20 routine patients with endogenous depression were investigated in a kinetic and 4 week treatment study. Steady-state plasma nortriptyline concentrations above 200 microgram/L were associated with a highly significant poorer therapeutic outcome. The correlations between the 24, 48 and 72 hour concentrations and steady-state concentration were very good (r = 0.81, 0.97, 0.94; p less than 0.0001) and better than the correlation between half-life and steady-state (r = 0.65; p less than 0.01). The Spearman rank correlations (Rs) between amelioration of depression measured by the Hamilton Rating Scale (HRS) and the 24, 48 and 72 hour concentrations were highly significant (Rs = 0.74, 0.79, 0.79; p less than 0.001) but for half-life (Rs = 0.33) the correlation was not significant. The single 48 hour plasma nortriptyline concentration following a single oral dose is recommended as a reliable simplified monitoring test suitable for a busy clinic. The test is useful for dosage adjustment to maximise antidepressant action and minimise toxicity. A tentative dosage adjustment schedule for individualising antidepressant treatment with nortriptyline based on the 48 hour or the 24 hour plasma concentration is proposed.