THE REVIEW FOCUSES ON INSTITUTIONALIZED OFFENDERS AND DEFINES CHILD MOLESTERS AS MEN WHO HAVE HAD PHYSICAL SEXUAL CONTACT WITH SOMEONE 13 YEARS OLD OR YOUNGER, WHO ARE THEMSELVES AT LEAST 16 YEARS OLD, AND WHO WERE AT LEAT 5 YEARS OLDER THAN THEIR VICTIM. TYPOLOGIES OF CHILD MOLESTERS, DANGEROUSNESS, SEXUAL PREFERENCES, AND PSYCHOLOGICAL TEST RESPONSES ARE DISCUSSED. THE REVIEW SUGGESTS THAT CHILD MOLESTERS ARE NOT A HOMOGENEOUS GROUP. COMPARED TO HETEROSEXUAL CHILD MOLESTERS, HOMOSEXUAL OFFENDERS CHOOSE OLDER (PUBESCENT) VICTIMS, ARE MORE LIKELY TO RECIDIVATE, AND ARE LESS NUMEROUS. INCESTUOUS CHILD MOLESTERS ARE ALMOST ALWAYS HETEROSEXUAL AND LEAST LIKELY TO RECIDIVATE. MOST CHILD MOLESTERS DO NOT PRESENT A PHYSICAL DANGER. HOWEVER, THERE IS LITTLE CONSENSUS ABOUT HOW TO IDENTIFY PERSONS WHO WILL USE FORCE IN COMMITTING A SEXUAL OFFENSE AGAINST A CHILD. PSYCHOPHYSIOLOGICAL MEASURES, PARTICULARLY PENILE RESPONSES TO VISUAL STIMULI, HAVE PROVEN USEFUL IN ASSESSING SEXUAL PREFERENCES. TWO TREATMENT APPROACHES--GROUP THERAPY AND BEHAVIOR MODIFICATION INCLUDING AVERSION THERAPY AND TRAINING IN SOCIAL SKILLS--ARE FOUND IN THE LITERATURE ON CHILD MOLESTERS. SEVERAL THERAPEUTIC INTERVENTIONS APPEAR USEFUL, BUT THERE IS A DEARTH OF STUDIES COMPARING TREATMENTS. RECOMMENDATIONS FOR TREATMENT AND RESEARCH EMPHASIZE THE FOLLOWING: TREATMENT OF RECIDIVISTS, HOMOSEXUAL OFFENDERS, AND OFFENDERS WHO COERCE THEIR VICTIMS; INCLUSION OF SEXUAL PREFERENCE ASSESSMENT AND MODIFICATION COMPONENTS IN TREATMENT PROGRAMS; INDIVIDUALIZATION OF TREATMENT; DEVELOPMENT OF MORE SOPHISTICATED OFFENDER TYPOLOGIES; AND RESEARCH ON ASSESSMENT OF SOCIAL SKILL DEFICITS AMONG CHILD MOLESTERS AND ON THE PREDICTION OF DANGEROUS POSTRELEASE BEHAVIOR ON THE PART OF INSTITUTIONALIZED CHILD MOLESTERS. A LIST OF REFERENCES IS INCLUDED. (AUTHOR ABSTRACT MODIFIED--LKM)