This paper uses prices that were directly observed at 27 gasoline stations in Guelph, Ontario, eight times per day for 103 days in late-2005, to examine several basic predictions of a theory of price cycles. It is found that price movements in Guelph are more consistent with the Edgeworth cycle theory than with other dynamic pricing theories. The data also identify some interesting (and somewhat systematic) pricing patterns that have not been identified in previous studies, and which would likely be overlooked with less complete data. These findings are not only of interest to applied economists and policymakers, but also to theoreticians who are interested in refining the theory to make more accurate predictions.