This paper reports the results of attempts to improve filberts by hybridization at the Experiment Station at Geneva, N. Y. The filbert project was started at Geneva in the spring of 1925 when a collection of varieties from American sources was established. In later years additional varieties from European and other sources were added until about 120 were under test. As soon as the varieties had fruited for several years it became evident that many of them were inferior and not adapted to New York conditions. A few exhibited considerable merit and the range of characteristics in the different varieties indicated that it might be worth while to start a filbert breeding project with the object of combining the desirable characteristics of the better sorts. [Footnote 6: Journal Paper No. 719, New York State Agricultural Experiment Station, Geneva, N. Y.] It was decided first to cross Rush, a selected form of _Corylus americana_, with the best varieties of _Corylus Avellana_, Rush contributing the hardiness of the native hazel, possible resistance to filbert blight, and the hybrid vigor that sometimes results from the crossing of two species. The European filberts were expected to furnish large-sized nuts as well as dessert and cracking quality. The first crosses were made in 1930 when two trees of the Rush variety growing on Dr. MacDaniels' place in Ithaca were pollinated with pollen of several varieties of _Corylus Avellana_ that was brought from Geneva. Additional crosses were made at Ithaca in 1931 and 1933. In 1932 the pollinations were made at Geneva, using a Barcelona tree covered with a tightly woven cloth. No pollinations have been made since 1933. In the spring of 1932, 535 seedlings were received from Willard G. Bixby, of Baldwin, Long Island, N. Y., which had resulted from crosses made by C. A. Reed of the United States Department of Agriculture, at Baldwin. Including these U.S.D.A. seedlings and those resulting from the breeding work at Geneva, 1,999 seedlings have fruited. The nuts from these crosses were stratified in sand in a cold frame, dug up, and planted in the greenhouse in early March. After one transplanting they were moved to the nursery to grow for two years, when they were moved to the seedling orchard. The nuts from one year's crosses were planted directly in the nursery but germination was low due to drought. The seedlings were spaced 10 x 5 feet in the orchard. This spacing was satisfactory if the trees came into bearing the fourth year, but if unfavorable weather eliminated the first or second crops the trees became too crowded to permit satisfactory fruiting. Usually, however, the trees fruited sufficiently to make it possible to evaluate them and remove the inferior trees so that the better seedlings would have enough room to remain for several additional crops.