SECT. 9] PROTEIN METABOLISM 1135 



the part of the embryo, for even if, as McCallum would have us 

 believe, the maternal sea water is practically pre-Cambrian, it is at 

 any rate as good as any other sea water for the disposal of nitro- 

 genous waste products, and from the embryonic point of view, a 

 boundless ocean. In other words, the continuous perfusion system of 

 the vivipara provides an artificial sea, and avoids the necessity of a 



uricotelic metabolism. Is it surprising, in view of these facts, that 

 fishes turn the ammonia from their protein breakdown into urea, birds 

 and reptiles into uric acid, and mammals once more into urea ? 



The thought may be stated in another way. Perhaps the saurop- 

 sida excrete their nitrogen mainly as uric acid because they had to 

 learn how to do so in order to pack their embryos into solid- and 



