In discussing the proportion of B in milk, Osborne and Mendel15 advise the use of a liberal amount of milk when this is depended on to supply any considerable proportion of this most necessary food factor. A case of this kind is that of infant feeding where it is customary to reinforce the supply of calories by diluting top milk and adding milk sugar. Under these circumstances the food contains a relatively smaller proportion of the water-soluble vitamin than does the original cow's milk. While milk thus modified may contain sufficient vitamin as long as the food intake is normal, if for any reason the child's appetite fails, the vitamin supply is reduced and endless dietary troubles may easily result. It is not improbable, they state, that a large part of the difficulties of artificially feeding babies is due to this cause, and that these can be obviated as successfully by securing an adequate supply of this indispensible constituent of a suitable diet as has been the case in feeding animals on artificial diets, since we have learned how properly to provide this food factor.

Drummond16 recommends modifying cow's milk with cream and whey only, and believes that if this is done there is little danger of deficiency of vitamins.

Bosworth has patented a method17 for preparing a "reconstructed milk " of the approximate composition of human milk, by separating the fat (with the vitamin A) and casein from the serum of cow's milk together with practically all the calcium and phosphorus, but without any of the dissolved vitamins B and C, and finally re-combining the resulting serum with a portion of the fat and with protein and calcium and phosphorus in the proportions present in human milk.

13 Eddy and Roper, Proc. 8oc. Exp. Biol. Med. 14, 62, 1016; Eddy, Proc. Am. Soc. Biol. Ghem. J. Biol. Ghem. 41, xxxiv, 1020. 14 Bachman, J. Biol. Ghem. 39, 236, 1010. 15 Osborne and Mendel, J. Biol. Ghem. 34, 644, 1018. 16 Drummond, Lane. 1018, ii, 482. 17 U. S. Patent 1,341,040.

As has already been pointed out various authorities recommend regular administration of additional antiscorbutic to all artificially fed infants. Chick, Hume and Skelton 18 and Barnes and Hume 19 place special emphasis on the necessity of giving an antiscorbutic ration to babies fed on heated or dried milk or any milk substitutes, on the ground that even in raw cow's milk this vitamin is near the minimum and that there is danger that any heat treatment may bring the amount down below the safety line.

Orange juice has long been used in infant feeding as a source of antiscorbutic. At University College Hospital, London, the minced orange, rind and pulp, is squeezed through a tincture press, and the juice collected. The essential oil from the rind acts as a preservative and the juice keeps for quite a long time. Doses of one teaspoonful in a little sugar water are given twice daily to infants under three months, and double that amount to babies over that age.20 Where infantile scurvy has developed Harden, Zilva and Still21 state that it may be successfully treated with lemon juice preparations from which the free citric acid has been removed by treatment with calcium carbonate and alcohol. When the irritant part was thus removed it was possible to give the equivalent of 6 to 12 lemons daily without any gastrointestinal disturbance. Hess 22 recommends canned tomato unreservedly as the most serviceable antiscorbutic for artificially fed infants. It is well borne, comparatively inexpensive, and readily available., According to Hess it may be used in doses of one ounce per day in infant feeding. The juice of raw swedes (yellow turnips) is recommended by the English authorities 23 as the best substitute for orange juice where this is found too expensive for use.

The antiscorbutic value of these vegetables is high 24 and they have the merit of being abundant and cheap. The clean cut surface of the raw turnip is grated on an ordinary kitchen grater and the pulp folded in a small piece of muslin and squeezed with the fingers, till the juice runs out. Barlow25 and Hess and Fish26 have reported favorably upon the use of potatoes as an antiscorbutic. This may be made into a cream by shaking up well cooked potato in water, and fed as a diluent of cow's milk.

18 Chick, Hume and Skelton, Lane. 1918, ii, 1.

19 Barnes and Hume, Bioch. J. 13, 306, 1919.

20 Barton, Lane 1919, ii, 348.

21 Harden, Zilva and Still, Lane. 1919, p. 17.

22 Hess, N. Y. State Jour, of Med. 20,209,1920.

23 Med. Res. Com. Rep. 38, p. 80.

24 Chick and Rhodes, Lane. 1918, ii, 774.

25 Barlow, Brit. Med. J. 1894.

26 Hess and Pish, Am. J. Dia. Child 8, 385,1914.

It should be noted that tomato and turnip juice are good sources of A and B as well as of C, so serve a triple purpose. Orange juice is about equivalent to milk as a source of B and may possibly contain A as well, but in very small quantities.

The fact that the proportion of children developing "deficiency diseases " on artificial diets which are unquestionably low in vitamins is small has led to the suggestion27 that the infant starts life with a reserve supply of these factors sufficient to maintain it until its milk diet can be supplemented with other foods. Even if this is the case, however, it may be questioned whether this reserve alone is sufficient for optimum development. Although no direct experimental evidence is available on this point the low mortality among breast-fed, as compared with artificially fed, infants during the first year of life suggests that their resistance may be increased by a higher vitamin content in human than in cow's milk. In the later stages of life, when the milk diet begins to be supplemented by other foods, the difference between the two classes tends to disappear.

Norman" states that he considers the superiority of breast-fed children as compared with bottle-fed ones is not so well marked at school age as one would be led to expect. It is during the early years of life, prior to school age, that the great advantage of breast-feeding over bottle-feeding is so apparent. The bottle-fed children of the poor that survive, appear to overtake their disadvantages, and by the time we find them in school, show no inferiority physically to their breast-fed fellows.

The variation in development of infants on identical diets may very possibly be due to variation in the reserve supply of vitamins which they are provided at birth, and this in turn may be at least partly dependent upon the vitamin content of the food of the mother during pregnancy.

27 Gibson and Concepcion, Phil. J. 80. B. 1910, ii, 119; Hopkins, Lano. 1919, ii, 979.

28 Quoted by Lane-Claypon, " Milk and its Hygienic Relations," p. 312.