﻿Relations 
  betioeen 
  JEther, 
  Matter, 
  and 
  Electricity. 
  373 
  

  

  directions 
  from 
  the 
  brushes. 
  The 
  contact 
  at 
  the 
  brushes 
  was 
  

   another' 
  source 
  of 
  difficulty. 
  The 
  lead 
  wires 
  were 
  led 
  side 
  

   by 
  side 
  from 
  the 
  rim 
  to 
  the 
  axis 
  of 
  the 
  wheel, 
  but 
  still 
  a 
  slight 
  

   alternating 
  current 
  existed 
  sufficient 
  to 
  give 
  a 
  throw 
  of 
  about 
  

   five 
  millimetres 
  on 
  the 
  galvanometer 
  scale 
  when 
  the 
  wheel 
  

   was 
  turned 
  quickly 
  through 
  a 
  half 
  -turn. 
  This 
  would 
  give 
  no 
  

   difficulty 
  at 
  high 
  speeds 
  unless 
  the 
  brush 
  were 
  thrown 
  off 
  

   periodically 
  so 
  as 
  to 
  act 
  as 
  a 
  commutator. 
  This 
  appeared 
  

   sometimes 
  to 
  occur 
  when 
  the 
  silver 
  tip 
  dug 
  a 
  small 
  cavity 
  in 
  

   the 
  plane 
  surface 
  of 
  the 
  spring 
  bearing 
  against 
  it. 
  This 
  was 
  

   overcome 
  by 
  frequently 
  smoothing 
  off 
  the 
  plane 
  surface 
  with 
  

   a 
  file. 
  This 
  difficulty 
  was 
  so 
  great 
  with 
  copper 
  contacts 
  that 
  

   it 
  was 
  necessary 
  to 
  use 
  silver, 
  as 
  was 
  mentioned 
  above. 
  

   Another 
  difficulty 
  lay 
  in 
  the 
  fact 
  that 
  the 
  inertia 
  was 
  so 
  

   great 
  that, 
  though 
  all 
  joints 
  were 
  carefully 
  soldered, 
  earth- 
  

   connexions 
  and 
  breaks 
  were 
  frequent 
  at 
  the 
  joints 
  or 
  in 
  the 
  

   wire 
  itself. 
  These 
  breaks 
  were 
  frequently 
  not 
  complete, 
  but 
  

   simply 
  gave 
  a 
  variable 
  contact 
  and 
  were 
  extremely 
  hard 
  to 
  

   locate 
  without 
  removing 
  all 
  connexions 
  and 
  replacing 
  with 
  new. 
  

   This 
  was 
  particularly 
  true 
  among 
  the 
  complicated 
  connexions 
  

   of 
  the 
  reversing 
  switch 
  in 
  the 
  later 
  experiments. 
  

  

  The 
  wheel 
  was 
  run 
  by 
  a 
  belt 
  from 
  a 
  one-sixth 
  horse- 
  power 
  

   electric 
  motor. 
  It 
  w 
  T 
  as 
  found 
  necessary 
  to 
  keep 
  the 
  frame 
  of 
  

   the 
  machine 
  in 
  electrical 
  connexion 
  with 
  the 
  gas 
  fixture, 
  in 
  

   order 
  to 
  avoid 
  violent 
  throws 
  of 
  the 
  galvanometer-needle 
  due 
  

   to 
  static 
  electricity. 
  The 
  speed 
  first 
  used 
  was 
  85 
  turns 
  per 
  

   second. 
  Later 
  a 
  larger 
  pulley 
  was 
  used 
  giving 
  125 
  turns, 
  

   and 
  this 
  was 
  sometimes 
  increased 
  to 
  150 
  by 
  shifting 
  the 
  

   brushes 
  on 
  the 
  motor 
  at 
  each 
  reversal. 
  If 
  this 
  was 
  done 
  

   after 
  a 
  high 
  speed 
  was 
  reached, 
  sparking 
  was 
  not 
  excessive. 
  

   In 
  the 
  early 
  experiments 
  the 
  motor 
  magnets 
  changed 
  the 
  

   zero 
  of 
  the 
  galvanometer 
  by 
  about 
  two 
  centimetres. 
  It 
  was 
  

   impossible 
  to 
  detect 
  any 
  unsteadiness 
  due 
  to 
  the 
  motor 
  when 
  

   running 
  ; 
  but 
  in 
  the 
  later 
  experiments 
  it 
  was 
  moved 
  to 
  the 
  

   farther 
  side 
  of 
  the 
  room 
  where 
  it 
  had 
  no 
  perceptible 
  effect 
  on 
  

   the 
  galvanometer. 
  

  

  In 
  taking 
  the 
  first 
  extended 
  set 
  of 
  readings, 
  the 
  wheel 
  was 
  

   run 
  in 
  one 
  direction 
  while 
  five 
  deflexions 
  were 
  taken. 
  Then 
  

   the 
  motor 
  was 
  reversed 
  by 
  reversing 
  the 
  current 
  in 
  the 
  

   armature, 
  so 
  as 
  to 
  affect 
  the 
  galvanometer 
  as 
  little 
  as 
  possible, 
  

   and 
  five 
  more 
  deflexions 
  were 
  recorded, 
  and 
  so 
  on. 
  A 
  

   reading 
  was 
  thrown 
  out 
  if 
  there 
  was 
  any 
  indication 
  that 
  it 
  

   had 
  been 
  affected 
  by 
  outside 
  influences. 
  The 
  wire 
  was 
  so 
  

   wound 
  on 
  the 
  wheel, 
  that 
  positive 
  electricity 
  lagging 
  behind 
  

   would 
  flow 
  to 
  the 
  galvanometer 
  through 
  the 
  end 
  of 
  the 
  axis 
  

   away 
  from 
  the 
  pulley 
  when 
  the 
  wheel 
  was 
  rotating 
  in 
  the 
  

  

  