﻿on 
  Electric 
  Waves 
  in 
  Short] 
  Wire 
  Systems. 
  605 
  

  

  of 
  the 
  screw 
  being 
  the 
  same 
  in 
  all. 
  Only 
  mean 
  readings 
  are 
  

   given 
  to 
  one 
  decimal 
  place; 
  in 
  no 
  case 
  did 
  individual 
  readings 
  

   differ 
  by 
  more 
  than 
  0*7 
  div. 
  j^ 
  g 
  

  

  In 
  figure 
  6, 
  a, 
  ft, 
  c 
  represent 
  the 
  three 
  

   equidistant 
  points 
  on 
  the 
  swinging 
  plate 
  at 
  

   which 
  measurements 
  are 
  made, 
  a\ 
  b\ 
  c' 
  being 
  

   three 
  adjacent 
  points 
  on 
  the 
  guard-ring 
  ; 
  

   the 
  readings 
  are 
  as 
  follows 
  : 
  — 
  

  

  a'=3-5 
  b'=Cvl 
  

  

  Before 
  weighing 
  a 
  = 
  5*9 
  b 
  =5'8 
  

  

  After 
  weighing 
  b 
  = 
  5*2 
  Z> 
  =4*9 
  

  

  The 
  mean 
  distance 
  reading 
  is 
  taken 
  as 
  5'51. 
  

  

  The 
  lever 
  calipers 
  measuring 
  the 
  distance 
  between 
  the 
  jaws 
  

   of 
  a 
  Brown 
  & 
  Sharp 
  vernier 
  calipers 
  set 
  at 
  30*00 
  mm., 
  

   reading 
  = 
  6'00. 
  

  

  Distance 
  between 
  plates 
  = 
  30*00 
  mm. 
  + 
  equivalent 
  of 
  (6*00 
  — 
  

   5*51) 
  divs. 
  The 
  accuracy 
  of 
  the 
  distance 
  measurements 
  is 
  

   much 
  greater 
  than 
  that 
  of 
  the 
  determinations 
  of 
  the 
  electro- 
  

   static 
  pull. 
  The 
  distance 
  between 
  the 
  plates 
  must 
  be 
  known 
  to 
  

   0*05 
  mm. 
  if 
  the 
  value 
  of 
  K 
  is 
  to 
  be 
  determined 
  to 
  1 
  p. 
  cent. 
  

   The 
  distance 
  is 
  really 
  known 
  to 
  a 
  value 
  well 
  within 
  this 
  limit. 
  

  

  Electrical 
  details. 
  — 
  The 
  lower 
  plate 
  of 
  the 
  electrometer 
  

   carefully 
  insulated 
  is 
  attached 
  to 
  the 
  insulated 
  terminal 
  of 
  a 
  

   large 
  induction-coil, 
  the 
  other 
  terminal 
  and 
  the 
  upper 
  plate 
  of 
  

   the 
  electrometer 
  being 
  earthed. 
  In 
  parallel 
  with 
  the 
  secondary 
  

   of 
  the 
  induction-coil 
  is 
  a 
  battery 
  of 
  twelve 
  large 
  leyden-jars. 
  

   The 
  primary 
  of 
  the 
  coil 
  is 
  fed 
  from 
  the 
  alternating-current 
  

   side 
  of 
  a 
  Westinghouse 
  converter, 
  run 
  as 
  a 
  motor 
  on 
  the 
  direct- 
  

   current 
  side 
  with 
  current 
  from 
  storage-cells. 
  The 
  frequency 
  

   of 
  the 
  alternations 
  is 
  about 
  50 
  per 
  second. 
  The 
  primary 
  

   circuit 
  includes 
  a 
  variable 
  resistance 
  the 
  last 
  part 
  of 
  which 
  is 
  

   a 
  straight 
  manganin 
  wire 
  with 
  a 
  sliding 
  mercury-cup 
  contact. 
  

   The 
  difference 
  of 
  potential 
  between 
  the 
  two 
  plates 
  is 
  indicated 
  

   by 
  a 
  Kelvin 
  electrostatic 
  voltmeter. 
  Every 
  precaution 
  is 
  taken 
  

   to 
  screen 
  the 
  balance 
  from 
  electrical 
  influence, 
  and 
  to 
  prevent 
  

   the 
  glass 
  plate 
  becoming 
  electrified. 
  

  

  A 
  weighing 
  is 
  made 
  in 
  the 
  following 
  manner. 
  With 
  the 
  

   beam 
  resting 
  on 
  its 
  knife-edge, 
  the 
  mass 
  in 
  the 
  pan 
  being- 
  

   greater 
  than 
  that 
  required 
  to 
  balance 
  the 
  swinging 
  plate, 
  the 
  

   pointer 
  is 
  brought 
  up 
  to 
  the 
  middle 
  mark 
  of 
  the 
  scale, 
  the 
  

   pointer 
  and 
  scale 
  being 
  observed 
  through 
  a 
  telescope. 
  The 
  

   primary 
  circuit 
  is 
  closed 
  and 
  the 
  variable 
  resistance 
  altered 
  

   until 
  the 
  pointer 
  of 
  the 
  balance 
  swings 
  over, 
  the 
  reading 
  of 
  

   the 
  voltmeter 
  being 
  taken 
  by 
  a 
  second 
  observer 
  at 
  the 
  moment 
  

   when 
  this 
  occurs. 
  Two 
  such 
  experiments 
  have 
  to 
  be 
  made, 
  

  

  Phil. 
  Mag. 
  S. 
  6. 
  Vol. 
  3. 
  No. 
  18. 
  June 
  1902. 
  2 
  S 
  

  

  