Much information of a general character as to the distribution of the vitamins has been gathered during the last decade, but our knowledge is still far from satisfactory. Only very recently has there been any attempt to make the tests quantitative and lack of standardized methods has made it impossible to compare with any degree of accuracy the results obtained by different experimenters. Steen-bock and Gross1a consider that the time is not yet ripe to warrant a general classification of foods into groups (1) rich in, (2) poor in, and (3) free from fat-soluble vitamin; as general methods of experimentation, especially in reference to vitamin content of the basal food ingredients,2 period of observation, and control of experimental animals, differ so greatly in different laboratories that the conclusions of the different observers are hardly comparable. Furthermore, there is much reason to believe that the variation in the fat-soluble vitamin content of naturally occurring food materials, even when harvested at the same stage of development, is tremendous.

No doubt, according to Steenbock and Gross, there lies here the foundation for many interesting correlations in the functional role of specific substances in both plant and animal kingdoms, the significance of which can scarcely be predicted.

The results of tests of the different foodstuffs so far as they have been obtained are collated in tables appearing in the appendix, with brief comments made by the observers, but, owing to the difficulties referred to above, no attempt has been made to compare the relative amounts present. The comparatively few researches in which more precise methods have been employed are discussed in greater detail in the text.

Fat-soluble A, as suggested by McCollum's nomenclature, is commonly associated with natural fats. The fact that it was first detected in butter-fat, and that this is a particularly convenient source, has led to what is perhaps over-emphasis upon the value of this foodstuff. While butter is commonly eaten in greater quantity than any of the other foods in which A is found, it is by no means essential as a source of this factor, since as will be seen later, certain vegetable products equal or surpass butter-fat in this respect, and on the other hand, Stccnbock, Boutwell and Kent3 have called attention to the wide variation possible in different butters and suggested a possible explanation.4

1 For tables, see Appendix.

1a Steenbock and Grots, J. Biol. Chem. 40, 503, 1910.

2 See also Drummond and Coward, Bioch. J. 14, 661, 1920.

"In the course of many experiments designed to demonstrate the vitamin content of butters we have been impressed with variations in the amount of vitamin present. These variations are not the result of heat treatment as in some cases the exposure to heat in the preparation of the fat was reduced to the minimum necessary to secure a clear sample. We surmised that possibly the rations on which the butter-fat had been produced might be responsible. On these general premises lard could be said to owe its deficiencies in vitamin to the poverty of the grains in this dietary essential. Similarly beef fat would owe its correspondingly greater vitamin content to the greater vitamin content of the roughages such as clover and alfalfa so extensively fed for beef and milk production. As yet experiments in this direction have not progressed sufficiently to warrant this generalization. One fact, however, is clear. Vitamin content of the ration on which butter-fat is produced is not the only factor to be considered. One sample of butter-fat obtained from a cow fed exclusively on alfalfa hay was found to contain no demonstrable amounts of vitamin. This butter had been kept in an unsalted condition in a poorly iced refrigerator for about three weeks.

In this time some molds had developed on its surface which necessitated discarding part of the material. . . . Questions of influence of feed, storage conditions, and temperatures used in the renovation of inferior products and even the method of use of the product in the home must be taken into consideration when depending upon butter-fat as a sole source of fat vitamin in the dietary".

It is an interesting fact that butters rich in pigments usually contain more vitamins than those less highly colored.5 According to Stephanson,6 however, extraction of the coloring matter of butter by treatment with wood charcoal does not affect the vitamin content.

Beef fat contains some A, the more liquid portion (oleo oil) having the highest concentration. This accounts for the presence of vitamins in oleomargarines made from animal fats, some of which have been reported by Halliburton and Drummond7 to be equivalent to butter fat in this respect. Oleomargarines from vegetable fats on the other hand have been found to be poor, if not entirely lacking in A.

3 Steenbock, Boutwell and Kent, J. Biol. Chem. 35, 519, 1918.

4 See also Dutcher, Kennedy and Ecklei, Sci. 52, 588, 1920, and Drummond and Coward, Bioch. J. 14, 668, 1920, on the influence of the diet of cows upon the fat-soluble and water-soluble vitamins in cow's milk.

5 Steenbock, Sell and Boell, J. Biol. Chem. 47, 89, 1921.

6 Stephenson, Bioch. J. 14, 715, 1920.

7 Halliburton and Drummond, J. Physiol. 51, 235, 1917.

The fat present in egg-yolks is rich in A, as is also cod-liver oil, and it has been suggested that the long-recognized restorative value of milk, eggs and cod-liver oil may depend upon their vitamin content.

Fig. 3. - The remarkable growth-promoting property of small amounts of cod liver oil as a source of fat-soluble vitamin is illustrated by the above growth curves. White Indian corn was used as a source of water-soluble vitamin as it has been shown in numerous experiments to lead to nutritive failure as a source of fat-soluble A and yet it furnishes plenty of the water-soluble vitamin for normal growth.

Courtesy of Prof. H. Steenbock, M. T. Sell, M. V. Buell and Journal of Biological Chemistry (47, 102, 1921).