

Heat by the Method of Mixtures. 277 



where 



ws ° 



In this equation all the quantities are known, at least approxi- 

 mately, or may be observed or calculated except K, B, and T . 

 The values of the constants K and B for metal plates in con- 

 tact with water are, according to Peclet (Kankine, p. 260), 



K = -00119 

 B — -1044 



I have not been able to find the data upon which these values 

 are based. They will of course vary with the condition and 

 nature of the cooling surface and will be larger when the 

 water is well stirred than when it is not. The best way is to 

 determine them, for the given body and given calorimeter, by a 

 series of preliminary observations on a? , the time required for 

 6 X to attain its maximum value, 0. In making these observa- 

 tions, t should be made about f(#— t) lower than /3, the tem- 

 perature of the surrounding air. Three such preliminary 

 observations (made in the same way as they would ordinarily 

 be made to determine s) would suffice (though a larger number 

 reduced by the method of Least Squares would be better) to 

 determine K, B, and T , this last quantity as determined from 

 (21) being practically the same (for such rapid cooling as 

 always takes place in a properly conducted calori metric experi- 

 ment) as that which would be determined from the theoreti- 

 cally correct but far more complicated expression (17). Hav- 

 ing thus found T , we may then find a first approximation to 

 the value of /3 from (13) ; A, the radiation factor of the calori- 

 meter being taken as given by Jamin, or better determined for 

 each calorimeter by experiment. 



^Yith this value of /3 a second set of observations are taken 

 for x and with this new value a second and closer value of /3 

 determined as before. Two such successive approximations 

 will suffice to determine /3 with the desired accuracy. 



The use of this method renders any precautions for prevent- 

 ing radiation useless. All that is necessary is to keep the tem- 

 perature of the surrounding air uniform, and at the determined 

 temperature (practically this is best done by adjusting the 

 initial temperature t of the calorimeter contents), and keep the 

 water in the calorimeter well agitated during the progress of 

 the experiment, while screening the calorimeter from direct 

 radiation from the observer's body. Then the maximum tem- 

 perature attained will be the theoretical maximum, 6, desired, 

 and radiation, instead of being a disadvantage, will become an 

 advantage inasmuch as it lessens the time of the experiment 



