MUSEUM OF COMPARATIVE ZOOLOGY. 313 



very trifling difference of location would determine whether the longi- 

 tudinal canal, by means of which the duct arises, should develop from 

 the mesoderm or from the ectoderm. It is to be remarked, however, 

 that such an explanation is not wholly satisfactory, since one would 

 expect on this hypothesis tha«t those forms in which the ectodermal 

 origin of the duct seems well attested would show evidence of close 

 genetic relationship, while those classes in which the duct arises from 

 the mesoderm ought to form an equally well defined group. This 

 condition, however, is by no means realized. On the other hand, the 

 force of this objection is materially weakened if we regard the duct as a 

 recent acquisition, which its absence in Amphioxus gives some justifica- 

 tion for assuming. The explanation seems to me, nevertheless, in a 

 measure unsatisfactory, and I have adduced it merely as a possible solu- 

 tion of the problem to which the apparently diverse relations of the duct 

 to the germ layers gives rise. 



An intimate relation is always very early established between the 

 excretory tubules and the cardinal veins. Such an arrangement is so 

 favorable for the process of secretion that there can be but little doubt 

 that this condition prevailed in the ancestors of all Vertebrates. There 

 does not appear to be any evidence which would indicate whether the 

 cardinal veins or the excretory tubules are the more primitive structures. 



In addition to the means of excretion afforded by the epithelial walls 

 of the tubules, the Vertebrate kidney-organs possess peculiar glomerular 

 structures. These, as I have already shown, are all formed on the type 

 of the pronephric glomus. In their primitive condition, they consist of 

 vascular tufts, which receive blood from the aorta and project into the 

 body cavity from the root of the mesentery. 1 The origin of such a prim- 

 itive glomerular structure is not far to seek. It is readily conceivable 

 that fluid may at first have simply exuded from the aorta, and, travers- 

 ing the small amount of tissue intervening between it and the body 

 cavity, may have reached the orifices of the excretory tubes prior to 

 the development of any specialized organ subserving a glomerular func- 

 tion. This process being once established, any modification of structure 

 which should allow a portion of the aortic current to be brought into 

 closer relations with the excretory tubules would be of obvious utility, 

 and would be preserved. 



The excretory system thus constituted would represent the proneph- 



1 The view of the excretory system here presented explains the double blood 

 supply of the kidneys of lower Vertebrates, and also the circumstance that the 

 Malpighian bodies always receive their blood by a direct branch from the aorta. 



