Streamflow characteristics are determined by a large number of factors of the meteorological and terrestrial environments. Because of lack of quantitative data to describe some of the factors and complex interrelations among them, complete analysis of the relations between streamflow and the various environmental factors is impossible. However, certain simplifying assumptions and generalizations made possible a partial analysis for the Delaware River region. For relations involving average runoff or low-flow parameters, average annual precipitation was assumed to be the principal meteorological factor, and geology (a complex of many factors) was assumed to be the principal terrestrial influence, except for that of basin size which was largely eliminated by expression of discharge in terms of unit area. As a first approximation, physiographic units were used as a basis for classifying the geology. Relations between flow parameters and precipitation are fairly well defined for some physiographic units, but not for those in which the geology varies markedly or the areal variation in average precipitation is very small. These relations provide a basis for adjusting the flow parameters to reduce or eliminate the effects of areal variations in precipitation and increase their significance in studies of the effects of terrestrial characteristics. An investigation of the residual effect of basin size (the effect remaining when discharge is expressed in terms of unit area) on relations between flow parameters and average precipitation indicates that such effect is negligible, except for very large differences in area. Parameters that are derived from base-flow recession curves and are related to a common discharge per unit area have inherent advantages as indicators of effects of terrestrial characteristics of basins, because the.y are independent of areal variations in average annual precipitation. Winter base-flow parameters are also practically independent of the effects of evapotranspiration from ground water. However, in many parts of the region these advantages are reduced or nullified by the difficulties of defining base-flow recession curves, particularly winter curves, with sufficient accuracy. In the absence of suitable base-flow recession data and a suitable basis for adjusting parameters, the ratio of the discharge equaled or exceeded 90 percent of the time to the average discharge (Qtt/Qa), or a similar duration parameter, probably is the best indicator of the influence of terrestrial characteristics, although the ratio may vary somewhat with average precipitation. In a part of the region where geologic differences are large and areal variations in average precipitation are small, values of Qm/Qa for each major geologic unit were determined from streamflow records. From these values and the percentage of area represented by each unit, a ratio for each gaging station was computed. Comparison of these computed results with the observed results indicates that nearly all of the variation in the ratio is associated with variation in geology. The investigation indicates that the original assumptions are correct; average precipitation is the principal meteorological influence and geology is the principal terrestrial influence. Together these two factors account for a very large proportion of the variation in average runoff and low-flow characteristics.