M.A. de Quatrefages on the Classification of the Annelides. 9 
less widely from the family with which they have been so long 
united, are, however, well distinguished by a striking degrada- 
tion both internal and external. Moreover, in proportion as we 
are acquainted with it, this family of Sy/lidea acquires more and 
more the physiognomy of a little world apart, in which organic 
variability is displayed within still more extended limits than in 
the rest of the class, and which of itself presents examples of 
some of the most interesting physiological phenomena. I refer 
to the facts of geneagenesis which have hitherto only been ob- 
served in this family and in some small species of Sedentari of 
which we cannot make a distinct family. 
To sum up, of the fifteen families which compose the order 
Erraticze, seven possess branchie, and eight are destitute of those 
organs. The advantage in favour of the latter increases consi- 
derably when we descend to the details of specics and genera. 
To the abranchiate types belong all those genera which are dis- 
tinguished by the number of their specics (such as Polynoé, 
seventy- seven species, Nereis, cighty-one specics). If we exa- 
mine the order Sedentarize from this point of view, we find that 
it is in quite a different case. Here, of eleven families, three only 
are deprived of branchial organs; eight possess well-marked 
branchie. Moreover, of the three abranchiate families (Cheto- 
pterea, Tomopteridea, Clymenea), there are two which together 
only include three genera with very few species; whilst among 
the branchiate families we find the richest in genera and species 
(Terebellea and Serpulea*). From this comparison we may con- 
clude that among the Erratic Annelides the type tends up toa 
certain point to be realized without special respiratory organs ; 
whilst among the Sedentary Annelides the opposite’ tendency is 
most distinctly manifested. 
In both orders we meet with species bearing branchiz on the 
head, and others bearing them on the body. But in the Erra- 
ticee the former form only a single family, composed of a small 
number of genera and species (Chloremeat); in the Sedentariz, 
on the contrary, the family which presents this peculiarity is 
much richer in genera and species (Serpulea). Moreover the 
Chloremea, by the totality of their organization, and especially 
by the entirely exceptional arrangement of their ‘digestive appa- 
ratus, constitute a truly aberrant group in the midst of the other 
families of the order. On the contrary, the Sedentarie with 
cephalic branchiz probably present the most complete realization 
of the type of the order to which they belong. 
If we were better acquainted with the organization of the Seden- 
tariz with abdominal somatic branchie, we might probably be 
* The Sabelle and allied genera belong, in my opinion, to this family. 
+ With me the buccal ring forms part ‘of the head. 
