242 FROM FISH TO PHILOSOPHER 



in their marine ancestors. As Prosser et al. {54} sum up 

 the situation for the invertebrates, 'Osmotic regulation 

 . . . comprises a group of very labile characters. Evolu- 

 tion of [the control of] osmotic function has proceeded 

 in many directions by many small changes.' 



The osmotic relations of the invertebrates are sum- 

 marized by Prosser, et al. {54}, and recent data have 

 been added by Robertson {58}. 



Reinterpretation of the structure and origin of the 

 vertebrate kidney vv^as foreshadowed by the vmter in 

 1930 {46}, as a result of studies on the regulation of 

 the composition of the blood in the elasmobranch and 

 teleost fishes. These studies indicated that the glomeru- 

 lar, fresh-v^^ater fish kidney is the older and primitive 

 type, and that the aglomerular kidney of the marine 

 fishes is a secondary specialization related to marine life. 

 Pursuing the implications of this view, Marshall and the 

 writer {37} presented a detailed re-examination of the 

 structure of the kidney in the fishes. Amphibia, reptiles, 

 birds, and mammals, and concluded that the glomerulus 

 was evolved as an adaptation to the fresh-water habitat 

 of the early vertebrates. The story of the evolution of 

 the structure and function of the vertebrate kidney then 

 began to unfold from the complementary pages of anat- 

 omy, physiology, and paleontology— a story that con- 

 tained many unanticipated impHcations with, respect to 

 vertebrate evolution as a whole. 



Nothing like the vertebrate glomerulus is found in 

 the excretory organ of any invertebrate, the nearest ap- 

 proach being the nephridium of the earthworm, Lum- 

 hricus {55}. To return to the ancestry of the vertebrates 

 and the artificial 'phylum' Chordata, no one has demon- 

 strated any specialized excretory organ in the Hemi- 

 chordata (balanoglossids) {49, p. 15}, in which excre- 

 tion apparently is a generalized fvinction of the gut and 

 skin; in the Urochordata (ascidians or tunicates), excre- 

 tion may be carried out by a mass of clear-walled cells 



