BACKGROUND: Using a population-based stroke registry system, we evaluated the relationship between ambient temperature parameters and stroke incidence in a Japanese population.Methods and Results:We analyzed data from the Takashima Stroke Registry, which records all stroke occurrences in Takashima City, Japan. The study period of 8,401 days was divided into quintiles of daily weather parameters, and the middle quintile was used as the reference category. Incidence rates (IR per 100,000 person-years) were calculated across the quintiles. Poisson regression analysis was used to calculate the effect of temperature parameters on stroke incidence. There were 2,405 first-ever strokes (1,294 men), including 1,625 ischemic, 545 cerebral hemorrhages, 213 subarachnoid hemorrhages, and 22 unclassified strokes. The stroke IR was higher in the middle quintile of average temperature, 357.3 (328.4-388.8), and for other parameters. After adjustment for age and sex, for all stroke, the incidence rate ratio (IRR) in the highest (Q5: IRR 0.81, 95% confidence interval (CI) 0.71-0.92) and the second-highest (Q4: IRR 0.80, 95% CI 0.71-0.91) quintile was lower than that in the middle quintile (Q3: Reference). Analogous results were observed for the minimum, maximum, and lag-days temperatures, also in the subtypes and across ≥65 years of age, also in females. CONCLUSIONS: Higher temperatures, irrespective of the parameter (average, minimum, or maximum), had a protective effect against stroke occurrence in Japan.