The purpose of this study was to investigate the relationship that exists between grades in engineering drawing, scores on a spatial relations tests, and in mechanical drawing. Two groups of 425 cadets at the U.S. Naval Academy were studied and comparisons were made by means of critical ratios, analysis of covariance, and item analysis. Cadets who had had previous training in mechanical drawing had, on the average, equal test scores with those who had not had such training, but the group with training received significantly better course grades in engineering drawing and the slopes of the regression lines for predicting the grades by the tests were significantly different.