THE STUDY WAS CARRIED OUT IN SOUTH AFRICA, ON THE CAMPUS OF THE UNIVERSITY OF THE WITWATERSRAND, WHICH IS, IN THE SOUTH AFRICAN CONTEXT, A LIBERAL INSTITUTION. TWO SIMULATED ASSAULTS WERE CARRIED OUT. THEY WERE ALMOST IDENTICAL EXCEPT FOR THE FACT THAT IN ONE, A WHITE ASSAULTED A WHITE, WHILE IN THE OTHER A BLACK ASSAULTED A WHITE. IN EACH CASE, 15 WHITE MALE AND FEMALE UNDERGRADUATE EYEWITNESSES WERE REQUESTED TO WRITE A REPORT DESCRIBING BOTH THE PERSONS INVOLVED AND THE SEQUENCE OF EVENTS. RELIABILITY OF TESTIMONY SCORES WERE CALCULATED FOR EACH SUBJECT. FOR THE REPORTS TAKEN AS A WHOLE, THE MEAN SCORE OF THE WHITE ASSAILANT GROUP WAS 39.7, AND THAT OF THE BLACK ASSAILANT GROUP WAS 42.8. FOR THE DESCRIPTION OF THE PARTICIPANTS INVOLVED, THE WHITE ASSAILANT GROUP HAD A MEAN SCORE OF 29.25, AND THE BLACK ASSAILANT GROUP HAD A MEAN SCORE OF 24.6. FOR DESCRIPTION OF THE SEQUENCE OF EVENTS, THE MEAN SCORE FOR THE WHITE ASSAILANT GROUP WAS 54.4, AND THAT OF THE BLACK ASSAILANT GROUP WAS 60. IN EACH CASE, THE DIFFERENCE BETWEEN THE SCORES WAS NOT SIGNIFICANT. WHILE THE RESULTS, WHICH POINT TO NO PRESENCE OF THE RACIAL BIAS, MAY HAVE IMPORTANT IMPLICATIONS FOR ACTUAL CRIMINAL TRIALS, THE FACT REMAINS THAT ENGLISH-SPEAKING UNIVERSITY STUDENTS SITTING IN A LECTURE THEATER ARE NOT IDENTICAL WITH ACTUAL WITNESSES MAKING STATEMENTS TO THE POLICE, WHICH WILL THEN BE USED AS EVIDENCE IN A COURT OF LAW. REFERENCES ARE INCLUDED. (AUTHOR ABSTRACT MODIFIED)