DESCRIPTIONS OF GENERA AND SPECIES 471 



among the aquatic Oligochaeta has usually been put down to the soft nature of their 

 food ; just as graminivorous differ from carnivorous birds by the increased muscularity 

 and internal thickening of the analogous organ. But the ' food ' of Pontodrilus would 

 seem to need a gizzard for its trituration more than that of any other earthworm, and 

 yet Pontodrilus, though it has a gizzard, has a very feebly developed one. I had 

 expected that the Bermuda specimens would show a greater development of gizzard 

 than those from the Mediterranean shores which feed upon soft sea-weeds, but this 

 was not the case. 



(3) Pontodrilus hesperidum, BEDDARD. 

 P. hesperidum, BEDDARD, P. R. Phys. Soc., 1893, p. 37. 



Definition. Length, about 25 mm.; diameter, 1-5 mm. Setae paired, but the individual seta 

 rattier distant ; setae 2 of segment XVIII absent. Septa V/XIII are thickened, and 

 particularly the last three. Gizzard entirely absent; intestine, begins in XV. Last 

 hearts in XIII. Sperm-sacs in XI, XII. Hah. Jamaica. 



This is the smallest species of the genus, but as the single individual examined 

 by myself was immature, the account I have given of it was not complete. There 

 is, however, no danger of confounding it with any other species of the genus. The 

 enormous thickness of some of the anterior septa is a noteworthy point about it. 

 I have mentioned in my description of the species that the spermiducal glands are 

 lined by a single layer of cells. This may be simply due to immaturity. . 



(4) Pontodrilus insularis (ROSA). 

 Cryptodrilus insularis, ROSA, Ann. k. Hofm. Wien, vi, 1891, p. 387. 



Definition. Length, 50 mm.; diameter, 3 mm.; number of segments, 100. Septa VI/XIII 

 very greatly thickened. Rudimentary gizzard in VII. Sperm-sacs in XI, XII. Sperma- 

 thecae in VIII, IX, without diverticula. Hal. Aru Island. 



This species was not definitely referred to the genus Cryptodrilus by ROSA, but 

 only ' provisionally,' pending a revision of the family ; there can be no doubt, I think, 

 that it is rightly referable to the genus Pontodrilus, as denned in the present work ; 

 the nephridia do not commence until the thirteenth segment, a character which is 

 found in all the species of the genus, though there is some variation in the actual 



