A 4-year study of the hydrology of the Wolf Branch sinkhole basin in Lake County, Florida, was conducted from 1991-95 by the U.S. Geological Survey to provide information about the hydrologic characteristics of the drainage basin in the vicinity of Wolf Sink. Wolf Branch drains a 4.94 square mile area and directly recharges the Upper Floridan aquifer through Wolf Sink. Because of the direct connection of the sinkhole with the aquifer, a contaminant spill in the basin could pose a threat to the aquifer. The Wolf Branch drainage basin varies in hydrologic characteristics from its headwaters to its terminus at Wolf Sink. Ground- water seepage provides baseflow to the stream north of Wolf Branch Road, but the stream south of State Road 46 is intermittent and the stream can remain dry for months. A single culvert under a railroad crossing conducts flow from wetlands just south of State Road 46 to a well-defined channel which leads to Wolf Sink. The basin morphology is characterized by karst terrain, with many closed depressions which can provide intermittent surface-water storage. Wetlands in the lower third of the basin (south of State Road 46) also provide surface water storage. The presence of numerous water-control structures (impoundments, canals, and culverts), and the surface-water storage capacity throughout the basin affects the flow characteristics of Wolf Branch. Streamflow records for two stations (one above and one below major wetlands in the basin) indicate the flow about State Road 46 is characterized by rapid runoff and continuous baseflow, whereas below State Road 46, peak discharges are much lower but of longer duration than at the upstream station. Rainfall, discharge, ground-water level, and surface-water level data were collected at selected sites in the basin. Hydrologic conditions during the study ranged from long dry periods when there was no inflow to Wolf Sink, to very wet periods, as when nearly 7 inches of rain fell in a 2-day period in November 1994, following an extended wet season. A comparison to long-term rainfall record (40 years) indicates that this range in hydrologic conditions during the 4-year study is representative of the range of conditions expected during a much longer time period. Two dye-trace studies conducted during the study indicated no direct connections between the sink and local wells. The path of a constituent entering the aquifer through Wolf Sink generally would be to the east, following the gradient of the regional ground-water flow system. The conductance of Wolf Sink (the rate at which the sink conducts water to the underlying aquifer) was estimated from streamflow data, ground-water levels, and water levels in Wolf Sink. The range of hydrologic conditions during the study provided a basis for the determination of a representative conductance value. The regression of streamflow as a function of head difference between the sink water level and the potentiometric surface at an observation well (an approximation of the potentiometric level beneath Wolf Sink) resulted in a significant relation r2=0.91, mean square error = 1.60 cubic feet per second); and the slope of the regression line, representing sink conductance, was 1.48 cubic feet per second per foot of head difference. Flow and storm-volume frequency curves for selected time periods (1-day, 7-days, 14-days, 21-days, and 30-days) were generated based on streamflow data from January 10, 1992, to September 30, 1995. These curves indicate that, based on the available record, the volume of water that would have to be stored (in the event that streamflow had to be diverted from Wolf Sink) during a 30-day period would be equal to or less than about 11 acre-fee 30 percent of the time and 161 acre-feet 80 percent of the time. The maximum volume that would be generated during a 30-day period, based on this study, would be about 570 acre-feet.