The existence of short-term links between urban air pollution and mortality is now well documented. In France, the Air and Health Monitoring Program (Psas) has been studying these associations for about a decade. The main objective of this study is to update, for the period 2000–2004, the quantification of the relationships between levels of air pollution indicators (NO2, O3, and particulate matter) in the ambient air of nine French cities and the risk of death from various causes. In addition, the links between the different particle size fractions (fine particles PM 2.5 and coarse particles PM 2.5–10) and mortality were studied for the first time in France. The analyses are based on the time-series method, which involves studying the links between short-term variations in pollution indicator levels and those in the number of deaths, after adjusting for potential confounding factors (long-term trends, seasonality, temperature, flu epidemics, holidays, etc.). They are conducted for each study area, and a combined relative excess risk (RER) associated with an increase in the pollution indicator is determined. Across all nine cities, significant associations were identified between all pollution indicators and mortality measures considered. Thus, the risk of all-cause non-accidental mortality was significantly associated with a 10 μg/m³ increase in pollution levels on the day of and the day before death, with SRRs ranging from 0.9% (95% CI=[0.4; 1.5]) for O₃ to 2.0% [0.8; 3.3] for PM 2.5-10. The observed effects were similar among people aged 65 and older. The results for cardiovascular and cardiac mortality were relatively close to those obtained for all-cause mortality, though with more pronounced effects, particularly in relation to particulate pollution indicators and among people aged 65 and older. This new analysis confirms the existence of significant links between ambient air pollution levels commonly found in nine French urban areas and the risk of death, particularly from cardiovascular causes. The standardization of methods across participating cities, as well as the consideration of potential confounding factors, are strengths of the study and lend the results a certain degree of reliability. The observed results are also consistent with those of other major North American and European multicenter studies, and with current knowledge regarding the short-term biological mechanisms by which pollutants affect health. The findings regarding the different particle size fractions contribute to the very limited existing knowledge on this subject and suggest that coarse particles, just like fine particles, may have an effect on mortality. The relationships between air pollution levels and mortality quantified in this study will enable assessments of the health impact of urban air pollution in French metropolitan areas. The evolution of the relationships observed here compared to previous PSAS studies underscores the need to maintain routine monitoring of these relationships in France. (R.A.)