II42 PROTEIN METABOLISM [pt. iii 



mammals. It is the conditions under which the embryo has to 

 live that govern what form of nitrogen shall be excreted throughout 

 the life-span. As Table 163 shows, the fishes and mammals (even 

 the whale) with their urea and the birds with their uric acid fit in 

 with the theory here propounded. The insects also are in perfect 

 accord, for, although we have no quantitative data concerning the 

 nitrogen partition of their urine, yet it has been generally recognised 

 for many years that uric acid is the most prominent constituent of 

 their excreta (see von Fiirth) , and there is doubt if urea has ever 

 even been shown to be present. A quite parallel case is that of the 

 hymenoptera and diptera, which excrete uric acid during metamor- 

 phosis into their fat bodies, according to Fabre; Schmieder and other 

 entomologists. Thus the insects, coming to live on land earher than 

 the reptiles, had the same chemico-embryological problem to face, 

 and solved it in the same way. This is a good biochemical instance 

 of the phenomenon known to zoologists as "convergence". It is 

 interesting that ovoviviparity occurs among the insects, but never 

 true viviparity (Holmgren; KeiUn), and one would like to know why 

 they never invented the placenta, and dropped their uricotelic 

 qualities, for they did invent the "private pond", or amnion. The 

 reptiles themselves form an interesting group in Table 163, for the 

 chelonia, standing as they do phylogenetically near the amphibia, 

 show a high ammonia and urea excretion, while the sauria with 

 their terrestrial embryos show a high uric acid excretion. It is, of 

 course, among the reptilian group that such a diflference would be 

 likely to show itself, as they were the first vertebrates to conquer the 

 land. In this connection it may be recalled that Tomita found urea 

 in the imperfectly cleidoic egg of a marine turtle. Clementi, again, 

 has found arginase in chelonian livers, an enzyme which does not 

 occur in the livers of uricotelic animals. 



The eggs of aquatic animals seem to divide into two classes. The 

 frog and the plaice develop within membranes which readily allow 

 the nitrogenous end products to escape; the trout and Ascaris 

 (Kozmina) do not. But hatching always occurs long before the end 

 of development in these aquatic forms, so that the permeability of 

 the egg-membranes is unimportant from the point of view of nitrogen 

 excretion. The trout at hatching gets rid of what has accumulated, 

 and for the rest of its embryonic life can excrete directly into the 

 water. The elasmobranchs, which excrete their urea into their yolk 



