Administration of lithium salts is used mainly as a treatment for recurrent endogenous affective disorders (review by Coppen, 1973). Lithium salts show a weak or moderate therapeutic action in already existing endogenous depression, they have a fairly strong therapeutic action in the manic phase of the disease, and when given as a maintenance treatment they exert a clear-cut stabilizing or prophylactic action so that further manic and depressive recurrences are attenuated or abolished. Discontinuation of the treatment with lithium salts leads to reappearance of the recurrences. This action pattern is not seen with any other drug, and the biochemical hypotheses proposed for the mode of action of the traditional antimanic and antidepressive drugs d o not seem to be adequate insofar as lithium salts are concerned. The ability of lithium salts to prevent both manic and depressive episodes indicates that Li+ acts on cerebral processes with a regulatory function, but these processes are as yet unknown, and no comprehensive and experimentally testable hypothesis has been advanced for the biochemical mechanism underlying the 'mood-normalizing' effects of lithium salts. Since the patients' brains cannot be subjected to direct biochemical examination, indirect approaches must be used. Among these is research on the biochemistry involved in the production of various extracerebral side effects of treatment with lithium salts: