I9II] A SNOWLESS VALLEY 191 
We pushed on till 9 p.m., descending slightly as we 
proceeded to the north, and camped on the glacier filling the 
upper end of the Dry Valley. The exploration of this 
glacier — which Scott had rapidly traversed in 1903 — was 
the work before us during the next fortnight. Captain 
Scott has honoured me by giving it the name of Taylor 
Glacier. 
I kept too near to the Kukri Hills on descending into 
the Taylor Glacier and we struck an extremely steep 
slippery surface consisting of clear ice cut into rounded 
hollows a foot across. This characteristic surface — like 
giant thumb-marks in a piece of putty — was full of small 
crevasses, and here the sledge repeatedly ' took charge.' 
We rolled about all over the place, and someone remarked 
that we had all the appearance of being drunk and none of 
the pleasure of it ! 
To our surprise, after five days' pulling over heavy 
snow in the Ferrar Glacier, we found no snow in the 
parallel Taylor Valley, only about 10 miles farther north. 
After lunching among the scattered blocks of the medial 
moraine we descended about a thousand feet, the sledge 
doing its own pulling. Debenham and I went on ahead 
with slack traces, while Evans and Wright enlivened the 
valley with what they were pleased to call ^ cheerful song ' ! 
A strong keen wind was blowing up the valley, but the 
most remarkable feature of this region prevented it from 
becoming obnoxious. There was no drift-snow ! 
Imagine a valley 4 miles wide, 3000 feet deep, and 
25 miles long without a patch of snow — and this in the 
Antarctic in latitude 77^° S. By this time we could see the 
