This study took popular music from the Top 30 charts and, in a pretest, evaluated its energy and joyfulness as musical qualities. The findings were used to create sets of musical selections that were either low or high in these qualities. In the experiment proper, respondents were placed in states of bad, neutral, or good moods and then, in an ostensibly independent study, provided the opportunity to freely choose from the sets of musical selections. The selections were offered by computer software that recorded individual exposure times by selection. To ensure selectivity, exposure time was limited to about one third of the total running time of all available selections. Consistent with predictions from mood-management theory, respondents in bad moods elected to listen to highly energetic-joyful music for longer periods than did respondents in good moods. Respondents in bad moods, moreover, were more decisive in exercising their musical preferences. Following the listening period, respondents' moods did not appreciably differ across the experimental mood conditions.