The rationale for using the Perceptual Ability Test (PAT) as a component in admissions decisions for dental schools is that candidates vary in an underlying aptitude that is predictive of degree of success in technique course performance and perhaps in clinical performance. There have been periodic attempts to identify tests that more directly measure manual dexterity aptitude that would supplement the predictive power of admissions decisions. Previous research has demonstrated that a commercially available "speeded" tweezer dexterity test (Johnson O'Connor Test #32022) is not associated with performance in dental school or dental practice. Our research investigated both Test #32022 and Test #18 that measure both speed and accuracy as potential predictors of dental school performance in technical and clinical courses. This article reports the results of a longitudinal, comparative study of tweezer dexterity scores for students at the University of the Pacific School of Dentistry during their first and last quarters in school. The goals of the study were to 1) evaluate the correlation between beginning students' scores on two different types of tweezer dexterity tests; 2) compare dental students' scores to normative data for the general population; 3) determine the effect of a dental curriculum on students' performance on Test #18; and 4) evaluate the two tests as potential dental school admission screening instruments in comparison to the PAT. Fifty first-quarter students were tested from a class of 134. Forty-nine of these students were retested on Test #18 during their final quarter. The predictor value of the initial scores for the two dexterity tests was assessed for seven outcome measures reflecting student technique performance. Analysis showed a significant correlation (r=0.318, p<0.05) between the two dexterity tests. The difference between the norm mean (41.58) and the dental student mean for Test #18 (40.42) was not significant (p>0.05). The correlation between the first and final quarter administrations for Test #18 was r=0.517 (p<0.01). The predictive power of these tests for the seven educational outcomes measures was weak. Results suggest that dental student tweezer dexterity is no different from that of the general population and is not changed by completing a dental school curriculum. The ability for an applicant to perform successfully in dental school will not be reliably predicted by tweezer dexterity score.