A study was conducted to examine the effect of gasoline sulphur content, catalyst variables and driving characteristics on the formation of hydrogen sulphide in catalyst cars drawn from the 1990-91 European fleet.The work was based on chassis dynamometer experiments involving continuous analysis of hydrogen sulphide in raw exhaust gas downstream from the catalyst. A preconditioning procedure was devised that allowed storage of sulphur on the catalyst in a consistent way. The experimental driving (test) cycle provided for wide variation of catalyst temperature and of driving conditions (idle, acceleration, cruise, deceleration). Supplementary time-resolved measurements of carbon monoxide, total hydrocarbon, pre- and post-catalyst gas temperatures and AFR (air-to-fuel ratio) were made as a function of the driving cycle.Using this overall approach, it was possible to determine the important mechanistic factors promoting H2S formation, which was seen to occur in well-defined peaks. The central role of catalyst age, temperature, good transient control of AFR (especially on the rich side), as well as that of gasoline sulphur content, could thus be demonstrated.Under the conditions of the study, it was shown that the risk of H2S malodour could be obviated under open-road conditions without recourse to extreme reduction of gasoline sulphur levels, even in vehicles with relatively unsophisticated emission control systems. Some work investigated the phenomenon of H2S at idle; this tended to be detectable only for the highest sulphur levels and with idle instability that gave the momentary richness required for H2S production.Gasoline sulphur content was thus shown to be one of several variables affecting H2S production. The substantial car-to-car variation observed in the study can be rationalised in terms of the catalyst and exhaust gas management strategy applied in each case.