592 
DE. W. FAIEBAIEN AND ME. T. TATE ON THE LAW 
1^-inch tube d, which was supported on a 
frame of wood e e, sliding up and down ver- 
tically on the larger frame yy, on which the 
apparatus was supported. The frame e e, and 
the tube d, were balanced by cords passing over 
pulleys in the top of the frame, and weighted 
by bags filled with small pieces of iron. In this 
way the tube d could be adjusted with the 
greatest facility, so as to maintain the upper 
level of the mercury column at a constant posi- 
tion. To read the lower and variable level of 
the column, a rod g g, graduated into tenths 
of an inch, was fixed vertically, carrying a 
finger and vernier, h, which could be made to 
coincide accurately with the top of the menis- 
cus of mercury, c. 
For heating the globe, a glass bath, k, con- 
taining oil, was provided, fixed in an outer iron 
mercury bath, 1. This Avas heated by a coil of t 
gas-jets, and the oil was stirred continually, 
excepting at the instant of adjusting the 
upper level of the column to the level of 
the platinum Avire. The globe was fixed in 
the oil-bath by a stufiing-box. 
The method of conducting the experiments 
was as follows. The globe a was filled with 
dry and warm mercury, the air-bubbles being 
extracted from time to time by an air-pump. 
It was then inverted to form a Torricellian 
vacuum. A small glass globule of water was 
then inserted, the platinum wire fixed in its 
place, and an india-rubber cap fitted over the 
extremity of the stem. It was then transferred 
to its place in the oil-bath, and fixed there. 
The india-rubber cap was replaced by an open 
glass cistern, so that the glass tube d could be 
elevated to its position. The gas-jets were 
lighted, and the temperature raised to 300°. 
From this point the levels of the column were 
read ofF at intervals of 50° until the temperature 
of saturation was reached. The levels were 
