for obtaining a Constant Temperature. 91 



ice to be easily packed under the bottom of C. If an ordi- 

 nary double bottom be used it will be found difficult to keep 

 ice there, and the temperature of the water is liable to rise 

 to 4° C. 



Within C is placed another copper vessel H, containing 

 whatever is to be heated to constant temperature. In this 

 experiment I used a large glass test-tube wrapped with 

 No. 22 German-silver wire. Some cork feet were waxed on 

 to the bottom of H, and the test-tube was suspended in H 

 by means of a cork top. Cork in the figure is indicated by 

 diagonal shading. 



A little raw cotton placed about the test-tube in H will 

 stop convection-currents and much improve the steadiness of 

 the temperature. 



On is next placed a metallic cover through which the 

 terminals of the wire wrapping are led out and connected to 

 the battery. This metal cap is covered with crushed ice, and 

 the whole inside region is thus surrounded by a wall at 

 zero. 



The test-tube was filled with water whose temperature was 

 measured with a sensitive Baudin thermometer on which the 

 degrees (Centigrade) were 10 millim. long ; so that with a 

 telescope there was no difficulty in reading to t J-q of a 

 degree. 



In series with the storage battery (in this case 12 cells) is 

 placed a loop, r, of German -silver wire on which are strung- 

 two wire connectors. A copper shunt, s, soldered across 

 these two connectors, as in the figure, makes of the whole a 

 simple and excellent rheostat by means of which changes in 

 the E.M.F. of the batteries, or temperature changes in the 

 resistance of any part of the circuit, may be compensated. 



A duplicate slider, parallel to s, should be added in order 

 that one may be clamped while the other is moved, thus never 

 breaking the current. 



A simple rheostat, devised by Mr. C. H. Bedall, and oc- 

 cupying much smaller space, is made by ploughing a number 

 of longitudinal grooves in a piece of ^ stuff and connecting 

 by U-shaped copper connectors. These grooves are then 

 filled with mercury, and the resistance varied by sliding the 

 connectors. 



As a source of constant current I have used Julien storage- 

 cells, and find that they leave little to be desired. 



One does not need a large current even for the production 

 of comparatively high temperatures; for the final tempera- 

 ture of the enclosed space increases as well with the thermal 

 conductivity of the packing between the hot and cold walls 



