and mentions the dry cured meat prepared by peasants in some places in the Alps, which, it is stated, is due to the greater drying power of the air. The reasoning is all but conclusive. There can be no doubt though, that the verdict of the soundest reasoning ought to be confirmed in the court of experimental evidence.

In 1907, Livingstone measured evaporation in the Santo Catalina mountains and found a decrease with altitude. This, perhaps, was not surprising, seeing that the series began in the desert below and extended to the cool mountain regions above.

During the past summer, I set up two series of instruments in the Selkirks at altitudes ranging from 800 m. to 2,900 m. Mountain flanks having a fairly uniform exposure were chosen, only short horizontal distances were involved, two stations were chosen at each altitude so that one might check the other. The season was favorable, being unusually warm and dry; one series was in perfect operation for twelve weeks. The results on the whole seem to exclude the idea that in the Selkirks evaporation increases with altitude. The maximum in each case was at the next lowest station, altitude of 1,100 meters. Above that there was a gradual and irregular diminution. It may be that the lower temperatures of higher altitudes more than offset the factors which make for increased evaporation. Indeed, it appears certain from the above data that for the Selkirks, such is the case.

The data in question, however, present only weekly totals. The possibility that excessive evaporation may take place during certain portions of the day, still remains to be studied. It must be ever kept in view, too, that it is not the absolute amount of transpiration which is so much of importance to the plant, as the balance between water supply and water loss. If the lower soil temperatures of higher altitudes make the obtaining of water more difficult, then the same or even a diminished evaporation rate might demand increased power of resistance on the part of the plant.

In trying to unravel the problems of mountain vegetation, it must never be forgotten that the plants of the present are descended from preexisting ones; facts of heredity are everywhere; phenomena that are found today not seldom hark back to conditions of the past.

In concluding, allow me to offer a plea for a service, which I have been thinking for several years would be one of the most helpful which could be rendered to this adolescent science of ecology, namely, that some one whose knowledge of physics and physiology fits him for such a task should overhaul and scrutinize our ideas and methods. Not counting minor and ephemeral papers, there can be no question that ecology, at the present time, contains not a little of discernible error. Rumors have been heard that zoologists are beginning to study ecology and looking to botanical methods for hints for developing their own. that if wide-awake the house in order.

Within the family, it may be said guests are coming, it is time to set Moreover, ecology is finding a large place in elementary text-books, and in this way errors are being propagated. The interest alike of science and education in this field could in no way be better served than by a relentless pruning.

DISCUSSION OF PROFESSOR SHAW'S PAPER.

DR. LIVINGSTON: A remark was made by Professor Shaw in the beginning of his paper, which suggests that we sometimes lose sight of physical facts. A certain plant is not killed or shut out by the fact that other plants are near it, but by the fact that light conditions or moisture conditions or evaporation conditions or temperature conditions are different from those in which they might live. It seems to me that to speak of biotic and physical conditions leads to confusion; the thing the plant feels must be a physical thing.

PROFESSOR SHAW: In my remark I meant to clear up certain cases where one might lose sight of the fact that plants had largely modified physical conditions. I presume the very case which gave rise to that sentence was this: there are some reasons for supposing that in the Selkirks, the zone of greatest moistness, where lichens and mosses flourish is at something like 4,000 feet in altitude. A forest has developed there, and the condition of low evaporation rate has come about on account of that forest.

SHORTER ARTICLES AND CORRESPONDENCE

PLEISTOCENE SWAMP DEPOSITS IN VIRGINIA

Buried swamp deposits of Pleistocene age are far more abundant in the Atlantic coastal plain than was suspected a few years ago. A large number of such deposits are exposed by the comparatively recent cutting of the waves along the western shores of the Chesapeake Bay and the estuaries of our southern rivers. These deposits vary greatly in botanical interest from point to point. In some no recognizable remains have been found, in others these are limited to the stumps of the cypress, while still others contain the remains of a considerable flora. It seems evident that quite a number of modern species are more restricted in their northward range than they were during certain interglacial periods or even post-glacial time, while others are extending their range at the present time. Among the former the red bud or Judas-tree (Cercis canadensis Linné) and the osage orange (Tokylon pomiferum Raf.) might be mentioned, both occurring in the interglacial beds of the Don Valley near Toronto in Canada. Another species abundant as far northward as New Jersey in inter- or post-glacial time was the bald cypress (Taxodium distichum Rich.), and the loblolly pine (Pinus taeda Linné) also appears to have withdrawn southward since the late Pleistocene. The water elm (Planera aquatica Gmel.), on the other hand, appears to have retreated southward in the late Pleistocene and to be readvancing at the present time.

Numerous Pleistocene swamp deposits are described in the recently published account of the Maryland Pleistocene1 and their contained fossil plants have been admirably exploited by Dr. Hollick, who enumerates over 40 species from beds of this age in Maryland.2 Similar deposits are to be found at many points in the Old Dominion, although very little attention has been thus far devoted to their description.

In 1906 the writer described3 Fagus americana (nuts) Vitis sp. (seeds), Hicoria glabra (fruit), Taxodium distichum (seeds

and cone-scales) and Nyssa biflora (seeds) from the Talbot or latest Pleistocene formation at Tappahannock, Va.

During the last two years peat deposits of late Pleistocene age have been discovered at numerous points along the Potomac, Rappahannock and James rivers. These usually contain characteristic stumps of the cypress, often of large size, and the associated peat furnishes recognizable plant remains, generally seeds, of Nyssa, Vitis, Fagus and Taxodium, the latter usually the most abundant. Associated with the plant remains are fragments of the elytra of beetles, occasional insect galls and molluscan remains. Evidently most of these Pleistocene peats indicate the presence of cypress swamps, but this is not always the case, since in some instances we find representatives of upland vegetation, while in other cases the fossils show that the vegetation was open and marsh-like with a scattering of trees of oak, birch and pine. Where quiet water conditions followed the subsiding forest-bed, deposits of clay are to be found and these often contain fragments of leaves along with casts of the shells of Unio if the locality is toward the head of the ancient estuary, or shells of Rangia cuneata if the waters were more saline. The latter is very abundant in Pleistocene deposits of this sort from Maryland southward and it is also a member of the modern fauna, confined, however, in the latter case to the Gulf of Mexico.

Fig. 1 shows the bed of massive hard peat a short distance above Tappahannock, Va., from which the above mentioned species were collected. It is exposed for a thickness of four or five feet and shows many cypress stumps in place with the "knees" protruding through the recent narrow sand-beach. Overlying the peat is a layer of drab plastic clay from one to four feet thick and carrying fragments of leaves. Above this there is from ten to fifteen feet of coarse sand. The Rappahannock has cut into the bluff along its south bank, exposing this old cypress swamp for a distance of over one half a mile.

Where the forest-bed was succeeded by waters which were agitated by wave action or swift currents the overlying deposits are sands or gravels laid down on the more or less planed surface of the old swamp. Fig. 2 shows such a condition of affairs in an exposure just above the junction of Parrotts Creek with the Rappahannock River. The line of apparent unconformity is sharply defined running across the center of the picture. Below are seen the cypress trunks and roots embedded in an impure peat which was planed off by wave or current action as the