Mexico City frequently experiences high levels of air pollution. This is due mainly to its topography and meteorology that suppress pollutant diffusion and dispersion. The atmospheric mixing is extremely poor, especially during the dry winter months. The levels of certain pollutants, such as particulate matter, are of concern since they have severe effects on public health. Visibility deterioration is one of the most noticeable effects in large cities. Biological effects of particulate matter on man and animals, ranging from mild eye irritation to death, have been reported. The effects depend on the size of the particles, their solubility, and toxicity. The main objective of this paper is to present the results of a chemical mass balance receptor model applied to a well-characterized data set of particulate matter collected in the Mexico City Metropolitan Area (MCMA). Samples of particulate matter were collected using a denuder and a Hi-Vol system for the respirable fraction and total suspended particles, respectively. In this paper the analysis of a database consisting of the chemical composition of 33 samples of respirable particulate matter (aerosols with diameter less than 2.5 μm) is presented. The 12-hour samples were acquired during day and night periods in a typical medium-income neighborhood from December 19, 1989 through February 5, 1990. The results show that the main contributors to suspended particles are vehicles without catalytic converters and heavy-duty diesel vehicles. The contribution of refineries, smelters, cement plants, resuspended dust, natural sources, and secondary aerosols were taken into account. In particular, the vehicles without catalytic converters represent the major contribution to PM2.5. They contribute with 50% during the day and 38% at night. Most of the source profiles were taken from the model library SPECIATE EPA. However, native profiles for soil, vehicles, and refinery were designed.