ALTHOUGH DEFINITIONS OF CRIME HAVE CHANGED AS SOCIAL SYSTEMS HAVE CHANED, THE CONCEPT OF CRIME ALWAYS HAS BEEN USED BY SOCIETY AS A MEANS OF KEEPING ORDER. HOWEVER, IT IS KNOWN THAT NOT ALL PUNISHABLE OFFENSES ARE DETECTED, THAT NOT ALL DETECTED OFFENSES ARE REPORTED, AND THAT NOT ALL OFFENSES ARE BROUGHT TO TRIAL. CRIMINOLOGISTS ARE CONCERNED WITH THE DISCREPANCY BETWEEN OFFICIALLY RECORDED OFFENSE STATISTICS AND ACTUAL CRIMINALITY. RESEARCH INTO THE HIDDEN AREAS OF CRIMINALITY IS DESIGNED TO SUPPLEMENT AND CORRECT THE PICTURE OF CRIMINALITY PROVIDED BY OFFICIAL STATISTICS. OTHER RESEARCH RELATES TO SELECTION PROCESSES -- POLICE AND PROSECUTOR DISCRETION -- WITHIN THE CRIMINAL JUSTICE SYSTEM. SELECTION PROCESSES ARE NECESSARY DUE BOTH TO THE LIMITED CAPACITY OF THE SYSTEM AND TO THE NEED TO MAINTAIN THE SOCIAL INTEGRATION FUNCTION OF THE CONCEPTS OF CRIME AND PUNISHMENT. STATISTICS ON CRIMINALITY ARE IMPORTANT IN THAT THEY PROVIDE GUIDANCE TO THE POLICE AND TO THE COURTS ON HOW TO OPERATE SELECTION PROCESSES; I.E., HOW TO DETERMINE WHICH MODES OF BEHAVIOR ARE SUFFICIENTLY ANTISOCIAL TO WARRANT THE EXPENDITURE OF THE CRIMINAL JUSTICE SYSTEM'S LIMITED RESOURCES. (LKM)