Four issues raised by Butler's (1989) commentary are addressed. The first issue is the possibility that the results of perceptual studies of tonal hierarchies can be attributed to task-specific strategies developed in response to particular stimuli. Such strategies cannot account for the convergence across experiments employing varied tasks and stimulus materials. The second issue is the correspondence between statistical summaries of music and perceptual data. The correspondence is shown to be quite general and to have implications for the acquisition of tonal knowledge. The third issue is the process listeners use to identify the tonal center. Patternmatching to tonal hierarchies is shown to be a plausible process contributing to key-finding, whereas a tritone rule has limited applicability. The final issue is the effect of temporal order on pitch perception. Principled temporal-order effects are found in many psychological experiments, but not in those focusing on the tritone relation.