That the curative power of cod-liver oil for rickets 32 is not associated with its content of A is indicated by the work of Phemiater, Miller and Bonar,33 who found that phosphorus and cod-liver oil and phosphorus alone had quite similar results. While it is not possible to make a definite statement without further investigation, the results so far obtained suggest that phosphorus may be as effective as the cod-liver oil alone.

McCollum and his associates 34 have published a preliminary communication on the production of skeletal derangements in rats by deficient diets. Certain of the pathological changes observed are identical with those seen in the bones of rachitic children, while others are similar to but not identical with these. They have proved that these conditions are undoubtedly due to the diet alone, but hesitate to ascribe them to one specific factor in the diet. From the observed results a low content of A, low calcium content, poor quality of proteins, and unsatisfactory salt combinations, acting together may all be contributory causes, and as the condition can be induced in several different ways it is suggested that there may be more than one kind of rickets.

Mackay 35 has carried on a clinical investigation of fifty-one rachitic children, the results of which, while not entirely accurate on account of the conditions under which the work had to be carried on (through the outpatient department of the hospital) are in accordance with the view that rickets is a dietetic disease, commonly developing in children receiving insufficient fat and too much carbohydrate, especially when bottle-fed.

An important contribution to this subject has been made recently by Paton and Watson.36 These experimenters found that a liberal allowance of milk fat, up to even 14 grams per kg. of body-weight does not prevent the onset of rickets in young dogs under ordinary laboratory conditions, whereas pups kept largely in the open air may escape the development of rickets on an intake of less than 1 gram of milk fat per kg. of body weight. With scrupulous care as to cleanliness it is, at least sometimes, possible to rear pups free of rickets in the laboratory on an intake of only about 0.5 grams of milk fat per kg. of body weight, along with bread, provided that the diet affords an adequate supply of energy. The energy value of the diet would seem to play a part in controlling the development of rickets, but that it is only a contributory part is shown by the development of rickets in pups with a high energy intake if confined without scrupulous care as to cleanliness. Milk fat may be reduced to about 0.3 grams per kg. of body weight, if its place is taken by an equal amount of lard, without the onset of rickets.

In view of these observations it seems necessary to conclude that both environment and diet are important factors in the development of rickets, sometimes one, and sometimes the other, predominating in a particular case.

The factor which plays a role in the etiology of rickets is possibly a fourth unidentified dietary factor or vitamin. (McCollum, Dental Cosmos, 63, 596,1021).

32 HowUnd and Park, Tr. Am. Pediat. Soc. 1020. 33 Phemister, Miller, and Bonar, J. Am. Med. Ass. 70, 850, 1021. 34 McCoUum, Simmondi, Parsons, Shipley and Park, J. Biol. Chem. 46, 333, 1920.

35 Mackay, Brit. Med. J. 1020, ii, 020.

36 Paton and Watson, Brit J. Expti. Pathol, a, 75-04, 1021.