The photochemistry of the troposphere over the South Atlantic Basin is examined by modeling of aircraft observations up to 12-km altitude from the TRACE-A expedition in September-October 1992. A close balance is found in the 0-12 km column between photochemical production and loss of O₃, with net production at high altitudes compensating for weak net loss at low altitudes. Simulation of H₂O₂, CH₃OOH, and CH₂O concentrations measured aboard the aircraft lends confidence in the computations of O₃ production and loss rates. Influx from the stratosphere is negligible as a source of O₃ or NO_x to the 0-12 km column. The primary sources of NO_x appear to be continental (combustion, lightning, soils) and include a major contribution from biomass burning. There is evidence that NO_x throughout the 0-12 km column is recycled from its oxidation products rather than directly transported from its primary sources. There is also evidence for rapid conversion of HNO₃ to NO_x in the upper troposphere by a mechanism not included in current models. A general representation of the O₃ budget in the tropical troposphere is proposed that couples the large-scale Walker circulation and in situ photochemistry. Deep convection in the rising branches of the Walker circulation injects NO_x from combustion, soils, and lightning to the upper troposphere, leading to O₃ production; eventually the air subsides and net O₃ loss takes place in the lower troposphere, closing the O₃ cycle. This scheme implies a great sensitivity of the oxidizing power of the atmosphere to NO_x emissions in the tropics.