AbstractStudies supporting Coloninger's original Type 1/Type 2 hypothesis for classifying the generic transmission of alcoholism have sometimes failed to control for important sources of potential bias. First, the environmental effects of parental alcoholism must be distinguished from the genetic effects of parental alcoholism. Secondly, antisocial personality disorder must be distinguished from alcohol dependence. Thirdly, to control for developmental effects a cohort of alcoholics should be followed into late mid‐life. The present report, a 50‐year prospective study of the development of alcoholism in two community samples of 456 disadvantaged youth and 204 Harvard graduates, addresses these three, potential sources of bias. In this report the age of onset of alcoholism and the degree of antisocial symptomatology was correlated with disturbed family environment but was quite independent of the presence or absence of a heredity positive far alcoholism. The reasons why such findings threaten the validity of the Type 1/Type 2 hypothesis are discus