HUMMOCKING. 
38'7 
the floe slowly part in the middle. The lines of pre- 
viously marked fissures rise up into gigantic tables. 
Tables of one side oppose those of the other, and the 
margins of the floes from which they have arisen are 
pressing on with renewed energies to fill uj) the par-' 
tial vacancy. Tables become more and more perpen- 
dicular ; the edges beneath meet again, grind, fight, 
rear themselves into fresh tables, thrusting over thosd 
first formed. New cracks rend the level ice. New 
curves fall into tabular masses ; and thus in a few 
minutes the tranquil surface of frozen snow is cover- 
ed by fragmentary barriers, grander and more massive 
than the Pharaonic rubbish of the Pamesium. 
Differences of resistance along the margin of the 
floes, owing to irregularities in their lines of junction, 
give, of course, every irregularity conceivable to this 
action;* and it is only after it has continued suffi- 
ciently long to break all protruding edges, that the 
axis of the hummock approximates to a right line. 
My sections exhibit great diversity in this ; hut we 
learned, by the direction of the forces ami the charac- 
* The tliickness of the ice, which the wood-cut on the following page is in- 
tended to represent, was between eight and nine feet. The heigiit of one ob- 
liquely-fractured table was sisioen feet. The whole mass was thrown up from 
a previously solid floe in less than fifteen minutes. It was one of those on 
which Brooks and I practiced balancing during the commotion of the 23d of 
.March. 
