436 PROFESSOR LANKESTER. 



purely mental abstractions, arbitrarily chosen from among 

 many possible abstractions, would be more justly termed 

 "subjective;" yet there is one objective classification possible, 

 which is no mental abstraction, but as corresponding with 

 the actual order in which the objects of classification were 

 brought into existence may claim to be the objective and 

 natural classification par excellence. This classification would 

 be exhibited in a complete statement of the pedigree of the 

 animal kingdom. 



It is certain that we shall never be able to set forth with 

 anything like completeness this real or objective genealogical 

 classification. Accordingly, all our attempts at it may be 

 called, as they have been, subjective classifications, since 

 they depart from the objective reality in proportion as our 

 imagination has to supply the gaps in our knowledge — in 

 proportion as in them inference is given a particular form 

 where our acquaintance with fact only justifies a general 

 form. But it is clear enough that in its broad outlines we 

 are likely enough as knowledge increases to make our genea- 

 logical classification correspond with the reality if we con- 

 sistently aim at that object. 



What there is of objective in such classifications is at any 

 moment capable of being tested and discussed, whereas 

 classifications which ignore genealogical speculations, though 

 they may claim indisputable objectivity as their justification, 

 yet fail to commend themselves equally to their authors and to 

 contemporary naturalists. Such classifications may be as 

 numerous as they are true, but, after all, their utility is mea- 

 sured by the approximation which they make to the formula- 

 tion of what they profess to ignore, namely, the pedigree. 

 The logical fact set forth in this kind of classification, how - 

 ever indisputable, is one of small or even over-estimated 

 consequence. 



In fact, if we agree with Mill that the higher kind of 

 classification is that which seeks " to arrange objects in such 

 groups, and those groups in such an order as will best con- 

 duce to the ascertainment and remembrance of their laws," 

 we cannot doubt that a classification of animals which keeps 

 the law of evolution steadily in view is more likely to be 

 that which will best conduce to the ascertainment and 

 remembrance of its operations than one which expressly 

 ignores that law. 



Homogeny and homoplasy — Progression and degeneration. 

 — In the attempt to draw up a true pedigree of the animal 

 kingdom it is clear enough that we are seeking to trace the 

 lines of heredity, and that we must proceed primarily on 



