ON THE BOILING POINTS OF AQUEOUS SOLUTIONS OF ELECTROLYTES. 199 

 The following are the corrected readings : — 



By barometer, 

 ,, thermometer, 



,, barometer, . 

 ,, thermometer, 



,, barometer, . 

 ,, thermometer, 



2-320 

 2-320 



3-179 

 3-179 



4-866 

 4-868 



2-435 

 2-435 



3-631 

 3-629 



4-876 



4-874 



2-525 

 2527 



4-036 

 4-034 



4-892 

 4-890 



2-888 

 2-888 



4-405 

 4-403 



These figures indicate only a slight difference between barometric and thermal 

 corrections for change in atmospheric pressure, which could not account for the large 

 values of the elevation constant which have been obtained, but might account for small 

 variations in its value for dilute solutions. 



(2) To find out if the high values were due to overheating, I took two courses. 

 First I determined the boiling point of the same solution several times, and found 

 the corresponding elevation constant to have, in all cases, approximately the same 

 large value. Had overheating been the cause of the large values of the constant 

 when compared with theory, one would not expect to be able to repeat a high value, 

 much less to repeat it several times, as the amount of overheating would be likely 

 to vary considerably. Secondly, in the case of each of several salts I made duplicate 

 series of observations of the elevation of the boiling point for solutions of different 

 concentrations, and plotted curves with grammes of salt added as ordinates and elevation 

 of boiling point as abscissge. The observations are given in a former paper,* but the 

 curves obtained are given below (figs. 2 to 4). In the case of each salt the duplicate 

 series gave different values for the elevation, and the curves are consequently not 

 coincident ; but whilst not coincident, they are smooth and parallel. If the values 

 of the elevation were vitiated by overheating, the amount of overheating in one 

 series must have exceeded that in the other by a constant amount — a quite improb- 

 able assumption. 



In connection with the above it may be interesting to note what Biltz t and Luther j 

 say. Biltz : " For the attainment of exact temperature adjustment it is necessary 

 to maintain an extremely energetic boiling.'' Luther, § quoted by Biltz: "Finds it 

 indispensable to maintain an energetic, and indeed stormy boiling, for the better 

 adjustment of temperature." Obviously, therefore, neither of these observers con- 

 sidered overheating as the cause of the discrepancy between the observed and the 

 theoretical values of the boiling-point constant. 



It would therefore seem that the high values obtained for the elevation constant are 

 not due either to error in correcting for change of atmospheric pressure or to overheating. 



The curves given above not only render the supposition of overheating improbable, 

 but also indicate the cause of the large and variable values of the elevation constant 



* Proc. Roy. Soc. Edin., 25, 960, 1905. 

 t Zeit.furphys. Chemie, 40, 185 (1902). 



Ibid. 



§ Ibid. 



S 



