THE ARCHAEAN ROCKS. 507 



ing horizontal strata, see Atlas Plate XIV, and the sections of Plate XXI of this vol- 

 ume. If the view, just indicated, that there are no folds concealed beneath the sand- 

 stone in the intervening valley, is the correct one, the thickness of the entire series must 

 be very great, and the amount of erosion that has taken place correspondingly great. 

 Fig. 23 indicates the present structure and relative positions of the ridges, and, by the 

 dotted lines above, the possible original structure, and the extent of the erosion that has 

 taken place. The figure is drawn to a natural scale, the line of section being the same 

 as that of Fig. II of Plate XX. The heavy black line represents the overlying Potsdam 

 sandstone. It is not impossible that the Valley between the ranges owed its existence, 

 to some extent, in the first place, to soft rocks intercalated between the harder quart- 

 zites. The hypothesis of Fig. 23 is not altogether satisfactory. The entire disappear- 

 ance of the other side of the great arch, as well as the peculiar ways in which the 

 ranges come together at their extremities are difficult to explain by it. It may be said 

 in this connection that the dip observations toward the west are not so satisfactory or 

 numerous as they might be. 



The irregular areas over which the Archaean rocks are at surface, are indicated, as ac- 

 curately as present knowledge will permit, on Atla*jPlate XIV. The greatest difficulty 

 in the tracing of the exact boundaries of the quartzite areas lies in the fact that rem- 

 nants of the horizontal sandstones which flank and cover them may be found at almost 

 any elevation upon the bluffs, so that no barometrical observations are of avail. The 

 areas, as indicated, are, however, very nearly accurate. There are places within them 

 where, without doubt, patches of the covering sandstone occur, but the quaitzite is in 

 eveiy such case but a short distance beneath. The peculiar features of these sandstones 

 and their relations to the other Silurian strata of the region are treated of on a subse- 

 quent page. 



Beginning tlae detailed descriptions at the best known, and at the same time one of 

 the most remarkable, points about the quartzite ranges, we note first the occurrences in 

 the vicinity of Devil's Lake, in T. 11, ranges 6 and 7 E., Sauk county. Here the 

 southern range is cut entirely through by a deep quartzite- walled valley or gorge, 500 

 feet in depth, and three-fourths of a mile in width. In its northern portion this valley 

 trends due north and south for about a mile; turning then abruptly at right angles it 

 extends eastward two miles and a half. In the north and south part lies Devil's Lake, 

 with a length of something more than a mile, and a width one-fourth less than this, its 

 surface being about 100 feet above the valley surface at the eastern end of the gorge, 

 more than 100 feet above the Baraboo river at Baraboo, and more than 200 above tlia 

 Wisconsin at Merrimack. It is held in this elevated position by two immense morainic 

 heaps of glacial drift lying at either end of the lake and rising more than 100 feet 

 above its level. The lake has a nearly level sandy bottom except near the shores, and is 

 over most of its area some 30 feet in depth. It has no outlet, and but one small stream 

 running into it. It is thus probably fed chiefly by springs, and maintains its level by 

 evaporation and by filtering through the heaps of gravel and sand which hold it in 

 place. Near the northwest corner a small stream running into the Baraboo passes 

 within a few rods of the lake, and possibly carries with it some of the lake water. 



As shown on the map of Plate XIX, on its west and south sides the lake washes the 

 bases of the bounding cliffs of the gorge. Both east and west cliffs are highest near 

 their southern ends, towards which they rise gradually from the north, following roughly 

 the dip planes, which also rise southward, and the edges of which can be seen quite well 

 marked on either wall of the gorge. By aneroid measurement the highest point of the 

 west bluff is 475 feet above the lake level or 860 feet above Lake Michigan, and the 

 southern portion of the east bluff but little lower. In their upper portions the cliffs are 

 vertical, sometimes for as much as 200 feet or more, but their lower parts are clothed 



