[1]

S. 4 56 80

36 flowers

4 5 1 4 4 6 1 5 2 [=] 32

Gartner
Bastard.

(36 crosses with vars of Peas 10 of Phaseolus & none quite fully fertile [illeg]

[1v]

Gartner Bastard.

1)

(Gærtner Bastarderzeugung) Variation & other subjects

p. 50 Lychnis diurna when growing in dry places sometimes has very small teeth or projections on corolla, like those in L. flos cuculi (so Dianthus

I wonder whether sport follow Laws, I suspect so from Nectarines & Peaches.

Birds with tufted heads in all classes. Hornless cows in all ages & conditions

p 76 Sports in Pelargonium, Achillæa, Viola, Cheiranthus Cheiri, Auricula. Pinks,— ⸮ are there not cases in all long domesticated flowers? (Indian Azaleas) Sp

p. 78 seeds of Dianthus chinensis & barbatus

(p. 208 & of other genera. vary much.

p 149 gives cases of animals from Blyth closely alike or identical in very different climates; & unlike in similar climates to show little relation to climate. Blyth has remarked that powers of distribution does not go with wide range: this I fancy applies to individual species (& not to the group) — the individual wd be governed by conditions of existence.

p. 157 seeds long retaining vegetative power. X

p. 158 on geological nature of plants before the Löess. —

(2

(Variation &c

p. 163 Gærtner Bastard. Definition of Species, or rather essence essence of Species

p 261 Hybrid verbascum reared in Pots become much more woolly on their leaves.—

p 304 cases of vars. very true or hereditary: white Lilac & white vinca. X

p 305 cases of sports in Lychnis diurna. — Mirabilis Dianthus barbatus, deltoides, mathiola annua, geranium pratense, Rosa centifolia &c &c Œnothera p. 308

p 309 with respect to Anagallis cærulea & arvensis. Watson in Cybele refers to Magazine of Nat. Hist. V. p 493 & VIII p. 634 & Phytologist I. p. 164.—

p. 361. asserts that in pure Dichogamous plants male organs sometimes imperfectly developed, yet can fertilise; but in male flower female rudiment never acquires power of concepcion.— different degrees of rudimentary structure.

p. 362. changes easier in monœcious than Diœcious. (have change less from aboriginal hermaphrodite) discussion to show that the rudiments are realities in the Dicho Dichogamous plant, & the difference in degree, I believe, affects the hybrids which is a new new way of testing essentiality of rudiment. —

p. 378 Gærtner Bastard. In fruitful years domestic animals have more births. — p 519 good instance of Argymone lessening seeds.—

p 445 p. 446 cause of variability in successive generations of hybrids, lies in act of generation, possibly aided by conditions.

(3

(Gærtner's Bastarderzeugung), (Variation)

[in margin:] I have not [illeg] collected all such facts, before p. 459.

There are many odd laws about p 445 p 452. variability in Hybrids

The amount differing Hybrids only according as sexes employed; & certainly according to complexity of cross p 516, 517 most when 4 species blended: — more variability excessive variability more sensibly variability (p 507) when a hybrid is crossed (see last reference on this page with pure father of being a 3d distinct species, but than when the mother is the 3d pure species & the father the hybrid.—

[Right margin:] This will come in Ch. I

variability goes with reduction per matrem or patrem; also in successive generation of simple hybrids. —

Certainly variability very often goes with sterility

p. 459. The shorter the period of absorption of one species into another, the less the variability.—

p 497. on variation in Peas, observed by Gærtner.

p 500 long-cultivated plants vary most

p. 501 variation tends to go back to parent-form no facts given.— (I see van Mons doubts about going back) Newest sub-vars. do, I think there can be no doubt—

p 542 on variability in plant in having annual or biennial from Koel living for 1022 years. (a Book which I cd not see.—

p 556. argues that avitism wd prevent new species being formed, yet admits, that local var. are constant as long as conditions present.

p 572 As Hybrids from first union of diverse male & female element do not vary; & as hybrids reduced per patrem or matrem do vary: as also when hybrid father & pure 3d species is mother the variability is impressed on both sexual element in the hybrid itself.—

(4

p 574. singular remark (by Puvis) that genera with single species do not vary discussed & cases given of certain genera & species (if applied to domestic plants false).— which do not vary; & some that do. Taraxicum said not to vary: (but I thought that this had been an eminently variable genus species.): case of variation in Platanus. Non-variation of Horse chesnut; but surely these are vars. sold?

p 577 case of white Dahlia, which was not hereditary. —

do.— vars. of Hollyocks constant.— X

do. Vars of Pæony described.—

p 578. sports in Cucurbitaceæ sp

p 578 Hyosciamus agrestis agrestis & albas proved same by seed.—

p 598. Remarks it is not known how long vars keep true; but Ægyptian which show how long: at least those who do not think that what a wheat is not an aboriginal form must admit this.—

p 607. enumerates points on which grafted plants do differ owing to sp being grafted. very trifling size, flavor of fruit.

p 620 Curious cases of striped grapes &c

p 628 Famous Laburnum case discussed.— Mem Herbert in Hort. Journal show that the returned branches are fertile

Gærtner Bëitrage p. 220 when many pistils, then number variable.—

― p 364. so about seeds. — Law.

6

variation

(7

(Variation)

Gærtner Bastard. p 459, there is less variability in proportion to the quickness of the period of reduction when hybrid crossed over & over again with pure parent.—

do. p. 440 —G. thinks the going back of varieties to one common var form proof of aboriginal forms, but many dispute nan Mons & Jourdain whether they do.—

Although G. says Hybrids always have something new, he maintains that really new characters do not arise, hence the extreme variability must depend in successive generation, on odd division of characters.— I shd infer same in mongrels.— Hence this variability something different from true variability from new conditions—

[7v]

Facts about Variation in Gærtner, Herbert & Kolreuter

Laws of Variation

Variation

Gærtner

It wd be good to go through large abstract & see what laws can be made out about sterility & variability

(1

Cases in which Gærtner artificially impregnated plants with their own pollen extracted from table, showing result.— (See Paper on Back)

1 Campanula medium (K)

2 Cobæa scandens (a) But as no cross with other species was effected; the pollen &c might have been bad, & therefore this case shd be excluded

2 Cucubalus alpina (=Silene inflata of Steudel) (K)

3 Dianthus barbatus (K)

4 ― superbus (K)

5 Fuchsia coccinea (a) 8 flowers

6 ― macrostemma (h) 3 flowers

Geranium sanguineum (a) But no cross effected with this spec

7 Gladiolus communis (a) do do. do

7 Lobelia cardinalis (K)

8 ― fulgens (K)

9 ― syphilitia (K)

10 ― splendens (ia)

11 Lychnis diurna (K)

12 ― vespertina (K)

13 Mimulus cardinalis (K)

14 Nicotiana pumila (K)

15 Passiflora cærulea O. 10 flower (Perhaps chance)

16 Potentilla atrosanguinea (K)

17 ― nepalensis (K)

18 Silene nutans (K)

19 ― orientalis (K) (g)

20 Tropæolum majus ig.

[1v]

Observe whether he anywhere says he puts plants which he artificially self crossed impregnated in Pots in chambers. — (I have now quite convinced myself that he did. —

(2

In this list of 20 plants artificially impregnated leaving out doubtful cases, so making case against me.)

14/20 produced (K) or normal number of seeds: some of the 6 failures may be attributed to chance or putting pollen (See Back) on wrong place.— Unfortunately I do not think it is anywhere said, whether these plants were (Yes) put in pots & put in House like those which he hybridised.

But as he was most careful observer, I think it probable that he did so & it is certain that he wd do so in the 29 cases of crossed varieties.—

If so it is scarcely possible to explain the few or no cases in which in crossing species, he attained (K) except that fertility is really always or almost always lessened in such crosses.— In the appended list of hybrids with very of first crosses with K Mathiola annua & glabra. And Lychnis diurna & vespertina, are only the sole cases; & the first is considered by R. Brown as varieties, & the latter by Henslow & some other Botanists as like-wise varieties.— If we in this

[2v]

To the 20, 29 cases of vars crossed may be added making 49 of which only 6 gave (K)— so that total number of 49 artificially impregnated with own pollen or that of undoubted var. & of these 49 only 20 gave (K) or normal number of seed. — But as of the 29 cases Phaseolus is included which wholly failed; 4 crosses of this genus may be subtracted leaving 45. (& this includes 12 crosses of Pisum & Leguminosæ are all easily hurt by castration) so that it wd be safest to say that 20/45 yielded (K).

It wd be safest leaving out doubtful cases, wherever no cross of any kind effected, & vars of Leguminosæ, we may say he experimentised on 33 cases of which only 17 gave (K) ie 17/33.

These are first crosses, & not hybrids. —

[3]

Cases of vars. as admitted by Gærtner artificially crossed & result

1 Lychnis diurna ♀ & L. diurna fl. albo (ia) (K) (1)

2 Phaseolus. 4 crosses absolute failures. (a) (b) (o) (o) (4)

3 Pisum sativum 12 crosses between its vars. & mongrels & vars yielded 7 results of (1a) 3 of (1B) & two of (a). (12) 12

4 Potentilla verna flora minore ♀ & majore ♂ (K) (1)

5 Zea mays 5 crosses, yielded four (1B) & one (1r) (5)

6 Verbascum Lychnitis 4 crosses of vars. & their mongrels all (K) (4)

7 Primula elatior communis ♀ & P. cl: calycantha ♂ (g) (h) 8 flowers & reverse or 8 flowers (fa) & (f) (2)

Here are 29 experiments, of which only 6 gave (K) or 13 (if Leguminosæ left out) of which only 3 gave (K)

3

consideration includes hybrids with own pollen, we have Lobelia cardinalis-fulgens & reverse, (with k) but in test (p. 385) even these seem rather less fertile than pure: as is likewise case with hybrid (p 385 text) Lychnis diurna-vespertina. Hybrid Malva mauritiano-sylvestris & Verbascum lychniti-albo-pyramidatum seem (perhaps in text somewhere contradiction cases of real (K) or normal fertility, yet in text he says never reach all these four certainly seem to lead to belief that crosses between species (& of Hybrids derived from them) are always really less fertile in some degree than the normal condition: on other haud Gærtner himself always rank lesser fertility as one of the least important characters of species (in which he is a vehement believer) & I do not see why he shd do so, if the evidence is really as good as he makes it out.— To get the normal standard of fertility must be very difficult)

(4

(& is admitted by him) it clearly wd not do to compare number of seeds in wild & cultivated plant. He admits the act of artificial impregnation tends to lessen fertility. As just said he admits somewhere that it is very difficult to ascertain normal fertility, & really when considering the paper appended, with fertility of mere varieties when crossed together entered, in which out of 29 experiments, only 6 gave (K) it makes one doubtful, yet in crossing true species there is not nearly are so few or no (K) that hardly this shakes me.— It is very evident

From G Koelreuter, Herbert, Sageret &c having succeeded where he Gærtner failed, that perhaps he was not very good gardener, & I suspect had no greenhouse. It is evident, as he admits, that numberless experiments must be tried to know the real fertility of any cross: yet it is strange that he shd have got 17/33 20/45 (K) when crossing vars. or pure species, & only one or two or

(5

perhaps none in crossing species.—

From considering these several facts, I must doubt whether it is possible to ascertain to such a nicety the fertility of crosses of two species compared with their normal condition. If not we must conclude that very many first crosses & Hybrids are really very fertile & so fertile that no trustworthy distinction can be drawn between their fertility & that of pure species; as was opinion of Herbert.— But if we do trust Gærtner, & certainly he was most careful, most candid, most laborious, admirable observer, then it will be very difficult to escape conclusion that some vars. are slightly sterile together.—

Gærtner wd probably say that the vars. of Phaseolus failed from difficulty of operation & that Pisum gave lessened fertility from same cause

(6

I know not whether he cd assign any reason for lessened fertility of the 5 experiments (9 22 flowers having been impregnated) on Zea mags vars.— No he admits more sterile

The case of Primula is very remarkable, (I have shown that even P. elatior calycantha & P. elatior communis produced no result) but as I think it certain that oxslip, cowslip & primrose are all vars., I may add up experiments, & I find during several years he crossed no less than 170 flowers!! & yet he only got one (ir) all the rest (h) (g) & less or no result whatever: as he got one fruit with some seed which but less than normal, it wd appear that he knew how to impregnate & there are several cases of (h) or half fructification, I know not whether to cannot attribute this to lesser fertility of vars. or but probably to some wrong treatment of plants.—

Lecoq says he has produced crosses, but no particulars given.—

Anagallis cærulea (or arvensis) & phœnicea, which I must consider vars. he tried on 19 flowers, which yielded only (a) & (h); probably to be accounted for as in Primula.

(7

There is always some doubt whether the plants experimentised on were rightly named by Gærtner, but all that can be said is that he was a careful man.—

Cucubalus (=Silene of Steudel) he crossed. Cu. alpinus, C. Belen angustifolius, C. Belen latifolius, & littoralis, & a mongrel from two of them which Steudel makes all = Silene inflata, but no doubt they must be vars. otherwise Gærtner wd not have given them different names. During several years he experimentised on 46 flowers. (& failing in two sets) got in all other (ia) (one 1b) ie one step less than normal fertility. Here he tried C. alpinus proporio polline & got (K); so that I must attribute if result can be trusted & if Steudel is right, some effect to vars. having been crossed.

Datura stramonium & tabula & racemosa produced (ia) see text p. 385 yet from Asa Gray & Dr. Bromfield I must consider them vars.—

(8

Digitalis lutea ♀ & media ♀ & reverse both gave only (ia) & yet Lindley thinks var. D. ocholeuca ♀ & media ♂ & lutea ♂ & their reverses all gave, likewise only (ia) & some authors think these only vars.

There is some case in Kolreuter on this genus.

Hyosciamus pallidus ♀ & agrestis (agrestis = albus. p. 699 Gærtner) ♂ gave (ia) but according to Steudel these are both same as H. niger. — as are likewise perhaps H. pictus & pallidus & these gave only (ia).— In all these cases no doubt they are vars. otherwise G. wd not have given them distinct names.—

The Lychnis case very interesting: most authors make L. diurna & vespertina distinct; if you assume this, you can prove that sometimes quite fertile together for L. vespertina ♀ & L. diurna ♂. crossed in 28 flowers gave (K) on other 9 flowers gave (ia) & so reverse

cross. (& so that Hybrid offspring of these crosses, with own pollen sometimes gave always (K)). but in text p. 385 fewer seeds than pure parents. On other

(9

hand if considered varieties it can be shown that they are slightly sterile when crossed together. on nearly as good evidence as can generally be obtained. — Here we have doubtful species & doubtful fertility.— perfect gradation which might be expected on my theory.

In Verbascum author in other work states that the colours of certain vars. of Verbascum influence fertility. See Sageret. —

Case of Sweet Peas

─ Hollyocks.—

I think we must admit either that vars. are sometimes sterile in slight degree together, or we must conclude, that it is quite impossible to deduce any safe conclusion on sterility & fecundity of crosses. —

(a

List of cases from Gærtner's Table, in which result of crossing 2 varieties with each other or second species as admitted by him to be vars, is given. (The cases of 2 vars. together, which in fact is self-impregnation done artificially are marked with red cross.)

(The list of Red crosses have been copied into other Paper.)

(This has given no particular result — & need not be looked over again.)

Aquilegia canadensis-vulgaris (hybrid) ♀

― vulgaris fl. fusca ia ♂

― ― ferrugina i a

― ― nisso i a

Cucubalus littoralis ♀

― belen latifolius i a

....... a

― belen angustifolius a

― i b.

Dianthus Chinensis ♀

― barbatus i (a) (f) (f)

― ― fol. angustus (fa) (f)

Lychnis diurna ♀ propi pollen (K)

― diurna flore albo. (ia) (K)

Malva rotundifolia Mauritiana ♀

― rotundifolia major (a) (f)

— ― minor (a)

Malva sylvestris ♀

― rotundifolia major (f)

― ― minor. (f)

(b

Nicotiana quadrivalvis ♂ male

― rustica ♀ (f)

― humililis ♀ (ir)

Nicotiana Langsdorfii ♂ male

― rustica ♀ (e) (f) (h) (g)

― ― humilis ♀ (f)

Oenothera nocturna ♀

― villosa ♂ (ia)

― nocturna var. striata ♀

― villosa ♂ (ia)

X Phaseolus several vars. when crossed all quite failed

Pisum sativum luteum ♀

― ― macrospermum ♂ (ia) 1

― viride ♂ (ia) 2

― sativum macrospermum ♀

― luteum ♂ (a) 3

― nanum ♂ (ia) 4

― viride (ia) 5

Pisum sativum nanum ♀

― viride ♂ (ib) 6

― viride ♀

― luteum ♂ (a) 7

― macrospermum (ib) 8

X ― nanum (ib) 9

(c

X Piso- vicia ♀ of Wagner. (Really a mongrel)

― luteum ♂ (ia) 10

― macrospermum ♂ (ia) 11

― ― viride ♂ (ia) 12

X Potentilla verna flore minore

― ― ― majore (K)

Primula acaulis ♀

X ― elatior calycantha ir

 ― communis (a) (h) (h) (a) (h) (h)

Primula elatior communis ♀

― acaulis

X But as the admitted vars. quite failed

I think this hardly goes for anything

X Zea 5 crosses four of them ib. one ir.

D

Verbascum austriacum* ♀ yellow ?? Yes p. 180

― Lychnitis fl. luteo (ib)

― ― fl albo (ia)

― blattaria fl. luteo (ia)

― ― albo (h)

Verbascum blattaria fl. albo* ♀

― lychnitis fl. albo (ib) (h)

― luteo (ir) (a)

V. blattaria fl luteo ♀

― lychnitis fl. albo. (a)(h)

V― ― blatteroides* ♀ — Colour?

blattaria fl luteo (g)

― fl. albo (g)

Lychnitis fl. luteo (g)

― albo (g)

V. cuspidatum (yellow) ♀

― blattaria fl. luteo (a)

 ― fl albo (a)

 Lychnitis fl luteo (a)

 ― albo (ir)

V. elongato-macranthum (hybrid) Probably yellow

― blattaria fl. luteo (f)

 ― fl albo (b)

(E

Verbascum Lychnitis* fl. albo ♀

X ― ― fl luteo (K)

― blattaria fl. albo (ia)

― fl luteo (ib)

V. Lychnitis* fl luteo ♀

X ― fl. albo —(K)

― blattaria fl. albo ib

(only 1 var here tried)

V.— Lychnitis albo-luteum* ♀

X ― Lychnitis fl. albo (K)

X ― fl. luteo (K)

V. macranthum ♀ yellow

― Lychnitis fl. albo (a)

 ― luteo (a)

V. nigrum* ♀ (yellow)

― Lychnitis-albo-luteum (i (a))

 ― fl luteo (ir)

 ― fl albo (a) i(a)

 blattaria fl. luteo 0 —— accidentally omitted?

 ― fl albo f

V. phlomoides (nemorosum)* ♀ yellow

― Lychnitis fl albo (ig)

 ― luteo (ib)

― ― blattaria fl. lute 0 — I think latter accidentally omitted

 ― fl. albo (ib)

V. phoeniceum* ♀ purple

 —Lychnitis fl. luteo (ia)

 ― fl albo (ir)

f

Verbascum pyramidatum* ♀ yellow

― Lychnitis fl. luteo (f)

— albo (ig)

V. speciosum

Lychnitis fl. luteo (a)

― albo (a))

V. thapsiform* yellow ♀

― blattaria fl. luteo (ir)

― albo (h)

Lychnitis fl luteo (ia)

― albo (ig)

― albo-luteum (ib)

V. Thapsus* yellow ♀

― Lychnitis fl. luteo (ia)

― ― albo (ib)

blattaria f. luteo (ig)

― f albo — (h)

If we trust the *, signifying that the males are arranged according to sexual affinity— yellow same coloured vars, with one exception do come nearest to species; but colour of 2 species not known.—

(13

p 445. In successive generations of simple hybrids more variability.

― Hybrids of this type Dianthus barbato-barbato-carthusianorum O most varieties of all.

X In viz Dianthus barbato-carthusia- more than barbato less variation.

p 446 Male more power in causing variability than female.

X cause of variability lie in act of generation, aided, perhaps, by conditions.

X X p 447. Nicotiana panicuto-rustica 3. not to be distinguished from pure rustica yet N. pan-rust (4) less fertile than pure parents —p. 449 (very curious) (Show how little

(good) depends on characters visible to eyes. C.D)

O p. 448 different hybrids species are converted or absorbed into the pure forms, at very different rates: but this varies even in same species.

p. 449. nicot. rustico-paniculata, (4) x when almost converted into pure paniculata, yet quite sterile in both sexes! This sterility commonest when a O form is with difficulty converted. Male side organs most affected.

p 450 Generally with successive paternal crosses fertility improves, & by successive self-impregnations, done artificially, (The more I think, the more I infer result of crossing) these hybrids improve in fertility.

p. 476 Precaution necessary, & approach of themselves to the paternal types, as in his rustico-paniculata (3) with own pollen.—

― asino-caballus horse with ass.

(Explain terms Equus caballo-asinus or ass with mare mother—father

(14

O p 452 maternal hybrids more fertile than corresponding paternal hybrids. Thus Dianthus barbatosuperbo-barbatus see last page this type less variable is more fertile yet some sterile than D. barbatosuperbo-superbus. see last page this type less variable

X These vary more (this is opposed to law of sterility & variability being connected). In further generalisations, when self impregnated, either naturally or artificially, become more fertile approach mother more, & less variable p. 454. The more O fruitful stick to the hybrid types, the less fruitful go back (of themselves) to mother type.

(Can there be any law that departure from parent type lessens fertility.— (opposed by varieties) This seems to hold with hybrids, ie whenever the offspring departs from middle resemblance it fertility is lessened.) nor is vars.n

O p. 455 It is certain that maternal hybrids of themselves go back to mother: G. thinks a proof of aboriginal (p 475 range thus) existence of species; but then it is general belief that vars do also. Bell stated so to me. No doubt that vars can be converted or absorbed.

If with truth the sterility is argued that differences between species & vars. on other haud all other points on which they resemble shd be borne in mind to show that not essentially different.—) (Reduction = absorption? which term? p. 476)

p. 458. The rate within narrow limits of absorption of one from into another varies according to the species, & the individuals employed. The shorter the period of absorption the the less variable. In reverse crosses hybrids (p. 359) though alike, the period of conversion (also p 465 example given) universally somewhat different. —{Fertility perfect always latest acquired.— X

(15

X 460 Law not only male organs first affected in hybridisation, but last restored (Mem. how little we understand of generation.)

p 461 extreme variability in fertility during the successive absorption, never acquired suddenly; depends much on individual. O The absorption varies on average in different species from 3- to 5-6 gen., which is the extreme required. — Long cultivation seems to affect the variability in the length of period of the absorption.—

p. 467. Fertility of hybrids not fixedly related to rate of absorption

p 468. Generally with nearer separate less elective affinity of pure parents, rate of #absorption quicker slower, (exception to p. 469)

X p 470. a species with power of producing a "decided type has stronger power of absorption. There is, also, relation in latter to systematic affinity (& we know some to in latter to elective sexual affinity) But the difference in rate of absorption in reciprocal crosses, shows no fixed relation to systematic affinity.

p 473. The returns approaches to parent-form through self-impregnation very slow.

p. 474. The goings back seldom (so sometimes is) observed in wild plants, but more common in cultivated plants, (therefore this is a form of variability)

We have also seen, that the variability in rate of absorption depends on culture. (This wd come in under the head of Variation.)

p 475 Even in reduced hybrids traces of parent.

X p 475 Thinks the absorption unequivocal proof of limit of species fixed. How curious that I shd think contrary. Perhaps opposed to Herbert's views.—

(16

p. 500 Remarks that those plants that do vary, have their inner nature or power of union much less affected. (So Herbert has remarked)

501. Long cultivated plants which vary much, likewise easily depart from normal bastard form.

p 504. Hybrid. Dianthus (a) barbatus ♀ + X barbato-superbus ♂ ✔

O all alike—

{(b) barbatus ♀ + superbo-barbatus ♂ ✔

{(c) superbo-barbatus ♀ + superbus ♂ ✔

It is only quantity of either blood that signifies. But when either side has a predominating influence, then there is difference. of ((a) (b) (c) types are all very variable & sterile)

507. Hybrids. Dianthus barbatus ♀

― arenaria-deltoides ♂ excessively variable & absolutely sterile

Compounded zusammengesetzten

X 507. Reverse of last, ie simple Hybrid the mother, always take after pure Father, (influenced by prepotent type) O fertility various but generally little quite sterile

509. When the 3 parents little allied, then most resemble pure parent & excessively sterile. The resemblance to Father G. remarks against ordinary law, accounts for it by potency of pure pollen. When mother pure species no tendency to her side.— These zusammen. are not so variable.

p. 514. Thinks hybrid pollen has preponderating influence over female in development force. If so how come it in reciprocal crosses that hybrids alike.

(17

p 283 particulars about Lychnis vespertina & diurna about day-sleep intermediate in the hybrids between the 2 parents).

p 515. Hybrids such Lychnis diurno-vespertina ♀ vespertino-diurna ♂.

XX good to try with Pigeons. O

Take after whichever type is preponderant, & are absolutely sterile!

p 516. 4 species united, produce excessively variable; no two O alike, & fertility lost.—    It is clear the more complicated the cross, the more variability & more sterility ensues.

p 517. Hybrids can be told from pure species by their variability in their succession & by lessened fertility in first cross, & in hybrids,— then greater luxuriance. But variability in offspring of hybrids is a haupt-charakter.

p 542. Hybrids longer lived attributed partly to sterility

p 528. Is opposed to luxuriance of hybrids being due to sterility, because most fruitful hybrids are most luxuriant. Innumerable flowers. (is this not like double flowers?)

(so Knight with Peas.)

p 533 absolutely sterile hybrids have flowers preserved longer when stigma dusted with pollen of parent: (so O none absolutely sterile).

p 539. does not think Köl. right about his fertile plants being only varieties.

p 540. Hybrids never give so many seeds as pure species. But as pure species are often sterile sterility cannot be taken as character of Hybrids.—

[17v]

cannot be called monsters Loudon does─ the regularity with which produced show not abnormal producing normal.

(18

p 549. Tropæolum majus & minus, (perhaps only vars. ig)

that some few flowers went back during summer to mother-form or minus.— Full particulars given.—

p 550 case of change in Hybrid ((in leaves) between Cereus speciosissima & phyllanthus.

p. 553 cases, of very few, of hybrids which have kept very constant one case for 10 generations, but with regularly lessened fertility specially in male organs. I wish I knew whether many plants, this seems probable. The one case, viz Dianthus arenaria-deltoides self-sown for 6-8 generation in garden.—

X p 554. Fertility more variable than any character: rarely become more fertile in 2d. generation, but generally even in fertile hybrids, still some much more sterile

O X p 551 Verbascum species most difficult to distinguish, [illeg] yet hybrids very sterile. together.

p. 565 Hybrids more inclined to double flowers than pure species

569 Male organs being easiest doubled, agrees with male organs first affected in Hybrids

p 572. It shows I think, how essential the disturbance in Hybrids is in the essence of generation that stamens nor pistil not modified, except O from being empty, & corolla (modified stamens) not in the least affected.—

X p 575 variability greater in mongrels than hybrids, & very close species do vary a little Even when the vars. themselves are constant. But if they do not vary, how are they known to be varieties & not true species.—

x x p 577 allows vars. of maize unite with great difficulty but offspring very variable & quite fertile. In Garden (p. 87 kept pure during a course of years.—

p. 577. Lychnis dioica & vespertina vary are in varying more like mongrels than in fertility & so too Petunia nyates & phœnicea.— but yet p. 582. these 2, when crossed with 3d give different offspring.

(19

X p. 578. O Mongrels generally intermediate, & this class is commonest (p. 282) in offspring of close species (important)

p do Thinks the greater fertility of mongrels, analogous has grander to greater luxuriance of Hybrids, (I do not see this). The mongrels are earlier. (are Hybrids? they are hardier. Yes p. 580 Hybrids do flower earlier)

p. 578 List of Kolreuter vars which he considers species even Alcea ficifolia & rotundifolia.

(Mem. all the argument against the luxuriance being due to sterility, if this be given up, & the cross be considered like change of conditions, then there will be analogy in two case. Here might be taken opportunity to discuss this curious point of luxuriance. Anyhow it show how healthy hybrids are.)

p. 580 The varieties of Hybrids in their successive generations & the mongrels can all be classed, I presume in same way.— says (p 581) that mongrels only differ from Hybrids in varying in first cross, ie when XX pure species are crossed & in their lessened fertility, in other respects resemble each other.

p 581. Thinks it very important that 2 close species when crossed with a third gave very different hybrids; but the O vars of tobacco when crossed with a third gave O very similar offspring. (Somewhere says Hybrid do when cultivated.)

p. 582. Mongrels more tendency to revert to parent-form than Hybrids — Yet somewhere this very reversion he considers proof of aboriginal existence of Species.—

(20

p 585. only a small proportion of Hybrids are exactly intermediate.

p 591. case of wild Hybrids of Verbascum, yet offspring excessively sterile

X p 679. Hybrid with seeds going back to 2 parent forms so I think, in Maize.

[20v]

[in margin:] In beginning of Book discussion on this & says he found no difference between wild & cult. hybrids —

(With respect to hypothesis that successive generations breaks down infertility, Gærtner shows that a plant does not become during its life time more or less fertile, nor do successive generations of hybrids, but it is quite another question whether pure species may not alter.

Primula is almost only case I can think of experimentising on.— Viola tricolor? Wild Columbine—) one might try whether wild cabbage wd come up true in Garden.—

N.B in London almost all the Dianthus are called "ornamental" & I do not see how to ascertain whether plants experimentised on had been cultivated. One might try whether seedlings of wild —

Caryophyllus were as easily crossed as cultivated vars.;—

this wd be my best case. —

All that can be said is that Herbert tried, on cultivated plants & Gærtner took, when he could, wild plants.—

Mingle pollens of common & red cabbage & place on wild stigma, & so with other varieties: this will show how far affected each other, & will anyhow indirectly show advantage of crossing

Do this with Dianthus. —

(1

I have only extracted on Hybridity not on other subjects.

Notes on some of the most important points in Gærtners Bastardz.

(on rereading Book June 1855)

p. 12, & 13 & p. 30 It is the first cross, which he seems to think is so invariably more sterile than natural impregnation; (but here comes in the ill effects of artificial impregnation) He admits states that some hybrids produce more seed, than are produced by the union of the two X pure parents—, as in Datura & Dianthus.

X p 64 important cases where crosses of pure species & Hybrids succeeded, when self impregnation wd not — can only be accounted for, as it appears to me, by advantage in crosses.—

p. 92. on varieties of Verbascum of like colour being crossed more fruitful.

p. 96 admit great difficulty of making scale of fertility in first crosses. (unimportant passage)

X p 97 p 97. Even species of same genus, absolutely sterile together.

p 102 In implication Mathiola arenaria & glabra, though marked (K) in Table ought to produce less seed than normal Fertility (i) found only amongst close species. —(unimportant)

p 116 The most nat. natural Families opposed to Hybridisation.

(2

p. 121. Capability for Hybridisation more special than any Family or generic character

p. 126. In same Family some genera are easily hybridised; other closely allied

X genera very difficult

p 135 universal law pollen of parents fructifies hybrids more than own.

p. 138. Even Herbert does not pretend that all species of genera can cross though when they do, they must belong to same genus.— case of "reverse crosses" & cases where a cross succeeds after years of attempts opposed to this law. (Genus ideal also)

p 139. There is he says a "sexual" & "systematic" relationship.

X p 140. Species of Silene though so close, neither he nor Köl. has succeeded in uniting any two: though he has one with Lychnis. It is false to infer because some species in genus will unite that others will—

X p 142. important statement from Sageret about non crossing vars. of Cucurbita contradicted by Girou.—

p 143. The sexual affinity can be predicted only of species, & often only of x individuals of a species.— x different time of flowering does not prevent crossing — best with heat.— annual & biennial.— evergreen with deciduous.— x (Can plants from tropics & cold countries cross?) x Fuchsia? see Lecoq & Verlot

Mem. crossing dependent on external conditions; can be effected in summer & not in autumn.

p 147. nicotiana Langsdorfii will impregnate several other species " [line illeg] but will not be impregnated by any kind. nic. acuminata will neither impregnate or be impregnated by any of the many nearly related kinds.—

X p. 149 some Hybrids do not vary even to 7' or 8' generations; are there any fixed mongrels. Is there not a Hybrid Swede Turnip? & a hybrid Trifolium.

(3

p 163. Close species united with sp. give very different hybrids: which he remarks as strange, if these close species are descended from common ancestor. But I do not see this.— It is odd whether applicable to var. or species. "The existence of species consists in fixed, sexual relations to other species."—

p. 164. Individuals somewhere of some species fail to be fertilised, dependent

O apparently on internal causes.

p. 166 x Girou de Buz. seems to speak of difficulty of vars. crossing. Vide p. 173 perhaps intermediates. (in ann. des Sc. nat. vol 30. p. 405.) des Sciences

p 167. O Pear & apple will not cross (p. 140. Tormentilla erecta & Potentilla verna will not cross.—

p. 168. Important sentence on malva sylvestris & mauritiana.— Mathiola arvensis & glabra, Delphinium ajacis & consolida — Datura stramon & tabula —Oenothera nocturna & villosa, being "biennale vollstandig" fertile together & offspring not variable, & so these like true species, though thought by some to be local vars.

p 168. Evidence for & against law, that species with O "similar habitus" cross easily.— few hybrids in great nat. families opposed to this.

But with respect to nat. fam. p 171. Gramineæ, Umbell. Compositæ very little tried — Leguminosæ practically difficult— Cruciferæ have failed — Labiatæ little tried (p. 172.─

p 173 on vars. of cucurbitaceæ crossing & not crossing. Sageret must be studied.

X p 174. Table of very close species which will not unite: I see here he includes Anagallis — & Table of very unlike species (Cereus) which will O X unite (see p. 179 better cases where same genus closest species do not always unite best. (see p. 178 also)

p. 176 O non-reciprocal fertility best evidence that union does not depend on affinity. N.B. This is against constitutional affinity, against

[3v]

This is var.

(4

any affinity, except sexual affinity, — p 177 Even where reciprocal crosses can take place, and where great resemblance of Habit, as in Malva sylvestris & mauritiana — Mathiola annua & glabra, Oenothera nocturna & villosa, Datura stramonium & tabula Petunia nyat. & phœnicea —yet the reciprocity is not quite equal. (this important in regard to Mathiola & Datura) (But ⸮ is there any evidence)

p 178 He quotes his Primula experiments, without doubt, & has elsewhere given this family, as one offering no difficulty in mechanical part of crossing.— He speaks of difference in x This is rare x Primula is reciprocal crosses; I can make out only one case. Table must be imperfect.

p 181. speaks of his experiments with Anagallis as "manyfold" with no result—

p 182. neither size of fruit, ovules, pollen, shape of pistil has O X no direct influence in preventing crosses. Herbert has given cases about seeds. —

p. 186. capacity of Hybridising a vital action, & allows that sexual affinity or "elective affinity" a mere word.

p 188 chemical constitution of plants sometimes differs greatly O X in same genus as in Stychnos: compares this to diversity in crossing.—

O p 189 gradation in effects when pollen of different species placed on stigma shows steps in shown by elective affinity of plants.

p. 194. no clear relation between facility of first cross & fertility of O X Hybrid thus raised (nothing shows more plainly what an obscure subject it is.)

(5

X p. 195. Examples of numbers of seeds in pure & Hybrid

p 197. p. 195 on sexual reciprocity. In slight degree different in Cucubalus belen, & Cu maritimus (this must be misprint for C. littoralis?) Malva, sylvans & maritimus Mathiola — Datura stramonium & tabula. There is a Silene maritima in Babington Lychnis diurna & X vespertina p. 218: These cases very important especially in Mathiola. — (If best Botanist to be trusted they are vars. & they are marked (K) yet p. 168 I presume not absolutely fertile.; yet M. glabra ♀ produces more seeds with M. annua ♂, than, M. annua ♀ with glabra ♂.—)

O Perfect chain of gradation in the greater inequality of the reciprocity.—

X p. 199. striking cases. Pollen of one species in genus sometimes has peculiar power of impregnating many species.

O X No result of the difference in reciprocity on the nature of Hybrid raised.—

p. 202 Hybrid offspring of (A+B) Hybrid + C, always closely like C, but sterile.

O Has there not been triple Hybrid at Zoolog. garden. Horse, ass, mule.

à [illeg] all this again

204 - 214. several pages on counting of seeds. I do not see that plant artificially self impregnated was placed in chamber (on contrary always wild plants taken as standard) because difference of such impregnation & natural impregnation is less than between nearest species.— I cannot quite satisfy myself. It is clear See Back

p. 215, 216 X Table, showing sexual affinity, in vars of Verbascum.—

p. 218. X do showing difference of reciprocal fructification of Lychnis diurna & vespertina

[5v]

I remember he somewhere says the races of Nicotiana very stable.

It wd be so monstrously unfair (& he is evidently cautious) to compare species artificially impregnated with own pollen in open air, with hybrids & mongrels done in chamber, (& as he often had other object in view) I think we may conclude that these experiments were done in chamber. In the Beiträge p 365 & p 332 (?) where he compares natural & repeated artificial impregnation he discusses whether effect of chamber can be cause: & answers that some are as fruitful, as in open air —

This shows that they were in chamber

(6

p 221. Reciprocity takes place generally when Hybrid is intermediate O X in character. but when hybrid takes after mother or father then reciprocity will not take place. The Hybrids from reciprocal

p. 285 contradicted. crosses alike.

p. 223 remarks that this differs from animals for sex has no influence on Hybrid: cases of difference in same litters or nests of eggs

O X p. 225 Exceptions to hybrids from reciprocal crosses being alike.

O X p 231 In essence there must be some difference in reciprocal crosses for they are are converted into other species at different rates.

(When one thinks of the wonderful difference in results of reciprocal crosses, one is led to conclude that no more important as character of species, than colour of flower.— No, because we know that flowers do vary in colour in varieties.)

X p. 232. Hybrids stable in first generation; but experimented as much as possible on wild plants

p 234. Hybrids from same parents are always alike. (Thinks this evidence of permanence of species, I do not see more than ordinary generation). Some exceptions to similarity, but then these exceptions p. 238 seem to follow law. The normal type keeps constant

O X in successive hybrid generations only in most fertile Hybrids. (so here is a gradation in character).

p. 243 The abnormal types often take after one parent. Similar (p 243) abnormal types occur in second generation of simple hybrids; in paternal or maternal hybrids; & in more complex crosses. — & in mongrels

O X p. 244. Abnormal types generally quite sterile (Curious)

p. 247. Slight variations in Hybrids, which seem accidental & are variable, O occur chiefly in species which are very close, & are held partially X as mere varieties, such as the Mathiola, Malva, sylv. & Maur. Lychnis diurna & vespertina, Primula elatior & officinalis. p. 168 directly contrary statement!! look again (Argument that these are varieties.)

p 249. These variations occur only in cultivated Plants.

(7

p. 251. not seldom Hybrid more resembles one parent than other, but this resemblance is constant; & is difficult to be decided on.

p. 256. striking cases where one parent almost lost, seldom hybrid resembles parents equally.

p 273. Datura stramonium & tabula give different hybrids with D. X quercifolia. Lychnis diurna & vespertina, with Cucubalus viscosus.

X on other hand. really vars. of nicotiana with a distinct species, gave very similar Hybrids. This shows that he decidedly thinks the above Lychnis & Datura really species.

p 275 mongrels in variability follow different law to Hybrids

do X Does not think the abnormal types in Hybrids can be due to external circumstances. (This is like what must have concluded in vars. Müller: law of generation)

p 282. In close species, as Malva sylv. & Maur. offspring O X half-way most commonly.— (1st Class)

p 282. Hybrids most generally hybrid (p. 289) resembles, (but not identical with) one species in (2d Class) one part, & the other species in another part.

p. 285. In another (3rd) class. the Hybrid more resembles one parent than other; but then the three classes of resemblances blend into each other.

Decided types, or those which take after either parent, generally sterile. Even when hybrid does strongly take after one parent, O X yet the two parents can be crossed reciprocally.— This contradiction to p. 221

p 289. some species in genus which have preponderating fructifying power X & some have preponderating power in structure of resultant O hybrid but these two powers do not go together. The latter are generally very distinct species. (This is what happens in zusammengesetze crosses)

p. 292. X no relation between facility with which A will cross. or be crossed by B, in different cases, with power of giving structure to offspring.

[7v]

Reference to vars of Anagallis}—

(8

p 296 even day-sleep of Lychnis blended.

p 307. White & yellow vars. of Verbascum crossed gave either colour pure

p 309 again says he has often attempted to cross the two Anagallis on this page I have given references on subject of Anagallis from Watson's Cybele.

p 329 Hybrids specially affected in male organs. exceptions p 339 p. 350 Emphatically repeated.— do p. 355 to p. 361

p 334. Even in most fruitful hybrids, not proper number of seeds: (as in first cross.)—

p. 353. With respect to lessened fertility in all Hybrids, says "mongrels "etwa" (perhaps, now & then, or nearly "alone excepted". This X etwa very remarkable.

p. 357. Wonderful cases of plants of which pollen good, yet wd not act on self. [illeg] X as Verbascum in own country; can only be introduced to illustrate advantage of crossing. p. 359 on same subject.

p. 360. There is wonderful parallelism between the 3 forms of sterility O X in Hybrids, & what occasionally occurs in pure species (specially exotics.—

p 363. In doubling male flowers always first affected.

p 364. accounts for Hybrids breeding easier with parent than with self by pollen being weaker & female organ weaker

p. 359 discussion by me whether the crosses of hybrids with pure parents O is owing to weakness of organs in Hybrids, or to advantage of crossing

X p 366 sterility of Hybrids, even from same capsule, very variable: this shows not dependent on external conditions so abnormal types — not on any law of species as real existences — but on something quite unknown in generative system. p. 367 insists on this O variability in fertility as very remarkable —p. do. some hybrids X will produce seed only at particular times of year.

(9

p. 369 Hybrid plants which produce extraordinary numbers of flowers are sterile. — same holds good with some pure species.

p 369. Female sterility of pure Dianthus Japonicus hereditary in crosses: (this is opposed to any relation to conditions.)

p. 373. Hybrids crossed with either parent, & assuming ancestral O X form, fertility (also very variable. (It is very odd, here & fro case of "suzammengesetze hybrid", that assumption of parent or pure form does not give (also p. 405) fertility p. 365. Hybrids after 8-10 generations lose have their generative power weakened; but then I think (see further on on this head) they ought to have been crossed with other individuals.—

N.B. G. always speaks of plants in "Ximmer", I do not think he had green-house or Hothouse; never mentioned.

p. 385 X number of seeds of hybrid Datura stram. & tab. & of hybrid Lychnis diurna & vespertina. (less fertile than pure species)

X p 387 case of hybrids from reciprocal crosses with different fertility (p. 407) also p. 388 says in table p 391 same Hybrid one year very fertile, next very sterile.

O X Hybrids differ in fertility out of same capsule (this shows p. 397 not external conditions, though external conditions have great effect on). p 393 great difference in number seeds produced by first flower O X. of Hybrids & later flowers; not this difference in pure species.

p 395 no art or culture improve sterile organs of Hybrids.

p 396. Fertility of Hybrids does not go through genera; & greatly depends not even on species, but on the individual.

p 405. most marked difference between fertility of first cross O X & fertility of Hybrid, resulting from this cross.—

p 406. When reverse crosses easily take place, then hybrids commonly O X fertile. Also when a cross between pure species is easy then generally can be reversed.—

[9v]

Here shows that natural fecundation is better than artificial because pollen of nature gets put on several times

(10

p. 407 Difference in fertility of Lychnis diurna-vespertina & reverse.

p 406. ✔with exceptions fertility of hybrids, depends on similarity in habitus in Hybrids.—or systematic affinity p. 408.

But (p. 409) so strong exceptions (for many close species will not unite, examples p. 410. &c) that he decries the rule. — Reurge the difference in fertility in individuals.

p 412 ✔ Even in quite sterile hybrids, yet pollen of either parent makes flowers endure longer.

p 414 List of all the most fertile species, yet of his own & Kolreuter experiment, yet says expressly not so fertile as pure species.

p 418 Ask Authors Most important passage that some Hybrids increase in fertility if artificially impregnated with own pollen; now this require castration (which is in itself injurious) & as castrated, pollen of another flower is required.

When a hybrid (as I suppose must be case) is left to itself, it would have either to stand by itself, or with other similar hybrids, & generally no doubt wd be crossed with own pollen, or perhaps wd have to be protected in most cases from pollen of other species or of parents. I shd expect in instances given, & from his remarks elsewhere on immense trouble of rearing hybrids for successive generations that such hybrids were isolated, & in these cases wd get only pollen of its own individual flower as insects wd be kept away, I can thus only account for increase in fertility, without indeed quantity of pollen & being put on successively may have something to do with this. —

(11

p 420 In 2d generation of simple Hybrids fertility variable & often less never O X more than in 1st or original generation.— Even takes place in 2d. generation, & this cannot be due to want of crossing.

p 421. when artificially impregnated for several generations, fertility sometimes increased, but Hybrid then goes back to parent form.

(p. 439 contradicted, as far as 2d. generation is concerned. p. 445) (How very odd that this going back is here accomp. x by increased fertility, fertility also increased by crossing with either parent, whereas in aboriginal hybrids, generally more sterile.)

p 421. some hybrids keep constant, but with x increasing sterility; most hybrids in 2d. generation vary much, some going back to father, more to mother

p. 423 offspring from a cross, varied more than from reverse cross, x O so here is another difference in reverse crosses.

p 424. The original abnormal hybrids, when fertile generally produce normal hybri offspring.

p 425 cases showing how many more seed hybrids yield O with parent seed pollen, than with own pollen.

p 426 when parent & own pollen mixed & applied to stigma, X O the latter no effect whatever. — Pollen of 3d & new X species (p 427) sometimes more effect than own pollen cases given.

p. 428. In Hybrids there is difference in seeds produced whether pollen of either one or other parent is used, & I fancy this has relation to the species which has most power of converting the hybrid int the pure parent first, will also produce most seed. —

[11v]

Perhaps variability goes with affected power of propagation whether increased or lessened.─

(12

p 430 By fathers pollen more seed generally raised than by Hybrids O own pollen, though generally fewer than in first cross X between pure parents. Like 2d gen. of simple hybrids these vary much, & in fertility, even out of same capsules

The more fruitful hybrids with paternal pollen vary less & go back more to father, but with exceptions. The reverse remark

X (p 431) made viz more sterile, more variable

(I am strongly inclined to think some degree of sterility & variability go together.) (only plants under very favourable conditions, which wd not be killed by little sterility can produce new forms.)— (Hardly because cabbage & wheat variable, yet very fertile). In some paternal hybrids, the O type which approaches most to father is sterile; the reverse case also happens

p. 435. general rule more type approaches father more suffer in fertility. But thinks the reverse probably really case: in fact no rule

Female organs first recover fertility p. 450 seems contradiction (which accords with males first suffering.) In these paternal hybrids (p. 436) much variability in fertility: when either artificially or naturally self-impregnated improve in fertility & of themselves tend to go to Father type

p 437. Maintains that simple hybrids in 2d generation certainly sometimes go back to mother or father form: this does not happen in O the (also p. 444) more fruitful hybrids.—

p. 439 In those going back in simple hybrid in 2d generation fertility not increased often lessened.

p 441. more rarely go back to father than to mother. Going back more common in some genera than in others.

p. 445. These going backs resemble the abnormal types in first cross of pure parents, & those abnormal are likewise very sterile.