A double-blind prospective study was carried out comparing the prophylactic effect of maprotiline and lithium carbonate over a period of one year in patients suffering from recurrent affective disorders. The average Affective Morbidity Index was lower, but not significantly so, in patients treated with lithium. A further analysis, based on dividing patients into those with no affective morbidity and those who showed some affective morbidity during the study, demonstrated lithium carbonate to be significantly superior to maprotiline both in the group as a whole and in unipolar depressives. A correlation between high plasma maprotiline concentration and low morbidity was observed and was in line with an earlier report. A highly significant negative correlation (r=-0-97; p less than 0-001) was found between plasma maprotiline concentration and body weight. Although the results showed lithium carbonate to be superior to maprotiline in the study, it should be emphasized that the plasma levels of lithium were constantly monitored and maintained at what is considered to be its optimum concentration, whereas the maprotiline treated patients were kept on a fixed dosage regime irrespective of plasma levels.