Fertilisation of Leschenaultia.—

As “F. W. B.” inquires, in your number of August 26, about the seeding of Leschenaultia, I will give my small experience.² During 1860 and 1862, I was led to make some observations on the fertilisation of L. formosa and biloba, from having read that with these flowers self-fertilisation was an inevitable contingency; and this, from what I had seen during many years, seemed to me highly improbable.³ I found, as “F. W. B.” states, that before the flower expands, the anthers open and the pollen is shed. This occurs in a considerable number of plants, as in most Leguminosæ, Fumariaceæ, &c.;⁴ but it can be clearly shown that this by no means necessarily leads to self-fertilisation. In Leschenaultia the pollen, when shed, is neatly collected in a cup-shaped indusium, the mouth of which is at first widely open, but soon closes. Thus far I can follow “F. W. B.;” but he will, I think, find, on further examination, that the pollen must, in order that the flower should be fertilised, be subsequently removed from the indusium, and then placed on an exterior stigmatic surface. This no doubt is effected by insects, tempted to visit the flowers by the copious supply of nectar. On the outside of the indusium there is a viscid surface, and when on two occasions I placed some pollen-grains on the surface, I found, after an interval of about 20 hours, that it was deeply penetrated by numerous pollen-tubes.⁵ I was so much surprised at this position of the stigma, that I asked Dr. Hooker to dissect some flowers, which he did with care, and he confirmed my conclusion with respect to L. formosa. He also examined two other species, and found no trace of a stigma within the indusium.⁶ I should here add that Mr. Bentham has subsequently described the structure of the parts in this genus, but I cannot at the present moment lay my hand on his paper.⁷ When the flower is fully expanded the lips of the indusium fit closely, and cannot be very easily opened. If, however, a finely-pointed, small camel-hair brush be held parallel to the pistil, and be gently inserted into the flower, so as to imitate the entry of an insect, the tip of the brush, by pressing against the slightly projecting lower lip of the indusium, opens it; and some of the hairs enter and become smeared with pollen. If the same brush be now successively inserted into several flowers, pollen-grains will be found left on the exterior viscid stigma.⁸ During the early part of the summer I treated in this manner several flowers, but with no result. Towards the end of July, however, five flowers were thus treated, and the germens of all soon became much enlarged. Two of them, after a time, shanked off, but three remained on till the autumn, and each contained about 25 seeds.⁹ My plant produced hundreds of flowers during two or three summers, but the germens of none spontaneously swelled, with the exception of two growing close together, which I imagined had been visited by some insect. These two produced some seeds, but fewer in number than in the above case. All the seeds were in external appearance good, but when sown they did not germinate. The flowers were necessarily fertilised with pollen from the same plant, but it would have been incomparably better if pollen from a distinct seedling plant could have been employed. This would have been all the more advisable, as the late Mr. Drummond, of Swan River, in Australia, to whom I wrote, asking him to observe in the proper season what insects visited the Leschenaultias, informed me that the species growing there in a state of Nature very rarely produce seed.¹⁰ It appears at first sight a surprising circumstance that in this genus and in some allied genera, the pollen, whilst the flowers are still in bud, should be scooped out of the anthers, in which it might have remained ready for use, and then be immediately enclosed in a specially contrived receptacle, from which it has afterwards to be removed, so as to be placed on the stigma. But he who believes in the principle of gradual evolution, and looks at each structure as the summing up of a long series of adaptations to past and changing conditions—each successive modification being retained as far as that is possible through the force of inheritance—will not feel surprise at the above complex and apparently superfluous arrangement, or the other still more complex arrangements, though they may all serve for one and the same general purpose. Any one desiring to learn how diversified are the means for preventing self-fertilisation, even within the limits of the same family of plants, should study Mr. Bentham’s short but extremely curious paper, just published, on the styles of the Australian Proteaceæ.¹¹ I cannot resist specifying one of the remarkable contrivances described by Mr. Bentham. In Synaphea the upper anther does not subserve its proper function of producing pollen, but has been converted into a short broad strap, firmly fixed to the edge of the stigmatic disc. By this means the stigma is held in such a position that it cannot receive pollen from the fertile anthers of the same flower; or, as Mr. Bentham puts the case, “the stigma thus held by the eunuch (i. e., the barren anther) is safe from all pollution from her brother anthers, and is preserved intact from any pollen that may be inserted * * by insects or other agencies.”¹²

Charles Darwin.