4 M.A. de Quatrefages on the Classification of the Annelides. 
by my predecessors. In every case where, in a collection of 
species, the differences have appeared to me to depend solely 
upon the more or less marked development of one or several 
characters, I have united them in a single generic group, 
confining myself to the establishment in the latter of tribes and 
sections fitted to facilitate investigation. Thus the genus 
Polynoé, for example, contaming seventy-seven species, has 
been ‘divided into two tribes and ten sections. 
In return, whenever I have noticed very distinct characters, 
I have not hesitated to establish a genus, even should it contain 
only a single species. This circumstance has occurred several 
times in the family of the Syllidea. ere the confusion of the 
two parts of the head, and the consequent non-distinction of 
the antennee and tentacles, had often caused the union of spe- 
cies which, when once the nature of these parts and organs was 
recognized, evidently required to be separated. 
The families once determined, it remained to group them in 
orders and suborders. ‘This distribution, attempted at different 
times, had led my predecessors to results which sometimes 
differed considerably. Without dwelling upon purely historical 
details, I shall confine myself here ‘a the indication of the 
course followed by me. 
If there be a group in which the employment of aux the 
characters is not only useful but necessary in the appreciation 
of zoological relations, it is most certainly the group of Annelides, 
and this in consequence of the extreme variability by which it 
is distinguished. But the more we attempt to grasp the 
characters, the more indispensable does it become to arrange 
them in the order of their importance. Now to judge of this 
importance the naturalist must choose between two modes of 
action which are very different, although often confounded— 
that of Cuvier and that of Jussieu. 
The former places himself at the physiological point of view. 
He seeks the dominating characters in the organs charged with 
the function which appears to him to be of the highest value. 
This mode of appreciation presupposes that each function is 
performed by means of a special organ. Now at the present 
day we know that this is by no means the case in a great 
number of Invertebrata. The method of Cuvier therefore re- 
poses on an @ priori which is true for the Vertebrata and for 
some groups of Invertebrata, but incorrect for the rest. The 
Annelides present frequent examples of this inexactitude, and, 
indeed, precisely in the order of the anatomical arrangements 
belongmg to one of the most important functions, to one of 
those which Cuvier placed in the foremost rank—that of respi- 
ration. It is scarcely necessary to refer to the fact that, in this 
