2S0 MR FRANK E. BEDDARD 



is demonstrably no connection between the " cardiac body" and the gut wall in the 

 former. This can, I think, be hardly regarded as an objection, though it has, of course, 

 to be proved that this connection does exist at some time or other in Mesenehytrceus and 

 Stercutus. 



It seems to me by no means impossible that the paired blood-glands of Phreodrilus 

 may have been originally paired diverticula — like the calciferous glands — and connected 

 like the latter with a dorsal and ventral vessel. The change of structure has obviously 

 been followed by some change of function, and I should consider that both the cardiac 

 body, and the structures which I describe in the present paper, have some relation to 

 the blood, as Claparede suggested. In relation to this matter I may refer to a highly 

 interesting paper by Weldon on the supra-renal bodies of Bdellostoma [18]. The con- 

 nection of a portion of the pronephros in that animal with the blood system, and its 

 almost complete separation from the rest of the renal organ, is a parallel instance of great 

 interest ; but a better analogy with the vascular glands of these Annelids is perhaps to be 

 found in the thymus gland wdrich, originally a diverticulum of the gut, is converted to 

 some function in relation to the vascular system, and entirely loses its connection with 

 the gut. The vertebrate spleen is another organ which may be possibly foreshadowed 

 in these Annelid structures. 



§ Nervous System. 



The supra-cesophageal ganglia lie between the first and second segments above the 

 dorsal vessel ; a strong nerve leads frOm the fore part of the brain to a patch of 

 modified epithelium upon the dorsal wall of the buccal cavity, just in the angle where 

 it becomes continuous with the epidermis of the prostomium. In my description of 

 the integument, I have not referred to this organ, which appears to be of a sensory 

 nature. 



The ventral chain commences in the Ilnd segment. In each segment three 

 pairs of nerves are given off at approximately equidistant intervals, which at once 

 perforate the integument, into which they can only be followed for a very short 

 distance; besides these, separate branches supply the dissepiments. The branches of 

 the nerve cord furnish characters which appear to be of a certain value for systematic 

 purposes. 



Three equidistant pairs of nerves have been stated to be given off in each segment 

 of Tubifex rivulorum [see d'Udekem, pi. i. fig. 8], and the same number in several of the 

 genera of Tubificidas described by Stolc [5] with the addition of dissepimental nerves 

 which were overlooked by d'Udekem in Tubifex, as Vejdovsky has pointed out [7, see 

 pi. viii. fig. 4]. In Lumbriculae, on the other hand, Vejdovsky could only discover a 

 single pair of nerves in each segment. 



Eisen has mentioned [3 J the very anomalous fact of the absence of any branches at 

 all from the ventral cord of Eclipidrilus, and has lately made the same statement with 

 regard to Sutroa [4]. 



