The estimation of flow quantiles is a major focus of current hydrologic research. Probabilistic representations of a few individual quantiles, notably the annual flood and low flow, have been devised, and recently a few of the larger quantiles have been modeled. This research describes and demonstrates the use of a probability model for all flow quantiles in a river. A flow duration curve represents the annual flow frequency characteristics of rivers by depicting the cumulative frequencies for average ranked flows in a river. A model requiring only five parameters is developed, by combining the principles of order statistics and traditional flow frequency analyses, and applied to flow duration curves for rivers in the province of British Columbia, Canada. Streamflows in British Columbia are generated from a number of distinct physical processes operating in highly variable environments. The model presented incorporates the physical generating processes of streamflow in both the statistical representation of flow duration curves and their interpretation. ¿ American Geophysical Union 1993