GLACIAL EKOSION ON KELLEYS ISLAND (345 



evidently was near at hand, for the sharp-cut details of striation indicate 

 freshness of the tools; (2) the country rock is not appreciably harder than 

 the tools used in carving it ; this conclusion is arrived at partly because of the 

 absence of crystallines in the neighboring drift, also because tools secured 

 from a more distant source probably would have suffered much from abrasion 

 in transit, thus losing the sharper outlines and becoming less liable to produce 

 tbe accurately defined striae above described; furthermore, there is positive 

 evidence in the drift that the local limestone was the chief source of the 

 debris; (3) these grooves indicate a concentration of erosive processes within 

 a narrow limit, or, better, a superimposed alignment of stria?, causing at first 

 a depression below the general level of the rock surface; (4) the basal ice 

 under pressure moulded itself to the slight depression thus made ; in conse- 

 quence there was further localization of mechanical work. That the ice con- 

 tinued to fit the grooves as they were deepened is shown in the fact that the 

 sides of the grooves are delicately striated, and in the further fact that in one 

 groove at least there was such a continued supply of tools as to cut the wall 

 back farther and farther, producing an overhanging condition. All parts of 

 this overhanging wall are striated. 



Summary 



Cumulative observation has shown (1) that glacier ice, of either the conti- 

 nental or alpine type, when moulded to or confined within a valley, changes its 

 subaerial and water-erosion profile to a U-shaped cross-section; (2) that this 

 ice continuing to occupy such a valley is competent to overdeepen it hundreds 

 of feet. 



The grooves on Kelleys island show that when tools are so localized in the 

 basal ice as to abrade continuously within a narrow limit, say a few inches, 

 the result shortly is a shallow elliptical depression, and, later, the action con- 

 tinuing to be localized, a U-shaped trough, which may become deeper than 

 broad. The weight of the ice mass keeps it moulded to the growing groove, 

 which is enlarged so long as cutting tools, even grains of sand, are present. 

 This behavior of ice is somewhat analogous to tbe response of a plastic sub- 

 stance under pressure ; in consequence the tools grind and scour laterally as 

 well as on the bottom of the groove. When ice feeds alpine-like through a 

 valley, or when the movement of an ice cap trends with a valley, overdeepen- 

 ing will inevitably follow if the ice carries tools in contact with the valley 

 walls. 



The paper was discussed by G. F. Wright. 

 Then was read 



CHALK FORMATIONS OF NORTHEAST TEXAS 

 BY C. H. GORDON 



[Abstract} 



Extending in a west to east direction across the southern part of Lamar 

 county, and thence northeast through Red River county to Red river, and hav- 

 ing a width of from 1 to 3 miles, is a belt of chalk known as the Annona chalk, 

 from the town in Red River county near which it outcrops. In the earlier pub- 



