aparine; that he has raised them in considerable quantities; that he has each year selected his seeds from such of his own seedlings as have shown any tendency to variation; and that this process has been carried on in different soils, in different situations, in different climates, and at different seasons. It is scarcely to be imagined that this has been done for so very uninteresting a plant; and yet, if any one of these precautions has been neglected, it cannot be said to be proved that the plant will never lose any of its characters. And, after all, what are these characters, so invariably reproduced? Not the want of hairs at the nodes, nor the narrowness of the leaves, for these he admits to be variable in his G. tenerum—besides, that such hairiness is often scarcely perceptible in the stoutest specimens of G. aparine—nor yet the glabrous or hispid fruit, for that is admitted to occur in both his species. There remain, first, the size of the plant, not more than a foot in G. spurium, often above three feet in G. aparine; but to which would he refer the numerous specimens occurring in some localities from 1 to 2 feet high? 2ndly. The articulations, swollen in G. aparine, but not in G. spurium, a mere result of the luxuriance of the former. 3dly. The size of the fruit, 4 to 5 millimetres in G. aparine, 3 or 4 times smaller, consequently 1 to 112 millimetres, in G. spurium. To verify this character, I have measured the fruits of numerous specimens, living and dried, of both forms, and I have never found the diameter quite so little as 2 millimetres; but from that size I have measured every intermediate from half to half millimetres, up to 5 millim., the largest I have met with. And 4thly. The hairs of the fruit, rising from a small tubercle in G. aparine, and no such tubercles in G. spurium. As to this point, if we take the hairy-fruited varieties of each form, I confess myself unable to discover any difference but what depends on size; the larger the fruit, and the larger the hairs, the more prominent are the tubercles at the base. Upon the whole, as far as my own experience goes, the results of cultivation constitute an item, but one item only, and that often a fallacious one, among the evidences on which the permanency of character is to be judged by inductive reasoning.
Even the proof of specific identity by cultivation is often liable to error. Such experiments are often several years in carrying on; it is not to be expected that they can be daily watched during the whole of that period, and all who have had the charge of gardens must be aware of the mishaps which may occur during a short absence, without being directly noticed—such as labels accidentally or intentionally destroyed or misplaced, or the sown seed failing, or the seedling perishing, and replaced accidentally by some common allied species or variety. The abnormal circumstances in which a plant under cultivation is placed, may also induce an apparent approach to some other species, without any real alteration of essential character. I have already instanced the Trifolium repens and hybridum, as one in which the supposed proof of identity by cultivation, notwithstanding my confidence in the experimenter, produces no conviction in my mind; and it is only with great
