56 Dr. R. T. Beatty on 



same current, and the results are shown in Table I. Fig. 5 

 (PL II.) is plotted from the last three lines of Table I., and 

 shows the intensity ratios as functions of the pressure. The 

 ratio for any two lines remains constant till the pressure falls 

 below a given value, and then the lines with the higher 

 ordinal numbers in the Balmer series become relatively more 

 intense. The photo-currents due to H 5 were very small, and 

 the curve showing H 5 4 /H 7 is not very trustworthy. 



Table I. 



Water Vapour. 



p in mm. 





•4 



•6 



•8 



1-0 



1-2 



9-8 



1-4 



9-8 



1-6 



9-7 



1-8 

 9-6 



2 

 9-35 



3 

 76 



4 

 6-0 



6 

 4-2 



8 

 2-9 



10 



2-2 



H2 



H a . 



7-4 



8-5 



9-2 



9-7 



a 



H* 



86 



91 



92 



94 



92 



88 



80 



77 



73 



51 



39 



27 



18 



14-1 



%. 



Hy. 



5-9 



5-9 



5-8 



5-8 



5-5 



5-2 



4-6 



4-2 



3-9 



2-2 



17 



1-15 



•77 



•59 



3 



o 



H 5 . 



•83 



•71 



5-8 



•70 



•55 



•47 



•50 



•38 



•39 



•20 



•15 



•15 



•08 



•051 



1 



o 



HjS/Ha. 



11-6 



11-7 



100 



9-7 



9-4 



9-0 



8-3 



8-0 



7-8 



6-7 



6-4 



64 



6-2 



6-4 



o 



H 7 /H 3 . 



•069 



•065 



063 



•062 



0-60 



•059 



•057 



056 



•054 



•043 



•044 



0-42 



•043 



•042 



Ph 



H5/H y . 



•14 



•12 



•10 



•12 



•10 



•09 



•11 



•09 



•10 



0-9 



•09 



•13 



•11 



•08 



Similar experiments were made with pure dry hydrogen 

 admitted through a heated palladium-platinum tube fused 

 into the glass. A side tube furnished with silver electrodes 

 was sealed on and a discharge passed so as to form a silver 

 deposit on the walls. When fresh, this deposit rapidly 

 absorbs mercury vapour * ; the green and yellow mercury 

 lines disappearing in the course of half-an-hour. The 

 many-line spectrum of hydrogen was now prominent, and 

 it was necessary to allow for it in measuring the intensities 

 of the Balmer lines. 



Table II. gives the results obtained with and without the 

 background to which the many-line spectrum gives rise. 

 Fig. 6 (PI. II.) shows the intensity ratio H^/Ha in the latter 

 case. The upward bend of the curve at low pressures takes 

 place as with water vapour. 



Fig. 7 (PI. II.) shows the photo-currents obtained with 

 water vapour at a pressure of 8 mm. Fig. 8 (PI. II.) 

 illustrates the corresponding curve with pure dry hydrogen 



* This excellent method of eliminating mercury vapour was suggested 

 to the author hy Mr. T.E. Merton, B.Sc, King's College, London. 



