32 
Captain Kater’s account of the 
pounds of distilled water, at the temperature of 62 0 of Fa- 
renheit’s thermometer, the barometer being at 30 inches ; 
consequently, for any difference from this state of the baro- 
meter, as well as of the thermometer, a correction must be 
applied. Water is 831 times heavier than air when the 
barometer is at 30 inches, and the thermometer at 62 0 , and this 
varies directly as the height of the barometer, consequently 
the correction due to a difference of one inch of the barometer 
will be -f— x — = — — part of the weight of water. 
The gallon weight being counterpoised with brass, and 
brass being about 8 times heavier than water, the effect of 
the buoyancy of the atmosphere upon brass will be only 
of that upon water, and this will tend to lessen the effect 
upon the water by one eighth part of the whole quantity. 
Therefore — * x -| x 70000 grains = 2,46 grains, is the 
number of grains by which the weight of the gallon of water 
will be increased by a depression of one inch of the barometer. 
Though the variation of the temperature of the air during 
the experiments was several degrees, the effect upon the 
correction for the buoyancy of the atmosphere would have 
been so inconsiderable as to be unworthy of notice upon the 
present occasion. 
As an error of one degree in the determination of the 
temperature of the water contained by the gallon would 
occasion an error ranging from four to six grains, it must 
be evident that a knowledge of the precise temperature of the 
water was of primary importance. For this purpose a thermo- 
meter was used which had been prepared with extraordinary 
care by Mr. Bate, and the scale was such that its indications 
were estimated without difficulty to tenths of a degree. 
