W rigut— Glacial Drainage round Montpelier Hill, Co. Dublin. 581 
brought down by the torrent pouring through the pass, and . 
checked at this point either by the ice-face, or by water held up 
by the ice-face, most likely the latter. The removal of the check 
would be followed by the subsequent erosion of the central portion 
of the deposit, leaving, as in the present case, mounds flanking the 
valley. 
As the straightness of the line in which the gravels end off 
upon the boulder-clay platform at the mouth of the Piperstown 
Valley is suggestive of a barrier of some sort since removed, and 
as the great thickness of the gravels seems to indicate that they 
were deposited in comparatively still water, it seems not unreason- 
able, if the former existence of an ice-sheet be once granted, to 
suppose it to have been the barrier in question, and to have held 
up in the Piperstown Valley a temporary sheet of water, with an 
outlet through the pass into the Killakee Valley. The gravels seem 
to have been washed into this lake partly off or out of the ice 
itself, and partly from the hills, the torrents running between the 
ice and the bare hillside, playing, no doubt, an important part in 
their transport. 
There is a complete absence of any trace of a shore line on the 
hill slopes above the gravels. This, however, is not to be wondered 
at when the smallness of the lake is considered. Professor Fair- 
child states that only the larger of the glacial lakes of western 
New York have shore lines, owing to the fact that it requires a 
wide expanse of water for waves to gather of sufficient force and 
volume to effect noticeable erosion. 
At a later stage the continued shrinkage of the ice-barrier 
seems to have opened a passage for the waters of the lake at a 
lower level round the north side of Montpelier Hill, and their 
former route across the ridge was then abandoned, but not before 
they had excavated the trough above described. In their new 
course tne waters have left their mark in a series of terraces and 
gashes in the hillside, which, in some cases at least, can be seen to 
be quite independent of the nature of the rock. These all slope 
from west to east, and seem to be more abundant, more steeply 
graded, and more sharply marked on the east side of the spurs 
which stretch north from the hill. They are all below the 1000 
feet contour, and therefore below the floor of the gap; above this 
level the hiil is comparatively smooth. On the north-west spur, 
