1911] Fernald.— Expedition to Newfoundland 159 
a high tide should throw some of this back, which it often does, the 
wind, by blowing this inward, at once begins to build up new hills." 
It is apparent, then, that during the maximum development of 
the last ice-advance (the Wisconsin of glacial geologists) a very large 
portion of our coastal bench must have been above the sea-level of 
that time, forming, as 1t were, a broad sand-bar off the mouth of the 
Gulf of Maine at least to the edge of Georges Bank. Opposite this 
to the east would have been another continuation of the sandy belt 
running along the Nova Scotia coast and across Sable Island Bank 
and the Banquereau to Cabot Strait, with St. Pierre Bank beyond. In 
fact, since the ice-action was very slight about the Gulf of St. Lawrence 
and there is no evidence of ice-action on the Magdalen Islands, it is 
safe to assume that the channel from the Gulf of St. Lawrence (Cabot 
Strait) was not then of nearly such depth and breadth as at the present 
time. Then when the ice began to recede and the water from its 
melting (impeded in its progress seaward by various barriers) began 
to return to the ocean, the deeper channels would first become filled 
and not until the ice-front had reached a comparatively high latitude 
would the more elevated portions of the coastal bench become sub- 
merged. 
Here, then, would be a strip of sands and similar soils stretching 
with only slight interruptions from our South Atlantic coast to New- 
foundland, and it is probable that not far in the wake of the receding 
ice and the more boreal plants and animals the Coastal Plain plants 
with the Newfoundland vole and the muskrat crossed on this sandy 
bridge to Newfoundland, though some other plants and animals which 
farther south occur with them or to which they are closely related — 
Sabatia dodecandra, Ilex glabra, etc.— failed to cover the entire dis- 
tance. Although many of the plants which seem thus to have reached 
Newfoundland belong to southern or even sub-tropical groups and on 
the continent find their great development from Cape Cod to Texas or 
even eastern Mexico, itis apparently unnecessary to assume that at the 
time of this migration the southeastern edge of the ice was far north 
of Newfoundland or even had fully receded from the island. For to- 
day, with arctic 1ce often fringing far into the summer the eastern and 
northern portions of Newfoundland, and snow and ice lying until 
mid-July or even through the drier summers upon the bleak slopes, 
Schizaea pusilla and Xyris montana, isolated representatives of large 
tropical families, and Bartonia iodandra, an essentially endemic 
