Early investigators differed widely in their estimations of the bromine content of normal blood (1, 2), and it was not until the development of a refined technique by Bernhardt and Ucko (3) that it was generally recognized that blood bromine varied around 1 mgrm. %. These authors gave a range of 1.0 to 1.6 mgrm. %. Their studies, however, were not extensive. During 1931–3 Zondek and Bier (4, 5, 6) published the results of a series of investigations into the bromine content of the blood of normal and psychotic patients. From a study of 150 mentally normal patients they considered that blood bromine usually varied between 0.8 and 1.0 mgrm. %, a range later extended to 0.731.10 mgrm. % as a result of further study. In a large number of psychotics examined by them, values lying within these limits were found with the notable exception that 85%-90% of 60 cases of endogenous manic-depressive psychoses gave figures 40% to 60% below their normal levels (Zondek, 1933) (6). These low figures appeared only to be associated with this psychosis, although of the 16 cases of schizophrenia examined by them, 5 gave figures lower than the normal. They stated that in these 5 cases the mental picture was characterized by more or less marked depression, and they considered the possibility of their ultimately turning out to be manic-depressive in character. In a later paper, three examples are cited of low blood bromine associated with organic brain disease, and thought to be due to interference with a bromine-regulating centre in the brain. The low values found in manic-depressive psychoses were shown to be independent of the phase, and it was stated that values in general were not subject to seasonal variations, to menstrual fluctuations, or to alterations due to variable salt intake.