Section /p. Errors in the Solutions and their Correction. 201 



no sulphates, 0.0006 per cent CI." The recrystallized salt was dissolved 

 in water of specific conductance 1.1 X 10 6 , and the concentration of the 

 solution was found, by evaporation with pure hydrochloric acid and gentle 

 ignition to constant weight, to be 112.2 milli-equivalents per kilogram.* 



79. ERRORS AFFECTING THE SOLUTIONS AND THEIR CORRECTION. 



The effect of carbon dioxide on the conductivity of ammonium hydrox- 

 ide solutions can be shown to be very large. Thus, the ammonium carbo- 

 nate formed by the addition of 0.01 per cent (in mols) of carbon dioxide 

 to a 0.1 molal ammonium hydroxide solution is not appreciably hydrolized, 

 on account of the large excess of ammonia present ; considering it there- 

 fore as being completely ionized, and taking the equivalent conductances 

 of NH 4 and CO s as 64 and 70 respectively, the increase in the specific 

 conductance of the 0.1 molal ammonia solution, caused by the addition 

 of the carbon dioxide, is found to be 2.7 X 10" 6 > or 0.9 per cent of that of 

 the ammonium hydroxide. That even such a small amount, which would 

 of course vary considerably, was not absorbed during the filling of the 

 bomb, is shown by the fact that successive determinations of the resistance 

 of the same solution agree at 18 within 0.1 per cent. 



The error due to carbon dioxide in the water used for making the solu- 

 tion or in the strong ammonia solution itself, is almost impossible 

 to determine. Water at 17 absorbs its own volume of carbon dioxide 

 at atmospheric pressure; ordinary air contains about 0.01 per cent C0 2 

 by volume, hence water in equilibrium with ordinary air will contain 

 17 X 10~ 6 mols of un-ionized H 2 C0 3 per liter. Using Walker'sf value 

 of 3040 X 10~ 10 for the ionization-constant of H 2 C0 3 into H + and 

 HC0 3 ~, and taking for the equivalent conductances of these ions 320 

 and 50, respectively, the specific conductance of this water should be 

 0.8 X 10" 6 . The specific conductance of the water actually used was 

 always less than 1.0 X 10~ 6 , usually less than 0.8 X 10" 6 ; but it is very 

 unlikely that this water was saturated, since it was condensed hot, and 

 afterward kept protected from the air. Hence the larger part of the 

 conductance found was probably due to organic bases which distil over 

 with the water, or to salts carried over mechanically by the current of 

 steam. The view that it is not due to carbonic acid is supported by the 

 fact that ammonium hydroxide solutions made from water varying in con- 

 ductance from 0.5 X 10" 6 to 0.8 X 10" 6 show, after subtracting the con- 

 ductance of the water, values for the equivalent conductance constant 



