The Madison and Minnelusa aquifers are two of the most important aquifers in the Black Hills area. Long-term estimates of recharge to the Madison and Minnelusa aquifers are important for managing the water resources in the Black Hills area. Thus, annual recharge from streamflow losses and infiltration of precipitation on outcrop areas is estimated for water years 1931-98. All estimates are for recharge that contributes to regional ground-water flow patterns and that occurs in outcrop areas connected to the regional flow system. Estimates exclude recharge to outcrop areas that are isolated from the regional flow system, which generally results in ground-water discharge to area streams. Streamflow recharge is calculated directly for 11 streams in the Black Hills area that have continuous-record gaging stations located upstream from loss zones, using available records of daily streamflow, against which estimated loss thresholds (from previous investigations) are applied. Daily streamflow records are extrapolated, when necessary, using correlations with long-term gages, to develop annual estimates of streamflow recharge for 1950-98. Streamflow recharge is estimated for a number of smaller basins using loss thresholds for miscellaneous-record sites. Annual recharge estimates are derived from synthetic records of daily streamflow for 1992-98, which are based on drainage-area ratios applied to continuous-record gaging stations. Recharge estimates are further extrapolated for 1950-91, based on the average percentage of streamflow recharge contributed by these basins during 1992-98, relative to overall streamflow recharge.Streamflow recharge also is estimated for small drainage areas with undetermined loss thresholds that are situated between larger basins with known thresholds. Estimates for 1992-98 are based on estimates of annual streamflow derived using drainage-area ratios, with assumed losses equal to 90 percent of annual streamflow. Recharge estimates also are extrapolated for 1950-91, based on the average percentage of streamflow recharge contributed by these basins.Precipitation recharge for 1931-98 is estimated using relations between precipitation and streamflow (or basin yield) for representative gaging stations. Basin yields are first normalized, relative to drainage area, by expressing in inches per unit of drainage area. Yields are further converted to yield efficiencies, by dividing by precipitation on contributing drainage areas. Relations between yield efficiency and precipitation are identified, which are developed for use in generically estimating annual yield for given areas, based on average yield efficiency and annual precipitation. The resulting annual yield is used as a surrogate for estimating annual recharge from infiltration of precipitation on outcrop areas of the Madison and Minnelusa aquifers. Annual yield (or recharge) efficiencies are estimated to range from about 2 percent to in excess of 30 percent, with corresponding average annual recharge estimates ranging from 0.4 inch in the southern Black Hills to about 8.7 inches in the northwestern Black Hills.Estimates of precipitation recharge for 1931-49 are used to estimate streamflow recharge for the same period, based on correlations between the two variables for 1989-98. Combined streamflow and precipitation recharge to both aquifers averaged about 344 ft3/s for 1931-98. Streamflow recharge averaged about 93 ft3/s, or 27 percent of combined recharge, and precipitation recharge averaged about 251 ft3/s, or 73 percent of combined recharge. Combined recharge ranged from 62 ft3/s in 1936 to 847 ft3/s in 1995. The lowest recharge amounts generally occurred during the 1930?s; however, a more prolonged period of low recharge occurred during 1947-61.For 1931-98, average precipitation recharge to the Madison aquifer is about 3.6 inches, compared with 2.6 inches for the Minnelusa aquifer. However, recharge volumes to these aquifers are nearly identical because th