﻿^-effect 
  in 
  Electrical 
  Oscillators. 
  431 
  

  

  oscillator 
  had 
  massive 
  spheres 
  of 
  about 
  4 
  cms. 
  diameter 
  the 
  

   action 
  on 
  a 
  resonator 
  first 
  ceased 
  at 
  about 
  11 
  metres 
  ; 
  while, 
  

   on 
  the 
  other 
  hand, 
  it 
  ceased 
  at 
  about 
  6 
  metres 
  when 
  these 
  

   spheres 
  were 
  replaced 
  by 
  hollow 
  ones 
  of 
  equal 
  external 
  

   diameter 
  and 
  about 
  1*5 
  mm. 
  of 
  wall-thickness. 
  Two 
  others 
  

   with 
  yet 
  decidedly 
  thinner 
  walls 
  gave 
  still 
  smaller 
  effects. 
  

   The 
  knobs 
  connected 
  to 
  the 
  machine 
  * 
  might, 
  however, 
  quite 
  

   as 
  well 
  be 
  hollow 
  as 
  solid, 
  and 
  also, 
  within 
  certain 
  limits, 
  

   might 
  be 
  of 
  less 
  or 
  greater 
  diameter 
  without 
  the 
  efficiency 
  of 
  

   the 
  oscillator 
  being 
  thereby 
  affected." 
  

  

  As 
  just 
  remarked, 
  these 
  are 
  the 
  only 
  references 
  to 
  the 
  

   subject 
  that 
  I 
  have 
  found. 
  Indeed, 
  considering 
  the 
  import- 
  

   ance 
  of 
  the 
  oscillating 
  doublet, 
  it 
  is 
  somewhat 
  surprising 
  

   that 
  so 
  little 
  has 
  been 
  done 
  in 
  the 
  direction 
  of 
  a 
  satisfactory 
  

   theory 
  of 
  it. 
  

  

  2. 
  The 
  Oscillators. 
  

  

  The 
  oscillators 
  were 
  of 
  two 
  types, 
  cylindrical 
  and 
  spherical 
  

   respectively. 
  The 
  cylinders 
  were 
  circular 
  in 
  section 
  with, 
  

   hemispherical 
  ends 
  ; 
  and 
  were 
  25 
  cms. 
  in 
  diameter 
  and 
  

   125 
  cms. 
  long. 
  For 
  exciting 
  the 
  oscillator 
  an 
  induction- 
  

   coil 
  capable 
  of 
  giving 
  a 
  12*5 
  cm. 
  spark 
  was 
  used. 
  It 
  was 
  

   placed 
  in 
  a 
  metal-lined 
  box 
  and 
  connexions 
  were 
  made 
  to 
  it 
  

   as 
  follows. 
  

  

  The 
  cylinders 
  C, 
  D 
  (fig. 
  1) 
  were 
  placed 
  along 
  the 
  focal 
  

  

  Fio-. 
  l. 
  

  

  line 
  of 
  a 
  parabolic 
  reflector 
  M, 
  made 
  of 
  sheet 
  zinc, 
  being 
  held 
  

   in 
  position 
  by 
  wooden 
  supports 
  hollowed 
  to 
  receive 
  them. 
  

   To 
  the 
  ends 
  of 
  the 
  wires 
  L, 
  L 
  were 
  soldered 
  strips 
  r, 
  r 
  of 
  

   brush 
  copper 
  about 
  5 
  mm. 
  wide 
  bent 
  into 
  circular 
  form 
  as 
  

   shown 
  in 
  G 
  (fig. 
  1). 
  The 
  ends 
  of 
  the 
  cylinders 
  were 
  slipped 
  

   into 
  these 
  copper 
  holders, 
  and 
  a 
  rubber 
  band 
  placed 
  about 
  

   the 
  holder 
  drew 
  it 
  closely 
  to 
  the 
  cylinder, 
  thus 
  ensuring 
  good 
  

   and 
  ample 
  contact. 
  The 
  wires 
  were 
  then 
  led 
  through 
  a 
  

   wooden 
  block 
  It 
  (as 
  shown 
  in 
  fig. 
  2, 
  p. 
  432) 
  , 
  into 
  which 
  they 
  

   were 
  securely 
  wedged, 
  and 
  then 
  through 
  small 
  glass 
  tubes 
  

   firmly 
  held 
  together 
  by 
  rubber 
  tape 
  ; 
  and 
  were 
  finally 
  joined 
  

  

  * 
  Influence 
  electrical 
  machine. 
  

  

  