Davis.] 40 [May 3, 



but farther inland and in the upper Mississippi region they are of 

 wide-spread occurrence. In the East, the old surface of the crys- 

 talline rocks was generally too rough to allow an equable, non-ero- 

 sive motion of the ice-sheet ; but in the west, the broad, comp- 

 aratively level country across which the ice advanced, permitted 

 it to pass over loose deposits without seriously disturbing them. 



J. G. Hinde gives a section at Scarboro', on the north shore of Lake 

 Ontario, made up of one hundred and forty feet of sand and clay, seventy 

 of till, ninety of sand and clay, thirty of till, and fifty of sand and clay at 

 the surface; he describes a striated pavement in the same neighborhood. 

 (Canad. Journ. XV, 1877, 388-413.) 



E. Andrews writes, " throughout central Illinois and probably in the 

 corresponding latitude of the adjacent States, the ancient Pliocene soil 

 still lies undisturbed beneath the boulder drift." "A breadth of some two 

 hundred miles along its (the drifts') southern border rests on Pliocene soil, 

 small patches of which are also found a hundred miles north of Chicago." 

 He quotes J. W. Dawson, Hurlburt and Kennicott to show that drift is gen- 

 erally absent on the Laurentian highlands to the north, but considers floods 

 and icebergs as the agents of transportation. (On the Western Boulder 

 Drift, Amer. Journ. Sci., xlviii, 1869, 172-179.) * 



N. H. Winchell, in summarizing the Drift Deposits of the Northwest, 

 describes a gravelly deposit below the " hard-pan " or till; it is not always 

 present, but is common enough to supply water to a wonderful series of 

 Artesian wells in northwestern Ohio; when it occurs the bed-rock does 

 not show so plainly the marks of glaciation, but when absent the rock is 

 nearly always scratched. (Pop. Sci. Monthly, in, 1873, 209.) 



W. J. McGee finds in north-eastern Iowa a boulder clay, averaging 

 twenty-five feet thick, generally resting on stratified blue clay or pebbly 

 clay and sand ; there are no striae, and few good roches mouionne.es and the 

 direction of glacial motion has to be estimated from thy position of the 

 " elliptical kames " (drumlins?). (On the Complete Series of Superficial 

 Formations in North-eastern Iowa, Amer. Assoc. Proc, xxvn, 1878, 198- 

 231. See also Amer. Journ. Sci., xvin, 1879, 301.) 



W. Upham describes the occurrence of decomposed gneiss and granite 

 and many beds of gravel under the drift in Minnesota. (Minn. Geol. Surv. 

 Rept. 1879, 35, 42, 48.) 



B. 10. Drumlins. The intermediate region between the centre 

 and the margin of the glaciated area is sometimes marked by the 

 occurrence of drumlins, or long, arched hills of compact boulder- 

 clay, with their axes parallel to the direction of glacial motion ; 

 their length maybe a mile, and their height one to three hundred 

 feet. We may suppose in a general way that these were formed 



