THE LOGIC OF ORGANIC EVOLUTION. 389 



same logical process was exemplified in the discovery of abdomi- 

 nal ribs in the human embryo. 



As long ago as 1801 Blumenbach inferred from the configura- 

 tion of its skull that ornithorhyncus laid eggs, and his deduction, 

 based on the known correlation of characters, was verified by 

 Caldwell's recent demonstration that the monotremes are egg- 

 layers. There was an unverified deduction that monotremes 

 must at some time have possessed normal teeth ; it was verified 

 by the recent discovery of calcified teeth in monotreme embryos. 

 " Thatcher showed in 1887 that fins of fishes are derived from two 

 pairs of lateral folds. In a paper on the significance of bone 

 structure Dwight commented on this and said that if evolution 

 were true he could see no reason why no vertebrate had more than 

 four limbs. He also said that he could see no reason why no 

 vertebrate had more than two eyes. In the same year Watase 

 demonstrated that the Japanese goldfish has eight limbs and 

 Spencer showed that all vertebrates have a third eye in the pineal 

 gland. Spencer predicted that fossils would be found in which 

 the pineal eye was functional ; Cope demonstrated in fossils the 

 orbit of this third eye and pointed out the attachment of muscles 

 for its movement. " * 



" Evolution suggested the annelids as remote ancestors of the 

 vertebrates. Kowalevsky was thus led to the discovery of germ- 

 layers among these animals — homologous with those of verte- 

 brates. In the same way, but on the other side, Semper was led 

 to the discovery of genuine worm kidneys (nephridia) in the lower 

 fishes. Starting with the theory of genetic affinity, we have dis- 

 covered a whole series of organs and tissues that had hitherto re- 

 mained unknown. For example, we have now the ' spinal gan- 

 glia ' in worms, the ' spinal nerves/ and, what is perhaps more 

 important, we have discovered the sense organs out of which the 

 sense organs of the vertebrates have been gradually built up." f 



One of the best-defined anticipations, and one that found con- 

 fident expression from different sources, in the way of special 

 predictions, long before it was verified, was the belief that primi- 

 tive mammals were generalized types. Cope wrote, March, 1874, 

 " I trust that I have made it sufficiently obvious that the jjrimi- 

 tive genera of this division of mammals (Mammalia educdbilia) 

 must have been bunodonts with pentadactyl plantigrade feet." J 

 " No perissodactyl or artiodactyl mammal was known at that 

 time to possess such feet, nor was any perissodactyl known to 

 possess tubercular teeth." Since the prediction was made, Cope 

 has described nine species of the Eocene genus Phenacodus, " prob- 



* Kingsley. f Whitman. 



\ Journal of Academy of Natural Sciences, Philadelphia, March, 18V4. 



