﻿414 
  Mr. 
  L. 
  Veo-ard 
  : 
  Research 
  

  

  & 
  

  

  es 
  upon 
  

  

  the 
  study 
  o£ 
  the 
  velocity 
  carves 
  we 
  come 
  to 
  the 
  following 
  

   conclusion 
  for 
  the 
  system 
  considered 
  : 
  — 
  

  

  (1) 
  1 
  he 
  lowering 
  of 
  the 
  reversion 
  point 
  is 
  accompanied 
  by 
  an 
  

   increase 
  of 
  the 
  leak. 
  

  

  (2) 
  As 
  long 
  as 
  we 
  have 
  a 
  comparatively 
  great 
  osmotic 
  leak 
  a 
  

   motion 
  in 
  the 
  characteristic 
  point 
  is 
  accompanied 
  by 
  a 
  motion 
  in 
  

   the 
  reversion 
  point 
  of 
  the 
  same 
  order. 
  But 
  ivhen 
  the 
  leak 
  

   diminishes 
  the 
  motion 
  of 
  the 
  reversion 
  point 
  compared 
  with 
  the 
  

   motion 
  of 
  the 
  characteristic 
  point 
  ivill 
  diminish, 
  and 
  with 
  a 
  very 
  

   small 
  leak 
  (eap. 
  II., 
  III.) 
  the 
  reversion 
  point 
  icill 
  keep 
  almost 
  

   invariable 
  for 
  the 
  considerable 
  changes 
  of 
  the 
  velocity 
  curve. 
  

  

  Then 
  we 
  see 
  that 
  by 
  diminishing 
  the 
  leak 
  we 
  approach 
  a 
  

   point 
  independent 
  of 
  the 
  variable 
  properties 
  of 
  the 
  membrane. 
  

   And 
  as 
  we 
  have 
  seen 
  in 
  § 
  3, 
  this 
  will 
  give 
  a 
  strong 
  support 
  

   for 
  the 
  continuity 
  of 
  7r 
  ' 
  up 
  to 
  the 
  point 
  of 
  perfect 
  semi- 
  

   permeability. 
  

  

  The 
  question 
  whether 
  the 
  pressures 
  measurable 
  by 
  our 
  

   membranes 
  will 
  be 
  able 
  to 
  give 
  an 
  approximate 
  value 
  of 
  

   the 
  osmotic 
  pressure 
  was 
  the 
  main 
  point 
  in 
  a 
  discussion 
  in 
  

   4 
  Nature 
  ' 
  * 
  during 
  the 
  year 
  1901), 
  called 
  forth 
  by 
  a 
  paper 
  of 
  

   Professor 
  Kahlenberg 
  f 
  . 
  At 
  the 
  general 
  discussion 
  upon 
  

   osmotic 
  pressure 
  that 
  took 
  place 
  in 
  a 
  meeting 
  of 
  the 
  Faraday 
  

   Society 
  last 
  year, 
  Pro 
  lessor 
  Kahlenberg 
  sums 
  up 
  his 
  opinion 
  

   by 
  the 
  following 
  statement}:: 
  — 
  

  

  "All 
  experimental 
  evidence 
  we 
  have 
  goes 
  to 
  demonstrate 
  

   (1) 
  that 
  there 
  is 
  no 
  such 
  thing 
  as 
  a 
  semipermeable 
  membrane 
  

   in 
  the 
  strict 
  sense 
  of 
  the 
  word, 
  and 
  (2) 
  the 
  more 
  nearly 
  a 
  

   membrane 
  is 
  semipermeable 
  in 
  character 
  in 
  practice, 
  the 
  

   greater 
  is 
  the 
  selective 
  action 
  ; 
  in 
  other 
  words, 
  it 
  is 
  the 
  pro- 
  

   nounced 
  selective 
  action 
  of 
  the 
  membrane 
  which 
  makes 
  it 
  

   approximately 
  semipermeable. 
  It 
  is 
  clear, 
  therefore, 
  that 
  

   such 
  a 
  membrane 
  as 
  thermodynamic 
  reasoning 
  postulates 
  can 
  

   never 
  be 
  realized 
  in 
  practice, 
  nor 
  can 
  we 
  hope 
  by 
  experiment 
  

   to 
  produce 
  even 
  approximately 
  such 
  a 
  membrane 
  ; 
  for 
  as 
  we 
  

   experimentally 
  approximate 
  toward 
  fulfilling 
  the 
  first 
  require- 
  

   ment, 
  semipermeability, 
  we 
  do 
  this 
  at 
  the 
  sacrifice 
  of 
  the 
  

   second 
  requirement, 
  passivity." 
  

  

  As 
  to 
  the 
  first 
  statement 
  it 
  is 
  probably 
  true 
  ; 
  and 
  it 
  is 
  the 
  

   more 
  so 
  as 
  a 
  statement 
  that 
  only 
  expresses 
  that 
  a 
  certain 
  

   process 
  in 
  nature 
  does 
  not 
  follow 
  a 
  mathematical 
  ideality 
  will 
  

  

  * 
  Earl 
  of 
  Berkeley, 
  ' 
  Nature/ 
  1906, 
  vol. 
  Ixxiv. 
  p. 
  7 
  ; 
  Earl 
  of 
  Berkeley 
  

   .& 
  E. 
  G. 
  J. 
  Hartley, 
  vol. 
  Ixxiv. 
  pp. 
  54, 
  245; 
  W. 
  E. 
  D. 
  Whethani, 
  

   vol. 
  Ixxiv. 
  pp. 
  54, 
  102, 
  295 
  ; 
  Professor 
  Henry 
  Armstrong, 
  vol. 
  Ixxiv. 
  p. 
  79 
  ; 
  

   Professor 
  L. 
  Kahlenberg, 
  vol. 
  Ixxiv. 
  pp. 
  19, 
  222 
  ; 
  Norman 
  E. 
  Campbell, 
  

   vol. 
  Ixxiv. 
  p. 
  79. 
  

  

  t 
  Professor 
  L. 
  Kahlenberg, 
  Journ. 
  of 
  Ph. 
  Chem. 
  vol. 
  x. 
  pp. 
  144-209. 
  

  

  J 
  Transactions 
  of 
  the 
  Faraday 
  Societ} 
  r 
  , 
  vol. 
  iii. 
  part 
  1, 
  1907. 
  

  

  