Lithium is an effective drug for both treatment and prophylaxis of bipolar disorder. However, the mechanism of lithium action is still unknown. The inositol depletion hypothesis is supported by biochemical and behavioral data in rats, but primate inositol levels are higher than in rodents and may obviate the effects of depletion. Inhibition of 5HT autoreceptors by lithium is supported by biochemical and behavioral data in rats but would seem more related to lithium's antidepressant than to its antimanic or prophylactic effects. Lithium induces increases in levels of the anti-apoptotic factor Bcl-2. This effect could be most relevant for treatment of neurodegenerative disorders. Lithium inhibits glycogen synthase kinase-3, which is involved in a wide range of signal transduction pathways. However, this lithium effect occurs at high concentrations and may be more relevant for its toxic effect. Lithium in low concentrations induces accumulation of PAP, which affects several cellular processes including RNA processing. However, PAP phosphatase is present more in peripheral tissues than in brain. This lithium effect could explain some of its peripheral side effects. Chronic lithium administration upregulates glutamate reuptake and thus decreases glutamate availability in synapse. Glutamate is an excitatory neurotransmitter and its reduction could exert an antimanic effect. Biochemical and clinical experiments are necessary to determine the key mechanism of lithium efficacy in treatment and prophylaxis of affective disorders.