Abstract. The effects of photochemical aging on emissions from 15 light-duty gasoline vehicles were investigated using a smog chamber to probe the critical link between the tailpipe and ambient atmosphere. The vehicles were recruited from the California in-use fleet; they represent a wide range of model years (1987 to 2011), vehicle types and emission control technologies. Each vehicle was tested on a chassis dynamometer using the unified cycle. Dilute emissions were sampled into a portable smog chamber and then photochemically aged under urban-like conditions. For every vehicle, substantial secondary organic aerosol (SOA) formation occurred during cold-start tests, with the emissions from some vehicles generating as much as 6 times the amount of SOA as primary particulate matter (PM) after 3 h of oxidation inside the chamber at typical atmospheric oxidant levels (and 5 times the amount of SOA as primary PM after 5 × 106 molecules cm−3 h of OH exposure). Therefore, the contribution of light-duty gasoline vehicle exhaust to ambient PM levels is likely dominated by secondary PM production (SOA and nitrate). Emissions from hot-start tests formed about a factor of 3–7 less SOA than cold-start tests. Therefore, catalyst warm-up appears to be an important factor in controlling SOA precursor emissions. The mass of SOA generated by photooxidizing exhaust from newer (LEV2) vehicles was a factor of 3 lower than that formed from exhaust emitted by older (pre-LEV) vehicles, despite much larger reductions (a factor of 11–15) in nonmethane organic gas emissions. These data suggest that a complex and nonlinear relationship exists between organic gas emissions and SOA formation, which is not surprising since SOA precursors are only one component of the exhaust. Except for the oldest (pre-LEV) vehicles, the SOA production could not be fully explained by the measured oxidation of speciated (traditional) SOA precursors. Over the timescale of these experiments, the mixture of organic vapors emitted by newer vehicles appears to be more efficient (higher yielding) in producing SOA than the emissions from older vehicles. About 30% of the nonmethane organic gas emissions from the newer (LEV1 and LEV2) vehicles could not be speciated, and the majority of the SOA formed from these vehicles appears to be associated with these unspeciated organics. By comparing this study with a companion study of diesel trucks, we conclude that both primary PM emissions and SOA production for light-duty gasoline vehicles are much greater than for late-model (2007 and later) on-road heavy-duty diesel trucks.