134 M. Agassiz on the Classification of Fishes. 



facts, which outgrew, beyond measure, the limits assigned by 

 Linnaeus to the extent of his incomparable diagnoses. Com- 

 parative anatomy, in particular, by investigating the inter- 

 nal structure of animals to the minutest details, furnished 

 to zoology more precise characters for fixing the limits of 

 classes, orders, and families. In place of simple diagnosis, at- 

 tempts were thenceforth made to form descriptive pictures of 

 the entire characters of all the natural sections which could 

 be circumscribed in a precise manner ; naturalists endeavoured 

 to arrange the characters according to their relative value in 

 the functions of life ; species were strictly compared with each 

 other ; all the facts relating to their manner of life, their re- 

 production, and geographical distribution, were carefully re- 

 gistered. It is to the immense influence which the works of 

 Cuvier have exercised on the development of the natural 

 sciences, that this new direction given to zoological studies 

 has principally to be ascribed ; and it may be affirmed that it is 

 in this same spirit that most of the great monographical works 

 which have been continually enriching science for the last 

 quarter of a century, have been conducted. There are few 

 classes which are without their monographs : the facts of 

 structure which have been studied and the species examined, 

 are now in general represented with so much exactness, that 

 we can form an accurate idea of them without ever having seen 

 them in nature. Such a detailed knowledge of species, and 

 such multiplied researches into the organisation of the prin- 

 cipal types of all classes of the animal kingdom, must neces- 

 sarily bring about great changes in classification. According- 

 ly, we have seen systems multiplied without end. Yet, not- 

 withstanding their number, they do not differ essentially from 

 each other, and in all of them we can more or less recognise 

 the influence of Cuvier's works ; the diff*erences which distin- 

 guish them consist principally in the respective position of the 

 great divisions relatively to each other, resulting from the difl'e- 

 rent principles which guided their authors, and the extension 

 assigned to these same divisions ; for it will be understood that 

 we cannot regard as particular systems all the systematic 

 sketches, in which, for the most part, there is nothing original, 

 and the outlines of which differ only in the order in which 



