MUSEUM OF COMPARATIVE ZOOLOGY. 2S3 



Mollier's researches and were written out before his paper came into my 

 hands, it seems to me that my confirmation of his position affords ex- 

 cellent evidence of the correctness of the view advocated. In one feature 

 alone our accounts of the earliest condition of the pronephros would seem 

 to differ widely, but I am confident that the difference is apparent rather 

 than real. Mollier states that each of the diverticula which form the 

 first indications of the nephrostomal canals emerges from a protovertehral 

 cavity. This statement, as I have already shown, does not in my opinion 

 accurately represent the actual conditions. In the stage under consid- 

 eration, the dorsal portion of the mesoderm is in the anterior region 

 divided by transverse planes into a series of metameric blocks ; the pro- 

 nephric thickening also is made up of metameric constituents, and is 

 continuous dorsally in Amblystoma with two, in Rana and Bufo with 

 three, of the blocks of mesoderm. As yet no definite line can be drawn 

 between the protovertebree and the lateral plates ; in a slightly older 

 embryo, however, the protovertebra? begin to be constricted off from the 

 lateral plates, and it is at once evident that the pronephric tubules have 

 to do with the ventral segment of the mesoderm. This difference in our 

 accounts seems to me then very trivial, and my only excuse for dwelling 

 upon it is the circumstance that Rlickert and Mollier seem to attach great 

 morphological significance to this feature of their-- account. This relation 

 to the protovertebra; seems to me quite untenable. 



Previous authors have been singularly reticent respecting the exact 

 position of the pronephros with reference to the body somites. Fiir- 

 bringer ('78 a , p. 5) states that the pronephros of Anura extends over 

 three, that of Urodela over two somites ; but I have looked in vain for a 

 statement which should show whether the nephrostomes are segmental or 

 intersegmental in position. Kellogg states that each nephrostome occurs 

 opposite the middle of a proto vertebrae. Marshall and Bles confirm this 

 statement, and contend that, in the case of liana, the nephrostomes lie 

 in the 2d, 3d, and 4th somites behind the auditory vesicle. According 

 to Mollier ('90, p. 213) the pronephros appears in Triton in the region 

 of the 1st and 2d trunk protovertebra? ; but since the most anterior two 

 protovertebra; are reckoned to the posterior region of the head, these 

 represent the 3d and 4th protovertebroe of the series. The enumeration 

 which I have given for Rana and Bufo is in precise agreement with that 

 of Marshall and Bles. For Amblystoma my account is in agreement 

 with that of Mollier for Triton. 



I am not aware that any definite attempt has thus far been made to 

 ascertain which of the three nephrostomes of Anura is unrepresented in 



