190 G. H. SQUIER 



or two it fails entirely. The figure does not exaggerate the 

 nearly straight course followed by the upper margin of the 

 deposit (to avoid misapprehension I may say that throughout 

 the article, in speaking of limestone, only that derived from the 

 Lower Magnesian horizon forming the caps of the higher bluffs 

 is regarded, the limestone found at lower levels and belonging 

 to the Potsdam series being carefully excluded). 



I would call attention to the fact that all the examples given 

 lie over against projecting angles of the primary bluffs. To 

 reach the position occupied, material must have moved in defiance 

 of the law which requires that it shall travel along the shortest 

 available course from a higher to a lower level. We are obliged 

 to account not for a few sporadic cases of such transportation 

 merely, but for such an abundance as really amounts to a con- 

 centration. I therefore feel justified in saying that under exist- 

 ing conditions it is quite impossible that the material should 

 have reached its present lodgment by rolling from the higher 

 bluffs. In the case of the deposits on the buttresses, I have 

 speculated as to the possibility of their having been earlier 

 accumulations antedating the formation of the circs. But the 

 hypothesis fails to explain all the facts even in these cases, and 

 is of no assistance whatever in cases like those illustrated in 

 Fig. 5 or Fig. 7. 



The only tenable hypothesis remaining, as it seems to me, is 

 that the valleys were filled with wind-drifted snow, which was 

 piled up around the higher bluffs, so that only their limestone 

 tops rose above. In such a case the limestone debris which 

 reached the upper edge of the drifts would sooner or later work 

 downward to a lodgment wherever the slope and other condi- 

 tions were favorable. This involves the further assumption that 

 the snow had been compacted to practically the consistency of 

 glacier ice. So far as I have been able to bring it to a test, this 

 hypothesis explains the various peculiarities remarkably well. 

 All the localities showing limestone debris thus far discovered, 

 are in places where, under the hypothesis, they would have 

 been most likely to occur. Compare Fig. 5 and explanatory 



