The studies reported here assessed pathophysiologic mechanisms that result from exposure to concentrated ambient particles (CAPs) in animals with and without cardiopulmonary compromise. These studies were carried out to determine the biologic plausibility of epidemiologic observations of increases in particulate air pollution associated with increases in human morbidity and mortality. Dogs were exposed two at a time to CAPs or filtered air via tracheostomy for six hours per day on three consecutive days. The electrocardiogram (ECG) and breathing pattern were recorded continuously, and indicators of inflammation were also assessed. In one experimental design, normal dogs were exposed in pairs to CAPs and subsequently to filtered air or to filtered air and subsequently CAPs (the double CAPs/double sham design). Comparisons were made between the CAPs measurements and each dog's own sham responses. In another design, one dog was exposed to CAPs while the chambermate received a sham exposure; these experiments were followed by crossover of the protocol the subsequent week (the crossover design). Comparisons were made between the CAPs exposure and both the chambermate's sham and each dog's own sham responses. The crossover experiments were conducted in normal animals and in animals who had undergone balloon occlusion of the left anterior descending (LAD) coronary artery to induce myocardial compromise. The effects of CAPs in animals with induced chronic bronchitis were part of the original specific aims; because these studies were not fully pursued, the results are presented only in Appendix A. In normal dogs, analyses of all double CAPs and crossover studies revealed low frequency (LF) and high frequency (HF) powers for heart rate variability (HRV) that were significantly higher for CAPs exposure compared to sham exposure. Variation in day-to-day exposure concentrations, aerosol composition, and pathophysiologic responses were also found. The crossover design, continuous measures of aerosol mass, and biologic responses were incorporated in the development of a statistical model that allowed isolation of changes associated with CAPs from changes due to animal variations. Comparison of individual exposures with this model revealed a range from no response in any measured parameter to statistically significant changes in cardiac autonomic balance, pulmonary air flow, and breathing pattern. On days in which dogs showed statistically significant changes in responses, the findings were consistent in both cardiac and respiratory parameters. Days associated with significant increases in LF and HF HRV, LF/ HF HRV ratio, and heart rate standard deviation (HR SD) were also associated with decreases in average heart rate. These same days had decreases in respiratory frequency, tidal volume, minute volume, and peak flows with corresponding increases in respiratory cycle times and enhanced pause (Pauenh), a measure of bronchoconstriction. These cardiac and respiratory changes suggest an effect mediated via both the sympathetic nervous system and the vagus nerve. Alternatively, days associated with increased heart rate had decreases in the HR SD; decreases or no change in HF and LF HRV; increases in respiratory flows and volumes; and decreases in breathing cycle times, all suggesting only sympathetic nervous system mediation. When all data from the crossover design experiments were assessed with this model, the heart rate and respiratory rate were significantly decreased in relation to both cumulative and actual exposure and the LF HRV, LF/HF HRV ratio, HR SD, and all other respiratory parameters were significantly increased (p < 0.0001 for all). When cardiac data were grouped by days in which the air mass trajectory came from the north or northwest (versus west, south, east, or northeast), significant increases in HR SD and HF HRV and significant decreases in average heart rate were associated with the northwest trajectory. (ABSTRACT TRUNCATED)