﻿Z. 
  DeForest 
  — 
  Reflection 
  of 
  Hertzian 
  Waves. 
  59 
  

  

  It 
  was 
  observed 
  that 
  reflection 
  with 
  change 
  of 
  phase 
  occurred 
  

   at 
  any 
  bridge, 
  bend, 
  or 
  sudden 
  change 
  in 
  the 
  nature 
  of 
  the 
  

   parallel 
  wires. 
  The 
  mathematical 
  theory 
  for 
  such 
  phenom- 
  

   ena 
  has 
  been 
  considerably 
  developed, 
  notably 
  by 
  Drude.^ 
  

  

  Theory 
  shows 
  that 
  the 
  waves 
  are 
  reflected 
  at 
  the 
  open 
  ends 
  

   of 
  the 
  parallel 
  wires 
  withont 
  change 
  of 
  sign 
  of 
  electrification. 
  

   The 
  first 
  node, 
  therefore, 
  would 
  be 
  expected 
  at 
  one 
  quarter 
  of 
  

   the 
  wave-length 
  from 
  the 
  ends. 
  If 
  capacity 
  of 
  any 
  sort 
  is 
  

   attached 
  to 
  the 
  ends, 
  however, 
  a 
  change 
  of 
  phase 
  would 
  occur 
  

   by 
  the 
  reflection 
  and 
  this 
  node 
  be 
  found 
  nearer 
  the 
  ends, 
  by 
  an 
  

   amount 
  depending 
  on 
  this 
  capacity. 
  Even 
  for 
  bare 
  ends, 
  

   where 
  no 
  capacity 
  has 
  been 
  assumed, 
  it 
  was 
  early 
  noticed 
  that 
  

   a 
  change 
  of 
  phase 
  did 
  occur 
  ; 
  that 
  the 
  distance 
  to 
  the 
  first 
  node 
  

   was 
  unexpectedly 
  small. 
  

  

  B. 
  B'3 
  1 
  Y)' 
  

  

  To 
  investigate 
  this 
  part 
  of 
  the 
  subject 
  was 
  the 
  aim 
  of 
  the 
  

   experiments 
  described 
  in 
  this 
  paper. 
  

  

  The 
  form 
  of 
  exciter 
  adopted 
  was 
  a 
  compromise 
  between 
  the 
  

   Lecher 
  and 
  the 
  Blondlot 
  forms. 
  In 
  the 
  latter 
  two 
  pieces 
  of 
  

   wire, 
  bent 
  semi-circularly, 
  take 
  the 
  place 
  of 
  the 
  primary 
  con- 
  

   denser 
  plates 
  of 
  the 
  Lecher 
  system. 
  To 
  one 
  end 
  of 
  each 
  wire 
  

   is 
  fastened 
  a 
  small 
  polished 
  zinc 
  ball, 
  forming 
  the 
  spark 
  gap. 
  

   To 
  the 
  other 
  ends 
  are 
  often 
  attached 
  a 
  pair 
  of 
  metal 
  plates, 
  

   placed 
  oppositely, 
  to 
  add 
  capacity 
  to 
  the 
  whole. 
  

  

  Outside 
  of 
  the 
  approximate 
  circle 
  thus 
  formed 
  is 
  bent 
  a 
  

   single 
  wire 
  for 
  the 
  secondary. 
  Its 
  two 
  ends 
  are 
  brought 
  near 
  

   together, 
  and 
  from 
  these 
  lead 
  out 
  the 
  long 
  parallel 
  wires. 
  

  

  This 
  Blondlot 
  system 
  has 
  a 
  greater 
  damping 
  than 
  the 
  

   Lecher. 
  Its 
  individual 
  period 
  is 
  less 
  marked, 
  and 
  the 
  frequen- 
  

   cies 
  of 
  the 
  waves 
  found 
  npon 
  the 
  parallel 
  wires 
  are 
  almost 
  

   altogether 
  dependent 
  upon 
  the 
  vibrating 
  systems 
  there 
  formed 
  

   by 
  bridges 
  arbitrarily 
  located. 
  A 
  great 
  advantage 
  resulting 
  is 
  

   that 
  a 
  large 
  range 
  of 
  wave-lengths 
  may 
  thus 
  be 
  had 
  from 
  one 
  

   oscillator. 
  

  

  In 
  the 
  exciter 
  used 
  throughout, 
  instead 
  of 
  wires, 
  two 
  narrow 
  

   zinc 
  plates, 
  A 
  A', 
  each 
  15^°' 
  long, 
  1*5'^°^ 
  wide 
  in 
  the 
  middle, 
  and 
  

   elliptical 
  in 
  shape, 
  formed 
  the 
  primary. 
  These 
  were 
  bent 
  as 
  

   shown 
  in 
  Fig. 
  1, 
  and 
  each 
  carried 
  a 
  zinc 
  ball, 
  8™"^ 
  in 
  diameter, 
  

  

  *Wied. 
  Ann., 
  lii, 
  p. 
  499; 
  liii, 
  p. 
  751. 
  

  

  