Thirty years of empirical research on visual aids have produced inconsistent results--perhaps because the quality and color of those visuals were inconsistent. The purpose of this study was to determine what effects quality and color of instructional transparencies used in an informative speech have on listener recall and attitudes toward speaker and speech. A total of 709 community college and university students in 36 intact classes were randomly assigned to one of four visual treatment groups (poor-quality black/white, high-quality black/white, poor-quality color, or high-quality color) or one of two control groups (no-speech or no-visuals). A videotaped speech was projected onto a large screen at the right of the room; visual aids (each shown for approximately 30 seconds) were projected onto another screen set immediately to the left. Recall was measured by a 10-item multiple choice test; attitude toward speech and speaker were each measured by six seven-item semantic differential scales. Analysis of variance indicated that the type of transparencies used in an informative speech have a definite effect on immediate recall and attitude toward the speech, but no effect on attitude toward speaker. All four treatment groups scored significantly higher on recall than the no-speech and no-visual control groups. Log percent of change showed poor color to produce the lowest scores (still 13% better than control) and quality color to produce the highest scores (19.5'% better). Analysis found listeners to have a more positive attitude toward the speech when quality color, quality black/white, or poor black/white visuals were used. It appears that any visual (even a poor quality one) produces better recall than no visuals. Speakers with the time to produce quality visuals should add color; speakers who pay little attention to quality would be advised to use black/white visuals. Implications for future research are suggested.