communication, and Materials for Experiments, will be received by the Secretaries, and may be transmitted to them previous to the Meeting.

It is also desirable that Memoirs intended to be read, or a short statement of their contents should be sent beforehand, in order to their being registered; and that any Memoir which may be too detailed to admit of being read at length, should be accompanied by an abstract of its principal contents.

On Monday, the 26th inst. the Managing Committee will receive, at the Museum, the names of Persons intending to be present; and will deliver Tickets for the Morning and Evening Meetings, and Dinners, and references for Lodgings. The Committee will think it right to pay regard to œconomy, as well as convenience, in these arrangements*.

The Apartments of the Society will be opened on Monday Evening; and the first Morning Meeting for scientific purposes will be on Tuesday, the 27th, at Twelve o'clock.

Yorkshire Museum,

Sept. 7, 1831.

To this account of the regulations of the Committee, it only remained to be added, that the number of scientific papers to be brought forward, was so considerable, as to demand the employment of the evenings as well as the mornings of the week, and the Committee recommended that the communications of the least abstract nature should be allotted to the evening Meet-

* On Tuesday a public dinner was provided at Twelve Shillings a Ticket; on the other days, during the session, ordinaries at from Five to Seven Shillings a head: venison, game, and fruit being contributed by Earl Fitzwilliam, the Earl of Carlisle, Paul Beilby Thompson, Esq. and Richard John Thompson, Esq. The Archbishop of York gave a public dinner to the Members of the Association on Friday. The Evening refreshments were furnished by a subscription among the Members of the Yorkshire Philosophical Society.

[page] 21

ings, as it was proposed to admit a more popular audience to these."

Mr. Phillips having concluded his statement,—Mr. William Harcourt again rose and read extracts from letters which had been addressed to him by Mr. Chantrey, Mr. Faraday, and Dr. Buckland, who had been reluctantly prevented from attending the Meeting by pressing engagements. Mr. Chantrey, he said, had given the Yorkshire Philosophical Society another proof of his liberal disposition to promote science, by presenting to its Museum on this occasion, a Cast of the celebrated specimen of Plesiosaurus in the Duke of Buckingham's collection. He then read a letter which he had received from the Duke of Sussex, who had been invited to honour the Meeting with his presence. The letter stated, that nothing would have afforded His Royal Highness greater pleasure than to have complied with the invitation, if he had not been unfortunately pre-engaged. "You will, therefore," His Royal Highness added, "be so kind as to express my regret on the occasion, accompanied with my best wishes for the success of so praiseworthy an object, and an assurance on my part, of my warm cooperation in promoting any measure which may be suggested, and sanctioned by such a respectable Meeting."

Mr. Harcourt then commenced his exposition of the OBJECTS AND PLAN OF THE ASSOCIATION.

"When we came to meditate, Gentlemen, on the means of giving stability and continuance to such Meetings as these, when we considered how little command of time men of science in this country enjoy, and how difficultly they are drawn from their occupations and homes, we could not but entertain a doubt whether the inducement of meeting one another, without a more imperative call, would be powerful enough to bring them annually together from distant parts of the kingdom. But, if there were objects of more essential consequence, which a yearly aggregate Meeting might propose to accomplish, objects now unattempted, and yet of the highest moment to the advancement of science, then we apprehended, that those who have any zeal to advance it would not lightly absent themselves from such a Meeting, and that thus the benefit of personal intercourse would follow in the train of still more important advantages.

Views of this extent, however, were not to be indulged without consultation; and, before we ventured to bring them forward, we inquired the opinions of several of the most distinguished among the lights of British science: from some of those who were consulted, we received warm encouragement and valuable suggestions, whilst to others we were indebted for cautions, of

[page] 22

which we also knew the value, and for a fair representation of the obstacles opposed to our success. These different opinions have been weighed with the attention which they deserved; and I present this plan to the Meeting, as one of which all the bearings have been considered, and of which the deliberate consideration has led us to hope, that a great preponderance of advantage may be derived from its adoption.

I propose then, Gentlemen, in the first place, that we should found a BRITISH ASSOCIATION FOR THE ADVANCEMENT OF SCIENCE, having for its objects, to give a stronger impulse and more systematic direction to scientific inquiry, to obtain a greater degree of national attention to the objects of science, and a removal of those disadvantages which impede its progress, and to promote the intercourse of the cultivators of science with one another, and with foreign philosophers.

On the first and most important of these objects, some difference of opinion may exist; a difference of opinion, I mean, as to the want in which we stand of a new Association, to give a stronger impulse and more systematic direction to scientific inquiry.

I do not rest my opinion, Gentlemen, of this want upon any complaint of the decline of science in England. It would be a strange anomaly if the science of the nation were declining, whilst the general intelligence and prosperity increase. There is good reason, indeed, to regret that it does not make more rapid progress in so favourable a soil, and that its cultivation is not proportionate to the advantages which this country affords, and the immunity from vulgar cares which a mature state of social refinement implies. But, in no other than this relative sense, can I admit science to have declined in England. What three names, if we except the name of NEWTON, can be shown in any one age of our scientific history which rank higher than those of men whose friendship we have enjoyed, by whose genius we have been warmed, and whose loss it has been our misfortune prematurely to deplore, the names of DAVY, WOLLASTON, and YOUNG! And there are men still remaining among us, individuals whom I must not mention, present in this Meeting, and absent from this Meeting, whose names are no less consecrated to immortality than theirs.

But it is not by counting the great luminaries who may chance to shine in this year, or that,—in a decade of years, or a generation of men,—that we are to inform ourselves of the state of national science. Let us look rather to the numbers engaged, effectually, though less conspicuously, in adding by degrees to Our knowledge of nature; let us look to the increase of scientific

[page] 23

transactions and journals; let us look, Gentlemen, at the list produced this day of Philosophical Societies which have grown up in all parts of the kingdom. The multiplication of these new and numerous institutions indicates a wide extension of scientific pursuits. The funds so liberally contributed to their support bear evidence of an enlarged disposition in the public to promote such pursuits.

It is on this very ground I rest the necessity and the practicability of establishing in science a new impulsive and directive force, that there are new and more abundant materials to be directed and impelled. The mining-field of discovery seems to me to show, on the one part, the ore breaking out on every side; veins of the precious metal scarcely opened or imperfectly wrought; and on the other a multitude of hands ready to work it; but no one engaging them to labour, or showing them in what manner they may employ their industry to the best advantage. And therefore it is that I propose to you to found an Association including all the scientific strength of Great Britain, which shall employ a short period of every year in pointing out the lines of direction in which the researches of science should move, in indicating the particulars which most immediately demand investigation, in stating problems to be solved and data to be fixed, in assigning to every class of mind a definite task, and suggesting to its members, that there is here a shore of which the soundings should be more accurately taken, and there a line of coast along which a voyage of discovery should be made.

I am not aware, Gentlemen, that in executing such a plan we should intrude upon the province of any other Institution. There is no Society at present existing among us, which undertakes to lend any guidance to the individual efforts of its members, and there is none perhaps which can undertake it. Consider the difference, Gentlemen, between the limited circle of any of our scientific councils, or even the Annual Meetings of our Societies, and a Meeting at which all the science of these kingdoms should be convened, which should be attended, as this first Meeting you see already promises, by deputations from every other Society, and in which foreign talent and character should be tempted to mingle with our own. With what a momentum would such an Association urge on its purpose! what activity would it be capable of exciting! how powerfully would it attract and stimulate those minds, which either thirst for reputation or rejoice in the light and sunshine of truth!

The eldest of our scientific Institutions contemplated, in its origin, the objects which we now propose to pursue. The foundation, Gentlemen, of the Royal Society was an attempt to

[page] 24

reduce to practice the splendid fiction of the New Atlantis*. The same comprehensive mind which first developed the true method of interpreting nature, sketched also the first draught of a national Association for undertaking, by a system of distributed and combined exertion, the labour of that work.

This philosophical romance was not composed by its great author to amuse the fancy, but to dispose the minds of the legislature towards the foundation of a public establishment for the advancement of science. His plan for its maintenance

* The actual and immediate effect produced by Bacon on the general spirit of philosophy has been underrated: His writings were quickly circulated through Europe, and their value was appreciated abroad even sooner than at home: he himself translated the New Atlantis into Latin, "in gratiam exterorum apud quos expeti inaudiverat," and his most important works were rendered into that language and into French before his death. His letter to Baranzon, who lectured on Natural Philosophy at Annecy in Savoy, and who, it appears, had consulted him on the substitution of his inductive method for the syllogisms of Aristotle, deserves attention not only as containing the most perspicuous summary of his views, but as showing how far the authority and influence of his writings had reached in 1621. It has been said by Playfair that Descartes, who became afterwards the head of so numerous a school, "does not seem to have been acquainted with Bacon's works;" and another eminent historian of philosophy, Dugald Stewart, has admitted that, "if he ever read them he has nowhere alluded to them in his writings." But the fact is, that in the correspondence of Descartes with Mersenne, published in 1642, there are, in several of his letters, passages in which he has referred to the works of "Verulam" with a respect which he yielded to no other writer, and has shown that be had both studied them and adopted the methods which they contain; so that there is no longer any difficulty in accounting for the remarkable coincidence with Bacon's views and language which Mr. Stewart has noticed in the principles laid down by Descartes for studying the phænomena of the mind. The passages to which I refer are these: "Scribis præterea velle te scire modum aliquem faciendi experiments utilia; ad quod nihil est quod dicam post Verulamium qui hac de re scripsit, nisi quod omissis minutioribus circumstantiis oporteret in qualibet materia potissimum facere generates observationes rerum omnium maxime vulgarium et certissimarum et quæ sine sumptu cognosci possint, ut, ex. gr. cochleas omnes in eandem partem esse contortas, atque utrum idem obtineat trans æquinoctialem; omnium animalium corpus esse divisum in tres partes, caput, pectus, et ventrem, et alia id genus, hujusmodi enim observationes ad veritatis investigationem certo deducunt." (Ep. LXV.) "Gratias tibi ago pro qualitatibus quas ex Aristotele desumpsisti; majorem illorum catalogum, partim ex Verulamio desumptum, partim a me collectum jam conscripseram, illasque imprimis conabor explicare." (Ep. C. V.) " Scripsisti ad me aliquando esse tibi notos viros quibus volupe erat scientiis propagandis dare operam, (these were probably the persons whose meetings at Mersenne's house laid the foundation of the French Academy,) adeo ut nullum non experimentorum genus propriis sumptibus se facturos profiterentur. Illorum siquis vellet conscribere historiam phænomenorum cœlestium secundum methodum Verulamii, atque omissis rationibus et hypothesibus accurate describeret cœlum prout nunc apparet, quem situm singulæ stellæ fixæ respectu circumjacentium obtineant, quæ sit aut magnitudinis, aut colons, autluminis, aut scintillationis, &c. differentia; item numquid ea consentiunt cum iis quæ de illis veteres Astronomi scripserunt, quave in re different, (neque enim dubito quin stellæ situm inter se suum aliquantulum mutent quamvis fixæ habeantur) hisque subjiceret observationes Cometarum, tabellam conficiens de uniuscunque motu, quemadmodum Tycho de tribus aut quatuor a se observatis fecit, denique variationes Eclipticæ et apogoeorum planetarum, opus esset utilius quam forte primo intuitu videatur, essetque mihi magnum operæ compendium; sed non spero id facturum quenquam." (Ep. LXVII.) If any one will compare these suggestions with the letter to Baranzon before referred to, he will find them almost a literal transcript of Bacon's request to the Savoyard philosopher to undertake this identical task. These extracts show the philosophical character of Descartes in a light somewhat different from that in which it is commonly regarded; like other great geometers before and since, he carried the use of abstractions and hypotheses too far and too soon into physical reasoning; but though he did not, with the wisdom of Newton, abide by the fundamental principle, laid down by Bacon, "non fingendum, nec excogitandum, sed inveniendum, quid natura faciat aut ferat," he was no stranger to the inductive method of collecting axioms from observation and experiment. In a letter addressed to Descartes, and prefixed to his celebrated treatise on the passions, a strong appeal is made to the public liberality to enable him to pursue those multiplied experiments for which he had occasion in order to carry on his investigations into nature. It is stated in this letter, that Gilbert had expended more than 50,000 crowns on the magnet alone, and that to execute all the experiments which Bacon had designed, would require more than the revenue of two or three kings. The writer (probably Mersenne) refers to "l'Instauratio magna et le Novus Atlantis du Chancelier Bacon, qui me semble estre de tous ceux qui ont escrit avant vous celuy qui a eu les meilleures pensees touchant la methode qu'on doit tenir pour conduire la Physique en sa perfection."
In England meanwhile an experimental school was forming, more faithful to the principles of the inductive philosophy. Foremost among the founders of the Royal Society, "Mr. Boyle, the ornament of his age and country, succeeded to the genius and inquiries of the great Chancellor Verulam1;" and he has left us no doubt as to the master by whom he had been taught; for in recording his experiments he has retained not only the method, but the peculiar idiom and technical phrases of Bacon. Thus this great interpreter of nature stood among philosophers like the pilot among the crew; he constructed the chart of knowedge, he marked upon it the place of the ship, he took the bearings of the land, he pointed out the variation of the compass, he noted the force and direction of the winds, and taught the adventurer to steer a certain course over the wide and trackless sea.

[page] 25

is detailed in 'A speech, touching the recovering of drowned mineral works, prepared for the parliament by the Viscount of St. Albans, then Lord High Chancellor of England.' For that end he would have proposed, by legislative enactment, 'to bring those deserted mineral riches into use by the assiduous labours of felons and the industry of converted penitents, whose wretched carcasses the impartial laws have dedicated, or shall dedicate, as untimely feasts to the worms of the earth.' 'By this unchargeable way, my Lords, have I proposed to erect the

1 Boerhaave.

[page] 26

academical fabric of this island's Solomon's house, modelled in my New Atlantis, and my ends are only to make the world my heir, and the learned fathers of my Solomon's house the successive and sworn trustees, in the dispensation of this great service, for God's glory, my prince's magnificence, this parliament's honour, our country's general good, and the propagation of my own memory.' From this speech it appears that the basis of the great Institution, which Bacon meditated, was a public provision for the maintenance and promotion of science. It was one of the defects noted by him in his masterly survey of the state of learning, that science had never possessed a whole man; and he exerted all the influence of his high station and commanding talents, to promote the supply of that defect. In a letter to the king respecting the foundation of the hospital at Dulwich by Allen the actor, he remarked, that though he was glad to see him play the last act of life so well, yet he thought Sir H. Savile's endowments of geometrical and astronomical Professorships of much greater necessity and more deserving of royal encouragement; and his own last bequest was one which, had it been executed, would have endowed two similar offices with salaries of two hundred pounds a year. In his opinion it was 'necessary to the progression of sciences, that those who are to generate and propagate them should be placed in such a condition as may content the ablest man to appropriate his whole labour, and continue his whole age, in that function and attendance;' and he added, 'there will hardly be any main proficiency in the disclosing of nature, except there be some allowances for expenses about experiments, whether they be experiments appertaining to Vulcan or Daedalus, furnace or engine, or any other kind; and therefore, as secretaries and spials of princes and states bring in bills for intelligence, so you must allow the spials and intelligencers of nature to bring in their bills, or else you shall be ill advertised.'

"These desiderata no means have yet been found of supplying in an adequate degree; and science, even to the present day, can scarcely be said to possess more than fractions of men. The Royal Society did not attempt to execute this part of Bacon's plan; but in other respects it copied as closely as possible, the model of the six days College. It was not then an association of individuals throwing their contributions casually into a common stock, but a body politic of philosophers acting in a corporate capacity and with systematic views, allotting to its members their respective tasks, and conjunctively debating and consulting for the advancement of knowledge. It had, in the figurative language of Bacon, its merchants of light, who

[page] 27

were dispatched in various directions at home and abroad, to gather information and bring back specimens of the productions of nature; it had its depredators who were deputed to examine histories of countries, and to question the travellers who had visited them, in order that queries might be framed which were then addressed to the Society's correspondents in foreign lands, among whom Consuls and Ambassadors were proud to be numbered. It employed some of its members as auxiliaries to the arts; to some it proposed the solution of the most important problems in mathematics, whilst it referred to others the charge of experimental researches, the mode of conducting which was discussed before-hand, and the results re-examined by a public Meeting. I may mention as examples of the effect of this system, that we are indebted to it, practically, for Evelyns History of Forest Trees, by which the planting of the country was so materially promoted, and, theoretically, for the determination of the law of the collision of bodies, simultaneously obtained from Huygens, Wallis, and Wren.

This was indeed to execute a noble plan in the spirit in which it was designed. The noise of works and inventions resounded on every side; new facts and original discoveries of the laws of the universe were daily brought to light; the conveniences and safeguards of life, the measurements of time, the construction of ships, the tilling and planting of the earth began to be rapidly improved. But the vigour of these exertions soon declined, and within thirty years we find Leibnitz suggesting to one of the original founders* of the Royal Society that it wanted new warmth to be infused into its constitution, and recommending that it should be remodelled after the example of the French Academy.

Leibnitz indeed had no right to consider a Society effete, which within a few years had elicited a work† from Newton, that eclipsed the fame even of the great German philosopher. Nor to this hour has it ever lost its title to public respect. It still embodies in its list every name which stands high in British science; it still communicates to the world the most important of our discoveries; it still crowns with the most coveted honours the ambition of successful talent; and when the public service requires the aid of philosophy, it still renders to the nation the ablest assistance, and the soundest counsel. Nevertheless it must be admitted, Gentlemen, that the Royal Society no longer

* Dr. Wallis.

† The Principia, as well as the Optics, of Newton were published at the solicitation of the Royal Society.

[page] 28

performs the part of promoting natural knowledge by any such exertions as those which we now propose to revive. As a body, it scarcely labours itself, and does not attempt to guide the labours of others.

Hence it happens, that when any science becomes popular, and those who interest themselves in its advancement perceive the necessity of working for it by united exertions, that science is detached from the central body; first one fragment falls off, and then another; colony after colony dissevers itself from the declining empire, and by degrees the commonwealth of science is dissolved. The new Societies distinguish themselves by their diligence and activity; the parts of knowledge which thus receive more distinct attention, and are propelled by more undivided labour, make rapid advances; and each separate undertaking justifies itself by the most promising appearances and undeniable fruits.

This is a new stage, Gentlemen, in the progress of science; a new state of things, which, whilst it is attended certainly with great advantages, has some consequences of doubtful aspect to the highest aims of philosophy. As the facts and speculations in any department of knowledge are multiplied, the study of it has a tendency to engross and confine the views of those by whom it is cultivated; and if the system of separate Societies shall encourage this insulation, science will be in the end retarded by them more than it is at first advanced. The chief Interpreters of nature have always been those who grasped the widest field of inquiry, who have listened with the most universal curiosity to all information, and felt an interest in every question which the one great system of nature presents. Nothing, I think, could be a more disastrous event for the sciences, than that one of them should be in any manner dissociated from another; and nothing can conduce more to prevent that dissociation, than the bringing into mutual contact men who have exercised great and equal powers of mind upon different pursuits; nothing more fitted to shame men out of that unphilosophical contempt which they are too apt to feel for each other's objects; nothing more likely to open to them new veins of thought, which may be of the utmost importance to the very inquiries on which they are more peculiarly intent.

I remember, at the Meeting of a foreign Society, to have heard a memoir read, in which a specific and original difference was inferred between two animals (commonly considered of one species), not from any difference in the higher and more essential parts of their organization, but from a dissimilarity of colour in

[page] 29

the skin or fur, and from minute anatomical distinctions; and I heard the error of the Zoologist corrected by a Botanist, one of the most eminent in Europe, who illuminated the whole subject of generic, specific and individual difference, by the light of a powerful mind which had been directed to the study of the question, considered in a different aspect, and with a more extensive survey. In like manner it is easy to conceive, on the one hand, how much advantage might be derived to geological debates from the presence of a sober and rigorous mathematician; and how, on the other hand, the abstract analyst and geometer might have his calculations restricted or promoted by listening to the detail of facts, which those could give him who cultivate the sciences more directly dependent on observation and experiment.

But there is a defect in these separate Societies, in respect to their own immediate objects, which I am sure no member of them would wish to dissemble, and which arises from the narrow basis on which they are of necessity built. It is not only that the constant converse of men, who, to borrow the expression of Goldsmith, have often travelled over each other's minds, is not half so effectual in striking out great and unexpected lights, as the occasional intercourse of those who have studied nature at a distance from each other, under various circumstances and in different views; but it is also, Gentlemen, that none of our existing Societies is able to concentrate the scattered forces even of its own science: they do not know, much less can they connect or employ that extensive and growing body of humble labourers who are ready, whenever they shall be called upon, to render their assistance. I have the pleasure of seeing here the President of the Geological Society of London; and I beg leave to ask him, whether in a science, the most complex of all sciences in its object, because it aims at deciphering the history of nature not only as it is but as it has been, in a science of which very few even among the lowest generalizations are as yet so settled as to be able to bear the weight of any theoretical superstructure whatever,—I ask him whether in the science of Geology there is not a multitude of facts to be ascertained in every district, on which he would be glad to see a much greater number of observers employed? And if it be so, let me remind him that we have heard today of nine Philosophical Societies in this county alone, which could doubtless find members ready to prosecute any local inquiry that this Meeting might, at his suggestion, request them to undertake. It is the same with all parts of Natural History, with Meteorology, and indeed with every science which is founded upon observation, or even upon

[page] 30

experiment; for the lower order of experiments, in subjects of the utmost ultimate abstractness,—such as the relations, for instance, of heat and of light,—are not only abundantly wanted, but by a moderate degree of industry and talent are by no means difficult to be supplied.

What numberless suggestions, what a crowd of valuable but abortive hints are continually floating in the thoughts of philosophers, for the pursuit of which time is wanting to themselves! Now I say, Gentlemen, that we have among us, scattered through the country, men willing to adopt these unexecuted hints, as they arise out of the profound and varied meditations of more experienced minds, men not incapable of surveying with accuracy a limited district, though they may not pretend to draw the general outline of the map, or fill up the whole of its details. Many such there are who only wait for instructions, and who require no other stimulus than that of being invited, to render the most essential service to researches and calculations of the highest order; and it is upon this ground especially that we venture to pronounce an Institution wanting, which shall not hesitate to make such invitations and to offer such instructions; it is upon this ground that if we now propose to revive in the nineteenth century a plan devised two centuries ago,— we see a difference, Gentlemen, in the probability of success. Scientific knowledge has of late years been more largely infused into the education of every class of society, and the time seems to be arrived for taking advantage of the intellectual improvement of the nation. Let Philosophy at length come forth and show herself in public; let her hold her court in different parts of her dominions; and you will see her surrounded by loyal retainers, who will derive new light and zeal from her presence and contribute to extend her power on every side.

Much, indeed, is not to be gained in the more recondite subjects of investigation from the first essays of inexpert inquirers; but let the number of those inquirers only be increased; collect around you, Gentlemen, a school fired with a zeal for truth, confess to them how little you know compared with what remains to be known, apprize them that there is not a subject to which they can apply themselves where new materials are not wanted to advance the fabric, or secure the foundations; let them see that the more multiplied have been your discoveries, the more additional openings to discovery have appeared,—and if you will then draw the precise line of what is, and what is not made out in every science, if you will indicate to them those promising points and inlets of inquiry which bid fair to lead to valuable results,—if you will thus put before them right sub-

[page] 31

jects, and at the same time suggest the right methods of treating those subjects; whatever more may be wanting to accurate and successful investigation, natural sagacity and a longer experience will not fail by degrees to supply.

But even the experienced in science will benefit by consultation with each other; for there are different degrees of experience, and no solitary industry or talent can ever hope to equal the power of combined wisdom and concerted labour. Above all, consider, Gentlemen, the excitement to exertion which will be felt by those who are solicited to undertake an inquiry at one of these Meetings, and pledged to produce the investigation at another. The greatest minds require to be urged by outward impulses, and there is no impulse more powerful than that which is exercised by publicly-esteemed bodies of men. Even Newton's papers might have remained unfinished, but for the incentive of such a solicitation. In a letter which I have lately received from Mr. Conybeare, and in which he expresses a deep regret at finding himself unexpectedly prevented from attending this Meeting, the benefit in these respects which may be looked for from a general scientific combination is described with the energy of his ardent and comprehensive genius. 'Your proposal,' he says, 'for ingrafting on the annual reunion of scientific men, a system for effecting such a concentration of the talent of the country as might tend more effectually to consolidate and combine its scattered powers, to direct its investigations to the points which an extensive survey thus generalized would indicate as the most important,—benefited by all the aids which the union of powerful minds, the enlarged comparison of different views, and a general system of intellectual cooperation could not fail to afford,—fills me with visions too extensive almost to allow me to write with sufficient calmness of approbation. The combined advantages, including at once the most powerful stimulus and the most efficient guidance of scientific research, which might emanate from such a point of central union, seem to me to be beyond calculation. If views like those you have sketched could be realized, they would almost give a local habitation and a name to the philosophical academy of Bacon's Atlantis, when "divers Meetings and consults" of the united body of DEPREDATORS, COMPILERS, PIONEERS, &c., suggested new experiments of a higher light and more penetrating nature to the LAMPS, and these at length yielded materials to the INTERPRETERS of nature.'

To that great model of a national Institution for the advancement of science, I have already adverted today, as I have formerly directed to it the attention of the Yorkshire Philoso-

[page] 32

phical Society: it is here referred to by Mr. Conybeare; and by a remarkable coincidence of ideas, we have the same reference from Mr. Harvey, who in a letter from Plymouth, which he has addressed to the Secretary of the Meeting, observes, that Bacon alludes to "circuits or visits of divers principal cities of the kingdom" as forming a distinguished feature of the New Atlantis. 'What Bacon,' he adds, 'foresaw in distant perspective, it has been reserved to our day to realize; and as his prophetic spirit pointed out the splendid consequences that would result generally from institutions of this kind, so may we hope that the new visions which are opening before us may be productive of still greater effects than have yet been beheld; and that the bringing together the cultivators of science from the North and the South, the East and the West, may fulfil all the anticipations of one of the greatest minds that ever threw glory on our intellectual nature.'

I have now laid before this Meeting the reasons for which I think it would be expedient to form a national Association, having for its first object to give a stronger impulse and more systematic direction to scientific inquiry. On the remaining objects which I have before mentioned, it is not necessary for me to enlarge much. It is not necessary to recommend the promotion of a more general intercourse among the cultivators of science, to those who have come in many instances from a great distance expressly to enjoy the gratification of meeting men of kindred minds and congenial pursuits. I shall content myself with remarking, that nothing can be better calculated to prevent those interferences, and reconcile those jealousies which sometimes disturb the peace of philosophy, than the mutual intelligence and amicable communion of such a Meeting as this.

On the grounds which subsist for seeking to obtain a greater degree of national attention to the objects of science, I have little to add to what the Chairman has said. In confirmation of his remarks on the obstacles which some of our fiscal laws oppose to the progress of knowledge, I may adduce the recent experience of this Museum. There is nothing more indispensable to the utility of such an Institution than a complete display of the specimens which it contains; and for that purpose, where the specimens are numerous, extensive glazing is required. Now there is a most serious impediment to this in the high price of glass, and of that price we find that two thirds consist in the duty paid to Government. So that more than one department of science is injured by this tax; the weight of the impost restrains the public exhibition of the objects of natural history; whilst the regulations of the Excise oppose an obstacle to the improvement

[page] 33

of astronomical instruments, still more to be regretted. Among the subjects to which a Scientific Association may be justly expected to call the public attention, I would particularly instance a revision of the law of Patents. The protection which is given to every other species of property is not given in the same extent to the property of scientific invention. The protection which it does receive must be bought of the state at a high price; an expense, varying from two to four or five hundred pounds, is first to be sustained. Then, after encountering the risk of this outlay, the Patentee is compelled to specify publicly and with legal precision, the particulars of his invention; thus it is exposed to be pirated, with the redress only of ruinous proceedings at law; and the consequence is, that no Patent is considered of any value till it has actually maintained a litigation; and though Patents are still taken out, their chief use is understood to be, not so much to secure a right as to advertise a commodity. Such is the present policy of our laws respecting the remuneration of practical science, a policy which seems to have no other end than to restrain the multiplicity of inventions.

With regard to the direct national encouragement which is due to scientific objects and scientific men, I am unwilling to moot any disputed or disputable question. There is a service of science to be rendered to a state with which it cannot dispense; and all, I think, must allow that it is neither liberal nor politic to keep those, who employ the rarest intellectual endowments in the direct service of the country, upon a kind of parish allowance. It would be difficult also to withhold our assent from the opinion that a liberal public provision would have a powerful effect in promoting those studies of abstract science which most require artificial encouragement; and that 'to detach a number of ingenious men from every thing but scientific pursuits; to deliver them alike from the embarrassments of poverty and the temptations of wealth; to give them a place and station in society the most respectable and independent, is to remove every impediment and to add every stimulus to exertion*.' But I will not, on this occasion, enter upon a subject on which any difference of sentiment can be supposed to exist, nor pretend to decide whether Playfair judged rightly of the degree in which a provision of this kind has actually improved the state of science in a neighbouring country, when he added, that 'to such an Institution operating upon a people of great genius and indefatigable activity of mind, we are to ascribe that superiority in the mathematical sciences,

* Second Dissertation prefixed to the Supplement to the Encyclop. Brit.

C

[page] 34

for which, during the last seventy years, they have been so conspicuous.'

One great benefit, at least, in addition to her maritime expeditions, England, as a nation, has conferred on the science of the world. She has had reason to be proud of her astronomical observations; though perhaps it is not equally gratifying to reflect that these observations have been turned to account, of late years, less by her own geometers than by the national school of mathematicians in France. But there are many other sciences, Gentlemen, on which the resources of states are no less dependent; and in them also there are physical data, which require to be ascertained by masters in science, with the most rigorous precision, and not without the most persevering labour. And I may be permitted to think with Mr. Herschel, that 'it may very reasonably be asked, why the direct assistance afforded by governments to the execution of continued series of observations, adapted to this especial end, should continue to be, as it has hitherto almost exclusively been, confined to Astronomy.'

The Chairman of the Meeting adverting to this subject, has said that 'there are enterprises in science which none but a nation can undertake;'let me add also, that there are establishments for science which none but a nation cap support. I remember, Gentlemen, to have heard the greatest philosopher of this age for variety and extent of attainments, M. de Humboldt, speak of Great Britain, as he was showing me the splen-did collections of natural history in the Louvre. What country in the world, said he, has such opportunities as England for collecting in her capital specimens of all the productions of the earth! I reflected, Gentlemen, on those unrivalled advantages,— but felt, I confess, no elation of national pride when I recollected the state of the British Museum. Since that time, however, one material step has been taken towards improvement; and when an adequate building shall have been prepared, let us hope that we may at length see a public school of natural history in London, so furnished, and so appointed, as not to be unworthy of the British nation. I am persuaded that even our statesmen would have no cause for regret, if, whilst the stores of this national repository were replenished by scientific missions judiciously employed, a more accurate knowledge were at the same time obtained of our distant possessions, and of their natural riches, than has been sometimes discovered in our diplomatic transactions.

All the remarks, Gentlemen, which I have this day made, have been made with an anxious desire to say neither more nor less than the truth. I have spoken both of scientific societies

[page] 35

and of the national policy with all freedom, because I take free speech upon points in which the interests of science are deeply concerned, to be one of the principal purposes for which we are now assembled; but I hope I have spoken also without any disposition to exaggerate the deficiencies which I have thought it right to notice, or to elevate a new institution by detracting from the merits of elder establishments. It only remains for me to lay before you the particulars of the plan by which we propose to accomplish the objects which I have stated; the subordinate details would be most advantageously revised by a Committee, but the material principles on which it is framed are points to which I would request the attention of this Meeting.

The material principles of the plan are included in the composition of the Association, in the constitution of its government, and in the selection of the work on which it is to be employed.

Having objects in view more extensive, and at the same time more specific than those of the German Association, we do not recommend the adoption of the same rules. It is not our desire in the general composition of the Society to separate writers from readers, the professor of natural knowledge from the student. A public testimonial of reputable character and zeal for science is the only passport into our camp which we would require. We propose, therefore, that all Members of Philosophical Societies in the British empire shall be entitled to become MEMBERS OF THE ASSOCIATION, on enrolling their names, and engaging to pay such subscription as may be agreed upon, the amount of which subscription, we think, ought to be low; and we propose that the members shall meet for one week in every year at different places in rotation; in order, by these migratory visits, to extend the sphere of the Association, to meet the convenience of distant districts in turn, and to animate the spirit of philosophy in all the places through which the Meetings may move, without rendering them burthensome to any.

But the governing or executive power of the Association, we think, should be vested in a more select, though still numerous body, and placed in the hands of those who appear to have been actually employed in working for science. We propose, therefore, that the GENERAL COMMITTEE shall consist of all Members present at a Meeting who have contributed a paper to any Philosophical Society, which paper has been printed by its order or with its concurrence; taking this as the safest definition of the class of persons intended, but leaving

C 2

[page] 36

power to the Committee to add to its own number, and to admit into the Association other Members at its discretion; and we propose that it shall sit during the time of the Meeting, or longer if necessary, to regulate the general affairs of the Association, to manage the business of the session, and to settle the principal scientific arrangements for the ensuing Meeting.

We recommend, however, that these arrangements should be first digested, and the particular advancement of every science specially looked to by SUB-COMMITTEES, which the general Committee shall appoint, placing severally on each those Members who are most conversant with the several branches of science. We propose that the Sub-Committees should select the points in each science which most call for inquiry, and endeavour, under the authority of the General Committee, to engage competent persons to investigate them; that where the subject admits of the cooperation of scientific bodies, the Sub-Committees should recommend application to be made for that assistance; and that they should attend especially to the important object of obtaining Reports in which confidence may be placed, on the recent progress, the actual state, and the deficiencies of every department of science.

On the last of these points I beg leave to quote the opinion of an able and zealous philosopher, the Professor of Mineralogy at Cambridge, who has been prevented by his public duties at the University from attending the Meeting, but who nevertheless takes the deepest interest in its objects. 'A collection of Reports,' says Professor Whewell, 'concerning the present state of science, drawn up by competent persons, is on all accounts much wanted; in order that scientific students may know where to begin their labours, and in order that those who pursue one branch of science may know how to communicate with the inquirer in another. For want of this knowledge we perpetually find speculations published which show the greatest ignorance of what has been done and written on the subjects to which they refer, and which must give a very unfavourable impression of our acquirements to well informed foreigners.'

I must add, however, to Mr. Whewell's remarks, that this want of knowledge is not by any means confined to our own country. I do not remember anywhere a more remarkable instance of it than that which occurred in France, to one of the most distinguished improvers of optical science*. As late as the year 1815 M. Fresnel re-observed Dr. Young's impor-

* M. Fresnel.

[page] 37

tant law of the interference of the rays of light; he re-constructed the same mathematical formulas for the application of that law to various phænomena, and he announced these researches as new, in a memoir read before the French Academy. One of the members of that eminent body*, better acquainted with the progress of optics than the writer of the memoir, happily preserved an author, to whose original and profound researches the science has been so largely indebted, from printing as his own the celebrated discoveries long before published by another philosopher; but had this information been earlier acquired, it would have saved all the time and labour which were lost in a retrograde inquiry. Even four years after this, when general attention had been drawn to the subject, and the prize offered by the Academy for the best Memoir on the diffraction of light was adjudged to M. Fresnel, the following animadversions were made by the Reporter† on the unsuccessful competitor, whom he nevertheless represents as an experienced physical inquirer ('physicien exercé'). 'L'auteur paraît n'avoir connu ni les travaux dont on est redevable au Dr. Thomas Young, ni le mémoire que M. Fresnel avait inséré en 1816 dans les Annates de Chimie et de Physique: aussi la partie de son travail qui se rapporte aux influences que les rayons de la lumière exercent ou semblent exercer les uns sur les autres en se mêlant, loin de rien ajouter à ce qui était déjà connu, renferme plusieurs erreurs évidentes.'

Having thus entrusted to the Sub-Committees, Gentlemen, the most active share in advancing their respective sciences, and considered them as the instruments by which, through the medium of the General Committee, the impulse of the Association must be principally directed, we recommend that they should not be dissolved with the Meeting at which they have been appointed, but continue in action till the Society re-assembles in the following year. We do not presume that the persons who may happen to compose them, far removed as they may be from each other, may often have it in their power to meet in the interval; but we conceive that they will feel themselves engaged individually to keep the objects, which they have agreed to forward, in their view, and that the correspondence which they may be induced to maintain between themselves, and with the Officers of the Association, may be highly conducive to that combined exertion, the introduction of which into science would save much labour and ensure a better progress.

The appointments to the HIGHER OFFICES of the Society,

* M. Arago.

† Rapport lu à l'Academie 15 Mars, 1819, par M. Arago.

[page] 38

we propose should be not only annual, like the rest of the machinery by which it is kept in motion, but annually changed; in order at once to extend the interest which they may be supposed to kindle, and to limit the burthen which they will impose; leaving it for future consideration, whether the appointment of more permanent SECRETARIES may not be necessary to secure a steady and uniform course: and we recommend that, when the General Committee is not sitting, the whole business of the Association shall be committed to the Office-bearers, assisted in scientific matters by the Members of the Sub-Committees, and in promoting the interests and objects of the Association in particular places, by the cooperation of LOCAL COMMITTEES.

I have now arrived at the last point to which it remains for me to advert—namely, the selection of the matter which is to engage the attention of the Meetings. It is evident that if the plan which I have thus far explained should be carried into effect, the deliberations of the Committee to be formed at the present Meeting will provide the chief materials for the consideration of the next. Those investigations and those surveys of science which shall have been suggested and procured by the Committees and Officers of the Association, will be entitled to the priority, though other communications may be accepted as far as the duration of the session will allow. Professor Whewell conceives 'that if this Meeting were to request from one or two among the most eminent men in the various branches of science, statements to be presented next year of the recent advances made in each department, and the subjects of research which they consider at present the most important and promising, such a request would be respectfully attended to.' Gentlemen, I do not doubt that it would; neither do I doubt that a request from this Meeting would be successful in procuring new researches to be made; and should the funds of the Association hereafter admit of its going further, and offering PRIZES for particular investigations,—then would another prolific source be opened from which the scientific materials of our Meetings would be derived."

This, indeed, would only be another and a very powerful method of carrying On the system which we recommend of advancing science in determinate lines of direction; a method, which, though scarcely practised in this country, has been found eminently successful abroad. Dr. Brewster, I believe, will confirm me in this statement; for he will recollect that we owe to a prize-memoir, the first announcement of that great optical discovery, that light may be polarized by reflection from the surface of transparent bodies, a discovery which has since

[page] 39

been productive of so many new and admirable observations, and which has been in no hands more fertile than his own; and he will remember also, that the first accurate investigation of all the phænomena of diffraction, and the first complete explanation of them by the doctrine of undulations, was contained in a memoir produced by a similar competition*. The award of medals, indeed, is an honourable encouragement not altogether withheld from successful researches in British science. But the principle on which they are given is of a more vague and general nature. The objects of these rewards have never been so distinct as to give a direct stimulus to specific inquiries. I may add, without imputing any mercenary feelings to men of science, that where the inquiry involves expense, a sum of money instead of a medal would, perhaps, be found a more useful and operative offer. It is well known that the important improvements which have been made in Chronometers have arisen, both in France and England, directly out of the public rewards munificently offered by the British Parliament; and I see no reason why adequate and well devised premiums should not be efficacious in the sciences as well as in the arts. No man, however high may be his literary or scientific pretensions, disdains to receive a pecuniary remuneration for the labour which he employs in the composition of his works; and there can be nothing derogatory to the character of a man of science in accepting a similar compensation for the successful exercise of his talents in researches especially which require an expenditure of money as well as time.

"Such, Gentlemen, are the provisions of the plan which we propose for your consideration; and you will perceive that the methods which it embraces are new in practice, though not in principle. How otherwise indeed, than by new methods, can we hope to exchange the present desultory and tardy progress of philosophy, for a more regular, energetic, and rapid advancement? There is a light in the distant horizon to which we have long eagerly looked, and complained that the current did not set us more quickly towards it; and the question now before you, Gentlemen, is no less than this: Whether you are satisfied still to float passively on the waters, or whether you will raise the sail, and ply the oar, and take the helm into your hands. The methods now proposed are new, and therefore cannot place us in collision with any other Society. It has never yet been seen in this country, that twenty Chemists for instance, or twenty Mineralogists, have met together, for the purpose of settling the

* This Memoir, written by M. Fresnel, gained the physical prize proposed by the French Academy for "a general examination of the phænomena of the diffraction of light," in 1819.

[page] 40

nomonolature of their respective sciences, or attemptig to fix with one consent the foundations on which they rest. It has never yet been seen, that the Chemical, Mineralogical, and Optical inquirers have assembled for the purpose of mutually explaining and learning what light the sciences of Chemistry, Mineralogy, and Optics are capable of reflecting reciprocally upon each other. You will perceive also, Gentlemen, that the Transactions, which we contemplate, are not to be collected by trespassing upon ground which was already occupied. In this respect there is on our part not only no design, but no possibility of interference. The course of an Association which meets once a year, and but for a few days, is necessarily different from that of more abiding Institutions; we have no time, if we wished it, to encroach upon the office, or to drain away the scientific resources of any other Society. It will be enough for us, if we can compress into the compass of a week's deliberations our own restricted objects,—specific investigations into fundamental points of science, reviews of its recent advances, and recommendations of subjects and methods for future research. Our plan contains within it a new power which may perhaps accelerate the wheels that are already in action; but its machinery is exclusively its own. The enlightened Institutions with which it hopes to be associated will regard it, therefore, not as a rival, but a coadjutor; and I trust it may prove such a coadjutor to them as the steam-engine has been to all other kinds of mechanism, in every mine, and in every manufactory; a coadjutor, by the aid of whose powerful movements all their operations have been facilitated, and their productions multiplied a hundredfold.

An enterprise like this has no danger to fear, but from a deficiency of zeal and union in carrying it into effect. It must undoubtedly fail, if it meets only with imperfect cooperation and cold support. But if it shall recommend itself to the full approbation of men of science, if it appears to you, Gentlemen, desirable to undertake it, the Association will have competent sponsors in the present assembly, who will stand pledged not only for its early encouragement, but for those future exertions which will be required to ensure its success. The Council of the Yorkshire Philosophical Society have not the presumption to dictate to this Meeting the course which it may be for the interests of Philosophy to pursue. They collected, in the first instance, the best opinions which they could obtain, before they proceeded to mature their plan; and they now wait for the opinion of the eminent persons who are here assembled, before they can assure themselves that it is as feasible in practice as it appears in theory. My own judgement waits with theirs, Gentlemen, on

[page] 41

that of the High Priests of the temple, in the porch of which I am only an humble worshiper,—'parcus Deorum cultor et in-frequens',—and I shall be the first to withdraw the resolutions which I am now ready to propose, unless I find them, by the deliberate and cordial concurrence of this Meeting, stamped with authority and endued with permanence."

A strig of Resolutions in which were embodied the Objects and Rules of the Association as stated in Mr. Harcourt's speech, were then moved by him seriatim, and seconded by Dr. Brewster, by Mr. Murchison, President of the Geological Society of London, by Dr. Pearson, Vice-President of the Astronomical Society of London, by Mr. Robison, Secretary to the Royal Society of Edinburgh, &c. It was resolved unanimously—"that an Association be formed, to be called The British Association for the Advancement of Science, the objects of which shall be to give a stronger impulse and more systematic direction to scientific inquiry, to promote the intercourse of those who cultivate science in different parts of the British Empire, with one another, and with foreign philosophers, and to obtain a greater degree of national attention to the objects of science and a removal of any disadvantages of a public nature which impede its progress." In the next Resolution, purporting "that the members of Philosophical Societies in any part of the British Empire may become members of the Association on enrolling their names and contributing a small subscription", several alterations were proposed; but it was finally passed, with the remaining Resolutions, subject to the revisal and report of a Committee, constituted, according to the proposed plan, of all members present who had contributed a scientific paper to any Philosophical Society, which paper had been printed with its concurrence.

The thanks of the Meeting were then voted to the Chairman, and to the Rev. Mr. Harcourt for his statement of the plan of the Association; and the further consideration of it was adjourned till the following day.

On Wednesday, at 12 o'clock, Viscount Milton was again called to the chair, and the Meeting resumed its deliberations.

The Rev. W. V. Harcourt, as chairman of the Committee, announced that the Resolution respecting the admission of members, in which alterations had been suggested on the previous day, had been revised, and that the Committee recommended that the following persons should be entitled to become members of the Association, upon subscribing an obligation to conform

[page] 42

to its Rules:—1st. allpersons assembled at the present Meeting: 2nd. the Fellows and Members of all Chartered Societies in the British Empire: 3rd. the Office-bearers and Members of the Council or Managing Committee of all Philosophical Societies: 4th. all Members of such a Society recommended by the Council or Committee thereof. They also proposed that the amount of the annual subscription should be One Pound, and that the composition for it should be Five Pounds; that the accounts should be audited annually by Auditors appointed by the Meeting itself, and that the Treasurer of the Yorkshire Philosophical Society should be Treasurer of the Association for the ensuing year.

Resolutions founded on these recommendations, having been moved and seconded by Sir Thomas Brisbane, Mr. Robison, Mr. Dalton, Dr. Daubeny, the Rev. Mr. Scoresby, Dr. Pearson, Mr. Murchison, Mr. Marshall, &c. were passed by the Meeting; and it was resolved that any further revision which the Rules might require should be left to the Committee.

On the motion of Mr. Murchison, seconded by Sir Thomas Brisbane, it was resolved, "that the Rev. Mr. Harcourt be requested to publish for the Association the exposition of its objects and plan which he delivered yesterday." Mr. Har-court, in assenting to the desire of the Meeting, asked permission to revise what he had said, previous to its publication.

The business of forming the Association being completed, the Chairman proceeded to announce the papers to be read that morning on subjects of Science, a report of the contents of which, as well as of the other communications made during the Session, will be found in the subsequent account of the Scientific Transactions.

On Thursday morning it was stated to the Meeting that the Committee had chosen Viscount Milton, the Rev.W.V. Harcourt, and the Secretaries of the Yorkshire Philosophical Society, to be the actual President, Vice-President, and Secretaries of the Association; that the Rev. Dr. Buckland had been chosen President elect; Dr. Brewster and the Rev. Professor Whewell Vice-Presidents elect; Dr. Daubeny and the Rev. Professor Powell, Secretaries elect; and that the next Meeting was appointed to be held at OXFORD.

On Saturday evening, the scientific communications and discussions having been closed by some remarks of Dr. Brewster, in which, adverting to a method of rendering visible the legends

[page] 43

of ancient coins, he stated that he had never been more struck than by observing on an old coin, which he had placed on hot iron, an inscription make its appearance which he could read in a dark room, bearing the words 'Benedictum sit nomen Dei" —Viscount: Morpeth rose, and addressed the Meeting as follows:—

"Ladies and Gentlemen, an office has been assigned to me, which, although most entirely without any qualification or pretension to fulfil, I nevertheless accept, and will discharge, to the best of my ability, with the utmost alacrity. To the character of a man of science I have, unfortunately for myself, no claim whatsoever; but I have the good fortune to be intimately connected with the county, and consequently with the city of York; and I feel that they have both received great benefit and additional credit from the Meeting which is now brought to a conclusion. I say this, both with reference to the positive instruction we have received upon so many most interesting and important subjects, and also to the circumstance of this town and this edifice, already so much indebted to the zeal, perseverance, and ability of our Vice-President, having been now selected as the birth-place of an Association, which, I trust, is destined to confer fresh lustre on British science, to give a new motive and a new guarantee to the friendly inter-course and continued concord of nations; to make further inroads into the untravelled realm of discovery, and glean fresh harvests from the unexhausted field of Nature; to promote the comforts and augment the resources of civilized man; and to exalt above and over all the wonder-working hand of Heaven. For it will always come out from the pursuit of knowledge as surely as from the rusty medal of which we have this moment heard, 'Benedictum sit nomen Dei.' Observe well, if you wish to appreciate rightly the true value and nobility of science, that while it proposes to itself distinct courses and definite spheres of its own, its general tendencies conduce to peace, and minister to piety. With these views and these hopes, it is natural and it is becoming that there should be mixed feelings of gratitude to those whose efforts have contributed so largely to our future progress. An assembly like that which I have the honour to address, will appreciate far more justly than I can pretend to do, the several papers and productions which have been submitted to our notice. I have no scruple in leaving to your more competent and accurate discrimination, the indications of enlightened and powerful thought which they have exhibited; but I feel sure that, if you pardon me for this intrusion of myself, the proposition I now make will

[page] 44

command, upon this occasion, both the grave assent of Science and the soft sanction of Beauty. I move that the thanks of this Meeting be given to Dr. Brewster, and the other authors who have favoured us with their communications."

Mr. Murchison then rose, and "on the part of Dr. Brewster and his other scientific friends, begged leave to return thanks for the high honour done to the contributors of scientific memoirs, and for the valuable aid which had been received from the residents of York and its neighbourhood, in the promotion of the objects of the Meeting." He explained the motives which induced the original promoters of it to select the city of York for their first assembly. "To this city," he said, "as the cradle of the Association, we shall ever look back with gratitude; and, whether we meet hereafter on the banks of the Isis, the Cam, or the Forth, to this spot, to this beautiful building, we shall still fondly revert, and hail with delight the period at which in our periodical revolution we shall return to the point of our first attraction." Mr. Murchison, after expressing his sense of the kind reception and hospitality which the strangers there collected had experienced from the Archbishop, and from all classes of the inhabitants of the city and neighbourhood, concluded with a motion of thanks as follows:— "That the cultivators of science, here assembled, return their most grateful thanks to His Grace the Archbishop of York, and the other Members of the Yorkshire Philosophical Society, for the very liberal manner in which, by the use of their Halls and Museum, and by their obliging and unwearied efforts to provide every accommodation and comfort to those who have visited York on the present occasion, they have so essentially contributed to the success and prosperity of this Association."

This motion was seconded by Dr. Brewster, and supported by Mr. Dalton. Mr. Harcourt, who was in the chair, then said, that "it was quite unnecessary, from the feelings which he knew to pervade the breasts of all, both strangers and residents, to put to the vote of the Meeting either of the proposals so eloquently brought forward. In the long period of its existence the ancient city of York had never greater reason to be proud, than of the genius and talent it contained within its walls at that moment, and of the honour it had acquired in being the birth-place of an Association destined, he firmly believed, greatly to enlarge the boundaries of science, and in so doing to advance the many interests of human nature which depend upon the improvement of knowledge." He then declared the Meeting to be adjourned to Oxford.

[page] 45

PROCEEDINGS

OF

THE GENERAL COMMITTEE.

THE General Committee was employed from the 27th of September to the 2nd of October in revising the Rules of the Association, in appointing the Office-bearers, in embodying the Local and Sub Committees, and in receiving the recommendations of the latter, and making arrangements for carrying their suggestions into effect.

The care of completing the objects which the Committee had in view, and of printing the results of its deliberations, together with the proceedings and transactions of the General Meeting, was entrusted to the Officers of the Association at York, who have drawn up from its minutes a summary of the Objects and Rules of the Association*; to which are subjoined the appointments of Officers and Committees. The particulars of the scientific business brought before the General Committee are included in the account of the Recommendations of the Sub-committees; and the success which has attended the applications, made in the name of the Association, to eminent individuals, requesting them to undertake the services in science which had been so recommended, has been stated in the Preface to the Report.

OFFICERS.

President.— Charles William, Viscount Milton, F.R.S. &c. President of the Yorkshire Philosophical Society.

president elect.—Rev. William Buckland, D.D., F.R.S. &c. Professor of Geology and Mineralogy, Oxford.

Vice-President.—Rev. William Vernon Harcourt, F.R.S. &c. Vice-President of the Yorkshire Philosophical Society.

Vice-Presidents elect.—David Brewster, LL.D., F.R.S.

* This summary, with the additional regulations since made, is printed in the present volume at the end of the Second Report.

[page] 46

L. & E. Corresp. Member of the Institute of France, &c. Rev. William Whewell, F.R.S. &c. Professor of Mineralogy, Cambridge.

Treasurer.— Jonathan Gray, Esq. York.

Secretaries.York.—William Gray, jun. John Phillips, F.G.S. &c. Secretaries of the Yorkshire Philosophical Society.

London.—Rev. James Yates, F.L.S., G.S. &c.

Dublin.—Rev. Thomas Luby.

Edinburgh.—John Robison, Secretary of the Royal Society of Edinburgh, &c.

Oxford.—Charles Daubeny, M.D., F.R.S. Professor of Chemistry, Oxford. Rev. Baden Powell, F.R.S., Savilian Professor of Geometry, Oxford.

LOCAL COMMITTEES.

London.—G. B. Greenough, F.R.S., Vice-President of the Geological Society. R. I. Murchison, F.R.S., President of the Geological Society. Rev. James Yates, F.L.S. &c.

Edinburgh.—James D. Forbes, F.R.S. E. &c. James F. W. Johnston, A.M. John Robison, Sec. R.S.E. &c.

Dublin.—W. R. Hamilton, F.R.S. &c. Astronomer Royal of Ireland. Rev. B. Lloyd, D.D., Provost of Trinity College, Dublin.

India.—George Swinton, Esq., Chief Secretary to the Government in India, has been requested to form a Committee at Calcutta, with the aid of Major Benson, J. Calder, Esq., Dr. Christie, J. Herbert, Esq., J. A. Prinsep, Esq. and Sir Edward Ryan.

SUB-COMMITTEES.

Mathematical and Physical Science.

David Brewster, LL.D., F.R.S. L. & E. &c. Sir Thomas Brisbane, K.C.B., F.R.S. L. & E., Corresp. Member of the Institute of France. James D. Forbes, F.R.S.E., &c. W. R. Hamilton, F.R.S. &c. Rev. William Pearson, LL.D., F.R.S. Vice-President of the Astronomical Society. Rev. Baden Powell, F.R.S., &c. Rev. William Scoresby, F.R.S. L. & E., Corresp. Member of the Institute of France. Rev. W. Whewell, F.R.S. &c. Rev. R. Willis, F.R.S. &c.

Chemistry.

Rev. James Cumming, F.R.S. Professor of Chemistry, Cambridge. John Dalton, F.R.S. President of the Literary and Philosophical Society at Manchester, Corresp. Member of the Institute of France. Charles Daubeny, M.D., F.R.S. &c.

[page] 47

Rev. W. V. Harcourt, F.R.S. &c. J. F. W. Johnston, A.M. Edward Turner, M.D., F.R.S. L. & E. Professor of Chemistry in the University of London. William West, Secretary of the Leeds Philosophical Society.

Mineralogy.

Thomas Allan, F.R.S. L. & E. Robert Allan, F.G.S. &c. David Brewster, LL.D.,F.R.S. &c. J. F. W. Johnston, A.M. Rev. W. Whewell, F.R.S. &c.

Geology and Geography.

Rev. William Buckland, D.D., F.R.S. &c. Rev. W. D. Conybeare, F.R.S. &c. Vice-President of the Geological Society, Corresp. Member of the Institute of France. Sir Philip Grey Egerton, Bart. F.R.S. &c. James D. Forbes, F.R.S. E. &c. G. B. Greenough, F.R.S. &c. William Hutton, F.G.S. &c. R. I. Murchison, F.R.S. &c. John Phillips, F.G.S. &c. Rev. Adam Sedgwick, F.R.S. &c. Woodwardian Professor, Cambridge. William Smith, Author of the Geological Map of England. Henry Witham, F.G.S. &c. Rev. James Yates, F.L.S. &c.

Zoology and Botany.

Charles Daubeny, M.D., F.R.S. &c. J. K. Greville, M.D., F.R.S. E. &c. Rev. J. S. Henslow, F.L.S. Professor of Botany, Cambridge. John Lindley, F.R.S., L.S. &c. Professor of Botany in the University of London. J. C. Prichard, M.D., F.R.S.

Mechanical Arts.

J. H. Abraham, F.L.S. &c. John Robison, Sec. R.S. E. &c. Benjamin Rotch, F.S.A. &c.

[page] 48

RECOMMENDATIONS

OF

THE SUB-COMMITTEES.

COMMITTEE OF MATHEMATICAL AND PHYSICAL SCIENCE.

Mathematics.

THE Committee recommend that the Vice-President of the Association residing at Cambridge be requested to use his utmost efforts to procure, from some competent individual, a Report to the next Meeting on the progress of Mathematical Science.

Astronomy.

That Professor Airy be requested to favour the Association with a Report on the state and progress of Physical Astronomy, together with such remarks on the improvements of Practical Astronomy, as he may deem it useful to add.

Theory of Tides.

That J. W. Lubbock, Esq. be requested to furnish a statement of the means which we possess, or which we want, for forming accurate tables for calculating the time and height of High-Water at a given place.

Meteorology.

That James D. Forbes, Esq. be requested to draw up a Report for the next Meeting, on the present state of Meteorological Science.

The Committee, considering that the science of Meteorology is in more want, than perhaps any other, of that systematic direction which it is one great object of the Association to give, has thought it advisable to propose the following points for investigation.

[page] 49

I. That the Association should employ all the means in its power to procure a Register of the Thermometer during every hour of the day and night, to be kept at some military or naval station in the South of England.

Note*. Until the phænomena and distribution of diurnal temperature are more thoroughly understood than at present, we can hardly hope that any very sure footing has been obtained in the study of meteorology. The hourly register kept for several years at the military station of Leith Fort, in lat. 56°, has shown that we want nothing but the combination of a sufficient number of trust-worthy observations, in order to obtain results of primary importance to the science, and which may one day enable us to arrive at the true form of the daily and annual curves of mean temperature with a precision almost mathematical. In order, however, to extend the benefit of such investigations, it is absolutely necessary that they should be pursued in different latitudes. The application to rendering available registers otherwise almost without value, from not being made at the proper hours, will be best illustrated by a reference to the account of the Leith observations. (Transactions of the Royal Society of Edinburgh, Vol. X.)

II. That the establishment of such an hourly meteorological register be pointed out as a highly interesting object, in reference especially to the important point of intertropical climate, to THE COMMITTEE OF THE ASSOCIATION IN INDIA.

III. That the Committee in India be requested to endeavour to institute such observations as may throw light on the phœnomena of the horary oscillations of the barometer, near the equator. Should the concurrence of the Committee on these points be obtained, it would probably be desirable that the Association should take measures for sending out delicate and accurate instruments.

IV. That Mr. Phillips and Mr. Wm. Gray, jun. of York, be requested to undertake a series of experiments on the comparative quantities of rain falling on the top of the great tower of York Minster, and on the ground near its base. The Committee have been induced to propose this specific question in consequence of the local fitness of the situation, and the facilities offered for its solution by the authorities; but it is to be wished that similar experiments should be made elsewhere,

* The notes appended to the Recommendations have been drawn up by some of the Members of the Committees since the Meeting.

D

[page] 50

that by an extended comparison of observations, light may be thrown upon the anomalies which have been observed at Paris and in other places.

V. That the Association should express its desire to receive a satisfactory exposition of the theory of the moistened bulb hygrometer, and that observers be also invited to institute series of comparative experiments on the indications of the moistened thermometer and the temperature of the dew point.

Note. These indications may be ascertained by Mr. Dalton's process, or by Mr. Daniell's Hygrometer, or by both. Notwithstanding the ingenious and laborious researches of Hutton, De Saussure, Leslie, Anderson, and Gay-Lussac upon this subject, scientific deductions drawn from more extended experiments are greatly wanted. The simplicity and certainty of the experiment by which the cold produced by the evaporation of water is measured, renders an accurate theory of the result peculiarly desirable. The experimenter would do well to consult Mr. Dalton's views on the theory of Hygrometry, contained in his Meteorological Essays, and in the Manchester Transactions, and to examine the investigations of Professor Leslie, (Relations of Heat and Moisture, and Supplement to the Encyclopædia Britannica, Article METEOROLOGY;) of Dr. Anderson (Edinburgh Encyclopœdia, Article HYGROMETER,) and of M. Gay-Lussac, (Biot, Traité de Physique, Tom. II.) A good series of observations at high temperatures will be found recorded in Nos. II. and III. of a Calcutta Journal, entitled, Gleanings in Science.

VI. That experiments on the Decrease of Temperature at increasing heights in the Atmosphere be recommended as an important subject for the contributions of observers.

Note. Series of observations for considerable periods of time on the mean temperature of the air at fixed hours, and at stations of which the difference of height has been accurately measured, are the most valuable. The best hours for observation are those which give most accurately the mean temperature of the period of observation. The hourly observations at Leith Fort have determined the hours which give the annual mean temperature in this country to be about 9¼ A.M. and 8½ P.M. Experimental balloons have lately been employed to assist the solution of this problem, which is one of the most interesting in Meteorology; but the investigation of it is nearly brought to a stand for want of sufficiently

[page] 51

numerous observations. The observer may be referred for information to Ramond, Mémoires sur la Formule Barometrique de la Méchanique Celeste; to the Researches of Humboldt; to Professor Leslie, Supplement to the Encyclopœdia Britannica, Article CLIMATE; to Pouillet, Elemens de Physique; to Mr. Atkinson's Paper on Refractions in the Memoirs of the Astronomical Society; and to Mr. Ivory's Memoir on the same subject in the Philosophical Transactions, and his Papers in the Annals of Philosophy.

VII. That the observation of the Temperature of Springs at different heights and depths should be pointed out as an object of great interest, in prosecuting which insulated inquirers may render essential aid to science.

Note. When springs are copious, a few observations in the course of the year suffice to give with great accuracy their mean temperature. The height of the springs above the mean level of the sea, and the depth of Artesian wells, should be carefully observed; and where the corresponding mean temperature of the air can be obtained, it should be stated. In two points of view these observations are important, independently of the inferences which they may furnish as to the decrease of heat in the atmosphere. The great interest attached to the phænomenon of the progressive increase of temperature of the globe, as we descend through the Strata, renders of value observations on the temperature of springs at considerable heights, of springs in mines, and of those brought to the surface from some depths by the process of boring. This question has been treated with great success by M. CORDIER, in several Memoirs, some of which have been translated into English. Again, the researches of Humboldt, Buch, Wahlenberg, and most recently Kupffer in a Memoir on Isogeothermal Lines, read before the Academy of St. Petersburg, in 1829, have shown that the temperature of the earth differs in many parts of the globe from that of the air, being generally in defect below lat. 56°, and in excess beyond it. The progressive increase of temperature with that of the depth in Artesian wells, and the deviation of the mean temperature of the Earth from that of the Air in different latitudes, have opened new fields for discussion; and by the zealous cooperation of observers cannot fail to present results, of which at present we can form but an imperfect idea.

D 2

[page] 52

Magnetism.

It appears to the Committee highly desirable that a series of observations upon the Intensity of Terrestrial Magnetism in various parts of England be made by some competent individual, similar to those which have recently been carried on in Scotland by Mr. Dunlop.

Should the Committee succeed in finding some individual ready to undertake the task, they propose that an application should be made to the Royal Society of Edinburgh, for permission to make use of the Standard Needle belonging to them, and constructed under the direction of Professor Hansteen of Christiania.

It appears to the Committee of considerable importance, that a certain number of observations should be made throughout Britain with the Dipping Needle, in order to reduce the Horizontal to the true Magnetic-Intensity.

Note. The time of three hundred vibrations should be observed, and the methods of observation and reduction should be the same as have been employed and described by Humboldt, Hansteen, and others.

Electro-Magnetism.

The Committee recommend, as an important subject for further prosecution, the examination of the Electro-Magnetic condition of Metalliferous Veins. The Committee would refer for the details of what has been already done upon this subject, to the Paper of Mr. Fox in the Philosophical Transactions for 1830; and would propose that the experiments should be extended to veins which traverse, as in some of our mines, horizontal and dissimilar strata.

Optics.

That Dr. Brewster be requested to prepare for the next Meeting a Report on the progress of Optical Science.

Acoustics.

That the Rev. Robert Willis be requested to prepare for the next Meeting a Report on the state of our knowledge concerning the phænomena of Sound, and the additions which have been recently made to it.

Heat.

That Professor Powell be requested to prepare for the next Meeting a similar Report respecting heat.

Electricity.

That Professor Cumming be requested to prepare for the

[page] 53

next Meeting a similar Report on Thermo-Electricity, and the allied subjects in which recent discoveries have been made.

CHEMICAL COMMITTEE.

It appears to the Committee of supreme importance, that Chemists should be enabled, by the most accurate experiments, to agree in the relative weights of the several elements, Hydrogen, Oxygen, and Azote, or, which amounts to the same thing, that the specific gravity of the three gases should be ascertained in such a way as would insure the reasonable assent of all competent and unprejudiced judges.

They think it highly desirable that the doubts which remain respecting the proportions of Azote, Oxygen, &c. in the atmosphere should be removed; that the proportions of Azote and Oxygen in nitrous gas and nitrous oxide should be strictly determined; and that the specific gravities of the compound gases in general should be more accurately investigated.

They recommend that the members of this Committee, and British Chemists in general, be invited to make experiments on these subjects, and communicate their results to the next Meeting at Oxford.

That Mr. Johnston be requested to present to the next Meeting a view of the recent progress of Chemical science, especially in foreign countries.

That Dr. Daubeny be requested to undertake an investigation into the sources from which organic bodies derive their fixed principles.

That Mr. Johnston be requested to undertake the inquiries which have been suggested to the Committee, into the comparative analysis of iron in the different stages of its manufacture.

That Mr. West be requested to pursue the experiments contemplated by him, into the combinations of gaseous bodies when passed through heated tubes.

That the Rev. W. Vernon Harcourt be requested to prosecute the inquiries contemplated by him, into the chemical phænomena from which the materiality of what are sometimes called ethereal substances has been inferred.

MlNERALOGICAL COMMITTEE.

The Committee recommend that the Rev. Professor Whewell be requested to present to the next Meeting a Report on the state and progress of Mineralogy.

[page] 54

GEOLOGICAL AND GEOGRAPHICAL COMMITTEE.

The Committee recommend that Geologists be requested to examine strictly into the truth of that part of the theory of M. Elie de Beaumont, in its application to England, Scotland, and Ireland, which asserts that the lines of disturbance of the strata assignable to the same age are parallel, and that a Report to the next Meeting on this subject should be procured.

That Mr. Phillips be requested to draw up, with such co-operation as he may procure, a systematic catalogue of all the organized fossils of Great Britain and Ireland, hitherto described, with such new species as he may have an opportunity of accurately examining, with notices of their localities and geological relations.

The Committee propose that Mr. Robert Stevenson, Civil Engineer, be requested to prepare a Report upon the waste and extension of the land on the East coast of Britain, and the question of the permanence of the relative level of the sea and land; and that individuals who can furnish observations, be requested to correspond with him on the subject*.

The importance which, especially of late years, has been attached to facts of this nature, in illustration of the sciences of hydrography and geology, and the mass of uncombined materials Which have recently been accumulating, have induced the Committee to make the present recommendation; and in doing so, it feels pleasure in being able to have in its view an individual whose practical acquaintance with the coast in general, and more particularly the minute survey made by him some years since, gives reason to expect from his Report much important and accurate information.

BOTANICAL COMMITTEE.

The Committee recommend that Professor Lindley be requested to prepare for the next Meeting an account of the principal questions recently settled, or at present agitated, in the philosophy of Botany, whether in this country or abroad.

That Botanists in all parts of Great Britain and Ireland be invited to compose and communicate to the Meetings of the Association, Catalogues of County or other local Floras, with indications of those species which have been recently introduced, of those which are rare or very local, and of those which

* Communications may be addressed to Robert Stevenson, Esq. Engineer to the Northern Lighthouse Board, Edinburgh.

[page] 55

thrive, or which have become or are becoming extinct; with such remarks as may be useful towards determining the connection which there may be between the habitats of particular plants, and the nature of the soil and the strata upon which they grow; with statements of the mean winter and summer temperature of the air and water at the highest as well as the lowest elevation at which species occur, the hygrometrical condition of the air, and any other information of an historical, (œonomical, and philosophical nature.

Note. If upon this plan a complete botanical survey of the British islands could be obtained, the results would be important when the Flora in the aggregate came to be compared with its relations of soil, climate, elevation, &c.

[page] 56

SCIENTIFIC TRANSACTIONS

OF

THE MEETING.

MONDAY EVENING.

MR. PHILLIPS, one of the Secretaries of the Yorkshire Philosophical Society, delivered an extemporaneous account of the most remarkable phænomena in the Geology of Yorkshire, illustrated by drawings and specimens selected from the Museum, and contributed by the visitors to the Meeting.

He observed, that though the principal design in opening the Museum that evening was to promote mutual acquaintance and friendly intercourse among those who were soon to engage in more important scientific labours, yet it was thought conducive to these objects that some observations should be offered by him from the Lecture Table, on the geological relations of the County in which they were assembled. In attempting, therefore, a rapid sketch of some of the more prominent and peculiar features in the Geology of Yorkshire, he was influenced by a natural desire to call the attention of the eminent individuals now assembled in York to the phænomena most worthy of observation in passing through the County. He should thus have the opportunity of illustrating the value of some remarkable specimens which within a few hours had arrived for the inspection of the Meeting, and offer the most appropriate welcome which the City and County of York, and the Institution they had founded for the advancement of science, could give to those who now came amongst them to lay the basis of a wide Association for the same important purpose.

The points embraced in the continuation of Mr. Phillips's address were the following:—

1. The peculiar character of the Carboniferous and Oolitic systems in Yorkshire,—both of these great systems of Calcareous Rocks being here diversified by large interpolations of sandy and argillaceous strata, with thin seams of Coal, and re-

[page] 57

mains of Plants. In both systems these interpolations thicken to the northwards. Thus the nearly undivided mass of limestone under Ingleborough becomes separated into many distinct calcareous beds, with sandstones, shales, and bad coal, in Teesdale, Tynedale, and Swaledale, which are still further modified by the introduction of coarse pebbly sandstones, and workable seams of coal, in the western and northern parts of Northumberland:—and the Oolites of Lincolnshire, diminished in thickness and debased in purity, are almost lost in several hundred feet of sandstone, shale, and coal, which form the north-eastern Moorlands of Yorkshire. Mr. Phillips referred these interpolations of sandstone, &c. to the originally littoral, or perhaps æstuary, situation of those parts of the calcareous deposit; while the thicker and more homogeneous limestone masses were probably produced under the deeper and more tranquil waters of the ancient oceans. The bearing of these deductions upon the important subject of the relative form and extent of the land and sea in this part of the globe at those periods respectively, was briefly illustrated.

2. The remarkable history of the deposit near Market Weighton (first observed by W. H. Dikes, Esq., Curator of the Hull Society, and afterwards more completely investigated by some of the members of the Yorkshire Philosophical Society), in which the bones of several kinds of quadrupeds, including species considered as extinct, were found mingled with many shells belonging to thirteen existing species of land, marsh, and fresh-water mollusca, and covered with gravel from the neighbouring hills, together with some larger stones from very distant localities.

3. The general character of the alluvial deposits, inclosing timber and many remains of quadrupeds, in the eastern part of Yorkshire, and the peculiar condition of some bones of deer obtained by Mr. W. Casson, from the Peat near Thorne. These bones appear to have been deprived of a large portion of their hardening earth, and are nearly in the state of leather,—quite flexible, and much altered from their original shape.

4. The traces of the action of the atmosphere in the rain channels which furrow the sides of the monumental stones of Boroughbridge, and form miniature valleys on the broad surfaces of the limestone scars on the mountains of Western Yorkshire and Westmoreland.

5. The occurrence of three specimens of unknown scaly fishes, with ferns and other fossil plants, in the ironstone bands in the lower part of the coal formation of Leeds and Bradford;

[page] 58

two of them indicating an individual of considerable size, the third smaller, and perhaps of a distinct species*.

TUESDAY MORNING.

This morning having been almost exclusively occupied in the business of forming the Association, the only communication read was the following extract of a letter from Geo. Harvey, Esq. F.R.S. L. & E.

"It was my intention, had I been able to enjoy the privilege of attending at York, to have drawn the attention of the Meeting to the very remarkable circumstance of the Geometrical Analysis of the ancients having been cultivated with eminent success in the northern counties of England, and particularly in Lancashire. The proofs of this may be gathered from a variety of periodical works, devoted almost exclusively to this lofty and abstract pursuit. I have now before me several exquisitely beautiful specimens of the geometry of the Greeks, produced by men in what, for distinction sake, we call the inferior conditions of life. The phænomenon (for such it truly is) has long appeared to me a remarkable one, and deserving of an attentive consideration. Playfair, in one of his admirable papers in the Edinburgh Review, expressed a fear that the increasing taste for analytical science would at length drive the ancient geometry from its favoured retreat in the British Isles; but, at the time he made this desponding remark, the Professor seemed not to be aware that there then existed a devoted band of men in the North, resolutely bound to the pure and ancient forms of geometry, who in the midst of the tumults of steam-engines, cultivated it with unyielding ardour, preserving the sacred fire under circumstances which would seem from their nature most calculated to extinguish it. In many modern Publications, and occasionally in the Senate-House Problems proposed to the Candidates for Honours at Cambridge, questions are to be met with derived from this humble but honourable source.

"The true cause of this remarkable phænomenon I have not been able clearly to trace. A taste for pure geometry, something like that for Entomology among the weavers of Spitalfields, may have been transmitted from father to son; but who was the distinguished individual first to create it, in the peculiar race of men here adverted to, seems not to be known.

* Since the Meeting, Mr. Phillips has had the opportunity of observing another specimen of a different species of fossil fish, in the possession of C. Rawson, Esq. from a still lower part of the coal strata at Halifax.

[page] 59

Surrounded with machinery, with the rich elements of mechanics in their most attractive forms, we should have imagined that a taste for mechanical combinations would have exclusively prevailed; and that inquiries locked up in the deep, and to them unapproachable, recesses of Plato, Pappus, Apollonius, and Euclid, would have met with but few cultivators. On the contrary, Porisms and Loci, Sections of Ratio and of Space, Inclinations and Tangencies,—subjects confined among the ancients to the very greatest minds, were here familiar to men whose condition in life was, to say the least, most unpropitious for the successful prosecution of such elevated and profound pursuits.

The contrast also between the Northern and Southern parts of England, in this particular, was most remarkable. In the latter the torch of geometry emitted but a feeble ray; while in the former it existed in its purest and most splendid form. The two great restorers of the ancient geometry, Mathew Stewart and Robert Simson, it may be observed, lived in Scotland. Did their proximity encourage the growth of this spirit? or were their writings cultivated by some teacher of a village school, who communicated by a method, which genius of a transcendental order knows so well how to employ, a taste for these sublime inquiries, so that at length they gradually worked their way to the anvil and the loom?"

TUESDAY EVENING.

MR. ABRAHAM delivered a Lecture on Magnetism, and particularly described several useful applications of this science, which he had employed for the advantage of the arts. He exhibited the model of a machine used for needle pointing, the labouring at which has been found so prejudicial to health, owing to the particles of steel inhaled during the process, that although the men were employed at it only six hours in the day, few ever attained the age of forty years, most dying at thirty or thirty-five, and several not surviving twenty-five. These deadly effects had been in a great measure obviated by Mr. Abraham's contrivance of placing several magnets around a mouth-piece, to attract the particles of steel as they came off in the process of grinding, or floated in the dusty atmosphere of the small apartments. This invention, for which the Society of Arts awarded their large gold medal, has not been so universally employed in the manufactories as its importance deserved, owing partly to the disinclination of the workmen to adopt methods which, by rendering their avocation less injurious to health, should lower the price of their labour.

[page] 60

Mr. Abraham exhibited another Magnetical Instrument, intended to guard the eyes of the grinders from the dispersion of fragments of steel, communicated several poles to the same magnetic bar, and detailed the method which he had found most effectual for communicating, combining, and increasing the magnetic influence.

He then exhibited his simple process for demagnetizing the steel balance-wheels of watches. Having dipped a balance-wheel, previously rendered magnetical, into iron filings, and thus discovered the situation of its poles,—he presented to one of these, at the distance of an inch, the similar pole of a small magnet. The filings immediately fell from the wheel, and it was found to be perfectly demagnetized. (Mr. Abraham's inventions having been presented to the Society of Arts, are described in their Transactions, Vol. XL. p. 135; Vol. XLIII. p. 48; Vol. XLIV. p. 19.)

WEDNESDAY MORNING.

DR. BREWSTER communicated a paper, which was read by Mr. Robison, presenting a general view of the progress of the science of mineralogy during the last thirty years, and of the principles of classification now adopted for minerals; and suggesting the propriety of adding to the four systems of crystallization now employed by Mohs and other mineralogists (the Rhomboidal, Pyramidal, Prismatic, and Tessular systems,) a fifth, viz. the Composite system, as combining a series of crystalline structures not included under the other heads, and mostly discovered by the agency of polarized light. This new system of crystallization, the Author proposes to divide into two classes, the first of which embraces those minerals in which the physical properties of the individual crystals are not altered by the combination; and the second, those minerals in which the physical properties of the individual crystals are altered by the combination. These classes were again divided into different orders, and the Composite minerals were enumerated, which the author proposed to place under each division.

The following Essay by DR. HENRY, was then read by Mr. Phillips.

An Estimate of the Philosophical Character of Dr. Priestley, by William Henry, M.D., F.R.S. &c. &c.

The principal source of the materials of the following pages, is the work, in which the discoveries of Dr. Priestley were originally announced to the public. It consists of six volumes

[page] 61

in octavo, which were published by him, at intervals between the years 1774 and 1786; the first three under the title of "Experiments and Observations on different kinds of Air;" and the last three under that of "Experiments and Observations relating to various Branches of Natural Philosophy, with a continuation of the Observations on Air." These volumes were afterwards methodized by himself, and compressed into three octavos, which were printed in 1790. As a record of facts, and as a book of reference, the systematized work is to be preferred. But as affording materials for the history of that department of science, which Dr. Priestley cultivated with such extraordinary success; and, still more, for estimating the value of his discoveries, and adjusting his station as an experimental philosopher, the simple narrative, which he originally gave in the order of time, supplies the amplest and the firmest ground-work.

In every thing that respects the history of this branch of experimental philosophy, the writings and researches of Dr. Priestley, to which I have alluded, are peculiarly instructive. They are distinguished by great merits, and by great defects; the latter of which are wholly undisguised by their author. He unveils, with perfect frankness, the whole process of reasoning, which led to his discoveries; he pretends to no more sagacity than belonged to him, and sometimes disclaims even that to which he was fairly entitled; he freely acknowledges his mistakes, and candidly confesses when his success was the result of accident, rather than of judicious anticipation; and by writing historically and analytically, he exhibits the progressive improvement of his views, from their first dawnings, to their final and distinct development. Now, with whatever delight we may contemplate a systematic arrangement, the materials of which have been judiciously selected, and from which every thing has been excluded, that is not essential to the harmony of the general design, yet there can be no question that as elucidating the operations of the human mind, and enabling us to trace and appreciate its powers of invention and discovery, the analytic method of writing has decided advantages.

To estimate, justly, the extent of Dr. Priestley's claim to philosophical reputation, it is necessary to take into account the state of our knowledge of gaseous chemistry, at the time when he began his inquiries. Without underrating what had been already done by Van Helmont, Ray, Hooke, Mayow, Boyle, Hales, Macbride, Black, Cavendish, and some others, Priestley may be safely affirmed to have entered upon a field, which, though not altogether untilled, had yet been very im-

[page] 62

perfectly prepared to yield the rich harvest, which he afterwards gathered from it. The very implements, with which he was to work, were for the most part to be invented; and of the merits of those, which he did invent, it is a sufficient proof that they continue in use to this day, with no very important modifications. All his contrivances for collecting, transferring, and preserving different kinds of air, and for submitting those airs to the action of solid and liquid substances, were exceedingly simple, beautiful, and effectual. They were chiefly, too, the work of his own hands, or were constructed under his directions by unskilled persons; for the class of ingenious artists, from whom the chemical philosopher now derives such valuable aid, had not then been called into existence by the demands of the science. With a very limited knowledge of the general principles of chemistry, and almost without practice in its most common manipulations;—restricted by a narrow income, and at first with little pecuniary assistance from others;—compelled, too, to devote a large portion of his time to other pressing occupations, he nevertheless surmounted all obstacles; and in the career of discovery, outstripped many, who had long been exclusively devoted to science, and were richly provided with all appliances and means for its advancement.

It is well known that the accident of living near a public brewery at Leeds, first directed the attention of Dr. Priestley to pneumatic chemistry, by casually presenting to his observation the appearances attending the extinction of lighted chips of wood, in the gas which floats over fermenting liquors. He remarked, that the smoke formed distinct clouds floating on the surface of the atmosphere of the vessel, and that this mixture of air and smoke, when thrown over the sides of the vat, fell to the ground; from whence he deduced the greater weight of this sort of air than of atmospheric air. He next found that water imbibes the new air, and again abandons it when boiled or frozen. These more obvious properties of fixed air having been ascertained, he extended his inquiries to its other qualities and relations; and was afterwards led by analogy to the discovery of various other gases, and to the investigation of their characteristic properties.

It would be inconsistent with the scope of this Essay to give a full catalogue of Dr. Priestley's discoveries, or to enumerate more of them, than are necessary to a just estimate of his philosophical habits and character. He was the unquestionable author of our first knowledge of oxygen gas, of nitrous oxide, of muriatic, sulphurous, and fluor acid gases, of ammoniacal gas, and of its condensation into a solid form by the acid gases?

[page] 63

Hydrogen gas was known before his time; but he greatly extended our acquaintance with its properties. Nitrous gas, barely discovered by Dr. Hales, was first investigated by Priestley, and applied by him to eudiometry. To the chemical history of the acids derived from nitre, he contributed a vast accession of original and most valuable facts. He seems to have been quite aware that those acids are essentially gaseous substances, and that they might be exhibited as such, provided a fluid could be found that is incapable of absorbing or acting upon them*. He obtained, and distinctly described†, the curious crystalline compound of sulphuric acid with the vapour of nitrous acid, or, more correctly, of sulphuric and hypo-nitrous acids, which, being of rare occurrence, was forgotten, and has since been rediscovered, like many other neglected anticipations of the same author. He greatly enlarged our knowledge of the important class of metals, and traced out many of their most interesting relations to oxygen and to acids. He unfolded, and illustrated by simple and beautiful experiments, distinct views of combustion; of the respiration of animals, both of the inferior and higher classes; of the changes produced in organized bodies by putrefaction, and of the causes that accelerate or retard that process; of the importance of azote as the characteristic ingredient of animal substances, obtainable by the action of dilute nitric acid on muscle and tendon; of the functions and œconomy of living vegetables; and of the relations and subserviency which exist between the animal and vegetable kingdoms. After trying, without effect, a variety of methods, by which he expected to purify air vitiated by the breathing of animals, he discovered that its purity was restored by the growth of living and healthy vegetables, freely exposed to the solar light.

It is impossible to account for these, and a variety of other discoveries, of less importance singly, but forming altogether a tribute to science, greatly exceeding, in richness and extent, that of any contemporary, without pronouncing that their author must have been furnished by nature with intellectual powers, far surpassing the common average of human endowments. If we examine, with which of its various faculties the mind of Dr. Priestley was most eminently gifted, it will, I believe, be found that it was most remarkable for clearness and quickness of apprehension, and for rapidity and extent of association. On these qualities were founded that apparently intuitive perception of analogies, and that happy facility of

* Series I. Vol. ii. p. 175.

† Series II. Vol. i. p. 26.

[page] 64

tracing and pursuing them through all their consequences, which led to several of his most brilliant discoveries. Of these analogies many were just and legitimate, and have stood the test of examination by the clearer light, since reflected upon them from the improved condition of science. But, in other cases, his analogies were fanciful and unfounded, and led him far astray from the path, which might have conducted him directly to truth. It is curious, however, as he himself observes, that in missing one thing, of which he was in search, he often found another of greater value. In such cases, his vigilance seldom failed to put him in full possession of the treasure upon which he had stumbled. Finding by experience, how much chance had to do with the success of his investigations, he resolved to multiply experiments, with the view of increasing the numerical probabilities of discovery. We find him confessing, on one occasion, that he "was led on, by a random expectation of some change or other taking place." In other instances, he was influenced by theoretical views of so flimsy a texture, that they were dispersed by the first appeal to experiment. "These mistakes," he observes, "it was in my power to have concealed; but I was determined to show how little mystery there is in the business of experimental philosophy; and with how little sagacity, discoveries, which some persons are pleased to consider great and wonderful, have been made." Candid acknowledgements of this kind were, however, turned against him by persons envious of his growing fame; and it was asserted that all his discoveries, when not the fruits of plagiarism, were "lucky guesses," or owing to mere chance*. Such detractors, however, could not have been aware of the great amount of credit that is due to the philosopher, who at once perceives the value of a casual observation, or of an unexpected result; who discriminates what facts are trivial, and what are important; and selects the latter, to guide him through difficult and perplexed mazes of investigation. In the words of D'Alembert, "Ces hazards ne sont que pour ceux qui jouent bien."

The talents and qualifications which are here represented as having characterized the mind of Dr. Priestley, though not of the rarest kind, or of the highest dignity, were yet such as admirably adapted him for improving chemical science at the time when he lived. What was then wanted, was a wider field of observation;—an enlarged sphere of chemical phænomena;— an acquaintance with a far greater number of individual bodies

* These charges, especially that of plagiarism, which had been unjustly advanced by some friends of Dr. Higgins, were triumphantly repelled by Dr. Priestley, in a pamphlet entitled," Philosophical Empiricism," published in 1775.

[page] 65

than were then known; from the properties of which, and from those of their combinations, tentative approximations to general principles might at first be deduced; to be confirmed or corrected, enlarged or circumscribed, by future experience. It would have retarded the progress of science, and put off, to a far distant day, that affluence of new facts, which Priestley so rapidly accumulated, if he had stopped to investigate, with painful and rigid precision, all the minute circumstances of temperature, of specific gravity, of absolute and relative weights, and of crystalline structure, on which the more exact science of our own times is firmly based, and from which its evidences must henceforward be derived. Nor could such refined investigations have then been carried on with any success, on account of the imperfection of philosophical instruments. It would have been fruitless, also, at that time, to have indulged in speculations respecting the ultimate constitution of bodies;—speculations that have no solid ground-work, except in a class of facts developed within the last thirty-five years, all tending to establish the laws of combination in definite and in multiple proportions, and to support the still more extensive generalization, which has been reared by the genius of Dalton.

It was, indeed, by the activity of his intellectual faculties, rather than by their reach or vigour, that Dr. Priestley was enabled to render such important services to natural science. We should look, in vain, in any thing that he has achieved, for demonstrations of that powerful and sustained attention, which enables the mind to institute close and accurate comparisons;— to trace resemblances that are far from obvious;—and to discriminate differences that are recondite and obscure. The analogies, which caught his observation, lay near the surface, and were eagerly and hastily pursued; often, indeed, beyond the boundaries, within which they ought to have been circumscribed. Quick as his mind was in the perception of resemblances, it appears (probably for that reason,) to have been little adapted for those profound and cautious abstractions, which supply the only solid foundations of general laws. In sober, patient, and successful induction, Priestley must yield the palm to many others, who, though far less fertile than himself in new and happy combinations of thought, surpassed him in the use of a searching and rigorous logic; in the art of advancing, by secure steps, from phænomena to general conclusions;—and again in the employment of general axioms as the instruments of further discoveries.

Among the defects of his philosophical habits, may be remarked, that he frequently pursued an object of inquiry too

E

[page] 66

exclusively, neglecting others, which were necessarily connected with it, and which, if investigated, would have thrown great light on the main research. As an instance, may be mentioned his omitting to examine the relation of gases to water. This relation, of which he had indistinct glimpses, was a source of perpetual embarrassment to him, and led him to imagine changes in the intimate constitution of gases, which were in fact due to nothing more than an interchange of place between the gas in the water and that above the water, or between the former and the external atmosphere. Thus he erroneously supposed that hydrogen gas was transmuted into azotic gas, by remaining long confined by the water of a pneumatic cistern. The same eager direction of his mind to a single object, caused him, also, to overlook several new substances, which he must necessarily have obtained, and which, by a more watchful care, he might have secured and identified. At a very early period of his inquiries (viz. before November, 1771), he was in possession of oxygen gas from saltpetre, and had remarked its striking effect on the flame of a candle; but he pursued the subject no further until August 1774, when he again procured the same kind of gas from the red oxide of mercury, and, in a less pure state, from red lead. Placed thus a second time within his grasp, he did not omit to make prize of this, his greatest, discovery. He must also have obtained chlorine by the solution of manganese in spirit of salt; but it escaped his notice, because, being received over mercury, the gas was instantly absorbed*. If he had employed a bladder, as Scheele afterwards did, to collect the product of the same materials, he could not have failed to anticipate the Swedish philosopher, in a discovery not less important than that of oxygen gas. Carbonic oxide early and repeatedly presented itself to his observation, without his being aware of its true distinctions from other kinds of inflammable air; and it was reserved for Mr. Cruickshank of Woolwich to unfold its real nature and characters. It is remarkable, also, that in various parts of his works, Dr. Priestley has stated facts, that might have given him a hint of the law, since unfolded by the sagacity of M. Gay-Lussac, "that gaseous substances combine in definite volumes." He shows that

1 measure of fixed air unites with 1 6/7 measure of alkaline air,

1 measure of sulphurous acid with 2 measures of do.

1 measure of fluor acid with 2 measures of do.

1 measure of oxygen gas with 2 measures nitrous, very nearly;

* Series II. p. 253.

[page] 67

and that by the decomposition of 1 volume of ammonia, 3 volumes of hydrogen are evolved.

Let not, however, failures such as these, to reap all that was within his compass, derogate more than their due share from the merits of Dr. Priestley; for they may be traced to that very ardour of temperament, which, though to a certain degree a disqualification for close and correct observation, was the vital and sustaining principle of his zealous devotion to the pursuit of scientific truth. Let it be remembered, that philosophers of the loftiest pretensions are chargeable with similar oversights;—-that even Kepler and Newton overlooked discoveries, upon the very confines of which they trod, but which they left to confer glory on the names of less illustrious followers.

Of the general correctness of Dr. Priestley's experiments, it is but justice to him to speak with decided approbation. In some instances, it must be acknowledged, that his results have been rectified by subsequent inquirers, chiefly as respects quantities and proportions. But of the immense number of new facts originating with him, it is surprising how very few are at variance with recent and correct observations. Even in these few examples, his errors may be traced to causes connected with the actual condition of science at the time; sometimes to the use of impure substances, or to the imperfection of his instruments of research; but never to carelessness of inquiry or negligence of truth. Nor was he more remarkable for the zeal with which he sought satisfactory evidence, than for the fidelity with which he reported it. In no one instance is he chargeable with mis-stating, or even with straining or colouring, a fact to suit an hypothesis. And though this praise may, doubtless, be conceded to the great majority of experimental philosophers, yet Dr. Priestley was singularly exempt from that disposition to view phænomena through a coloured medium, which sometimes steals imperceptibly over minds of the greatest general probity. This security he owed to his freedom from all undue attachment to hypotheses, and to the facility with which he was accustomed to frame and abandon them;—a facility resulting not from habit only, but from principle. "Hypotheses," he pronounces, in one place, "to be a cheap commodity;" in another to be "of no value except as the parents of facts;" and so far as he was himself concerned, he exhorts his readers "to consider new facts only as discoveries, and to draw conclusions for themselves." The only exception to this general praise is to be found in the pertinacity with which he adhered, to the last, to the Stahlian hypothesis of phlogiston;

E 2

[page] 68

and in the anxiety which he evinced to reconcile to it new phænomena, which were considered by almost all other philosophers as proofs of its utter unsoundness. But this anxiety, it must be remembered, was chiefly apparent at a period of life when most men feel a reluctance to change the principle of arrangement, by which they have been long accustomed to class the multifarious particulars of their knowledge.

In all those feelings and habits that connect the purest morals with the highest philosophy (and that there is such a connection no one can doubt), Dr. Priestley is entitled to unqualified esteem and admiration. Attached to science by the most generous motives, he pursued it with an entire disregard to his own peculiar interests. He neither sought, nor accepted when offered, any pecuniary aid in his philosophical pursuits, that did not leave him in possession of the most complete independence of thought and of action. Free from all little jealousies of contemporaries or rivals, he earnestly invited other labourers into the field which he was cultivating; gave publicity in his own volumes to their experiments; and, with true candour, was as ready to record the evidence which contradicted, as that which confirmed, his own views and results. Every hint, which he had derived from the writings or conversation of others, was unreservedly acknowledged. As the best way of accelerating the progress of science, he recommended and practised the early publication of all discoveries; though quite aware that, in his own case, more durable fame would often have resulted from a delayed and more finished performance. "Those persons," he remarks, "are very properly disappointed, who, for the sake of a little more reputation, delay publishing their discoveries till they are anticipated by others."

In perfect consistency with that liberality of temper which has been ascribed to Dr. Priestley, it may be remarked also, that he took the most enlarged views of the scope and objects of Natural Science. In various passages of his works he has enforced, with warm and impressive eloquence, the considerations that flow from the contemplation of those arrangements in the natural world, which are not only perfect in themselves, but are essential parts of one grand and harmonious design. He strenuously recommends experimental philosophy as an agreeable relief from employments, that excite the feelings or overstrain the attention; and he proposes it to the young, the high-born, and the affluent, as a source of pleasure unalloyed with the anxieties and agitations of public life. He regarded the benefits of its investigations, not merely as issuing in the acquirement of new facts, however striking and valuable; nor

[page] 69

yet in the deduction of general principles, however sound and important; but as having a necessary tendency to increase the intellectual power and energy of man, and to exalt human nature to the highest dignity, of which it is susceptible. The springs of such inquiries he represents as inexhaustible; and the prospects that may be gained by successive advances in knowledge, as in themselves "truly sublime and glorious."

Into our estimate of the intellectual character of an individual, the extent and the comprehensiveness of his studies must always enter as an essential element. Of Dr. Priestley it may be justly affirmed, that few men have taken a wider range over the vast and diversified field of human knowledge. In devoting, through the greater part of his life, a large portion of his attention to theological pursuits, he fulfilled what he strongly felt to be his primary duty as a minister of religion. This is not the fit occasion to pronounce an opinion of the fruits of those inquiries, related as they are to topics, which still continue to be agitated as matters of earnest controversy. In Ethics, in Metaphysics, in the philosophy of Language, and in that of General History, he expatiated largely. He has given particular histories of the Sciences of Electricity and of Optics, characterized by strict impartiality, and by great perspicuity of language and arrangement. Of the Mathematics, he appears to have had only a general or elementary knowledge; nor, perhaps, did the original qualities, or acquired habits, of his mind fit him to excel in the exact sciences. On the whole, though Dr. Priestley may have been surpassed by many in vigour of understanding and capacity for profound research, yet it would be difficult to produce an instance of a writer more eminent for the variety and versatility of his talents, or more meritorious for their zealous, unwearied, and productive employment.

APPENDIX.

Since the foregoing pages were written, I have added a few remarks on a passage contained in a recent work of Victor Cousin, in which that writer has committed a material error as to the origin of Dr. Priestley's philosophical discoveries. "La chimie," he observes, "est une création du dixhuitième siècle, une creation de la France; c'est I'Europe entière qui a appelé chimie Francaise le mouvement qui a imprimé à cette belle science une impulsion si forte et une direction si sage; c'est à I'exemple et sur les traces de Lavoisier, de Guyton, de Fourcroy, de Berthollet, de Vauquelin, que se sont formes et que marchent encore les grands chimistes étrangers, ici Priestley et

[page] 70

Davy; là Klaproth et Berzelius." (Cours de l'Histoire de la Philosophie, tom. i. p. 25.)

It is to be lamented that so enlightened a writer as Victor Cousin, yielding, in this instance, to the seduction of national vanity, should have advanced pretensions in behalf of his countrymen, which have no foundation in truth or justice. Nothing can be more absurd or unprofitable than to claim honours in science, either for individuals or for nations, the title to which may be at once set aside by an appeal to public and authentic records.

It was in England, not in France, that the first decided advances were made in our knowledge of elastic fluids. To say nothing of anterior writers, Dr. Black had traced the causticity acquired by alkalies, and by certain earths, to their being freed from combination with fixed air; and Mr. Cavendish, in 1766, had enlarged our knowledge of that gas and of inflammable air. In England, the value of these discoveries was fully appreciated; in France, little or no attention was paid to them, till the philosophers of that country were roused by the striking phænomena exhibited by the experiments of Priestley. Lavoisier, it is true, had been led, by an examination of evidence derived from previous writers, to discard the hypothesis of phlogiston. The discovery of oxygen gas by Dr. Priestley not only completed the demonstration of its fallacy, but served as the corner-stone of a more sound and consistent theory. By a series of researches executed at great expense, and with consummate skill, the French philosopher verified in some cases, and corrected in others, the results of his predecessors, and added new and important observations of his own. Upon these, united, he founded that beautiful system of general laws, chiefly relating to the absorption of oxygen by combustible bodies, and to the constitution of acids, to which, alone, the epithet of the Antiphlogistic or French theory of chemistry is properly applied. Of the genius manifested in the construction of that system, and the taste apparent in its exposition, it is scarcely possible to speak with too much praise. But it is inverting the order of time to assert, that it had any share in giving origin to the researches of Priestley, which were not only anterior to the French theory, but were carried on under the influence of precisely opposite views. This, too, may be asserted of the discoveries of Scheele, who, at the same period with Dr. Priestley, was following, in a distant part of Europe, a scarcely less illustrious career.

It is the natural progress of most generalizations in science, that at first too hasty and comprehensive, they require to be

[page] 71

narrowed as new facts arise. This has happened to the theory of Lavoisier, in consequence of its having been discovered that combustion is not necessarily accompanied with an absorption of oxygen, and that acids exist independently of oxygen, regarded by him as the general acidifying principle. But after all the deductions that can justly be made on that account from the merits of Lavoisier, he must still hold one of the highest places among those illustrious men, who have advanced chemistry to its present rank among the physical sciences. It is deeply to be lamented that his fame, otherwise unsullied, should have been stained by his want of candour and justice to Dr. Priestley, in appropriating to himself the discovery of oxygen gas. This charge, often preferred and never answered, would not have been revived in this place, but for the claim so recently and indiscreetly advanced by M. Victor Cousin. To the credit of Dr. Priestley it may be observed, that in asserting his own right, he exercised more forbearance than could reasonably have been expected under such circumstances. In an unpublished letter to a friend, he thus alludes to the subject of M. Lavoisier's plagiarism. "He" (M. Lavoisier) "is an Intendant of the Finances, and has much public business, but finds leisure for various philosophical pursuits, for which he is exceedingly well qualified. He ought to have acknowledged that my giving him an account of the air I had got from mercurius calcinatus, and buying a quantity of M. Cadet while I was at Paris, led him to try what air it yielded, which he did presently after I left. I have, however, barely hinted at this in my second volume*." The communication alluded to was made by Dr. Priestley to M. Lavoisier in October, 1774; and the Memoir, in which the latter assumes to himself the discovery that mercurius calcinatus (red oxide of mercury) affords oxygen gas when distilled per se, was not read to the Academy of Sciences before April, 1775†. In evincing so little irritability about his own claim, and leaving its vindication with calm and just confidence to posterity, the English philosopher has lost nothing of the honour of that discovery which is now awarded to him, by men of science of every country, as solely and undividedly his own.

WEDNESDAY EVENING.

MR. R. POTTER, Jun. read a description of his new construction of Sir Isaac Newton's reflecting microscope, and exhibited

* Letter to the late Mr. Henry, dated Calne, Dec. 31, 1775.

† See an Abstract of this Memoir in the Journal de Rozier, Mai, 1775.

[page] 72

the instrument, with a variety of finely executed elliptical mirrors, &c. In the reflecting microscope of Sir Isaac Newton, the object is placed directly in the focus of the speculum, and the image is formed in that of the eye-glass; and thus, having only one additional surface in the essential parts of the instrument, it must be considered as next in simplicity to the singlelens microscope.

In this construction the object must be placed in the axis of the tube, where it is difficult to provide sufficient illumination, and it is this defect which the new construction is intended to obviate. A large hole is cut in the tube between the object and the speculum, to allow the light to fall upon the former when it requires to be viewed as an opake object, and all the lower parts of the tube are lined with black velvet to absorb the irregular light. A large lens is also occasionally employed to concentrate the light. Transparent objects require a small oval mirror to be placed immediately behind them; this mirror receiving a concentrated light from a lens, fixed in a sliding piece on the side of the tube, reflects it through the object to the speculum. The objects, to be placed in the centre of the tube, are attached to thin brass wires in wooden handles, and kept separately in a box.

This construction of the reflecting microscope has a great advantage in point of distinctness, from there being only one necessary reflection between the object and the image, and will be found particularly suitable for the examination of opake objects, on account of the large aperture of the speculum, compared to its focal length.—It is therefore recommended as an excellent working tool to the scientific inquirer, who will disregard the little trouble required in its management.

The Secretary then read a description by Dr. Brewster of an Instrument for distinguishing Precious Stones and Minerals. The object of this instrument is to distinguish mineral bodies by the relative quantity and colour of the light reflected from their surfaces, when placed in contact with fluids of different refractive powers. The surfaces employed for this purpose may be either natural or artificial, so that the method is equally applicable to regular crystals, and to gems cut into artificial forms. If a fluid, of a given refractive and dispersive power, is placed on the surface of a mineral of the very same refractive and dispersive power, there will be no light whatever reflected from their separating surface; but in proportion as the fluid and the solid differ in these respects, in the same proportion will the quantities of light differ which are reflected at

[page] 73

the separating surface, and its colour will undergo corresponding changes.

The principal part of the instrument is a triangular prism of glass, between the lower surface of which and the upper surface of the mineral the oil is placed. This oil will form a parallel film, but, by the mechanism of the instrument, the two surfaces which, bound this film can be inclined to each other, so that an eye looking into the prism will see at once the images of a luminous body, such as the sun, &c., reflected from the separating surface of the oil and the prism, and from that of the oil and the mineral. The first of these images is constant both in its colour and quantity of light, while the oil is the same, but the second will vary with the mineral. The comparison of the colour and quantity of light obtained from different minerals furnishes the nicest tests for discriminating them.

The author illustrated his explanations of the principle of the instrument by means of diagrams, and the instrument itself, as constructed by Dollond, was exhibited to the Meeting.

THURSDAY MORNING.

MR. DALTON read a paper written for the Literary and Philosophical Society of Manchester, containing a series of expertments on the quantity of food, taken by a person in health, compared with the quantity of the different secretions; with chemical remarks and deductions.

Mr. Dalton, whose regular habits of life and uniform good health enabled him to make these Experiments upon himself to great advantage, commenced them about 40 years since at Kendal, and prosecuted them for periods of a week or a fortnight at various seasons of the year, to ascertain the proportion beween the weight of food, and the ordinary evacuations. Particular observations were made on the effects occasioned by drinking an infusion of Carbonate of Potash, and a train of experiments was continued for three weeks to determine the loss of weight by perspiration for the whole day, and for certain hours of the morning, afternoon, and night. The mean daily loss by perspiration was 37½ oz.

From these experiments, the state of organic chemistry 40 years since did not permit Mr. Dalton to make any deductions, but he was now enabled to return to the subject with the powerful aid of exact analysis. He showed that the quantity of Carbon contained in the solid and liquid food taken into the

[page] 74

stomach daily was 11½ oz.
of which there passes off sensibly 1
Leaving for the waste by insensible perspiration 10½ oz.

Mr. Dalton had ascertained from experiments on his own respiration that the quantity of Carbonic acid gas expelled from his lungs contained of Carbon 10¼ oz.

His daily loss by perspiration of aqueous vapour from the lungs was at the same time found to be 20½ oz. to which adding 10¼ oz. Carbon, we have the total loss by perspiration from the lungs, 30¾ oz., which taken from 37½, leaves 6 ¾ oz. per day for the insensible perspiration of the skin, of which 6½ oz. are water, and ¼ oz. is Carbon.

The element Azote, of which 1½ oz. per day was taken into the stomach, appears to have passed off by evacuation. Of the 6 lbs. of aliment taken in a day, 1 lb. consists of Azote and Carbon, and 5 lbs. of water, and nearly the whole quantity of food taken into the stomach enters the circulation,—the residual part constituting only 1/18 of the whole,—of which about half is thrown off by the kidneys, more or less according to season and climate, another part passes off by insensible perspiration, 5/6 being perspired from the lungs, and 1/6 from the skin.

MR. R. POTTER read a paper containing remarks on a theory of the late M. Fresnel concerning the reflection of light from the surfaces of bodies.

M. Fresnel in a paper read before the Academy of Sciences in Paris, and of which he published an abstract in the Annales de Chimie for 1820, proposed to account for the reflection of light at the surfaces of bodies, on the undulatory theory, by its impinging on the ponderable particles. He appears to have afterwards in some measure modified his views, but not, to the writer's knowledge, ever to have formally renounced his former proposition. Hence the subject may fairly be considered as still open to discussion: and the manner of considering reflection, as caused by the light striking the ponderable parts of bodies, being the one which almost every person would recur to, on first commencing the study of physical optics, it will perhaps be considered not entirely useless, on this account also, to enter into a regular examination of M. Fresnel's hypothesis.

Now if reflection were caused immediately by the ponderable matter in any surfaces, it ought to be some function of the quantity of matter in the bodies furnishing such surfaces; but even a superficial view of the small quantity of light reflected

[page] 75

at a perpendicular incidence in transparent bodies, compared with the large quantity reflected at the same incidence in metals, is sufficient to convince us that reflection has no relation to the densities of bodies. To remove, however, an objection which might be urged, that the extension of the particles of bodies may not bear an invariable ratio to their weight, it will be necessary to examine cases where a metal, by combining with a new element or elements, has acquired the property of transparency, and thus possesses an evident refractive power. By knowing the comparative weights of the metal in the two states, it is easy to calculate the relative numbers of similar particles in equal surfaces, and of course to calculate the relative quantities of light which ought to be reflected, if caused only by the ponderable particles of metal. Experience is so much at variance with the hypothesis under examination, that the other elements in the compound may be considered even as lending no assistance at all.

The results obtained by Photometry show that the metals, with the exception, perhaps, of two or three, reflect two thirds and upwards of the light incident perpendicularly on them. For the reflective power of the transparent bodies we may use the analytical formula of M. Poisson, (which was admitted by M. Fresnel,) to calculate it from the refractive index, though it gives most probably, in all cases, quantities too large and of course proportionally favourable to the controverted hypothesis.

Proceeding in this manner for glass of Antimony, the reflection, according to Fresnel's hypothesis, should have been at least 46 rays of every 100 incident, whilst the quantity given by the analytical formula is only 19·3 rays.

In the white oxide of arsenic or arsenious acid the reflection should have been 31·9 (taking even the old number for the specific gravity of the metal,) in place of less than 8·3, which it really is.

In the red silver ore it should have been at least 37·5 rays if the hypothesis were correct, instead of less than 19·2 as determined by the analytical formula.

If the metals of the alkalies and earths might be assumed of equal reflective powers with the other metals, and it is most likely they are, the chloride of sodium would form one of the strongest cases which could be brought forward; for whilst it really reflects only about two per cent., it ought, according to the controverted hypothesis, to have reflected upwards of 60 rays of every 100 incident, from the metal in the chloride being almost as dense as in its proper state.

[page] 76

MR. WM. HUTTON read a paper upon the Whin Sill of Cumberland and Northumberland.

A bed of Stratiform Basalt is well known to occur extensively in connection with the Mountain Limestone Rocks of the North of England, and is called in Alston Moor and the adjoining mining districts, "the Whin Sill." This bed has naturally a good deal of geological interest attached to it, from the circumstance that rocks of its class are generally found under conditions which indicate that their production is entirely independent both as to antiquity and origin of that of the Strata which they divide,—and upon which, at the points of contact, they have produced mechanical and chemical changes which afford the most conclusive evidence of their violent intrusion since the deposition and consolidation of those Strata.

The Whin Sill is visible in many of the streams running into the South Tyne from the West, and may be seen in the bed of the Tyne itself at Tyne-head. It occurs in the bed of the Wear, in Teesdale, where it is extensively developed, and in the Lune; in short, throughout the whole district wherever the water-courses, or the operations of the miner, pierce deep enough: its basset-edge may also be traced almost uninterruptedly from Helton in Westmoreland to Tindale Fell in Northumberland.—Here the whole carboniferous formation is broken through by the "great Stublick Dyke," which throws down the Whin Sill along with the other beds of the formation to an immense depth; its edge reappears on the north side of this dyke at Wall Town Crags, near Glenwhelt, in Northumberland, rising rapidly to the North, and from this spot it can be traced almost throughout the county of Northumberland to the sea-coast near Newton; it occurs again with other beds of the carboniferous formation in consequence of a general depression of the strata a little south of Bamborough, from whence it sweeps round by Belford to Kyloe on the coast, near to which place it finally disappears.

In the course of this bed northward from Alston Moor, it appears to rise in the series of strata, from 'the putting in' of new beds, as the miners term it; and of course it is found in contact with all the varieties of rock composing the carboniferous formation. It is generally in one bed, sometimes in two, and once at least it occurs in three beds.

The action of heat in hardening the rocks near it and rendering the limestone crystalline, can generally be observed accompanying this bed, but nowhere to such an extent as in High Teesdale.

After an attentive examination of the appearances exhibited

[page] 77

throughout its whole course, the conclusion of the writer is, that this bed of Basalt was produced by an overflowing of lava during the deposition of the Mountain Limestone Group, after those beds, which are found below, and prior to those above it.

After Mr. Hutton's paper had been read,

MR. MURCHISON rose to bear testimony to the general value and accuracy of the memoir. His own observations, however, on the evident violence which in High Teesdale accompanied the arrangement of the basaltic matter, the altered character of the limestone, sandstone, and shale both above and below it, and the occasional ramification of its substance through the contiguous and superior Strata, led him to confirm the opinion of Professor Sedgwick, that the Whin of that district had not been injected into the Carboniferous Limestone till after the deposition of that whole system of rocks. He thought it very desirable that the views of Professor Sedgwick respecting the Whin Dykes of Durham should be further pursued, with reference to this theory, to ascertain whether they were emanations from the great Whin Sill, or were posterior to it. Some of these Dykes break off into various branches, all pointing to the Whin Sill, and thus appearing at least to be related to it in age.

Mr. Phillips had formerly examined the whole range of the Whin Sill, and was happy in being able to agree in opinion with both the author of the paper, and the President of the Geological Society. The definite geological situation, between the same Limestone bands, of the great portion of the Basalt, its wide lateral extension, the general limitation of the effects of its heat upon the contiguous rocks to the lower surface of the mass, its course for miles together without furnishing a single dyke, or even entering at all into the many natural fissures of the Limestone, and its division by metallic veins, obliged him to infer that a large portion of the Whin Sill was formed by periodical submarine eruptions of Lava, at intervals during the deposition of the Carboniferous Strata with which it is associated. On the other hand, the instances described of violent eruption and local intrusion of the Basalt into the Strata above its general range, seemed to show that Teesdale had been the theatre of more than one such eruption. The views of Professor Sedgwick and Mr. Hutton were, therefore, by no means irreconcileable; and it might be very possible hereafter to fix upon the foci or centres from which, as probably at Caldron Snout in Teesdale, the principal basaltic coulees had flowed.

[page] 78

MR. JOHNSTON gave an account of the Metal Vanadium and its ores.

He stated that this metal was first observed, though without being distinctly made out, by Del Rio in Mexico 25 years ago, and afterwards rediscovered by Seftström, and nearly at the same time by himself, about the end of last year. He exhibited and described the mineral from Wanlock-head, in which he found it, detailed the process for extracting it, enumerated its most interesting properties when in the metallic state, and the characters by which in all its states it may be distinguished from Chromium, the only other metal with which the analytical chemist is likely to confound it. He exhibited various compounds and salts of the metal, among which were some beautiful crystals of Vanadic Acid, which were transparent, of a brown colour by transmitted light, but reflecting a purplish tint. They were in the form of flat prisms of a high degree of lustre, and had been ascertained by Dr. Brewster to possess a refractive power approaching that of the diamond, to have at least one axis of double refraction, and to belong to the prismatic system of Mohs.

MR. WITHAM read a paper on the general results of botanical investigation concerning the character of the Ancient Flora, which by its decomposition has furnished the materials of coalseams. He described the discoveries which the art of slicing and polishing the fossil stems of plants had enabled him to make, concerning the internal structure of these large coniferous trees which especially abound in the lower part of the Carboniferous Series of Berwickshire; and stated that Geologists are now agreed that the plants of these ancient periods are of more diversified and complicated types than a distinguished foreign writer supposed.

The following Notice of a fact observed in the torrcfaction of Yellow Copper Pyrites at Amlwch, in Anglesey, by DR. HENRY, was read by the Secretary.

"When on a visit, a few years ago, to the Copper-Mines and Works at Amlwch in Anglesey, a fact was mentioned to me by an intelligent superintendent of the processes carried on there (Mr. Joseph Jones), which struck me to be interesting and curious. The poorer part of the ore (a native mixture of yellow copper pyrites with so much foreign matter as to contain only about 5 per cent. of copper,) is roasted in kilns on the spot, in order to expel a considerable part of the sulphur. The combustion, after the kiln has been once lighted, is supported by

[page] 79

the inflammable matter of the ore itself, and a smouldering heat, never, I was assured, sufficient to occasion fusion, is kept up for several months. On examining the lumps of roasted ore, small nodules are observed of an iron grey colour, with some lustre, resembling in appearance the vitreous copper ore (cuivre sulfuré of Haüy). These nodules have been found, when assayed, to be very much richer in copper than the original ore. Their specific gravity I found to be 4·6 very nearly that of vitreous copper ore. By a few general experiments, I ascertained that they are not entirely soluble in heated nitric acid, and that the solution contains a small proportion of peroxide of iron, with a much larger one of oxide of copper. The yellow copper pyrites, in its natural state, was determined by Mr. Richard Phillips, to consist of

2 atoms of protosulphuret of iron,

1 atom of bisulphuret of copper.

It should appear, therefore, that during torrefaction the bisulphuret of copper, by parting with one atom of sulphur, is converted into protosulphuret, which, by its aggregative attraction, is collected into small nodules. This fact furnishes an additional example, to the few which were before known, of changes of molecular arrangement in bodies heated below their point of fusion; with this further circumstance, that the attraction, which causes the aggregation of the particles, is sufficient to overcome the obstacle of interposed matter of a different kind. The only other instances of similar facts that at present occur to me, are presented by crystallites; and by the products of Mr. Gregory Watt's experiments on Basalt, some of the appearances of which he supposes to have taken place after the fused mass had returned to a solid state.

In the second volume of Breislac's Institutions Geologiques, I was pleased to observe, two or three years after the foregoing fact had occurred to me, that a precisely similar observation had been made by Brocchi on the roasted copper ore of Agardo, which is also the yellow pyrites, and of quality, as to its proportion of copper, not exceeding that of the Parys Mountain. On breaking the lumps of roasted ore, similar nodules were observed; and these, when assayed, were found to contain two thirds of their weight of copper, while the surrounding ore had lost greatly of its original proportion of that metal. In the central parts of some of those nodules, small fibres and plates of metallic copper were visible, an appearance which I have not observed in the roasted ore of Anglesey. The nodules, thus enriched in their proportion of metal, are picked out, and subjected to reducing processes.

M. Breislac adds, that the torrefaction of the ore, at Agardo,

[page] 80

is carried on with the careful avoidance of such a heat as would occasion the fusion of the ore; for this is ascertained to be very injurious to the subsequent operation."

It was remarked by Mr. Johnston, that a similar observation had been made some years ago by Professor Brodberg, of the school of Mines at Fahlun, and was detailed by him in a paper published in the Swedish Transactions, and reprinted in the Edinburgh Journal of Science.

THURSDAY EVENING.

On Thursday Evening MR. SCORESBY gave An exposition of some of the laws and phœnomena of Magnetic Induction, and of the mutual influences of magnets on each other, with an account of a method of application of the magnetic influences for the determination of the thickness of solid substances not otherwise measurable.—In the introductory observations, Mr. Scoresby considered the general nature, as far as it is understood, of the magnetic principle, and described a magnetic bar as a battery of magnetic particles, the arrangement of which being regular and consistent, transmits to the poles, like the galvanic pile, the general aggregate of their individual energies. He defined induced magnetism, "as the development of the latent magnetism in iron or steel by the juxtaposition of any substance in a magnetic condition." His investigations on the law of induced magnetism extended to the different qualities of iron and steel; to the proportion of influence acquired by similar masses at different distances from the proximate magnet; and to the relation of capacity in masses in all other respects similar except as to thickness. The proportion of influence was shown, at different distances, of the magnetism induced upon the nearer and more remote ends of a bar of soft iron, and the quantity transmitted, compared with the portion directly induced, was numerically stated. A bar of very soft iron, placed directly over a magnet of similar dimensions, was found at the distance of 5 inches, to acquire 1/30th of the power of the magnet. At 4 inches above, the inductive influence of the magnet was 1/19th of its own power; at 3 inches, 1/13th; at 2 inches, 1/7th; at 1 inch, ¼th. At ¼ of an inch above, the power induced was equal to ½ that of the magnet, and at the distance of 1/16th, it amounted to 2/3rds. Several new and curious illustrations of the phænomena of induced magnetism were then exhibited to the Meeting.

FRIDAY MORNING.

MR. SCORESBY concluded his account of his Magnetical experiments. He described the action of magnets of different

[page] 81

dimensions on the needle of the compass, and the investigation of the law of the deviation at different distances, and with magnets of various sizes. With a pair of three-feet bar magnets, he was able to produce a very perceptible action on the compass, through a variety of intervening solid substances, at a distance of more than 61 feet; and to measure with tolerable precision various masses of solid rock of from 3 feet to more than 40 feet in thickness by the magnetic deviations. The phænomena now communicated to the Association, with their different important applications, were the results of original investigations and discoveries, accomplished, for the most part, within the last ten months.

A paper by DR. BREWSTER on the structure of the crystalline lens in fishes, birds, reptiles, and quadrupeds, was read by the Secretary, and illustrated with drawings and models by the author. After giving an account of the previous observations of Leeuwenhoek and others, the author explained the method in which he conducted his inquiries. The lenses of almost all animals are composed of distinct fibres, and when any of the laminæ are removed, the surface appears fibrous or grooved. In large lenses, the direction of these lines may be easily traced by the microscope alone, but in many cases this is quite impracticable. In order to get rid of this difficulty, the author observed the image of a candle or bright luminous object, when reflected from a fresh surface of the lens, and he found this colourless image invariably accompanied by coloured images on each side, as in mother-of-pearl, and in Mr. Barton's Iris ornaments. As the direction of the fibres is necessarily perpendicular to the line joining these coloured images, and as the distance of the coloured images varies inversely with the diameters of the fibres, Dr. Brewster was able to trace these fibres to their points of convergency or terminations, even when the fibres themselves were no longer visible. When the crystalline lens is dried, a furrowed impression of its surface may be taken upon wax, and the impression will, like that from mother-of-pearl, exhibit the same coloured images.

By the process now described, Dr. Brewster has examined many hundreds of the lenses of animals brought from all parts of the world, and has found that there are five different modes in which the fibres are arranged, the same mode being invariably found in the same animal. These different arrangements of the fibres were illustrated by elegant drawings from the pencil of Dr. Greville, and by wooden models, which ex-

F

[page] 82

hibited all the inflexions of the fibres and the dimensions of their diameter as they approached to their termination.

In the greater number of lenses the structure is perfectly symmetrical in relation to the anterior and posterior surfaces, or to the poles of the axis of vision: but Dr. Brewster discovered in some lenses a remarkable deviation from this symmetry; the anterior surface having the fibres arranged according to one law, and the posterior surface according to another. The object of this singular structure he conceived to be to obtain a more perfect correction of the spherical aberration of the eye to which it belongs.

MR. MURCHISON, President of the Geological Society, communicated, verbally, observations on certain accumulations of clay, gravel, marl, and sand around Preston in Lancashire, which contain marine shells of existing species.

The marine shells of existing species in this district were first noticed by Mr. Gilbertson of Preston; and Mr. Murchison was desirous of calling the attention of the Meeting to the merits of that able naturalist.

He had this year visited the localities, and found the deposit in question to consist, near the surface, of clay with boulders of distant rocks, covering great thicknesses of marl, gravel, and sand, the sand usually being the lowest. These accumulations are not only spread over the broad delta extending from the coast at Blackpool to Preston in the interior, but they rise at the latter place into considerable eminences extending in plateaux on the banks of the Ribble and the Darwent, for several miles inland.

In certain places the marls, sands, and gravels contain shells of existing species (Mr. Gilbertson enumerates about 20 species), not differing from those of the adjoining sea, above which they were found at various heights from 80 to 300 feet. The accumulations seldom offer proofs of regular bedding or tranquil deposit, but rather resemble the detritus formed upon an agitated shore; although from their diversity of structure they present distinct evidence of having been heaped up during a long protracted period. Seeing the height above the sea at which these shells are found, and that they are usually buried under a cover of clay, containing large boulders of Cumbrian rocks, Mr. Murchison infers that one of the last elevations of the central ridge of the North of England is thereby proved to have taken place after the creation of existing species of animals.

[page] 83

The deposit was described as resting on inclined and contorted strata of millstone grit, and shale, and an overlying red sandstone, (banks of the Ribble and Darwent,) and upon the edges of the productive coal measures near Chorley.

This communication was followed by a discussion, in which Mr. Greenough, Mr. Murchison, and Mr. Phillips took part, on the application of these observations to resolve the question of the change of level on the coast of Lancashire, and on that of Yorkshire, where gravel deposits containing marine shells of existing species have been described by Mr. Phillips as diluvium.

DR. DAUBENY gave a Lecture on the connexion of Hot Springs with Volcanos.

Hot springs, he observed, are met with for the most part in one of these situations. 1st. In the vicinity of volcanos. Of this kind of position, the active volcanos of Iceland, Italy, and Sicily, and the extinct ones of France, Hungary, and the Rhenish provinces, afford numerous examples. In these cases, it cannot be doubted that the heat of the springs is derived from the volcanos contiguous to them.

2nd. At the foot of chains of mountains which have been uplifted. Now as the elevation of such chains may with some probability be referred to a volcanic cause, it seems most natural to attribute the occurrence of their hot springs to the same; and this is confirmed by observing that they are found for the most part either near the line at which the elevation seems to have commenced, or else near the axis of the chain, in places where the valleys penetrate to the greatest depth. Of both these positions, the Pyrenees afford abundant examples.

3rd. Hot springs occur in some cases at a distance from any great chain of mountains, but then there is in these cases often strong evidence of some fracture or dislocation of the strata, such as may reasonably be attributed to a volcanic cause. Instances of this are supplied by the hot springs of Clifton in this country, Carlsbad in Bohemia, and Pfeffers in Switzerland.

It appears, then, that the great majority of hot springs are attributable to volcanic action, and this is confirmed by considering the gaseous products which they evolve, for these are the same as those given off by volcanos.

The first of these is sulphuretted hydrogen, which is common also to volcanos, especially when in a state of languid action.

Another kind of gas given off by many hot springs, is

F 2

[page] 84

carbonic acid, which abounds also in cold springs; but when this is the case, the latter often exists in valleys of elevation, to use Dr. Buckland's nomenclature, which in the structure of the beds surrounding them, bear evidence of sudden uplifting. Such are the springs of Pyrmont in Westphalia, and of Tunbridge in this country.

The third gas given off by hot springs, is nitrogen. It had been previously found at Bath and Buxton; but Dr. Daubeny has likewise detected it in several other tepid springs in Derbyshire, and in that of Taafe's Well near Cardiff in Glamorganshire. He met with it also in a state of purity in the hot springs of St. Gervais, Cormayeur, St. Didier, and others, on the skirts of the Alps, and accompanying carbonic acid in those of Mont Dor, St. Nectair, and Chaudes Aigues in France; and from these observations of his own, combined with those of others, he concluded that nitrogen is disengaged from the generality of hot springs.

The presence of nitrogen is also an argument for adopting that chemical theory of volcanic action, which supposes it to arise from a species of combustion or oxidation, in preference to the mechanical hypothesis which regards it merely as a consequence of the law of distribution of temperature within the earth, and excludes the idea of chemical agency altogether.

On the latter part of this paper, it was remarked, that the gases mentioned by Dr. Daubeny are evolved from decompositions known to be going on at the surface, and at various depths from the surface of the earth, independently of hypothetical causes.

With respect to the occurrence of remarkable dislocations in connexion with mineral springs, Mr. Smith observed, that in the neighbourhood of the Bath waters, the dislocations must have been occasioned in very ancient geological æras; since the strata of the lias series, through which the hot springs rise, are unaffected by the disturbances of the coal and limestone series beneath.

FRIDAY EVENING.

MR. POTTER communicated the following observations on Electrical Phœnomena, exhibited in the Torricellian vacuum.

Though early experimenters had directed their attention to the phænomena of Electricity shown in passing through space as void of matter as they were able to procure, yet the question whether electricity can pervade a perfect vacuum, or can not do so, is still far from being decided. The experiments with

[page] 85

the air-pump would lead us to conclude that electricity would pervade an actual vacuum without sensible resistance, and without exhibiting light. But Mr. Walsh, Mr. Morgan, and other experimenters, had asserted that electricity could not pervade a perfect Torricellian vacuum. Sir Humphry Davy has maintained that the Torricellian vacuum is permeable to electricity, with an exhibition of more or less light according to the temperature. But as the appearances he describes are similar to those given by Mr. Morgan, when a very minute portion of air remained in his tube, it must be considered a question still open to further investigation.

From the writer's experiments, made with the object of learning something which might throw a light on the nature of Aurora Borealis, it has appeared, in conformity with the previous experiments of Sir Humphry Davy, Mr. Morgan, and others, that the passage of electricity through space containing only a very minute portion of air, was attended with a very considerable display of light, and this when the mercury in the tube stood scarcely perceptibly lower than that in a good barometer.

DR. WARWICK exhibited the method of making a powerful temporary Magnet, by coiling round a horse-shoe of soft iron a quantity of copper wire, connecting the poles of a galvanic battery, as originally performed by Professor Moll of Utrecht.

DR. DAUBENY exhibited a new instrument composed of finely reticulated wire, intended to illustrate the effects of capillary attraction.

A description of the New Volcanic Island, by Mr. OSBORN, communicated by Captain Hotham, R.N. was read to the Meeting.

SATURDAY MORNING.

MR. DALTON, President of the Manchester Society, read his Physiological Investigations arising from a consideration of the mechanical effects of atmospheric pressure on the animal frame.

In this essay Mr. Dalton endeavours to answer the question, how the animal body is enabled to resist the pressure of the external atmosphere, which varies in amount from 15 to 20 tons on a middle-sized man, without being sensible of the whole or any part of this enormous and fluctuating pressure.

The average specific gravity of the human body being taken

[page] 86

according to Robertson's experiments at 0·9, Mr. Dalton observes that the mean specific gravity of all the solids and fluids which are in it is about 1·05. The air contained in the lungs and other receptacles of the body is estimated at 150 cubic inches; the average bulk of the body 4500 cubic inches, of which consequently 4350 cubic inches are solid and liquid parts. The mean specific gravity of these parts, taken separately when dead, being 1·05, their total weight should be equal to 4567 cubic inches of water; but it was found by actual weighing, when alive, equal to 4044 cubic inches,—a difference of weight equal to 523 cubic inches of water, or more than 1/9th of the whole weight of the body. The general conclusion deduced by Mr. Dalton from these data, combined with other considerations, is, that the whole substance of the body is pervious to air, and that a considerable portion of air constantly exists in the body during life, subject to increase and diminution according to the pressure of the atmosphere, in the same manner as it exists in water: and further, that when life is extinct, this air in some degree escapes and renders the parts specifically heavier than when the vital functions were in a state of activity. (This Paper has since been printed in the Manchester Memoirs, Vol. V.)

MR. ALLAN communicated a Notice of a magnificent specimen of aqua-marine in the possession of Don Pedro.

The largest mass of precious beryl known to mineralogists is an aqua-marine belonging to Don Pedro; it is nearly as large as the head of a calf, its extreme length being 9 1/8 inches, its breadth 6 5/8 inches; it weighs 225 ounces Troy, or eighteen pounds nine ounces. On one side there are slight indications of the plane of a crystal; but it is otherwise entirely water-worn. Its surface is consequently dull; but beneath it the mass is perfectly clear and transparent, and, large as it is, without a flaw. It is of a beautiful pale bottle-green colour.

MR. ROBISON, Secretary of the Royal Society, Edinburgh, described and illustrated by diagrams the principles and mode of construction of his Linseed Oil Barometer, and detailed the mechanical processes by means of which he had been enabled entirely to free the oil from atmospheric air and other gaseous admixtures.

MR. FORBES read to the Meeting his Essay on the Horary Oscillations of the Barometer near Edinburgh.

In the. former part of this communication, the Author, after a short view of the progress of his researches on this subject,

[page] 87

states the circumstances under which his observations commenced. The place of observation is 4 miles S. W. of Edinburgh, lat. 55° 55′ 20″ N. long. 12′ 57·5″ west of Greenwich, at an elevation of 410·5 feet above the mean level of the sea. Five observations were made daily of the barometer and attached thermometer, from 8 to 8½ A.M., at 10 A.M., about 4 P.M., and at 8 and 10 P.M., in order to detect the morning and evening maximum and afternoon minimum. The number of observations was 4410, which, being reduced to a standard hour (10 P.M.), by methods described at length in the paper, yielded the following results.

The maximum of oscillation occurred in spring and summer at 8 or 8½ A.M. and 10 P.M., in autumn and winter at 10 A.M. and 8 P.M.

Taking these hours, and selecting the actual maxima, the amounts of Oscillation are found to be

Morning. Evening.
In Spring ·0213 ·0202
Summer ·0181 ·0151
Autumn ·0136 ·0079
Winter ·0031 ·0031

After comparing the results of his observations with an extensive collection of those of other observers in various latitudes, Mr. Forbes proceeded to investigate formulae which should express with the least error the general amount of oscillations in different latitudes, at the level of the sea. For this the paper itself, which will appear in the Transactions of the Royal Society of Edinburgh, must be consulted, as also for various indications of the influence which elevation above the sea, the season of the year, and the absolute mean temperature of the place, have in modifying the amount and period of occurrence of the phænomenon.

The following Extract of a letter from SIR JAMES SOUTH to Dr. Brewster, dated Observatory, Kensington, Sept. 29, 1831, was read by the Secretary.

"Should the York Meeting not have terminated its labours, and should you think it worth the trouble, I wish you would call the attention of any astronomers that may be there, to the anomaly which sometimes attends the transits of the satellites of Jupiter, over the planet's face. Generally speaking, the satellite may be seen to glide on the face as a bright planetary disc, and remains so till it has proceeded one sixth or one eighth of the planet's diameter. It then becomes invisible till it approaches the opposite limb within one sixth or

[page] 88

one eighth of the diameter of the planet, when it may be again detected as a bright disc, and remains so till it passes entirely off the face of the planet.

This anomaly I have witnessed three or, I believe, four times, and the last on Saturday fortnight, when the satellite itself instead of being invisible, except near the limbs, was perfectly visible as a black disc, and with its attendant shadow, was distinctly seen with the five-feet equatorial. The point to be settled is,—why should this fact be presented sometimes and not always?

The approaching disappearance of Saturn's Ring, and even its present situation, will be very advantageous for obtaining a knowledge of the various phænomena of the Satellites, and of the actual figure of the planet. Of the former we know next to nothing; of the latter but little that is satisfactory. Sir William Herschel's observations of the planet's figure are entirely at variance with mine.

If Lord Oxmantown, or any person possessing a large Reflector, would turn it every fine night on the Georgium Sidus, it would be well; for although Sir William Herschel expressed to me orally his doubts as to the accuracy of his observations, which assigned to that planet two rings perpendicular to each other, still I know not if this suspicion of his has even been promulgated. One Satellite has certainly been seen with my Achromatic, and one also by Mr. Herschel, myself, and Struve with the twenty-feet Reflector. Laplace doubted the existence of more than two. If the others are not greatly more faint than the one I have seen, Lord Oxmantown will certainly detect them instantly with the immense quantity of light afforded by his Reflector."

SATURDAY EVENING.

DR. DAUBENY read an account of Experiments on the combustion of Coal Gas performed at York, by the Rev. W. Taylor, from which it appeared that by regulating the quantity and mode of admission of the atmospheric air to the flame of a common Argand gas burner, the quantity of light might be much increased without increased expenditure of gas, while the colour of the light so produced varied according to circumstances. He referred the effects to the principle laid down by Sir H. Davy as to the luminosity of flame depending on the amount of solid matter maintained in a state of ignition at any given time.

THE REV. WM. VERNON HARCOURT showed a Lamp constructed upon a new principle, and explained the principle and

[page] 89

construction of it. He gave it the name of an oil gas lamp; not because it was lighted by gas formed at a temperature below that of flame, for this was common to all lamps, but because, as in the gas lights of the streets, the gas issued from a reservoir, and owed the perfection of its combustion not to an ascending current of hot air, but to the force with which it was propelled from the reservoir and carried the air along with it. It differed, however, from the common gas lights in these circumstances,— that the reservoir formed part of the burner; that the gas was formed as it was consumed; and that it was propelled, not by a vis a tergo and in a state of condensation, but by the expansive force of its own heat. In consequence of this circumstance the current of the gaseous jet was more rapid in proportion to the quantity of matter contained in it than in the common gas lights, whilst it was also at a much higher temperature, so that it could issue with a greater velocity without being liable to blow itself out. The practical difficulty of the construction consisted in the obtaining a steady supply of oil, especially with the cheap oils. This difficulty had been in great measure surmounted, but the instrument was still imperfect, and had been charged by some accident that evening with a vegetable oil, from which a clear light could not be obtained.

An Essay by DR. BREWSTER on a new Analysis of Solar Light was read by Mr. Phillips.

According to Sir Isaac Newton's Analysis of Solar Light by the prism, white light consists of seven different colours, each of which has a peculiar range of refrangibility occupying distinct spaces in the prismatic spectrum, "to the same degree of refrangibility ever belonging the same colour, and to the same colour ever belonging the same degree of refrangibility."

While examining the specific action of different coloured bodies in absorbing particular portions of the prismatic spectrum, Dr. Brewster was led to observe that rays of two different colours in the same spectrum had actually the very same degree of refrangibility, the one colour being superimposed upon the other. By extending this inquiry, and availing himself of the aid of various methods of insulating rays which the prism could not separate, he was conducted to the new Analysis of Solar Light, which it was the object of this paper to explain and establish. The following propositions contain a general view of the results.

1. White Light, whether it be that of the sun or of artificial flames, consists of three simple colours only, Red, Yellow, and Blue, by the union of which all other colours are composed.

[page] 90

2. The solar spectrum and that of artificial flames, whether formed by prisms of transparent solids and fluids, or by grooves in metallic and transparent bodies, or by the diffraction of light passing through a narrow aperture, consist of three spectra of equal length, beginning and terminating at the same points (viz. a Red, a Yellow, and a Blue Spectrum), and having their maximum of illumination at different points of their length, and their minimum at their two extremities.

3. All the seven colours in the solar spectrum, as they were observed by Newton and Fraunhofer, are compound colours, each of them consisting of Red, Yellow, and Blue Light in different proportions.

4. A certain portion of White Light incapable of being decomposed by the prism, in consequence of all its component rays having the same refrangibility, exists at every point of the spectrum, and may at some points be exhibited in an insulated state.

(Since this paper was read, an abridgement of it has been published in the Edinburgh Journal of Science, No. X. New Series, pp. 197—207; and the original Memoir, illustrated with coloured drawings, will appear in the next Part of the Transactions of the Royal Society of Edinburgh, Vol. XII. Part I.)

MR. WM. GRAY, jun. read the Translation of a memoir by Professor Gazari of Florence, on a method of detecting the traces of writing which has been fraudulently erased.

The Author of this paper having been frequently appointed by the Tribunals to give professional evidence in trials of this nature, instituted experiments on the subject, which, by showing him the possibility of removing entirely not only the ink, but also the materials employed in its removal, proved that cases might arise, when the fraud could not be detected in any other manner than by examining the condition of the paper or other material written on. For this purpose optical means were tried in vain, and immersion in water did not show such a difference in the absorptive power of the written and unwritten parts as happens in the employment of certain sympathetic inks; but on exposure of the suspected paper near to a moderate fire, the paper, which in consequence of the corrosive effects of the ink, was in those parts altered in its nature, was unequally acted on by the process of carbonization, and thus the number and length of the lines, and often the whole of the erased portion was distinctly revealed.

[page] 91

EXHIBITIONS, &c.

Mr. Gould, Member of the Zoological Society of London, exhibited select specimens of Birds figured and described in his work on the Ornithology of the Himalaya Mountains, and copies of this work were laid upon the tables for inspection.

Mr. R. Havell exhibited Drawings of Birds for Mr. Audubon's great work on American Ornithology.

Mr. Hey, Curator to the Philosophical Society of Leeds, showed some remarkable specimens of Fishes from the Yorkshire coal district, which belong to the Museum of that Institution.

Mr. Williamson, Keeper of the Museum to the Philosophical Society of Scarborough, brought for examination a series of the reliquiæ of fossil Crustacea recently discovered in the strata of that coast.

Mr. Wm. Gilbertson, of Preston, displayed an instructive suite of Crinoidal remains with other remarkable fossils from the vicinity of Clithero, and marine shells belonging to existing species from the gravel deposit on the banks of the Ribble.

Copies of recent publications lay upon the tables from Dr. Boswell Reid, from Mr. John V. Thompson of Cork, from Mr. Ashley of Edinburgh, Mr. Harrison of Barton, and others.

Mr. Smith, Author of the Map of the Strata of England, showed a Geological Map of the district round Hackness.

Mr. Murchison, President of the Geological Society, showed coloured Maps representing the Transition Rocks, the old Red Sandstone, and Carboniferous Limestone, on the border of Wales; the basin of new Red Sandstone (as Mr. Murchison has determined it to be,) in the Vale of Clwydd; and other Maps, Sections, and Notices relating to parts of South Wales, Lancashire, Durham, and Yorkshire.

[page 92]

[page 93]

SECOND REPORT

OF

THE BRITISH ASSOCIATION

FOR

THE ADVANCEMENT OF SCIENCE.

1832.

[page 94]

[page 95]

SECOND REPORT.

PROCEEDINGS

OF

THE GENERAL MEETING.

1832.

ON Monday, the 18th of June 1832, the British Association commenced its sittings at Oxford, in the rooms of the Clarendon Buildings, and proceeded to the election of candidates, recommended by the General Committee. In the Evening a numerous assembly of Members met together in the same apartments.

On Tuesday, at one o'clock P.M., the Chair was taken in the Theatre, by the President, Viscount Milton, who opened the business of the Meeting by a speech to the following effect:

"Gentlemen,

Feeling as I did, when called upon to preside over the first Meeting of this Association, the insufficiency of the individual chosen for that office, and knowing that the choice was to be attributed not to any merits or any desire of my own, but to the circumstance of my official connexion with the Society, by whose invitation that Meeting had been collected,—how much more must I feel sensible of the difficulty of the situation in which your kindness has placed me, when I find myself called upon to address you in this Theatre, within the walls in which are transacted the most important concerns of this great and august University! That difficulty, however, is much lightened by the consideration, that on the present occasion my almost only duty is to resign the office with which I have been invested, which will now devolve into hands more competent to wield it, and on shoulders whose strength is more capable of bearing its burthen. The difficulty also is lightened to me, Gentlemen, by the consideration that I am addressing an audience upon whom it

[page] 96

is unnecessary even to endeavour to press the importance of Associations, which have for their object to extend the bounds of human knowledge, and to give man a larger empire over nature.

I should, however, ill discharge the duty which is placed upon me as your organ at the present moment, and I should ill satisfy the wishes of my own mind, if I surrendered this office into the hands of my reverend and learned friend, who sits near me, without expressing the gratitude which I feel, in common I believe with every other Member of the Association, to the constituted authorities of this University, who have so kindly welcomed us here. Confident I am that they will never have reason to repent of the favour which they have shown us, in permitting the Meeting to assemble within these walls, but will reflect with well-founded satisfaction on the encouragement now afforded to an institution, the object and tendency of which is to promote the highest and most important interests of man;—I say, Gentlemen, his highest and most important interests: for were I to be asked what is the chief use of any new facts which we may be enabled to add to the stock of our knowledge, or what is the greatest value of any new inference which may be deduced from those facts of which we are already in possession, I should answer, that the principal use of such knowledge and such reasoning is to lead man to lift up his mind and his heart to his Maker; and in comparing his own inability (of which the greater is his knowledge, the deeper must his conviction be), in comparing the inability of the creature with the stupendous works of creation, to imbibe a deeper feeling of religious awe, and acquire a stronger sense of the reverence and duty which he owes to the power, the wisdom, and the beneficence of the Creator. It is on this ground especially that every reflecting mind will rejoice in the advancement of Science; and it is, I doubt not, to similar views of the value of every improvement in the knowledge of nature, that we are to ascribe the reception with which our Association is honoured in this ancient seat of learning and religion."

The President elect, the Rev. Dr. Buckland, then took the Chair, and addressed the Meeting in the following manner:

"My Lords, and Gentlemen,

I cannot enter on the duties of my office without acknowledging, with the deepest gratitude, the honour which the Association has conferred upon me, and without thanking my noble predecessor for the undeserved compliment which he has been pleased to pay me. The objects of this Meeting have been so fully, philosophically, and eloquently set forth, in the Report

[page] 97

of the Proceedings at York, by my reverend friend the late Vice-President, who has had so large a share in founding this Association, that I think it needless to occupy your time, either in explaining them, or in proving their importance. If any argument were necessary to justify the attempt now made to stimulate and combine the energies of science; if a doubt has existed on any man's mind as to the probability of its success,—I would only ask him to look round upon the present audience, and observe with how many and what manner of persons this Theatre is filled. Such an attendance leaves no room to fear that the Meeting should fail of its intended objects. Your presence, Gentlemen, adds an indisputable sanction to the proceedings of last year, and fulfils the warmest hopes which the promoters of the Association had indulged.

I will detain you no longer from the scientific business of the Meeting than may be necessary to apprise you of the regulations which the Committee has adopted for carrying it on in such a. method as may enable it to accomplish all its objects. The greatest facilities have been afforded to our proceedings by the authorities of the University, who have granted us every possible accommodation, and shown in all instances the most liberal disposition to promote our views. The Chancellor, Lord Grenville,—a name which I can never utter without grateful veneration,—when I had first the honour of communicating to him the proposal of the Association to hold the present Meeting in this place, was pleased instantly to reply, that it was his ardent desire to be enrolled among its Members. From that moment he has by all the means in his power seconded our wishes; and within these few days,—from his retirement, from that calm and peaceful retirement, in which, after the labours of an arduous political life, after exercising the highest functions of the statesman, he now enjoys the dignified repose of the scholar and philosopher,—he has expressed to me his heartfelt regret that he is debarred by the infirmities of advanced life from being present in Oxford at a moment so interesting as this. The Vice-Chancellor, Gentlemen, and the other Governors of the University, have placed at your disposal the Theatre in which we are now assembled, and the numerous rooms which have been lately fitted up for academical and scientific purposes in the Clarendon Building. These will be found very conveniently adapted to answer the exigencies of a Meeting, before which so great a variety of matter is to be brought: the business of your several Committees and Sectional Meetings will be separately distributed in them; the Committee of Mathematics and General Physics will meet in one room; that of Chemistry and Mineralogy in another; that of

G

[page] 98

Geology and Geography in a third; and that of Natural History and Physiology in a fourth. The Committees of these sciences will transact their own especial business, and hold their consultations, from ten to eleven in the Morning; and in the interval from eleven to one, Sectional Meetings of the whole body will be held, at which there will be read, before such Members of the Association as choose to assemble in any of the rooms, by the Secretary of each Committee, the papers which he has received on the subjects included under its denomination. At one P.M. the Meeting will daily adjourn to this Theatre; and here the Reports on the state and progress of different sciences will be read, which have been drawn up at the request of the Association. In the Evening, at nine o'clock, the Sectional readings will be resumed in the Clarendon Buildings, except on Thursday and Saturday, when Lectures will be delivered in the Music room on the late discoveries in Magnetism, and on Chemical and Geological subjects. Thus, Gentlemen, we hope to conduct the multifarious business of the Meeting, so as to accomplish three objects: first, to lay before the whole assembly the general views of the condition of science, to which it is desirable to invite the attention of all; secondly, to enable every one to listen to, and to join in, those scientific details in which he may be more particularly interested; and thirdly, to give instruction of a more popular nature, to a more miscellaneous audience. On Thursday morning, the University of Oxford will avail itself of the present opportunity to express the deep respect which it entertains for the improvers of science, by conferring on four Members of the Association, of preeminent celebrity in different branches of Philosophy, the highest distinction which it has the power to bestow; and when the ceremonial is concluded, in the afternoon of the same day, I would beg leave to offer to any of the Members who will do me the honour of accompanying me on an equestrian excursion, such familiar illustrations of Geology as the country round Oxford is able to afford."

Dr. Buckland then proceeded, after stating the arrangements which had been made for the personal accommodation and hospitable reception of the Meeting*, to call upon Professor Airy for his Report on the recent progress of practical and

* On Tuesday a public dinner was given to the Association by its Oxford Members, in the Hall of New College, granted for that purpose by the Warden and Fellows. On Wednesday Morning a public breakfast was given to it by the Vice-Chancellor, the Rev. Dr. Jones, in the Hall of Exeter College. Ordinaries, at five shillings a-head, were provided daily, to which venison was contributed by the Archbishop of York and the Duke of Buckingham. Refreshments were furnished to the Evening Meetings in the Clarendon Buildings, at the expense of the Oxford Members of the Association.

[page] 99

physical Astronomy, undertaken at the request of the former Meeting of the Association at York.

Professor Airy gave an account of the contents of his Report, and read those parts of it which he considered as possessing the most general interest.

Mr. Lubbock's Report on the present state of our knowledge respecting the Tides being next in order, the substance of it was delivered to the Meeting, in the absence of the author, by the Rev. William Whewell, and illustrated by the exhibition of a Map of the World on which Mr. Whewell had drawn the co-tidal lines passing through the points where it is high water at the same time.

On Wednesday, at one o'clock, the Meeting having re-assembled in the Theatre, the Chairman of the four Sub-committees read the minutes of the transactions of the Sectional Meetings.

At the conclusion of the minutes of the Geological Section, the President requested the Meeting to allow the Wollaston Medal, which had been awarded by the Geological Society to Mr. William Smith, to be delivered to him in the presence of the Members of the Association. The President of the Geological Society, Mr. Murchison, having in consequence presented the medal to him, in the name of that Institution, as a testimony of respect to the acknowledged "Father of English Geology," Mr. Smith expressed his gratitude for the high honour which had been conferred upon him in the Assembly of the British Association, and in the public Theatre of so distinguished a University,—an honour, he said, which was the more grateful to his feelings, from the circumstance of Oxfordshire being his native county. He little thought in his youth that so proud a moment as the present would ever arrive; and he trusted that his example and success would stimulate others to follow in the same course. In devoting himself to his geological pursuits, and opening a new page of knowledge, he had had the satisfaction of procuring the good will of many kind and indulgent friends; he hoped that he had served his country, and in so doing he had endeavoured to serve his God.

Professor Cumming being then called upon by the President, read a Report on Thermo-electricity.

Mr. Forbes gave an account of his Report on the present state of Meteorology, and read extracts from it.

Mr. Willis delivered a verbal Report on the present state of the Philosophy of Sound, illustrated by diagrams and musical experiments.

In the Evening, at nine o'clock, a Meeting was held in the

G 2

[page] 100

Music Room, and two lectures were delivered, the first by Dr. Ritchie on Magnetic Electricity, with reference especially to the recent discoveries of Mr. Faraday; the second by Dr. Turner on the General Principles of Chemistry.

On Thursday morning the University held a Convocation in the Theatre, for the purpose of conferring the honorary degree of Doctor in Civil Law on the under-mentioned Members of the Association:—

Sir David Brewster, K.H. LL.D. F.R.S. L. & E. Instit. Reg. Sc. Paris. Corresp.

Robert Brown, F.R.S. V.P.L.S. Instit. Reg. Sc. Paris. Corresp.

John Dalton, F.R.S. Instit. Reg. Sc. Paris. Corresp.

Michael Faraday, F.R.S., Instit. Reg. Sc. Paris. Corresp.

The Regius Professor of Civil Law, Dr. Phillimore, in presenting them to the Convocation, adverted to the distinguished services which they had respectively rendered to different departments of science, and the celebrity which they had acquired by their successful labours, not only in Great Britain, but throughout Europe, and expressed the high satisfaction felt by the University of Oxford in enrolling such illustrious names in the catalogue of her Members.

After the degrees had been conferred, a party of the Members of the Association accompanied Professor Henslow on a botanical excursion; and a numerous assemblage attended the President, to hear his Lecture on the Geology of the neighbourhood of Oxford. In the course of the Lecture Dr. Buckland took occasion to enforce the importance of Geological Science, as connected with agricultural improvement, and suggested that there might be great utility in an appointment, by the Geological Committee, of a Sub-committee to devote its attention to this object. He pointed out many defects in the ordinary system of drainage, and explained in what manner large tracts of land might in many cases be permanently drained at a small expense, by methods depending entirely on a knowledge of the structure of the strata. He adverted to the possibility of reclaiming the peat bogs in Ireland, distinguishing those which are capable of being reclaimed, from those where the outlay of capital must exceed any profitable return; and in speaking of Artesian wells*, suggested the advantage which

* The name of Artesian wells has been recently applied to those wells in which water is obtained by boring down and introducing tubes, through strata destitute of water, into a subjacent stratum, which is charged with it, in such a manner that it ascends through the tubes almost, or entirely, to the surface, forming in the latter case perpetual fountains, such as are made, and designated by the name of Blow wells, on the eastern coast of Lincolnshire. The practice is most available in low situations, where the upper stratum is a thick bed of clay, and has been of late years introduced in the neighbourhood of London. It is much used in Artois, the ancient Artesium, whence is derived the appellation of Artesian wells.

[page] 101

might be derived from a more general application of them in the neighbourhood of London.

On Friday, the President having taken the chair in the Theatre at the usual hour, the minutes of the Sectional Meetings were read by the Chairmen of the Committees. Mr. P. Duncan gave notice, that there were laid upon the table some Original Manuscripts from the Ashmolean Museum, recording the early Proceedings of the Philosophical Society which met at Oxford during the Civil Wars, and subsequently gave birth to the Royal Society.

An Abstract of a Report on the progress of Optical Science, by Sir David Brewster, was then read by one of the Secretaries.

Mr. Johnston read his Report on the progress which Chemical Science has recently made, especially in foreign countries.

Professor Powell read his Report on the state of our knowledge respecting the phænomena of Radiant Heat.

The Rev. William Conybeare gave a general account of the contents of his Report on the recent progress of Geology.

The Rev. Dr. Bliss and Mr. John Taylor were appointed to audit the accounts.

On Saturday, the Association having assembled in the Theatre for the last time, Mr. John Taylor made a Report on the state of the accounts, which was approved. The minutes of the proceedings in the Sectional Meetings to their close were read by their respective Chairmen.

Mr. Brunel gave a history of the attempt to carry a Tunnel under the Thames, illustrated by Drawings.

The Rev. William Whewell gave a sketch of the views contained in his Report on the recent progress and present state of Mineralogy.

An essay by Dr. Prichard, on the application of Philological inquiry to the Physical History of Man, was read by the Rev. William Conybeare.

The President then announced, that the place which had been fixed upon for the next Meeting was Cambridge; that the President who had been chosen was Professor Sedgwick; that

[page] 102

the Vice-Presidents elect were Dr. Dalton and Professor Airy; that Professor Henslow and Mr. Whewell had undertaken the duties of Secretaries for Cambridge, and the late Vice-President, Mr. Harcourt, those of General Secretary; and that Mr. John Phillips had been appointed to the office of Assistant Secretary. He added, that a Council had been nominated to direct the affairs of the Association during the interval which would elapse before the next Meeting of the General Committee.

The Rev. Professor Sedgwick said, that it would be at all times and in all situations one of his greatest pleasures to contribute his assistance to the British Association, and that he was willing to give any pledge for the zealous performance of the gratifying but arduous duty which had been imposed upon him, as far as his ability extended. He might have been overwhelmed, indeed, by the prospect of such a task, did he not feel confident in the cooperation of many distinguished Members of the University of Cambridge, possessing much greater powers than his own, and did he not believe that before the next Anniversary the organization of the Society would be so complete that his duties would be light when compared with those of the present President, whom he would take this opportunity of publicly thanking for the delightful manner in which he had presided over the Meeting, bringing the various elements of which it was composed into order and harmony, and diffusing sunshine through all its proceedings. He knew not how to express strongly enough the satisfaction which he had derived from this Meeting, or the delight which he had felt at being associated in such a place with such men as Dalton and Brewster, and Faraday and Brown, in honouring whom the University of Oxford had done honour to itself. Studies like those which had lately occupied the Society, consecrated by the principles which have pervaded it, could not but tend to elevate and purify the mind, to engender mutual friendship, mutual forbearance, mutual kindness and confidence; they prevented the growth of any bad feelings, and caused those which were good to germinate with the greatest luxuriance compatible with our nature. He looked forward with full assurance to the happy results of this union between men of similar sentiments and similar pursuits, who possess one common object,—the improvement of mankind by the promotion of truth; and he thanked the Association most cordially for the honour which it had conferred upon him in electing him to the high office of its President. At Cambridge they would endeavour to follow, though they could scarcely hope to rival, the example of hos-

[page] 103

pitality which had been set them at Oxford, and would receive the Meeting not merely with the forms of courtesy, but with the right-hand of fellowship. In one respect he believed that they would be more advantageously circumstanced as to the means of offering accommodation; for at the time when the Association would meet again, at the latter end probably of the month of June, they should be enabled, he trusted, to receive a large proportion of its Members within the College walls.

The Rev. Wm. Whewell said, that his services had been fully and freely given to the Association, and would be so given as long as they could be useful. He hoped that all who were assembled there would accept the offer of a cordial welcome to Cambridge; he hoped that all the Members of the Association in every part of the empire would equally accept it. "We are desirous," he added, "of seeing as many as possible from as many places as possible; we ask for the company of all who are cultivators of science or interested in its objects. I trust all such persons have felt here that they are united by one common tie; and with so large and united a body of zealous and active men, I think I may reasonably prophesy that the next Meeting will be a worthy successor to the brilliant and successful one which now soon must close."

The Marquis of Northampton moved the Thanks of the Meeting to the Vice-Chancellor, the Heads of Houses, and the other resident Members of the University, for the great hospitality and attention with which they had received the Association. The Resolution was seconded by Sir David Brewster, supported by the President of the Geological Society of London, Mr. Murchison, and passed unanimously.

The President said, that he could not allow the Meeting to quit the public Theatre, without adding on his own behalf, and on behalf of his fellow-academics who are Members of the Association, the expression of their grateful thanks to the Vice-Chancellor and the Heads of Houses, for the kind assistance which they had rendered them in promoting the objects and providing for the accommodation of the Meeting.

The Vice-Chancellor replied in the following manner. "On my own part, Sir, and on the part of the University, I beg leave to assure you, that we are most sensible of the advantages afforded to us by this visit of the British Association for the Advancement of Science, and that we have been most happy to have had it in our power to offer it any accommodation; and I will add, that whilst the pursuit of truth and the advancement of knowledge are its objects, and whilst it pursues those ends by the judicious rules which at present regulate its proceed-

[page] 104

ings, it cannot fail to command the good wishes, the respect and admiration of all, and most especially of those, whose institutions have connected them with the duties, and taught them to appreciate the value, of public instruction."

In the evening a Lecture was delivered in the Music Room, by the President, on the Fossil Remains of the Megatherium, recently imported into England from South America. The lecture was illustrated with Drawings by Mr. Clift.

Dr. Buckland began by stating, that the history of this animal is very remarkable. The Megatherium is most nearly allied to the Sloths, a family which presents an apparent monstrosity of external form, accompanied by many strange peculiarities of internal structure, which, before the discoveries of the immortal Cuvier, were but little understood. "Gentlemen," said he, "I cannot utter the name of Cuvier, and associate with it the term 'immortal', without being at once arrested and overwhelmed by melancholy and painful recollections of mortality. We have at this moment to deplore, in common with the whole philosophical world, the loss of the greatest naturalist and one of the greatest philosophers that have arisen in distant ages, to enlighten and improve mankind. The names of Aristotle, and Pliny, and Cuvier, will go down together through every age, in which natural history and physical sciences, in which philosophy and learning, and talent, and everything which, next to religion and morality, gives dignity and exaltation to the character of man, shall be respected upon earth. Gentlemen, I need not state to you how voluminous are the works of that exalted and most illustrious naturalist, whose recent and irreparable loss we now deplore. For nearly thirty years he has been the leader of that branch of natural philosophy which comprehends the structure and relations of all the kingdoms of animated nature. It was the genius of Cuvier that first established the perfect method after which every succeeding naturalist will model his researches; and which laid the foundation of that analytical process of investigation, of that most philosophical and accurate and uniform system of reducing every organ in every species to a fixed and certain type, which will enable his followers to extend their inquiries over the almost boundless regions of the organized world. He has shown that the frame and mechanism of every animal present an uniformity of design and a simplicity of purpose, which prove to demonstration that every individual, not only of existing species, but of those numerous and still more curious races which have lived and perished in distant ages, and of which our knowledge is due to discoveries in geology,

[page] 105

were framed and fashioned by the same Almighty hand, and were designed and contrived by the same Almighty mind. Gentlemen, to this great and good man not only are the sciences of natural history profoundly indebted, but the higher science of morals also owes a debt of deep and everlasting obligation, for that he has proved to demonstration the high and solemn truth to which I have alluded, viz.—the unity and universal goodness of the great Creator. Of this great man, so lately torn away from us by the mysterious and incomprehensible counsels of the Almighty, we now lament the loss; in the vigour of his mind, and almost in the fulness of his strength, at the age of sixty-three, he has been suddenly summoned to the grave, and the tears of Europe have not yet ceased to stream over his funeral. The gratitude of the great nation to whose philosophic fame his genius has added so bright a wreath, has already displayed itself by a liberal provision for his family, and has fixed his widow, during the remainder of her mortal life, in that honoured and well-known mansion, in the Jardin des Plantes, which during a quarter of a century has ever been open, in noble and friendly hospitality, to every son of Science assembled at Paris from every nation under heaven. The French nation has placed her there, a brief and perishable monument of their gratitude: they have resolved also that a splendid and more lasting monument shall be raised to her immortal husband, and have invited the whole philosophical world to partake in the honour of contributing to its erection. He has raised to himself a monument 'ære perennius,' a monument which will endure even when the pyramids are crumbled into dust;—but this does not absolve ourselves from the pleasing and pious duty of contributing our humble share to that monument of marble which a grateful nation and a grateful world are about to consecrate to his memory. —Gentlemen, I fear my feelings of respect, and love, and gratitude, have transported me beyond the limits which the task I have undertaken should impose on me; still I cannot but rejoice in the opportunity which this august assembly affords of inviting you to partake in this great and glorious work, and thus publicly to record your gratitude to that immortal man, whose friendship I have ever counted among the most distinguished honours of my life, and whose genius will, even by those who have not enjoyed this high and enviable privilege, be ever followed as their guide in the paths of science, so long as science shall be cultivated, or virtue venerated upon earth."— Returning to the subject of the Lecture, Dr. Buckland stated, that this monstrous animal, the Megatherium, has been brought to England by Woodbine Parish, Esq., His Majesty's

[page] 106

Consul at Buenos Ayres. It was discovered by a peasant who, passing along the river Salado in a dry season, threw his lasso at something he saw ·half-covered with water, and dragged on shore the enormous pelvis of this animal; the rest of the bones, consisting of the greater part of the skeleton, with many of the claws and teeth, were obtained by turning aside the current by means of a dam. Dr. Buckland proceeded to illustrate the peculiarities of its structure and mode of life, by reference to the peculiar organization and contrivances in its skeleton. This animal, and its kindred monster the Sloth, have been considered by Buffon and other naturalists to afford the greatest deviations from the ordinary structure of quadrupeds—deviations which they have viewed as indicating imperfection in their organization, without any compensating advantage. The object of the Professor's Lecture was to show, that these anomalous conditions and deviations are so far from being attended with inconvenience to the class of animals in which they occur, that they afford striking illustrations of those rich and inexhaustible contrivances of nature by which the structure of every created being is precisely fitted to the state in which it was intended to live, and to the office which it was destined to perform. The peculiarities of the Sloth which render its movements so awkward and inconvenient upon the earth, are adapted with much advantage to its destined office of living upon trees and feeding upon their leaves; the peculiarities of the Megatherium are not less wisely adapted to its office of feeding upon roots.

This animal was about 8 feet high and 12 feet long; its teeth, though ill adapted for the mastication of grass or flesh, are wonderfully contrived for the crushing of roots, with the further advantage of keeping themselves constantly sharp set by the very act of performing their work. The fore feet, nearly a yard in length, and exceeding a foot in breadth, were armed with three gigantic claws, each more than a foot long, and forming most powerful instruments for scraping roots out of the ground. The head and neck and anterior part of the trunk were comparatively light and small: its posterior proportions much exceeded the bulk of those of the largest elephant. The object of this apparently incongruous admixture of proportions, was to enable the creature to stand at ease on three legs, having the weight of its body chiefly supported by the hinder extremities, and one of its fore paws at liberty to be exercised without fatigue in the constant operation of digging roots out of the ground. A further peculiarity consists in the fact of its sides and back having been armed with a coat of mail like the

[page] 107

armadillos, which also obtain their food by the act of continual digging in the ground; this coat of mail exceeds an inch in thickness. The Professor suggested his opinion, that one use of the bony armour is to prevent the annoyance which this class of animals would feel, without some such protection, from the constant presence of sand and dirt with which the act of digging and scratching for their daily food would otherwise fill their skins; its further use may be to afford protection against the myriads of insects that swarm in the regions frequented by these animals, and also against beasts of prey.

The Professor concluded by stating, that this was but one of the many examples afforded by comparative anatomy of the inexhaustible richness of contrivances whereby Nature has adapted every animal to a comfortable and happy existence in that state wherein it was destined to move; and added, that the researches of Geology tended not only to afford similar examples of contrivance, indicating the wisdom, and goodness, and care of the Creator over all his works, but afforded also to natural theology a powerful auxiliary, showing from the unity of design and unity of structure, and from the symmetry and harmony that pervade all organic beings in the fossil world, as well as in the present, that all have derived their existence from one and the same Almighty and Everlasting Creator.

Professor Babbage then rose, and said, that before the Meeting separated he wished to express a feeling which he believed was general among the Members of the Association, that in the selection of the places at which the Annual Meetings were to be held, attention should be paid to the object of bringing theoretical science in contact with that practical knowledge on which the wealth of the country depends. "I was myself," said Mr. Babbage, "particularly anxious for this, owing as I do a debt of gratitude for the valuable information which I have received in many of the manufacturing districts, where I have learned to appreciate still more highly than before, the value of those speculative pursuits which we follow in our academical labours. I was one of those who thought at first that we ought to adjourn for our next Meeting to some large manufacturing town; but I am now satisfied that the arrangement which has been made will be best adapted to the present state of the Association. When, however, it shall be completely consolidated, I trust we may be enabled to cultivate with the commercial interests of the country, that close acquaintance which I am confident will be highly advantageous to our more abstract pursuits."

[page] 108

Mr. Murchison called the attention of the Meeting to the peculiar obligation which it lay under to one of its Members. "At our first Meeting, Gentlemen, in York, when the Institution was in its infancy and every difficulty hung around us, a Professor of this University came forward and undertook, on his own responsibility, that Oxford would open its gates to receive us. Delighted as we have been with the reception which we have experienced, and sensible how much the Association has been consolidated by this Meeting, we owe an acknowledgement of gratitude to Dr. Daubeny as the primary cause of our having assembled here."

Dr. Daubeny said, that from the situation which he occupied in the University, it was naturally to be expected that he should regard with peculiar interest the Meeting, in that place, of an Association which he considered calculated to form an important epoch in the history of British science. "The attachment," said the Professor, "which I entertain for the cause of science implies in my case no extraordinary merit, placed as I am in a situation of comparative independence, by my connexion with one of the great ecclesiastical establishments of the country. It is to those only who have pursued such studies without partaking of the advantages derived from academical institutions, or that patronage of Government which in other countries supplies their place, to whom the praise is due of a high degree of disinterestedness in preferring the attractions of philosophy to those of emolument. With respect to my office of Secretary, any credit which may be attached to the discharge of it belongs equally to my colleague Professor Powell, and the other Secretaries of the Association, and amongst them to one who I regret to find has been prevented by illness from attending this Meeting, I allude to Mr. Phillips, Secretary for York, an individual whom I regard with peculiar friendship, and to whom the Society was more deeply indebted at a much earlier stage of its progress than to me."

The Marquis of Northampton said, that the Meeting had that evening received from its President an excellent exemplification of the utility of scientific knowledge. "We have seen, Gentlemen, that an animal which has been represented even by Buffon as imperfect in its constitution, and almost incapable of enjoyment, appears, on a nearer and more accurate view of its habits and anatomy, to be no less happily framed, and adapted to its peculiar manner of living, than the other parts of the creation. Thus even in those cases which present difficulties in the way of superficial knowledge, a higher degree of acquaintance with nature is sure to find a satisfactory solution. I rejoice, Gen-

[page] 109

tlemen, in the success with which this great and important effort for the advancement of science has been crowned. Long may our Association flourish, and produce fruits for the benefit of others as well as of ourselves! I rejoice in its success, not only in an intellectual but in a moral sense; for I believe it may be the means of binding together all the portions of this great empire, and even of uniting other parts of the world in the same bond. It is a refreshing thing for a person like myself, to come from the metropolis, from the turmoils of political life, and meet with eloquence and ability dedicated entirely to the promotion of good-fellowship and truth, to see discussion deprived of its sting, and those who are elsewhere opposed, brought here into cordial intercourse with each other, and made to feel that on many points they are able to agree. The pursuits of science have a tendency to associate together the whole human family; and I cannot but remark with pleasure, that we have had at least one eminent individual from the United States of America among us at this time. I hope, Gentlemen, that our next Meeting at Cambridge may have more.We must remember, and I trust our Transatlantic brethren will remember, that they and we are sprung from the same race; that we speak the same language; that we equally rejoice in the possession of free though of different institutions; that their ancestors as well as ours were fellow-countrymen of Bacon and Newton, Milton and Shakespeare, and many other great men who have preceded us in science and literature. I hope that these feelings of mutual sympathy will ever exist between us and them; and I hope that the interests of science will form a bond of intercourse and union between us and all the other nations of the world, that wars and tumults at last may cease, and that our only emulation may be, who shall become the wisest, and who become the best."

The President then said, "Gentlemen, the hour is come for the adjournment of this most happy Meeting. I congratulate the University of Oxford on the compliment that has been paid it by the presence of so many distinguished and illustrious strangers, who have honoured us with their company on this ever-memorable occasion. I congratulate the Association on the perfect harmony which has pervaded its Meetings, and on the vast and inestimable utility which is likely to result from its operations; I congratulate the British nation that it possesses such a Society, comprehending a host of individuals not only qualified, but prompt and ready, to come forward and promote the general interests of science. Gentlemen, I congratulate each individual here present, on the enjoyment of what I con-

[page] 110

sider one of the highest gratifications of which our nature is capable,—the enjoyment of that personal knowledge and familiar intercourse, which this Meeting has afforded, with those whose kindred minds and congenial pursuits have been long familiar to us through the medium of their works; the enjoyment of being thus brought into friendly contact and brotherly association, with those whom we have long esteemed and loved and venerated from a distance; the enjoyment of being thus enabled, though but for a short, yet a most delightful week, to hold sweet counsel and communion together in these our palaces of peace. Gentlemen, it is now my painful duty to announce, that the moment of separation is arrived; it is my more grateful task to remind you, that we are to re-assemble at Cambridge in the latter part of the month of June next year."

[page] 111

PROCEEDINGS

OF

THE GENERAL COMMITTEE.

ON Monday the 18th of June the General Committee made the necessary arrangements for transacting the business of the Meeting; appointed a Treasurer; recommended Candidates for election; and granted to Foreigners and to certain other individuals gratuitous tickets of admission to the Meetings.

On the Saturday following, the Treasurer reported the state of the accounts. The General Committee appointed the Trustees in whom the property of the Association was to be vested; selected the place of meeting for the ensuing year; chose the new Officers of the Association; and nominated a Council to transact the business in the intervals between the Meetings. The Recommendations of the Committees of Sciences were presented to the General Committee, and adopted. A Report of the Proceedings of the Meeting was ordered to be published.

TRUSTEES OF THE ASSOCIATION.

Charles Babbage, K.H. F.R.S. Lucasian Professor of Mathematics, Cambridge.

R. I. Murchison, F.R.S. President of the Geological Society.

John Taylor, F.R.S. Treas. G.S. &c.

OFFICERS.

President.—Rev. William Buckland, D.D. Canon of Christ-church, F.R.S. G.S. &c. Professor of Geology and Mineralogy, Oxford.

President elect.—Rev. Adam Sedgwick, F.R.S. G.S. &c. Woodwardian Professor of Geology, Cambridge.

Vice-Presidents.—Sir David Brewster, K.H. LL.D. F.R.S. L. & E. &c. Instit. Reg. Sc. Paris. Corresp. Rev. William Whewell, F.R.S. G.S. &c. Secretary to the Cambridge Philosophical Society.

Vice-Presidents elect.—G. B. Airy, F.R.S. Professor of As-

[page] 112

tronomy, Cambridge. John Dalton, F.R.S. Pres. of the Lit. and Phil. Soc. of Manchester, Instit. Reg. Sc. Paris. Corresp.

Treasurer.— John Taylor, F.R.S. Treas. G.S. &c.

General Secretary.—Rev. William Vernon Harcourt, F.R.S. G.S. &c.

Assistant Secretary.—John Phillips, F.G.S. &c. Secretary to the Yorkshire Philosophical Society.

Secretaries for Oxford.—Charles Daubeny, M.D. F.R.S. Professor of Chemistry. Rev. Baden Powell, F.R.S. Savilian Professor of Geometry.

Secretaries for Cambridge.—Rev. J. S. Henslow, F.R.S. Professor of Botany. Rev. William Whewell, F.R.S.

Secretaries for Edinburgh.—John Robison, Sec. R.S.E. &c.

Secretary for Dublin.—Rev. Thomas Luby.

COUNCIL.

Sir David Brewster, K.H. LL.D. F.R.S. L. & E. Instit. Reg. Sc. Paris. Corresp. Robert Brown, D.C.L. F.R.S. V.P.L.S. Instit. Reg. Sc. Paris. Corresp. M. I. Brunel, F.R.S. Instit. Reg. Sc. Paris. Corresp. William Clift, F.R.S. &c. J. R. Corrie, M.D. F.G.S. James D. Forbes, F.R.S.E. &c. Davies Gilbert, D.C.L. V.P.R.S. &c. J. H. Green, F.R.S. Sir John Herschel, K.H. F.R.S. &c. W. R. Hamilton, F.R.S. Astronomer Royal of Ireland. W. J. Hooker, M.D. Professor of Botany, Glasgow. Rev. B. Lloyd, D.D. Provost of Trinity College Dublin. Rev. George Peacock, F.R.S. J. C. Prichard, M.D. F.R.S. Rev. William Scoresby, F.R.S. L. & E. Instit. Reg. Sc. Paris. Corresp. J. S. Traill, M.D. F.R.S. N. A. Vigors, D.C.L. F.R.S. &c. Ex officio members,—The Trustees and Officers of the Association.

Secretaries.— Edward Turner, M.D. F.R.S. L. & E. Rev. James Yates, F.L.S. G.S.

COMMITTEES.

I.Pure Mathematics.Mechanics, Hydrostatics, Hydraulics.Plane and Physical Astronomy.Meteorology, Magnetism, Philosophy of Heat, Light, and Sound.

Chairman.—Davies Gilbert, D.C.L. V.P.R.S. F.G.S. &c.

Secretary.—Rev. H. Coddington, F.R.S. G.S. &c.

G. B. Airy, F.R.S. &c. Professor of Astronomy, Cambridge. Charles Babbage, F.R.S. Lucasian Professor of Mathematics, Cambridge. Sir David Brewster, K.H. LL.D. F.R.S. L. & E. &c. Lieut. Gen. Sir Thomas M. Brisbane, K.C.B. F.R.S. L. & E.

[page] 113

Instit. Reg. Sc. Paris. Corresp. David Forbes, F.R.S. L. & E. Mr. E. W. Gill. William Gray, jun. Sec. Y.P.S. Rev. R. Greswell, F.R.S. W. R. Hamilton, F.R.S. &c. Astronomer Royal for Ireland. George Harvey, F.R.S.E. F.G.S. Eaton Hodgkinson, Memb. Manch. Soc. Rev. Thomas Jarrett, Professor of Arabic, Cambridge. Robert Murphy, F.C.P.S. Rev. George Peacock, F.R.S. F.G.S. Rev. Baden Powell, F.R.S. Savilian Professor of Geometry, Oxford. Richard Potter, jun. S. P. Rigaud, F.R.S. Savilian Professor of Astronomy, Oxford. Rev. Robert Willis, F.G.S. Rev. Robert Walker, F.R.S. Charles Wheatstone. Rev. William Whewell, F.R.S. F.G.S.

II.Chemistry, Mineralogy, Electricity, Magnetism.

Chairman.—John Dalton, D.C.L. F.R.S. Instit. Reg. Sc. Paris. Corresp.

Secretary.—-James F. W. Johnston, M.A.

Henry James Brooke, F.R.S. F.G.S. John George Children, F.R.S. Rev. James Cumming, F.R.S. F.G.S. Professor of Chemistry, Cambridge. J. F. Daniell, F.R.S. Professor of Chemistry, King's College. Michael Faraday, D.C.L. F.R.S. Charles Daubeny, M.D. F.R.S. F.G.S. &c. Professor of Chemistry, Oxford. William Gregory, M.D. F.R.S.E. William Snow Harris, F.R.S. Rev. W. V. Harcourt, F.R.S. F.G.S. &c. W. H. Miller, F.G.S. Professor of Mineralogy, Cambridge. William Prout, M.D. F.R.S. Rev. William Ritchie, LL.D. F.R.S. Professor of Natural Philosophy, University of London, and the Royal Institution. Rev. William Scoresby, F.R.S. L. & E. Instit. Reg. Sc. Paris. Corresp. Mr.W. Sturgeon. Edward Turner, M.D. F.R.S. Sec. G.S. Professor of Chemistry, University of London.

III.Geology and Geography.

Chairman.—R. I. Murchison, F.R.S. &c. President of the Geological Society.

Secretary.—John Taylor, F.R.S. Treasurer of the Geological Society.

Sir T. D. Acland, Bart. F.H.S. Rev. William Buckland, D.D. V.P.R.S. Professor of Geology and Mineralogy, Oxford. Rev. W. D. Conybeare, F.R.S. V.P.G.S. Instit. Reg. Sc. Paris. Corresp. Viscount Cole, F.R.S. G.S. Joseph Carne, F.R.S. G.S. M.R.I.A. Sir Philip de Malpas Grey Egerton, Bart. F.R.S. G.S. W. H. Fitton, M.D. F.R.S. G.S. Robert W. Fox. G. B. Greenough, F.R.S. L.S. G.S. William Hutton, F.G.S. Sir Charles Lemon, Bart. F.R.S. G.S. G. Mantell, F.R.S. G.S. The Marquis of Northampton, F.G.S.

H

[page] 114

Rev. A. Sedgwick, F.R.S. G.S. Professor of Geology, Cambridge. William Smith. Major-general Stratton, C.B. F.R.S. Rev. James Yates, F.L.S. G.S.

IV.Zoology, Botany, Physiology, Anatomy.

Chairman.—Rev. P. B. Duncan, F.G.S. Keeper of the Ashmolean Museum, Oxford.

Secretary.—Rev. J. S. Henslow, F.L.S. F.G.S. Professor of Botany, Cambridge.

R. Brown, D.C.L. V.P.L.S. &c. William John Burchell, F.L.S. S. D. Broughton, F.R.S. G.S. Rev. William Clark, F.G.S. Professor of Anatomy, Cambridge. William Clift, F.R.S. G.S. Rev. W. L. P. Garnons, F.L.S. C. Henry, M.D. Rev. L. Jenyns, F.L.S. John Kidd, M.D. F.R.S. Reg. Professor of Medicine and Anatomy, Oxford. J. C. Prichard, M.D. F.R.S. Joseph Sabine, F.R.S. Richard Taylor, Sec. L.S. F.G.S. N. A. Vigors, D.C.L. F.L.S. G.S. Sec. Z.S. G. Williams, M.D. Professor of Botany, Oxford.

[page] 115

RECOMMENDATIONS

OF

THE COMMITTEES.

COMMITTEE FOR MATHEMATICS, &c.

THE Committee recommend that REPORTS should be applied for by the Association on the following subjects:—

I. On the recent theoretical and practical history of the Pendulum.

II. On the present state of the analytical theory of Hydrostatics and Hydrodynamics; stating how far the leading problems recently discussed have been solved theoretically,—on what suppositions,—and what remains wanting to complete the solution.

III. On the present state of our knowledge of Hydraulics as a branch of Engineering; stating whether it appears from the writings of Dutch, Italian, and other authors, that any general principles are established on this subject; what they are; and what are the points contested among authors.

IV. On the present state of our knowledge of the Strength of Materials; whether from the experiments of various authors any general principles have been obtained; what these are; how modified in their application to different substances; and what are the differences of opinion which prevail among authors on this subject.

V. On the state of our knowledge respecting the Magnetism of the Earth.

The Committee recommend as a subject for examination, the law of the Resistance of Water to bodies in motion.

The Committee recommend,—That a request be made to the Portsmouth Philosophical Association, through its deputies, to institute a series of hourly meteorological observations at that place.

That the Secretaries of the Yorkshire Philosophical Society

H 2

[page] 116

be requested to continue their observations on the Quantity of Rain which falls at different heights.

That the invitation to investigate fully the theory of the wet bulb Hygrometer be renewed.

That persons who may have opportunities of travelling on mountains or of ascending in balloons, be invited to observe the state of the Thermometer and the dew point Hygrometer below, in, and above the clouds; and to determine how the different kinds of clouds differ in these respects.

That the fourth Sub-section of this Committee be requested to procure Reports and Researches to be made on the following subjects.

I. On the connexion of Vaporization with Pressure.—On the improvement of the Thermo-barometer as to accuracy and portability.—On the correction for Humidity in the barometrical measurement of heights.

II. On the improvement of Thermometers for rendering them sensible to minute impressions of radiant heat.—On the alleged Polarization of Heat.—On the effect of transparent Screens upon Heat.

III. On the phænomena considered as opposed to the undulatory theory of Light.—On the improvement of the Achromatic Telescope.

COMMITTEE FOR CHEMISTRY, &c.

The Committee recommend,—That Dr. Dalton and Dr. Prout be requested to institute experiments on the specific gravities of Oxygen, Hydrogen and Carbonic Acid, and to communicate their results to the next Meeting.

That Dr. Turner be requested to extend his researches into the atomic Weights of the elementary bodies, and to report to the next Meeting, on the progress recently made in this branch of chemical science.

That Mr. Johnston be requested to report to the next Meeting, on any additional evidence which may be deduced from his own experiments, or those of others, in support of the new theory of the Sulphur Salts.

That Professor Miller be requested to determine the form and optical characters of those Crystallized Bodies which have not been examined by Mr. Brooke and Mr. Levy, and that Chemists be invited to send him specimens of perfect artificial Crystals.

That the Rev. Wm. Whewell, Mr. Brooke, Professor Miller, and Dr. Turner, be requested to cooperate in prosecuting and

[page] 117

promoting the following inquiries, with a view to examine the theory of Isomorphism, and the connexion between the crystalline forms and chemical constitution of Minerals:—

I. To determine whether the angles of varieties of the same species (in the usual acceptation of identity of species,) are identically the same, under various circumstances of colour, appearance, and locality; and if not, what are the differences.

II. To determine the chemical constitution of such varieties, —the specimens, mineralogically and chemically examined, being in all cases the same.

III. To determine what quantity of extraneous substances may be mixed with a crystalline salt, without altering its form.

IV. To determine the angles of the various species or varieties of isomorphous or plesiomorphous groups,—and their respective chemical composition.

That a list be drawn up and printed of substances considered isomorphous* and of those considered isomeric†.

List of simple substances and binary compounds presumed to replace each other.

image

* The lists of isomorphous substances have been drawn up for the Committee by Professor Miller.

† For the list of isomeric substances, see the Report on Chemistry.

[page] 118

image

Isomorphous Groups.

O. Octahedral. Q. Square prismatic. R. Rhombohedral. P. Prismatic. P′. Oblique prismatic. P″. Doubly oblique prismatic.

image

[page] 119

image

[page] 120

image

[page] 121

image

[page] 122

image

COMMITTEE FOR GEOLOGY, &c.

The Committee recommend,—That Mr. John Taylor be requested to prepare for the next Meeting, a Report on the state of our knowledge of Mineral Veins.

That the attention of Geologists be invited to the Faults or Dykes in the carboniferous rocks in Flintshire, with a view to ascertain whether some remarkable differences in their character may not be observed, as compared with that of veins and dykes in other districts.

That the attention of Geologists be invited to those coal districts in the midland counties of England, where the Carboniferous Limestone and Old Red Sandstone being deficient, the coal measures rest immediately on the Grauwacke and Transition rocks;—with a view to discover whether any circumstances connected with the physical structure of that part of the island can be stated, explanatory of the local absence of the two great formations above mentioned.

That Mr. John Taylor be requested to collect detailed sections of the Carboniferous series of Flintshire, with a view to a comparison with the same series in other parts of England;— with a view also of ascertaining the circumstances under which the Mountain Limestone is developed, after its suppression in certain coal-fields in the central parts of England.

That sections and plans should also be collected of the Coal-

[page] 123

fields of Worcestershire, Shropshire, Staffordshire, Cheshire, Lancashire, and the South Western part of Yorkshire.

That inquiries should be made as to the existence of the Wealden formation in the midland counties of England, and particular attention paid to the character of the Organic Remains.

That a detailed examination of any River in Great Britain should be instituted;—with a view to compare the outline of its bed, with that of the valley in which it runs.

That the quantity of Mud and Silt contained in the water of the principal rivers in Great Britain should be investigated; distinguishing, as far as may be possible, the comparative quantity of sediment from the water at different depths, in different parts of the current, and at different distances from the mouth of the river; distinguishing also any differences in the quality of the sediment, and estimating it at different periods of the year;—with a view of explaining the hollowing of valleys, and the formation of strata at the mouths of rivers.

That the experiments of the late Mr. Gregory Watt, on the fusion and slow cooling of large masses of Stony Substances, should be repeated and extended by those who, from proximity to large furnaces, have an opportunity of trying such experiments on a large scale; and that trial should be made of the effect of long-continued high temperature, on rocks containing petrifactions, in defacing or modifying the traces of organic structure.

That the Members of the Association be requested to institute experiments, for the purpose of ascertaining whether the continued action of steam or of water at a high temperature, is capable of dissolving or altering minerals of difficult solution.

That a request be made to the Board of Ordnance, that the scale of colours adopted in the geological colouring of the Ordnance Maps may be published for general information.

COMMITTEE FOR ZOOLOGY, &c.

The Committee recommend, That REPORTS should be applied for by the Association on the following subjects:

On the recent progress and present state of Zoology.

On the recent progress and present state of Animal Physiology.

[page 124]

TREASURER'S ACCOUNT.

Dr. Cr.
1831. £. s. d. 1831. £. s. d.
Compositions received from 49 Members 245 0 0 Payments at York by Jonathan Gray, Esq. provisional Treasurer, as audited at the Annual Meeting at Oxford, viz.
Annual Subscriptions from 47 Members 47 0 0 Printing, Stationery, 95 12 5
Carriage of Reports, 3 16 8
Balance paid to John Taylor, Esq. 192 10 11
£292 0 0 £292 0 0
1832. £. s. d. 1832. £. s. d.
Balance from preceding Account 192 10 11 Paid on Account of Expenses at Oxford;— (particulars tobe stated and audited next year) 112 4 0
Compositions received from 99 Members 495 0 0 Paid to London Secretary 1 8 9
Annual Subscriptions from 435 Members 435 0 0 Postages, &c. to Bankers 0 7 1
Purchase of 1000l. s percent consols, vested in the names of the Trustess 836 5 0
Composition not yet received 5 0 0
Balance in hand, Oct. 1832 167 6 1
£1122 10 11 £1122 10 11

JOHN TAYLOR, Treasurer.

[page] 125

TRANSACTIONS.

Report on the Progress of Astronomy during the present Century. By G. B. AIRY, M.A., F.R. Ast. Soc., F.G.S., Fellow of the American Academy of Arts and Sciences; late Fellow of Trinity College, Cambridge; and Plumian Professor of Astronomy and Experimental Philosophy in the University of Cambridge.

THE "Committee for Mathematical and Physical Science" of the British Association having done me the honour to desire of me a Report on the progress of Astronomy, I could not but cheerfully comply with their request as far as I was able, though conscious that my Report must in many respects be imperfect. And I must beg the indulgence of the Society for any omissions or erroneous views; and must request them to attribute such, in part, to the circumstance that my own connexion with Astronomy is of short standing, and that since that connexion began I have been much occupied with the minute duties attached to the care of an Observatory, as well as with other official business of a very different kind.

I propose to take a brief survey of the progress of Astronomy since the beginning of the present century. That time must always be regarded as one of the most important epochs in the history of Astronomy. The English theodolite and the French repeating circle had been several years invented, and the advantages of circles were generally recognised; the principal part of the Mécanique Céleste was published, and the theory of perturbations, and especially of inequalities of long period, was beginning to be well understood. But, besides that I have the advantage of commencing from an almost definite point in the history-of the science, I feel that the progress of Astronomy since that time has been such that a correct statement of it must be highly interesting. I will not say that any discoveries of observation within the present century can be compared in general importance to the discovery of aberration and nutation;— that any theoretical discovery of the present century is equal to that of the great inequality of Jupiter and Saturn, or the

[page] 126

acceleration of the Moon's motion,—or that any single effort has been made like that which sent expeditions to Peru and to Lapland: But I will undertake to say, that in no similar period has greater progress been made in the increased number and excellence of observations; in the accuracy of the methods of treating them; in the examination and extension of theory; in the improvement of our powers, both instrumental and mathematical; and, finally, in the diffusion of accurate knowledge, in the increase of the number of persons who are interested in the science, and in the facility of communication among Astronomers.

I shall arrange my Report under the following principal heads:

I. A short general history of institutions and periodical publications.

II. An account of some of the instruments principally in use.

III. A statement of the improvements in the catalogues of fundamental stars, including the discussions of the various corrections.

IV. An account of the more extended star-catalogues, with the tables for facilitating the corrections.

V. Notices upon the measures of double stars, the observations of nebulæ, &c.

VI. An account of the principal observations, tables, &c. of the Sun and Moon, the old planets and their satellites.

VII. History of the new planets and periodical comets: and of comets generally.

VIII. Account of measures whose object is to determine the figure of the earth.

IX. General history of physical theories.

X. Comparison of the progress of Astronomy in England with that in other countries.

XI. Suggestion of points to which it seems desirable that the attention of Astronomers should be directed.

I. At the beginning of the century, the Observatory of Greenwich was the only one (I believe,) in which observations were made on any regular system. The thirty-six stars selected by Dr. Maskelyne, the Sun, and the Moon, were observed on the meridian with great regularity; the planets very rarely, and only at particular parts of their orbits; small stars, or stars not included in the thirty-six, were very seldom observed. A mass of observations was thus accumulating which, though confined in its object, surpassed in regularity and accuracy, and perhaps in general value, any other observations made at that time. The observations also were published in the form in which

[page] 127

they were made, and this circumstance alone gave them great value. So little had this been formerly understood, that Bradley's original observations, and the first part of Maskelyne's, were considered private property; great delays had consequently occurred in the printing of them, and only the first part of Bradley's Greenwich observations was at this time published*. The concluding part, however, was published in a few years: and this work reflects honour on the liberality of the University of Oxford and the care of the Savilian Professors who superintended it. An uninterrupted series of observations thus existed, made on the same plan and at the same place, and published with the fullest details. But this statement cannot be extended to any other astronomical institution. Observations had been made at Oxford, and transcribed, but not published: observations had been made at Armagh, and a standard catalogue had been deduced. I know of no other public observatories in activity in this country; and few observations seem to have been made by private persons. On the Continent, the several observatories of Paris and that of Berlin were the most important. At the national Observatory of Paris observations apparently were not made on any regular plan, and were only published in the Connaissance des Temps; the immense collection of observations of small stars made principally at the Observatory of the Ecole Militaire, was however completely published in the Histoire Céleste. Many irregular observations made at Viviers, Montauban, Mirepoix, &c., were also published, seldom with great detail, in the Connaissance des Temps. In like manner the observations made at Berlin and the various German observatories were imperfectly published in the Berliner Jahrbuch and (I believe,) in the Vienna Ephemeris†: those made in Italy were abstracted in the Effemeridi di Milano‡, &c. Thus, besides the Greenwich Observations, there existed no regular repository of observations, or discussions of observations, except these Ephemerides and (occasionally) the Transactions of the Societies of London, Paris, Berlin, Petersburg, Turin, Modena, &c. The necessary

* Within a few months, the observations made by Bradley before his residence at Greenwich have been published, under the superintendance of Professor Rigaud, at the expense of the University of Oxford. They include the observations by which the most important of his discoveries were made.

† I have not been able to procure a copy of this work.

‡ As I shall have occasion frequently to cite these Ephemerides, it is proper to mention that the Connaissance des Temps and the Berliner Jahrbuch have generally been published three years in advance, and the Effemeridi di Milano one year in advance.

[page] 128

want of detail in these publications has in many instances deprived the observations of much of their value.

With the year 1800 commenced the publication of Zach's Monatliche Correspondenz: it continued without interruption to the end of 1813 (a year in which the order of almost every continental scientific publication is interrupted). Lindenau's Zeitschrift für Astronomie commenced in 1816, and finished with 1818; Zach's Correspondance Astronomique commenced in 1818, and terminated in 1826; and Schumacher's Astronomische Nachrichten, which commenced in 1821, exists still as an astronomical periodical. These works were published at intervals of one or two months (the last of them, as often as matter to fill a sheet could be obtained): and nothing perhaps has contributed more to the progress of the science; especially in those parts (as the observations of comets,) which were useless without immediate circulation. "What," asks Lindenau, "would have been the fate of the small planets, if the Monatliche Correspondenz had not then existed?" But besides the rapid communication of information, these journals also allowed of the publication of observations with greater detail, and of fuller exposition of theoretical or physical views. And in fact nearly all the astronomy of the present century is to be found in these works or in the Ephemerides of Berlin, Paris, or Milan. It is owing, I suppose, partly to political events, and partly to our small acquaintance (in general) with the German language, that the three most valuable of these periodicals and the Berlin Ephemeris have till lately been little known in England.

In 1814 the regular annual publication of the Königsberg Observations (by Bessel) was begun; as well as that of the Dorpat observations (by Struve); in 1820 that of the Vienna observations (by Littrow); and in 1826 that of the observations at Speier (by Schwerd). These are all the Observations regularly published on the Continent with which I am acquainted. One volume. comprising several years' observations, has been published at Paris, but it does not seem likely to be followed by any more; one also has been published at Turin. Several very valuable volumes of observations have also been published at Palermo (by Piazzi and Cacciatore), and, I believe, two at Abo (by Argelander). In all these the original observations are given as fully as in the Greenwich Observations, and some steps of the reductions are in general much more completely explained.

Nor has our own country in the mean time been idle. Soon after the present Astronomer Royal (Mr. Pond) succeeded to Dr. Maskelyne, the regular annual publication of. observations

[page] 129

was changed for quarterly publication. The establishment of assistants at the Greenwich Observatory has been gradually increased, till it now exceeds in numerical strength every other observatory (so far as I know,) in the world. The mass of observations which it produces, of a very laborious kind, but of the very highest value for their accuracy, exceeds those which any other institution has put forth. The plan of these observations is rather confined, but by no means so much as under Dr. Maskelyne: their results have been occasionally published, but without any intermediate step. The Observatory of Dublin, under the direction of Dr. Brinkley, assumed the highest importance. The observations have not been regularly published, but the results (accompanied sometimes with the original observations,) have appeared in various memoirs by Dr. Brinkley. These are confined to observations of the principal stars: but other observations I believe have also been made. In 1823 and 1824 an Observatory was erected at Cambridge: it has been placed successively under the super-intendance of Professor Woodhouse and of the author of this Report. Though at present it is only in part furnished with instruments, the regular publication of observations has commenced, and four volumes (the result of as many years' labour, commencing with 1828,) have appeared. The only difference between the plan of these and that of the others which I have described, is that the reductions are given at greater length; the observation of planets is made one of the principal objects of this Observatory. In 1826 the original observations of Tobias Mayer were published at the expense of the British Government, under the superintendance of M. Mosotti. Within three years the regular publication of observations made at Armagh by Dr. Robinson has been begun: they are nearly on the same plan as the Greenwich Observations.

In 1821 the British Government determined to found an Observatory at the Cape of Good Hope: and Mr. Fallows was immediately sent out with some small instruments. The erection of the Observatory was not completed till 1828: and the two most important instruments arrived in 1829. The observations made by Mr. Fallows have not yet been published: but I have seen them in manuscript, and I can assert them to be most valuable. His successor, Mr. Henderson, has it in contemplation not only to continue the independent observations peculiar to a southern latitude, but also to observe regularly in concert with European astronomers. Let us hope that the publication of Mr. Fallows's observations will not be delayed, and that provision will be made for the regular appearance of

I

[page] 130

his successors', in the same manner as those of European observers.

One addition to our astronomical establishments, the gift of an individual, is yet to be mentioned. In 1822 Sir Thomas Macdougall Brisbane, soon after his appointment as governor of New South Wales, founded an Observatory at Paramatta, and furnished it with excellent instruments. By his personal attention, and by the activity of the assistants whom he procured, a series of valuable observations has been produced. On his leaving the station, he presented the instruments, &c. to the British Government, on condition that the Observatory should be maintained in an efficient state. The condition was accepted, and an observer (Mr. Dunlop) is now maintained by the British Government at this distant station. Among all the instances that we have mentioned, there is not one which reflects higher lustre upon the motives which caused its establishment.

Observatories have also been founded by the East India Company at Madras, Bombay, and St. Helena. The observations made at the first of these places have been published by the Company.

I regret that I cannot attempt to give any accurate history of the increased number and the improvements of Continental Observatories during this period. Several new ones have been erected; several have been much improved both in the character of their buildings (the situation being in some instances changed from an upper story to the ground floor,) and in the nature of the instruments, especially by the general introduction of circular instruments.

At none of these (excepting Königsberg and Vienna, and perhaps one or two more,) is the system of observation so regular as at Greenwich.

The following list includes all the public Observatories with which I am acquainted:—Greenwich, Oxford, Cambridge, Edinburgh, Dublin, Armagh, Cape of Good Hope, Paramatta, Madras, Bombay, St. Helena, Paris, Marseille, Geneva, Turin, Milan, Padua, Bologna, Modena, Naples, Palermo, Abe, Altona, Bremen, Christiana, Dorpat, Copenhagen, Königsberg, Berlin, Gotha, Mannheim, Speyer, Munich, Göttingen, Vienna, Cracow, Warsaw, Wilna, Ofen, Kremsmünster. The Observatories of Brussels and Cadiz, and perhaps some in the above list, are not yet in full activity. I am not aware that there is any public Observatory in America, though there are some able observers.

In 1820 the Astronomical Society of London (now the Royal Astronomical Society,) was founded: and this event, whether

[page] 131

considered as an indication of public feeling on the subject of Astronomy, or as a means for promoting the science, must be considered as most fortunate. Astronomy in England has undoubtedly received a strong impulse from the institution of this Society; and the four volumes of Memoirs which it has published contain some of the most valuable contributions to Astronomy that any country has yet produced.

I must here observe, that nothing appears to me to prove more strongly the extension of accurate science than the increased demand for original observations. Astronomers are now sensible that though observations may be reduced, and the results exhibited in the form most valuable at the time of publication, future researches will generally give the means of improving them; and that the opportunity of doing this will be lost, except the observations are published in the shape in which they are made.

I have spoken above of the Astronomical Ephemerides, without allusion to that which forms their distinctive character. The Nautical Almanac had been followed by most of the others in the system of giving with accuracy only the places of the sun and the moon and principal stars, and such quantities as were necessary for nautical observations; the places of the planets being exhibited very roughly. In two works published by Schumacher (beginning with 1822,) the places of the planets were given more accurately. In the Berliner Jahrbuch for 1830 not only was the accuracy of the solar and lunar places increased, but the places of the planets were given with the same accuracy. Mean time was also adopted in every part, to the exclusion of appaŕent time. The comparison of observations with tables becomes thus an easy work. The English Government, by the advice of the Astronomical Society, have determined on making this improvement in the Nautical Almanac: and the volume for 1834 will appear with these alterations. Among the many additions made by Encke to practical astronomy, the example which he has thus set is not the least.

II. At the epoch from which this Report commences, the mural quadrants were still the only instruments (assisted by the use of the zenith-sector,) for observing zenith-distances, at the Greenwich Observatory, and at most of the Continental Observatories. At Palermo, however, a reversible circle or "altitude and azimuth instrument," of six feet diameter, was in the hands of Piazzi: and a similar instrument was in preparation for the Dublin Observatory, and was mounted early in the century.

I 2

[page] 132

A smaller instrument of the same construction, well known as the Westbury Circle, was in the hands of Mr. Pond: and by the use of this, it may be asserted, the errors of the large Greenwich quadrants were first completely established. In 1809 Mr. Troughton published in the Phil. Trans. an account of his method of dividing circles; and this may be considered as the greatest improvement ever made in the art of instrument-making. The general principle is to make a number of temporary points very near the places of many of the graduations, to compare by microscopes the distances between every pair, and when the errors are found numerically, to set off by a simple apparatus the permanent points at the proper distances from the temporary points*. In 1812 the first mural circle (by Troughton) was erected at Greenwich; and this is an important epoch in the history of Astronomy. I conceive that no instrument but the reversible circle can compete with Troughton's mural circle; and between these I cannot presume to decide. It must be observed that, as the mural circle was first intended to be used, the objects of these instruments were somewhat different. The reversible circle could be turned round its vertical axis in a few minutes, and the deviation of the axis from perfect verticality could be ascertained by the plumb-line; and the body under observation being observed in both positions of the circle, its zenith-distance was directly found (a small correction being necessary to obtain its meridian-zenith-distance, as both observations could not be made on the meridian). But the mural circle, like the mural quadrants, had no reference to the zenith: it could give only the polar distance of heavenly bodies, the position of the instrument corresponding to an observation of the celestial pole being found by observing circum-polar stars above and below the pole. To remedy this want (sometimes felt as an inconvenience,) Mr.Pond introduced (about 1821,) the system of observing sometimes the image of the heavenly body seen by reflection from the surface of mercury. In 1825 another instrument of exactly the same kind was erected; and now the system may be said to have reached its perfection. Whenever the weather permits, the same object is observed directly with one circle, and by reflexion with the other. The determination of zenith-distances from the combination of these observations, though laborious, must I think be unrivalled in point of accuracy.—Several circles on Troughton's plan have been sent to continental observatories.

* Several methods, slightly different, have been founded on this of Troughton's. I do not know what method the continental artists employ.

[page] 133

It had long been thought that an instrument might be constructed which would enable an observer to take at once the transit and the zenith distance of a celestial body. Several such had been made in England; one is particularly described in Wollaston's Fasciculus; and one is well known as the instrument used by Mr. Groombridge. It consists of two parallel circles, firmly braced together, and fixed to an axis similar to the axis of a transit instrument: the telescope also passes through this axis and between the two circles, and it rests near both its extremities on the braces connecting the two circles. The graduations on the circles are read (as in the large instruments,) by microscopes with moveable wires. This I should conceive to be an excellent instrument. In Germany, however, a very different instrument has been constructed with the same object, and (principally through its use by Bessel,) has obtained considerable celebrity. Reichenbach's circle* was first constructed (I believe) about 1820, and a considerable number of instruments of this construction have since been made for observatories in all parts of the Continent. In a journey in the North of Italy in the year 1829, I saw and examined several exactly similar to Bessel's. To the extremity of the axis of a transit instrument, a graduated circle is fastened; this circle turns accurately around another circle carrying on its circumference four verniers (the line of separation between the two circles being almost imperceptible): and this vernier-circle has a fixed as well as a reversible spirit-level, to show how much it deviates from a fixed position. One pivot of the graduated circle passes within that of the vernier-circle, and the latter rests upon the Y: at the opposite end the pivot of the graduated circle rests immediately on the Y. To prevent friction, each extremity of the transitaxis is supported by a lever-counterpoise, and the vernier-circle is supported by an independent lever-counterpoise: and, to prevent flexure of the telescope, each end is supported by a lever-counterpoise whose fulcrum is at the transit-axis. An instrument of this kind would I conceive be below mediocrity unless the workmanship were most exquisite (the German workmanship is very fine); and when made in the best possible way, I cannot but think that its mechanical structure is extremely weak. The first thing to be provided for in an instrument for measuring zenith distances is, that the circle and the telescope shall move together. In Troughton's mural circle this is insured by firmly fixing the telescope by its extremities to the limb of the circle, without any close connexion at the centre.

* I have seen several circles by Reichenbach constructed on different plans that which is used by Bessel is generally known by this name.

[page] 134

In Reichenbach's circle, the connexion is by three weak joints, where a failure of any one will spoil the instrument: and one of these is particularly liable to be strained, on account of the friction which must in time take place between the two circles. I ought not to omit that the Germans consider it a great advantage that the circle is cast in one piece.—It will easily be seen, that the use of this circle requires reversion: this cannot be done readily (as in the Dublin and Palermo circles), and therefore it is only done occasionally.

I have particularly described this instrument, because it is little known in this country, and because it will give a very good idea of the peculiarities of the German school of instrument-making. Its distinguishing features are these:

Telescopes are always supported at the middle, not at the ends.

Every part is, if possible, supported by counterpoises.

To these principles, every thing is sacrificed. For instance, in an equatorial the polar axis is to be supported in the middle by a counterpoise: this not only makes the instrument weak, (as the axis must be single,) but also introduces some inconvenience into the use of it. The telescope is on one side of the axis: on the other side is a counterpoise. Each end of the telescope has a counterpoise. A telescope thus mounted must, I should think, be very liable to tremor. If a person who is no mechanic, and who has not used one of these instruments, may presume to give an opinion, I should say, that the Germans have made no improvement in instruments except in the excellence of the workmanship.

The French repeating circle has lost much of the credit which it once enjoyed. Reichenbach's repeating circles have been much used, but in most instances by rejecting the principle of repetition; which converts them, in fact, into altitude-and-azimuth instruments. Of Reichenbach's universal-instrument, and several others invented here and on the Continent, I shall say nothing, because they do not seem likely to produce any influence on Astronomy.

Among those, however, which have been made as auxiliaries to the principal instruments, I must not omit to mention Capt. Kater's vertical collimator. The object of this instrument (whose construction is too well known to require description here,) is to supply a mark which shall be visible like a star, revolving in a very small circle round the zenith. By observing this in its north and south distances from the zenith, the reading of a circle corresponding to an observation of the zenith may be found, and thus the zenith distance of any heavenly body may be immediately obtained. Strong testimony has

[page] 135

been produced as to the accuracy of the results which may be expected from using this instrument.

In the equatorials of Dorpat and Paris, a clock-work motion has been given, I believe with great success, so as to keep the telescope steadily pointing on the same star. I have not had an opportunity of seeing either of these in a state of action.

Within a few years considerable improvements have been made in achromatic telescopes. A telescope of nine inches aperture was made by Lerebours: many small telescopes of great excellence, and one of more than nine inches aperture (for Dorpat), were made by Fraunhofer: two refractors by Cau-choix, of eleven or twelve inches aperture, have been imported into this country. All these are made on the common principle of the achromatic telescope. Mr. Barlow has turned his thoughts to the construction of telescopes in which the place of the flint-glass is supplied by a fluid lens (of sulphuret of carbon): and having succeeded with telescopes of six and eight inches aperture, proposes to attempt larger dimensions. I believe that none of these surpass in power or clearness the twenty-feet telescopes which Sir William Herschel and Sir John Herschel were in the habit of using; but the science has undoubtedly gained much by the diffusion of these powerful instruments.

In clocks I do not know of any improvement. Hardy's clock is found very useful with transit-instruments, for the loudness and sharpness of its beat; but for steadiness of rate it is probably inferior to the dead-beat which was in general use at the beginning of the century. The execution of chronometers (without any novelty of principle,) has been very greatly improved.

In the use of many of these instruments an improvement (as I consider it,) has very generally been introduced. It is now the rule at many observatories not to attempt mechanically to remove all the errors of the instrument, but to measure them (which can be done more accurately), and to apply numerical corrections to the observations. This innovation is due principally to the Germans.

III. At the beginning of the century the only good catalogue of stars was that of Dr. Maskelyne for 1790. In the last volume of his observations appeared his catalogue of the right ascensions and declinations of thirty-six stars for 1605. These were by far the most accurate places that had ever been produced. The amount of precession (combined for each star, with the proper motion of that star,) was determined by com-

[page] 136

paring these places with those obtained from Bradley's observations, as reduced by Dr. Maskelyne.

The determination of the mean declination of a star is independent of other observations, and depends only on the refraction and nutation at the time of observation (as in a series of observations the effect of aberration is nearly eliminated). For the mean Image the observations are not independent; the position of the different stars with respect to each other is the subject of one determination, and that of all with respect to the sun, when he has a particular zenith distance, is the subject of another: the values of Æ are affected, therefore, by errors of refraction, as well as by nutation, and a complication of errors of observation. Maskelyne had used Bradley's table of refractions, and had used 9".55 as the coefficient of nutation in declination. The right ascensions were found by comparing all the other stars with α Aquilæ, and comparing α Aquilæ with the sun.

At the beginning of the century Cagnoli determined independently the place of Capella, and founded on this determination a catalogue of stars of which I shall speak again.

In 1806 Mr. Pond gave a catalogue of N.P.D. founded on his observations with an altitude-and-azimuth instrument, but using the same corrections as Maskelyne.

In 1807 Piazzi published, in the sixth volume of the Palermo Observations, his catalogue of 120 principal stars observed with great care, and a greater number of stars on which less attention was bestowed. All the places were referred to Procyon and α Aquilæ; and these stars were compared immediately with the sun. The table of refractions used by Piazzi was one deduced by himself from observations of circumpolar stars in different parts of their diurnal circles, and differed little from Bradley's. This catalogue therefore was strictly independent. In 1814 it was extended so as to include 7646 stars, and published separately; and this large catalogue is at this time referred to by all observers as a standard catalogue.

The subject of refraction had in the last century attracted considerable attention, and had been treated theoretically by Oriani, Kramp, and Laplace. The object of these writers was, from an assumed law of constitution of the atmosphere, founded as far as possible on experiment, to determine, à priori, the magnitude and law of refraction. The experiments made by the French chemists and opticians had determined the relation between the pressure, and density of the air (subject to a very small doubt as to the effect of heat) and the quantity of refraction, without reference to any astronomical observations. Some

[page] 137

doubt remained as to the law of temperature in ascending in the atmosphere: this they attempted to remove by means of observations of horizontal refraction. On these grounds a table was calculated by Delambre (assisted partly by Piazzi's and his own observations), and published in 1806 by the French Board of Longitude; and this table has been always highly esteemed.

In 1806 Carlini published (in the Milan Ephemeris) his astronomical investigations of refraction. His object, besides giving a table of refractions generally, was to show that the refraction on the north side of the zenith at small altitudes was sensibly different from that on the south side. For refractions on the north side he used observations of circumpolar stars; as, the law of refraction as far as the pole being well known, the true zenith distances below the pole are immediately found. For refractions on the south side he referred to observations made at Palermo, where the same stars were seen with 7° greater elevation. In this manner a difference of a few seconds was found below 85° zenith distance. In the position of Milan this difference is quite conceivable: I know not whether the idea has been taken up by any other astronomer.

In 1810 Mr. Groombridge published in the Phil. Trans. a table of refractions founded cn observations of circumpolar stars.

In 1813 and 1815 Mr. Pond published catalogues of the N.P.D. of the principal stars, determined by observations with the Greenwich circle. In reducing these, Bradley's refractions were still used.

In 1813 Bessel published (in the Berliner Jahrbuch for 1816,) the table of refractions obtained from Bradley's observations. In 1818 he published the Fundamenta Astronomiæ pro Anno 1756, deduced from Bradley's observations. This work has always been considered one of the most valuable contributions to our Astronomy. It exhibits the result of all Bradley's observations of stars, reduced on a uniform system, and is always referred to by succeeding astronomers as the representative of Bradley's observations. Various disquisitions on refraction and the mathematical theory of the other corrections are contained in it. From Kramp's and Laplace's theories and Bradley's observations, a new table of refractions was formed. No alteration was made in the coefficient of nutation.

In 1816 Lindenau announced in the Zeitschrift für Astronomie, that by the discussion of 810 Greenwich observations of the right ascension of Polaris, he had found the coefficient of nutation to be rather less than 9"·0. No details of this investigation were given, and none (I believe) have been pub-

[page] 138

lished elsewhere. This, I think, is unfortunate; for I cannot consider the Greenwich transits of Polaris (observed in general on only one wire,) to be very accurate. This coefficient, however, has been adopted by Bessel and all the German astronomers in every subsequent investigation that I know. (Dr. Brinkley's sidereal investigations, and Cacciatore's investigations from the obliquity of the ecliptic in different years, agree in giving something very near 9".3 for the coefficient.) In the Dorpat Observations for 1822, and the Milan Ephemeris for 1819, &c. are many observed right ascensions of Polaris, undertaken with the same object; and I have observed about 400 with much care.

In the Berlin Memoirs for 1818 and 1819, Bessel published a very valuable paper on the right ascension of Maskelyne's 36 stars, from observations with a transit by Dollond and a circle by Cary. This memoir may be cited as a model for all succeeding investigations of the same kind. In the volume for 1825 he published another paper on the same subject, the results being founded on five years' observations with Reichenbach's circle.

In the Konigsberg Observations for 1821 and 1822 appeared a number of observations made by Bessel for ascertaining the amount of refraction near the horizon. They consisted of observations of stars which when on the meridian passed so near to the zenith that there could be little uncertainty about their absolute places. From these observations, and from observations of fifty-nine circumpolar stars on the meridian above and below the pole, he formed a new table of refractions, differing a little from that given in the Funddmenta. This table he applied (amongst other things,) to the solution of a curious diffeculty. Every astronomer (Mr. Groombridge excepted), who had observed the sun's zenith-distance at the solstices, had deduced from the summer solstice a greater obliquity than from the winter solstice. It was impossible that this could arise from any planetary perturbation; and several hypotheses were invented to explain it. Piazzi (Memorie della Società Italiana, 1804,) conceived that solar refraction might depend on something besides the barometer and thermometer, as for instance on the electricity of the air, and that the peculiar state of the air during the prevalence of the sirocco might affect it. Legendre ascribed it to something like nutation of the earth. Olbers thought that the sun's centre of figure might possibly not coincide with its centre of attraction. The general belief, however, was that it depended upon some fault in the tables of refraction, or the method of using them. Now Bessel showed

[page] 139

that upon using his Table the obliquities, from his own observations, came out equal from the two solstices. He had remarked (Berliner Jahrbuch, 1825,) that Bradley's observations gave equal obliquities, and in a paper in the Zeitschrift für Astronomie, vol. i., he endeavoured to show that the observations of all the different observers make the obliquities equal. The difficulty depends (probably) on one of the nicest points about refraction, namely, the thermometrical correction. It is perhaps not easy to ascertain the exact temperature of the air at the time of the sun's passage; and perhaps the difference between the temperature within and without the transit-room (that stumbling-block to astronomers and theorists,) may then be considerable. Perhaps also the colours on the sun's limb, produced by atmospheric dispersion, may produce some doubt. On the whole, I conceive that this question cannot yet be regarded as settled. The subject is well discussed in Cacciatore's Observations.

In the Transactions of the Royal Irish Academy, vol. xii. for 1815, Dr. Brinkley published investigations on refraction, principally astronomical. In vol. xiii. he extended them to observations near the horizon: tables were formed from these materials.

In the Phil. Trans. 1823, Mr. Ivory published a theoretical investigation of refraction: it proceeded principally on the supposition that, on ascending uniformly, the temperature of the air decreases uniformly: the result of this inquiry was given in tables. In the Phil. Trans. 1824, Dr. Young proposed a simple formula for the relation between the density and pressure of the air, which corresponded nearly to Mr. Ivory's.

The most remarkable, by far, of all the investigations of refraction that I have seen, is one by Mr. Atkinson in the Memoirs of the Astronomical Society, vol. ii. To discover the law of the decrease of temperature, this gentleman collected a number of observations of the thermometer, made at various elevations by different persons; and fixed at last upon this law: That uniform decrements of temperature correspond to increments of height which are in arithmetical progression. For the effect of temperature on the density of air, and for the whole refraction of air, the best experiments were referred to. The calculation was effected by a method of quadratures, the air being (for low refractions) supposed to be divided into sixty-four strata. The result is given in tables. It is to be regretted that the untimely death of the author prevented the completion of a second paper on nearly the same subject.

Before quitting the subject of refraction I may point out two

[page] 140

papers in Zach's Correspondance as worthy the attention of the astronomer. In vol. i. the effects of radiation from the thermometer bulb (which had been pointed out by Fourier,Annales de Chimie, vol. vi.)are insisted upon; and the plan proposed by Fourier,namely to use two thermometers of which one has a blackened bulb and the other a clean one, and to apply to the indications of the latter a proportional part of the difference between the two, is recommended to observers. In vol. ii.,Flaugerges proposes to inclose the thermometer in a case consisting of a bright inside surface, and a bright outside surface, with a thin lamina of some nonconducting substance between; the form of the case to be such as will permit the free passage of air. He also observes, that probably the correction for temperature will depend on the hygrometer. It seems to me likely that these precautions might prevent many of the discordances that have been noticed.

Among the various essays which, though less known, are well deserving of attention,may be mentioned, A theory of refraction,by Schmidt; elaborate investigations by Plana, Littrow, Struve, and Schwerd,in their Observations; terrestrial refractions observed by Tralles (Berlin Memoirs, 1804); refractions near the horizon observed by Mechain (Conn.des Temps, 1807), by Oriani (Effemeridi di Milano, 1816), and by Foster (Phil. Trans. 1826); Lee's remarks on atmospheric dispersion,—a valuable paper (Phil. Trans. 1815); and Brinkley's remarks on the observations of the same stars at Dublin and at Paramatta (Memoirs of the Astronomical Society, vol. ii. and Ast. Nachrichten, No. 78.). Rumker's observations also (in his Preliminary Catalogue) are intended, partly, for comparison with European observations,in order to obtain the amount of refraction.

While this subject was pursued at home and abroad, two discussions were going on, confined almost entirely to England. In the Memorie della Società Italiana for 1805 and 1809, and the Berliner Jahrbuch for 1814, Piazzi, Chiminello and Calandreili conceived that they had found a sensible parallax in several stars. Bessel, however, could find no trace of it in Bradley's observations. In the Transactions of the Royal Irish Academy, 1815, a paper appeared by Dr. Brinkley on the same subject; but the parallaxes whose existence he considered to be established, differed considerably from those of Piazzi. In the Philosophical Transactions for 1817, Mr. Fond stated, that with the Greenwich circle no such parallax was discoverable. He proposed however that telescopes should be immoveably fixed to stone piers, for the purpose of observ-

[page] 141

ing stars of nearly the same declination and different right ascensions. This was done at Greenwich, and still the observations gave no indication of parallax. Dr. Brinkley still maintained its existence; and for several years each successive volume of the Phil. Trans, contained a paper on one or the other side of the question. In the course of this discussion, the defects of the different instruments and of the different corrections were closely examined, and in this respect Astronomy has certainly been advanced by the controversy. In the Phil. Trans. 1821, Dr. Brinkley investigated from observations, by the method of equations of condition, the quantities of aberration, nutation, and parallax, and found sensible values of parallax for several stars, especially for α Lyræ. Mr. Pond still doubted of this result; and in vol. i. of the Mem. Ast. Soc. Dr. Brinkley endeavoured to show that the Greenwich observations themselves afforded evidence of parallax. In the Irish Transactions for 1825, Dr. Brinkley attempted to settle the question by an instrumental investigation of the most delicate kind. The quantity of solar nutation (which had never before been extracted from observations,) being smaller than that which he attributed to parallax, it seemed that if the observations were competent to exhibit the former, they might assuredly be relied on for the latter. By equations of condition, therefore, he investigated the quantities of aberration, solar nutation, and parallax, for different stars, and obtained consistent results for the solar nutation, and sensible quantities for parallax. This result would have appeared to me decisive, but for a difficulty which another investigation has made to appear. In the Ast. Soc. Mem. vol. iv. is a most valuable investigation of the quantity of aberration as deduced from a vast number of Greenwich observations, by Mr. Richardson, Assistant at the Greenwich Observatory. The mean result (20"·50) appears to me the most accurate that has yet been obtained. But from different stars different values are obtained; and it is remarkable that the difference between the values for γ Draconis and η Ursæ Majoris,—stars less affected than almost any other by uncertainty of refraction,—is of opposite kinds in the two determinations of Brinkley and Richardson, and that the discrepance far exceeds the quantity which Dr. Brinkley had proposed to investigate. The existence of sensible parallax appears to me, therefore, to be yet undecided.—A few observations and remarks on parallax may be found in Bessel's Fundamenta, and Struve's Observations.

Mr. Pond in the mean time had remarked (Phil. Trans. 1823), that many fixed stars appeared to have an accelerated

[page] 142

motion towards the south. Dr. Brinkley found no traces of this, and in several volumes of the Phil. Trans. the reasons for and against this opinion are fully stated. This question, like the other, seems to be still in doubt.

These discussions, while they have improved the accuracy of the places of the fixed stars, have naturally given rise to different fundamental catalogues. The thermometrical correction of refraction affects the determination of the equinox, because the temperature of the two equinoxes is not the same. As Bradley's refractions are still employed at Greenwich, the Greenwich right ascensions will differ from Brinkley's or Bessel's. The declinations are immediately affected by the refraction. Mr. Pond, several years since found that the right ascensions of all the stars of Dr. Maskelyne's catalogue ought to be increased 0"·31 of time; but he has since thought that 0"·20 is sufficient: no investigations were published with these statements. Several catalogues have been published by him in the Greenwich Observations and in the Nautical Almanac: the last (included in the large catalogue published with the Greenwich Observations for 1829,) is undoubtedly excellent. Dr. Brinkley's catalogue is in the Irish Transactions for 1828. Bessel's catalogues are in various volumes of his Observations, and in the Berliner Jahrbuch. In the Astronomische Nachrichten, vol. v. and vi., are elaborate comparisons of the deelinations of different observers, by Mr. Pond; and in the Berliner Jahrbuch, 1825, by Bessel. In Plana's Observations (Turin Memoirs, 1828), and in Schwerd's Observations, there appear to be good catalogues of declinations.

By comparing late declinations with those of Bradley, &c. the quantity of lunisolar precession is found from different stars, and the difference of each from the mean is held to be the proper motion of that star. In like manner, by comparing late right ascensions with those of Bradley, &c, the quantity of general precession is found. Different values of precession and proper motion are of course obtained by different catalogues. Piazzi's and Maskelyne's depend sensibly on the nutation employed. In the Memorie dell' Istituto nazionale Italiano, 1804, Piazzi gave the proper motion of 300 stars; in the Berlin Memoirs 1801, and the Phil. Trans. 1805 and 1806, Prevost, Maurice, and Sir W. Herschel explained some on the supposition that the sun was moving towards λ Herculis: their Calculations appear very plausible, but the result has not been generally received. Pond's proper motions have been given with his. catalogues; Bessel's precession in the Berlin Mernoirs, 1819 and 1825, and the Astronomische Nachrichten, No. 92,

[page] 143

Brinkley's in the Irish Transactions, 1828. The latter is founded solely on the supposition that (as appears from observations,) the three stars, α Orionis, α. Cygni, and Rigel, have the same relative position as in Bradley's time, and therefore probably have no proper motion. The result as to precession is exhibited as the mean of results from thirteen different stars, but this, taken literally, may convey an erroneous notion to the reader; the result is obtained from one star only, (which, assuming the direction of motion of the pole, is theoretically sufficient,) and any one of the three stars whose relative positions have not altered, or any other star whose place is corrected so as to refer it to the same position with regard to them, would give just the same result. An ample discussion of proper motions may be found in Cacciatore's observations: a valuable paper on the same subject has also been lately communicated by Mr. Baily to the Royal Astronomical Society. The places of some stars in the southern hemisphere have been determined independently by Mr. Rumker.

IV. The places of a number of principal stars being well established, those of other stars are easily established by comparison with them. Catalogues of small stars may therefore be made by astronomers whose instruments are not competent to fix independently the places of fundamental stars. The principal original catalogues in use at the beginning of the century were, one of 387 stars deduced from Bradley's observations, and published in the Nautical Almanac 1773; one by Mayer, of 992 stars; and three by Lacaille*, (of which one included 1942 southern stars; one, 515 zodiacal stars; and one, 307 not confined to any part of the heavens). Besides these there were many compilations (as Wollaston's general catalogue); and many of less authority, and principally catalogues of smaller stars, were published in the Continental Ephemerides, especially in the Connaissance des Temps. In 1800 Wollaston's Fasciculus Astronomicus appeared, including the circumpolar regions to 25° N.P.D. In 1801 the Histoire Celeste was published, comprising observations of 50,000 stars, made principally by Lefrancais Delalande at the Ecole Militaire. Tables for the convenient reduction of these observations, on a principle suggested by Bessel in the Astronomische Nachrichten, No. 2, have since been published by Schumacher (in 1825). These stars were for the most part observed but once. About the same time appeared Bode's Charts of the Heavens, including a con-

* In the Memoirs of the Astronomical Society, Mr. Baily has discussed and republished Mayer's and Lacailie's catalogues.

[page] 144

siderable number of telescopic stars, and his catalogue of 17,000 stars from all authors; it was followed in 1805 by his smaller catalogue. In 1803, Cagnoli's catalogue of 500 stars appeared in the Memorie della Società Italiana (published separately in 1807, with tables for computing the aberration and nutation); and in the same year Piazzi's catalogue of 6748 stars, founded on Maskelyne's catalogue of 1790, was published in the Palermo Observations. The last-mentioned catalogue, revised from the author's fundamental places, and (in his opinion) improved, and extended to 7646 stars, was published as a separate work in 1814. This may well be considered as the greatest work undertaken by any modern astronomer; as not only was every star so frequently observed as to determine its place well, both in right ascension and declination, but every observation was reduced, and the results and their comparison with those of former astronomers exhibited in a clear form. It is still the standard accurate catalogue, as the places deduced from the Histoire Celeste are still the standard approximate catalogue for small stars. In 1806 Zach's Tabulœ Spectales was published. The object of this work, besides giving a catalogue of 1830 zodiacal stars principally from Zach's observations, was to supply facilities for applying the astronomical corrections of aberration and nutation to 494 of the principal stars. And in 1812 the same astronomer published the Tables nouvelles d'Aberration§c. for 1404 stars. In the Greenwich Observations for 1816, Mr.Pond published a catalogue of 400 stars. In 1818 (as I have mentioned,) Bessel published the Fundamenta Astronomiæ, exhibiting the results of all Bradley's observations of stars in a catalogue of 3222 stars; these were compared with Piazzi's, but no means of applying aberration and nutation were given. In 1822, Harding's Atlas Cœlestis was published, comprising a series of charts including every star on the observation of which any dependence could be placed, as far as the 30th degree of south declination.

The principal defect in Piazzi's selection, is the want of stars near the pole; to supply this, Struve at Dorpat observed (in right ascension only)many circumpolar and other stars; he also fixed the places of many minute stars in the neighbourhood of large ones: these observations are contained in the Dorpat Observations. Schwerd also (at Speyer) has observed many circumpolar stars both in right ascension and declination, and has published a chart of this part of the heavens.

Besides these, the following less complete catalogues have been published. Right Ascensions, by Littrow, in the Vienna Observations, and by the writer of this paper in the Cambridge

[page] 145

Observations. Declinations by Schwerd, in the Astronomische Nachrichten. Southern stars by Rumker, in several parts of the same work; and by Fallows and Rumker, in the Phil. Trans. for 1824 and 1829. Declinations by Bianchi, in the Milan Ephemeris 1830. And very lately a catalogue of 632 southern stars by Rumker, with constants for the reductions, has arrived in this country.

In the Greenwich Observations 1829, Mr. Pond published a catalegue of 720 stars lately observed at Greenwich; it is understood that he is now occupied with the extension of it. Mr. Groombridge observed a great number of stars, principally circumpolar, which have been reduced at the expense of the British Government, but are not yet published. Dr. Robinson is, I believe, now employed in re-observing the whole of Bradley's stars.

Bessel has for several years been much employed in observing (with Reichenbach's circle) all stars as far as the 9th magnitude in zones. In general, the transits of these stars are observed at only one wire; their places therefore are only approximate. The unreduced observations, with the elements of reduction, are published in the Königsberg Observations. These zones have been extended from 15° south declination to 45° north declination.

In 1825 the Berlin Academy invited astronomers to join in forming charts of the region of the heavens 15° on each side of the equator (Astr. Nachr. No. 88). It was proposed that each observer should take one hour of right ascension; that having formed a chart including all the stars of the Histoire Céleste and Bessel's Zones, he should put in, by estimation only, all the stars that could be seen with one of Fraunhofer's telescopes of 34 lines aperture. Some parts of this are in progress: three hours (one by Mr. Hussey, one by the Padre Inghirami and Capocci, and one by Göbel,) have been completed and are engraved.

Tables for the reduction of stars were published by Dr. Pearson in 1824; by Mr. Baily in 1827; by Mr. Groombridge and Sir John Herschel, in the Mem. Astr. Soc. vol. 1, and by several others. But in the Astr. Nachr. No. 4. Bessel proposed a method which for most purposes seems likely to supersede every other. The reduction of the right ascension as well as of the declination of a star may be expressed by the sum of four products, one term of each depending only on the year and day, the other depending only on the place of the star. The numbers depending on the day have been published in the German periodical works; and in Schumacher's Hülfstafeln

K

[page] 146

for 1822, the numbers depending on the star's place were given for 500 stars. (Numbers are also given in the Berliner Jahrbuch, depending on the day, and adapted to a different system of reduction.) One of the first important acts of the London Astronomical Society, was to publish (under the super-intendance of Mr. Baily,) a catalogue of 2881 stars, founded on the observations of Bradley and Piazzi, accompanied by the numbers depending on the place of the star; to be used in a method identical (in all essential parts,) with Bessel's. Since that time the numbers depending on the day have been published in the Supplement to the Nautical Almanac: and with these the Astronomical Society's Catalogue, though somewhat less accurate than those founded on late observations (as Mr. Pond's), is far more convenient for use than any other.

Within a short time, a volume of Tables has been published by Bessel under the title Tabulœ Regiomonfanœ, which must have a powerful influence on the state of Astronomy. Besides all tables wanted for ordinary reductions, this volume contains all the numbers depending on the year and day, which are necessary for reducing observations from 1750 to 1850. The advantage of being able to reduce on a uniform system, and by easy methods, all the accurate observations that have been made, can be easily conceived by those who have had occasion to discuss distant observations.

V. At the beginning of the century, our only accurate knowledge of double stars and nebulae was founded on Sir W. Herschel's observations, made nearly twenty years before. A few measures are to be found in Wollaston's Fasciculus. In the Phil. Trans. 1802, Sir W. Herschel published a catalogue of 500 new nebulae of various classes, with remarks on the constitution of the heavens; and in the Phil. Trans. 1803, a paper "On the changes in the relative situation of double stars in 25 years." This may be considered as the epoch of the creation of the science in the form in which it now exists. In the same work for 1804, he continued the subject. In 1811, he published a paper on nebulæ, and on the constitution of the heavens; in 1814 one on the same subject, in which he noticed the breaking up of the Milky Way in different places, apparently from some principle of attraction; and in 1817, one on the local arrangement of stars, and on the Milky Way. These memoirs contained those remarkable ideas on the distribution of the stars in our own cluster between two parallel planes, and on the connexion between stars and nebulae, (the former appearing sometimes to be accompanied by the latter, sometimes to have absorbed a

[page] 147

part, and sometimes perhaps to have been formed from them,) which have since been generally received. The additions made to the subject since that time consist of little more than accumulations of observations (with the exception of one set of deductions, of which I shall speak presently). Sir W. Herschel's last paper was a catalogue of 145 new double stars, without accurate measures, communicated to the Astronomical Society in 1822, and printed in the first volume of their Transactions. In the Dorpat Observations, some measures of the positions and distances of double stars were given, and a catalogue of the places of 795 double stars, from all authorities, but without measures. Some were also observed by Bessel, and a catalogue of 257 is given in the Königsberg Observations part 10, and in the Ast. Nachr. No. 88, with estimations of distance, but no angles of position. In 1827, Struve published his Catalogus Novus, containing the places of 3112 double stars; not measured, but classified by estimation of their distances. This work (the fruit of two years' labour, with Fraunhofer's large telescope,) contains all the double stars of a certain description to 15° S. declination. And though the want of measures renders it inapplicable for the speculations which had been and are now grounded upon measures only, yet this must always rank as a very valuable catalogue. While this was going on, Sir John Herschel and Sir James South published (in the Phil. Trans. 1824,) accurate measures of 380 double and triple stars. Sir James South published (Phil. Trans. 1826,) measures of 458, and a reexamination of 36 of the former list; and Sir John Herschel added remarks on the changes apparently going on. There can be no doubt of the very great value of these determinations. Amici however, in Zach's Correspondance, vol. 8, has called in question the accuracy of some of the measures. In the Mem. Ast. Soc. vol. 2. is a comparison, by Struve, of his own measures with those of Herschel and South. In 1826, Sir John Herschel presented to the Astronomical Society a catalogue of 321 new double stars, the distances and positions being given by estimation, with remarks on the great nebulæ of Orion and Andromeda; in 1827, one of 295 stars; in 1828, one of 384. About the same time Mr. Dunlop published measures of 253 southern double stars (Ast. Soc. Mem.), and remarks on the southern nebulæ(Phil. Trans.). In 1830, Sir John Herschel communicated good measures of 1236 stars, made with a 20-feet reflector; and lately, in vol. 5. of the Ast. Soc. Mem. he has given accurate measures of 364 with an achromatic telescope. This last paper is the most interesting that has been mentioned, exhibiting all the most striking results as to the motion of double

k2

[page] 148

stars that have yet been obtained. In the Monatliche Correspondenz, vol. 26, Bessel had speculated on the relative motion of the stars of 61 Cygni; and in the Phil. Trans. 1824, maps of the apparent relative motions of 61 Cygni and ξ Bootis were given. In many cases it is doubtful whether the apparent motion may not be produced by the motion of our system (supposing the stars unconnected and at very different distances), and whether a part of it may not depend on annual parallax. But in this paper it is shown that ζ Cancri and ξ Ursæ Majorishave nearly completed an entire revolution since they were first observed; that η Coronæ has probably made more than a revolution; and that η Castor, γ Virginis, σ Coronæ, 70 Ophiuchi, 61 Cygni, and others, are undoubtedly connected as binary systems, and have changed their position remarkably.—Other private observers, I believe, in this country, are employed on the measures of these objects.

The belief in the connexion of double stars by some law of attraction naturally excited a desire of reducing their orbits to calculation. Every attempt that has been made has assumed the law to be that of gravitation. In the Conn. des Temps 1830, Savary gave a method requiring four complete observations of distance and position, which he applied to determine the: relative orbit of the two stars of ξ Ursæ Majoris. (In the history of methods it is remarkable that one of the distances actually used by him for £ Ursæ Majoris was concluded from the others by the ratio of the angular motions.) In the Berliner Jahrbuch 1832, Encke gave a method, also requiring four complete observations, which he applied to 70 Ophiuchi. But Sir John Herschel has lately communicated to the Astronomical Society a method which, for elegance and practical utility, must I think be placed above every other that has appeared. For reasons of which only an observer can judge, he rejects entirely the measures of distances; and from the observed angles of position only (of which any number can be used, and the more the better,) he obtains, by a singularly skilful mixture of graphical construction and numerical calculation, all the elements of the orbit. This method has been applied to γ Virginis, σ Coronae, Castor, 70 Ophiuchi, and ξ Ursæ Majoris; and an ephemeris of the first (whose position will change very rapidly in the next few years,) is now published in the Supplement to the Nautical Almanac. This is really a new step in science.

A very extensive series of observations of nebulae, it is understood, is nearly completed by Sir John Herschel; but nothing has yet been published.

Among the changes in nebulæ that have been suspected, one

[page] 149

of the most remarkable is that pointed out by Cacciatore (Ast. Nachr. No. 113). In a place where Lacaille, Piazzi, and Cacciatore himself, had formerly seen a star, Cacciatore in 1826 saw a nebula; and this nebula has since been observed by Capocci and Dunlop. The only doubt is whether the telescopes with which it was seen before were good enough to discriminate between a star and a nebula; and on this point I cannot pretend to decide.

In 1826 a new star (extremely small) appeared in the great nebula of Orion, and was observed by Struve and Herschel (Ast.Nachr. No. 138,Ast. Soc. Trans. vol. 2). Whether it is a periodical star or a new star does not appear certain.

Nothing remarkable, so far as I know, has been added to our knowledge of variable* stars. Many remarks are to be found scattered in the German periodicals, and some in the English Transactions, but none which appeared to be worth extracting.

VI. The planetary tables in highest repute about the year 1800 were those published in Lalande's Astronomy. A review of these will show that astronomers hardly yet expected Tables to represent the places of the heavenly bodies with accuracy, but rather confined their use to approximate prediction; in fact, the theories of perturbation were used no further than was necessary for this purpose. The Solar Tables (calculated by Delambre,) were founded on Laplace's theory and the Greenwich observations; the Lunar Tables (Mason's of 1780, with very small alterations,) were founded on Mayer's theory; but the coefficients of the inequalities were obtained from observation. The Tables of Jupiter, Saturn, and Uranus, (by Delambre,) were founded on Laplace's theory, as the magnitude of their equations made it impossible to dispense with them; but those of Mercury, Venus, and Mars, (by Lalande,) had no effects of perturbation. The Tables of Jupiter's satellites (by Delambre,) were founded on Laplace's theory and one thousand observations. Besides these Tables, however, there were others by Zach, Oriani, Triesnecker, &c, which were also much esteemed.

In several volumes of the Berliner Jahrbuch at the beginning of the century, formulae are investigated for the perturbations of Mercury, Venus, the Earth, and Mars. In the Berlincr Jahrbuch for 1806, Wurm investigated the correction of the mass of Venus from the perturbations of the earth; he found

* I have lately found that the star 42 Virginis, which was observed by Flam-steed and described as of the 6th magnitude, but which was lost in the last century, still exists in the same place, but is not brighter than the 11th magnitude.

[page] 150

that the mass ought to be increased. But as no other errors. were taken into account, no great value can be attributed to this result.

In 1804, Zach published complete Tables of the Sun, founded on the observations made at Gotha. In 1806, the French Board of Longitude published Delambre's Solar Tables, which (till within a short time,) have been generally adopted. They were founded on observations by Bradley, Maskelyne, and Delambre, and on Laplace's theory; the masses of Venus and Mars, as well as the other elements, being determined by the discussion of the observations. In 1809, Zach published his Tables abrégées et portatives, differing little from the larger Tables except in the arrangement, which, giving more trouble to the computer, required less space. In the Milan Ephemeris 1810 and 1811, Carlini published his Solar Tables. By a new arrangement (making the difference of successive values of the arguments the same as the alteration due to one day), he has diminished very much the labour of calculating a solar ephemeris; though for the calculation of an independent place, his system gives no particular facilities. The elements of these Tables are the same as those of Delambre's. In the Conn. des Temps for 1816, Burckhardt gave the results of a comparison of Delambre's Tables with a great number of Maskelyne's observations (far greater than the number on which they were founded). It appeared that the epoch, the perigee, and the eccentricity, required sensible alterations, and that the mass of Venus ought to be reduced about 1/9th, and that of the Moon to be sensibly diminished. In Lindenau's Zeitschrift for 1817, Littrow arrived at nearly the same results, except that he diminished Mars considerably. In the Phil. Trans. 1826, Sir James South gave 86 observations of the Sun, compared with the Tables; which I discussed in the Phil. Trans. 1827. In 1827 the writer of this paper compared Delambre's Tables with 1200 Greenwich observations made with the new transit, and deduced from them the corrections in the elements. These agreed closely, in general, with Burckhardt's, excepting that a diminution of Mars appeared necessary. Some discordancies however led him to suspect the existence of an inequality that had escaped the sagacity of Laplace and Burckhardt, and a new term was at length found and calculated. This was announced in the Phil. Trans. 1828. Corrections founded on these alterations of the elements have for some years been published in the Nautical Almanac. In the Astronomische Nachrichten,Nos.133 and 134, (March 1828,) Bessel gave the result of a discussion of Bradley's and his own observations. Adopting Burckhardt's masses of Venus and Mars, and a mass of the moon nearly corresponding to Lin-

[page] 151

denau's value of nutation, (and therefore smaller than any other received mass,) he put in a tabular form the corrections to be applied to Carlini's Tables. (Some numerical results of these had been published about half a year before.) These have been adopted in the Berlin and other ephemerides; in that of Milan they are adopted, excepting the mass of the Moon, for which mine is substituted. Nothing was added by Bessel to the theory. In Nos. 172, 179, and 217 of the Ast. Nachr. the corrected Tables are compared with observations; in the last place Bessel conceives that something is still wanting to the theory. I have also compared more than 200 Cambridge observations with the Berlin Ephemeris, and I think that this suspicion is well founded. It is understood that Bessel is employed on more complete solar Tables.

The change in the obliquity of the ecliptic, and the length of the solar year, are obtained from discussions of solstices and of solar Tables. The former of these are scattered about very much; but a most able discussion of all the valuable conclusions, with reference to both these objects, is contained in Cacciatore's observations. The annual diminution of obliquity is now almost fixed at 0"·45. The mass of Venus given by this number agreesnearly enough with that obtained from the inequalities of the Sun's longitude.

Little has been done in observing the solar spots, &c. Some observations are contained in the Conn. des Temps 1805, and the Berhiner Jahrbuch 1828; one of the best papers is perhaps that by Mosotti in the Milan Ephemeris 1821. In 1827, the Frankfort Society published some figures &c. of spots observed by Sömmering.

During this century, several astronomers, (in the German periodicals,) from comparisons of the duration of the sun's transit with the difference of zenith distance of the upper and lower limbs, had been led to the conclusion that the Sun's figure is that of a prolate spheroid. As two observers seldom give the same duration to the Sun's passage, this notion seemed in itself to deserve little attention. In the Milan Ephemeris, however, for 1821, is a series of observations with a divided object-glass by M. Mosotti, which seem to establish the sphericity of the Sun.

In 1806, the French Board of Longitude published Burg's Lunar Tables. In these the arguments of the inequalities were taken from Laplace's theory, and the coefficients from the Greenwich observations. In one instance only were so few as 668 equations of condition used to determine the value of a coefficient. They were compared with observations, and received the prize of the French Institute. In these, for the first time, Laplace's (or rather D'Alembert's) equation of long period

[page] 152

was introduced, the coefficient of which was entirely empirical. In 1809, Zach put them in a portable form. In 1812, Burckhardt's Tables appeared, differing a little in the numbers, and a little in arrangement, from Burg's, and containing also a greater number of equations; still the coefficients were obtained from observation. It was felt that the lunar theory was imperfect so long as appeal to observation for more than the six fundamental elements was necessary, and the Institute offered a prize for the lunar theories and Tables which should borrow nothing more. Two able theories were produced, and on one of these Damoiseau's Tables (published 1824,) are founded. They include a greater number of equations than the former, and are simpler in arrangement. In the Conn. des Temps 1828, they are compared with observations. Laplace's long equation is here entirely rejected. In fact, Burckhardt (Conn.des Temps 1824,) shows that other equations will do as well; Carlini (Effemeridi di Milano 1825,) on trying four equations depending on different arguments, shows that some are preferable to Laplace's, but that the best of all is a term depending on the square of the time. The theory of the Moon therefore appears still defective. In the Milan Ephemeris 1827, is a most valuable paper by Carlini on the Moon's variation; in which, after comparing theory with his own observations, he arrives at the conclusion that they are not perfectly reconcileable. Among the materials that have been produced for correcting the lunar Tables, I may mention 215 occupations calcultations by Triesnecker (Göttingen Transactions 1800); 100 by Carlini, and some by Oriani (Milan Ephemeris 1812 and 1814); several for the Moon's diameter, by Wisniewski (Petersburgh Transactions torn. 8); the comparisons by the French Board (Conn. des Temps, and Introduction to the French Tables); comparison of Greenwich observations, by order of the English Board of Longitude; a few by Mr. Henderson (Ast. Nachr. No. 176); some by Carlini (Milan Ephemeris 1830); and nearly 200 right ascensions, and several occultations compared by me (Cambridge Observations). Of unreduced observations, none can be compared with the uninterrupted series made at Greenwich. For distant times, Mr. Baily, on the eclipses of Thales and Agathocles (Phil. Trans. 1811), and Oltmanns (Berliner Jahrbuch 1823 and 1824), and Wurm on 20 ancient eclipses (Zeitschrift vol. 3,) are worth consulting. In No. 102 of the Astr. Nachr. are observations of declination at Paramatta, for the Moon's parallax.

In the Conn. des Temps 1822, is a discussion by Nicollet of 124 observations of one of the lunar spots, for the phænomena of libration. On comparison with theory he is led to the re-

[page] 153

markable conclusion, that the Moon is not homogeneous, and has not the form which it would have, had it been originally fluid.

In 1813, Lindenau published Tables of Mercury. They were founded principally on a discussion of 17 transits over the sun's disk. Lindenau concluded from these that a considerable increase of the mass of Venus was necessary to reconcile theory with observation. The Tables for perturbations are arranged on Carlini's system.

In 1810, Lindenau published Tables of Venus. They are founded entirely on Bradley's observations, and on continental observations of the present century, with the three observed transits. Lindenau would doubtless have preferred a continuous series of observations made at Greenwich, but the observation of Venus has been almost entirely neglected there. The secular variations of the orbit deduced from these observations do not agree with those given by Laplace's (or Lagrange's) theory; and Lindenau thinks that the mass of Mercury ought to be much increased. In Zach's Correspondance, vol. 13, Plana asserts that this difficulty is at present insuperable. Olbers (Monat. Corr. vol. 22,) prefers the theoretical variations. I may mention that it appears from my comparison of observations with the Berlin Ephemeris, that these Tables admit of sensible improvement. These Tables of Venus, (and Bouvard's of Jupiter,) were compared with late Greenwich observations, by order of the English Board of Longitude. In 1811, Reboul published Tables founded on the elements given by Lindenau, Mon. Corr. voL 10. Elaborate discussions of the transits of 1761 and 1769 have since been published by Encke, in separate works.

In the Milan Ephemeris 1801, and the Mon. Corr. vol. 2, are discussions of the elements and perturbations of Mars by Oriani and Wurm. In 1811, Lindenau published Tables of Mars. The Greenwich observations were used as far as possible; but as the observation of Mars was finally abandoned there, he had recourse to continental observations. The variations of the elements agree nearly with Laplace's. In the Ast. Nachr. No. 191, are physical observations of Mars at the opposition of 1830, by Beer and Mädler; they have fixed his time of rotation at 24h 39m. Many physical observations of Mercury, Venus, and Mars, by Schröter, at the beginning of the century, are to be found in his works and in the Berliner Jahrbuch. In the Phil. Trans. 1831, are remarks on Mars by Sir James South; two obsecrations on stars seen very near the planet, lead him to doubt the existence of any extensive atmosphere. In No. 29 Ast. Nachr. are observations by Rumker at Paramatta, for the parallax of Mars. Tables of Jupiter and Saturn, founded

[page] 154

on Laplace's theory, were published by Bouvard in 1808, but were soon suppressed, as it was found that in Burckhardt's addition to Laplace's theory, several terms had been applied with the wrong sign (in consequence of mistaking the perihelion for the aphelion). A new set of Tables was therefore published in 1821, with the improved theory, and founded on all the good observations of modern astronomy. In discussing these, values are obtained for the masses of Jupiter, Saturn, and Uranus.

In the Ast. Nachr, No. 97 and 139, are micrometrical measures of Jupiter and Saturn by Struve. He determines the flattening of Jupiter to be 1/3·7, and the inclination of Saturn's ring to the ecliptic to be 28° 5′. Bessel (Berlin Ephemeris 1814 and 1829,) had made it about 28° 22′. These values are considerably less than that formerly received (about 31° 20′). In No. 189 Ast. Nachr. are measures of Saturn by Bessel, with a divided object-glass.

In the Phil. Trans. 1805, 1806, and 1808, Sir W. Herschel gave observations of Saturn's figure. It appeared that about latitude 45° the planet projected above the elliptic form. I think it worth mention that I have myself witnessed an instance in which a person, who had never heard of this observation, on seeing the planet very distinctly, made spontaneously the same remark. I have many times seen the planet with extreme distinctness, and have on one occasion thought that it certainly had this shape; and on another, have been equally convinced that it is rather flattened at latitude 45°. The shape assigned by Sir W. Herschel (See Monat. Corr. vol. 15, and Cambridge Transactions, vol.2) cannot be reconciled with theory.

In 1821, Bouvard published Tables of Uranus (in the same volume with those of Jupiter and Saturn). With respect to this planet a singular difficulty occurs. Seventeen observations of Uranus have been found in the observations of Bradley, Mayer, &c. (for discussions of which see the Zeitschrift, the Conn. des Temps, &c.); and since its discovery as a planet in 1781, observations have not been wanting in any year. Now it appears impossible to unite all these observations in one elliptic orbit, and Bouvard, to avoid attributing errors of importance to the modern observations, has rejected the ancient ones entirely. But even thus the planet's path cannot be represented truly; for these Tables, made only eleven years ago, are now in error nearly half a minute of space.

Delambre's new Tables of Jupiter's satellites (for eclipses), published in 1817, were founded on all the observations that he could collect from 1662 to 1802, and on Laplace's theory; and will probably want little alteration for some years. It is to be

[page] 155

regrettedthat no measuresof the elongations of these satellites have been made, as they would throw much light on the mass of Jupiter, upon which (as I shall mention hereafter,) there is at present considerable doubt.

The motions of Saturn's largest satellite have lately attracted some attention. In the Berliner Jahrbuch 1814, is a discussion of Bessel's; from the motion of its apse he concluded the mass of Saturn's ring to be 1/213 that of the planet. In the Zeitschrift 1817 be predicted a series of conjunctions which it was desirable to observe. In the Ast. Nachr. Nos. 193, 194, 195, and 214, he gave new investigations (from observation of conjunctions and of the passage of its shadow on Saturn); he condueled the mass of the ring to be 1/118 that of Saturn, and found for saturn's mass a value agreeing nearly with Bouvard's. In No. 208 of the same work, is a prediction of eclipses by Mädler.

On the satellites of Uranus nothing is known except what was published by Sir W. Herschel, Phil. Trans. 1815; though it is understood that his conclusions as to the positions and dimensions of their orbits have been verified by Sir John Herschel.

A new method of giving the places of planets has been introduced, principally by Gauss (Monat. Corr. 1812, and Theoria Motús), namely, of giving the place, referred to the sun, by rectangular coordinates, two of which are parallel to the earth's equator. The sun's place, referred to the earth, being given in the same way, the coordinates of the planet referred to the earth are found by simple addition, and from these the right ascension and declination are found with great ease. This method is generally used for comets: in the Astron. Trans. vol. 3, Littrow proposed to use it for planets: and Weisse in 1829 published Tables for all the planets. These Tables admit of the introduction of secular change of the elements, but not of periodical perturbations: and on this account I think that they will now be little received.

A vast number of observations of planets is to be found in the Transactions, the Ephemerides, and the astronomical periodicals. Their object however is generally rather confined. The inferior planets are little observed: the superior, little except at opposition. At the regular observatories they have been much neglected. In the Berliner Jahrbuch 1816, it is remarked that in two years there were only six observations of planets at Greenwich. The foreign observations are sometimes given without any comparison: sometimes however (especially in the Milan Ephemeris,) they are compared with the

[page] 156

Tables, and even the equations of condition for correcting the elements are formed (as in Milan Eph. 1822). In reflecting on these circumstances, it appeared to me desirable that one set of good instruments should be devoted to the observation of planets: and when the Cambridge Observatory was put under my care, I determined on making the planets my principal object. I hope in a few years to collect a mass of observations directed to this point that will possess great value. I have already obtained and compared with Tables about 1100 right ascensions of planets, besides numerous observations of the sun and moon.

VII. At the beginning of the century the only bodies recognised as belonging exclusively to the solar system were Mercury, Venus, the Earth, Mars, Jupiter, Saturn, Uranus, the satellites of these planets, and Halley's comet. As to Lexell's comet of 1770, whose orbit appeared to have been changed by the action of Jupiter from a parabola to an ellipse of short period, it was generally believed that by again passing near to Jupiter it had been so much deflected that probably it had completely left the system.

On Jan. 1, 1801, Piazzi discovered a moveable body. It was generally observed in Europe during 41 days, in which time it described an arc of 3 degrees; when it was lost from its proximity to the sun. The calculation of its orbit was taken up entirely by the German astronomers. They soon found that the supposition of a parabolic orbit (which was the only one that had usually been made,) could not be applied with the least success: and Gauss invented a new method (which with some alteration was afterwards published in his Theoria Motús). He at length announced that this body was a planet, moving in an orbit rather more eccentric and more inclined to the ecliptic than those of the old planets, and intermediate in distance from the Sun to Mars and Jupiter. Its discoverer gave it the name of Ceres Ferdinandea. The joy of the German astronomers at this discovery was undoubtedly increased by the circumstance, that the mean distance of the new planet gave continuity to a curious law empirically established (as a rough representation of the distances of the successive planets,) by Bode, in which one was wanting between Mars and Jupiter. Their essays are generally headed, "On the long-expected planet between Mars and Jupiter," or with some similar title. So accurate were Gauss's elements, that in the beginning of December of the same year it was found again, and has since been regularly observed at most observatories (at least the

[page] 157

continental ones). At every fresh opposition the German astronomers have corrected the elements of the orbit: the perturbations have been regularly applied: and the place is now predicted with very great accuracy. The principal information respecting this is contained in the German periodicals: but much will be found in the Milan Ephemeris, and some in the Connaissance des Temps.

In March 1802, Olbers, in the examination of stars near Ceres, discovered another planet (Pallas), smaller than the former and moving in an orbit much inclined to the ecliptic. The general history of the discovery and improvement of its elements is the same as that of Ceres: but one curious consideration was suggested by the comparison of the two orbits. Their major axes were so nearly equal, (the order of magnitude being sometimes changed by the perturbations of Jupiter,) and their orbits approached so near at the intersection of their two planes, that Olbers started the hypothesis of their having been originally parts of a larger planet. If this were true, it seemed probable that there might be other parts; and if these were describing orbits round the sun, the intersection of their planes must fall nearly at the same point. By examining the parts of the heavens corresponding to the two intersections, such planets must infallibly be found.

On this principle, the German astronomers proceeded in a systematic look-out for new planets. Olbers in particular examined, once in every month, a certain portion of the heavens. In September 1804, Harding discovered Juno: and in March 1807, after monthly examinations during three years, Olbers discovered Vesta. No others have been found, though the same system of examination was long kept up. In Lindenau's Zeitschrift, vol. 1, is a notification by Olbers, that he had examined the same parts of the heavens with such regularity that he was certain no new planet had passed between 1808 and 1816. Nothing can give a more forcible idea of the perseverance which led to these discoveries*.

The elements of all these orbits have been successively improved (entirely by the Germans); the perturbations are calculated; and the places for some time before and after opposition are now given in the Berlin Ephemeris. I have lately observed, and compared with the Berlin Ephemeris, the right ascensions

* In the Berlin Ephemeris 1817, is a list of eight lost stars, none of which is either of the new planets; and in the Monat. Corr. vol. 15, Harding states that he misses 24 stars of the Histoire Celeste, and that he has six times observed stars which he has not been able to find again. One such instance (apparently quite free from doubt,) has occurred to myself.

[page] 158

of Juno and Vesta, and I find that they are rather more accurate than those of Venus.

Of the successive steps in the theories of these planets, the following are the principal.

In the Milan Ephemeris 1803, Oriani gave formulae for the perturbations of Ceres, on two suppositions of the value of the major axis; also for the perturbations of Pallas, as far as the third power of the inclination, and the second dimension of the eccentricity and its combination with the inclination. In the Berlin Ephemeris 1805, Schubert gave expressions for the perturbations of Ceres by Jupiter; and in the volume for 1809, Pfaff gave similar expressions for the effect of Saturn. In the Monatliche Correspondenz, vol.7, Gauss gave Tables for the perturbations of Ceres, In tom. 1. of the Göttingen Transactions, the same writer discussed the elements of the orbit of Pallas, taking no account of perturbations. In the Memorie della Società Italiana for 1810, Santini gave Vesta's secular variations and formulae for her periodic inequalities to the first order of small quantities, on two hypotheses of the value of the major axis. In the Milan Ephemeris 1815, Carlini gave Tables for the equation of the centre and the reduction of Ceres; in 1816, expressions for the equation of the centre of Vesta. Lindenau remarked, in vol. 1. of the Zeitschrift, that Carlini's Tables for the equation of the centre would be of little use, because the enormous perturbations produced by Jupiter would alter the eccentricity so much that the term depending on a given variation of the eccentricity, would soon be found inaccurate. In the Milan Ephemeris 1819, Carlini gave the equation of the centre for Pallas and Juno, with two values of the eccentricity, together with the alteration for each depending on alteration of eccentricity. In the Monatliche Correspondenz, vol. 28, Burckhardt had given formulae for the perturbations of Vesta, on two suppositions as to the magnitude of its semi-major axis: in the Mem. della Soc. Italiana, Santini gave elements deduced from observations, and complete Tables, including those for the perturbations to the first order of small quantities. In the Connaissance des Temps for 1818 and 1820, Daussy gave very complete Tables for the perturbations of Vesta, including 40 equations. These are still considered standard, except that the Germans prefer calculating the perturbations produced by Jupiter, by the method of quadratures. In the Berlin Ephemeris 1826, Nicolai gave a short paper containing results of great importance deduced from the discussion of observations on Juno. In all the calculations hitherto made, the mass adopted for Jupiter was either that assumed by Laplace (founded on

[page] 159

Pound's observations of the elongations of Jupiter's satellites), or that given by Bouvard (from the perturbations of Saturn), differing little from the other. Now Nicolai stated, that the observations of Juno at 15 oppositions required an increase of about 1/80th in the mass of Jupiter; but that even then the observations could not be well represented; and that he conceived the absolute attraction of Jupiter on Juno, must be different from that upon the Sun. The last conclusion, attacking one of the most important principles in the theory of gravitation, required further examination. In the Berlin Memoirs 1826, Encke discussed all the observed oppositions (fourteen) of Vesta, separating the perturbations produced by Jupiter into two parts, one being Jupiter's attraction on the Sun, and the other, Jupiter's attraction on Vesta, and considering the assumed mass of Jupiter in these two attractions, as liable to two separate errors. The result was, that the absolute attraction of Jupiter on Vesta did not differ from that on the Sun, by more than 1/10000 of the whole, and that Nicolai's mass ought to be increased about 1/300 of the whole. Encke remarks however, that Nicolai's mass will represent the observations very nearly as well; and Gauss has found the same for Pallas. Nicolai's mass is generally adopted by the German astronomers.—In the last-mentioned paper, and in the Berlin Ephemeris 1827, the reader will find an account of the method of quadratures used by the Germans (to which I intend to refer hereafter). In the Astronomische Nachriciten, No. 165, Heiligenstein has given the outlines of the calculation of the perturbations of Ceres for the opposition of 1830.

The methods of determining from observations the orbits of comets may be divided into those which assume parabolic motion, and those which do not: of the former, at the beginning of the century, Olbers's was best known on the Continent, and Lagrange's and Boscovich's in this country: of the latter; Laplace's was the only received one. In the Berlin Mem. 1801, is a method by Trembley. In 1806, Legendre published his methods, (the last Supplement appeared in 1820,) which began without the parabolic assumption, but finally adopted it. It is curious that the only two examples which he has taken for the parabolic orbit, are comets now known to move in very short ellipses, and in which the preference of an elliptic to a parabolic orbit was shown, at the time, by Gauss and Bessel. This is a striking instance of the danger of making our calculations on too restricted suppositions. In 1809 appeared Gauss's Theoria Motûs, still considered in Germany the classical work on this subject. Of the variety and contrivance in the methods given there, it is impossible to give any idea; parabolic motion

[page] 160

is not assumed, (the methods being best adapted to the small planets,) and the tentative part of the operations differs from that commonly used in this respect, that two unknown quantities must be tried. In the Göttingen Transactions, tom.2, is a parabolic method by Gauss. In the Phil. Trans. 1814, Mr. Ivory gave a parabolic method, amounting to the same as Olbers's. In the Berlin Ephemeris 1820, Olbers has given a method of correcting the approximate elements, and introducing the supposition of ellipticity; this, however, had been done by Laplace in the Mecanique Céleste. In the Milan Ephemeris 1817, and the Berlin Ephemeris 1824, Mosotti and Littrow have given methods. Pontécoulant has given a parabolic method in his Théorie Analytique. In the 5th book of the Mécanique Céleste, Laplace pointed out an alteration in his own method, and showed that the preliminary calculations (whose difficulty and inaccuracy had been considered the most formidable objection,) might in fact be made very easy and accurate. Lagrange left some remarks on the orbits of comets, which are published in the last edition of the Mécanique Analytique. In the Mem. Astr. Soc. vol. 4, Mr. Lubbock has shown that supposing the orbit to be parabolic, or supposing the major axis to be known, the equation may be reduced to a quadratic; and in a Supplement he has increased the accuracy of the method, so as to make it applicable to observations at a greater interval. In this method, after an approximate determination of the orbit, on the supposition that it is parabolic, the major axis may be easily found, and may be applied to determine more exactly the orbit; in the present state of the science of comets, this is an important point. Finally, in the Berlin Ephemeris 1833, Olbers has made some additions to his old method.

All these methods (except Laplace's,) require three complete observations, and can use no more; and in every part of the calculations they require accurate numbers for those observations, and calculation with 7-figure logarithms. Laplace's can use any number of observations, and after the preliminary calculations requires no extreme accuracy in any part. The general methods (including Pontécoulant's and Mr. Lubbock's,) tail when the apparent geocentric path passes nearly through the Sun's place.

The calculation of the true anomaly for a given time, by the common elliptic formulæ is troublesome and liable to error when the ellipse is very long. In the Monatliche Correspondentz vol. 12, Bessel gave Tables for finding the true anomaly in a long ellipse; Posselt, in vol. 5 of the Zeitschrift, has also given Tables.

In the Mon. Corr. vol. 14, Gauss found apparently a short

[page] 161

period (1731 days) for the last comet of 1805, which it has since been ascertained is periodical, though with a rather longer time. In 1810, Bessel's "Untersuchungen" &c. on the comet of 1807 was published; and this is an important epoch in the science of comets. The accuracy and long continuation of observations on this comet, seemed to show clearly that an elliptic orbit, though of great length, must be adopted. The author then took into account the perturbations, by methods invented for that occasion and now generally adopted. He then estimated the greatest possible deviation which the determination admitted of; by giving to every observation the greatest error that he thought it could bear, and giving to each such a sign that all their effects were positive or all negative. His conclusion was, that after the comet had left the sensible disturbances of our system, its periodic time could not be less than 1404 years nor greater than 2157 years, and that 1543 years was most probable. In the Berlin Ephemeris 1815, Bessel has found a period of 3383 years for the great comet of 1811. Argelander has published a treatise on this comet; he finally fixed on 2888 years. The most remarkable however of these long comets is Olbers's of 1815. All the calculations of different observers agreed in giving a period of between 72 and 77 years. In a masterly paper printed in the Berlin Memoirs for 1813, Bessel, after correcting the Sun's place, discussing all the observations, calculating the perturbations during and after the time of observation, &c. has fixed on 1887, Feb. 9, as the time of its next return to perihelion. Since that time many periods have been found for comets, of which some have been afterwards rejected. The second comet of 1819, as calculated by Encke, has a period of about 5½ years; and the fourth comet of 1819, a period of 3¾ years. These numbers may be correct, (though these bodies have not been seen again,) as many comets are so small that they can be seen only when near the earth. In the Zeitschrift, vol. 2, Encke gives a period between 66 and 76 years to the comet of 1812, and it seems impossible that it can exceed 100 years; in the Nctchrichten, No. 22, the same writer gave 194 years to the second comet of 1822, which he soon extended to 1550 years; in No. 37, Rumker fixed on 1817 years for the third comet of 1822; in No. 90, Hansen gave 556 years to the fourth comet of 1825: in Zach's Correspondance, vol. 7, Mosotti thought that the first comet of 1822 moved in an ellipse with a period of three or four years, but it was finally judged to be a parabola: in vol. 14, a period of 265 years was given to a comet of 1825. The orbit of the comet of 1824 (Ast. Nach. No. 69,) appears

L

[page] 162

to be hyperbolical. None of these determinations, I suppose, deserve much credit, except where the comet has been long observed and very ably discussed.

In 1818 and 1819, in examining the coincidence of the observed places of a comet discovered by Pons, with the places given by parabolic elements, Encke* found that the supposition of an ellipse of very short period was absolutely necessary; and his first calculations gave 1310 days. It was soon remarked that its elements were similar to those of the first comet of 1805, in calculating which Bessel had remarked (Mon. Corr. vol. 14,) that a parabolic orbit would not represent the observations. The interest excited by the discovery that we had a real periodic comet of short period will best be gathered from the successive parts of Zach's Correspondence, vol.2. Olbers pointed out its identity with that of 1795, on which he had long before remarked (Berlin Ephemeris 1814,) that different calculators had found very different elements. Encke, in the Berlin Ephemeris 1822, showed that the sums of the squares of errors of observation in the comet of 1795, were reduced to less than half by taking an ellipse of 1200 days instead of a parabola. In the same volume, the perturbations for these periods were given. Olbers soon after pointed out that the same comet had been observed before; and this discovery is very curious. In the Conn. des Temps 1819, are given two observations of a comet in 1786; from these alone no orbit could be determined. But Olbers found from the approximate elements, that these were certainly observations of the new periodic comet. Thus a series of observations extending through 33 years, or 10 revolutions of the comet, was established. After very short examination, Encke found (Berlin Ephemeris 1823,) that the periodic time given by the late observations was shorter than that from the earlier, or that the comet was gradually approaching the sun; which would seem to prove the existence of a resisting medium. He however predicted its place approximately for 1822, when, on account of its southern declination, it could not be seen in Europe; happily the Observatory at Paramatta

* In a French elementary work, it is stated that M. Arago first remarked the similarity of the elements of the comet of 1819, with that of 1805. But the discovery was certainly made by Encke in the manner stated in the text. That M. Arago may have conceived there was some similarity, (not much, as may be seen on examining a table of comets,) is quite possible; but nothing followed from this conjecture. Every calculation respecting this comet (except one by Damoiseau, which was a duplicate of one of Encke's,) has been made by the German astronomers.

[page] 163

was established, and it was observed by Rumker. The approach to the sun was confirmed by this observation (Berlin Ephemeris 1826). Damoiseau however, rejecting the earlier observations, found in the later ones no proof of resistance (Conn. des Temps 1827); and Encke himself (Ast. Nachr. No.123,) acknowledged that the supposition of resistance would not reconcile all the observations. It was predicted and generally observed in 1825; and so anxious were astronomers to discover it, that two new comets were found in looking for it; but this return was not favourable for deciding on the question of resistance. Finally, it was predicted and generally observed in 1828 and 1829; and now at last the point was cleared up. The axis of this comet's orbit lies nearly in the plane of Jupiter's orbit, and its aphelion is very near to Jupiter's orbit. Consequently, when Jupiter is in that part of his orbit while the comet is at aphelion, the perturbations of the comet are excessive; and if an erroneous mass is used for Jupiter, its calculated place will be very erroneous. This was nearly the situation of Jupiter between the appearances of 1819 and 1822 (when the perturbation produced by Jupiter in one revolution of the comet retarded the perihelion passage nine days); and the mass assumed for Jupiter by Encke and Damoiseau, in their calculations, was that of Laplace. Upon proceeding in the equations of condition with a term for the determination of Jupiter's mass, a value was found very nearly agreeing with that which Nicolai had found from the perturbations of Juno, and Encke from those of Vesta; and now with the supposition of a resisting medium everything was reconciled. The magnitude of the resistance is such as to diminish the periodic time about 1/10000 of the whole at each revolution; a quantity so large that there can be no mistake about its existence. The history of this discovery is undoubtedly the most curious that modern astronomy has presented. An abstract is given in the Ast. Nachr. No. 210 and 211, and the first part of a longer paper in the Berlin Memoirs has lately arrived. The place of this comet is predicted for the present year; it must be difficult to observe it in Europe, (I know not whether it has yet been seen,) but it has probably been observed at the Cape of Good Hope.

In 1826, M. Biela (a military officer at Prag,) discovered a comet which it appears he had partly expected. Calculation showed that its path was elliptic, and it was soon found that its elements agreed with those of the comet that passed its perihelion about the first day of 1806, (for which Gauss had found a short period.) The elements of a comet of 1772 agreed so nearly, that in 1806 Gauss had thought it probable they might

L2

[page] 164

be the same. It was now found that the later observations might be reconciled by supposing a periodic time of 2460 days, but the earlier observation required 2469 days. There seems no doubt of the identity of the three comets; but as the earlier perturbations have not been computed, it is doubtful whether this difference depends on perturbation or resistance of a medium. In an elaborate paper in the Mémoires de l' Institut, torn. 8, Damoiseau has calculated the perturbations of the mean anomaly and axis major from 1806 to 1826, and those of all the elements from 1826 to 1832; and an ephemeris for the present year, grounded on these, is printed in the Supplement to the 'Nautical Almanac. This comet will pass in the present year, within 20,000 miles of the earth's orbit. The motions of the three new periodical comets (including Olbers's of 74 years,) are in the same direction as those of the planets. The motion of Halley's comet, however, is retrograde.

Much labour has been employed in calculating the elements of Halley's comet for 1835. In the Ast. Nachr. No. 180, Rosenberger has deduced from observations the elements at the last appearance; and in No. 196, the elements at the appearance of 1682. In the results he has given the effects of an error in the assumed value of the major axis. In the Turin Memoirs 1817, is a most elaborate paper by Damoiseau on the perturbations of its elements between 1682 and 1759, and also between 1759 and 1835. I am not aware that the whole of these (which are undoubtedly the best materials,) have been combined to give a prediction for 1835. In the Conn. des Temps 1833, Pontécoulant determines the elements for 1835 by a similar calculation of perturbations applied to the elements which Burckhardt had obtained (Conn. des Temps 1819,) for -1682 and 1759.

A great number of old comets have been calculated, principally by Burckhardt and Olbers, but I know of no interesting result. In the Mémoires de l' Institut 1806, is an elaborate paper by Burckhardt on Lexell's comet of 1770. There seems no doubt that, from the perturbations of Jupiter, its parabolic orbit was changed into an elliptic orbit of about 5½ years, and that this was much altered by the earth's perturbation: but the further' history of the comet is unknown. Burckhardt is inclined to think that it may possibly still be a periodic comet; or possibly a satellite of Jupiter, as it would not at the distance of Jupiter be visible to us.

On the physical constitution of comets we have learnt nothing, except that they appear to be wholly gaseous. In the beginning of the century there were many discussions in Germany re-

[page] 165

specting a comet which some observers conceived they had seen upon the sun's disk. In 1826 M. Gambart found that a comet would cross the sun's disk: he watched the sun most carefully at the time predicted, but nothing was visible. The dilatation of Encke's comet as it receded from the sun, has given rise to some speculations on the nature of the ether pervading space.

I may mention in this place that the method of minimum squares and estimation of probable errors, though applicable to almost all physical calculations, have been most extensively used in calculations for comets, and were in fact first proposed in treatises on comets (Legendre's Nouvelles Methodes, and Gauss's Theoria Motûs). I will not undertake to say that I think the method of minimum squares is unexceptionable in all its applications, or that I attach much more than a relative value to the estimation of probable errors. But I think there is no doubt that these methods have contributed much to the accuracy of modern astronomy, and that in many doubtful cases they have been admirable assistants to the astronomer's judgement.

VIII. The materials upon which a knowledge of the earth's figure was grounded, at the beginning of the century, were the following. The arc measured in Peru by Bouguer, Lacondamine, &c.; that measured in Lapland by Clairaut, Maupertuis, &c.; that in America by Mason and Dixon, &c.; that from Rome to Rimini by Boscovich; and that from Barcelona to Dunkirk, measured by Delambre and Mechain. Besides these there were some others, as one in Piedmont by Beccaria, one in Austria by Liesganig, and one in India by Reuben Burrows, to which little credit was given; and there was Lacaille's measure at the Cape of Good Hope, which could not be reconciled with the others. One arc of parallel had also been measured in France: and one of much greater value in England. The pendulum experiments (serving, with the help of Clairaut's theorem, to determine the proportion of the earth's axes,) were principally scattered observations by De la Croyére, Campbell, Mairan, Bouguer, Godin, Maupertuis, Lacaille, Legentil, Phipps, Malaspina, and Borda. The last of these (confined to Paris,) were the only ones from which great accuracy could be expected; of the others, the only set in which a series of considerable geographical extent were observed by the same persons and with the same instrument, was Malaspina's. The observations of the attraction of Schehallien, and Cavendish's experiments with leaden balls, had given a pretty good knowledge of the earth's mean density.

[page] 166

In the years 1801, 1802, 1803, the arc measured in Lapland (which, according to the calculations of Clairaut and Maupertuis, seemed to present a strange anomaly,) was remeasured and extended by Ofverbom, Svanberg, and others, so as to embrace an amplitude exceeding 1½ degree. For the geodesic part, as well as for the astronomical determinations, the new repeating-circle was used. The conclusions at which they arrived differed from those of Maupertuis, and are more in accordance with those given by other measures. But they did not succeed in pointing out the cause of their difference; and, as far as their measures admitted of comparison, they confirmed greatly the accuracy of the former measure. The former measure has lately been much discussed, especially by M. Rosenberger in various numbers of the Ast. Nachr.; and the general opinion I think is now, that the first measure was the best, and that its anomaly depended only on the ruggedness of the country. In the Phil. Trans. 1803, is an account of the English measure of an arc from the south-eastern part of the Isle of Wight to Clifton in Yorkshire. The bases were measured with Ramsden's steel chain, and the horizontal angles with a large theodolite: the astronomical observations were made with Ramsden's zenith-sector. There is no doubt that, for its length, this was the most accurate arc that had been measured. Yet a point near the middle of this arc presented an anomaly in regard to the direction of gravity. The measure was afterwards extended to Burleigh Moor: and it thus comprehends an arc of nearly four degrees. Two arcs (of which the details are to be found in the Asiatic Researches,) were measured by Colonel Lambton in India. The first of these, near Madras, was of 1½ degree: the other, beginning near Cape Comorin, nearly 10 degrees. The latter has lately been extended by Captain Everest to nearly 16 degrees. The methods adopted in these measures differ in no respect from those of the English measure: and this arc is undoubtedly the best that has ever been surveyed. The French arc from Dunkirk to Barcelona has been extended by Biot and Arago to the little island Formentera in the Mediterranean (near Iviza), and its whole length is now nearly 12½ degrees. Of the excellence of the geodetic part of this there is no doubt; but there seems some reason to doubt the goodness of the astronomical determinations, though no labour was spared by the observers. The account of this forms a conclusion to the Base du Systeme Metrique. The Piedmontese arc of Beccaria has been re-measured with much care by Plana and Carlini: and the account is published in the Operations Géodésiques et Astronomiques en Piémont et Savoie. It is clearly proved that the astronomical part of Beccaria's

[page] 167

measure was erroneous: but the result of MM. Plana and Carlini's measure is still anomalous; perhaps not more so than the form of the country would lead us to expect. I may mention here that Zach, in the Monatliche Correspondentz and in the Correspottdance Astronomique, has shown clearly that Liesganig's measure is worth nothing. An arc has been measured by Gauss from GÖttingen to Altona, of 2 degrees; the astronomical observations being made with Ramsden's zenith-sector: some accounts of it will be found in the Ast. Nachr., and in a small work entitled Bestimmung des Breitenunter-scheides,§c. An arc of 3½ degrees has been measured by Struve, the northern extremity being on an island in the Gulf of Finland. In many parts of this operation, new instruments and new methods have been used: in particular, for the determination of the latitudes, great reliance was placed on the method of observing stars with a transit instrument whose motion is confined to the prime vertical: accounts of this measure are in the Astronomische Nachrichten, The distance on the arc of parallel between Dover and Falmouth having being ascertained in the course of the English survey, and the difference of longitude between them being determined, by Dr. Tiarks, by the transportation of chronometers, the length of an arc of parallel for one degree in a definite latitude is found, and this determination assists much in determining the Earth's figure. But a far longer arc of parallel has been measured, on the Continent, from Marennes (near Bordeaux) to Padua. The geodesic part of this measure had been nearly completed by the French Government, while the country was in their possession; all that was wanting was to connect the surveys on opposite sides of the Alps. This was effected (though not without difficulty,) by Austrian and Sardinian officers. It was then necessary to determine the difference of longitude of the extremities. This was done by dividing the arc into six portions, in each of which a point could be found visible at both its extremities, and observing at each extremity the absolute time at which small quantities of gunpowder were fired at the middle point. The French part was undertaken by MM. Nicollet and Brousseaud: the rest by MM. Plana and Carlini. The result thus obtained is perhaps liable to considerable doubt, as the errors of all the different observations are accumulated. It is unfortunate that the difference of longitude of the extremities has not been determined without any intermediate determination.

The above, as far as I am aware, are all the measures that have actually been made within the present century. But there

[page] 168

are others to which we may look as not far distant. The survey of Ireland that has lately been and is now going forward, is, I suppose, in accuracy and in excellence of arrangement, (I am not speaking of the minutiae of the map, but of the principal triangles, by which the great distances north and south or east and west are to be measured,) superior to every preceding survey. Little is now wanting for the measure of an arc of meridian but the observation of zenith-distances of stars at its extremities. The country is also favourable for the measuring an arc of parallel of considerable extent: and a new method of producing intense light, introduced into practice by one of the gentlemen employed on the survey, will probably give the means of determining the differences of longitude on a long arc without the errors produced by intermediate stations. It is also understood that our Government have long contemplated the repetition or extension of Lacaille's measure at the Cape of Good Hope: and several circumstances lead me to hope that this undertaking, which would perhaps contribute more than any other to our knowledge of the earth's figure, will ere long be seriously taken up. The extension of Struve's arc is in contemplation.

I may state here (though not immediately connected with the subject,) that a vast number of latitudes and longitudes have been determined, accounts of which are to be found in the Transactions and periodicals. Of the longitudes, one of the most important is that of Paris, determined by instantaneous signals as above described (see Phil. Trans. 1826 and 1827). The method of determining longitudes by transits of the moon has been pretty generally introduced (for which in this country we are indebted principally to the zeal of Mr. Baily); and the longitude of Paris has been determined by this means also (Conn. des Temps 1825). Surveys also, of different degrees of merit, have been going on in almost every part of the Continent.

Of pendulum experiments, the most valuable series is that made by Captain Sabine in almost every practicable latitude. Invariable pendulums which had been observed in London (to ascertain the number of vibrations made per day,) were observed in the same manner at all the stations, and again in the same manner on returning to London. In this manner, without ascertaining the absolute force of gravity at any one place, the proportion at different places is found probably with greater accuracy than by any other method. This is the method commonly adopted by the English experimenters. Experiments were previously made at several places in Britain by Captain Kater; and others have been made in different parts of the

[page] 169

world by Captain Hall, Sir Thomas Brisbane, Mr. Goldingham, &c. A vast number of most careful observations by Captain Foster, in his last voyage, has been received in England, and is now (I believe) preparing for the press. Advantage has also been taken of our repeated expeditions to the North Seas to observe pendulums at high latitudes. The method commonly used by the French philosophers was, to observe the absolute length of the seconds pendulum at each station: thus they experimented at several stations in France and Italy, in the Mediterranean, and in Britain. An extensive series, however, made in Freycinet's voyage, and a few in Duperrey's, were made with invariable pendulums. In the course of experiments for ascertaining the absolute length of the seconds pendulum by a new method, Bessel found that the correction applied in all former experiments for the buoyancy of the air was defective. This has been fully confirmed by Captain Sabine's experiments in a vacuum; and Mr. Baily has been actively employed in determining, with superior accuracy, the correction that ought to be adopted. This error, however, produces very little effect on the determinations of the proportion of the force of gravity at different places.

A series of pendulum experiments was made by Carlini, at the Hospice of Mont Cenis, to ascertain the diminution of gravity at the height of a thousand toises. The account of these is given in the Milan Ephemeris for 1824. The result obtained for the mean density of the earth agrees pretty well with that generally received; but the changes which experiment has shown to be necessary in the elements of reduction, throw a little doubt upon its value. The mountain Schehallien (on which Maskelyne's observations of attraction were made,) has been surveyed, and some alteration made in the numerical results: the calculations of Cavendish's experiments have also been corrected. See various volumes of the Phil. Trans.

In the theory, no improvement has been made, I believe, since the time of Clairaut. No satisfactory rule has been given for taking into account the elevation of the station: perhaps the considerations suggested by Dr. Young in the Phil. Trans. 1819, may be regarded as the most useful.

It is generally thought that the measures of arcs give an ellipticity of nearly 1/300 to the earth; some persons considering it a little greater, and others a little smaller. The pendulum experiments, with Clairaut's theorem, give an ellipticity rather greater, though not without remarkable anomalies.

IX. About the year 1800 the following may be considered

[page] 170

AS nearly the state of physical astronomy. The method of investigating the perturbations of the radius vector and longitude and latitude of a planet, and of expressing them by means of a single function, was well understood. The treatise in which this (and nearly everything relating to planetary perturbation,) is given with the greatest extension, is the Mécanique Céleste, a work which contains, without any acknowledgement, a vast quantity of the labours of preceding and contemporary writers. The method commonly referred to by the Germans is Klügel's, given in vols. 10 and 12 of the Göttingen Transactions: it differs little from that of the Mécanique Céleste. The general conception of the variation of elements had long been formed, and expressions had been given for each variation; but as they depended on differential coefficients of the perturbing function with regard to the coordinates of the disturbed planet, and not with regard to the elements themselves, they could not easily be applied to the planets. Still it was possible to use them, and Laplace has used them in one instance. The theory of the secular variations of the elements, the limits of variation of the eccentricity and inclination, the unlimited variation of the perihelion and node, and the permanency of the axis major, were (to a certain degree of approximation,) well understood. The perturbations depending on the second order of the disturbing force were well understood by Laplace. The long inequality of Jupiter and Saturn (a discovery which has been stated, though not quite correctly, to have "banished empiricism from astronomy,") had been calculated, and even the terms of the second order had been included (by proper application of the expressions for the variation of the elements): the acceleration of the moon's mean motion had also been explained, and the inequality depending on the sun's parallax had been pointed out, as well as that depending on the earth's ellipticity. And (which appears to me the greatest step of all,) the remarkable relation between the motions of Jupiter's three first satellites, which exists in consequence of their mutual perturbations, and depends on the second order of the disturbing force, had been explained. These theories had been numerically applied to all the planets, the terms depending on the second and third powers of the eccentricities being (unnecessarily) developed by a method different from that used for the first powers. The lunar theory was almost perfect. The general methods of computing the perturbations of comets had been well explained by Lagrange. With regard to the figure of the planets, Laplace's remarkable theory had appeared. The theorems for precession, change of obliquity of ecliptic (depending on the change of both the

[page] 171

equator and the ecliptic), &c., were almost complete. It will be seen that comparatively little has been added since the beginning of the century.

In 1808, Laplace presented to the French Bureau des Longitudes a Supplement to the third volume of the Mécanique Celeste. Lagrange immediately after produced equivalent results obtained in a different way (Mémoires de l' ζ Institut 1808). These essays may be considered as completing the theory of planetary perturbations. Their object was to express the variation of all the elements by differential coefficients of the perturbing function (supposed to be expanded in terms of the elements and the time), taken with respect to the elements only, and multiplied only by functions of the elements. The perturbations of the elements can therefore be found from the usual expansion of the perturbing function; and then the true position in longitude and latitude can be found by using the elements corresponding to that time as if they were invariable. It has been objected, that whereas we want only three quantities (perturbations of radius vector, longitude, and latitude,) we in fact investigate six (those of the six elements). I believe, however, that in any case the investigation is not more difficult, and in many cases the saving of time is very great. For instance, in an inequality of long period, which is always accompanied by other terms; if the method of variation of elements is used, the development of one term only of the perturbing function is sufficient: if the original methods were used, the development of several terms would be necessary, and the treatment of each of these would be more troublesome. But it is principally with regard to terms of the second and higher orders of the disturbing force that its advantage is felt: it is necessary to substitute in the expressions values of the elements as near as possible to the true ones, and the method therefore becomes a very simple successive approximation, no reference to the longitude &c. being necessary till the whole is completed.

To reduce the calculation of perturbations to a mere me-chanftca]operation, nothing was wanting but the expansion of the perturbing function. This was given in part by Burckhardt in the Mémoires, 1808: the terms depending on the inclination were not included; but those depending on the eccentricities and their combinations were given to the sixth order.

On the variation of constants generally (in mechanics as well as astronomy), and the secular variation of the elements, the most able papers are by Lagrange, Mémoires, 1808 and 1809; Poisson, Journal dc l' Ecole Polytechnique, vol. 8, and Mé-

[page] 172

moires, 1816. The point to which greatest attention is paid is the variation of the axis major. Laplace had previously shown that it contained no permanent terms to the third powers of eccentricity, &c. and the first order of the disturbing force: Lagrange had extended this to all terms of the first order of the disturbing force: Poisson now extended it to the second order of the disturbing force, as far as fourth powers of eccentricity, &c.; and Lagrange showed that the same theorem is true generally to the second order of forces, whether we consider the perturbing body to be itself liable to perturbation, or not.

In the Göttingen Transactions 1816-1818, Gauss investigated the secular variations of the elements, supposing the disturbing body extended over the line of its orbit, the proportion of the thicknesses at different points being the same as that of the time actually occupied in describing a given length. The ingenuity of the transformations, &c., deserves notice, but the theory of perturbations has gained nothing.

In the Memoirs of the Astronomical Society, vol.2, M. Plana made some remarks on the correctness, in point of form, of Laplace's investigation relative to the constant alteration in the axis major, and on the accuracy of his results as to the effect of the attraction of the stars. In the Conn. des Temps 1829, Laplace made some alterations in his investigation of the latter.

In Nos. 166, 167, 168, and 179 of the Astronomische Nackrichten, Hansen has presented the theory (with reference to practical applications,) in a form that well deserves attention. Instead of determining the true longitude by means of the usual elements, all which (including the mean longitude corresponding to any given instant,) are variable, he assumes that the true longitude shall be determined by the usual expression for longitude applied to invariable elements, the mean longitude only (at any fixed epoch) being considered variable. He assumes also that the true radius vector shall be determined by applying the usual formula for the elliptic radius vector to the same invariable elements and variable epoch of mean longitude, and adding to this expression certain variable terms. This method was probably suggested to its author by the observation that, in the great inequalities of long period, the variation of epoch is much more important than the other variations. At all events, it is a form particularly well adapted to the construction of astronomical Tables, and the more so as Hansen found, in application, that the convergence of the terms in this method, especially for the higher orders of the disturbing force, was more rapid than in any other. Laplace's or Lagrange's expres-

[page] 173

sions for the variation of the elements are of course assumed as a foundation for the first investigations.

In the Berlin Memoirs, 1824, Bessel has given a method of investigating separately the effects of perturbation produced by a planet's action on the sun and its action on another planet. This was done in consequence of the agitation of the question, to which I have before alluded, whether the absolute force of the planet on these bodies was the same; a question first started (I believe,) by John Tobias Mayer, Gött. Trans, 1804-1808. The physical investigation consists merely in taking the two terms of the perturbing function separately: this paper however is remarkable for the mathematical part of the process, which by a mixture of general integration and definite integration, assisted by special Tables, seems well adapted to the accurate calculation of planetary inequalities. The subjects of investigation are the perturbations of radius vector, longitude, and latitude (that of the longitude being expressed independently of the radius vector,) to the first order of the disturbing force.

In the Phil. Trans. 1830, 1831, and 1832, Mr. Lubbock has given four papers on the general problem of perturbations. The object of the first of these is to give expressions for the variation of the elements which shall be true to all orders of the disturbing force, (which however holds with regard to Laplace's and Lagrange's expressions,) together with equations in which the eccentric anomaly is the independent variable. In the second it was shown that the perturbations of the reciprocal of the radius vector might be found more readily than those of the radius vector itself. The rest of these papers (relating to perturbations in general,) is occupied with expansions, and with theorems equivalent to those of Laplace, but in a different form. In the Phil. Trans. 1832, Mr. Ivory has also given an investigation of the perturbation of elements, and Mr. Lubbock has shown the identity of the results obtained by perturbation of the elements and by perturbation of the co-ordinates: it is not the object of these papers to extend the theory of perturbations. In the München Denkschriften and the Turin Memoirs, Pfaff and Cisa de Grésy have given various expressions, which however are only equivalent to those of preceding writers.

In the Milan Ephemeris for 1818, and the Memoirs for 1823, Carlini and Laplace have shown that some of the series by which a planet's place is expressed in terms of the mean longitude, cease to be convergent when the eccentricity exceeds 0·62. In the Berlin Memoirs 1816-1817, Bessel has expressed

[page] 174

the co-efficients of these series by means of definite integrals: and in the Conn. des Temps 1825, Poisson has done nearly the same thing. This principle, I believe, has lately been extended by Cauchy. In the Conn. des Temps 1828, Laplace has given means of estimating the value of distant terms in the expansion of the perturbing functions.

I am not acquainted with any other additions to the theories of elliptic motion or perturbation in general.

With regard to the solar theory, Nicolai, in the Berliner Jahrbuch 1820, investigated the secular variations of the Earth's orbit, as a verification of those given by Lagrange and Laplace. In the Phil. Trans. 1828, the author of this Report announced the discovery of a small inequality of long period in the Earth's motion produced by the action of Venus, and a corresponding inequality in the motion of Venus produced by the Earth: the details of the calculation are given in the Phil. Trans. 1832. And in the Milan Ephemeris, 1830 and 1831, (the latest volumes that I have been able to procure,) Carlini has given an investigation, not yet completed, of an inequality in the Earth's motion, depending on the Sun's distance from the Moon's perigee. It has commonly been thought sufficient to consider the motion of the centre of gravity of the Earth and Moon the same as if their masses were united there: but it is quite conceivable that a small error in this may grow up into a sensible inequality; and this, I believe, is the subject of the investigation.

The lunar theory has been much discussed. In the Conn. des Temps 1813, is a paper by Laplace on the inequality of long period, which, from observation, seems to exist in the Moon's motion. In the Mécanique Céleste he had been disposed to attribute it to a term independent of the Earth's form, which had been pointed out by Dalembert; in this paper he inclines to that depending on the difference of the northern and southern hemispheres. This inequality, with an empirical coefficient, was adopted in Burckhardt's Tables. In the Conn. des Temps 1823, Laplace re-investigated the equations depending on the Earth's ellipticity, and on comparing their values with those found by Burg and Burckhardt from observations, fixed on 1/306 as its value. The Institut having offered a prize for a complete lunar theory, in which the values of the co-efficients should be calculated from theory only, two that were sent in were deemed worthy of the prize, one by Damoiseau, the other by Carlini and Plana. The latter, I believe, is not printed; the former is in the Savans Etrangers. The general method pursued by Damoiseau is the same as Laplace's in the

[page] 175

Méc. Cél.; the rule for retaining terms being directed principally by the magnitude of their numerical values, and not by the order of small quantities. The immense calculations of this theory are given with great clearness and attention to order. In the Conn. des Temps 1823, Laplace gave some remarks on the two memoirs, giving generally the preference to Damoiseau's, partly because he had followed Laplace's method. Carlini and Plana replied in an elaborate paper in Zach's Correspondance, vol. 4. Without attempting to analyse it, I shall only remark that I think no one can have an idea of the delicacies and difficulties in a theory of the Moon in the present day, without examining this reply. In another paper in the same volume, they considered one of the most troublesome equations, depending on twice the distance of the perigee from the node. Damoiseau, as well as Carlini and Plana, found that the equation, depending on the difference on the hemispheres, would probably be insensible: and Laplace (Conn. des Temps 1823,) assented to this. In the Conn. des Temps 1824, Laplace has given the investigations of several lunar inequalities of long period. In the same volume, Burckhardt, after discussing several occupations, maintains the necessity of some equa tion not yet given by theory. In the Milan Ephemeris 1825, Carlini suggests an equation depending on six times the distance of the perigee from the node diminished by the Sun's mean anomaly: the period of this would be 1760 years. In Mr. Lubbock's papers, before alluded to, the lunar theory is considered. The author has commenced the investigation in a manner different from that of Laplace, Damoiseau, Carlini, and Plana, by making the time the independent variable in the equations; and has given Tables for facilitating the research of the terms arising from the combination of other terms. He has also given developments of the perturbing function adapted to this case.

In the theory of Mercury, a discussion of an insignificant numerical quantity has taken place between Laplace and Plana, Mem. Ast. Soc. vol. 2, and Conn. des Temps 1829. In the theory of Venus, I believe, the only addition is the term investigated by the writer of this paper, (before alluded to,) and depending on the difference between eight times the mean longitude of Venus, and thirteen times the mean longitude of the Earth. This inequality is small; but as the corresponding inequality of the earth has the opposite sign, and as Venus at inferior conjunctions is very near the earth, the effect of the inequality at those times will be very sensible. In the Conn. des Temps 1820, Burckhardt gave the equations for the prin-

[page] 176

cipal perturbations of Jupiter to the sixth order of eccentricities, &c. In the Mem. Ast. Soc., vol. 2, Plana gave remarks on Laplace's investigations of the perturbations of Jupiter and Saturn depending on the second order of the disturbing force. As the greater part of Laplace's investigation was suppressed, it was only possible to compare the results, and to examine the correctness of some equations given by Laplace. Plana's results differed much from Laplace's; and a simple equation between the perturbations of Jupiter and those of Saturn, given by Laplace, appeared to be incorrect. Laplace in answer (Conn. des Temps 1829, published 1826,) allowed that his equation was not perfectly correct, but maintained that Plana's error was much greater. In the Turin Memoirs 1827, Plana in reply said that Laplace's answer did not enter sufficiently into details. In the Conn. des Temps 1831, Poisson pointed out some terms omitted by Plana. In the Turin Memoirs 1830, Plana again made some calculations, and still obtained results differing from Laplace's. Finally, in a memoir of which an extract is printed in the Conn. des Temps 1833, M. Pontécoulant stated that he had found errors in the calculations both of Laplace and of Plana, and that on correcting these, both determinations agreed. In the Turin Memoirs 1831, Plana acknowledges that this is true. And thus the discussion of these terms appears to be finished, and physical astronomy has gained much from this inquiry, prosecuted at first by M. Plana under the repulsive circumstances of comparing a final result with Laplace's without an intermediate step. An elaborate investigation of the theory of Jupiter and Saturn by Hansen, on the principles which I have described as peculiar to him, has lately been received in this country.

On the theory of satellites, little has been done. In the Conn. des Temps 1821, Laplace has investigated the effect of the long inequality of Jupiter and Saturn on the other bodies of the system, and has shown that they are sensible only in the motions of Jupiter's satellites. In the Ast. Soc. Mem., vol. 2, Plana objected to Laplace's theory, in reference to the seventh satellite of Saturn. Laplace, in the Conn. des Temps 1829, maintained its correctness, which Plana (Turin Mem. 1827,) has again denied. In the Conn. des Temps 1831, Poisson has shown that both methods produce the same results: and here, I believe, the question rests. In the Ast. Nachr., No. 193, are expressions, by Bessel, for those variations of the elements of Saturn's sixth satellite which do not depend on its position in its orbit: the permanent variations and variations of long period, in fact, analogous to the secular equations of the planets.

[page] 177

Of the methods used by the German astronomers for the calculation of the perturbations of the small planets, I can give no complete account. I regret this the more, because the magnitude of their perturbations is far greater than those of any other planets. For though it may not appear, as far as their general theory has yet been carried, that they have equations as large as the great inequality of Saturn, which however is 450 years in passing from one extreme value to the opposite, yet the magnitude of their perturbations in a given time, one year for instance, and the consequent irregularity of their motion, is very much greater than that of Saturn. This only I can state, that the Germans do not generally compute the perturbations of longitude, latitude, and radius vector, but the perturbations of the elements of the orbit; and these, I believe, entirely by mechanical quadratures; in other words, by summation instead of integration, in a method analogous to that which they use for comets. Perhaps in some calculations for Vesta, as in part of those by Encke, Berlin Memoirs 1826, they may use Tables and apply the perturbations directly to the radius vector, &c.: but even in this instance, the most important part of the perturbations, namely, those produced by Jupiter, are computed by quadrature, the elements being corrected for perturbation: and Encke conceives this to be more accurate than the use of Tables. The intervals used here are of forty-two days each, and the fresh corrected elements are used after every sixth or seventh interval.

The groundwork of Lagrange's method for the perturbation of comets (Méc. Cél. tom. 4. liv. 9.) consists in estimating the disturbing forces resolved in the direction of three rectangular co-ordinates, finding the effect of these on the elements of the comet's orbit, and performing the integration by quadratures. The method given by Bessel in the Untersuchungen über die scheinbare und wahre Bahn des im Jahre 1807 erschienenen grossen Kometen, referred to by the Germans as the standard work, consists in resolving the disturbing forces in the direction of the comet's radius vector, a perpendicular to the radius vector in the plane of the orbit, and a perpendicular to the orbit. The inclination and node are referred to the ecliptic. From these quantities, expressions are found for the variations of the elements, which are integrated by quadratures. This is the method used by Bessel in the calculations relative to Olbers's comet, Berlin Mem. 1812-1813. For the comet of 1807, Bessel has calculated the quantities for every thirty days; and for Olbers's comet, he has taken intervals of twenty-five days during its visibility, of one year from 1815 to

M

[page] 178

1883, and of two years from 1833 to the time of its reappearance, 1887. In the Ast. Nachr. Nos. 210 and 211, Encke has described the method by which he calculated the variation of elements of the periodical comet bearing his name, undoubtedly the most troublesome of all. Using the same general methods, the perturbations produced by Mercury were computed for every four days; those of the Earth and Venus for every twelve days; and those of Mars, Jupiter, and Saturn, for every thirty-six days. These calculations were carried on till the comet reached a certain distance from the planets, and then its place was referred to the centre of gravity of the sun and planets. For some details Encke refers to Argelander's treatise on the comet of 1811, a work which I have not been able to procure. For Encke's comet the effect of a resisting medium, whose density is inversely as the square of the distance from the sun, was calculated by the same method. This had been done analytically for several laws of density by Plana, in Zach's Correspondence, vol. 13; it is also noticed by Mosotti,Mem. Ast. Soc. vol. 2. In the Conn. des Temps 1832, Damoiseau has described his own method. Me refers the co-ordinates to the original plane of the comet's orbit, (taking its original axis-major for the axis of one ordinate,) and resolves the disturbing forces in these directions, and finds the variation of elements in terms of these forces, which he integrates by quadratures. As the ordinates of the comet are conveniently calculated by means of the eccentric anomaly, he calculates the variations for given intervals of that angle. This is the method that he has adopted in the Turin Memoirs 1817-1818, for Halley's comet, varying the eccentric anomaly by 1° each time for the perturbations of Jupiter, by 2° for those of Saturn, and by 6° for those of Uranus. He has also used it in the calculations for the comets of short period,Conn. des Temps 1827 and 1830, and Memoires 1826. In the Conn. des Temps 1883, is an extract from Pontéoulant's Memoir on the same comet; he refers generally to Lagrange's method, and states that having with the first elements computed by quadratures the perturbations of the elements through 30° (of eccentric anomaly, I suppose), he has then used the elements so corrected in the calculations for the next 30°, when he has again changed the elements from the result of these calculations; and so on for each successive 30°.

The following additions have been made to the theories connected with the figure of the earth, &c. In the Phil. Trans. 1809, is a paper by Mr. Ivory, of which the most important part is the very beautiful theorem for finding the attraction of a

[page] 179

spheroid generally, on a point without it, from the attraction of a spheroid on a point within it. In the Memorie della Società Italiana, 1810, and the Göttingen Transactions 1811-1813, Plana and Gauss have given theorems founded on the same kind of integration, and adding little to our knowledge of the subject. In the Journal l' deζEcole Polytechnique, tom. 8, is a paper by Lagrange on a difficulty in Laplace's general theory; and in the Phil. Trans. 1812 and 1822, Mr. Ivory has pursued this objection, and given a method of his own, of very great analytical elegance. In the Cambridge Transactions, vol. 2, the author of this Report supported Laplace's correctness with respect to the point objected (as Laplace had done himself in the fifth volume of the Méc. Cél.), and pointed out what he considered to be another defect in Laplace's reasoning. In a discussion on the figure of the earth, Phil. Trans. 1826, I gave a theorem analogous to Clairaut's, admitting of extension to all powers of the ellipticity. In the Conn. des Temps 1829, Poisson gave a very able memoir on the attraction of spheroids. In the Phil. Trans. 1824, and in a later volume, Mr. Ivory introduced a new equation in the consideration of the equilibrium of fluids of which the particles mutually attract each other; the necessity for this has not been generally allowed, and was explicitly denied by Poisson in a paper (treating also on other points,) in the Conn. des Temps 1831. In the French Memoirs 1817 and 1818, Laplace has applied his general method to the case of an irregular nucleus covered by a fluid; the most general case that can be conceived, and the case that comes nearest to the state of the earth, but which analysis has not yet completely mastered. In the Conn. des Temps 1821, he gave as a consequence of this theory, that the gravity on a continent reduced to the level of the sea by the factor depending only on the distance from the earth's centre, follows the same law as at the surface of the sea. In the Journal de l' Ecole Poly technique, tom. 8, Poisson investigated the motion of the earth's axis of rotation within the earth itself (considering the motion of the axis in space as completely treated in the Méc. Cél. liv. 5). He found that neither the place of the axis nor the velocity of rotation is permanently altered. In the Mémoires, 1824, he has treated of the earth's motion about its centre generally (by variation of constants), and has compared his numerical results for the obliquity, &c., with observation. In the Conn. des Temps 1827, Laplace has alluded to the combined effect of change in the plane of the ecliptic and precessional motion of the earth's axis; and has shown that in consequence of the latter, the limits of the diminution of obliquity are very much contracted.

M 2

[page] 180

In the volume for 1821, Poisson has treated the precession of the equinoxes by Lagrange's method of the variation of constants. In the same volume is a paper by Laplace on the effect which the sea produces on the earth's motion round its centre. In the volume for 1823, he has shown that, supposing the earth's dimensions to have altered by cooling, the effect on the length of the day would not be sensible. In Zach's Correspondance, vol. 14, Plana has deduced from Lindenau's nutation a value for the moon's mass, which however does not agree with that generally obtained from it.

In the Conn. des Temps 1821 and 1822, Poisson has treated of the libration of the moon. His special object is to determine the inequalities in the inclination and node of the moon's equator, depending on her secular inequalities.

I have been obliged almost to confine myself to a bare enumeration of the titles and subjects of these works, partly by the fear of occupying too much space, and partly because it is impossible to give an opinion on the methods and accuracy of many, without having worked through every line of the investigations; a degree of acquaintance with them which, I suppose, no person living can pretend to possess.

I may mention that treatises, of a more elementary kind than the originals, and embracing different parts of the subject of this section, have been published in England, France, Italy, and Germany.

X. In the preceding sections I have endeavoured to give materials for estimating the steps which Astronomy has made in this century, and for understanding its present state, at least in all the important parts. But I cannot forget that the Association which I have the honour to address, while it is a Philosophical Association, is also a British Association, and that while it is anxious to promote science abstractedly, it is also jealous of our national scientific character. I feel therefore that my Report would be incomplete if I did not, in some degree, give means for answering the questions, What has England contributed to the progress of Astronomy? and, How have the knowledge and practice of Astronomy advanced generally in England?

I fear that the answer to the first of these questions will not be very satisfactory. While I allow that in some important parts of Astronomy we have done much, I cannot conceal that in other parts, especially those which cast a lustre on the conclusion of the last century, and those which are peculiarly distinctive of the present century, we have done nothing.

[page] 181

A subject so complicated as Astronomy, may be divided in several different ways, and thus different comparisons may be made as to the progress of its various parts. I shall here view the subject in two different manners, and I will assert:—

First, That in those parts which depend principally on the assistance of governments or powerful bodies, requiring only method and judgement, with very little science, in the persons employed, we have done much; while in those which depend exclusively on individuals, we have done little.

Secondly, That our principal progress has been made in the instrumental and mechanical parts, and in the lowest parts of Astronomy; while to the higher branches of the science we have not added anything.

I must of course refer generally to what has gone before for materials to justify these assertions; but I may here point out a few of the leading facts which have induced me to bring forward these opinions.

With regard to the first, I can assert that we have contributed more than all the rest of the world to furnish materials for ascertaining the figure of the earth. This praise is to be divided, I suppose, between our Government and the East India Company. Be that as it may, I conceive that nothing which has been done by other nations can be put in competition with the arcs of meridian and parallel in England, the great arc of meridian in India, and the pendulum expeditions of Kater, Foster, Sabine, &c. To some of the latter, objections have been made which are in my opinion groundless; but if they were ever so well founded, they would detract nothing from the merit of originating these expeditions. But these expeditions, though they require care and prudence in the persons who conduct them, demand very little science. The vast improvement of chronometers is entirely due to the encouragement offered by our Government. I may also assert that the observatories depending on our Government are maintained with an extent of establishment which few governments would be willing to allow. And in speaking of this, I cannot forbear alluding to one Institution, which I hope some future reporter on Astronomy will be able to describe as having been beneficial to the science. The Observatory at Cambridge was built, not from any fund bequeathed of old for the purpose, nor with the assistance of any other body, but partly by grant of the University as a corporate body, when its funds were ill able to support such an expense, and partly by the private subscription of its members. It was built and is to be furnished on a plan which will enable it to stand in competition with any other at home or

[page] 182

abroad. Whatever may be its success, none is more creditable to the body which founded it.—Now if we examine what has been done by individual attempts, we shall find it small. We have discussed theories of refraction and aberration, perhaps quite as much as our share in the science requires; but we have done nothing in examining the past state of the heavens, or making it subservient to a knowledge of their future state: the reduction of Bradley's observations was left to a foreigner; the formation of Tables of the Sun and Moon, from British observations, even when the theory was put in a distinct shape, was left to foreigners; and, as if we had determined to leave the present state of the heavens also in obscurity, our own observations have too generally been cast on the world unreduced, with a hope, I suppose, that others would have the zeal to reduce them. The observations that require only moderate instruments, with patience and zeal on the part of the observer, as the discovery and observation of comets, and the observation of the small planets, (which on the Continent have generally been made with unmounted telescopes,) have been little attended to. Of the latter, some observations by Mr. Groombridge, some at Greenwich, and a few by myself, constitute, I believe, the whole amount.

I will not deny that there are some exceptions to my general assertion; and in one of these my hearers will anticipate me. I think that I can fix on only two discoveries, the results of combined theory and observation, which are original in the present century, and one of these belongs to an Englishman. New planets and periodical comets had been discovered in the last century; abstract theory of every kind and observations of almost every kind had been produced: but the existence of a resisting medium was established in this century by Encke, and the practical prediction of the phases of double stars is due to Sir John Herschel. Nor can I omit to mention Sir Thomas Brisbane and Mr. Baily, and (for several investigations connected with the physics of Astronomy,) Mr. Ivory, and lately Mr. Lubbock. But after every credit has been given to their labours, it will, I believe, be allowed that the part in which England has contributed most to Astronomy, and which is likely to be mentioned with greatest gratitude by future historians of the science, is that in which she has contributed as a nation.

In proof of the justice of my second assertion, the following remarks may be sufficient. Our instruments I conceive (though a German would not allow it,) to be superior to those of any other nation. The observations at our observatories are conducted, I imagine, with greater regularity and greater steadiness

[page] 183

of plan than those of foreign observatories. This, indeed, is the character which gave (in some respects) preeminent value to the Greenwich observations of last century, and which makes those of the present century highly valuable. In the reduction of these observations we begin to fall off. Though Dr. Brinkley has investigated from observations a new Table of refractions, and applied it to his own observations, yet Bradley's Table, known twenty years since to be sensibly erroneous, is still the standing Table of refractions at Greenwich. The discussion of the reduced observations has been, I think, confined absolutely to the proper motion of stars. On one or two occasions a number of observations of the moon have (by order of the Board of Longitude,) been compared with the then existing Tables, but not with a view of improving the Tables. I have had occasion to mention the correction of the elements of the earth's orbit made by myself (from Greenwich observations), and the discovery, in consequence, of a new equation in the perturbations of the Earth and Venus. As far as I have been able to ascertain, this was the first improvement in the solar Tables made by an Englishman since the time of Halley, and the first addition to the solar theory since the time of Newton. From English observations of planets it has been impossible to extract a result, because scarcely any have been made. To show the extent of this deficiency, I will mention a mortifying circumstance that has occurred to myself. In order to verify completely the equation above alluded to, I was desirous of collecting observations of Venus near her inferior conjunction. In examining the Greenwich observations I found that no opportunity of making this observation was omitted by Bradley or his immediate successor Bliss; soon after the accession of Maskelyne it was wholly neglected; and from that time till several years after his death scarcely an observation is to be found: several conjunctions have been passed over by the present Astronomer Royal; five however have been completely observed. Under these circumstances, (though the deficiency for the latter part of the time only might be supplied from scattered foreign observations,) considering how desirable it is, in a research of some delicacy, to use observations made at the same place, I believe that I shall be compelled to abandon it entirely. The superior planets have been more frequently observed, and those but very little. And generally as to the comparison of theory with observation, and its immediate consequences, the reducing of complicated phænomena to simple laws, or the showing that new supplementary laws are necessary, forming altogether the most glorious employment for the intellect of man, I may state,

[page] 184

in one word, to the best of my knowledge nothing has been done in England. In the lunar and planetary theories we have done nothing, not even in the way of numerical application. In the theory of the new planets and the periodical comets, we not only have done nothing, but we have scarcely known what others have done. With regard to the latter points, the distinguishing discoveries of the present century, our humiliation is great. Some of the new planets are very faint, and all are subject to excessive perturbation. If Astronomy had been confined to England, we never should have rediscovered them, even if we had once made out their orbits. If Astronomy had been confined to England, the paths of the comets would never have been traced, and the consequences deduced from the appearances of Encke's comet, the brightest discovery of the age, would have been lost. While Germans, Italians, and Frenchmen, have emulously pushed on the theory and the observation of these bodies, Englishmen alone, of all the nations professing to support a high scientific character, have stood still.—I am glad to turn from this dispiriting subject.

There are other points to which I can scarcely allude without introducing a degree of personality which cannot be admitted in a public Report. They can be understood perhaps only by those who know the state of observation here, and who have seen the interior of foreign observatories. Of the latter, I can only profess personally to be slightly acquainted with those of France and those of the North of Italy. The characteristic difference between the spirit of the proceedings in England and on the Continent may be stated thus.—In England, an observer* conceives that he has done every thing when he has made an observation. He thinks that the merely noting the passage of a star over one wire and its bisection by another, is all that can be expected from him; and that the use of a Table of logarithms, or anything beyond the very first stage of reduction, ought to be left to others. In the foreign observatories, on the contrary, an observation is considered as a lump of ore, requiring for its production, when the proper machinery is provided, nothing more than the commonest labour, and without value till it has been smelted. In them, the exhibition of results and the comparison of results with theory, are considered as deserving much more of an astronomer's attention, and demanding greater exer-

* I am far from asserting that this is the character of every English observer, and I am equally unwilling to point out any individual to whom it is applicable. My object is merely to explain what I conceive to be the kind of difference which exists between English observers generally and foreign observers generally.

[page] 185

cise of his intellect, than the mere observation of a body on the wire of a telescope. As an instance of the extent to which the reductions are carried there, I may mention that in one Italian observatory where the planets were considered the principal object, not only were the observations freed from instrumental errors and astronomical corrections, but the tabular places were computed by direct use of the Tables, (the ephemeris attached to Schumacher's lunar distances not having reached that country,) and the equations of condition were regularly prepared for the correction of the elements. I suppose such a thing has never been done in England. This system must however contribute powerfully to produce that strong connexion between physical theory and practical observation, which is general on the Continent, but which does not exist in England.

I believe that in the actual state of our institutions, reasons might be found which would seem to render it improbable that there ever can be so strong a connexion; and I can only hope that my view may be incorrect. There is one point with regard to the foreign astronomers to which I cannot help alluding, without however intending to draw any distinct inference. It is, that they have first obtained distinction while in the lower departments of the observatories. Encke's reputation was first acquired, not when he became Astronomer at Berlin, but when he was assistant at Seeberg: and Bessel became known in every part of Europe, not as Astronomer at Königsberg, but as assistant at Lilienthal. Walbeck and Argelander, in similar situations, have arrived at considerable eminence.

I now proceed, and with great pleasure, to consider the second question. And this leads me to explain my opinion on a point respecting which I am anxious that I may not be misunderstood. I am not one of those who have joined in the cry of "the decline of science in England," nor do I believe that in this science there is any foundation for that cry. On the contrary, I assert without hesitation, that it is now and has been for some years rapidly advancing in this country. That there has been a decline, thirty or forty years ago, or rather that we have not kept up with the advances made by foreigners at that time, I am willing to admit. Perhaps this arose from political separation; perhaps in some degree from our pertinaciously retaining a system of mathematics which was insufficient for the deep investigations of Physical Astronomy, (for it was in this principally that we were behind our neighbours). And I have not disguised my opinion that in all the important branches of science we are still behind them. But in all with which I am acquainted a rapid progress has lately been made. In Physical Astronomy more

[page] 186

has been done in England within the last five years than in the preceding century; and this not only with regard to the additions actually made by Englishmen to the stock of results drawn from that science, but also with respect to the number of persons who understand its principles, and who at some future time may be expected to contribute to its progress. In the University with which I am best acquainted, the study of this subject has made great advances. Of the amount and excellence of our geodetic measures and pendulum experiments, and of our discussions of refraction and aberration, I have already spoken. In accuracy of examination and correction of instrumental errors, perhaps something has been gained. In the extension of our star catalogues, much more has been done within a few years than in the whole previous time which followed Bradley's death. In the observation of planets, and the regular comparison of observations with Tables, (the first essential step to the improvement of the latter,) it is hoped that a great advance has been made. The observation of occultations and eclipses has extended; the exhibition of the results also, both for terrestrial and celestial determinations, has increased; and the regular publication of them in the Memoirs of the Astronomical Society, saves from oblivion the past and insures more completely the observation of the future. In the observation of double stars very much has been done. In all this I see grounds for exultation at "the advance of science in England." And when I remark the growing intermixture of physical with observing science, I indulge in the hope that the character as well as the extent of our Astronomy is improving, and that the time is approaching when a person will not in England be considered a great astronomer because he can observe a transit or measure a zenith-distance correctly.

XI. In conclusion, I shall suggest a few points to which it seems desirable that some attention should be directed. In this part however, more than in any other, the judgement of an individual must be considered fallible.

1. After all that has been done in respect to refraction, I suppose that there is no subject of such continual application, in the theory of which so many difficulties occur, and in whose application there are so many discrepancies. It seems not improbable that some of the latter may depend upon anomalies in the indications of the thermometer, owing perhaps to the effects of radiation, the nature of which till lately has been little understood; and scarcely recognised among astronomers. In some of Mr. Fallows's pendulum experiments it appeared to account for many

[page] 187

discordancies. 1 think that astronomers would be glad to receive simple directions for placing a thermometer so as to indicate correctly the temperature of the air at the place of observation. I omit all discussion of the difference of the external and internal thermometer, as I think the only way of overcoming the difficulty is, to make the external and the internal temperature as nearly as possible equal.

2. In the theory of refraction, the following questions present themselves as only to be solved by experiment. What is the law of the decrease of temperature, or rather of density, in ascending? How does this vary at different times? Can any means be contrived for indicating practically at different times the modulus of variation? (The last question is suggested by the remarks in Mr. Atkinson's valuable paper, Mem. Ast. Soc. vol. 2.) Does the refractive power of air depend simply on its density, without regard to its temperature? Is it well established that the effects of moisture are almost insensible? From Carlini's Tables, as well as from general reasoning, it seems likely that refraction may be different in different azimuths, according to the form of the ground: can any rough rule be given for estimating its effect? Finally, when the atmospheric dispersion is considerable, what part of the spectrum is it best that astronomers should agree to observe?

3. I have already stated that I think Lindenau's constant of nutation has been adopted by the German astronomers on insufficient grounds. The value which I should certainly prefer is that determined by Dr. Brinkley, and which Mr. Baily, with his usual judgement, adopted for the catalogue of the Astronomical Society. The Greenwich circles have now been erected, and in a perfect state, long enough (or nearly so,) to determine this constant; and the mass of excellent observations which they have produced, applicable to this question, vastly exceeds any other that has been used for the same purpose. It is highly desirable that the coefficient of nutation should be investigated from the Greenwich circle-observations.

4. Bradley's observations of stars were nearly useless till Bessel undertook to reduce them. In like manner Bradley's and Maskelyne's observations of the sun and planets are still nearly useless. At different times observations of the sun have been reduced (by Delambre, by Burckhardt, and lately by Bessel or Schumacher), and probably much labour has been wasted. A reduction of these observations on a uniform plan (adopting, for instance, Bessel's Tabulœ Regiomontanœ,) would be invaluable in the application of the planetary theories. Many observations of the moon have been reduced

[page] 188

and published; but, for the sake of uniformity of system, it would be desirable to re-compute them.

5. I have mentioned that the perturbations of the small planets and of Encke's comet give good reason to believe that the mass of Jupiter adopted by Laplace is too small. Laplace's estimation is founded on Pound's measures of the elongation of Jupiter's satellites; and I am not aware that any measures have been made since that time. It is extremely desirable that they should be measured, at least those of the fourth satellite. It would be sufficient, in observing the transit of Jupiter, to observe also the transit of this satellite, one or two days before and one or two days after the time of its greatest elongation, as the theory of the satellites could be applied without difficulty to this measure.

6. The dimensions of the orbit of Encke's comet, as investigated by Encke, depend upon the assumed law of density of the resisting medium. In fact, by assuming a law, he has established a relation between the diminution of the aphelion distance and the diminution of the perihelion distance, which would not hold with any other law. It will be interesting now or at some future time to investigate separately from observations the diminution of these two elements, as a means of proving Encke's law, or of suggesting a new one.

7. The perturbations of Biela's comet have not been calculated, I believe, for the interval between 1772 and 1806, nor those of the node and inclination from 1806 to 1826. It is desirable that this should be done, both for ascertaining the identity of the comet of 1772 (which is not perfectly established), and for examining whether this comet, like Encke's, gives any indication of a resisting medium.

8. The most laborious part of the expansions in physical Astronomy is completely dispensed with by the use of Burckhardt's formulae in the French Mémoires for 1808. But Burckhardt expressed himself very doubtful as to their accuracy; and they do not comprehend any terms depending on inclination. It is desirable that they should be verified, and extended to the terms depending on the inclination.

9. The theory of the perturbations of Pallas has so often and so vainly been proposed, that it would seem useless to urge it again, and still more so to propose the theory of the perturbations of Encke's comet. Yet I conceive this to be the problem which at present demands the efforts of physical Astronomy. It is plain that there is no hope of solving it by any of the usual methods, as the series, which in other cases are convergent, here either diverge or converge so slowly as to be

[page] 189

useless. But it may be possible to choose functions for these expansions, by the use of which the series may become convergent (such perhaps as sin θ/1 + b cos θ instead of sin θ). At all events this may be fixed on as being at present the difficult problem of Physical Astronomy.

In the preceding suggestions I have endeavoured to fix on definite points for the attention of astronomers. I need not mention that there are other subjects (the theory of Uranus, for instance,) in which the existence of difficulties is known, but in which we have no clue to their explanation.

G. B. AIRY.

Observatory, Cambridge,

May 2, 1832.

Report on the Tides. By J. W. LUBBOCK, V.P.§ Treas. R.S.

THE connexion between the Tides and the motion of the moon was known to the ancients; but we are indebted to Newton for the discovery of the mechanical principles which regulate these phænomena. Newton contented himself with explaining the most obvious results of observation, and left all the details open to future inquiries. The subject was next taken up by Bernoulli, Euler and Maclaurin, about the same time, in their several treatises which participated in the prize awarded by the Academy of Paris in 1740. Laplace afterwards undertook this difficult investigation, and succeeded in forming the differential equations from which the explanation of the phænomena is to be derived. The integration of these equations presents, however, so many difficulties, that he confined his attempts to a very simple case, namely, that in which the depth of the ocean is constant, and the solid nucleus but little different from a sphere. Even in this case, his analysis is far from complete, and contributes but little to unravel a question which he has characterized, as "la plus epineuse de l'Astronomie Physique."

Finally, Laplace had recourse to the following indirect consideration, namely, "that the state of any system in which the primitive conditions have disappeared through the resistances which its motion encounters, is periodical with the forces which act upon it." Hence he concludes, that if the system is dis-

[page] 190

turbed by a periodic force expressed by a series of cosines of variable angles, the height of the tide is represented by a similar series of which the arguments are the same, but the epochs and the coefficients different. The adoption, however, of the preceding principle must be considered rather as an evasion of the difficulties by an indirect method, than an accurate and complete solution of the problem.

Lately, the Academy of Sciences of Petersburg has proposed the problem of the Tides for a prize question. The programme may be seen in the Number of the Annales de Chimie for February of the present year (1832).

Since the publication of the researches of Laplace, the theory of the integration of partial differential equations has been very materially improved by Fourier, and by MM. Poisson and Cauchy. The small undulations of an incompressible fluid, acted upon by gravity, which were not previously understood, were completely made out by the latter mathematicians, about the same time, in 1815. This case however, in which the force acting upon the fluid is constant, and in parallel lines, is the simplest which can be proposed; while the problem of the Tides in which the motions of the fluid are due to the action of a force of which the intensity and direction are continually changing, presents more serious difficulties, which are further increased by the circumstance that the bed of the ocean is far too irregular to be represented even approximately by any algebraic curve surface, and by the effect of the resistance and friction of the water against the shores, which cannot be considered as insensible.

The attention of Laplace does not appear to have been directed to the construction of Tide Tables for predicting the time and height of high water at any port; and indeed up to the present time the Table for this purpose published in the Annuaire du Bureau des Longitudes, is deduced, by a very slight alteration of form, from that given by Bernoulli in his prize essay. Nor does the subject appear, until very lately, to have met with any attention in this country, no attempt having been made previously to ascertain how far the theories of Bernoulli or Laplace can be reconciled with the results of observation on our coasts.

Formerly, the time of high water at London Bridge was obtained by adding a constant quantity, three hours, to the time of the moon's southing. As the mean interval is now very little more than two hours, we may infer that the time of high water in our river has been considerably accelerated; and this circumstance shows the importance of continual observations, and the

[page] 191

necessity of renewing from time to time these determinations. This method of adding a constant quantity was somewhat improved by Mr. Phillips in 1668, who gave, in the Philosophical Transactions for that year, a Table showing the variations in the interval between the time of the moon's southing and the time of high water. Shortly afterwards, Flamsteed having frequent occasion to pass between London and Greenwich by water, and having caused above 80 high waters at Tower Wharf and Greenwich to be observed, found that the greatest and least differences betwixt the moon's true southing and the high waters were not, as Mr. Phillips had placed them, at the full or new and quarter moons, but the greatest nearer to the neaps, and the least to the highest spring tides. Previously the Tide Tables had only shown the time of that high water which next follows the moon's southing; Flamsteed introduced both in his Tide Table.

In the earliest years of the Royal Society, some attempts were made to set on foot observations of the Tides by some of the active members; but these phænomena have long ceased until lately to excite any attention whatever in this country. Since the establishment of the various Docks at London, the times and height of high water have indeed generally been registered there in books kept for the purpose; and to these, which scarcely deserve the name of observations, must we have recourse if we wish to determine the various constants of the expressions which furnish the means of calculating the time and height of high water, or of ascertaining the agreement between theory and the tides in our river. Nevertheless by taking the mean of an immense number of observations, the error is almost eliminated; and in the unfortunate alternative of being obliged to relinquish the question altogether, or of making use of these imperfect data, I have preferred the latter, and, with the assistance of Mr. Dessiou of the Admiralty, have discussed a great number of observations made at the London Docks. I have found that when the effects of changes in parallax and declination are neglected, the agreement between the results of observation and the theory of Bernoulli (which so far coincides with that of Laplace,) is very remarkable.

With respect to the effects due to changes in the moon's parallax and declination, I am not yet able to speak so positively, although it is certain that the height of high water is about a foot more, at the London Docks, when the moon is nearest to the earth than when she is furthest, of course cœteris paribus. When the moon is in the equator, the time of high water is retarded about half an hour from what obtains when she is in her greatest declination; the height is also about six inches less.

[page] 192

The height of high water appears to vary very little at different months of the year. The difference between spring and neap tides at the London Docks is about 3 feet, and the rise about 19 feet. Mr. Dessiou has undertaken the laborious task of discussing 6000 observations more, in addition to those of which the results are given in the Philosophical Transactions, in order to obtain with greater accuracy the corrections due to changes in the moon's parallax and declination.

As the times and heights of high water at London Bridge are in future to be inserted in the Nautical Almanac, and as Tables for the prediction of these phænomena are to be given in the new edition of the Requisite Tables, we may hope shortly to see this question rescued from the neglect in which it has lain so long and so undeservedly. With this view it is of great importance that observations should be originated in various parts of the world, and with greater attention to accuracy than any which are yet carried on, except perhaps those at Brest, which were instituted by the French Government in 1806, at the request of Laplace, and have been continued ever since. In August last year, at the request of the Council of the Royal Society, directions were sent from the Admiralty to the masters attendant at Woolwich, Sheerness, Portsmouth and Plymouth, to cause observations of the Tides to be made, and to forward reports quarterly. This order has been complied with, and the observations are in the possession of the Royal Society. I have not been able to ascertain that any observations are made on the coast of Scotland and Ireland; and in this country, with the exceptions I have noted, and at Liverpool, these interesting phænomena pass away unheeded and unrecorded. I trust that the influence of the British Association will be exerted to remove in some degree this national reproach.

Observations of the Tides should record particularly,

The time and height of high water.

The time and height of low water.

The direction of the wind and the height of the barometer and thermometer should be noted, and the direction and velocity of the current should also be described.

The circumstances of high water are more interesting, and admit generally of more accurate observation, than those of low water.

The height of the water must be given from some fixed mark or line*, which should be described accurately, so that it

* I consider this of particular importance, and I allude to it because it has not been complied with in some observations transmitted to the Royal Society. Observations of the rise are useless.

[page] 193

may be easily recovered. It should also be carefully stated whether the time in which the observations are given is mean or apparent, and how obtained.

The name of the observer, or his initials, should be attached to each observation. The simplest method of observation appears to be by means of a staff, carefully graduated, connected with a float, and working through a collar where the height is read off. The staff must be kept in a vertical position by means of friction rollers; the float should be in a chamber to which the water has access by a small opening, in order that the ripple may be as much diminished as possible. It would be convenient to have a clock close to the tide gauge; and if made to strike minutes, so much the better. The observer should note the height of the water at the end of every minute, for half an hour before the expected time of high water, and until there can be no doubt that the time of high water is past. The minute at which the water stood the highest, or the time of high water, is then easily seen. This process is tedious, and it might be imagined that it would suffice to note the time when the water reaches a certain height shortly before high water, and the time when it reaches the same line in its descent; but the water rises and falls by jerks, and much too irregularly for this plan to be adopted with safety, at least in our river.

Mr. Palmer has described, in the Philosophical Transactions, a self-registering machine which is intended to give the time and height of high water; and I believe it is intended to set one up at the London Docks, but I have not heard that it is yet in operation. The principle consists in a style, or pencil, which is moved horizontally by the tide along the summit of a cylinder, which is turned round slowly and uniformly; the pencil describes a curve upon paper wound round the cylinder, which curve indicates the fluctuations of the water. The motion of the tide being originally vertical, is changed by a common mechanical contrivance of the simplest kind.

When it is intended to make a long series of observations, it is of course very desirable to adopt every precaution to ensure accuracy; but many persons have it in their power to make observations, which may be useful in determining the establishment of a port, or the mean interval between the moon's southing and the time of high water, without any expensive apparatus.

For this purpose the observations during one lunation, or even less, may suffice, where, as in the river Thames, the rise is considerable and the tides little subject to irregularities. In the open ocean, where the rise on the contrary is small, the tide often hangs half an hour at high water, and the phænomena

N

[page] 194

take place very irregularly. At St. Helena the rise in springs, according to Dr. Maskelyne, is 39 inches, and in neaps 20 inches; and I apprehend that less information could be elicited from a year's observations there, than from a month's observations at the London Docks. When a few observations only are made with a view of determining the establishment, they should not be used to determine that quantity absolutely, but they should be compared with observations at some place of which the establishment is accurately known, or where observations are continually carried on. It would be very desirable for those who are able, to combine so as to effect the monography or detailed description of the tides through some short extent of coast, such as that which has been effected by M. Daussy for the coast of France.

M. Daussy has determined with great care, by means of observations executed by the "Reconnaissance hydrographique des côtes de France," undertaken by the body of ingénieurs hydrographes under M. Beautems Beaupré, the establishment of all the principal places on the coast of France between Ouessant and the coast of Spain. M. Daussy finds that the influence of the wind upon the height of high water is insensible. I have found that the direction of the wind (unless in violent gales,) has no effect upon the phænomena of the tides in the river Thames; but this I attributed to its comparatively sheltered situation; and I should have thought that the tides in the Bay of Biscay would be much affected by gales sweeping over the surface of the Atlantic. M. Daussy has shown beyond doubt that such is not the case, and that the irregularities of the tides there must be due to more remote causes. He has also shown that the atmospheric pressure has considerable influence upon the height of the tide: an inch of rise in the mercurial column depresses the tide fourteen inches. This fact is very remarkable. I have ascertained, however, that in the river Thames the influence of the fluctuations of the barometer upon the tide is insensible, or very nearly so. Beyond the coasts of France, our knowledge of the progress of the tide-wave is very imperfect; and it is difficult at present to trace satisfactorily the course of high water throughout the globe, owing to the paucity of even bad observations. It is generally high water at any given instant at a series of points which form the crest of the tide-wave, and which I have called, at the suggestion of Mr. Whewell, cotidal lines. If the ocean were not intersected by continents, the tide-wave would proceed from east to west; and if the luminaries moved in the equator, the cotidal lines would be meridians.

[page] 195

The continents of Africa and South America may be considered as immense dams in the course of the tide-wave which completely change its direction, so that it is high water at the same time on the opposite shores of the Atlantic. The rudiments of the cotidal lines which would obtain in the case of a perfect spheroid probably exist round the south pole, interfered with, as they must be, by the great continent of ice in those regions. Owing to the obstructions I have mentioned, it is high water nearly at the same instant at the Cape of Good Hope, off the Straits of Gibraltar, off the coast of Scotland near the Murray Frith, and in the river Thames. The wave takes six hours in proceeding form the Land's End to the North Foreland, being at the rate of about 70 miles an hour, and in a direction contrary to the course of the luminary. If the ocean completely covered the solid nucleus of the earth, it would only be high water at the same instant at places of which the longitude differed by 180°; and at the equator the tide-wave would travel at the rate of about 500 miles per hour. The motion of the crest of the tide-wave must be carefully distinguished from that of the particles of water themselves, which forms a current the velocity of which seldom exceeds a few miles per hour: these currents are modified by others due to changes of temperature. The analytical investigation of the motions produced by changes of temperature, and of the propagation of heat in fluids, is one of extreme difficulty, and has not been yet attempted. In order to approach this important question with any chance of success, it seems necessary to consider the problem in the first instance in its most simple form, and one in which the results of theory can easily be compared with those of observation.

Works of navigation and sailing directions supply much information with respect to the velocity and direction of the currents; while the time of high water appears to have been carefully ascertained at very few points only on the earth's surface. Yet the phænomena of the tides are of extreme interest. Laplace says, "Les marées ne sont pas moins intéressantes à connôitre, que les inégalités des mouvemens celestes. On a negligé' pendant longtemps de les suivre avec une exactitude convenable, à cause des irrégularités qu'elles présentent; mais ces irrégularités disparaissent en multipliant les observations." There is indeed no branch of Physical Astronomy in which so much remains to be accomplished.

[page] 196

Report upon the Recent Progress and Present State of Meteorology. By JAMES D. FORBES, Esq. F. R. S. L. & E. F. G. S. Member of the Royal Geographical Society,, of the Society of Arts for Scotland, and Honorary Member of the York-shire Philosophical Society.

I FEEL that, in undertaking a Report upon the recent progress and present state of Meteorology, I have engaged in a task of greater difficulty than most persons are probably aware of; greater, too, than attaches to sciences of which the fabric is more deeply founded and massive, but at the same time more connected.

In the science of Astronomy, for example, as in that of Optics, the great general truths which emerge in the progress of discovery, though depending for their establishment upon a multitude of independent facts and observations, possess sufficient unity to connect in the mind the bearing of the whole; and the more perfectly understood connexion of parts invites to further generalization.

Very different is the position of an infant science like Meteorology. The unity of the whole, or of the individual greater divisions of which it is composed, is not always kept in view, even as far as our present very limited general conceptions will admit of: and as few persons have devoted their whole attention to this science alone, or the whole exertions which they did bestow, to one branch of so wide a field,—no wonder that we find strewed over its irregular and far-spread surface, patches of cultivation upon spots chosen without discrimination and treated on no common principle, which defy the improver to inclose, and the surveyor to estimate and connect them. Meteorological instruments have been for the most part treated like toys, and much time and labour have been lost in making and recording observations utterly useless for any scientific purpose. Even of the numerous registers of a rather superior class, which monthly, quarterly, and annually are thrown upon the world, how few can be expected to afford, or are even intended to afford, specific information upon any one leading doctrine or fact of the science! These hardly contain one jot of information ready for incorporation in a Report on the progress of Meteorology: such of them as are fitted for undergoing an analysis must previously have furnished the raw material, as it were, for the construction of some arbitrary general laws to connect phænomena, when duly combined with

[page] 197

similar elements already in store. The amount of detached facts is absolutely appalling; and the consequence is, that not only in registers of individual observations, but in those where the results are presented in a condensed shape and the arithmetical means have actually been taken, but a few points have been applied in practice to the elucidation of any one theoretical difficulty in the science.

The most general mistake probably consists in the idea that Meteorology, as a science, has no other object but an experimental acquaintance with the condition of those variable elements which from day to day constitute the general and vague result of the state of weather at any given spot; not considering that while such heterogeneous elements can be of little avail, when viewed simply as a group of facts, towards forwarding any one end of the science, or giving us any precise knowledge regarding it, yet that the careful study of the individual points, when grouped together with others of the same character, may afford the most valuable aid to scientific generalization. If instead of aiming at a rude approximation to the mean numerical elements furnished by meteorological instruments for a particular spot, some individual branch were selected and pursued under the most favourable circumstances, a result would be obtained at no more expense of labour,—an insulated one it is true, but capable, by combination with others, of making a real addition to the deductions of the science. As this appears to me the place to insist upon a total revision of the principles upon which meteorologists have hitherto very generally proceeded, I shall explain my views a little more particularly.

It is in the first place worthy of remark, that the most interesting views which have been given in this science, and the most important general laws at which it has yet arrived, have for the most part been contributed by philosophers who, in pursuit of other objects, have stepped aside for a moment from their systematic studies, and bestowed upon the science of Meteorology some permanent mark of their casual notice of a subject which they never intended to prosecute, and which they soon deserted for other and more favoured paths of inquiry. Mr. Dalton descends for a moment from his chemistry in the abstract, to illustrate the constitution of the atmosphere and the theory of vapour. Laplace, viewing nature with the eye of a master, introduces into his Mécanique Céleste an investigation of the mechanical structure and laws of equilibrium of the gaseous envelope of our planet: he applies Meteorology to one of its great objects,— the laws of atmospherical refraction; and gives to the scientific world a new formula for the measurement of heights by the

[page] 198

barometer, which greatly exceeds in accuracy those which had previously been proposed. Yet may the speculations of these philosophers, and the discussions to which they give rise, be more important to the science than the labours of a professed meteorologist, who has made, with minute scrupulosity, all the ordinary entries in his Journal, daily for a life-time.

We must not be supposed to give to theory a pre-eminence over observation. Had the meteorologist just supposed, instead of observing all the ordinary intruments, perhaps not upon the best construction, and at hours dictated by convenience or by accident, directed his attention even to the merely mechanical examination of any one phænomenon, not to the mixed result of a chaos of heterogeneous principles;—had he, with an eminent chemist of our own time, determined the specific gravity of air every day, and watched the unsuspected variations which that amount undergoes;—had he directed his observations to the detection of the lunar atmospheric tides;—had he examined by reiterated experiments, under every varied condition, the solution of the beautiful problem of the barometrical measurement of heights;—had he taken advantage of lofty and mountainous situations to study the formation and dissolution of clouds and the influence of humidity and temperature in their phases, or of a low and flat country for determining the amount of solar and terrestrial radiation in sheltered spots and under different aspects of the heavens;—had he in any of these, or in one of a hundred other equally fertile paths of inquiry, added to our knowledge of the connexion of cause and effect in this intricate subject, he would have conferred, at perhaps even less expense of time and labour, an infinitely greater boon upon the science which he wished to advance.

The mere local meteorology of a country may frequently be a very interesting object in relation to its physical geography and agriculture, and as such may be prosecuted by the systematic establishment of Registers on a small scale; but for the great facts of the science the adequate support of a few great Registers in any country would suffice, provided such be sustained on the most liberal scale and on the most accurate principles, by great Societies, or, still better, by Governments. Instruments must be provided on the best possible construction, placed in the best situations, and observed at the best times, and with undeviating regularity, by fit observers. The critical hours vary in different climates, and should be determined with the greatest care by preliminary experiments; and no greater error has been committed in the establishment of such Registers than the indiscriminate observation of all instruments which

[page] 199

have no bond of connexion in their diurnal changes, at the same hours.

Great numerical accuracy is always of extremely difficult attainment; and it is hoped that the good sense of observers will dismiss from Meteorology, as well as from some other branches of physical science in which it has prevailed*, that superfluity of decimal places, which when they exceed to a great extent the compass of the instrument to verify, create rather a distrust in the observer than confidence in his observations. Even within very moderate limits it is clear that, where accuracy so entirely depends upon the extreme precision of instruments and attention to their condition, and upon perfect regularity and consistency of observation, there are few individuals who can furnish the numerical data now required for the advancement of the science. Five or six Registers in Great Britain and Ireland, carried on by learned Societies or by Government, would afford the great normal quantities required for establishing the numerical data of the climate of this kingdom with regard to the great elements of temperature, pressure, and humidity in relation to that of other parts of the globe. And while we would by all means encourage the continuance and the extension of local Registers aiming at no very high degree of precision, in illustration of the particular climate of different parts of the island, these Registers would be of a very simple description, and might be confided to the hands of merely mechanical observers, under the occasional superintendance of persons of greater acquirements.

Let us conceive for a moment the gain to science, which such a saving as would thus be effected in cude and unprofitable Registers would produce. The whole class of those who profess to study Meteorology, either as an occasional pursuit or as a more constant occupation, would be left almost free to pursue individual objects of inquiry which, though not so simple as the vague mechanical task to which at present they generally devote their time, might in many cases be rendered nearly as much so, and might add every year a stock of information, which, instead of being looked upon with the coolness and indifference with which an ordinary Register is generally glanced over, might be hailed by fellow-labourers in the same field as throwing new light upon their several branches of inquiry.

We have already adverted to some subjects which afford ample scope for judicious and well-directed experiment: it would be needless formally to enlarge such a list of desiderata; but in the course of this Report we shall have an opportunity of point-

* For example, in Chemistry.

[page] 200

ing out some of the more important, and of the methods which should be taken for supplying them. It is much to be desired that other bodies as well as the British Association should use their influence to direct individual effort into the channel most likely to contribute to the advancement of science. In a period like the present, when the stream of knowledge seems to diminish in depth as it increases in diffusion, it is above all necessary for influential bodies to retard the progress of that tendency to merely superficial study, which has injured nothing more than the science of Meteorology,—one which, though in many respects apparently simple and abounding in palpable results, really consists, in its very nature, of a most elaborate piece of mechanism, delicate in its parts, and of which the connexion is anything but obvious.

The true basis of the science rests upon several branches of physics, which are only at the present moment rising to their true level of importance in the scale of human knowledge; and there are few of the sciences which are not more or less directly connected with the progress of Meteorology. Astronomy bears not only the great relation of taking cognizance of the causes of change of season, but it gives the data for estimating the influence of the heavenly bodies in raising tides in our atmosphere, and indicates the causes of alteration of climate which some of their longer periodic motions present*. Geology teaches us the probable state of cooling from an intensely high temperature in which our globe now exists, which most likely exerts a material though till lately unsuspected influence upon climatology. Chemistry analyses the composition of that gaseous atmosphere, the modifications of which it is the principal object of Meteorology to investigate. Pneumatics furnishes us with the grand laws which connect the pressure and density of the air with height, which gives a key to many of the variations indicated by the barometer, and by means of that instrument enables us to attain an accurate comparison of different elevations in the gaseous medium. To the science of Electricity we must look not merely for the explanation of those phænomena which more obviously indicate its presence and action, but likewise of many which at present are almost veiled in obscurity, or can be but partially explained by other agencies.

But most of all is the science of Heat the very basis of all accurate knowledge in Meteorology. No one department is exempt from its influence; no one substance in nature seems independent of the action of this subtile element. Impalpable though it be, yet since we possess such accurate means of in-

* See Sir John Herschel on Astronomical Causes affecting Geology, Geol. Trans. N.S. iii.

[page] 201

vestigating many of its laws, it is surprising how very imperfect are the notions entertained by mankind at large, and even by the scientific world, as to the importance of the part which it assumes in the œconomy of nature. To attempt to study Meteorology without it, is like trying to read a cipher without previously mastering the key. The laws, so far as they are known to us, by which it is regulated, though generally simple in their enunciation, rise, when pursued into their consequences, to highly complicated deductions, and soon (as is the case with every science rising above the limits of first generalisations of facts, and empirical laws,) require all the resources of mathematical analysis to eliminate general laws and to re-descend to the prediction of phænomena*. The propagation of heat in solid bodies, which forms the first problem in the theory of heat to be solved, and one of the greatest importance in the consideration of the globe as a heated mass in the course of cooling, has occupied the attention of some of the first philosophers of France. At an early period in this century M. Biot pointed out the expression for the condition of a solid bar with regard to temperature, receiving a constant supply of heat at one end, and parting with it towards the other by conduction and radiation, which gave rise to a partial differential equation, which has since undergone repeated discussion†. Laplace took up the question, and removed some analytical difficulties in which it was involved. He was succeeded by Fourier and Poisson, who gave greater generality to the solution, and extended it to bodies of various figures‡. Fourier, in his great work the Théorie Analytique de la Chaleur, has extended his profound inquiries to a vast number of problems in the propagation of heat, most important to our present subject, and which, in special relation to the temperature of the mass of the earth, will shortly be noticed more particularly.

A variety of points connected with the relation of many substances to heat have of late years been determined, though there is yet much to be done in this important field. The constants which regulate the passage of heat through various bodies, and which have been termed by Fourier "external conducibility," or penetrability, and "internal conducibility," or permeability§, have been determined for several bodies, but a

* "L'étude approfondie de la Nature est la source la plus féconde des découvertes mathématiques."—Fourier, Théorie de la Chaleur, Disc. Prel. xiii. This beautiful discourse gives some fine views on the application of mathematical reasoning to physical questions.

Traité de Physique, iv. 669. Fourier, Théorie, chap. i. sect. v.

Mémoires de l'Institut: Journal de l'Ecole Polytechnique: Connaissance des Tems, &c.

§ Théorie Analytique, Art. 30, 37.

[page] 202

much smaller number than would be desirable*. Fourier's "thermometer of contact†," intended for examining the constant of permeability, has not come into general use, as it probably will do when the calculations required to be made from its results are adapted to variable thicknesses of the bodies under experiment.

The specific heats of substances have also formed a subject of nice and successful investigation. MM. Dulong and Petit have determined those of a great number of solid bodies, and rendered it highly probable that the specific heats of their ultimate atoms are the same. De la Roche and Berard, and De la Rive and Marcet have signalized themselves in the more arduous investigation of the specific heats of the gases. From its direct application to the condition of our atmosphere, and to the probable cause of cold as we ascend through its strata, this subject and the whole relation of the gases to heat has offered a most important and interesting field for investigation during the present century. Laplace discussed it in the tenth book of the Mécanique Céleste; the experimental part was taken up by Gay-Lussac and Welter, by Clement and Desormes, by De la Roche and Berard, by De la Rive and Marcet, by Mr. Haycraft, and finally by M. Dulong. Though this question as relating to different gases cannot be considered settled, the most probable result is that obtained by De la Rive and Marcet, and by Mr. Haycraft,—that equal volumes of the different gases have the same specific heat. The consequences to which the variable specific heat of the gases under different pressures give rise, and especially the extrication of heat which attends their compression, have been studied and analysed by Ivory‡, Poisson§, Leslie‖, Avogadro¶, and others.

Our information upon the expansion of solids has not of late years much increased. Several fluids however have been reexamined, and the anomalous expansion of water and its point of greatest density have been elaborately investigated by Hallström, Müncke, and Stampfer. M. Erman has added to our knowledge of the anomalous effects of heat upon several other substances, and upon various phænomena of liquefaction**.

* See Experiments of Despretz in the Annales de Chimie et de Physique, tom. xix. His result for platinum is generally considered erroneous; and some experiments which I have recently made will, I believe, demonstrate it to be so.

Annales de Chimie, tom. xxxvii.

Philosophical Magazine, 3rd series, vol. i.

§ Annales de Chimie, tom. xxiii.

Encyclopœdia Britannica, Sup., art. CLIMATE.

Mémoires de l'Académie Royale de Turin, tom. xxxiii.

** Annales de Chimie, xxxviii. xl.

[page] 203

One of the most universally admired investigations of a physical law by which science has been recently benefited, is that of MM. Dulong and Petit into the law of cooling, published in the Journal de l'Ecole Polytechnique, and in the Annales de Chimie; one to which, from its universally acknowledged beauty and importance, we need do no more than allude. The radiation of heat, which has been so powerfully illustrated, and whose general laws are so well determined by these experiments, forms one of the most important elements of the science of Meteorology. Baron Fourier has recently deduced from theory the law of radiation experimentally proved by Professor Leslie,— that the calorific rays decrease proportionably to the sines of the angles they make with the radiating surface; and he has drawn some interesting conclusions*. The same author, considering our globe as a radiating body placed in indefinite space, and as having reached a condition of temperature sensibly invariable, has deduced the temperature of planetary space to be —50° cent.† Swanberg, arguing from the observed decrement of heat in the atmosphere, has arrived at almost the same result‡.

Considering heat as the power by which liquids are converted into vapour, the science of Hygrometry has received of late years important additions, not merely from several researches upon the theory of vapour, but from the elaborate experiments, undertaken with praiseworthy zeal under the superintendance of the French Academy of Sciences, upon the force of vapour at different temperatures§. Mr. Faraday, on the other hand, has pointed out the existence of a limit to vaporization‖.

I have thought it necessary briefly to point out the prodigious obligations under which Meteorology lies to the science of Heat, because the truly philosophical procedure of arriving at the great truths of the former seems to be too much overlooked. The results just enumerated have every one been attained by constant and assiduous labour, some by a course of most arduous experiment, others by the application of the most refined mathematical analysis. Till we have the laws of heat more completely unravelled than at present,—till the most important yet profoundly difficult problem of its relation to a gaseous atmosphere of varying density shall have been adequately solved,— Meteorology will stand upon an uncertain basis, and will abound

* Mémoires de l'Institut, tom. v.

Annales de Chimie, xxvii.

Bibliotheque Universelle, xliii. 367; and Edinburgh Journal of Science, N.S. iii. 13.

§ Annales de Chimie, xliii. 74. The Commission was composed of MM. De Prony, Arago, Ampère, Girard, and Dulong.

Philosophical Transactions; and Journal of the Royal Institution.

[page] 204

in empirical laws and inconsequential reasoning. Let therefore those whose time is too much wasted in a vague study of the chaos of conflicting phænomena which is presented to us, fit themselves, by suitable physical and mathematical inquiries, for grappling with the difficulties individually. Never, we may be assured, will Meteorology attain the true dignity of a science till that of Heat is fully mastered,—till the laws which regulate its distribution generally are recognised, and its peculiar relations to the materials of our globe and the component parts of its atmosphere are ascertained,—till, in short, the motto of Fourier's great work is fulfilled, Et ignem regunt numeri.

It may be proper now to mention the particular course which is to be adopted, in the remainder of this Report, in endeavouring to give a general view of what has actually been done in Meteorology for some time past, and what points most require elucidation.

I shall not be particular in inquiring what are the precise limits to be assigned to the science of Meteorology, nor proceed to discuss the subject with the formality of arrangement which would be required in a treatise. I shall chiefly confine myself to those branches which admit of systematical cultivation, and which have assumed some consolidation of parts, without which any attempt at general views would be premature. On this account, I shall very slightly allude to what are commonly called atmospheric phænomena, unless where their circumstances have been sufficiently classified to admit of being treated, in a general view, as groups of facts connected by some law, whether deduced by reasoning, or empirical.

After alluding to such systematic works as have appeared of late years upon the science, I shall briefly notice any general views which have been presented as to the constitution of the atmosphere: I shall then successively consider the three great elements of Temperature, Pressure, and Humidity; and finally, under the head of Atmospherical Phænomena and Precipitations, notice such points as may especially claim attention, upon Electricity, Auroræ Boreales, Winds, Rain, &c.

With an earnest desire to render my exertions as useful as possible, I conceived that in giving an idea of what has recently been done in Meteorology, I should very inadequately fulfil the object by analysing merely the few works which may be published separately, or the longer papers scantily scattered through volumes of Transactions: I propose therefore to myself, after mentioning simply, in each department, the great steps by which it has been brought to its present condition, to refer to the papers of any interest which touch upon the subject in question in the periodical literature of the last five years, that is, from

[page] 205

1827 to 1831 inclusive. This will preclude my attempting to analyse these papers, unless where they are most important and comprehensive: but I conceive that in the present state of the subject it will be more important to bring together the results, in the simplest connected form, of the mass of floating periodical information, and give means of reference to enable any one to inquire for himself into the steps by which individual authors arrived at the conclusions which may be announced. When opportunity offers, I shall add to this, hints for the prosecution of the subject in the best lines of direction.

I am sensible that in sciences further advanced, a different style of Report would be more useful and more agreeable. In Meteorology however, where the literature is almost solely fragmentary, I believe that by the mode which I adopt I shall produce a more satisfactory work, though at the expense of greater labour of compilation.

The only meteorological work which I know of as having appeared in this country during the last few years, is Mr. Daniell's volume of Meteorological Essays. This book has been too long in the hands of the public to require any extended criticism or analysis. Mr. Daniell has very justly considered that, in the present state of the science, detached Essays were better suited to its imperfections than a more systematic plan of composition: this likewise affords us the opportunity of taking up the particular subjects of which he has treated, when they contain any important novelty, in their regular place in our Report. It is perhaps to be regretted that Mr. Daniell has mixed so much purely theoretical matter with the interesting practical conclusions in which his book abounds: it has certainly had the distinguished merit of directing public attention strongly to the science, and of eliciting some further very interesting experimental inquiries upon topics touched upon in these Essays.

In France several systematic works have appeared in Meteorology, but chiefly of a character purely popular, and with no great pretensions to scientific interest. M. Pouillet's work*

* Elemens de Physique et de Méteorologie, 8vo, 2 tom. 1827-30. Baron Humboldt has recently presented to the Academy of Sciences two new systematic works, of which the authors are M. Schubler and M. Kämtz, both of whom are already known by their contributions to Meteorology. Bulletin de la Société Philomathique, Mars, Avr. 1832.—N.B. December 1832. Since writing the above, though I have not seen the work of M. Kämtz, I have received such information as leads me to believe that it will take a distinguished position among systematic treatises. The first volume only is published. The second will contain many original researches of the intelligent author, with whom I have been fortunate enough to become acquainted.

[page] 206

is the only exception which occurs to me. In the last volume he has given a valuable though somewhat brief and incomplete view of the present state of Meteorology, and particularly of the electricity of the atmosphere,—a subject to which he has carefully attended, and upon which he has published some valuable papers. We shall occasionally avail ourselves of the information contained in this work, as well as of a useful compendium of facts contained in the article METEOROLOGY in the Encyclopedia Metropolitana now in the course of publication.

Constitution of the Atmosphere.

The opinion, formerly general, that the atmosphere is a chemically combined compound gaseous fluid, consisting of nitrogen, oxygen, carbonic acid, and aqueous vapour, has gradually given way to the views entertained by Mr. Dalton, that these ingredients exist merely in mechanical union, and each in precisely the same condition as if it formed a simple atmosphere without foreign admixture. The important consequences to which this theory leads, have been developed by Mr. Dalton in an interesting paper in the Philosophical Transactions for 1826, part ii. p. 174, of the conclusions of which we shall now give a sketch nearly in the words of the distinguished author.—He conceives a mixed atmosphere composed of a heavy gas such as carbonic acid, and a light one such as hydrogen; and after showing the consequences, generally, which would result from the intermixture of equally elevated sections of two independent atmospheres of these gases placed side by side, each exerting a pressure of thirty inches of mercury, he shows that the two gases would be mixed in equal volumes at the earth's surface; that the carbonic acid would diminish rapidly in density, in ascending, and terminate at twenty-eight or thirty miles of elevation; whilst the hydrogen, diminishing slowly in density, would attain the superior elevation of eleven or twelve hundred miles.

Mr. Dalton considers these views established by three experimentally determined facts. 1st, That two gases combined in whatever proportions in a close bottle, are equally diffused through one another. 2nd, That if different gases be placed together in a bottle with water, and shaken, no pressure of one gas upon its surface can confine another gas in the water, each acting as a simple and independent atmosphere. 3rd, That the quantity and force of vapour of any kind will be the same whether there be any air present or none, being entirely regulated by temperature.

"From these three facts," adds Mr. Dalton, "but more especially by the two last, it appears to me as completely demonstrated as any physical principle, that whenever two or

[page] 207

more such gases or vapours as we have been describing are put together, either into a limited or an unlimited space, they will finally be arranged, each as if it occupied the whole space and the others were not present; the nature of the fluids, and gravitation, being the only efficacious agents*." Upon this principle, then, Mr. Dalton conceives that the total weight of the gases existing in the atmosphere which they compose, is proportional to the volumes existing at the surface of the earth. Thus taking the pressure of the nitrogen and oxygen together at thirty inches, he conceives that the particular pressure exerted by each is as 79 to 21, being the ratio of their volumes; consequently 23·7 inches of pressure result from the atmosphere of the former, and 6·3 inches from that of the latter. The weight of the aqueous atmosphere is variable, and may be assumed at 0·4 inch, and that of carbonic acid at ·03 inch. Mr. Dalton computes the height of the respective atmospheres to be fifty-four miles for nitrogen, thirty-eight for oxygen, carbonic acid ten miles, and aqueous vapour fifty miles. He justly observes, that the condition of the earth's atmosphere may be much modified by the disturbance to which it is subjected, and suggests the inquiry as an experimental rather than a purely theoretical one. It is to be hoped that the experiments which Mr. Dalton has promised to publish on the subject, may soon be given to the world.

Mr. Daniell's Essay on the Constitution of the Atmosphere† will require little notice here, both because it has been a considerable time before the public, and because, its object being an extension in some detail of Mr. Dalton's original views, it does not readily admit of abridgement. Mr. Daniell has successively considered the habitudes of a gaseous atmosphere, one of aqueous vapour, and a mixed atmosphere such as the globe actually possesses. He has illustrated at great length what he conceives to be the particular course of phænomena, chiefly by means of Tables, which he has carried to a considerable extent. These tables are intended to give a general idea of the influence of temperature, the rotation of the globe, and other circumstances in producing currents, of which Mr. Daniell has endeavoured to establish the velocity and other characters, and has applied them to the explanation of various meteorological phænomena.

M. Theodore de Saussure has published an extended memoir J upon the variations of the quantity of carbonic acid in the atmosphere, upon which he has made an elaborate series of

* Phil. Trans. ut supra, p. 184.

Meteorological Essays, p. 1—137.

Annales de Chimie, xliv. 1—55.

[page] 208

observations. He has determined that the upper strata of the atmosphere contain more carbonic acid than the lower ones; that the quantity undergoes a sensible diurnal variation, being greater during the day than during the night; and that the quantity generally is greater in dry weather than in damp, when it is absorbed by the moisture of the soil. The proximity of the ground probably accounts in that manner for the fact of its being less in quantity in the lowest strata of the atmosphere, which otherwise would be hostile to Mr. Dalton's views.

Temperature.

The thermometer is certainly the most perfect of our meteorological instruments. The range of natural temperatures being confined on the surface of our globe within comparatively narrow limits, namely, 96° centigrade or 172° Fahrenheit in the shade*; the indications of the mercurial thermometer may be considered as absolutely accurate. Notwithstanding, too, the difficulties of procuring tubes of perfect calibre or making due allowance for its variation, the deviations of thermometers made by different makers may, when particular care is taken in their construction, be confined within very narrow limits. This I have recently had the means of particularly observing in some comparisons of standard thermometers belonging to the Royal Society of Edinburgh, by Professor Christison and myself†. The relation of the mercurial to the air thermometer has been investigated since Gay-Lussac's and Dalton's experiments, by MM. Dulong and Petit, who extended the examination to a great range of temperature. More recently M. Auguste has taken up the subject, and given a formula of comparison‡. M. Parrot has re-investigated the subject of the fixed points of thermometers§. He finds that the purity of the water employed has a sensible influence on its point of congelation; and has observed one tenth of a degree of Reaumur between that of the water of the river Neva and distilled water. He has likewise determined that the maximum heat of water in a state of ebullition occurs at and below seventeen lines under the surface.

Nothing has lately been done in the way of materially improving self-registering thermometers: Rutherford's are still the best. Magnus has proposed one acting by the expulsion of

* Arago, Annuaire du Bureau des Longitudes, 1825, p. 186.

† Captain Sabine (Account of Experiments with the Pendulum, &c. 4to,) found the difference of above a degree in two standard thermometers by the same maker. Such an error however cannot be considered unavoidable.

‡ Poggendorff's Annalen, 1828.

§ This he has published in a Latin pamphlet, 4to: Petropoli, 1828.

[page] 209

mercury from the end of the tube*,—a proposal made, in the middle of the last century, by Lord Charles Cavendish, and since that revived by Mr. Blackadder†. I have elsewhere taken occasion to point out the principal defects to which it is liable‡. In connexion with the thermometer we must mention an elegant method devised by M. Bessel of finding the correction to be applied to any thermometer on account of inequalities in the bore; which consists in breaking off columns of various lengths of mercury in the tube, causing them successively to traverse the length of the scale; and by noting the spaces occupied by each at the different points, an equation for the scale of the particular instrument may be formed§.

A very interesting discovery has recently been made with regard to the early history of the thermometer, by Signor Libri of Florence‖. In 1829 were found at Florence a large number of the original alcoholic thermometers made under the direction of the Academia del Cimento, which enabled Sig. Libri to restore the true scale of these early instruments so as to afford a direct comparison with those of modern times. The scale was divided into 50 degrees. The zero corresponded to —15° of Reaumur, the 50th degree with 44° R; and it stood at 131/2° in melting ice. The latter fact is interesting because it shows that no sensible change had taken place in the freezing point of those instruments during the lapse of nearly two centuries; for it is recorded that in melting ice the liquid marked 131/2°. It is well known that, from whatever cause, many old thermometers indicate a temperature somewhat above 32° Fahr. when plunged in melting ice. In some cases however the change has not been noticed, and this is one of the most remarkable examples. Some years ago the question excited considerable discussion; but as nothing has been added to our knowledge for some time, I shall merely refer in a note to the papers which have treated of it¶. By an accident almost as fortunate as the recovery of the thermometers, some registers nearly complete for sixteen years and kept by Raineri, a pupil of Galileo, have come to light; and by the discussion of them, with a knowledge of the true scale, Signor Libri has been

* Poggendorff's Annalen, xxii. 146.

Edinburgh Transactions, vol. x.

Edinburgh Journal of Science, ix. 300.

§ Poggendorff's Annalen, 1826. Also Bulletin des Sciences Mathematiques, viii. 42. and Philosophical Magazine, vol. lxiii.

Annates de Chimie, xlv. 354.

¶ Flaugergues, Bibliotheque Universelle, xx. 117, xxi. 252. De la Rive and Marcet, Ibid. Avr. 1823. Bellani, Giornale di Fisica, v. Arago, Annales de Chimie, xxxii. Moll, Edinburgh Philosophical Journal, ix. 196.

O

[page] 210

able to show that no sensible change has taken place upon the climate of Florence between that period and the present, which had been suspected.

The metallic thermometers of M. Breguet have not been so much used as might have been expected; they are however rather adapted to delicate experiments on heat than to Meteorology in general. The same remark applies to the beautiful thermo-multiplier recently invented by Signor Nobili*, and which he has applied to the investigation of some of the most delicate phænomena of radiant heat†.

In the use of the thermometer, the sources of error from terrestrial and solar radiation have not been sufficiently attended to. Some hints on the subject may be found in the Article THERMOMETER in the Edinburgh Encyclopœdia. The mode of exposure of the thermometer of the Observatory at Paris is described by Pouillet in his Elemens de Physique.

These various sources of error go far to diminish our confidence in the nice accuracy of thermometric results, where they have not been the subject of particular attention. It is surprising at what a distance a sensible portion of heat is conveyed from soil and walls, or even from grass, illuminated by the sun; the maxima of temperature are thus generally too great, and from the near contact in which thermometers are generally placed with large difficultly conducting masses, such as walls, the temperature during the night is kept up, and the minima are thus also too high.

This however has been by no means the greatest difficulty in determining the mean temperature of a given spot. Since it is difficult to have the thermometer observed oftener than twice or thrice in a day, it becomes an object of great importance to determine those hours the mean temperatures of which will give that of the whole twenty-four. This however involves an accurate determination of the curve of diurnal temperature; and as this varies with the seasons, its connexion with the curve of annual mean temperature must also be assigned. It is obvious that for these objects a very extended scale of observations is requisite, but that when once attained, the results will be subject to the same general law throughout a considerable extent of country. The mean of the maximum and minimum temperature measured by a register thermometer is one of the best approximations; it is however by no means absolutely accurate. The multiplication of hourly observations has only in one or two instances been resorted to for filling up this blank; but it

* Bibliotheqve Universelle, N.S. ii. 225.

Annales de Chimie, 1831.

[page] 211

is to be hoped that the facilities which military stations offer for fulfilling this most important object will not be neglected.

The only thermometrical register of great extent which has been undertaken, except that of Toaldo and Chiminello at Padua, and of Neuber at Apenrade, was executed by the military at Leith fort, under the superintendance of the Royal Society of Edinburgh; and an account of the results of complete hourly observations during the years 1824 and 1825 has been published by Dr. Brewster*. The results I consider most important to science; and they afford a proof, which I hope will not be overlooked,—that meteorological observations have only to be conducted upon a right scale in order to afford results to a degree of precision scarcely exceeded by any of the physical sciences. The result I particularly allude to is the following. One principal object being to establish the particular hours, the mean of the temperatures of which for the whole year should equal the mean of the whole twenty-four hours, it is obvious that one of these critical times must occur in the morning, the other in the evening. The observations for two years have given the following extraordinary coincidence:

Hour of Morning. Hour of Evening.
Mean Temp. Mean Temp.
1824 9h 13′ 8h 26′
1825 9 13 8 28
Mean 9 13 8 27

Such a series of normal observations ought to be made in every extensive country; for the critical hours vary materially with the latitude, and also with the height above the sea. At Paris, for example, the mean temperature occurs before 9 o'clock in the morning. At Padua the critical hours were 8h 41′ A.M. and 7h 52′ P.M. But notwithstanding this considerable variation, occasioned by a difference of 11° of latitude, the interval elapsed between the morning and evening mean is remarkably constant. At Padua it was 11h 14′; at Leith 11h 12′; at Apenrade, in Denmark, 11h 11′†. It is to be hoped that the exertions already made by the British Association for the establishment of an hourly register near the equator, and also one in the South of England, will be successful, as the results would be of the highest interest for science.

Some of the other most important consequences deducible from the Leith observations are the following.

* Edinburgh Transactions, vol. x.

Schouw Beiträge zur vergleichenden Klimatologie. 1827.

O 2

[page] 212

The mean hour of minimum temperature for the year is 5h 0′ A.M; that of maximum temperature, 2h 40′ P.M.

The deviation of any pair of hours of the same name* from the mean of the day, is less than 0°·5 Fahr. And of all pairs of hours, 4 A.M. and P.M. are the most accurate.

The reduction of the results of any register made at regular hours, in this climate, to the mean temperature of the day, is readily deduced.

The mean annual temperature of any hour never differs more than 3°·2 from the mean of the day for the whole year.

The mean daily range is a minimum at the winter solstice, and a maximum in April. The mean daily range in this climate is 6°·065. The result of the two years agree within 1/1000 of a degree, though the mean temperature was considerably different.

By dividing the curve of daily temperature into four parts, by the maximum and minimum points and the points of mean temperature, the intermediate portions of the curve may be accurately represented by parabolic arcs.

On the whole, these most interesting results give us an insight into what may be done, by multiplying observations, towards bringing the science of Temperature under calculation.

An interesting series of comparative observations were undertaken at Christiana under the superintendance of Professor Hansteen, during the warmest and coldest month of the year 1827†. The result is very striking. The daily variation of temperature at Christiana is in February 12°·01 F.; in July, 12°·09. At Leith, in the former month it is only 3°·57; in the latter, 9°·68. The annual variation is also immensely greater at Christiana than at Leith:

Christiana. Leith.
Mean of February. 16°·224 40°·621
———July 61 · 690 60 · 361
Difference. 45 ·466 19 ·740

These are striking illustrations of the difference of a continental and an insular climate.

The curves of diurnal and annual temperature have been investigated by Professor Hallström of Abo, but before the appearance of the above-mentioned observations; he was led to

[page] 213

the following equation for reducing the mean of the temperatures at 10 A.M. and P.M. to the daily mean at any part of the year:

* "Heures homonymes," Humboldt. This author has observed that the same results are deducible from the Padua observations. Fragments Asiatiques, ii. 420.

Edinburgh Journal of Science, ix. 309.

v=1/2 (x f + x e)—0·33 + 0·41 sin [(n — 1) 30° + 124° 8′]

where v is the mean temperature; 1/2(xf + xe) the mean of the morning and evening temperatures at 10; and n the number of the month, reckoning from January (=1). This is intended to apply all over Europe*. M. Poggendorff inquires with some justice, whether it would not be better to avoid all calculation by inclosing the thermometer in a difficultly conducting medium, by which the daily variations of temperature might be diminished.

A mechanical mode of taking the mean of an infinite number of temperatures has been proposed by M. Grassman, by observing the change of rate caused by the influence of temperature upon the uncompensated pendulum of a clock†. The idea is a good one, but was proposed long ago by Dr. Brewster‡.

M. Bouvard's valuable paper upon the meteorological observations at the Observatory of Paris, contains much information, deduced from the registers of many years, upon the form of the annual curve of mean temperature at Paris§. He observes that the days of greatest and least temperature in the year are the 14th January and the 15th July, differing only a day from an accurate interval of six months; and each follows the corresponding solstice at an interval of twenty-five days. Baron Humboldt has observed the remarkable symmetry of the curve on either side of its maximum ordinate‖. The same author has pointed out the near coincidence of the days of mean temperature observed, even in the too short continuance of registers which we possess¶; these are

Buda, 18th April and 20th October.
Milan, 13th—— and 21st ——
Paris, 22nd—— and 20th——

These interesting inquiries lead to the general subject of Climatology, which, since the publication of Humboldt's masterly Essay on Isothermal Lines**, has assumed a more satis-

* Poggendorff's Annalen, 1825.

Ibid.

Edinburgh Encyclopœdia, Art. ATMOSPHERICAL CLOCK.

§ Mémoires de l'Institut pour 1824. M. Bouvard has also given an equation for the diurnal curve depending upon the sine of the angle corresponding to the time from noon.

Fragments Asiatiques, ii. 422.

¶ Ibid. p. 426.

** Mémoires d'Arcueil, tom. iii. p. 462—603.

[page] 214

factory character than perhaps any other branch of Meteorology. That work has been too long in the hands of every one interested in the particular subject, or in the skillful generalization of groups of facts, to require any notice here; and in touching upon what has been more recently added to our knowledge of the subject, we must confine ourselves within the narrowest limits.

The opinion that the climate of a particular place, or of the globe generally, has materially changed during historic records is improbable; and all the force of that precise and circumstantial evidence which ought to carry weight with it, is against the idea. The eminent naturalist M. Schouw, who has recently published upon several points of interest connected with Meteorology*, has written an Essay, replete with curious matter, upon the supposed change of climate since ancient times, part of which has been translated into English†, which goes to show that we have no authority for assuming such a change. A similar result has been arrived at by M. Arago, who has collected a number of curious facts relative to great colds which have occurred at Paris, showing, in opposition to an opinion which had been started, that we have no reason to believe the climate to have been deteriorating for some centuries past‡. We have already quoted the results obtained by Sig. Libri from the registers of Raineri at Florence.

The old formula of Mayer for expressing the mean temperature of any place in terms of its latitude, which made the temperature of the equator 85°, and of the pole 32°, though a respectable generalization for the time at which it was made, was not calculated to stand the experience of another century. Captain Scoresby, I believe, first pointed out its great inaccuracy; and Dr. Brewster, in a paper presented to the Royal Society of Edinburgh and printed in their Transactions§, showed that the deductions of Humboldt in his Essay on Isothermal Lines would be far better represented by the simple formula

τ = 81°·5 cos L,

τ being the mean temperature of a place in latitude L. This

* Particularly connected with the geography of plants. See an Essay on the Geographical Distribution of the Vine, Edin. Phil. Journal; and an interesting brochure, entitled Specimen Geographies Physical Comparativœ, Hauniæ 1828, which contains some interesting comparative views of the climate of the Alps, Pyrenees, and Scandinavian range, the position of the snow line, &c. We have already quoted his work on Comparative Climatology.

Edinburgh Journal of Science, viii. 311.

Annuaire du Bureau des Longitudes, 1825.

§ vol. ix. p. 201.

[page] 215

indeed applies very well to observations made in the meridian of Europe. A more extended view of the subject, and a comparison of the results of Captains Scoresby and Parry, showed, however, that there must exist two poles of maximum cold, one in America, the other in Asia, round which the isothermal lines circulate: and he has lately pointed out, in a letter to Baron Humboldt*, the remarkable analogy of these to the isodynamical lines of magnetic intensity, developed by Professor Hansteen. In a private letter dated 27th March 1830, Dr. Brewster communicated to me his new formula under this form,

t = (T + τ). (sin1/2 δ. sin1/2 δ′) + τ,

t being the mean temperature of a place of which the distance from the two cold poles is δ and δ′: T the maximum temperature of the globe, and τ the minimum. In the meridian of maximum heat which passes through Europe δ = δ′, and the formula becomes

t = (T + τ). sin δ + τ,

which nearly coincides with the formula T = 81°·5 cos L given above, and represents observations extremely well. I was much pleased to see a formula which took into consideration the actual distance† from the cold poles, because it had always struck me that the modifications of the isothermal lines depending upon the accidental figure of the continental masses, it would be better to discard at once the arbitrary coordinates of latitude and longitude, the essential connexion of climate with the latter being nothing, and with the former modified by an infinity of perturbing causes. This seems, in the arbitrary formula just quoted, to be well effected by making the mean temperature a function of the distance from two imaginary poles of greatest cold. Perhaps the modifying circumstances produced by the physical geography of continents are too complex ever to enter expressly into a formula which should exhibit the relation of the temperature, as an effect, to its really active causes.

Mr. Atkinson has published, in the Memoirs of the Astronomical Society‡, an examination of the results of Humboldt's researches, with a view to obtain an accurate expression for the law of climate. He has however considered it merely as a function of the latitude, which can never represent universally the phænomena. His equation

t = 97·08 cos3/2 l — 10°·53

* Poggendorff's Annalen, 1831: i. 323.

† By employing two distinct formulæ for the two cold poles, Dr. Brewster had before introduced the angle of simple distance. Sec Edin. Trans. ix.

‡ vol. ii.

[page] 216

(where t = mean temperature, l = latitude), it will be observed gives the temperature of the equator = 86°·55, and of the pole —10°·53. The former is decidedly too great, and has been opposed by Baron Humboldt* himself and Dr. Brewster†. It is probable that the mean temperature of the equator does not exceed 81°·5 or 82°.

The temperature of the arctic regions has received the greatest elucidation by the recent labours of Scoresby, Parry, and Franklin. The two last enterprising travellers have established the existence of a degree of cold quite unsuspected in the northern part of America. From admirably conducted observations, embracing a large portion of the year, the following mean temperatures have been established:

Lat. Mean Temp. Observer.
Melville Island 743/4° - 11/2° Parry.
Port Bowen 731/4 + 4 ditto.
Igloolik 691/3 + 7 ditto.
Winter Island 661/4 + 91/2 ditto.
Fort Enterprize. 641/2 + 151/2 Franklin.

To Captain Beechey also we owe some interesting meteorological results‡.

M. Arago had concluded§ from the results of Scoresby, Parry, and Franklin, that the mean temperature of the pole is —25° cent., = –13° F. This however is upon the idea that the cold is at a maximum at the pole, which is not probable; it cannot however be much short of that intense degree. The objection to such a result on account of the supposed increase of ice, which would constantly take place if the temperature were below the freezing point of sea water, I have lately endeavoured to combat, and to show that observation presents no opposition to theory‖.

A gradual accumulation of facts all over the globe is paving the way for a very accurate knowledge of the mean temperature of its surface; and in a few years more our mass of observations -will probably be doubled. Great Britain has done most by her arctic expeditions; and it is earnestly to be desired that with the means she possesses of extending this branch of science likewise in equatorial regions, in the vast continent of India, this great and interesting object will yet meet with some attention,

* Annales de Chimie et de Physique, Septembre 1826.

Edinburgh Journal of Science, vi. 117.

‡ Beechey's Voyage to the Pacific, and Behring's Straits, 4to edit. vol. ii.

§ Annuaire, 1825, p. 186.

Edinburgh Journal of Science, N.S. v. 17; and Bibliotheque Universelle, 1831.

[page] 217

although hitherto totally neglected. In France an excellent register, one of the standard ones in Europe, is kept up, by the assiduity of M. Bouvard, at the Observatory; and several valuable series of observations have been produced by French expeditions in tropical regions. In Switzerland, Meteorology flourishes more than in almost any country in Europe; and though its small extent gives little room for contributions to general climatology, other problems of the greatest consequence have been successfully investigated, as we shall immediately have occasion to notice. From Russia much is to be hoped for, in the prosecution of the science of Mean Temperature, which we believe is even now obtaining daily accessions of facts from observations in the remote regions of Siberia*; the zeal which has established magnetic observatories in various parts of that vast country†, will not, it is to be expected, neglect the union of some of the most interesting meteorological observations with that of phænomena to which they are so intimately allied. But we wish particularly to allude to the exertions making in the United States of North America to elucidate the mean temperature of that important part of the globe,—one of the most interesting points, indeed, which can at present be examined with a view to rectify our knowledge of the course of the isothermal lines, which, except at the equator, are hardly at all known in the new continent. A great number of Academies scattered over this widely extended country, make annual reports of observations on the mean temperature, fall of rain, and natural phænomena, to the Legislature of New York, and the military stations have afforded extensive series of valuable results‡.

Baron Humboldt has recently published an interesting Essay on the Causes of the Inflexions of the Isothermal Lines§. Without containing much of novelty, this little work gives some general and philosophical views upon climatology, pointing out the nature of the hourly and daily variations of temperature, the variable absorbent and emissive powers with regard to heat of the materials of which the visible surface of the earth is com-

* See Humboldt's Address to the Petersburg Academy of Sciences, 28th November, 1829.

† Humboldt, Kupffer.

Edinburgh Journal of Science, viii. 303. x. 267, &c.—In alluding to the exertions of different Governments for the advancement of meteorological science, I am happy to be able to add that of Prussia. I have had the good fortune to meet this summer (1832), in the Alps, M. Kämtz, a zealous member of the University of Halle, who had been sent on a scientific mission by the Prussian Government to establish some most curious facts in Meteorology.— J. D. F. Dec. 1832.

§ In his Fragments Asiatiques, tom. ii. p. 398.

[page] 218

posed; and the changes which cultivation and other circumstances may gradually effect in the elements of climate, especially with regard to the isotheral* and isocheimal† lines, even when the mean annual temperature suffers little variation. The author next considers, at some length, the particular influence of the configuration of soil in modifying the climate, contrasts the continental arrangements of the two hemispheres, and the characters of a terrestrial and marine climate at the equator. He points out in a clear manner the influence of forests upon temperature, from the shadow they produce, from the coolness created by the evaporation at their surfaces, and from the extended radiating surface which they present. Such are the principally interesting points to which this Essay alludes. In the same volume, Baron Humboldt gives some views upon the climate of Asia, which he has collected during his late journey.

The subject of the decrease of heat as we ascend above the surface of the earth, has excited, and especially in Britain, much less attention than it deserves. We hardly know a finer problem for complete solution in Meteorology. Little has been done towards it during the last few years, but we cannot pass it over quite without notice. The principle upon which this diminution of temperature takes place in the higher regions of the atmosphere, is now universally allowed to be the increased capacity for caloric of air when it is rarefied. The first question for solution is, therefore, the specific heat of air under different degrees of condensation,—a point of by no means easy investigation, and which has engaged the attention of some of the first philosophers of the day, as we have already noticed. It has been investigated experimentally by Dalton, and Leslie, De la Rive and Marcet; and theoretically by Laplace, Ivory, Poisson, and Avogadro. The object is the more important, as it is intimately connected with the amount of astronomical refractions. We do not consider, however, that the total effect can be expressed simply by the law of specific heats varying with height; and it appears that the experimental data which have been sought by observations on the atmosphere, have not generally been conducted on principles which can lead to conclusive results. I do not allude merely to the use of insulated observations of temperature at great elevations, which in my opinion can lead to no general result‡, because of the innumerable accidental causes always at work, which can only have their influence multiplied by a long series of observations,—but to the fact, that observations of temperatures at great heights,

* Equal summer temperature.

† Equal winter temperature.

‡ Such as the collection of insulated observations in Ramond Sur la Formule Barometrique de la Mécanique Céleste, p. 189.

[page] 219

whether on insulated points or in balloons, have been directly compared with those close to the massive heating surface of the earth, the radiation of heat from which probably exerts far more influence in producing a decrease of temperature in the very lowest strata of the atmosphere, than the general law which is sought for, and which is usually alone considered. The true law of decrease of temperature, such as it would be if the earth was removed, must be sought for probably by successive stages of balloon observation*, commencing at a considerable height above the surface; or else we must find the means of estimating accurately the calorific influence of the earth by radiation and conduction. There cannot be a question that in the lowest strata the diminution of temperature appears much greater than at higher elevations, if the observations be not made on a naturally inclined surface of soil, but at two stations, one nearly vertically above the other†.

In order that the terrestrial influence may be as much equalized as possible, observations on the mean temperature of table-lands at considerable elevations, (but not covered with perpetual snow, which introduces a new element,) are perhaps the most satisfactory. And hence the reason that Humboldt's equatorial observations are by far the best that we possess‡. Humboldt's general result gives 121 toises of ascent for a diminution of 1° R. The admirable comparative observations at Geneva and St. Bernard, give a surprisingly near approach to this, the difference of mean temperature of the stations being 8°·64 R. for 1069 toises: this gives 1231/2 for 1° R. or 352 feet for 1° Fahr. This probably is the most correct mean result which we can adopt. The influence of the seasons is very considerable, as the following results of M. Guerin at Ventoux near Avignon, lately published, prove§.

Summer, 80 toises for 1° R., or 156 metres for 1° cent.

Winter, 100 ———— or 195 ————

Some good observations made in summer upon the Rigi by M. Eschmann of Zurich, give 97 toises of ascent for 1° R.‖ I have succeeded in establishing, in latitude 56°, a system of observation on this interesting point, from which I hope in due time to be able to obtain the most important results.

* A few observations with balloons have been published by Lord Minto (Edinburgh Journal of Science); it will be necessary however to carry them on upon a much more extended scale, before any general conclusions can be attained.

† See some observations, remarkably confirmatory of this view, by Sir Thomas Brisbane, (made in New South Wales,) in the Edinburgh Journal of Science.

Observations Astronomiques, 4to, i. 126.

§ Annales de Chimie, xlii. 128.

Bibliotheque Universclle, 1827.

[page] 220

Several formulae have been proposed for a general expression of the law of gradation. Lagrange, and many after him, considered the decrement of heat to be simply proportional to the height, and the best observations may be tolerably represented by

t = n H.

Where t, the decrement of temperature, is expressed in degrees of Fahrenheit, and H in English feet, the coefficient n will be a fraction nearly equal to 1/270, or between that and 1/300 for small heights where the decrement proceeds most rapidly: the observations of Humboldt and those at the Grand St. Bernard would, we have just seen, make n about 1/350 Euler considered the progression an harmonical one. Professor Leslie, from experiments upon the heat absorbed by air in rarefaction, proposed theoretically the formula 25 (1/θ — θ) for the diminution of temperature on the centigrade scale; where θ represents the density of the air at the upper station. This formula was first proposed without demonstration*; afterwards the nature of the experiments upon which it rested was explained†: from its principle, this formula only takes cognizance of the influence of the change of specific heat in the atmosphere, without any reference to the effect of the mass of the earth. Professor Leslie's formula has given rise to several discussions, to which it will only be necessary to refer in a note‡.

Mr. Atkinson, in discussing the subject of Astronomical Refractions§, has examined with great care all the actual observations to which he had access, and from them he has deduced the following formula:

h = [251·3 + 3/2 (n — 1)]n,

where h is the height of the station in English feet, n the depression of Fahrenheit's thermometer.

An ingenious attempt was made by M. Mathieu‖ to deduce the law of decrease of temperature in the polar regions, by analysing two observations by M. Swanberg upon the amount of refraction, by means of the formula of the Mécanique Céleste. The result at which he arrived was 243 metres of eleva-

* Leslie's Elements of Geometry.

Encyclopœdia Britannica, Supplement, ArticleCLIMATE. See also Thomson On Heat, p. 122.

‡ By Mr. Ivory, Philosophical Magazine, 1821; by an anonymous writer, Edinburgh Journal of Science, vol. v.; and a paper by myself relative to the last-mentioned one, Id. N.S. vol. v.

§ Memoirs of the Astronomical Society, vol. ii.

‖ Humboldt, Observations Astronomiques, i. 155.

[page] 221

tion for a decrease of 1° cent. This however is at variance with the only direct observation we have on the subject. Captain Parry found in latitude 69° 21′, by means of a kite, that the thermometer indicated no diminution of temperature at a height of 400 feet; it stood at — 24°. Much however cannot be inferred from a single observation.

Some observations have recently been made by M. Kupffer in the range of the Caucasus, by means of the temperature of springs, but too few to admit of any satisfactory conclusions*.

We must now touch upon a point of the greatest importance, and which daily increases in interest,—I mean the proper temperature of the globe itself. We have already pointed at the fine views of Baron Fourier relating to this subject; and from the experimental confirmation which they receive every year, there is little doubt that they will soon be established on an immoveable basis. He considers the globe as a mass in the process of cooling from an intense temperature. He has proved that the heat may be very intense at a short distance from the surface, and yet, from the extremely bad conducting power of the crust, that it may exert no sensible influence on the climate: he actually computes it as not amounting to 1/30th of a centigrade degree. Towards the centre the heat may be of the most extreme intensity, and the phænomena of earthquakes and volcanos may be imputed to its influence. The process of cooling, though at first of course comparatively rapid, may now be considered to have reached an asymptotic condition. It is well known that the influence of the seasons, or the total difference of the effect of solar radiation in summer and winter, affects the temperature of the soil to a comparatively minute depth. Experiments with thermometers, sunk to different depths, have been made at Zurich by M. Ott, near Edinburgh by Mr. Ferguson, and at Strasbourg by M. Herrenschneider†. The influence of the solar rays decreases rapidly; and it is probable, from experiments made at Paris‡, that at about 30 metres, or 100 feet, it is almost extinct. The position where this takes place is called by Fourier the "couche invariable," or invariable stratum; all variations above this plane are imputed to the influence of radiation, all below to the native or primaeval heat of the globe.

A successive influx and efflux of heat is constantly going on

* Voyage au Mont Elbrontz, 4to, Petersburg.

† See a resumé of these experiments by M. Pouillet, Element de Physique, ii. 642.

‡ The excellent observations regularly made in the caves under the Observatory are regularly published in the Annales de Chimie et de Physique.

[page] 222

in the strata above that of invariability. The heat of the solar rays is constantly acting on it during the day, and with an intensity depending upon the absorbent power of the surface, and the latitude. M. Pouillet, from some ingenious experiments, concludes that the solar rays which reach the surface of our globe in the course of a year, have sufficient intensity to melt a complete stratum of ice over its whole extent of 14 metres in thickness*. The amount of solar radiation in various parts of the globe, presents almost an open field for investigation. Mr. Daniell, in his Meteorological Essays, started the apparently paradoxical opinion that the force of solar radiation increases from the equator to the poles; and though his reasonings have been opposed by an eminent French philosopher†, and by meteorologists at home‡, we think he has at least had the merit of pointing out the fact, that the force of radiation is much less in the equatorial and much greater in the polar regions than might have been anticipated. Dr. Richardson has made some very interesting though not quite decisive experiments on this subject in the late Northern expedition§. I am indebted to Mr. (now Sir John) Herschel, for the remark, that observations on solar radiation seem generally to have been made upon an erroneous principle; the true indication of the force of the solar rays, not being the statical effect upon the thermometer, but their momentary intensity measured by the velocity with which they communicate heat to an absorbent body. We may confidently point out the subject of radiation as one which will reward the researches of the Meteorologist.

When the immediate calorific cause no longer acts, the surface of the globe of course begins to radiate the superfluous heat which it had received, and exactly in proportion to the facility with which it received it, the absorbent and emissive powers of surfaces being equal. Hence a nocturnal radiation of heat takes place from the soil, occasioning that cold which, according to the laws known to regulate this process, is materially affected by the purity of the sky; for it is perfectly certain that clouds or any interposed body, or even the finest films, have a sensible influence in intercepting the rays of invisible heat. To the coolness thus produced in bodies exposed at night to a clear sky the phænomena of dew have been accurately attributed by Dr. Wells in his Essay on that subject,—a work undoubtedly

* Elemens de Physique, ii. 704.

‡ Edinburgh Philosophical Journal, xiv.

§ Franklin's Second Journey, 4to edit.

Annales de Chimie, Aug. 1824.

[page] 223

both philosophical and important, but which it appears to me has received fully its due share of commendation.

Let us now see what experimental proofs we have of native heat below the invariable stratum. It is nearly a century since it was first suspected that the temperature of the earth increased as we descend. The proof however has been reserved for our own day, by the multiplication of observations which might annihilate every plausible objection,—for many such there undoubtedly are. Nor can we assert, that before the late researches of M. Cordier* truth has decisively been made to appear. The general consequences which have resulted from his inquiries, and which are substantiated by an abundant collection of facts observed in Cornwall, Saxony, Brittany, Switzerland, America, and other points, are as follow:—That the temperature of any stratum below that of invariability is absolutely the same all the year round†;—That in all strata so situated the temperature increases as we descend, without any exception;—That though the results which have been obtained are far from giving the same law of increase for different countries, which from the imperfection of the observations it was impossible to expect, yet the general progression may be stated at from 25 to 30 metres of descent for an increase of one degree centigrade, or from about 37 to 44 feet for one degree Fahr.‡ M. Cordier has elaborately and successfully refuted the idea that these effects could be produced by the lamps of the miners, though he has shown the nature and amount of the influence they actually exert. A more refined objection, imputing the heat to the condensation of atmospheric air descending into the mines, has been satisfactorily answered by Mr. Fox§, whose scientific observations on the mines in Cornwall, and especially on their temperature, and the electro-magnetic properties of their metalliferous veins, promise so much towards the advancement of science. M. Magnus, whose register thermometer we have already alluded to, and who applied it to the present object, has

* Annales du Museum d'Histoire Naturelle, 1827. See also Bulletin des Sciences Mathématiques, ix. III; Edin. New Phil. Journal, vol. v. & vi.; and De la Beche's Manual of Geology (Introduction).

† See proofs of this in Saxony, Annales de Chimie, xiii. 211.

‡ M. Kupffer has lately deduced 36°·81 English feet for 1° F. Poggendorff's Annalen, xv. Edinburgh Journal of Science, April 1832. The recent experiments of M. Gherard in Prussia, communicated by Baron Humboldt to the Academy of Sciences, give 180 feet of descent for 1° R . Bulletin de la Société Philomathique. Mars—Avr. 1832.

§ Philosophical Magazine, 1830. Both to Mr. Fox and to Mr. Henwood we are indebted for some excellent original experiments. Edinburgh Journal of Science, vol. x.

[page] 224

recently published some good observations on a boring near Berlin*. The temperature of the earth, measured by that of deep-seated and copious springs, has conducted to a similar result; for their variation of temperature as connected with latitude is confined within smaller limits than the mean temperature of the air, indicating a proper temperature of the earth, which at the mean depth of springs diminishes that of those near the equator, and increases it near the poles; the mean point, or where the ordinary temperature of the earth and that of the air are the same, being about latitude 56°. Von Buch pointed out some years ago this interesting fact†, which was deduced from his own observations and those of others, especially of Wahlenberg in Sweden‡. The question has lately been treated in its greatest generality by M. Kupffer, who has established to a considerable extent the course of what he calls the isogeothermal lines, and has given formulæ for their computation as well for longitude as latitude; for, like Humboldt's isothermal lines, he finds that they do not regularly follow the parallels of latitude, but are subject to anomalous inflections. The form of the expression given by M. Kupffer is

ab sin2 l=t,

a and b being constants which vary with the meridian of the place, and which he has computed for a range extending from 85° W. to 60° E. of Paris§. Near the equator, the ground at 25 metres depth appears to have a temperature 2° R. below the mean temperature of the air, whilst in Lapland it is as much above it. I do not think that the connexion of this remarkable fact with the proper temperature of the globe has been pointed out with sufficient distinctness, but this is not the place to insist upon it more particularly.

It is obvious that the temperature of the ocean, which covers so large a portion of the surface of our globe, must have a great influence in the modification of its climate. This subject has therefore occupied particular attention. The most active observers have been Humboldt, Scoresby, Parry, Ross, Sabine, Hall, Davy, and Duperrey. The variations of temperature of the sea being comparatively small, the climate is subject to much smaller fluctuations than on continents. As might be expected, the maximum temperature of the air is greater than that of the surface water; the mean temperature however, it appears from

* Poggendorff's Annalen, xxii. 146.

† See the Edinburgh New Philosophical Journal, vi. 166.

‡ Wahlenberg's observations were originally published in Gilbert's Annalen der Physik.

§ Voyage au Mont Elbrontz.

[page] 225

the observations of Duperrey, is somewhat less. The temperature, as it varies with depth, is of course amenable to the laws of the maximum density of water; though with respect to salt water, this phenomenon still requires elucidation*. Within the tropics, the temperature constantly diminishes as we descend. Towards latitude 70° this decrease vanishes, and an opposite series of phænomena take place; the temperature increasing as we descend. As these points, though interesting in a high degree, are not so intimately connected with Meteorology as those we have been discussing, and those which yet remain, we shall merely give references to those works where general views have lately been given on the temperature of the sea:—Arago, Annuaire pour 1825; Humboldt, Rélation Historique, iii. 514—530, and Fragments Asiatiques, ii. 556 ; Pouillet, Elemens de Physique, ii. 684; Recent Observations by M. Lens, who accompanied Kotzebue; Poggendorff's Annalen, 1830; and the Voyages of Beechey and Duperrey.

From the great and universal importance of the subject of Temperature, and the more general views which admit of being taken of it, we have been induced to extend our review of its different branches to a greater length than we can permit ourselves in the remainder of the subjects which are before us. We proceed therefore to give a brief view of the subject of

Atmospheric Pressure.

Notwithstanding the beauty of the Torricellian method, the barometer must be admitted to be far from that state of constancy, simplicity and perfection, which could be wished, for the purposes to which it is now applied. My attention has long been greatly devoted to the improvement of this instrument, and to the careful study of its desiderata; but I believe we must be content to admit, that on every plan which has yet been proposed, and with any modification of such plans, the barometer will remain liable to considerable objections.

Whilst a barometer is immoveably fixed, its capability of precision is much greater than when it is constructed with a view to portability; indeed there is not perhaps a more difficult problem in philosophical mechanism than a satisfactory portable barometer. I have lately had occasion most particularly to consider the construction of the instrument in both forms, having last year been requested by the Royal Society of Edinburgh to give designs for, and order a standard barometer to be placed in

* From Erman's experiments it would appear that sea water has no point of greatest density above its freezing point. Annales de Chimie, xxxviii. 287.

P

[page] 226

their apartments; and more lately I have had constructed for my own use, a portable barometer intended for a projected tour on the Continent, in which I have endeavoured to unite the great requisites of accuracy, portability, and security from accident. These properties have been considered so much opposed, that hardly any of the various inventions, or modifications of inventions, which are constantly brought before the scientific world, can prefer a claim to all three. Perhaps the most portable barometer susceptible of any considerable accuracy which is in use, is that of M. Bunten, a modification of Gay-Lussac's, having a safety cavity in glass half way up the tube for stopping the progress of any air which may pass the syphon at the bottom*. The instrument is well mounted and graduated; but the principal defect, as well as in Gay-Lussac's, arises from the friction of the mercury in the shorter leg of the syphon, where it never fails to oxidize,—and from the contraction of the scale, which is necessarily much shortened†. I am informed however, by my friend Captain King, that he found it on the whole a satisfactory instrument and extremely portable‡.

An important source of error in portable barometers is the difficulty of finding the actual temperature of the mercury; with a syphon barometer of the kind just mentioned, Signor Bellami and M. Legrand have pointed out methods of converting the instrument into a temporary thermometer, and thus showing its own temperature§.

Among other ingenious devices for diminishing the risk of breakage, one has been proposed by Mr. Jones of Charing Cross, by constructing the tube wholly of iron; such an one has, I believe, been actually completed, but I have not heard of its success. Mr. Robinson, of Devonshire Street Portland Place, London, has lately constructed a barometer in which the tube consists of two parts, capable of being screwed together at the moment of observation.

The adjustment of the lower level of the barometric column is one of the most difficult parts of the apparatus. I am convinced that the French method of bringing up the mercury in a transparent cistern till a fine fixed point impinges on its surface, will gradually come more into use in this country, whenever really good barometers become an object of attention, which at present can hardly be said to be the case. A most beau-

* A good account of this barometer, with M. Arago's Report upon it, made to the Academy of Sciences, will be found in Ferussac's Bulletin des Sciences Mathématiques, x. 187.

† In fact it is generally reduced to one half.

‡ Since writing the above I have used a barometer upon Gay-Lussac's construction with great satisfaction among the Alps. Dec. 1832.

§ Bulletin des Sciences Mathématiques, tom. x.

[page] 227

tiful modification of this plan has been invented and adopted by Dr. Prout in his admirable standard barometer, one of the finest philosophical instruments I ever had the pleasure of seeing; by means of which he informs me he can ascertain the lower level of the mercury to much greater precision than he can read off upon the scale, which is divided to single thousandths of an inch. As no account of it has been published, it would perhaps be out of place to give any description of it here.

Dr. Jacob proposed a cistern in which the mercury assumed a constant level by merely being permitted to overflow*. Mr. John Adie of Edinburgh has contrived a mode of adjusting the level of the mercury without a leather bag, which in great hygrometric extremes may become unmanageable†, by substituting a glass plunger with a stuffing box‡.

The difficulties arising from want of portability, have sometimes brought the instrument back to its earliest stages of simplicity. Some observers now strenuously recommend the practice of constructing a temporary barometer at the place of observation, by filling a tube with mercury, thus dispensing with the precaution of boiling. A distinguished Russian philosopher, M. Kupffer, recommends that air be left above the mercury, and its effect computed§. Neither of these plans can we approve, more especially the latter, as the effect of temperature, the difficulty of ascertaining which we have already noticed, becomes tenfold more important. Upon the whole, we cannot flatter ourselves that the barometer as an instrument has made much progress towards perfection for some time past. In stationary instruments simplicity and solidity are important requisites; and there is one interesting fact which though frequently suspected, can hardly he said to have been substantiated till lately by the observations of Mr. Hudson, the indefatigable observer to the Royal Society of London,—that the sensibility of barometers depends much upon the bore of the tube, which he has found to have a sensible effect even when it is by no means small‖.

The manometer of Hooke was revived about twelve years ago by Mr. Adie of Edinburgh, under the name of the Sympieso-meter, and he has conferred upon it the most essential improvements and the means of giving indications of very con-

* Dublin Philosophical Journal, No. IV.

† As Captain Hall found near the cataract of Niagara.

Edinburgh Journal of Science, N.S. i. 338.

§ St. Petersburg Transactions, 1830; and Journal of the Royal Institution, N.S. vol. i.

‖ Published (since the above was written) in the Philosophical Transactions for 1832, Part II.

P 2

[page] 228

siderable accuracy. The action of this instrument I have examined with great labour and in great detail, being convinced that with some ameliorations it may be made yet more valuable to science, and capable of general adoption, especially by geologists; for it must be admitted, that at present, without the strictest care, it may most seriously mislead the observer. Its portability far exceeds that of any other barometric instrument*.

One of the last donations of Dr. Wollaston to science was a Differential Barometer for measuring minute differences of pressure; but, not being intended for purposes of Meteorology, need not here detain us.

Considering the great attention which is required in conducting continued series of barometrical observations, and the care that is absolutely requisite in having accurate and comparable results by means of really good instruments, we cannot wonder that in but very few places is the mean pressure of the atmosphere accurately known. Much less can we pronounce upon the question whether the general mean pressure over the globe is the same†. The mean height at Paris for fourteen years, being one of the best-determined points, is exactly 756 millimetres. When Meteorology shall have taken its due place among the sciences, and observations are assiduously carried on at several points in connexion in fixed observatories, we shall have some data for determining likewise whether the pressure remains the same from age to age,—a point upon which at present we are wholly in the dark. Professor Schön, from observations at Wurtzburg, thinks that the pressure has increased during the last fifty years‡. But such generalization is quite premature. Some curious anomalies however with regard to mean pressure seem pretty well established, and demand accurate observation. Von Buch observed that the mean pressure on the shores of the Baltic was less than in France, and imputed the difference to what he calls a vallée atmospherique. A similar fact, not less extraordinary, is established by the observations of M. Erman in the East of Siberia§. His barometrical observations would place Jakuzk below the level of the sea of Ockozk, yet the river Lena flows down from Jakuzk to the North Sea, which must be almost if not precisely on a level with the sea of Ockozk. It is well to observe that the mean temperatures of this part of Asia differ very abruptly at short distances, which are probably intimately connected with the phænomenon. M. Erman allows for

* My papers have been published in the Edinburgh Journal of Science, x. 334; N.S. iv. 91. 329.

† See Humboldt, Rélation Historique, 4to edit. tom. iii.

‡ Kastner, Archiv, viii. 475.

§ Poggendorff's Annalen, Oct. 1829.

The variations of pressure may be considered as periodical and accidental.

[page] 229

the effect of air in his barometer, which, considering the variable temperatures to which it must have been exposed, rather diminishes our confidence in the observation. Captain King communicated to me a remarkable barometric anomaly observed by him at Port Famine in the Straits of Magellan. The instrument was compared with that at Greenwich on leaving and on returning to this country, and was observed five times a day, with all care, for five consecutive months. The result gave the very small mean height of 29·462 inches at five feet above high-water mark*. This fact seems intimately connected with those just mentioned.

Of the periodical variations, that which first demands our attention is the horary oscillation. This phænomenon, somewhat indistinctly pointed at by observers in the tropics above a century ago, has within the last thirty years acquired great interest. Baron Humboldt by his observations near the equator gave an impulse to inquiry, and the observations have been pursued with assiduity and success throughout a great range of latitude. The general fact that the barometer attains a maximum in the tropics at 9 A.M. and P.M., and a minimum at 3 or 4 A.M. and P.M., it must be hardly necessary to recall. Nor does it fall within our province to recapitulate the labours of Ramond†, or the individual results of the earlier observers. It is sufficient to mention the well known names of Humboldt, Caldas, Horner, Boussingault and Rivero, and Simonoff, as observers in the tropics; and of Marqué-Victor, Billiet, Gambart, and Herrenschneider in Europe. Of the recent contributions which fall more particularly under our notice, M. Bouvard's Memoirs are the first. In an excellent analysis of the meteorological observations made at Paris‡, he has determined with great accuracy the amount at that station, which gives for the morning period 0·76 millimetre, and for the evening 0o37, by the mean of eleven years. In a later paper he has analysed the law of diminution of the oscillation from the equator to the poles, and likewise the influence of height and of seasons§. He has adopted the Table given by Humboldt in his admirable Essay on this subject‖, and enlarged it by new observations, especially the manuscript ones

* The results have been published, but without any remark, in the first Number of the Royal Geographical Society's Journal, p. 172.

Mémoires de l'Institut, 1812; and his excellent Mémoires sur la Formule barometrique de la Mécanique Céleste.

Mémoires de l'Institut pour 1824.

§ Bibliotheque Universelle, 1820.

Relation Historique, 4to edit. tom. iii.

[page] 230

of Duperrey and Freycinet; and where reduction on account of the hours employed has been necessary, he has introduced a formula similar to that for temperature depending upon the sine of the arc corresponding to the time from noon. Such a formula has also been employed by Carlini* and Hallström. The general result to which M. Bouvard's interesting inquiry led him was, that at the equator the amount of the oscillation is proportional simply to the temperature, on the centigrade scale, of the period during which the oscillation is observed at the given spot, the oscillation and temperature at the level of the sea being unity;—that in any other latitude the same law is to be modified by introducing the additional proportionality to the square of the cosine of the latitude. Or representing by m and t the oscillation and mean temperature at any place and for any period in latitude 0, and by m′ and t′ those quantities at the equator, we have, according to M. Bouvard,

m′ = t′/t m/cos2θ

or, the latitude and temperature being given, to find the oscillation

m = t cos2θ/tm

When the temperature (on the centigrade scale) becomes negative, whether by change of latitude or from height, the oscillation will become negative also and take place in an opposite direction; this inference M. Bouvard confirms by the fact, that the mean oscillation at the convent of the Grand St. Bernard is actually negative.

The observations by Mr. Goldingham at Madras, printed (not published) by the East India Company, confirm the results of former observers. As an attached thermometer seems to have been neglected, the amount of oscillation cannot be depended on, but the critical hours are extremely well fixed.

Dr. Russell at Berhampour (lat. 24° N.), and Mr. Prinsep at Benares (251/2°) have also added to our list of recent observations†. In more northern climates we are also lately indebted for valuable observations to the Royal Society of London,—an abstract of the results of which has recently been published by Mr. Lubbock‡, giving an oscillation equal to 0·57 millimetre§;—

* Memorie della Società Italiana, tom. xx.

Philosophical Transactions, 1828.

Ibid. 1831, p. 223.

§ Since the publication of Humboldt's Essay on this subject, millimetres have become the more usual reference for the measure of the oscillations.

[page] 231

to M. Carlini*, whose Memoir is rather upon the general subject of the variations of the barometer, than containing minute original observation, his register having been continued only for a short time;—to the Geneva and St. Bernard Observatories, where the registers are kept with a regularity and precision worthy of the greatest commendation, and which annually are affording data of the highest value for science†. I have already noticed the important fact that the mean annual oscillation is actually reversed at the St. Bernard, 8000 feet above the aea. M. Eschman has since noticed that it is almost extinct on the summit of the Rigi‡.

Professor Hansteen has published some observations, continued however for only six months, at Christiania, where he gives 0mm·53 for the morning oscillation, and 0mm·40 for the evening§.

M. Hallström from observations at Abo, (likewise of too short continuance,) has endeavoured to deduce the principal oscillation, which he states at 0mm·44; and he has attempted to assign in a general way the law of diminution from the equator to the poles; it is however formed on imperfect data‖. V being the oscillation in latitude l, he gives

V = 0·3931 – 2·3536 cos l + 4·5687 cos2 l

for millimetres. This would give a positive oscillation, at the pole, of 0mm·39 which is quite improbable.

In 1828 I published some observations made by myself at Rome the previous year ¶. Though continued only for a short time, yet, as I frequently made twelve or fourteen observations in a day, I was enabled to trace out very well the diurnal curve of variation, establish the critical hours of morning and evening maxima and afternoon minimum, and give an approximation to the amount.

Since that, I have investigated with great care, during the years 1827–30, the oscillation in latitude 56°,—the most north-

* Memorie delta Società Italiana, tom. xx. In this paper, which is of considerable length, M. Carlini has aimed at giving a type of the mode of treating such observations generally, rather than affording extremely accurate results by the analysis to which he has submitted his own, which were only continued for a few weeks in summer, and again in winter. These observations, however, having been made during part of the time every two hours, are well worthy of being consulted on their own account.

† The annual means are regularly published in the Bibliotheque Universelle.

Bibliotheque Universelle, 1827.

§ Bulletin des Sciences Mathématiques, ix. 32.

‖ Poggendorff's Annalen, 1826. Bulletin des Sciences Mathématiques, ix. 190.

Edinburgh Journal of Science, January and April 1828.

[page] 232

erly point in Europe at which any observations of long continuance on this subject have been made. The results have been published at length in the Edinbuvgh Transactions*; and I have also entered into an analysis of all the existing information on the subject. The following are the general results at which I have arrived.

1st, That near Edinburgh, in lat 56°, the mean annual oscillation between 10 A.M. and 4 P.M. is ·0106 inch, or 0mm·27.

2nd, That the hours of maxima are further from noon in spring and summer than in autumn and winter; and that the amount of oscillation of both the diurnal periods diminishes regularly through the seasons from spring to winter. These conclusions, derived directly from my own observations, I have shown to be the most probable for all parts of the globe, as far as existing observations guide us.

3rd, That the St. Bernard observations, and those of Captain Parry in the arctic regions, both indicate a true negative oscillation, though the second result has been overlooked by M. Bouvard.

4th, That M. Bouvard's hypothesis and formula mentioned above, are founded upon too hasty generalization. This I have shown upon various grounds, but especially from his own quotation of the St. Bernard observations, where, as the mean temperature is much above 0° cent. in summer and below it in winter, the oscillation should be distinctly positive in the former case, and negative in the latter. This I have shown to be precisely the reverse of the fact.

5th, Availing myself of M. Bouvard's excellent Table, with such additions as I could make to it, I proceeded to investigate, from observations made near the level of the sea alone, the influence of latitude in modifying the oscillation; and from a careful combination of the best results, by reducing the squares of the errors to a minimum, I obtained the following equation, which represents wonderfully well the existing observations:

z = 3·031 cos5/2 θ – ·381

for millimetres, z being the oscillation in latitude θ This gives for the equatorial oscillation 2mm·650, and for the poles—·381. The latitude where the oscillation changes its sign, or is = 0, is 64° 8′.

* Vol. xii. The title of the paper is, "On the Horary Oscillations of the Barometer near Edinburgh, deduced from 4410 Observations; with an Inquiry into the Law of Geographical Distribution of the Phænomenon." An abstract has been printed in the Edinburgh Journal of Science for April 1832.

[page] 233

6th, In the course of this investigation, having selected the observations at Cumana and Toulouse, (both being places where the oscillation is positive,) for obtaining approximate values of the constants in the formula, I found to my surprise and satisfaction, that from these observations alone, we might have inferred, à priori, not merely a negative oscillation in the arctic circle, but one not differing sensibly in amount from the actual observation of Captain Parry*.

7th, I have determined from the formula the mean atmospheric tide from the equator to the pole to be equivalent to the weight of a stratum of air 101/2 metres in thickness; and the mean for the whole surface of the earth to be 16 metres, the air being considered under the usual pressure and temperature.

I hope I shall be excused for dwelling so long upon this paper, as it contains not my own observations merely, but the results of all those which I could collect, made up to the present time, in every part of the globe.

In the southern hemisphere, the excellent observations of Captain King at Port Famine† have given us the amount of the oscillation in a much higher latitude than any previous experiments. They correspond very well with my formula.

With regard to the cause of this remarkable and very general phænomenon, extending from the equator to the poles, we are very much in the dark. We must be content to wait for much more complete information before hazarding conjectures. The connexion of it however with other meteorological changes, by whatever means related, seems certain. The diurnal cycle which it so strictly follows, and its modification by the seasons, show the influence of the sun. M. Bouvard's views are certainly so far correct, that temperature appears to be intimately connected with its variations. M. Dove, who is known by several essays on subjects connected with Hygrometry, has pointed out the connexion of the horary oscillations with the state of humidity of the atmosphere‡; and Mr. Snow Harris of Plymouth has kindly put into my hands the results of a number of original experiments, which show in a very striking manner the diurnal changes in the force of the wind,—a subject quite in its infancy, and to which almost no attention is paid,—which correspond closely with the barometric oscillations; the mean force of the wind being much greater at the period of afternoon minimum than at the morning and evening maxima. This is

* See Art. 15. of the paper.

† The observations are given in the Royal Geographical Society's Journal, No. I., and in my paper, Art. 21.

‡ Poggendorff's Annalen, 1831.

[page] 234

analogous to the usual influence of the wind on the barometer, a point not yet quite satisfactorily elucidated, and presents a connexion not to be lost sight of. Mr. Harris, with his usual diffidence, suggested it to me merely as a coincidence worthy of notice, not as the foundation of any hypothesis, and has permitted me to mention his observation here.

The other principal oscillation strictly periodical which we have to notice, is one of which even the existence is hypothetical; I allude to the lunar atmospheric tides. It is remarkable that the observations most lately made, and now before us, are of the most contradictory character. M. Bouvard, by his deductions from the observations at the Paris Observatory, has been led to the conclusion that the minimum pressure takes place at new and full moon, the maximum at the quadratures*. Mr. Lubbock, by the discussion of observations carried on for three years at Somerset House, concludes that the maximum takes place at the syzygies, the minimum at the quadratures, or precisely the reverse†. M. Flaugergues, on the other hand, to complete the contrariety of opinion, has stated, in a memoir on his own observations, the maximum to take place at the last quadrature, and the minimum half-way between the first quadrature and full moon‡. He states the difference of height at lmm·48. Laplace is disposed to consider the lunar influence as not yet established§. On the whole, we must be content to leave this interesting question quite open to discussion.

Among the variable causes which affect the barometer, we shall first notice the direction of the wind. Upon this point observations are more at one. Both the observations at Paris and at London, just referred to, indicate a maximum of pressure when the wind is N.E., decreasing in both directions of azimuth till it reaches a minimum between S. and S.W. This fact may therefore be considered quite established in this climate. The difference of extreme heights amounts at Paris to no less than seven millimetres; at London (from a smaller number of observations,) it amounts to above 5/10ths of an inch, or nearly eight millimetres. Burckhardt, from the observations of Messier, made it 5mm·146‖. The fact of the rise of the barometer in this country with an east wind, is one of the commonest subjects of remark. It is probably in a great measure owing to the cold which accompanies our east winds

* Mémoires de l'Institut pour 1824.

Philosophical Transactions, 1831, p. 227,

Bibliotheque Universelle, xl. 265.

§ Mécanique Céleste, tom. v. Supp. p. 30.

Connaissance des Tems, 1805.

[page] 235

in spring, connected as they probably are with the melting of the snows in Norway*. Mr. Meikle, however, has lately remarked, and we think with justice, that the circumstance of their opposition to the direction of rotation of the earth, will cause an atmospheric accumulation by diminishing the centrifugal force of the aerial particles†.

The accidental variations of barometric pressure are greatly influenced by latitude. At the equator they may be said to be almost reduced to nothing; for it rarely happens that any change takes place to interfere with the regular course of the diurnal tides. A hurricane creates almost the only exception. The amount of variability increases towards the poles, in a great measure owing probably to the irregularity of the winds beyond the tropics. The mean amount of variation may be stated at the equator at two lines, in France at ten lines, in Scotland at fifteen lines, throughout the year; but this quantity has its monthly oscillations. Hence, a series of lines of equal variation of pressure, or isobarometrical lines as they have been termed, may be constructed‡. These do not appear to follow the parallels of latitude, but, like the isothermal lines, undergo inflections, and are stated to have a striking similarity to the isoclinal magnetic lines of Hansteen. If so, it is probably by the medium of temperature that these two are connected.

The great extent of country over which the accidental variations of the barometer take place, is one of their most striking features; and in a future and more advanced state of Meteorology we may be able to draw the most interesting and important conclusions from the great atmospheric tidal waves which are thus perpetually traversing oceans and continents. The best example we possess of a systematic examination of these great progressive fluctuations, is, it is to be regretted for the present character of the science, of rather old date. The Meteorological Society of the Palatinate was set on foot in 1780, and by the distribution all over Europe of instruments of the best construction then known, made at one common establishment, founded a set of observatories which annually afforded comparable results of the most intrinsic interest§. It is to be lamented that a system which at the present time has no successor, should have lasted only ten years, having ceased with the life of the Secretary of the Society: not

* See Mr. Marshall "On the Causes of the East Winds in Spring," in the Edinburgh Journal of Science.

Edinburgh New Philosophical Journal, iv. 108.

‡ Kämtz, Jahrbuch der Physik und Chemie, 1827; and Bulletin des Sci. Math. x. 199.

§ Published at Mannheim under the title of Ephemerides Societatis Meteorologieœ Palatinœ.

[page] 236

only has it afforded many important results, especially upon the course and progress of barometric fluctuations, but has left a model of a scheme of combined exertion which the savans of the nineteenth century would do well to imitate. Some account of this Society and of the results of their labours, with projected charts of the barometric oscillations as a specimen, have been given to the world by Mr. Daniell in an interesting article in the second edition of his Essays*.

The connexion of barometric changes over large districts is very important in the determination of heights by simultaneous series of observations carried on for a considerable time at points even very distant. Examples of such an application at Paris and Clermont are given by Ramond†; but to avoid incidental derangements, the continuation of the observations for some time is desirable.

More lately, a comparison of the barometric changes at some principal points in Europe has been given by Prof. Schouw‡ , who has been followed by M. Kämtz in pointing out the connexion of the winds with such changes, and who has illustrated the influence of the prevalent aerial currents which traverse Europe, though not with apparent regularity, yet at least subject to some general laws§.

Of all the problems in Meteorology, few appear to me so intrinsically beautiful as that suggested by the fertile genius of Pascal,—the application of the barometer to the measurement of heights. It should, I think, be an object of ambition to bring this elegant method to the utmost degree of perfection of which it is susceptible. The laborious and praiseworthy experimental exertions of Roy, Shuckburgh, De Luc, Saussure, and Ramond, united to the theoretical skill of Laplace, have brought the method to a degree of precision which a century ago might well have been considered unattainable: but we are by no means arrived at the point at which improvement becomes hopeless, nor do we think that all has been done which might have been accomplished since Ramond's last determination of the coefficient of height, and his consequent improvements upon the barometrical Tables.

One important element neglected in the investigations of Laplace (at least only approximately estimated,) has lately begun to acquire the importance it deserves. I allude to the correction for moisture. The degree of Saussure's hygrometer was indeed made an element of calculation in some pretty early

* Meteorological Essays, p. 541.

Mémoires sur la Formule Barometrique.

Bibliotheque Universelle, xxxix. 260.

§ Ibid.

[page] 237

Tables; but the means of measuring the force of vapour with accuracy being only lately attained, the due correction has but within a few years become an object of adequate attention. Dr. Anderson, who has bestowed great attention upon the subject of Hygrometry, wrote a paper on this correction a few years ago*. Mr. Galbraith, in his Mathematical and Astronomical Tables, has followed Dr. Anderson in giving the correction, and has facilitated its application by the use of Tables.

Considering the problem as one of the highest interest accurately to solve, we approve of the introduction of every correction established upon sound theory and accurate experiment conjoined, even though in amount it may be less than the errors of observation or the unavoidable uncertainties arising from the interference of imperfectly understood active causes. By this process uncertainty will gradually be cleared away; and though there will undoubtedly be a limit beyond which no human perseverance can carry the approximation to truth, and a much wider one within which not one observation in a hundred will come, yet still truth will be separated from error, and the actual anomalies unaccounted for will be eliminated with precision. We do not therefore blame the superfluous accuracy (practically considered,) at which the formula of Laplace appears to aim. And for every-day observations it is easy to substitute those simple expressions which in most cases will give almost as good an approximation to the truth†. But in all experimental investigations where the arrival at truth within certain limits is the object, too great care can hardly be taken to avoid the intrusion of causes always acting in one direction, or which in the mean of a number of observations do not compensate themselves. Such, in fact, are some of the minuter corrections of Laplace, as those for latitude and for diminished gravity in a vertical direction.

The configuration of the ground has a considerable effect on the measurement of heights by the barometer, as has also the season of the year and time of the day. This last point has lately been a more especial subject of attention, although along with the former it was investigated by Ramond with his usual assiduity, who pointed out noon as the best hour for the experiment. It is very clear that the horary oscillation will in the first place affect the barometric measurement, but this is in Europe a very minute quantity and easily allowed for. At the equator, being almost the only variation to which the mercurial

* Edinburgh Philosophical Journal, vol. xii. xiii.

† Among the numerous forms which these have received, there is perhaps none more comprehensive and satisfactory than that given by Mr. Baily in his most valuable portable volume of Astronomical Tables, Lond. 1827.

[page] 238

column is subject, when it is allowed for, the height at the level of the sea may be considered known at any instant,—a most material advantage to the observer between the tropics, and which has conferred much of their accuracy upon Humboldt's beautiful barometric levellings and sections*.

The great obstacle to the accuracy of barometrical measurement, and the most influential change produced by the hour of the day, is to be found in the variable temperature of the strata of air intervening between the stations, as it is clear that the mean of the upper and lower temperatures may often deviate greatly from the true mean of the intercepted column. The currents produced from the plains to the mountains during the diurnal revolutions of temperature are extremely considerable; and hence the errors arising from the hour of the day greatly exceed those which the horary oscillation would produce. This has been pointed out and experimentally investigated by M. Horner, a Swiss Meteorologist of great activity, who made the mountain of the Rigi the scene of his operations†. Still more lately M. Gautier, Professor of Astronomy at Geneva, has published some interesting observations on the same spot, by which he found the error from the hour of the day alone, to amount to 14 toises upon a height of 700, the corresponding observations being made at Zurich‡.

In my papers on the application of the sympiesometer to the measurement of heights, already alluded to, I have given the results of many comparisons of this method with the geometrical one in several parts of Scotland, and a number of heights from original trigonometrical operations§.

I cannot dwell either upon the construction of portable barometers, or upon the precautions required in observation; but I strenuously recommend the subject to the scientific meteorologist, as one which will repay his labour, and which is yet open to most important ameliorations. When we consider the accuracy and extensive knowledge we have arrived at in the position of points of interest on the surface of the globe, with regard to the coordinates of latitude and longitude, and how little has been done for the third coordinate of elevation, we shall have a field before us open to cultivation at every corner. The results to physical geography of what has already been done by the use of the barometer, excite our warmest hopes of its extension. To mention only one instance;—the singular discovery

* Voyage aux Régions Equinoctiales; ATLAS, and OBSERVATIONS ASTRONOMIQUES. Baron Humboldt has recently circulated a beautiful "Carte hypsometrique" of the Cordillera.

Bibliotheque Universelle, 1831, N.S. iv. 337.

Ibid. v. 337.

§ Edinburgh Journal of Science, N.S. iv. 91. 329.

[page] 239

by MM. Parrot and Engelhart, of the depression of the Caspian and Lake Aral below the Mediterranean*, and the not less extraordinary extension of this anomalous fact by MM. Humboldt, Rose, and Hoffmann, to an immense territory about 18,000 square leagues in surface†. These conclusions, most important for physical geography, might never have been attained but for the barometer; and at the suggestion of Baron Humboldt the Academy of St. Petersburg have undertaken to prosecute the inquiry with the same instrument, to institute "barometrical soundings," as they have been aptly termed, over this vast crater-like depression, and establish the lines of equal altitude.

A most important synopsis of what has been done in Europe in this department will be found in the Orographie de l' Europe,—a collection of above 7000 heights, formed by the industry of M. Brugiere‡ to whom the scientific world is most deeply indebted, and whose work has been deservedly approved and rewarded by the Geographical Society of France. A vast proportion of the determinations of heights in this volume are due to the barometer.

Humidity.

Hygrometry, scientifically considered, has only had justice done to it within a very short period. Till Mr. Dalton established the true views of the connexion of temperature and the tension of vapour, meteorologists had vague ideas of the true expression of degrees of moisture. The labours of Saussure, though most meritorious, were destined to be superseded by a more elaborate analysis of the subject; indeed his views of hygrometry were in some respects so very imperfect, that he was not aware of the fact, that the coolness produced by the evaporation of water from porous bodies, was independent of the rate at which the moisture was carried off by currents of air, —a want of knowledge which gave him much trouble.

On the general principles of Hygrometry I have no intention of dwelling; I shall chiefly confine myself to a notice of the latest additions to the subject. It will be necessary however to premise one or two observations on the general state of the question.

If the views of Mr. Dalton, noticed in an early part of this Report, be true, with regard to the condition in which vapour exists in the atmosphere,—views, which are now universally admitted,

* The depression of the Caspian is 334 English feet below the Mediterranean, and it has recently been ascertained by Captains Duhamel and Anjon that Lake Aral is 117 feet above the Caspian.

† Humboldt, Fragmens Asiatiquets, tom. i. pp. 9. 91. 136.

Mémoires de la Société de Géographie, tom. iii. Paris 1830.

[page] 240

in so far as they consider the tension of vapour as totally independent of that of air, or the presence of air at all,—we must banish all confused notions from our minds about "saturation of air with moisture," "solvent power of air," &c., which are to be found even in the very writings in which Mr. Dalton's principle is assumed as established. This should be guarded against with care, because it may insensibly lead in practice to the most innaccurate ideas regarding the influence of the presence of gaseous matter. We think nevertheless that in some cases the rage for purifying our scientific nomenclature has been carried too far, where even the results of reasoning are arraigned because they include the use of terms suggesting perhaps hypothetical views, but the adoption of which conventionally, need not be objected to.

The easiest way of obtaining a distinct, simple, and accurate knowledge of the hygrometric state of the atmosphere at any moment, is to ascertain by some means the temperature at which the vapour then existing can no longer maintain its aeriform state, or, in other words, to find the temperature of the dew-point. Then being furnished with a Table of the elasticities of aqueous vapour at different temperatures, the elasticity is of course equal to that of vapour which can just subsist at the temperature of the dew-point; whence the weight of grains in a cubic inch may be easily computed from the experiments of Gay-Lussac, and the expression of the sensible state of humidity of the atmosphere at its own proper temperature, must be obtained by the ratio of the vapour actually existing in a cubic inch, to what might have existed without deposition, in the same space.

Such, in few words, is the rationale of the dew-point experiment. Let us see now the means we have of arriving at this result. Regarding instruments, the simplest form of the experiment is that which Mr. Dalton employs. The dew deposited on the surface of a glass of cold water has been observed from the earliest times, and has been particularly alluded to by ancient authors; let therefore the cold liquid be transferred from one glass to another till the deposition ceases,—the temperature then measured will give the dew-point. Mr. Daniell's elegant instrument is too well known to require minute description: he has applied the principle of the Cryophorus of Wollaston to obtain the requisite cold for the production of dew upon a ball of dark-coloured glass, the temperature of the ether inclosed, being measured by a delicate thermometer inserted. This instrument has come into very general use, and notwithstanding some delicacy required in the management of

[page] 241

it, and an occasional difficulty of arriving at a precise result, we may expect the most valuable results from its application to hygrometry.

The dew-point hygrometer of Mr. Jones*, though more simple and compact, is not so satisfactory in its results; it consists of a thermometer with a cylindrical bulb turned upwards and half-covered with muslin, which is cooled by pouring ether upon it, and the deposition of the dew is observed on the upper part of the bulb. A different form has lately been brought forward by Mr. John Adie of Edinburgh†, who incloses the bulb of a very delicate thermometer in an exterior ball of glass, the interval being filled with mercury; and he observes the deposition of dew on that portion of the outer ball from which the covering of muslin for receiving ether has been removed. By agitating the instrument at the time of deposition, Mr. Adie has been able to get results more closely agreeing with Dalton's experiment, than by the hygrometers of Daniell and Jones. The principle of the instrument is very obvious; from the small size of the ball of the thermometer the temperature is more accurately found than in Mr. Jones's apparatus; and the small bulk and better conducting power of the medium interposed between it and its outer case, render it perhaps more sensible than the instrument of Mr. Daniell. To do the latter instrument justice however, (and from my experience of dew-point instruments in their simplest forms, I think the remark of importance,) the temperature at which dew appears should not only be noticed, but that at which it disappears. The errors of the two must almost always be in opposite directions, and the mean should be taken.

A dew-point hygrometer, in some respects resembling Mr. Adie's, has been proposed in America by Mr. Hayes‡.

Another instrument,—and though I have not tried it, I confess it appears to me a very elegant one,—has recently been proposed by M. Pouillet§. He places a delicate thermometer vertically with its ball upwards, which passes into a small cup of polished silver. Ether is poured into the cup till it covers the ball, and when by the coolness produced by its evaporation the deposition of moisture is produced on the silver, the temperature is noted.

There is probably no instrument which gives the dew-point with so much accuracy as Mr. Dalton's simple experiment when

* Philosophical Transactions, 1826.

Edinburgh Journal of Science, N.S. i. 60.

Silliman's Journal, xvii. 351.

§ Element de Physique, ii. 732.

Q

[page] 242

the glasses employed are thin. This is the result at which Dr. Thomson of Glasgow has arrived*.

The next point is the Table of the elasticities of vapour at different temperatures. Mr. Dalton's excellent Table, or that calculated by Dr. Young from the experiments of Dr. Ure, will be quite sufficient for the range of atmospherical temperatures. The new Table derived from the meritorious labours of the French savans, whose experiments have been carried up to a pressure of steam amounting to 24 atmospheres, will probably become the standard reference on this subject, at least in the case of high pressures†. The formula at which they have arrived, and which bears a striking analogy to that of Dr. Young, is the following:

e = (1 + 0·7153 t)5,

where e is the elasticity in atmospheres (reckoned at =0m·76), and t the temperature reckoned from 100° and computed in centigrade degrees.

From any such Table of elasticities, with Gay-Lussac's result for the specific gravity of aqueous vapour, the weight in a cubic inch under any circumstances may easily be computed. It is hardly necessary, however, to repeat that the expression for the degree of humidity is not the actual weight of moisture in a given space, but the proportion which that bears to the weight which might exist without deposition under the circumstances of temperature and pressure.

Great as are the advantages of simplicity of calculation, which the dew-point experiment affords, there is a less direct experiment which offers great facilities in performance and likewise the means of self-registration. I allude to the moistened bulb hygrometer, in which the coolness produced is a function of the dryness of the atmosphere, without bearing any relation to the force of wind or other circumstances which affect the rate of evaporation. Under the simplest form of two thermometers, one of which had its ball moistened, it was employed by Dr. James Hutton; and afterwards Professor Leslie adapted it to the principle of his differential thermometer; a change perhaps not contributing to the simplicity of the instrument, which still requires a detached thermometer to determine the temperature of the air. Accordingly, the instrument in its first and simplest form (in which for years we have been in the habit of using it,) has recently been reproduced by M. Auguste, under the high-

* Thomson On Heat, p. 256.

Annales de Chimie, Janvier 1830, tom. xliii. p. 74.

[page] 243

sounding name of the Psychrometer*. This instrument has been employed by Baron Humboldt in his recent journey in Asia, where he had occasion to observe a very high degree of dryness, the coolness by evaporation amounting to 11°·7 cent., the temperature of the air being 23°·7†. It were to be wished that, for the improvement of the theory of the instrument, he had at the same time ascertained the dew-point by experiment. M. Auguste is himself the author of the formula by which the tension of vapour is deduced. He has published not only a paper expressly on this subject‡, but an essay (which I have not been able to meet with,) upon the progress of Hygrometry in modern times§.

The perfection of the method of the moistened thermometer forms an important and an interesting problem. Mr. Leslie's solution, which was the first, offers a near approximation to the truth, but at the higher temperatures will require modification, especially as instead of adopting any of the Tables of the force of vapour now in use, he has contented himself with the general result of some original experiments, that the "capacity of air for moisture," to use his own phrase, is doubled by the increase of temperature by every 15° of the centigrade scale. This leads him into inevitable errors at higher temperatures‖. Dr. Anderson's elaborate investigation contained in an able article on Hygrometry in the Edinburgh Encyclopœdia, to which we can do no more than allude, appears also to be faulty in the higher parts of the scale, if we can depend upon some experiments recently made by an anonymous writer in India¶. Undoubtedly the most valuable application of Professor Leslie's hygrometer will be, by rendering it self-resistering on the simple principle proposed by the Rev. Mr. Gordon, which is similar to that of Rutherford's minimum thermometer**.

* Bulletin des Sciences Mathématiques, vii. 379.

Fragments Asiatiques, ii. 378. At Geneva, in August last, I observed a coolness by evaporation amounting to 20° Fahr., the thermometer in the shade being at no less than 92°. I then found it quite impracticable to obtain a deposition upon Daniell's hygrometer.—Dec. 1832.

‡ Poggendorff's Annalen, 1828. There is a paper by Brouwer upon Auguste's instrument, in the Amsterdam Journal, entitled "Bijdragent ot de Naturkundige Wetenschappen" 1831, p. 272. See also Bull. des Sci. Math. x. 302.

§ In German. Read to the Society of German Naturalists in 1828.

‖ Professor Leslie's researches are contained in a tract upon "Heat and Moisture," Edin. 1813: and in the article METEOROLOGY in the Supplement to the Encyclopœdia Britannica.

¶ See two clever papers in a periodical work entitled "Gleanings in Science,1* Nos. II. and III. Calcutta 1829. The Author points out the great difficulty of using Daniell's hygrometer in warm climates, from the deterioration of ether.

** Edinburgh Encyclopœdia, Art. METEOROLOGY.

Q 2

[page] 244

The researches of Gay-Lussac upon the scale of Saussure's hair hygrometer, are too well known to require notice here; we only mention them to observe that the analogy to the abscissae of a hyperbola, of the tensions of vapour, the ordinates representing the degrees of the hygrometer, has been further extended by Signor Melloni in a long paper recently published on this subject*.

The hygrometer proposed by M. De la Rive indicating the temperature evolved by the Combination of a film of sulphuric acid with the moisture of the atmosphere, has not as far as we know come into general use.

The distribution of vapour in the atmosphere is a most curious and difficult problem, of which the data are only now beginning to be collected. We know the mean tension of vapour at very few points on the surface of the globe, which, from the influence of temperature, varies exceedingly, and will one day be the subject of connected and scientific discussion as satisfactory as the isothermal lines are at present. Dr. Anderson has given some interesting views upon what we may believe to be the distribution of vapour from the equator to the poles†, and the same subject has been taken up by Mr. Daniell in his Essay on the Constitution of the Atmosphere‡. As to its variation with height, we are almost equally in the dark, but we are certain that intense dryness reigns in the higher regions of the atmosphere‖. The law of decrease is probably not a regular progression: it appears probable from many circumstances, and in particular from some experiments of Captain Sabine, that the dryness is pretty constant for a certain height, and then rapidly diminishes. In fact there is certainly a stratum of air at the height of from 1 mile to 4 miles, which is more frequently saturated with vapour than any other, and which constitutes the region of clouds.

The annual and diurnal variations of temperature produce effects in the distribution of humidity, analogous to those which we observe in passing from one latitude to another. Even with our extremely limited views of the nature and extent of these changes, we can trace, with a little care, the influence of the great fundamental law of hygrometry, in producing clouds, mists, and other phænomena, which, in hilly countries espe-

* Annales de Chimie, xliii. 39.

† Article HYGROMETRY in the Edinburgh Encyclopœdia.

Meteorological Essays, p. 73, &c.

‖ The interesting researches of M. Kämtz in the higher Alps, promise to throw the greatest light on this important point. I had the satisfaction of witnessing last summer along with him, at the height of 8,500 feet, a degree of natural dryness unexampled, I believe, in the annals of hygrometry.— Dec. 1832.

[page] 245

cially, produce such grand and varied spectacles, and indicate in the most beautiful manner the constancy of the laws by which the temperature and variable conducting powers of the materials on the surface of our globe, modify the distribution of atmospheric vapours*. The subject is one of wide extent, and may at a future period disclose very interesting results; it does not appear however to present such definite points of investigation as to be reckoned among the first objects of the scientific meteorologist in search of general laws.

The nomenclature of the clouds, adopted by Mr. Howard, is a happy specimen of a conventional system, and is well calculated to stamp with a definite character the future results of observation; each species of cloud very probably is attended with a characteristic hygrometric condition, and most likely exists between fixed limits of altitude. I do not recollect to whom we are indebted for a suggestion which well deserves attention, but which cannot be accomplished without that essential condition which it seems the fate of Meteorology to want,—Cooperation. If by a series of little maps of the state of the sky we could represent the daily condition of the atmosphere over a large continent such as Europe, what curious results might not be unfolded! The determination of the existence of immense clouds covering whole countries for days together, while others were under sunshine—the watching of the progress of these clouds, not so much by the influence of wind, as by a gradual process of hygrometric dissolution and recomposition, day after day, would give us more insight into the operations of the higher atmosphere on the large scale, than a thousand insulated observations.

The diurnal extremes of the hygrometric state are of course limited by those of the temperature of the atmosphere; the minimum temperature causing a deposition of moisture when it exceeds a certain amount, and the limit within which the maximum temperature of the air is kept, (86° Fahr. being the maximum over the ocean at any point of the globe†,) preventing the existence of vapour beyond a certain degree of tension. Dr. Anderson has in an elegant paper shown the connexion which is hence established between the dew-point at any time of the day, and the minimum temperature of the same period‡.

We cannot propose to meteorologists a finer problem for complete solution, than that of the moist bulb hygrometer; which will require a close analysis of all that has hitherto been

* See Sir Humphry Davy "On the Formation of Mists," Phil Trans.

† Arago, Annuaire, 1825 p. 186.

Edinb. Phil. Journal, vol. xi.

[page] 246

done on the subject. M. Gay-Lussac, who has communicated some valuable observations towards its attainment, thought that it would never repay the labour of complete investigation. In the present state of the science, however, we look upon it otherwise, and feel strongly assured that in a few years the more direct method of the dew-point will be banished altogether*.

Atmospheric Phœnomena and Precipitations.

In the introduction to this Report I have stated my intention of by no means undertaking to examine the wide field which this subject opens to us. My remarks will be confined to one or two individual points, upon which some general views have been entertained. Those I have selected are Wind, Rain, and Electrical Phænomena, including the Aurora Borealis.

The direction and force of the Winds we have already seen to be intimately concerned in the modification of climate, and in the distribution of temperature in the atmosphere. The periodical winds of the equator and tropics correspond in regularity to the uniform course of the seasons and the limited range of the barometer characteristic of that part of the globe: nor has anything particular been added to our knowledge of these great currents, of late years, which requires notice here†.

In temperate climates the irregularity of the wind in general seems so great as to baffle inquiry. There are a few leading points, however, which show that there is something to be seized in this question, and that an analysis of it may one day lead to more general results. It is undoubted that the south-west is the predominant wind of Europe, and the east winds in spring may be considered as almost accurately periodical in the climate of Britain‡. M. Schouw has gone a step further, and has shown that the prevalence of westerly winds diminishes as we advance towards the east of Europe§. The west winds at London exceed the east in the ratio of 1·7 to 1. At St. Petersburg this is diminished to 1·3 to 1‖.

M. Erman has determined the mean direction of the wind in Russia and Siberia to be as follows:

* In addition to the references already given, we may mention, for the use of those who may pursue the subject, a paper by Mr. Meikle on this point, in the Edinburgh New Phil. Journal, ii. 22; and some articles in Poggendorff's Annalen for 1829, by M. Dove, whose remarks on the connexion of the horary oscillations of the barometer with humidity we have already noticed.

† There is a paper by Captain Hall on this subject in the second edition of Mr. Daniell's Meteorological Essays.

‡ See Mr. Marshall "On the Causes of East Wind in Spring," Edin. Journal of Science.

§ Beiträge zur vergleichenden Klimatologie.

Bulletin des Sciences Mathématiques, x. 201.

[page] 247

St. Petersburg S. 41° W.
Moscow S. 35 W.
Kasan S. 52 W.
Tobolsk S. 47 W.

Baron Humboldt thinks that the western winds diminish in frequency from St. Petersburg towards Central Asia, though they increase towards the North of Siberia*.

The direction of winds is greatly affected by the configuration of a country, their general direction being modified, so as to coincide with its natural lines of elevation and depression. It is probably on this account that in Egypt the winds are generally either north or south, the former prevailing nine months in the year†. Where the climate is tolerably regular, as in the South of Europe, the direction of the wind makes all possible difference in its character. The transition from a sirocco to a tramontana at Rome or Naples is as great as the effect of ten degrees of latitude. It is surprising therefore, that, powerful elements as these aerial currents are, they have been so imperfectly studied.

I have now before me the results of a register of the force of the wind by Lind's anemometer for the year 1826, kept with great assiduity by my friend Mr. Snow Harris of Plymouth. I have already alluded to the connexion which he has pointed out between the force of the wind and the horary oscillations of the barometer, which has not before been remarked: indeed the observation of the anemometer is so rare, that there are few meteorologists who have persevered in the use of it. This renders the register of Mr. Harris the more valuable. The mean force of the wind for the whole year at 9 A.M. was 0·855; at 3 P.M. 1·107; and at 9 P.M. 0·605. Mr. Harris informs me that he has found the anemometer of Lind a more satisfactory instrument than it is usually considered. The improvement of anemometers has been almost abandoned for some time; indeed it may be doubted whether, with an element so momentarily variable, insusulated observations can be of very great value. M. Leroy has proposed what he calls an Eolian clock, which by means of machinery is intended to measure the direction and force of the wind‡. I think that if the anemometer is ever to become an available meteorological instrument, it must be on some principle of self-registration such as I proposed about two years ago§. Either by a piece of clock-work or some simple movement put in action

* Fragments Asiatiques, ii. 353, note.

† Belzoni, Researches and Operations in Egypt and Nubia, vol. i.

Bulletin des Sciences Mathématiques, ix. 32.

§ Edinb. Journal of Science, January 1830.

[page] 248

by the wind itself, I proposed that small spherules of wood or other light matter, or even shot, should be let fall through a free space, suppose of three feet, and that the force and direction of the wind should at once be measured, at every interval of the falling of a spherule, by the amount and direction of the deflexion produced, and which should be ascertained by the dividing into compartments a platform arranged to receive them. I have made some experiments on the subject, and have every reason to believe that the method admits of great accuracy, and that it is perhaps the most satisfactory mechanical one that has been proposed. I conceive that Professor Leslie's ingenious plan of measuring the force of the wind by the cooling of a thermometer exposed to it, is the most satisfactory indirect method, and has not met with the attention which it deserves*.

We have seen that the direction of the winds exerts an important influence on the height of the barometer. There is another source of action which it creates, and which is less understood. M. Schubler has shown, in an interesting paper, that the winds have each their characteristic electric power The precipitations during the winds from the northern half of the circle of azimuth, have a ratio of positive to negative electricity which is a maximum; and in the other half it is a minimum, the negative precipitations when the wind is south being more than double the positive ones. The mean intensity of electricity, independent of its sign, is greatest in north winds. We must refer to M. Schubler's paper for his reasoning upon these facts†.

It is a fact to be attended to, that the progress of a wind and the storm which may accompany it is not always in the direction in which it blows. M. Pouillet terms the modes of propagation of wind by "impulsion" and by "aspiration‡." In the latter case, a vacuum or diminution of pressure being at any point effected, the air which flows to fill it up commences, of course, its motion nearest the point of deficient equilibrium, from which the current gradually retires. Franklin long ago compared it to the flow of water through a canal upon opening a sluice. Mr. Mitchel in America has discussed this view of Franklin, which he thinks will not always apply; indeed the case is one of difficulty, and, unless we can ascend to the first active causes, would only lead into unprofitable speculation. In many cases the deficiency assumed on Franklin's hypothesis cannot be proved, and in some is untenable; but we

* Essay on Heat, p. 284.

Jahrbuch der Chemie und Physik, 1829, Heft iii.; Bibliotheque Universelle, Nov. 1829; Edinb. Journal of Science, N.S. iii. 116.

Elemens de Physique, ii. 715.

[page] 249

are not sure that we shall gain much by assuming the gyratory theory of Mr. Mitchel*.

We would recommend to any meteorologist taking up this subject, to endeavour to establish observations at two stations, one considerably elevated above the other, and to trace the course of the wind at the two points when it is changing its direction: we know that currents in various directions several times superposed often coexist in the atmosphere, and it is probable that changes of wind generally commence at considerable elevations.

Of all the columns of that too often unprofitable work, a Meteorological diary, one of the most profitless has generally been that devoted to the direction of the wind, as in its usual form it does not admit of having any average taken, and therefore remains an undigested mass of insulated observations. In order to draw any useful conclusion from this observation, we would therefore recommend the adoption of Lambert's numerical form, in which the south is denoted by 0°, and the angle is measured round the horizon by the W., N., and E. In this way S.W. is denoted by 45°, W. by 90°, &c.

On the subject of RAIN,—a very important one in a practical point of view,—we have not lately obtained much new information. The theory of Dr. James Hutton remains nearly unaltered, only strengthened and enlightened by the clearer views of the nature of deposition which we now possess. The connexion of rain with the fall of the barometer has met with one elucidation from Mr. Meikle† which is worthy of notice, because the change of pressure, it is shown, may be a cause as well as an effect. He observes that the expansion of air accompanying diminished pressure being productive of cold, will diminish the elasticity of the existing vapour, and cause a deposition.

M. Arago has collected many interesting facts in the phænomena of rain‡. He has traced the progress of decrease in the annual amount from the equator to the poles. It is now known that on the Malabar coast in lat. 111/2° N., not less than 123·5 inches of rain fall in a year; whilst in lat. 60° it is reduced to 17 inches. The law of decrease is not known with accuracy. The author of the article PHYSICAL GEOGRAPHY in the Supplement to the Encyclopœdia Britannica§, has proposed the following formula for the fall of rain in inches:

75 (rad. – sine lat.) + 8,

* Silliman's American Journal, xix.

Royal Institution Journal.

Annales de Chimie, xlii. 360; Annuaires du Bureau des Longitudes pour 1824 et 1825.

§ vol. vi. p. 163.

[page] 250

which however but imperfectly represents the observations. The causes which regulate the amount of rain in different latitudes, have been well pointed out by Dr. Anderson in the Essay on Hygrometry, before alluded to*.

A less explicable variation takes place in the fall of rain at different heights. And here a distinction, not always enough attended to, must be pointed out. The quantity of rain which falls on high grounds exceeds that at the level of the sea; but the amount at stations abruptly elevated above the surface of the earth diminishes as we ascend. For example, at Kinfauns Castle, Perthshire, by a mean of five years, 25·66 inches of rain fell; whilst on a hill in the neighbourhood, 600 feet higher, no less than 41·49 inches were collected by a mean of the same period†. On the other hand, at Paris, whilst 56·37 centimetres of rain fall in the court of the Observatory, according to Arago, only 50·47 fall on the tower at a vertical height of 28 metres. The former fact may readily be explained by the influence of a hilly country in retaining clouds and vapours; but the latter seems yet to have met with no satisfactory explanation, nor has any theory having even novelty to recommend it been recently proposed. The interesting observations established at York Minster, at the suggestion of the British Association, and under the active superintendance of my friends Mr. Phillips and Mr. Gray, jun., will soon, I am certain, afford us valuable information on this curious subject.

The very interesting comparative registers kept at Geneva and at the Convent of the Grand St. Bernard, have not failed to illustrate the influence of a mountainous country on the fall of rain. From the results published in the Bibliotheque Universelle, it appears that the amount at the latter point is double that at the former‡. Mr. Dalton, in an interesting paper upon these observations, which has just appeared§, points out in a clear manner the influence of hot currents of air ascending by the surface of the ground into the colder strata which rest upon a mountainous country. The consequence is, that although neither the hot nor the cold air was accompanied with more moisture than could separately be maintained in an elastic state, when the mixture takes place, the arithmetical mean of the quantities of vapour cannot be supported in an elastic state at the arithmetical mean of the temperatures, since we have seen that the weights of vapour which can exist in a given space, increase nearly in a geometrical ratio when the temperatures follow an arithmetical one.

* Edinburgh Encyclopœdia, vol. xi.

Ibid, Art. METEOROLOGY.

Bibliotheque Universelle, Mars 1828.

§ Manchester Transactions, New Scries, v. 233.

[page] 251

At Geneva by a mean of 32 years the annual fall of rain is 30·7 inches; at the Grand St. Bernard by a mean of 12 years it is 60·05 inches.

The variation in the amount of rain with the seasons follows in a great measure the same law, founded on hygrometric principles, which causes the difference in different latitudes. The greatest quantity falls in summer, the least in winter. The influence of the lunar periods has also met with some attention. The popular belief of the influence of the moon upon the weather is probably too strong and too universal to be totally without foundation. At one time I attended a good deal to the subject, and my observations led me to believe that there was some real connexion between the lunar phases and the weather. The old writings of La Cotte and Toaldo contain some curious observations on this subject, which has more lately been resumed by M. Flaugergues, who has observed the weather at Viviers with great assiduity for a quarter of a century. He has marked the number of rainy days corresponding to the lunar phases, and he finds them at a maximum at the first quadrature, and at a minimum at the last*. This agrees pretty nearly with his corresponding observations on the height of the barometer which we have already recorded.

A similar question to that which has been put in every other branch of Meteorology, whether there is any secular variation, —has been asked in the case of rain; and we are quite as unable as in the other instances to afford any satisfactory reply to it. There are several causes which may tend to change the amount of rain on a particular spot without forming part of any general law; among such changes will be found the destruction or the planting of forests, the inclosure and drainage of land, and the increase of habitations. M. Arago has shown† that the fall of rain at Paris has not altered sensibly for 130 years; and in order to show that the conclusion drawn by M. Flaugergues at Viviers, that the amount of rain is on the increase, cannot be a general one, he has quoted the case of Marseilles, where the amount of rain appears to have undergone a striking decrease in 50 years. M. Arago justly observes that it is very difficult to know how many years of observation are necessary to get a mean value of the fall of rain, the amount being extremely variable: thus at Milan, where an increase of rain has been thought to be decidedly proved by observations for 54 years, the extremes of the annual results between 1791 and 1817 were 24·7 and 58·9 inches‡.

* Bibliotheque Universelle, xl. 283.

Annuaire, 1824.

Annuaire, 1825, p. 155.

[page] 252

It is worthy of remark that notwithstanding the enormous annual fall of rain at the equator, particular instances of a great depth of rain in a short time have occurred (though rarely) in Europe, which probably have seldom been equalled by authentic observations in any part of the globe. At Geneva, on the 25th October 1822, there fell thirty inches of rain in one day*. An example equally extraordinary has recently been quoted by M. Arago, and which is perfectly authentic. At Joyeuse in the department of the Ardêche, on the 9th October 1827, there fell 29 inches 3 lines French measure (above 31 inches English) of rain in 22 hours†.

The subject of ATMOSPHERICAL ELECTRICITY excited in the middle of the last century an unexampled degree of interest in consequence of the fine discoveries of Franklin; and the application of thunder-rods produced a more vehement spirit of discussion among all classes than is usually to be met with on any purely scientific question. This excitement was naturally succeeded by a degree of apathy; and it must be admitted, that whilst every department of the noble science of electricity has been illustrated with triumphant success by Coulomb, Davy, Oersted, Faraday, and many others, its application to Meteorology has been strangely neglected, and in fact, on this important subject almost everything has yet to be done. On the general subject of atmospherical electricity, the principal contributions which we have to notice are those of M. Pouillet, to whom we owe some very interesting experiments in electrical science.

One great question in this subject is the source of the vast amount of electricity which seems, as it were, perpetually created in the atmosphere, and which, notwithstanding the constant recombinations which are going forward, remains sensible, according to the experiments of Lemonier, Saussure, and others, during the most steady and cloudless weather. M. Pouillet has very happily shown two causes in constant operation which create this abundant supply‡. The first of these is vegetation. M. Pouillet has proved by direct experiment that the combination of oxygen with the materials of living plants, is a constant source of electricity; and he has shown that a surface of 100 square metres in full vegetation disengages

* Pouillet, Elemens de Physique, ii. 758.

Annales de Chimie, xxxvi. There are some interesting collections relative to the fall of rain at different places, in Schouw's Specimen Geographiœ Physicœ Comparativœ 4to, 1828.

Annales de Chimie et de Physique, 1827. See also his Elemens de Phyue, liv. ix. chap. 5.

[page] 253

in the course of a day as much vitreous electricity as would charge a powerful battery. The second source is evaporation. The experiments of M. Pouillet went to prove the unexpected fact, that the conversion of pure water into vapour excites no electric tension. As however this applies only to water and other fluids in a state chemically pure, it makes no difference in the efficacy of this change as productive of atmospheric electricity. It is needless to observe how extensive and powerful must be the result of this action.

It is hence easy to conceive how the electricity, produced by these and other sources, must vary in different climates, seasons and localities, and at different heights in the atmosphere. The general principle of the formation of electrical clouds, and the production of thunder and lightning, is easily apprehended; but the fact of our almost total ignorance of any one step of the process cannot be disguised, and, as M. Pouillet frankly admits, "il faut avouer si la principe de la formation des nuages orageux ne présente pas des difficultés, les applications en présentent, parceque nous n'avons pas assez de données sur la formation des nuages elle-mêmes."

It is generally believed that in fine weather the electricity of the air is positive, and increases in intensity as we ascend. Upon these points however observers are by no means agreed, and the subject opens a wide field for experiment. From the observations of M. Schubler, it would appear that in the climate of Europe the electricity of precipitations is more frequently negative than positive in the ratio of 155: 100, but the mean intensity of the positive electricity is greater than that of the negative in the ratio of 69: 43*.

There is a subject intimately connected with electricity which we are unwilling totally to pass over in this place, although little has lately been added to our knowledge upon it, because we think that it has not excited the attention in this country which it deserves;—we mean the phænomenon of Hail. The difficulty of accounting for the retention of masses of ice in the free atmosphere till they attain in some cases the diameter of several inches, is certainly very great. Perhaps no hypothesis more satisfactory, certainly none more ingenious, has followed that of Volta, who conceived from the highly electric condition of the atmosphere, almost universally attending the production of hail, that the frozen masses were kept in a state of reciprocating motion between two clouds oppositely charged with electricity, until the increase of the mass rendered the force of gravity predominant, or the electric tension of the clouds was exhausted by mutual reaction. It

* Bibliotheque Universelle, xlii. 203.

[page] 254

were easy to multiply objections to this hypothesis, and some of the reasonings of the author relative to the production of cold are almost certainly erroneous; but at the present moment it would be difficult to point out any explanation more plausible*.

From the rarity of the occurrence of hail storms in this country, the subject has met with little attention compared to what it has received in most parts of the Continent†. In our Indian territories, however, the finest opportunities occur for the investigation of facts connected with the subject. Dr. Turnbull Christie has recently published, on this subject‡, a short notice in reply to some theoretical views of Prof. Olmsted, an active American meteorologist§. We hope that among the scientific objects which engage Dr. Christie's attention since his recent return to India, this will not be forgotten.

This question has appeared of so much importance and interest on the Continent, that the Academy of Sciences at Paris has recently proposed the theory of hail as the subject of a prize memoir.

Before concluding this Report, I am anxious to advert to the very interesting subject of the AURORA BOREALIS,—one which appears intimately connected with the science of Electricity, and upon which we cannot but hope soon to acquire new and extended views.

I shall not dwell for a moment upon older observations, but proceed to state that Mr. Dalton has been led, from numerous and very interesting observations which he has collected upon auroral arches, to conclude that their average height above the surface of the earth is about 100 miles‖,—a conclusion not differing much from what he had long before been led to¶. The frequent occurrence of these beautiful phænomena of late years, has rendered them an object of general observation, and many descriptions have been published by different authors in the periodical works of the day. The one of which Mr. Dalton deduced the height in the most satisfactory manner, was that of the 29th March 1826. The most striking examples which have since occurred, were on the 29th Sept. 1828, and the 7th Jan. 1831**. The last is perhaps the most extensively observed on record.

* See on this subject a paper by M. Arago in the Annuaire for 1828.

† See however an interesting account, by Mr. Neill, of a remarkable hail storm which occurred in Orkney some years since. Edinburgh Transactions, ix. 187.

Edinburgh New Philosophical Journal.

§ Published in Silliman's Journal, 1830.

Phil. Trans. 1828, p. 291.

¶ In his Meteorological Essays.

** On the last may be consulted papers by Mr. Christie, Mr. Harris, and Prof. Moll, in the Royal Institution Journal, N.S. vol. i. On that of 1828, Mr. Gilbert, Phil. Mag. N.S. iv. 453; Capt. Kater, Ibid. 337; Mr. Harvey, Edin. Journal of Science, x. 146.

[page] 255

The opinions of Mr. Dalton on the height of the aurora have not been received without contest. Mr. Farquharson of Alford in Aberdeenshire has published his views upon this subject in a paper containing many curious facts, and from which he draws the conclusion that the aurora is not elevated above one or two miles*. His estimate however appears to me to be founded on a species of observations somewhat vague, and by no means comparable, as scientific deductions, to the trigonometrical measures of Mr. Dalton. Indeed, the fact of the immense distances of points at which these arches have been seen at the same instant, is alone sufficient to throw great doubt upon any theory which assigns to them a low position in the atmosphere. So strong is this objection, that Mr. Farquharson has been obliged to suppose that the different observers were viewing different parallel auroral bands,—a supposition surrounded with difficulties. The only striking actually observed fact appearing to demonstrate that the aurora sometimes approaches the surface of the earth, is that, related by Captain Parry, of a beam of the aurora appearing to shoot down between the observers and a rising ground only 3000 feet off†. This very extraordinary and unique observation, certainly appears to me more attributable to an optical illusion, than as fitted to become the basis of extensive induction. At least it is very conceivable that a beam of the aurora shooting downwards, as is described, with all the brilliancy peculiar to that meteor in arctic climates, might, as it passed behind an eminence, appear from the quickness of its motion to continue its former course, and shoot across the obstacle which actually intercepted it from view. Such at least seems to me a highly natural explanation:—be this as it may, a single observation cannot, in the face of all those to the contrary, limit the bounds of the aurora to the lower strata of the atmosphere.

I readily admit however that some phænomena of electrically illuminated clouds, such as I remember particularly to have observed on the 10th Sept. 1827‡, are of difficult explanation. Should it however be admitted that these were "clouds highly electrified," as I have stated in the memorandum just referred to, I would beg to draw a very broad line between these and true auroral nebulæ or arches. The evidence which convinced me that these were truly clouds, was especially the fact that "they

* Phil. Trans. 1829, p. 105.

† Parry's Third Voyage.

Edinb. Journal of Science, ix. 138.

[page] 256

obscured bright stars," for the auroral arches which I have observed, generally allowed even minute stars to be seen through their mass; but this I admit to be a question of degree. Very recently an interesting article by Prof. Jameson has been published in the new edition of the Encyclopœdia Britannica, in which the views of Mr. Farquharson on this point are strongly supported*.

In the paper just alluded to, Mr. Farquharson justly observes, that the motion of the auroral arches is from north to south, or rather N.W. to S.E.: he adds, that he never heard of an arch observed whilst low in the north, and traced in its course up to the zenith, and thence southward. I beg to refer to some circumstantial details of an arch observed by myself, 21st January, 1826†, traced almost from its origin in the north till it disappeared close to the southern horizon. This arch had also the peculiarity of moving, not in the direction of the magnetic meridian, but from N.E. to S.W., and diametrically against the wind.

Mr. Potter has recently given some interesting views regarding the height of the aurora, and pointed out a method by which (certain postulates being admitted,) its height may be calculated from observations at one station‡. His results coincide generally with those of Mr. Dalton.

The influence of the aurora upon the magnetic needle has for some years afforded a fertile subject for discussion; and it is to be regretted that no continued series of observations has been undertaken in Britain, adequate to the solution of the question, or indeed materially contributing to our knowledge of the state of the earth's magnetism. M. Arago of Paris gave an account, some years ago, of the connexion, which his observations established, between the phænomenon of the aurora and the irregular motions of the variation and dipping needles§. Prof. Hansteen coincided in the truth of this result; and added the observed anomalies of the magnetic intensity under the same circumstances‖. M. Arago then remarked, that this variation of the horizontal intensity might only arise from the irregularity occasioned in the dip, of which the former is a function¶. Meanwhile, M. Arago's general conclusions were warmly opposed by Dr. Brewster, who considered the fact as not sufficiently established by such obser-

* Encyc. Brit., Art. AURORA BOREALIS.

Edinburgh Journal of Science, ix. 129.

§ Ibid. N. S. v. 23.

§ Annales de Chimie, 1825.

Jahrbuch der Chemie und Physik, xviii. 353.

Annales de Chimie, 1827.

[page] 257

vations as were published*. In support of his views he quotes the observations of Parry and Foster, particularly on the occasion of the luminous beam to which we have just alluded, which appeared to exercise no energy on the needle. Unaccountable, however, as the discrepancy may be between travellers so well qualified to judge, and under such favourable circumstances, Capt. Franklin and Dr. Richardson detected the most decided proofs of magnetic action of the aurora†. It must be obvious therefore, that independent of the difficulties of the observation, and the delicacy of the instruments required, there must be some innate source of difficulty in the subject.

Mr. Farquharson has endeavoured to point out some explanation of these anomalies in a recent Memoir in the Philosophical Transactions‡; where he states that from his observations on the variations of magnetic intensity with auroral phænomena, and also of the dip and variation of the needle, he has found the effect to be a maximum when the streamers reach the plane of the dip, or when they pass through that region of the heavens to which the south pole of the dipping-needle points. Dr. Richardson had remarked that the effect on the needle was greatest when the streamers passed to the south of the zenith. This observation of Mr. Farquharson must therefore be considered one of importance, though it does not quite explain some anomalies in the circumstances of the observations, especially of those not made in high latitudes. As far as the observations of Mr. Farquharson himself go, they confirm the results obtained on the Continent‖.

In almost every part of the continents bordering on the arctic circle have observations to the same effect been recently made. Prof. Kupffer has observed the influence of the aurora in the most striking manner at St. Petersburg, Nicolajew, and Kasan; and the results of contemporary observations at these points are well worth consultation§. Prof. Hansteen has detected the same at Tornea in Lapland and M. Riess of Berlin has lately observed with care the influence of the aurora upon the mag-

* Edinburgh Journal of Science, viii. 189.

† See Franklin's Second Journey; the Edinb. New Phil. Journal; and the Bulletin des Sciences Mathénatiques, xi. 293. These observations have given rise to some criticisms by Dr. Brewster, which have been replied to by Mr. Christie in the Journal of the Royal Institution.

‡ For 1830, p. 97.

‖ For an abstract of Mr. Farquharson's papers, see the Edinb. New Phil. Journal, vi. 392; and the Encyclopœdia Britannica, (New Edit) art AURORA BOREALIS.

§ Poggendorff's Annalen, 1831.

Ibid. 1827; and Bulletin des Sci. Math. ix. 252.

R

[page] 258

netic intensity*. In America also, experiments have led to the same result†.

On the occasion of the great aurora of the 7th January 1831, M. Arago observed the magnetic needle powerfully affected, whilst Mr. Sturgeon of Woolwich could not notice it at all‡: on the 19th April 1831, Mr. Christie of Woolwich, in company with Mr. Faraday, observed the most unequivocal signs of auroral action§. This observation, made by two philosophers perfectly habituated to such experiments, must be considered probably the most complete evidence yet obtained in this country. On the whole, it seems undeniable that the aurora borealis, frequently at least, exercises the most marked action on the magnetic needle, with regard to variation, dip, and intensity. The circumstances under which it does not take place, require however the most careful scrutiny, and we hope that Mr. Farquharson will pursue unremittingly his observations. Unfavourable as is the sky of Britain for many kinds of experiment, her geographical position is in other respects highly important as concerns scientific undertakings. Among these especially rank Meteorology and Magnetism; and it were deeply to be desired that she should lead the way in the prosecution of these too much neglected sciences. There can be no reason why experiments should not be as well conducted here as in the cabinet of M. Arago; and when Baron Humboldt boasted to the French Academy of the wide distribution of his "maisons magnetiques," or magnetic observatories, from Paris, the centre of civilization, to the wilds of Siberia, and to Pekin itself, whose gates have been so long shut against the approaches of science, —it is a humiliating fact that he could not with truth have mentioned Britain as possessing a solitary establishment of this description, either within her own limits, or probably even in the range of her much more widely extended dependencies.

It may not be superfluous to add in conclusion, should more errors (especially those of omission,) be found in the preceding Report than might seem to be inseparable from the nature of the work,—that it has been drawn up within a very limited space of time, and under the pressure of a variety of preparations for an extended scientific tour on the Continent.

June 1832.

* Communicated to the Academy at Paris by Baron Humboldt, 10th Oct. 1831.

Silliman's Journal, 1828.

Philosophical Magazine, N.S. ix. 151.

§ Journal of the Royal Institution, Dec. 1831. p. 271.

[page] 259

Report on the present State of our Knowledge of the Science of Radiant Heat. By the Rev. BADEN POWELL, M.A. F.R.S., Savilian Professor of Geometry in the University of Oxford.

IN attempting to give a condensed account of the present state of our knowledge of the science of Radiant Heat, it appears to me that I shall be best consulting the design of such a Report by offering, in as brief a form as possible, a sketch of what has been formerly done in this department; and thence proceeding to a more detailed survey of what is now doing. And we shall proceed with greater clearness if we distinguish the several different departments into which the subject divides itself, agreeably to certain known distinctions in the properties and species of heat acting under peculiar circumstances. All these have been too commonly confounded together under the general and vague name of Radiant Heat, whence not unfrequently the most erroneous views have resulted. By distributing our subject, however, under the few well-marked divisions which the scanty results of observation as yet supply, we shall at once secure perspicuity in our views, and be treating the subject in a way most accordant with the inductive process; which, it must be distinctly avowed, has not yet enabled us to advance to any such comprehensive knowledge of the facts as can warrant us in generalizing them, or in ascribing to a common principle the radiation of heat from a mass of hot water, from a flame, and from the sun.

We shall take each of these principal divisions separately, and under each shall consider what is known in reference to those properties to which experiment has been directed.

DIVISION I.

Radiation of heat from hot bodies below the temperature of luminosity.

a.) Radiation (or communication of heat to sensible distances,) is distinct from its conveyance by conduction through the air; since,

1.) It takes place perpendicularly downwards:

2.) Only in elastic media.

The relative cooling in different media is seen in the following experiments. (Rumford's Essays, ii. 425; Torricelli; Murray's Chem. i. 328.)

R 2

[page] 260

Thermometer cooled from 212° to 32° Fahrenheit:

In Vacuo in 10m 5sec
Air 7 3
Water 1 5
Mercury 0 36

Dulong and Petit, in their elaborate researches on the cooling of bodies, have investigated the law of cooling in the most perfect vacuum they could form: but they admit that there was always a minute portion of air present. The radiation therefore of heat in an absolute vacuum is by no means conclusively established. (See Annals of Phil. vol. xiii. p. 241.)

3.) Professor Leslie ascertained, That the effect from a mass of given size is nearly proportional to the angle which it subtends at the thermometer; and that the heat suffers little or no diminution in its passage through the air.

The radiation is most copious in the direction perpendicular to a plane surface of the hot mass, and is proportional to the sine of its inclination to the direction of the thermometer. (Inquiry into the Nature and Propagation of Heat, p. 51, &c.)

For the same position the effect is proportional to the excess of temperature of the hot body above that of the air.

4.) Pictet made an attempt to estimate the velocity with which heat radiates, by means of concave reflectors at 69 feet distance. The effect on the focal thermometer was absolutely instantaneous. (Essais de Phys.)

b.) Reflexion of simple heat from nonluminous hot bodies.

1.) The general principles are established by Professor Leslie. (Inquiry, pp. 14, 51.)

2.) He shows that the quantity of heat reflected is proportional to the sine of incidence on a plane surface.

3.) It is affected by the polish of the surface. (Leslie, Inquiry, pp. 81, 20, 98, 106.)

4.) The most exact experiments are those made with conjugate concave reflectors; a ball of iron below luminosity in one focus, a thermometer in the other: a glass of boiling water may be substituted for the iron ball. In either case a great effect is produced in the opposite focus, though little out of it. (Saussure, Voyages, t. iv. p. 120; Sir W. Herschel, Phil. Trans. 1803, p. 305.)

Professor Leslie made extensive use of reflectors, but ob-

[page] 261

served that there was a very considerable degree of aberration in the focus from an exact position; considerably nearer to the reflector than the true focus, the effect continued undiminished. (Inquiry, p. 64.)

5.) Alleged reflexion of cold.

An account of the earliest experiments will be found in the Memoirs of the Florentine Academy, (Waller's Transl. p. 103; also Gærtner, 1781.)

Pictet with conjugate reflectors found the thermometer sink when ice was in the opposite focus. (Essais de Phys. p. 82.)

Count Rumford employed a tube, a frustrum of a cone, open at both ends; placing ice at the small end, the thermometer at the large end sunk very little. The ice being at the small end, the thermometer at the large end fell considerably. Rays reflected by the inside of the tube from the body at the large end, would be concentrated on that at the other.

6.) M. Prevost (Essai sur la Calorique rayonnant, Geneva 1809, and Recherches sur la Chaleur, p. 15,) proposes a theory of radiation, that heat is a discrete fluid every particle of which moves in a straight line, and such motions are constantly taking place in all directions, whether there be more or less heat present. Hence all bodies, whether of a higher or lower temperature, are supposed to be continually radiating heat; and this going on mutually tends to bring them all to an equilibrium of temperature.

On this theory explanations are given of the apparent radiation of cold.

The thermometer in the conjugate focus, when nothing is in the other, remains stationary, because the rays reflected from all the surrounding space so as to cross at the focus of the opposite mirror, and be reflected in a parallel state to the other, and thence on to the thermometer in the focus, are exactly equivalent to those which the thermometer radiates. But when a mass of ice is placed in the opposite focus, it intercepts and absorbs a portion of the rays which would otherwise have fallen on the first mirror, and so have reached the thermometer, which in consequence radiates more than it receives, and therefore sinks.

A similar explanation applies to Count Rumford's experiment. (See Thomson On Heat, &c. p. 163.)

In the Quarterly Journal of Science (June 1830, p. 378,) some observations are given on this subject, and an explanation offered, which, though very ingenious, appears somewhat complicated.

[page] 262

It may not be improper to observe, that if the above be a correct view of Prevost's theory, it can hardly be conceived as otherwise than partially hypothetical. The idea, viz. that bodies even of a lower temperature than those about them actually give out a small degree of heat, is extremely difficult to conceive; and it does not appear absolutely essential to the explanation of the facts.

Without reference to any theory, I venture to propose the following as the simple experimental law:

All bodies of unequal temperature tend to become of equal temperature; if in contact—by conduction; if at sensible distances—by radiation, of the excess of heat: and (in the latter case) whether the radiation reach the cooler body directly or by an intervening reflexion.

This appears sufficient to include the facts of Pictet's and Rumford's experiments.

7.) Alleged polarization of simple heat by reflexion.

Mons. J. E. Berard (Mémoire sur les Propriétés des différentes Espèces de Rayons qu'on peut séparer au moyen du Prisme de la Lumière solaire," Mém. de la Société d'Arcueil, Paris 1817, tome iii. See also Annals of Phil. O.S.ii. 164; Biot, Traité de Phys. iv.) tried experiments for the polarization of heat. His apparatus was the same as Malus's, having the axis of revolution vertical; but no precautions of screening, &c. are mentioned. He used an air thermometer containing a bubble of alcohol in the tube, in the focus of a reflector moving round along with the second glass: a ball of copper about two inches in diameter was in the focus of a reflector, placed in the position for polarization of light. (His experiments on heat with light will be referred to in another place.) He tried the effect with the metal heated below luminosity, and assured himself that there was a difference in the degree of heat reflected in the two rectangular azimuths of the second glass.

I have attempted to repeat these experiments with the same kind of apparatus, carefully screened and arranged with the tube horizontal; but could produce no diminution in the proper position. (Edinb. Journal of Science, N.S. vol. x. p. 207.)

I also tried the experiment with a delicate mercurial thermometer, comparing this case with others (referred to in their proper place), in which light accompanied the heat; but in the former could detect no difference in a long series of repetitions.

The total effect is in all cases extremely small, and the disturbing causes considerable, especially the heating of the

[page] 263

glasses, &c. The whole experiment was very unsatisfactory. Edinb. Journal of Science, N.S. vol. vi. p. 297.)

c.) Effect of the nature of surfaces on the emission of simple heat.

1.) Count Rumford (Nicholson's Journal, ix. 60,) employed two similar vessels of hot water of the same temperature; one naked, the other coated with linen, glue, black or white paint, or smoked with a candle: the results were,

Naked vessel cooled 10° in 55m
Coated ————— 10 — 361/2

Mr. Murray supposes a relation between radiating and conducting powers. (System of Chem. i. 326, 334. See Phil. Trans. 1804, p. 90, &c.)

2.) The most complete investigation of this and other parts of the subject has been made by Professor Leslie in his Inquiry into the Nature and Propagation of Heat, 1804.

He first used hot water in a globe of tin, in which the inserted thermometer fell a given quantity, with the tin bright, in 156m; with the tin coated with lamp black, in 81m.

The difference was greatest in still air, and diminished with the violence of its motion:

Time of Cooling.
Wind. Bright. Blackened.
Gentle. 44m 35m
Strong 23 201/4
Violent 91/2 9

Hence the effect is different from conduction by air.

3.) The most exact series of experiments was that in which he used conjugate reflectors, a differential thermometer having one bulb in the focus, and a cubical tin canister of hot water (the temperature of which was seen by the projecting stem of a thermometer), and each side of which could be coated with a different substance, and presented successively towards the reflector.

The following results collected together afford the best view of the general nature of the conclusions relative to the influence of the state of the surface on the radiation of heat. (Inquiry, pp. 81, 90, 110.)

Lampblack 100*
Water (estimated) 100
Writing-paper 98*

[page] 264

Rosin 96
Sealing wax 95
Crown glass 90
China ink 88*
Ice 85
Minium 80*
Isinglass 80
Plumbago 75
Thick film of oil 59
Film of jelly 54†
Thinner film of oil 51†
Tarnished lead 45
Film of jelly, (1/4 of former quantity) 38
Tin scratched with sand-paper 22
Mercury 20
Clean lead 19
Polished iron 15
Polished tin, gold, silver, copper 12
Thin lamina of gold, silver, or copper leaf on glass 12‡

* From comparing the results marked, it appears that the effect follows no relation to colour. Softness probably tends to increase radiation.

† Thickness of film increased beyond a certain limit does not increase the radiation.

‡ The tenuity is not sufficient to produce any diminution of effect, which probably would take place if thinner films could be applied.

4.) The effect of the surface on radiation is beautifully exemplified in the laws which regulate the formation of dew as developed by Dr. Wells. (Essay on Dew, 1814. See also Dufay, Mem. Paris 1736, p. 352; and Harvey on Dew, Quarterly Journ. of Science, No. 33; Edinb. Journ. of Science, i. 161.)

5.) Dr. Ritchie (Edinb. Phil. Journ. xxiii. 15,) explains his theory of the mode in which the radiating power of surfaces is increased by making them rough, or furrowing, &c. He contends that it is not owing to the increase of surface, but to the quantity of heat reflected by the sides of the furrows.

He adopts the hypothesis of material caloric, and that its molecules are mutually repulsive.

The effect of surface is an essential distinction between radiation and conduction by air; the latter being shown by Dulong and Petit to be absolutely independent of the nature of the surface. (Annals of Phil. xiii. 322.)

[page] 265

d.) Effect of surface on the absorption of heat from non-luminous hot bodies.

1.) De Saussure and Pictet, with the apparatus before described, found that the thermometer rose in two minutes,

Plain 41/8° Fahr.
Blackened 31/8

2.) By the same apparatus as before described, Prof. Leslie found that on coating the bulb of the thermometer with the different substances, the absorptive powerwas very nearly in the same proportion as the radiative; and by making the same modifications in the surface of the reflector, he found that reflective power is inversely as the radiative or absorptive. (Inquiry, pp. 19, 81, 98.) He also gives a very precise set of experiments on the effect of coatings of jelly of increasing thicknesses. (p. 106.)

3.) Dr. Ritchie has devised a very elegant mode of showing that the absorptive power of surfaces is precisely proportional to their radiating power. (Royal Inst. Journ. vol. v. p. 305.)

The instrument consists of a large differential thermometer, whose bulbs are chambers of considerable size, presenting large and equal plane surfaces on the sides which are towards each other: of these one is plain or polished, the other coated. Midway between them is placed a canister having equal plane surfaces facing each of the former respectively, and one polished, the other coated with the same pigment as before; this canister is filled with hot water, and is capable of turning on a vertical axis; thus the coated surface of the canister can be turned to the coated bulb or to the polished: in the former case a great effect is produced on the coated bulb, and a very small effect on the plain: in the second case the better radiating surface is directed to the worse absorptive one, and the worse radiating to the more absorptive, and the liquid in the tube remains perfectly stationary: the exact equality, therefore, of the absorptive and radiating powers is established. The whole is on a large scale, and can be exhibited to a Class.

4.) The most recent and curious researches on this part of the subject (and extending, as we shall see, to other parts also,) are those of MM. Nobili and Melloni. (Annales de Chimie, Oct. 1831; Recherches sur plusieurs Phénomènes calorifiques, &c.)

The authors commence by describing their thermo-multiplier, by the aid of which their researches were carried on. This

[page] 266

consists in a thermo-electric combination, susceptible of excitation from the feeblest conceivable application of heat, and connected with a delicate galvanometer, which gives a measure of the effect produced, and consequently of the heat.

The pile is in a case coated with the smoke of a flame when used for radiant heat, but left naked wven for heat of temperature, on account (as they observe) of the bad conducting quality of this coating.

They applied this instrument to the examination of the different reflecting, absorbing, and radiating powers of surfaces.

They confirmed in general the results of Leslie and others already mentioned. They found that polish augments the reflecting power much less than usually supposed. Non-metallic substances possess scarcely any reflecting power, whatever be the state of their surfaces.

They examined the absorptive power of different substances, taking laminæ of equal thickness and similarly fixed, &c.: these having been heated for a few minutes in the rays of the sun, were placed in pairs on apertures at the opposite sides of the thermo-multiplier, and in this way the order of their absorptive powers was considered to be obtained by the degree of heat they respectively radiated; and the results were, that the effect increased by blackness of colour and with roughness of surfaces. Also the following surfaces were in this order,—silk, wool, cotton, flax, hemp, (all white,) which is the inverse of their conducting powers. In like manner, with metals of nearly the same colour and polish, the order was—copper, silver, gold, steel, iron, tin, lead, exactly in the inverse order of the conducting powers;—the same with several woods and minerals.

On these experiments I must remark, that the heat acquired from the sun's rays is so obviously dependent on colour, that it is astonishing that any experimenter should adopt this as affording any ground for making conclusions respecting the comparative absorbing or radiating powers for heat in general. The later results, when the surfaces were all of the same colour, are extremely important. Supposing they all acquired the same degree of solar heat which was thus converted into heat of temperature, and then radiated from the surfaces as simple heat, the real conclusion established is, that the RADIATING powers of surfaces for simple heat are in the inverse order of their conducting powers.

[page] 267

e.) Effect of screens on heat from nonluminous hot bodies.

1.) Pictet found a difference in the interceptive effect, according as the plain or the silvered side of a glass screen was towards the source of heat.

Towards Hot Body. Ratio of Effects on Thermometer.
Glass 5
Amalgam 35
Amalgam, blackened 92
Amalgam removed,—glass blackened 180

(Essai, &c., p. 72.)

2.) He tried to refract simple heat, without effect.

Sir W. Herschel tried with a lens, and supposed it effected: this has been refuted by Sir D. Brewster. (Vide infra; Phil. Trans. 1800, Part II. No. 15. Exp. 19, 20.)

3.) Prof. Leslie's experiments on screens are perhaps the most valuable portion of his inquiry.

He found the effect of a screen increase rapidly with its distance from the source (p. 28), and less so with its thickness (p. 38).

Different substances appear to have a different interceptive power; but this upon examination appears always to be dependent on their conducting power, and the absorptive nature of their surface jointly.

The most decisive experiment on this point was that made with two panes of glass, each having one side coated with tin foil: according as the plain or coated sides were placed in the contact, the compound screen had a greater or less apparent interceptive power; that is, a greater or a less power of absorbing and subsequently radiating the heat. Again, either might be used separately, or the two at an interval. (p. 35.)

4.) Prevost concluded that a certain portion of heat is directly transmitted through transparent screens, by employing moveable screens which continually presented a fresh surface, so that it was supposed all communication of heat and conveyance by way of secondary radiation would be prevented.

But it must be considered that it is impossible to prevent entirely any portion of a screen in the most rapid motion from acquiring heat:—no such experiments therefore can be strictly conclusive.

Dr. Ritchie tried experiments with the same view, by means of a film of liquid adhering to threads stretched across a frame

[page] 268

continually renewed. (Phil. Trans. 1827, Part II. p. 141.) But to this a similar objection must apply.

5.) The results of Prof. Leslie do not apply to temperatures above those of boiling water.

This extension of the inquiry formed the subject of the researches of De la Roche. The complete account of these is given in its proper place; at present we have to consider them only as far as relates to bodies below luminosity. He tried the effect of a screen of glass, first transparent, and then with one surface blackened, on the heat radiating from mercury at 180° centig. and at 346° when it was boiling. (Biot, Traité de Phys. iv.640.)

The results were as follows:

Rise of focal thermometer (centig.) in 1m.

No Screen. Transparent Screen. Blackened Screen.
Mercury at 180° 3°·94 0°·22 0°·07
— at 346 16 ·33 1 ·36 0 ·17

He hence infers a partial transmission of heat at these high temperatures; and the more so, viewing these results in connexion with the rest of the subsequent series (considered in another place).

These are the only ones of his experiments referring really to simple radiant heat; and the inference of an actual transmission in the way of direct radiation, is open to several objections.

6.) The blackened screen causes a greater diminution of heat than the transparent, and it was therefore inferred that a portion of heat radiates freely through the transparent screen, and is stopped by the opake one: but there are several circumstances which show that this is not a necessary conclusion.

The coating was towards the source of heat, and rendered this screen more absorptive of heat where exposed to it, that is, at its central part,—and a better radiator towards the edges without the area of the incident rays; so that it radiated its heat most copiously on the side away from the thermometer. With the plain screen there was no such tendency to radiate more on one side than on the other; and hence the greater effect on the thermometer.

This explanation I suggested in the Annals of Philosophy, xlv. 181.

Some observations bearing upon this subject, occur in Sir David Brewster's elaborate paper on "New Properties of Heat," &c. in the Phil. Trans. 1816, Part I. His 40th propo-

[page] 269

sition is directed to prove that radiant heat is not susceptible of refraction, and is incapable of permeating glass, like the luminous rays. The truth of this is demonstratively shown from the curious properties examined in the previous parts of the paper, and shown to be communicated by heat to glass; and by the progress of which, the passage of the heat through the glass may be as clearly traced as if the heat itself were visible.

He applies this conclusion to the experiment of Sir Wm. Herschel, in which the concentration of simple heat by a lens appears to be proved. The thermometer must have received the heat radiated by the lens itself; and from the circumstance that the edges will cool first, the most copious radiation of heat will be in the direction of the axis.

In connexion with the same point he also examines the conclusions of MM. De la Roche and Prevost, and observes: "The ingenious experiments of M. Prevost of Geneva, and the more recent ones of M. De la Roche, have been considered as establishing the permeability of glass to radiant heat. M. Prevost employed moveable screens of glass, and renewed them continually, in order that the result he obtained might not be ascribed to the heating of the screen: but such is the rapidity with which heat is propagated through a thin plate of glass, that it is extremely difficult, if not impossible, to observe the state of the thermometer before it has been affected by the secondary radiation from the screen.

"The method employed by M. De la Roche of observing the difference of effect when a blackened glass screen and a transparent one were made successively to intercept the radiant heat, is liable to an obvious error. The radiant heat would find a quicker passage through the transparent screen, and therefore, the difference of effect was not due to the transmitted heat, but to the heat radiating from the anterior surface. The truth contained in M. De la Roche's fifth proposition, is almost a demonstration of the fallacy of all those that precede it. He found that a thick plate of glass, though as much or more permeable to light than a thin glass of worse quality, allowed a much smaller quantity of radiant heat to pass. If he had employed very thick plates of the purest flint glass, or thick masses of fluid that have the power of transmitting light copiously, he would have found that not a single particle of heat was capable of passing directly through transparent media."

7.) I have further attempted a direct experimental examination of the question in a paper inserted in the Phil. Trans. 1826, Part III. p. 372.

[page] 270

The substance of my observations is as follows:

De la Roche found, that if radiant heat be intercepted by two transparent screens, the additional diminution of effect occasioned by the second, is proportionally much less than that produced by the first; and the same conclusion is extended to any number of screens. This was explained by the supposition that the heat in its passage through the first glass undergoes a certain modification, in some respects analogous to polarization, by which it is enabled to pass, with very little diminution, through the second and subsequent glasses.

In those cases where the source of heat is luminous, such phænomena would receive an obvious explanation on the principle investigated in my other paper. Vide infra.

But if the same effect is still observable below the point of luminosity, we must have recourse to some other principle of explanation. That deduced by De la Roche appears at least plausible; and though it should be considered proved, that, in general, heat is incapable of being radiated directly through glass, it perhaps would not necessarily follow, that it might not, under peculiar circumstances, have a power of doing so communicated to it. Though on the other hand it must be confessed, that in the present case some difficulty would attend such a supposition.

It certainly would not be easy to conceive such a property to be communicated to the heat, by the mere act of being conducted through the first glass. Again; a new property of heat is thus introduced, which, it must be conceded, is not absolutely and exclusively established.

It appeared to me therefore a point of some interest to examine, in the case of non-luminous heat,—in the first place, the accuracy of the fact; and secondly, if verified, whether there might not be circumstances observable in the conditions of the experiment by which it might be accounted for, without the necessity of supposing any peculiar property of heat, or a direct transmission even through the second glass.

My apparatus in following up this inquiry was similar to that described by M. De la Roche, and consisted of two tin reflectors;—in one focus the bulb of a thermometer coated with Indian ink, and in the other an iron ball two inches diameter, which was heated to redness, and then cooled till it ceased to be visibly red in the dark, at which point it was placed on its stand, and a thick screen withdrawn. The indications were observed, first, for the direct effect; secondly, with one glass screen interposed; and thirdly, with two. The temperature of the screens was observed by means of a small thermometer

[page] 271

attached to the face of each away from the ball, towards its central part; the bulb being kept in contact by the spring of a wire with which the thermometer was fastened.

The results are: 1st, That the additional diminution occasioned by the second screen, is proportionally much smaller than that occasioned by the first. Thus De la Roche's conclusion is shown to hold good, not only in the case of luminous, but also of non-luminous hot bodies; which is perhaps of consequence, as I believe doubt has been entertained respecting it; and it may be remarked, that here the greater thickness of the second screen would be against such a result. 2ndly, If the progress of the indications of the direct effect be followed, it appears that the rise in the first 30 seconds is the greatest, and that those in the subsequent periods gradually diminish. 3rdly, With one screen, the effect in the first period is equal to, or even less than those in the subsequent ones; and if we follow the temperature of the first screen, it appears to sustain a rapid increase at first, and afterwards continues gradually to rise till some time after the focal thermometer has become stationary. The progress of the focal thermometer exactly accords with what must be the heating effect of the screen as a source, viz. rising slowly at first as the screen acquires heat sufficient to supply it, and remaining stationary so long as the still increasing temperature of the screen could balance its loss of heat. 4thly, With two screens, there is no rise till the second half-minute, when it is not greater than in the next half, after which the thermometer becomes stationary; and this trifling effect exactly accords with what the temperature of the second screen should produce. It does not begin till the second screen has acquired a higher temperature, and it is stationary while the temperature of the screen continues to increase; and the temperature of the second screen is such as is clearly accounted for from the heating effect of the first. It does not begin to rise till after that of the first has risen; it continues stationary some time after the first has begun to cool, as the first screen did when the iron was cooling. But as in this case the source of heat was cooling during the whole time of the experiment, whilst in the other it was heating during the first part of the time, it follows, that a greater proportional temperature should be communicated to the second screen by the first, than to the first by the iron ball.

Other circumstances will partially cooperate in producing this effect,—as the greater proximity of the second screen to the thermometer; also more heat might be lost in communicating an equable temperature to the first screen from its central and

[page] 272

more heated part; whilst the heat would be thus more equally radiated to all parts of the second without such loss.

Thus it appears that the fact stated by M. De la Roche is fully substantiated; while on the other hand it is satisfactorily accounted for, without supposing any new property of heat, or any direct radiation through glass.

In some unpublished experiments of my own, I found upon observing the temperature acquired by a screen exposed to iron below luminosity, first plain, and then coated with Indian ink towards the source of heat, the thermometer being in contact at the central part on the outside, that it rose rather more on the plain, than on the coated screen.

8.) MM. Nobili and Melloni, in the Memoir before quoted, applied their instrument to estimate the effects of transparent screens. Over the thermo-multiplier were placed successively transparent screens of glass, sulphate of lime, mica, and of water, oil, alcohol, and nitric acid (inclosed between plates of glass?), and also of ice.

The source of heat was a ball of iron, heated to a point below, luminosity, suspended, or rather passed rapidly, at a certain distance above the screen.

The index indicated an instantaneous effect, greater or less in all cases except those of water and ice, in which none was produced, even when the iron was kept a longer time over the instrument, or even heated to redness, and the screen reduced in thickness.

9.) A set of experiments presenting some important results with respéct to the absorbing and radiating properties of surfaces, as well as the action of screens in air and in vacuo, are given by Mr. W. R. Fox, in the Phil. Mag. and Annals, New Series, No. 65, p. 245. A brief statement of the results is as follows:

A cylindrical tin vessel of hot oil with its surface polished, and another similar, painted black, had their times of cooling a certain number of degrees observed under a receiver first highly exhausted, and then full of air; the cylinders being respectively 1st exposed, and 2ndly inclosed in one and sometimes more tin cases with intervals; the outer and inner surfaces being one or both polished or blackened. From all the different combinations of these results, of which he states in detail, I collect the following general inferences:

I. In vacuo; (1) the polished vessel had its cooling always accelerated by the cases; and in this order—

[page] 273

Case
Inside. Outside.
Most accelerated bright black.
black black.
bright (3 cases) black.
bright (3 cases) bright.
bright (1 case) bright.
Least accelerated black bright.

(2.) The coated vessel had its cooling in all cases retarded; and in this order—

Inside. Outside.
Least accelerated black black.
bright black.
black bright.
Most retarded bright bright.

II. In air: both vessels in all instances had their cooling retarded by the cases.

Mr. Fox also found the boiling of water in a bright vessel before a fire accelerated nearly doubly by a case blackened externally.

He considers the results inexplicable, except on the hypothesis of an attraction between matter and heat.

Mr. Fox has also communicated to me in manuscript, an account of some further experiments of the same kind on iron raised to a red heat, but which nevertheless are of such a nature as properly to come under this division of the subject.

The precise temperature to which the iron was raised in each experiment was estimated by the remarkable cessation of its action on a magnetic needle at a certain stage of incandescence.

The iron was inclosed in tin cases of two different sizes, within which the air could be exhausted, the inside being either plain or coated with lamp-black.

The whole was immersed in water, and the temperature communicated to the water in a given time, noted. After observation the iron was plunged in water, and the residual heat thus communicated to the water, noted.

The general results were,—that in the smaller case the cooling was more rapid than in the larger; and in either the internal coating accelerated the cooling; in no case was any material difference produced by exhausting the air.

10.) Dr. Ritchie (Edin. Phil. Journal, xxii. p. 281,) has shown that when a hot nonluminous body is placed between the two bulbs of a differential thermometer, blown out very large and thin, and both remaining plain, the liquid is stationary: the

S

[page] 274

outside half of ohe being coated with black, the liquid sinks from that side.

Hence he infers that the coating has here stopped the heat, which otherwise radiates freely through the very thin glass.

He varied the experiment by using portions of glass blown thin as screens over an aperture: when blackened in a flame or coated with silver-leaf, they intercepted heat; when transparent, not. That this was not from increase of thickness, was shown by using three thicknesses transparent, then removing the middle one, and blackening the inner surface of the others.

He explains the subject by the theory of material caloric and mutual repulsion of its particles.

The same author in another paper (Ann. of Phil. 2nd Series, xii. 123,) gives a variation of the experiment: the hot body is placed between two large and very thin bulbs; one of the hemispheres of one bulb, formed by a plane passing through the centres of both, is coated with China ink; as are also two of the alternate quarters of the other, formed by a plane cutting the former at right angles.

A greater effect is produced on this second bulb.

This is an argument against the effect being due to greater radiation, from the outer surface of the bulb.

Dr. Ritchie has also maintained the same conclusions in his paper before referred to, (Phil. Trans. 1827, Part II. p. 142,) by varying the distance of the screen, which he found to produce no sensible difference in the effect; though with screens of moderate thickness it diminishes rapidly with the distance, according to Leslie's experiments.

DIVISION II.

Terrestrial luminous hot bodies.

a.) Nature of radiation.

The earliest observers noticed differences between this case and that of heat from nonluminous bodies.

The heat from flame, &c., at least in part, passes through air, &c., without heating it.

Scheele observed this with a fire, and that currents of air did not change the direction of the rays. (Treatise on Air and Fire, &c.)

Cavallo (Phil. Trans. 1780,) found a blackened thermometer affected by the light of a lamp.

Leslie (inquiry, p. 448,) found a fire affect his photometer;

[page] 275

also candles, &c. (p. 447),— a distinction pointed out between this and the solar rays, (p. 83, 54.)

The light from putrescent substances does not appear to be accompanied with any appreciable degree of heat, according to Dr. Hulme. (Thomson's Chem. i. 414, 4th edit.) But the effect, if any, must be so small that we cannot positively assert there is none.

The same remark may apply to many other very faint lights.

b.) Reflexion of heat.

1.) Mariotte collected the heat of a fire in the focus of a reflector. (Mem. Acad. of Sciences, 1682.)

Lambert, with burning charcoal in the focus of conjugate reflectors, found a combustible body kindled in the other focus. (Lambert, Pyrometrie; Saussure, Voyage, iv. 119.)

Scheele (On Air and Fire, p. 67-71,) observes that a glass mirror, though it reflects the light of a fire, does not reflect the heat (it is not stated by what means the heat was estimated); but the mirror becomes heated. A polished metallic mirror reflected both the light and heat, and did not become much heated itself; if blackened, it was soon hot.

Pictet extended the experiments with conjugate reflectors to this case, by placing a candle in one focus. The thermometer rose nearly 10° in 6 minutes (Essais de Phys. p. 63.).

Sir W. Herschel (Phil. Trans. 1800, p. 297,) placed a candle at 29 inches from a concave metallic reflector: the focal thermometer in 5 minutes rose 31/4°; another out of the focus was not affected.

The same took place with a fire, and with red hot steel.

2.) Polarization by reflexion.

Berard (Memoir before cited,) tried the polarization of heat from luminous sources, and found a considerable diminution in the position when the light ceases to be reflected.

There was of course here no distinction drawn between the heat accompanying the fight, and the simple heat: of the latter nothing is proved; the former may be merely an effect of the absorption of light, and if so, the term polarization is applied to the heat without any proof.

I repeated these experiments, and, after all precautions, thought there was a small perceptible effect, (when the simple heat was cut off by a glass screen,) which was diminished in the position of non-reflexion for the light; when the whole heat was admitted, no proportional diminution took place. (Edinb. Journ. of Science, vi. 303.)

S 2

[page] 276

c.) Effect of surfaces on emission of heat.

Nothing ascertained under this head, unless we except some remarks in the Edinb. Journ. of Science, No. ii. p. 302.

d.) Effect of surfaces on absorption of heat.

All experimenters have usually blackened their thermometer. (Cavallo, Phil. Trans. 1780.)

Prof. Robison exposed a thermometer on charred oak under a glass cover to the rays of a fire, when it rose to 212° Fahr. (Black's Lect. i. 547; Thomson, i. 127.)

e.) Effect of screens.

1.) Mariotte interposed a glass screen between the fire and concave mirror, and found the heat no longer sensible at the focus. (Biot, iv. 606; Mem. Paris, i. 344.)

Scheele interposed a glass screen in the experiment before mentioned, and found the heat of a fire so much intercepted as to be no longer sensible to the hand: not even sensible in the focus of a reflector.

Pictet with the conjugate reflectors interposed a glass screen. The focal thermometer, which had risen 10°, fell 7° in 9 min.; on removing the screen it rose again. (Essais de Phys. p. 63.)

2.) Sir W. Herschel tried experiments on this point, (Phil. Trans. 1800.) Two moveable objects illuminated by a lamp were viewed by the eye, one through an open hole, the other through a hole covered successively by different transparent media. One object was moved to greater or less distance, till they appeared equally bright; the interceptive power was estimated directly as the illumination required to produce the equalization, that is, inversely as the square of the distance.

Two equal thermometers inclosed in a box, with apertures over the bulbs (which were plain), one open, the other covered successively by the different transparent media, were exposed to different sources of heat, and the interceptive effects compared together and with those of the same media for light. Thus among the results were the following:—

common Fire. Candle.
Light. Heat. Light. Heat.
Coach glass 0 750 86 625*
Dark red glass 999 573 999 526

* Out of 1000.

[page] 277

3.) Refraction by lenses.

Lambert collected the rays of a fire by a large lens, and found the heat scarcely sensible to the hand.

Sir W. Herschel (Phil. Trans. 1800, pp. 272, 309, 327,) received the rays of a candle on a lens, with a pasteboard screen, having an aperture nearly equal to that of the lens; the thermometer in the focus rose 21/2° Fahr. in 3 min.;—the same with the rays from a fire, and from a mass of red hot iron.

Mr. Brande found the rays of a flame, concentrated by a lens, produced an effect on a blackened thermometer in its focus; the lens did not become heated. (Phil. Trans. 1820, Part I.)

4.) Dr. Ritchie found that if Leslie's photometer be placed opposite a ball of iron heated almost to redness, no effect whatever will be produced; but if the temperature of the ball be raised so as to shine in the dark with a dusky red colour, the fluid in the stem of the black ball will sink a considerable number of degrees. If the temperature of the ball be raised still higher, it.will produce a greater effect upon the instrument than the flame of the finest oil-gas, though the one possesses a much greater illuminating power than the other.

Dr. Turner, and Dr. Christison have found that Leslie's photometer, "is powerfully affected by heat" when placed "before a ball of iron heated so as not to be luminous, or even before a vessel of boiling water." The opposite result of Dr. Ritchie may possibly be owing to some difference in the surface, substance, or thickness of the black bulb employed. (Edinb. Journ. of Science, iv. 321.)

I have found differences, which I am at a loss to account for, between the effects on a differential thermometer with the bulbs of equal height, and one in which they are in a vertical line.

5.) That there exist essential differences between the constitution of the heating power of luminous hot bodies, and that of the same power proceeding from those which are non-luminous, was remarked by former experimenters. But it is a point which does not seem to have excited any close or systematic inquiry until the subject was taken up by M. De la Roche, whose researches are justly entitled to the high celebrity they have acquired. The Report of the French Institute upon them will be found in the Annals of Phil. O.S. ii. 161; and a full account of the experiments in Biot's Traité dc Phys. iv. 640.

The whole series of results is as follows:—

[page] 278

Rise of hermometer in 1 min, centig.
Soverce of Heat. No Screen. Transparent Sereen. Blackened Screem.
1. Vessel of mercury, temp. 180° cent. 3°·94 0°·22 0°·07
2. Vessel of mercury, boiling, 346° 16·33 1·36 0·17
3. Iron, 427° 32·8 1·70 0·31
4. Copper, 960° (1.) 38·97 11·83 0·40
5. Ditto (2.) 71·54 21·41 0·21
6. Argand lamp–no chimney 21·12 7·29 0·21
7. Argand lamp–chimney 23·44 12·82 0·23

The two first experiments of this series have been already considered. The 3rd, or iron at 427° centig., was at a red heat, its temperature of luminosity in the dark being about 400°. This, therefore, and the subsequent part of the series are affected by the consideration that light was emitted, which materially alters the case, as we shall presently observe.

De la Roche infers from these experiments, that a portion of simple radiant heat is transmitted directly in the way of radiation through glass; and that this increases as the temperature is raised.

A thick glass, though very transparent, stops heat more than a thin glass less so: the difference is less as the temperature is raised.

A portion of the heat having been intercepted by one screen, a proportionally much less diminution is caused by the introduction of a second; hence he infers that the rays emitted of a hot body are of several kinds, possessing different degrees of power to pass through glass.

He views the results, when the source of heat is raised to the temperature of luminosity, as forming one connected series with those below that point, and thus conceives a gradual advance in the radiant matter or agent, from the state of simple heat towards that of light or "luminous heat."

6.) The theory adopted by De la Roche, as well as by Biot (Tratité de Phys. iv. 640,) and Leslie, is that of one simple agent, which, as the temperature of the source is raised, is gradually brought more into the state of light, which on absorption is reconverted into heat. At low temperatures it is wholly or nearly all stopped by transparent screens. At in-

[page] 279

creasing intensities more of it is enabled to pass in the way of direct radiation.

In order to establish this theory, it would be necessary to show that whatever may be the particular law of relation to the surfaces of bodies by which the action of the "igneous fluid" is determined at any stage of its evolution, the portion transmitted by a screen should act upon any two given surfaces in precisely the same ratio as the part intercepted, or as the whole. Such a ratio will obviously differ at different stages of incandescence or inflammation; but at the same stage it ought to be found exactly the same—only diminished in the actual magnitude of its terms when the glass screen is interposed,—as when there is none.

But no such experimental proof had been offered by any of the experimenters before named. It was obviously called for to Support or refute their theory, and was capable of being easily supplied by experiment. That the conclusion is not a necessary one, will be evident by merely observing that the phænomena may just as well be explained by supposing two distinct heating influences, one associated in some very close way with the rays of light, carried as it were by them through a glass screen without heating it; the other being merely simple radiant heat stopped by the screen, exactly as in the case of a nonluminous hot body.

To ascertain by experiment which of these suppositions was the true one, was the object of an inquiry which I communicated to the Royal Society, and which is published in the Phil. Trans. 1825, Part I. p. 187. I also gave an abstract of the results, accompanied by other illustrative remarks, and some theoretical views in a paper in the Quarterly Journal of Science, No. XIX. p. 45. Some remarks also on the experiments are made in the Edinb. Journ. of Science, N.S. No. VI. p. 304.)

These experiments combine the examination of the effect of screens with those of surfaces. It is assumed, on the authority of previous experiments, that simple heat affects a thermometer in proportion to the absorptive nature of its surface: for example, a surface washed with a paste of chalk is rather more absorptive than one coated with Indian ink; and this kind of heat is stopped by transparent screens of ordinary thickness. It would seem from some experiments already mentioned, that from luminous hot bodies the effect is greater in reference to the darkness of colour of the surface, and is transmitted through glass. But when a body is heated to luminosity, how does this change in its properties take place? Are its relations gradually altered in themselves? or are there two sorts of heating

[page] 280

effect emanating from it at the same time? These are the questions which my experiments were directed to answer, and the mode of trying the point is extremely simple; it is only to ascertain whether of the total heating effect from a luminous hot body, the portion intercepted by a transparent screen is of the same nature as, or different from, the part transmitted, in its relation to the surfaces on which it acts.

The experiments were conducted simply by having two thermometers, one coated with smooth black, the other with absorptive white, observing the ratio of the effects when they were exposed together to the direct influence of a luminous hot body, and comparing it with the ratio similarly observed when a glass screen was interposed.

The screen acquiring and therefore radiating heat from the first moment of the experiment, will affect the thermometers in a ratio (as before observed,) differing little from equality; and these equal quantities added to the terms of the ratio of the direct effects of the luminous body will of course diminish the inequality of that ratio. This cause of error may not have operated to any great degree, but its tendency is obviously to a diminution of the ratio.

Notwithstanding this, the observed result in all cases, with a lamp, or with iron raised to a bright red heat, was, that the ratio of the effect on the black to that on the white thermometer was increased by the interposition of the screen.

A summary of the results of two sets of experiments (conducted with some slight variation), and in the second of which the temperature acquired by the screen was carefully noted, is as follows:

Rise of Thermomoeter (centing.) in 1 min.
White. Black. white. Black.
Iron bright hot (1.) 1°·25 2°·75 7°·0 8°·75
(2.) 0·6 1·25 2·95 3·75
Argand lamp (1.) 0·6 2·0 1·8 3·4
(2.) 1·3 2·35 2·35 3·2

These numbers are the means of several repetitions.

The necessary conclusion from this difference in the ratio of the direct and screened effects, is, that the portion of heat which has the property of permeating the screen has also the property of affecting the two surfaces in a ratio different from that in which the part intercepted acts upon them.

As in researches of this kind great numerical precision is unattainable, I was especially, at every step of the inquiry,

[page] 281

anxious to devise as many variations of the experiment as possible;—these all tended to confirm the results just given.

Thus I used a large differential thermometer having its bulbs differently coated, and exposed each of them in turn to the luminous source of heat, the other being completely screened, and invariably found the ratio of the effects on the black and white bulbs considerably greater when affected only by the transmissible part of the heat, than when exposed to the whole. As before, the part added on the removal of the screen was of a nature tending to add to the terms of the former ratio, quantities in a ratio much nearer equality; viz. that which the effects o£ simple radiant heat would give when acting respectively on the two bulbs.

Other variations of the fundamental experiment, were as follows:

A differential thermometer having one bulb black, was exposed to the radiation from luminous hot bodies, first with and then without the interposition of a glass screen; the same position being preserved.

If the screen had no influence, it is evident that in whatever proportion the radiant matter affects the two bulbs, if it be of one simple kind, the only difference on removing the screen will be that its intensity will be increased, but will act on the two bulbs in the same proportion as before. Consequently an increase of effect, or motion of the liquid in the tube in the same direction as before, must take place.

In various experiments of this kind, after using several precautions against the influence of the screen, I never found an increase, and generally a decrease; that is, the action on the other bulb was now increased, or the portion of heat before intercepted and now admitted has a different relation to sur* faces from that transmitted. (Quarterly Journal of Science, xix. p. 45.)

Similar experiments were tried with the two bulbs in a direct line from the hot body, each placed nearest alternately, with and without a screen. The difference of ratios in the two cases was very striking. (Annals of Phil. June 1825, p. 401; see also Edinb. Journ. of Science, No. IV. 323.)

Upon the whole, the unavoidable conclusion is, that if the total direct effect were the result of one simple agent, the intervention of the glass would, by intercepting some portion of it, produce no other alteration than a diminution of intensity; the ratio of the two effects would remain unchanged: but the reverse being the case, it follows that there are two distinct agents or species of heat acting together.

[page] 282

Upon combining these results with those of previous experimenters, we are led to the following general statement of the case:—

When a body is heated, at lower temperatures, it gives off radiant heat stopped entirely by the most transparent glass, and affecting bodies in proportion to the absorptive texture of their surfaces.

At all higher temperatures it continues to give off such radiant heat distinguished by exactly the same properties.

At a certain temperature it begins to give out light: precisely at this point it begins also to exercise another heating power distinct from the former; this is capable of direct transmission through glass, and affects bodies in proportion to their darkness of colour.

This second species appears to agree with what the French philosophers have called "calorique lumineux," or the "igneous fluid" of Prof. Leslie; but they seem to have considered it as constituting the entire effect.

The distinction thus established easily applies to the explanation of De la Roche's results before stated. On inspection it appears that the numbers in the column belonging to the blackened screen are almost exactly in the same ratio to the first or direct effect throughout the whole series.

Upon the principle here laid down, the effects with the blackened screen would be those arising from the absorption and subsequent radiation of both species of heat; these in each instance being absorbed in the proportions in which they existed in the original radiation, produce a secondary effect proportional to the primary.

The effect with the transparent screen does not follow any proportion to the primary; and this is explicable as due to the glass intercepting the one kind of heat, which follows no proportion to the other, this last being wholly transmitted. Also by comparison of the latter experiments with the two first of the series, it is probable that, throughout, a certain degree of heat was in this case also absorbed and radiated again by the screen.

The existence of this distinction, and the proportion between the two species of heat in the radiation from different sources, as various kinds of flame, metal at successive stages of incandescence, &c., afford many topics of inquiry, on some of which I attempted some rough determinations, confessedly very imperfect. (Annals of Phil. N.S. liii. 359; liv. 401.) The distinction applies to some results of Mr. Brande on the flames of different gases, (Phil. Trans. 1820, Part I, p. 22,) and

[page] 283

of Count Rumford, on increased intensity of combustion, and on the coalescing of several flames. (Essays, i. 304.)

7.) Melloni states, (Ann. de Chim. Dec. 1831, p. 385,) that by using his thermomultiplier he has found the permeability of transparent bodies to heat to be also dependent on their refractive power. He has compared twenty such media, and finds the order of permeability constantly the same, whatever be the temperature of the source. Chloruret of sulphur has the greatest power, oil next, and water least; he exposed them to the rays of a candle, an Argand lamp, or the sun. He finds the differences of permeability less, the higher the temperature. The full account is promised in another memoir.

All this obviously applies only to luminous hot bodies.

MM. Melloni and Nobili, in their former paper, (Annales de Chimie, Oct. 1831, p. 211,) also speak of the heat from phosphorus having been by these means found sensible, though it is often supposed to give light without heat.

8.) For information on various points connected with the subject, and on the theories of the evolution of light and heat, the following references may be useful.

Wedgewood, Phil. Trans. 1792, p. 28, thinks that light from attrition is produced by a heat of from 400° to 600° Fahr.

Dizé on Heat as the Cause of Shining, Journ. de Phys. xlix. 177. Gilbert, Ann. iv. 410.

Fordyce on Light from Inflammation, Phil. Trans. 1776, p. 504. Morgan, Phil. Trans. 1785, p. 190. M. Hermstaedt, Nicholson's 4 to Journal, v. 187.

Mr. Davies on Flame, Annals of Phil. Dec. 1825.

Mr. Deuchar on Flame, Edinb. Phil. Journ. iv. 374.

M. Seguin on Heat and Motion, &c., Edinb. Journ. of Science, xx. 280.

DIVISION III.

Heat of the sun's rays.

Speaking according to our ordinary sensations, we are accustomed to say that the sun communicates both light and heat. Light is transmitted in a way which we term radiation. The heat from nonluminous hot bodies is transmitted to a distance in a way closely analogous; and to which the same name has been applied.

In the first instance, we might suppose that the sun sends out two separate emanations, one of light, and another distinct from it, and similar to that of radiant heat from a mass of hot water; and this, perhaps, was the first view taken of the sub-

[page] 284

ject, though a confused idea of some very close and intimate connexion subsisting between the solar light and heat appears to have prevailed.

This subject, as might naturally be expected, attracted the early notice of experimenters. A very slight examination sufficed to show that the rays of solar heat (whatever their nature might be,) differed essentially in many properties from those of terrestrial heat, whether radiated from luminous or nonluminous bodies. Whether there existed a separate set of heating rays distinct from those of light, and at the same time differing in many respects from rays of terrestrial heat; or whether these differences depended on some unknown property of the rays of light, was a question which for a long time remained without any direct investigation, and on which even now we have, perhaps, no very precise ideas.

I. Solar rays in their natural state.

a.) Nature of radiation.

1.) The solar heat is transmitted through the air without heating it.

It invariably accompanies the light.

Scheele conceived that the sun's rays of light produced heat not when in motion but when stopped by the interposition of solid bodies. (On Air and Fire, &c.)

Mr. Melville seems to have adopted nearly the same theory, and to have conceived reflexion at an opake surface to be the cause of an excitation of heat from the sun's rays. (Evans on the Calorific Rays, &c. Phil. Mag. June 1815.)

In general, for light of the same composition the heat appears nearly proportional to the illuminating intensity.

2.) Measures of radiation.

Theory of the sensibility of thermometers especially for experiments of this kind. (Sir W. Herschel, Phil. Trans. 1800, Note, p. 447.)

Leslie contends for the exact proportionality of intensity of light and heating power. (Inquiry, pp. 160 and 408.)

Theory and construction of his "Photometer" ch. xix. p. 403.

Ritchie's "Photometer" of the same kind. Phil. Trans. 1825, Part I. p. 141. See his Remarks on Leslie's Photometer, Edinb. Journ. of Science, No. IV. 321, and V. 104.

Mr. Daniell in his work on Meteorology has collected a great number of observations on the heating power of the sun's rays in different latitudes from the polar to the equatorial regions.

[page] 285

Most of these observations were made by comparing two thermometers, one of which was kept in the shade, whilst the other, having its bulb blackened, was exposed to the direct rays of the sun; but, as Dr. Ritchie observes, no correction seems to have been made for the variable causes which abstract caloric from the blackened ball of the exposed thermometer. (Edinb. Journ. of Science, v. 107.)

In the same paper is described the method proposed by Sir J. F. W. Herschel; his object was to ascertain, by direct experiment, the relative heating power of the sun's rays; this he did by exposing in a glass vessel, or large thermometer, at different times and places, a deep blue liquid, for a given time, to the direct rays of the sun,—noting the increase of temperature, which was purposely rendered very small by properly adjusting the capacity of the instrument, then shading the sun's direct rays, and leaving it exposed for an equal time to the free influence of all the other heating and cooling causes, radiation, conduction, wind, &c., and again noting the effect of these. The same difference of these, according to their signs, was the effect of the mere solar radiation. Dividing this by the time of exposure, he had the momentary effect or differential co-efficient, which is the true measure of the intensity of radiation.

Professor Cumming has been engaged in researches, the object of which was to obtain a measure of the total heating effect of the sun's rays. He has communicated for this Report an account of his investigations, of which the following is the substance.

His instrument consists of a bent tube in the form ⋒ one side terminating in a black bulb containing ether, or sulphuret of carbon; the other a graduated tube closed at the bottom; into this, on exposure to the sun, some of the liquid is distilled over from the bulb; and the quantity measured on the scale is proportional to the amount of radiation, when all interfering causes are allowed for; and these are estimated by comparative observations.

The experiments have been varied by exposing the bulb and screening the other part, or by exposing the whole instrument equally to the sun; and by making contemporaneous observations with the instrument wholly uncovered, or covered totally or partially by a glass to protect it from currents of air.

The Professor has endeavoured to make a standard scale by registering the sun's radiation on clear days every half hour, or hour, in the usual manner, and comparing them with the contemporary distillation; or by placing the two sides of the instrument in two vessels of water at unequal temperatures, and

[page] 286

noting the distillations in given times by ascertained differences of temperature.

The instrument is filled with ether in the same manner as Wollaston's Cryophorus (from which the suggestion was taken); but there is an inconvenience, arising from the circumstance of the difference of pressure under which the instrument is hermetically sealed, which renders two instruments not strictly comparable; this he proposes to remedy by sealing a standard instrument when exhausted to a known pressure by the air pump.

The ether or sulphuret of carbon employed must be perfectly pure, or there is a re-absorption. The circumstance of being exposed to the air, or covered, makes great differences in the indications; especially in windy weather. To avoid an inconveniently long scale, there should be two instruments constructed, one for winter and the other for summer. The Professor has kept for nearly a year a register of sunshine.

b.) Reflexion of solar heat.

1.) It takes place exactly by the same laws as that of the light.

The heat is collected in the focus of concave reflectors along with the light.

2.) The sun's rays reflected from the moon, are probably much too feeble to allow of any heat being made sensible.

Dr. Howard however states, that with a peculiar differential thermometer he has obtained an effect. (Silliman's American Journal, vol. ii. 329.

MM. Melloni and Nobili (with the apparatus before described) tried to detect heat in the moon's rays, but without success; they mention however that terrestrial radiation interferes greatly with such experiments, and do not describe fully their contrivances for obviating this cause of error. (Ann. de Chimie, Oct. 1831, p. 210.)

3.) Berard (Memoir before cited,) tried the polarization of the solar heat; that is, polarized the sun's light; and in the position of non-reflexion found that the heat had disappeared with it. (See Edinb. Journ. of Science, vi. 297.)

c.) Under this head nothing known.

d.) Effect of surface on the absorption of solar heat.

1.) I am not aware of any experiments directly showing how

[page] 287

far the same relation to the texture of surfaces which has been found in absorption of simple heat, may hold good in regard to the sun's rays. But for surfaces of the same texture it has been incontrovertibly established that the effect in this case increases in proportion to the darkness of colour, or in proportion to the absorption of light; and it would seem most probable that this relation is the only one which really holds good, the texture of the surface being probably quite indifferent except so far as it tends to the better absorption of the light.

2.) Among the earliest experiments on the subject, if not actually the first, were those of Mr. Boyle, on the different degrees of heat communicated by the sun to black, white, and red coloured surfaces.

He caused a large block of black marble to be ground into the form of a spherical concave speculum, and found that the sun's rays reflected from it were far from being too powerful for his eyes, as would have been the case had it been of any other colour; and although its size was considerable, yet he could not set a piece of wood on fire with it; whereas a far less speculum of the same form, made out of a more reflecting substance, would presently have made it inflame.

It was remarked by Scheele, that the thermometer when filled with alcohol of a deep red colour, rose more rapidly when exposed to the sun's rays than another filled with the same kind of spirit uncoloured; but that the fluid rose equally in both when dipped together into the same vessel of warm water. (On Air and Fire, &c.)

Dr. Franklin found that the hand when applied alternately to a black and to a white part of his dress in the sun, would feel a great difference in their warmth.

He observed that black paper was sooner fired by exposure to the focus of a lens than white.

His well known experiment of placing differently coloured pieces of cloth on the snow in the sun, and observing them sink deeper in proportion to the darkness of colour, was first suggested by Dr. Hooke.

3.) Cavallo observed that a thermometer with its bulb blackened, stands higher than one which had its bulb clear when exposed to the light of the sun, or even of the clouds. (Phil. Trans. 1780.)

Pictet made a similar observation, observing that when the two thermometers remained for some time in a dark place they acquired precisely the same height. He also found that when they had both been raised to a certain point, the clean one fell

[page] 288

much faster than the coated one. (Sur le Feu, ch. iv. Thomson, i. 126.) This last statement is so contrary to all other experiments, that we must suppose some mistake.

De Saussure received the sun's rays into a box lined with charred cork, containing a thermometer with a glass front; it rose in a few minutes to 221°, when the temperature of the air was 75°. (Voyages, ii. 932.)

Professor Robison in a similar experiment employed three vessels of flint glass within each other at 1/3rd of an inch distance, set on a base of charred cork, and placed on down in a pasteboard cylinder; the thermometer within, in clear sunshine rose to 230°, and once to 237°. (Black's Lect. i. 547. Thomson, i. 127.)

Sir H. Davy took several small disks of copper of equal weight, size, and figure, on one side painted respectively white, yellow, red, green, blue, and black. A mixture of oil and wax, which became liquid at a temperature of 76° Fahr., was attached to the other surface of each disk; and on exposing the coloured surfaces together to the sun's rays, the length of time elapsed before the mixture on each began to be affected, was in the order in which they are above enumerated. (Beddoes's Medical Contributions, p. 44.)

4.) The experiments of Sir E. Home (Phil. Trans. 1821, Part I.) are particularly deserving of attention, as exhibiting what might at first sight be considered an exception to the above remarks; a greater effect being produced in some instances on a white, than on a black surface. A more attentive examination, however, will show us that these experiments prove thus much: The heat occasioned by the rays of the sun when received directly, or when in some degree intercepted, as by thin white cloth, on the skin, is greater than that communicated by conduction to the same skin, through a black cloth in contact with it, which is itself, in the first instance, heated by absorbing the rays.

He observes also that a white skin is scorched, and that of a negro is not, in 10 minutes, by the direct rays of the sun; that is, as before, the outer coat of the skin allows some of the direct rays to pass through and affect the sentient substance beneath; whereas in the case of the black, the rays are absorbed and converted into heat of temperature, which diffuses itself equally and does not produce the effect of scorching.

5.) The most singular facts connected with the absorption of the sun's rays, are those exhibited by the substances called "phosphori" or "pyrophori". (Thomson's Chem. i. 17.)

[page] 289

The general fact is, that after exposure to the sun, on being removed into the dark they give out light, but it is after a time exhausted; it is given out more copiously and exhausted sooner if heat be applied. Many solar phosphori will always emit light of one colour only, to whatever coloured ray they may have been exposed. In a short notice given by Dr. Young, in his valuable Catalogue of authors, it appears that M. Grosser found that such phosphori as emitted red light only were made to shine most by exposure to blue light. (Rozier, xx. 270.)

Beccari, in a memoir "de Phosphoris" extracted in the Phil. Trans. 1746, p. 81, gives as one of his results, that the light emitted was brightest when the surface of the mass was of a rough texture; those which were smooth and polished, retained little or none, but (supposing the colour the same,) a rougher surface would evidently absorb more light than a smooth one, and therefore might emit more.

Mr. T. Wedgewood compared two pieces of phosphorescent marble, one naked, the other painted black; on applying uniform heat, the coated marble gave out no light, though the other did. (Phil. Trans. 1792.)

But the coating increased the radiating power, and it therefore probably did not retain heat enough to cause the extrication of light.

Mr. Morgan (Phil. Trans. 1785,) after examining many of the phænomena of phosphorescence, generalizes his views by maintaining that all phosphori emit light proceeding in order from violet to red, in proportion as the process is effected by the application of an increasing degree of heat.

This is a very curious subject, as connected with the whole theory of the relations of light and heat. Some valuable information might probably be obtained as to the degree of heat necessary, and whether there is any loss of heat when light is evolved, compared with cases when no light is evolved; as there should be on the hypothesis of conversion of heat into light, or on that of heat becoming latent in the light.

In Mr. Wedgewood's paper above cited, is an account of the principal researches on the subject.

e.) Effect of screens.

1.) That no diminution of the effect of the sun's rays on a blackened thermometer, is occasioned by a transparent screen, was remarked by several experimenters, particularly De la Roche. (Biot, iv. 611.)

2.) I tried the point by two thermometers, (as in the case of terrestrial heat), and found no perceptible difference in the ratio,

T

[page] 290

with, and without the screen, of the black and white thermometers. (Annals of Phil. xli. 321.)

The same result was found with a differential thermometer, with a glass screen over the bulb; which was not blackened; no difference was observable between the indication under these circumstances, and when both were exposed. (Annals of Phil. xlii. 401.)

Hence, I think we are entitled to conclude, that there does not exist in the solar beam, in its natural state, any simple radiant heat (as before defined); but that the whole emanation consists of the other species, distinguished by the two characteristics of affecting substances with heat in proportion to the darkness of their colour, and being wholly transmissible through glass without heating it; and inseparable from the rays of light.

This applies to the rays of the sun which come within the reach of our examination. It must, however, be admitted, as by no means improbable, that the sun may originally give out a separate radiation of simple heat. None of this kind reaches us, but we must consider the very different degree in which any medium, as air, absorbs or intercepts the passage of those two sorts of radiant agents. The heat from a hot body will not be perceptible at a short distance, while its light will traverse an amazing extent of length; and thus at different distances the ratio between the two sorts of heating effect will be very different. Some degree of simple heat, therefore, may actually be initially radiated by the sun, and be lost before it reaches us. We do not know that there is any medium between the different parts of the solar system capable of absorbing heat. The highest regions of our atmosphere into which observation has penetrated, are uniformly the coldest; but they are known to have a greater capacity for heat. Thus, though it is possible that some heat may reach to that distance, and be absorbed without becoming sensible to us, its quantity must be very small; if, therefore, we suppose any simple heat to be initially radiated from the sun, it must be all, or nearly all, absorbed by some parts or appendages of that luminary exterior to the part where it is generated.

3.) The concentration of the sun's heat by a lens is a familiar experiment.

Sir W. Herschel (Phil. Trans. 1800, Exp. 23,) concludes that there is a focus of greatest heat further from the lens, than that of light; sealing-wax was scorched in the same time when in the luminous focus, and at half an inch further from the lens; —this affords no proof of its being separated from the light.

[page] 291

That the heat is found to accompany the rays of light in the most constant and inseparable manner through various refractions, as in the instance of the four lenses in the eye-piece of a telescope after reflection, is also remarked by Sir W. Herschel (Phil. Trans, 1800, Exp. 11).

II. Solar rays subjected to analysis by the prism.

1.) The different heating powers belonging to different parts of the spectrum, were probably first observed by the Abbé Rochon. (Phil. Mag. June 1815; and Biot, Traité de Phys. iv. 600.) He found the maximum in the yellow-orange rays: the prism was of flint glass: his thermometer was filled with spirits, probably therefore tinged red: this may account for his result.

I tried some experiments with the bulb of the thermometer painted red, which appeared to agree with his result. (Annals of Phil. li. 201.)

Prof. Leslie applied his "photometer" to these experiments. (Inquiry, p. 454.)

Dr. Hutton observed the different heating powers, and that they are not proportional to the illuminating. (Diss. on Light and Heat, p. 38.)

Landriani found the maximum in the yellow rays, as also did Senebier. (Volta, Lettere, &c. 136.)

Berard (Mém. d'Arcueil, iii.; Ann. de Chimie, lxxxv. 309,) repeated the experiment with a heliostat. He found the maximum in the red, but some heat beyond. He repeated the experiment in both the spectra formed by Iceland spar.

2.) Sir W. Herschel (Phil. Trans. 1800, Part II.) first observed the maximum of heat beyond the red end of the visible spectrum, and considered the effect as due to essentially invisible rays of a separate kind from those of light.

Yet he found them subject to the same laws of refraction, and their dispersion corrected by another prism: they were concentrated by a lens (Ibid. p. 317), and by reflexion (pp. 298, 302).

Leslie objects to the conclusion of invisible rays, and tries to account for it as owing to an optical cause. (Inquiry, Note, p. 559; see also Nicholson's Journal, 4to, iv. 344 and 416.)

Sir H. Englefield (Nicholson's Journal, iii. 125,) found heat beyond the visible red; it does not appear whether it was there at a maximum: the rays were such as to be concentrated by a lens, and he compared the effects on a black and a white bulb.

T 2

[page] 292

The exterior effect on the white bulb was in a much less ratio to that within the visible spectrum, than on the black.

Sir H. Davy repeated these experiments in the clear atmosphere of Italy, and with thermometers of extremely minute size, to secure an instantaneous effect: he found the maximum beyond the red.

These experiments were also tried by Ritter and by Prof. Wünsch (Magazin der Gesellsch, &c. Berlin 1807). He used prisms of different substances; with alcohol, oil of turpentine and water, the maximum was in the yellow; with green glass in the red; and with yellow glass on the extreme boundary.

3.) But by far the most important and conclusive researches on this subject are those of Dr. Seebeck, who in a memoir read to the Royal Academy of Berlin, after discussing the conclusions and views of previous experimenters, proceeds to an elaborate series of experiments of his own, in which he has discovered the cause of all their discrepancies. The position of the maximum heat in the spectrum depends entirely on the nature of the medium employed,—a circumstance almost wholly unnoticed by former experimenters.

The heating intensity is very small towards the violet extremity; it thence gradually increases in prisms of water, alcohol, or oil of turpentine; the maximum is in the yellow space: in those of solution of sal-ammoniac and corrosive sublimate, or sulphuric acid, it is in the orange; in crown glass and common white glass, in the middle of the red: in those glasses which contain much lead, it is in the limit of the red: and in flint glass, beyond the visible boundary, but nearer to it with Bohemian than with English glass. In all cases it gradually diminishes from the maximum, and is perceptible to some distance beyond the visible boundary. (Schweigger's Neues Journ. x. 129; Annals of Phil. Sept. 1824; Abhandl. der Königk. Acad. Wissenschaften in Berlin, 1818-19, p. 305; Phil. Mag. Nov. and Dec. 1825; Edinb. Journ. of Science, No. II. 358.)

4.) Analysis of the solar rays by the absorption of media.

In respect to light, the remarkable variety in the absorption of different rays exhibited by different media has been well established, and affords a new sort of analysis of light.

In regard to the solar heat, similar researches have been made, though as yet to little extent. The first observations of the kind were those of Sir W. Herschel (Phil Trans. 1800). He found the absorption of several kinds of glass for his invisible rays and for the middle red, to be proportional to the following numbers out of 1000 rays incident:

[page] 293

Invisible rays. Red rays
Flint glass 000 143
Coach glass 143 200
Crown glass 182 294
Dark red glass 000 692

5.) Sir D. Brewster has lately been engaged in some researches on this subject, an abstract of which he has kindly communicated in manuscript for this Report. Agreeably to the view he has established of the solar prismatic spectrum as consisting of spectra of three primary colours superposed, and having their maxima at different points, he regards the heating power as due, in like manner, to another primary spectrum superposed in the same way; and similarly the chemical rays. He makes the following statements with respect to the heating rays.

1st, There is no proof whatever of the existence of invisible rays of any kind beyond the red or the blue extremity of the spectrum. Sir W. Herschel's experiments prove the existence of heat beyond the visible extremity of the spectrum which he used; but Sir D. Brewster has succeeded in rendering the spectrum visible at every point where any heat was produced.

By particular processes he has traced the light at that end greatly beyond the place where Frauenhofer makes the spectrum terminate.

The same he considers established in regard to the blue end of the spectrum and of the deoxidizing rays. He thinks it extremely probable that the heating and illuminating rays are different rays; but they have never yet been found in a state of complete separation.

2ndly, Until it is proved, therefore, or rendered probable, that the same intensity of light of different colours, as it proceeds directly from the sun, is accompanied with different degrees of heat, we must assume it as true that the heating power is proportional to the illuminating power of the different rays of solar light.

3rdly, It appears from Dr. Seebeck's experiments on the water spectrum, that this relation holds generally in it, as he found the maximum of heat to be in the yellow rays, or coincident with the maximum of light. Hence Sir D. Brewster draws the important conclusion, that water has the same degree of transparency for the solar heating rays that it has for light, which is the same as all colourless transparent media have for light; that is, water absorbs equally all the different rays of solar heat, in the same manner as it does all the different rays of solar light.

[page] 294

4thly, It has been found by experiment, that with prisms of crown glass the maximum heating effect is in the middle of the red space. Unfortunately the relation between the maximum heat in the water spectrum and in the crown glass spectrum has not been ascertained. If we suppose them equal, it appears that the crown glass must have exercised a greater absorptive action than the water upon the more refrangible rays, and a less absorptive action upon the less refrangible rays; in the same manner as is done by red glasses upon light.

A prism of sulphuric acid gives the maximum ordinate of heat in the orange space; or the fluid absorbs more of the red rays than crown glass, and less of the rays on the other side of the orange.

In flint glass, where the maximum heat is at the very extremity of the spectrum, scarcely any of the red rays are absorbed, while great proportions of all the others are.

Dr. Turner (Chem. p. 84, 3rd edit.) says, that it is difficult to account for Seebeck's results without supposing that different media differ in their power of refracting caloric (i.e. the heating rays of the sun).

Sir D. Brewster considers that the true explanation is that Which the above principles afford, viz. that colourless transparent bodies, in acting upon the solar heat, exercise the same sort of absorptive action upon it, that coloured transparent bodies do upon light; the maximum ordinate shifting its position with the nature of the body. Coloured media give sometimes two or more maxima of light, with large spaces and small lines entirely defective of light, in consequence of the absorption being total at those places.

In like manner he is persuaded it will be found that there are defective spaces and lines in the spectrum of solar heat; these he thinks may possibly be detected by using as thermometers the minute natural cavities in topaz, &c., filled with fluid or vapour, and not more than 0·001 inch in magnitude.

5thly, These views are exactly accordant with the results of Sir W. Herschel above stated.

They are equally consistent with the facts, whether the curve of heat terminate abruptly at the extremity of the red space, or continue beyond the visible spectrum.

Sir D. Brewster has by particular methods of condensation succeeded in detecting both heat and light at considerable distances beyond the maximum of heat, with a flint glass prism; that is, rays undergoing very little refraction.

He considers it highly probable that the deoxidizing rays will be found to be subject to the same laws of absorption as

[page] 295

those of heat and light; the media we commonly use may absorb them copiously, whilst others may be found which may transmit them more abundantly.

Similarly with the magnetizing rays. And thus we may account for the contradictory results hitherto obtained on this point, by supposing that some ingredient rendered one prism absorptive of these rays, and another not so.

6thly, Sir D. Brewster extends these views to the analogies between solar and terrestrial heat.

He considers those rays of the solar spectrum just mentioned, which undergo little refraction, to be analogous to those thrown off by bodies slightly heated. The waves of heat are broad and slow in their motion; as the temperature is raised they are thrown off with more velocity, and become smaller and suffer a greater refraction. When the velocity is such as to give them a refraction equal to that of the red rays, then red light is produced; and successively the other colours are added, till at a very high temperature white light is radiated.

He proposes to examine what transparent body transmits most heat, and by converting it into a lens, expects to find a series of foci at different distances, beginning from that of the violet rays to that of those corresponding to rays of very little refrangibility.

7thly, He applies these views as affording an explanation of De la Roche's result before mentioned, viz. that a second screen intercepts a much smaller proportion of the heat, after passing a first, than the first did of the whole effect: this De la Roche ascribed to something analogous to polarization.

On the principle just stated, the explanation is very simple. The first plate intercepts those rays which it has a tendency to absorb, and transmits the rest: the second, being of the same kind, of course will transmit these with scarcely any further diminution.

He observes, that thick masses of colourless fluid or of glass transmit scarcely any radiant heat in a way analogous to that in which thick masses of coloured glass are opake to all rays of light.

He conceives that substances may be found which are opake to light and yet transparent to heat. These should be carefully sought for, as they would be of great practical value. Red glass, for example, which scarcely transmits any light or 1 ray in 2000, transmits all the invisible rays of Herschel, 692 of the 1000 red rays, 606 rays out of 1000 of solar heat, and 630 of "culinary" heat, according to Sir W. Herschel. We may expect therefore to find an opake metallic glass, or thin plate of metal,

[page] 296

which, though quite opake for light, may transmit heat copiously.

Sir D. Brewster considers Sir W. Herschel's experiment on the refraction of "culinary" heat by lenses, to be very unsatisfactory, as before noticed. He recommends a lens composed of zones, so as to have no greater thickness in the middle than towards the edges, a construction which he has described in his "Optics," p. 322 (Cabinet Encyclop.), and made of glass, which unites the highest refractive power with the smallest absorptive power for heat.

It is also important to find, as sources of heat, bodies which do not become luminous till at extremely high temperatures.

6.) The researches of M. Melloni have also been extended to this part of the subject. (Annales de Chimie, Dec. 1831, p. 388.)

From known observations on the spectrum, he remarks that there exists on opposite sides of the maximum, isothermal points; one in a coloured part, the other without the red end of the spectrum.

On causing the different rays to pass though a plate of water, and noting the effect on the thermo-multiplier; the heat of the violet ray was undiminished, but its isothermal totally intercepted.

That of the indigo slightly diminished; its isothermal not totally intercepted.

Proceeding in this way with the other rays, he found in general that the portions of heating power intercepted in the coloured rays, and those which are transmitted in their isothermal rays, increase in proportion as they approach the position of the maximum, where of course upon the whole the interception is greatest; or, in other words, the rays of the calorific spectrum undergo an interception by water in proportion as their refrangibility is less.

He gives a Table of the numerical results. He views his results as precisely according with and explaining those of Seebeck. With a water prism the heating orange and red rays are more intercepted than the yellow; in this therefore the maximum appears.

Conclusion.

We have thus far taken as close a survey as is consistent with the limits of a Report like the present, of the successive and varied researches which have been made with the view of tracing the laws of radiant heat. In the present state of our knowledge, it must upon the whole be avowed, that we have little to contem-

[page] 297

plate but an assemblage of facts, or alleged facts, determined with more or less accuracy; few indeed with any great precision, many resting upon very vague evidence, and in several instances the results of different observers exhibiting a wide discrepancy or even direct contradiction: whilst, with very few exceptions, any general laws can hardly be said to be established with that certainty which can substantiate their claim to be received as legitimate physical theories.

In offering suggestions for the advance and improvement of this branch of science, the first and most essential point to which attention ought to be directed, is the improvement, or rather invention, of the means of obtaining accurate indications of radiant heat, down to its most minute and feeble effects. In reference to this point, good determinations are much wanted of the degree to which the expansion of the bulb influences the accuracy of air thermometers. The improvement of mercurial thermometers so as to produce an instrument of extreme sensibility to the minutest effects of heat, is an object the attainment of which would probably be more important than that of any other means for accomplishing the end in view. But other methods founded on good principles should be diligently sought for and tried; for example, it might be matter of inquiry whether we could render available to this purpose the incipient melting or softening of some substances by a very slight increase of heat, or the evaporation of volatile liquids.

But it is more particularly desirable that the instrument of MM. Nobili and Melloni should be tried, and a precise examination set on foot of its real accuracy and the causes of error to whose influence it may be liable. This is the more necessary from the very remarkable character of many of their results; whilst the alleged sensibility of the instrument, as they describe it, is such as almost to exceed belief.

When we shall have succeeded in obtaining that prime requisite, an unexceptionable measure of minute effects of radiant heat, we may then proceed with some hopes of success to examine the points on which there at present prevails so wide a discrepancy between different experimenters.

The polarization of heat is perhaps the question which of all others requires the most extreme sensibility in our thermometer, or rather thermoscope, in order to its satisfactory determination. It may be tried either directly with the simple heat from nonluminous hot bodies; or with luminous sources, with and without a glass screen, comparing the total compound result with that due to the transmissible part or heating power of light alone, and thence deducing the part due to simple heat. The

[page] 298

main difficulty is that of getting any indication at all, after two reflexions from plane surfaces.

Another point which requires further investigation is the apparent transmission of simple heat through very thin transparent screens, but not through opake. This should be examined in connexion with the acute remark of MM. Nobili and Melloni, that a thin stratum of soot may retain its low conducting power, and thus intercept the effect. This of itself would form a subject for an accurate series of experiments; viz. whether the ratios of the conducting powers of substances remain the same for all thicknesses.

The very nature of the transmissive aad interceptive powers of screens is little understood. Supposing simple heat transmitted without diminution, how far is the mode of such transmission analogous to that of light? what time is required for a body to commence radiating heat after it has begun to acquire it? whether it acquires it from a distant source instantaneously? how the heat distributes itself upon or through a screen? what is precisely the effect of a coating on one side of the screen in relation to the last question? upon what the singular exceptions and anomalies pointed out by Melloni and Nobili depend? whether any other such apparently anomalous cases can be found?—These are a few of the most obvious questions which arise out of the slightest survey of the present state of our knowledge, and on which accurate determinations are wanted before we can be said to possess even the elements of a scientific theory.

May it not be the law, that if a body be placed in the rays from a source of heat, it will be acquiring and giving out heat, till the intensity of radiation at the points before and behind it, resumes its original proportionality?

The time in which this takes place will depend on the extent of the body, its thickness, its conducting power, its capacity for heat, and the state of both its surfaces.

These may be such that the effect may be sensibly instantaneous, and the radiation therefore appear to go on without interruption. In this case also the distance of the screen from the source (within moderate limits,) may make no sensible difference; though if any of the above circumstances retard the effect to a sensible amount, then there will be a difference with the variation of distance. In this way we may as it were regard the medium between the source and the thermometer, as merely a compound, of which the screen is one portion, and the air the other.

Another class of questions respecting which little if anything

[page] 299

is accurately known, may be put with regard to the modifications (if any) which radiant heat may undergo, in passing through small apertures; this will again be connected with the interceptive power of net-work. A very curious and delicate subject of inquiry is the repulsion exerted between heated bodies at sensible distances, of which a short notice is given in the Quarterly Journal of Science, xxxix. 164.

The reflexion of heat has been little examined, except in the single case of its concentration by spherical reflectors; and here (according to Leslie,) it is not brought to the same focus as light: this requires examination, as well as the simpler case of plane surfaces, and the proportion of heat reflected at different incidences. There will probably in all cases be a very large deduction to be made for the heat acquired by the reflector and radiated again.

But another class of such questions yet remains in connexion with that fundamental point which was the object of my first inquiries. The conclusion from my experiments, viz. that luminous hot bodies are sending forth at the same time two distinct species of heat distinguished by different properties, is the unavoidable conclusion from the experiments, depending on the mathematical truth, that if a ratio be altered by the addition or subtraction of quantities from its terms, the quantities added or subtracted must be in a different ratio from the original one. I here repeat this because the nature of the reasoning has not been perceived by some persons. This conclusion undoubtedly introduces a complexity into the view we must take of the phænomena; whereas if we were at liberty to adopt the simpler theory of De la Roche and others, many of the apparent anomalies would be reconciled. Hence the verification of my results becomes a point of considerable importance. If any experimenter with more accurate apparatus shall succeed in showing them to be erroneous, he will achieve an important step towards simplifying the theory. In this instance again the improvement of the thermometer is a primary requisite.

I may here mention that I have recently had a more delicate apparatus made, with which I have repeated my former experiments, still with the same result; it consists of two thermometers mounted together as before described. They were contrived for me by Mr. Cary, so as to have very large degrees for a small part of the scale a little above ordinary temperature. 1° Fahr. occupies about half an inch; but the bulbs are large, which is unfavourable to the rapid communication of the effect. These experiments are of a very tedious

[page] 300

nature to repeat with precision, owing to the necessity of waiting between each repetition for the thermometers to cool and become stationary.

But it should be observed that there is nothing in my results which contradicts the idea that simple heat may have in a very slight degree a power of transmissibility through glass: all I have assumed is, that it is sufficiently distinguishable in this respect from the heating power which accompanies the light, and which undergoes no diminution. Connected with these points, again, is the question, whether if simple heat can radiate through solid transparent media, it cannot also commence radiating IN them. It is commonly asserted that radiation can only take place, or commence, in elastic media. This, then, is an inquiry which will lead into a wide field of research, and may be found connected with the intimate nature of radiation. It will also be a question, whether, and how far, radiant heat passes through elastic media without heating them, and what support this gives to Leslie's theory of pulsations. The whole subject should be viewed in connexion with the admirable remarks of Sir J. Herschel in his Discourse on the Study of Natural Philosophy, p. 205.

The radiation of heat in vacuo is another point on which further inquiry is much wanted. The greater capacity of air for heat, as it is more rarefied, would occasion a more rapid abstraction from the hot body; and thus in an atmosphere of extreme rarity the cooling ought to be extremely rapid, and this must be accurately estimated in measuring the radiation. But it appears from the experiments of Gay-Lussac, (see Edinb. Phil. Journ. vi. 302,) that when air is reduced to the most extreme degree of rarefaction possible, a very considerable compression makes so little difference in its actual density, that the giving out of heat which ought to take place from diminishing its capacity is absolutely insensible.

But even in this case it is very questionable whether so complete an approach to a real vacuum is obtained as to warrant inferences respecting the radiation of heat in an actual vacuum.

In fact, we want a connected series of determinations to show the order and increase of conducting powers, as connected both with the radiation in and through different media, and the interception which they offer to its passage.

In solids it is presumed no radiation can commence; it is disputed whether it can continue even partially: but conduction goes on rapidly.

In liquids it has been disputed whether there can be radiation; and they are worse conductors than solids.

[page] 301

In elastic media radiation can commence and continue; but they are still worse conductors.

In vacuo it might be presumed by analogy that a yet more free radiation might take place; yet some experiments (as we have seen,) show the contrary; and here there is no conduction.

With regard to that portion of the heat which accompanies or belongs to light, the theory which I originally suggested, (merely as an hypothesis representing the facts) viz. that it was simply the latent heat of light, developed of course when the light was absorbed, is connected with the hypothesis of the materiality of light; but it may be worth inquiry whether it does not apply even better to the elastic æther, in whose undulations light is now proved to consist.

Report on Thermo-electricity. By the Rev. JAMES CUMMING, F.R.S., Professor of Chemistry in the University of Cambridge.

IN communicating to the members of this Society an outline of the progress and present state of Thermo-electricity, I congratulate both them and myself on the allied branches of science having fallen into such able hands, that I should not be justified, even if it were my wish, to extend this Report beyond its immediate subject.

On one point more particularly I am happy,—since we are not so fortunate as to receive instruction from the discoverer himself,—that Dr. Ritchie has undertaken to exhibit and explain to us the recent researches of Mr. Faraday. The continuous electrical currents, now made known to us by these experiments, seem so much more nearly connected with those in the thermoelectric circuit, than with those peculiar either to the common or galvanic electricity, that I should otherwise have thought it incumbent on me to make the notice of them a part of this Report. Divested, as the subject will thus be, of all extraneous matter, I shall therefore be enabled to say all that I think to be really necessary, and yet detain you but a short time from more important communications.

On a review of the labours of different experimentalists on Thermo-electricity, it soon became evident to me, that, to give anything like a luminous account of them, it would be necessary to make some attempt at a classification of their objects. This, I must confess, was no very easy matter; for in this, as in some other branches of experimental inquiry, I have found it difficult,

[page] 302

after reading a detail of an elaborate series of experiments; to discover what object was intended by them.

The first and most obvious inquiry seems to be into the circumstances which are necessary for thermo-electric excitation, or which modify its action. In this respect the original experiment of Seebeck left much room for further investigation. When he had found that a brass wire coiled round the ends of a bar of antimony exhibited magnetic action by the application of heat to one of the extremities of the bar, it was still doubtful whether this effect might not depend either on some peculiarity in one or the other of these metals, on their contact, or on the mode of their juncture.

The remarkable effects produced by helices in the hydroelectric circuit made it not improbable that much might depend on the wire being coiled round the bar. This was soon shown not to be the case, and that a circuit, however formed, provided it were composed of perfect conductors, was all that was necessary.

Reasoning, again, from the analogy of the galvanic circuit, it might have been imagined that as three elements were necessary in the one, so two metallic elements with heat acting the part of the third might be required in the other; but it appeared from some of the earliest experiments, that metallic bars heated in contact with wires of the same metal gave considerable deviations with the galvanometer needle, and therefore that one metal alone sufficed for the development of thermo-electricity. The experiments of Dr. Trail in 1824 may be referred to this class; since, though made with slips of copper attached to the bar of antimony, yet, as the circuit was not completed through the copper, they properly exhibit the thermo-electricity of a single metal. One result, which is too important to be overlooked, is that the application of ice or heat to the centre of the bar produced opposite deviations in two needles placed between the centre and the extremities; whence he infers that "the direction of the compass needle may be considered as the resultant of two forces, the magnetism of composition of the earth, and its thermo-magnetism, which tends to place the needle east and west." How far this coincides with subsequent experiments I shall have occasion to point out to you hereafter.

But the most important researches on the thermo-electricity of a single metal were those made by Yelin in the same year; from which we learn, that all metallic bodies acquire magnetic properties when unequally heated, and that the series of their magnetic intensities when thus excited, is bismuth, antimony, zinc, silver, platina, copper, brass, gold, tin, lead:—that a metal acts differently according as the hot or cold part of it is placed

[page] 303

under the needle;—and that the magnetic action of metals unequally heated depends upon the form given them in casting; for which purpose masses of each metal, in the different forms of prisms, cylinders, and spheres both hollow and solid, were heated successively in different points, and examined by applying magnetic needles to their surfaces. From the different directions of the magnetism in cylinders of bismuth as they were cooled slowly or rapidly, he infers that there is some relation between the crystallization of metals and their magnetic properties. I may observe that I had previously shown that no difference either in the nature or quantity of the deviation could be detected in bismuth under similar circumstances, when forming a circuit with copper wires, &c.; the modification induced by slow or rapid crystallization is confined to the direction of the currents in the bar itself; and since fluid mercury is capable of becoming a thermo-electric element, crystallization is evidently not a primary agent in thermo-electric excitation or conduction, however it may modify its progress.

The latest experiments connected with this branch of the subject, are those of Mr. Sturgeon in 1831, on the thermoelectricity of homogeneous bodies, and the connexion between crystalline arrangement and thermo-electricity. The objects of his two papers appear to be to trace the directions of the magnetic currents in masses of metal, varying the form and the point of excitation; and so far they agree with those of Yelin. With these it appears that Mr. Sturgeon was not acquainted, as he says he is not aware that any experiments are yet before the public, illustrative of thermo-magnetic action in one solitary piece of metal. As a general result, it may be stated that, whether the mass of metal were in the form of a rectangular prism, a cylinder, or a cone, upon heating a point in the periphery of one extremity, the current proceeded longitudinally from the heated point on the same side of the axis, and returned on the other side, accompanied with transverse currents passing in opposite directions nearly at right angles to the longitudinal ones. With a large rectangular plate of zinc, when the heat was applied at one of the angles, the electric current was in the direction of the diagonal advancing, and returned along the edges. In this and all similar experiments, it seems that the direction of the electricity to or from the heated point, depends upon some peculiarity in each metal, which remains to be discovered; but that the course of the currents afterwards, with reference to the figure of the mass, depends solely upon the figures; and I think may be accounted for, by considering the whole as a congeries of wires, from which, ac-

[page] 304

cording to their position with respect to each other and the exterior surface, the heat is conducted away with more or less rapidity;—each portion, so far as the heat extends, acting both as a thermo-electric element and as a conductor of heat, but beyond that space acting simply as a conductor. The effects of crystalline structure in modifying thermo-electric action Mr. Sturgeon considers as arising from the laminæ of each crystal being only in juxtaposition, and that therefore the heat passes more readily through the parts of each than from one to another. This hypothesis it is obvious is inapplicable to metals devoid of crystalline structure, as wires of copper or silver, and still more so to metals in a liquid state: but, by conceiving each wire to be divided into an indefinite number of circular laminæ, we may suppose each of these to act as a layer of cold particles upon the laminæ on one side and of hot upon those on the other, and the total effect of the whole to depend on their aggregate action; each bar or wire acting as an assemblage of an indefinite number of small plates, as the common magnet may be conceived to be composed of an indefinite number of atomic magnets. Still, admitting this mutual action of the metallic particles, the original induction of electricity by heat and its subsequent propagation remain to be explained. This Becquerel conceives may be accounted for on the hypothesis that, whenever a particle of a metal is heated, part of the neutral electric fluid which is attached to it is decomposed, the vitreous fluid being retained, and the resinous driven off and passing into the adjoining particles. In proportion as the heat extends by communication from particle to particle, similar effects take place in each of those that are acquiring heat, and the contrary in those that are losing it. Thus the first effect is only to produce an oscillatory movement of the electric fluid between the adjacent particles; but if the source of heat be permanent, the retrograde movements are prevented, and a continued current takes place. I can only observe as to this theory, that the hypothesis appears to assume the very point that was to be established. I am not aware of any experiments to prove such a decomposition of the electricities of an uninsulated particle of metal.

The next class of experiments to be mentioned are those which relate to the transmission and augmentation of thermo-electricity.

Reverting to the original experiment of Seebeck, the brass wire connecting the extremities of the bar of antimony might act simply as a conductor, or might modify at the same time that it transmitted the electricity, according as the susceptibility to this species of electric excitation was confined to antimony, or

[page] 305

was common to all the metals. In the latter case it was to be expected that the metals might be formed into a series similar to the voltaic; and if such a series were formed, that the order might coincide with that obtained by arranging them according to some of their properties previously discovered. The results have been so far successful as to determine the thermo-electric relations of the metals to each other, (subject to anomalies which it is not now necessary to mention,) but not to connect them with any other of their properties; unless the conjecture of Becquerel be verified, that they are in the order of the specific heats,—in which case the remark of Dr. Ritchie on galvanic electricity will be applicable to this branch of it, and the whole theory of electricity may be intimately connected with that of latent heat.

The real gain to science from the knowledge of this thermoelectric series, is in the increased effect from the proper apposition of the metals; and as it appears that bismuth and antimony in conjunction, as originally proposed by Fourier and Œrsted and myself, are the furthest removed from each other of the available substances, it is not probable that any single pair of elements will be found more efficacious. On endeavouring to obtain an increase of power by augmenting the number of elements, it was soon discovered that this was limited by the want of tension in this species of electricity; in consequence of which, nearly as much was lost by transmission through a number of elements as was gained by their united action; and for the same reason the galvanometer, which had been so efficacious in multiplying voltaic action, was found to be of little advantage by the earlier experimenters. It appears, however, that by the ingenious contrivances of Nobili and Melloni, this difficulty has been overcome, and that they have been enabled to construct a thermo-electric thermometer of almost incredible delicacy. If there be any tension in thermo-electric currents, we may hope that it will be detected by this instrument, in which there may possibly, by the transmission of the electricity through its numerous elements, acting as more or less perfect conductors in their substance and at their junctures, be induced a tension, similar to that in the voltaic series by the passage from metals to fluids. As yet the only semblance of it is that obtained by Fourier and Œrsted in 1823; which was manifested by a slight convulsion of the muscles of a frog placed in the circuit. I am not aware that the experiment has been repeated; and, with all due deference to such able observers, I must doubt its accuracy. Wherever there is tension, however feeble, I should expect it to be attended either with the power of producing

U

[page] 306

heat, or of inducing magnetism. In the magneto-electric currents discovered by Mr. Faraday, which resemble the thermo-electric in the difficulty with which they are transmitted through even metallic conductors, tension has been manifested, accompanied with the faculty of inducing magnetism in steel by the helix; and 1 cannot but suspect that it will be found capable of heating small wires introduced into its circuit. On the other hand, no such effects have been obtained from the thermo-electric current; as I have repeatedly failed (and I believe others have not been more successful,) in magnetizing a needle by a thermoelectric current, of greater deviating power than that from a pair of galvanic elements, which succeeded without difficulty. Similarly, in regard to their heating power, by passing a wire through the axis of Briguet's thermometer, I have detected an increase of temperature in it when connected with a minute voltaic series, but none with the thermo-electric. I cannot therefore but think this a subject worthy of further inquiry.

Having noticed the thermo-electric experiments that have been made on a single mass of metal, and on two metals partly in contact, it only remains for me to call your attention to some very interesting researches on the thermo-electric action of two metals symmetrically united throughout. This part of the subject was first considered by Dr. Trail in 1824; who ascertained that if two plates of different metals were applied together throughout, (which in fact resolves itself into the case of two very short bars connected in the usual manner,) these would form a thermo-electric arrangement. This experiment was varied, and very important consequences deduced from it, by Mr. Christie in 1827. Acting upon a suggestion in the Cambridge Transactions, that as the diurnal variation of the compass needle seemed in some measure dependent on the position of the sun, it might possibly have a thermo-electric origin, he has shown in a paper in the Philosophical Transactions for 1827,—to which I cannot too strongly call your attention,— that these natural phæomena, so far as they can be imitated experimentally, accord with the supposition that the earth and its atmosphere may form a thermo-electric combination put into action by the sun. Imitating this arrangement by a circular ring of copper surrounding a plate of bismuth, and applying heat to a point in the ring, he found that the characters and extents of the deviations, were such as would arise from the polarizing the plate in lines nearly at right angles to the axis of heat,—contrary poles being opposite to each other in the two surfaces; and applying this to represent the state of the equatorial regions of the earth, we should have two magnetic

[page] 307

poles on the northern, and two poles similarly posited on the southern side; the poles of different names being opposed to each other on the contrary sides of the equator.

I cannot refuse myself the pleasure of quoting another result from an apparatus still more resembling the connexion of the earth and its atmosphere. When heat was applied to a point in the equator of a copper shell surrounding a sphere of bismuth, it resulted that the deviation of the end of the needle of the same name as the latitude, was always towards the west when the place of heat was above the horizon, and towards the east when it was on the meridian below. It is unnecessary for me to say to this audience, how perfectly this and the preceding results accord with observation.

I should be happy, if it were in my power, to bring before you a series of researches of equal interest with this paper by Mr. Christie; but, as I am not aware of any of recent date that are worthy of your notice, it only remains for me to point out to you what I conceive to be the present state of our knowledge of thermo-electricity.

Thermo-electricity may be developed in a homogeneous metallic mass, or in two distinct masses of the same metal unequally heated at their point of contact. In the former case the direction of the electricity in the same metal is influenced by the figure; in the latter, as when a metallic disk is touched by a heated wire, the action is anomalous, varying with the metal.

A metallic bar affords as many circuits as there are portions into which it can be divided. Therefore dividing it indefinitely, it may be conceived that each particle has its distinct current, the positive and negative electricities being separated by heat:

The currents between the extremities of a metallic bar are not affected by varying the directions of the intermediate currents.

Charcoal, plumbago, and some, possibly many, of the metallic sulphurets, are capable of thermo-electric excitation. This species of electricity is therefore not confined to the simple metals.

The metals compose a thermo-electric series of which bismuth and antimony are the extremes, and are the most efficacious within certain limits.

A battery may be formed of pairs of such elements, the action of each depending in some degree on the surfaces in contact; and the effect is greater when they are arranged in sequence, as in the voltaic series, than when connected as in Hare's calorimeter; but in neither case increases in the same ratio as the number of elements.

The ratio between the temperatures and the corresponding

U 2

[page] 308

deviations of a needle, in the thermo-electric circuit, is not invariable, the deviations increasing slower than the temperature, the law of which deviation is unknown; hence a pyrometer constructed on thermo-electric principles gives inaccurate indications.

The action of alloys is not intermediate between those of their constituents, and in some instances changes not only its quantity but its character at different temperatures. The action of iron with other metals is anomalous, being positive or negative with the same metal, according to the temperature.

The thermo-electric order of conducting powers in metallic wires, is silver, copper, gold, zinc, brass, platina, and iron; but compared with galvanic electricity, it is conducted with difficulty.

Thermo-electric rotation may be produced by the action of an exterior magnet, but differs from electro-magnetic rotation in not being the result of a continuous action. The action of a magnet upon an indefinite thermo-electric conductor, varies inversely as the distance: hence the same laws seem applicable to hydro- and stereo-electric currents.

The thermo-electric current appears to be incapable of passing through fluid non-metallic conductors, of heating wire placed in its circuit, or of either magnetizing steel, or forming a thermo-electric magnet; and therefore seems to have no assignable tension.

Hence it is somewhat beyond the limits of probability to suppose (as some have fancied,) that by these currents the metals themselves will be decomposed, and that the great revolution in chemistry, commenced by the pile of Volta, will be completed by that of Seebeck.

Report on the recent Progress of Optics. By Sir DAVID

BREWSTER, LL.D. F.R.S. &c.

THERE are few branches of science which have been so irregular in their progress as that which treats of the nature and properties of light. With the exception of some general notions respecting its rectilineal propagation, the equality of the angles of incidence and reflexion, and the refraction of light towards the perpendicular in passing from a rare to a dense, the ancient philosophers contributed little to the advancement of physical

[page] 309

optics. It was not till the second century of the Christian æra, that Claudius Ptolemy, the celebrated astronomer of Alexandria, laid the foundation of this branch of the science, by an accurate examination of the phænomena of refraction. This distinguished individual, to whom Astronomy owes such deep obligations, measured with singular exactness the angles of refraction in water and glass for various incidences from 0° to 80°, and he determined the same angles when the light passed from the one medium into the other. These inquiries were no doubt undertaken in reference to the refractions of the atmosphere, in which, as an astronomer, he felt the greatest interest; and such was the success with which he applied them to this important object, that he gave a theory of astronomical refractions more complete than that of any other astronomer before the time of Cassini. While Tycho in the fifteenth century believed that the refraction of the atmosphere terminated at an altitude of 45°, and while others placed the zero of refraction at the pole of the ecliptic, Ptolemy had shown, 1200 years earlier, that the refraction increased gradually from the zenith to the horizon, and that all stars were elevated by it above their true place: nay, he uses the very same diagram upon which Cassini has founded his theory, and he employs almost the same reasoning as the French astronomer in order to determine the quantity of refraction.

Although the "Optics" of Ptolemy, in which these discoveries are recorded, is known to have existed in the time of Roger Bacon, yet the work appears to have been lost sight of by his successors, till two copies of it were lately found, one in the Savilian Library at Oxford, and the other in the Imperial Library at Paris.

The "Optics" of Alhazen and of his disciple Vitello, though. written a thousand years after that of Ptolemy, contain but very trifling additions to the science. Vitello, indeed, obtained more accurate measures of the deviation of the refracted ray in glass and water; but though his numbers were sufficiently exact for the purpose, he did not discover the constant relation which exists between the sines of incidence and refraction. The discovery of this important law was reserved for Willebrord Snellius, professor of mathematics at Leyden, a young man of the finest genius, who died at the age of 35, before he was able to give his own account of it to the world. The attempt which was made by Descartes to appropriate to himself this discovery is unparalleled in the history of Science; and the means which have been taken in later times to deprive the accomplished Snellius of his legitimate and single claim, present to us one of the most striking examples of national partiality.

[page] 310

The same discovery was undoubtedly made a few years afterwards by Mr. James Gregory; but though he never saw the MSS. of Snellius, as Descartes is believed to have done, neither he nor his countrymen have endeavoured to disturb the laurels which have so long and so justly decorated the Flemish philosopher.

The next important step in the history of Physical Optics was the discovery of double refraction by Bartholinus, a Danish Professor, who published an account of his experiments in 1669. After a careful examination of the phænomena as they appeared in Iceland spar, he discovered that one of the refractions was performed according to the ordinary law of Snellius, and the other according to an extraordinary rule which had not been recognised by philosophers. About nine years after the publication of this work, the celebrated Christian Huygens directed his attention to the same subject. He had long maintained the doctrine that light consists in the vibrations of an elastic medium, and he had succeeded in applying it to explain the rectilineal propagation of light;—the equality of the angles of incidence and reflexion;—the constant ratio of the sines;— the total reflexion of light, and the relation between the reflective and the refractive forces. While he was engaged in these researches, he became acquainted with Newton's discovery of the different refrangibility of light; and though it had a tendency to unsettle his previous opinions, yet he viewed it with an unfavourable eye, and remained firmly attached to the undulatory hypothesis.

In order to explain the phænomenon of two separate refractions, Huygens imagined an hypothesis before he had made a single experiment on the subject. He conceived that the ordinary refraction was produced by spherical emanations of light similar to those which take place in the elastic medium that pervades all transparent bodies, while the extraordinary refraction was produced by spheroidal emanations propagated indifferently through the elastic medium and the solid particles of the body. This singular idea was immediately submitted to the test of experiment; and Huygens had the satisfaction of finding that it represented in the most accurate manner all the phænomena which had been observed by Bartholinus and himself. This law, confirmed by every subsequent inquiry, and certainly one of the most remarkable in physics, is now the law of double refraction for all crystals with one axis; and it deserves to be especially noticed, that it was discovered by the hypothetical method which Kepler had employed with such brilliant success.

[page] 311

It will afford a curious insight into the character of great minds, and at the same time a striking proof of the fallibility and even the weakness of the loftiest intellects, to study the conduct of Newton and of Huygens in reference to their mutual discoveries. We have already seen that Huygens rejected Newton's doctrine of the different refrangibility of light, and also his analysis of the spectrum; and it is well known that he opposed the theory of universal gravitation. The discoveries of the Dutch philosopher were received by our illustrious countryman with equal if not greater distrust. Newton not only rejected the undulatory theory which Huygens had so ably expounded, but he rejected also the mathematical law of extraordinary refraction, which was established by direct experiment,—a law the truth of which was independent of the hypothesis from which it was deduced. Anxious, no doubt, to avoid controversy, Newton did not formally attack the law of Huygens, nor did he call in question the accuracy of his experiments; but without noticing this law he brings forward a new one of his own, without the sanction of any general principles, and without a single experiment in its support. "The unusual refraction," says Newton, "is performed by the following rule," which he proceeds to describe minutely with the aid of a diagram. Now this rule, investigated by the first genius of the age, and with all the powers of the inductive philosophy, has been universally rejected as erroneous; while the law of Huygens, explored by a less gifted mind, and originating in a bold hypothesis, enjoys the splendid triumph of having not only laid the foundation of one of the noblest branches of knowledge, but of having led, by its application, to the establishment of that very hypothesis from which it sprung. Historical truth, however, requires us to add that Huygens did not see the universality of his own law, and that in his investigation of the double refraction of rock crystal, both his experiments and his hypothesis utterly failed him. He announced that the double refraction of this mineral was regulated by two spherical emanations, one of which was a little slower than the other. This strange mistake had however a very natural origin. Taking it for granted that the extraordinary ray must always be the least refracted one, as in Iceland spar, Huygens appears to have measured only the refraction of the least refracted ray, which he found to be regulated by the ordinary law of Snellius. This result was perfectly correct; but he had been working only with the ordinary ray, having assumed most improperly that the other ray would have the same properties as the corresponding one in Iceland spar. By

[page] 312

this oversight Huygens failed in making that fine discovery which was reserved for M. Biot, that the double refraction of quartz differed from that of calcareous spar only in its being regulated by a prolate in place of an oblate spheroid.

After Huygens had established his law of double refraction, and even after he had drawn up his treatise on the subject, he discovered what he calls "a wonderful phænomenon" namely, the polarization of the two pencils of light formed by Iceland spar; and he confesses that other suppositions besides those which he has made will be required to explain it. He acknowleges, however, that he was unable to form any satisfactory conjecture respecting this new property of light, and therefore left the investigation of it to future inquirers. Sir Isaac Newton followed Huygens in this difficult research; but he only stated the fact in another way, when he said that the different sides of the ray had acquired, in passing through the first crystal, different properties which either favoured or prevented its passage through the second.

The subject of double refraction and polarization remained in the state in which we now leave it for one hundred and twenty years, without having received any accession of the least importance. The current of optical discovery, however, was not stopped; its direction only was changed, and during the next century it continued to flow in a more practical and useful channel.

The discovery of the different refrangibility of light which caused Sir Isaac Newton to despair of the improvement of common object-glasses, led himself and his contemporaries to perfect the reflecting telescope. The mistake which he had committed in supposing that all bodies produced spectra proportional to their mean refraction, was detected a few years after his death by Hall and Dollond, who discovered the different dispersive powers of bodies, and who were both led, by independent inquiries, to the invention and construction of the achromatic telescope. At a later period the discovery of the irrationality of the coloured spaces in the spectrum by Clairaut and Boscovich, furnished Dr. Blair with the general principle of the aplanatic telescope, and enabled him to construct fluid object-glasses, in which a perfect correction of colour was effected. These two instruments were doubtless the most valuable gifts which one science has presented to another; and the kindred subjects of navigation and practical astronomy exhibit in the perfection of their methods the great benefits which they have thus received.

After having slumbered for a hundred years, Physical Optics began to revive about the close of the eighteenth century. The events of the French Revolution had summoned from a state

[page] 313

of hybernation all the genius and talent of France. Men of high endowments and lofty intellect found an elevated place in society; and the establishment of the National Institute, the mightiest organization of intellectual power that history records, gave a new and a vigorous tone to scientific research. Even amid the convulsions and atrocities of that awful period, Science shot forth some of her brightest radiations; and in the moral and religious darkness which prevailed, her evening star was the only surviving emblem of heaven. The impulse thus given to knowledge was propagated over all Europe; and the quarter of a century which succeeded was one of the brightest periods in the history of philosophy.

Although Sir Isaac Newton had determined the law of the colours of thin plates, and had begun to examine the phænomena of inflexion, yet his experiments on both these subjects, but especially the last, were left imperfect. The young and ardent minds that were now ambitious of following Newton in his career of discovery, chose the inflexion of light as the subject of their first achievements. Lord Chancellor Brougham, Dr. Young, and Mr. Jordan gave an account of their respective labours on this subject in elaborate and valuable Memoirs. Lord Brougham and Mr. Jordan were warm admirers of the Newtonian theory of light, but Dr. Young had adopted the undulatory hypothesis of Huygens; he was therefore led to examine the phænomena under a different aspect, and was conducted to an experimental demonstration of the general law of interference, which was first observed by Grimaldi, and had been subsequently applied by Dr. Hooke to the explanation of the colours of thin plates. About the same time Ritter had discovered the deoxidating rays of the spectrum; Sir William Herschel had observed the invisible rays of heat; and Dr. Wollaston had detected fixed lines in the spectrum, and had confirmed by new experiments the Huygenian law of double refraction. During the same period the Marquis Laplace, who was a keen supporter of the Newtonian hypothesis, had referred the deviation of the extraordinary ray in doubly-refracting crystals to those attractive and repulsive forces by which the ordinary refraction and reflexion of light are performed; and by considering the force which acts upon the extraordinary ray as a function of the angle which the refracted ray forms with the axis of the crystal, he obtained formulae perfectly coincident with the Huygenian law.

The attention of philosophers was now anxiously directed to the subject of double refraction, and in 1808 the Institute of France proposed it as the subject of their Physical Prize. This prize was adjudged in 1810 to M. Malus, Colonel of the Imperial

[page] 314

Corps of Engineers, who not only composed a most valuable Memoir on double refraction, but enriched the subject with a discovery which laid the foundation of a new science. Having accidentally turned a doubly-refracting prism to the windows of the palace of the Luxemburg, which were at the time illuminated by the setting sun, he was surprised to observe that one of the double images of the windows vanished alternately during the rotation of the prism; and after various fruitless speculations on the cause of this singular phænomenon, he was conducted to the great discovery, that light reflected at a particular angle from transparent bodies is polarized like one of the rays produced by double refraction. This singular result opened a wide field of inquiry to philosophers: and the successive labours of Malus, Arago, Biot, Fresnel, and Cauchy in France; Seebeck and Mitscherlich in Germany; and Young, Herschel, and Airy in England—present a train of research "than which," as a distinguished philosopher remarks, "nothing prouder has adorned the annals of physical science since the development of the true system of the universe."

It would be impossible in a brief Report like the present, to convey even a general idea of the relative labours of these eminent philosophers,—of the new laws which they have established,—of the splendid phænomena which they have brought to light, or of the valuable applications which they have made of their discoveries. But, without giving offence to those who survive, we may distinguish two names, Malus and Fresnel, already marked out by the melancholy preeminence of a short and brilliant career,—-names which will ever be united in the history of Science by the extraordinary similarity of their lives and labours, of their honours and their misfortunes.

Devoted from their earliest years to the study of the sciences, they entered with ardour on the same field of inquiry,—the double refraction of light. Versed in the same abstract acquirements, they were both called to the situation of Examiner of Natural Philosophy and Geometry in the Polytechnic School. Their sovereign conferred upon them the same honours, the Cross of the Legion of Honour. The early discoveries of each were crowned by the Physical Prize of the National Institute. Their latest labours were rewarded by the Royal Society of London with the Rumford Medals; and, as if Providence had invigorated their exhausted frames to enable them to receive On their death-bed this brilliant trophy of their success, they both sunk beneath its banners, the one in the 37th and the other in the 39th year of his age. Thus triumphant in the same field, crowned with the same laurels, and doomed to the same early

[page] 315

grave, their names will be indissolubly embalmed in the sympathies of their countrymen; and in recounting their brilliant discoveries, the philosophers of every age will mingle their tears with their admiration.

Heu fortunati ambo!

Nulla dies unquam memori vos eximet ævo.

As the discoveries of Fresnel and his contemporaries have been fully described in several works in our own language, it would be an unprofitable task to give any account of them at present. The nature of this Report, however, requires me to notice those more recent discoveries which are less accessible and less generally diffused; and to endeavour to point out those new paths of discovery which the young philosopher may most successfully pursue, and those applications of optical science which are likely to be most useful in extending the power of man, in throwing light upon other branches of knowledge, and in investigating the structure and properties of organized matter.

I regret that the first part of this task should embrace the researches of so few labourers; and I fear that the future holds out but little prospect of any increase either in their number or in their efficiency. Pursuits more popular and more generally appreciable, and employments incompatible with scientific inquiry, have allured from Optical research many of those distinguished individuals who were most able to grapple with its difficulties; and within these few months Science has lost Dr. Seebeck* of Berlin, one of the most able and successful discoverers of the present century.

The only individuals who have been recently and actively engaged in the higher departments of physical optics, are Mr. Airy of Cambridge, and M. Cauchy of the Academy of Sciences.

In examining the two rays produced by the double refraction of quartz, Mr. Airy was led to a discovery which we consider as one of the most important in its results, and one of the most beautiful in its phænomena that has yet been made in this branch of optics. The circular polarization of the two rays along the axis of quartz had been studied by different philosophers, and had been explained by Fresnel with singular ingenuity on the principles of the undulatory theory. No attempt, however, had been made to account for the existence of this property only in the rays which pass near the axis of the crystal, or to define the limit where the circular polarization ended and the

* Dr. Thomas John Seebeck was born at Reval on the 9th of April 1770. He died at Berlin on the 10th of December 1831, in the 62nd year of his age.

[page] 316

plane polarization commenced. Fresnel, and all who have written on the subject, seem to have shrunk from this difficulty; but Mr. Airy saw that the two kinds of polarization must have some connecting link, and by the aid of theory and experiment he succeeded in discovering it. In place of the two rays in quartz consisting of plane polarized light, as was universally believed, Mr. Airy has shown that they both consist of elliptically polarized light, the greater axis of the ellipse for the one ray being in the principal plane of the crystal, and the greater axis of the other perpendicular to that plane. One of the rays he found to be right-handed elliptically polarized, and the other left-handed elliptically polarized. The proportion of the axes of the ordinary ray is more nearly one of equality than the proportion of the axes of the extraordinary ray, each proportion being one of equality when the direction of the ray coincides with the axis, and becoming more inequal with the inclination according to a law not yet discovered. The results calculated from the theory are in perfect accordance with those which Mr. Airy has obtained from very nice and difficult experiments; so that we may regard this beautiful and singular property of the two rays of quartz as perfectly established.

Mr. Airy has still more recently discovered a remarkable modification of Newton's rings, when they are produced by a lens laid upon a polished metallic surface. This modification possesses much interest when considered only as a detached fact; but its importance is greatly enhanced by its direct bearing upon the two rival theories of light. On the Newtonian hypothesis the colours of thin plates are produced solely by the light reflected from the second surface of the plate; whereas, according to the undulatory hypothesis, they depend on the interference of the light reflected at the second surface, with the light reflected at the first surface. Hence, if we can by any means destroy the light reflected at the first surface, the rings ought to vanish, according to the undulatory hypothesis; while they ought still to appear, according to that of Newton. Mr. Airy conceived the happy idea of using polarized light, which was freely reflected from the second surface, while it was incapable of being reflected from the first; and upon trying the experiment, he found that the rings disappeared,—a result which he regards as "perfectly inexplicable on any theory of emission, and as affording satisfactory evidence that the rings are produced by interference only." We have no hesitation in admitting that this experiment is inexplicable on the Newtonian hypothesis of fits, and that the action of the two reflected pencils, either on each other or on the retina, is necessary to the

[page] 317

production of the rings; but, as Dr. Young and others have allowed that the doctrine of interference is reconcileable with the doctrine of emission, the disappearance of the rings is not necessarily inexplicable on any theory of emission.

I regret that I am unable to give any satisfactory account of the very important optical discoveries of M. Cauchy, and I am not aware, indeed, that he has published any detailed account of his researches. In one of the Memoirs which he has lately printed, among those of the Academy of Sciences, he refers to three important results, which he has obtained from the undulatory theory.

1. The deduction of the law of the tangents which connects the polarizing angle with the refractive power of the body.

2. The explanation of the phænomena of dispersion.

3. The existence of a triple refraction.

The inability of the undulatory theory to explain the phænomena of inequal refrangibility, is almost the only exception to its universal application in accounting for the most complicated phænomena of light. Various attempts, though not very successful ones, have been made to remove this difficulty. Dr. Young supposes that the material particles of transparent bodies are susceptible of permanent vibrations, somewhat slower than the undulations which produce them, and that the velocity of the original undulation will be diminished in proportion to their frequency. The Rev. Mr. Challis, adopting Dr. Young's idea, has endeavoured to explain the manner in which the undulations of the æther within bodies are modified by their material atoms. He supposes that a sensible reflexion takes place at every interruption of continuity in the medium; and he infers that the mean effect produced by a retarding cause proportional to the reflective power of the atoms, will be to make the condensation corresponding to a given velocity, greater in a certain proportion than in free space, and to diminish the velocity of propagation in the same proportion. Mr. Airy has more recently endeavoured to remove this difficulty, by supposing that in refracting media there may be something depending on time, which alters their elasticity, in the same manner as in air the elasticity is greater with a quick than with a slow vibration of particles.

An anonymous writer, in a very recent Number of The Annals of Philosophy, has proposed another hypothesis for obtaining a difference of elasticity. He supposes that the æther accumulates itself round the particles of transparent media, and forms spheres of a density increasing towards their centres; and he infers that a succession of vibrations communicated through a

[page] 318

medium thus constituted, will give rise to new vibrations propagated with various velocities corresponding to those of the different rays in the spectrum.

The complete removal of such a difficulty from the undulatory theory by the analysis of M. Cauchy, must be regarded as one of the greatest steps in physical optics; and philosophers will look forward with the most intense anxiety to the development of that part of the same theory which renders necessary the existence of a triple refraction.

Such is a very brief notice of some of the most recent discoveries and views in physical optics. It would now be a pleasing task to point out the desiderata of this branch of science, and to indicate the locality of those rich mines which yet remain to be explored, did we descry any young and active labourers who were likely to gird themselves in earnest for so difficult a work. But when we see those who are best fitted for such inquiries, either abandoning altogether the study of light, or pursuing it in professional harness as a sort of contraband adventure, we almost despair of seeing acquired for our country the glory of any fresh achievements. Could we count on the unfettered labours of two of our most eminent natural philosophers, who have already evinced such high capacity for optical discovery, we might still cope with foreign genius, even though it does repose under the sunshine of Royal favour and of Academic ease.

There is scarcely any branch of the subjects of double refraction and polarization which does not afford the richest fields of discovery. Even the theory of undulations, with all its power and all its beauty, is still burthened with difficulties, and cannot claim our implicit assent. It has not yet brought under its dominion the phænomena of elliptic polarization in all its varieties, from the rectilineal polarization of transparent bodies, to the almost circular polarization of pure silver. It has not explained the singular influence of the force of double refraction over the force which polarizes reflected light; and it has great difficulties to struggle with, in accounting for certain phænomena of absorption, to which I shall presently have occasion to refer.

The determination of the physical data (or the physical constants, as Mr. Babbage calls them,) of these departments of science, constitutes a new and almost untrodden field, which may be successfully cultivated by almost every variety of talent. The refractive indices of the two pencils in all crystallized bodies, measured in reference to fixed points in the spectrum, as has been lately done by Rudberg;—the angles at which light is polarized by reflexion from crystallized and uncrystallized

[page] 319

surfaces;—the inclination of the resultant axes of crystals having double refraction, for different rays of the spectrum;—the dimensions of the ellipse which regulates the polarization of metals and their alloys;—the circularly polarizing forces of fluids and solutions;—and the refractive and dispersive powers of ordinary solid and fluid bodies, measured according to the method of Fraunhofer,—are some of the points to which we would call the attention of young and active observers.

But important as these determinations would be in a scientific point of view, and particularly in the renovation of mineralogy, the application of the principles of double refraction to the examination of structures, is pregnant with a still higher interest. The chemist may perform the most dexterous analyses;—the crystallographer may examine crystals by the nicest determination of their forms and cleavages;—the anatomist and the botanist may direct the dissecting knife, and use the microscope with the most exquisite skill;—but there are still structures in the mineral, the vegetable, and the animal kingdom, which defy all such methods of examination, and which will yield only to the magical analysis of polarized light. A body which is quite transparent to the eye, and which appears upon examination to be as monotonous in its structure as it is in its aspect, will yet exhibit under polarized light the most exquisite organization, and will display the result of new laws of combination, which the imagination even could scarcely have conceived. Like the traveller who has visited an unknown land, polarized light emerges from bodies bearing with it the information it has acquired during its passage, and indicating the structures through which it has passed, when put to the question of optical analysis. As an example of the utility of this agent in exploring mineral, vegetable, and animal structures, I may refer to the extraordinary organization of Apophyllite and Analcime; to the symmetrical and figurate deposition of siliceous crystals in the epidermis of equisetaceous plants; and to the wonderful variations of density in the crystalline lenses, and the integuments of the eyes of animals.

One of the finest fields of optical inquiry, and one almost untrodden, is that of the absorption of definite rays of the spectrum by the specific action of the material atoms of those bodies through which light is transmitted, or from which it is reflected. The discovery of dark lines in the solar spectrum, is certainly one of the finest which has been made in the present century, whether we view it in its theoretical bearings, or in its practical application to the construction of the achromatic telescope, and to the determination of all optical data which depend

[page] 320

on coloured light*. Fraunhofer found that the spectrum formed by solar light is crossed with numerous dark lines of different thicknesses, while the spectrum of artificial white flames contains all the rays which are thus wanting. Fraunhofer counted about 590 of these lines; and in a fine map of the spectrum which he has published, he has inserted the strongest of them, amounting to about 354. Some of these lines he found to be entirely black, while others were darker than the rest of the spectrum. From various experiments to which he submitted them, he concluded that they have their origin in the nature of the light of the sun, and that they cannot be attributed to illusion, aberration, or any other secondary cause. Sir John Herschel, taking a wider view of the subject, remarks, "that it is no impossible supposition that the deficient rays in the light of the sun and stars may be absorbed in passing through their own atmospheres; or, to approach still nearer to the origin of the light, we may conceive a ray stifled in the very act of emanation from a luminous molecule by an intense absorbent power residing in the molecule itself; or, in a word, the same indisposition in the molecule of an absorbent body to permit the propagation of any coloured ray through or near them, may constitute an obstacle in limine to the production of that ray."

For reasons which I may have an opportunity of explaining in another communication, I conceive that the original light of the sun is continuous from one end of the visible spectrum to the other, and that the deficient rays are absorbed by the gases

generated during the combustion by which the light is produced. But whatever be the manner in which these dark lines are occasioned, it is manifest that while they are of the highest value as affording fixed points in the spectrum, they render the sun's light absolutely unfit for experiments on absorption. We cannot, for

* In the spectrum formed by a narrow "beam of day-light," Dr. Wollaston had, previously to the year 1802, discovered seven lines, which he has designated by the letters A, B, f, C, g, D, E, the first line being, according to his observations, the extreme boundary of the red rays of the spectrum, and E the extreme boundary of the violet rays. The correspondence of these lines with those of Fraunhofer, I have, with some difficulty, ascertained to be as follows:
A, B, f, C, g, D, E, Wollaston's lines.
B, D, b, F, G, H, Fraunhofer's lines.
There is no single line in Fraunhofer's drawing of the spectrum, nor is there any in the real spectrum coincident with the line C of Wollaston; and, indeed, he himself describes it as not being "so clearly marked as the rest." I have found, however, that this line C corresponds to a number of lines half way between b and F, which, owing to the absorption of the atmosphere, are particularly visible in the light of the sky near the horizon.
In order to have seen the lines B and H of Fraunhofer, especially the last, Dr. Wollaston's "beam of day-light" must have come from a part of the sky very near the sun's disc.

[page] 321

example, examine the action of absorbent media on any one of the 590 rays which are deficient; and there is no possible way of recognising them, for the purpose of examination, in the spectra of those white artificial flames where they all exist.

This difficulty, however, has been completely removed by the discovery which I have lately made, of a gaseous substance which produces more than a thousand dark lines in the spectrum of ordinary flames, and thus renders artificial light more valuable even than that of the sun for the determination of optical data, while it enables us to study the action of material bodies upon all the defective rays of solar light. I have mentioned this experiment at present, in order to point out its bearings upon the two rival theories of light. On the Newtonian hypothesis of emission, the fact may be thus stated:—When a beam of white light is transmitted through a certain thickness of a particular gas, a thousand different portions of that beam are stopped in their passage, in consequence of a specific action exerted upon them by the material atoms of the gas,—an action which is powerfully assisted by the simple application of heat. Such a specific affinity between definite atoms and definite rays, though we do not understand its nature, is yet perfectly conceivable; and we may render it more easy of reception by hazarding the conjecture, that the particles of light itself are identical with the ultimate molecules of bodies, and that similar atoms in each may again unite when brought within the spheres of their mutual attractions.

In the language of the undulatory theory, the same fact may be thus expressed. A thousand different waves or rays of light of different velocities or refrangibilities, are incapable of propagating undulations through the æther of a transparent gas, while all waves or rays of intermediate velocities and refrangibilities are freely transmitted through the same medium: that is, a wave of red light, the 250 millionths of an inch broad, and another wave of the same light the 252 millionths of an inch broad, are capable of transmitting vibrations freely through the gas, while another red ray the 251 millionths of an inch produces vibrations which are entirely stopped by the medium. There is no fact analogous to this in the phænomena of sound, and I can form no conception of a simple elastic medium so modified by the particles of the body which contains it, as to make such an extraordinary selection of the undulations which it stops or transmits. We may suppose, indeed, that æther is a compound medium, consisting of other media, whose particles are the ultimate atoms of matter, and that the undulations of the same æther in transparent bodies are somehow affected by the affinity of similar atoms in the æther

X

[page] 322

and in the refracting body*; but this only removes the difficulty a step further, and leaves the mind impressed with the conviction that the production of such a system of defective rays by the action of a gaseous medium presents a formidable difficulty to the undulatory theory.

But whatever hypothesis be destined to embrace and explain this class of phænomena, the fact which I have mentioned opens an extensive field of inquiry. By the aid of the gaseous absorbent, we may study with the minutest accuracy the action of the elements of material bodies in all their variety of combinations, upon definite and easily recognised rays of light, and we may discover curious analogies between their affinities and those which produce the fixed lines in the spectra of the stars. The apparatus, however, which is requisite to carry on such inquiries with success cannot be procured by individuals, and cannot even be used in ordinary apartments. Lenses of large diameter accurate heliostates, and teleseopes of large aperture are absolutely necessary for this purpose; but with such auxiliaries it would be easy to construct optical combinations, by which the defective rays in the spectra of all the fixed stars down to the tenth magnitude might be observed, and by which we might study the effects of the very combustion which lights up the suns of other systems.

Report on the Recent Progress and Present State of Mineralogy. By W. WHEWELL, M.A., Fellow and Tutor of Trinity College, and late Professor of Mineralogy in the University of Cambridge.

MINERALOGY may be said, in a certain sense, to have continued to be a popular science ever since the time when Werner and Haüy inspired their pupils with so much enthusiasm and activity. During the course of the subsequent years very many persons have employed themselves in making collections of minerals, public and private; in arranging and naming the specimens; in referring their forms and characters to the types of acknowledged species. In England, as well as elsewhere, our best

* This supposition is countenanced by the remarkable fact which I have placed beyond a doubt,—that there are in different parts of the spectrum two or more sets of rays which have the same refrangibility, or which undulate with exactly the same velocity; and yet one of these sets of rays will freely permeate certain transparent bodies, or excite undulations in ita æther, while the other sets re absorbed, or are incapable of propagating undulations through the body.

[page] 323

chemists have frequently analysed mineral specimens; and we have had here persons at least as skilful as have appeared in any other country, who have disentangled the crystalline forms and examined the optical properties of minerals, and have thus established differences and identifications among certain varieties of crystalline bodies.

But on the other hand, mineralogy cannot be said to have been of late a popular science, in that higher sense in which we use the phrase, when we apply it to the sciences in which striking advances in theory, new and widening views, and the bright promise of future progress, attract the attention of all, learned and unlearned,—draw to them the energies of those who feel within themselves the vocation of discovery,—and communicate a feeling of scientific exultation and hope, even to those who have the most imperfect knowledge of the nature of the acquisitions, and of the grounds on which more is expected. Such sciences, in our own time, optics, geology, and chemistry have been and are; such, at least in our own country, mineralogy of late has not been.

It is not difficult to point out some of the reasons of the comparatively slow and undistinguished advance of mineralogy for the last few years. Nothing could be more brilliant than the prospects which appeared to open themselves to this science a few years back, at the epoch of Werner and Haüy. The German Professor gave a fixity and clearness to the determination of minerals by external characters, far exceeding anything which had been taught before; he introduced a system of classification which appeared to lend itself very happily to the known relations of minerals; and he announced the possibility of distinguishing, by the mineral characters of the mountains of the earth, the place which their strata occupied in an invariable chronological series, their meaning as the record of remote but ascertainable epochs in the physical history of the globe; —an application of mineralogy, which of itself was sufficient to give to the study a most attractive dignity and interest.

The French crystallographer, on the other hand, laid before his hearers a science which detected the most beautiful symmetry, simplicity, and constancy, in the midst of apparent complexity and instability; which undertook to determine the forms and laws of aggregation of the component atoms of bodies; and which boasted that, in the most remarkable manner, its predictions and suggestions, founded on differences which the unassisted eye could not appreciate, had been confirmed by the testimony of chemical analysis, summoned as a witness for that purpose. It appeared therefore in the highest degree probable, that mineralogy would be found to be, on the one

X 2

[page] 324

hand, a necessary vestibule to geology, and on the other, an entrance to passages, by which a new way was to be opened to the most recondite questions of chemistry and physics.

This promise has undoubtedly hitherto not been fulfilled; on the contrary, the mineralogist appears in some measure to have been disappointed of the advantages anticipated from both his allies, the geologist and the chemist. The former is now far from considering the mineralogist as his main supporter: conchology, zoology, botany, hydrography and general physics; are held to be at least as important as mineralogy, to the examination of the strata of the earth; and our geological teachers, in a playful spirit of exaggeration, have sometimes said that a person may be too good a mineralogist to be a good geologist. In his appeals to the chemist, the student of the mineral kingdom has always had his claims to assistance allowed; but chemistry is very far indeed from having done for him what he might have hoped it would do; not to mention, that the mere chemist seldom bestows a close and technical attention on that peculiar train of characters, which is the basis of the mineralogist's knowledge. Instead of our knowing exactly the chemical constitution of every mineral species; of finding chemistry ever ready to confirm the arrangements and classifications otherwise made, or if not, to offer something steady and unexceptionable in their place, we find that now, forty years after Haüy began to compare the results of crystallography and chemistry, we have very few minerals of which the chemical constitution is not liable to some dispute;—scarcely a single species of which the rule and limits are known, or in which two different analyses, taken at random, might not lead to different formulæ;—and no system of classification which has obtained general acceptation, or is maintained, even by its proposer, to be free from gross anomalies.

Berzelius has given to one new mineral species an appellation derived from the Greek word for shame, (αισχυνη,) acknowledging a sort of disgrace to fall upon science from the analysis of this mineral; inasmuch as two of its elements of very different natures (titanic acid and zirconia,) cannot be separated so as to determine their relative quantities. If we were to give this name to all the kinds of minerals of which the chemist cannot tell us the exact constitution, eschynite would be a large family instead of a single species.

This decided check in the progress of the science has, I think, without question, very much damped the interest with which mineralogy, as a branch of natural philosophy, has been looked upon in England. Indeed this feeling appears to have gone so far, that all the general questions of the science excite with us scarcely any notice whatever. The value of a method

[page] 325

of classification seems to be looked upon as a point not worth discussing; any one method is considered as good, or as bad, as any other. This opinion indeed is openly maintained by some of our best mineralogists. Their labours have been employed solely and exclusively in the crystallographical and chemical analysis of particular species; and I am not aware that any attempt has been made among us to establish any proposition including a class of species of minerals, with the exception of Sir David Brewster's optical researches.

Such is the state of the case in England.—But a more forward and hopeful spirit appears to have prevailed for some time in other countries, especially Sweden, Germany, and more recently France. It may therefore be of service to point out what is the progress which has been made in this branch of knowledge, and what are the views respecting it which have been opened during the last few years in Europe at large.

The subject may be conveniently considered under the following heads.

1. The Physical characters of minerals; as hardness, specific gravity, lustre, &c.

2. Crystallographical speculations, and their application to minerals.

3. The Optical properties of minerals.

4. The Chemical constitution of minerals.

5. The Classification of minerals.

6. Miscellaneous researches and observations; as the discovery of new minerals, the identification or discrimination of old ones, the determination of their crystalline forms, &c.

I purposely omit all that refers to the localities of minerals, as more properly pertaining to the domain of geology; and all that regards their economical uses, and the processes of metallurgy, as forming in itself a distinct subject. The study of the properties which we have now some hope of referring to general laws,—the optical and chemical properties of minerals, —is a topic sufficiently ample for the present occasion.

1. Physical characters.

The discrimination of minerals by their most obvious properties of colour, lustre, weight, hardness, was naturally attended to in the first attempt to obtain some distinct and connected knowledge with regard to those substances. The study of such characters has now been prosecuted far enough and long enough to show that a systematic and solid mineralogy cannot be formed by attention to these alone; and that crystalline form and chemical constitution must be the main elements of our mineralogica! science. Still, the more vague external characters by no

[page] 326

means deserve to he neglected; nor will they be so by any who study the actual minerals with persevering and close observation.

From the nature of this portion of mineralogical study, it is scarcely susceptible of much speculative progress. The increase of the personal skill and sagacity of the observer by practice is the best result of its cultivation. Yet some improvements in method may be pointed out as having been recently made. Werner, eminently acute in his observation of sensible qualities, gave fixity to his discriminations by the introduction of an appropriate nomenclature. His pupil Mohs, formerly his successor at Freiberg, and now Professor at Vienna,—one of the most gifted of his disciples in the same way,—has attempted to fix one of the most important characters, hardness, by a numerical scale. In this scale, the hardness of common talc is 1, of gypsum 2, of calc spar 3, of fluor spar 4, of apatite or asparagus stone 5, of felspar 6, of quartz 7, of topaz 8, of corundum 9, of diamond 10;—thus, to say that the hardness of a mineral is 51/2 indicates that it scratches apatite and is scratched by felspar. Prof. Breithaupt of Freiberg, the pupil and successor of Mohs, has proposed to put 12 degrees in this scale instead of 10, introducing mica between gypsum and calc spar, and sodalite between apatite and felspar, as intermediate degrees. It has been observed by others, that the hardness of several minerals is different in different parts, and even in different directions; thus kyantte gives a different value of the hardness, according as we scratch it along or across the direction of the axis.

The specific gravity has also been scrupulously attended to by the same school of mineralogists, and both Mohs and Breithaupt have determined very minutely the value of this element for very extensive series of minerals. Beudant also has paid great attention to this subject; he has ascertained by experiment (1. x. 331.) that large crystals, and especially bacillary masses, have a smaller specific gravity than small crystal; and he hence recommends us to reduce minerals to powder previously to finding their specific gravity, in order to avoid the influence of these differences in the mode of aggregation. Magnus found that garnet and similar minerals when melted and again solidified in a glassy but uncrystalline state have their density diminished; the Greenland garnet, for instance, was in this manner reduced from sp. gr. 3·9 to 3·05.

In the observation of the colour and lustre of minerals, we have hitherto been left to the unassisted eye and judgement. It was the object of an instrument described and exhibited by Sir David Brewster, at the last Meeting of the British Asso-

[page] 327

ciation, to make this observation more precise and delicate. The principle of the instrument is, to observe, not the whole light reflected from the surface of the mineral, but the excess of light which remains undestroyed when we apply to the surface a lamina of liquid differing but slightly from the mineral in its refractive power: the differences of lustre and colour in minerals, may thus become much more sensible than when the whole effect is compared.

The different kinds of lustre,—glassy, fatty, pearly, adamantine, metallic,—undoubtedly depend upon optical differences in the surfaces, which differences have not however as yet been clearly explained. Professor Breithaupt is in the habit of showing, by the superposition of a number of watch-glasses, that the pearly lustre results from the lamellar structure of a transparent mass. The very curious difference between the optical properties of the surfaces of metals, and of transparent bodies, has been traced, on different roads, by Sir David Brewster and by Professor Airy; and both agree in considering the optical properties of the diamond as intermediate between the transparent and the metallic character; though they do not agree in their representation of the peculiar laws which the diamond discloses. When the connexion of these properties with those of other bodies is clearly made out, we shall probably learn more distinctly than we now can, what is the precise distinction of metallic, adamantine, and vitreous lustre.

The more distinct cleavages of minerals are among their most important characters, and the less distinct are also of value. Sir David Brewster has suggested a method of obtaining cleavages too indistinct to be made visible in any common way, by tearing the surface of the mineral with a dry file. In this manner he made obvious a cleavage of calc spar in the direction of the long diagonal of each of the rhombic faces.

We may notice here, also, the ingenious mode of mechanical analysis described by M. Cordier, and successfully employed by him in the examination of rocks of various kinds which had been considered as homogeneous substances, but which are in fact aggregates of small crystalline portions of various simple minerals. The specimen is reduced to minute fragments, rather by pressure than by trituration*, and the particles of different kinds, being separated by differences of specific gravity or appearance, are examined in various ways, and especially by means of the blowpipe. This method was found to be particularly applicable to the discrimination and discussion of rocks of a trappean character.

* Journ. Phys. 1816, pp. 82, 83.

[page] 328

2. Crystaliegraphy.

Though no change has since been made with reference to the crystalline forms of minerals which has excited so much popular notice as Haüy's establishment of the fixity of forms and the laws of their derivation, the subject has undergone a complete change since his time; and a principle of classification of forms has been introduced, so scientific and yet so simple, that it is irresistibly superseding the older Haüyian arrangements, and the more so, as it is strikingly confirmed by the optical properties of crystals. I speak of the division of forms into systems of crystallization; namely, the tesseral; the tetragonal, or square pyramidal, or pyramidal of Mohs; the rhombic, or oblong pyramidal, or prismatic of Mohs; the rhombohedral of Mohs, or hexagonal of Naumann; and the monoklinohedral, diklinohedral, and triklinohedral of the last-named writer. Some notion may perhaps be formed of the nature of these distinctions from the following representation.

If we conceive a square steeple with all the four sides of the walls and roof exactly alike, so that every slope and face which occurs on one side, occurs similarly on the other three; we have before us a form belonging to the square pyramidal system.

If instead of this we imagine a house of which the two ends are like each other, and the two sides also precisely like each other, but different from the former, this will belong to the oblong prismatic or rhombic system.

If again we conceive a triangular pillar, as an ancient tripod, its three sides being similarly cut and ornamented; this will belong to the rhombohedral system*. In fact, its three faces may be terminated by slopes which may meet and form an apex resembling in all respects the apex of a rhombohedron. And if each of its three faces be formed into an edge by planes sloping to the right and left, the form may be thus converted into a six-sided pillar with no loss of its regularity.

If we conceive the form of the house of which we spoke as representing the prismatic system to be made less regular by sloping its end walls in the direction of one end; we have the monoklinohedral system; and if the side walls slope also, we may have thus the diklinohedral and triklinohedral forms.

The tesseral or tessular system includes the forms which are derived from the regular solids of geometry, the cube, the octahedron, the dodecahedron.

This distinction of different kinds of forms, is one founded on the most general relation of their parts, and regulated by the

* The rhombohedral and rhombic systems are quite distinct. A rhombic prism has its base a rhomb;—a rhombohedron has all its sides equal rhombs.

[page] 329

degree and kind of their symmetry. The claim of priority in introducing this classification of forms has been a subject of controversy between Prof. Mohs and Prof. Weiss. However this question may be decided, the merit of this valuable simplification rests between them; and all must allow the propriety with which Prof. Naumann of Freiberg, the author of the best recent system of Crystallography, has dedicated his work "to Mohs and Weiss, the Coryphæi of German crystallographers."

The distinction of systems is now generally adopted. Thus Germar, (1830,) one of the most recent authors, has the tessular, pyramidal, prismatic, and hexagonal systems, each subdivided into homohedral and hemihedral, as all or half the faces occur; —the oblique prisms are considered as hemihedral and tetartohedral right prisms, according to the method of Mohs, whose notation also is retained. In England, the distinction of systems of crystallization has not been explained, so far as I am aware, except in Mr. Haidinger's translation of Mohs.

Crystallography is essentially a mathematical subject. The striking mixture of simplicity and complexity which here, as in other parts of nature,—but yet more here than in any other part of nature,—offers itself to our notice, depends upon the combination of the primary forms belonging to the above systems with the geometrical and numerical laws by which other forms are derived from these. To trace the properties of such derived forms, and of their combinations, necessarily requires some considerable portion of mathematical calculation, which may however be of several kinds. Spherical trigonometry, solid geometry, and analytical geometry of three dimensions, may any of them be made to answer the purposes of the crystallographer. Haüy and Mohs, proceeding in the manner which, of the three, implied the least extended acquaintance with mathematics, employed in most instances particular constructions and calculations founded on solid geometry; and though they thus want the conciseness, beauty, and generality of other methods, they are perhaps, in consequence of this, intelligible to a wider circle of students. Monteiro, Bournon, Cordier, Soret, and others, have followed the method of Haüy; and denominations and notations borrowed from it are still common in our catalogues. Phillips also, so far as he referred to any method, employed that of Haüy; but his extraordinary merits consisted rather in determining the angles and forms of individual specimens and species, than in referring them to any general law.

Prof. Hausmann of Göttingen, a pupil of Mohs, has laboured in the spirit and according to the method of his master; as has another distinguished mineralogist from the same school, Mr. Haidinger.

[page] 330

Mr. Brooke has, in a great measure, employed the formulæ of spherical trigonometry, in which he has been followed by others. This method has the great advantage of enabling us immediately to perform all our calculations by the help of logarithmic Tables.

The most scientific mode of treating the subject, which coists in reasoning by means of the equations to the planes according to the methods of analytical geometry, was employed from the first by Weiss. It has been adopted by Mr. Levy, and by a number of German writers, as G. Rose, Kupffer, Köhler. Naumann in his Principles of Crystallography, published in 1826, employed processes much resembling those of Mohs: but in a much enlarged and improved work on the subject which appeared in 1830, he states, with great candour, that at the former period "he was not acquainted with the great advantages of an analytical-geometrical mode of treating the subject," and that he has now "arrived at the conviction that this is and must be the simplest and most natural of all methods." This is a conviction which will probably be more widely diffused as the subject is more studied. M. Naumann has by this method calculated all the formulæ which are likely to be needed in a very clear and complete manner, and has exhibited the results of the most common combinations in a tabular form. Ratzeberg has published a similar synoptical Table, with figures of the crystalline forms and their combinations according to the method of Weiss; a very convenient mode of presenting the subject.

Geometrical truth has generally several aspects, each of which by constant contemplation appears to the individual reasoner to become the most luminous possible; and this is especially the case with regard to a system of truths so complex and multiplied as those which the solid geometry of crystals offers to our notice. It is not surprising, therefore, that other authors besides those above mentioned, should have taken other views of the best mode of treating the subject, and should have brought forwards these as considerable discoveries. Thus. Mr. Grassmann (Stettin 1829,) published a Treatise "On Physical Crystallonomy," in which he develops the connexion of forms by means of "a mathematical discipline hitherto never pursued." He determines the position of a plane by means of a "radius constructor" or line perpendicular to it, and assuming three fundamental radii of this kind, he deduces the number and mutual relation of the others by the combination of the relations of these fundamental radii. Neumann (a different person from Naumann,) has also endeavoured to simplify the subject by the introduction of normals and "index planes," as I learn from

[page] 331

Mr. Hessel, who has followed him in the use of these terms. Mr. Hessel himself (Professor at Marpurg,) in his work entitled "Crystallometry" (Leipzig 1831), has adopted several other new denominations and modes of considering forms. Thus, by way of example, he states concerning the rhombic dodecahedron, that its twelve faces "are perpendicular to doubly-two-membered normals, (the edges are doubly-one-membered, like-sided, unlike-ended,) which are perpendicular to doubly-one-membered four-and-three-spaced rays," &c.

The principle of this and similar methods of treating this subject consists in the permutations and combinations of various kinds of symmetry in lines, surfaces, and solids. One kind of symmetry, which occurs frequently in crystals, is not easily described by any common expression; and Mr. Hessel, who justly attaches much importance to the consideration of it, has introduced a peculiar term to designate it. The symmetry here spoken of is that which is seen in comparing the two ends of an oblique prism; and they are called by him "gerenstellig" gore-wise-placed, in opposition to "gleichstellig" alike-placed. One or two new phrases in such cases may perhaps be introduced with advantage: but the systems to which I here refer are so far laden with new phraseology and new views of the relations of space, that they will probably not be found by many a convenient mode of mineralogical study. It may be readily allowed, that when a person has mastered the fundamental views of these methods, the application of them to crystallometry may have some advantages of order or of generality; but this is for most readers too long and indirect a road to the results: and if crystallography leads to new views on the subject of elementary geometry, the prosecution of these will interest the pure mathematician, but the mineralogist will find it necessary to confine himself to investigations more peculiarly professional.

The consideration of the faces of crystals as distributed into zones, points out a mode of transition from one system of crystallization to another. Thus, if a rhombic dodecahedron be placed so that an axis is vertical which is terminated by three plane angles at each vertex, we may then, by prolonging or contracting the axis, make the form pass from the regular to the hexagonal system. But if the vertical axis be one which is terminated by four planes at each end, its prolongation or contraction converts the same form into one belonging to the tetragonal system. This mode of deriving one set of forms from another has been followed up by Breithaupt, who has thus derived all forms from the octohedron; the axis being, in the regular octohedron 720, and in the other cases greater or less.

[page] 332

Particular questions of crystallometry (as the mathematical part of crystallography has been termed by some writers,) have been examined by various persons. Mr. Haidinger in the Edinburgh Journal of Science for 1824, gave an excellent series of papers on twin crystals; in which he pointed out the various laws of combination, and analysed the resulting forms, in each of the systems of crystallization, and in the most important species. The general principle which governs these various combinations is, that the two parts of the twin crystal are such that one would come into the position of the other, by making a half-revolution round a certain axis; and this axis is always a real line in the series of crystalline forms belonging to the species which presents these phænomena. This general law in particular cases gives rise to occurrences as curious and as perplexing to the mineralogist as double and monstrous flowers are to the botanist. These are now for the most part understood.

One of the most common and yet most curious of these cases, is that of the interposed films in calc spar. These films, which were early noticed as giving rise to remarkable optical properties, were shown by Sir David Brewster to consist of crystalline plates of a thickness from 1/1000th of an inch upwards, in a position transverse to that of the crystal. He proved this by an analysis of the optical properties, and also synthetically by imitating those properties by means of crystalline plates purposely interposed.

A question has been raised, whether the oblique prism and the forms referable to it should be considered as a peculiar system, or as a right prism with only one half the number of sides extant in one case (hemihedral), and one fourth in the other (tetartohedral). Thus the twin-crystallization of pyroxene and of wolfram appears to indicate that though they appear as oblique prisms, they have rectangular axes. Yet the more general opinion and evidence seem to be in favour of the existence of a monoklinohedral or hemiprismatic system. And thus wolfram may be an oblique prism of an angle of 90° 1′, or 90° 0′ 1″. Naumann expresses nearly the same thing by saying that it is qualitatively monoklinohedral, quantitatively rhombic. The question must be decided by determining which mode of considering such crystals gives simple numbers and relations for the individual forms and twin crystals which really occur.

The part of Haüy's views which most caught the popular attention was the supposed exhibition of the real structure of crystals as built up of molecules of known shape, the crystalline faces being formed by given laws of decrement in the courses of

[page] 333

this sort of masonry. This doctrine was readily accepted, because it pretended to offer to our curiosity the ultimate analysis of the constitution of bodies; but by reason of this very boldness of promise, it was unlikely to bear the test of time and trial. At present this doctrine is probably not maintained, as a physical truth, by any one who has examined the subject; for though its assumptions may appear obvious à priori, they are not confirmed by observation. In order to support the theory of integrant and subtractive molecules, the facts of the case ought to be quite different from what they are. If indeed, in all minerals, the cleavage planes were such as to bound forms which would join so as to fill space, and if the forms of the crystals could always be referred to these planes with great numerical simplicity, the theory would still be a good mode of grouping the facts. But it appeared very early that it could not claim this praise; and when the author of it was driven (as in the case of fluor spar,) to conceive crystals made up of solids hanging together by their edges, we had an example of a theory in which difficulties were solved by suppositions directly contradicting the only reasons which could be assigned why the theory should be accepted.

Any theoretical mode of representing in general the ultimate structure of crystals, as consisting of elementary particles, whether as an aggregate of plane-faced solids, spheres, or spheroids, will probably not be of great value to the science in its present stage. But it must be considered interesting to know how far that numerical simplicity in the relations of the faces of crystals which led to the hypothesis of decrements, is really found in nature. The greater part of the faces of the most usual crystals are expressible by laws of which the ratios are very remarkable for simplicity. But in not a few cases the numbers run considerably beyond what was supposed to be the admissible limit in the earlier stage of the study. Thus in arragonite we not only have the numbers 2, 3, 4, 5, but 7, 9, 10, 19, occurring in the ratios. In carbonate of lead we have 13, 19, 21, 28. In galena Naumann has 12012, and 36036, according to his notation. In certain crystals of gold from the Ural, Rose finds a face which is (a, a/11, a/19) in the notation of Weiss. It would be important if any one could decide whether there is any limit of magnitude or simplicity to such ratios.

The most generally useful result which has followed from the modern methods of treating the subject of crystallometry, has been a great simplification in the mode of deducing the laws of formation of faces, when we find them on the crystal.

[page] 334

Instead of requiring the trigonometrical and algebraical calculations of Haüy. the law can, in a great majority of cases, be inferred from properties which are obvious to the eye, especially from the parallelisms of the edges of the faces. This mode of reasoning, introduced by Mohs, has been very successfully cultivated by succeeding writers, and especially by Naumann. The fertility and convenience of this resource is greater than any one not acquainted with it can easily imagine. With a collection of diagrams representing the binary combinations of forms, such as Naumann and others have given, the crystallometrical analysis of a very complex crystal becomes comparatively easy.

Still, to determine the laws of all the faces which commonly occur in the known species of minerals, is a task which has necessarily required the labour and skill of many persons. The early labourers in this province have a particular claim to our gratitude. Haüy did much, but he left also much to do. Weiss (Berlin Trans.) first successfully discussed some of the more difficult and complex cases, as gypsum, felspar, epidote. Professor Moh's Treatise contains a vast treasure of such determinations, and has only left for more recent crystallometers the task of supplying special deficiencies. And of such contributions we have excellent examples in recent times, among which we may mention the examination of the crystallization of felspar by Hessel, and of the blue carbonate of copper by Zippe of Prague.

In speaking of crystallometry, it is necessary to say something of notation, a subject which is repulsive to many in consequence of the multiplicity and complexity of the symbols which have been promulgated and which is yet absolutely indispensable to the mineralogist who would economize time, labour, and thought. Perhaps it may be found that the discrepancies of different authors are not so great as they at first sight appear. The notation of Haüy, indeed, belongs to so imperfect a knowledge of the subject, contains so much that is arbitrary, and is so incapable of being rendered either simple or symmetrical, that its reign ought by this time to have become only a matter of history, although in fact, among the disputes of its successors, it retains here and there some little show of authority. But the systems of notation of Weiss, Mohs, and Naumann, have better claims to our notice. That of Weiss is simple, according to his view of the subject, which, it will be recollected, consists in using the equations to the planes of the crystal; his symbol for any plane consists merely of the three coefficients of the equation, included in brackets. Of this method, the main defect is, that it is too general, and does

[page] 335

not very obviously exhibit those relations of the forms and of their edges of combination, which are so useful, as we have already said. Moreover three coefficients are more than we need; for the ratios of one of the three to each of the others are all that we have occasion for. Accordingly Mohs uses two indices only, by which he indicates any plane; and so far his method has an advantage. But he has been most peculiarly unfortunate in the mode he has selected of combining the indices with his fundamental letter. It is quite inconceivable how a mathematician, having to annex the two indices 2 and 3 to the letter P, should, without any support whatever from mathematical analogy, choose to connect one index with the letter by the sign + so as to convert the symbol into a binomial P + 2, and then use the other index as the exponent of a power of this binomial, as (P + 2)3. The violation of all mathematical significance, and the anomalous and useless complexity thus incurred, make such a system truly forbidding. Mr. Naumann has been more fortunate; and indeed his notation is indisputably almost as simple as it is possible for a crystallometrical notation to be; for two indices being necessary, and all that is necessary, he puts one before and one after his fundamental letter, and thus obtains a simple and convenient symbol. Moreover, the mode in which the laws of derivation are treated in his system is such as to bring very well into view the most important relations of the forms*: and as he has both published an excellent treatise on crystallography, and a compendious system of mineralogy, in which all the known forms of individual crystals are exhibited in this notation, we may hope that in time this system, or one resembling it, will supersede more complex and imperfect ones. It may be added that the systems of Weiss and of Naumann approach near to each other, and the notation of the one is very easily translated into the other. They predominate over a great part of Germany, and stamp the language of a great number of the best publications on the subject.

3. Optical Properties

Malus examined many mineral substances in the course of his inquiries concerning double refraction; but he does not ap-

* In comparing, however, the notation of Naumann and of Weiss, it ought to be taken into the account, that Naumann's two indices have often a more complicated appearance than Weiss's three indices. Thus we may compare Naumann's symbols 3/2 P3, 5/2P∞, 15/7 O 15/11 with Weiss's equivalent symbols, (3a, 1b, 6c), (5a, 2b, ∞ c), (a/7, a/11, a/15).

[page] 336

pear to have noticed any differences between the double refraction of crystals of different forms. Biot found grounds for separating doubly refracting crystals into two classes, characterized by opposite properties, which he called attractive and repulsive, or positive and negative; quartz is of the former, calc spar of the latter class. But no constant external distinction of such substances could be detected: the only fact connecting the form of crystals with their optical properties, was that observed by Haüy,—that substances having the cube, octahedron, dodecahedron, &c., for their primary form, had no double refraction.

Sir David Brewster must be considered as in a great degree the creator of the science which studies the mutual dependence of optical properties and crystalline forms; and he not only gave the first impulse to this study, but has enriched it with a vast quantity of most curious and interesting observations;—so great indeed, that all which has been done by other labourers in this field, bears as yet no proportion to the amount of his contributions.

Some of Sir David Brewster's first results* appeared, however, to contradict the general fact which we have just mentioned, as the only one then known on this subject, (Edinb. Trans. viii. 1815.) He found that some specimens of muriate of soda, fluor spar, and diamond, which according to the law just stated should have no optical axes, did, when they were obtained in considerable thicknesses, exhibit the colours which had already been found to indicate double refraction. The crystals seemed to consist of complementary parts, the effects of which nearly neutralized each other, but left in certain different parts of the crystal a small excess of action on one side and on the other.

His next observations were on calc spar. He had already shown that the colours which appear in the specimens crossed by films are produced, not by these films as thin plates, but by the properties of polarized light; and he now found that these films have a crystallization of a position opposite to that of the rest of the crystal, as has already been stated in speaking of twin crystals.

Another of Sir David Brewster's memoirs belonging to this period, is remarkably interesting. It had appeared by his

* I do not dwell on the discovery,—one of the first announced by Sir David Brewster on such subjects,—that doubly refracting crystals have two dispersive powers corresponding to their two refracting powers; which discovery has recently been re-stated as a novelty by Rudberg, and would have been again so restated by Mr; Cooper if he had not learnt from Sir David Brewster these previous publications of it.

[page] 337

experiments, that compression and dilatation give to transparent bodies a structure which produces the same effects as the crystallization which is associated with double refraction. He now tried the effects of compression and dilatation on crystals; and found that in this way the optical phænomena were variously affected, the rings deformed, the colours altered, the curves multiplied, &c. And, as the law which regulates the influence of this mechanical tension on the previous crystalline tension of the substance, he found that "positive crystals, compressed so that the axis (or direction) of compression is parallel to the axis of the crystal, have the order of the tints raised." The terms "negative," "dilated," "perpendicular," and "depressed," of course alternate in the enunciation of this law, with "positive," "compressed," "parallel," and "raised."

The striking and valuable generalization, however, which has established for ever a close connexion between crystallonomy and photonomy,—a connexion rich in the discoveries which it has already given us, and richer still in those of which it gives us no doubtful promise,—is found in the Phil. Trans. for 1818, (the memoir having been sent in 1817). In this Sir David Brewster states that the extreme perplexity of the subject, and the difficulty of procuring proper specimens, had prevented him hitherto from doing, what he has there done,—" reducing under a general principle all the complex appearances which result from the combined action of more than one axis of double refraction."

This law, so far as we are concerned with it as mineralogists, (for I am not now to speak of pure optical investigations,) is this:—That all crystals with one optical axis belong to the hexagonal or the pyramidal system (using the terms already explained; which are equivalent to those of Sir David Brewster,) —that all crystals with three optical axes belong to the tessular system;—and that all the crystals with two optical axes crystallize in other forms. It thus appeared that there was an exact correspondence between the degree and kind of the symmetry of the optical properties and of the crystalline forms.

This important principle was not hastily snatched from a few observations, as men, judging of great discoveries after the event, and struck by their simplicity, are always ready to think might have been done. It was carefully collected from an examination of many minerals, including 23 species with one axis, and 81 with two axes; and there were not wanting some apparent exceptions and difficulties in the application of the rule. These have for the most part disappeared under a closer

Y

[page] 338

examination. Perhaps one of the most striking instances of such an occurrence is the history of a mineral which has been termed by Mr. Brooke the sulphato-tricarbonate of lead. It was placed both by Count Bournon and by Mr. Brooke among the rhombohedral or hexagonal forms, and therefore ought to have had but one optical axis. Sir David Brewster however found that it presented the phænomena belonging to two axes. The difficulty was solved when Mr. Haidinger subjected the substance to an exact crystallometrical examination. It then appeared that a figure which had been supposed to be a right hexagonal prism had not the exact dimensions which the symmetry of that figure implies. Its sides, instead of making an angle of 90° with its ends, make an angle of 90° 29′; and instead of making angles of 120° with each other, they make angles of 120° 20′. The crystal had in fact precisely one of those forms from which its two optical axes would, by the rule, result.

Sir David Brewster had indeed already discovered a similar case in sulphate of potass; which had been arranged as a rhombohedron by previous mineralogists: but when its optical examination had indicated two axes, he found that the apparent bipyramidal dodecahedron of the rhombohedral system was composed of three prisms with angles of 114°.

The memoir of 1818 of which we have spoken, contains also Sir David Brewster's very happy detection of the remarkable' optical law on which the form of the curves seen in biaxal crystals depends; but on this and the other contents of this valuable memoir I must, for the reason already referred to, forbear to dwell.

The properties of doubly refracting crystals which affect all colours similarly, have now been reduced to a theory of singular beauty, which explains the most complex and apparently anomalous parts of their details: but the properties of such crystals, which seem to select certain colours for their action, remain still to be traced to their most general laws. Here also, however, Sir David Brewster has done a large proportion of all that has been done. His memoirs on the Absorption of Light by Crystals (Phil. Trans. 1816,) contain many curious facts on this subject. It is well known that certain tourmalines polarize the light which passes through them, to such an extent that they are commonly used as the easiest mode of obtaining polarized light. Agate and other substances were found, in like manner, to polarize light by transmission. But it appears that these are merely instances of a more general fact: many doubly refracting crystals, perhaps all coloured ones, affect the ordinary and extraordinary pencils with different colours. Thus beryl,

[page] 339

by Sir David Brewster's experiments, when exposed to polarized light, transmitted different colours (blue and greenish white), as the axis of the crystal was perpendicular or parallel to the plane of polarization.

Other species of mineral crystals were found to possess similar properties, and biaxal crystals exhibit it also with certain modifications. Sir David Brewster's list of cases is, as usual, considerable. He found also that many minerals absorb certain portions of common light, the transmitted portion being more or less polarized; so augite, epidote, produce upon light an effect partly of the same kind as tourmaline. Smoky quartz produces the effect strongly; but it is to be observed that a prism of quartz and one of tourmaline polarize in planes, the one at right angles to the axis, and the other parallel to it. Babinet has recently (1832,) enunciated as the general rule of such cases, that one or the other occurs as the crystal is of the attractive or repulsive class: but as it appears in fact by Sir David Brewster's previous researches, that the results in the two positions do not differ as dark and bright merely, but occur by a selection of colours, the general rule thus asserted must require, if true, to be differently expressed.

After these discoveries concerning the optical structure of crystalline substances, we might have here supposed that we could form some conception of the extent of the variety of nature in this class of phænomena. In such cases, however, nature is more fertile than our conjectures. It was soon found that many crystals possessed a structure far more complex than the mere number of axes of a single crystal could give them. This discovery also, and the accumulation of cases in which it is exemplified, are due to Sir David Brewster. It appears from his researches that many kinds of crystals must be considered as composed of a most curious mosaic work of crystals, in various positions, arranged in an order highly complex yet perfectly symmetrical. Thus he found in 1817, and announced in 1819, that amethyst consisted of different portions, which act differently on light in an alternate and complementary manner; these portions being generally wedges, with their vertices to wards the axis of the crystal, or a series of V's one within another, exhibiting the outlines of such wedges.

Again, it appeared that apophyllites from Iceland and from Ferroe were composed of a most curiously tessellated structure, capable of being visibly resolved into its elements by the transmission of polarized light.

And in 1818, (Edinb. Trans.) Sir David Brewster published his representation of the optical structure of analcime, which is in

Y 2

[page] 340

some respects even more curious and complex than the preceding cases. The icositetrahedron, which is the usual form,—a figure belonging to the class of crystals, which exhibit none of the properties connected with polarized light,—is in this species distinguishable into 24 solids, of which the boundaries have peculiar optical properties.

This phænomenon of the composition of a crystal, apparently simple, of portions exhibiting different optical relations, appears in fact to be very common. Thus nitre and arragonite often contain such portions; and in the second volume of the Cambridge Transactions, Sir David Brewster has shown, that the Brazilian topaz possesses a tessellated structure, a central lozenge being surrounded with a border of a different kind, sometimes with additional variations.

There would be something utterly perplexing in this complexity in the structure of objects apparently so simple, if we were to conceive such a kind of composition as formed of independent portions adhering together; but we ought, probably, rather to conceive these relations of parts as the result of a peculiar state of the equilibrium of the elastic æther which exists within the body, and on which its optical properties depend.

An additional principle, still further complicating the apparently inexhaustible phænomena of crystals, was discovered and fully discussed by Sir J. Herschel (Phil. Trans. 1820). The deviation of the succession of colours which many crystals exhibit from that scale of tints which Newton established by observations on thin plates, and which since his time has always been the alphabet of the higher optics, attracted Sir J. Herschel's attention, and he found that it could be fully explained by conceiving the direction of the axis of double refraction to be different for different colours. In biaxal crystals, such a deviation is almost universal, as in Rochelle salt, in which it is very prominent. Bicarbonate of potash, indeed, is said to be the only biaxal crystal yet examined, in which the axes for all colours are found to be strictly coincident. (Herschel, Light, 923.)

But this deviation from Newton's scale of colours perplexed observers more in the first instance, when it was seen in the case of uniaxal crystals. Sir J. Herschel (Camb. Trans. vol. i. Part I.) found certain varieties of apophyllite in which the doubly refracting structure was positive for the red, and negative for the violet rays, while for the intermediate indigo rays there was no double refraction at all. This remarkable circumstance was confirmed by the most decisive experiments, and now offers ho difficulty when viewed in connexion with the undulatory theory.

[page] 341

A somewhat similar circumstance has been discovered by Sir David Brewster in some specimens of glauberite. They are biaxal for red rays, the resulting axes being 5° asunder; but the axes for violet rays coincide, and for such light the crystal is uniaxal. This remarkable peculiarity was detected by the use of homogeneous light.

We have still another fact to notice equally striking, equally unexpected, and having also the name of Herschel associated with its discovery. There existed an optical law which had already attracted the attention of philosophers as being entirely anomalous and sui generis, and a crystallographical peculiarity equally curious. Sir J. Herschel, with singular sagacity and felicity, showed that these two circumstances were constantly conjoined. I speak of the circular polarization of light to the right or left, and the plagihedral crystallization of quartz. In both these eases we had, instead of the geometrical symmetry by which the laws of nature are usually marked, a set of appearances suggesting the idea of progress round a circle to the right or left hand; the deviation of the plane of polarization, as shown by the succession of colours on increasing the thickness of the transparent plate, being the optical fact thus governed, and the oblique position of certain faces of the crystal the mineralogical fact. It was proved that right-handed polarization always accompanies right-handed plagihedral faces, and left-handed polarization left-handed faces. This was established by Sir J. Herschel from the examination of thirteen crystals, and has since been fully confirmed by other observers.

It does not properly belong to our subject to dwell upon Prof. Airy's theory of the circular and elliptical polarization of the rays of light in quartz, by which an extremely complex and apparently unsymmetrical collection of phænomena are reduced to the most complete simplicity and regularity. But we may mention the experiments of the same observer on the optical properties of diamond. It appeared from these, that diamond, instead of completely polarizing light reflected at a certain angle, as other transparent substances are found to do, presents, at the angle at which light is most nearly polarized, phænomena resembling rather those of a surface of metal, than of a diaphanous medium.

Among optical inquirers, several persons have employed themselves in researches on the causes of the play of colour which is seen in Labrador felspar, as Hessel and Genf; and Nordenskiold has attempted a mathematical explanation of these colours, of which however I am not able to give any further account. Sir David Brewster has examined these

[page] 342

curious phænomena; and it appears from his inquiries that the colours are produced on the principles of the colours of thin plates, by cavities bounded by parallel plane surfaces.

In the application of polarized light to the examination of the properties of minerals and other crystals, we have acquired a new instrument, of a use far more extensive and instructive with regard to the structure and differences of substances, than anything which had before been dreamt of. Physical optics and crystallography are for the future two coordinate portions of a vast province of science, of which the limits as yet have not been caught sight of. In the optical examination of minerals there remains much to do; and it would not be difficult to point out branches of inquiry which are of evident importance to the present state of our knowledge of crystals. It would, for instance, be very satisfactory to know the difference of optical symmetry which exists between a right and an oblique prism; —whether the additional deviation from geometrical symmetry, which occurs in the latter case, corresponds, in the optical properties, with the fact of our no longer finding the poles of the lemniscates in the same plane, as would seem to be the case from some experiments of Sir J. Herschel,—or whether this case is marked by some yet unguessed peculiarity.

But while we look forward with hope to the augmentation of the stores of observation in this most interesting department of the study of nature, it is desirable that we should be aware of the treasures already in existence. The discoveries already mentioned as published by Sir David Brewster, and many others which might have been added to the enumeration, form, I believe, but a part of the facts bearing on optical crystallography, which that indefatigable observer and acute philosopher has in his possession. He has long led the mineralogical world to hope to receive from his hands a Treatise on Mineralogy, on optical principles, in which it may be presumed he will state all the most remarkable facts and laws with regard to the relations of mineral species to light, which have come under his notice; and certainly no acquisition could be more interesting to the mineralogist, or more likely to give a fresh impulse to the progress of this science*.

The different optical properties of minerals have been theoretically expressed by speaking of the different elasticity of the crystal in different directions. This induced Savart to examine, by his ingenious methods, whether the acoustical elasticity was also governed by similar differences. We ought not to overlook,

* I am sorry to learn from Sir David Brewster, that he does not contemplate the immediate publication of this long-desired work.

[page] 343

in this comparison, the circumstance that the term "elasticity" is here applied to two different classes of phænomena, referable to different principles;—the acoustical phænomena depending on the elasticity of the parts of the solid, the optical on that of the optical æther. Savart found that some of the acoustical differences of elasticity correspond to the optical relations, but also that there are other acoustical differences following a different law of symmetry. In a common crystal of quartz, a transverse plate (perpendicular to the axis,) has the elasticity of all its diametral lines equal; but though all the plates cut parallel to the axis have the same optical properties, their acoustical properties have a relation to the edges of the prism; such, however, that any three plates at angles of 120° have equal acoustical elasticity; he found also, that by the acoustical properties he could determine the cleavage planes of quartz; he made like observations on calc spar, and some of equal interest on metals. These, it appears from his researches, have a structure neither regularly crystallized nor altogether uncrystalline; their properties are different in different directions, and they give, by their vibrations, corresponding differences of note, amounting to a tone; yet, as it is found that parts taken from the whole have not properties identical with those of the whole, the composition is not a repetition of that of small parts, as in regularly crystallized bodies. It is extremely interesting to see the sciences of colours and of sound thus uniting to give us that information indirectly concerning the internal structure of minerals, which we have so long attempted in vain to obtain directly.

4. Chemical Mineralogy.

In entering upon the Chemistry of Mineralogy, we come to that part of the subject in which undoubtedly the greatest labour has been employed and the least progress made. That this is not too unfavourable a judgement will be clear, I conceive, when it is considered how numerous and operose have been the analyses of mineral substances executed by all the best chemists during the last century, and yet how scanty and unconnected our knowledge on this subject still is. Not only are there no general and generally recognised chemical laws, capable of being predicated concerning extensive classes of mineral bodies, but the constitution of any particular mineral species, with the exception of a few, is a matter of doubt and dispute; and we shall hardly be contradicted when we say that there are very many cases in which, if we were to state to the chemist the ingredients of a substance and their quantities, he would be unable to tell us what mineral species the substance was.

[page] 344

We can hardly flatter ourselves, therefore, that we are at present close upon the discovery of the nature of the connexion between chemical constitution and mineral character. In the mean time, the chemists have been far from idle in the only road which, under such circumstances, offers itself; and a number of excellent analyses of particular substances have been perpetually accumulating. Still, such labour is naturally and inevitably pursued with less energy and connexion than would probably show themselves, if the analysers were tempted or rewarded by the prospect of some general law to be extended or verified,—some anomalous cases to be included in a well-established analogy.

Perhaps, however, such prospects are already opening. Mitscherlich's Law of Isomorphism, published about twelve y ers ago, promises far more fairly than any previous portion of chemical knowledge, to relieve chemical mineralogy from its stationary and helpless condition. According to this law, the ingredients of a given species of minerals are not absolutely fixed as to their kind and quantity; but one ingredient may be replaced by an equivalent (not necessarily an equal) portion of some similar ingredient,—generally some elementary body in the same degree of oxidation. Thus in amphibole, or in pyroxene, the lime may be in part replaced by portions of protoxide of iron, or of manganese, while the form of the crystal and the angle of its cleavage planes remain the same;—or in such cases the angles may vary slightly, while the other properties remain so far unchanged as to establish a strong mineralogical connexion in the group thus related: so the different kinds of felspar vary only by the substitution of one alkali for another; and the carbonates of lime, magnesia, protoxide of iron, protoxide of manganese, and their mixtures, agree in many respects of form, &c., while the angle varies through one or two degrees only. Several other such groups might be mentioned; as garnet; olivine; the carbonates of baryta, strontia, lead, and lime (arragonite); the sulphates of the same bases; the sulphates of iron and of cobalt;—again, the sulphates of zinc, nickel, magnesia; various phosphates and arseniates; and several other compounds.

It is obvious from the mere enunciation of such propositions that we have here chemical laws of a more general and scientific character, than any that can be founded on the analysis of a single specimen of each species. And as this principle is of the greatest importance to the mineralogist, it is interesting to observe that the mineralogist was the first person to perceive the necessity of such a principle. Thus Breithaupt, writing in 1818 (Auswahl der Dresd. Gesellsch. vol. ii. p. 142;) considers

[page] 345

tourmaline as containing potassa and soda along with lime and magnesia; these alkaline substances taking each other's places, so that there may be less of the one in proportion as there is more of the other. And he observes that Fuchs calls such elements "vicarious" with respect to each other,—a term since used, and with propriety, to designate the relations of isomorphous ingredients.

I do not know whether all our English chemists fully acknowledge the reality of the isomorphous groups of minerals; but those who do, will probably agree, that one of the most important objects which the chemist or the mineralogist can at present propose to himself, is to extend such grouping to as many minerals as possible. We have at present a mere mob of species; by brigading them under a system of isomorphism, they may become a well-ordered army.

It cannot be denied that there is something formidable in the prospect of the labour which is thus found to be incumbent upon those who would learn the constitution and relations of mineral species. The exact analysis of one or two specimens of each species has been considered, and justly, as a business requiring no small skill and sagacity, and great care and sacrifice of time. Even the most patient and most industrious of chemists, Berzelius and the Germans, complain of the employment on this ground. But it appears from the isomorphous doctrine, that we cannot hope to understand the chemical constitution of any mineralogical species or group, without subjecting to careful analysis not one or two specimens only, but many, from different localities and forming different varieties. It is only thus that we can obtain the character, common to the whole group, which may be taken as its type or formula.

In expressing the constitution of bodies, many chemists have found it necessary to call in the aid of notation; and the algebraical system introduced by Berzelius is now pretty generally diffused, though modified in parts, by some of his followers, as, for instance, Beudant. Such a notation is convenient, I conceive, in other parts of chemistry; but it is indispensable in mineralogy, where the composition of bodies is often much too complex to be intelligibly expressed by the resources of the language of modern chemistry. The doctrine of isomorphism gives us an additional reason for the employment of such a notation; for the constitution of an isomorphous group can be most conveniently expressed by means of a formula in which one of the letters is subject to be replaced by others indicating the vicarious ingredients.

I say nothing here of the merits or defects of different systems of chemical notation; for though I cannot but think it

[page] 346

unfortunate that algebraical symbols should be used in a manner contradictory of the first principles of algebra, as is the case in the notation of Berzelius, it is probable that the general acceptation of a notation of any kind will be mainly influenced by the amount and value of the information which it is employed to convey; and in this respect the Swedish system possesses an advantage in which it cannot easily or soon be matched by any rival system.

In speaking of the connexion of chemistry and mineralogy, I ought to mention the general law announced by M. Kupffer. This law professes to give the dependence of the crystalline form on the specific gravity and atomic weight of the body, and thus, if well established, would be a principle of a very high and comprehensive character in our science. I do not think, however, that any one who examines M. Kupffer's Memoir critically, will be satisfied with the kind and quantity of facts from which this induction is held to be collected. The selection of substances belonging to one particular system of crystallization, (for instance, the rhombohedral,) for comparison with each other, seems to be quite an arbitrary step, and is in no way explained by the law so asserted. But not to insist on this objection, the mode in which the dimensions of the primary form are compared with the other quantities is such as would enable the author to prove almost any law with equal facility; for he holds it to be an unimportant matter whether he takes what is usually considered the primary form, the primary rhombohedron for instance, or any other rhombohedron which can be derived from it. Thus in comparing calc spar with rhombohedral iron oxide, he takes the number expressing the axis of the primary form of the one substance, but in the other substance he multiplies the axis by four, thus substituting for the primary rhombohedron that which arises from truncating its edges. And it is by using a similar license in other cases that he exhibits an approximate verification of the formula which he states. There appears to be little hope of any valuable result to be obtained by comparison of numerical results, except the properties which the numbers express be clearly the same property in the different cases which are compared. If the cleavage rhombohedron of one case is analogous to anything, it must be analogous to the cleavage rhombohedron in the other.

Still it is impossible for any one to take an interest in this portion of science, without seeing that the connexion of chemical composition with crystalline form is one of the great problems to be solved; and it is very natural that those who feel this should be tempted to hazard a guess concerning the solu-

[page] 347

tion of this problem. Thus in a Memoir which appears in the Transactions of the Wernerian Society of Dresden for 1818, Breithaupt conjectures that boron is the ingredient which gives the electrical and crystalline polarity which he attributes to boracite, tourmaline, anatase, and axinite. Hitherto, however, conjectures and researches on this subject have had little success. In the course of last year, Magnus endeavoured to detect the chemical difference of garnet, a tessular, and idocrase a pyramidal substance; yet after many analyses and conjectures he was obliged to acknowledge that it had escaped him. Berzelius (Brewster's Journal, N.S. iii. 188.) finds no difference between the composition of hexahedral and prismatic iron pyrites. More recently still, Ampech has analysed a number of tessular minerals, spinelle, pleonaste, gahnite, franklinite, and chromic iron oxide: and in this instance he seems to have had some success in giving a common type to their chemical formulæ, as there is a common type in their crystallization. According to him, they consist of ingredients of two classes: the one class containing alumina, protoxide of chrome, peroxide of iron or manganese; the other class containing protoxide of zinc or iron or magnesia; and the rule of composition is, that the ingredient of the former class contains three times as much oxygen as that of the latter. If this law be true, we cannot doubt that many similar laws exist both for tessular and for other forms, and we may hope that after one has been detected others will soon appear.

The discovery of artificial crystals in the slags of furnaces was not unimportant to the chemistry of mineralogy. One of the first and most extraordinary instances was the detection of perfect crystals of titanium in the Welsh iron slag, by Dr. Wollaston and Professor Buckland. It has appeared by examination that these accidental products are more free from any admixture of iron than it is easy to obtain titanium by the ordinary chemical processes. In 1825 Mitscherlich found in the Swedish furnaces bisilicate of iron (pyroxene), mica, and other mineral species. About the same time, Berthier in France obtained in the furnace, by direct synthesis, regulated by the atomic theory, crystals similar to those found in nature. Professor Miller of Cambridge has examined several slags from the furnaces in Wales, and it appears that the crystals in those assume the form of olivine. It is satisfactory thus to find that the same substances affect the same crystalline form in our furnaces and laboratories, and in the great laboratory of nature. Indeed nothing can be more likely to help us in obtaining a knowledge of the chemical laws of crystalline forms, than to have the power of verifying our conclusions synthetically by forming crystals, as well as analytically by destroying them.

[page] 348

In the same point of view, the examination of crystals formed from solutions is of great value to mineralogy; as, for instance, the many excellent measures of artificial salts by Mr. Brooke, Mr. Haidinger, and others. Such crystals may often be obtained in much greater abundance and perfection than natural crystals, and especially than natural crystals of similar chemical composition; and thus they widen very much the field of facts to which our inquiries lead. In former times the mineralogist was professedly restricted to substances which occur in nature; but we may venture to say that a line so arbitrary and accidental cannot be the true boundary of the science. Where-ever crystalline forces act, the crystallographer is called upon to pursue his speculations: these speculations, whether we call them mineralogical or not, are such as give interest and promise to our study. In this point of view mineralogy possesses not only the importance which belongs to its ancient subjects, but also an importance of another kind, which belongs to it as a necessary supplement to chemistry; for it takes into consideration those physical characters of chemical compounds (crystallization, specific gravity, hardness, fracture, lustre,) which belong to them as solid bodies, and which indicate the law and intensity of the corpuscular forces by which each combination is bound together. The study of artificial crystals, therefore, whether obtained in the wet or in the dry way, may be recommended as very useful to the mineralogist.

Haldat (Ann. de Chim. Jan. 1831,) has shown a mode of obtaining artificial crystals of iron oxide by the decomposition of water; and these resemble the natural crystals of "fer oligiste" from Elba. So Becquerel has obtained the oxides of copper, lead, zinc. But by far the most valuable and important of such experiments appear to be those of M. Becquerel on the sulphurets, iodurets, and bromurets of metals, which he has obtained by artificial chemical action in a perfectly crystalline form. The agency which he employs is very weak galvanic tension; and he has succeeded thus in producing sulphuret of silver in small octohedral crystals resembling the native mineral, and sulphuret of copper, also closely resembling the native sulphuret. The sulphurets of zinc and iron require additional precautions, but are also obtained like to the native species; and iodurets, bromurets, and seleniurets of various metals are procured as crystals by similar processes. (Ann. de Chim. Oct. 1829.) These important steps in synthesis will probably throw a new light upon known analytical results.

M. Beudant has made a number of interesting experiments on the subject of another class of causes which modify the forms of crystals, and of which the general laws are, if pos-

[page] 349

sible, more unknown and obscure than those which determine different fundamental forms to different compounds. He has examined the circumstances which determine the various modifications which a given fundamental form undergoes; and has proceeded so far as to be able to produce at will one or other of certain possible modifications. Thus (Mineralogie, i. 190.) common salt crystallizing in pure water affected almost always the cubical form; if it crystallized in a solution of boracic acid, it assumed the form of the cube with truncated angles. Alum in nitric acid had the same form; in muriatic acid it was a figure of twenty sides, the octohedron and dodecahedron combined, the faces of the former being much the larger. An addition of alumine to the liquor, produced, in addition to the former faces, those of the cube; in pure water this salt is the simple octohedron. Sulphate of iron has commonly a simple form; by adding a few drops of sulphuric acid, more complex forms are obtained; and this rule respecting the effect of the addition of acid appears to be extensively true. The sulphate of iron mixed with sulphate of copper has its simple form, an oblique rhombic prism: the mixture of sulphate of nickel produced the same effect, but that of zinc an opposite one, the crystals becoming less simple. It has long been known that common salt mixed with urea, affects the octohedron instead of its usual form, the cube; and that in similar circumstances, sal ammoniac becomes the cube instead of the octohedron. Alum in a concentrated solution of alumine assumes the cubical form; an octahedral crystal of alum placed in such a solution soon assumes a cubical form; by being placed again in a solution adapted to give octohedral crystals it may be made to assume the octohedron. It is impossible not to be tempted to refer phænomena similar to hese, occurring, as they so often do, in natural crystals, to similar circumstances which have prevailed when the crystals have been forming.

Several statements of a curious kind have been made concerning the recent crystallization of substances which we cannot cause to crystallize in our laboratories. Thus (Brewster's Journal, vol. x.) Repetti observed quartz in a pasty state and in the act of crystallizing. The same kind of occurrence is said to have been observed of various other substances, as beryl, opal, heavy spar.

The chemical changes to which minerals are subject have been well described by Haidinger, (Brewster's Journal, vol. x.) who applies the name of parasitic minerals to those which retain the form of a substance while the substance has undergone a change, particle by particle. This change is of various kinds; weaker

[page] 350

affinities give way to stronger; carbonates are changed into sulphates; metallic substances are oxidized; copper is replaced by iron, &c.

I have not spoken of improvement in the methods of analysing minerals, of which many made in recent times might easily be enumerated; since these processes-seem rather to belong to the history of chemistry. Nor have I attempted to give the results of analyses of particular minerals; for these, though valuable materials of mineralogical knowledge, cannot be introduced into a general view like the present one, till they have been connected by some principle of dependence or relation.

5. Classification.

I. Distinction of species.—It will probably give to common hearers and readers a strong impression of the confusion still prevailing in the science of mineralogy, when we state that it is still a matter of dispute what are the limiting conditions and definition of Species in general; and that very wide and numerous differences of opinion prevail as to the identity and diversity of species in particular cases. Indeed it would be almost difficult to mention a species which is free from such doubts. This uncertainty is however not so fatal to real science as might at first sight appear. The formation of definitions, and the establishment of unerring distinctions, are among the last, and not the first, steps of systematic knowledge.

Haüy's definition of a mineral species, "the same ingredients combined in the same form," acquired a kind of celebrity at the time, and it has been adopted by many succeeding mineralogists. The definition which seems to be recognised in the crystallometrical school of more modern times is, "the same primary form with the same fundamental angles of cleavage, combined with an approximate identity of chemical and physical characters." But both these definitions were announced as axioms when they should have been tried as guesses. It was impossible to know, independently of experience, that the sensible differences of minerals corresponded universally to determinate differences in their ingredients. It is now certain that they do not: for, without calling in the doctrine of isomorphism, we know that scarcely two analyses of minerals of the same kind give identical results; so that the Haüyian definition of species is inapplicable without some reformation of its terms, and to make it unexceptionable will be found no easy task. In like manner it was imposible to know, independently of experience, whether minerals which resembled each other so as to have no constant

[page] 351

distinction to the eye of the common observer, would coincide exactly in their angles when tried by the severe test of the goniometer. Whether they do so or not, may be considered as a question still under decision. We know that we have, in some instances, groups in which the angle varies slightly in correspondence with certain variations of the physical properties and ingredients. And the question is, whether this is a variation per saltum,—so that the carbonate of lime has one invariable angle, that of iron and lime another, that of iron a third, and so on;— or whether it is a variation by insensible degrees, the angle passing from one magnitude to another by gradations corresponding to the minutest gradations in the proportions of the ingredients. Of these opposite opinions Mr. Brooke maintains the former, M. Beudant the latter. The most exact and multiplied observation alone can decide the point.

It may be observed that the groups of which we have already spoken as isomorphous groups, are not, in all cases, such according to the etymological sense of the term; for the forms and angles of the members of these groups are near to each other, but not equal. They have hence been termed plesiomorphous groups by Professor Miller, who has examined some of them; and as the majority of the so-called isomorphous groups are certainly of this character, it may perhaps be questioned when there are any groups strictly isomorphous (those belonging to the tessular system being of course excepted in all such assertions). Thus Rose has shown that several varieties of minerals, crystallizing like pyroxene, agree in composition by the vicarious substitution of one mineral for another; and Bonsdorff has shown the same thing for amphibole. But I am not aware that in these instances the exact identity of the angles of the crystals compared was ascertained by means of the goniometer, so that slight differences in the angles of the different kinds of pyroxene or of amphibole which were compared, may still have existed, as we know that such differences do exist in felspar.

M. Breithaupt has carried still further this scepticism concerning the constancy of crystalline angles. He finds by measurement that crystals of calc spar, apparently equally pure, from different localities, vary in their angles to the amount of half a degree. What is a still more unexpected blow to the fixity of these angles, he finds (1829,) that corresponding, or, as a geometer might call them, homologous angles of the same crystal, which ought to be precisely equal by the law of symmetry, are perceptibly different under the goniometer. Thus two pairs of opposite planes of the square pyramid of anatase were found

[page] 352

to differ in their inclination no less than 3°. Similar irregularity was found to exist in the crystallization of idocrase, and of tetragonal copper pyrites; and at last, as he says, it seemed to be time to satisfy himself that there really was such a thing as a regular square pyramid in nature. This he fortunately discovered in a perfect crystal of zircon. In attempting to systematize these anomalies he has been led to introduce a number of new terms and laws, which perhaps may be less necessary when we have a fuller view of the facts of these and similar cases.

All this unexpected uncertainty shows us how assiduously we ought to measure and compare crystals, if we wish to bring mineralogy into the form of a science either certain or systematical. The invention of the reflecting goniometer by Dr. Wollaston, was an invaluable gift to the crystallometer; and every step of our progress makes us more sensible of the importance of this elegant and well-devised instrument. But unless we can acquire some knowledge of the laws of constancy and of change in the angles which we measure, this instrument is valuable only as an ingenious means for an undiscovered purpose, a precise expression in an unknown tongue.

Perhaps we cannot deny that at present we have still to learn the true place of the isomorphous or plesiomorphous groups, and that we are ignorant on what step of the ladder of classification they ought to stand;—whether they correspond to species, or to some higher division;—and whether they contain, within the groups themselves, a further definite subdivision into subordinate members.

The confusion and perplexity in which this branch of the subject is still involved, may be judged of in some measure from the fact, that in the course of last year Gustavus Rose published a very interesting memoir "On the necessity of uniting Augite and Hornblende into one species"; thus throwing a doubt on the distinction of two minerals which had hitherto been supposed to be as clearly separated by their form and physical properties as any two species composed of similar ingredients. His grounds for maintaining their identity are not slight;—they are, the possibility of reducing the one form to the other, the resemblance of their chemical composition, but especially the mode of their occurrence in combination, and the fact that melted hornblende crystallizes as augite.

Another indication of the same kind may be found in Mr. Kobell's memoir on Diallage and Hypersthene. He conceives that these minerals may be traced to an agreement with pyroxene, both as to their crystallometrical and their chemical pro-

[page] 353

perties. Both consist of bisilicates of a mixed base, the ingredients of the mixture being lime, magnesia, protoxide of iron and of manganese; but in the case of pyroxene the lime is equal to the sum of all the other ingredients of the base; in diallage the magnesia takes this predominant place; and in hypersthene, or bronzite, an intermediate composition prevails.

Nor does it appear that the optical properties of minerals, valuable as they are, can help us out of this uncertainty. It may hereafter be found that the distinctions marked by such properties are more clear and constant than any others. But this is at present uncertain, and there exists a possibility, as yet not disproved, that a very minute change of composition may affect the optical properties much more than it affects any others. However this be, it is clear that the optical distinctions cannot take the place of the familiar divisions which the mineralogist is accustomed to use. The kinds of Apophyllite which have been termed Tesselite and Leucocyclite, in consequence of the curious optical phænomena which they exhibited to Sir D. Brewster and to Sir J. Herschel, do not appear to be distinguishable by the eye, or by the common tests of chemistry, from other kinds of the same mineral. Sir J. Herschel found in a single fragment of a crystal of this substance,three portions, each possessing distinct and peculiar properties. Sir David Brewster's tessellites had, in like manner, a difference of optical structure in different points. It is clear that species, discriminated by such differences, cannot easily be employed in classifying mineral specimens; and it is at present difficult to foresee the place which these differences, when they are more fully known, will occupy in our mineral systems.

The angle between the axes of topaz from different localities, is also said to vary very considerably. But one of the kinds of minerals in which this perplexity appears to be greatest, is the micas. Substances which have been referred to this species (or group, of whatever kind it be,) have been found (Beud. ii. 149,) to have, some of them a single attractive, some a single repulsive axis: others (the more common kinds,) have two axes, and the angle between the resulting optical axes has been found to vary from 50° to 76°; and though the chemical analyses of these different kinds of topaz have given results sufficiently variable, it does not appear that any steady connexion between the composition and the optical properties has yet been discovered. Sir D. Brewster (Journal, 1825, ii. 205,) has shown that a particular kind of mica which he examined (lithion mica), contained a combination of both uniaxal and biaxal crystals.

z

[page] 354

Among the properties which have been proposed as bonds of connexion between the members of plesiomorphous groups, may be mentioned Naumann's formulæ expressing certain simple algebraical relations between the dimensions of the different axes; thus the right prismatic primary forms of carbonates of baryta, strontia, &c. have all the relation 2 a = b + c, though the magnitudes of a, b, c somewhat vary.

II. Systems of classification.—While the lowest member of our mineralogical classification, the Species, is in this state of uncertainty and confusion, it cannot be surprising that the superior departments should be not yet satisfactorily adjusted. The fact has been, that in England the imperfection and inutility of the systems commonly put forwards have been so obvious, that a general impression has established itself among our mineralogists, that a system is a useless source of perplexity, and that any system, however arbitrary, is nearly as good as any other. On the Continent, however, the case is widely different; and an extraordinary number of mineralogical systems, published of late years in Germany, Sweden and France, show how earnestly foreign philosophers have struggled with the difficulties of the problem. Much indeed of the past failures and present apathy of our countrymen on this subject, is to be attributed to their having undertaken the task somewhat hastily and lightly. If any one endeavours to construct a complete classification on that which is perhaps the most obvious principle, the leading ingredient, he will soon be led into endless inconsistencies, and will obtain none of the great advantages which a good system will certainly afford. It might easily be shown by examples that this is what has occurred; and any attempts to patch up the manifest defects of such undertakings would probably lead us, after various wanderings and struggles, to the point at which the Swedish, German, and French chemists have already arrived.

There has taken place in Germany a somewhat vehement dispute respecting the merits of two apparently opposite methods of classification;—that which proceeds by external characters, and that which depends on chemical composition.

In Werner's system, chemical differences were recognised as the great leading divisions; but the subordinate distinctions were established by means of that nice discrimination of external characters which Werner so successfully applied and taught. It was, I believe, with Professor Mohs, the successor of Werner at Freiberg, that the design originated of demonstrating the possibility and the propriety of founding our mineralogical classification on external resemblances and differences alone, borrowing nothing from chemistry; so that mineralogy

[page] 355

should in this respect be brought into agreement with the other provinces of natural history. The general analogy of natural history supplied one motive to this attempt, but another arose from the conviction that we cannot compare the chemical and physical characters of minerals, without first studying them independently. The chemist cannot assert that arragonite does or does not contain strontia, except the mineralogist can tell him what is or is not arragonite.

To the execution of this very remarkable undertaking Professor Mohs brought a consummate acquaintance with the mineral kingdom. The plan of his task was regulated in a great measure by the analogy of the science of botany, in which system has been so successful; and the publication in which he first gave to the world the results of his labour was his "Characteristik," (Dresden, 1820,) a work of not many pages, but one which excited a very extraordinary interest in Germany.

The "characteristik" in any branch of natural history is a portion of the science distinguished from the "systematik," in-asmuch as the latter arranges individuals and species into their classes by a consideration of all their properties; while the former selects certain marks by which we may easily recognise in each instance the class in which any given individual has been placed. Thus the varieties and the species of the genus Lamium in Botany are placed together because of their general affinities; but in the "characteristik" of the science, the genus is distinguished from other flowers of similar form by a small tooth in the outline of the corolla on each side; and the species Lamium album (White Dead-nettle,) is further distinguished by having the tube of the calyx shorter than these teeth.

To devise characters of this kind which should mark a series of successive and subordinate distinctions in the mineral kingdom, was a very curious and difficult task, and was executed with no small skill. We cannot here go into any detail on this subject; but as examples of the method we may take the characteristics of the order Pyrites, the genus Copper Pyrites, and the species Octohedral Copper Pyrites*.

Immediately after the publication of the "Characteristik"

* X. Order Pyrites.
Metallic.
H = 3·0 to 6·5.
G = 4·1 to 7·7.
If H ss 4·5 and less, G is under 5·3.
If G = 5·3 and lew, colour is yellow or red.
V. Genus Copper Pyrites.
Tessular or Pyramidal.
Colour brass-yellow, copper-red.
H = 3·0 to 4·0.
G =4·1 to 5·1.
I. Species Octohedral Copper Pyrites.
Tessular.
Cleavage, octohedron very indistinct.
Colour copper-red.
H = 3·0; G=4·9 to 5·1.

Z 2 V. Genus

[page] 356

of Mohs, in 1820, appeared also a "Characteristik" by another excellent mineralogist, likewise a pupil of Werner,—Breithaupt. Whether this coincidence is to be looked upon as an indication of the general tendency of thought in the school of Freiberg, and how far it resulted from any more direct communication, it is not necessary here to determine. Since that time Professor Mohs has removed to Vienna, to superintend the Imperial Cabinet, and Professor Breithaupt has succeeded him at Freiberg; so that the "natural history method" is now taught to no small or insignificant portion of the mineralogical students of Germany; and the translation of Professor Mohs's Treatise by Mr. Haidinger (1825), enriched as it is with much valuable additional mineralogical information, has done all that could be done for the diffusion of the system in this country,

Berzelius may be considered the head of the chemical classifiers, as Mohs is of the natural history classifiers. In 1816 he published his Essay to establish a purely Scientific System of Mineralogy by means of the Application of the Electro-chemical Theory and the Chemical Doctrine of Definite Proportions. In this Essay he proceeds upon the great principle of his school, that the relation of electro-positive and electro-negative is the foundation of all chemical relations. Of the strict truth of this principle chemists must decide; the application of it to Mineralogy was made with great consistency. Minerals were arranged into families according to their electro-positive element, and these families disposed according to the place of this element in the general electro-chemical scale. Thus there was a family for sulphur, another for azote, another for carbon; others for each of the electro-negative metals; other families for each of the electro-positive metals; others for each of the bases of earths and alkalies. Each of these families was subdivided according to the electro-negative elements. Thus, Copper had the subdivisions,—1. Pure Copper; 2. Sulphurets of Copper, of which there are nine or ten (including mixtures of

[page] 357

iron, lead, tin, bismuth); 3. Oxides of Copper; 4. Sulphates of Copper;—and so on.

Nothing could be more complete and symmetrical than such a system, if the object had been to arrange the products of the laboratory, or substances as definite and distinct as those. But it very soon appeared that the system contained in itself germs of inevitable confusion. Very many minerals are so complex and so imperfectly known as to their chemical composition, that it remained doubtful where they were to be placed; and the parts of the system which appeared to have equal claims to them were widely removed from each other. Nor did any approximation of substances apparently of the same kind, or any analogies and rules with regard to the association of chemical formulae, come into view by means of this classification, and thus give it the air of a successful conjecture.

The same may be said of the system of Brongniart, the successor of Haüy, which proceeded on the chemical principles then (1806) generally recognised, and of that of Leonhard; in both of which there may be observed what was considered as a blemish in the first system of Berzelius,—a struggle between the "scientific" or electro-chemical principle, and the ancient customary views which tended to place similar minerals together. Thus tellurium, like sulphur, is a mineralizing substance; and hence there is no more reason, on the electro-positive principle, for making a family for tellurium than for sulphur. Yet both Brongniart and Leonhard have made such a family, and have arranged the telluriurets in it, while the sulphurets stand each under its metal. This inconsistency is one of the marks of the unsatisfactory impressions produced by a rigid chemical system on such principles, and of the difficulty of adapting such a system to the mineral kingdom.

When in the due course of time the examination of the chemical difficulties of mineralogy had led to the doctrine of isomorphism (1821), the untenable nature of Berzelius's first system, and of all similar ones, became more obvious. With a candour and alacrity worthy of his elevated position in the world of science, Berzelius was himself one of the first to acknowledge the necessity of some change. In 1824 he published in the Transactions of the Academy of Sciences at Stockholm, a Memoir "On the Alterations in the Chemical Mineral System which necessarily follow from the property exhibited by Isomorphous Bodies, of replacing each other in given proportions." In this Memoir he gives a classification of mineral substances according to their electro-negative element; explaining that the fact that one isomorphous electro-positive element may take

[page] 358

the place of another with no definite change of external character, renders an arrangement, such as he had before proposed, impossible in practice. Thus, in this new system one of the families is sulphur, and under this are arranged the sulphurets of zinc, of iron, of cobalt, of nickel, of copper, &c He acknowledges, however, that by this means the difficulty arising from isomorphism is only diminished and not removed: for Mitscherlich had found that electro-negative as well as electropositive elements may replace each other isomorphously;—for instance, arsenic and phosphoric acid.

In 1824 Beudant also published his Traité Elementaire, which professes to found its arrangement on two leading max ims: 1. The electro-negative element imparts its character to the combination more frequently than the electro-positive one, and hence is rather to be taken as the principle of classification; 2. The electro-negative elements are to be arranged in a circular series according to their natural relations. The circular series thus adopted, is founded upon certain views of Mr. Ampere, according to which elementary substances are divided into three great classes, termed, 1°,Gazolytes; 2°, Leucolytes; 3°, Croicolytes; according as they form gases, uncoloured solutions, or coloured solutions. In this system also the form of the chemical symbol of composition is taken into the account. Thus under the family of sulphurets we have (a) simple sulphurets, (b) double or multiple sulphurets.

The doctrine of isomorphism, however imperfectly developed at present, undoubtedly promises much fairer to disclose to us the true chemical relations of minerals than the views previously entertained. It was probably in consequence of this promise that an extraordinary number of new systems of mineralogical classifications on chemical principles were published about the period of which we are speaking. Gmelin (1825) was the author of one which appeared about the same time as that of Berzelius, and like that founded its leading distinctions on the electro-negative or formative element of bodies; but besides this, it took into account the numbers which occur in the chemical formula, distinguishing, for instance, simple, double, and triple silicates. The method of Nordenskiöld (1827) has a similar groundplan, but proceeds still more decidedly upon the proportions of the number of atoms of the different elements; and thus, as Berzelius observes, (Jahr Bericht, viii. 188.) presupposes a complete knowledge of the chemical composition of each species.

Bonsdorff's Essay on this subject was printed at Abo in Finland, in 1827; but its publication was prevented, the im-

[page] 359

pression being destroyed in the calamitous conflagration of that city (Sept. 1827). Its plan was in some respects similar to the, preceding ones; but it made the number of ingredients, and not that of the atoms, its basis. Keferstein also published a mineral system in the fourth volume of his Geognostiches Textschland; he has imitated Beudant, who formed his system upon Ampere's circular arrangement of simple bodies, having, constructed for himself another circular arrangement of eight members, on which his method is grounded.

Notwithstanding the confusion which may naturally be expected to arise from the conflict of so many different plans of arrangement, we may see, I think, a tendency in the chemical and mineralogical methods to approach towards each other. It has, been now proved that neither course can by itself lead to a satisfactory classification. Those who wished to arrange by external characters alone, trusted much to the acuteness of their senses, and believed that they could, by a sort of instinct, make out which of the perceptible properties of substances were most important. The chemists, on the other hand, deeming that their knowledge of the constitution of a mineral must be sufficient to determine its nature and place, did not consider that we require observation to teach us in what mode such knowledge is to be applied;—for what but the comparison of external characters can teach us, for instance, that tellurium, sulphur, selenium, arsenic, discharge similar functions in the composition of minerals, and must be similarly employed in our classification?

It will probably be allowed that a system of arrangement proceeding on strictly chemical principles, which should bring together in all cases the substances which most resemble each other in external properties, would satisfy the requisitions of the science, and that nothing short of this would do so. Such a system is not at present within our reach; but it will perhaps be useful to look upon all methods of classification now proposed, as attempts to approximate to such a perfect system, whether they be founded upon external characters or on chemical principles. Our knowledge of neither of these branches of the subject is as yet complete enough to lead us to expect from it a system which shall be exact according to both; but we may be held to have made some progress in the requisite series of trials and conjectures, when we have constructed any chemical classes which consist of substances of similar character and properties.

That some progress of this kind has already been made cannot be denied. The new system of Berzelius, or that of Beudant, or indeed any of the new chemical systems, would

[page] 360

produce a grouping of substances which would at once be recognised as far more natural than that of Haüy or Phillips.

The new system of Berzelius has been adopted in the arrangement of the minerals of the British Museum in their new apartment, under the intelligent and industrious superintendence of Mr. König; and every one will probably be struck by the evidence which the aspect of the collection offers, of the advantage of this over the ancient mode of arrangement.

Besides the natural historical and the purely chemical, we may observe that there are some which may be called mixed systems of classification. These, proceeding as if our knowledge were as yet too incomplete to allow us to apply any one principle with logical severity, borrow their resources from various quarters, and may thus perhaps make the nearest approximation to their object which is now possible.

One of the best of such systems appears to be that of M. Naumann, of whose crystallographical labours we have already spoken. His Mineralogy (Berlin 1828,) contains, in a very comressed form, very full and systematic accounts of the different kinds of minerals, their properties and crystalline forms; and the species are there arranged in classes and orders, which bear chemical titles, and which bring together similar bodies. Thus the Silicides are— Unmetallic Hydrous Silicides (the zeolites),Unmetallic Anhydrous Silicides (felspar, &c),Mixed (of metallic and unmetallic)Anhydrous Silicides (pyroxene, amphibole, &c.),Metallic Hydrous Silicides (dioptase, silicate of zinc, &c.); and so on. In the sulphurides Naumann retains the distinction of pyrites, glance, and blende, though its chemical signification has not yet been discovered; and this seems to be done not without reason, for the difference of octohedral copper pyrites (cuivre pyriteux, =2 (2 cu +s) + fe + s), and rhombic copper glance (cuivre sulfuré, =2 cu + s,) is of the broadest kind: the sulphuret of zinc (zinc blende), and the sulphuret of iron (iron pyrites) have scarcely any resemblance.

I may also notice the work of M. Kobell, who has published a "Characteristic (Nürnberg 1830,) founded on mixed physical and chemical characters, as those of Mohs and Breithaupt were on physical alone. M. Kobell has however, in some parts of his classification, returned to the arrangement by the electro-positive element, which appears to be a retrograde step on our road to a permanent system. It may be mentioned here that the Mineralogy of M. Beudant contains an excellent chemical "characteristic," that is, certain and universal formularies of tests for determining the place of a given mineral in his arrangement. This part of the work has been adopted by Naumann.

[page] 361

III. Nomexclature.—Some of the authors of mineralogical classifications have endeavoured to introduce a systematic nomenclature into the science. Mr. Mohs in particular has given a series of names almost wholly new to the species of the mineral kingdom. There is no doubt that mineralogical nomenclature has long been in such a state of perplexity and disorder, so defective in all valuable qualities, and so overloaded with useless synonyms, that any reformation of it would be a most important service. It appears reasonable to suppose also, that the introduction of a right and consistent classification ought to be attended in Mineralogy, as it was in Botany, with the introduction of a reformed and simplified language. But we may well doubt whether we have yet reached the point at which such a systematic reform is possible. Our genera and orders are probably too unstable to be made the basis of permanent names. Some of the groups of species, indeed, are pretty well characterized, and have in some degree influenced the common names; thus,Iceland spar, or calc spar, fluor spar, heavy spar, iron spar, seem, by these names, to be referred to a natural order Spar. In like manner native gold, native platinum are connected by the form of their names, as they are by the simplicity of their constitution. The oxides, as red copper oxide, red iron oxide, are a chemical order which can generally be recognised by their appearance, though at one extremity the metallic silicides approach near to them. The orders Pyrites, Glance, Blende have been already noticed. Names referring to such groups as these seem more likely to be permanent than any others, though it must be acknowledged that such groups are often vague; nor can we at present draw their boundary lines. In forming the names of species, the crystallization seems to give one of the best, because the most certain and definite, grounds of nomenclature:hexahedral iron pyrites, rhombic iron pyrites, hexagonal iron pyrites are names which admit of no confusion.

The disorder of our mineralogical nomenclature has been much increased by the facility with which new species have been assumed, and new names applied to them. The rebuke of Berzelius (J. B. vii. 180,) has not been uncalled for. "The mineralogists par excellence," says he, "that is, they who do not trouble themselves about the internal nature of minerals, appear to hold new names for an essential thing; for they hasten to give them before they can possibly know whether they have before them a combination already known or not." In another place (J. B. v. 197,) he appears to consider our countrymen as peculiarly given to offend in this respect. "It is the fashion in England," he says in 1824, "to seek new forms by crystal-

[page] 362

lometrical examination, to give a new name to each conjectural new mineral, generally the name of a living person, and then to give the mineral to Mr. Children to try with the blowpipe, if the pieces are sufficient."

The best method (as appears to me,) to control in some degree this inconvenient multiplication of names and species, will be to require that the name of the species should contain, besides the distinctive term, the name of the order to which it appears to belong (as Spar, Oxide, Pyrites), and an adjective designating the system or some peculiarity of the crystallization (as Hexahedraliron pyrites). The termination of the word, where it is a new one, might also be made to imply some distinction. This has been considered a matter of indifference. The termination ite has hitherto been most common: but ose, ine, and various others, appear to be coming into favour. Thus we have in Beudant's last edition (1830,) not only leadhillite, lanarkite, chamoisite, proustite; but also scolexerose, opsimose, argyrose, argyrythose, exanthalose, rhothalose; diacrase, panabase, neoplase; neoctese; rhodoise, stibiconise, crocoise, malaconise; marceline, wilhelmine, carbocerine, mysorine; exitele, and many more of the same kind; and these terminations are employed without any regard either to any etymological principle or to any difference in the minerals. Even where these names are not superfluous, or superfluously long, or superfluously learned, they are superfluously varied; and to make the variety depend on caprice alone, is to throw away a resource of which chemical nomenclature may teach us the value.

M. Beudant himself has pointed out the advantage which would result from retaining in the names of species a substantive marking the order to which the substance belongs (vol. i. 525); thus he would say silicate stilbite, silicate chabasie, silicate scolezite, &c.; carbonate calcaire, carbonate witherite; sulfate couperose, &c. Mr. Mohs had long before, in the nomenclature which he proposed, founded on characters altogether different, proceeded upon this as an indispensable condition of the specific designations.

6. Particular Discoveries and Researches.

Several of the labours of mineralogists which would naturally come under this head have already been referred to in speaking of the general views which they illustrate; and to speak of the examinations of particular minerals would lead us into too long an enumeration. Such inquiries are best conducted when the chemist and mineralogist are joint labourers, as in the ad-

[page] 363

mirabe examination of the ores of manganese by Dr. Turner and Mr. Haidinger. For similar reasons I shall not dwell on the researches which have been carried on with regard to other physical properties of minerals; as, for instance, Köhler's on their electricity, and Sir D. Brewster's on their pyro-electricity. The discovery of a new metal,vanadium, by Sefström, is closely connected with mineralogy, but will probably appear as part of the history of chemistry, and therefore need not here be dwelt upon.

In reviewing the account which has been given of the recent labours of mineralogists, it is impossible not to be struck with the small share which Englishmen have taken in all that relates to System in this science. With regard to optical researches, we have already mentioned that one person in our own country has done incomparably more than all the experimenters of the Continent together; and in the measurement of the angles of crystals, the goniometer, without which no measure would have any value, is the invention of an English philosopher; and Mr. Phillips and Mr. Brooke have contributed to the stock of crystallography, observations more numerous and exact, probably, than any other two names could rival. Yet in the adoption of new generalities we have been slow: the distinction of the crystalline systems is not commonly employed among us; the doctrines of isomorphism are contested by some and applied by few Englishmen*; and no attempt has recently been made, nor any interest excited, with regard to scientific views of the classification of minerals. This prosecution of details, and apathy or contempt with respect to methods, appears to be a part of the intellectual character of this country. Men here appear to feel no interest with regard to rules and systems till they are so complete, so clearly developed as to principle, their apparent difficulties so far explained, that the general rule will bear a strict application in each particular instance. They are disposed to despise the dim glimmerings of dawning principles, in cases where, though a connexion may be probable or certain,the asserted connexion is clearly not exact. Our countrymen thus often lose much of the pleasure and honour which belong to those who labour to unfold an obscure and imperfect truth: but yet, on this very account, their discoveries, when made, have a more positive character and a more original tone than they might otherwise possess. The step to which mineralogy at

* It will be seen by reference to the proceedings of the Chemical Section, that Dr. Turner, Mr. Brooke, and Professor Miller have undertaken to bring before the next meeting of the Association the result of experimental researches on this subject.

[page] 364

present owes the best portion of its scientific character, was made by an Englishman,—the doctrine of Definite Proportions: and if Englishmen seriously propose to themselves the task, we are justified by the history of Science in asserting that none are more likely than they to solve the great problem of mineralogy which now offers itself,—the connexion of chemical composition and crystalline form. Besides this great problem, it has appeared in the course of this Report, that various other questions of narrower extent remain to be decided by experiment. We will recapitulate a few of these.

1°. To determine the optical differences on which depend the distinctions of the different kinds of lustre,metallic, adamantine, vitreous, resinous, pearly.

2°. To determine whether the oblique rhombic prism is a real system, or is a hemihedral form of a right prism.

3°. To determine the limits of magnitude and simplicity in crystallometrical ratios.

4°. To determine whether chemical groups are strictly isomorphous, or only plesiomorphous,

5°. To determine whether the angles of plesiomorphous crystals are separated by definite or by indefinite steps.

6°. To determine what are the differences of chemical composition corresponding to differences of optical structure in resembling minerals, as apophyllite, tesselite, leucocyclite.

I will further add, that the formation of good collections of well-crystallized minerals, (in which should be included suites of artificial Crystals, both from fusion and solution,) and all arrangements which make such collections accessible to the working mineralogist, are circumstances highly important to the progress of mineralogical knowledge.

The determination of many of the above and similar very essential questions, must depend on observations made on crystals which are generally difficult to procure sufficiently perfect and transparent for such researches, there being often only a few known specimens in the world which would answer the purpose, and these having an enormous and fantastical money value affixed to them as rarities. I conceive, therefore, that all persons and Societies possessing splendid and beautiful minerals, if they are desirous that such possessions should be of use to the advancement of science, cannot in any other way have nearly so good a chance of furthering this object, as by placing these treasures at the disposal of the intelligent and skilful optical or chemical experimenter. While the unique crystal stands on its shelf unmeasured by the goniometer, unslit by the optical lapidary, unanalysed by the chemist,—it is merely a piece

[page] 365

of furniture, and has no more right to be considered as anything pertaining to science, than a curious china tea-cup on a chimney-piece:—given up to the mercy of the philosopher, there is no possibility of determining what valuable information it may not convey,—what grand series of truths it may not originate or establish.

Report on the Progress, Actual State, and Ulterior Prospects of Geological Science. By the Rev. W. D. CONYBEARE,F.R.S. V.P.G.S. Corr. Memb. Institute of France, &c. &c. &c.

IT cannot be necessary, before an assembly like the present, to expatiate on the interest of the science to which I have now to call your attention; a science which by investigating the traces indelibly impressed on the surface of our planet by the successive revolutions it has undergone, proposes to elucidate the history of these stupendous physical actions; and thus fully to develop what may be termed the archæology of the globe itself,—a science which associating itself to those branches of our knowledge which relate to organized nature, to zoology, and to botany, affords to each the important supplementary information of numerous species which have long vanished from the actual order of things;—thus unexpectedly extending our views of the various combinations of organic forms; and in many instances supplying links, otherwise wanting, in uniting the different terms of this series in a continuous and unbroken chain.

Nor, if from these higher views of scientific interest we advert to the more practical considerations of utilitarian importance, and applicability to those œconomical arts on which our national wealth and strength depend, can we think meanly of a science which guides us in the full development of our mineral resources; which, (to mention only a single instance,) in indicating the proper line in which researches for coal may offer the prospect of success, extends, facilitates, and œconomizes the supply of this article, the great element not only of domestic comfort, but of mechanical power.

In tracing the progressive development of this science, it might have been interesting, had the bounds necessarily prescribed by an occasion like the present permitted, to have commenced our examination with the records of classical antiquity. We might have noticed the apparent connexion of many of the

[page] 366

cosmogonical tenets of the philosophical schools with our subject, and indicated the allusions in the writings of historians, geographers, and naturalists, to some of the more striking phænomena of geology; those especially which, by the occurrence of marine remains in the midst of our continents, attest the displacement of the ocean from the regions which it must once have covered: even at this early period we should have seen volcanic agency referred to, as affording the most probable explanation of these striking facts, and the elevation by these forces of considerable tracts, such as that near Methone, pointed out as analogous cases.

On the revival of literature and science, we might have observed that Italy, the earliest and most active country in that bright career, by no means neglected this subject. Numerous and interesting are the anticipations of subsequent discoveries in this science, which may thence be gleaned. From the age of Boccaccio the subject had there received frequent attention, and long before a similar spirit had extended to other countries we find Steno in possession of many of the fundamental facts of geology,—a distinct recognition of various successive formations; of the dislocations and fractures of the strata; of the orderly distribution of organic remains, &c.: but to these subjects I can now only briefly allude; and I regret this the less as I have years ago submitted to the public a concise statement of these historical particulars, and the outline I then offered has been far more ably filled up by subsequent writers, especially by Mr. Lyell, to the early chapters of whose important work I would particularly refer those who may desire satisfactory information on this part of our subject.

I will only therefore allow this earlier period of the progress of geological science to detain us a very few moments while I point out the claims of one most distinguished philosopher, the universal Leibnitz, who honoured this branch of physical speculation by devoting to it a portion of his attention, and anticipated, with the prophetic sagacity of a powerful mind, its future progress, and the very methods of investigation which would most effectually contribute to its successful development. I am induced to pause on the consideration of the geological treatise of this most eminent writer (his Protogœa), because I am persuaded that its merits have been seldom sufficiently appreciated, and admit of being most prominently exhibited by being brought into immediate contact with the subsequent discoveries of our science.

In the 4th section of his Protogœa, Leibnitz presents us with a masterly sketch of his general views; and perhaps, even in

[page] 367

the present day, it would be difficult to lay down more clearly the fundamental positions which must be necessarily common to every theory, attributing geological phænomena in great measure to central igneous agency. He attributes the primary and fundamental rocks,—"id enim potissimum de prima tantum massa, ac terræ basi, accipio,"—to the refrigeration of the crust of this volcanic nucleus;—an assumption which well accords with the now almost universally admitted igneous origin of the fundamental granite, and with the structure of the primitive slates; for the insensible gradation of these formations appears to prove that gneiss must have undergone in a greater, and mica slate in a less degree the same action, of which the maximum intensity produced granite*. The dislocations and deranged position of the strata (phænomena for which he cites the writings of Steno,) he attributes to the breaking in of the vast vaults which the vesicular and cavernous structure assumed by masses during their refrigeration from a state of fusion, must necessarily have occasioned in the crust thus cooling down and consolidated. He assigns the weight of the materials, and the eruption of elastic vapours, as the concurrent causes of these disruptions;—"denique vel pondere materiæ vel erumpente spiritu fractâ fornice:" to which we should perhaps add, that the oscillations of the surface of the still fluid nucleus may, independently of any such cavities, have readily shattered into fragments the refrigerated portion of the crust; especially as, at this early period, it must have been necessarily very thin, and resembling chiefly the scoriae floating on a surface of lava just beginning to cool. He justly adds, that these disruptions of the crust, must, from the disturbances communicated to the incumbent waters, have been necessarily attended with diluvial action on the largest scale,—"maximæ secutæ inundationes." When these waters had subsequently, in the intervals of quiescence between these convulsions, deposited the materials first acquired by their force of attrition, these sediments formed, by their consolidation, various stony and earthy strata: —"Nec dubito postea materiam liquidam in superfìcie telluris procurrentem, quiete moxredditâ, ex ramentis subactis ingentem materiæ vim deposuisse, quorum aliæ varias terræ species formarunt, alia in saxa induruere, e quibus strata diversa sibi

* Whatever theory be entertained as to the origin of these rocks, a gradual transition of character from granite to gneiss and mica slate, assuredly exists; and it is foreign to our present purpose to pronounce on the more minute question, concerning their origin,—further than to observe, that if we admit the igneous origin of granite, this gradation of character appears to indicate a gradation of igneous action on gneiss and mica slate also.

[page] 368

superimposita diversas præcipitationum vices atque intervalla testantur" Thus, he observes, we may recognise a double origin of the rocky masses, the one by refrigeration from igneous fusion, (which, as we have seen, he considered principally to be assignable to the primary and fundamental rocks,) the other by concretion from aqueous solution:—"Unde jam duplex origo intelligitur primorum corporum, una, cum ab ignis fusione refrigescerent, altera, cum reconcrescerent ex solutione aquarum." We have here distinctly stated the great basis of every scientific classification of rock formations. By the repetition of similar causes (i. e. disruption of the crust and consequent inundations,) frequent alternations of new strata were produced, until at length, these causes having been reduced to a condition of quiescent equilibrium, a more permanent state of things emerged:—"Redeunte mox simili causa strata subinde alia aliis imponerentur, et facies teneri adhuc orbis sæpius novata est." Have we not here clearly indicated, the data on which what may be termed the chronological investigation of the series of geological phænomena must ever proceed? But I would particularly invite to the following clause the attention of those writers of the present day, who appear to assume it as an essential condition of their theories, that the same physical causes can never, under any former circumstances, have acted with more intense energy than they actually exert:—"Donec quiescentibus causis, atque æquilibratis, consistentior emergeret rerum status."

The beginning of the following section is very remarkable, as exhibiting a clear anticipation of the importance and of the prospects of the new science, of which he foresaw the dawn:— " Hæc de incunabulis nostri orbis semina continent scientiæ novæ quam Geographiam Naturalem appelles." Leibnitz proceeds even distinctly to indicate the line of future research into the geographical distribution and extension of the various formations, which might be expected to place this new science on a firm basis:—"Rectius tamen omnia definient posteri, ubi curiositas mortalium eo processerit, ut per regiones procurrentia soli genera et strata describant." And then, after making judicious remarks on the distinctions of general and local causes, he modestly and prudently adds, that before we are able to determine to which of these the phænomena we observe are to be attributed, we must wait patiently until the whole surface of the planet shall have been more accurately examined:—"sed quid privatts imputandum sit, aut publicis causis, facilius aliquando statuet poateritas, exploratâ melius humani generis sede." How much of the Wernerian doctrine of universal formations would not a proper attention to this caution have

[page] 369

spared us! Nor can we even yet pronounce that our knowledge of the general structure of the earth's surface is sufficiently advanced, or a hasty spirit of speculation sufficiently moderated, to render the same caution inapplicable or superfluous in the present day.

Another distinguished philosopher of the same age, well worthy of being mentioned in connexion with Leibnitz, our countryman Hooke, (who may almost be considered as having anticipated Newton in his application of the great principle of gravitation to the mechanical system of the universe,) affords an additional example that the greatest minds of this period fully appreciated the high importance of geological inquiries. Much of his posthumous works is dedicated to this subject, especially to the investigation of the arguments derived from geological phænomena in favour of the hypothesis of the volcanic elevation of our continents.

For the reason already assigned, I must pass equally rapidly over the many other interesting topics connected with the earlier history of our science, until Werner—closely, however, treading in the steps of his countrymen Lehman and Fuchsel—at length combined the results previously obtained into a more methodical and systematic arrangement, and, by the ardour of his genius and the influence of his popular lectures*, attracted to geology a degree of general attention which it had assuredly never before received. The previous labours, however, of

* We are chiefly indebted to the reports of his pupils, especially to those of Jamieson, for our knowledge of Werner's general views as fully developed in his lectures, and these only; for his own two short publications, the Kurze. Klassification and Essay on Veins, are confined to partial subjects. Prom these reports of his lectures, I feel convinced that it is to him we are indebted for the first general announcement, that the various species of organic remains grouped together in the rock formations bear a constant relation to the age of those formations: the Italians much earlier, and more recently Rouelle in France, had recognised their regular distribution in certain associated groups; but the distribution to which they referred appears to have been, according to their views, rather topographical than stratigraphical, whereas Werner clearly regarded it in the latter light; thus he characterizes the transition limestone as containing corallites, encrinites, &c., which though not absolutely confined to this formation, yet gradually disappeared in the newer rocks, becoming replaced by other species which never appeared in the transition series. The organic remains of the floetz rocks he regarded as increasing in quantity and variety, the newer the formation; he particularly specifies the most characteristic fossil shells of the gryphite limestone, the muschelkalk, chalk, &c. We should also mention that during the progress of Werner's observations, Saussure, in the excellent geological agenda published at the conclusion of his "Voyages," suggested the solution of the same great problem in the following terms, which state its conditions with the most admirable clearness and precision, "Constater s'il y a des coquillages fossiles qui se trouvent dans les montagnes les plus anciennes; et non dans celles d'une formation plus récente, et classer ainsi, s'il est possible, les ages relatifs et les époques de l'apparition des différentes espoces."

2 A

[page] 370

Saussure in the Alps, of Palassou in the Pyrenees, and of Ardouino Ferber and Fortis in Italy, had at that time collected such a rich store of materials as required only the intervention of a compiler and digester to apply them at once to the purposes of a more comprehensive system. In a former publication, before referred to, I have endeavoured to show what England had contributed to this store; but I am happy to find that at the present moment this interesting subject is about to receive a much fuller illustration from the pen of my friend Dr. Fitton*.

The progress of geology from the period at which it thus began to assume the systematic character of a regularly digested science, may be considered as having presented three marked stages, distinguished by three successive schools; each of these schools has selected for the more especial object of its attention a single member of the three great geological divisions in the series of formations, i. e. the primitive, secondary, and tertiary; and the succession of these schools has, by a singular coincidence, followed the same order with that of the formations to which they were devoted: it may also be observed that the leaders of each school have been distinguished geologists of three different nations,—Germany, England, and France. The first, or German school, is that of Werner: this directed its attention principally to the primitive and transition formations†, in which the distinctions of mineralogical character assume the greatest importance; and the imbedded minerals, from their variety, and relations to the rocks containing them, become the chief objects of the geologist's notice. The second, or English school, has distinguished itself by the ardent and successful zeal with which it has developed the whole of the secondary series of formations: in these the zoological features of the organic remains associated in the several strata, afford characters far more interesting in themselves and important in the conclusions to which they lead, than the mineral contents of the primitive series. This school generally recognises the masterly observations of Smith, first made public in 1799, as those which have principally contributed to its establishment; although the regular distribution of organic remains had before

* Now published in the 1st and 2nd vol. of the Lond. and Edin. Phil. Mag.

† In the early works of one of the ablest British disciples of this school, whose meritorious labours undoubtedly contributed very largely to the diffusion of an ardour for geological inquiries in this island, there occurs a curious illustration of -the exclusive attention to the older rocks. In the general view of geology contained in the Introduction to Professor Jamieson's Account of the Hebrides, 1800, after a sufficiently full detail of the various primitive formations, we find the whole secondary group dismissed in these few vague words: " They consist of limestone and argillite, with numerous petrifactions; also basalt, porphyry, pitchstone, greenstone, wacke, and the various coal strata."

[page] 371

been recognised in Italy by Steno, and in France by Rouelle; and although Werner in his lectures, and Saussure as above quoted, appear to have indicated generally, that the laws of this distribution bore a relation to the geological age of the formations containing them, yet a degree of vagueness hung over the whole subject, which precluded any extensive or useful application of this great principle, until the acute observations of Smith first brought it prominently forward in all the precision of exact detail as applied to a vast succession of formations, including the most important portion of the geological series; and as from his situation in life we must consider the discoveries of Smith as the extraordinary results of native and untaught sagacity of intellect, they must on this account be held to challenge a still warmer tribute of approbation, and may be regarded as strictly original in him, even where faint traces of anticipation may be found in Continental writings little likely to have fallen beneath his observation. The third school, or that of Tertiary Geology, owes its foundation to the admirable Memoir on the Basin of Paris, published by Cuvier and Brongniart, 1811. Although this school was certainly subsequent in point of date to that of Smith, yet those who had already directed their attention to such pursuits at this period, must well remember that the Wernerian school of primitive and mineralogical geology having previously obtained an undisputed and exclusive ascendancy in the minds of most of those who possessed any influential station in the scientific world, the observations of the individual alluded to had little chance of recommending themselves at first to public notice, and that in fact the knowledge of them appears to have been for ten years chiefly confined to a small circle in the neighbourhood of Bath,—-until the high scientific distinction of Cuvier, and the striking and interesting nature of the facts developed in his brilliant Memoir, excited a marked sensation and commanded the general attention of men of science; for none such could peruse with indifference those masterly descriptions, which exhibited the environs of one of the great metropolitan cities of Europe as having been successively occupied by oceanic inundations and fresh-water lakes; which restored from the scattered fragments of their disjointed skeletons the forms of those animals, long extinct, whose flocks once grazed on the margins of those lakes; and which presented to our notice the case of beds of rock only a few inches in thickness, extending continuously over hundreds of square miles, and constantly distinguished by the same peculiar species of fossil shells.

The public mind being thus fully awakened to a perception of the vast importance of zoological geology, as superadded to

2 A 2

[page] 372

mineral geology, became thus ripely prepared to appreciate the value of the materials previously collected by the unassisted acuteness and industry of Smith, which had illustrated the whole secondary series of formations in the same spirit as Cuvier and Brongniart had applied to a portion of the tertiary class, and which thus, after an interval of neglect, assumed their just place and rank in the geological system*.

From this period the views of the zoological school were universally adopted by the most active and efficient labourers in the progress of English geology, and were by them from time to time greatly extended.

The establishment of the Geological Society of London in 1808, afforded also, about the same time, a central point of reunion to those engaged in this pursuit,—an establishment eminently calculated to stimulate their endeavours by the promotion of mutual intercourse, and the comparison of the information individually obtained,—a point in every science very important, but most emphatically indispensable in one which can never be effectually advanced without the steady cooperation of numerous independent observers. Besides accomplishing this, the Geological Society was also most useful as affording the facility of publication to the researches thus prosecuted: indeed it has been well observed, that if we consider our philosophical Societies merely in the light of publishing engines, we shall have no mean idea presented to us of the very important advantages which they yield to science.

The first volume of the Transactions of the Geological Society was published in 1811, and it well illustrates the actual state of the science at that date: the greater part of its contents obviously belong to the Wernerian school, which we have characterized by its almost exclusive attention to primitive and mineralogical geology. The paper by Dr. Berger on the Geology of Dorset, Hampshire, and the Isle of Wight, will well exhibit the low state of secondary geology at that period; but another paper of Mr. Parkinson on the Organic Remains of the neighbourhood by London, including a comparative view of Cuvier's then recent discoveries in the basin of Paris, sufficiently evinces the dawn of a more intelligent system; and it deserves remark, that the introduction of this, the first respectable paper on secondary geology

* As the bases of this advanced geological system mainly depend on an exact knowledge of the zoological characters of the remains contained in the strata,— a knowledge extending to the most minute specific differences,—this could scarcely have been attained anteriorly to the considerable additions made by the French systematic writers, especially Lamarck, to the arrangements of the Linnæan school.

[page] 373

published by the Society, expressly refers to the discoveries of Smith, as the great basis on which all sound and really scientific researches on this subject must be established.

We cannot better illustrate the rapid march of geology from the period when this new light burst in upon the system, than by comparing the Memoir on the Isle of Wight and the Dorsetshire Coast, published by Mr. Webster in the second volume of the Geological Transactions, with the meagre notices of the same district by Dr. Berger, already alluded to as having appeared in the former volume. In this paper Webster ably follows the admirable model presented by Cuvier and Brongniart's Memoir of the Basin of Paris; with the geological structure of which he shows that of the Isle of Wight closely to agree, both districts exhibiting the very same alternations of marine and fluviatile beds* reposing on the chalk; while in one respect the phænomena observed in the Isle of Wight are rendered even more interesting than those of the Parisian basin, by the violent convulsions which have here dislocated the strata and thrown a large portion of them from a horizontal into a vertical position. If we compare this Memoir of Mr. Webster with the preceding one of Dr. Berger, they at once show themselves to belong to two very distinct eras of science; and it is difficult to believe that the interval which elapsed between their respective publication was only three or four years.

The publication in 1815 of Smith's general geological map of England†, succeeded by his more detailed separate county maps, illustrated by the work of the same author on "the English Strata identified by Organic Remains," and by the contemporaneous production of Sowerby on Mineral Conchology, filled up the whole great outline of English geology, and left to those who followed little more than the task of condensing and concentrating what was already ascertained, and enlarging and rendering more precise the detail. I should speak, however, in more distinguished terms than these, of the great geological map of England drawn up by Mr. Greenough, and published by the Geo-

* The same anoplotheria, &c., have subsequently been found in the fluviatile formations of the Isle of Wight as in those of the basin of Paris.

† It is quite erroneous, however, to attribute, as has been sometimes done, to Smith the earliest attempt to execute such maps; their construction was originally proposed by Lister, 1684; in 1746 Guettard published many such maps, although their execution was necessarily at this early period vague and imperfect; and before 1796 Dr. Maton had thus delineated the geology of the southwestern counties; and the various Reports of the Board of Agriculture had included similar representations of Yorkshire, Nottinghamshire, Derbyshire, Kent, and Devonshire.

[page] 374

logical Society in 1819. This map, as compared with the earlier publication of Mr. Smith, will be found to present, in all the districts occupied by formations older than the lias, corrections of the most material description; and in the more recent formations, where both maps generally agree, that agreement is in itself important as a confirmation of the accuracy of each, as that of the Geological Society was in no instance a copy from its predecessor, but entirely the result of independent observations collected during frequent, extensive, and laborious journeys. Those who have never seen the immense collection of materials in Mr. Greenough's most valuable manuscript geological note-books, can have little idea of the immense labour which he bestowed upon this object: his library also contains a vast collection of materials, equally important, in illustration of Continental geology; and it is greatly to be regretted that these still remain entirely unpublished. To no one individual does the progress of our science stand more deeply indebted than to the first President, and I may well add principal founder of the Geological Society, which, without his unwearied zeal and unstinted devotion of his talents, time, and pecuniary resources, could never have struggled through the numerous difficulties which embarrassed the first years of its existence*.

While geology on the Continent was advanced by the labours of Von Buch in Germany and Scandinavia, and by the able general systematic works of Daubuisson and the universal Humboldt, we may here pause to observe what had been accomplished by our own London Society in its earlier years before the close of its first series of Transactions in 1821. Already had its contributions completed to a high degree of perfection all the most important details of English geology; and besides this, the eastern half of Ireland had been very exactly described in the Memoirs of Dr. Berger and Mr. Weaver,—the latter especially well deserving the highest consideration, both from the copiousness and precision of its details, and the extent and beauty of its graphic illustrations†. The primitive districts and the West-

* Considering that this Report was originally delivered within the walls of the theatre of Oxford, I cannot refrain on this occasion from repeating the acknowledgement which I formerly made in the Introduction to my "Outlines,"—"that we owe the introduction of these pursuits into our University to lectures delivered between 1805 and 181O by my much valued friend Dr. Kidd, whose more private exertions in encouraging the rising talents of others were as successful in effect as liberal in design."

† The same author has since, in 1831, communicated to the Geological Society a similar Memoir of the south-western counties; so that the north-western portion of Ireland is all which remains undescribed at the present time.

[page] 375

ern Islands of Scotland had also received very important illustrations in several Memoirs communicated by Dr. MacCulloch* In relation to Continental geology, the very able Memoirs of Mr. Strangways, which so greatly extend our knowledge of the physical structure of an important portion of the Russian territory, claim especial notice; and the first series of our Geological Transactions also contained some valuable papers on portions of our Indian empire, on Ceylon, and on Madagascar.

I have been principally induced, in the present summary of the progress of geological science, to draw a line at the close of the first series of our Geological Transactions in 1821, because an author already alluded to has asserted in a recent publication, that "since that year geology has received scarcely any valuable additions, and not a single fundamental one." Drawing a line at this point, therefore, I shall endeavour to give a slight sketch of the contributions which have really marked the progress of the science during this supposed period of inanition, leaving it to your judgement how far they really deserve the above depreciating character.

Now although previously to this period the main features of English geology had been very amply illustrated, yet even in this province, where least remained to be accomplished, our additions have neither been few nor unimportant; and if we turn to Continental Europe, we shall find that what was then comparatively a blank, has been now filled up to such a degree that we are actually in possession of nearly as good materials for a general geological map of Europe at the present day as we were for one of England only at the former date; and to this, observers from our own country have contributed no less than their ablest Continental brethren. Nor let it be imagined that this only supposes an extension of our knowledge in insulated details: it is in truth far otherwise; since extensive comparative geology affords the only materials for obtaining the fundamental facts of our science. It is by this inductive process alone that we can hope to collect and combine the data which exist for what may be termed a general geological chronology. It is thus only that we can ascertain to what extent and under what modifications the same geological causes have acted at the same epochs. It is thus only that we can learn, what have been the violence, extent, and epochs of the disturbing and ele-

* These Memoirs were embodied in the work of that author on the Western Islands, published in 1819. His treatise on the Classification of Rocks, published in 1821, also claims notice as a very useful manual. Those who may have looked into his recent System of Geology, will feel why, in kindness to his reputation, his friends must here wish to close their survey of his publications.

[page] 376

vating forces which have affected the strata,—whether similar groups of organic remains universally, and in the most distant countries, characterize contemporaneous geological deposits,— or whether those zoological species are not rather restricted (like most of the species of the actual period,) to different geographical districts. All these are evidently questions at the very root of any sound geological theory, whenever the time shall be fully ripe for constructing such a theory; and although it were assuredly premature to assert that this time is even yet completely arrived, we may nevertheless boldly assert that no eye at all capable of appreciating these problems, or the appropriate evidence tending towards their solution, can glance over the discoveries of any single year since 1821 without observing a very rapid accumulation of the most valuable materials for their elucidation. During the same period, moreover, our knowledge of the principal volcanic districts, both those which are still in activity and those now extinct, has been advanced to the greatest degree of precision; and the whole of that which is perhaps the most important geological series*,—that of the tertiary formations, with the lower members of which alone the previous researches of Cuvier, &c. had made us acquainted, —has within the few last years received an additional development, no less important than that which, in an earlier stage of geological progress, the secondary system of the Wernerians received from the discoveries of Smith.

To confirm this general statement, it will be necessary to enter more minutely into the detail of the recent progress of geological discovery. To begin, then, even with that part of our subject which we have admitted to have been far the least promising of interesting novelty; with reference to the series of English strata alone, the corrections and additions to our previous information since 1821 have not only supplied such details as were of local interest, but such as were moreover often pregnant with important general consequences: the rectification, for instance, of the previous arrangement of the subcretaceous sands has brought to light in the Weald of Kent† a fresh-water formation, previously unknown as such, between these sands and

* The tertiary period is especially important in systematic geology, inasmuch as since, on every hypothesis, the geological causes must have acted during this period under conditions most nearly approximating to those which belong to the actual order of things: the formations of this age therefore afford the most essential link in connecting our actual experience with our speculations on the former state of our planet.

† The lowest bituminous clays of this formation have also been noticed on the opposite side of the Channel in the Boulonnois; and traces of them are said to have been observed, in our midland counties, in Buckinghamshire.

[page] 377

the subjacent Purbeck limestone; thus showing that the alternation of oceanic and lacustrine deposits was certainly not confined to the tertiary epoch, but had equally occurred in the more ancient periods*. Messrs. Murchison and Sedgwick have observed similar lacustrine deposits in still older rocks in the Isle of Sky. The elaborate Memoir of Prof. Buckland and Mr. De la Beche on the Weymouth district is not only valuable as having imparted minute and accurate precision to our knowledge of the interesting geological phænomena exhibited in the western extremity of the tract affected by the great convulsions which have elevated the chalky ranges of Purbeck and the Isle of Wight, but it informs us that this tract also, like the Weald, furnishes facts pregnant with remarkable consequences, as to the circumstances of the general surface at the period when the strata of Portland limestone were deposited; for we find interposed between these and the superincumbent Purbeck limestone, a bed of black vegetable mould, full of the stems of Cycadeæ and of large Coniferæ, many of their roots being fixed upon and still adhering to the subjacent limestone; so as to evince that they must have originally grown in their present position: the surface of the Portland limestone must therefore at that time have been dry land, bearing a thick growth of tropical vegetables†.

In the inferior oolite of Yorkshire, associated with the coaly

* A similar conjecture had indeed been previously entertained concerning the fluviatile origin of still older rocks, including portions of the coal measures; but the evidence resting only on the occurrence of obscure shells, referred perhaps too hastily to the fluviatile genus Unio, must be regarded as very insufficient, and appears opposed by the undoubtedly marine shells of the associated carboniferous limestone. It is however certainly by no means improbable that the coal strata have in part at least originated in the drift timber of vast æstuaries like that of the Mississippi; and in such localities this intermixture of fluviatile shells might naturally be expected.

† It must surely be unnecessary to insist on the fundamental importance of a fact thus affording direct evidence of the repeated oscillations by which the relative level of the ocean and land has been affected; for the Portland beds, subsequently to their having been thus exposed as a dry continental surface, appear to have been again submerged, first by an æstuary in which the fluviatile deposits prevailed, although with a partial intermixture of marine fossils as is shown in the Purbeck beds; in the adjoining district of the Weald the fluviatile character of the beds is more unmixed; a second oceanic submersion must have produced the vast mass of the cretaceous superstrata. And lastly, the alternating oceanic and fluviatile deposits of the Isle of Wight seem to attest a recurrence of similar oscillations. Prevost indeed, in a Memoir on this question, opposes the idea of reiterated oceanic submersions, and endeavours to explain the phænomena of the basin of Paris by the hypothesis of a basin originally oceanic, but converted by the gradual subsidence of the sea level successively into an æstuary and inland lake of brackish water, subject occasionally to accidental irruptions of the oceanic water on one hand and of the land floods on the other; but it does not appear possible thus to explain the vast fluviatile and oceanic deposits above described: and whereas this writer builds much on his attempt to prove that no bed in the geological series can be pointed out which appears to represent an ancient continental surface on which vegetables once grew, &c.; the facts stated in the text present a complete answer to this negation.

[page] 378

beds of the eastern moorlands, Equisetaceæ were observed by Mr. Murchison under circumstances exactly parallel, so as to warrant similar inferences.

The very valuable details concerning these Yorkshire oolites, published by Mr. Phillips, will, if we compare the series of rocks as there exhibited with the characters of the same series at its opposite extremity on our southern coast, sufficiently illustrate the great changes which took place in different parts of the very same deposits, inasmuch as the calcareous sands of the inferior oolites in Yorkshire present nearly the characters and mineral constitution of the rocks associated with the older coal formation. We thus collect a series of facts calculated to throw considerable light on the modification of circumstances which may have concurred in different ages to produce carboniferous deposits; and we see convincing proof how far we must depart from the doctrine of universal formations, (if that term be supposed to convey the notion of anything like an identity of character,) in order to approach to the truth of nature*.

The comparative view of the contemporaneous rocks of the Scotch oolitic coal-field at Brora, by Mr. Murchison, is equally important; indeed in many respects it may be considered as having suggested the line of examination pursued by Mr. Phillips†. Still more remarkable is the discovery, by the same geologist and Prof. Sedgwick, of an entirely new formation (seemingly occupying the relative position of our own carboniferous group,) in Caithness, of bituminous schist containing fish‡ and shells apparently fluviatile.

The disruption and disturbance of these newer strata by the elevation of the subjacent granite (at a period, however, evidently

* We can scarcely feel authorized from analogy to conclude that distant portions of a contemporaneous geological deposit in Dorsetshire and Yorkshire should possess absolute identity of mineral character, any more than that the mud banks deposited at the present moment on the coasts of the two counties should so correspond. At the same time, however, we must allow that geological causes appear to have acted on a much greater scale, and homogeneous depositions to have prevailed to a very considerable extent, subject however to material local modifications.

† Mr. Murchison first observed at Brora a considerable number of species of fossil shells previously unknown in the oolite; these were figured in Sowerby's Mineral Conchology, and subsequently the same species were observed by Phillips in Yorkshire.

‡ The vertebrae, teeth, and radii of fish have also previously been observed in carboniferous and even in transition limestone.

[page] 379

subsequent to the consolidation of the latter rock,) presents a fact of great importance, inasmuch as it clearly evinces that the adjoining primitive chains of mountains must have been subject at the least to two æras of disturbance; the first when the injection of the granitic mass, yet in a fluid state, rent the incumbent micaceous slates, injecting veins of its substance into their fissures*; and a second at a much later period, subsequent not only to the refrigeration of the granite, but even to the depositioh of the Brora oolites, which partook in the motions occasioned by this latter elevation, and have been in places shattered by this convulsion into fragments which have been reunited into a brecciated conglomerate.

The next geological group which requires our notice beneath the lias and oolites, is that which is universally characterized by the new red or variegated sandstone, the grés bigarre of the French, and bunter sandstein of the Germans, which is associated in its lower portion with the magnesian lime or zechstein, and (on the Continent) in its upper part with the muschelkalk and keuper. Geology stands much in need of a convenient name for this group; and I will venture therefore to propose the term Pœcilite (from the Greek ποιхιος), as expressing its characteristic rock the grès bigarré, and hence denominate the group, pœcilitic. Brongniart has already adopted the Gallicised form Pœcilien.

The elaborate Memoir of Prof. Sedgwick on the Magnesian Limestone of the Northern Counties is doubly valuable as showing at once the variations and also the identities presented by the comparative view of the same formation in distant points; while in our southern counties this formation exists only in the form of a conglomerate, derived from the debris of the older carboniferous lime united by a dolomitic paste, thus illustrating the original mode of its formation; in the northern counties it becomes fully developed in a regular series of calcareous beds, distinguished by peculiar organic remains, exactly corresponding with the zechstein and rauchwacke of the contemporaneous German deposits; while the organic remains contained are very important as forming a link between the types of the older subcarboniferous and successive newer rocks. Prof. Sedgwick has well shown how, if we take into account the intermediate formations of muschelkalk and zechstein, so amply developed in Germany, but not yet discovered in these islands, we may trace a regular graduation in the

* I believe, indeed, that at the Ord of Caithness, where the granite is in contact with the oolite, the mica slate is absent; but there is surely no reason to believe this particular granitic mass different in age from the other granites of the same portion of the Highlands, which are thus related to the mica slates.

[page] 380

types of the imbedded organic remains, thus almost observing a law of continuity between the carboniferous lime and lias. I cannot conceive that Dr. MacCulloch can ever have read these remarks; otherwise common prudence must at once have shown him the necessity of cancelling his negation, that geology had recently received any fundamental additions*.

The comparative view of our northern coal-fields† has equally extended our knowledge of the varying modifications affecting contemporaneous formations. The geology of this important district is now fully illustrated by a series of elaborate Memoirs by Messrs. Wood, Winch, Witham, Buddle, and Hutton, accompanied by detailed and accurate sectional views representing the whole Northumbrian coast, &c., published in the first volume of the Transactions of the Newcastle Philosophical Society, a work which reflects the highest credit on one of our youngest provincial Societies, and without which no geological library can be esteemed complete. It was indeed previously known that the millstone grit and limestone shale of Derbyshire became in Northumberland complicated into an extensive series of alternating limestones, shales, sandstone, and coal-beds; but an important addition to this fact has been now distinctly established, for we find that to the north

* Prof. Sedgwick has also fully illustrated the beds of sandstone lying beneath this conglomerate as seen at Pontefract, &c., which he has fully shown to be equivalent to the rothe todte liegende of Germany (an identity originally suggested by the author of this Report): a dolomitic conglomerate like that of the southern counties is often interposed between this sandstone and the regular magnesian limestone; and indeed in the southern counties, wherever the dolomitic formation is most fully developed, the uppermost beds are finely grained and gradually pass into a compact limestone. The Pontefract sandstone is completely unconformable to the subjacent coal measures, and partially unconformable also to the superincumbent magnesian lime. M. Elie de Beaumont has observed the same unconformity between the equivalent grès de Vosges and the coal measures on the one hand and the zechstein on the other. A comparison of the equally able Memoirs of the French and English geologists will be found very interesting. I cannot, however, entirely agree with Prof. Sedgwick in assigning to the Heavitree quartzose or porphyritic conglomerate a place in the series younger than the rothe todte, and equivalent to the dolomitic conglomerate; a comparison of the German and Heavitree conglomerates has convinced me of their close connexion. Mr. Hutton in the Newcastle Phil. Trans., has greatly extended our knowledge of this rock.

† The labours of Prof. Buckland, in which I had the honour of being associated with him, had before illustrated the Somersetshire and Gloucestershire portions of our south-western coal-fields, with every desirable copiousness of detail. I am at present actively employed in completing our survey of the South-Welsh portion of these districts. It is greatly to be regretted that we have hitherto no good account of the extensive coal-fields of Lancashire: in the vicinity of such philosophical institutions as those of Liverpool and Manchester, surely this desideratum ought not to be permitted to remain.

[page] 381

of Cross Fell the lower series of the carboniferous limestone, locally termed Scar lime, is subdivided in like manner, and alternates with sandstone and carboniferous shale*; the limestone decreasing and the coal beds increasing as the strata approach the transition chains of the Scotch border. The coal measures occupying the great valley of the Scotch Lowlands on the north of these chains, are with great probability referred to the same lower series; but this Scotch district has as yet been very imperfectly described.

The immense faults and dislocations of our great northern coal-field have also received the fullest illustration in the Memoirs above cited; and no geological subject can be considered more pregnant with fundamental information than this for it is only by a careful and detailed examination of the phænomena attendant on these great convulsions that we can ever hope to be enabled to speculate satisfactorily on the causes which have produced them. Now in the Newcastle district we find the strata shattered at every two or three miles interval, with fissures extending to many leagues distance, and producing subsidencies of occasionally not less than 140 fathoms, which, if they affected equally the configuration of the surface, would produce precipitous escarpments near 1000 feet high; yet is the actual level of the surface found absolutely uniform, and affording no trace whatever of the vast subterraneous disturbances;—a most striking proof of the vast mass of materials which must have been removed subsequently to their occurrence. Here we remark one of those great problems for the

* Thus in the two important groups just noticed (viz. the pœcilitic and carboniferous series), recent observations have enabled us to supply those great lacunae which previously occasioned the appearance of an abrupt transition (per saltum) from one order of geological products to another totally different; the connecting links before missing are now restored, and seem to establish a graduated and continuous order in this most important portion of the geological series. Thus in the transition formations we see alternations of slate, coarse gritty grauwacke, and beds of entrochal limestone, often associated with seams of anthracite, and, according to Mr. Weaver's late Memoir, with regular coal in the South of Ireland. In the carboniferous group we have precisely similar alternations, only that the limestone frequently prevails most in the middle regions, and the coal in the uppermost; the organic remains of the limestones approximating very nearly to those of the preceding transition series. In the succeeding pœcilitic group the organic remains still in many instances belong to the same class, and partake of the earliest type, although species of a more recent character begin to be introduced: the subjacent Pontefract sandstone (rothe todte) also exhibits a regular approximation to the coal grits. We have before seen how the graduation is continued from the magnesian limestone through the muschelkalk to the lias and oolite, and we shall hereafter learn that the same intervention of gradual links exists between the cretaceous and tertiary groups at Maestricht and in the eastern Alps.

[page] 382

solution of which we require the action of diluvial currents on the most vast and violent scale*. Trap dykes are so common in this district, and so frequently associated with the lines of fault, (see especially Mr. Wood's description of the great Statlick fault,) that we cannot but refer those dislocations principally to volcanic agency.

To proceed to the lower formations: The transition districts† of our island, for a long period after the introduction of the more modern schools, were, as if in revenge for the exclusive attention devoted to the older rocks by the Wernerian school, abandoned to comparative neglect. These, however, have recently received important elucidation from the researches of Prof. Sedgwick, who has already fully described the transition chains of the Cumbrian lake district, and is at present engaged in prosecuting a similar examination of North Wales; while his friend Mr. Murchison is simultaneously exploring the junction of the transition and secondary districts along the Welsh border. In his description of the adjoining Island of Anglesea, Prof. Henslow had already presented us with an Essay, which may well serve as a model of the manner in which such investigations ought to be conducted‡.

* On this subject I would quote an interesting note from Mr. Greenough's "Examination," p. 156.
"Mr. Hutchinson, who wrote about 1750. of whose geological opinions I have more than once had occasion to speak with much respect. was the first by whom this important fact was noticed. His words are, It is extremely rare to find a lifted edge of strata standing up above the general surface. The faults, however large the rise which they occasion, being rarely discernible by any sudden inequality of the ground, numerous ascliffs, facades, mural ascents, or precipices are, very few of them are owing to faults; in general the matter has been carried off.'"
Mr. Greenough gives other similar references to the works of Catcot, Williams, Desmarest, Playfair, Deluc, Richardson, and Farey.

† Here geology again stands in need of a term less barbarous than grauwacke slate, which would conveniently denominate the characteristic rock of this æra. Might not clasmoschist (from the Greek χλασμα,) be conveniently adopted? It would afford a term well contrasted to mica schist, the characteristic rock of the primitive group. We should thus obtain a series of convenient denominations for the various geological groups which are principally distinguished: the primitive we might call the mica schistose group; the transition the clasmoschistose group; the denomination of the carboniferous group is already sufficiently established; for the new red sandstone with the associated magnesian zechstein and rothe todte in its lower, and muschelkalk and keuper in its superior portion, I have already proposed the term, the pœcilitic group. The oolitic group, for the lias and oolites; the cretaceous group, for the chalk and subjacent greensand; and the supercretaceous group, for the tertiary formations, are appellations already commonly received.

‡ This Memoir, published in the first volume of the Cambridge Phil. Trans., is peculiarly valuable for its accurate description of the phænomena of the numerous trap dykes, and the changes and crystallized minerals which they have produced in the rocks traversed by them.

[page] 383

Mr. De la Beehe has recently been appointed to colour geologically the Government trigonometrical survey of Devon; and a complete geological map of Cornwall is at length promised by Mr. Henwood, under the auspices of the Geological Society of that county: when these works shall be executed, we may trust that the history of the older formations of this island will receive as full and satisfactory an elucidation as our secondary series has long since obtained from the labours and acuteness of our geologists.

If it be said that these questions as to our older rocks are only questions of detail, be it remembered that the boldest and happiest generalizations of science must rest on such details in the first instance. While we remain imperfectly acquainted with the various modifications exhibited by our earliest formations, (variations which must have resulted from corresponding changes in the causes which produced them,)—while we are as yet unable precisely to distinguish the disturbing forces and intrusive ignigenous masses of this period,—how can the bases of any geological theory be securely laid? Mr. Weaver has recently added to his former important paper (before noticed) a continuation which completes the geology of the South of Ireland, and now leaves only the north-western portion of that island a desideratum. The most important general feature of this paper appears to be the having ascertained the fact that the coal beds of the South-east of Ireland present an older carboniferous formation than any previously known, being associated with the transition rocks.

Before we advert to that wider and more important field, the comparative geology of the Continent, it is most gratifying to remark, that as we shall there find our countrymen distinguishing themselves no less than at home, so in return the geology of our own island has been indebted for many valuable contributions to the labours of the ablest Continental observers. The indefatigable and acute Boué (whose name we shall have such repeated occasion to cite as connected with almost every department of our subject,) commenced his career by exploring Scotland, concerning which he has presented the public with a most masterly sketch, ably condensing every important previous observation before spread over diffuse and voluminous works, and adding original materials of at least equal value*.

* A second edition of this work, incorporating the important subsequent observations of Messrs. Murchison and Sedgwick, would be very desirable; but it would require a previous detailed examination of the coal-fields of this part of our island to render it complete. Should this be accomplished, we might hope that Mr. Weaver might be induced to complete and condense into one volume his most important Memoir on Ireland. I trust also shortly, with the assistance of my distinguished friend Prof. Sedgwick, who has so largely contributed to the elucidation of our older rocks, to complete the survey of England in a second volume of the "Outlines." We should then possess a complete geological history of the British Islands, reduced within the manageable compass of four portable volumes.

[page] 384

Those eminent French geologists to whom the important task of preparing a grand geological map of their own country is intrusted by its Government,—Brochant, Elie de Beaumont, and Dufrenoy,—also began their labours by carefully examining all the most important points of English geology, as affording an ascertained basis for comparative observations on the general structure of Europe; the published results of their journey are indeed rather statistical (relating to the position, extraction, and preparation of our mineral ores,) than strictly geological; but the spirit of these comparative examinations will fully appear in the several valuable Memoirs they have published from time to time on the Central and South-eastern districts of France, the countries to which the survey has been first directed*. Still more recently two of the first German geologists, Messrs. Oeynhausen and Decken, have visited our island, and contributed several important Memoirs on the Granitic Veins of Cornwall, on several of the Scotch Islands, &c.

This general intercourse of observers of different nations is not only, from the liberal spirit which has ever on all sides pervaded it, most gratifying in itself, but it is also especially important to the advancement of a science in which all the great general views require the most widely extended comparative observations for their establishment and development.

Before I proceed to submit to your attention an outline (necessarily brief and slight,) of the rapid progress which geology has recently made in developing the structure of foreign countries, it may be convenient here to premise the general geographical order which it is my intention to adopt in adverting to the investigations thus successfully pursued. I shall begin by those constituting or bordering the great European basin, which I shall take in the following order:—France, the Alps, Germany, the Baltic coasts and Scandinavia, ending with Russia. Next I shall proceed to the countries connected with the Mediterranean basin,—the Spanish Peninsula, Italy, Turkey, and the African coasts. The other quarters of the globe will

* I would especially refer to M. Elie de Beaumont's Memoirs "On the Formations in the Vosges intermediate between the Coal and Lias," or what I would call the pœcilitic group, (Annales des Mines, 1827); and " On the Uniformity of the Jurassic Zone environing the Basins of London and Paris," (Annales des Sciences Nat. 1829); also to those of Dufrenoy "On the Upper Beds of the Lias in South-west France," (Annales des Mines, 1827); " On the Central Platform of France," (Ibid, 1828), and " On the Chalk of the South of France," (Bulletin de la Soc. Géol 1830).

[page] 385

follow,—Asia as divided into the northern and central provinces explored by the Russian Government, and those of India by our own nation. I shall conclude with North and South America: but in all these instances I shall reserve what relates to the two great points of tertiary and volcanic geology, as demanding a distinct notice rather in their relations to the general questions of the science than to the geographical distribution of formations.

To begin with France.—The geological map of this country now in progress has been already alluded to, and some of the preparatory essays of those to whom its execution is intrusted cited. The scientific publications of that country also contain many other most important Memoirs, of which I would especially mention those of Boué on the South-west of France, of Roget on the district of Boulogne and on the Ardennes, and of Voltz on the two departments of the Rhine, as presenting the most important contributions to comparative geology. The work of Charpentier on the Pyrenees is excellent as a descriptive essay, but in many points connected with the secondary rocks appears to belong to the older rather than more modern geological school.

English geologists have ably contributed to the elucidation of the comparative structure of the two countries,—especially, as was to be expected, of those districts of France which bordering on the Channel present the direct prolongations of our own formations;—we may particularly refer to the examination of the Boulogne district by Dr. Fitton, and of the Norman coast by Mr. De la Beche. The institution of a Geological Society in France, in 1830, cannot fail to promote the development of our science equally with its English prototype; and the travelling geological class of M. Boubée, whose members are conducted successively over the most interesting districts, may be considered as advantageously introducing a peripatetic school in geology.

In the Netherlands, before the late disturbances, the Government had in like manner proposed to undertake the publication of a geological map. The scientific commission consisted of that venerable and zealous veteran in our science Omalius D'Halloy, so well known for his important Memoirs on these districts and the adjacent portion of France, published at an early period of modern science, when such communications were among its first models*. With him were associated Van Breda and Von Gor-

* Omalius D'Halloy has since published a small outline geological map of France, and more recently a general work on the classification and history of the various formations.

2 B

[page] 386

kum. With respect to this country the most valuable recent Memoirs are those of Oeynhausen and Dechen on the carboniferous and transition districts of the South; with these we may compare Cauchy's Essay on the province of Namur, to which a prize was awarded by the Brussels Academy; and the communications of MM. Dumont and Davreux on the province of Liege, in which it appears that the coal of that district is, like that on the northern edge of our Nort