CLASSIFICATION A TREE-LIKE STRUCTURE. 6 1 



ing the nearest approximation to the truth, it be- 

 comes evident that the divisions thus recognized have 

 an unequal value. Indeed, one of the best of modern 

 scientific books does not attempt to arrange the 

 animal kingdom into sub-kingdoms, realizing the 

 futility of all such endeavors, but simply takes up in 

 different chapters animals which seem to the author 

 to deserve separate rank. No less than twenty-six 

 different types are thus considered, although their 

 value is not by any means the same, the author rec- 

 ognizing that his division is only provisional. The old 

 idea of types has thus been supplanted by the more 

 modern view, that types are not rigid norms, or that 

 if they do exist we know nothing about them, not 

 even their number, for the animal kingdom can be 

 divided with almost equal justice into five, seven, 

 eleven, or even more different groups. 



It appears then that no rigid lines can be drawn 

 separating from each other the so-called types. Va- 

 rious attempts to draw such lines have been made. 

 It was urged for a long time that no homologies 

 exist between members of the different types ; that 

 the parts of a vertebrate can be homologous with 

 those of any other vertebrate, but never with a mol- 

 lusk or a radiate. If this were true it would certainly 

 define the types. This subject, therefore, has been 

 the scene of a long battle, which is not entirely set- 

 tled even to-day. It is now recognized, however, 

 that it is possible to discover certain homologies 

 between members of the so-called types, though 

 naturally not so exact as between animals more 

 closely related. The wkole series of homologies is 



