318 Refractive Indices, &c, } of Solutions of Hydrogen Chloride, &c. 



and potassium, and also with water. In both the rubidium solutions 

 a mould had developed, and therefore no observations were made with 

 these two solutions. 



The observations were made in the same way as those already 

 described, except that (1) the gas regulator was removed from the 

 water bath containing the coil, and the temperature was allowed to 

 rise, and (2) all the measurements were made with the prism in one 

 position. 



The temperature of the prism rose slowly, about 2 in an hour ; the 

 prism was not reversed in order to avoid exposing its surfaces to the 

 cooling action of the air of the room. The object of the experiments 

 being to ascertain the relative values of the index of a solution at 

 different temperatures, and not the absolute value at any particular 

 temperature, the result would not be affected by any small error due to 

 the observations having been made with the prism in one position of 

 minimum deviation and in one position only. 



The values obtained were plotted, one unit in the fifth place being 

 represented by 1 mm., and the rate of change, the " sensitiveness," 

 ascertained from the plotting. 



The results are given in the table : in the second and fifth columns 

 the temperatures between which the observations were made, in the 

 third and sixth the number of observations, and in the fourth and 

 seventh the decrease of the index for an increase of 1. 



The table shows that between 10 and 20 the " sensitiveness " of 

 the normal solutions is greater than that of the semi-normal solutions ; 

 and further, that it increases with the increase of the refractive 

 index of the solution. 



The value for water between 10 and 20 calculated from the values 

 given by Dufet* is 0-000071, or slightly less than that obtained in 

 these experiments. December 31, 1898.] 



* Loc. cit. 



