Duck and Wind lakes, their relation to the igneous rocks being most perfectly exemplified. Sometimes this red quartzite becomes micaceous and also felsitic, as may be seen at Wind lake. The great extent and the more intense metamorphism of this red formation, in the country lying to the north and west of Lake Superior, accompanied by larger belts of the igneous rock, more coarsely crystalline, not only show that the seat and source of the igneous action was there instead of in the basin of the lake, but also that it was longer continued. It implies also, that a similar modification of these beds may be looked for throughout the northwest, wherever the formation is known to have been upheaved by igneous forces, although the igneous rock itself may be wanting.

Northwest of Lake Superior the igneous rock forms the main watersheds, rising in two main ridges, or ranges of mountains, that run southwestwardly, one known as the Mesabi, and one as the Sawteeth mountains, though the former term is not restricted to this belt of high land. The width of the belt of metamorphic red shales and sandrock, associated with the igneous rock, is about thirty miles in a right line, extending from the headwaters of the Brulé and Temperance rivers to the shore of Lake Superior. The Sawteeth range of mountains, which is that nearer the Lake Superior shore, dies away in passing to the southwest, and the Mesabi belt of igneous outflow approaches the lake shore, appearing at Duluth in the form of the "Rice Point Granite." The tilted red shales, conglomerates and sandstones at Fond du Lac, a few miles west of Duluth, are the same as those seen associated with the igneous rock all along the shore. They lie there on a white-quartz, pebbly conglomerate, of a few feet in thickness, which lies unconformably on the roofing slates of the Huronian, the same formation that succeeds to the red rock formation toward the northwest, at Ogishike Muncie and Knife lakes, northwest of Grand Marais.

The mineralogical characters of these belts of igneous rock, which form some of the main features of the topography, seem to ally them to the Norian rocks of T. S. Hunt, and to the labradorite rocks of Canada. At least, if they be not the western extension of those formations, then those formations have not yet been discovered in Minnesota. But several traverses have been

hill ranges.

made of the country northwest of Lake Superior, for the purpose of geological examinations without finding anything that is at all comparable to those formations, if it be not the rock of these The rock consists generally of some felspar, which at Duluth has been found to be labradorite in large per cent., and at some places constituting over ninety per cent. of the mass, with varying proportions of augite and magnetite, or magnetic menaccanite, with various accessory ingredients, or minerals that result from change. It is massive, firm, dark-colored, and rises. in low mountain ranges, as already stated. If its relation to the red granites and gneisses with which it is accompanied were not so evident, by simply noting the changes from the lake shore northwestwardly, it would hardly be presumed to be a parallel of the igneous rocks of the coast, any more than the red gneisses and quartzites would be of the shales and sandstones that are interbedded with them at the coast. So far as yet examined, these labradorite rocks contain no bands of limestone, which is due probably to the fact that the Cupriferous Series in the northwest is not known to contain any beds of limestone. In the absence of this element, and in this only, so far as can be judged by the writer, these labradorite rocks seem to differ from the labradorite rocks of the "Upper Laurentian" of Canada.

Inferentially, therefore, the so-called "Upper Laurentian," containing Eozoon Canadense, seems to parallelize with the igneous rocks of the Cupriferous Series, or rather with the modified interbedded sedimentary portions of it, and hence the Eozoon, instead of being truly a Laurentian organism, seems to be one of the Cambrian or Lower Silurian. The abundant graphite of the "Laurentian" which pointed the way to the prediction of organisms in that formation, is also found in the modified quartzites and shales of the Cupriferous Series in perhaps equal abundance in the state of Minnesota.

AN ATTEMPT ΤΟ ESTIMATE APPROXIMATELY THE DATE OF THE CLOSE OF THE GLACIAL EPOCH, FROM AN INSPECTION OF THE KAMES AND KEettle-holes of NEW ENGLAND. By G. FREDERICK WRIGHT, of Andover, Mass.

[ABSTRACT.']

The distance from The rim is composed

AFTER referring to the evidence connecting these phenomena with the closing period of the glacial epoch, he presented, as illustrative of innumerable other cases, the facts concerning a kettle-hole near Pomp's pond in Andover. rim to rim of the hole described is 380 feet. of gravel. The height of the rim above the surrounding plain is from forty to fifty-two feet. The height above the peat bog in the centre of the depression is fifty feet. The distance across this bog is ninety-six feet. The nature of the material composing the rim is such, that the depth of the depression could never have been more than twenty-four feet greater than now. Twenty-four feet of sediment in the bottom of a conical depression would only equal eight feet upon the present surface. The question is how long a time would be required for the wash of the rains, the dirt brought by the winds and the solid material collected by vegetation, to fill this depression to the amount indicated. No definite answer can be given, but if one hold, with Mr. Croll, that the great glacial period closed 80,000 years ago, he must believe that only an inch of sediment would accumulate upon the area of this peat bog in 1,000 years. Whereas, if he bring the close of this period down to 10,000 years ago, the rate of accumulation would seem sufficiently slow to tax severely even a credulous imagination.

1The paper with a plate is published in full, in the American Journal of Science for February, 1881.

THE IRON ORES OF THE BRANDON PERIOD. BY HENRY CARVILL LEWIS, of Germantown, Pa.

[ABSTRACT.]

THE theory that a great portion of the iron ores of our lower Silurian limestone valleys are of a tertiary age was first proposed by Prof. E. Hitchcock, but has been rejected by many geologists. The present paper describes in full recent discoveries, made by the writer, of lignite associated with limonite iron ores in the limestone valley of Montgomery county, Pa., and shows their relation to the deposit at Brandon, Vt., and their bearing upon a theory of the age of iron ores in similar positions in the Atlantic States. The lignite of Brandon, lying within beds of plastic clay, kaolin and iron ore, was shown by Lesquereux to be of tertiary age. Lesley afterwards described strata of lignite in a similar position at Chambersburg, Pa., but regarded them as local deposits of late date. More recently Prime has found lignite in a plastic clay at Ironton, Pa., and supposed it to have been transported by a glacier. The present paper shows that in each of these cases the lignite lies below the surface drift, and that, as at Brandon, the latter lies unconformably upon the plastic clays containing the lignite.

The occurrence of lignite, in connection with limonite iron ore, plastic clay, kaolin and firesand in a number of places in Montgomery county, Pa., is described, and it is shown that these localities lie in a line corresponding to the line of strike of all the iron ores of the valley. Overlying the plastic clay which contains the lignite is what appears to be a decomposed lower Silurian hydromica slate, and for this reason the iron ores had been supposed to be of primal age. It is shown that this decomposed material and the underlying iron ores have been originally derived from lower Silurian slates, and have been re-stratified in an age intermediate between Triassic and Upper Tertiary.

The iron ores of this region are divided into four classes: (1) Gneissic Ore; (2) Primal Ore; (3) Tertiary (Brandon) Ore; (4) Drift Ore. The last two classes of ore are often found at the

The Iron ores and Lignite of the Montgomery Co. Valley, by the writer. Proc. Acad. Nat. Sciences, Phila., 1880.

same locality; the latter lying unconformably upon the former. The paper discusses at length the age of the drift containing the latter. Notwithstanding the fact that a region of triassic red shale lies north and east of the valley, not a single fragment of such rock occurs in this drift. The pebbles are composed almost wholly of Potsdam sandstone, -a material now in great part eroded away in this vicinity. The evidence is strong that this drift was not caused by any flood from the north. That it is older than the Glacial Epoch is also indicated both by the great amount of erosion it has suffered, and by the fact that in the adjoining triassic region no trace of drift occurs. It was perhaps formed at a time when hills of Potsdam sandstone, since eroded, stood as a barrier between the limestone valley and the triassic rocks to the north. It is of interest to find that a large proportion of the pebbles of the sub-cretaceous clays of New Jersey are also formed of Potsdam. The four different gravels of the Delaware valley are described, and it is shown that the drift ore of the Montgomery county valley belongs to the oldest of these, and is proba bly of tertiary age.

It follows that the strata containing iron ore and lignite, which underlie this drift unconformably, are yet older. Some facts point to a Wealden age, but the identity of the deposits with that at Brandon, in which Tertiary plants are found, indicates a middle Tertiary, perhaps Oligocene age. Since an exact geological age cannot at present be assigned to these deposits, it is thought best to group them together under the name of the Brandon Period.

Attention is directed to another deposit of lignite and iron ore near Augusta, Ga., whose geological situation and the section given are so remarkably similar to those of Brandon, Chambersburg, Ironton and the Montgomery county valley, that it appears to belong to the same formation, and to indicate that in this Brandon Period there was an inland fresh-water formation in eastern America of large extent and importance.