Magnitude judgments of the speed of self-motion were examined. The principal independent variables were edge rate, global optical flow rate, and the type of texture (grid or dot). Results indicated that edge rate and global optical flow rate had additive effects on magnitude judgments, with edge rate accounting for a larger portion of the variance. Effects were independent of texture type. Secondary variables examined were viewing condition and task load. Attempts were made to control the availability of flatness cues. Evidence indicates that the effectiveness of global optical flow rate varied with the control of flatness cues. A secondary running auditory Sternberg task (S. Sternberg, 1969) was used to prevent edge counting; the presence of this task did not reduce the effect of edge rate. These results replicate and extend previous work by D. H. Owen and colleagues (1984 and 1987).