Tlte Classification and Disirihution of Eartliioorms. 253 



Observations. — Apart altogether from the nephridia, it is 

 necessary to include these families in one group : they are in 

 every case so closely connected. The more typical Pericha'- 

 tida3 seem sharply marked off from any others, but BcAno- 

 clrilus is an almost exactly intermediate form between 

 Periclimta and Acanthodrilus. It has more than eight setai 

 in each segment, and a clitellum like that of Perichceta. The 

 male reproductive apparatus is like that of AcanthodriluSy but 

 in P. stuarti of Bourne ^ — a form which I have ventured to 

 distinguish generically — we have also four tiibular atria. 

 Moreover, in Perichceta ccylonica^ there are indications of an 

 approach to Acanthodrilus, though that species requires 

 further investigation. 



With regard to the Cryptodrilidse,^ such a form as Ponto- 

 drilus is very distinct from Perichmta, and in the absence of 

 any knowledge of intermediate forms would have to be 

 separated into a very distinct family. This has been done 

 by Vejdovsky;* but, at the time when he v/rote, the two 

 genera Microscolex and Photodrilus, as well as the Aus- 

 tralian genera described by Fletcher,^ were unknown. 

 These form collectively a family, which is chiefly defined, 

 however, by negative characters. I exclude from this 



separation is still adhered to, but the presence of more than one gizzard is 

 dropped out as a part of the definition. It appears to me also necessary to 

 omit the characters of the clitellum as a definition, since in Acanthodrilus 

 OMncdens — a species with paired nephridia — the clitellum extends beyond 

 segment xix. It may be useful, however, to adopt Dr Michaelsen's separation 

 of Acanthodrili with diffuse nephridia into a distinct genus, as the genus is 

 even now getting inconveniently large. 



^ Preliminary Notes on Indian Earthworms — pt. i. : On Earthworms from 

 the Shevaroys and Nilgiris (Proc. Zool. Soc, 1886). 



^ Notes on some Earthworms from Ceylon and the Philippine Islands, 

 including a description of two new species — Ann. and Mag, Nat. Hist., 1886, 

 p. 89. 



^ I name this family Cryptodrilidse, though on the grounds of priority it 

 ought to be called Megascolididse ; after the recent discussion in Nature 

 (Feb. 13, 1890) about the correct writing of terms borrowed from the Greek, 

 I have not the courage to introduce so awkward a term, and therefore fall 

 back upon Cryptodrilidse. This word has the advantage of being pronounce- 

 able, and in calling attention to the fact that Cryptodrilus is the most, 

 prominent genus of the family. 



4 Log. cit. (on p. 238). 



^ Notes on Australian Earthworms— Proc. Linn. Soc. N.S. W., 1886-88. 



