uml M.ARK CllAl)^ AND 'ROUND VADDKV 195 
side aiul the side of the defile on the other. Throui^h this 
we earrieti our ]\ieks ; through this in the other direeticii 
the seals iiuist have lahorious]\' era\vK'>.l io die tar inlaiul. 
We eould not see the sea, hut h)uiKl the elefile oeeupied 
hv a fro'/.en lake a mile long. There were dr)' gia\'elly 
banks around this hd-;e and here we pitelu'd tlie tent. We 
had brought no lloor-elotli, but after the wet and ie)' floor 
of the ' Aleove ' eanii^— wliere W right had slept in a pool 
of water three inelies deep — we louiul the warm gravel 
mc^st eomfortable. We liad t)ur irugal meal, washed down 
by cold water from the la ke ad jat en t . 'V\\r kit I i-r was 
distinctly medicinal and had no outlet, so ignoring climatic 
differences we unanimously christened it Lake Chad. 
I was quite worried to know what hail become ui (lie 
broad stony valley which ShackletoiTs men had seen Irom 
the coast in 1908, and wondered if we were side-tracked in 
some tributary valley. So after dinner Mvans- who 
was always eager lor extra \v<M'k — acconi|>aiiied me to llie 
top of the ridge immediately soutli of the tenl. it was a 
stifT ascent of 1600 feet to a ilat bare expanse obviously 
planed by bygone glaciers. 'I'o my surprise I saw tJuit a 
much- larger rounded valley lay immediately north of lids 
ridge, but this ^ Round ' Valley, unlike the di-lile, tliti uot 
connect with the Taylor glacier. To the east some ten 
miles beyond a broad debris-strewn plain lay the sea, and 
in the far distance we could see the glaciers on the slopes 
of Erebus and the pyramid of Heaidort Island. 
Early on the 5th I'! vans and I started lor 1 he coast, while 
Debenham and Wright invest igati-d the rocks anti glaciers 
near the defile. We proceeded S.l^., j^assing sevrrrd 
O 2 
