This study investigates the relationship between ambient fine particle pollution and impaired cardiac autonomic control in the elderly. Heart rate variability (HRV) among 56 elderly (mean age 82) nonsmoking residents of a retirement center in Baltimore County, Maryland, was monitored for 4 weeks, from July 27 through August 22, 1998. The weather was seasonally mild (63-84 degrees F mean daily temperature) with low to moderate levels of fine particles (PM2.5 < 50 micrograms/m3). Two groups of approximately 30 subjects were examined on alternate days. A spline mixed-effects model revealed a negative relationship between outdoor 24-h average fine particulate matter (PM2.5) and high-frequency (HF) HRV that was consistent with our earlier Baltimore study for all but 2 days. These 2 days were the only days with significant precipitation in combination with elevated PM2.5. They were also unusual in that back-trajectory of their air masses was distinctly different from those on the other study days, emanating from the direction of rural Pennsylvania. Mixed-effects analysis for all 24 study days showed a small negative association of outdoor PM2.5 with HF HRV (-0.03 change in log[HF HRV] for a 10 micrograms/m3 increment in PM2.5) after adjustment for age, sex, cardiovascular status, trend, maximum temperature, average dew point temperature, random subject intercepts, and autocorrelated residuals. After excluding study days 4 and 5, this association was strengthened (-0.07 change in log[HF HRV] for 10 micrograms/m3 PM2.5, 95% CI -0.13 to -0.02) and was similar to that obtained in an earlier study (-0.12 change in log[HF HRV] for a 10 micrograms/m3 increment in outdoor PM2.5, 95% CI -0.24 to -0.00) [Liao D., Cai J., Rosamond W.D., Barnes R.W., Hutchinson R.G., Whitsel E.A., Rautaharju P., and Heiss G. Cardiac autonomic function and incident coronary heart disease: a population-based case-cohort study. The ARIC Study. Atherosclerosis Risk in Communities Study. Am J Epidemiol 1997: 145 (8): 696-706]. Acute (1 to 4 h) previous PM2.5 exposure did not have a stronger impact than the 24-h measure. A distributed lag model incorporating the six preceding 4-h means also did not indicate any effect greater than that observed in the 24-h measure. This study is consistent with earlier findings that exposures to PM2.5 are associated with decreased HRV in the elderly.