RESEARCH WAS CONDUCTED TO CONTRAST THE EFFECTS OF TWO TYPES OF TEACHERS, LIKED AND DISLIKED, ON THE LEARNING BEHAVIOR OF THEIR STUDENTS. TEACHERS PRESENTED MESSAGES BY FILM, BY TAPE, AND IN PERSON IN EXPERIMENTAL CLASSROOMS TO STUDENTS FITTED WITH FINGER ELECTRODES. CHANGES IN ELECTRICAL RESISTANCE WERE RECORDED OF GALVANIC SKIN RESPONSES. ACHIEVEMENT TESTS WERE ALSO ADMINISTERED. MEASUREMENTS INCLUDED (1) PHYSIOLOGICAL AROUSAL, (2) RATINGS OF THE TEACHER, (3) RATINGS OF THE SUBJECT MATTER, (4) SCORES ON ACHIEVEMENT TESTS, AND (5) SCORES ON TESTS OF INFERENCE. RESPONSES, RATINGS, AND SCORES OF BOTH COLLEGE AND HIGH SCHOOL STUDENTS WERE STUDIED. AMONG THE SEVERAL FINDINGS WERE (1) STUDENTS RECEIVED SIGNIFICANTLY HIGHER SCORES ON TESTS BOTH OF FACTS AND OF INFERENCE UNDER THE"DISLIKED" TEACHERS WHEN SUBJECT MATTER WAS PRESENTED IN PERSON AND ON FILM, AND (2) STUDENTS ACHIEVED HIGHER SCORES UNDER "LIKED" TEACHERS ONLY WHEN SUBJECT MATTER WAS PRESENTED BY TAPE RECORDING. FOR ANY GIVEN PRESENTATION, NO RELATIONSHIP WAS FOUND BETWEEN HOW STUDENTS RATED THE TEACHER AND AMOUNT LEARNED. SUBJECT MATTER RATINGS APPEARED TO HAVE MUCH MORE BEARING ON ACHIEVEMENT THAN DID THE FACTOR OF TEACHER LIKEABILITY. (RS)