682 
Journal of Agricultural Research 
Vol. V, No. is 
was omitted from the average. Close duplicates on leucin and on the 
next control substance, casein, were obtained easily and by the identical 
methods that failed to produce as close duplicates on hydrolyzed meat. 
The casein (casein sample i) hydrolyzed for the control determinations 
was prepared in the laboratory in the usual manner, from separator 
skim milk. The dry protein contained 14.87 per cent of nitrogen and 
0.10 per cent of ash. Although small amounts of impurities were prob¬ 
ably present in this preparation, it compared favorably with those used 
by other workers. The hydrolysis of the casein, distillation of ammonia, 
and determination of amino nitrogen were carried out exactly as with 
hydrolyzed meat, except that 5 gm. of the dry casein were used instead 
of 25 gm. of meat. The following results were obtained: 71.37, 72.81 
per cent (boiled for 24 hours), and 71.37, 71.73 per cent (boiled for 48 
hours). The average of these four, 71.8 per cent, given in Table V, is very 
close to the figure (72.1 per cent) obtained by Van Slyke (1912, p. 297) 
and other investigators. The various determinations made with casein 
sample 1 indicate that the methods used were essentially correct and 
would yield close duplicates on materials to which they were applicable. 
It was thought possible that the fats or their hydrolytic products mi ght 
interfere with the amino-nitrogen determination, and for this reason 
determinations were made on beef and veal samples 1.1. These were 
dry, almost fat-free meat powders, prepared early in the work from 
beef and veal samples 1 by treating the hashed meats with alcohol and 
ether (see p. 685). The hydrolysis and determinations were made on 
these materials as usual, but no better duplicates were obtained. The 
figures for beef sample 1.1 were 74.1, 75.3, and 75.7 per cent; average, 75 
per cent; for veal sample 1.1,74.1, 74.1, 75.7, and 76.3 per cent; average, 
75.1 per cent. The difference between the highest and the lowest figure 
for veal sample 1.1, 2.2 per cent, corresponds to a difference of 0.6 c. c. of 
nitrogen gas under the conditions of the determinations, in which the 
volume of gas actually measured was about 20 c. c. 
A sample of tyrosin labeled “Tyrosin, pure, synthetic, Schuchardt” 
was also used for control determinations. It contained 1.66 per cent of 
moisture. Calculated to the dry basis the total nitrogen content by the 
Kjeldahl method was 93.5 per cent of the theoretical. The figures for 
amino nitrogen were 95.6, 96.0, and 95.3 per cent of the theoretical total; 
average, 95.6 per cent. In the first determination the gas after being 
measured was passed back into the absorption pipette, where it remained 
overnight. As usual, there was a slight diminution in volume—from 
95.6 to 95.3 per cent. 
It is believed that close duplicates on beef and veal have not been 
obtained, for reasons inherent in the material; the method used gave 
good results on comparatively pure leucin, casein, and tyrosin. The 
comparison between the amino-nitrogen content of beef and that of veal 
having been made under similar conditions, the data in Table V, although 
