It has been 60 years since Cade first described patients who responded to antimanic lithium treatment. Two decades later, responders to lithium stabilization emerged in larger numbers. The responses of many severely ill bipolar patients to lithium were striking and called for an explanation. Remarkable reactions to a simple ion generated hope for an uncomplicated laboratory test of response and an extensive search for suitable biological markers ensued. But despite promising reports, particularly from molecular genetics, we are still waiting for a biological elucidation of the stabilizing effects of lithium. The most useful predictor of lithium stabilization has to date been the patient’s clinical profile, based on a comprehensive clinical assessment: complete remissions and other characteristics of episodic clinical course, bipolar family history, low psychiatric comorbidity and a characteristic presenting psychopathology. In brief, the responders approximate the classical Kraepelinian description of a manic-depressive patient. But the most intriguing findings have recently emerged from prospective observations of the next generation: the children of lithium responders, their counterparts coming from parents who did not respond to lithium and controls. Overall, they indicate that parents and offspring suffer from a comparable brain dysfunction that manifests clinically in distinct stages. If the child’s predicament starts early in childhood, it presents with varied, nonaffective or subclinical manifestations that are usually nonresponsive to standard treatments prescribed according to the symptoms. The next stage then unfolds in adolescence, first with depressive and later with activated episodes. The observations have a potential to markedly enrich the prevailing understanding and management of mood disorders.