This study was designed to allow systematic comparison of exposure on public (40-seater buses) and private (four passengers cars) transport modes for carbon monoxide (CO), methyl-tertiary butyl ether (MTBE), and benzene by carrying out simultaneous measurements along the same routes. There were statistically significant differences (p < 0.05) in the concentrations of all target compounds among the three microenvironments; inside autos; inside buses; and in ambient air. The target compounds were significantly correlated for all the three environments, with at least p < 0.05. The in-vehicle concentrations of MTBE and benzene were significantly higher (p < 0.0001), on the average 3.5 times higher, in the car with a carbureted engine than in the other three electronic fuel-injected cars. On the other hand, the CO concentrations were not significantly different among the four cars. The in-auto MTBE levels (48.5 micrograms/m3 as a median) measured during commutes in this study was 2-3 times higher than the New Jersey and Connecticut's results. The in-auto concentration of CO (4.8 ppm as a median) in this study was comparable with those in later studies in some American cities, but much lower than those in earlier studies in other American cities. The in-bus CO concentration was 3.6 ppm as a median. As a median, the in-auto concentration of benzene was 44.9 micrograms/m3, while the in-bus concentration 17.0 micrograms/m3. The in-auto/in-bus exposure ratios for all the target compounds was 31-40% higher than the corresponding concentration ratios, due to the higher travel speed on buses in the specified commute route as compared to the autos.