84 Mr. F. E. Beddard on a 



The spermatotliecse have a somewhat unusual form. There 

 are two pairs of these organs in segments vii. and viii. The 

 pouch is lined with a single layer of columnar cells, covered 

 by an exceedingly delicate layer of muscles. The pouch 

 was invariably much crumpled ; it was filled with an ahnost 

 homogeneous coagulated mass, in which no spermatozoa could 

 be detected. The pouch communicates with the exterior by 

 a very long and slender duct (see fig. 4) ; the length of this 

 duct is only paralleled among earthworms by the genus Monili- 

 gaster] it has thick muscular walls. The sperm atothecfe are 

 quite unprovided with diverticula. 



3. Gordiodrilus elegans, sp. n. 

 (Ph VI. fig. 1 ; PI. VII. figs. 6 A, 7.) 



From the same locality as that which produced the last 

 species were a number of small slender worms^ which I took 

 at first for immature examples of Gordiodrilus rohustus. 

 They prove, however, to belong to a distinct though closely 

 allied form. 



The principal points of difference apart from the shape 

 concern the sperm atothecse and the alimentary canal. 



The pores of the atria and of the vasa deferentia appear 

 also to be on different segments ; in the present species they 

 certainly lie upon segments xviii. and xix. : that is to say, 

 the first pair open on to segment xviii., the second pair 

 and the vasa deferentia open, independently of each other, on 

 to segment xix. In Oordiodrilus rohustus it is the xviith 

 and xviiith segments which appear to bear these pores ; but 

 as the sections were rather broken, owing to the presence of 

 grit in the alimentary canal, I may have made a mistake of 

 one segment. 



In any case, this species can be distinguished by the entire 

 absence of a gizzard and by the much shorter duct of the 

 spermatothecEe [cf. figs. 4 and 7). 



External Characters. 



The shape of Gordiodrilus elegans as compared with the 

 last two species can be seen in fig, 6 ; it is a much more 

 slender worm than G. rohustus, and is not nearly so long as 

 G. tenuis. As in both these species^ the male pores open 

 upon an area which is marked off from the rest of the body. 

 When the body-wall is examined as a flat preparation in 

 glycerine, the area (see fig. 1) is seen to be circumscribed by 

 a ridge with an undulating outline. The appearance of 

 this area is at first sight remarkably like the area surrounding 



