The melting of glacier ice always begins at the surface of the glacier grains. A mass of glacier ice at a temperature near its melting point is therefore an aggregate of crystals separated by a film of liquid. The electrical conductivity of ice subjected to differential melting differs considerably according to whether the fractions come from the grain surfaces or from the crystal centres, being much the higher in the former case. It is concluded therefore that the glacier grain consists of a crystal of pure ice with a surrounding film of brine, thus verifying the hypotheses of Buchanan and of Quincke on the formation of natural and artificial ice.

The growth of the grain is facilitated by the elimination of the saline skin between neighbouring crystals, the resulting coalescence having the character of autogenous welding. On the other hand, the intergranular film of liquid may act as a lubricant in the as yet inadequately explained mechanism of glacier flow.

IN 1939 it was decided to leave a book in the Ben Nevis Hut in order that visitors might record the snow conditions there. The Scottish Mountaineering Club gave the necessary permission and much help. The register was brought down early in 1946, inspected and returned to the hut for further entries.

Many authorities have demonstrated that the hardness of ice varies inversely as the temperature, which is, of course, to be expected by analogy with metals. Recently Dr. E. Blackwelder1 caused some experiments to he made at a temperature of −78.5° C., the temperature of solid carbon dioxide. At this temperature the hardness of ice was found to be approximately 6, or that of orthoclase felspar. Hitherto it had been assumed that aerial corrosion of rocks close to glaciers and snow-fields was in the main due to rock dust. It now seems likely that when air temperatures fall very low, as for instance during blizzards, ice fragments or snow could abrade certain limestones and shales and even some igneous rocks. Ice and snow may therefore play a more important part in aerial erosion than had been suspected.

This paper concludes the earlier researches of the author and his colleagues on the development of glacier ice from the snow crystal. It traces the crystal growth down to the end of the glacier and records the distribution of large crystals in glacier tongues. Various agencies are suggested which might help in the proposed investigation of the mechanisms of growth of the crystals of glacier ice.

These glaciers are in a district that was seldom visited before the great expansion of recreational tramping and mountaineering in New Zealand during the past twenty years, but the record extends back for eighty-five years. During this time the chief glaciers have receded considerably but irregularly, and for one of them the records indicate a vertical downwasting and thinning. Although this glacier shows no sign yet of recovery it is fed from the same snowfield as that which supplies another glacier descending west from the main divide and which may have begun to advance. An appreciable re-advance of the Franz Josef Glacier has already been given notice in the, Journal of Glaciology, and it may be that the steep gradients of the west-flowing glaciers of the Southern Alps enable them to respond to short-term climatic fluctuations, whereas the flatter east-flowing streams continue to shrink, in keeping with the world-wide trend.