ZOOL.— Vol. II.] RISEN— OLIGOCHMTA. IO9 



sizes. The largest of them are never longer than the short 

 diameter of the nucleus of the lymphocyte, but they are 

 generally very much smaller. The majority of microcytes 

 show a central dark-staining point, while others are appar- 

 ently homogeneous, with only a shaded center. These 

 microcytes are scattered everywhere in the lymph and even 

 among the tissues, and now and then are found even in 

 the cytoplasm of the amoebocytes. The smallest micro- 

 cytes in some respects resemble bacteria, but the great 

 variation in siz^ precludes the possibility of their being 

 such. It is possible that the microcytes are homologous 

 with plasmocytes (Eisen 20). I believe they will be found 

 to constitute regular elements of the lymph. 



OCNERODRILIN^. 



General Remarks. 



There can be no doubt, as Michaelsen has pointed out, 

 that Beddard's genus Ilyogenia is a true Ocnerodrilide, 

 the only distinguishing character being the absence of pros- 

 tates. The same character distinguishes the author's genus 

 Phoenicodrihis, which, being later than Beddard's genus, 

 must be withdrawn. 



The prostates in Ocnerodrilus vary so much in size and 

 development that it is difficult to determine when they are 

 entirely absent. . In Phoemcodrilus tepicensts a trace of the 

 prostate remains as an atrial chamber near the male pore. 

 In the closely related species P. taste there is no trace of 

 this muscular part of the prostate. The former species thus 

 stands intermediate between Ocnerodrilus and Ilyogenia or 

 Phoenicodrihis taste. PygtncBodrihis has already been with- 

 drawn by Michaelsen, but I prefer to retain both Ilyogenia 

 and PygnicBodrilus as subgenera, to which I have added five 

 others, each subgenus being defined by several characters. 



Most of the African species belonging to group V seem 

 to differ from all the other species in having the male pores 

 approximated, a fact which is interesting to note in connec- 

 tion with Michaelsen's suggestion that the family of Eudril- 

 inae has descended from Ocnerodrilide ancestors. 



