NATURAL HISTORY AND PHYSIOGRAPHY OF NEW BRUNSWICK. 197 
feet drumlins, one of them conical, that I have seen in New Bruns- 
wick. Below this the river continues very rough with much fall 
and many ledges for nearly two miles, where again the banks rise 
into rocky ledges, and at Lyon Stream the junction of river and 
stream is in a rocky post-glacial valley, some 30 feet deep. Below 
this the banks again fall off and the country gradually opens out ; 
the river has much fall and flows for the most part over a flat con- 
glomerate-ledged bottom. Gradually it broadens and develops 
some intervales and low terraces, but below Hardwood Creek the 
banks again rise and the valley is once more post-glacial, with 
banks often of vertical rock some forty or fifty feet high ; then 
these fall off and the river reaches a wide valley in which the 
Yoho unites with the Oromocto. 
Reviewing now this part of the river it seems plain that from 
the lake to Yoho stream, this valley is all post-glacial, and that it 
cuts directly across three, and perhaps four, low ridges* and their 
intermediate shallow valleys, which in pre-glacial times drained 
from northwest to southeast, probably into the present Piskahegan 
and Shin Creek. There is probably a low ridge just east of the 
lake, forming the eastern boundary of the old valley now occupied 
by the lake, and east of that lies the shallow valley across which 
the river now wanders. It is very likely that this valley drained 
through Little and Peltoma lakes into the Piskahegan and Mag- 
aguadavic (compare the map) in pre-glacial times. The next 
ridge to the eastward would be that extending from Roach Settle- 
ment, crossing the river at North-Branch Falls and extending 
between Shin Creek and the Piskahegan. East of that comes 
another valley in which probably Lyon Stream belongs, the pre- 
glacial position of which must have been either farther east or 
farther west, doubtless the latter; and it is likely that this stream, 
( formerly extending through the gap at Harvey to Cranberry 
Lake), flowed in pre-glacial times through the present Otter 
Brook valley into Shin Creek, while earlier than that, there is 
good reason to believe, it flowed across the southern highlands 
into the Lepreau (as discussed in Note 75). East of this lies 
another ridge and then a small unimportant valley occupied only 
by Hardwood Creek, also doubtless emptying pre-glacially into 
Shin Creek, and east of that another ridge bounding the Yoho 
Stream. If we ask now why this part of the river took this direc- 
tion across the ridges, the answer would seem to be this, that these 
shallow valleys in the glacial period were filled with drift, and the 
new stream, turned by a glacial dam from its old course into 
* The rocks of these ridges are nearly horizontal, showing they are ridges of erosion and 
not anticlines. 
