t See Rev. George Henslow. Jour. Linn. Soc. Lond. 29: 485-528. 1893.
adapted for specialization in this direction, it was the one upon which the office devolved. Every botanist knows what an endless variety of forms and special adaptations of particular foliar parts have arisen in the course of evolution which was inaugu rated when this setting aside of the leaf to bear in future the weight of the assimilative function took place, or rather when this additional function was placed upon it, for the old protective function has always been retained, though it has become less no ticeable as the new function has overshadowed the old.
There has been in the line of vegetable descent a progressive development of the foliar organ, and a history of this devel opment, together with that of other organs, if it were obtainable, would give us a complete phylogeny of the flowering plants, and leave no morphological problem unsolved,* but as the geological record is very incomplete, and we have in the lower Cretaceous an already well developed and much differentiated angiospermous flora of the earlier history of which almost nothing is known, we must seek other sources of information in determining the homol ogies of parts. At this juncture we may safely follow the exam ple of the zoologists and turn to embryology for the evidence which geology, as yet, refuses to give except in fragments Among animals, as the phylogeny and ontogeny are found to par allel one another, so we may feel confident they will be found to do among plants when the geological record shall be more com pletely unearthed.
It has become a well established part of the theory of evolu tion that each individual organism epitomizes more or less fully in its development the historical steps in the evolution of the type to which it By the application of this law of re capitulation to the development of plants we may arrive at valu able and trustworthy conclusions. The question would at once be asked, where shall the embryology of the flowering plants be studied, and the answer would naturally be, in the development of the seed in the ovary. And here indeed, we trace in outline an epitome of the course of development from the simple unicel lular organism, represented by the fertilized egg-cell of the ovule to the highest thalloid form, the " embryo," with its bud (plu mule) which is to develop into the full-formed plant perfect in all its parts. For a summary of the further development of the
Angiosperms we must look to the growing bud which is the essential reproductive organ of the sporophyte stage and, doubt less, a more primitive one than the seed, for it is common among the more ancient Pteridophytes and these have no seed. The embryo of flowering plants does, however, correspond pretty closely to that advanced stage of development of the egg-cell of some of the higher Pteridophyta now generally spoken of as the embryo and should be regarded as a young plant in a state of arrested development. In this state it remains during a period of rest, in a highly specialized environment in the seed, await ing favorable conditions for farther growth. Because of the highly specialized environment of the embryo, it has itself be come correspondingly specialized and has been variously modi fied to suit the special conditions of its surroundings. The pin mule cannot then be regarded as any longer representing a prim itive form of bud and its development is so altered by secondary modifications that the series of phylogenetic changes is disguised and imperfectly represented. A parallel case is found among animals in the development of Echinoderms, in which the changes that have taken place through secondary modification are so great that the relationship of the group cannot be satisfactorily deter mined by developmental evidence.
It is not then in the seedling that we should expect to find rep resentations of primitive leaf-forms, though later ancestral forms paralleling those of fossil leaves, of which we shall speak, are found in some seedlings, as for example in Liriodendron. But it is in the growth of the less specialized buds developing under more primitive conditions that we should expect to find them. Such buds are the ordinary leaf-buds of perennial plants, and especially those occurring on basal and subterranean portions which I con ceive to develop under conditions somewhat more primitive than is the case with aerial buds. But in both these the recapitulation of the development of leaf-forms may be traced with a consider able degree of confidence, from the primitive sheathing protective scale to the most highly differentiated and complex of modern leaf-forms.