Counselor supervision has been examined in many ways; from the angle of the supervisor, the supervisee, and the supervisory dyad. An area that has not been researched as solidly is that of the effect of supervisor race and gender on White trainees' perceptions of supervision. The present study was designed to examine those variables. The independent variables included supervisor race (Black female or White female), trainee gender, and trainees' levels of White racial identity. Dependent variables included perceptions of supervisor expertness, attractiveness, and trustworthiness, and supportive supervisory behavior and evaluative supervisory behavior scores on the Expectations for Supervisor Behaviors Questionnaire. The expectation was that supervisor race, trainee gender, and trainee's level of White racial identity would serve as predictors of perceptions of supervisors and their behaviors.Participants were 50 Caucasian masters level counseling and counselor education students at two Midwestern universities who volunteered for the study. Caucasian female research assistants were utilized to collect the data. Trainees were provided with one of two biographical sketches describing a hypothetical female supervisor; the information in each sketch was identical except for race (Black female or White female). Trainees then listened to a short audiotape of a simulatedsupervision session. Finally, they completed the following surveys: the Supervisor Rating Form (short version), Expectations of Supervisor Behaviors questionnaire, the White Racial Identity Scale, and an author-generated demographic sheet.A canonical correlation was performed to answer the major hypotheses of this study. The results indicated that supervisor race and two subscales of White racial identity, Disintegration and Autonomy, were significant predictors of perceptions of supervisor attractiveness and evaluative supervisory behaviors. Of five canonical roots calculated, this was the only one that was significant.A oneway MANOVA was also computed to test the significance of the supervisor race. The results indicated that the Black supervisor received higher scores on both the supportive and evaluative supervisory behavior subscales than did the White supervisor. Finally, only two other trends were noted. First, female trainees scored both supervisors significantly higher than did male trainees on perceived expertness, attractiveness, and trustworthiness. Second, male trainees scored the Black supervisor significantly lower than the White supervisor on perceived expertness, attractiveness, and trustworthiness.The significance of these findings for research and practice, and the limitations of the present study are discussed in the last chapter of this dissertation.60 / Department of Counseling Psychology and Guidance Services