18 Mr. E. H. Griffiths and Miss Marshall on the 



potential-difference of 3 Clark cells 



'd6{ 



\ dt / 3eg ' 



/a%\ 



Hence ^ dt ^ (A) 



(^l\ -( d ll\ 

 \ dt kes \ dt ), 



gives the ratio of the heat supplied by the stirring to the 

 heat-supply when the potential-difference was that of 3 Clark 

 cells ; and as the latter can be calculated if 1^ is known, the 

 value of Q s can be obtained without any assumptions as to 

 the thermometric scale, the capacity for heat of the calori- 

 meter, &c. 



The individual experiments were in close agreement with 

 each other and give the following results : — 



At 50°, (-rr 1 ) ='14533 expressed in mm. of bridge-wire 

 \ at he* wnere rate of s ti rr i n g =5'380, 



r^l \ =-001155, at rate 5'380. 



Hence Q s (at rate 5*380) ='003404 ; 



but (see Paper W, p. 330) £ =a constant = 



Q, s -003404- 



hence Q s =-003206 at rate 5'300. 



A second determination gave 



Q s =-003202 at rate 5-300. 

 A third gave 



Q B = '003205 at rate 5'300. 

 We therefore assume 



Q s = -00321 + {r* - 789) x '0000041, 

 as sufficiently accurate. 



