Recent evidence has suggested that woodsmoke exposure in developed countries is associated with acute and chronic health impacts. Accordingly, it is increasingly important to investigate the much higher woodsmoke exposures associated with the use of wood and other biomass for cooking and heating in developing countries. Particulate concentrations were measured in rural Mexican kitchens using biomass combustion for cooking. To investigate differences in indoor particle concentrations between kitchens using different fuels and stove types, measurements were made in eight kitchens using only biomass, six using only liquefied petroleum gas (LPG), six using a combination of biomass and LPG, and three using biomass in ventilated stoves. Outdoor samples were collected at the same time as the indoor samples. PM(10) and PM(2.5) measurements were made with inertial impactors, and particle light scattering was measured continuously with an integrating nephelometer. Nephelometer and particulate mass measurements were highly correlated (r(2) of 0.9 and 0.83 for PM(2.5) and PM(10), respectively), indicating that the light scattering measurements could be used to estimate shortterm concentrations. PM(10) and PM(2.5) concentrations (mean concentrations of 768 and 555 mu g m(-3) respectively) in the kitchens burning only biomass were greater than in all other types (biomass > biomass + LPG > ventilated > LPG > outdoor). A similar trend was evident for the indoor/outdoor concentration ratio. Based on the short-term measurements estimated from the nephelometer data, PM(10) and PM(2.5) cooking period average and ii-min peak concentrations were significantly higher (p < 0.05) in kitchens using only biomass than in those using LPG, a combination of LPG and biomass, or a ventilated biomass stove.