GLACIAL EROSION ON KELLEYS ISLAND 643 



ually pared off along the shearing planes developed over its surface. It is 

 possible that in the course of time these shearing surfaces may be so lowered 

 as to gradually carry away the entire mass of stagnant ice. During the period 

 of stagnation the surface immediately covered was protected from further ice 

 action. In accordance with the above discussion, this surface could not have 

 been much scored in the process of overloading ; but previously it may have 

 been striated. Furthermore, it is possible that the gliding surface once estab- 

 lished niay persist, and the overloaded mass become, in effect, a part of the 

 bed over which the glacier continues to move. 



So far we have discussed only the conditions observed in a longitudinal 

 cross-section of the ice. The basal ice on either side of this overloaded lens. 

 immediately contiguous, lost velocity; here, too, shearing planes were devel- 

 oped, and the ice laterally moved with the general basal trend. If this ice 

 were properly shod with rock, the surface beneath it would suffer abrasion ; so 

 under these conditions there might be an area on all sides of which the rock 

 surface has been smoothed and striated. Remembering, however, that debris 

 in an ice-sheet is not apt to be homogeneously distributed, such general erosion 

 is seldom the case. 



As the ice may carry a plentiful supply of tools in a narrow longitudinal 

 band at its base, and adjacent to this band be relatively free of rubbish, we 

 might find just such a relation of surfaces as are shown in figure 1, plate 108. 

 The fact, however, that glacially scored surfaces are interrupted in the line of 

 ice motion by unscored surfaces gives much significance to the theory of up- 

 ward and downward movement of currents within the ice-mass. We have 

 long known that the erosive work of rivers is accomplished largely by such 

 secondary currents, which are accessory to the main flow of the river. It 

 seems not unreasonable that in an ice-sheet there may exist similar secondary 

 currents, occasioned also by irregular topography, and that when these lines 

 of accentuated flowage bear a reasonable amount of tools they may do unusual 

 erosive work. 



Direction of Ice Motion 



Data on the direction of ice motion in this region was collected on Marble- 

 head, on Kelleys island, and on Point Pelee island, about 10 miles north of 

 Kelleys island. For part of this information I am indebted to Mr C. R. 

 Stauffer. 



In the first area the irregular channels shown in figure 2, plate 109, read 

 south 76 degrees west, conforming with the general direction of the major 

 joints. 



On Kelleys island, at the North quarry, the grooves read south 73 degrees 

 west. At the recently stripped area south of this quarry two sets of striae are 

 conspicuous, the older varying from 1 degree to 8 degrees south of west, with 

 very many east-west readings, while the latest ice movement varies from 23 

 degrees north to 23 degrees south of west ; thus there is some divergence in 

 both the older and more recent direction of ice motion. 7 At the West quarry 

 the general direction of stria? is south 75 degrees west. A surface between the 



7 P. Leverett gives readings of other striated areas in this part of Ohio. Monograph 

 XLI, U. S. Geological Survey, 1902, pp. 423-424. 



