PREFACE TO THE FIRST EDITION. ix 
The classes and orders were not only not sufficiently con- 
formed to the intimate nature of animals to serve conve- 
niently as a basis to a treatise on comparative anatomy, but 
the genera themselves, although mostly better constituted, 
presented but inadequate resources, on account of the species 
not having been arranged under each of them, in conformity 
with these characters. Thus in placing the Sea-cow (Mana- 
tus, Cuv.) in the genus Morse (Trichechus, Lin.), the Siren 
in that of the Eels, Gmelin had rendered any general propo- 
sition relative to the organization of these two genera imposs- 
ible, just as by approximating to the same class the same 
order, and placing side by side the Sepia and the fresh- 
water Polypus, he had made it impossible to say any thing in 
general on the class and order which embraced such different 
beings. 
The examples above cited are selected from the most strik- 
ing of these errors ; but there existed an infinitude of them, 
less sensible at the first glance, which presented difliculties 
not less real. 
It was not enough then to have imagined a new arrange- 
ment of classes and orders, and to have properly placed the 
genera there; it was also necessary to examine all the species 
in order to be assured, whether they really belonged to the 
genera in which they had been placed. 
Having come to this, I found species not only grouped or 
dispersed, against all semblance of reason, but I remarked 
that several had not been positively determined; neither by 
the characters assigned to them, nor by their figures and de- 
scriptions. 
Here, one of them, by means of synonymes, represents seve- 
ral in one single name, and often so different from each other 
that they should not be placed in the same genus 3 there, a 
single one is doubled, trebled, and successively reappears in 
several subgenera, genera, and sometimes in different orders. 
What shall we say, for instance, of the Trichechus mana- 
tus of Gmelin, which in one single specific name comprises 
three species and two genera ; two genera, differing in almost 
Vou. I.—(2) 
