The purpose of this study was to assess the contributions of period and birth cohort effects to changes in cerebrovascular disease (CVD) mortality in Spain over the period 1955-1992. Poisson regression models were fitted to age-and sex-specific CVD mortality rates obtained from National Vital Statistics. In the period 1955-1975, CVD mortality remained stable. In the period 1975-1992, CVD mortality declined by 54% (rate ratio, RR: 0.46; 95% confidence interval, CI: 0.43-0.49) in males and 62% (RR: 0.38; 95% CI: 0.34-0.42) in females. The cohort effect was very small up to the generation born in 1905, moving clearly downward thereafter. CVD mortality for subjects born in the period 1945-1949 was lower than for those born in the period 1905-1909 by 68% (RR: 0.32; 95% CI:0.16-0.63) in males and 82% (RR: 0.18; 95% CI: 0.07-0.45) in females. Among the possible partial explanations for these effects are the decline in ischemic heart disease and rheumatic fever mortality, the drop in salt and alcohol intake, the reduction in smoking among males and blood pressure among females, and the widespread use of antihypertensive treatments in Spain over the last 20 years.