INTRODUCTION. 



THE METHOD. 



The method of measuring the conductivity of the solutions, employed throughout 

 this work, was essentially that of Kohlrausch. The bridge used in most of the work 

 was the latest improved form made by Leeds and Northrup, consisting of a manga- 

 nine wire between 4 and 5 meters long, wound around a marble cylinder. The wire 

 was calibrated by the method of Strouhal and Barus.* 



The resistance coils were standardized against a rheostat which had been corrected 

 by the United States Bureau of Standards. A number of forms of telephone receivers 

 were tried, and finally a sensitive form furnished by Leeds and Northrup was adopted. 

 The very satisfactory inductoria were also made by Leeds and Northrup. 



Three separate readings on the bridge were made for each solution at each temp- 

 erature, different resistances being, of course, used for each reading. The average 

 of the conductivities obtained by these measurements, which differed only slightly 

 from one another, was taken as the true conductivity of the solution. The meas- 

 uring flasks and burettes used in this work were generally calibrated by the method 

 of Morse and Blalock.f For the work from to 35 the measuring apparatus was 

 all calibrated at 20, and the results at lower and higher temperatures multiplied by 

 the proper factor. For the work from 35 to 65 the measuring apparatus was 

 usually calibrated at 50, and the proper correction inserted into the results at the 

 lower and higher temperatures. 



The conductivities are all expressed in terms of potassium chloride solutions 

 which were used for standardizing the cells. 



CONDUCTIVITY CELLS. 



The form of cell used in this work is shown in fig. 1. The glass tubes carrying the 

 electrodes are sealed firmly into the tops and bottoms of the ground-glass stoppers, 

 and these tubes are sealed down tightly on to the platinum plates serving as elec- 

 trodes. The plates are thus held firmly in position, and the distance apart is fixed 

 for any given cell. 



In making a series of readings at any given temperature, as many cells were 

 used as there were solutions of different concentrations of the salt in question to 

 be measured. Eight such cells constituted a set, and the distances between the 

 plates and the sizes of the plates were adapted to the concentrations to be studied. 



The conductivity of the water was determined in a cell especially constructed for 

 this purpose. It consisted of two concentric platinum cylinders, about 1 mm. apart 

 and 6 cm. long, shown in fig. 2. Glass tubes carrying platinum wires were sealed 

 down on to the tops of these cylinders. These glass tubes were firmly sealed into 

 the top and bottom of the ground-glass stopper. 



: Wied. Ann., 10, 326 (1880). fAmer. Chem. Journ., 16, 479 (1894). 



