is an eruptive rock, which has been intruded into the quartzite in the form of dikes or sheets. If this is the case, it corresponds to an igneous rock forming a dike in the quartzite north of Corson, S. Dak. In favor of the latter view may be urged the evident unevenness of its surface. For example, it has been struck at the depth of 500 feet, or 850 feet above the sea, in sec. 25, T. 103 N., R. 61 W., while at Mitchell, less than 5 miles away, at a depth of 710 feet, or 590 feet above the sea, the quartzite, which according to the first view must lie above it, had not been penetrated; or, in other words, the granite had not been reached.

Besides the instance already mentioned, the granite has been struck in two or three wells about 5 miles north of Farmer, and a little beyond the north line of this area near Hitchcock. In the latter case it seemed to be overlain by several feet of quartzite.

The granite from the Budlong and Motley wells north of Hitchcock and from wells north of Farmer, in Hanson County, is a fine-grained, light-gray rock abounding in transparent feldspar, while that from the wells southwest of Mitchell is darker and coarser.

From wells at Huron and near Esmond it appears that below the water-bearing rock there are several feet of secondary deposits from granite, such as arkose, impure kaolin strata, and the like, then weathered granite before the sound rock is reached.

The deepest well at Huron, city well No. 4, opposite the college, according to the report of Mr. F. H. Holton, who drilled it and submitted specimens of the lower strata, shows the following record · Record of deep well at Huron, S. Dak.

Coarse sand with rounded grains about the size of No. 6 shot, largely reddish-brown concretionary grains with some white quartz-like material struck at corresponding levels in the Wilcox well and city well No. 3. The brown grains are apparently siderite, or carbonate of iron

Gray sandy rock with green and brown specks and whitish opaque grains like weathered feldspar (arkose), and lumps of coarse red granite....

Soft. fine-grained, kaolin-like clay.

1,138

Hard rock, probably granite

1

1,139

The Sioux quartzite, which is exposed in the southeastern port of the area under consideration, is commonly a dense and firmly sil. fied rock. It is usually fine grained and thick bedded, but in so cases a few feet of it contains numerous pebbles, while not inf quently the rock is so thin bedded as to be worthless for buildi purposes. Moreover, at some points it is so imperfectly consolidat that it may be dug with a pick. A case of this sort occurs southwe of Bridgewater, in the valley of Wolf Creek. This soft rock, ho ever, is very limited in extent, the normal hard quartzite occurrin within a few feet of it. Associated with the quartzite and interstrat fied with it are occasional thin beds of a red, hardened clay, calle "pipestone." When this has been exposed to the weather it become chalk-white. Examples of the latter condition are found at the Wol Creek locality.

The peculiar interest of the quartzite in connection with the subject under consideration is the fact that it is the bed rock over all the area. The water-bearing strata rest upon it and its presence marks the lowest horizon at which a flow may be obtained. Hence it is of importance to be able to recognize it and to have a general knowledge of its depth below the surface.

In drilling, the quartzite may be distinguished from pyrite, which is of about the same hardness, by its thickness, the latter being rarely more than a few inches thick. Greater thickness also serves to distinguish it from some of the hard layers of the Dakota formation, which are rarely over 4 or 5 feet thick. Moreover, the Dakota rocks are usually cemented by carbonate of lime, which effervesces with acids, or by iron oxide or carbonate of iron, which is of a dark or rusty color, while the quartzite is uniformly of a light pinkish shade. However, in some cases a compound microscope is necessary to detect the difference. When so examined quartzite is recognized by clusters of sand grains cemented together so firmly by glassy silica that usually the original grains divide along the fractures as easily as they separate where cemented. Similar examination suffices to show the presence of granite, which, as has already been stated, has been struck at several points in this area.

BED-ROCK CONTOUR.

In general the quartzite underlies the whole area under consideration and has the configuration shown by the contour lines in Pl. III. The most conspicuous feature of the bed-rock surface is a high ridge, having a breadth of about 12 miles, which enters the east side of the area, affording surface exposures in the vicinity of Spencer and Bridgewater. Its outcrop area narrows rapidly toward the west, so that the westernmost appearance of the rock on the surface is on Enemy Creek, in the western part of T. 102 N., R. 59 W. In the triangle between the limits indicated it lies next underneath the drift and shows frequently in the bottom of the valleys of the larger

streams. It has an altitude of 1,380 feet at Spencer and about the same near Bridgewater. On Enemy Creek it rises to only about 1,200 feet. Its most extensive exposure is in the bottom of James River Valley at Rockport. It has been traced farther west at lower levels in a zone of considerable width, and about a mile southwest of Mitchell has been struck at a depth of nearly 258 feet, or about 1,087 feet above sea level. At Plankinton it was found at a depth of 800 feet, or 700 feet above sea level.

This ridge is not a smooth anticline, but is composed of strata dipping in different directions at from 2° to 5°. It seems to be much eroded in places, where portions of the original surface are left as peaks between the valleys. Differences of elevation of 300 feet or more have been found within the distance of a mile. The northern slopes of this ridge seem to be generally rougher than the southern, and the valleys are correspondingly deeper and narrower. Toward the north the general surface of bed rock drops 600 feet in 8 or 10 miles from the last-observed outcrops. Then by a gentler slope it reaches a depth of 1,000 feet, or 300 feet above sea level, and continues northward till it rises slightly near the northwestern part of this area. Southward the slope is more gentle and a depth of 600 feet is not reached within 20 miles of the last outcrop, which carries the observation much beyond the limit of this area. These general facts are derived from well records, as the wells strike the quartzite very frequently in the vicinity of the ridge, but rarely far from it. The various sections and maps published in this paper exhibit such details as are available concerning the relations and distribution of the quartzite.

PALEOZOIC GAP.

In this region there are no traces, even in the numerous borings which have reached the Algonkian, of any Paleozoic formations or of any Triassic or Jurassic strata. The nearest occurrences of Paleozoic rocks that have been discovered are in the borings at Ponca, Nebr., and Sioux City, Iowa. During Paleozoic times, when great masses of sediments were being deposited in many other regions, this area was probably an elevated land surface. It is possible that soils and vegetation may have extended over it, and possibly some thin deposits which were removed by the advance of the sea during Cretaceous and earlier times. At any rate no traces of such deposits have been found in this vicinity on the surface of the quartzite. Since several hundred feet of strata of marine origin representing all the ages of Paleozoic times are found in the Black Hills, the shore of the Paleozoic sea probably extended across South Dakota from north to south, west of the present course of Missouri River. Moreover, the deep erosion of the intensely hard Sioux quartzite indicates that it was subject to erosion for a great length of time.