312 BULLETIN OF THE 



seems to me entirely in harmony with pliysiological requirements ; and 

 in this earliest fragment of the excretory system we have, in my opinion, 

 a remnant of the primitive mode of formation of the segmental duct. 



The question at once arises whether there is any indication of this 

 mode of oi'igin preserved in the development of the posterior portion of 

 the duct. A free backward growth, such as is maintained for many 

 Vertebrates, is evidently far removed from the primitive mode of forma- 

 tion, and is to be regarded as an adaptation to the needs of the proneph- 

 ros. The origin of the duct in situ from a somatopleural proliferation 

 is without doubt a modified condition ; yet it suggests a mode of origin 

 which is in agreement with that observed in the anterior region. I have 

 already emphasized the circumstance that in Amphibia the duct arises 

 from a mass of cells which is perfectly continuous with that from which 

 the pronephric tubules are difterentiated ; and it is possible that both 

 regions represent disguised nephridial evaginations of which those in the 

 posterior region are never ditferentiated as actual canals except in such 

 portions as are converted into the duct. Further evidence in favor of 

 this view is afforded by the occasional occurrence of supernumerary pro- 

 nephric tubules such as have been observed by Mollier ('90, p. 224) and 

 myself (page 253). The acceptance of this interpretation would necessi- 

 tate a modification of our conception of the relations between pronephros 

 and mesonephi'os, since we should be obliged to regard the mesonephric 

 tubules as a second generation of tubules, the first generation having 

 been employed in giving rise to the duct. On the other hand, it is quite 

 possible that the entire backward growth of the duct is a wholly secon- 

 dary process to meet the needs of a prematurely developed portion of the 

 primitive excretory organ. This is the only interpretation which seems 

 admissible in those cases where the duct has been found to grow back- 

 ward free from adjacent tissue. 



The conception of the phylogeny of the duct which I have just pre- 

 sented offers a partial explanation of the contradictory evidence which 

 has been advanced respecting the germ layer from which the duct arises, 

 With a narrower conception of the phylogeny of the duct, it is difficult 

 to understand why the ectoderm should participate in the formation of 

 the excretory system in one group, but not in another, and why the 

 posterior end of the duct should in some cases be formed at the expense 

 of a germinal layer different from that which gives rise to its anterior 

 portion and to the nephrostomal canals wherever they appear. If, how- 

 ever, we assume the existence of a phylogenetic stage in which a series 

 of nephridia open directly to the exterior, it is at once evident that a 



