SINCE THE PUBLICATION of Bumke's 1 paper in 1902, paraldehyde has been generally accepted as a useful drug in the treatment of delirium tremens. With the introduction of the phenothiazines, these drugs have gained clinical acceptance, but there are few controlled studies reported in the literature. Friedhoff and Zitrin 2 demonstrated that patients receiving paraldehyde recovered more quickly than those treated with chlorpromazine. Hart 3 found no significant difference in the recovery time of delirium tremens patients treated with promazine compared to those treated with paraldehyde. In patients without delirium tremens, the group treated with promazine showed a shorter duration of anxiety and agitation. Laties and colleagues, 4 in a double-blind study comparing promazine and chlorpromazine in delirium tremens, stated that "the drugs were essentially indistinguishable in performance." Gruenwald and associates 5 concluded that patients with mild to moderate symptoms responded promptly to either promazine or triflupromazine, but patients in