Cuvier's first objection was of course determined to some extent by the imperfection of the palaeontological knowledge of his time. But even at the present day the objection has a certain force, for although we have definite evidence of many serial transformations of one species into another along a single line, for example, Neumayr's _Paludina_ series, yet at any one geological level the species, the lines of descent, are all distinct from one another.[66] Cuvier recognised very clearly that there is a succession of forms in time, and that on the whole the most primitive forms are the earliest to appear. Mammals are later than reptiles, and fishes appear earlier than either. As Deperet puts it, "Cuvier not only demonstrated the presence in the sedimentary strata of a series of terrestrial faunas superimposed and distinct, but he was the first to express, and that very clearly, the idea of the gradual increase in complexity of these faunas from the oldest to the most recent" (p. 10). He did not believe that the fauna of one epoch was transformed into the fauna of the next. He explained the disappearance of the one by the hypothesis of sudden catastrophes, and the appearance of the next by the hypothesis of immigration. He nowhere advanced the hypothesis of successive new creations. "For the rest, when I maintain that the stony layers contain the bones of several genera and the earthy layers those of several species which no longer exist, I do not mean that a new creation has been necessary to produce the existing species, I merely say that they did not exist in the same localities and must have come thither from elsewhere."[67] It was left to d'Orbigny to teach the doctrine of successive creations, of which he distinguished twenty-seven (_Cours elementaire de palaeontologie stratigraphique_, 1849). Cuvier, however, can hardly have believed that all species were present at the beginning, since he does admit a progression of forms. Probably he had no theory on the subject, for theories without facts had little interest for him. At any rate it is a mistake to think that Cuvier was a supporter of the theological doctrine of special creation. His philosophy of Nature was mechanistic, and he dedicated his _Recherches sur les Ossemens Fossiles_ to his friend Laplace. He admitted the idea of evolution at least so far as to conceive of a development of man from a savage to a civilised state.[68] He refused to accept the extravagant evolutionary theory of Demaillet and the somewhat confused theory of Lamarck (whom he joins with Demaillet),[69] just as he rejected the transcendental theories of Geoffroy St Hilaire, because they seemed to him not based upon facts. [41] _Lecons d'Anatomie Comparee_, tome i., pp. 10 _et scq._, 1800. [42] _Lecons d'Anatomie Comparee_, i., p. 18. [43] _Loc. cit._, i., p. 13. [44] _Lecons d'Anatomie Comparee_, tome i., Articles iii.-iv., 1800. [45] _Lecons d'Anatomie Comparee_, i., p. 47.