. r >2<; PROFESSOR W. E. AYRTON, MR. T. MATHER AND MR. F. E. SMITH 



3 millims. distant from the bases of the cylinders and are practically earth-connected. 

 If the mean potentials of the suspended systems were +31 volts and 31 volts 

 respectively, relatively to the earth, then the total electrostatic effect should be nil. 

 The resistance of the whole circuit was 110 ohms, that of the balance 71 ohms, and 

 the E.M.F. of the battery was 110 volts. By earthing the battery between the 19th 

 and 20th cells, counting from that end directly connected to the balance coils, the 

 potentials were approximately as indicated and there was no measurable effect on 

 the balance when the current was reversed. Except for the first few observations 

 the battery was earthed at this point No error was introduced by the omission, as 

 the electrostatic effect occurs twice with opposite signs in the observations. 



Normal Procedure. In making a determination of the strength of a current, the 

 following scheme was adopted : 



(1) The commutators and plugs were set so that the current circulated through 

 the balance coils in the order : suspended left, fixed left, fixed right, and suspended 

 right, and so that the total force was the sum of the direct and secondary forces 

 (D + S) (see p. 508). Observations for the determination of the balancing mass were 

 then made and repeated when 



(2) the current through the fixed coils was reversed ; 



(3) the current through the whole of the balance circuit as typified by (2) was 

 reversed ; 



(4) the current through the fixed coils was reversed, that in the suspended coils 

 being as in (3) ; 



(5) the current through the whole of the balance circuit as typified by (4) was 

 reversed. 



Each of these arrangements as indicated by two letters, one denoting the position 

 of the main commutator M, fig. 23, and the other that of the commutators on the 

 multiple commutator and plug board P, as described on p. 522. After these obser- 

 vations a similar set was made when the direct and secondary forces opposed one 

 another, thus determining (D S). The order of making the observations in each set 

 was rigidly adhered to, but the (D S) observations sometimes preceded and sometimes 

 followed the (D + S) observations. 



After the first few determinations of E.M.F. had been made, the current which it 

 would be necessary to pass through the circuit to balance the cadmium cell was 

 estimated from the secular change in resistance and the temperatures of the coil 

 and cell ; the balancing mass was then calculated and the position of the rider 

 decided on, so as to give, together with the weights, the required mass. Previous to 

 observations of any kind being made, the circuit through the manganin coils was 

 completed for an hour or more, after which an examination of the steadiness of the 

 current was made by one of us, and observations of the sensitiveness of the balance 

 and stability of the rest-point of same were made by another. In accordance with 

 the scheme on p. 522, the multiple commutator was appropriately set, the balancing 



