and the powers of assimilation, is checked by natural selection, which, by placing at a disadvantage, brings about the ultimate elimination of multicellular organisms, in which the tendency is displayed.
On the other hand, this or that cell-descendant of the germ may vary from the normal in a direction which is not towards the ancestral type, but in a different direction, and its descendant cells then form a mass or group which also differs more or less from the normal. The bud variations which have been observed on several cultivated plants, by taking advantage of which gardeners have sometimes been able to establish new varieties, probably result from this cause.
Secondly, each successive germ cell must differ from the ancestral conjugating unicellular organism more and more, in that different lines of its cell-descendants differentiate more and more from one another in structure and function, whence arise in high animals differentiated and specialized tissues such as muscle, bone, skin, gland, nerve, &c.
Moreover, just as each germ cell during the phylogeny differed more and more from its prototype, the unicellular organism, in that it was the starting-point of a more and more complex organism (mass of cell-descendants), so each embryo during the whole process of its ontogeny differs from its prototypes in the phylogeny, in that it carries within it the potentiality of developing beyond those prototypes. We cannot, however, discern in the macroscopic or microscopic appearances of germ or embryo any peculiarities of structure which imply this potentiality. They lie beyond the ken in the minute structure of the cells, probably in those portions of them which are known as the nuclei.
It may be objected, (1) that it is impossible that during the short period of the development of the
