292 CRUISE OF THE NEPTUNE 



is deflected to the westward along the southern shores of the 

 island, and finally enters the southern part of the Gulf of St. 

 Lawrence, while the remainder is soon melted in the warmer 

 waters of the Atlantic south of Newfoundland. 



Ice commences to form in the smaller bays of the northern 

 parts of Hudson bay and Hudson strait early in October, and 

 by the end of that month the northern harbours are frozen over. 

 The more southern ones, especially those at the mouths of the 

 larger rivers, do not close until late in November or early in the 

 following month. By the beginning of January, James bay is 

 frozen across, and at the same time solid ice usually extends 

 from the east coast of Hudson bay to the outer line of islands, 

 some sixty or seventy miles from that coast. In other parts of 

 Hudson bay, and in Hudson strait a margin of solid ice usually 

 extends from one to five miles from the shore, except where the 

 coast is high and bold with deep water close to the base of the 

 cliffs. In such places, especially at headlands, solid ice does 

 not form, and the natives in winter often have to make long and 

 difficult detours inland to pass them. 



The main body of Hudson bay does not freeze solid, and the 

 same may be said of Hudson strait. Although this is the case, 

 these waters are quite unnavigable for ordinary ships during 

 the winter and spring months owing to the great sheets of heavy 

 ice borne backwards and forwards by the tides and currents, 

 and drifted about by the winter gales. There is little doubt 

 that a specially constructed ship for ice navigation might pass 

 through Hudson strait at any season, but the voyage would be 

 a long one, and the difficulties and dangers would be great. 



The ice of Hudson bay and the greater part of that of Hudson 

 strait is of local origin, being formed by the freezing of the 

 surface of the sea near-by. Observations on the growth in thick- 

 ness of the ice were made in Fullerton harbour throughout the 

 winter of 1903-04, and a record of the weekly observations is 



