We have performed the tourniquet-twitch test of Roberts and Ryan in normal and in malignant hyperthermia (MH) patients and relatives. This test measures the ratio of electrically induced thumb twitches noted after 10 min of ischemia with those noted immediately prior to the ischemia. We found no significant differences in this ratio between normal subjects and those who have had MH reactions, or relatives of such individuals. Furthermore, we have observed no significant differences in tourniquet-twitch ratios between those with normal caffeine-halothane contractures and persons with caffeine-halothane contractures tests positive for MH. These findings do not agree with those of Roberts and Ryan, who reported that tourniquet-twitch ratios were higher in MH patients than in normal patients. We have, however, determined that subjects with tourniquet-twitch ratios greater than or equal to 1.8 are substantially younger than those with tourniquet-twitch ratios less than or equal to 1.0. Therefore we do not believe that the tourniquet-twitch test is useful as a diagnostic, or even as a screening test for MH.