As Mr. Darwin points out in the passage I have cited, and still more at large in his work "On the Different Forms of Flowers," a large proportion of existing plants are fertilized by the action of the wind; and again, many plants bear two kinds of flowers, the one conspicuous and attractive to insects, the other inconspicuous and which never open to admit the activity either of insects or of the wind. Moreover, there are various other agencies called into play. Some plants, such as the Hypericum perforatum, one of the commonest of the St. John's Worts, and probably the bindweed, are, it seems, fertilized by the withering of the corolla, which naturally brings the stamens into contact with the style, and so transfers the pollen grains from the one to the other.[8] Other plants, again, such as the common centaury (Erythræa centaurium) and the Chlora perfoliata, are fertilized by the closing of the corolla over the anthers and stigma, not in the death but in the sleep of the plant.[9] In the brilliant autumnal Colchicum, and in the Sternbergia, again, according to Dr. Kerner, Nature has recourse to a more complex machinery: the corolla first closes over the anthers, which are at a lower level than the stigma, and takes off some of the pollen; a growth of the corolla carries the pollen dust to the level of the stigma, and a second closing of the corolla transfers the pollen to the stigmatic surface. The pollen has been made to ascend to its proper place by an arrangement which reminds one of the man-engine of a Cornish mine.[10] A similar arrangement is described as occurring in the bright-flowered Pedicularis.[11]
Let us take another group of beautiful flowers which adorn our greenhouses and our tables: I mean the Asclepiadæ, to which the Stephanotis and the Hoya belong. The former is distinguished by the beauty of its scent as well as of its flowers. Both present flowers not merely conspicuous in themselves from their size, form, and colour, but conspicuous also by reason of their grouping. Here, if anywhere, we should expect that beauty should justify itself by its utility. But the facts appear to be just the other way. The pollen is collected together into waxy masses, which are arranged in a very peculiar manner on the pistil; and the pollen tubes pass from the pollen grains whilst still enclosed within the anthers, and so bring about fertilization without the intervention of insect agency. It is difficult to suppose the Asclepiadæ can have become beautiful for the sake of an agency of which they never avail themselves.
Our common Fumitory has not very conspicuous flowers, but still they have considerable attractiveness of form and still more of colour, due both to the individual blossom and to their grouping together; and yet Fumaria is said to be self-fertile.[12]
A much more brilliantly coloured member of the same family is the Dicentra (Diclytra) spectabilis, so familiar in our gardens. Any one who examines the flowers of this species will continually find the pollen grains transferred to the stigma without the slightest trace of the flower ever having opened so as to allow of insect agency. Dr. Lindley[13] has given an account of the mechanism for self-fertilization; and this flower has recently been the subject of an elaborate study by the German botanist, Hildebrand,[14] and he concurs in the view that the anthers inevitably communicate their pollen to the pistil, and that as the result of a very complicated and subtle arrangement of the parts, which it would be useless to attempt to describe without diagrams. But he believes that in addition to the arrangements for self-fertilization, another arrangement exists for producing cross-fertilization by insects; but as the plant has never produced seed under his observation, he is unable to tell whether one mode of fertilization is more useful than the other. I think the evidence of the self-fertilization is far clearer than that of the cross-fertilization.
Now, if the Dicentra has become beautiful in order to attract insects, it must have done so through a long series of developments, for its adaptation to their agency is of the most complex kind. It is difficult to suppose either that, side by side with this development for cross-fertilization, there has been also developed another complex arrangement for self-fertilization, or that an earlier complex arrangement for self-fertilization should have survived through the changes necessary to render the flower fit for insect fertilization. The co-existence in one organism of two complex schemes for different objects, and the interlacing of those two schemes in one beautiful flower (which, if Hildebrand be right, occurs in the Dicentra), seem to be things very improbable if the beautiful flower has become what it is in the pursuit of one only of those objects. These speculations may be premature as regards the particular flower; but the co-existence of two modes of fertilization is not peculiar to Dicentra and seems to furnish material for important reflection.
Yet one more plant must be considered. The Loasa aurantiaca is a creeper which grows freely in our gardens, and has large and brilliantly coloured scarlet flowers turned up with yellow. Its seeds set freely in cultivation. The means by which fertilization is effected are—unless my observations have misled me—very peculiar. When the flower first unfolds, the numerous stamens are found collected together in bundles in depressions or folds of the petals; after a while the anthers begin to move, and one after the other the stamens pass upwards from their nests in the petals, and gather in a thick group round the style; subsequently a downward and backward movement begins, which brings the anthers against the pistils, and restores the stamens nearly to their old position, but with exhausted and faded anthers. I have never seen any insects at work on the flowers, and yet I find the plant to be a free seeder.
So long ago as 1840 M. Fromond enumerated several conspicuous flowers in which, according to his observations, fertilization was effected without the agency of either the wind or insects.[15] And much more recently an American writer, Mr. Meehan, has given a list of eleven genera, amongst others, in which he has observed the pistils covered with the pollen of the plant before the flower has opened, and in the one case which he submitted to the microscope, it was found that the pollen tubes were descending through the pistil towards the ovarium.[16] Amongst the genera he names were Westaria, Lathyras, Ballota, Circes Genista, Pisum, and Linaria.
The instances which I have given are mostly from plants familiar in our fields, our gardens, or our greenhouses. They are, I think, sufficient to make us pause before we conclude that all conspicuous flowers are fertilized by insect agency. It may be that Bacon's warning to attend as carefully to negative as to affirmative instances has been a little forgotten. Moreover, these instances seem to show that it would be a great error to suppose that all flowers are fertilized either by insects or by the wind; and it is probable that the more the subject is considered the more complex will the arrangements for fertilization be found to be.
The agencies to which I have last referred exist, it will be observed, in beautiful and conspicuous flowers; and yet act independently of that beauty and that conspicuousness: so that in each instance these facts are, on the utilitarian theory, unexplained and residual phenomena. They, therefore, demand earnest inquiry. For the existence of a single residual phenomenon is notice to the inquirer that he has not got to the bottom of his subject; that his theory is either not the truth or not the whole truth.
Do the facts justify us in concluding that insect fertilization is more beneficial to the plant than fertilization by the wind or any other agency? Do they afford any sufficient cause for that change from the one mode of fertilization to the other which has been suggested? The facts bearing on these questions are very remarkable; for, as we have already seen, many plants produce two kinds of blossom, the one conspicuous and the other inconspicuous; the one visited by insects, the other self-fertilizing. Recent observation shows that these cleistogamous flowers, as they are called, are present in a great variety of plants.[17] In the violet they are found to exist, being seen in the summer and autumn, when all the more brilliant flowers have gone. The one flower has everything in its favour—honey and a beauty of colour and of smell that has passed into a proverb—and it opens its blue wings to the visits of the insect tribe in the season of their utmost jollity and life. The other has everything against it: it is inconspicuous, scentless, ugly, and closed. And yet, which succeeds the better? which produces the more seed? The cleistogamous, and not the brilliant flowers: the victory is with ugliness, and not with beauty.