A recent publication extends previous findings on mortality associated with hot-weather episodes and reports estimates of average hot-weather-related mortality in 44 U.S. metropolitan areas. A geographic pattern emerges in the results: the highest hot-weather-related mortality rates are in northern metropolitan areas, even though average Summer temperatures are higher in southern metropolitan areas. This paper reports the results of regression analysis used to identify weather and socioeconomic characteristics of the 44 metropolitan areas that may explain the differences in hot-weather-related mortality. The results show that variability in minimum daily Summer temperatures may be one of the most important factors. This finding suggests that biological or behavioral adaptation occurs in areas that are consistently hot, but not where minimum daily temperature variability is greater. The results also suggest that differences in the availability of air conditioning, standards of living and housing quality contribute to differences in hot-weather-related mortality, but that these factors explain a much smaller share of the variation in hot-weather-related mortality than variability in minimum daily temperatures. The results suggest that whether climate change would result in higher hot-weather-related mortality may depend on the effect on the variability of minimum daily temperatures as well as on the change in absolute temperatures.