A thifd okss is composed of gravel and sand, often arranged in ter- 

 races, which in some cases occur at different levels, following the 

 sinuosities of the bays and headlands of the seacoast ; in others 

 ramif3'ing into the interior, along the sides of deep cavities occu- 

 pied by freshM'ater lakes. 



The well-known " parallel roads " of Lochaber offer the most 

 striking example of terraces at different levels above a series of ex- 

 isting lakes, and their explanation has been long the subject of cori- 

 troversy. For many years it was the favourite hypothesis, based, 

 however, upon the supposition of iheiv perfect parallelism, that these 

 lines of shingle VA^ere the shore lines of the lakes when they stood 

 at higher levels, from which they have been successively let off by 

 the breaking down or wearing away of their barriers. Though 

 supported by several good observers, this view has always presented 

 great difficulties as to the demand upon our belief in the wearing 

 away and destruction of enormous barriers. Very recently, indeed, 

 the expounders of a terrestrial glacial theory have at once obviated 

 all difficulty, by constructing in their imagination enormous Walls 

 of ice 2000 feet high, by which such lakes were formerly sup- 

 ported, and by the sudden melting of which they have been let off. 

 Unable to screw my courage up to the belief in this glacial expla- 

 nation, I will not now repeat the many objections which must be 

 raised against it, but will simply join those who prefer to invoke the 

 more rational, and, as it appears to me, perfectly satisfactory hypo- 

 thesis, that all these terraces of gravel (including those of the 

 parallel roads) are nothing more than ancient lines of beach, which 

 ar6 so many marks of the successive rise of the land. This vievi' is, 

 that, as you all know, which was so ably sustained by Mr. C. Dar- 

 win, who, in pointing out their analogy to raised beaches in other 

 countries, has also shown, that as similar materials occur in the 

 great chasm of the Caledonian Canal at still greater altitude (900 

 feet above the sea), it was impossible to i-efer them to any other 

 cause than submarine elevation. I have too long entertained the 

 same opinion respecting most of the great gravel accumulation of 

 our isles, to doubt that this is the true explanation. I would add, 

 however, that the southern shores of the Murray Frith, which are 

 of Course open to the wide ocean, offer to my eye still more con- 

 vincing proofs of the accuracy of this view, than the patches of 

 gravel along the Caledonian Canal. The terraces of gravel, sand and 

 boulders which there occur at different levels, may, in fact, be traced 

 from the slopes of the mountains of Morayshire and Elgin, to those 

 heaps which lie in the great gorge of the centre of the Highlands ; 

 and therefore I maintain, that all these accumulations must have 

 been once connected, and were all originally formed under the sea. 



No one who has read M. de Beaumont's report on the memoir of 

 M. Bravais, can fail to be struck with the strong, nay even direct 

 analogy which the gravel and shingle beds of the marine bays and 

 freshwater lochs of the Highlands of Scotland bear to the terraces 

 of the Norwegian fiords ; and as the latter author has now by patient 

 observation brought the phaenomeha observed among the latter under 



