Are the arbitrary “stages” of the embryologists--the ovum, blastula, gastrula, etc., phases in a system in the above sense, the only sense in which the process can be regarded as capable of physico-chemical analysis? What precisely is the embryo at the close of the process of segmentation? It is an harmonious equipotential system, that is to say, an assemblage of discrete organic parts or cells, each of which has all the potentialities that every one of the others has. _Any_ cell in the blastula may become a cell, or a series of such, in _any_ part of the gastrula or pluteus larva. This is what the parts are in potentiality, but actually their individual fates are different. The system is an harmonious one, and each of its parts, although able to do whatever any other part can do, yet does one thing only: it becomes an endoderm cell, or an ectodermal cell, or a part of the skeleton, and so on; what it does depends on its position with regard to the other cells. An extensive manifoldness or diversity is produced, but this was not the consequence of a preceding extensive manifoldness, for in the preceding stage _all the parts of the system were the same_. The manifoldness of the ovum or blastula--that potential manifoldness which became actual in development--must be an intensive manifoldness, and admitting this we must abandon the comparison of the ontogenetic (and, of course, phylogenetic) processes with the phases of a physico-chemical system in process of transformation. _Evolution is the transformation of an intensive into an extensive manifoldness._