SIDS cases were defined by examining all death certificates, in which sudden deaths were expected to be found from the years 1969-80 from the Central Statistical Office of Finland. The age limits were 28-364 days. If the death certificate did not give enough information as to whether the cause of death was explained or unexplained, autopsy records and microscopic specimens were examined. If the death was sudden, but no autopsy was done, no microscopic specimens were taken, or there were some slight findings which could have partly explained the death were classified as borderline cases. The mean annual incidence of SIDS in Finland was 0.41/1000 livebirths in 1969-80. In 1969-74 and 1975-80 the incidences were 0.31 and 0.51, respectively. The increasing tendency of SIDS was partly due to more borderline cases in the first period and partly due to more twins, and infants with small birth weight, dying of SIDS in the second period. Deaths at weekends and sleeping with parents in the second period were more common than in the first study period. In the SIDs group the young maternal age, low social class, family type unmarried couple or single mother, maternal anemia during pregnancy were more common than in the control group. Mothers of SIDS infants had more previous children and fewer visits and later first visit to prenatal clinics than control mothers. The duration of gestation was shorter and the mean birth weight and length were smaller in the SIDS case than in the control group. Twins were more common among SIDS infants than in the common population. The most important risk factor of SIDS was maternal smoking during pregnancy. The epidemiological results conform with the hypoxia hypotheses.