194 GENERAL GEOLOGY 



when these did occur, they gave one some idea of the amount of rock flaked off 

 by the frost-action. Mount Cis, the small parasitic cone on the lower slopes of 

 Erebus, was flanked on one side by such a scree, and the essentially angular nature 

 of the fragments, together with the fact that they were all of the one very 

 local type of rock, precluded the possibility of the debris being of glacial origin. 



The highly jointed nature of the lavas at Cape Royds greatly accelerates and 

 facilitates this action, and Flagstafi" Point in particular is seamed by huge cracks, 

 and it is always possible, and even probable, that at no very distant date the western 

 half of this cliff may fall into the sea bodily, for there is one crack extending nearly 

 north and south which is in places 3 or 4 inches wide, and in one place it was 

 impossible to touch the bottom of the open part with the end of a 10 feet bamboo pole. 



The reason for the disappearance of glacial grooves and strise at Cape Royds, 

 and indeed from the surfaces of all the land-moraines we visited in Antarctica, 

 is jjrobably largely explained by this frost-weathering. Almost all of the boulders 

 which have their striated surface exposed must have this surface flaked off, owing 

 to differential expansion and conti-action resulting in a type of concentric weather- 

 ing. Many of the holocrystalline boulders at Cape Royds were surrounded with 

 these arc-shaped flakes. Thus the absence of striation on the surfaces of the 

 blocks of the moraines at Cape Royds is amply accounted for, and it is only in 

 places from which the ice has recently retreated, or on the under surfaces of partly 

 embedded blocks, that we are likely to find a large proportion of striated material. 

 The survival of striated roches moutonnees, and of striated blocks, in temperate 

 and tropical regions which have been strongly glaciated in former eras, may 

 perhaps be in part explained by the fact that, once the period of maximum glaciation 

 is over, the region from which the ice has retreated is only subjected to the rapid 

 disintegrating processes of frost-weathering for a comparatively short time. 



Another fact which may account for a large number of blocks retaining their 

 striated surface in old glacial deposits is that they are often enclosed in stiff' 

 boulder clay, which would exercise a preservative influence on the rocks. It follows 

 from the facts observed, in the region of the Antarctic explored by ourselves, that 

 the only three places where it is likely that striated boulders and rock surfaces 

 should appear in and under the deposits of the present geological period are, (i) 

 on the bed of the Ross Sea, where the material is protected from the action of the 

 frost by the water ; (ii) in the deposits at the mouths of the larger glaciers, where 

 the tine sediment brought down by the thaw on the glacier accumulates and pro- 

 tects the boulders ; and (iii) in the upper portions of the glacial valleys of the main- 

 land, which are not likely to be exposed to sub-aerial denudation until the very end 

 of the Polar glacial period. 



Impressive evidences of the effect of frost-weathering are to be seen in the Ferrar 

 Glacier Valley. 



Against all the cliffs bordering the glacier, except those from which the debris 



