Long-term aerosol measurements have been conducted at two sites in Rome, Italy, April 2001 through March 2003, in a traffic-oriented site, and at an urban background site, close to the city center. The main objective was to establish validated and consistent data sets of particle number concentrations (PNC) in Rome to be used for epidemiological analyses of cardiovascular health effects. Particle number concentrations were measured by a condensation particle counter (CPC 3022A, TSI). Other pollutants (PM10, PM2.5, CO, NO2, NO, NOx, O3) were simultaneously measured at the traffic-oriented site. During the study period, the mean (standard deviation) 24-h PNC values were 4.69 x 10(4) (1.99 x 10(4)) cm(-3) and 2.46 x 10(4) (1.10 x 10(4)) cm(-3), respectively, at the traffic-oriented site and at the urban background site. Mean (standard deviation) 24-h mass concentration of PM2.5 was 23.1 (11.9) microg m(-3), while for PM10 it was 41.3 (17.9) microg m(-3). Higher values for all the pollutants, except ozone, were recorded during the winter period in comparison with the summer period, and a higher variability of the results was also observed during cold months. The comparison between the daily PNC measured at the two sites showed a good correlation (r = .74). CO (r = .77), NO (r = .82), and NOx (r = .83) were all highly correlated with PNC (simultaneous obs. number 576). The diurnal and seasonal pattern of PNC can be attributed to the combined effect of motor vehicle emissions and meteorological conditions.