Prof. E. Claparéde on the Structure of the Annelida. 351 
- In contrast to this radically false description, let us see how 
the circulation is effected in the normal branchia of an Annelide. 
There cannot be in a regular way any mixture of arterial and 
venous blood ; in fact the artery travels as far as the extremity 
of the branchia, where it bends round to return as a vein. The 
vein and the artery are exactly parallel to each other. Through 
the whole length of the branchia these two vessels are put in 
communication by a double series of vascular loops, which pass 
into the subcuticular layer, and which are subjected with the 
greatest facility to the action of the water charged with oxygen, 
through the very thin cuticle. As to the contraction of the 
supposed ampulle, there is nothing of the kind. Some genera, 
- such as the Terebelle and the Telethuse, for example, certainly 
present rhythmical contractions of the whole branchia, but not 
of the vessels themselves. This fact, however, is exceptional. 
The family Serpulea alone presents in the structure of its 
branchiz a distant resemblance tothe description of M. de 
Quatrefages. In these Annelides the artery is continued directly 
into the vein at the base of the branchie, and from their point 
of union starts a single vessel, which penetrates into the 
branchia and sends a cxcum into each branch of it. But 
M. de Quatrefages describes in the secondary branches of the 
branchize of the Serpulea all his apparatus of ampulle, of which 
not the least trace exists. The cecal vessel does not present 
any ramification ; it is simply cylindrical and contractile, as de- 
scribed by MM. Grube and Kolliker*. In these branchiz the 
blood exhibits an alternating circulatory movement; but this is 
the only exception}; in all the other families the branchial 
circulation constantly takes place in the same direction. Cecal 
vessels with alternating circulation are met with also in the 
tentacles of the Spiodea, Amphictenea, and Pherusea, and in a 
part of the so-called branchial filaments of the Cirratulea; but 
the latter organs are not respiratory (unless perhaps lymphatic). 
How could M. de Quatrefages commit an error so manifest 
and so frequently repeated? This is easily explained. The 
branchiz are in general not cylindrical, but slightly compressed. 
Now, in the position which they must naturally take under the 
microscope, the artery exactly conceals the vein, and one might . 
* M, Milne-Edwards, ignoring these observations, erroneously attributes 
to the Tubicolous Annelida lymphatic branchize exclusively (Lecons sur 
PAnat. et la Physiol. tome ii. p. 103). 
+ I think I have a right to speak thus categorically. Of the twenty-six 
families of Annelida admitted by M. de Quatrefages, I have studied twenty- 
five anatomically, by the dissection of numerous species or individuals. As 
to the twenty-sixth, that of the Hermeliea (Sabellaria), it is too nearly re- 
lated to the Amphictenea and Terebellea to allow us to suppose that it 
differs much from them. 
