One need not fail to appreciate its logical completeness, its symmetry, and the skill with which it has been defended, and yet one need not be blind to the fact that it has not been a stimulating guide for its friends. It has been conservative rather than progressive. Founded on a definite morphological conception of the ultimate constitution of living substance, it has not adapted itself plastically to the rapidly changing conditions in biological science. The considerable amendment it has received in the last eighteen years has only made it so cumbersome and complex that it is now little more than a mere formulation of the facts it attempts to explain.
Time will not permit us to explore thoroughly the mass of evidence on which this criticism has been based. While differentiation according to the determinant hypothesis assumes qualitative divisions of the chromatin in the nucleus, numerous investigations have shown that at least five divisions of the egg in some animals may occur before there is any recognizable difference between the cells thus formed. Each of the first sixteen is competent to develop the entire adult structure. The only way to account for such a result in terms of morphological determinants is to assume that a complete outfit passes to each cell with each division of the nucleus, obviously a serious burden for the determinant hypothesis to bear. Further, among these phenomena of development which are conveniently investigated under the head of regeneration, similar difficulties have so constantly recurred, requiring similar assumptions of reserve determinants, that the theory has long since ceased to interest investigators in this field. It follows, rather than leads, investigation. Finally, in the field of heredity, just that characteristic of Mendelian inheritance—namely, the segregation of parental characters in second generation hybrids—which at first seemed to give the strongest support to the conception of a germ plasm composed of morphological determinants, has now been resolved far more satisfactorily, because more simply and workably, in terms of chemical substances.
These cases lay emphasis upon the distinction between morphological and physiological conceptions that defines the essential difference between modern preformation and modern epigenesis. Instead of a congeries of morphological determinants, the epigenesist finds in the germ a problem in physical and chemical relations. He is interested in the dynamic aspects of development, in the energy transformations. He does not seek to construct a scheme of the ultimate organization of living substance, but he does seek to control its operations, to predict its behavior.
In this new form, the problem of differentiation presents many interesting aspects and is being encouragingly developed. By way of illustration, recent investigations indicate that color differentiation is based essentially on a well-known chemical process, the oxidation,
