Matthew I. Rorabaugh (U.S. Geological Survey, Louisville, Kentucky, August 27, 1945)—In regions where snow is not a factor there is some advantage in making forecasts based on the effects of travel time and base flow. However, the runoff which results from precipitation during the period of forecast usually constitutes a substantial portion of the flow in many streams, and accordingly these predictions may be highly speculative.

The authors state that on March 1 an average of 52 per cent of the winter flow of Parker Creek has passed the gaging station on that creek while only 36 per cent of the winter flow of the Salt River has reached Roosevelt Reservoir. The forecast is based on this lag, which would approximate 16 per cent of the winter flow of the Salt River or about 25 per cent of the flow during March to May. In giving reasons for the lag the authors minimize the effect of accumulated snow, because less than 15 per cent of the Salt River Basin is thought to have accumulated snow. It may be, however, that the snow on this portion of the basin is the principal reason for the lag. The snow cover accumulates at high altitudes, where precipitation is heaviest and temperatures lowest. As losses will be proportionally less in these low‐temperature areas than in the remainder of the basin, it follows that storage of 25 per cent of the flow could result from snow accumulation on less than 15 per cent of the basin.