temperature of the ocean and atmosphere , &c. 279 
In all the experiments on the density of sea-water, the re- 
sults of which are recorded in the Journal, the water used 
was taken from the surface of the ocean, in a large clean 
bucket. The results introduced before we passed the equator 
the first time, were procured at sea ; the remainder, from 
o° 1 2" south lat. to Ceylon, were obtained on land from expe- 
riments made on specimens of water preserved in well corked 
phials. In the experiments on board ship, as soon as the 
water was drawn, its temperature was ascertained, and then 
it was immediately weighed. The balance employed was not 
very delicate, for a very delicate instrument does not answer 
at sea, on account of the ship’s motion ; however, it was pretty 
readily affected by of a grain. The glass vessel in which 
the water was weighed, was such a one as is commonly used 
at home ; its capacity was equal to about 300 grains. In the 
experiments on shore, the same vessel was used, but a diffe- 
rent balance, one of a more delicate construction. I have 
chosen the temperature 8c° Fahrenheit, for which I have 
calculated all the results, because it is nearly the mean annual 
temperature of this place, and nearly the mean at sea, in the 
intertropical regions. 
The experiments made at sea I do not of course value so 
much, as those made on land : considered, however, merely 
as approximations to the truth, which I am sure they are, 
the results favour the general conclusion already formed by 
some philosophers, that the ocean resembles the atmosphere 
in being (caeteris paribus) of nearly the same specific gravity 
throughout. 
And farther, they lead to the conclusion, that the slight 
mdcccxvii. O o 
