Emission indices were derived from in-flight measurements of CO, nonmethane hydrocarbons (NMHCs), H2O, and nonvolatile condensation nuclei in the exhaust plumes of the Deutsches Zentrum f¿r Luft- und Raumfahrt VFW 614 (ATTAS) and NASA DC-8 experimental aircraft. CO emission indices, EIs(CO), of the ATTAS Rolls Royce M 45H Mk501 engines were determined concurrently by two independent techniques: monitoring of exhaust emissions using a customized Fourier transform infrared spectrometer (FTIR) and by simultaneous continuous fast CO and CO2 measurements. The EIs(CO) determined by FTIR were systematically 28% lower than those derived from the CO/CO2 concentration ratios. The EIs(CO) of the newer and larger CFM 56-2C1 engines, used on DC-8, were substantially smaller than those of the ATTAS engines. The emission behavior of CFM 56-2C1 engines is very similar to CFM 56-3 engines frequently used on Boeing 737 aircraft. In-flight derived EIs(CO) of the ATTAS engines were strongly dependent on the fuel flow rate and agreed well with those calculated from ground-based measurements. Emission indices for individual NMHCs were determined from the concentration ratios of NMHC/CO in the plume of ATTAS and DC-8 and from the EIs(CO) determined by FTIR or derived from the concentration ratios of CO/CO2. The EIs(NMHC) are highest for alkenes and alkynes generated by a cracking of larger fuel molecules and for benzene from unburnt fuel, and they depend strongly on the power setting of the engines. As with EIs(CO), the EIs(NMHC) of the CFM 56-2C1 engine tend to be smaller than those of the Rolls Royce M 45H Mk501. ¿ 2001 American Geophysical Union