beneath, with none or few yellow scales; male aments one to three inches long, densely pubescent; styles three, stout, glabrous, divergent; fruiting involucre with stout divergent spines (Fig. 11) one-half to one inch long,

FIG. 10. LEAVES AND NUT OF CASTANOPSIS CHRYSOPHYLLA.

subverticillately many branched; nut usually solitary, obversely triangular, six lines long."-"Geological Survey of California," Botany, Vol. II, p. 100.

"This handsome broad-leaved evergreen tree is indigenous to the elevated regions, from Monterey, California, northward to the Columbia river in Oregon. It is also common in the Sierra Nevadas at elevations of six

thousand feet, but in its southern limits rarely below ten thousand feet elevation."-C. S. Sargent ("Woods of the United States").

In the warmer and drier regions of California it is a mere shrub two to six feet high, and these dwarf forms have, in some instances, been described as varieties. As, for instance, Castanea chrysophylla, var. minor, Bentham; C. chrysophylla, var. minor, A. de Candolle; and C. chrysophylla, var. pumila, Vasey. But northward, where the cli

mate is more moist, it becomes a large tree fifty to one hundred and twenty feet high, with a stem two to three feet in diameter. In its wide variation in habit of growth, this western chinquapin is similar to our Eastern dwarf chestnut, which is mainly a low shrub in the more Southern States, but becomes a fair-sized tree in the Middle States, or near its northern limits.

I have introduced the Western chinquapin here among the nut-bearing trees, not with the idea that it will ever be extensively cultivated for its edible nuts, but because it is a beautiful broad-leaved evergreen tree, and of which we have far too few kinds in cultivation to give warmth and a cheerful aspect to our gardens and pleasure grounds in winter. It is true that, so far as can be learned at this time, no extended experiments have ever been made to introduce or cultivate the Castanopsis in the Atlantic States, consequently nothing pos

!

itive is known as to whether it will succeed here or not. In its northernmost range it thrives in forests among many kinds of trees and shrubs that are already common in our gardens, and this leads me to think that specimens or seeds of this tree procured from the mountains of northern Oregon will withstand the rigors of our climate.

Mr. S. B. Parsons writes me that he first saw Castanopsis chrysophylla in Kew Gardens (Eng.) thirty-five years ago, and procured specimens, which were planted in his gardens at Flushing, N. Y., but they failed, presumably because not hardy. It may be that his specimens were raised from nuts procured in the warmer part of California, and, as with many other Pacific coast plants, proved to be tender, while later introductions of the same species collected in colder localities have proved hardy here. In my experience I have found a great difference in the hardiness of trees and plants obtained from the higher and lower levels of the mountains from Colorado westward to the Coast range, for in those regions acclimation extending over thousands of years has developed and fixed certain physiological attributes, which enables them to readily adapt themselves to similar conditions elsewhere, especially in the line of tenperature. It may make no difference to those who want plants for warm climates, whether they are obtained from mountain or valley, but it certainly does to those who value hardiness above all other merits.

In horticultural matters we are supposed to confine ourselves within certain natural lines in making experiments, but if we fail in one, or one hundred, it proves little beyond the bare fact that we have not been successful. I have experimented enough to have become somewhat wary of deciding that a thing cannot be done, or is impossible, because of my own and others' failures. Every practical horticulturist can call to mind many

productions which had evaded the pursuit of experimenters for decades and even centuries.

For specimens of the nuts, burs and plants of this handsome nut-bearing tree I am indebted to Mr. J. J. Harden, of Stayton, Oregon, who informs me that it grows in the mountains near by to a very large size, and among such well-known kinds of shrubs and trees as Rhamnus Purshianus, Cornus Nuttalli, Corylus rostrata, and various species of conifers which are now more or less common in our Eastern gardens and parks. The twigs and leaves are shown in Fig. 10, and below a nut, and in Fig. 11 a bur, all of natural size. The small conical nut is slightly triangular, with a rather firm, brittle shell, not fibrous as in the acorn and chestnut. The burs are produced singly, but sometimes several on a twig, and when mature, instead of opening by valves, as in the true chestnut, they break up irregularly. The kernels are sweet and excellent flavored, and are sought for by various kinds of birds, as well as by all the squirrel tribe, and for this reason it is very difficult to procure specimens, unless gathered before they are fully ripe. The nuts do not mature the first season, but pass the winter in a partly developed stage, usually ripening the second year about midsummer or, in northern Oregon, in July.

It is quite probable that this Castanopsis, when planted in the Atlantic States, will require a little shade or protection, like the American holly and similar broadleaved evergreens, and while it may not thrive anywhere north of Delaware and Maryland, it is worth trying, as the sole native representative of a genus containing several species of noble evergreen trees.

CHAPTER V.

THE CHESTNUT.

Castanea, Tournefort. The ancient classical name derived either from Castanis, a town in Thessaly, or one in Pontius, as historians disagree in regard to its derivation. The genus belongs to the order Cupuliferæ.

Male flowers irregularly clustered in long, naked, cylindrical catkins from the axils of the leaves and on the new shoots of the season. Calyx five or six parted; stamens or pollen-bearing organs seven to fifteen; anther two-celled. On old, mature trees, the male catkins are usually crowded near the end of the short new twigs, as shown in Fig. 12, the terminal one productive; but on young thrifty trees, wide apart. Female flowers always on and near the base of a late-developed male catkin, sometimes two or three together, or even six or eight on the chinquapins,-oval or ovoid, scaly, prickly, two- to four-valved involucre or bur; calyx usually with a four to six-lobed border crowning the three- to sevencelled ovary; stigmas bristle-shaped, and as many in number as there are cells in the ovary. Shell of the nut leathery, not brittle, ovoid, two or more together in the larger species, in others solitary, or only one in a bur. Kernel very thick, fleshy, and somewhat plaited, sweet and edible.

Both male and female flowers appear late in spring, the males usually exceedingly so, exhaling a slightly nauseating odor. The productive male catkins appear the latest, their base becoming the rachis or stalk supporting the burs, this rather anomalous arrangement appear