PARALLEL ROADS OF LOCHABER. 107 
sea charts. It appears also that in the Arctic regions such gravel banks 
and lines of boulders are occasionally due to the action of “pack ice.” 
In one of the long circular banks which stretch across this flat valley, 
there is a singular breach with an arrangement of boulders, which suggest the 
idea of injury by an iceberg, or more than one, coming across it from the west, 
and breaking through it, discharging cargoes of boulders at this place and 
beyond it. (See fig. 2 on Plate XIV.) 
If these embankments are all due to one cause, I cannot conceive anything 
else to explain them, than sea currents bearing pack ice flowing in upon their 
concave sides. These currents, to produce the effects observed, must. have 
flowed from the N.N.W. up Glen Spean. 
To the same conclusion I have been led by a study of the boulders on 
these banks. The vast majority of the boulders are on the concave slopes 
of the banks, and seem to have been dropped there by the ice on which they 
floated, being stopped by the banks in its farther progress eastward. 
There are several knolls on this extensive flat district, which stand up 
from 50 to 60 feet above the general surface, whose tops are thickly crested 
with boulders. The tops of these knolls are generally rock, in which case the 
top, especially on its N.W. side, is smoother than any other. One of these 
knolls I found consisted entirely of detritus ; five large boulders were on its 
top. It is represented in the following woodcut. 


SSeS = = 
SSE oe 
STOO 
Knoll covered by boulders. A, Section ; B, Ground Plan. 
The following woodcut of an angular boulder about 8 feet square, resting on 
a steepish hill facing down Glen Spean, leads to the same conclusion. Former 
