ld be found to be of the nature of isolated islands.
Those familiar with the adjacent land, however, were all in favour of it
being continental--a continuation of the Victoria Land plateau. The land
lay to the south beyond doubt; the problem was to reach it through
the belt of ice-bound sea. Still, navigable pack-ice might be ahead,
obviating the need of driving too far to the west.

"Ice on the starboard bow!" At 4 P.M. on December 29 the cry was raised,
and shortly after we passed alongside a small caverned berg whose
bluish-green tints called forth general admiration. In the distance
others could be seen. One larger than the average stood almost in our
path. It was of the flat-topped, sheer-walled type, so characteristic
of the Antarctic regions; three-quarters of a mile long and half a mile
wide, rising eighty feet above the sea.

It has been stated that tabular bergs are typical of the Antarctic as
opposed to the Arctic. This diversity is explained by a difference in
the glacial conditions. In the north, glaciation is not so marked and,
as a rule, coastal areas are free from ice, except for valley-glaciers
which transport ice from the high interior down to sea-level. There,
the summer temperature is so warm that the lower parts of the glaciers
become much decayed, and, reaching the sea, break up readily into
numerous irregular, pinnacled bergs of clear ice. In the south, the
tabular forms result from the fact that the average annual temperature
is colder than that prevailing at the northern axis of the earth. They
are so formed because, even at sea-level, no appreciable amount of
thawing takes place in midsummer. The inland ice pushes out to sea in
enormous masses, and remains floating long before it "calves" to form
bergs. Even though its surface has been thrown into ridges as it was
creeping over the uneven land, all are reduced to a dead level or
slightly undulating plain, in the free-floating condition, and are still
further effaced by dense drifts and repeated falls of snow descending
upon them. The upper portion of a table-topped berg consists, therefore,
of consolidated snow; neither temperature nor pressure having been
sufficient to metamorphose it into clear ice. Such a berg in old age
becomes worn into an irregular shape by the action of waves and weather,
and often completely capsizes, exposing its corroded basement.

A light fog obscured the surrounding sea and distant bergs glided by
like spectres. A