﻿264 
  

  

  Mr. 
  G. 
  B. 
  Bryan 
  on 
  the 
  Determination 
  of 
  

  

  Water. 
  — 
  The 
  ordinary 
  water 
  from 
  the 
  laboratory-still 
  was 
  

   used. 
  This 
  had 
  a 
  comparatively 
  high 
  conductivity, 
  and 
  was 
  

   in 
  fact 
  a 
  dilute 
  solution. 
  The 
  gilded 
  zinc 
  electrodes 
  were 
  

   employed, 
  and 
  to 
  keep 
  the 
  conduction 
  as 
  much 
  as 
  possible 
  to 
  

   the 
  space 
  between 
  the 
  plates 
  they 
  were 
  surrounded 
  by 
  a 
  close- 
  

   fitting 
  inner 
  cylinder 
  of 
  glass, 
  as 
  shown 
  in 
  fig. 
  1. 
  

  

  Before 
  putting 
  the 
  plates 
  into 
  position 
  they 
  were 
  cleaned 
  

   with 
  dilute 
  acid 
  and 
  alcohol. 
  

  

  The 
  resistance 
  was 
  measured 
  by 
  method 
  3, 
  two 
  Daniells 
  

   being 
  used 
  to 
  give 
  the 
  E.M.F. 
  

  

  A 
  large 
  number 
  of 
  readings 
  were 
  taken, 
  and 
  they 
  all 
  agreed 
  

   in 
  showing 
  that 
  the 
  apparent 
  specific 
  resistance 
  was 
  consider- 
  

   ably 
  greater 
  in 
  the 
  thin 
  layers. 
  The 
  plates 
  were 
  removed 
  and 
  

   cleaned 
  several 
  times, 
  but 
  the 
  results 
  were 
  always 
  the 
  same. 
  

   The 
  following 
  is 
  one 
  set 
  of 
  readings 
  : 
  — 
  

  

  Table 
  VIII.— 
  Distilled 
  Water. 
  

  

  Temp. 
  12° 
  0. 
  

  

  Turns 
  of 
  

   screw. 
  

  

  mm. 
  

  

  r, 
  ohms. 
  

  

  Sp. 
  K, 
  

   C.G.S.xlO- 
  11 
  . 
  

  

  ; 
  

  

  Differences 
  Resistance 
  of 
  

  

  between 
  1 
  turn 
  = 
  T27 
  mm. 
  

   the 
  nos. 
  in 
  from 
  last 
  

   col. 
  3. 
  column. 
  

  

  •1 
  

   •3 
  

   •5 
  

   1-0 
  

   20 
  

   40 
  

   70 
  

  

  127 
  

   •381 
  

   •635 
  

   1-270 
  

   254 
  

   5-08 
  

   8-89 
  

  

  659 
  

  

  132 
  

  

  202 
  

  

  375 
  

  

  740 
  

   1438 
  

   2440 
  

  

  1055 
  

   704 
  

   646 
  

   600 
  

   592 
  

   575 
  

   559 
  

  

  661 
  330 
  

  

  70 
  350 
  

  

  173 
  346 
  

  

  365 
  365 
  

  

  698 
  349 
  

  

  1002 
  334 
  

  

  

  

  

  3457 
  mean. 
  

  

  The 
  values 
  of 
  the 
  specific 
  resistance 
  in 
  the 
  fourth 
  column 
  

   show 
  a 
  large 
  falling 
  off 
  as 
  the 
  thickness 
  increases. 
  In 
  the 
  

   fifth 
  column 
  are 
  given 
  the 
  differences 
  between 
  consecutive 
  

   values 
  of 
  the 
  resistance 
  in 
  column 
  3, 
  and 
  these 
  are 
  multiplied 
  

   by 
  constants 
  to 
  give 
  the 
  resistance 
  of 
  a 
  thickness 
  of 
  *1 
  27 
  mm. 
  

   The 
  approximate 
  equality 
  of 
  these 
  numbers 
  shows 
  that 
  the 
  

   great 
  increase 
  in 
  specific 
  resistance 
  in 
  the 
  thin 
  layers 
  is 
  

   probably 
  due 
  to 
  the 
  plates 
  not 
  being 
  perfectly 
  clean 
  to 
  begin 
  

   with 
  ; 
  to 
  polarization, 
  and 
  changes 
  at 
  the 
  electrodes 
  produced 
  

   by 
  the 
  current 
  ; 
  and 
  to 
  errors 
  of 
  adjustment. 
  As 
  the 
  end 
  

   effect 
  was 
  probably 
  rather 
  different 
  with 
  different 
  thickness, 
  

   the 
  following 
  experiment 
  was 
  made 
  to 
  find 
  the 
  conductivity 
  

   at 
  each 
  position 
  of 
  the 
  plates. 
  

  

  At 
  each 
  position 
  the 
  resistance 
  was 
  measured, 
  then 
  the 
  

  

  