248 THE CANADIAN NATURALIST. [A-Ug. 



each of whicli generic names there may be several species. It 

 is evidently not an arbitrary arrangement of ours thus to group 

 species : they naturally arrange themselves in such groups, under 

 the action of our comparing powers. 



3. Genera and Higher Groups. 



In comparing species with each other for purposes of classifica- 

 tion, there are four distinct grounds on which such comparison can 

 be made. These are : — 1st. intimate structural or anatomical resem- 

 blance; 2nd. Grade or rank; 3rd. Use or function; 4th. Plan 

 or type. All of these may be, indeed must be, used in classifica- 

 tion, though in very difierent ways. 



1. Intimate structural relationship is the ground on which we 

 frame Genera. Two or more species resemble each other structurally 

 to such an extent that the same definition will in many important 

 points apply to both. Such species we group in a genus. It is 

 most important to observe, as Agassiz has well pointed out, that 

 this close resemblance in structure is really our main ground for 

 the formation of genera. But for this very reason it is not to be 

 expected in our higher groups. It is the mistaken application of 

 this criterion to classes, which constitutes the leading def>_ct of a 

 work otherwise very valuable, and which I cordially recommend 

 to students, — Huxley's " Lectures on Classification." 



2. Grade or rank refers to degree of complexity of structure, 

 or to the degree of development of those functions that are the 

 highest in the animal nature. A coral polyp is more simple in 

 structure than a fish, and is therefore lower in rank. A fish is less 

 highly endowed in brain, sensation, and intelligence,than a mammal, 

 and is therefore of lower rank. An egg or an embryo is simpler 

 than the adult of the species to which it belongs ; and when one 

 animal resembles the embryo of another, it ranks lower in the 

 scale. A worm ranks lower than an insect whose larva it resem- 

 bles. 



We use this difference of grade or rank in grouping genera in 

 Orders ; but it occupies a very subordinate place in the construc- 

 tion of other groups. Many grave errors have arisen from its 

 indiscriminate application ; most heterogeneous assemblages being 

 formed when we construct groups larger than orders merely 

 on the ground of lower grade : and when, on the other hand, we 

 separate the lower members of natural groups on the ground of 

 simplicity of structure, we fall into an equal mistake of another 



