This study was designed to determine if a female physical educator's appearance of body fatness affects 1) the cognitive performance of junior high school students on a test of health-related fitness knowledge, and 2) student attitudes toward the instructor. Ninety-three seventh through ninth grade students from two schools, participated in a six-week instructional course on health-related fitness knowledge. Instruction was identical with one exception: In one group, the instructor wore an under garment to present an image of an obese physical educator. A pretest-posttest health-related fitness examination was administered to the students. Additionally, student attitudes toward the physical educator were assessed during week one, week three, and week six. Analysis of covariance revealed a significant group x posttest score interaction, F [subscript 1, 88] = 4.408 p = 0.039. The student attitude questionnaire (SAQ) sums were analyzed for all participants and detected no significant group x time interaction. However, an SAQ sum score main effect was found to exist, F [subscript 2, 75] = 10.056, p = 0.000. Paired t-tests found significant differences between all combinations of SAQ mean scores, except for the values of week one to week three. Within group paired t-tests were computed for the SAQ to detect differences independent of group. Analysis revealed significant differences in attitudes occurred between week one and week three and week one and week six, for the control group t (32) = -3.323, p = 0.002; t (32) = -3.800, p = 0.001). Similarly, for the experimental group significant difference occurred between week one and week six, t (44) = -2.517, p = 0.016. Results indicate instructor physical appearance is related to student performance on a cognitive based health-related fitness test. In contrast, instructor physical appearance does not appear related to student attitude toward the instructor. (Contains 1 table and 4 figures.)