The chemical NOx budget in the upper troposphere over the tropical South Pacific is analyzed using aircraft measurements made at 6–12 km altitude in September 1996 during the Global Tropospheric Experiment (GTE) Pacific Exploratory Mission (PEM) Tropics A campaign. Chemical loss and production rates of NOx along the aircraft flight tracks are calculated with a photochemical model constrained by observations. Calculations using a standard chemical mechanism show a large missing source for NOx; chemical loss exceeds chemical production by a factor of 2.4 on average. Similar or greater NOx budget imbalances have been reported in analyses of data from previous field studies. Ammonium aerosol concentrations in PEM-Tropics A generally exceeded sulfate on a charge equivalent basis, and relative humidities were low (median 25% relative to ice). This implies that the aerosol could be dry in which case N2O5 hydrolysis would be suppressed as a sink for NOx. Suppression of NO2O5 hydrolysis and adoption of new measurements of the reaction rate constants for NO2 + OH + M and HNO3 + OH reduces the median chemical imbalance in the NOx budget for PEM-Tropics A from 2.4 to 1.9. The remaining imbalance cannot be easily explained from known chemistry or long-range transport of primary NOx and may imply a major gap in our understanding of the chemical cycling of NOx in the free troposphere.