were intentionally fertilized with pollen from local sweet chestnut trees. Nearly one hundred hybrid seedlings resulted from the work in two succeeding seasons, some of which came into bearing in 1908, just as the Endothia blight began to invade New Jersey. The hybrids between the chinquapins and native and European chestnuts were quickly infected, but those with Japan varieties appeared far more resistant. All work with the susceptible native and Europeans ceased, but crosses with Japans and the Chinese chestnut, Castanea Molissima, have been continued until now there are over eight hundred in existence. In late years we have used the Southern creeping chinquapin, C. alnifolia, as a seed parent to some extent, as it appears more resistant than the common species, C. pumila, though it cannot be considered immune. The southern chinquapin is hardy in the North, bears good-sized, sweet nuts for its type, but is very late in ripening.

The Rush chinquapin, and other large-fruited, tall growing varieties have also been used to some extent. The resulting hybrids make handsome trees of rapid growth and bear profuse crops of very attractive nuts, but are greatly injured by blight. As experience accumulated it was found that the extreme caution used in the earlier trials to keep out foreign pollen were scarcely needed and that merely covering the pistillate blooms as soon as they could be distinguished with cotton batting is all that is necessary, and also that hybrids may be produced with considerable certainty in open pollination if the tree or branch is kept entirely free from staminate tassels and the selected pollen is promptly applied as soon as the stigmas become receptive.

Quite a number of chance or self-pollinated seedlings from choice hybrids have been raised in the hope that their good qualities might be perpetuated and the trouble and expense of grafting largely obviated but, as with most other hybrids between distinct species, the seedlings lacked sufficient uniformity to be of especial value. A few individuals turned out superior to the parent but on the whole degeneracy, from the nut-producers standpoint, appears among seedlings of hybrid chestnuts.

In 1909 the unfruited chinquapin hybrids, 68 in number, were transferred to Arlington Farm, Virginia, and two years later Bell Experiment Plot was established near Glendale, Maryland, largely for the purpose of developing blight-resistant varieties of chestnuts as far as this can be done by breeding and testing of wild forms. There are now over 2000 hybrids and seedlings of species at Bell ranging from one to ten years of age. Of the original trees planted at Arlington about 20 remain. They have formed handsome trees twenty feet high with tops almost as wide in diameter and have borne many profuse crops of nuts mostly of good quality and from three to six times as large and heavy as those of the parent chinquapin. All have been attacked by blight, the most promising one only this spring, after thirteen years of resistance to this virulent disease. All the hybrids carrying blood of native or European chestnuts were quickly killed, but those with the Japanese species as a pollen parent are still growing vigorously and bearing well, though considerably disfigured by blight.

Of the various species used the native sweet chestnut, Castanea Americana, and the European, C. vesca, appear entirely useless in breeding for disease resistance, as the hybrids are destroyed by the blight fungus as soon as, or even before, they reach bearing age. The tall, or tree, forms of native chinquapin, sometimes grouped under the botanical name of Castanea arborea but which appear to be only natural hybrids between the sweet chestnut and the bush chinquapin, may also be regarded as useless for the purpose. The hybrid progeny show slight powers of recuperation but, in our plantings, do not sufficiently recover to make useful trees. The Rush chinquapin sometimes resists infection under natural conditions for several years but quickly succumbs when attacked, but its hybrid seedlings develop practically no resistance. The common bush or dwarf chinquapin, Castanea pumila, widely distributed over the Atlantic States, is not as readily infected by blight as the chestnut, many individuals under cultivation and in the wild resisting attack for an indefinite time, while the creeping species of Florida and South Georgia, C. alnifolia, appear practically immune in nature but succumb to artificial inoculation with the blight virus The smooth bark and shrubby forms of these dwarf chinquapins probably account to a very great extent for the limited damage caused by blight under natural conditions.

Next in degree of resistance comes the wooly-twigged Chinese chestnut, C. molissima. There are established at Bell Experiment Plot over nine hundred Molissima trees grown from nuts collected near Tien-Tsin, China, in 1911. These trees in their eighth year of growth have borne excellent nuts, rather larger than those of our native species, in some quantity for three successive years though, owing to extensive locust injury last season, there is practically no crop this year. The trees average twelve or more feet high and are thrifty growers when not too greatly afflicted by blight. No summary of disease injury has been taken, but probably over 80 per cent of the trees show infection many of which are making attempts to heal which are often very successful. This species is native to Eastern China and has long been accustomed to the Endothia fungus, developing in the course of time a very considerable degree of resistance to it. From present appearances the Chinese chestnut may be grown in orchard form with no greater loss from disease than the pear from its particular form of blight. It hybridizes well with the Japan chestnut and both of the dwarf chinquapins, but this progeny is not yet sufficiently developed to warrant judgment.

Castanea crenata, the familiar Japan chestnut, appears everywhere to show greater blight resistance than any other species that has been tried out and is therefore the most hopeful parent to be used for developing a useful race of disease-resistant hybrids and cross-breeds. It has the further merit of bearing very profuse crops of large nuts at an early age, but they are often lacking in quality, being usually harsh to the taste in the raw state though palatable when cooked. A few varieties bear well-flavored nuts, but these appear to be hybrids with our native species and are notably less resistant to blight. Pure Japan varieties grown from imported nuts are rarely injured by blight, and by many are regarded as immune, but those grown from nuts produced in situations exposed to the effects of native pollen are occasionally attacked and even killed outright by the Endothia fungus. It has considerable power to transmit its resisting qualities to its hybrids with the chinquapins, and a few individuals among the latter appear to retain resistance to such a degree that we may yet find among them some of the best nut-producing chestnut varieties of the future.

From the purely horticultural standpoint these hybrids between chinquapins and chestnut species must be considered as most striking successes. If this terribly destructive disease, probably the most virulent that afflicts any tree in the temperate climate, could be controlled there would be little need to look further for varieties suited for commercial and home culture, some of which can be as readily grown as peach trees and come into bearing as young. As the situation stands we must search further for individuals that combine good cropping capacity with practical disease-resistance.

At this writing the most promising outlook appears among selected seedlings of pure Crenata blood, or hybrids of this species that have again