The effect of heating, drying, and storing upon vitamins is of serious import in connection with the methods used in preparing foods. It is generally conceded that fresh foods are more valuable than those which have been preserved, and that a diet should if possible include a certain amount of uncooked food, but the basis for this belief must be investigated.

The vitamins, which are known to be highly sensitive to reagents, are the constituents of foods which would seem most likely to undergo undesirable alteration during heating and drying. Unfortunately here, as in many other questions connected with the vitamins, authorities are by no means agreed.1

Chick and Hume and Davis have referred to the unsatisfactory character of much of the data given in the literature.1a Chick and Hume in experiments on the effect of temperature on vitamins note that the temperatures given are usually those registered by the autoclave or steamer used, no measurement being made in the interior of the substance heated. This is an important point, as the latter temperature remains for a surprisingly long time far below the former, especially if the material investigated has a low conductivity, which is true of most foodstuffs, particularly if fairly dry. Davis remarks that for comparative purposes with cooked food it is important to know how it was cooked, how long, the state of division, the amount cooked at a time, the amount of water used if boiled, the temperature if baked, and the interval between cooking and consumption. The data concerning commercial preserved food is interesting from an economic standpoint but is of little scientific value since the conditions under which is was prepared are not accurately known.

1 A summary of the facton influencing the vitamin content of foods is given by Dutcher (J. Ind. Eng. Chem. 1021, 1102).

1a Chick and Hume, Proc. Roy. Soo. (London) 90 B, 60. Davis, J. Home Econ. 12, 206, 1020.

Vitamin A was reported by the earlier investigators as very stable to heat. McCollum and Davis2 found it present in the ether extract of boiled egg yolks. Osborne and Mendel 3 stated that passing live steam through butter-fat for two and a half hours or longer did not destroy its nutritive efficiency, and McCollum4 supported their evidence by the statement that the A in butter-fat was not destroyed by heating to 100° for over an hour. A curious fact was noted by Osborne and Mendel5 when they separated butter-fat into a high and a low melting portion by fractional crystallization from alcohol. In the butter "oil" in which vitamin A is more concentrated, deterioration occurs on keeping, even in absence of light, to such an extent that within a year the potency is almost completely lost, while some of the same lot of butter-fat from which this oil had been prepared was still effective.

Drummond 6 found that low grade whale and cod liver oil which had been prepared by steam digestion of the tissues were still rich in A and concluded that the vitamin was therefore thermo-stable. In a later experiment he used a high grade whale oil which had been treated with soda to remove the free fatty acids, and subsequently clarified with fuller's earth. It was rich in A as demonstrated by experiments with rats, while a similar sample hardened by the usual process, involving exposure to hydrogen gas at 250° C. for 4 to 6 hours was found entirely deficient in A. In order to test whether destruction is brought about by reduction of some unsaturated substance or simply by high temperature the oil was heated to 250° C. for four hours. It proved destitute of A after this treatment. Next, samples of whale oil were heated to 100° and 150° for four hours respectively. In both cases destruction of A was almost complete. Eventually he concluded that the vitamin in both butter-fat and whale oil7 was readily destroyed by short exposure (one hour) to 100°. Destruction occurs, but is less rapid, when exposed to temperatures ranging from 50 to 100°. Exposure of oil to a temperature of 37° for several weeks may cause destruction of A if contact with air or oxygen is extensive.

Commenting on these results, Osborne and Mendel * remark:

2 McCollum and Davis, Proc. Soc. Exp. Biol. Med. 11, 101, 1919. 3 Osborne and Mendel, J. Biol. Chem. 20, 381, 1915.

4 McCollum, Am. J. Publ. Health 8, 191, 1918.

5 Osborne and Mendel, J. Biol. Chem. 24, 37, 1915. 6 Drummond, J. Physiol. 52, 95, 1918-19.

7 Drummond, Bioeh. J. 13, 81, 1919; Drummond and Coward, lb. 14, 734, 1920.

8 Osborne and Mendel, J. Biol. Chem. 41, 557, 1920.

In considering Drummond's experiments one is struck by the fact that even with six per cent of unheated butter-fat in the diet his control rats grew at much less than the normal rate. This is contrary to the experience of several investigators, including ourselves, and raises a question as to the value of the untreated butter-fat or the food intake of the animals used by him. At any rate the reason for the discrepancies between us is not apparent.

In confirmation of their earlier experiments they heated dry butter-fat in an air bath at 96° for 15 hours without destroying sufficient A (if any) to render the butter-fat noticeably less efficient. They point out, however, that their butter-fat was fed in amounts which were probably well above the minimum and that a certain amount of destruction might therefore have escaped detection.

Steenbock, Boutwell and Kent 9 found that 12 per cent of butter-fat which had been heated to 100° for four hours was less efficient for the nutrition of rats than 5 per cent unheated butter-fat incorporated into a similar basal ration. It is to be noted that their experiments showed a very considerable variation in different butter-fats which had been subjected to the minimum heat necessary for separation of the fat. While 5 per cent in the diet was usually sufficient as a source of A there were numerous instances in which this amount proved unsatisfactory.