GEOLOGICAL OBSERVATIONS 



banks of hard rock, the ice, expanding as it formed, was 

 forced to curve itself upwards in order to make room for 

 itself, as it could not burst the sides of the rock basin. 

 Thus its surface was frequently convex upwards. In this 

 process of upward bulging of the ice towards the lake's 

 centre each successive fresh layer of ice as it froze beneath 

 the older and overlying layers, would buckle them and 

 bend them. The latter would at last crack open, and so 

 in the final stage of freezing of a small shallow lake, by 

 the time that the whole of the water had frozen from top 

 to bottom, the basin would be occupied by a biconvex 

 cracked lens of ice, the cracks being widest at the top and 

 tapering away to nothing below. 



A curious feature which we observed in the lake ice 

 was the presence of what we termed " snow tabloids." 

 We found that in some cases these were merely empty 

 bubble-like spaces in the ice filled with air. In other 

 cases, however, where the bubbles were larger, 3 to 6 in. 

 in width, they were occupied by snow. In some cases 

 patches of thin rippled snow were inter-stratified in this 

 lake ice. 



Most of this fresh-water lake ice exhibited at its sur- 

 face a very beautiful structure, which we termed coralloidal 

 structure. 



The mode of origin of this curious structure will be 

 discussed in the Scientific Memoirs of this Expedition. 



(b) Action of Wind in Relation to Antarctic 

 Glacial Phenomena 



An explanation has already been given of how vast 

 quantities of finely divided rock material, chiefly in the 

 form of sand, are constantly being blown on to the sea ice 

 by the wind. For some distance seawards from the shore 

 such wind-blown material must form an appreciable 



Vol. II.— 20 305 



