144 PHILOSOPHY OF ZOOLOGr. 



set of character s is easily recognised, and keeps together at 

 great number of animals, in which the systems of organs 

 exhibit the same general structure. So apparent, indeed, 

 are the characters of the former groups, that they have been 

 recognised by ARISTOTLE, RAY, and many modern authors. 

 When we employ a threefold method of primary divi- 

 sion, as LINNAEUS, or a fourfold as CUVIER, we are obliged 

 to make use of more characters than one in their construc- 

 tion, and depart from that unity of principle to which the 

 twofold method can exclusively lay claim ; in other words, 

 it becomes impossible to assign any limits to the number of 

 the primary classes. If we employ one character, it must ei- 

 ther be positive or negative. The negative is indivisible, but 

 the positive is susceptible of inferior arrangement, from the 

 numerous modifications of the organs from which it has 

 been derived. If, instead of making divisions, from the 

 modification of the positive character, or from its negative, 

 we make each of our primary divisions from positive charac- 

 ters, in that case, their numbers must necessarily equal the 

 number of systems of organs which animals possess, and 

 the same animal must have a place assigned it in each class 

 (or in several of them,) a method of classification which, 

 we have already said, is natural, but which has been de- 

 clared inadmissable. 



The primary divisions of LINNAEUS, which are three in 

 number, may serve to illustrate some of these remarks. 



1. COR biloeulare, biauritum ; sanguine calido, rubro. 

 (Mammalia, AveSj *.) 



2. COR uniloculare, uniauritum - y sanguine frigido, rubro. 

 (Amphibia, Pisces.) 



3. COR uniloculare, inauritum ; sanie frigida, albida, 

 (Insecta, Vermes.) 



* Systema Naturee. Holmiae, 176G, i. p. 79. 



