130 



WORK OF E. G. MAHIN. 



attributed to the non-applicability of the dilution law of Ostwald to solutions of 

 cadmium iodide; or may be explained by the assumption of molecular association 

 and simple ionization, or of association accompanied by the formation of complex 

 ions, the latter being the most probable cause. 



The result of the work upon the molecular conductivity of cadmium iodide in 

 acetone shows a close analogy to the case of lithium nitrate. There is almost no 

 change in molecular conductivity with dilution, until a concentration of 0.0002 

 normal is reached. From this point the increase is quite rapid, but it has not been 

 found possible to reach the maximum molecular conductivity, though this point 

 has been, we believe, closely approached. The Ostwald constant, calculated for 

 these solutions, shows a very rapid decrease until V = 5,000, where it is constant 

 until V = 50,000. From this point it increases quite rapidly. The large initial 

 value of K, with its subsequent decrease, is probably due to association of the salt. 

 The later rise in the value of K is, no doubt, due to the fact that dissociation is not 

 complete at 7 = 400,000, so that we have used too small a value for jU M in the cal- 

 culation of the dissociation at greater concentrations. 



Table 100. Ionization Constant for Cadmium Iodide in Acetone. 



It is thus evident that in acetone the dissociation is not complete at either tem- 

 perature, although it is more nearly so at than at 25. This is to be expected 

 from a knowledge of the temperature coefficient, which is negative below a dilution 

 of 10,000 liters and has a very small positive value at higher dilutions. Since the 

 change in conductivity, corresponding to a given rise in temperature, is the resultant 

 of two factors which are probably opposite in their effect, i. e., decrease in viscosity 

 and probable decrease in ionization, it is seen that there must be a very large 

 decrease in the ionization of cadmium iodide in acetone as we pass from to 25. 

 Thus, the molecular conductivity curve for cadmium iodide in acetone-water mix- 

 tures at suggests the fluidity curve for the solvents, more plainly than at 25, and 

 it seems probable that, could the conductivity be accurately measured at dilutions 

 as high as 600,000 or 700,000 liters, the curves of molecular conductivity would 

 become similar to those of fluidity, so that molecular conductivity and viscosity 

 would be inversely proportional, as is the case with lithium nitrate and most other 

 electrolytes. There is no means of knowing what the numerical value of the product 

 of viscosity and conductivity would be. Taking the highest conductivity that was 

 obtained in each case, the product is irregular, as is seen in table 101. 



