242 P110CEEDINGS OF THE GEOLOGICAL SOCIETY. [Jail. 21, 



little delta of the same kind in Glen Collarig, and also a large one at 

 the mouth of the Rough Burn in Glen Spean. 



A further way of testing the matter presented itself. 



Had the lakes been freshwater, they should each have had an out- 

 let for the discharge of their excess, and we ought to be able to 

 point out the channels along which the streams descended to a lower 

 level. Now this can be done for all the three Glen Roy lines, 

 and likewise for that of Glen Gluoy. It has been ascertained that 

 the latter, which is the highest of all, coincides in level with the 

 watershed at the top of its glen*; so that were Glen Gluoy filled to 

 the height of its line, the water would pour over this col or summit- 

 level down a hollow, called Glen Turrit, into the upper part of Glen 

 Roy. The fall from this col to the highest line in Glen Roy is, 

 according to Messrs. Milne-Home and Chambers, nearly thirty feet. 

 The overplus of the Glen Gluoy lake would, therefore, run into that 

 of Glen Roy. That this was actually the case is well confirmed by a 

 huge old delta at the mouth of Glen Turrit, coinciding in its upper 

 surface with the lowest of the Parallel Roads. This delta is out of 

 all proportion to the size of the stream. Standing at the junction of 

 the Turrit and the Roy, I estimated the water in the latter at fully 

 three times as much as that of the Turrit. Seeing then that the 

 channels of the two streams are similar in character, the Roy should 

 have had much the larger delta. Instead of that, the delta of the 

 Turrit is more than twice as big as that of the Roy. It is a consider- 

 able mass, some furlongs wide, with a flat surface, and a front rising 

 at its centre ninety feet above the bottom of the valley — a striking- 

 proof how long the lake must have existed. Deep terraces of delta- 

 matter fill the glen here at all the lateral openings, and the scene 

 strongly reminds one of Dr. Hooker's sketch of the terraces in the 

 Yangma valley (Himalayan Journal, vol. i. p. 242). 



With regard to the uppermost Glen Roy line, it was ascertained 

 by Sir T. Lauder-Dick, and the fact has been further verified by Mr. 

 Darwin, that it coincides with the watershed at the head of Glen 

 Roy. So that were Glen Roy filled with water to the level of its 

 uppermost ' road,' the lake would have discharged its excess over 

 this col into Lochan Spey. 



Again, the middle line of Glen Roy coincides (as was first pointed 

 out by Mr. Milne-Home) with the col at the head of Glen Glaster, 

 a little valley which branches off to the eastward from the lower 

 end of Glen Roy. This point I verified myself, and found the line 

 to be just on a level with the watershed, or, more accurately, a foot 

 or two above it. The parting of the waters is in a flat swampy hol- 

 low, over which peat is now growing. Here a small stream rises, 

 and flows eastward down a hollow into the Rough Burn, which en- 

 ters Glen Spean opposite Loch Treig. The fall, or descent, of the 

 stream from this Glen Glaster col to its junction with the Rough 

 Burn is, by an aneroid observation, about seventy feet. The trace 

 of an old channel is here visible, that looks to have belonged to a 



* More accurately, it is a foot or two above it ; see Darwin, Phil. Trans. 1839, 

 p. 42. 



