EPA's Integrated Air Cancer Project has produced comprehensive winter time ambient air data bases from three widely separated airsheds (Albuquerque, NM, Raleigh, NC, Boise, ID), which were anticipated to be heavily impacted by residential woodsmoke and motor vehicle emissions, but little else. Because of the simplicity of the airsheds a relatively simple multiple linear regression method using tracers based on potassium and lead or bromine has been successful in apportioning ambient fine particle extractable organics and mutagenicity. A comparison of receptor modeling results from all three cities shows a general consistency between the three cities, and with earlier mutagenic potency measurements from direct source testing.