Table IV gives a summary of the results of the compari- 
son of diet 107 with diet 108 containing 5 per cent. more but- 
ter fat, diet 109 containing 5 per cent. more casein and diet 110 
with an excess of 1 per cent. salt mixture over that in diet 107. 
TABLE IV. 
ge 
& & MS) a rs) 
gue y & re 
buys «& a O = Aver. age at death Gain for 
aE ‘3 oo = re or when killed 8 weeks 
2 68 6 . Be Ses 
iam) MsEk 4 <2 QaB 
107, 0 11 47gms. 36gms. 54 days at death 
108 0 10 47 “ 36“ Born aie: At ie 
109 0 10 47 “ 36 OO ME 
110 0 9 47 * 34“ Ck is 
107 7 cc. 9 48 “ 69 “ Killed at 12 weeks 21 gms. 
108 7cc. 10 47 “ 84* “ 1 dead at 64 days 37“ 
Rest killed at 12 wks. 
109 7cc. ll 49 “ 73 Killed at 12 weeks 23“ 
110 7cc. 10 46 “ 70 “ oe ITD ee 24." 
107 Bcc. 10 47 “ 82 “ ZS SP 35. 
108 Bcc. 12 48 “ 81 “ EEO CET ZY SE 33“ 
109 Bcc. 11 47 “ 82 “ eee | os 35“ 
110 8cc. 11 47 * 78 “ eas ees 31 “ 
*84 gms. is the average weight of 9 rats at 12 weeks of age. 
The results of the experiments tabulated in Table IV are 
the same as those in the previous series (Table IT). It ap- 
pears from the data that the basal rations 94 and 107 furnish 
optimum amounts of the essential food factors, protein, fat, 
mineral salts and vitamin A. 
DETERMINATION OF THE MOST ADVANTAGEOUS 
DOSE FOR MEASURING HEAT DESTRUCTION 
OF VITAMIN B IN SKIMMED MILK POWDER. 
Feeding the Milk Separately from the Rest of the Diet. It 
was next necessary to determine the amount of milk to feed in 
order to detect most readily any change in the water soluble 
B content upon heating the milk. Increasing amounts of milk 
were superimposed on the basal diet and a study made of the 
growth curves in relation to size of the dose. The milk was 
12 
