468 J. W. SPENCER — PLEISTOCENE SUBMERGENCE. 



rier, even far away toward the south as well as toward the north, that their 

 occurrence demands explanation by other than local causes. 



The highlands of the Ontario peninsula do not form nilometers reaching 

 more than 1,700 feet above the sea ; but in Potter county, in western Penn- 

 sylvania, 100 miles south of Lake Ontario, they develop a water-shed, rising 

 to 2,680 feet above tide, with the Genesee river flowing northward to Lake 

 Ontario; the Alleghany to the Ohio river; and Pine creek to the Susque- 

 hanna. About the highest flattened knob, of only a few acres in extent, and 

 rising to within 20 feet of its summit, there is a low ridge of small, well 

 water- worn gravel, nearly free from sand. Mr. Carvill Lewis speaks of it as 

 kame-like, but its structure and form are not difierent from that which may 

 be a true beach. This is emphasized by the occurrence of a zone of bowlders, 

 forming a pavement a few feet below the gravel ridge — a feature so com- 

 monly developed in front of the deserted beaches of the lake region. This 

 gravel ridge rests upon the highest point of, and at the very front of, the 

 "terminal moraine" of Mr. Lewis, with the land declining to the north, as 

 well as falling away to the south. These gravels form, a superior deposit, 

 resting upon till charged with angular shingle of local Carboniferous sand- 

 stone, and it is out of this material that the pebbles were formed. 



There are similar superficial gravels on other, but of course inferior, knobs 

 along the very foremost portions of the " terminal moraine ; " but the drain- 

 age from these ridges is toward the north, and Mr. Lewis emphasized the 

 fact that there is no drift in the small streams flowing toward the south.* 

 The theoretical importance of this observation will be noted later. 



Besides these highest of all thfe superficial gravels south of the Great Lakes, 

 which I have examined, I have also visited the high terraces of the Genesee 

 river, flowing northward from the deposits just described. Here several 

 pauses in the receding waters are recorded. These are notable from an 

 elevation of 1,900 feet downward. At this high altitude, the valley is nearly 

 a mile wide and now 250 feet below the terrace. Our knowledge of ^these 

 elevated and disconnected water deposits is yet very scanty, but certainly 

 very suggestive when supplementing the surveys of the lower coast mark- 

 ings in the lake region. 



A very interesting terrace remains in a valley three or four miles east of 

 Horseheads, New York. The altitude of the terrace is 1,200 feet above tide, 

 while the gravel-covered floor of the valley, at Horseheads, is only 900 feet. 

 This last valley is over a mile wide, and it is that connecting the trough of 

 Seneca lake with the Susquehanna valley. 



Similar elevated terraces have been noted by Professor I. C. White along 

 the upper Potomac valley facing the Atlantic, and along the adjacent trib- 

 utaries of the Monongahela, which drain to the westward. These deposits 



*" Terminal Moraine," by H. C. Lewis: Geol. Surv. of Pa., rept. Z, 1884, p. 143. 



