304 G. K. GILBERT — CRESCENTIC GOUGES ON GLACIATED SURFACES 



The features called crescentic cracks are vertical fractures of the rock 

 without the removal of fragments. They are usually curved in plan, with 

 the concavity turned forward. Their orientation thus relates them to the 

 chatter-marks, but they are independent of grooves. 



An approximate understanding of the glacial chatter-mark is easily 

 reached, because the phenomenon is intimately related to the chatter- 

 mark of the machinist, from which it is named. The plowing of a 

 groove in a brittle substance is not a continuous process, but is accom- 

 plished by making a series of fractures, each one of which separates a 

 fragment of the substance. Each fracture is preceded by a condition of 

 strain and stress, and these are relieved by the fracture. The resistance 

 to the grooving tool is thus essentially rhythmic, and if the tool is slender, 

 or is not firmly supported, a vibratory motion is set up (with chattering 

 sound) and the groove becomes a succession of deep scars. When the 

 grooving tool is a hard boulder held in a slow-moving body of ice, and 

 the thing grooved is a brittle rock, the remaining condition for rhythmic 

 action is probably found in the elasticity of ice and rock, which permits 

 the development of strain and stress before each fracture. 



The crescentic crack, being vertical, is presumably a result of tensile 

 stress parallel to the rock face. As the glacier moves forward it tends, 

 through friction, to carry the bed-rock along with it. If the friction on 

 some spot is greater than on the surrounding area, the rock just beneath 

 that spot is moved forward in relation to the surrounding rock through a 

 minute but finite space. This relative movement involves compression 

 about the downstream side of the affected rock and tension about its up- 

 stream side, the magnitude of the stresses depending on the differential 

 friction, and rupture ensuing when the tensile stress exceeds the strength 

 of the rock. Exceptional friction may be given by the passage in the ice 

 base of some substance which has a high coefficient of friction in relation 

 to the bed-rock ; for example, if the glacier base contains a pocket of sand 

 surrounded by clear ice, the coefficient of friction between the sand and 

 the bed-rock will probably be much higher than between the ice and the 

 bed-rock. 



The crescentic gouge is less easy to understand, and it is the purpose 

 of this communication to put forward a hypothetic explanation. 



Crescentic gouges have been observed in granite and other massive 

 plutonic rocks, in sandstone, and in limestone. My own observations 

 have been made chiefly in the granite district of the High Sierra, where 

 opportunities for the study of glacial sculpture are exceptionally good. 

 In some localities the gouges are abundant, and in most districts where 

 glacial polish and-striation are preserved they can be found by a few 



