Nothing is yet known concerning the chemical structure, and very little concerning the properties of the fat-soluble vitamin. Most of the observations that have been made on it have to do with its solubility.1 McCollum and Davisla found that ether extracted this substance from egg-yolk and butter-fat, but McCollum and Simmonds 2 failed to extract it from plants by such solvents as ether, chloroform, benzene, and acetone, and therefore concluded that it was not found in any fats or oils of plant origin. McCollum, Simmonds and Pitz 3 state that the residue from corn meal after extraction with ether is more effective in causing a slow but long-maintained upward trend of the curve of growth than is corn oil. Ether extraction of plant tissue therefore does not necessarily remove the vitamin. They suggest that the fat-soluble A is in chemical union in the plant tissues, in some complex which is not soluble in fat or in ether. In digestion and absorption it is set free, and being readily soluble in fats thereafter accompanies the fats in the animal body. Owing to the large content of waxes, etc., extracted from plant leaves they have not been very successful in feeding ether extracts from these sources.

Osborne and Mendel 4 however, obtained a very effective extract from spinach and clover with ether. Steenbock and Boutwell 5 found that there was great variation in different substances as to the ease with which the fat-soluble vitamin was yielded to solvents. While ether in general was rather a poor solvent, it was nevertheless possible to obtain a fairly satisfactory ether extract of alfalfa but not of carrots. Chloroform and carbon disulphide remove some of the fat-soluble vitamin as found in carrots, while alcohol and benzene remove considerable amounts of it. The results were not entirely satisfactory, however, as the extracted carrot residues were almost as effective as the original carrots when fed in small amounts as the sole source of A. Evidently while carrots contain considerable amounts of A, for some reason or other no satisfactory quantities of the vitamin could be accumulated from them. These investigators suggest that this may be on account of the presence of the colloidal pectins and hemicel-luloses or possibly on account of the occurrence of the vitamin in molecular combinations. Fat solvents applied to alfalfa gave far more satisfactory results. Alcohol was found to be the most satisfactory solvent, but benzene was also effective. Water proved very poor.

With corn, ether failed to extract the vitamin, but alcohol,4 cold or hot, removed it quantitatively. Although dried tomatoes contain considerable amounts of the fat-soluble vitamin, the ether extract had no growth-promoting power.7

1 LaMer considers vitamins from the standpoint of physical chemistry. (J. Ind. Eng. Chem. 1921, 1106).

1a McCollum and Davis, J. Biol. Chem. 1913, 15, 167.

2 McCollum and Simmonds, Ibid, 32t 33, 1917.

3 McCollum, Simmonds and Pitz, Am. J. Physiol. 41, 361, 1916.

4 Osborne and Mendel, Proc. Soc. Exp. Bi. Med. 16, 98, 1918-19.

5 Steenbock and Boutwell, J. Biol. Chem. 4a, 131, 1990.

Drummond 8 found that A is not extracted from oils by water or dilute acid but that it may be removed in small quantity by cold extraction with alcohol. He could find no evidence which suggests that A is not a single substance, but was unable to identify it with any of the recognised components of fats, such as glycerol, saturated or unsaturated fatty acids, cholesterol, lecithin, phosphatides, or lipo-chromes. He suggests that it may be a labile substance of ill defined construction.

According to Drummond,9 saponification in non-aqueous media of oils containing the accessory factor results in the disappearance of the growth-stimulating properties, although an earlier experiment10 had shown that a high grade whale oil which had been treated with soda to remove the free fatty acids and subsequently clarified with fuller's earth was still rich in A.

Steenbock and Boutwell 11 found that A as extracted from alfalfa with alcohol resists saponification with alcoholic potassium hydroxide in the cold. From such solution on dilution with water it can be extracted with ether. On fractionation by differential solubility there can be obtained a portion soluble in petroleum ether, containing A in large amounts, along with carotin, and a dilute alcohol-soluble portion which contains xanthophyll, but little or no A.12 This is in accordance with the observation of McCollum and Davis13 that the property of inducing growth in rats which have grown as far as possible on a fat-free ration can be conferred on olive oil by shaking the latter with a solution of the soaps prepared by completely saponifying butter-fat in a non-aqueous medium with potassium hydroxide. According to McCollum and Davis, the results obtained leave no room for doubt as to the stimulating effect of this oil as compared with the untreated oil.

6 See also McCollum, Simmonds and Pitz, J. Biol. Chem. 28, 153, 1916. 7 Osborne and Mendel, J. Biol. Chem. 41, 549, 1920. 8 Drummond, Bioch. J. I3, 81, 1919. 9 Drummond, Bioch. J. 13, 81, 1919.

10 Drummond, J. Physiol. 52, 344, 1918.

11 Steenbock and Boutwell, J. Biol. Chem. 42, 131, 1920. 12 Steenbock and Boutwell, 1. c.

13 McCollum and Davis, J. Biol. Chem. I9, 245, 1914.

More recently, Steenbock, Sell and Buell 13a have investigated the stability of A as found in animal fats, and state that no appreciable destruction of the vitamin could be observed after boiling the fat for 1/2 hour with 20 per cent alcoholic KOH.

In recent years a large amount of vegetable oil has been treated by the hydrogenation process to form solid fats known as hardened oils. The process as usually carried out involves the treatment of the oil with hydrogen gas in the presence of a nickel catalyzer. The temperatures generally employed range from 150° to 190° C. The time of treatment with hydrogen for the hardening operation ranges from one to eight hours.14 In view of the effect of high temperatures on vitamins it is to be expected that their activity would decrease during the hardening process. This is apparently the case as is indicated by the investigations of Halliburton, Paton, Drummond and others15 who found fat-soluble vitamins to be absent in oils treated by the hydrogenation process.