The Workman-Reynolds effect was investigated during the phase change of dilute (about 2 × 10-4 N) KCl solutions into single crystals of ice. The ice crystals were oriented with the c-axes either parallel or perpendicular to the growth direction. The solute distribution in the liquid phase. near the interface (within 10 mm), was obtained with a wire-type conductivity cell. For a crystal growth rate 8.8 μm/s the freezing potentials were + 10.0 V and + 6.0 V and the specific charge séparations were 1.3 ± 0 1 × 10-6 C/g of ice and 1 4 ± 0.1 × 10-6 C/g of ice for growth parallel and perpendicular. Respectively, to the c-axes of the ice crystals . The equilibrium solute distribution coefficient was found to be 4 × 10-3 for KCl solutions for both crystal orientations. An “apparent” (because of convection in the liquid phase) distribution coefficient ranged from 0.031- 0.074. The “apparent” diffusion coefficients ranged from 1.3–4.9 × 10-3 mm2/s and varied linearly with growth rate. The ionic distribution coefficients. K+ and K-, were approximately K+ - K- = - 2 × 10-5 and K+ + K - = 8 × 10-3 for the KCl solutions.

During drilling in the Athabasca Glacier in April 1968, a cavity containing water was punctured at a depth of 9.2 m below the ice surface. Upon removing the drill, water gushed from the bore hole for about 55 s indicating an excess pressure of at least 0.25 bar within the cavity. The surrounding ice was slightly below the pressure melting point, and the excess pressure was apparently generated by the reduction in volume of the cavity caused by freezing of some of the water within it.

Thickness variations above an ice fall in the terminal tongue of Coleman Glacier on Mt Baker, Washington, have been quite small (±5 m) in contrast to the rapid expansion and shrinkage of the tongue itself during the last 20 years. The tongue has not responded as actively to recent, larger thickness changes above the ice fall as it did during its period of most active expansion between 1954 and 1956. There is no apparent relationship between the glacier thickness and the type of flow in the terminal tongue.

The results of two surveys of the frontal margin of a valley glacier in West Greenland are reported. If they indicate a constant or decreasing mass loss rate, the glacier shows similar behaviour to that of the nearby ice lobes on the ice sheet.

Recovery of the Muldrow Glacier in the area below the lower Muldrow ice fall is progressing at a rapid rate. About one-third of the lost ice thickness at the base of the ice fall has been replaced during the 10 years since the surge. The ice velocity in this region is apparently quite rapid, but quickly decreases down-stream and is barely detectable at McGonagall Pass, 6.5 km from the base of the ice fall. Muldrow Glacier still offers an excellent opportunity to study the recovery from a surge, and the flow of a normal wave into stagnant ice.

An interpretation is proposed of the phenomenon of splitting of ice dendrites growing from supercooled water and of their orientation at an angle α to the basal plane. Both phenomena are related to the instabilities which develop on the dendritic cap when the ratio Vc/Va of the growth velocities in the direction of the c- and a-axes exceeds a certain limit. The different interface supercoolings, (δTi)a and (δTi)c at the tip and at the sides of the dendrite cap are considered. Although Vc/Va increases with the bath supercooling ΔT it is found that Vc ≪ Va for (δTi)c = (δTi)a through the whole range investigated (o < ΔT ≤ 18 deg). The increase of α with the concentration of solutes is also qualitatively explained. Considering the kinetic theory of the molecular growth mechanism, the results obtained for Vc as a function of (δTi)c would indicate that most of the studied interval must correspond to the transitional regime of growth.

Segregation of chemical impurities in ice formed from very dilute solutions was observed autoradiographically, using radioactive 22NaCl or H36Cl as β-ray tracers. When water containing such radioactive solutes was frozen, the segregation of solute occurred primarily along grain boundaries and partly in grains. Successive autoradiography enabled us to observe the heterogeneous distributions of segregated impurities in ice.

Finger-rafting, similar to that reported in sea ice, is common in thin (2-15 mm) new ice on Lake Superior. It is produced by wind traction alone; neither wave action nor brine is necessary for its initiation or propagation. Fingers have been observed to progress at rates of 0.5-5 cm/s.

Three potholes and a narrow channel cut into bedrock in a side-hill position were observed beneath an Irish Sea till west of Wolverhampton, England. The potholes and the channel are believed to have been cut by subglacial or latero-glacial streams flowing beneath or immediately beside the Irish Sea ice sheet. They were later choked by sand and gravel from this ice sheet and capped by till which ended the glaciofluvial deposition. Deposits below and above the till have been 14C dated at 30 655 and 13 490 years B.P. at localities between 13 and 27 km north of the trench section described.

Much of the history of British geological thought in the second quarter of the nineteenth century centered on problems which are now explained by reference to the events of the Ice Age. This paper reviews the data and theories then current among British geologists as the background of the British response to Louis Agassiz’s “modern” theory of a glacial epoch. Today, as we read Agassiz’s amazing speculation, our own sympathy for the striking accuracy of his ideas masks from us the difficulty they faced in gaining acceptance. By first examining the context into which the glacial theory was introduced, we can then appreciate the novelty of Agassiz's efforts and understand the long delay in their achieving prominence. The present examination suggests that this delay was due to the unfortunate merger of Agassiz’s new ideas with the older drift theory of Charles Lyell.