182 Dr. Marcet on the specific gravity and temperature 
tioned in the annexed tables, that this separation of saline 
matter does not exclusively belong to ice actually formed ; 
but that it also prevails, more or less, in water which is only 
approaching to the state of ice, or has just passed to the liquid 
state; so that (as appears from specimen 64), there are cir- 
cumstances under which water may be found on the surface 
of the sea almost entirely deprived of its saline contents; 
which fully accounts for the great difference of density ob- 
served in the northern ocean, between the surface of the sea 
and its lower strata. This separation of the saline matter 
had long been shown, by experiments upon a small scale, 
to take place during the freezing of sea water ; and Mr. 
Nairne, who ascertained this point so far back as the year 
1776* states that this congelation takes place at the tem- 
perature of about 28,5°, the water thus frozen being almost 
entirely freed from its salt. Upon trying a similar experi- 
ment with the air pump, in the mode invented by Mr. Leslie 
for artificial congelation, I found that I could, without the 
least difficulty, and in the course of a few minutes, freeze 
sea water of the specific gravity of 1026, The water con- 
gealed when the thermometer reached 27®; then it rose to 
28°, and remained at that temperature. This experiment 
being repeated with another portion of the same water, but 
more slowly and with weaker sulphuric acid, the temperature 
gradually sunk to 20 0 , when the whole mass froze at once, 
the ice being quite smooth and not at all frothy, though it 
did not exhibit the dry snowy surface which is observed in 
the freezing of fresh water under similar circumstance. The 
thermometer, as in the former case, rose to 28° instantly, and 
* Philosop. Trans. Vol. 66 , 
