Liquefied petroleum gas (LPG) is the main fuel used in the residential sector of the Metropolitan Zone in the Valley of Mexico (MZVM). LPG represents 16% of the total fuel consumption in the MZVM and its demand increased 14% from 1986 to 1999. Propane and butanes, the main compounds of LPG, constituted 29% of all non-methane hydrocarbons found in the air of Mexico City. Some researchers have reported that LPG losses are a significant cause of high ozone concentration in MZVM. Three analyses are carried out in this work to estimate LPG's share of responsibility for MZVM pollution problems. First, the correlation between LPG consumption and three ozone pollution indicators was calculated for the period of 1986-1999. The non-significant correlation of these indicators with LPG consumption in a monthly basis suggests that LPG associated emissions are not the foremost cause of ozone formation. Second, a simulation model is applied to three LPG related emission control strategies to estimate the reduction in the maximum ozone concentration. The most noticeable effect was obtained when both hydrocarbon (HC) and oxides of nitrogen (NO,) emissions associated with LPG use were totally reduced. The other two strategies, that only reduce HC emissions, had a minimum effect on the ozone concentration. Third, organic compounds consumption in air samples captured and irradiated in outdoor smog chambers is used to determine the chemical loss rate of LPG associated species and aromatics in the MZVM. The smog chamber results showed that 70% of propane and n-butane remain at the end of a 1-day irradiation, therefore they remain in the MZVM atmosphere for several days being the reason for the high concentration of these compounds. LPG associated compounds only account for 18% of ozone formed but aromatics contribute 35% to ozone in smog chamber. (C) 2004 Elsevier Ltd. All rights reserved.