We investigated the association between rice burning and daily asthma hospitalizations in Butte County, California, from 1983 to 1992. Eighty-two percent of planted rice was burned, with a mean of 555 acres burned on days when burning was permitted. For 60% of the days during this period, no rice burning occurred. Peak burning occurred in fall and spring but was not correlated with criteria pollutants. Asthma admissions averaged 0.65/day and peaked in March. In the basic Poisson model with daily asthma hospitalizations as the outcome of interest, burn acreage showed a small but statistically significant elevation of risk for hospitalization per acre of rice burned [relative risk (RR) = 1.0001; 95% confidence interval (CI), 1. 00004-1.0002], after adjusting for maximum daily temperature, seasonal factors, and yearly population. In this model, burn acreage showed a dose-response effect as acreage burned increased. Days with the greatest acreage burned (>499 acres) had the largest risk of hospitalization (RR = 1.23; CI, 1.09-1.39), and days with moderate burning (between 100 and 499 acres) had a slightly lower risk of admission (RR = 1.2; CI, 1.05-1.37). Elevations of air pollutants were not associated with days of increased rice burning; however, rice burn acreage was shown to have a small but statistically significant effect on asthma morbidity in Butte County. This evidence suggests that further limitations on the daily amount of rice straw permitted to be burned should be considered to reduce pulmonary morbidity related to asthma.