Part I of this paper was reviewed in the Journal of Glaciology, Vol. 1, No. 2, p. 79. Part 2 was to have dealt with the upper air data, but Dr. Matthes died before he could write it. Actually it is doubtful whether sufficient material yet exists for such a study to be fruitful. But he left a number of further notes on surface weather in Greenland, and the junior author has collected and arranged these and added some later material. This forms a valuable summary of the weather prevailing over an inland ice cap, and it leaves no doubt that the persistent anticyclonic conditions, and the maintenance of the ice sheet entirely by low-level condensation, which constitute Hobbs’s theory, are a great exaggeration if not an illusion. Quiet weather prevails for less than half the time, and the main element in maintaining the ice sheet is clearly seen to be the snow brought by barometric depressions.

This must not be taken as implying that the great cold mass of Greenland does not influence the general weather situation—far from it. The normal sequence of events seems to be that depressions approaching Greenland, generally from the south-west, are diverted, and their centres travel northwards along either the west or the east coast. But a vigorous depression has a diameter of well over a thousand miles, and its wind system extends far across the inland ice, giving south-westerly winds on the west or easterly winds on the east. The easterly depressions are the more vigorous and hence the greatest accumulation of snow lies well to the east. These winds are frequently strong enough to sweep away the surface layer of cold air and bring a complete reversal of wind direction and a great rise of temperature, especially in winter. Weaker storms are not able to sweep away the surface air, and in these the cyclonic circulation overrides the cold air and snow or freezing rain falls through it. This condition, occasionally though rarely met with in England, where it brings the “glazed frost,” is probably quite common in Greenland. It is fair to say that the centres of most storms avoid the inland ice, but very occasionally an exceptionally intense depression is deep enough to ignore the ten thousand foot plateau and pass directly across Greenland from west to east—instances of this are quoted. The controversy exists because observations are so sparse and the width of the country is near the critical limit between non-interference with the tracks of depressions and complete blocking.

One other interesting point emerges. Greenland is not a simple dome, but has at least three centres of elevation, each of which has its own system of outflowing surface winds. This topography is not possible by Hobbs’s theory, according to which it would be rapidly smoothed out, but it is to be expected if the main source of supply is cyclonic snowfall, for then each separate dome would tend to be maintained by the increased local snowfall to which it gave rise.