ICE OBSERVATIONS. 195 
1} in. is a convenient size.* In the case of boulders, note should 
be taken as to whether their surfaces are scratched, grooved or 
polished. 
Blocks of sandstones, limestones and clays, should be carefully 
searched for fossils, for which purpose, whenever possible, the whole 
of the boulder should be broken up. One specimen of each such 
kind of rock present should be collected, and a note taken of the 
relative number and size of the different kinds, 
4, Berg Forms.—Some idea as to the age of bergs can be obtained 
from their forms. The bergs may be assumed to begin, in the great 
majority of cases, as fairly flat-topped blocks of land ice, and, as they 
float away from their place of origin, they are worn into peaks and 
pinnacles. Information should therefore be collected as to the dis- 
tribution of flat-topped and pinnacled bergs. 
5. In the case of stratified icebergs, the thickness and texture of 
the different layers should be noted, and the general average texture 
of the berg above sea level. This information enables an estimate to 
be made as to the total thickness of the icebergs, which cannot be 
directly determined unless they are aground, when soundings beside 
them show their full depth. A stratified appearance may be due to 
either of two causes—(a) the inclusion of layers of clay, grit and 
stone, which are generally restricted to the lower part of the glacier, 
seldom exceeding, according to Chamberlin, about seventy-five feet 
in the Greenland glacier (double that amount being a possible maxi- 
mum), with clean ice above; or (2), an alternation of layers of blue 
(solid) and white (more porous) ice, which would probably imply 
little more than that the mass had come from a large glacier. 
6. Floe Ice. 
(1) The thickness of ice formed in a single season, the rate of its 
increase in thickness, and the maximum thickness f of floe ice met 
with should all be noted. 
(2) The extent to which newly formed thin sheets of ice are 
* As it may sometimes be practically impossible, owing to difficulties of carriage, to 
bring away specimens of this size, it may be well to remark that a fragment, little more 
than a cubic inch in volume, may be quite sufficient to determine the species of a rock; 
provided always, the sample be carefully selected to show the average character of the 
rock, ie. it must represent the least decomposed part, be free from veins or adven- 
litious minerals, but should show any mineral banding, like stratification, if such 
there be.—T. G. B. 
t+ The maximum thickness of the ice in the North Polar Sea observed by Nansen 
during the drift of the Fram, from April 10, 1894, to Feb. 6, 1895, just exceeded 
nine feet; but it attains to a greater thickness by packing of the floe—* Farthcst 
North,’ vol. i. pp. 404-406. 
O2 
