The baroclinic transport of the Antarctic Circumpolar Current through Drake Passage, within limits set by the substantial variability of the baroclinic flow, has been steady between 1975 and 2000. The baroclinic transport above and relative to zero at 3000 m over this period (14 sections) is 107.3±10.4 Sv. For six hydrographic sections along the WOCE line SR1b taken between 1993 and 2000 the baroclinic transport relative to the bottom is 136.7±7.8 Sv. The majority of this transport occurs in two fronts of the Antarctic Circumpolar Current: the Polar Front 57.5±5.7 Sv and the Subantarctic Front 53±10 Sv. South of the Polar Front, the Southern Antarctic Circumpolar Current Front carries 9.3±2.4 Sv and between this front and the Polar Front a rich eddy field carries around 16 Sv, though the eddies themselves have large barotropic velocities and recirculating transports up to ±25 Sv. For the six sections the Polar Front is observed at two distinct latitudes separated by 90 km. In its northerly position the net baroclinic transport through Drake Passage is 141.7±1.8 Sv and in its southerly position is 131.9±8.6 Sv. Associated with this bimodal distribution of Polar Front position is the temperature and absolute flux of Antarctic Bottom Water. For those years when the Polar Front is northerly/southerly the Antarctic Bottom Water is cold/warm and has a westward/eastward flux. Sir George Deacon in 1935 suggested an intimate link between the deep circulation and Polar Front position. In 1996, 1997 and 2000 lowered ADCP observations were made at each CTD station. De-tided bottom track data give near bottom reference velocities to calculate the absolute transport. The bottom reference velocities (error ~0.2 cm/s) have a similar distribution in the three years. A westward jet with velocities exceeding 20 cm/s is found over the Antarctic continental shelf and the Polar Front has eastward velocities between 15 to 20 cm/s. The net contribution to the absolute transport of the near bottom reference velocities is -28 to 50 Sv. We review a series of papers from the International Southern Ocean Studies Programme that determine the absolute transport through Drake Passage from a year-long array of current meters and pressure gauges: from these the average year-long absolute transport in 1979 is 134 Sv (the canonical value) and the standard deviation of the average is 10.3 Sv. We highlight that the careful estimates of the average transport error given in those papers is 15 to 27 Sv. Finally, we show that variability in the total transport is partitioned equally between the baroclinic and barotropic components, so that a monitoring programme would need to measure both these components to accurately determine the variability of the total transport.