OBJECTIVE To study the articular morphology (radius of curvature), (diameter, depth, circularity, and percent of circle) of the capitate, proximal lunate, and the lunate fossa of the distal radius using both magnetic resonance imaging (MRI) scans and plain radiographs. The correlation between plain radiographs and MRI scans for these measurements will also be assessed. METHODS Twenty MRI scans and 17 sets of radiographs of asymptomatic volunteers were evaluated. Standardized surface landmarks were digitized and measured in both the sagittal and coronal planes. The parameters of interest were calculated from the digitized data using specialized software. RESULTS Using MRI data, we determined the radius of curvature of the capitate to be only 37% +/- 10 of the lunate fossa of the distal radius on the coronal (anteroposterior) view and to be 57% +/- 10 on the sagittal (lateral) view. In both planes, the proximal lunate had a significantly larger diameter and radius of curvature than did the capitate. The ratio of the radius of curvature of the proximal capitate to the proximal lunate on the coronal projection ranged from .366 to .811, and on the sagittal projection the values ranged from .46 to .71. Plain radiographs were not sufficiently accurate to determine the radius of curvature ratio of the capitate to the lunate or to the lunate fossa of the distal radius on the coronal view based on a comparison with MRI data. Plain radiography did not correlate with MRI for most clinically relevant parameters. CONCLUSIONS The articular morphology of the capitate does not closely correspond with that of the lunate fossa when compared with the proximal lunate articular surface. Based on observed variations in capitate morphology and the potential for associated alterations in joint contact forces after proximal row carpectomy, preoperative MRI may facilitate the selection of patients with more favorable capitate morphology.