74 Bulletin American Museum of Natural History. [Vol. XXVII, 



cularian" and "trinitarian" principles are maintained by Swainson at great 

 length. The idea of "wheels within wheels" is worked out in such detail 

 that the reader, reminded of the endless cycles and epicycles of the Ptole- 

 maic astronomy, or of the metaphysical arguments used by Copernicus to 

 establish the sphere as the universal figure of the heavenly bodies, becomes 

 lost in the bewildering labyrinth of "affinities" and analogies. Neverthe- 

 less, the idea which was dimly adumbrated in Blumenbach's "Palmata 

 Glires, "Palmata Ferae" etc., namely that similar functional types, such as 

 the "rasorial," "scansorial," "natatorial," "ghriform," "vermiform," 

 occur in different orders, and the related idea of the parallelism of series, 

 as well as the prolonged analysis of "analogies" vs. "affinities," all fore- 

 shadow the modern discovery of parallel evolution and adaptive radiation; 

 while the whole movement of "philosophical zoology" was of great value, 

 not only in stimulating search for the causes of resemblances and differences 

 among animals, but also because, in one of its less extreme forms it guided 

 de Blainville to the remarkable classification which may now be considered. 



DE BLAINVILLE, 1816. 



'Prodrome d'une nouvelle distribution systematique du regne animale.' 

 Bull, de la Soc. philom. pour I'annee 1816, p. 105. 

 Journ. de phys., t. 83, p. 244. 



The labors of the long series of naturalists from Ray to Cuvier, whose 

 systems have been examined above, had brought to light before 1816 many 

 of the fundamental problems of mammalian taxonomy. The relations of 

 the aquatic mammals to each other and to their terrestrial congeners, the 

 problem of the edentates, the arrangement of the ungulates, the relations of 

 the monotremes and marsupials to other mammals and to the lower verte- 

 brates, the greater problem of the essential nature of "natural" groups, — all 

 these had been formulated, and many contradictory answers had been given. 

 But the net result, in so far as expressed in Cuvier's system, was an ordinal 

 classification still very artificial. The latter was merely a development of 

 the systems of Storr, Blumenbach, Vicq d'Azyr and GeofFroy; and although 

 more brilliant in form and improved in many details, did not withal rise much 

 above these in underlying principles. Cuvier, as remarked by Gill (1907, 

 p. 497) made but little use of his wide knowledge of anatomy in the construc- 

 tion of his ordinal arrangement of the mammals, but followed his predeces- 

 sors in selecting as prime criteria of classification characters of the kind 

 now regarded as relatively plastic and unstable, such as the number of 

 the three sorts of teeth, the number of the digits, the various modifications 

 of the extremities. 



