ALTHOUGH BLACK PRISONERS WERE PAROLED IN APPROXIMATELY THE SAME PROPORTION AS WHITE PRISONERS, BLACKS WERE EVALUATED BY DIFFERENT CRITERIA. MOST BLACK PRISONERS HAD TO MEET AN ADDITIONAL REQUIREMENT NOT IMPOSED UPON WHITE PRISONERS -- PARTICIPATION IN INSTITUTIONAL TREATMENT PROGRAMS. THE RESULT WAS THAT MOST BLACK PRISONERS WHO WERE PAROLED HAD SERVED SIGNIFICANTLY LONGER PROPORTIONS OF THEIR SENTENCES THAN HAD WHITE PAROLEES. THERE WAS AN APPARENT TENDENCY ON THE PART OF THE PAROLE BOARD TO FAVOR OLDER BLACK PRISONERS AND BLACK PRISONERS WITH PRIOR CONVICTIONS AND TO PENALIZE YOUNGER BLACK PRISONERS AND THOSE WITH NO PRIOR CONVICTIONS. THE FEW BLACK PRISONERS WHO WERE PAROLED WITHOUT HAVING PARTICIPATED IN TREATMENT BENEFITED IN COMPARISON BOTH TO OTHER BLACK PAROLEES AND TO WHITE PAROLEES. IN COMPARISON TO OTHER BLACK PAROLEES, THESE FEW WERE OLDER, MORE LIKELY TO BE PROPERTY OFFENDERS, AND SLIGHTLY MORE LKELY TO HAVE PRIOR CONVICTIONS. THE DATA SUGGEST THAT THE PAROLE BOARD'S BIAS MAY NOT HAVE BEEN RACIAL PER SE BUT ACTUALLY A BIAS AGAINST PERCEIVED MILITANCY IN YOUNGER BLACK PRISONERS AND THOSE CONVICTED OF DRUG AND VIOLENT OFFENSES. THE PAROLE BOARD MAY HAVE VIEWED PARTICIPATION IN TREATMENT PROGRAMS NOT AS AN INDICATION OF REHABILITATION BUT AS EVIDENCE THAT THE OFFENDER WAS NO LONGER MILITANT. THE APPARENT LENIENCY DISPLAYED TO OLDER BLACKS, DESPITE THEIR MORE EXTENSIVE CRIMINALITY, SUPPORTS SUCH AN INTERPRETATION. TABULAR DATA ARE INCLUDED. (LKM)