Extant physical models allow the hypothesis that seeding may produce either increases or decreases in hailfall. Present technology for delivery of seeding materials permits seeding with higher concentrations and greater accuracy of placement with airborne equipment than is possible with ground equipment. Evaluation of experience to date supports the hypothesis that seeding at rates of less than 1000 gm hr−1 per storm may stimulate convection and increase the number of individual hail events, but that heavier seeding at rates of 2000–3000 gm hr−1 per storm is effective in reducing hailfall damage by reducing the total impact energy from hailfalls. Limited observational evidence suggests that the effects of cloud seeding on hailfalls may extend to clouds not directly affected by the seeding material. Abstract Extant physical models allow the hypothesis that seeding may produce either increases or decreases in hailfall. Present technology for delivery of seeding materials permits seeding with higher concentrations and greater accuracy of placement with airborne equipment than is possible with ground equipment. Evaluation of experience to date supports the hypothesis that seeding at rates of less than 1000 gm hr−1 per storm may stimulate convection and increase the number of individual hail events, but that heavier seeding at rates of 2000–3000 gm hr−1 per storm is effective in reducing hailfall damage by reducing the total impact energy from hailfalls. Limited observational evidence suggests that the effects of cloud seeding on hailfalls may extend to clouds not directly affected by the seeding material.