SEVERAL CONCLUSIONS FROM RESEARCH ON VISUALS ARE DISCUSSED. THE PARTICULAR STUDIES ON WHICH THESE CONCLUSIONS ARE BASED ARE TWO MULTIVARIATE EXPERIMENTS WHICH USED PROGRAMED SCIENCE DEMONSTRATIONS PRESENTED OVER TELEVISION AND PROGRAMED VERBAL MATERIALS ON THE SAME SCIENCE PRINCIPLES BUT PRESENTED IN SELF-PACED BOOKLETS. THE LEVEL OF THE PRESENTATIONS WAS DESIGNED FOR EIGHTH GRADERS. THE CONCLUSIONS ARE--(1) IT IS FEASIBLE TO BRING STUDENT RESPONSES UNDER CONTROL OF VISUAL PRESENTATIONS. (2) THIS KIND OF CONTROL IS BETTER ESTABLISHED BY HAVING STUDENTS ACTIVELY PRACTICING RESPONSES TO FEATURES OF THE PRESENTATION RATHER THAN PASSIVELY WATCHING THE PRESENTATION. (3) IF VISUAL PRESENTATIONS ARE USED AS A MEANS OF TEACHING CONCEPTS, IT IS PREFERABLE TO HAVE THE VISUAL PRESENTATION PRECEDE A VERBAL PRESENTATION. (4) WHEN THE VISUAL PRESENTATION DESIGNED TO TEACH CONCEPTS DOES PRECEDE THE VERBAL PRESENTATION, IT IS PREFERABLE THAT STUDENTS SELECT RESPONSES FROM AMONG PICTORIAL OPTIONS RATHER THAN FROM VERBAL OPTIONS. IN OTHER WORDS, CONCEPT ACQUISITION IS BETTER IF STUDENTS RESPOND FIRST TO CONCRETE SITUATIONS AND THEN TO MORE ABSTRACT TREATMENT OF THE SAME PHENOMENA. (5) IT IS POSSIBLE TO INTEGRATE FIXED-PACED VISUAL PRESENTATION, WHETHER ON FILM OR TELEVISION, WITH SELF-PACED PROGRAMED VERBAL MATERIAL IF THE FIXED-PACED DEMONSTRATIONS AND THE SELF-PACED VERBAL PROGRAMS WHICH FOLLOW THEM ARE BROKEN DOWN INTO SEGMENTS. THIS PAPER WAS PRESENTED TO THE DAVI CONVENTION (MILWAUKEE, APRIL 1965).