100 DR. C. L. BOULENGER ON 



short, narrow duct (text-figs. 21 R and 22, and PI. YIII. figs. 2 

 and 4), also lined by a ciliated epithelium ; the latter, however, 

 consists of very small cells with long cilia, and seems continuous 

 with the epithelium lining the coelom. The diameter of this 

 tube is less than half that of a mature ovum such as one finds 

 in the uterus ; it is, however, surrounded by transversely 

 arranged muscle-fibres, which no doubt allow the dilatation of 

 the tvibe necessary for the passage of such ova. into the oviduct. 



The above account shows that, at least in M. costatum, the 

 oviduct is quite a distinct structure, and I have therefore used 

 for it the name employed by the older investigatoi's of Myzo- 

 stomid anatomy. I cannot follow Wheeler and v. Stummer- 

 Traunfels, who refei' to this organ as a median posterior branch 

 of the uterus. 



JVephridia. 



The nephridia resemble those described by Beard (2), Wheeler 

 (17), and V. Stummer-Traunfels (16) in other species. They have 

 the form of a pair of bent tubes opening antei'iorly into the dorsal 

 coelomic space (text-fig, 22), and sloping outwards and backwards 

 towards the ventral surface so as to come to run parallel with the 

 alimentary canal (text-fig. 21) ; they open posteriorly by separate 

 openings into the commencement of the rectum (text-fig. 19). 



Along the greater part of their coiirse the nephridia are of 

 considerable diameter and are lined by large glandular cells with 

 long cilia ; they, however, narrow considerably towards their 

 anterior ends, where the lining cells become much smaller and, 

 as a matter of fact, the nephridial epithelium passes gradually 

 into that of the cojlom. In this i-espect I am in agreement with 

 Maidl (12), who in a recent paper described much the same in 

 M. glabrum. 



The nephrostomes are situated on either side of and very close 

 to the internal opening of the oviduct, although not quite in the 

 same plane as this structure, being slightly dorsal to it ; all three 

 openings can, however, be seen in one horizontal section. 



The nephridia contain numerous immature and fragmentary 

 ova (text-fig. 21 C), but, like Wheeler and other investigators since 

 Nansen, I have never seen complete, ripe ova in this situation. 



The similarity in structure and position between the nephridial 

 and oviducal openings in M. costatum is very striking, and has, 

 perhaps, some bearing on the homologies of these organs. Nansen 

 (15), who described a very similar arrangement in M. carpenteri, 

 referred to them as three oviducts, and we must remember that 

 before his discovery of the " lateral oviducts " Beard (1) had 

 suggested that "the oviduct opening into the cloaca may perhaps 

 originally have been formed from two segmental organs, for the 

 cloaca is an epidermic invagination, and if it were obliterated the 

 oviduct would open on the median line." That partial fusion of 

 a pair of segmental oi'gans can take place has been shown by 

 Wheeler (17), who described the nephiidia of M. glahrum as 



