MU!6EUM OF COMPARATIVE ZOOLOGY. 2G9 



cord of cells acquires a lumen, either by a rearrangement of the cells, or 

 by destruction of the axial ones. Anteriorly the structure now opens 

 into the body cavity. The anterior portion elongates and becomes more 

 and more convoluted up to the time of "post-embryonic" development. 

 Opposite each of the peritoneal funnels are formed glomeruli [glomi] as 

 processes from the radix mesenterii. They are covered by a pigmented 

 layer of peritoneum. Salensky does not seem to me to have been very 

 clear upon the earliest development, which was studied mainly by surface 

 views, and I am of opinion that these stages would show very different 

 conditions if more recent technical methods were employed. The most 

 interesting feature of the development, as described by Salensky, is the 

 occurrence of a glomus in the position which is typical for Amphibia alid 

 Petromyzon. 



The excretory system has probably been studie<l more carefully in 

 Selachii than in any other group. The independent researches of Bal- 

 four ('75 and '78) and Semper ('74 and '75) are in substantial accord, 

 and have formed the basis for all subsequent investigations. For our 

 purpose, the most prominent feature of the development as described 

 by these authors is the absence of any structure which demonstrably 

 r^resents the pronephros. According to Balfour, the first trace of the 

 excretory system appears as a solid knob springing from the ''interme- 

 diate cell mass " near the level of the hind end of the heart. From this 

 anterior proliferation a solid cord of cells grows backward between ecto- 

 derm and mesoderm. The posterior portion is the fundament of tlie 

 segmental duct ; the anterior knob persists in adult females as the 

 ostium abdominale of the oviduct. According to Balfour, this solid 

 knob represents a rudimentary pronephros. 



Very recently the early development of the excretory organs has been 

 placed in a new light by the researches of RUckert ('88) and van "\^'ijhe 

 ('89). According to Riickert, the development begins with the forma- 

 tion of a pronephros as an outgrowth towards the ectoderm from the 

 ventral portions of several -protovertehroe, extending from the third or 

 fourth trunk somite backwards for a distance of four to six somites. 

 The thickening extends venti-ally in each somite to the region where the 

 segmented mesoderm passes into the unsegmented lateral plates. The 

 proliferation, in the formation of which the somatic layer is alone con- 

 cerned, shows on careful study a metameric character. From the pos- 

 terior end of each protovertebra a narrow canal can be traced outwards 

 and backwards, where it unites with a similar canal emerging from the 

 next following somite. The pronephric mass fuses for a time with the 



