50 



Mr. S. R. Milner and Prof. A. P. Chattock on 



yet to keep the heating currents through them the same as 

 those through their external portions the arrangement shown 

 in fig. 3 was adopted, the lettering in this figure corresponding 



Fig. 3. 



To Galv. 



with fig. 2. The heating current flows first through the coil 8 a, 

 then through a stretched wire A A' of very low resistance, and 

 then through the second heating-coil «' 8' in series. Parallel 

 with these are resistances d b, b a, a 1 b 1 , V d' of resistances ten 

 times greater than those of the corresponding portions of the 

 heating-coils. db &c. are of german-silver immersed in 

 oil. The points d, d' are connected to two stretched wires X 

 and Y of about 2 ohms each, on which good contact can be 

 made at any point with the main current by a screw. Leads 

 from /3 and b, /3' and V, the contact-maker D and a a', are 

 taken to a mercury commutator by which these pairs of points 

 may be connected either directly or through a galvanometer. 

 /3 and b are brought to the same potential by connecting them 

 through the galvanometer and altering the position of the 

 contact on X, similarly with /3' and V by altering Y. On 

 now short-circuiting /3 to b and /3' to V the currents will be 

 the same all along the line of conductors 8 8', and by trans- 

 ferring the galvanometer to D and a a! balance may be 

 obtained with respect to the four arms /3 a, a! 0, b o, u' V by 

 moving D. It is true that the balances of potential at /3 b 

 and fi 1 b' are destroyed by the alteration of the water thickness 

 m one of the vessels, and this must produce a difference in 

 ihe heat generated per sq. cm. in a /3 and ft 8 respectively. 

 In the actual readings, however, the effect of this on the value 

 of the conductivity was found to be an error of less than 

 01 per cent. ; so that it was sufficient to make the adjust- 

 ments on X and Y with an average value of the heating 

 current, and not necessarily to readjust during the experiment 

 when the water thickness was altered. 



