ee a ee 
444 
THE DECOMPOSITION PRESSURE AT 445°. 
The next experiments were made in sulphur vapor, the bath being 
arranged as in the last experiment. At this temperature also, the first 
attempts were made with an indirect method. The simple method 
previously used seemed hardly likely to succeed here, as a change of equi- 
librium in the tube would probably occur during the process of cooling 
from the temperature of boiling sulphur. A new method was therefore 
adopted. The construction of the apparatus is indicated in fig. 1 by 
BCHI. A thick-walled, glass tube of not more than 2 mm. internal 
diameter was drawn out to form a long capillary C. The tube B on one 
side of the capillary was filled with silver oxide and closed at the free 
end. The tube was again drawn to a capillary at I and sealed, leaving 
about 3 or 4 em. between H and [. During this construction a small 
piece of tar was placed loosely in the tube where it narrowed to a capil- 
lary at H. When the apparatus had been in the bath long enough for 
equilibrium to be established a hot glass rod was brought for a moment 
in contact with the tube at H, thus melting the tar and causing it to run 
into the capillary. This heating was purely local and had no appreciable 
effect on the temperature of the gas in the tube H I. After the tar had 
solidified the capillary was broken at C leaving in the tube H I the orig- 
inal pressure. The capillary I could then be opened under water and 
the pressure calculated from the volume of gas evolved and the volume 
of the tube HI. 
While this method was being used, experiments were also made with 
a direct method, a manometer being finally constructed which gave very 
satisfactory results? It is shown as E F G in fig. 1. BC HE F G was 
drawn from a single tube of the kind used in the preceding method. At 
E it was drawn into a series of small bulbs, as these proved to be stronger 
than a single straight tube, and finally it terminated in a long capillary 
which had a nearly uniform bore (about 0. 1 mm.) in the region F G. 
The pressure was registered by a small mercury column standing at E 
at atmospheric pressure, and at a point F, about 10 cm. from the end 
of the capillary, when the pressure reached its maximum. On breaking 
the tube at C after the experiment, the mercury column invariably re- 
turned to its initial position at E. The capillary point at G was then 
broken off within one or two tenths of a millimeter from the end, and 
the volumes F G and E G were determined by drawing in mercury suc- 
cessively to F and to E and weighing the quantities drawn in. The 
smaller amount being only a few milligrams was weighed on a fine assay 
balance. I believe that this manometer gives pretty accurate results. 
The error caused by the capillarity of the mercury is not over one-tenth 
of an atmosphere, nor is it likely that the change in volume of the eapil- 
lary tube when the pressure is exerted from within is large enough to 
cause any great error. 
~~ ee 
