774 PROFESSOR J. STEPHENSON ON 



through the Enchytrieidse, Naiclidse, and Tubificidse, to that presented in the higher 

 families of the Order. Again, speaking broadly, it is scarcely feasible to read this 

 series downwards, as a number of successive stages in a process of degeneration ; it 

 would, for example, be difficult to suppose that a developed dorsal vessel has united 

 with the gut-wall and finally dissolved in the vascular network ; and it would be im- 

 possible to contend that the contractions of an originally distinct dorsal vessel, which 

 were by chance in the same direction and of the same nature as those of the underlying 

 intestine, united by a series of stages with those of the gut in proportion as these 

 structures became gradually united anatomically, until finally there was a physiological 

 fusion and solution, if I may use these terms, similar to the anatomical fusion and 

 solution. 



It may, of course, be urged that the lateral loops might nevertheless undergo de- 

 generation while the rest of the system continued on the upward path. But it is 

 unlikely that in a system which has been shown to be generally progressive so 

 important a part should develop in a contrary direction, and the burden of proof must 

 lie on those who suo-gest it. 



(2) The fact that there are a certain number only of lateral loops, and these in the 

 anterior part of the body, in the Naididse and Enchytrseidse, is in harmony with the 

 idea of a progressive rather than a regressive development. We have seen that 

 the anterior region is the region of greatest differentiation of the vascular as of the 

 associated alimentary system ; the dorsal vessel first appears here as a distinct vessel ; 

 when formed throughout the length of the body, it separates from the alimentary canal 

 here first ; an indication that the ventral vessel develops in a similar way is seen in 

 jEolosoma ; the supraintestinal and subintestinal vessels are developed first, or only, 

 in the anterior region ; and as regards the transverse channels, it may be recalled that 

 in Chsetoguster the parallel transverse vessels in the alimentary wall are most regular, 

 numerous and conspicuous on the crop, less so on the stomach, and become indistin- 

 guishable on the intestine. We should, I think, naturally expect to find the separated 

 lateral commissures at first few and anterior, as in Nais, Pristina, Dero, etc. ; then 

 more numerous and extending further back ; and finally in regular series throughout 

 the body, as in Branchiodrilus among the Naididse, in the Tubificidse, and in the 

 earthworms. 



It appears to me, however, to be impossible at present to come to a definite 

 decision on the question ; and accordingly the above statement of the arguments 

 on both sides is all that will be attempted here. 



The Course of the Circulation. 



The relations between the contractions of the alimentary wall and those of the 

 dorsal vessel, previously described, explain to us the fact that the course of the blood 

 in the dorsal vessel of the Annelida is from behind forwards. Even where this vessel 



