A meta-analysis procedure was used to review research on pictorial effectiveness which focused on the use of static iconic visuals in instructional materials. The purpose of this exploratory study was to provide a means for forming future hypotheses based upon a quantitative aggregation of past research. The study was concerned with differential instructional effectiveness, as measured by cognitive dependent outcomes attributable to illustration iconicity. Both automated information retrieval and manual search procedures of selected indices and references were used to identify a final sample pool of 121 data sets. Four major variables--illustrations, pacing, grade level, and achievement--and five physical attributes of treatment illustrations--production, shading, context, embellishment, and chroma--were coded. A jackknife technique was used to average 2,607 effect-size values across all studies and all variables. Results showed that illustrated treatments were more effective than verbal treatments, illustrations were most effective with secondary students, and externally paced illustrations were more effective than internally paced illustrations; color illustrations were more effective than black and white. (LMM)