MUSEUM OF COMPARATIVE ZOOLOGY. 311 



ticula primitively terminated, whether they opened on the surface or 

 joined a longitudinal duct, it will be necessary to consider the pronephros 

 alone, since the segmental duct is already present before the mesonephros 

 is formed, and we cannot expect to find an adequate criterion for deter- 

 mining whether the union of the mesonephric tubes with the duct be 

 primitive or secondary. In the pronephros there is in most cases no 

 evidence of a mode of termination more primitive than that of com- 

 municating with a duct. Two arguments, however, occur to me, which 

 seem to indicate that a series of direct outlets to the exterior may have 

 been early present. In the first place, the pronephric diverticula have 

 frequently been observed to enter into intimate union with the ectoderm. 

 Thus Riickert ('88, p. 217) was led to believe that the pronephric thick- 

 ening of Selachians even received a contribution of cells from the outer 

 germ layer. The most natural explanation of this condition seems to me 

 to be, that the fusion of the diverticula with the ectoderm is the re- 

 capitulation in the ontogeny of a phylogenetic stage, which possessed 

 nephi'idia provided with direct openings to the exterior. Secondly, 

 Amphioxus, according to the most recent investigations, is provided with 

 a series of nephridia opening into the atrial chamber, which latter we 

 are, in my opinion, justified in regarding as a simple infolded portion of 

 the exterior. Accepting the homology of the nephridia of Amphioxus 

 and those of Craniotes, it seems to me probable that the ancestors of 

 Vertebrates possessed nephiudia which resembled those of Amphioxus in 

 opening directly to the exterior. 



If separate diverticula leading from the coelom to the exterior be the 

 primitive condition of the Vertebrate excretory organs, we have still to 

 seek the origin of the segmental duct. On this point, the pronephros 

 alone can afford evidence. The participation of the ectoderm maintained 

 by many authors for the posterior end of the duct affords the suggestion 

 that it may have first been formed as a groove of that layer, or that a 

 primitive anterior opening was gradually shifted back to the cloaca. It 

 may be objected to this view, (1) that in many Vertebrates no participa- 

 tion of the ectoderm occurs, while in none has it been shown that the 

 mesoderm does not play a part in the formation of both anterior and 

 posterior portions of the duct; and (2) that the longitudinal canal of the 

 pronephros, which forms the anterior prolongation of the duct, in no case 

 arises in this way. In the pronephros the longitudinal canal arises, as 

 testified by a large number of recent investigators for divers groups, and 

 as confirmed by my own observations on Amphibia, by the fusion of the 

 distal ends of the pronephric diverticula. This mode of development 



