APPENDIX. 751 



D. — Glacial Epoch (p. 535). 



Glacial scratches occur in Pennsylvania on the top of Penobscot Knob, 3000 

 feet above the sea-level, and on Peter's Mountain, near Harrisburg, 2000 feet. 

 On the former, where they cover a naked face of rock at the extreme summit, 

 three sets of scratches cross each other, diverging at angles of 25 and 30 degrees. 

 On Peter's Mountain, horizontal scratches occur at the summit on the upright 

 wall of a notch in the rock thirty or forty feet deep (Lesley). 



In the valley of Wyoming, Pa., there is a conformity between the direction 

 of the scratches and the course of the valley (H. D. Rogers). 



On Catskill Mountain, N.Y., according to Ramsay, the scratches are nume- 

 rous, and continue up to the plateau on which the hotel stands, 2850 feet above 

 the sea ; and all but a few of the highest run from north to south along the 

 flanks of the escarpment, or in the direction of the Hudson River valley, and 

 not from toest to east down the slope of the hill. The chief grooves run between 

 S. 22° E. and S. 55° W. ; among them one runs S. 22° E., two S. 10° E., two 

 N. and S., one S. 10° W., six S. 22° W., one S. 30° W., two S. 55° W., one W. 

 10° N. Ramsay observes that the variations seem connected with bends and 

 other irregularities in the face of the great escarpment. The course S. 55° W. 

 was found at the top, near the hotel ; and in the plateau on the summit of the 

 water-shed there are "numerous main grooves passing across the hill at right 

 angles to most of those observed during the ascent" (Quart. Jour. Geol. Soc, xv. 

 209). 



See further, for information on scratches in New York, a table by Mather, in 

 his New York Geological Report, pages 199-206. 



W. B. Dwight, in a recent communication to the author, mentions that at 

 Cherry Valley, N.Y., there are two systems of scratches, nearly at right angles 

 to each other, and none between the two : the directions are (1) S.S.W. to S. by 

 E. and (2) E. by N. These courses, as he says, do not follow the slopes of the 

 minor valleys of the region ; but they do appear to correspond to the grander 

 slopes of the land. The town lies near the summit-level, in the vicinity of the 

 head-waters of the Susquehanna, and also on the south border of the Mohawk 

 valley ; and he suggests, with good reason, that one system is that of a great 

 glacier moving southward along the wide Susquehanna slope of the plateau of 

 southern New York, and the other system that of a glacier moving eastward down 

 the Mohawk valley. The latter probably had its independent movement when 

 the Glacial epoch was on the decline. 



These interesting observations sustain the conclusion that the features of the 

 land have guided the courses of the great glaciers. On page 544, evidence has 

 been stated as to a Connecticut-valley glacier and a Hudson-valley glacier, and, 

 also, on page 545, of a Penobscot-Bay glacier ; and now we have evidence of a 

 Susquehanna-valley glacier and a 3 lohawh -valley glacier. The absence of gla- 

 cier-phenomena about many of the heights bounding the Hudson and Connec- 

 ticut valleys, and the irregularity in the courses of scratches about those other 

 summits on which they occur, serve to define the outline of the independent, or 

 partly independent, glacier-streams. 



In Missouri, unstratified Drift of the Glacial epoch abounds north of the 



