The average growth curves on these various diets were 
practically identical and the average duration of life varied by 
only four days. Table II briefly summarizes the results of 
this series of experiments. 
TABLE II. 
Summary of average results on diets 94, 102, 103, 104, 105. 
Number Average Average Average 
of initial weight age at 
Diet rats weight at death death - 
D4 nbc aack 12 47 gms. 34 gms. 61 days 
102), wcceeeiss 8 46 “ 31 4 65 “ 
1 eee 8 47“ 34 “ 64“ 
lOS sccwvascas 8 47 “ 34 “ 63 “ 
O04 ieee: 8 46 “ 33“ 62 “ 
It was thought that the results might be different if vita- 
min B were added to the diets, in other words if this vitamin 
was not the limiting factor in the diet. At about this time in 
the course of the work, it was also thought wise, merely as a 
factor of safety, to replace 2 per cent. of the butter fat in the 
diet by cod liver oil. Each rat in this comparison received 
either diet 107, 108, 109, or 110 with an addition of 7 cc. or 
8 cc..of reconstructed skimmed milk. These doses were used, 
since from experiments reported later in the paper it became 
apparent that one or the other of these amounts would prob- 
ably give the best results in the study of the destruction of 
water-soluble B by heat. 8 cc. was later decided upon 
as the best dose but the results for both are here given, both 
being valuable in this connection. 
TABLE III. 
The following diets were used in the second comparison: 
IPE eecoaeetucants be aaeraieen 107 108 109 110 
% % % % 
Casein apis dade dkeae, 18 18 23 18 
Buttér fat 3; scakeas aan hes 8 13 8 8 
Cod liver oil ............000- 2 2 2 2 
Starch saciseenes ae cone agers 68 63 63 67 
Salt mixture (O. & M.)*..... 4 4 4 5 
