190 
ME. W. EAIEBAIEN AKD 1^£R. T. TATE OX THE DEXSITT 
By at once superheating the steam in the globe and then slowly reducing the tempe- 
rature until the maximum temperature of saturation is reached, we secure the following 
advantages : — The cohesion of the water to the surface of glass being once overcome, 
that force of cohesion, it may be presumed, cannot be regained until the glass again 
becomes wet, which can only occur on condensation, that is, by the reduction of the 
temperature below that which corresponds to the maximum temperature of ^saturation. 
Moreover the observation of the columns at different temperatures of superheating, not 
only supplies us with data for determining the law of expansion of the superheated 
steam, but also data of verification or correction, as the case may be, of the point of 
maximum saturation determined by direct observation. 
The results of the experiments show, Jirst, that the density of saturated steam at all 
temperatures, above as well as below 212°, is invariably greater than that which is 
derived from the gaseous laws ; second, that the law of expansion of a peiffectly elastic 
fluid does not hold strictly true for superheated steam. At the maximum temperatui-e 
of saturation, and for some degrees above it, the rate of expansion of the steam gi-eatly 
exceeds that of a perfect gas. 
Section I. — Exjperhnents at Pressures varying from fifteen to seventy pounds jjer 
square inch. 
In these experiments the thermometers were, it udll be seen, exposed naked in the 
steam of the copper boiler, and this method offered the great advantage of rapidity in 
the indications of change of temperature round the globe, and enabled the observers, by 
constant and vigilant attention to the regulation of the supply of gas, to keep the 
temperature in the copper bath as nearly as possible uniform. But the thermometers, 
from exposiu’e to pressure, require a correction. For this pui'pose the follo-sving expe- 
riments were obtained with simple water pressure. 
Experiment 1. — Thermometer A. 
Pounds pressure per square inch. 
Temperature indicated. 
Eise from pressm-e. 
0 
48-2 
20 
48-8 
0-6 
40 
49-1 
0-9 
60 
49-5 
1-3 
Here the mean rise is 0°'022 for each pound of pressure above the atmosphere. 
Experiment 2. — Thermometer B. 
Pounds pressure per square inch. Temperatui’e indicated. Eise from pressure. 
0 198-0 
50 198-9 
Here the rise is 0°-018 per pound pressui-e per square inch. 
0-9 
