We review the paper by Andrew Clay McGraw, noting that it represents an interesting and valuable contribution to the study of music in cognition in its informed exploration of non-western musical perceptions. We raise a number of concerns about the methods used, and make suggestions as to how the issues that were empirically addressed in the paper might have been tackled in ways that would have enhanced the interpretability of its findings. THIS fascinating paper takes as its starting point the idea that time in music and time in general, may be bound by culture in our conceptions, in particular, by processes of cross-domain mapping of the types that have been postulated as operational by ethnomusicologists, music theorists, and music psychologists. It explores by means of three experiments Becker's hypothesis that Balinese gamelan music reflects and perhaps embodies emic conceptions of time. The first uses an experimental paradigm known as the Implicit Association Test to investigate, by means of response times, the strengths of associations between musical and/or dance terms, music and/or instrument terms, and time and/or position terms, among Balinese musicians. It found evidence for a slight tendency for time-in music to be 'implicitly iconic' of time-in-general, though noted that the results showed high variability. The second experiment explored Balinese musicians' abilities to identify objective durations while listening to gamelan music; it found that subjects tended to rate examples as somewhat shorter than they actually were, a finding that contrasts with the results of a limited study of American musicians unfamiliar with gamelan who judged examples as rather longer than their actual durations. A final experiment investigates Balinese musicians' judgments concerning tempo transformations (rallentandi), finding that these listeners' preferences were equally for patterns of tempo change that previous literature suggested should be regarded as complex or 'unnatural.' These findings are discussed in the light of measurements of Balinese gamelan rallentandi that illustrate the complex, 'terraced' nature of closural tempo change that appears to predominate in this music. We must say at the outset that this is precisely the type of research that requires to be done; it takes empirical methods that have been shown to have value in elucidating the dynamics of mind in western laboratory contexts and employs (and adapts) these to explore 'cognition in the wild' in a non-western cultural context. It poses questions about relationships between emic and etic conceptions of time and music, and about the real-time experience of temporal structures in music among Balinese musicians, from what appears to be a background of real understanding of, and engagement with, the musical culture that is being explored. The provisional nature of the results, and many of the limitations of the experimental design employed, are acknowledged. The response that the paper provokes is one of admiration, and any criticism that follow in this comment should be viewed as constructive and offered with the intent of precipitating debate about just how research of this type can, or should, be conducted. To start with, while the author himself notes that he is 'not a psychologist', some assistance in designing the experiments and in analyzing and presenting the results would have been helpful. The 'readability' of the paper might have been aided by consulting on appropriate methods for presenting results. For example, Table 1 might have been presented as a boxplot, which would have aided a reader greatly, and it might be helpful to have the excerpts from experiment 2 available online for reference. The lack of statistical analysis of the results severely limits their interpretability; as they stand they are indicative and suggest lines for future research (of considerable value in itself), but the author's arguments