The purpose of this paper is to integrate the findings of studies undertaken by Dwyer (1978) and colleagues which represent the only extended and concerted investigation of pictorial stimulus complexity as it relates to static instructional visuals. After scrutinizing all available published and unpublished reports of research in this series for appropriate data, 23 sets of complete data were found to be available. Those studies which contained data sufficient for the application of meta-analysis techniques developed by Glass (1977) comprise the sample for this investigation. Effect size measures were computed, and regression analysis was performed to examine the relationships of effect size with presentation pace, grade level, type of learning, and type of illustration. Results of the analysis are reported, and it is concluded that, although it is likely that more thorough aggregation will modify these findings, the data do provide indications of possible trends which may ultimately be sustained by more complete analyses, and which may help to generate experimental investigation in the interim. A 92-item bibliography, tables and figures referred to in the text, and a list of the publications which comprised the sample are attached. (Author/CHC)