ONE OF THE more important questions regardI ing an aptitude test is whether or not training affects scores. This is true not only for the theorist who is concerned about the innateness of abilities, or the definition of them, but also and especially for the practical test-user who will select prospective trainees on the basis of test performance. The pur pose of the present study is to examine the relation of training to the Drake Musical Aptitude Tests (1) in a public school system, a setting which makes possible a controlled comparison of differing levels of musical experience. Published research on the topic of training and musical aptitude is meager. Music students at the Eastman School of Music were tested at three-year intervals with intensive training interpolated and showed little change in scores on the well-known Seashore measures (4). Drake (1) found low corre lations between number of years of musical train ing and scores on the Drake Musical Aptitude Tests. Although both findings were interpreted as indicat ing little training effects, the question is still in doubt. For one thing, there were no control groups against which to compare scores after some tr a in ing interval. Further, the music students in the Eastman study represent a highly select group, most of whom have probably had considerable musical experience prior to their first test. If training ef fects are, like the typical teaming curve, negative ly accelerated, which is likely, little d if fe rene e would be expected with highly-trained subjects. In any event, these results shed little light on the e f fects of training upon scores of unselected subjects such as beginning and prospective music students in the elementary and secondary schools. One investigator recently attempted a controlled comparison of musical aptitude using test-score changes as a function of training. Gordon (2) pr e tested high-school students with the Drake T ests, trained them for one month on similar material,and [ compared score changes with a no-training group. | He found differences between preand posttests reflecting training effects on Musi cal Memory scores. The differences failed of significance,how ever, probably due to an extremely small number of subjects. In the present study, the relation between train ing and Drake Test scores is examined. The design, utilized in the context of a public elementary and secondary school system, has these features: (a) the lower range of level of musical training is stud ied; (b) the interval between preand posttest ex ceeds two years; and (c) a no-training group is pro vided for comparison. These features offer the best possiblity for demonstrating a training compo nent in test scores if it exists, and consequently of providing not only the aptitude theorist but the test user with useful information regarding inter pretation of the tests. Specifically, the questions not previously answered but addressed here are these: Does early training raise scores? How do these changes compare with no-training-group changes? Do training effects decrease with amount of training?