the west, in a very undulating course, and marked by several transverse depressions, to a high isolated summit,' adjoining the north line of the east of the North Ponds' (Salisbury)." (P. 129). In a paper read before the American Association in 18733 Professor J. D. Dana quotes Percival as stating that the mica schist in which he found garnets in the township of Salisbury, is below the "Stockbridge or Canaan Limestone," but giving it as his own view that the schist is the overlying rock. This observation of Percival has considerable interest, for though the "Stockbridge or Canaan Limestone" has been shown to consist of two members, one of which is below and the other above the Staurolite-bearing rock, it is probable that Percival discovered a locality at which the Riga Schist comes out from below the Egremont Limestone. On the map accompanying Professor Dana's paper entitled Taconic Rocks and Stratigraphy, a number of schist areas are represented within the area here treated, which he correctly described to be, in some cases at least, "isolated within the limestone area, as isolated as islands in a sea." 5 He mentions eleven of them in Salisbury and eight in the part of Sheffield township just north. He believed that there is but one schist horizon, which overlies the limestone, and described three localities, nearly or quite within the area studied, to sustain his views. These are, (1) the hill three miles north of Gallows Hill (locality 4, 1. c., p. 213) where the schist "overlies the limestone"; (2) Turnip Rock (locality 5, l. c., p. 213) where schist overlies limestone in a shallow synclinal; and (3) Tom's Hill in Salisbury, which is described as a very flat trough of schist toward the north, but developing farther south into an overturned synclinal with its axis dipping east (1. c., p. 214). The observations made by Tom's Hill. 2 Twin Lakes. 3 On Staurolite Crystals and Green Mountain Gneisses of the Silurian Age, by J. D. DANA. Proc. A. A. A. S., 22d (Portland) Meeting, 1875, p. B25. 4 American Journal of Science, Vol. XXIX., June, 1885. 5 Amer. Jour. Sci., Vol. XXIX., March, 1885, p. 211. the writer accord with those of Professor Dana in the second instance only, which relates to the upper or Everett schist member. As will be fully shown below, the other mentioned localities have a much more complicated structure than was supposed by Professor Dana. LITHOLOGICAL CHARACTERS OF THE HORIZONS. As has already been stated, the horizons outcropping within this area all belong to the Mt. Washington series, viz.: The Canaan Dolomite, the Riga Schist, the Egremont Limestone, and the Everett Schist. The Canaan Dolmite seems to be for the most part a dolomite or dolomitic limestone, with more or less admixed quartz. A white pyroxene or salite is found to be common in it in the vicinity of Canaan, and in the belts extending east and northeast from that point. It has also been found at several localities in the vicinity of Lime Rock, but is only rarely seen west and southwest of that place. Tremolite is also found in this horizon, but as will be more fully explained beyond, this is largely restricted to a zone bordering the Housatonic River on the east. Masses of Canaanite are also found in this horizon, and as neither pyroxene nor tremolite has been found in the Egremont Limestone, their presence here is useful for purposes of identification. The Riga Schist within this area has the characters which distinguish it on Mount Washington. In most of the ridges where it occurs, garnets alone or garnets and staurolites have been found in it. They are most abundant and of largest dimensions in the ridge south of Twin Lakes Station, the ridge south of Chapinville Station, in Tom's Hill and Mile's Hill, in Mt. Prospect (south of the area here mapped), and in Barnard Mt. and Johnny's Mt.' near Sheffield. The mica is often a silvery These minerals were described from this locality in 1824 by Dr. Chester Dewey. Am. Journ. Sci., Vol. VIII., p. 7. 2 Professor Dana has specially mentioned them from many of these localities. (1. c., p. 440). The increase in size of garnets and staurolite from Mt. Washington to the Housatonic, as described by him, has not been confirmed by this study. The largest that have been noted are from the south end of the ridge south of Chapinville Station. sericite and considerable graphite is sometimes associated with it. The Egremont Limestone resembles that found along the east base of Mt. Washington, its principal impurities being muscovite and quartz. It contains locally important layers of calcareous mica schist. In the vicinity of Twin Lakes, two distinct beds of the latter are made out, one immediately below the Everett Schist-a transitional zone-and the other lower down near the middle of the horizon. A third, less important and less constant, zone forms a transition from the Riga Schist to the Egremont Limestone. The upper of these layers forms the cap of Babe's Hill (northeast of Washining Lake). The middle layer is also found in the same hill along the southwest base, and the lowest layer may be seen above the Riga Schist at the first road-corner northeast of Chapinville. Graphitic phases are found as a transitional zone between this horizon and the overlying Everett Schist in the northeastern part of the area, particularly in areas 16 and 25. The Everett Schist is not chloritic to any marked degree, as is so often the case on Mt. Washington, but is frequently sericitic, usually porphyritic from rounded eyes of feldspar, and frequently passes downward into graphitic schist. EXPLANATION OF MAP. The map which accompanies this paper (Plate V.) is based on the Sheffield and Cornwall sheets of the topographical atlas of the United States, by the U. S. Geological Survey, and is drawn on the same scale-1: 62,500, or one inch to the mile. It overlaps by about one half mile the map which accompanies the Mt. Washington paper. To bring as much of the area ast possible on the page, the narrow northern portion is placed in one corner, its actual position being roughly indicated by the positions of the Housatonic Railroad and the large marsh to the west of it. Fig. 5 also extends the map some distance to the south. The area covered by the Egremont Limestone is left blank, while the Riga and Everett Schist areas are shaded, the |