52 Dr. J. A. Harker on 



centration of this solution by escape of steam. Leaving 

 the salt-bath, the steam now passed through a wider 

 vertical tube in which was placed a second thermometer. 

 At the lower end of this tube was a large three-way tap, 

 which made communication either with an ordinary metal 

 Liebig's condenser or the calorimeter. The inclined tube 

 leading to the calorimeter was provided near its centre 

 with a union joint, which permitted the parts beyond to 

 be detached when the condensing-worm had to be 

 removed for weighing. 



The thermometer by which the temperature of the 

 steam leaving the salt-bath was measured was a very 

 small one of Jena glass, by Geissler, with milk glass scale 

 reading from 50 — 105 , and passed through a piece of 

 sheet rubber, leaving only one or two degrees of the scale 

 projecting through it. It could easily be read to "05 

 degree, and its boiling point was controlled during the 

 experiments. 



The water for the calorimeter was measured in a cali- 

 brated copper vessel with glass neck, holding approximately 

 4,900 grms. up to a certain mark. Its volume was 

 carefully determined in three different ways, and a table 

 constructed of its capacity at different temperatures. The 

 condenser was weighed before and after each experiment 

 on an Oertling balance to milligrammes, it having been 

 found that the grease used on its flange in making a 

 water-tight joint could easily be wiped off to within this 

 amount without any special precautions. 



Its weight when empty was about 250 grms. In 

 the three first experiments of the series the temperature 

 of the steam as it entered the calorimeter was measured 

 by means of a nickel-copper therm oj unction placed in a 

 narrow tube let into the stopper of the inclined piece of 

 ebonite fastened to the calorimeter. The other junction 

 was placed in a similar tube surrounded by steam at ioo°, 



