AQUEOUS SOLUTIONS.—ARCHIBALD. 297 
tube, about 35 em. long and 4 em. in diameter, bent so as to 
form three sides of a square. Two vanes fixed at an angle of 
45° near the bottom of the bath, to a vertical axis, which was 
turned by a small hydraulic motor, kept the water of the bath 
well stirred. The thermometer used was graduated to fiftieths 
of a degree, and could easily be read to hundredths. Its readings 
were compared with those of another, whose errors had recently 
been determined to hundredths of a degree at the Physikalisch- 
Technische Reichsanstalt, Berlin. With this apparatus the 
temperature of the bath could be kept constant to within a 
fiftieth of a degree, for half an hour at atime. A variation of 
one-fiftieth of a degree might cause an error of 0.05 per cent in 
the determination of the resistance. 
That one might be sure that the temperature of the solution 
to be measured had come to be that of the bath, two or more 
determinations of the resistance were always made at intervals 
of about five minutes, and that reading taken which was found 
to be the same for successive intervals. 
Data for the Calculations. 
For the simple solutions the ionization coefficient (@) was 
taken to be equal to the ratio of the specific molecular conduc- 
tivity to the specific molecular conductivity at infinite dilution. 
Kohlrausch’s values for the specific molecular conductivity at 
infinite dilution were used. They were taken to be 1280 x 10°° 
and 1060 x 10-*for Potassium and Sodium Sulphate respectively, 
as determined by him.* 
The value of p in the above formula was found by density 
measurements before and after mixing. These measurements 
were carried out with Ostwald’s form of Sprengel’s pyknometer. 
Measurements, accurate to one in the fourth place of decimals, 
which was beyond the degree of accuracy required, could be 
made without much difficulty. The value of p was found to be 
practically equal to unity for the most concentrated solutions 
examined. 
* Wied. Ann., Vol. xxXvlI., p. 204. 
