186 PROCEEDINGS OF THE AMERICAN ACADEMY. 



were found upon careful qualitative analysis, but here as in the other 

 cases a trace of sulphuric acid was detected. Two analyses gave 

 identical results, showing that the solution contained 1.307 percent 

 of solid hydroxide, thus possessing the concentration LiOH, IOO.4H2O. 



It will be noted that in some cases, particularly in the case of lithiuna, 

 the solution was not exactly of the concentration MX 100 H2O. 

 A moment's consideration will show, however, that the specific heat 

 of such a solution is different by only a negligible amount from one of 

 the exact strength. For example, the specific heat of NaOH 100 H2O 

 is 0.9664 and that of NaOH 200 HoO is 0.9827. Evidently then 

 the addition of a single H2O to the 100 HoO already present will make 

 a difference of less than 2 in the fourth decimal place, and O.-l H2O 

 will cause a difference of less than 1 in that place, — an amount beyond 

 the range of accuracy of the experiments. 



In making the calculations the following international atomic weights 

 have been assumed O = 16.000, H = l.OOS, CI = 35.458, Br = 79.916, 

 I = 126.99, N = 14.008, K = 39.09, Na = 23.00, Ag = 107.88. 



The Execution of the Determinations. 



Having thus discussed briefly the details of the apparatus and of 

 the purification of the substances employed we may turn to the execu- 

 tion of an experiment. In order to follow the description reference 

 to the figure on page 175 is necessary. 



The "submarine" vessel was first secured in position with its lid ofl", 

 the outer bath was filled almost to the rim of the vessel, and the 

 temperature of its crude dilute alkali was brought to the point at 

 which the determination was to start. The liquid whose specific 

 heat was to be determined was then placed in the platinum calori- 

 meter, brought to the desired temperature, rapidly weighed to the 

 nearest centigram, and placed in position within the "submarine." 



The cover of the "submarine," temporarily suspended from a 

 stand in the neighborhood of the calorimeter, had been already fitted 

 with its various attachments, ready to be clamped into place. First, 

 the open cap (E) of the platinum bottle had been arranged in such a 

 way that the rest of the apparatus could be attached below it. Then 

 the receptacle for the alkali, (F) made water-tight by means of a rubber 

 stopper (r) placed in the hole in the lower part, was fixed in place. 

 After the calorimeter had been prepared as described above, the 

 required portion of caustic soda solution was measured into its recep- 



