To clarify to what extent death rate reflects the incidence of colorectal cancer, we calculated the age-adjusted death rates for colorectal cancer among municipalities in Aomori Prefecture for the period between 1983 and 1987, and examined the correlation with the age-adjusted incidences of the disease during different 5-year periods. In males, the age-adjusted death rate of rectal or colon cancer most correlated with the age-adjusted incidence in the preceding period by three or two years, respectively; however in females, the age-adjusted death rate best reflected the age-adjusted incidence of the same period. In males, significant correlation was lost if the interval exceeded seven years in both rectal and colon carcinomas. In females, the correlation became insignificant after an interval of four and five years in rectal and colon cancers, respectively.