It is instructive to find that between Serres and Milne-Edwards there existed the same antagonism as between von Baer and the German transcendentalists. Milne-Edwards was a constant critic of the law of parallelism which Serres continued to uphold with little modification for over thirty years, just as von Baer was a critic of that form of the doctrine which was current in the early part of the century. As early as 1833, Milne-Edwards, through his studies of crustacean development,[319] had come to the conclusion, independently of von Baer, that development always proceeded from the general to the special; that class characters appeared before family characters, generic characters before specific. In an interesting paper published in 1844,[320] he discussed the relation of this law of development to the problems of classification, and arrived at results almost identical with those set forth by von Baer in his Fifth Scholion. Like von Baer he rejected completely the theory of parallelism and the doctrine of the scale of beings; like von Baer he held that the type of organisation--of which there are several--is manifested in the very earliest stages and becomes increasingly specialised throughout the course of further development; like von Baer, too, he sketched a classification based upon embryological characters. These views were further developed in his volume of 1851, and also in his _Rapport_ of 1867. They brought him into conflict with his confrere in the Academy of Sciences, Etienne Serres, who in a number of papers published in the 'thirties and 'forties,[321] and particularly in his comprehensive memoir of 1860, still maintained the theory of parallelism and the doctrine of the absolute unity of type. His memoir of 1860 shows how completely Serres was under the domination of transcendental ideas. Much of it indeed goes back to Oken. "The animal kingdom," he writes, "may be considered in its entirety as a single ideal and complex being" (p. 141). His views have become a little more complicated since his first exposition of them in 1827, and he has been forced to modify in some respects the rigour of his doctrine. But he still holds fast to the main thesis of transcendentalism--the absolute unity of plan of all animals, vertebrate and invertebrate alike,[322] the gradual perfecting of organisation from monad to man, the repetition in the embryogeny of the higher animals of the "zoogeny" of the lower.