304 Mr. W. Sutherland on the 



But at high temperatures the high-pressure steam projected 

 the hot water with such violence into the calorimeter that it 

 was difficult to regulate the amount of hot water so that the 

 free surface of the water in the calorimeter should be within 

 the graduated neck ; Eegnault, therefore, had to modify his 

 method slightly. He admitted enough hot water not to raise 

 the surface into the graduated part of the neck ; but then 

 added enough of the cold water which he had withdrawn to 

 bring the level of the water amongst the graduations. This 

 small weight of water never exceeded 100 grammes; let it be 

 denoted by ir, then the equation becomes 



»(P-P +p-r).(T-0-(Po-i»+»)(*-0- 



Thus Eegnault conducted two series of experiments by 

 slightly different methods ; but he includes all his results in a 

 single table without giving any information as to which belongs 

 to which series. It is easy to see, however, that the second 

 method was used in precisely the thirteen experiments in which 

 the differences occur, namely, those numbered 3, 11, 12, 13, 27, 

 29, 34 to 40, the last seven corresponding to the highest tem- 

 peratures ; because, while in all the other experiments the quan- 

 tity of hot water ranges from 9904'4 gr. to 10075'9 gr., in these 

 it is, in all cases except one, more than 10100 gr. But accord- 

 ing to Eegnault' s description of his second method, he uses 

 less hot water in these experiments than in the others ; hence 

 we must admit that in copying the results of his second 

 method into their appropriate places in the table of results of 

 the first method, Eegnault must have transferred a wrong 

 series into the column headed " weight of hot water ;" pro- 

 bably he copied his values of p — it instead of P — P +^> — ir, 

 the true weight of hot water. At all events it will now be 

 shown that he must have used the correct series of numbers 

 in calculating the values of the specific heat. 



For if we assume that in these thirteen experiments the 

 level of the water in the calorimeter at t always stood at the 

 zero mark of the neck, we can estimate from Eegnault' s table 

 the quantity of water P minus the small quantity contained 

 in n divisions of the neck. Subtracting the weight of cold 

 water P —p + tt given by Eegnault, we get the weight of hot 

 water added minus the small quantity contained in n divisions 

 of the neck. If we substitute this for P — P +js— ir in the 

 above equation we shall get values of x a little too large ; if 

 then the values for the specific heat given in these thirteen 

 cases by Eegnault are a little smaller than our calculated 

 values, we can allow that he has calculated his values from the 



