the Americans. It will be remembered that, according to Poe's account,
Pym and Peters passed through the 'great white curtain' on March 22d.
Peters says that this statement is probably correct. That date
corresponds to their autumnal equinox--about. Three months later
corresponds to our summer solstice--their midwinter. By the latter time,
and for weeks before, the antarctic sun never rose above the horizon.
But this season was in Hili-li the most beautiful and enjoyable period
of the year. The open crater of almost pure white boiling lava which I
have described, and which presented a surface of the most brilliant
light, covering an area of more than 150 square miles, was amply
sufficient to light islands from 45 to 75 miles distant. Hili-li
received some direct light from a hundred or more volcanic fires--two
within its own shores; but by far the greater illumination came from the
reflected light of the great central lake of boiling lava. The sky,
constantly filled with a circle of high-floating clouds formed of
volcanic dust, the circumference of which blended away beyond the
horizon, but in the centre of which, covering a space the diameter of
which was about thirty miles, was a circle of light of about the same
brilliancy as that of the moon, but in appearance thousands of times
larger. From this overhanging cloud (the City of Hili-li lay under a
part of its circumference) came during the antarctic winter a mild and
beautiful light, whiter than moonlight, and lighting the island to many
times the brilliancy of the brightest moonlight, though quite subdued in
comparison with that which would have been derived from the sun if
directly in the zenith. Peters says that the illumination in Hili-li at
its midwinter was about as intense as with us on a densely cloudy day;
the light not, however, being grayish, but of a pure white, now and
again briefly tinted with orange, green, red, blue, and shades of other
colors, caused by local and temporary outbursts of those colors in the
enormous crater fires.