MUSEUM OF COMPARATIVE ZOOLOGY. 277 



pronephros as detailed in the preceding pages. Moreover, Felix pro- 

 duces no evidence to show that they stand in any genetic relation what- 

 ever to the Wolffian duct, or to the pronephric structures described by 

 other authors. In the present state of knowledge his interpretation 

 seems to me untenable. 



(3.) The remaining views all have the common feature that they regard 

 certain rudimentary canals in connection with the anterior end of the 

 Wolffian duct as pronephric. The views are somewhat divergent, but I 

 have been able to compile from them a general statement which will in 

 a measure explain their conflicts. In bringing the observations of each 

 author under this general scheme, I shall frequently be driven to regard 

 his results as incomplete, but I shall as far as possible avoid questioning 

 his statements from an a priori standpoint. 



In general three regions of the embryonic excretory organ may be 

 distinguished : the pronephros, an intermediate region, and the meso- 

 nephros. For criteria of these regions, I shall use in the main glomeru- 

 lar structures : those of the pronephros are glomi wholly external to the 

 tubules; those of the intermediate region are transitional glomeruli, 

 which develop in peritoneal canals, but project through the nephrostomes 

 into the body cavity ; those of the mesonephros are typical glomeruli, 

 which have only a mediate connection with the body cavity through the 

 tubule. 



It now remains to consider the results of the observers whom I have 

 placed in my thii'd gi-oup. The work of Gasser and Siemerling ('78, '79), 

 subsequently carried on by Siemerling ('82), i-elates to Birds alone. These 

 authors recognize two distinct portions of the Wolffian duct: a portion 

 lying in front of the fifth somite, and a posterior portion. The former 

 shows many irregularities, is broken up into discontinuous fragments, 

 and early atrophies ; the latter develops more slowly, but more regularly, 

 and persists as the duct of the Wolffian body. The first indications of 

 tubules consist of the so-called primary coi-ds, which are continuous with 

 the coelomic epithelium by means of fimnel-shaped ostia, while they are 

 distally in contact with the duct. Gasser and Siemerling maintain that 

 they belong to the most anterior part of the mesonephros, a portion which 

 early atrophies. They are quite similar to the S-shaped canals of Kolli- 

 ker ('79). In front of the region of the " primary cords" similar evagi- 

 nations occur, but these never reach the duct. A typical glomus, which 

 may be single or may be divided into parts, projects from the radix 

 mesenterii opposite the openings of these evaginations. In embryos 

 of this stage the space between the most anterior Wolffian tubule and 



