224 . Scientific Intelligence. 
course neglected; but the air, which soon begins to stream over 
in consequence of volatilization of the materia) introduced, is col- 
lected over water in an ordinary graduated tube. e experiment 
is soon finished, and the stopper @d must then be removed, to pre- 
vent any recession of the water in the trough. 
It remains only to measure the volume of the air in the gradu- 
ated tube with the usual precautions. For this purpose the tube 
with its contents is transferred to a tall cylindrical glass vessel of 
water, and held by a clamp in a vertical position, so that the 
water is at the same level within and without the tube, and, 0 
t, from which we can caicu- 
moisture, and therefore, that in order to find the true tension of 
the confined air, we must subtract from the reduced height of the 
barometer the maximum tension of the vapor of water at the tem- 
perature ¢. Representing by / this tension (which will be found 
in Regnault’s tables), we have for the weight of the air displaced 
by the vapor 
' H-A 273 
W’=0'001293——_—_ ,. —____ 
: 76 273+¢ 
and if W represents the weight of the substance used 
V; 
Sp. Gr. = a 
For convenience of logarithmic calculation these formule are 
easily combined into the following form: 
log (Sp. Gr.) = 2°3330 + ar. co. log (H~A) + log (273+) + 47. 0 
log V + log W. 
In these perectiantions it is important that the amount of sub- 
have separately. In the case of a heavy vapor, no considerable 
vessel—might reasonably be expected i with sur 
been Pp . ence it was 
prise that thie tnetbod was found to be applicable to light, vapoF* 
