﻿Minute 
  Metal 
  Particles 
  for 
  Light-waves. 
  

  

  403 
  

  

  when 
  heated, 
  the 
  spectrum 
  appearing 
  as 
  in 
  fig. 
  2, 
  b. 
  I 
  do 
  not 
  

   feel 
  sure 
  that 
  I 
  have 
  determined 
  the 
  exact 
  cause 
  of 
  the 
  per- 
  

   manent 
  colour-changes, 
  due 
  to 
  heating. 
  It 
  doubtless 
  depends 
  

   on 
  the 
  partial 
  evaporation 
  of 
  the 
  metallic 
  particles, 
  and 
  we 
  

  

  3 
  

  

  must 
  bear 
  in 
  mind 
  that 
  the 
  period 
  of 
  vibration 
  depends, 
  not 
  

   only 
  on 
  their 
  size, 
  but 
  on 
  the 
  distance 
  between 
  them 
  ; 
  both 
  

   of 
  which 
  quantities 
  will 
  be 
  affected 
  by 
  evaporation. 
  The 
  

   matter 
  is 
  probably 
  further 
  complicated 
  by 
  the 
  fact 
  that, 
  in 
  

   some 
  cases, 
  I 
  have 
  had 
  films 
  made 
  up 
  of 
  particles 
  of 
  two 
  

   different 
  sizes. 
  In 
  this 
  case 
  it 
  is 
  not 
  difficult 
  to 
  imagine 
  

   that 
  the 
  smaller 
  ones 
  are 
  completely 
  vaporized, 
  while 
  the 
  

   larger 
  ones, 
  which 
  previously 
  were 
  of 
  too 
  great 
  size 
  to 
  produce 
  

   colour, 
  are 
  brought 
  down 
  to 
  dimensions 
  small 
  enough 
  to 
  

   enable 
  them 
  to 
  resonate. 
  This 
  might 
  explain 
  the 
  change 
  

   from 
  purple 
  to 
  blue. 
  Experimental 
  evidence 
  is, 
  however, 
  

   rather 
  against 
  this 
  hypothesis 
  ; 
  and 
  for 
  the 
  present 
  I 
  prefer 
  to 
  

   leave 
  the 
  cause 
  of 
  the 
  permanent 
  change 
  in 
  colour 
  an 
  

   open 
  question. 
  

  

  The 
  changes 
  produced 
  by 
  cooling 
  are 
  much 
  more 
  startling, 
  

   and 
  of 
  greater 
  interest, 
  for 
  the 
  cause 
  has 
  been 
  discovered. 
  

   We 
  may 
  summarize 
  these 
  effects 
  something 
  as 
  follows 
  : 
  — 
  

   Pink 
  and 
  purple 
  films 
  become 
  blue 
  when 
  cooled, 
  the 
  original 
  

   colour 
  returning- 
  as 
  soon 
  as 
  they 
  are 
  warmed 
  up 
  again. 
  A 
  

   pale 
  green 
  film 
  becomes 
  deep 
  violet, 
  while 
  in 
  some 
  cases 
  I 
  

   have 
  found 
  that 
  films, 
  originally 
  blue, 
  become 
  almost 
  per- 
  

   fectly 
  transparent 
  when 
  cooled, 
  the 
  blue 
  colour 
  returning 
  

   subsequently. 
  These 
  effects 
  are 
  best 
  shown 
  by 
  touching 
  the 
  

   outside 
  of 
  the 
  bulb 
  with 
  a 
  small 
  piece 
  of 
  ice. 
  At 
  first 
  sight, 
  

   it 
  seems 
  as 
  if 
  the 
  only 
  possible 
  effect 
  of 
  cooling 
  would 
  be 
  a 
  

   slight 
  reduction 
  of 
  the 
  size 
  of 
  the 
  particles, 
  and 
  a 
  contraction 
  

   of 
  the 
  supporting 
  wall. 
  That 
  this 
  should 
  be 
  sufficient 
  to 
  

   affect 
  the 
  period 
  of 
  their 
  vibration 
  was 
  almost 
  unthinkable. 
  

   If 
  it 
  were 
  so, 
  a 
  further 
  lowering 
  of 
  temperature 
  should 
  

   produce 
  further 
  changes, 
  but 
  I 
  found 
  that 
  it 
  was 
  quite 
  

   immaterial 
  whether 
  the 
  bulb 
  was 
  touched 
  with 
  ice, 
  or 
  with 
  a 
  

   mixture 
  of 
  solid 
  carbonic 
  acid 
  and 
  ether 
  ; 
  the 
  resulting 
  colour 
  

   being 
  the 
  same 
  in 
  both 
  cases. 
  I 
  have 
  frequently 
  obtained 
  

  

  