14 WORK OF C. M. STINE. 



into both the upper and lower walls of the stopper. For the more concentrated 

 solutions, cells of the type used by Jones and Getman 1 were employed. The elec- 

 trodes were treated in the usual manner (coated with platinum black, etc.), in order 

 to obtain a sharper tone minimum or point of silence in the readings. 



The cells for the more dilute solutions were standardized by determining the cell 

 constant by means of N/50 potassium chloride. The value fc = 0.0015133 was 

 taken as the conductivity of N/50 potassium chloride. For the more concentrated 

 solutions the cell constants were determined with a N/2 solution of potassium 

 chloride, for which A; = 0.03365 at 0. 2 The potassium chloride employed in deter- 

 mining cell constants was recrystallized five times, the last two times from conduc- 

 tivity water, and was free from all appreciable impurities. The cell constants were 

 frequently redetermined, but very little change in them was noted from time to time. 

 The cells were filled with distilled water when not in use, and were always carefully 

 cleansed, dried, and rinsed with the new solution when a change in solutions was 

 made. All conductivities were measured at C. The temperature of the zero bath 

 was always within a tenth of a degree of C, as determined by an accurately stand- 

 ardized thermometer. All conductivity measurements are the average of four 

 readings with different resistances in circuit. 



The water used in this work was purified as in former work. 



It is well known that the conductivity method is not an accurate measure of 

 dissociation in concentrated solutions. Especially is this the case at the concentra- 

 tions with which we dealt, since the values of {!<*>, which we used in our mixtures,, 

 were determined for each salt by diluting the single solution to a maximum value of 

 fj,,,. This is, obviously, a source of additional error, for the conductivity of the ions 

 resulting from a completely dissociated molecule in a very dilute solution must be 

 different from the conductivity of those ions in a more concentrated solution. In 

 the more dilute solution a larger atmosphere of the solvent is attached to the ions, 

 and this must be dragged along by them. There is present a much smaller number 

 of particles of salt (ions or molecules) in the dilute than in the more concentrated 

 solution. Consequently, the friction between the ions in their movements must be 

 different. The viscosity of the solutions in which fi x can be determined is very 

 much less than the viscosity of the solutions at the concentrations employed. These 

 are all important factors which affect the conductivity of any dissolved substance. 

 The method is, however, an approximate one for measuring such dissociations, and 

 is the best available at present. All that can be said is that it probably gives values 

 of the right order of magnitude. 



CALCIUM CHLORIDE. 



A strong solution of calcium chloride was prepared and standardized. All solu- 

 tions were made by diluting the calculated amount of the original solution to the 

 required volume. Table 2 gives the data as to freezing-point measurements, con- 

 ductivity measurements, weight-normal corrections (or number of grams of solvent 

 in 1,000 c.c. of solution), and hydrates. 



Zeit. phys. Chem., 46, 254 (1903). 



'These values are calculated from the results of Jones and West. See Amer. Chem. Journ, 34, 381 (1905). 



