This Graduate Record Examination (GRE) study assesses: (1) the relative contribution of a vocabulary score (consisting of GRE General Test antonyms and analogies) and a reading comprehension score (consisting of GRE sentence completion and reading comprehension sets) to the prediction of self-reported undergraduate grade point average (GPA); and (2) criterion-related validity patterns for item-type part scores on the GRE quantitative and analytical measures. Data from GRE files for 9,375 examinees in 12 fields of study representing 437 undergraduate departments from 149 colleges and universities were standardized within each undergraduate department, and then pooled for analysis by field. There were differences by major field in average performance on the various item-type part scores within each test. The reading comprehension subtest carried most of the predictive load in the GRE verbal measure. Item-type part scores on the other measures also exhibited differential patterns of relationships with the self-reported undergraduate grade point average. The findings suggest that the different item types within the respective broad ability measures may be tapping somewhat unique skills and abilities and that further exploration of their potential contribution is in order. (Author/BS)