Aerosol black carbon concentrations have been measured between 1992 and 1995 at Halley station using an aethalometer. The complete record is dominated by a number of events of high concentration that are clearly due to contamination from the station generators. However, using detailed meteorological data, it is possible rather satisfactorily to remove these events from the record. The events can be used to show that local contamination has no significant effect on sulfate or other major ion data, The remaining data then show a clear seasonal cycle, with monthly mean values of 0.3–2 ng m−3, slightly higher than those recorded at south pole. The cycle peaks in summer, with a possible doublet, and an overall maximum in October. This pattern is similar to that of south pole and of mineral dust at the coastal Neumayer station. The pattern seems to be controlled by the timing of biomass burning in the tropics, strongly modulated by the efficiency of transport to Antarctica. This transport seems to be poorly represented in model simulations. The concentrations are too low to have any significant effect on snow albedo. The similarity of the Halley and south pole data suggest that ice cores should give a historical record of black carbon that is rather representative of the Antarctic as a whole and therefore indicative of trends in biomass burning throughout the southern hemisphere.