XXII.] CLASSIFICATION. 291 



that are superior to themselves iu organization, rather than 

 members that are inferior ? — for this is implied in the fact 

 that groups generally approximate by their lowest members. 

 In other words, the earliest formed species or classes of a 

 group are generally the lowest; thus, according to the 

 theory detailed in the last chapter, the Batrachia were the 

 first-formed groiip of air-breathing Vertebrata, and have 

 given origin by descent to the groups above them. The 

 question now before us i.s, why progress is upward, and not 

 downward ; why the later-formed groiips are in most cases 

 of higher organization than their ancestors. I shall have 

 to consider this question in the next few chapters. 



I have said that the leading branches of the organic tree An ex- 

 diverge and re-diverge into classes, orders, genera, and ^a"sifica- 

 species. To make the enumeration complete, however, I tion would 

 should have added races, families, and individuals. Ifn^tonty 

 it is true that community of descent is the real bond of fpecjes 



pr. • Ti 1 . T .„ . but indi- 



organic alnnity, a really exhaustive classification would viduals. 

 include not only every species, but every individual that 

 has ever existed; just as a really exhaustive cosmology 

 would include not only every nebula, every star and planet, 

 and every continent and ocean, but the position of every 

 pebble and the fall of every rain-drop. 



NOTE. 



It has been urged against the validity of the argument from 

 classification in favour of the origin, of species by descent with 

 modification, that crystaUine species also are naturally classified Crystal- 

 into groups, and into groups of groups, and yet there is no bond '^ ^P*^ 

 of descent between crystalline species. I think this is sufii- species. 

 ciently answered by the fact that the unformed but formative 

 material of organisms can only, so far as we know, be produced 

 by organisms ; but this is not true of crystals, as their formative 

 material has a merely chemical source : so that the cases are not 

 parallel. 



U 2 



