The core assumption of the typological approach is that there are a number of distinct types or groups of offenses and offenders that can by identified and studied. This view has prevailed in various degrees of sophistication throughout the history of criminology. The rationale for the development of typological schemes in criminology is to facilitate etiological analysis of criminal behavior and the development of rehabilitative approaches appropriate for various offender types. For a typological system to be valid, it must be clear and explicit so that actual offenders can be assigned to categories. Typologies should also consist of mutually exclusive types so that actual offenders or crime events fall into only one of the categories. Also, the number of types should be relatively small so as not to be too unwieldy. Finally, the types must be empirically congruent with actual offenders or offenses. An assessment of three comprehensive typologies (Clinard and Quinney, Glaser, and Gibbons), including one developed by the author (Gibbons), indicates that the degree of patterning of offense behavior and other dimensions assumed in these typologies is much greater than what exists in the real world of criminal behavior. The behavioral diversity of offenders apparently defies systematic classification. Forty-five references are listed.