524 CRliSSWELL SlIKAREK. 



As the following observatious apply only to the male worm, 

 1 have never observed tlie fifth nepliridiuui which, according 

 to Harmer (8), is present in the female. Schimkewitsch (23) 

 has confirmed this observation on the White Sea species, so 

 there can be no doubt of the presence of this fifth pair in the 

 female. Nelson (19), in an American species obtained like 

 Korschelt^s (11) from an aquarium tank, has been unable to 

 find any trace of the nephridia; this is the more remarkable 

 on account of their almost constant presence in other species. 

 They can so readily be observed in D. tseniatus that it seems 

 almost impossible they could have been overlooked if present 

 in this American species. In D. gyrociliatus,^ D. vorti- 

 coides and D. tgeuiatus there are five pairs of nephridia in 

 the female, and this would seem to bear out Harraer's sugges- 

 tion that possibly the body iu the genus Dinophilus is com- 

 posed of five metameres. What we know of the disappearance 

 of nephridia and the apparent ease with which they can be 

 dropped from various segments, as, for instance, in the 

 anterioi' segments of the Arenicola larva, the presence of 

 more than one pair to a segment, as in Capitella, and their 

 very erratic behaviour in the development of Oligocheets, 

 renders it doubtful whether they offer trustworthy evidence 

 as an index to the number of segments, and their total 

 absence may possibly exemplify this in the case of the Ameri- 

 can species of Dinophilus. Again, Ctenodrilus, a form 

 usually classed Avith the Archiannelids, possesses only one 

 pair of nephridia although the animal is plainly divided at 

 least into seven metameres." On the other hand, in llis- 

 triobdella, according to Foettinger (4), there are five pairs 

 in the female as in Dinophilus, Harmer has emphasised 

 this as indicating another point of relationship between them. 

 I have, however, re-examined the nephridia in Foettiuger's 



' D. gyrociliatus according to llepiachoff (21) is identical willi D. 

 apatris of Korschelt's (11) paper. 



^ I have couCrmcd Kennel's (10) and Zejipelin's (27) observations on a 

 species of Ctenodrilus found at Naples, and there is only one pair of these 

 structures present. 



