Asbestos emissions from automobile brakes were measured under conditions simulating downtown city driving. Reference data for city driving was obtained by measuring vehicle speed, time of brake application, cool-down time between applications, brake hydraulic-line pressure and pad temperature. Data from 1800 braking applications were then analyzed to provide a statistical distribution of representative braking cycles.We constructed a computer-controlled brake emission test rig that simulated road braking operation of a front wheel disc brake and collected airborne wear debris. Representative braking cycles were programmed on this system to experimentally estimate brake emissions under a variety of braking conditions.Realistic braking operations produced particle and asbestos emission rates in close agreement with those measured by Williams and Muhlbaier. Asbestos concentration was not correlated with mechanical work done in braking.Application of the asbestos emission rates to air quality models confirmed Williams and Muhlbaier's finding that about 1% of asbestos fibers in city core districts originated from disc brake wear.