NEW GENERA AND SPECIES OF EARTHWOEMS. 263 



glandular portion is^ as in Gordiodrilus and Ocnero- 

 drilusj lined by a single layer of large columnar glan- 

 dular cells. 



The calibre of the tube is, however, rather greater than in 

 either of those two genera ; but still there is no indication of 

 the double layer of cells such as is found in Acanthodrilus, 

 &c. As already mentioned, no penial setse exist. There are a 

 single pair of sperraatothecse in Segment 9. One of these is 

 shown in fig. 16, partly in optical longitudinal section; the 

 pouch itself is large and somewhat oval in form ; the duct is 

 long, but not so narrow in proportion to its length as in Gor- 

 diodrilus tenuis. The pouch is lined by a single layer of 

 columnar cells (fig. 16, a) which are irregularly defined at the 

 free extremities ; they are here prolonged into short processes, 

 and contain clear vesicles : this is no doubt associated with the 

 production of a slightly granular substance occupying the 

 interior of the pouch. There were no spermatozoa to be 

 found. The duct of the spermatotheca is lined by an epithe- 

 lium of a non-glandular appearance, and has a muscular 

 investment which is present as an excessively thin layer upon 

 the pouch itself; there are no diverticula. 



I refer this worm to the genus Pygraseodrilus chiefly on 

 account of the form and position of the oesophageal append- 

 ages. In certain other points it difi'ers from P. quilima- 

 nensis, P. bukobensis, and P. affiuis, the remaining 

 species of the genus, and from Gordiodrilus, with which 

 Pygmseodrilus has affinities. 



The investment of the end of the vas deferens by muscular 

 fibres does not form a very important point of resemblance 

 between P. lacuum and P. quilimanensis, since the same 

 differentiation occurs in Microdrilus (nov. gen.). Michaelsen 

 says nothing about the structure of the atria in the other species 

 of the genus. The form of the spermatothecce is very dif- 

 ferent in P. lacuum from what is found in the three remain- 

 ing species. It is rarely the case that diverticula are present 

 and absent in the same genus, or even in the same family. 

 The only instances that I can at present recall are Acautho- 



