Little or no regard has been paid to this reaction; the textbooks on children's diseases give this point no consideration whatsoever, expressing the salt content merely in terms of total ash. By this means ideal conditions are furnished for the destruction of a large part of the vitamine, and if this factor is not present in excess, scurvy will readily develop. It is to these various circumstances--the deficiency of milk, the alkalinity and the heating--that the scorbutic quality of proprietary foods is to be ascribed. The most flagrant example of an infant's food leading to scurvy is "malt soup"--an alkaline preparation of malt, which is prepared with flour and a small amount of milk.[16] This food brings about scurvy almost invariably unless an antiscorbutic is added to the dietary. That this effect is due to the alkaline potassium carbonate was shown by a recent test which is illustrated in chart. (Fig. 2). Here we see that in a case of scurvy, when the food was altered and prepared with the same amount of milk and alkali, the disorder did not abate (although the carbohydrate had been discontinued), but when the potassium carbonate was omitted a gain in weight ensued, and, we may add, the symptoms disappeared. [16] Malt soup is prepared by dissolving the alkaline malt soup extract in water, and mixing wheat flour with milk. These two mixtures are then boiled together for about three minutes. According to some directions, 11 ounces of milk are used; in our formula 16 ounces were employed. [Illustration: FIG. 2.--Human scurvy: weight curve of a baby that developed scurvy on a diet of malt soup (period 1). During period 2 the flour and malt soup were omitted from the diet, the same amount of milk and potassium carbonate being continued. The disorder did not abate. The only change in period 3 was a discontinuation of the potassium carbonate. This brought about a gain in weight and cure, showing the destructive effect of the alkali on the vitamine.] A test of this kind once more raises the question whether carbohydrates lead to the development of scurvy. As previously stated, there are many who believe that the carbohydrates, especially starch, have this harmful effect by requiring a large amount of the various vitamines for their metabolism. This relationship was first brought forward by Bradden and Cooper in regard to beriberi. It is a suggestion which cannot be disregarded in view of the great amount of scurvy which comes about on diets rich in carbohydrates, especially those containing malt sugar. We do not believe, however, that the harmful effect which this foodstuff exerts can be explained on the above hypothesis, as experience has shown that infants may receive for many months equally large amounts of carbohydrates--cane sugar, flour, or a combination of both--and nevertheless not manifest a similar tendency to develop scurvy.