Some years ago the senior author, in collaboration with Norah Dowell Stearns, undertook to make a monthly inventory of the water‐supply of the Pomperaug River Basin, in Connecticut, from a study of data obtained by A. J. Ellis from 1913 to 1916. For this purpose approximate determinations or estimates were made of the ground‐water runoff, that is, of the part of the daily discharge of the river that consisted of water derived from the zone of saturation by effluent seepage. The determinations or estimates were essentially based on the assumption that after a week of fair weather the storm‐water was all discharged and that until the next rain the runoff was essentially all ground‐water (U.S. Geol, Surv. Water‐Supply Paper 597, pp. 107–116 and pi. 19, 1929). In the discussion of the methods and results of this investigation, the following statement was made (Water‐Supply Paper 597, pp. 145–146):

“In this investigation the estimates of ground‐water runoff were based on the discharge of the Pomperaug at Bennetts Bridge—that is, on the discharge during the periods between rains, when there was virtually no direct runoff left in the stream‐system. Much better results could be obtained by basing the estimates on periods beginning as soon after rains as all of the direct runoff has reached the streams. With this method the ground‐water runoff during any particular day would be the total runoff minus the decrease in stream‐storage. The decrease in stream‐storage could be estimated by maintaining gages at several points on the trunk‐stream and on selected tributaries and making surveys of the stream‐system showing the approximate water‐areas of different parts of the system at different gage‐heights.”