GEOLOGICAL AND HISTORICAL ASPECTS OF CLIMATIC CHANGE 
equatorial oceans. The result was a permanently large 
temperature difference between the seas north and 
south of Gondwanaland, giving rise to a permanent 
wind from south to north across the continent. The 
latter consisted of a lofty plateau or series of high 
mountain ranges extending well above the snow line. 
For most of its track the temperature of the air must 
have been below its dew point, causing an almost 
permanent cloud cover. In such circumstances glaciers 
could descend the mountain slopes and coalesce to 
form continental ice sheets, protected by the cloud 
cover from the tropical sun. On the other hand, the 
warm seas to the north encouraged the growth of 
rich vegetation. It is probable that the coal forests 
preceded the maximum glaciation, because when the 
1015 
The Late Palaeozoic Ice Age, like the Quaternary, 
was not continuous, but was broken up into glacial 
and interglacial periods. This suggests that either geo- 
graphic conditions were not continuously favourable 
for glaciation or more probably that some other factor 
was superposed. This point is discussed on p. 1016. 
Humphreys’ Theory of Volcanic Dust. In 1913 Humph- 
reys [14] showed that the presence of volcanic dust 
in the atmosphere lowered the mean temperature of 
the surface. The dust particles are large enough to 
scatter solar radiation, returning part of it to space 
unaltered, but too small to have any appreciable effect 
on long-wave terrestrial radiation. He calculated that 
during the period of instrumental observations explo- 
sive volcanic eruptions have lowered the mean tempera- 
ORT! 
LAA, qi 
ATLANTIS# 
s| ICE SHEETS Bea 
« MOUNTAINS == 
WARN 
CURRENTS 
L_@@LD -===e: 
CURRENTS 
Fre. 3.—Geography, ocean currents, and ice sheets of the Late Palaeozoic. 
ice sheets reached the sea the melting ice lowered the 
temperature of the ocean waters by almost 10F and 
spread the local cooling over the whole world. This 
fits in with the impoverishment of the fauna and 
flora and the period of drought which set in during 
the Permian. 
The warmest conditions apparently occurred in 
Europe and western Asia, where the influence of the 
Tethys Sea was greatest, and here there is no good 
evidence of even small glaciers. In North America, 
where there was some local cooling due to a return 
circulation from north to south, there were large moun- 
tain glaciers, represented especially by the famous 
Squantum Tillite, but as the Arctic Ocean was appar- 
ently ice-free, the climate of the lowlands was favour- 
able enough for coal forests. 
ture of the earth by about 1F. At present the amount 
of vulcanicity is small compared with some of the 
geological periods, and it is possible that voleanic dust 
has contributed towards the decrease of temperature 
in glacial periods. On the other hand, the complete 
absence of volcanoes could not alone have raised the 
temperature by more than 1/—a negligible amount. 
Vulcanicity can never have been more than a contribu- 
tory cause; an attempt to estimate its effect quantita- 
tively is described on p. 1016. 
Variations of Carbon Dioxide. Carbon dioxide ab- 
sorbs long-wave radiation and so helps to maintain 
the temperature of the earth’s surface above that at 
which it would otherwise be in equilibrium with solar 
radiation. The amount of CO, in the atmosphere must 
have varied greatly during geological time, being de- 
