Parallel Terraces, 237 



notwithstanding the numerous sinuosities of the valleys. The 

 upper part of Glen Spean, as far as the part opposite Loch 

 Treig, presents only one of these terraces, which surrounds 

 Loch Laggan, and is prolonged round Loch Treig. This same 

 terrace extends along the left side of the valley nearly down 

 to the Bridge of Roy ; on the right side it penetrates into 

 Glen Roy, of which it makes the round, following all its sin- 

 uosities, and, continuing always at the same level along the 

 lower part of Glen Spean, it terminates, on this side also, op-, 

 posite the Bridge of Roy. It is worthy of remark, that the 

 level of the col^ separating the valley of the Spean, which runs 

 to the west, from that of the Spey which runs to the east, is 

 only a few feet above that of this first terrace. As far as Loch 

 Treig, the valley of the Spean is surrounded by one terrace 

 only ; but lower down there are two others at different levels ; 

 which follow, in the same manner, all the sinuosities of the 

 lower part of Glen Spean and of the whole of Glen Roy ; which, 

 taken at equal elevations, correspond perfectly on both sides of 

 the valley ; and both of which likewise terminate at the same 

 point as the first, but at different levels, viz. near the Bridge of 

 Roy. The first, or the least elevated of these three terraces, is 

 972 English feet above the level of the sea ; and, as it is hori- 

 zontal, its height above the bottom of the valley depends on the 

 point of observation. The second is 212 feet above the first, and 

 the third 82 feet above the second. It is to be remarked, that 

 the two upper terraces make the round of Glen Roy, whereas, 

 in Glen Spean they do not extend higher than the opening of 

 the valley of Loch Treig. 1 noticed them on the left side of 

 Glen Spean between Loch Treig and the Bridge of Roy, as 

 well as on the flanks of Glen Roy ; and I mention this particu- 

 larly, because they are not indicated at that point in the maps 

 which represent their position (Plate IV. Fig. 1). It is evi- 

 dent that these terraces indicate levels of water. The next 

 enquiry is, if the barriers which restrained these lakes have 

 disappeared, or if the valley has been elevated at different times 

 above the level of the water ? The perfect horizontality of 

 these terraces, at three different levels, appears to me irrecon- 

 cilable with the idea of a repeated soulevement of the sur- 

 face. The ablation of a rocky barrier seems impossible^with- 



