352 PROCEEDINGS OF THE GEOLOGICAL SOCIETY. [Feb. 15, 



from three to three and a half miles wide, of boidders and pebbles 

 is found along this hollow, from Dunkeld, by Blair Gowrie, eastward 

 to the sea at Forfar, and that, although its course is followed by 

 no great river, it is marked everywhere by lakes and ponds, sur- 

 rounded by gravelly ridges 50 to 70 feet high. Sir Charles further 

 remarks the absence of marine fossils, and admits the difficulty he 

 found of accounting for the arrangement of this upper gravel of 

 Forfarshire, proposing as the most feasible explanation, that an estu- 

 ary had extended by Blair Gowrie to Dunkeld, and that these over- 

 lying ridges of sand and gravel might have been formed one after 

 the other in the same manner as the bar of sand and shingle that 

 now crosses the mouth of the Tay. 



I may here remark that if an estuary had extended up in this 

 manner, a similar one would have stretched far up the valley of the 

 Tay, and this watershed would have formed a land-strait between 

 these two arms of the sea : and Mr. Darwin, in explaining his theory 

 of the origin of the parallel terraces in Lochaber, has shown that 

 such land-straits have a tendency to be silted up, the more so in pro- 

 portion to their narrowness, and on the recession of the sea would 

 coincide with a terrace or raised beach at corresponding levels. 



Now here, instead of a silting up, we find remarkable denudation ; 

 instead of belts of terraces, we have tumuli and ridges of gravel along 

 the midst of the depression. Further, in the neighbourhood of the 

 Kirk of Caputh, to the eastward of the Pass of Birnam, we see tails 

 of detritus stretching from behind the rocky masses of Stenton, that 

 present denuded fronts looking up the valley. 



About Killiecrankie, in the narrows of the Tummel, the hill-sides 

 are swept to a remarkable degree ; and on other rivers like pheno- 

 mena will occur to the recollection of most readers. 



In tracing the water-rolled gravel up to such narrow passes, 

 I have observed that its materials become coarser, and its arrange- 

 ment more tumultuous and irregular as it approaches such points, 

 until we find it in their immediate neighbourhood containing large 

 boulders many feet in diameter. I was particularly struck with this 

 in examining the Moor of Dinnet, before alluded to. This moor con- 

 Fig. 5. — Section of the Moor of Dinnet. 



W. Camus-o-may. E. 



1. Granite. 2. Boulder-drift. 3. Water-rolled gravel. 



sists of a wide plain of drift covered by a sheet of this rolled gravel, 

 which in its eastern portion is spread out in a horizontal manner, 

 containing many seams of sand interstratified with fine pebbly shin- 

 gle ; but as I approached the rocks of Camus-o-May, where the 

 valley is narrowed by a riclge of granite protruding from Culbleen 

 that almost bars the passage, I found (as I expected from similar 



