﻿[ 
  863 
  ] 
  

  

  LXXIV. 
  On 
  the 
  Coefficient 
  of 
  Diffusion. 
  By 
  Basil 
  W. 
  Clack, 
  

   B.Sc, 
  Lecturer 
  in 
  Physics 
  at 
  Birkbeck 
  College*. 
  

   [Plate 
  XXV.] 
  

   S 
  1. 
  Introduction. 
  

  

  2. 
  Theory. 
  

  

  

  § 
  3. 
  Apparatus. 
  

   § 
  4. 
  Method 
  of 
  Experiment. 
  

   § 
  5. 
  Determination 
  of 
  Concentration. 
  

   § 
  6. 
  Determination 
  of 
  g. 
  

   § 
  7. 
  .Results 
  for 
  KOI 
  and 
  KN0 
  3 
  . 
  

   § 
  8. 
  Earlier 
  Method 
  of 
  Experiment 
  TMethod 
  A]. 
  

   § 
  9. 
  Length 
  of 
  the 
  Tube. 
  

   § 
  10. 
  Conclusion. 
  

  

  1. 
  Introduction, 
  

  

  THE 
  following 
  paper 
  contains 
  an 
  account 
  of 
  a 
  series 
  of 
  

   experiments 
  on 
  the 
  diffusion 
  of 
  various 
  salts 
  through 
  

   water. 
  The 
  objects 
  with 
  which 
  they 
  were 
  undertaken 
  were 
  : 
  — 
  

  

  1st. 
  To 
  test 
  the 
  practicability 
  of 
  a 
  new 
  method 
  for 
  the 
  

   determination 
  of 
  the 
  coefficient 
  of 
  diffusion; 
  

  

  2nd. 
  To 
  find 
  how 
  the 
  value 
  of 
  the 
  coefficient 
  varies 
  with 
  

   the 
  concentration 
  of 
  the 
  solution. 
  

  

  Various 
  methods 
  have 
  been 
  employed 
  from 
  time 
  to 
  time 
  

   to 
  determine 
  the 
  value 
  of 
  the 
  coefficient 
  of 
  diffusion. 
  In 
  

   some 
  cases 
  careful 
  chemical 
  analyses 
  are 
  necessary 
  to 
  obtain 
  

   the 
  amount 
  of 
  the 
  substance 
  diffused 
  ; 
  in 
  others 
  it 
  is 
  the 
  

   density 
  of 
  the 
  solution 
  which 
  must 
  be 
  accurately 
  determined 
  

   at 
  different 
  points. 
  Graham's, 
  and 
  kindred 
  methods, 
  involve 
  

   serious 
  disturbance 
  of 
  the 
  solution 
  in 
  commencing 
  an 
  ex- 
  

   periment, 
  and 
  also 
  in 
  withdrawing 
  the 
  various 
  specimen 
  

   layers 
  for 
  analysis. 
  

  

  The 
  present 
  investigation 
  may 
  be 
  called 
  a 
  gravitational 
  

   method, 
  since 
  the 
  amount 
  of 
  the 
  substance 
  diffused 
  is 
  deter- 
  

   mined 
  by 
  means 
  of 
  the 
  balance 
  ; 
  but 
  the 
  method 
  employed 
  

   involves 
  little 
  disturbance 
  at 
  the 
  commencement 
  of 
  an 
  ex- 
  

   periment, 
  and 
  during 
  its 
  progress 
  the 
  disturbance 
  is 
  quite 
  

   negligible. 
  Moreover, 
  neither 
  a 
  series 
  of 
  density 
  determi- 
  

   nations 
  nor 
  of 
  chemical 
  analyses 
  is 
  required 
  to 
  obtain 
  the 
  

   amount 
  of 
  diffusion 
  which 
  has 
  taken 
  place. 
  

  

  The 
  method 
  may 
  be 
  looked 
  upon 
  as 
  an 
  extension 
  of 
  that 
  

   employed 
  by 
  Fick 
  (Pogg. 
  Ann. 
  xciv. 
  p. 
  59, 
  1855) 
  and 
  by 
  

   Graham, 
  but 
  the 
  method 
  of 
  obtaining 
  the 
  quantity 
  of 
  salt 
  

   diffused 
  is 
  essentially 
  different, 
  as 
  they 
  determined 
  this 
  amount 
  

   by 
  chemical 
  analysis. 
  

  

  Again, 
  it 
  does 
  not 
  appear 
  that 
  Fick 
  took 
  any 
  account 
  of 
  

   * 
  Communicated 
  by 
  the 
  Physical 
  Society 
  : 
  read 
  May 
  22 
  7 
  1908. 
  

  

  