Leaves that show paraheliotropic movement are inclined at a shallow angle with respect to the horizontal near dawn and dusk, and at a steep angle near solar noon. However, the mean inclination angle is only a first approximation to the distribution of leaf angles, and, as such, is not a good measure of the complexity of the underlying distribution. In order to ascertain the accuracies of statistical summaries of angular data used to measure the orientation of plant leaves for intercepting light, I used cellulation plots to analyze the distribution of leaf angles or cosines of the angle of incidence. Examination of these plots shows that inclinations of leaves in the top canopy layer of individual heliotropic plants are distributed over a wide range of angles during a period of several hours before and after solar noon, but that there is a sharp lower limit to this distribution of angles. Experimental observations show that near midday leaves are inclined at least at some minimum angle with respect to the horizontal. In order to interpret these data, I used a simple model of plant canopy photosynthesis and optimized the distribution of leaf angles to maximize total canopy photosynthesis. The results of this model suggest that the existence of a limit angle may be a function primarily of stress factors, and that the distribution of leaf angles more vertical than the limit angle may be important primarily for maximization of whole—plant photosynthesis.