168 Trans. Acad. Sci. of St. Lows. 
A U-shaped water gauge was mounted in another stop- 
per of the flask. In the third stopper of the flask a tube 
with a bulb containing calcium chloride was mounted, 
this tube being provided with means for connecting the 
confined air with the outer air. The condensers were 
wholly removed from the machine. The glass bottle was 
placed upon a sheet of heavy plate glass. 
When the discharge-knobs of the machine were near 
enough together so that the brush discharge between 
them was accompanied by faint disruptive effects, the 
pressure within the flask could be increased by about 
two grams-weight per sq. em., the effect of the negative 
electrification being somewhat greater than that of the 
positive. When the knobs were far enough apart to pre- 
vent disruptive discharges, no luminous effects being 
observed within the flask, the change in pressure due to 
either terminal was reduced to between one and two mm. 
as shown by the water gauge. After the pressure due to 
electrification of the air had reached a final limit, a trans- 
fer of the many pointed collector to the front of the other 
terminal, would result for a time in a slow decrease in 
pressure, and then in a slow increase to the former 
limit. The decrease in pressure did not begin until the 
discharge from the other terminal had begun. Heat ef 
fects had been practically eliminated. 
Tf the gas contains moisture a permanent decrease in 
pressure at once results, due to condensation of vapor 
upon the sides of the vessel. 
These results seem to indicate that there is an electri- 
eal condition of the gas, for which the Boyle-Gay-Lussa¢e 
constant is a minimum. When this condition is reached, 
the second term of equation (1) is zero, as applied to 
molecules of gas, 
An attempt has been made to determine whether oF 
not the value of the gravitation term of Eq. (1) can be 
affected by electrical action, when the effect represented 
by the final term is eliminated, The apparatus used was 
