Casein nitrogen has been shown to account for roughly 76 per cent, of the total nitrogen in over 200 samples of normal milk. Samples low in solids-notfat have shown a casein nitrogen percentage below this figure, and by assuming a 76 per cent, level of casein nitrogen as standard, it is suggested that milk may be considered as made up of two fractions, (a) a true-milk fraction and (b) an isotonic diluent. From the distribution of the nitrogen in whole-milk samples it has been possible to calculate on this basis the amount of dilution and the nitrogen distribution in the diluent. Results are given for 38 samples of milk. The chloride content of the diluent has also been approximately arrived at and is approximately isotonic with a hypothetical non-lactosecontaining milk.

The similarity in composition of the diluent either to a modified blood serum or lymph serum or a casein- and lactose-free lacteal secretion has been discussed. Its high chloride content favours its similarity either to lymph serum or a “transudate” or “exudate” closely allied in composition to lymph serum from pathological secretory tissue. The possibility of after-effects of attacks of mastitis causing a chronic catarrhal condition of the udder being the cause of the secretion of this diluent is also discussed.