﻿Measuring 
  large 
  Molecular 
  Masses. 
  509- 
  

  

  and 
  interpretation. 
  The 
  data 
  for 
  the 
  casemates 
  were 
  obtained 
  

   by 
  Saekur 
  (Ztschr. 
  /. 
  physih 
  Chem.lxl 
  1902) 
  at 
  25°. 
  For 
  

   Na 
  casein 
  thev 
  give 
  A^,/2=2S'l5, 
  A 
  j/2+A 
  0o 
  =219 
  , 
  3j 
  and 
  

   &== 
  1-678. 
  The 
  sum 
  of 
  28*15 
  and 
  219*3 
  is 
  247*4, 
  while 
  X 
  for 
  

   NaOH 
  at 
  25° 
  and 
  infinite 
  dilution 
  is 
  247*2. 
  Kohlrausch's 
  

   value 
  for 
  A 
  j 
  2 
  at 
  25° 
  is 
  25'6, 
  so 
  perhaps 
  28'15-25'6 
  repre- 
  

   sents 
  a 
  tendency 
  for 
  A 
  02 
  for 
  casein 
  to 
  make 
  its 
  appearance. 
  

   The 
  following 
  comparison 
  shows 
  the 
  applicability 
  of 
  (14) 
  to 
  

   Na 
  casein 
  : 
  — 
  

  

  1 
  1000;;; 
  .. 
  

  

  40 
  

  

  80 
  

  

  160 
  

  

  320 
  

  

  640 
  

  

  X 
  exp 
  

  

  46-5 
  

  

  513 
  

  

  562 
  

  

  630 
  

  

  69-5 
  

  

  X 
  calc 
  

  

  46-6 
  

  

  51-1 
  

  

  56-4 
  

  

  625 
  

  

  69-7 
  

  

  For 
  XH 
  4 
  casein 
  A 
  o6 
  /2 
  = 
  38*4, 
  A 
  oi 
  /2 
  + 
  A 
  oa 
  = 
  231*5, 
  and 
  

   6=1-634. 
  For 
  Hardy's 
  NH 
  4 
  globulin 
  at 
  18° 
  the 
  corre- 
  

   sponding 
  values 
  are 
  29, 
  208*3, 
  and 
  2*3, 
  this 
  last 
  number 
  

   differing 
  unaccountably 
  from 
  the 
  3*1 
  for 
  Na 
  globulin. 
  

  

  It 
  is 
  necessary 
  to 
  form 
  some 
  conception 
  of 
  the 
  way 
  in 
  

   which 
  half 
  of 
  the 
  positive 
  ions 
  in 
  these 
  solutions 
  are 
  exempt 
  

   from 
  the 
  resistance 
  of 
  viscosity 
  of 
  electric 
  origin. 
  The 
  case 
  

   becomes 
  clearer 
  if 
  we 
  consider 
  first 
  the 
  simpler 
  one 
  of 
  the 
  

   HC1 
  compound 
  of 
  globulin 
  for 
  which 
  (Joe. 
  cit.) 
  

  

  l/\ 
  = 
  1/384*6 
  + 
  0-9 
  mi 
  (15) 
  

  

  Here 
  at 
  infinite 
  dilution 
  the 
  conductivity 
  is 
  384*6, 
  which 
  is 
  

   almost 
  the 
  384*0 
  for 
  pure 
  HC1, 
  while 
  0'9 
  for 
  b 
  is 
  so 
  large 
  

   that 
  it 
  must 
  be 
  due 
  to 
  large 
  globulin 
  ions. 
  It 
  seems, 
  then, 
  

   that 
  HC1 
  in 
  water 
  ionizes 
  the 
  globulin 
  molecule 
  into 
  a 
  

   positive 
  and 
  a 
  negative 
  globulin 
  ion 
  which 
  we 
  shall 
  denote 
  

   by 
  the 
  symbols 
  G/, 
  and 
  G 
  fi 
  . 
  The 
  solution 
  of 
  globulin 
  in 
  HC1 
  

   is 
  a 
  mixture 
  of 
  solutions 
  of 
  G^Cl 
  and 
  G«Xa, 
  which 
  are 
  both 
  

   completely 
  ionized 
  at 
  all 
  concentrations, 
  so 
  that 
  at 
  

   infinite 
  dilution 
  the 
  conductivity 
  is 
  the 
  sum 
  of 
  those 
  for 
  

   H, 
  CI, 
  G 
  a 
  , 
  and 
  Q 
  b 
  . 
  But 
  A 
  0a 
  -f 
  A 
  & 
  is 
  too 
  small 
  compared 
  with 
  

   A 
  ol 
  4-A 
  o2 
  for 
  HC1 
  to 
  be 
  found 
  from 
  ordinary 
  conductivity 
  

   experiments, 
  and 
  therefore 
  does 
  not 
  appear 
  in 
  (15). 
  But 
  to 
  

   find 
  the 
  effect 
  of 
  the 
  globulin 
  ions 
  on 
  the 
  value 
  of 
  L 
  we 
  may 
  

   reason 
  in 
  the 
  following 
  manner. 
  So 
  far 
  I 
  have 
  not 
  considered 
  

   the 
  molecular 
  conductivity 
  of 
  mixtures 
  in 
  general, 
  though 
  

   the 
  subject 
  ought 
  to 
  illustrate 
  some 
  of 
  the 
  details 
  of 
  visco- 
  

   sities 
  of 
  electric 
  origin. 
  But 
  for 
  the 
  matter 
  in 
  hand 
  we 
  may 
  

   formulate 
  a 
  simple 
  approximate 
  theory 
  for 
  our 
  mixture 
  of 
  

   globulin 
  compounds 
  thus: 
  — 
  As 
  the 
  ions 
  pass 
  one 
  another 
  we 
  

   shall 
  have 
  cases 
  where 
  G& 
  has 
  G« 
  for 
  its 
  neighbour, 
  and 
  also 
  

   others 
  where 
  CI 
  is 
  its 
  neighbour, 
  and 
  similarly 
  with 
  Q«. 
  It 
  

   is 
  as 
  if 
  we 
  had 
  to 
  do 
  with 
  a 
  mixture 
  of 
  G,,G^ 
  G&C1, 
  G 
  a 
  Na, 
  

  

  