2 M.A. de Quatrefages on the Classification of the Annelides. 
divide it into subordinate groups. Many attempts have been 
made in this direction: I myself, as early as 1849, proposed a 
distribution which, dividing the Vermes into two series com- 
posed of corresponding terms, allows us to appreciate and dis- 
tinguish the relations of analogy and the relations of affinty*. 
This mode of conception of this embarrassing group, which every- 
thing seems to me to justify more and more, led me from that 
period to separate from the class of Annelida two great groups 
which had been united therewith by Cuvier, Lamarck, and their 
successors, namely the Lumbricina and the Hirudinea, which to 
me constitute two distinct classes, that of the Erythrema and that 
of Bdellea. 
Thus reduced, the class of Annelida, as I understand it, no 
longer contains either the armed Gephyrea, which have been 
placed among the Cheetopod Annelides by several naturalists, or 
the Leeches and Lumbricina. It 1s composed entirely of the 
Annélides dorsibranches and Annélides tubicoles of Cuvier (A. né- 
réidées and A. serpulées of Savigny ; A. errantes and A. tubicoles 
of Audouin and Milne-Edwards, and of most authors ; Rapacia, 
Limivora, and Gymnocopa of Grube). 
As by most of my predecessors, the totality of species here to 
be arranged is divided by me into two orders ; but the conside- 
rations which have led me to this result differ from those which 
have generally been followed. Hence result considerable differ- 
ences in the formation of the orders themselves and of the sub- 
orders, and in the number and arrangement of the families. 
The latter first occupied my attention. In my eyes they con- 
stitute the fundamental element of every systematic classification. 
Essentially they are only the Linneean genera better understood 
and better defined. The species once distributed into really 
natural families, their grouping in divisions of a higher order 
pecomes at once easier and more certain, and in any case we 
must pretty nearly get correct aud distinct notions upon the. 
totality of the class. 
It is because I am deeply convinced of the truth of the pre- 
ceding statements that I set myself especially, and in the first 
place, to limit my families strictly, not placing in them any but 
% here reproduce the table which I published in the * Institut’ (No- 
816) :— . 
Dic:cIOoUS WORMS. Monccious WORMS. 
Annélides. Eryihrémes. 
Rotateurs. 
Géphyriens. 
Malacobdelles. Bdelles. 
Myocaleés. Turlellariés. 
Neématoides. 
Cestoides. 
