An empirical examination of previous alcoholism classifications that have used the Minnesota Multiphasic Personality Inventory reveals that one particular subtype emerges with considerable convergence. This article describes this subtype and outlines some of its implications for alcoholism theory and research.The last two decades have yielded a number of studies that have attempted to establish useful typologies of alcohol abusers by using multivariate classification strategies (see reviews by Morey & Blashfield, 1981;Skinner, 1982). Although some consistencies have emerged from these studies, disagreement remains regarding the stability and replicability of the subtypes that have been suggested.Many of these studies have sought to differentiate alcoholics according to features of personality and psychopathology, using personality inventories such as the Minnesota Multiphasic Personality Inventory (MMPI) as a basis for classification. Reviewers of this literature, however, have drawn substantially different conclusions in attempting to integrate typologies reported across studies. For example, some authors (e.g., O'Leary, Donovan, Chaney, & O'Leary, 1980) have suggested that two clusters (a high 4 profile and a 2-7-8 profile) appear consistently in alcohol typology studies. Other reviewers, though, have interpreted previous studies as being consistent with four (Conley, 1981), five (Zivich, 1981), or seven (Nerviano & Gross, 1983) replicable subtypes.Recently, Morey and others (Morey, Skinner, & Blashfield, 1984;Roberts & Morey, 1985) have proposed a different approach to the classification of alcoholics. The basis of this classification is the Alcohol Use Inventory (AUI; Wanberg, Horn, & Foster, 1977), a multidimensional instrument that taps domains such as preferred style of alcohol use, unfavorable consequences of alcohol use, and perceived benefits from drinking. Using an extensive validation design that compared 23 different statistical classification strategies, three subtypes emerged that were distinguishable by different stylistic features of alcohol use. Although these subtypes were very different with respect to severity of drinking, there were a number of qualitative indices that suggested these types reflected distinct constructs as well as different points on an alcoholic continuum (Morey, Blashfield, & Skinner, 1983). In addition, these subtypes have been found to replicate across samples of alcoholics as well as providing differential descriptions of personality and psychopathology between types.