In March--April 1995, as part of the World Ocean Circulation Experiment section A23, we completed 49 hydrographic stations across the Weddell Gyre and southern Antarctic Circumpolar Current, from the Antarctic continental shelf (72.5 ¿S, 16.5 ¿W) to South Georgia (55 ¿S, 34.5 ¿W). Chlorofluorocarbon (CFC-11, CFC-12, and CFC-113) data collected at these stations reveal that distinct sources renew the Antarctic Bottom Water (defined as waters with potential temperatures less than 0 ¿C) of the Weddell Gyre. Weddell Sea Bottom Water (defined as waters with potential temperatures less than -0.7 ¿C) formed in the western Weddell Sea has CFC concentrations about 5 to 6 times higher in the eastward flowing northern Weddell Gyre than in the westward flowing southern limb. Our CFC measurements suggest that distinct sources of Weddell Sea Bottom Water exist in the western Weddell Sea, in agreement with previous descriptions based on potential temperature and salinity signals. In the northern Weddell Gyre, high CFC concentrations in Weddell Sea Deep Water, potential temperatures between 0 ¿C and -0.7 ¿C, confirm the long-recognized sources for this water mass in the western and southwestern Weddell Sea. In the southern Weddell Gyre at about 20 ¿W and along the Antarctic continental slope, Weddell Sea Deep Water with potential temperatures around -0.45 ¿C shows a deep CFC maximum about 1000 m above the seafloor. CFC concentrations in this deep southern core are about 80% of those of new Weddell Sea Deep Water in the northern Weddell Gyre near 30 ¿W. The A23 CFC and hydrographic data are not consistent with the hypothesis that Weddell Sea Deep Waters are derived from a single source in the western Weddell Sea. Instead, these tracers suggest that an important portion of the Weddell Sea Deep Water in the southern Weddell Gyre originates outside the western Weddell Sea, probably near the Amery Basin and environs, around 75 ¿E. These features of the circulation and renewal of the deep Weddell Gyre should be carefully considered in simulations dealing with fluxes, pathways, and formation rates of Antarctic Bottom Water. ¿ 2000 American Geophysical Union