or THE LAEVAL POLYPTEKUS. 335 



two series had a very similar appearance, but I was unable to show their actual origin. 

 I have now indicated the origin of the genital duct in the female. 



On re-examining my sections of the young male 9 cm. in length, I find an indication 

 . that the posterior part of the genital duct is formed in the same way, namely, by a 

 folding-off of a groove in the coelomic wall on the outer side of the genital ridge 

 (PI. XXXV. figs. 9, 10, g.d.). 



Now this is precisely the region where in the female I have found the nephrostomes 

 opening, and I can but suppose that the larval male possesses them too in early life. 

 That is to say, in all probability, in the larva of the male Polyi^terus there will 

 be found nephrostomes opening into that portion of the coelom which becomes 

 folded off to form the male genital duct. If, then, nephrostomes were retained in the 

 adult, we should have formed a series of tubules connecting the genital duct with the 

 glomeruli of the kidney; and if the genital duct ceased to perforate the wall of the 

 urinary sinus the genital products could find their way out through the kidney, 

 and the nephrostomes would be converted into vasa efferentia, similar to those of 

 Lejndosteus and other Ganoidei, of Elasmobranchii, Dipnoi, Amphibia, and Amniota. 



I suggest, then, that since I have previously given reasons for supposing that the 

 genital ducts in the male and female Polypterus are homologous, the discovery of these 

 nephrostomes opening into the female duct precludes the possibility of their being 

 homologous with Miillerian ducts, but indicates a possible homology with the longitu- 

 dinal canal of the testicular network of those forms in which such a network occurs, 

 and the possible origin of the vasa eff"erentia, leading thence to the kidney, from 

 nephrostomes. 



Upon this view the genital ducts of Polypterus, Ganoidei, and Teleostei have arisen 

 from a condition similar to that of the Cyclostoraata, where the generative products are 

 shed into the coelom and pass into the ureter through genital pores. In Polypterus 

 similar genital pores opening into the ureter upon a papilla are very early formed in 

 male and female. From this condition, by the formation of first a groove, then a canal 

 opening into the body-cavity anteriorly, and finally a canal continuous with the genital 

 gland, the genital products were conducted outwards. Into this canal there opened a 

 number of nephrostomes, which in the male of all forms except rolypterus and the 

 Teleosteans took up the spermatozoa and became the vasa efferentia (text-fig. 5). 



It is true that there is very little ontogenetic evidence of this mode of the formation 

 of the vasa efferentia. But the fact that in all forms where nephrostomes leading from 

 the ccelom to the kidney persist they are absent in the region of the testicular network, 

 would iit in with the view that the nephrostomes in this region have been used to 

 form the vasa efferentia. 



Jungersen in his work on the genital ducts of Teleosteans ^ cites evidence brought 



^ Loc. cif. 



