84 Mr. J. H. Grindley. A?i ExpeririURtal Investigation of 



specific heat exists it will be very small compared with the variation 

 Avith temperature, such examination indicating that the value of the 

 specific heat is sensibly independent of the pressure. 



The law of cooling followed by the wiredrawn steam is slightly 

 different from that obtaining in many other gases, \az., that the fall of 

 temperature varies directly as the difference of pressm^e. The rate of 

 cooling was foimd to diminish with increase of initial temperature. 



The curves showing the pressure-temperature relations of the super- 

 heated steam wiredra^vn from definite initial pressures, seem to follow 

 for a short distance the law of boiling points, and the experiments show 

 that this coincidence always exists in saturated steam, and may well 

 be mistaken for evidence of w^etness in the steam. 



Tables showing the fall of temperature with pressure in the ^vire- 

 drawn steam, of the total heat of the steam under certain pressures 

 and temperatures, and of the mean value of the specific heat at con- 

 stant pressure of superheated steam at definite pressures and between 

 definite temperatures, accompany the paper. 



Part III. 



In this portion of the paper the two properties of steam deduced 

 directly from the experimental figures, viz., the specific heat and 

 the cooling effect Wjhp or c, are more directly considered. In the first 

 place, the cooling efi'ect c is found to be inversely proportional to r-^-^, 

 where r is the absolute temperature. 



It is then shown that the following formula 



f (K^)- -^(.K,.) 



is capable of strict proof from thermodynamical principles, the inter- 

 pretation of the formula being that the variation of K^, with the 

 pressure at constant temperature is equal to the variation of the pro- 

 duct cKp AAdth the absolute temperature at constant pressure, but of 

 opposite sign. 



Applying this to steam when superheated, it has been shown in 



I^u't II of the paper that the variation ^(K^^) is zero to the degree 



op 



of accuracy to which the experiments have been taken. It follows, 



therefore, from the above formula, that the variation g- (cK^) should 



equal zero ; hence, the values of the product cKp have been tabulated 

 for different pressures and temperatures, and so far as the results go, 

 it is clearly shown that the product cKp is an absolute constant, which 



means that the variations ^ {cKp) and (cK^) ^"^^ both zero. 



