any great depth to be warmer than that at the surface.

On the 28th we tacked to avoid an extensive stream of sailing ice. The
temperature of the water fell to 39.5 degrees when we were near it, but
was at 41 degrees when at the distance of half a mile. The thermometer in
the air remained steadily at 40 degrees. Thus the proximity of this ice
was not so decidedly indicated by the decrease of the temperature of
either the air or water as I have before witnessed, which was probably
owing to the recent arrival of the stream at this point and its passing
at too quick a rate for the effectual diffusion of its chilling influence
beyond a short distance. Still the decrease in both cases was sufficient
to have given timely warning for a ship's performing any evolution that
would have prevented the coming in contact with it had the thickness of
the weather precluded a distant view of the danger.

The approach to ice would be more evidently pointed out in the Atlantic,
or wherever the surface is not so continually chilled by the passing and
the melting of ice as in this sea; and I should strongly recommend a
strict hourly attention to the thermometrical state of the water at the
surface in all parts where ships are exposed to the dangerous concussion
of sailing icebergs, as a principal means of security.

The following day our ship came near another stream of ice and the
approach to it was indicated by a decrease of the temperature of the
water at the surface from 44 to 42 degrees. A small pine-tree was picked
up much shattered by the ice. In the afternoon of the 30th a very dense
fog came on; and about six P.M. when sailing before a fresh breeze we
were suddenly involved in a heavy stream of ice. Considerable difficulty
was experienced in steering through the narrow channels between the
different masses in this foggy weather, and the ship received several