We made a clinical and statistical evaluation of the occurrence of acute myocardial infarction with respect to the relation between its occurrence and the meterology of the cold inland area of Hokkaido (the Kamikawa Basin) over a period of 10 years (1976-1985). A total of 581 cases were studied. Monthly fluctuation of incidence was not found to be statistically significant. A cold period in the Kamikawa Basin was defined in this study as the period when ordinary mean atmospheric temperatures were below 0 degree C (from 7 Nov. to 16 Apr.). Canonical discriminant analysis was applied to 10 meterological factors between the days with occurrences and those without occurrences (245 days vs 245 days) in the cold periods of the investigated 10 years, and between the days with outdoor occurrences and those without occurrences (37 days vs 37 days). In order to compare the regional difference, this analysis was done on the same 10 factors for the cold periods over 3 years in Yamagata (46 days vs 46 days). The F values of 0.0003, 0.0155 and 0.0098 respectively in the above 3 analyses were small (much less than F 1(9) (0.25) = 1.51). A circadian rhythm of 2 cycles/day was recognized concerning the time of occurrence by power spectral analysis of the data of 562 patients for whom the time of the onset of myocardial infarction was known. Subdividing the patients into 2 groups according to physical activity just before the occurrence, the group who experienced an occurrence at rest showed a rhythm of 1 cycle/day, and the group who experienced an occurrence on effort showed a rhythm of 2 cycles/day. Therefore, the 10 meterological factors could not discriminate the probabilities between the days with occurrences and the days without occurrences of myocardial infarction in the cold periods. On the other hand, it was suggested that biological intrinsic rhythm participates in triggering the occurrence of myocardial infarction.