M. A. de Quatrefages on the Classification of the Annelides. 5 
class, certain groups have well-developed branchie, whilst other 
groups, sometimes very nearly allied to the former, do not 
present the least trace of special respiratory organs. Cuvier’s 
principle, and the rules which he deduced from it, are therefore 
inapplicable to this class. 
Jussieu kept strictly to observation. With him the most 
essential character is that which persists in the largest number 
of species and groups. This rational and wise manner of 
appreciating the value of characters is that which I have thought 
it necessary to adopt. 
It has led me to recognize that one of the fundamental 
principles taught by Blainville had in this case a very decided 
value, and that it was in the modifications of the external form 
that we should seek for the bases of the distribution of the 
families. 
Thus the Annelides are essentially dicecious animals, com- 
posed of segments which repeat themselves, and bear on each 
side a perfectly characteristic organ—a foot armed with exsertile 
and retractile sete. 
It was natural enough to think that the modifications bearing 
on this general type must have a great value in relation to the 
present matter. In particular, every exception to the law of 
repetition appeared necessarily to take a place in the first rank, 
and to be the more important in proportion as it reached a 
greater number of secondary groups. 
In fact, when we examine the Annelides from this point of 
view, we find that they divide at once into two groups. In one 
of these the same parts are repeated from one extremity of the 
body to the other. Hence the animals present no distinct re- 
gions. ‘This group constitutes our first order, that of the AN- 
NELIDE ERRATICA. It is composed almost entirely of species 
belonging to the Dorsibranches of Cuvier, the Hrrantes of MM. 
Audouin and Milne-Edwards, and the Rapacia of Grube; I 
have only added to them the Chloremea and the Polyoph- 
thalmea. 
In the second group the law of repetition of parts is suddenly 
interrupted in particular places, and the body is thus composed 
of distinct regions, in each of which the segments resemble 
each other, whilst they differ from one region to the other. This 
constitutes is for me the order of the A. sypENTARIA. It includes 
all the Tubicoles of Cuvier and of Audouin and Milne-Edwards— 
that is to say, the Serpulées of Savigny, the Limivora of Grube. 
I also place with them a certain number of the Hrrantes of the 
former, some Rapacia of the latter of these naturalists, and the 
Tomopterides (Gymnocopa, Gr.). 
Each of these two orders is divided into two suborders by 
