154 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.55. 



glands; moreover, the two ducts of either side of the animal unite 

 into one duct, and there is thus but one pair of spermiducal pores, 

 and these are entirely distinct from the prostate gland pores. This 

 is a condition similar to that which is normal in what are considered 

 to be more primitive forms of Megascolecidae, such as some of the 

 Acanthodrilinae and Diplocardiinae, but with one great difference, 

 namely, that in the latter groups the sperm ducts open far posterior 

 (17-19) to the position in which they open in the new form, which is 

 on the anterior part of 13 and presumably anterior to septum 12/13, 

 which is shifted posteriorly, as described above. Considered with 

 reference to internal metamerism, the sperm duct openings corre- 

 spond in position quite closely with those of Pelodrilus violaceus 

 Beddard, in which the two sperm ducts of either side open separately 

 but quite near each other and nearly midway of the length of 12. In 

 most species of Pelodrilus the two pairs of spermiducal pores are on 

 adjacent somites. 



In the extensive spenn sacs which extend posteriorly from the 

 sperm vesicles and are included in the ovisacs, we have a character 

 in which the new species differs from the earthworms and strongly 

 resembles certain limicoline forms. The unique feature of this sperm 

 sac development is their inclusion within the ovisacs, but perhaps 

 no great stress should be placed on this character, since it does not 

 seem to be found in the Haplotaxidae, from which the earthworm 

 ancestors are assumed to be derived. 



Notwithstanding the important differences above mentioned, it 

 seems to us reasonable to include the new species in the family 

 Moniligastridae, at least tentatively, because of the various important 

 characters in which it is allied with Desmogaster. This will entail 

 some modification of the definition of that family as will be seen by 

 a comparison of that given at the beginning of the discussion with 

 the one of Michaelsen. It seems to us desirable to recognize a sub- 

 family Syngenodrilinae for the new genus and a subfamily Monili- 

 gastrinae for those previously known. 



An attempt to arrange Pelodrilus^ SyngenodrUus^ Desmogaster^ 

 and a primitive Acanthodriline form in a phyletic series has met 

 with difficulty because of the peculiar condition in Desmogaster in 

 which the sperm ducts of either side are separate and yet open into 

 prostate glands instead of independently of them. If Desmogaster 

 is considered as derived from Syngenodrilus then we must assume 

 that it was before the union of the sperm ducts in the latter form 

 or else that after such union they again separated and later came 

 into connection with the prostate glands. 



The family Haplotaxidae is represented by a considerable number 

 of species of Pelodrilus in the South African, Australian, and Xew 

 Zealand regions. If these regions actually were connected in the 



