130 MR. W. B. HEMSLEY ON THE 
the first place we give a free translation of Maximowicz’s account, 
relating chiefly to the Eastern and Northern parts. 
The winter in Tibet, observed three times in different years and 
months, and therefore the best known season, is cold, practically 
snowless, and characterized by a very dry, tempestuous atmo- 
sphere. The mean temperature of December and January is 
between +6°6 and +2°3 Fahr., but the nocturnal temperature 
in October falls to —9°4, in November and December to ~—22°0. 
In January the lowest temperature was observed, —28°3, and 
soon afterwards the mercury froze; at 1 p.m. it varied from —0°'4 
to +11%6, and even higher; on sunny days occasionally +39°0 
was reached, and once even +46°0 in consequence of the presence 
in the air of a copious dust heated by the sun’s rays. During the 
day the wind, often violent, always blew from the west. Thus, 
both in October and November there were ten stormy days, in 
December fourteen, in January eighteen. Swamps and rivulets 
were frozen in Octeber; the larger rivers in November; the 
lakes are so salt that they never become ice-bound. Later on, 
on account of the extreme dryness of the air, the ice on the 
swamps evaporates, the greater part of it disappears, and they 
are dried up; similarly the rivers become waterless. Herbs 
also become so dried up that they crumble when touched and 
are ground to powder beneath the feet. The Yak and other wild 
animals were observed to lick up their food rather than detach 
it by their teeth. 
Snowstorms are very frequent, and always tempestuous on 
account of the west wind. Nevertheless, snow falls sparingly 
and in very small quantities and usually disappears on the 
following day, dissipated by the wind-blast and the heat of the 
ground. Hence level places and the flanks of the mountains 
looking south were destitute of snow, and only on the northern 
slopes did the snow remain for any length of time. The 
inhabitants state that in some years the snowfalls are heavy, 
but they do not remain long, or all the animals would perish. 
The limit of perpetual snow in Tibet seems to correspond to 
the lowest limit of the glaciers, and lies at 16,500-17,000 ft. ; 
it is evidently much higher than in Amdo around the upper 
Hoangho and on the south side of Nan Schan, where perpetual 
snow begins at 15,700 ft., whilst on the northern slope of the 
same range it descends to 14,700 ft. 
Spring in Tibet is cold and tempestuous, and disturbed by 
frequent snowstorms. 
