584 TAXONOMY. 



and it seeks in their arrangement to express this interlinking in their 

 subdivision, and, thereby to produce the proof of the correctness of the 

 opinion. It thus creates a system which therefore must be natural, 

 that is to say, such as appears expressed in nature itself. This 

 system can, however, only be constructed, when not only all the forms, 

 but also all the ideas which express themselves in these forms, are 

 known, and when it is seen that every form has absolutely a thought as 

 its foundation, and that it is not an accidental but a necessary one. 

 But the idea to which the study of natural bodies leads is their 

 gradual development, to exhibit which is the proposition of the 

 systematist. 



Both subdivisions, for the system is also a subdivision, have this 

 in common, that they form groups, the members of which possess 

 certain characters, and by collecting these groups together by means of 

 still more general characters, a survey of the entire contents is effected. 

 They nevertheless sometimes lead to very different results, by separating 

 a division that connects the rest, and vice versa. But system has the 

 advantage of not regarding solitary characters only, but all collectively, 

 and can therefore only separate and connect where nature itself has 

 marked a separation and connexion. 



320. 



The methods whereby both attain the goal are different, for artificial 

 subdivision proceeds from the characters of the last group, which we 

 generally call species, and collects similar species under one common 

 character, and thence forms the genus ; the characters common to 

 genera give those of the higher groups, the orders and their common 

 characters combine to form those of the classes. It depends, there- 

 fore, upon every classifier how far he will proceed in separation 

 and subdivision. Indeed, much difference of opinion exists upon the 

 determination of the groups between the species and the order, whence 

 have arisen the several definitions of sub-genus, genus, and tribe. In 

 fact, opinions will never harmonise upon the claims of genera, because 

 no universal principle for the structure of genera in an artificial sub- 

 division can be given. This principle is in itself exceedingly capricious, 

 and if one maintains thus far a genus extends, and another thus far, 

 both are certainly right, if only every group which they distinguish as 

 genera are distinguished by similar and exclusive characters. 



The natural system, the object of which is to discover analogies and 



