ugh the porthole,
filled the cabin with dim shadows. Toward midnight, mingled with the
noises of the ship, another and more ominous sound became audible--the
grinding of the ice in the channel outside.

I threw on my clothes and went on deck. The tide was running flood, and
the ice was moving resistlessly past the point of the cape. The nearer
ice, between us and the outer pack, was humming and groaning with the
steadily increasing pressure. By the light of the moon we could see the
pack as it began to break and pile up just beyond the edge of the
ice-foot outside us. A few minutes later the whole mass broke with a
rabid roar into a tumbling chaos of ice blocks, some upheaving, some
going under, and a big rafter, thirty feet high, formed at the edge of
the ice-foot within twenty feet of the ship. The invading mass grew
larger and larger and steadily advanced toward us. The grounded piece
off our starboard beam was forced in and driven against the big ice
block under our starboard quarter. The ship shook a little, but the ice
block did not move.

With every pulse of the tide the pressure and the motion continued, and
in less than an hour from the time I had come on deck, a great floe-berg
was jammed against the side of the _Roosevelt_ from amidships to the
stern. It looked for a minute as if the ship were going to be pushed
bodily aground.

All hands were called, and every fire on board was extinguished. I had
no fear of the ship being crushed by the ice, but she might be thrown on
her side, when the coals, spilled from a stove, might start that horror
of an arctic winter night, the "ship on fire." The Eskimos were
thoroughly frightened and set up their weird howling. Several families
began to gather their belongings, and in a few minutes women and
children were going over the port rail onto the ice, and making for the
box houses on the shore.

The list of the _Roosevelt_ toward the port or shore side grew steadily
greater with the increasing pressure from outside. With the turn of the
tide about half-past one in the morning, the motion ceased, but the
_Roosevelt_ never regained an even keel until the following spring. The
temperature that night was 25 deg. below zero, but it did not seem so very
cold.

Marvin's tidal igloo was split in two, but he continued his
observations, which were of peculiar interest that night; and as soon as
the ice had quieted down Eskimos were sent out to repair the igloo.

Strange to