The primary objective of this study was to improve the characterization of particulate matter emissions from burning incense. Emissions of particulate matter were measured for 23 different types of incense using a cyclone/filter method. Emission rates for PM2.5 (particulate matter less than 2.5 microm in aerodynamic diameter) ranged from 7 to 202 mg/h, and PM2.5 emission factors ranged from 5 to 56 mg/g of incense burned. Emission rates were also determined using an electrical low pressure impactor (ELPI) and a small electrostatic precipitator (ESP), and emission rates were compared to those determined using the cyclone/filter method. Emission rates determined by the ELPI method were consistently lower than those determined by the cyclone/filter method, and a linear regression correlation was found between emission rates determined by the two methods. Emission rates determined by the ESP method were consistently higher than those determined by the cyclone/filter method, indicating that the ESP may be a more effective method for measuring semivolatile particle emissions. A linear regression correlation was also found between emission rates determined by the ESP and cyclone/filter methods. Particle size distributions were measured with the ELPI, and distributions were found to be similar for most types of incense that were tested. Size distributions by mass typically ranged from approximately 0.06 to 2.5 microm in aerodynamic diameter, with peak values between 0.26 and 0.65 microm. Results indicated that burning incense emits fine particulate matter in large quantities compared to other indoor sources. An indoor air quality model showed that indoor concentrations of PM25 can far exceed the outdoor concentrations specified by the US EPA's National Ambient Air Quality Standards (NAAQS), so incense smoke can pose a health risk to people due to inhalation exposure of particulate matter. Emissions of carbon monoxide (CO), nitric oxide (NO), and sulfur dioxide (SO2) were also measured for seven types of incense. Emission rates of the gaseous pollutants were sufficient to cause indoor concentrations, estimated using the indoor air quality model, to exceed the outdoor concentrations specified by the NAAQS under certain conditions. However, the incense samples that were tested would fill a room with thick smoke under these conditions.