1819.] and the Specific Gravity of Atmospheric Air. 341 
weight at 60° indicated by the following proportion, 1-00063 : 
1 :: 13562°6 : 13554-06; from this last number, 0°03 is to be 
taken, as the correction for buoyancy of air. 
Hence the ratio of water at 60 to air at 60° and 30° is 
— = 827:472: 1, or thereabouts. 
If we estimate the specific gravity of oxygen at 1-111], and 
that of hydrogen at 0:0694 (both at 60° and 30°), and suppose 
water to be composed of hydrogen and oxygen united in the 
proportion of two volumes of the former to one of the latter, the 
specific gravity 827-472 will indicate the formation of water to 
be effected by the union of 1324 volumes of hydrogen with 662 
volumes of oxygen—the whole condensed into one volume, the 
specific gravity of which will be 827-4338. This comes so near 
the former determination that I have little doubt but the assigned 
is the correct, or at least very nearly correct composition. The 
specific gravity of air to water would in this case be represented 
by @-001208555: 1; and reckoning the weight of a cubic inch 
of water at 252°525 gr.a cubic inch of air will weigh 0°30519035 
gr. or 0°30519 gr. From this latter number, deducing the weights 
of the two component gases, the composition of water, weight 
of a cubic inch, and specific gravity, willstand thus : 
, Cubic inches. Sp. Gr. Weight Gr. 
Hydrogen.... 1324 ...... penal OR ODL, earch «. 28°04256 
Oxygen...... GE rss 5st 795°S482 1.1290 «.- 224°48195 
1986 827-4338 252'52451 
I am disposed to think the above a very close approximation 
to the truth, and that when the thermometer stands at 60° and 
the barometer at 30, we may estimate 100 cubic inches of dry 
atmospheric air to weigh 30°519 gr.; one cubic inch of distilled 
water 252°525 gr.; the specitic gravity of water to air as 827-437 
: 1, or reckoning water as unity as 0:00120855: 1; from these 
data I will endeavour to make the necessary corrections in Sir 
George Shuckburgh’s experiments on the weight of a given 
velume of water. (Phil. Trans. 1798.) I am aware that the sub- 
ject has been handled in Nicholson’s Journal by a gentleman of 
great abilities; and am the more confident in the estimates 
obtained by my calculations, as they approach very near his 
determinations, and the numbers to which Dr. Thomson’s adop- 
tion has added so much weight. 
I shall only notice the experiments made with the brass 
sphere: in these the experimenter himself seems to place most 
confidence ; and they were undoubtedly conducted with very 
little liability to error. As the contents of the sphere seem to 
me to have béen estimated at 64° Fahr. (the temperature at 
which the length of the bar was valued) to contain 113-519147 
cubic inches, and its expansion being for each degree of Fahr. 
