162 EXPERIMENTS ON THE SPECIFIC 
10.38 cubic inches. To ascertain the exact ca- 
pacity, the ball was first filled with mercury, 
and the weight of the fluid then ascertained. 
The tube was then filled and the additional 
weight of mercury determined. I next calcu- 
lated how many grains of water the ball and 
tube would contain, the specific gravity of the 
mercury being considered 13.5, this was deter- 
mined with each instrument, the trials being 
made independently of each other, and the 
weights found to agree. The tube and ball ha- 
ving been well cleaned, carbonate of potash 
previously made red hot, and put into a small 
tube of filtering paper, was introduced into the 
open end of the tube, which was then inverted 
into a cup of mercury. It was suffered to re- 
main in this situation until the mercury stop- 
ped rising in the tube. The ball and tube 
were then filled with mercury, well boiled, 
and poured while hot into the tube through 
a paper funnel dried for the purpose. When 
filled, every bubble of air was carefully de- 
tached by a slender strip of whalebone: the 
tube inverted into a basin of mercury was of 
course a barometer, the comparative accuracy 
of which was judged of, comparing it with a 
good standard instrument. 
