It has for some time been obvious that far-reaching results could be obtained by penetrating right through a glacier to its bed. The expense however has always been prohibitive. Recently however Dr. H. Carol, of Zürich, had the ingenious but simple idea of going down through a glacier by means of a glacier pot-hole and has described his experiences in Les Alpes, 1945 (pp. 180–84).

The prevention of snow accumulation on the tracks of railways in Great Britain. Various methods of removal, once lines have been blocked, are discussed in the light of the comparatively small snowfalls in the British Isles.

The typical U-shaped cross-section of glaciated valleys is well known but little understood. If a climatic change results in a glacier occupying a deep river valley, and some erosion is assumed to occur wherever the ice makes contact with the bedrock, then a partial change from the “V” to “U” section is to be expected. A more potent agency in such a change is probably the sapping of the valley sides by the alternate freezing and thawing of melt-water flowing down to, and under, the glacier. This has been briefly suggested by de Martonne and worked out somewhat more fully in the case of cirques1

The author describes the ice in Crown Prince Gustav Channel on the east coast of Graham Land, Antarctica, based mainly on observations made by members of the Falkland Islands Dependencies Survey in 1945–48. He considers that the ice in the northern half of the Channel is landfast sea ice which may persist for more than one season, and, that in the southern half it is shelf ice of the same origin as the Larsen Shelf Ice. He concludes that this shelf ice in the southern part of the Channel is part of the Larsen Shelf Ice.