The safe and economical design of infrastructure in and near waterways and the effective management of flood-hazard areas require information on streamflow that may not be readily available. This report provides estimates of flow-frequency characteristics for gaged streams in Vermont and describes methods for estimating flow-frequency characteristics for ungaged streams. The flow-frequency characteristics investigated are the magnitude of peak discharges at recurrence intervals of 2, 5, 10, 25, 50, 100, and 500 years, and the magnitude of daily-mean discharges exceeded 25, 50, and 75 percent of the time. Peak-flow frequency characteristics for gaged streams were computed following the guidelines in Bulletin 17B of the U.S. Interagency Advisory Committee on Water Data. To determine the peak-flow exceedance probabilities at stream-gaging stations in Vermont, a new generalized skew coefficient map for the State was developed. This new map has greater resolution and more current data than the existing National map. The standard error of the new map is 0.269. Two methods of extending streamflow record were applied to improve estimates of peak-flow frequency for streams with short flow records (10 to 15 years) in small drainage areas (sites less than 15 square miles). In the first method, a two-station comparison, data from a long-record site was used to adjust the frequency characteristics at the short-record site. This method was applied to 31 crest-stage gages--stations at which only instantaneous peak discharges are determined--in Vermont. The second method used rainfall-runoff modeling. Precipitation and evapotranspiration data from 1948 to 1999 for numerous climate data-collection sites were used as input to a model to simulate flows at 10 stream-gaging stations in Vermont. Also, methods are described to estimate flow-frequency characteristics for ungaged and unregulated rural streams in Vermont. The peak-flow estimating methods were developed by generalized-least-squares regression procedures with data from 138 U.S. Geological Survey stream-gaging stations in Vermont and in adjacent areas of New York, New Hampshire, Massachusetts, and Quebec. The flow-duration (daily flow exceeded a given percentage of the time) estimating methods were developed by ordinary-least-squares regression procedures with data from 81 stream-gaging stations in Vermont and adjacent states.