Three groups of 16 cars each or a total of 48 fleet vehicles will be driven in normal urban and suburban modes of operation over a 24,000 mile interval under controlled fuel conditions and maintenance. Each group of 16 cars is being operated on the same base fuel. Two of the three group's test fuels contain a gasoline additive which is added to the base fuel. Exhaust emission measurements were made as the vehicles started the test and at each subsequent 4,000 mile interval. An analysis of variance was conducted on each of the 19 emission parameters for the zero and 4,000 mile interval. A major conclusion reached on the basis of these analyses is that: no significant differences in emission parameters were determined at the 4,000 mile point on the basis of fuel employed. A secondary conclusion drawn concerns the make factor. That conclusion is: all emissions parameters vary significantly between makes; however, no single make is consistently the highest or lowest valued across the spectrum of the 19 emissions parameters measured. The VW vehicles were distinctly different to such a degree that their data were not included in the final analysis. (Author)