A study of the inorganic elements in nutrition was carried out by Osborne and Mendel by preparing a large variety of salt mixtures in which one or more of the elements was omitted and replaced by increments of the remaining ones so as to maintain the balance of the acids and bases as nearly as possible. This was fed to the rats on an otherwise salt-free diet, and the effect on their growth noted. Good growth was possible on foods containing only traces of magnesium, sodium and chlorine, less than 0.04 per cent of either sodium or chlorine in the food being sufficient to enable rats to complete their growth. When both sodium and potassium were lacking growth stopped, but when only one of these elements was missing growth was nearly or quite normal. Restriction of either calcium or phosphorus causes prompt restriction of stoppage of growth. Magnesium could not replace calcium even when given in large quantities. Osborne and Mendel, commenting on the fact that the growing animal can fully supply from inorganic sources its requirements for various of the elements emphasizes anew that it is unnecessary to consider the presence of calcium, phosphorus, and iron, for example, in natural food to the degree that is currently believed.

Any shortage of an essential inorganic element can be suitably remedied under ordinary conditions by the use of its salts. For feeding farm animals, where the lack of calcium and prosphorus in their grain rations is always encountered, the demonstration that complete nutrition can be attained upon diets in which the inorganic ingredients are supplied in the form of their commercial salts has a significance that is just beginning to be appreciated.

Some interesting observations have been made by Sherman and Pappenheimer (J. Exptl. Med. 34, 180, 1021) respecting the effect of phosphates and lime salts on rickets. A diet of 95% patent flour, 3% calcium lactate, 2% common salt and also a diet having the same proportion of flour, 2.9% calcium lactate, 2% salt and 0.1% iron citrate regularly induced rickets in young rats. The substitution of 0.4% potassium phosphate for a small portion of the calcium lactate in the diet very curiously prevented the development of rickets. An examination of the content of calcium in the bodies of the rats showed those receiving the potassium phosphate had assimilated a greater proportion of calcium than those which developed rickets. The results obtained perhaps may not be due to the greater assimilation and deposition of calcium. There may be a quantitative relationship between the lime and phosphorus rather than an actual deficiency of either of them.

66 McCollum and Davis, J. Biol. Chem. 19, 260, 1914. 67 Osborne and Mendel, J. Biol. Chem. 34, 131.

A pathological condition bearing fundamental resemblances to rickets of the human being resulting from diets low in phosphorus and fat-soluble A has been observed by Shipley, Park, McCollum and Simmonds (Bull. Johns Hopkins Hosp. 32, 160, 1921). Two diets, low in fat-soluble A and phosphorus produced in the majority of the young rats placed upon them pathological conditions of the skeleton having a fundamental resemblance to rickets. The changes are not identical, however, with that disease as it usually manifests itself in the human being. The chief difference consisted in the presence of scattered or irregular deposits of calcium salts in the cartilage and metaphysis. When the deficiency in phosphorus is compensated by the addition of a complete salt mixture containing P04 ion, the deficiency in fat-soluble A still existing, no pathological changes of a rachitic nature developed. The addition of the P04 ion to the diets deficient in it and in the factor A prevented, therefore, the development of any changes of a rachitic-like nature. Thus the P04 ion in the diet may be the determining influence for or against the development of rickets. If the P04 ion content of the diet is sufficiently high, a deficiency of A cannot cause these changes in the skeleton.

A deficiency in A is held not to be the sole cause of rickets. Conversely, it is necessary that the diet be low in its content of phosphorus, all other factors, except A, being optimal, for these conditions to develop. Since the addition of the P04 ion to the diet prevented the development of these changes in the skeleton, but had no effect in preventing xerophthalmia, it seems permissible to infer that xerophthalmia and rickets do not have an identical etiology. The above results do not exclude A from consideration as an etiological factor in the production of rickets and kindred diseases, since the level of the blood P04ion is, in all probability, determined in part by the amount of A available.