TECHNICAL NOTES AND BIBLIOGRAPHY 237 



physiologist, I cannot see how the comparative anatomy 

 of the kidney, the urea-retention habitus of the elasmo- 

 branchs, and related matters can be reconciled with the 

 belief that the vertebrates acquired their definitive ana- 

 tomical and biochemical pattern in salt water, and my 

 preference for the time being therefore remains with 

 the fresh-water hypothesis. 



In identifying the Chamian with the Grand Canyon 

 disturbance in North America, we are following Schu- 

 chert and Dunbar {17, p. 90 f}. 



The hypothetical structure of the protovertebrate 

 given here follows closely that visualized by Chamber- 

 lain {30}. 



Where the invertebrates had armored themselves with 

 shells made of chitin or calcium carbonate, the ostra- 

 coderm armor was composed of calcium phosphate, laid 

 down by fibrous tissue and around vascular channels in 

 concentric layers. 



A. S. Romer, who proposed the eurypterid theory of 

 the origin of this armor {40, 42}, accepts fresh water as 

 the habitat of the ostracoderms, and admits that the 

 glomerular structure of the vertebrate kidney is a cogent 

 argument in favor of the fresh-water origin of the verte- 

 brates as a phylum, but he dismisses the fresh-water ori- 

 gin of armor on the grounds that, 'As any section of the 

 plates and scales of the ostracoderms will show, the thick 

 and relatively impervious part of the armor lay beneath 

 a superficial porous region close to the surface in which 

 there was a rich blood supply— a type of construction 

 which would destroy most of the value of the armor 

 under such a theory.' {42, p. 46.} But as Bryant {29} 

 describes the armor of Astraspis, it consisted of small, 

 tuberculated plates, each of which contained three zones: 

 an inner, very thick, laminated zone penetrated by verti- 

 cal vascular canals; a middle, dense zone in which the 

 vertical extensions of the underlying vascular canals, 

 now comparatively few in number, were surrounded by 



