262 BULLETIN OF THE 



glm.') occur, which constitute the earliest fundament of the glomerulus. 

 The nephrostomes, howevei", have not opened at this stage. 



In the region between pronephros and mesonephros (Plate VII. Fig. 

 54) certain masses of cells are found on the median side of the duct in 

 the same position as that occupied in the posterior region by the meso- 

 uephric tubules. These cells do not form a continuous mass, but are 

 interrupted at intervals. The cords of cells thus formed do not, how- 

 ever, appear to correspond in their arrangement to the metamerism of 

 the body. It is possible that they represent rudimentary nephridial 

 tubules, but the evidence in favor of this interpretation must be regarded 

 as far from satisfactory. 



1 have been unable to ascertain the precise mode of origin of the 

 mesonephric tubules, having sought in vain for nuclear mitoses which 

 "should throw light upon this question. There are in younger stages 

 many retroperitoneal (subperitoneal) cells which might be collected and 

 rearranged so as to produce the tubules ; or, again, the fundaments of 

 the tubules might be formed by proliferation from the peritoneum. The 

 cells of the tubule have evidently undergone very rapid division, as is 

 indicated by the complete consumption of the yolk ; and this circum- 

 stance seems to me to favor the second view. Furthermore, I have found 

 nuclear mitoses (Fig. 51) in the region immediately in front of the meso- 

 nephros which indicate that the cords of cells in this region arise from 

 the peritoneum. Although I am unable to assert that the mesonephric 

 tubules arise from the peritoneum, I am inclined to regard it as probable 

 that they do. There is no evidence, however, of a definite invagination 

 of the wall of the body cavity. 



This is the oldest stage of Amblystoma which I have examined, and 

 with it I close the descriptive part of this paper. 



III. General Discussion. 



Having presented in a purely descriptive manner the facts of develop- 

 ment as yielded by my own studies, I shall now endeavor to use these 

 observations as a basis for the criticism of the results of other investiga- 

 tors, and in closing shall point out certain general conclusions which seem 

 to me warranted by such a review. 



Eecent researches have extended greatly the number of animals in 

 which a homologue of the pronephros is known, so that it may now be 

 fairly assumed that the organ appears in the ontogeny of all Vertebrates. 



In view of much recent evidence (Hatschek, -88 b , Rabl, '88, Ayers, '90) 



