JS'ote  on  Talbot's  Lines.  531 
feeble  polarization  current  or  other  signs  of  electrolysis  have 
been  observed,  we  have  both  metallic  and  electrolytic  con- 
ductivity associated  in  the  same  substance. 
These  experiments  were  carried  out  at  the  Cavendish 
Laboratory,  aud  I  gladly  take  this  opportunity  of  thanking 
Professor  Thomson  for  his  advice  and  interest  during  the 
investigation. 
XLIV.  A  JSrote  on  Talbot's  Lines. 
By  James  Walker,  M.A.,  Oxford  *. 
IT  is  sometimes  felt  that  Stokes's  explanation  of  the 
"polarity  "  of  Talbot's  lines  is  mathematical  rather  than 
physical,  so  that  those  who  are  unable  to  follow  the  analysis 
still  require  an  adequate  reason  for  the  fact  that  the  retarding 
plate  must  be  inserted  on  the  one  side  of  the  aperture  rather 
than  on  the  other. 
Prof.  Schusterf  has  indeed  supplied  this  want  by  considering 
the  source  of  light  to  be  due  to  a  succession  of  impulsive 
velocities  ;  but  there  is  perhaps  still  room  for  an  elementary 
explanation  of  the  phenomenon  on  the  old  familiar  lines  of 
regarding  the  light  as  resolved  into  a  congeries  of  mono- 
chromatic constituents. 
1.  When  a  stream  of  monochromatic  light,  coming  from 
a  distant  slit,  falls  upon  the  object-glass  of  a  telescope  that 
is  limited  by  a  rectangular  aperture  with  its  sides  parallel  to 
the  luminous  line,  it  is  easily  shown  that  the  diffraction- 
pattern  in  the  focal  plane  of  the  telescope  is  characterised  by  a 
series  of  dark  lines,  arranged  at  equal  distances  on  either  side 
of  the  geometrical  image  of  the  slit.  When  half  the  aperture 
is  covered  by  a  retarding  plate,  the  minima  of  an  even  order 
retain  their  former  positions,  but  those  of  an  odd  order 
are  displaced  towards  the  side  on  which  the  plate  is  placed 
by  an  amount  depending  upon  the  retardation  introduced  by 
the  plate. 
Let  us  now  suppose  that  the  primitive  light  is  white  and 
that  its  monochromatic  constituents  have  been  separated 
before  reaching  the  aperture  by  a  prism  or  a  grating,  so  that 
the  different  colours  occupy  different  angular  positions  in  the 
field  of  view. 
The  minima  of  an  even  order  will  now  disappear  ;  for  on 
account  of  the  dispersion  there  will  be  a  gradual  shift  of  the 
*  Communicated  by  the  Physical  Society  :  read  February  23, 1906. 
f  Au  Introduction  to  the  Theory  of  Optics,  p.  329 ;  Phil.  Mag.  [0J 
vol.  vii.  p.  1  (1004). 
