HOMOLOGY. 73 



mental features of both. Now this means absolutely 

 nothing, for it necessitates the assumption that 

 there once existed a vertebrate with only one pair 

 of appendages ; and no one has the hardihood to 

 make such a claim. In such serial appendages we 

 have, therefore, an homology which is not open to 

 the explanation given to homologies in general, and 

 the fact that the explanation will not apply here 

 makes it at least a legitimate question whether it is 

 sufficient in the other cases. At all events, unless 

 some other way of accounting for serial homologies 

 can be discovered, it can hardly be claimed that the 

 reason for homologies is understood. The whole 

 argument for descent, derived from comparative 

 anatomy, is, in a measure, weakened unless these 

 cases be met. 



Spencer has attempted to answer this question in 

 accordance with the theory of evolution. He 

 would distinguish two sorts of serial homologies 

 which have had different origins. The first is that 

 which is found in Crustacea and other animals, 

 where the whole body consists of a series of equiva- 

 lent segments, each having essentially the same 

 parts. This sort of segmentation has arisen, accord- 

 ing to Spencer, by multiple division. Many low 

 animals possess the power of dividing into two 

 parts, each of which is like the other. Every organ 

 shares in this division. If now, as frequently is the 

 case, these two segments, instead of completely sep- 

 arating from each other, remain connected after the 

 division is nearly complete, we should have a com- 

 pound animal of two segments, in which every part 



