COSMIC PHILOSOPHY 



habits of life, with six or seven of the forms into 

 which the original didelphian had at an earlier 

 date branched off. And we are also obliged 

 to maintain that the remarkable shoulder-struc- 

 ture of the pterodactyl, in which it agrees so 

 closely with the carinate birds, was independ- 

 ently evolved and has a purely physiological sig- 

 nificance. That is to say, the resemblance of 

 the pterodactyl to carinate birds is a secondary 

 adaptive resemblance, like the less marked re- 

 semblance of bats to birds, or like the resem- 

 blance of a porpoise to a fish. And this view, 

 which seems to be Professor Huxley's, is ren- 

 dered probable by the fact that in wing-structure 

 the pterodactyl differs from birds in much the 

 same way that a bat does. 



We are now extricated from our imbroglio 

 with regard to classification, but we are still left 

 confronted with the difficulty of supposing that 

 the natural selection of casual variations can so 

 often have resulted in producing whole orders 

 of closely resembling animals from distinct an- 

 cestral orders. Other facts, brought up by Mr. 

 Mivart, still further increase the apparent diffi- 

 culty. The most important of all these relate to 

 the development of the higher organs of sense 

 in the three sub-kingdoms of annulosa, mol- 

 lusks, and vertebrates. Coincidences between 

 the members of any one of these sub-kingdoms 

 and the members of the others are not to be 



76 



