Research was reviewed to determine the validity of aptitude tests when administered to high ability examinees, those comprising the top one percent of the general population. Data on the Scholastic Aptitude Test (SAT) indicated that the ceiling effect was not serious. In some studies, the error of measurement was smallest at the highest scores; however, the highest discrimination was not necessarily provided in these situations. Validity was examined in several long-term follow-up studies, using the Army General Classification Test, the Graduate Record Examinations, the Miller Analogies Test, the Owens Bennett Mechanical Comprehension Test, The Concept Mastery Test, College Board Achievement Tests, and the Scholastic Aptitude Test. It was concluded that, in most situations, aptitude tests do discriminate for high ability students. In theory, the ceilings of the tests can be sufficiently high to provide adequate discrimination, by including enough difficult items. On a practical level, this is uneconomical and unnecessary. Test reliability was not necessarily lower for high aptitude than average students. Results indicated that substantial validity is possible when: (1) the test is appropriate for the sample and the predictive task at hand; (2) a relevant criterion measure is available; and (3) the correlation is not reduced by criterion ceiling effect, restricted range effect, or multiple predictor effect. It is likely that aptitude measures account for fully as much variance as other measures.