﻿410 
  Mr. 
  L. 
  Vegard 
  : 
  Researches 
  upon 
  

  

  the 
  state 
  of 
  the 
  systems 
  just 
  after 
  the 
  apparatus 
  had 
  been 
  

  

  put 
  up, 
  and 
  the 
  observations 
  commenced 
  at 
  the 
  lowest 
  

  

  pressure. 
  It 
  is 
  the 
  same 
  mode 
  of 
  operation 
  as 
  used 
  in 
  

  

  experiment 
  I. 
  The 
  course 
  of 
  the 
  curves 
  is 
  also 
  quite 
  similar. 
  

  

  We 
  have 
  an 
  inflexion 
  point 
  in 
  the 
  interval 
  0<7r<7r', 
  and 
  on 
  

  

  the 
  other 
  side 
  of 
  the 
  reversion 
  point 
  we 
  have 
  a 
  point 
  of 
  

  

  dX' 
  

   discontinuity 
  in 
  the 
  function-^ 
  with 
  a 
  fall 
  in 
  the 
  value 
  by 
  

  

  increasing 
  pressure. 
  After 
  passing 
  this 
  point 
  the 
  curves 
  

   bend 
  upwards, 
  and 
  now 
  much 
  more 
  rapidly 
  than 
  in 
  the 
  

   first 
  case. 
  

  

  The 
  position 
  of 
  the 
  characteristic 
  point 
  for 
  the 
  two 
  curves 
  

   is, 
  however, 
  very 
  different. 
  For 
  the 
  curves 
  fig. 
  11 
  it 
  has 
  

   the 
  coordinates 
  7r 
  = 
  61, 
  \' 
  = 
  312. 
  Or 
  we 
  get 
  it 
  first 
  for 
  a 
  

   pressure 
  more 
  than 
  twice 
  the 
  equilibrium 
  pressure. 
  For 
  

   the 
  curve 
  fig. 
  12 
  the 
  characteristic 
  point 
  lies 
  close 
  up 
  to 
  the 
  

   reversion 
  point, 
  the 
  coordinates 
  being 
  7r 
  = 
  303, 
  V 
  = 
  13'5. 
  

  

  We 
  also 
  notice 
  tbat 
  in 
  the 
  interval 
  0<7r<7r 
  / 
  the 
  absolute 
  

   value 
  of 
  the 
  velocities 
  is 
  greater 
  in 
  experiment 
  III. 
  

  

  The 
  two 
  other 
  curves, 
  drawn 
  as 
  dotted 
  lines, 
  were 
  observed 
  

   after 
  the 
  apparatus 
  had 
  been 
  placed 
  under 
  pressure 
  for 
  about 
  

   two 
  days. 
  We 
  see 
  tbat 
  in 
  both 
  cases 
  the 
  curves 
  have 
  

   undergone 
  a 
  considerable 
  change. 
  

  

  The 
  change 
  can 
  be 
  characterized 
  thus 
  : 
  — 
  

  

  (1) 
  The 
  characteristic 
  point 
  has 
  moved 
  towards 
  the 
  

   reA 
  T 
  ersion 
  point. 
  

  

  (2) 
  The 
  velocities 
  near 
  the 
  pressure 
  -tt 
  = 
  have 
  consi- 
  

   derably 
  diminished. 
  

  

  By 
  this 
  motion 
  of 
  the 
  characteristic 
  point 
  towards 
  the 
  

   reversion 
  point 
  the 
  latter 
  is 
  in 
  both 
  cases 
  very 
  little 
  affected. 
  

   In 
  experiment 
  III. 
  we 
  have 
  a 
  small 
  increase 
  in 
  the 
  reversion 
  

   pressure 
  during 
  the 
  first 
  twenty-four 
  hours, 
  but 
  it 
  keeps 
  later 
  

   nearly 
  constant. 
  

  

  Let 
  us 
  imagine 
  that 
  we 
  place 
  one 
  of 
  the 
  two 
  plates 
  upon 
  

   the 
  other 
  so 
  that 
  the 
  it 
  axis 
  and 
  zero 
  points 
  coincide. 
  Then 
  

   we 
  should 
  find 
  that 
  all 
  the 
  curves 
  have 
  a 
  considerably 
  different 
  

   course, 
  mainly 
  due 
  to 
  the 
  different 
  position 
  of 
  the 
  characteristic 
  

   point. 
  In 
  spite 
  of 
  this 
  variation 
  in 
  the 
  velocity 
  curves 
  we 
  see 
  

   that 
  all 
  of 
  them 
  very 
  nearly 
  cut 
  the 
  it 
  axis 
  in 
  the 
  same 
  point. 
  

   This 
  is 
  the 
  more 
  remarkable 
  as 
  the 
  curves 
  show 
  great 
  differ- 
  

   ences 
  quite 
  up 
  to 
  the 
  point 
  of 
  intersection. 
  At 
  all 
  events 
  the 
  

   variations 
  in 
  the 
  reversion 
  point 
  are 
  very 
  small 
  in 
  comparison 
  

   to 
  the 
  variations 
  of 
  the 
  curves 
  close 
  up 
  to 
  this 
  point. 
  

  

  