32 MICHIGAN SURVEY, 1908. 



inclined at a high angle, this was not their original position; in all 

 probability they were formed in a nearly horizontal position by fissure 

 eruptions under the sea, because the lavas are interbedded with shales, 

 sandstones and conglomerates. In thickness these lava beds vary 

 from a few inches to hundreds of feet. The narrow beds often show 

 upper and lower surfaces filled with small cavities (amygdules) in 

 contrast with the denser central part. These cavities were formed by 

 gas or vapor while tlie lava was hot, aud leave such ai rock porous 

 and less resistant to disintegrating agencies and to erosion. The same 

 principles also hold for the thicker beds of lava; the outer parts are 

 more porous and softer than the central part. This structural difference 

 is clearly shown in the topography of the island; the ridges mark the 

 central or more resistant parts of the truncated lava beds, while the 

 valleys, in general, have been worn into the softer outer j)arts of the 

 lava and into the interbedded sedimentary rocks. These beds are of 

 Keweenawan or pre-Cambrian age; their formation ceased with an ele- 

 vation of the land from the sea and their destruction was begun by 

 the agents of subaerial erosion. These processes continued until the 

 titled strata were truncated and reduced to a base level. Again the 

 region was depressed and upon this eroded surface were deposited un- 

 conformably those red sandstones and conglomerates which now char- 

 acterize the Siskowit Bay region and to the southward, and are of 

 Cambrian age. Once more the region was elevated, titled and subjected 

 to prolonged erosion and the strata truncated as had been done with 

 the Keweenawan. Similar processes continued until the marked eleva- 

 tion of the land, which took place at the close of the Tertiary, and 

 which initiated the repeated glaciations of the Ice Age. 



With the extension of the last or Wisconsin ice sheet in the Superior 

 basin, Isle Eoyale was completely overridden by the movement of an 

 ice sheet from the northeast that moved almost parallel to the ridges, 

 but was somewhat more inclined from the east (Lane '98, p. 1S3). For 

 this reason there was a tendency to plane down the southeastern slopes 

 and to preserve the steeper ones which had been formed on the north- 

 western side (Foster & Whitney, '50, p. 202). As the island has a 

 topography which indicates subaerial rather than marine erosion, it 

 must have had at one time a residual soil, which, unless it had been 

 swept away by a former ice invasion or the waves of some body of water, 

 was probably removed at this time with the minor inequalities of the 

 surface. In this manner the Superior lobe buried the island under 

 several thousand feet of ice and continued its movement far to the 

 southward, leaving a glacial desert in its wake. This condition of affairs 

 lasted until the return movement broke up the great ice sheet into lake 

 basin lobes and brought the receding ice front into the Superior basin. 

 As soon, however, as this lobe wasted away from the margin of this 

 basic, the water from the melting ice accumulated before it and formed 

 a lake which, overflowing the rim, found its way through the St. Croix 

 valley to the Mississippi river, as indicated in Fig. 51. But, as the ice 



