224 Mr. F. B. Pidduck on a Method of Measuring 



where R is calculated at the frequency used in the experi- 

 ments. In designing the high frequency circuit we should 

 have comparatively few turns in the deflecting coil, in order 

 to reduce self-capacity, which causes inequality of current 

 in the various parts. It is nevertheless easy to obtain suffi- 

 cient deflexion, and it was found convenient to work with a 

 DuddelPs thermo-ammeter at a current of about 0*05 ampere. 

 The circuit is shown in the figure. The deflecting coils DD 



Fio-. 1. 



formed an approximate Helmholtz pair, each having 50 turns 

 of cotton-covered S.W.G. 20 wound on a hard wooden core 

 of 13 cm. diameter, the mean outside diameter being 

 13'8I cm. and the separation between centres 7*12 cm. 

 A cardboard cylinder C was pressed tightly between the 

 coils, and was cut for the insertion of thin mica windows. 

 The ring E, with small mirror attached, was hung by a 

 galvanometer suspension from an ebonite rod, which passed 

 through a plug fitting loosely into the cardboard cylinder. 

 The thermo-ammeter A was normally inserted between the 

 coils. Experience showed that the disturbing effect of 

 the generating coils GG might be considerable, and they 

 were consequently removed to a distance of 3 metres, with 

 a connexion of twisted flexible wire. The condenser K was 

 of 1 mmfd. maximum capacity, giving oscillations of wave- 

 length 3250 metres. The triode T was of the French type, 

 with a 4-volt heating battery and 200 volts on the anode. 

 Flexible leads were also brought from the heating circuit to 

 a key near the scale, by which the oscillations of the ring R 

 could be conveniently controlled. 



