Crime is distinguished from noncriminal behavior on qualitative grounds. Wolfgang and Sellin pioneered research aimed at devising a set of scores for criminal acts by getting members of the public to attach numbers to crimes according to degree of seriousness. Students at two Philadelphia universities and a group of judges and police were asked to rate various sets of descriptions of criminal acts according to their seriousness. Subsequent research by Wellford and Wiatrowski in 1975 and Turner in 1978 confirmed the findings of Wolfgang and Sellin and established general criteria for crime seriousness rating. Crimes involving physical harm against persons are consistently rated as most serious; violations of administrative rules are regarded as least serious. Alternative ways of measuring seriousness correlate highly. Considerable agreement on which crimes are considered more serious exists across cultures, classes, age groups, and educational levels. Finally, results of later studies support earliest findings. Limitations of seriousness scores involve the fact that a legal definition of crime does not address the etiology of the criminal acts in question, and the most serious and least serious crimes are not necessarily on a continuum of criminality. In addition, the seriousness measures have not been accepted in all sectors of the social science community. Other limitations include controversies over the meaning of seriousness, the fact that different theories of crime react differently to the crime seriousness studies as exemplified by sociologists' assessments of homosexuality and abortion, and by radical criminological theories regarding the criminal act as a social control device. Despite the admitted lack of convincing evidence to support the position, it is argued that the findings of crime seriousness research reflect the normative structures of society. The authors conclude that there is little to be gained by the use of crime seriousness scores in the evaluation of criminal justice programs, because such scores cannot substitute for offense-specific crime rates. Future research should focus on more common offenses, and probability samples should be employed. Extensive endnotes and 25 references appended.