﻿28 
  Prof. 
  Chattock 
  and 
  Mr. 
  Tyndall 
  on 
  Changes 
  

  

  horizontal 
  axis 
  — 
  not 
  all 
  the 
  positives 
  first 
  and 
  then 
  all 
  the 
  

   negatives. 
  

  

  Curves 
  II. 
  

  

  Byne-Ctn. 
  

  

  3* 
  

  

  2-£ 
  

   24 
  

  

  N 
  ls 
  - 
  

  

  

  \(+) 
  

  

  

  

  

  

  

  

  hs 
  

  

  

  

  

  

  

  

  

  

  5-1-2; 
  

  

  H 
  

  

  

  

  

  

  \ 
  

  

  

  

  3~Ql 
  

  

  

  

  

  

  

  A 
  

  

  J\ 
  

  

  

  *-* 
  

  

  

  

  

  

  

  

  ^ 
  

  

  ^r^^k^^.j/ 
  

  

  

  

  

  

  

  

  

  

  Cr 
  

  

  \ 
  — 
  ■<****$[ 
  

  

  

  

  

  a 
  ° 
  

  

  11 
  SDJ 
  

  

  v 
  5 
  y 
  

  

  ■ 
  f2 
  

  

  D 
  

  

  D 
  

  

  45 
  

   fc*S. 
  

  

  As 
  a 
  rule 
  the 
  time 
  between 
  each 
  observation 
  and 
  the 
  next 
  

   was 
  about 
  a 
  quarter 
  of 
  an 
  hour, 
  but 
  at 
  special 
  points, 
  as 
  when 
  

   a 
  night 
  intervened, 
  the 
  longer 
  interval 
  is 
  given 
  in 
  hours 
  -by 
  

   the 
  number 
  below. 
  At 
  the 
  points 
  marked 
  by 
  D 
  a 
  discharge 
  

   of 
  negative 
  electricity 
  (2 
  microamperes) 
  was 
  kept 
  up 
  for 
  

   about 
  half-an-hour, 
  the 
  corresponding 
  values 
  of 
  p 
  being- 
  

   unobtainable 
  on 
  account 
  of 
  the 
  drift. 
  The 
  break 
  in 
  the 
  curves 
  

   indicates 
  a 
  fresh 
  filling 
  of 
  the 
  apparatus 
  with 
  "pure'* 
  hydrogen 
  

   after 
  pumping 
  out 
  sufficiently 
  to 
  reduce 
  the 
  original 
  impurity 
  

   about 
  80 
  times. 
  

  

  It 
  will 
  be 
  seen 
  that 
  with 
  the 
  first 
  filling 
  a 
  wait 
  of 
  some 
  

   hours 
  is 
  always 
  accompanied 
  by 
  a 
  rise 
  in 
  the 
  values 
  of 
  p. 
  This 
  

   is 
  due 
  to 
  the 
  diffusion 
  into 
  the 
  discharge 
  vessel 
  of 
  unpurified 
  

   hydrogen 
  l'rom 
  the 
  space 
  above 
  the 
  mercury 
  of 
  the 
  pressure- 
  

   gauge, 
  connecting 
  tubes, 
  &c. 
  Discharge, 
  on 
  the 
  other 
  hand, 
  

   brings 
  the 
  values 
  down 
  as 
  it 
  should. 
  

  

  With 
  the 
  second 
  filling 
  the 
  gas 
  was 
  so 
  much 
  purer 
  to 
  start 
  

   with 
  that 
  these 
  effects 
  are 
  almost 
  absent. 
  

  

  In 
  neither 
  case 
  would 
  it 
  be 
  easy 
  to 
  say 
  which 
  sign 
  of 
  dis- 
  

   charge 
  gives 
  the 
  higher 
  value 
  of 
  p, 
  and 
  by 
  the 
  end 
  of 
  the 
  

   second 
  curves 
  the 
  two 
  signs 
  are 
  obviously 
  about 
  equally 
  

   effective. 
  The 
  common 
  value 
  of 
  p 
  is 
  then 
  roughly 
  0*3 
  dyne 
  

   per 
  sq. 
  cm., 
  which 
  corresponds 
  with 
  N=l*l. 
  

  

  This 
  residual 
  contraction 
  of 
  0*3 
  is 
  a 
  rough 
  average 
  of 
  what 
  

   appears 
  to 
  be 
  a 
  limit 
  — 
  a 
  rather 
  variable 
  limit 
  it 
  is 
  true 
  — 
  below 
  

  

  