buds, as I have already mentioned, contain the young leaves of the following year; the flowers of the Horse Chestnut or Maple (figs. 8 and 9) may be found in the bud in the preceding October; in some Conifers the development of the leaf even occupies two years,

FIG. 10.-YOUNG SHOOT OF LIME (Tilia). Reduced. St, stipules of terminal bud; B, axillary buds, the upper of which replaces the terminal bud on the fall of the latter.

so that if we open a bud in the autumn we may find the rudiments not only of next year's leaves, but even of those of the spring following.

There is a remarkable point about the Lime and some of our other forest trees and shrubs, which Vaucher (2) seems to have been the first to notice, namely, that the terminal buds die, and that very early. Fig. 10 represents a twig of Lime drawn at the end of May; the terminal shoot and stipules (St) are very small, and easily drop off. If a branch be examined a little later, it will be found to be terminated by a scar, left by the true terminal bud, which

has dropped away, so that the one which is apparently terminal is really axillary.

Fig. 11 represents the end of a shoot of Hornbeam (Carpinus Betulus), taken in July, and shows how snugly the bud nestles between the stump of the terminal shoot and the petiole of the leaf.

P,

upper part having been cut off; b, bud; t, terminal shoot, the upper part has already dropped off; s, scar of stipule.

The same thing occurs in the Elm, Birch, Hazel-Nut, Lilac, Willow, &c. In these and many other species the bud situated apparently at the end of the branchlets is in reality axillary, as is shown by the presence of a terminal scar, due to the fall of the true terminal bud. I have found that even at the end of May the terminal buds of the Lime have almost all died and fallen away.

But why do the terminal buds wither away? In some cases the bud contains a definite number of leaves,

but in the genera above mentioned the number is indefinite—more than can come to maturity; and yet the rudiments, which are constructed to produce true leaves, cannot modify themselves into bud-scales. Thus, in the Ash, Maple, Horse Chestnut, and Oak, which have true terminal buds, there are comparatively few leaves; while in the Elm there are about seven, Hornbeam eight, Lime eight, Willow fifteen, and Lilac fifteen.

In the above species it is generally the uppermost lateral bud or buds which develop, but in some cases, as in Viburnum Opulus (the Guelder Rose), Gymnocladus, &c., these also perish, and as a rule only the lower ones grow, and the upper part of the stem dies back.

The arrangement of the leaf in the bud influences, and sometimes determines, the form of the leaf.

This consideration explains, I think, the curious fact that the first leaves, or cotyledons, often, indeed generally, differ altogether in shape from the true leaves. They offer an immense variety of form; not quite so innumerable, indeed, as those of true leaves, of which Linnæus truly observed that 'Natura in nullâ parte magis fuit polymorpha quam in foliis,' but still immense. They may be large or small, broad or narrow, entire or much divided.

Now, why should the first leaves differ so much from their successors? The reason, I believe, is that while the forms of leaves often depend greatly on the 'Philosophia Botanica.

buds, those of cotyledons are even more often influenced by the shape of the seeds. Let me give two instances in illustration.

I will take first the Common Radish (Raphanus sativus), with which, as regards the cotyledons, the Cabbage and Mustard closely agree. The seed of the Radish is shaped as in fig. 12. What regulates the shape

FIGS. 12-14.-RADISH (Raphanus sativus).

12, outline of seed, x4: m, micropyle; h, hilum; 13, embryo in vertical section, × 4, showing the folded cotyledons and root (r) lying between them; 14, germinating seedling, showing the cotyledons still folded, x 2.

of the seed is another question, into which I will not now enter. The young plant, consisting of two leaves, a small root, and a minute bud, occupies the whole interior of the seed. Each leaf is folded on itself like

a sheet of notepaper, and one lies inside the other. To this folding the emargination is due. If a piece of paper be taken, folded on itself, cut into the form shown in fig. 12, with the fold along the edge from m to h, and then unfolded, the reason for the form of the cotyledons becomes clear at once.

Now let us test this explanation by another case. The Wallflower has a seed of similar shape to that of the Radish, though thinner. Now, are the cotyledons of the same form as in the Mustard or Radish? Not at all. Those of the Mustard, as we have seen, are kidney-shaped; those of the Wallflower are racket-shaped, as in fig. 15. At first this seems a FLOWER. Two thirds nat. difficulty; but on looking closer the difficulty vanishes, for while the cotyledons of the Mustard are folded, this is not the case with those of the Wallflower, which lie flat in, and conform to the shape of, the seed, as shown in fig. 16. Thus the difference, which at first sight seemed a difficulty, is really a confirmation of the explanation suggested.

FIG. 15.-SEEDLING OF WALL

size.

But we may even carry the matter a step farther. Why are the cotyledons of the Radish folded, and not