854: Mr. E. W. B. Gill: Comparison of - 



derived from it should be different for different negative 

 electrodes. 



As all the original values of o had been obtained bj the 

 use of zinc plates, a preliminary experiment was made using 

 a copper negative electrode. 



The same method was used for finding a as in all the 

 previous determinations when ultra-violet light was used as 

 the source of ionization (see Townsend, ' Electricity in Gases,' 

 p. 274), the only difference being in the arrangements to 

 produce the ultra-violet light. It was hoped that more 

 constancy could be obtained by running the spark-gap 

 not off an induction coil, as previously, but off a trans- 

 former whose primary current was supplied by a rotary 

 converter. This certainly effected some improvement, but 

 not perhaps as much as had been hoped. The gas used was 

 air which had been dried for some time over phosphorus 

 pentoxide. The results obtained were : 



100 200 300 



P 



Copper electrode. — '69 27 4"4 



P 



Zinc electrode ± 72 2*6 44 



The old values of - found with the zinc are put in the 

 P 

 third column, and it will be seen that the new and old 

 results are the same within errors of experiment, though, as 

 will be seen later, copper differs widely photo-electrically 

 from zinc. 



4, A final test was then made with a specially constructed 

 apparatus. 



A parallel plate apparatus was constructed in which as 

 usual the lower plate had a hole cut in it to admit the light 

 to the top plate, the field being kept practically uniform by 

 a thin wire grid covering the hole and level with the top of 

 the lower plate. The hole was some little way from the 

 centre of the plate. The top plate could turn about a 

 vertical axis and was composed of three sectors — copper, 

 zinc, and iron ; this was very accurately turned in the 

 lathe. When set up with the usual split insulation, it 

 was possible to bring each sector in turn over the hole 

 in the lower plate and thus allow the light from the same 



