To the Editor: Shitij Kapur, M.D., Ph.D., F.R.C.P.C., and Philip Seeman, M.D., Ph.D., F.R.S.(C.), seemed to suggest in their article (1) that antipsychotic drugs that have atypical properties have more rapid rates of dissociation from dopamine D2 receptors than those with typical properties. However, this conclusion is incompatible with their own report (2) that sertindole, which most certainly is an atypical antipsychotic drug because of its very low propensity to cause extrapyramidal symptoms, has the same koff as haloperidol, the prototypical typical neuroleptic drug. Despite this clear failure, their theory might have had credibility if they had offered as proof even one atypical antipsychotic drug that had been developed on the basis of this theory. To my knowledge, they have not done so. By contrast, risperidone, olanzapine, quetiapine, ziprasidone, and iloperidone, all clearly atypical antipsychotic drugs, were explicitly selected for clinical development on the basis of the "serotonin 5-HT2A/D2 hypothesis"that Drs. Kapur and Seeman reject and were then shown to be atypical antipsychotics in both basic research models and clinical studies. In addition to the atypical antipsychotic drugs mentioned, there are at least a half dozen other agents of diverse chemical classes selected on the basis of the 5-HT2A/D2 hypothesis during their early development. Most people engaged in drug development agree that a valid theory should be capable of generating new chemical entities.