Data are supplied to show that the “blue” surface ice zone encountered in the Skelton Glacier, Antarctica, results from horizontal compressive forces. Examples of the densification of surface snows of the Ross Ice Shelf by this means are also given. Such density changes should be expected in any area where either of the principal horizontal strain-rates is negative.

Small glaciers lying in funnel-shaped hollows on steep slopes have been variously described as wall-sided glaciers, cascade glaciers, cliff glaciers and niche glaciers. This type of glacier and the landforms associated with them in Bünsow Land, central Vestspitsbergen are described and their possible mode of origin is discussed. The theory that they were formed by an overspill of plateau ice and that they are thereby genetically related to hanging glaciers is rejected. It is suggested that they are formed by an accumulation of snow in an initially water-worn gully and that as the small glacier develops the original form of the gully is modified by nivation to a rounded, funnel-shaped hollow. Niche glaciers may thus be considered as glacier types genetically related to cirque glaciers, forming an early stage in corrie development.

The rate of movement of the Ross Ice Shelf has been determined at its terminal face to the south-east of Scott Base. An average rate of movement of 23 cm./day in a direction 270° True was determined for the period March 1957 to October 1958. The terminating face of the ice shelf in McMurdo Sound is only about 3 metres thick compared with a thickness of several hundred metres in the Ross Sea. Altimeter heights are used to demonstrate a thinning of the shelf which is attributed to melting from below.

Curious interlocking thrust structures, described in the Journal of Glaciology, Vol. 3, No. 23, 1958, p. 173–75, are shown to be a common and typical deformation feature of young sea ice, and a suggestion is put forward as to their formation.

The Gorge du Guil (Hautes-Alpes) might be regarded as a typical example of a post-glacial gorge. In fact, this is not the case. Systematic and very detailed observations based on maps of the scale of 1 : 5,000 have revealed that the greatest part of the excavation of the defiles, cut across the bands of massive rocks, was accomplished by subglacial melt waters. Remnants of glacial deposits containing well-rounded pebbles, but whose finer constituents have been largely removed, occur a few metres above the present valley floor in each of these defiles. The down-cutting was very powerful, being facilitated by the crevassing of the ice, which was at the most 300 m. thick, over the ridges and in the constricted sections of the valley, and by the load carried in the englacial melt water. The sheltered and sunny climate, and the situation of the area below the limit of permanent snow also favoured erosion by melt water. By comparison, post-glacial down-cutting has been slight, not exceeding some ten metres in the massive strata, and appears to have been far less rapid than the subglacial erosion.

Separated etch pits can be produced by a simple operation using a plastic replica film. Combining this etching technique with Schaefer’s replica method, it is possible to record both the size distribution and the orientation of crystal axes of individual grains of ice.

The velocity of flow of a glacier was measured at two places where the ice margin had no moraine debris between clear ice and unfragmented rock. In one of these places the point at which the measurement was made was only about 1 m. from the solid rock. It was found that at both sites the ice was flowing past the rock at a velocity of the same order of magnitude as the velocity of flow in the centre of the glacier in this region. Although conditions at the side of a glacier are not the same as at the bed, this large amount of slip is evidence for a high rate of slip all along the ice-rock interface, and this is in accordance with current ideas of the flow of ice below ice falls.

A microscopic method for the determination of particle-size distribution of pulverized snow was worked out. The method gives satisfactory distribution curves, presenting the number of particles as a function of their cross-sectional “areas”. The measurements were made by means of a filar micrometer eyepiece, the snow particles being placed on a ruled glass slide, which was submerged in silicone oil to prevent evaporation. The time for the determination of a distribution can be appreciably shortened by estimating the size of the particles instead of measuring them, though the accuracy is not so high in this case.

A model is proposed to explain the sliding of any glacier whose bottom surface is at the pressure melting point. Two mechanisms are considered. One is pressure melting and the other is creep rate enhancement through stress concentrations. Neither of the mechanisms operating alone is sufficient to explain sliding. If both mechanisms operate together appreciable sliding can occur.

During the summer of 1957 a party visited the Salmon Glacier in British Columbia. A meteorological station was established in the accumulation area at an altitude of 1700 m. From 12 June to 16 August pressure, temperature and humidity were recorded continuously and the daily totals of precipitation, ablation and percolation were obtained. Periodic measurements of snow density were made; on the first occasion to a depth of 4.7 m. and later near the surface only. Occasional observations of snow temperatures to a depth of 12 m. were also made. Towards the end of the season the heat balance of the snow surface was examined in detail. The various contributions to the melting are evaluated for several short periods and their relative magnitudes compared.

The “earthquake avalanche” theory was invoked by Tarr and Martin in 1914 to explain anomalous turn-of-the-century glacier advances in Yakutat Bay, Alaska. The concept was given credence by the fact that this bay was the epicenter of a series of severe earth tremors in the autumn of 1899.

In considering alternative explanations, Tarr and Martin dismissed a record of excessive precipitation along this coast in the late 1870’s and 1880’s in the belief that the measurements were in error due to a “peculiarity in the method of measurement”. The disputed data are now known to be consistent with observed and inferred regional climatological trends in the latter half of the 19th century. The case for the climatological control of glacier fluctuations in the St. Elias District is therefore reassessed. Reference is made to new and extensive evidence from other districts in southern Alaska where the earthquake influence was negligible. It is concluded that the fluctuation pattern in the Yakutat area was not unique and that the diastrophism to which these particular resurgences have been attributed was actually but a supplemental and generally minor factor in a widespread set of glacial advances initiated by climatological causes.

The theoretical implications of changes in load conditioned by climatological parameters over the broad névés of the St. Elias Mountains are briefly discussed with respect to orogenic activity in this tectonically sensitive region.

Observations of the snow depth at 21 sites at Resolute were made twice weekly during the winter of 1957–58. As a result of these observations, and of other observations on snow made for the National Research Council, it is shown that the snow depth and the water content of the snow did not continue to increase during the winter as the snow fell. Rather the strong winds eroded the snow surface and the increase in depth was irregular and relatively slow. Furthermore, the observations on the density of the snow cover lead to the conclusion that attempts to measure the density in similar regions with an accuracy greater than ± 0.05 g. cm.−3 are not warranted.

Winter conditions on the Mackenzie River and Delta in North-west Canada are discussed. Daily observations during the break-up period of the Mackenzie River for 1954 are recorded, as well as a general record of the break-up of ice in the delta and detailed observations of the break-up in one of the main channels.

The general characteristics of the coastal regions between long. 45° E. and long. 80° E. are described. The features and conditions are similar to those found along the coasts of the Australian sector further to the east. Measurements of accumulation, snow transport, ice flow and ablation are described and results are given.

Accumulation, measured from stakes and pits, is 1.0 × 1014 gm./yr. in a 1 km. wide strip running 850 km. inland from the coast of MacRobertson Land. The methods of gauging drifting snow and extrapolating the results are given and a meridional mass transport of 0.16 × 1014 gm./km. yr. is deduced. Iceberg calving rates given in a previous paper are again quoted, although they are now felt to be too low. Net ablation is 0.053 × 1014 gm./km. yr. and additional evaporation above the firn limit accounts for 0.045 x 1014 gm./km. yr. The estimates are compared with old and new observations from other parts of Antarctica and the problem of bottom melting beneath ice shelves is discussed.

A distinction is made between the meteorological water budget for Antarctica and the glaciological mass balance for the ice sheet. Mass budgets for the sector between long. 45° E. and long. 80° E. and for the whole of Antarctica are drawn up. In each case a surplus of accumulation over losses appears, but it is felt that the data are insufficient to claim that the ice sheet is growing at the present time. The drift snow and ablation losses are added to the net accumulation to give a figure of 14 cm. of water as the mean annual precipitation over Antarctica, a value lying between the estimates of Meinardus and Kosack.

Melting at the bottom of floating ice shelves may represent an important item in the mass economy of ice sheets. Some earlier studies of the behaviour of fresh-water ice in sea-water at a temperature below 0° C. are quoted.

A number of measurements of ice flow in the coastal regions of Antarctica are given. Observations show that the general outward movement of the continental ice, termed “sheet flow”, is locally accelerated where “ice streams” are formed. Estimates indicate that ice streams, which only occupy a small fraction of the total length of coast, are responsible for the removal of more ice from the continent than the “sheet flow” over the remaining length of coast. Further estimates suggest that the great bulk of Antarctic icebergs are produced by ice shelves, but that data on ice shelf movement are inadequate at the time of writing.

Eight different ice budgets for the Antarctic Ice Sheet are examined. Five of these budgets call for rates of increase of the ice which are in surprisingly good agreement considering the wide dispersion of individual components of the budgets. The observed rise in sea-level of the world’s oceans would appear to contradict the removal of water required to nourish the Antarctic Ice Sheet at the claimed rates. The thermal expansion of oceanic columns, caused by “climatic” warming of the oceans, has been invoked to resolve this contradiction but it appears that this is not large enough judging from the 30 yr. difference in water temperatures measured by the Meteor and the Crawford in two profiles across the tropical Atlantic Ocean.