TREES (continued) Platanus occidentalis Juglans nigra Fraxinus sp. Prunus serotina HERBS Dryopteris noveboracensis Epiphegus virginiana Polygonum virginianum Here the trees outnumber the shrubs and herbs, and there are more vines than in any other habitat in the region. This preponderance of trees and vines seems to be characteristic of river banks and alluvial swamps in many other parts of the world.* Rivers as a rule are bordered by vegetation approaching the climax, but at this altitude of 3,000 feet there is still so much erosion going on that the normal succession is retarded, which probably accounts for the abundance of four evergreens. Few if any of the species in this list can be considered as peculiarly Appalachian. Nearly all of them are common in the Piedmont region from Pennsylvania to Alabama, as well as in the Mississippi valley; and several are still more widely distributed. In the gravelly and muddy beds of the same streams, which must be covered with water half the time, the following herbs find a congenial habitat: The fact that four of these, or 40 per cent., are monocotyledons, is probably not without significance. All of them are pretty widely distributed, mostly northward. Near Davis Gap (sometimes called Pigeon Gap), about three miles east of Waynesville, and near Balsam Gap, about seven miles southwest, are the only wet meadows which I made note of in the region under consideration. Both are about 3,300 feet *See Ann. N. Y. Acad. Sci. 17: 67-73, 103-104. 1906. above sea-level. The cause of the treelessness of such areas, and their relations to other habitats in the neighborhood, are unsolved though perhaps not very difficult-problems. With the exception of Acer rubrum and Salix longipes, scattered along stream channels at Balsam Gap, the vegetation is entirely herbaceous, as follows: All of these are just as common outside of the mountains as they are here, if not more so. Most of them can be found in wet meadows in New England, and a still larger proportion along the head-waters of East Meadow Brook, near Hempstead, Long Island; and all range at least as far south as Middle Georgia, about 100 miles farther south and 2,500 feet lower. All the species seem to be perennial, but none are evergreen in the ordinary sense of the word. Nearly half the angiosperms are monocotyledons. There are no Ericaceae among them. The weeds of the mountain region are found principally along trails and roads and in pastures and abandoned fields. They are all or nearly all herbs, and mostly dicotyledons. The following list is doubtless very incomplete. The species are arranged approximately in order of abundance, as usual. Juncus tenuis * Potentilla canadensis Rumex Acetosella Lobelia inflata Verbascum Thapsus * See Gray, Am. Jour. Sci. 42: 41.1842. Chrysanthemum Leucanthemum Polygonum Hydropiper Trifolium repens Oxalis stricta? Of these weeds about 28 per cent. are supposed to have been introduced from Europe and 2 per cent. from Asia, while the remaining 70 per cent. are considered indigenous by nearly all systematists. And yet all the supposed natives, with five or six exceptions, are confined to unnatural habitats, exactly like the introduced species, from which there is no possible way of distinguishing them without the use of botanical literature, such as a manual, and even that is not infallible. At least half, perhaps two thirds, of the species in the above list evidently belong to that class of native weeds (mutants?) which I discussed just before going to North Carolina.‡ COLLEGE POINT, NEW YORK MAGNOLIA AT FLORISSANT § By T. D. A. COCKERELL The Miocene flora of Florissant, Colorado, includes so many genera living today in the southeastern states, that the apparent absence of Magnolia has seemed remarkable. During the past summer, however, a leaf which may I think be referred to this * See Gray, Am. Jour. Sci. 42: 27. 1842. † See Gattinger, Fl. Tenn., 107. 1901. Bull. Torrey Club 35: 347-360. July, 1908. Illustrated with the aid of the Catherine McManes fund. genus with confidence, has been found by Mr. Terry Duce, and is herewith recorded. Magnolia florissanticola n. sp. Leaf apparently thick, shaped as in M. grandiflora; apex lacking, but length apparently about 130 mm.; broadest about 42 mm., from base; base broad-cuneate, slightly inequilateral, FIGURE 1. Magnolia florissanticola; Miocene shales of Florissant. from a very stout (3 mm. diam.) twisted petiole, which is about 16 mm. long, arising from a clasping base; width of blade about 50 mm., tapering apically, so that at 80 mm. from base the width is 38 mm.; margin entire; venation as in M. grandiflora, the strong lateral veins averaging about 5 mm. apart. Miocene shales of Florissant. (Terry Duce.) REVIEWS Spalding's Distribution and Movements of Desert Plants* The author has divided his problem into seven divisions, under as many headings. Five of these appertain to various phases of his problem, the last two are mainly recapitulative. Tumamoc Hill and its environs, near the desert botanical laboratory, Tucson, Arizona, was the place chosen where "prolonged observational and experimental work could be undertaken." The first section of the paper (pp.5-27. pl. 1–12) is taken up with a clear and logical account of the plant associations and habitats as they have appealed to the author. Appended to this is an account of the lichens of the region, written by Dr. Bruce Fink. Leaving the section on plant associations and habitats which, though valuable, is necessarily becoming more and more stereotypic in each succeeding ecological paper, we come to the most interesting part of the whole work. In this second chapter (pp. 29-66. pl. 13-24), the author gives an account of the local distribution. He writes: "Dealing more in detail with constituent species of the associations, the attempt to trace cause and effect is carried a step farther. Certain species have been carefully mapped and their habits have been more thoroughly studied with reference to differences of soil and aspect.' notes. The species selected for this study are plants " with a remarkable definiteness of habitat preference"; they are Encelia farinosa, Larrea tridentata, Cereus (why not Carnegiea?) giganteus, Cercidium Torreyanum, and Prosopis velutina. A distribution-map for each of these species is included, and they form a series of invaluable Each map is practically a graphic census of the individuals of the species under discussion. Nothing could have been found to indicate so well the relative density of these plants. The various soil formations are critically studied, and following as they do the various distribution-maps mentioned above, they are at least a suggestion of the factors the author credits with the *Spalding, V. M. Distribution and Movements of Desert Plants. Pp. 1-144. 22 Oct. 1909. Carnegie Institution of Washington, Publication No. 113. pl. 1-31. |