wide back of a large fish diving down into the concave
side towards the trough. This looked very much like porpoises or
dolphins jumping in a heaving sea; only that in my memory picture the
real dolphins always jump in the opposite direction, against the run of
the waves, bridging the trough.

In other places a fine, exceedingly delicate crest-line would spring up
from the high point of some buried obstacle and sweep along in the most
graceful curve as far as the eye would carry I particularly remember one
of them, and I could discover no earthly reason for the curvature in it.

Again there would be a triangular--or should I say
"tetrahedral"?--up-sweep from the direction of the wind, ending in a
sharp, perfectly plane down-sweep on the south side; and the point of
this three-sided but oblique pyramid would hang over like the flap of
a tam. There was something of the consistency of very thick cloth about
this overhanging flap.

Or an up-slope from the north would end in a long, nearly perpendicular
cliff-line facing south. And the talus formation which I have mentioned
would be perfectly smooth; but it did not reach quite to the top of the
cliff, maybe to within a foot of it. The upsloping layer from the north
would hang out again, with an even brow; but between this smooth cornice
and the upper edge of the talus the snow looked as if it had been
squeezed out by tremendous pressure from above, like an exceedingly
viscid liquid--cooling glue, for instance, which is being squeezed out
from between the core and the veneer in a veneering press.

Once I passed close to and south of, two thickets which were completely
buried by the snow. Between them a ditch had been scooped out in a very
curious fashion. It resembled exactly a winding river bed with its water
drained off; it was two or three feet deep, and wherever it turned, its
banks were undermined on the "throw" side by the "wash" of the furious
blow. The analogy between the work of the wind and the work of flowing
water constantly obtrudes, especially where this work is one of
"erosion."

But as flowing water will swing up and down in the most surprising forms
where the bed of the river is rough with rocks and throws it into choppy
waves which do not seem to move, so the snow was thrown up into the most
curious forms where the frozen swamp ground underneath had bubbled,
as it were, into phantastic shapes. I remember several places where
a perfect circle was formed