376 TJie American Geologist. June, 189^ 
north temperate and frigid regions of our continent and Eu- 
rope which are drift-co-vered. 
An observatory for weather observations was estabHshed 
on the sunnnit of Ben Nevis in 1883, and hourly records dur- 
ing both day and night are taken there for comparison with a 
meteorological station close to the sea level at Fort William, 
only five miles distant to the west. The mean annual precipi- 
tation of rain and snow (the latter being reduced to its equiva- 
lent of rain) recorded during the first ten years at the Ben 
Nevis observatory was 142.34 inches, being the greatest 
known at any locality in Scotland; while for the same period 
at Fort William it was 75.79 inches. In comparing this rain- 
fall, as Jamieson has done,^' with that of other parts of Scot- 
land, we find its eastern half to receive much less rainfall 
yearly, decreasing eastward from 40 to 25 inches; but in west- 
ern Scotland, from Gare loch north-northwest to the Isle of 
Skye, a wide belt of the Highlands has 80 inches and upward 
of mean yearly rainfall. 
Upon this tract of very abundant rainfall and snowfall, 
probably the earliest part of the Scottish ice-sheet in the 
Glacial period was amassed. It gradually filled the valleys 
and glens, and finally overtopped the mountains, excepting 
apparently a few of the highest summits. Meanwhile the ice 
accumulation extended far outward over the whole country, 
into conflvience with the ice-sheet of Ireland, the ice-fields 
of the Southern Uplands, of the mountains in the English 
Lake District, and of Wales; and, on the east, it became con- 
fluent with the great ice-sheet deploying from the mountain- 
ous Scandinavian plateau on the wide low plain that is now 
covered by the shallow North Sea. 
When the Glacial period ended, the confluent European 
ice-sheet in its departure doubtless became again divided, as 
during the early states of growth, into separate parts flowing 
outward from the great central tracts of maximum snowfall. 
The courses of glacial striae, and of the dispersal of drift, give 
clear evidence of the areas which thus preserved the latest 
remnants of the formerly confluent icefields. In the British 
Isles, probably the last remnant to melt away was in western 
Scotland, lingering somewhat longer, on account of the altl- 
*Quart. Jour. Geol. Soc, XLVIII (1892), 5-28. 
