Changes in Geography and Climate. 4! 
ought to give a fairly accurate idea as to the temperature 
of the water. No doubt a large iceberg may travel a 
long distance through comparatively warm water before it 
entirely melts away; but shore-ice, such as forms every 
winter in the Arctic Regions, once fringed our south coast, 
and beset the shores of Brittany and of the Channel 
Islands. When, in the spring, the ice became detached, 
it transported its burden of included rocks hither and 
thither, even across the Channel. We thus find on Selsea 
Bill erratics weighing several tons, but undoubtedly derived 
from Bognor or from the Isle of Wight. Others, equally 
large, have come from the Channel Islands and the coast of 
Brittany ; one block of granite is like that of Cornwall. 
The transportation of large erratic blocks for distances of 
at least a hundred miles, shows that the temperature of the 
water in the spring, though sufficiently high to dislodge 
the ice, was yet too low to melt it rapidly. Even witha 
strong wind a flat mass of shore-ice would take several 
days to cross the Channel. In order to compare this ice- 
laden English Channel with existing seas, it is necessary 
to travel northward, till we cross the isotherm of 32° F,, 
and are near the Arctic Circle. 
Thus far we have dealt solely with the temperature of 
the sea. We will now turn to the evidence as to the 
temperature of the air during the same period in the South 
of England ; and for this we can employ both physical and 
biological data. The country north of the Thames and 
Severn, buried under ice, must have been bordered by a 
wide strip of barren land, with dwarf birch and willow, but 
without trees. In this belt flourished also a mammalian 
fauna like that now inhabiting similar belts in the Arctic 
Regions, for in the area lying between the ice-sheet and 
the ice-cold English Channel it would be impossible to 
have a mean temperature much above the freezing point. 
