Three patients had recurrent episodes of thrombocytopenia that resembled drug purpura, but the drug history in each case did not support the diagnosis. Although the patients specifically denied taking quinidine, serologic testing with this drug was done because the patients had access to it, and it is the commonest cause of drug purpura. Highly specific quinidine-dependent antiplatelet antibodies were found in the sera of all three patients. After being informed of the laboratory findings, the patients have had no recurrences of purpura. Serologic tests for quinidine- or quinine-dependent antibodies can help elucidate some obscure cases of purpura that may be self-induced.