590 THE JOURNAL OF GEOLOGY. 
toward the direction of the prevailing winds and away from the direc- 
tion of their progression. ‘The middle and most advanced part of the 
parabola tends to travel faster than the flanks, which are held back more 
effectively by the protecting growth of dune plants. In fact this center is 
often blown away entirely, and the flanks then remain as isolated nearly 
straight ridges which run approximately parallel with the direction of 
the prevailing winds. Such nearly straight and parallel ridges are there- 
fore not to be regarded as independent formations, but rather as 
remnants of earlier loop-shaped ridges. The author also observes that 
the trend of the dune ridges changes in different localities with the 
changing proportions of the strongest prevailing winds. The direction 
inland, also, is a little different from the direction near the coast, but 
the cause for this difference is not yet very clear. It is shown that the 
dunes, in traveling over the land, do not always leave a deposit along 
their course. In several places it was noticed that the winds in the rear 
of the dunes have eroded the ground a few feet below the original level 
seen in front of the advancing sand. ows UW. 
A Great Pre-Glacial River in Northern Canada. 
At the annual meeting of the Royal Society of Canada held in 
Ottawa on May 15, Professor Robert Bell of the Geological Survey 
read a paper on the above subject, illustrated by a map which he had 
prepared as the result of much study and extensive observation in the 
northern regions. We take the following report from the O/fawa 
Journal. Itwas, he said, generally conceded by geologists that just before 
the advent of the glacial epoch, the continent of North America stood 
at a considerably greater elevation than at present, the difference, 
according to some authorities, amounting to two or three thousand 
feet, if not more. The difference was greater towards the south, as 
compared with the present general altitudes. The inevitable result of 
this would be to greatly alter the river systems. We should find in 
northern Canada a wide central drainage area equal to about one-third 
of the present land surface of the continent, the center of which would 
be in the region now covered by Hudson Bay. 
This great inland sea does not average 4oo feet in depth, and it 
would be all dry land even with a very moderate elevation. 
Hudson Strait is much deeper and it would either form a long bay 
or a river valley, according to the amount of the continental elevation. 
