280 CHEMISTRY: SHERMAN, ET AL. Proc. N. A. S. 
at 110° C. in shallow trays freely exposed to air was found sufficient. 
To the mixture of oats and skimmed milk powder were added pure fresh 
butterfat and sodium chloride. The proportions adopted were: 
Ground whole oats 59 per cent 
Heated skimmed milk powder 30 per cent 
Butterfat 10 per cent 
Sodium chloride 1 per cent 
By using heated skimmed milk and fresh butterfat instead of heated 
whole milk, the absence of antiscorbutic vitamin is equally well ensured 
and the palatability of the diet and its fat-soluble vitamin content are 
improved. This diet supports excellent growth up to the time of the 
onset of scurvy. 
Experimental Animals. — Our experience with about 200 experimental 
animals indicates that young guineapigs from six to eight weeks old 
and weighing 300 to 350 grams are best used for studies of vitamin C. 
Placed upon the above ration at this age and size they usually eat about 
18 to 20 grams per day and continue to grow for about 15 days, then lose 
weight rapidly and die of scurvy in from 26 to 34 days after being con- 
fined to the diet. If much older, the animals are somewhat less suscepti- 
ble ; if much younger, the results are less regular. 
Symptoms, Survival Period, and Autopsy Findings. — The first symptoms 
appear after about 12 days on the basal diet. In our cases on the above 
basal diet only, the survival periods were, respectively, 33, 26, 27, 28, 
34, 28, 27, 34, 26, 29, 32, 32, 32, 31, 34 days. The nature of the symptoms 
and autopsy findings agreed so closely with the descriptions of Cohen and 
Mendel and of Hess as not to require further discussion here. At autopsy 
the findings which proved most significant were: looseness of teeth, 
fragility of bones, enlargements and hemorrhages of joints and of rib 
junctions. When the animal receives some antiscorbutic vitamin but 
not enough for protection from scurvy, life is prolonged, but to a rather 
uncertain extent, and some of the symptoms and autopsy findings may 
become even more severe than in the animals that receive no antiscorbutic 
and die more quickly. 
Quantitative Rating. — From what has just been said it necessarily follows 
that exclusive dependence upon either the survival period or any one set 
of symptoms or autopsy findings might easily prove misleading. All 
of these are, therefore, considered in each case. This was done with 
several series of animals, some receiving the basal ration only, and others 
measured amounts of filtered canned tomato juice up to the amount which 
gave complete protection from scurvy. By comparison with the standard 
protocols thus established, for data of which reference must be made to 
the fuller account of the work to be published elsewhere, it becomes possi- 
ble to form a quantitative estimate of the degree of protection (if any) 
