300 BULLETIN OF THE 



as is shown by the relations of the glomus and the glomeruli respectivel}- 

 to the aorta, and by the existence of transitional glomeruli (Birds, Croc- 

 odilia, Chelonia). On the other hand, the features in which the glomus 

 differs from the glomerulus may be summarized as follows : the glomus 

 lies in the body cavity, instead of projecting into the lumen of a spe- 

 .cialized excretory tubule, and it is a continuous structure, instead of 

 consisting of a number of separate parts. 



Turning now to the tubular portions of the two glands, one can recog- 

 nize a number of common characters. In both can be distinguished a 

 longitudinal conduit and transverse canals, the latter communicating with 

 the body cavity by means of ciliated nephrostomes. The longitudinal 

 canal of the two glands is in reality a continuous structure, the segmental 

 duct. Since the pi'onephric and the mesonephric tubules are similarly 

 related to this continuous duct, it is evident that they must themselves 

 lie in approximately equivalent regions of the body. The metamerism 

 of both glands primitively corresponds to that of the body somites; this 

 feature is apparent from my account of the Amphibian pronephros, and 

 has been proved for the most anterior mesonephric tubules in Ambly- 

 stoma (see page 261), as well as for the entire series in Selachii and cer- 

 tain other groups. Finally, the cardinal veins give rise to a mesh work of 

 vascular spaces which bathe in a like manner the tubules of the proneph- 

 ros and mesonephros. In addition to the different ways in which their 

 tubules are related to glomerular structures, the pronephros and meso- 

 nephros are imlike, in that the tubules of the former develop in continuity 

 with the duct, while those of the latter join the duct secondarily. The 

 character of the convolution also is different in the two glands. As is 

 evident from the reconstructious (PlatesJV. and VIII.) of the pronephros 

 in Eana and Amblystoma, the complication is here mainly due to the 

 convolution of the longitudinal canal (common trunk) ; whereas in the 

 case of the mesonephros the longitudinal canal (segmental duct) traverses 

 the gland as an almost straight duct, the transverse tubules alone being 

 highly convoluted. 



Tlie pronephros and mesonephros, then, present many striking ana- 

 tomical features of resemblance, but also ditier in several respects, I 

 am however of opinion, that the similarities of structure are sufficiently 

 great to make it probable that pronephros and mesonephros have de- 

 velo])ed from a common beginning. I do not think, however, that such 

 tabulation of the resemblances and differences gives an adequate insight 

 into the true relationships of the structures. In the search for ances- 

 tral characters, it is a matter of indifference whether the organ in ques- 



