454 STRATIGRAPHICAL GEOLOGY. 



small angle, yet must be inverted along their eastern border (Fig. 71); and 

 they may possibly fold over the quartzite by the school-house opposite 

 Williams's (Fig. 72). The quartzite range is also inverted ; — it hes between 

 two hornblende bands, all three dipping south-easterly. Hence it must 

 constitute an inverted synclinal. There is the possibility, however, that 

 there may be a band of the Bethlehem gneiss between these hornblende 

 areas. I have so called them upon the map, but would prefer the con- 

 clusions of Prof. Dana, that we must not rely too strictly upon lithology; 

 and, as by his conclusion (3) gneiss and quartzite are rocks of the most 

 intimate relations, we may say that the very quartzose gneiss of this range 

 is not of Bethlehem but of Coos age. As once before intimated, the horn- 

 blende schist of Northfield may be regarded as the repetition of two bands 

 of the same rock to the east, and of the Leyden and Shelburne Falls 

 groups farther west, though some of them may be apparently monoclinal. 

 Furthermore, many of the easterly-dipping mica schists in Hinsdale and 

 Winchester may be regarded as inversions, particularly near their eastern 

 border. I cannot now state all. the overturns existing in this area. 



8. It is not so easy to be sure of the repetitions of the quartzite, yet 

 the following seem probable : On the east we have the range from the 

 Perchog river in Winchester through Northfield. This must pass under 

 mica schists, and come up in the range at Doolittle's saw-mill, at South 

 Vernon and West Northfield, which passes south-westerly into Bernards- 

 ton and Gill. These ranges are connected naturally and obviously, the 

 dips not being inverted. The next repetition is the range near a school- 

 house, a mile north of Bernardston. Though it seemed to point directly 

 to the quarry back of Williams's, the discovery of the northward contin- 

 uation of the latter makes any connection between them improbable, save 

 by a synclinal fold beneath the valley of Fall brook. Then, if there is 

 but one band of quartzite, there must be a repetition of it upon the Wil- 

 liams section (Fig. 73). The dips that have been observed show a ten- 

 dency of the rocks to conform to this position. The north-westerly dip 

 of this rock in the north-west part of Gill indicates a connection between 

 the Williams Hill range and that east of Fall brook (Fig. 74). We find 

 here, also, that all the dips along the eastern border of this largest range 

 are not inverted. 



9. Next will follow the most singular of all our conclusions, the proba- 



