116 Mr ^QOVQ^hy on the Impregnation of Wood with Sea-Water, 
in the sea, in augmenting its specific gravity, by impregnation 
with sea-water. In these experiments, however, some of the water 
was observed to escape out of the wood, on its being removed 
from pressure, by the expansion of the compressed air contained 
in its pores, — a circumstance that prevented me from ascertaining 
the highest degree of impregnation of which the wood was sus- 
ceptible. A mode of obviating this inconvenience occurred to 
me during my last voyage to the polar seas ; and this mode also 
promised to shew to what extent, and under what degi^ees of 
pressure, sea-v^ater might be forced through the pores of wood. 
Not having any metallic vessel suited for the purpose, I employ- 
ed a strong wine-bottle. I ground the inside of the neck (for the 
cork) perfectly circular, by means of a cone of wood with sand 
and water, and reduced it to such a form that a piece of wood, 
in the form of a frustum of a cone, fitted the neck through the 
extent of an inch in length, and formed a perfectly air-tight 
plug. This plug was of very dry ash, and two inches in length. 
It had a square head, of somewhat greater diameter than the 
rest of the plug, so that the cone terminated by a kind of shoul- 
der, touching the extremity of the neck of the bottle, to prevent 
the pressure from thrusting it farther in, and bursting the glass. 
The neck of the bottle being now heated, the plug, first coated 
with sealing-wax, was introduced, and, the heat being sufficient 
to render the wax fluid, it was worked down to the shoulder. 
The plug and the glass being thus intimately united by a thin 
intermediate coat of sealing-wax, there could be no doubt that 
it was perfectly tight. 
In this state, the bottle was sent to the depth of 125 fathoms, 
and, after remaining a quarter of an hour, was hauled up. 
About two ounces of water were found to have penetrated the 
pores of the wood. The bottle unopened was then sunk a second 
time to the same depth, and a small additional quantity of wa- 
ter was found to have entered within the bottle, at this second 
sinking. 
Now, by this process, I expected, that on pressure being ap- 
plied to one end only of the wood, instead of every part, as in 
my former experiments, the flow of water through the pores 
would force all the air contained in the wood into the bottle, and 
not confine it by compression, as had before been the case ; and 
