126 WITH SCOT J: THE SILVER LINING 



direction, and we soon saw we should be justified in devoting 

 some hours to its examination ; while Debenham came along 

 later and collected the varied rocks in the vicinity. As we 

 approached the northern slopes the surface of the Ferrar 

 Glacier altered in character, and gave place to large lake-like 

 areas of ice, which exhibited most beautiful figures on close 

 examination. In the upper layers of the ice were included 

 radiating designs which resembled a miniature Hampton Court 

 maze in porcelain embedded in glass. These intricate 

 patterns — which are characteristic of glacier ice — I termed 

 " Arabesques." They are due, I imagine, to some variation 

 in the solidifying water, perhaps owing to air being squeezed 

 into the latest ice formed — or again show where stones have 

 sunk deep into the glacier. 



Somewhat nearer the shore there were more unpleasant 

 surfaces met with — large dome-covered ponds into which 

 we fell at frequent intervals. We decided that a tramp over 

 the Crystal Palace would give rise to the same sensations. 

 Bounding the glacier and separated from the debris slopes by 

 a wide stream was an avenue or colonnade of gigantic ice 

 pinnacles thirty feet high. These were traversed by narrow 

 crevasses, down one of which I had to climb to rescue an ice 

 axe. The sun glistening on the icy minarets and beautiful 

 icicles made a most impressive sight. This ice ridge is due to 

 pressure from the glacier piling the ice against the cliff higher 

 up. This crenellated selvage to the more level central level 

 centre of the glacier — moves to the sea with the main body, 

 and so preserves its lateral position, though no pressure can 

 exist where we saw it — for it is many yards from the rock. 



Between the pinnacle ridge and the slopes was the water- 

 bearing channel which invariably accompanies a large glacier in 

 these regions. This physiographic feature is one of the most 

 interesting and most important in connection with the char- 

 acteristic topography of Antarctic valleys. The small valley 

 bounded by ice on one side and rock on the other is 

 conveniently termed the Lateral moat. Hereabouts it was 

 rather complex, but further up the main glaciers its valley 

 occupied merely a simple V. After crossing the pinnacles we 

 had to negotiate a stream in which the water lay in pools 

 several feet deep — though its flow was comparatively small. 

 Then over a silt moraine and so across another slight 



