Dr. J. W. Spencer — Subsidence versus Glacial Dams. 267 



points, owing to subsequent terrestrial deformation ; and tlie inter- 

 ruptions occasioned by topographic features and subsequent obliter- 

 ation by erosion. All of these difficulties are greatest in the higher 

 regions, for there the beaches must be looked for amongst islands 

 and detached mountain knobs. 



The absence of marine remains seems perhaps the greatest obstacle 

 to the acceptance of a sea-level formation of the beaches, as marine 

 organisms are found only up to 520 feet.' But the Pleistocene 

 gravels occur in Georgia and Alabama, in positions facing the sea, 

 at altitudes of 700 or 800 feet, and higher up the greater valleys at 

 1500 feet,^ without their containing any marine remains. Even 

 where marine Pleistocene beaches occur on the coast of Norway, 

 there are very few localities where shells are found. How many of 

 the older geological formations are unfossiliferous ? How many 

 of those ancient beach deposits now represented by conglomerates, 

 porous sandstones, and, indeed, many clays, are entirely barren ? 

 Under such conditions have we a right to pronounce judgment on 

 the freshness of waters based on the absence of aqueous organic 

 remains ? This question will be referred to again in considering 

 the glacial dam theory. 



As to the personal equation, it ought not to pass beyond the limit 

 of conservatism into the province of obstruction : but it is quite 

 proper that it should be considered ; for, as Prof. Geikie has said, 

 when controversy ceases, the interest in the investigation declines. 



Glacial lakes are of two kinds, those whose waters are retained 

 by morainic barriers, and others sustained by ice barriers alone. 



The former class is represented in several valleys in the Alps, 

 where lateral glaciers enter and cross greater valleys ; sometimes 

 the glacier carries its lateral moraine across the valley, and builds 

 a more or less permanent earth dam. Such lakes remain long after 

 the glacier has melted away, and, even when drained, show evidence 

 of their origin. A consideration of this class of glacial lakes does 

 not enter into the subject of this paper. 



In Switzerland, Greenland, and Alaska, other glacial dams are 

 now well known. These are retained by the ice alone. When 

 glaciers, free from morainic materials, descend lateral valleys, and 

 cross other valleys, they do not obstruct the river, for it continues 

 to flow beneath the ice. However, there are many places where 

 glacial lakes occur between the ice and the sides of the valleys; 

 especially is this the case where two glaciers meet at the end of 

 a mountain spur, like Lac Tacul in Switzerland. Small glacial lakes 

 sometimes occur where lateral valleys unite with the glacial -filled 

 channel, like the Marjelen See. All modern glacial lakes are of 

 small size. One of the largest lakes described in Greenland is not 

 over three or four miles long, and a mile wide.^ 



Such lakes, when they exist above sea-level, are evanescent. Mr. 

 H. Topham described some glacial lakes of Alaska which discharge 

 by a tunnel eight miles long, under 600 feet of ice.* Mr. I. C. 



' At Montreal. ~ On the Upper Etowah River of Georgia. 



3 Medellelser om Groenland. * Proc. Eoy. Geogr. Soc. 1888, p. 424. 



