54,6 



Mr F. B. Young on the Field at 



of the needle A in which the proof-plane C shared as though 

 an integral part of the needle, being in metallic connexion 

 with it through the key K. The screen having reached the 

 required potential, the keys M and H were closed in rapid 

 succession. When the screen was discharged the induced 

 charge on the needle was released ; the greater part of that 

 charge returned to earth, but the portion residing on the 

 proof-plane, finding its way of escape cut off by the opening 

 of K, distributed itself over the proof-plane, wire and 

 quadrants, and produced a throw of the electrometer-needle. 

 A similar throw was produced in the opposite direction by 

 charging the screen negatively instead of positively, and the 

 average of these and two similar throws was taken. The 

 sphere was next rotated to bring the proof -plane successively 

 to various angular distances from its central position, and at 

 each position the above operations were repeated. In this 

 way was obtained a series of mean deflexions which, provided 

 the capacity of the wire E and the potential of the electro- 

 meter-needle remained constant, were proportional to the 

 density of the charge upon the proof-plane in its successive 

 positions. 



To correct for any possible want of symmetry in the needle 

 a second series of observations was made, in which the sphere 

 was rotated in such a direction as to lower the proof-plane on 

 the opposite side, whilst to guard against a possible distortion 

 of the field by surrounding objects a third and fourth series 

 were made for which the proof-plane revolved in a different 

 vertical plane. As a further precaution the • sign of the 

 charge on the electrometer-needle was reversed for two of the 



Table I. 



9. 



(7. 



A. 



B. 



C. 



D. j 

 1 



Mean. 





















1-0 



•262 



•995 



•993 



•995 



•988 



•993 



•524 



•967 



•971 



•978 



•968 



•973 



•785 



•937 



•933 



•933 



•934 



•935 



1-047 



•880 



•885 



•883 



•884 



•883 



1-309 



•804 



•807 



•805 



•818 



•809 



four series. The complete set of observations is given in 

 Table I.; the consistency of the series seems to show that there 



