The discussion whether assay of bloodlevels of antidepressants is useful should be divided into the question whether a relationship between bloodlevel and therapeutic effect exists and whether this relationship can be used for the treatment of the indi- vidual patient. According to all experts a relationship between bloodlevel and therapeutic effect has been shown to exist for imipramine (plus its active metabolite desipramine), for desmethylimipramine (desipramine) and for nortriptyline. Concerning amitriptyline and clomipramine the evidence is more equivocal. For other classical and for the modern antidepressants no relation between bloodlevel and therapeutic effect has been established. It is shown that the probability of a therapeutic effect with imipramine can be increased when bloodlevel assays are used in clinical practice. It is concluded that the dose with treatment with imipramine, desipramine or nortriptyline should be based on bloodlevels. We are inclined to advise similarly for amitriptyline and clomipramine. With all other antidepressants dosing based on bloodlevels does not appear advisable and it can actually be counterproductive.