220 ESSAY ON CLASSIFICATION. 



order to be natural, even though, in practice, they did not 

 always perceive the true value of the characters upon 

 which they established their standard of relationship. 

 Linnaeus, the first expounder of the system of animals, 

 already distinguished by anatomical characters the classes 

 which he adopted, though very imperfectly; and ever since, 

 systematic writers have aimed at drawing a more and 

 more complete picture of the classes of animals, based 

 upon a more or less extensive investigation of their struc- 

 ture. 



Structure, then, is the watchword for the recognition of 

 classes, and an accurate knowledge of their anatomy the 

 surest way to discover their natural limits. And yet, with 

 this standard before them, naturalists have differed, and 

 differ greatly still, in the limits which they assign to classes, 

 and in the number of them which they adopt. It is really 

 strange, that, applying apparently the same standard to 

 the same objects, the results of their estimation should so 

 greatly vary; and it was this fact which led me to look 

 more closely into the matter, and to inquire whether, after 

 all, the seeming unity of standard was not rather a fancied 

 than a real one. Structure may be considered from many 

 points of view : first, with reference to the plan adopted 

 in framing it; secondly, with reference to the work to be 

 done by it, and to the ways and means employed in 

 building it up ; thirdly, with reference to the degrees of 

 perfection or complication exhibited, which may differ 

 greatly, even though the plan be the same and the ways 

 and means employed in carrying out such a plan may 

 not differ in the least; fourthly, with reference to the 

 form of the whole structure and its parts, which bears no 

 necessary relation, at all events no very close relation, to 

 the degree of perfection of the structure, or to the man- 



