Correspondence. 333 
been a cold body of water into which glacif-rs discharged bergs. It is 
not easy to understand the presence of the boreal mollusca upon any 
•other basis. The submergence lessened as time progressed, for the Leda 
■clays are overlain by the Saxicava sands. With tiiis arctic sea I would 
associate the h)cal glaciers of the White and Green mountains, as well 
as those described by me in Maine, and others described by Canadian 
:geologists in New Brunswick and Quebec. The fossiliferous strata are 
in some places covered by a till with large, more or less rounded boul- 
ders, on the coast of Maine and on the sides of the St. Lawrence valley in 
Quebec. Sir William Dawson speaks of it as a "second boulder drift as- 
sociated with the Saxicava sand, and apparently resting on the terranes 
cutout of the older clays." If we consider that Dawson confines iiis de- 
scription of the "Canadian Ice Age" to the continent as it was in this 
Champlain glacial epoch, we can accept his conclusions as truthful. It 
was a cold ocean, with floating ice and glaciers discharging from both 
the north and the south. 
Ells and Chalmers, of the Canadian Geological Survey, have repre- 
sented that the elevated land of New England and adjacent parts of 
Canada and New Brunswick was a center of dispersion for glaciers, and 
that no ice ever passed from the St. Lawrence valley over New England. 
Not to be misunderstood, I will quote from Ells, in the Canadian Kejjort 
for 1886, page 44J: "The theory of a universal ice-sheet of many hun- 
dreds of feet in thickness does not appear to meet with much support as 
applied to this region. Proceeding southeast from the St. Jjawrence 
basin, three principal ridges with elevations from 1,000 to nearly 4,- 
OOO feet above the sea, would have to bi; surmounted, which would re- 
quire a propelling force imparted to the glacier, the source of which 
cannot be found in any great continental elevation related to the St. 
Lawrence valley. The great diversity also observable in the direction 
of the stri;e at different points would appear to be opposed to this the- 
ory, for over a great portion of the eastern Cambro-Siluriaii area, there 
is a general course eitlier to the southeast or northwest. If we accept 
the former course as that in whicli the ice passed, we must explain tiie 
manner in which the ice-sheet overcame the gradual ascent from the 
valleys of the Massawippi and St. Francis rivers, whicli have an eleva- 
tion of .")r)0 feet above tlie sea level, tothe iKMglit of land on the Maine 
border, which reaches an elevation of from 1,800 to 15, 800 feet. The theory 
which ignon>s for the most part the existence of the great continental 
ice-sheet presupposes the presence of local glaciers which formed along 
the summits and crests of the principal mountain ranges, from which 
the ice descended in eilhi>r direction, inlliienced largely by existing to- 
pographical features." 
The following notes from Mr. Ells' report for 1887 (!)age lOOK) fur- 
ther illustrate his views: "Among the most interesting surface features 
in this section is the presetici' of scallered boulders of Luurentian rocks, 
gneiss, labnidorile, limestones, etc." These have been used for build- 
ing a church; and the elevation of the boulders was from 4.')0to (500 feet 
above the St. Lawrence. Similar boulders i)ccur farther inland, both 
