Mr Hugh Miller on Boulder- Glaciation. 169 



the whole of the sheltered portion was rough-grained and 

 sub-angular. Instances may occasionally be met with in 

 which both the upper and the under side is striated. These, 

 however, are few enouQ-h to be accidental. Whatever was the 

 nature of the glaciating agent, it is certain that the boulders 

 had undergone striation before they came to anchor in the 

 clay, and faced the glacial tide ; and they were liable to be 

 torn from their moorings and turned over. 



Depth at which they ivere imhedded. 



The surface-glaciation of boulders is common to all parts of 

 the till. In the long railway cutting of the Suburban Kail way 

 at Colinton, for instance, I observed pavement-boulders at 6, 

 10, 15, 20, 25, and 30 feet from the surface. The Granton 

 group of boulders to which I have referred lie evidently at 

 the very base of the till, where it rests upon shale, and is, in 

 fact, a kind of shale-mash ; and although, as was pointed out 

 by Mr Arch. Geikie, the boulder-pavements on the shore 

 further to the east are not now actually covered by till, there 

 can be no doubt, from the presence of a small scar of till 

 etched out by the waves here and there along the edge of 

 the beach, that they, too, have lain — to use the words of 

 Eobert Chambers — " deep down in the entire bed," and have 

 only been laid bare by the sea. 



The Boulder-Pavements. 



These boulder-pavements differ from other groups of 

 striated boulders by their partial arrangement into strips of 

 glaciated causeway. 



During the thirty years that have elapsed since they were 

 last described, the pavements at FiUyside have considerably 

 altered. They have been better cleared out by the waves, 

 and now, at half tide, they break the margin of the coast for 

 about a quarter of a mile with a succession of low spits 

 advancing obliquely seawards (PL VI.). There seems, thirty 

 years ago, to have been just two. There are now at least six, 

 and indistinct trains of imbedded boulders indicate at least as 

 many more in an imperfect condition. The best of them are 



