Count Lacepede (m the Natural History of Fishes. 269 



mate, and who could speak of their habits, the advantages which 

 they afford, and the injuries which they occasion. To appre- 

 ciate these testimonies, he must know all the circumstances of the 

 authors which he consults, their moral character, and their de- 

 gree of instruction ; he must be able to read almost all lan- 

 guages. The historian of nature, in a word, cannot overlook any 

 of the resources of criticism (that art of finding out the truth so 

 necessary to the historian of man), and he must moreover join to 

 it a multitude of other talents. 



M. de Lacepede, when he composed his work on fishes, was 

 far from being placed in circumstances under which the resources 

 of which we speak were entirely at his disposal. The anatomy of 

 fishes was not sufficiently advanced to furnish him with the basis 

 of a natural distribution. A general war had established an al- 

 most insurmountable barrier between France and the other coun- 

 tries ; it had shut up the seas against us, and separated us from 

 our colonies. Foreign books, also, did pot reach us ; nor did 

 travellers bring home those collections, so numerous and so rich, 

 which arrived among us as soon as the seas were open ; even 

 Peron himself, who had been employed in a voyage of discovery 

 during the war, had not arrived when the work was finished. 

 The author, therefore, could only take for the subjects of his 

 observations, the individuals collected in the Royal Cabinet be- 

 fore the war, and those afforded by the cabinet of the Stadthold- 

 er, which was brought to Paris after the conquest of Holland. 

 Of the naturalists who had preceded him, he selected Gmelin and 

 Bloch as his principal guides, and perhaps he followed them too 

 faithfully, punctual as he was in observing the same courtesy 

 toward authors as to society. The drawings and manuscript 

 descriptions of Commerson, and paintings formerly made by 

 Aubriet, after drawings by Plumier, were almost the only un- 

 published resources which it was possible for him to have access ; 

 and yet, with materials so poor, he succeeded in collecting up- 

 wards of 1500 fishes, whose history he traced ; and estimating 

 at the highest the number of species described more than once, a 

 kind of error unavoidable in such a work, and which sometimes 

 he fell into, there would remain from 1200 to 1300 undoubted 

 and distinct species. Gmelin had only about 800, and Bloch 

 in his great work did not exceed 450 ; and there are not more 



