The effect of randomized cloud seeding on daily depth of runoff, during 27 years of experiments in central Israel, is estimated through data for a small watershed. The estimation is carried out with respect to depths of precipitation over the watershed and at two control areas. Double ratio and linear regression models are applied for this estimation. Distribution of rainfall depths by intensities and of seeding results with respect to wind direction are also examined. Differing results are obtained for the four individual experiments, each carried out over a period of 6 years. These reflect sampling variations in the properties of precipitation and runoff. A comparison of variables at the control areas, between seeded and unseeded allocations, indicates that an equipoise has existed over the entire experimental period. The results for this period are a 19% increase in the number of runoff days under seeded allocation, a similarity in the lower halves of the distributions of daily runoff depths along with marked differences in the upper halves, a 15% increase in the mean daily depth of immediate runoff and a 36% enhancement of the total depth of runoff under seeded allocation. These values are considerably lower than those obtained in former studies on three small watersheds located in northern Israel. Volumetric ratio of runoff to rainfall on days with northwesterly winds is higher than that on days with southwesterly winds. This reflects a relation between wind direction and properties of rainfall associated with generation of runoff.