A relationship between seasonal climate changes and the occurrence of stroke has been postulated. We reviewed the seasonal occurrence of stroke in 2,960 patients seen at the University of Iowa from 1978 through 1985. Stroke was classified as cerebral infarction (CI; n = 1,357, 46%), transient ischemic attacks (TIAs; n = 913, 31%), subarachnoid hemorrhage (SAH; n = 476, 16%) and intracerebral hemorrhage (ICH; n = 214, 7%). Local climatological data were obtained from the National Oceanic and Atmospheric Administration. Poisson regression was used to analyze the data. The occurrence of TIA and SAH was not influenced by seasonal climatic variables. There was a significant increase in the rate of referral for CI during warmer months (p = 0.027). The amount of rainfall did not influence the rate of CI. Conversely, the rate of referral for ICH was significantly less during warm weather (p = 0.027) and rainy weather (p = 0.014). A possible inverse seasonal relationship in temperate climates between CI and ICH deserves more investigation.