found from forty to fifty miles south of that line. The great facility with which this rock is recognized, its character so peculiar, contrasting so strongly with all the rocks of New York, its layers so thick at Oriskany Falls, make it a useful rock in investigating the history of the ancient flows of water in that section of the country." ""*

Going south from the ledge, diagonally across the valley, the southeast valley wall is reached at a distance of two miles. The summit of the range (D. map. p. 219) is three-fourths of a mile further and 600 feet above the ledge. The kame deposits of the moraine are here at least a mile wide, banked against the southeast side. They are transversely cut by a small valley heading near the southeast wall of the main valley. This depression in the moraine was doubtless kept open by a small ice tongue and near its head, but still on the moraine, is a profuse assemblage of large Oriskany bowlders (at B), probably the last of the train to come to rest. The hill-slopes rising directly above the moraine (C) are strewn with the same large fragments. The larger masses and by far the greater number lie on the lower 200 feet of the hillside, at an average altitude of 1300 feet, continuously for several miles to the southwest, (CEFG). In good numbers and up to a diameter of four feet, they rise to the top and are occasionally carried over the summit and through gaps in the range, to the further slopes of another valley which lies several miles to the southeast, about Hubbardsville (N).

At several points on the lower hill-slope referred to, 100 to 150 bowlders ranging from 2 to 7 feet in diameter can be counted upon a single acre. Three miles from the ledge (E) is found an Oriskany block measuring 19×12x8 feet. Its altitude is 1416 feet. On the slope 150 feet above Madison village (F) on a plot of three to four acres, 86 Oriskany masses were counted, whose dimensions range from 6 to 15 feet. Of these about 15 have lengths and breadths of 10 and 12 feet. All would probably average six foot cubes. Many smaller bowlders, not included in the count, range from five feet downward. Six four foot limestone slabs and one Archæan of like size, were found in this field; but the Oriskanys had been sorted and grouped to a surprising degree. The valley from Oriskany Falls to Pecksport has a bearing of S. 60° W., or at least 40° more westerly than the general ice movement. The hillsides on the northwest of this section of the valley are free from Oriskany bowlders. In the general southward movement

* For other early discussions of New York bowlders see James Hall, Geol. 4th Distr., pp. 332-341, and W. W. Mather, Geol. 1st Distr., pp. 163-197. The latter adds a review of bowlders in New England, the Western States and foreign countries, with numerous references to the earlier literature.

the bowlders were swept diagonally across the valley, a moderate number were carried to, or over the summit of the range, but by far the greater number and almost all the larger fragments were swerved from 0° to 28° westward by the valley wall and caught on the lower slopes. It is a good illustration of the law stated for such cases by Professor Chamberlin.* In further harmony with the passage cited, it is to be noted that the Oriskany valley is neither so broad and shallow as to fail of diversion of the basal currents, nor so narrow and deep as to cause a decided cross current. There can be little doubt that the bowlders in question were moved while the ice was deep and its action vigorous, rather than by an ice tongue at a later and decadent stage.

Near Hamilton the valley bends toward the east and hence the Oriskany bowlders begin to appear on the western slopes. East of Hamilton near the base of the hill, an Oriskany block measures 21x12x10 feet. On the campus of Colgate University, eleven miles from the ledge, the bowlders are abundant, the largest measuring 15X12x10 feet. From Hamilton to Earlville, masses with dimensions of 6 to 8 feet are not uncommon. From Earlville to Smyrna the number is much smaller and but three or four were seen having a diameter of four feet. Near Norwich the track of the ice was crossed for 10 miles. But two Oriskany pieces were seen, each about one foot in diameter, on the lower hill-slope, one mile southwest of Norwich. The last Oriskany fragment recognized, was an 8 inch piece in the valley 1 miles north of Oxford, or 40 miles from the nearest outcrop.

Vertical distribution.-Oriskany material is infrequent in the kames and gravel trains of the valleys, which, being of somewhat later date, may have buried such of the Oriskany blocks as rested in the valleys. The greatest abundance is on the lower slopes facing the approach of the ice, with moderate carriage to the summits, dropping off to the lower levels, however, as we go southward. When we turn to the question of amount of glacial elevation, it must be noted that Oriskany bowlders occur on the hill range north of the ledge for five miles (to O). Portions of the Oriskany terrane were swept from the summit, whose floor is now of lower Helderberg limestone. Taking into account the dip and comparing the highest northern extension of the preglacial ledge with the greatest height attained by the bowlders in Madison, we find that 248 feet in 8 miles gives the largest rate of ascent which the facts compel us to accept. Several considerations make it probable that the most important assemblage of blocks (at CEFG) was derived from the ledge now seen at Oriskany * Rock Scorings of the Great Ice Invasions, 7th Ann. Rep. U. S. G. S., pp. 197–200.

Falls. This would give them an average elevation of 250 feet. in an average distance of 4 miles. The single mass before noted (E) would thus have been raised 366 feet in 34 miles. Upham notes the uplifting of Niagara bowlders to a height of 100 to 200 feet within three or four miles.* C. H. Hitchcock reports the lifting of bowlders 4,000 feet in the White Mountains within moderate distances.†

As already suggested, the east and west extension of the central New York terranes vitiates conclusions as to lateral distribution, or "fanning out" such as has been described by Professor Shaler in Rhode Island and Massachusetts. The Oriskany sandstone is to a considerable extent exceptional, however. In the passage above cited, Vanuxem alludes to local differences and recognizes the Oriskany drift as from Oriskany Falls. For the only outcrop in the next valley to the eastward he notes the presence of a distinctive character. At Munnsville, 6, miles W. N. W. the Oriskany is not present, the two limestones being continuous, and the bowlders are nearly absent from Munnsville to Pecksport. If the Oriskany exists in the valley at all it is a thin wedge south of Pratts, carried by the dip beneath the valley filling. At Morrisville, 10 miles W. S. W., a deep boring apparently found no trace of Oriskany, except quartz grains in the limestone. The ledge at Oriskany Falls thins 4 to 5 feet southward in three-quarters of a mile. The indications are that it thinned northward also. It is a shore formation, which we should expect to find variable within short distances. We probably have a thin lens of sediment, whose maximum thickness is at the present exposed ledge, and the southward bowlders are very surely to be connected with it. In the adjacent valleys, east and west, the Oriskany bowlders, with slight exception, do not occur, or are found so far southward that they can have come across from the Oriskany valley.

The fragments fan out to a width of six miles in a distance of eight miles. This cannot, however, be taken as a measure of normal dispersion on a plane surface, for the diagonally disposed valley has widened the train rapidly to the westward.

The Oriskany bowlders were observed with a view to ascertaining the comparative amount of disintegration at the ledge and southward. There has been appreciable loss from the sur

Eskers, near Rochester, N. Y., Proc. Roch. Acad. Sci., vol. ii, p. 196, and Bull. G. S. A., vol. v, p. 76.

Bull. G. S. A., vol. v, p. 37. See also this Journal, III, vol. xxviii. p. 233; Ibid., vol. v, pp. 218, 219. T. C. Chamberlin, Jour. Geol., vol. i, pp. 50–51.

C. S. Prosser, "The Thickness of the Devonian and Silurian Rocks of Central N. Y.," Bull. G. S. A., vol. iv, p. 96.

§ Cf. "Bowlder Train from Iron Hill," Bull. Mus. Comp. Zool., vol. xvi, No. 11, pp. 196-202.

faces only upon relatively flat-topped masses, where water and frosts would act effectively, and along fossil zones, where a retreat of three to six inches from the general surface is not uncommon. On the whole it is thought that the surfaces at the ledge average slightly fresher than those to the southward. On the other hand the impression given is, that all the Oriskany masses belong to the same general movement, and that movement a late one. No buried specimens of large Oriskany bowlders have been found. They have the appearance of being left behind, upon the surface of the hills, in the course of a rapid retreat. They are not often associated with morainic accumulations, but lie upon thinly masked linear and often drumloidal elevations. The relations of the ledge, the bowlders associated with the valley kames at B, and the bowlder fields of the succeeding hills, strongly suggest that we are dealing with one piece of work so far as the Oriskany material is concerned.

But it does not seem probable that the carriage and letting down of the Oriskany bowlders can belong to the general advance to the southern limit of glaciation and the retreat therefrom; for on this supposition they should have reached to Binghamton and beyond. The texture, indeed, is not so firm as that of the Oneidas, but the comparatively abrupt manner in which they terminate renders their connection with a long continued and far southward movement questionable. It is suggested as a matter for farther inquiry, whether the termination of some such vigorous and rapid forward movement as seems indicated by these bowlders, may not be marked by the kames and other valley accumulations between Oxford and Greene, just southward from the last found pieces of the sandstone. This section belongs to a "Collateral belt of moraines " described by Chamberlin, in this region.*

Corniferous Limestones.-This formation overlies the Oriskany sandstone at Oriskany Falls, and has a thickness of at least 50 feet. For three or four miles it is quite freely mixed with other rocks in the hillside drift. Northwest of Hamilton, where the turn of the valley brings the Oriskany drift upon its western wall, Corniferous bowlders are found far exceeding those seen elsewhere. One slab near the valley bottom measures 19×11×3 feet. Farther up, other and more rounded. masses were seen, one of which measured 6×6×9 feet. Near Eaton village also is a tract where limestones exceed all other materials of the bowlder drift. This illustrates a quite general fact in the distribution of the more local drift, that the blocks of a given kind are often found in groups, dependent upon local conditions of exposure, plucking and transport. South

* Term. Mor. Sec. Gl. Exp, 3d Ann. Rep. U. S. G. S., p. 372,

of Hamilton, limestones are rare above the kame and terrace limit, for which 100 feet is an average figure.

The writer has not found a fully satisfactory interpretation for the distribution of the limestones. The case may be stated as follows: from 8 to 12 feet of sandstone are intercalated in the midst of 200 or more feet of limestones at Oriskany Falls. The whole mass was attacked by the glacier. The amount of limestone left on the hills southward bears no comparison to the amount of sandstone. The limestone is practically absent from the hill summits beyond six miles. The Oriskany continues in good force for over 20 miles. That the Lower Helderberg continues far south in the valley train, might be expected, from prolonged erosion in the valley bottom at Oriskany Falls, during the presence of the ice front there, after work upon the higher sandstones had become ineffective. But how shall we get the limestone deposited in kames, with some striations still preserved, at Oxford, 40 miles south, without carrying the fragments of the same terrane to the hills above? We might suppose a valley tongue to account for the kames and the absence from the hills, but the distance is too great, with the slight descent, to admit of this view. Owing to the solubility of the limestones we may believe them less capable of distant carriage by glacial action, the pebbles of the valley being hastened forward by sub-glacial water currents. But we have records of limestone carriage for at least 60 miles.* Professor Shaler notes the disappearance of crystalline limestones within five or six miles. But here again Emmons records the finding of many bowlders of "Primary" limestone south of Rome, New York, and one two feet in diameter is reported from Clinton, New York. These latter fragments have with little doubt traveled as much as 60 miles.

We may perhaps find relief in the history of the gathering of the materials by the ice. The Corniferous overlaid by the perishable Marcellus shale and breaking up easily owing to its large quota of flints and the solubility of the lime, may have receded along the lines of its ancient outcrop, leaving considerable areas of sandstone exposed. A short and vigorous ice movement might pluck and carry off extensive masses of the sandstone, before action became highly effective upon the limestone ledges in retreat above, or those protected below. We may thus suppose the Oriskany to be well started on its jour ney before the migration of the limestones was fairly begun.

* Chamberlin and Salisbury, Driftless Area of the Upper Mississippi Valley, 6th Ann. Rep. U. S. G. S., p. 267; also, A. P. Low, Glacial Geology of Labrador and Quebec, Bull, G. S. A., vol. iv. pp. 419-421.

206.

Bowlder Train from Iron Hill, Bull. Mus. Comp. Zool., vol. xvi, No. 11, p.

Agriculture of N. Y., vol. i, p. 260.