DURING the past 20 years a very large number of observations have been published on the total cholesterol content of blood or of serum from pathological cases. As a result, the blood picture is described, in any one case, as a hypercholesterolaemia or a hypocholesterolaemia, in comparison to a socalled normal value of total cholesterol in human blood. It is, however, to be noted that relatively few determinations have been made on normal subjects, and that values of the normal, obtained from these sparse data-not too concordant in the different researches-have been quoted from author to author.It is generally agreed that in certain forms of nephritis, irregularly in diabetes mellitus and also in the later months of pregnancy, a hypercholesterolaemia is found, and that in infective conditions a hypocholesterolaemia occurs. As regards other diseases it is impossible in the present state of our knowledge of the physiological limits of cholesterol content of the blood to decide if there is any constant alteration in the level of this value. Denis [1917] considers that at present the total cholesterol content of the blood is practically of no diagnostic significance.Though much work has been done on the total cholesterol content of blood or serum, comparatively little attention has been directed to the relative amounts of so-called free or non-esterified cholesterol and cholesterol in ester form. This unfortunate position is to some extent explained by the historical development of our knowledge of this subject. Fat, cholesterol and allied substances were obtained together by the extraction of tissues and fluids of the body by means of ether and other solvents, and in early days no effective methods were available for differentiating, quantitatively at any rate, between free and ester cholesterol. Consequently cholesterol came to be regarded as intimately connected with fat metabolism.This idea has persisted throughout the history of the subject and has been upheld largely by the French observers. Terroine [1919] considers that there is a constant lipaemic coefficient, i.e. a constant ratio of total cholesterol to total fatty acid in individual animals. Chauffard and his school [1920]