of the Malpighian Tubules in the Arthropoda. 291 



" coxal glands " and " green glands " are modified nephridia. 

 Sedgwick remarks {he. cit. p. 119) that witii the exception 

 of these structures there are no nephridia in Arthropods 

 recognizable as such ; and this conclusion probably represents 

 the opinion of most comparative anatomists. The claim of 

 the Malpighian tubules to be looked upon as nephridia has 

 been more and more ignored. Gegenbaur (Com p. Anat., 

 Engl, transl. by Bell, p. 276) carefully abstains from discus- 

 sing the morphology of these organs. Lankester (' Notes on 

 Embryology and Classification,' 1877, p. 33) remarks that 

 " in tracheate Arthropods the ]\Ialpighian filaments possibly 

 are the nephridia." Balfour (Comp. Embr. vol. ii. p. 568) 

 doubtfully compares them to the anal vesicles of the Ge- 

 phyrea. Now these structures are so widely spread among 

 the Tracheata (even if the tubes of the Amphipoda are not 

 of the same nature) that they must be regarded as among 

 the most characteristic organs of that group. The fact that 

 they do not occur in Peripatns might perhaps be regarded as 

 evidence that the Malpighian tubes have arisen within the 

 group ; but, on the other hand, their absence from Peripatus 

 may be reasonably accounted for by the persistence of un- 

 modified nephridia performing the same function : in any 

 case a similar argument must be applied to account for the 

 great reduction in the nephridia of the Crustacea ; they are 

 nephridia, and they are reduced in number, in accordance 

 wnth the reduction of the coelom. 



If, then, the absence of the Malpighian tubules in the most 

 primitive known Arthropod, Peripatus^ is not necessarily a 

 real break of continuity, the segmented worms are naturally 

 the animals in which one might expect to find the beginning 

 of these organs, especially in the more primitive segmented 

 worms, for the Gephyreans must be regarded as greatly modi- 

 fied Annelids. 



In a species of Acantliodrilus^ which I refer, at present 

 with some little doubt, to Acanthodrilus multiporus (Beddard, 

 *' On the Specific Characters and Structure of certain New- 

 Zealand Earthworms," Proc. Zool. Soc. 1885, p. 813), the 

 last segments of the body are almost entirely filled with 

 nephridial tufts, which, as I have elsewhere stated, open by 

 numerous pores on to the surface of the body and by nume- 

 rous ciliated funnels into the coelom. The gut in this region 

 of the body has a ytrj narrow lumen and is lined by tall 

 columnar cells, which are not ciliated, as in the intestine 

 generally, but covered with a delicate chitinous cuticle. 

 This region of the gut is probably proctodaium. At irregular 

 intervals minute cacal diverticula arise from the gut ; these 



20* 



