i iq6 Ward. Some Plant Formations from the 
I have remarked this scorching blast in all the valleys alluded to, par- 
ticularly on the three biggest rivers, the Kin-sha, Mekong, and Salween, 
strongest of all perhaps on the Mekong, where it is more confined than any- 
where else. Throughout the day it increases in ferocity, till about four 
o’clock in the afternoon it reaches a climax, then gradually calms down 
towards nightfall ; and though rather of the nature of a sirocco, it is always 
extremely welcome after the intense heat of the day. 
This wind is of course caused by the cold air sweeping down from the 
surrounding peaks to fill the partial vacuum caused by the great heating of 
the shut-in valleys during the day, and it greatly intensifies such desert con- 
ditions as are already imposed owing to the proximity of the mountains, 
which act the part of rain-screens. It is indeed a peculiarly desiccating 
wind, since it has already rid itself of its moisture, and throughout the 
summer it comes up the valleys like the breath from a hot furnace, till the 
withered vegetation on the cliffs seems gasping for water. Thus the severity 
of the conditions increases automatically as one travels northwards, for the 
intensity of the wind itself increases with the dryness produced by its action. 
On the Salween and Mekong rivers an abrupt transition from a rainy 
region to one of extreme aridity corresponds to a general elevation above 
the snow-line of the dividing watersheds ; the only break in the dreary 
scene is where a mountain torrent has flung out an alluvial cone which, 
terraced, and green with cultivation, forms a little oasis in the wilderness. 
Travelling up the Mekong valley I have frequently noticed a long 
ribbon of blue sky faithfully following the valley, while the clouds, though 
perhaps clear of the mountain peaks, are massed in great puffs of cumulus 
on either hand, showing plainly enough that columns of hot air are ascending 
from the valley. 
Once in the Mekong valley I watched the clouds in the south, from 
which snow and rain were actually falling at the time, trying to force their 
way up the valley into the arid region beyond, where blue sky prevailed ; 
however, they failed signally to cross the dividing line, and it was as though 
some invisible barrier was forcibly holding them back, the truth being that 
the hot air rising from the bare rocks prevented any cloud existing as such. 
Again, in the Salween valley I have watched the clouds gathering 
black in the west over the Irrawaddy, watched them sweep slowly over the 
valley, and reinforced still further, burst again over the Salween-Mekong 
divide to the east, a few scattered drops hissing on the scorched rocks of the 
Salween valley alone announcing their passage of that river. 
It is abundantly clear then that the arid valleys of Western China owe 
their inception to just those causes which, on a much larger scale, give rise to 
continental deserts. 
The scarcity of rain, which probably does not exceed five inches in 
a year, would of itself be a serious hindrance to plant life, and this is 
