Understanding the role that nymphal and female ticks, Ixodes scapularis, have in the epidemiology of Lyme disease is essential to the development of successful prevention programs. In this study, the authors sought to evaluate the seasonal and annual relations between tick densities and patients > or = 16 years of age diagnosed with erythema migrans (EM), the rash associated with early Lyme disease. Ticks were collected weekly by drag sampling throughout most of the year from 1991 to 1996 in Westchester County, New York. The number of EM cases was based on patients diagnosed at the Westchester County Medical Center using Centers for Disease Control and Prevention (CDC) criteria. No patients with EM were diagnosed from January through April, when only adult ticks were active. Correlation analysis between monthly tick densities and EM incidence was significant for nymphs (r = 0.87, p < 0.01), but not for adult ticks (r = -0.57, p > 0.05). There was a strong, although not significant, correlation between peak annual number of patients with EM and peak nymphal tick abundance (r = 0.76, p = 0.08). These data indicate that bites from adult I. scapularis only rarely result in Lyme disease, and that annual nymphal tick abundance determines exposure. This suggests that annual fluctuations in Lyme disease case numbers are largely due to natural changes in tick abundance and, therefore, that control of nymphal I. scapularis should be a major component of Lyme disease prevention efforts.