Little or no attention was given by the earlier explorers to the markings of ice-striae and other glacial phenomena, and the only records of the movement of the glacial ice noted by them was the distribution of erratic boulders. These observations have been summarized by Dawson as follows:
‘Along the Arctic coast, and among the islands of the archipelago, there is a considerable volume of evidence to show that the main direction of movement of erratics was northward. Thus, boulders of granite supposed by Prof. Haughton to be derived from North Somerset are found 100 miles to the northeastward, and pebbles of granite, identical with that of Granite point, also in North Somerset, occur 135 miles to the northwest. The east side of King-William land is also said to be strewn with boulders like the gneiss of Montreal island, to the southward. Prof. Haughton shows the direction and distance of travel of some of these fragments by arrows on his geological map of the Arctic archipelago, and reverts to the same subject on pages 393-394, pointing out the general northward movement of ice indicated, and referring the carriage of the boulders to floating ice of the glacial period.’
‘Near Princess Royal island, in Prince of Wales strait, and also on the coast of Prince of Wales island, the copper said to be picked up in large masses by the Eskimos may be supposed to be derived from the Cambrian rocks of the Coppermine river region to the south, as it is not probable that it occurs in place anywhere in the region of horizontal limestone where it is found.’
‘Dr. Armstrong, previously quoted, notes the occurrence of granitic and other crystalline rocks, not only on the south shore of Baring land, but also on the hills inland. These, from what is known of the region, can scarcely be supposed to have come from elsewhere than the continental land to the southward.’
‘In an account of the scientific results of the Polaris expedition, it is stated of the west coast of Smith sound, north of the Humboldt glacier, that “wherever the locality was favourable the land is covered by drift, sometimes containing very characteristic lithological specimens, the identification of which with rocks of South Greenland was a very easily accomplished task. For instance, garnets of unusual large size were found in latitude 81° 30´, having marked mineralogical characters by which the identity of some garnets from Tiskernaces was established. Drawing a conclusion from such observations, it became evident that the main line of the drift, indicating the direction of its motion, runs from south to north.” ’
Dr. Bell in his report on the geology of Hudson bay and Hudson strait, 1885, draws attention to the flow of the ice from the land on both sides of the strait into that body of water, while the striæ on the islands in the strait show that a great stream of ice passed eastward through the strait from Hudson bay into the north Atlantic. These observations have since been confirmed by observations of the striæ on other islands of the strait.
Tyrrell’s observations on the glacial phenomena of the barren-land region west of Hudson bay show that the country was intensely glaciated; that the centre of glaciation was on a nearly level plain now elevated some 400 or 500 feet above sea-level, there being no evidence to show that it was much more elevated during the period of glaciation. The centre of ice distribution was situated close to the western shores of Hudson bay, and the moisture sufficient to allow of such an accumulation of ice was probably derived from an open Arctic sea. The glacier moved south and southwest from this centre up a gradual grade to Manitoba, where morainic accumulations are found on the summits of the Duck mountains at elevations from 1,800 to 2,400 feet above present sea-level. Striæ evidently formed by moving ice from this centre have been found by Dr. Barlow and the writer on the branches of the Moose river to the south of James bay, where the movement was from the northwest.
There is little doubt that the ice also moved northward from the centre of glaciation, and that the evidence quoted above of the erratics found in the western Arctic islands is proof of this.
From a study of the different sets of glacial striæ, Tyrrell concluded that the centre of glaciation was, in the early part of the glacial period, somewhere to the north and west of the head of Chesterfield inlet; that later, when the ice increased in thickness, the centre of dispersion moved to a position southwest of Baker lake; while as the glacier diminished the centre moved nearer the seashore, and the final stage was probably represented by the ice-cap breaking up into a number of distinct glaciers, each with local movement of its own.
These conclusions of Tyrrell as to the southern movement of the centre of glaciation are borne out by the writer’s observations of the striæ along the shores of Roes Welcome, where as many as six sets of striæ were found at Whale point, the usual number being three. The oldest set, found only at Whale point, showed that the ice movement was from the northwest. The next in age were from N. 50° E., or almost at right angles to the oldest; following in order of age come striæ from N. 25° E., N., N. 30° W., and N.W. The last three sets are found in a number of places between Winchester inlet and Whale point; the others only at Whale point. The direction of the above sets of striæ apparently shows that the earliest accumulation of ice in the region north of the western side of Hudson bay was somewhere to the northwest; this was followed by an abrupt change in the ice-movement, which was next from almost northeast, after which the centre of movement of the ice-cap gradually shifted, by way of north, to northwest. It would also appear that the centres of dispersion were much greater in area than the limits placed upon them by Tyrrell.