﻿812 
  Mr. 
  H. 
  L. 
  P. 
  Jolly 
  on 
  the 
  

  

  potential 
  difference 
  available 
  for 
  breaking 
  down 
  the 
  spark- 
  

   gap 
  falls 
  below 
  the 
  necessary 
  value, 
  and 
  the 
  discharge 
  

   ceases. 
  

  

  This 
  process 
  may 
  be 
  repeated 
  many 
  times 
  in 
  a 
  given 
  cycle, 
  

   bat 
  when 
  the 
  spark-length 
  is 
  approaching 
  the 
  maximum 
  

   permissible, 
  the 
  number 
  of 
  such 
  peaks 
  occurring 
  in 
  each 
  

   half-cycle 
  may 
  be 
  as 
  small 
  as 
  3, 
  2, 
  or 
  even 
  1. 
  Thus 
  we 
  

   cannot 
  be 
  certain 
  of 
  reproducing 
  the 
  conditions 
  at 
  two 
  

   different 
  times 
  when 
  it 
  might 
  be 
  desired 
  to 
  compare 
  the 
  

   energy 
  of 
  one 
  line 
  in 
  the 
  spectrum 
  with 
  that 
  of 
  another. 
  

   The 
  fact 
  that 
  each 
  of 
  the 
  peaks 
  represents 
  a 
  current 
  passing 
  

   for 
  a 
  very 
  small 
  fraction 
  of 
  a 
  second 
  (i. 
  e., 
  a 
  fraction 
  of 
  quite 
  

   a 
  different 
  order 
  from 
  the 
  1/90 
  sec. 
  characteristic 
  of 
  the 
  

   current 
  supply), 
  was 
  established 
  by 
  observing 
  them 
  in 
  an 
  

   apparatus 
  with 
  7 
  small 
  mirrors 
  fitted 
  to 
  a 
  drum 
  which 
  

   was 
  caused 
  to 
  rotate 
  at 
  about 
  50 
  revolutions 
  per 
  sec. 
  This 
  

   arrangement 
  would 
  have 
  been 
  able 
  to 
  detect 
  a 
  duration 
  of 
  

   the 
  current 
  over 
  an 
  interval 
  of 
  time 
  as 
  small 
  as 
  10~ 
  4 
  sec. 
  

   No 
  such 
  duration 
  was 
  apparent, 
  and 
  the 
  image 
  was 
  as 
  

   depicted 
  in 
  fig. 
  8 
  (PL 
  XIII.) 
  . 
  

  

  A 
  single 
  cycle 
  is 
  spread 
  out 
  by 
  such 
  an 
  image 
  into 
  a 
  band, 
  

   only 
  a 
  verv 
  small 
  portion 
  of 
  which 
  can 
  be 
  seen 
  in 
  any 
  one 
  

   mirror. 
  This 
  portion 
  may 
  or 
  may 
  not 
  contain 
  one 
  of 
  the 
  few 
  

   peaks 
  belonging 
  to 
  each 
  cycle. 
  The 
  next 
  mirror 
  as 
  well 
  as 
  

   several 
  succeeding 
  ones 
  will 
  still 
  show 
  a 
  portion 
  of 
  the 
  same 
  

   cycle, 
  but 
  only 
  certain 
  isolated 
  ones 
  will 
  show 
  a 
  peak. 
  

  

  We 
  thus 
  have 
  a 
  peak 
  shown 
  by, 
  say, 
  the 
  Nth 
  mirror, 
  the 
  

   N 
  + 
  Pth, 
  the 
  N 
  + 
  P 
  + 
  Qth, 
  and 
  so 
  on. 
  The 
  positions 
  of 
  these 
  

   peaks 
  will 
  depend 
  on 
  the 
  constant 
  frequency 
  with 
  which 
  one 
  

   mirror 
  takes 
  the 
  position 
  of 
  the 
  previous 
  one, 
  and 
  the 
  

   irregular 
  frequency 
  of 
  recurrence 
  of 
  a 
  condenser 
  discharge. 
  

   They 
  therefore 
  appear 
  quite 
  at 
  random 
  in 
  the 
  field 
  of 
  view, 
  

   both 
  above 
  and 
  below 
  the 
  line 
  (fig. 
  8) 
  . 
  

  

  The 
  discliarge-tuhe 
  which 
  was 
  used 
  as 
  source 
  of 
  radiation 
  

   was, 
  in 
  these 
  later 
  experiments, 
  of 
  quartz. 
  Fig. 
  9 
  is 
  a 
  

   diagram 
  of 
  two 
  of 
  the 
  tubes. 
  Each 
  electrode 
  in 
  the 
  larger 
  

   tube 
  consisted 
  of 
  a 
  thin-walled 
  brass 
  tube 
  with 
  a 
  lump 
  of 
  

   zinc 
  soldered 
  on 
  to 
  the 
  end. 
  

  

  They 
  were 
  also 
  water-cooled 
  so 
  as 
  to 
  prevent 
  the 
  fusing 
  of 
  

   the 
  sealing-wax 
  joints. 
  The 
  bulbs 
  were 
  about 
  4 
  cm. 
  in 
  

   diameter, 
  and 
  the 
  capillary 
  portion 
  about 
  1*8 
  cm. 
  long 
  and 
  

   0*7 
  mm. 
  internal 
  diameter. 
  

  

  The 
  smaller 
  tube 
  had 
  given 
  good 
  results, 
  but 
  owing 
  to 
  the 
  

   thinness 
  of 
  the 
  capillary 
  portion 
  and 
  consequent 
  puncturing 
  

   on 
  the 
  passage 
  of 
  a 
  heavy 
  discharge, 
  it 
  had 
  to 
  be 
  abandoned 
  

   in 
  favour 
  of 
  the 
  larger 
  tube. 
  Unhappily, 
  after 
  only 
  two 
  

  

  