﻿170 
  Mr. 
  W. 
  Sutherland 
  on 
  Ionization, 
  

  

  determinations 
  of 
  Kohlrausch 
  (Belli, 
  xxv.) 
  for 
  strengths 
  from 
  

   ra 
  = 
  '0001 
  to 
  m='l 
  at 
  18° 
  C, 
  with 
  which 
  we 
  can 
  join 
  on 
  his 
  

   previous 
  determinations 
  from 
  m=l 
  to 
  m 
  = 
  o. 
  Kohlrausch 
  

   estimates 
  the 
  specific 
  molecular 
  conductivity 
  of 
  NaCl 
  at 
  

   infinite 
  dilution 
  in 
  water 
  as 
  108*99 
  cm 
  -1 
  ohm 
  -1 
  . 
  We 
  need 
  

   give 
  then 
  only 
  A/X 
  in 
  the 
  following 
  table 
  along 
  with 
  t) 
  /tj 
  

   found 
  by 
  interpolation 
  from 
  Hosking's 
  data 
  (Phil. 
  Mag. 
  [5] 
  

   xlix.), 
  the 
  viscosity 
  of 
  water 
  at 
  18° 
  being 
  taken 
  as 
  '01059 
  c.G.s. 
  

  

  Table 
  I. 
  — 
  NaCl 
  Solutions. 
  

  

  m 
  l/oo 
  -0001 
  -001 
  01 
  l 
  1 
  2 
  3 
  4 
  5 
  

  

  wi/8 
  ... 
  00 
  -0464 
  -10 
  -2154 
  -464 
  1-0 
  T26 
  1*442 
  1'587 
  1710 
  

  

  XA 
  1-0 
  -9918 
  -9770 
  -9354 
  8443 
  -6822 
  -594-517 
  *450 
  -391 
  

  

  rjjri 
  TO 
  1-0 
  10 
  -9995-995 
  927 
  '832 
  -736 
  645 
  -558 
  

  

  ^=W/ 
  ... 
  10 
  -9918 
  -977 
  -936 
  -849 
  -736 
  -715 
  -701 
  697 
  '700 
  

  

  jwcalc.. 
  10 
  983 
  -964 
  -925 
  -851 
  -741 
  -712 
  -701 
  -697 
  -699 
  

  

  * 
  is 
  put 
  for 
  l/il-Cl-Ko/KO^i/D-Cl-Ko/K^^/D} 
  

   which 
  for 
  the 
  small 
  values 
  of 
  m 
  differs 
  so 
  little 
  from 
  1*0 
  that 
  

   xi 
  may 
  be 
  taken 
  to 
  give 
  an 
  approximate 
  value 
  of 
  i 
  for 
  very 
  

   dilute 
  solutions. 
  In 
  the 
  second 
  row 
  values 
  of 
  m 
  1 
  / 
  8 
  have 
  been 
  

   given, 
  because 
  Kohlrausch 
  found 
  several 
  years 
  ago 
  that 
  for 
  

   dilute 
  solutions 
  \/\ 
  is 
  a 
  linear 
  function 
  of 
  m^ 
  3 
  , 
  so 
  that 
  xi 
  

   must 
  also 
  be 
  nearly 
  a 
  linear 
  function 
  of 
  ??i 
  1/3 
  when 
  m 
  is 
  small. 
  

   But 
  the 
  rest 
  of 
  the 
  data 
  show 
  that 
  xi 
  reaches 
  a 
  minimum 
  

   value 
  at 
  nearly 
  the 
  strongest 
  solution. 
  In 
  the 
  graph 
  of 
  these 
  

   data, 
  given 
  in 
  the 
  figure 
  (p. 
  171) 
  with 
  m 
  l 
  ' 
  z 
  as 
  abscissa 
  and 
  xi 
  as 
  

   ordinate, 
  we 
  see 
  that 
  the 
  graph 
  instead 
  of 
  being 
  a 
  straight 
  

   line 
  for 
  small 
  values 
  of 
  m 
  1/3 
  is 
  a 
  curve 
  with 
  a 
  point 
  of 
  inflexion 
  

   separating 
  two 
  parts 
  of 
  small 
  opposite 
  curvature. 
  Some 
  

   graphs 
  given 
  by 
  Whetham 
  (Proc. 
  Roy. 
  Soc. 
  1900) 
  for 
  \/\ 
  

   show 
  the 
  same 
  effect. 
  Kohlrausch' 
  s 
  straight 
  line 
  is 
  the 
  simplest 
  

   approximation 
  to 
  this 
  part 
  of 
  the 
  curve 
  ; 
  but 
  the 
  whole 
  graph 
  

   looks 
  like 
  a 
  parabola 
  which 
  at 
  small 
  values 
  of 
  m 
  l 
  3 
  gets 
  

   distorted. 
  The 
  cause 
  of 
  the 
  distortion 
  is 
  probably 
  this. 
  

   Kohlrausch 
  pointed 
  out 
  that 
  in 
  experiments 
  on 
  molecular 
  

   conductivity 
  for 
  dilute 
  solutions 
  it 
  is 
  necessary 
  to 
  subtract 
  

   from 
  the 
  conductivity 
  of 
  the 
  solution 
  that 
  of 
  the 
  solvent, 
  

   assuming 
  that 
  the 
  conductivity 
  of 
  the 
  solvent 
  is 
  not 
  altered 
  

   by 
  the 
  addition 
  of 
  the 
  solute. 
  Now, 
  for 
  a 
  liquid 
  like 
  water 
  

   which 
  contains 
  (H 
  2 
  0)3 
  an< 
  i 
  (H 
  2 
  0) 
  2 
  in 
  delicately 
  equilibrating 
  

   amount, 
  it 
  is 
  not 
  safe 
  to 
  assume 
  that 
  its 
  conductivity 
  is 
  not 
  

   altered 
  by 
  the 
  mere 
  presence 
  of 
  a 
  small 
  amount 
  of 
  electrolvte. 
  

   The 
  fact 
  that 
  the 
  electrolyte 
  is 
  dissociated 
  makes 
  it 
  probable 
  

   that 
  the 
  slight 
  ionization 
  and 
  conductivity 
  of 
  water 
  are 
  

   altered 
  by 
  the 
  presence 
  of 
  the 
  electrolyte. 
  In 
  fact 
  it 
  is 
  well 
  

  

  