54 Dr. J. A. Harker on 



numbers, 640*9, expresses the total heat necessary to raise 

 one gramme of water from o° to ioo° and convert it into 

 steam at that temperature under a pressure of 760 mm. of 

 mercury, the unit of heat being the amount necessary to 

 raise one gramme of water through one degree of the 

 mercury scale of the French hard-glass thermometer in 

 the neighbourhood of 17 , the mean temperature of the 

 calorimeter in my experiments. 



Assuming with Regnault that 100*5 of such units are 

 required to heat unit mass of water from o° — ioo°, we have 

 then from these experiments 540*4, a number differing but 

 little from the usually accepted number of Regnault, the 

 difference being in the opposite direction to that given in 

 the preliminary experiments. As it is quite possible that a 

 trace of the water which was always found condensed 

 in the ebonite "leading-in" tube ought to have entered 

 the calorimeter, thereby increasing the weight of the 

 water found and lowering L, this small difference might in 

 this way be easily accounted for. 



During the progress of my experiments, Mr. E. H. 

 Griffiths* has been conducting an elaborate investiga- 

 tion on the latent heat of steam at temperatures below the 

 boiling point. From the values he obtains from experi- 

 ments at 30 and 40 C, Mr. Griffiths calculates by 

 extrapolation that at ioo° the latent heat would be 536*6 

 mean calories, assuming from Regnault's experiments that 

 L is a linear function of the temperature. This result is 

 a remarkable confirmation of the accepted value of 

 Regnault obtained nearly fifty years ago. It is to be 

 hoped that Mr. Griffiths will extend his observations over 

 the wider range he originally intended, io° — 6o°, since 

 there seems to be some doubt as to the accuracy of 

 Regnault's values for L at temperatures other than ioo°. 



* Phil. Trans., Vol. 186, 1895 A, pp. 261-341. 



