666 OLIGOCHAETA 



This genus has been hitherto invariably placed in the family Lumbricidae a 

 situation which is not, in my opinion, at all justified by its anatomical structure. 

 ROSA remarks of the genus that it is ' in fondo un' Allolobophora modificata e in 

 parte degenerata in seguito a perfetto adattatnento alia vita acquatica' a statement 

 which is almost word for word that of BENHAM. 



Its resemblances to Allolobophora are the following: 



(1) Presence of spermatophores. 



(2) Male pores on xv, surrounded by swollen lips. 



(3) Sperm-sacs and sperm-ducts. 



Of these only the first is of any importance ; the second has lost the importance 

 which it once had owing to the discovery of a Kynotus with male genital pores 

 opening upon this segment. 



That the sperm-sacs are like those of Allolobophora is not a matter of first-rate 

 importance. 



As to the presence of spermatophores they are undoubtedly unknown elsewhere 

 than in Allolobophora and Lumbricus with the peculiar form that characterizes them 

 in Criodrilus. Spermatophores, as I have pointed out, do exist in Benhamia and 

 Polytoreutus. It is quite likely that they will be found elsewhere. 



The principal reason which leads me to refer the genus to the family Geoscoliciclae 

 is the extensive clitelluin ; in no Lumbricus or Allolobophora does the clitellum 

 begin so far forward and extend so far backwards ; on the contrary, a clitellum 

 of this kind is perfectly characteristic of the Geoscolicidae, though it is undoubtedly 

 extensive even for that family. 



Other facts in the structure of Criodrilus are not opposed to its inclusion in the 

 present family. It has been pointed out that the absence of spermathecae is a 

 point of resemblance to Allolobophora eiseni and A. constricta ; but there are 

 Geoscolicids (Anteus, Siphonogaster) in which these organs are not present. 



The ornamentation of the setae in Criodrilus seems to be rather doubtful. It is 

 figured by VEJDOVSKY (24), but not by either ROSA or OEBLEY. COLLIN (1) does not 

 refer to it. 



If characteristic, the ornamentation of the setae with transverse ridges is a 

 marked point of resemblance between Criodrilus and the Geoscolicidae. 



Our knowledge of the cocoons of earthworms is not very extensive ; so that it 

 is premature to base anj r affinities upon the few facts that we do know. I call 

 attention, however, as bearing on the matter in hand, to the peculiar form of the 

 cocoon in Criodrilus; it is very long, of comparatively small diameter, and is drawn 

 out at both ends. The only other worm which has a cocoon anything at all like 



