[page] 81

course with all the flowers of any species whatever. As it is difficult to prove without troublesome experiments the falsity of the belief of regular fertilization in the bud, I here notice this subject. An estimable and laborious observer*, resting his belief on the usual kind of evidence, states that in L. Austriacum (which is dimorphic and is considered by Planchon¹ as a variety of L. perenne) the anthers open the evening before the expansion of the flowers, and that the long-styled stigmas are then almost always fertilized. He asks whether this precocious fertilization in the several species of Linum and in other plants is not one cause of the short duration of their flowers. Now we know positively that, so far from Linum perenne being fertilized by its own pollen in the bud, its own pollen is as powerless on the stigma as so much inorganic dust.

Linum flavum.—To recur to our more immediate subject, in the long-styled form of this species the pistil is nearly twice as long as in the short-styled form; and the stigmas are longer with the papillæ coarser. In the short-styled form the stigmas diverge and pass out between the filaments. The stamens in the two forms differ in height, and, what is singular, the anthers of the longer stamens are shorter; so that in the short-styled form both stigmas and anthers are shorter than in the other form. The pollen of the two forms does not differ. I have not been able to try any experiments on this species; but a careful observer, Mr. W. C. Crocker, intends proving their reciprocal fertility next summer. As this plant is propagated by cuttings, I have generally found that all the plants in the same garden belong to the same form. On inquiry I have never heard of its seeding in this country; but to anyone wishing to raise seedlings, in all probability the path is now open, namely, by carrying pollen from one form to the other.

I have now shown that three species of Linum are dimorphic, besides several races of L. perenne, esteemed by some botanists to be distinct species, such as L. montanum, L. Sibiricum, and L. Austriacum. According to Vaucher†, L. Gallicum, L. maritimum, and L. strictum are in the same manner dimorphic, as likewise is, according to Planchon‡, L. salsoloides. This latter botanist is the only one who seems to have been struck with the importance of the subject; and he acutely asks whether this dimorphism has not

[page] 82

seven dimorphic species of Linum; but as this structure has been overlooked in such common garden-flowers as L. grandiflorum and L. flavum, it is probably of frequent occurrence.

All the species, however, are certainly not thus characterized. I have examined many specimens of L. catharticum, and found in all that the stamens and stigmas were of nearly equal height and the same in all the plants. So, again, I looked, near Torquay, at many flowers of the wild L. usitatissimum or angustifolium (I know not which), and there was no trace of dimorphism. Again, I raised 111 plants from seed sent me from Kew, incorrectly named L. Austriacum; the plants were tall and straight, having a rather different aspect from the wild species seen at Torquay, with extremely fugacious blue flowers: in all these plants the stigmas stood on a level with the anthers or projected a very little above them. I protected the flowers from insects; but every one of the 111 plants produced plenty of seed. I mention this fact because it had occurred to me that possibly a species might be dimorphic in function, though not in structure.

Lastly, Linum Lewisii, which is ranked by Planchon as a variety of L. perenne, but which, now that we know the meaning of reciprocal dimorphism, surely deserves specific honours, must not be passed over. According to Planchon*, the same plant bears some flowers with anthers and stigmas of the same height, and others with styles either longer or shorter than the stamens; so that the same individual plant is trimorphic. This, as far as I know, is a unique case. From analogy we may pretty safely predict the function of the three kinds of flowers: those with stigmas and anthers of the same height will be self-fertile; those with these organs of unequal height will require reciprocal fertilization. A plant of L. grandiflorum or of the other dimorphic species, growing by itself, could no more perpetuate its race than could one sex of a diœcious plant, nor could any number of plants without the aid of insects. A single plant of Linum Lewisii, on the other hand, in all probability could propagate itself, even if no insects were present, as probably sometimes occurs in its Arctic home. If insects visited the plant, the flowers which were dimorphic would be fertile one with another or with those on any neighbouring plant. Thus the plant would receive the advantage of a cross.

[page 449]

Plants propagated by seed in a garden during twenty-five years have kept constant, excepting that in some cases the flowers varied a little in tint and size‡. Nevertheless Mr. Hewett C. Watson and Dr. Bromfield state§ that "exceptional instances to all the characters, taken singly, by which this plant is distinguished from P. vulgaris and veris" may be occasionally detected; it remains to be discovered whether these intermediate forms are not hybrids between P. elatior and veris, which often grow together. With respect to differences in function, both the long- and short-styled forms of P. elatior are more sterile when fertilized by their own pollen than the corresponding forms of the Cowslip and Primrose when similarly fertilized.

Mr. H. Doubleday, who I believe first called attention to the existence of the Bardfield Oxlip in England, kindly sent me several living plants, which I subjected to trial for the sake of ascertaining whether they differed in their reproductive power from the common Oxlip. I did not think it worth the time and labour to

[page 450]

[not transcribed, crossed by Darwin]

[page 449]

Plants propagated by seed in a garden during twenty-five years have kept constant, excepting that in some cases the flowers varied a little in tint and size‡. Nevertheless Mr. Hewett C. Watson and Dr. Bromfield state§ that "exceptional instances to all the characters, taken singly, by which this plant is distinguished from P. vulgaris and veris" may be occasionally detected; it remains to be discovered whether these intermediate forms are not hybrids between P. elatior and veris, which often grow together. With respect to differences in function, both the long- and short-styled forms of P. elatior are more sterile when fertilized by their own pollen than the corresponding forms of the Cowslip and Primrose when similarly fertilized.

Mr. H. Doubleday, who I believe first called attention to the existence of the Bardfield Oxlip in England, kindly sent me several living plants, which I subjected to trial for the sake of ascertaining whether they differed in their reproductive power from the common Oxlip. I did not think it worth the time and labour to

[page 450]

[not transcribed, crossed by Darwin]