BETWEEN THE FREEZING AND BOILING-POINTS. 
207 
was by the glass side tube in the inflow end. The egress was by a glass-tube placed 
through a bole in the rubber stopper closing the outflow end. 
The calorimeter was held in place in the water-jacket by heavy rubber caps specially 
made to withstand hot water. On the inflow end of the calorimeter the cap was 
placed at the end so as to include the whole length of tube in the water-jacket. On 
the outflow end the cap was drawn up to the vacuum-jacket. Side tubes, cemented 
into the rubber caps, served to hold them on the calorimeter. The calorimeter was 
shoved lengthwise through the water-jacket and the caps sprung into place over 
the ends. 
The length of outflow tube protruding from the jacket was heavily lagged with 
flannel strips wound round it. As the outflowing water was made to flow the 
complete length of the outflow tube over the wires leading in the electric current, 
and as the tube itself was well protected from outside influences, the loss of heat from 
the water in the outflow tube was made as small as possible. This was shown very 
effectively by withdrawing the outflow thermometer, when the water was heated 
through about 8°, and determining the temperature at different points down 
the tube. 
The glass-tube placed through the stopper closing the outflow end of the calori¬ 
meter was connected with a short rubber tube to the coil of tubing in the water 
cooler, which in turn was connected in a similar way to the switch-over device. 
Water-jacket and Circulating System ivitli Electro-thermal Regulator .—The 
water-jacket was an oval tube of g^-inch copjDer, 2 ft. 9 in. long, with two lateral 
tubes 1 inch from each end on the under side. The jacket was 6 centims. wide and 
8 centims. fligh- On the other narrow end of the oval two other lateral ojjenings 
were made, one in the middle for a thermometer to obtain the temperature of the 
jacket water, and the other, which could be closed or opened at will, for an exit for 
accumulated air from the circulating water. The water in the jacket was circulated 
by means of a centrifugal pump run by, a water-motor attached to the high-pressure 
mains in the laboratory. The water was drawn from the bottom of a large 10-gallon 
copper tank through the jacket to the pump, when it was thrown back again into 
the top of the tank. The whole system of circulating tubes formed a chain round 
which the water was constantly circulating. No water either left or entered the 
system, except that lost by evaporation, and that was exceedingly small except for 
the higher points. The circulating tubes were about 4 centims. in diameter, and the 
pumping was sufficient to supply a solid column of water from the tubes into the 
tank. The tank as well as the water-jacket and circulating tubes were all heavily 
lagged. For the higher temperatures a device was fixed to the tank to make up for 
the evaporation from the hot water, and to keep it always at the same level. This 
was most important, to prevent the exposure of part of the bulb of the thermo¬ 
regulator by the lowering of the water level in the tank. 
In order to maintain the jacket at a constant temperature, a thermo-regulator was 
