In 1980-81, carbon monoxide (CO) exposures were measured inside a motor vehicle during 88 standardized drives on a California urban arterial highway, El Camino Real (traffic volume of 30,500-45,000 vehicles per day), over a 13-1/2 month period, yielding the first known data set of vehicular exposures over an entire year. Because of the highly standardized nature of the approach, it was possible to repeat the entire experiment in 1991, more than a decade later. Similar driving patterns were maintained, and data were collected on the Surrounding Vehicle Count, a measure of traffic congestion at intersections along the way. Over the 11-year period, the exposures have decreased by 53%. A vehicular emissions model developed for the highway predicted a decline in emissions of approximately 52%, which agrees well with the change in exposures. In contrast, total emissions for the region, which takes into account increases in vehicle-miles due to additional highways, declined only 33%. Ambient concentrations decreased approximately 34% over the 11-year period. The results suggest that CO exposures measured on highways are a sensitive indicator of the effectiveness of vehicular emission control systems.