124 G. K. GILBERT — GLACIAL SCULPTURE IN WESTERN NEW YORK 



region. When not coincident with the cliff, the boundary of the lime- 

 stone could not be mapped in detail, as it is largely covered by the drift, 

 3 -T^-,______^ but its approximate position 



was inferred from the records 

 of wells and the distribution 

 of limestone boulders in the 

 drift. 

 The first question for which 



Figure Z.— Typical Profiles of the Niagara Escarpment. aUSWCr WaS SOUght at the be- 

 The profiles are drawn with and without the Niagara ginning of the WOrk WaS the 

 limestone. 1, Niagara shale. 2, Niagara limestone. 3, radical One,whether the Niag- 

 ara escarpment belonged to 

 the preglacial topography or had been created by ice erosion, and this 

 question seems to be answered by the phenomena already described. 

 The occasional absence of the limestone from the crest of the escarpment 

 suggests that the action of the ice ma}^ have tended even to obliterate 

 the cliff; and certainly if it had created the cliff by rapid erosion of the 

 shale where unprotected by the limestone, we could not expect it to re- 

 verse the process and uncover the shale by carrying away the protecting 

 armor. Moreover, the southward slope of the plain on the back of the 

 limestone — a slope in the direction of the dip, but opposed to the general 

 descent of the region — is characteristic of subaerial and not glacial ero- 

 sion, and the failure of minor ridges of that surface to conform in trend 

 to the direction of ice motion indicates that the surface received com- 

 paratively little modification from the ice. All the more general features 

 of the limestone belt thus seem to be preglacial. 



From the Niagara river eastward to Lockport, a distance of nearh' 20 

 miles, the edge of the limestone is almost continuously exposed to view, 

 and with slight exception it coincides with the escarpment. The general 

 course is east and west, but the outline is diversified by salients and re- 

 entrants, so that the cliff may be divided into two sets of courses, the 

 one trending south of west and the other south of east. The general 

 direction of ice motion in the region was toward the southwest, and it 

 was thus diversely related to the two cliff trends. Where the cliff face 

 trends toward the southwest, or approximately in the direction of ice 

 motion, its contours are simple. Where it trends toward the southeast, 

 so as to make a wide angle with the direction of ice motion, its contours 

 are deeply inflected or serrated, the axes of serrations being parallel to 

 the glacial striae. In other words, the portions of cliff whose faces were 

 directly opposed to the ice advance are characterized b}^ furrows, and the 

 portions met by the ice at a small angle are free from furrows. The fur- 

 rows have smooth sides and bottoms, the latter rising: slowlv to the level 



