MUSEUM OF COMPARATIVE ZOOLOGY. 307 



ber, from which a portion of the nephrostomes ("inner" nephrostomes) 

 emerge. Each nephrostomal canal, however near the nephrostomal end, 

 is joined by a branch which communicates with the pei-manent body 

 cavity by means of an "outer" nephrostome. According to Semon, the 

 pronephric chamber, as well as the cavity of a mesonephric Malpighiau 

 capsule, is a diverticulum of the ccelom ; and the nephrostomal canal 

 which joins the glomerular portion of a mesonephric tubule with the 

 body cavity is represented by those canals of the pronephros which 

 emerge from the open body cavity. The mesonephros is to be regarded 

 as a " genei'ation " of excretory tubules younger than the pronephros, 

 and the latter may be conceived to have primitively extended thi'oughout 

 the entire trunk. In many features Semou's view is similar to that ex- 

 pressed in the preceding pages. The point of diflerence which I would 

 here emphasize is the diiferent way in which the nephrostomal canal of the 

 mesonephros is explained. According to my opinion, this canal is a rem- 

 nant of the communication hetiveen the x>rotovertebral cavity and the secondary 

 body cavity, and is not represented in ilie tuhidar portion of the proneph- 

 ros. Semon, on the other hand, claims that it is the homologue of the 

 outer series of nephrostomal canals in the pronephros of Ichthyophis. 

 Considering the relations of the glands in that form alone, this view 

 seems well justified; but it neglects the significant relation which has 

 recently been shown to exist between the mesonephros and the com- 

 municating canal ; and I am of opinion that the view as applied to other 

 Vertebrates is untenable, unless it can be shown that the outer nephro- 

 stomal tubule of the Gymnopliionian pronephros also develops from that 

 canal. The latter interpretation is, I must admit, at least possible ; 

 but we must await further researches on the development of these 

 Amphibia before accepting such a conclusion. 



The closing section of this discussion will be devoted to a consider- 

 ation of the evidence which the development of the excretory system as 

 a whole throws on the origin of Vertebrates. 



Two methods of investigation, which are mutually dependent, yet 

 quite unlike in their application, may be employed in attempting to 

 draw phylogenetic conclusions. One of these methods is peculiar to 

 embryological research ; it is dependent upon the principle that on- 

 togeny is in part an abbreviated recapitulation of phylogeny ; its 

 method is to eliminate coenogenetic characters ; it accomplishes this 

 largely by the aid of a physiological estimate of the influences of lar- 

 val and embryonic environment, and it is comparative only throughout 



