﻿Distribution 
  of 
  Energy 
  in 
  the 
  Spectra 
  of 
  Gases. 
  815 
  

  

  manner. 
  It 
  is 
  somewhat 
  remarkable 
  that 
  the 
  Carve 
  2 
  A 
  should 
  

   be 
  linear 
  oyer 
  so 
  great 
  a 
  portion 
  of 
  its 
  length, 
  for 
  this 
  does 
  

   not 
  generally 
  happen, 
  as 
  is 
  shown 
  by 
  Carve 
  2 
  B, 
  which 
  is 
  

   another 
  of 
  the 
  same 
  sort. 
  The 
  calibration 
  of 
  the 
  electroscope 
  

   is 
  given 
  in 
  Table 
  I. 
  On 
  the 
  curve 
  obtained 
  from 
  the 
  dis- 
  

   charge 
  current 
  we 
  find 
  that 
  a 
  deflexion 
  of 
  39' 
  9 
  divisions 
  

   corresponds 
  with 
  a 
  potentiometer 
  resistance 
  of 
  7780 
  ohms. 
  

   Therefore 
  128 
  volts 
  correspond 
  with 
  

  

  128/172 
  x 
  7780 
  ohms=5790 
  ohms, 
  

  

  against 
  which 
  we 
  find 
  on 
  the 
  curve 
  a 
  deflexion 
  25'3 
  divisions, 
  

   agreeing 
  to 
  within 
  2 
  per 
  cent, 
  with 
  the 
  25*7 
  obtained. 
  

  

  Corresponding 
  to 
  86 
  volts 
  is 
  86/172 
  x 
  7780 
  = 
  3890 
  ohins, 
  

   and 
  the 
  deflexion 
  should 
  be 
  7*1. 
  The 
  observed 
  value 
  is 
  7'3. 
  

   If 
  the 
  deflexion 
  is 
  greater 
  than 
  15 
  divisions, 
  the 
  curve 
  shows 
  

   that 
  we 
  may 
  compute 
  the 
  current 
  from 
  the 
  formula 
  

  

  z 
  = 
  l/r(3100rf 
  + 
  18000), 
  

  

  where 
  d= 
  electroscope 
  deflexion, 
  r= 
  potentiometer 
  resistance. 
  

   Thns 
  a 
  certain 
  current 
  gave 
  d= 
  18*3, 
  when 
  r 
  was 
  8000, 
  

   while 
  on 
  decreasing 
  r 
  to 
  5000, 
  d 
  was 
  only 
  2*1. 
  From 
  the 
  

   accurate 
  part 
  of 
  the 
  carve 
  we 
  obtain 
  

  

  = 
  1/8000(3100 
  x 
  18-3 
  + 
  18000) 
  = 
  13*1 
  milli-ampe 
  

  

  res. 
  

  

  Making 
  use 
  of 
  the 
  small 
  deflexion 
  2*1 
  division, 
  the 
  curve 
  

   shows 
  that 
  this 
  would 
  have 
  been 
  given 
  in 
  the 
  previous 
  case 
  by 
  

   3120 
  ohms. 
  This 
  corresponds 
  with 
  (172/7780) 
  x 
  3120 
  volt, 
  

   i.e. 
  with 
  (172/7780) 
  (3120/5000) 
  x 
  1000 
  = 
  13'8 
  milli-amperes, 
  

   which 
  agrees 
  well 
  enough 
  with 
  the 
  value 
  13' 
  1 
  obtained 
  from 
  

   ihe 
  accurate 
  part 
  of 
  the 
  curve. 
  

  

  Table 
  I. 
  

  

  Voltage. 
  

  

  Deflexion. 
  

  

  128 
  

  

  172 
  

   86 
  

  

  19-3] 
  o,- 
  

  

  45-0 
  ;-°' 
  

  

  59*0 
  } 
  39 
  " 
  9 
  

  

  19-11 
  „.., 
  

   26-4 
  ) 
  l 
  ° 
  

  

  