brary, while the laboratories of the Shaw School of Botany, at Washington University, are in close relationship to the garden. Much important research, principally taxonomy, has been prosecuted. Publications consist of seven volumes of Annual Reports and nine "Contributions from the Shaw School of Botany."

5. The Botanical Garden of the Michigan Agricultural College was begun in 1877. There are now about three acres under high cultivation, exclusive of the arboretum and decorative grounds, which together cover several acres. There are several small greenhouses, an herbarium of about sixty thousand specimens, a good botanical library, and extensive, well equipped laboratories.

6. The University of California, at Berkeley, has a botanical garden of several acres, established some years ago, in which a large number of plants are grown. It furnishes a valuable adjunct to the work of the botanical department, which has well appointed laboratories, a working library, and a large herbarium.

7. The University of Pennsylvania has recently established a garden of about three acres in the immediate vicinity of its building in Philadelphia, and has many species under cultivation. The extensive and well appointed laboratories of its School of Biology, good library facilities and a small herbarium afford capital opportunity for research, especially in physiology and morphology.

8. Smith College, at Northampton, Mass., has also recently established a botanical garden on the campus.

9. The Buffalo Botanical Garden, in South Park, Buffalo, N. Y., was commenced in 1893, and has since made rapid and encouraging progress. A small range of greenhouses has been built, and others are planned. A beginning has been made in accumulating a library and herbarium, and much permanent planting has been accomplished.

10. The New York Botanical Garden. The establishment of the New York Botanical Garden was authorized by the Legislature in 1891, and the enabling act was amended in

1894. The enterprise was inaugurated and the legislation. procured by a committee of the Torrey Botanical Club, appointed in 1889. The act of incorporation provided that, when the corporation created should have raised or secured by subscription a sum not less than $250,000, the Commissioners of Public Parks were authorized to set apart and appropriate a portion of one of the public parks, not exceeding two hundred and fifty acres, and the Board of Estimate and Apportionment was authorized to issue bonds, aggregating the sum of $500,000, for the construction and equipment, within the grounds, of the necessary buildings. The subscription of $250,000 required by the Act of Incorporation was completed in June, 1895, and the Commissioners of Public Parks, in the following month, formally appropriated two hundred and fifty acres of the northern part of Bronx Park for the purposes of the Garden. Since that time the preparation of plans for the development of the tract has been steadily progressing, including designs for the museum building and a large horticultural house. This planning is still in progress, in charge of a commission of architects, engineers, gardeners and botanists, who will complete their work within a short time and be ready to submit a complete scheme to the Board of Managers during the coming autumn. Meanwhile, much preliminary work has been accomplished in clearing the ground, in grading, in the planting of borders, in the establishment of an extensive nursery, and in the accumulation of herbarium, museum and library material. Through a cooperative agreement entered into with Columbia University, the herbarium and botanical library of the University will be deposited with the Garden, and most of the research and graduate work of the University in botany will be carried on in the Museum Building.

The endowment fund has been materially increased, and about four hundred and thirty persons have become annual members of the Garden, contributing ten dollars a year each to its support. The publication of a Bulletin has been commenced by the issue, in April, of the first number of Volume I.

THE GLACIAL OR POST-GLACIAL DIVERSION OF THE BRONX RIVER FROM ITS OLD CHANNEL.

BY J. F. KEMP.

As one of the Scientific Directors of the recently organized New York Botanical Garden, the writer has had frequent occasion to visit Bronx Park in the last two years. In one of the earliest of these visits the anomalous relations of the Bronx River to what is its natural line of drainage were noted, and in subsequent ones attempts have been made, not, it must be admitted, with altogether satisfactory results, to explain the present channel. The facts are briefly as follows:

The Bronx River takes its rise a few miles above White Plains and flows southward for thirty miles into the western extremity of Long Island Sound. For much the greater part of its course, it occupies a valley, excavated in a belt of crystalline dolomite that is almost continuous to the salt water. The valley is similar to the usual type of valley in Westchester County, and doubtless owes its depressed character to the easy erosion of the dolomite. The depression is used by the Harlem Railroad from a point just below Morrisania, northward until it crosses into the drainage basin of the Croton River. The Bronx, however, at a point a half a mile or so below Williamsbridge, and just above Bedford Park Station, and in the upper portion of the area assigned to the Botanical Garden, abruptly leaves its old valley and breaks across the enclosing ridge of gneiss, in a gorge 75 feet deep. For nearly a mile it occupies this gorge and then reaching more open country, with a rocky fall at Bronxdale and another at West Farms, it makes its way to the sound.

Just below Williamsbridge it flows against the west side of the valley, and immediately alongside of the railroad. It then leaves this and passes diagonally to the south, being diverted in part by a broad and flat terrace of coarse, rounded, cobble stones, up to one foot or more in diameter and with comparatively little sand intermingled. The cobbles have

been exposed for a thickness of at least 20 feet, by the excavations, for the new Williamsbridge sewer. The river passes along the foot of the ridge on the east side and divides the gravel terrace in two, so as to leave a small remnant on the eastern gneisses. It then runs against a westerly spur of the ridge and cuts through it, in a pronounced gorge, diagonally across the foliation of the rock, which is a hard micaceous gneiss. From the entrance to the gorge a swampy depression extends westward to the railroad and has all the characteristics of an abandoned channel. The railroad has crossed it by an embankment and culvert. Just east of the culvert there is gneiss but a few feet below the soil, and at this point the old stream evidently surmounted a reef. The depression continues southward just west of the track, as far as Morrisania; it is then crossed by the track, which, traversing a low divide into the next limestone valley to the west, follows this to Mott Haven, and then passes over the river to Manhattan Island. The depression runs south from Morrisania and enters the East River opposite Randall's Island. In all its extent there is no natural barrier, although many streets have been filled in across it. At its highest point, near Bedford Park Station, it is not more than ten or twelve feet above the present surface of the Bronx River at the gorge, probably less. The river is about five feet deep at this point. All these relations are shown on the accompanying map, which is reproduced from the Harlem sheet of the United States Geological Survey. The excess of streets and railways on the original map mask the contours, which can only be traced on it with difficulty. Accordingly everything but the contours, the river and one or two railway lines have been omitted in redrawing, and for the same reason no attempt has been made to put in the geology. The authorities of the Botanical Garden have a map of its area on a very large scale and with five foot contours. The writer has used this also in the preparation of the paper, although all the material points are well illustrated on the smaller scale.

By observing the map it will be seen that both the river

and the old channel are below the 60-feet contour just south of Williamsbridge. The terrace of cobblestones is well shown at this point by the projecting 60-feet contour, just under the number 60. The location of the upper part of the old channel where the little brook comes into the Bronx is also shown, and the present almost unrecognizable divide may be identified between it and the brook that flows south into the East River along the old channel. This latter brook is now practically obliterated by street improvements. It cuts the 40-feet contour about a mile and a half south of Williamsbridge, and the 20-feet contour three miles south of the same point. The Bronx, however, passing into the gorge, crosses the 40-feet contour a quarter of a mile from the entrance and the 20-feet contour at West Farms. The new route to salt water is shorter than the old by some two miles, being three as against five, and therefore the fall of 50 feet is the more accentuated under present conditions.

The gorge is somewhat open at its upper end, but it soon closes in and has steep and jagged walls. Some 65 to 75 feet from the bridge at its entrance, and on the west bank, is one of the two potholes, which were described by Dr. N. L. Britton in the Transactions of the New York Academy, Vol. I., p. 181, 1881. It is broken down on the outer side, as is usual with potholes below which the creating stream has cut. The inner half remains, however, and is about 12 feet deep. The bottom of the bowl is quite perfect and shows that the hole must have been 5 or 6 feet in diameter. The bottom is, by aneroid, about 25 feet above the river surface. The walls are still smooth and scarcely decayed at all. A few paces up the hillside to the southeast is another, that is even more impressive. This is only about 5 or 6 feet deep, but much more of it remains than of the lower one. It is 6 feet 2 inches in diameter, and contains a large rounded bowlder, 4 feet by 3 feet by 2 feet 6 inches. A tree a foot in diameter sprouts out from beneath the bowlder. This pothole is 30 feet above the lower one and therefore over 50 feet above the river. Back of it, the hill rises at its summit some 20 feet higher. About