neio African Earthworm. 459 



is situated near tlip stalU-oiul of tlie cell. In the middle 

 cliaaiber the cells (fig. 5, b) are niucli taller and narrower, 

 and liave a very short " cup '^ at the free end, a much 

 narrower stalk, and a smaller nucleus, situated further from 

 tlie l)aseincnt-menibrane. 



Tlie siKM-m-containing diverticula (figs. 3 & 4, Div.), 

 varying in number from one to three, which project on the 

 anterior and outer surface of the organ, open into the cavity 

 of the middle chamber by a narrow passage a little above its 

 junction with the muscular duct. When more than one of 

 these diverticula or pockets are present, their ducts appear 

 to unite into a single canal of small calibre. There is a 

 muscular coat external to the epithet iuin lining the pockets, 

 which ])robably serves for the expulsion of the sperm at the 

 ai)propriate moment. 



The pockets and their ducts, like the rest of the sperma- 

 tlieca, are lined with a glandular epithelium, the cells of 

 -tthicli are similar to those of the middle chamber, but 

 smaller. The contained sperm is massed together into a 

 solid ball — so much so that, on teiising ont the mass of 

 sperm from the pocket, I at first imagined that it was 

 enclosed in a membrane. On the examination of sections, 

 however, this does not appear to be the case. It is probable 

 that the spermatozoa were swimming in a fluid mediam, 

 which has become sf)lid as the result of preservation. 

 Possibly it is the function of the cells of the epithelium to 

 ])roduce a special fluid for the nutriment of the spermatozoa 

 during their retention in the pocket. If they were " glued ■" 

 together, in the manner described, during life, it is difficult 

 to understand how they could ever escape from the ])ockets 

 again, since the way of egress is so narrow. 



The muscular duct of the spennalheca seems invariably to 

 contain a tough cuticular lining of peculiar shape. This is, 

 in fact, the lower, trumpet-shaped portion (fig. 6, T.) of the 

 " spcrmatophore," and is almost exactly like that described 

 by Bcddard in the case of D. avsteni. To the more detailed 

 study of this portion of the apparatus I shall return later. 

 The upper part of the sijcrmatophore consists of a spherical 

 capsule (fig. 0, Cap.) of a fibrous or parchment-like con- 

 >istency, which "when fully developed occupies the middle 

 chamber of the sperraatheca, and fits into the trumpet- 

 shaped upper end of the cuticular tube. This capsule and 

 the trumpet-shaped tube together make up the complete 

 '' spcrmatophore." 



The wall (;f the upper capsule is composed of numerous 

 lasers of some substance which is apparently non-cellular 



