446 OLIGOCHAETA 



Later in the paper, which contains the definition of Notoscolex (on p. 570), 

 Cryptodrilus is thus defined, though in a more informal fashion : ' postclitellian, has 

 eight rows of setae arranged in a peculiar manner, has three or four pairs of accessory 

 gland-pores, two pairs in front of, and the others behind, the male pores which are on 

 segment xviii, and one gizzard.' 



In a later paper, Fletcher (4, p. 614) came to the conclusion that the differences 

 between these two genera were not as originally stated ; he says : ' Hence it seems 

 to me that the chief difference between the genera Notoscolex and Cryptodrilus is 

 becoming narrowed down to the question of segment xviii being included in the 

 girdle. I begin to suspect, therefore, that when I come to revise the species already 

 described, with additional material to work upon, it may .be necessary to include the 

 latter in the former genus, or, at any rate, to regard it as a subgenus.' 



The difiiculty of distinguishing these two genera was still further commented upon 

 by Fletcher in the following year (1888). He writes as follows: 'They include . . 

 eleven species of Cryptodrilus, which, with four ah-eady described, make a rather 

 heterogeneous collection of postclitellian worms (with four couples of setae to 

 a segment, male pores on xviii, a single gizzard, and, in such cases as they have been 

 visible, a single pair of vasa deferentia), among which it is possible to pick out 

 a well-marked group of which G. mediterreus may be taken as the type, and 

 a second smaller group of the type of C. saccarius ; the remainder differ among 

 themselves, and from these so much as not at present to permit of their being 

 satisfactorily sorted into sections. As every considerable acquisition of new material 

 throws fresh light on this matter, it would be premature just now to attempt to 

 separate any of them as types of new genera. Nevertheless, worthy of mention are 

 G. fastigatus, with three pairs of nephridio-pores to a segment, and two pairs of 

 prostates, the two prostatic ducts of each side uniting to form a single genital duct ; 

 G. unicus with a single median series of spermathecal and male pores — an intermediate 

 condition between this and the rather widely separated pores in forms like G. sloanei 

 being offered in C. vianifestus ; and C. singularis with but single pairs of testes, 

 ciliated rosettes, and vesiculae — a condition which obtains also in Megascolides 

 (Notoscolex) illawarrae: . . . characters other than those of the clitellum, which will 

 satisfactorily separate forms like Megascolides {Notoscolex) and Cryptodrilus, are still 

 desiderata.' 



I shall attempt to divide up the large number of species which have been referred 

 to these two genera in any way which appears to me to recognize the principal 

 structural differences. 



The three most salient points of difference, as it appears to me, between the 



