Vol. 58.} CONDITIONS OF THE PLEISTOCENE EPOCH. 4] 
cooled continents attract them during the winter, but they are not 
regularly formed over the Polar regions, where the weather is nearly 
as variable as in our latitudes, with moving cyclones and anti- 
cyclones. Thus we have no reason to believe that per- 
manent anticyclones were formed over the ice-sheets of 
the Great Ice-Age. And, in fact, as remarked above, the ice 
would have disappeared relatively soon during such a state of 
weather, as it would not then have been maintained by snowfall. 
(2) It remains to determine the influence of the ice-cap on the 
glaciated land and its neighbourhood. 
We know that the amount of rain or snowfall generally increases 
with the height of the ground above sea-level, and proportionately 
more so if the ground is covered with snow. This is verified both 
by theory and by observation. Thus, the observations recently 
made by Dr. Axel Hamberg in Lapland have given a yearly 
precipitation of 80 to 120 inches of water at a height of 6500 feet 
above sea-level, whereas the amount in the neighbouring meteoro- 
logical stations at heights of 650 to 1000 feet is only about 
15 inches. 
The formation of an ice-sheet one or more miles thick will, 
therefore, increase the snowfall enormously and tend to reinforce 
itself, so long as the temperature remains sufficiently low. More- 
over, the height of the upper surface of the ice and the thermal 
properties of snow and ice will lower the temperature considerably. 
In this way the enormous extension of the ice-sheets may be 
explained. 
As to atmospheric circulation, the effect on it during the winter 
was probably similar to that observed in a cold winter nowadays. 
Such a winter begins with frequent snow-storms caused by cyclones, 
which are deviated into a more and more southerly track as the 
winter advances, whereas an anticyclone is often formed in the 
north over the cooled area. Such an anticyclone, however, is not 
stationary, for it generally moves eastward just as the cyclones do, 
though as a rule more slowly and irregularly. Owing to the creat 
contrasts of temperature and copious condensation of water, snow- 
storms in the Ice-Age were probably much more violent and frequent 
than at the present time. Also the summer must have been 
generally cool and stormy, with frequent fogs. The present climate 
of Cape Horn or Kerguelen Island will probably give some faint 
idea of the weather prevailing in Europe and North America during 
the Great Ice-Age. 
As in our present cold and stormy winters the cyclonic tracks 
are much more southerly than in the mild winters, I conclude—in 
agreement with Mr. Harmer—that the southern part of North 
America and Europe, as well as the northernmost part of Africa, 
were much more rainy than now. The great anticyclones now 
situated between 20° and 40° lat. N. were also probably displaced 
southward, and the thermal equator of the earth, which now is 
to be found at about 10° N. in the longitudes of Europe and 
