KARA-SOU, A SOURCE OF THE EUPHRATES. 373 
that is, with a greyish tint, fantastically marked with red, blue, 
or yellow, and bound with red. 
Our course was N. 70° W., the road level and excellent, on 
a gradual descent into the middle of the valley. It is fruitful, 
well-watered, and at the greatest width stretches about ten 
miles. The mountains to the north were not very lofty, but yet 
formed a grateful shelter to the many villages at their base. 
Having marched two miles, we passed the little hamlet of Khan ; 
a couple more brought us to that of Guz ; and shortly afterwards 
we came to a rapid and pretty stream, which accompanied us to 
the village of Elija ; a place called two agatches from Arzeroom. 
At a short distance from the village we descried the Kara-sou, 
which we were approaching, flowing past the mountains to the 
north. This is one of the western sources of the Euphrates, 
and even so near its fountain-head, occupies a large channel; a 
bridge across it, of six arches, being then very discernible. This 
source lies about thirty miles from where I now saw its stream, 
in that branch of the Tchiller mountains called the range of 
Auggi-Dahgler: the country about them bears the name of 
Keldir. Close to Elija a hot spring was pointed out to me, 
where I saw about fifteen or twenty boys, like as many Cupids, 
playing about under three or four buffaloes, who appeared to 
enjoy the bath as much as themselves. The valley narrowed 
from this place very considerably, and a little farther to the 
westward, almost closes to the width of the Kara-sou, which 
flows through it. We had continued our way along its left 
bank for about three miles, when we passed two villages on its 
opposite or northern shore; they were called Guy-opi and Tip- 
kere. Another mile, and we reached that of Anguerore. For 
a long time we had left all rural beauty behind us: not a tree 
4 R 
VOL. 11. 
