BACKGROUND Coronary heart disease is the major cause of mortality in the United States. Factors associated with coronary risk are important to identify. Coronary mortality is greater during the winter months. OBJECTIVE To investigate whether declining coronary mortality has been accompanied by a change in the seasonal pattern and to investigate the hypothesis that diminishing exposures to environmental cold and heat have affected the seasonal pattern. METHODS We used published data on coronary mortality by year to evaluate the time trend in the seasonal pattern. We fit a sine curve to the monthly frequency of deaths in each year and examined the trend over time in the ratio of the peak to the trough of the curve. METHODS We used monthly coronary deaths in the United States from 1937 through 1991. Deaths by cause and month were not available by geographic area within the United States, but we were able to examine total monthly deaths in 2 regions with contrasting climates, New England and the South. METHODS We used the yearly peak-to-trough ratio as our primary outcome and assessed its trend over time by linear regression. We also depicted the time trends using polynomial smoothing. RESULTS The peak-to-trough ratio diminished by about 2% per year until around 1970, when the trend reversed. In New England, the decline was steeper than in the South, as measured from all deaths. CONCLUSIONS Seasonal patterns in coronary mortality in the United States have changed with time. These changes are compatible with the gradual expansion of adequate heating and the subsequent increased use of air-conditioning. Microclimatic influences on coronary mortality could explain in part the socioeconomic gradient of cardiovascular mortality.