TRANSACTIONS OF SECTION A. 



533 



Compared with the Work of other Observers. 



The vfilue of the specific heat obtained by Regnault,' and so universally adop ed, 

 i.e., -03332, is found to be in agreement with our determination at 13°'3, and very 

 close to the value at the ordinary temperature of laboratory work. The value was 

 obtained over a range 10"^ to 100" in terms of a thermal unit at 10°, and is con- 

 siderably in excess of our value for tlie same mean temperature. E.Kperimenls to 

 determine the variation with temperature indicated an increase in specitic heat 

 with increasing temperature. 



iOO' 



© fibservatioiis of Barnes and Cooke, 19U2. 



X CihscrvatioMs of Callcndar and Barnes, 1807. 



\ V FcirnnUa proposed by Barnes and Cooke. 



W Fornnila proposeil by Winkelmaun. 



JI Formula proposed by Jlllthaler. 



Winkelmann- carried out an extensive series of experiments by the method of 

 mixture, to determine the temperature coefficient, and was the first to show that 

 the specific heat decreased as the temperature increased. This result being so ex- 

 ceptional at the time, and so contrary to the results for other liquids, the author 

 was inclined to regard it as doubtful until he finally verified the first experiments 

 by a second series, in which he eliminated certain sources of error. Two ranges of 

 temperature were taken, the first between 20" and 50° C, and the second between 

 26° and 142° C. The average variation from these measurements was found to 



' Poffg. A7m., vol. 11. p. 44 (1840), and vol. li. p. 238 (1840). 

 ' Ibid., vol. clix. p. 152 (1876). 



