The re-ingestion of toxic or odorous gases exhausted from rooftop stacks of a building may be a cause of indoor air quality problems of the same or an adjacent building. Although many experimental studies have been carried out to investigate the dispersion of exhaust from low-rise buildings, relatively little work has been conducted for high-rise buildings.  The present study examines the dispersion of pollutants from rooftop stacks on high-rise buildings and their effect on adjacent buildings. The water flume of the Building Aerodynamics Laboratory (BAL) has been used to carry out flow visualization experiments to identify building configurations that may produce exhaust re-ingestion. Results from the water flume were verified in the boundary layer wind tunnel of the BAL using the tracer gas technique. General flow patterns are discussed in terms of dilution contours. Thirteen empirical equations of the minimum dilution variation with different building configurations have been derived based on a significant amount of experimental data. The effects of various factors are investigated. The dilution measurement results are compared with prediction from ASHRAE dilution model and those from other recent similar studies.  It was found that the distance of stack to wall inlet and the exhaust momentum ratio affect the exhaust dilution dramatically. However, the stack location does not make any significant difference on dilution within the wake cavity zone with the same stack distance. Higher stack provides higher wall dilution. The gap between emitting and adjacent buildings affects the distribution of dilution, but it does not affect the value of the minimum dilution