576 OLIGOCHAETA 



so many Eudrilids is correlated with the fact, that in thiis family the ova undergo 

 their whole development in the egg-sacs. The first description of the egg-sac of an 

 Eudrilid was given by myself for Eudrilus ; in this genus I described the egg-sac 

 as an ovary, on account of the presence of germinal cells in all stages of develop- 

 ment. A study of immature examples of Polytoreutus kilindinensis (84) showed 

 that the ovary is apparently transferred to the egg-sac en niasse before the germinal 

 cells had commenced to become differentiated in the direction of eggs ; in a young 

 specimen of this worm the egg-sacs were filled with cells, unmistakably generative 

 cells, while there was no clearly recognizable trace of an ovary. 



In the simpler forms of the family the ovaries retain their freedom, and are in 

 no way related to the oviducts ; in the genera Eudriloides, Platydrilus, and 

 Megachaeta (?) both the ovaries and the funnels of the oviducts hang freely into the 

 thirteenth segment ; in all other genera the ovaries are enclosed in a peritoneal 

 sac. The complication of the system of peritoneal sacs surrounding the female 

 reproductive organs varies in degree. The simplest form is Eudriloides; here the 

 arrangement of the different organs is precisely as in other earthworms, except 

 for the fact that the spermathecal sac opens on the thirteenth segment and is of 

 considerable length. The next stage in the complication of the various organs 

 concerned is seen in the genus Platydrilus; here we have, as already mentioned, 

 the ovaries and the oviducal funnels quite free, but the oviduct appears to com- 

 municate with the median spermatheca ; this matter requires looking into again, as 

 the said communication was only seen in Platydrilus lewaensis, where no egg-sac 

 was observed. In all other genera at present known, the formation of peritoneal 

 sacs which enclose the female organs of reproduction have reached a greater degree 

 of development. There are many differences of detail, but apart from these, which 

 are described under the descriptions of the genera, we can distinguish two main 

 types as it appears to me. In one series the ovaries are enclosed in sacs which are 

 prolonged into a median spermathecal sac often surrounding the gut, and reuniting 

 on the dorsal side of the gut into a backwardly directed sac of variable dimensions ; 

 these sacs do not communicate directly with the exterior on their own account ; but 

 they enclose and perhaps originally open into the egg-sac into which opens the 

 funnel of the oviduct; within the peritoneal sacs lies a spermatheca which opens 

 into the exterior in the middle line, but does not communicate with the peritoneal 

 sac in which it lies. This type is represented by my genera Hyperiodrilus and 

 Heliodrilus; it is exceedingly hard to understand how fertilization can take pla,ce 

 in these two genera on the hypothesis (possibly a perfectly gratuitous one), that the 

 method of procedure is as in the British Lumhrici. If sperm is emitted from one 



