462 



DE. K. B. DENISON AND DR. B. D. STEELE ON THE 



calculated from our specific mobilities (columns 3 and 5), whilst the agreement 

 between the found and calculated values is remarkably close in the majority of 

 instances. A few cases, where the differences are larger, call for special comment. 

 For example, the calculated and observed mobilities of the hydrogen ion in hydro- 

 chloric and nitric acid solutions agree very well indeed, whereas the agreement is by 

 no means good for the velocities of the anions of these acids. This is explained by 

 the fact that the velocity of the hydrogen ion is about five times as great as that of 

 the chlorine ion (or nitrate ion), and as a consequence the latter only moves about 

 10 millims., whilst the former moves through the whole length of the tube. The 

 result of this is that there is a very much larger percentage experimental error 

 introduced in the measurement of the mobility of the anion than in that of the cation 

 of an acid. The same applies, but in the opposite sense, to the measurement of the 

 mobilities of the anion and cation of an alkali. 



The observed values for /10 sulphuric acid are quite different from the calculated 

 values. Whether this is due to the inaccuracy of the transport number, or to the 

 occurrence of complexes in the system, it is not possible to say, but the cause of error 

 seems to have completely disappeared in the n/50 solutions. 



Interesting results are obtained when we compare the velocities of the same ion in 

 equimolecular solutions of different salts. KOHLRAUSCH'S law of the independent 

 wandering of the ious depends on the assumption that, in solutions sufficiently dilute 

 for any variation in " electric friction " to be neglected, the same ion has the same 

 velocity in whatever solution it occurs. If this is so, and remembering that it is only 

 the specific mobility, u, that is invariable, and that the actual mobility U = au, we 

 see at once that it is only in solutions of salts that are equally ionised that we can 

 expect to find the same actual mobility for the same ion, and, on the other hand, 

 where we do find this, we have strong evidence that the salts in solution are equally 

 ionised. 



On comparing the velocities of the potassium ion in equimolecular solutions of all 

 the potassium salts that we have measured (see Table V.), we find that this velocity 

 is identical in solutions of potassium, chloride, bromide, and iodide, but is a little 

 smaller in the other solutions. This identity of mobility is manifested not only in 



TABLE V. Velocity of K ion in various Salt Solutions. 



