Two experiments investigated the interfering eects of a manual tracking task on timing performance. Subjects generated a series of 5-s temporal productions under control (timing only) and experimental (timing + pursuit rotor tracking) conditions. Timing was more variable under experimental conditions, a ®nding consistent with attentional models which argue that timing tasks compete with concurrent distractor tasks for limited processing resources. A pretest-posttest paradigm was employed to evaluate the hypothesis that practice on the tracking task would lessen its attentional demands and thereby attenuate the interference eect. Experiment 1 involved single-task practice (tracking alone). Single-task practice leads to automaticity, the ability to perform a skilled task using fewer processing resources. Pretest-posttest comparisons showed that such practice reduced interference in timing. Experiment 2 involved dual-task practice (timing + tracking). Dualtask practice promotes the development of timesharing, the ability to eciently switch attention between multiple tasks. In this case, practice failed to reduce the interference eect in timing. The results suggest that eective strategies for timing in dual-task situations must allow one to closely monitor the ongoing¯ow of temporal events.