SoLiAS — A Map to show the Distribution of Eslcers in Ireland. 789 



definitely mentions " ozars," only to explain them as clue to " the constant breaking 

 of the waves." 



The same omission to connect the Ohio ridges with esters occurs in later 

 writings by this author.* At the same time there can be no manner of doubt that 

 the features which Professor Winchell had under observation are true eskers, and 

 therefore we must justly assign to him a share of the credit of first attributing to 

 them a glacial origin. 



In the same year Mr. James Shawf described gravel hills in Illinois (which 

 appear to be eskers, though this is not expressly stated),, and attributed them to 

 the combined action of glaciers and water. 



Dr. D. HummelJ was the first to propose and elaborate a complete and 

 consistent theory of eskers. An excellent summary of this is given in Professor 

 J. Geikie's Ice Age.^ The following account is taken from Dr. Hummel's memoir. 

 The eskers on which the author bases his theory are those of Southern Sweden, 

 and particularly those of the high ground, more than 400 feet above the sea-level, 

 and so beyond the limits of the glacial submergence ; in their case, consequently, 

 an appeal to the action of the sea is excluded. The only other theory (with 

 which the name of Tornebohm is most closely associated) making any pretence to 

 explanation presupposes the existence of great plains of mud, through which 

 rivers cut their way, forming valleys in which the gravel and sand of the eskers 

 were deposited. By the subsequent removal of the mud, the eskers acquired their 

 existing shape. Hummel remarks of this, that it is inconceivable that so much 

 mud should have existed without leaving some trace behind ; particularly as the 

 form of the surface of the Swedish plateau is such as would be likely to prevent 

 its complete removal. Traces of the mud-plain might certainly be expected to 

 occur between the esker deposits and the imderlyiug boidder-clay ; but though 

 the contact of the two latter can be plainly seen, there are no signs of the mud — 

 on the contrary, the boulder-clay and the esker material pass into each other. 



Stratified deposits covering the eskers are of rare occurrence and of inconsider- 

 able extent : in tracts where they generally occur, one is accustomed to distinguish 

 an interior part, or kernel, from an outer part or shell ; the kernel is evidently 

 contemporaneous with the ancient ice-sheet. This is shown by the intimate 

 association of the Yoldia clay with the shell of the esker in the lowlands, and the 

 occurrence of boulder- clay immediately below the kernel. 



* The Geological and Nat. Hist. Survey of Minnesota. — First Annual Eeport for the year 1872 (1873), 

 p. 62 ; and Second Annual Eeport, 1873 (1874), p. 194. 



f Geological Survey of Illinois, vol. v., p. 108, 1873. 



X Om Rullstenshildningar, by D. Hummel. — Bihang till K. Svenska Yet. Akad. HandUngar, Bd. H., 

 No. 11, p. 3, 1874. 



§ The Great Ice Age, 3rd edition, 1894, p. 172. 



