FORMATION OF BERGS. 
113 
to a parallelopipedon, its remaining side could not 
have been less than 1000 feet. 
The symmetrical character of this great body of ice 
allowed me to estimate its magnitude and weight. 
Applying the recognized proportion of 8.2 below wa- 
ter for 1 above, and assuming, as Scoresby's experi- 
ments seem to justify, that thirty-five cubic feet of 
water in the Greenland seas have a weight of one ton, 
we have more than 2135 millions of cubic feet as the 
solid contents of the berg, and 61 millions of tons for 
its weight. It was therefore at least one third larger 
than the one which Scoresby measured on the eastern 
coast (Scoresby's Jour., p. 233). But great as it was, 
we saw others afterward still more stupendous, one 
of which I measured topographically. 
Many of the bergs were covered with detritus. 
From one which had thawed down to the water's 
edge, I obtained some specimens of different rocks, 
which were found adhering to its upper face. They 
all belonged to the primary series — quartz, gneiss, sy- 
enite, augitic green-stone and clay slate. Some of 
them were marked with well-defined strise, without 
angular crossings, smooth, and occasionally polished 
even highly ; others were cut in facets of more or less 
regularity. They varied in size from large blocks to 
mere pebbles, conglomerated in the ice with finely- 
powdered gneissoid material. The berg had evident- 
ly changed its equilibrium ; and it seemed as if these 
rocks had been cemented in its former base, and had 
there been subjected to attrition during its rotary os- 
cillations against the bottom of the sea. 
Others of them bore unmistakable marks of the mo- 
raines through which they had passed. The depos- 
ited material had a linear arrangement, as if dropped 
