OF THE PARALLEL ROADS OF LOCHABER. 
721 
always the base on which the frontal moraine reposes.”"' Elsewhere he speaks of the 
moraine r £>vofonde as “ the blocks, pebbles, and detritus held between the ice and its 
bed.”t These definitions are conflicting, and although it is evident that geologists 
have used the term in a modified sense, and that an ice-sheet acts under different con¬ 
ditions to an ordinary glacier, I think it better not to use a term in a so much wider 
significance than allowed by its author or by glacialists generally. The terms simply 
of “moraine detritus” or “sub-glacial detritus” are not open to this objection. 
It is only when the glaciers descend into wider valleys with small gradients and are 
able to expand laterally that they cease to plough their bed, and spread over detrital 
beds instead of uprooting them. But when from any cause the glacier again undergoes 
lateral compression, owing to another contraction of the channel or to the confluence 
of other glaciers, the abrading and propelling power recommences. Consequently the 
ice may in some places over-ride the terminal moraine, or it may in others force it to 
higher levels, until stopped by interferences presently to be noticed, pounding and 
grinding the ground in its furthur travel, and leaving as the product of this mixed 
origin the great body of unfossiliferous and unstratified Till.J 
From this mode of formation there would necessarily result that irregular distribu¬ 
tion of the Till exhibited in hilly and mountainous countries, and those lodgments of it 
in exceptional positions which arose from the antagonism of the extending glaciers, 
and from the blocks and heaping up of the ice described in § 8. In such cases the 
moraine detritus likewise would be stopt and piled up in some places, and spread out 
in others, whether in valleys or on slopes, at the mouth of glens or on mountain passes. 
In this way it was, I imagine, deposited in the broad entrance of the Spean Valley, in 
the narrow gorge of the Boy, on the slopes of Creag Dhu, and on the many hills 
elsewhere in Scotland. 
It must also have often happened during the growth of the ice-sheet that valleys 
were crossed by glaciers descending from lateral ravines, thereby stopping the drainage 
and giving rise to basins, which for a time received the waters and debris carried down 
by the river and tributary streams. There would thus be formed lake-basins with 
sand and shingle spread over their bed, and extending to the Till of the blockading 
glaciers; and as the ice-streams closed in, their moraines, pushed forward and over the 
sand and shingle beds, would overlap the lacustrine beds. This process might be 
repeated for a time, so that when ultimately, owing to the increasing intensity of the 
cold, the ice-streams passed over and buried the lake basin, we should have local strati¬ 
fied deposits of freshwater sand and gravel embedded in and overlaid by the unstrati¬ 
fied masses of glacier Till. Although the increasing cold would in most cases so 
strengthen the ice barriers that they became permanently established, there must have 
been cases where the volume of the river was for a time too great for the barriers to 
* ‘Materiaux pour l’Etude des Glaciers,’ vol. v., p. 201. 
f Ibid., p. 219. 
J See also on tbis subject the ‘ Great Ice Age,’ 2nd edit., cliaps. vi. and vii. 
4 Z 2 
