Dowling — New Form of very High Resistance with Electrometers. 31 



veniently have a capacity of one -thousandth of a microfarad and be 

 subdivided into smaller fractions. 



The vibration of the spring S is controlled by an electromagnet, which is 

 excited by an intermittent current from a tuning-fork or other automatic 

 interrupting device. 



§ 5. The action of the apparatus is obvious. On opening the earthing 

 key K, the electrometer system (Ed) charges up to a steady potential V, such 

 that the intermittent current nc 2 V drawn off by the vibrating contact is equal 

 to the current i being measured, which passes into the electrometer system 

 from the ionisation vessel / (or other source of current) . This is, of course, 

 correct only if the potential to which c 2 is charged at each contact is 

 approximately the same as the electrometer potential. For this to be 

 the case, it is simply necessary that c 2 should be small compared with c,. 

 This condition is readily fulfilled in practice, since c 2 rarely exceeds 

 •001 microfarad. 



If the above condition is not fulfilled, we sacrifice the simplicity of the 

 relation between the electrometer deflection and the current. As against 

 this drawback we would have the advantage that the electrometer would 

 take up its deflections the more rapidly the smaller we made c x . On putting 

 this to a practical trial, however, in the case where c, and c 2 were equal, it 

 was found that another difficulty was met with. Owing to slight variations 

 in the times of contact of S with A and B respectively, the electrometer 

 deflection was unsteady. As it was not found easy to remedy this, further 

 trials of the apparatus in this way were abandoned. 



§ 6. The writer has tested the method over a considerable range, using it 

 to measure ionisation currents in gases drawn from a flame. Each measure- 

 ment was repeated immediately afterwards by the old " rate of charge " 

 method. In all cases a very good agreement was found, provided that the 

 condition mentioned in the previous paragraph was fulfilled. There was one 

 difficulty met with. The electrometer was somewhat unsteady at first. This 

 was found to be due to the faulty action of the contact-breaker ; but on 

 approaching the screws A and B very near together, so as to limit the 

 vibration of S to a very small amplitude, the trouble disappeared. 



The largest " equivalent resistance " tried was limited by the apparatus 

 used by the writer, namely : (1) the smallest condenser (c 2 ) ; (2) the slowest 

 interrupting device. The smallest condenser was about 10"* microfarad, and 

 the clockwork interrupter gave two interruptions per second. The equivalent 



resistance was, therefore, ~ — ^cfTo ohms, thai, is 5000 megohms. 



The lowest equivalent resistance tried was 10 7 ohms, or 10 megohms. 

 This was obtained by using a capacity c 2 = -002 microfarad and a tuning-fork 



