﻿L. 
  DeForest 
  — 
  Reflection 
  of 
  Hertzian 
  Waves. 
  71 
  

  

  the 
  third, 
  deflections 
  23 
  to 
  6 
  respectively. 
  Rotating 
  the 
  first 
  

   analyzer, 
  the 
  average 
  results 
  shown 
  in 
  the 
  following 
  table 
  were 
  

   obtained. 
  

  

  Angle 
  of 
  strips 
  

  

  

  Equivalent 
  

  

  

  from 
  vertical. 
  

  

  Deflection. 
  

  

  deflection. 
  

  

  Cosine 
  of 
  angle 
  

  

  0- 
  

  

  26 
  

  

  1-0 
  

  

  1^0 
  

  

  22-5° 
  

  

  18 
  

  

  •7 
  

  

  •9 
  

  

  45-0 
  

  

  9 
  

  

  •35 
  

  

  •7 
  

  

  67-5 
  

  

  4 
  

  

  •16 
  

  

  •4 
  

  

  90-0 
  

  

  

  

  •0 
  

  

  •0 
  

  

  Using 
  two 
  gratings, 
  the 
  segments 
  of 
  the 
  one 
  nearest 
  the 
  coherer 
  

   lying 
  horizontal, 
  and 
  those 
  of 
  the 
  one 
  in 
  front 
  at 
  an 
  angle 
  of 
  

   45° 
  to 
  these, 
  an 
  occasional 
  deflection 
  was 
  secured. 
  More 
  often, 
  

   however, 
  the 
  resolved 
  vertical 
  component 
  from 
  the 
  first 
  was 
  

   too 
  weak 
  to 
  penetrate 
  the 
  second. 
  With 
  both 
  gratings 
  vertical 
  

   good 
  deflections 
  were 
  secured. 
  

  

  Thus 
  the 
  earlier 
  assumption 
  that 
  a 
  beam, 
  as 
  it 
  were, 
  of 
  plane- 
  

   polarized 
  displacement 
  traveled 
  on 
  beyond 
  the 
  ends 
  of 
  the 
  

   two 
  wires, 
  was 
  justified. 
  It 
  is 
  also 
  shown 
  very 
  strikingly 
  that 
  

   it 
  is 
  the 
  component 
  'peryendicular 
  to 
  the 
  elements 
  of 
  the 
  grat- 
  

   ing 
  that 
  passes 
  through 
  it. 
  At 
  a 
  very 
  little 
  distance 
  from 
  the 
  

   ends 
  this 
  plane-polarized 
  beam 
  is 
  lost 
  in 
  the 
  expanding 
  (spheri- 
  

   cal) 
  waves, 
  or 
  in 
  those 
  diffusely 
  reflected. 
  

  

  Yery 
  much 
  might 
  be 
  determined 
  as 
  to 
  the 
  exact 
  nature 
  of 
  

   the 
  field 
  in 
  the 
  neighborhood 
  of 
  the 
  wires 
  by 
  such 
  use 
  of 
  the 
  

   coherer 
  — 
  devising 
  proper 
  shields, 
  analysers, 
  resonators, 
  etc., 
  

   although 
  as 
  now 
  used 
  it 
  is 
  only 
  roughly 
  quantitative. 
  

  

  My 
  best 
  thanks 
  are 
  due 
  Professor 
  Wright, 
  for 
  the 
  facilities 
  of 
  

   the 
  Sloane 
  Laboratory 
  put 
  at 
  my 
  disposal, 
  and 
  for 
  kindly 
  in- 
  

   terest 
  and 
  valuable 
  suggestions 
  throughout 
  the 
  course 
  of 
  this 
  

   work, 
  and 
  to 
  Prof. 
  Gibbs 
  for 
  helpful 
  advice. 
  

  

  Sloane 
  Physical 
  Laboratory, 
  

  

  Yale 
  University, 
  May 
  31, 
  1899. 
  

  

  