716 L. L. MILLER 
and the appearance of stoichiometric amounts of urea: 1*, the glutamine has dis- 
appeared well before the end of the perfusion. The increases in leucine-isoleucine, 
valine and alanine, seen in Fig. 17, are striking, and the maintenance or actual 
increase of glycine and serine is obvious. 
Fig. 18 reveals changes incidental to the addition of a single large dose (1.5 mmoles) 
of L-arginine to the isolated perfusion of an 18-h fasted rat liver. Here again, on 
the basis of independent chemical determinations, the arginine was completely 
cleaved to measured quantities of ureal®. At the end of the perfusion, not only are 

RLP 646-6 i RLP 646-0 + 
Fig. 15. Rat liver perfusion (RLP 646). 18-h fasted male Wistar alloxan-diabetic rat (weight, 
218 g). Urine, 44 for glucose and 4+ acetone before operation. Dose of DL-[1-'Cjleucine and 
elucose the same as in Fig. 14. Liver weight, 9.6 g. 
leucine, isoleucine and valine prominent but alanine and glutamic acid spots are 
also increased in intensity. These results call to mind the probable role of trans- 
amination from glutamic acid, glutamine and glutamic acid derived from arginine 
via ornithine in the elaboration of the non-essential amino acids whose concentra- 
tions were either enhanced or at least maintained during the course of liver perfusions 
as depicted in Figs. 15-17. 
Fig. 19 affords, for the sake of comparison, changes in plasma free amino acids 
seen after a 5-h survival of an eviscerated rat. In this particular experiment, a dose 
of glutamine had been given intravenously at the outset of the experiment. In 
contrast to the isolated liver perfusions, it is at once clear that many of the amino 
acids, with the striking exceptions of leucine, isoleucine and valine, have increased in 
References p. 721 
