The purpose of this study is to investigate ways to control the wind dispersal of contton-wrapped seeds of such poplars and willows as Poplus alba, P. tomentiglandulosa, P. euramericana, P. deltoides, and Salix pseudo-lasiogyne. These trees are hated by many people because of their seeds blowing all over the place during May. These cottony seeds can be nuisance to various types of outdoor activities, pose safely threat to drivers, become fire hazards during prolonged spring dry spells, and cause many types of health hazards of allergy such as sneezing. rhinitis, asthma, and skin rashes. Four control methods can be used to resolve the problem. First, pruning can be a solution, but it is unsatisfactory in terms of costs and outcome. Second, planting of male trees only can be a solution, but it is hard to identify sexes of saplings. Third, female trees can he replaced with other species. But it requires high costs and takes at least ten years to functionally replace the removed ones. As an alternative to such unsatisfactory control methods, the possibility of applying plant growth regulators has been investigated since 1983. During the pre-test, various concentrations and mixtures of them were either sprayed or injected, but failed to achieve any promising results. But the injection of a mixture made up of 0.75g of 2-chloroethane phosphonic acid with 0.2 mg of GA in 300cc water in the end of March produced premature falling of almost all aments and capsules of treated poplars and willows. It was found that the effect of the injection lasts two years. The results of the main experiment of 1987 can be summarized as follows ; First, the injection of the mixture of 2-chloroethane phosphonic acid and GA increases the premature abscisin of aments and capsules, thus reducing the wind dispersal of the cottony seeds of S. pseudolasiogyne, P. tomentiglandulosa, and P. euramericana 1644.09, 1200.61, and 1485.11 times, respectively, than that of the naural abscisin, It is estimated that the average number of wind-blown seeds reduced are approximatively 6,185,100, 4062,900, and 2,830,670 per tree, respectively. Second, the treatment causes no observable side effects on the growth of the samples tested.