FIGURE ON ICE. 
83 
continue in greater quantity directly over its source, than on 
any other part. From the roughness of the ice it is evident, 
that when the frost came on, the surface had been gently 
agitated or rippled by the wind every where, excepting that 
place occupied by the oily film, which would necessarily be 
smooth and at rest. It is also evident, that this part would not 
congeal quite so soon as that in a state of motion, as it has 
been long known, that water when at rest may be cooled many 
degrees below the freezing point without losing its fluidity, but 
on the slightest agitation the congelation is instantaneous. 
Why the figure of the hat was not represented in the ice, is 
a further corroboration of the foregoing opinion, as any oily or 
greasy fluid could not readily find a passage through a substance 
of so close a texture as felt. 
The white line bordering the figure seems to have been 
occasioned by the abrupt termination of the confused crystalliza- 
tion, and the transparency of the ice by the tranquil state of 
the water in the act of congelation, by which the extraneous 
particles, with which it was mechanically mixed, would be 
deposited. 
With respect to the snow first covering the figure, and after- 
wards disappearing it might either have been swept off by the 
winds, or, more probably, melted during the day when exposed 
on so smooth a body, while that on the other parts would be 
protected by the inequality of the surface. 
it is inferred 
from the 
roughness of 
the ice that the 
wind had 
agitated the 
water previ- 
ous to the con« 
gelation : 
except where 
defended (>y 
the oily cover* 
ing. 
The figure of 
the hat want* 
insr : 
the white 
border ; 
and the 
transparency 
accounted for. 
Why the 
figure was 
bare of snow. 
I am, 
Sir, 
Colham , April Jth, 
Your obedient humble Servant, 
ROBERT HARRUP. 
813 . 
ANNOTATION. W. N. 
Upon a subject which still seems to be obscured by difficulties, Some facts 
I should, probably, have been inclined to leave the farther ^dicate^that 0 
discussion to my correspondents, instead of making any remarks the pheno- 
myself upon the ingenious speculations of Mr H. if I vveie not ^^fonedTy 
led by his invitation to point out a few circumstances which 0 j]y matter, 
appear to militate against his inductions. 
G 2 1. When 
