﻿Properties 
  of 
  Artists' 
  Colours. 
  731 
  

  

  burner. 
  The 
  light 
  reflected 
  from 
  the 
  white 
  paper 
  formed 
  

   the 
  standard 
  spectrum. 
  

  

  It 
  is 
  difficult 
  to 
  determine 
  the 
  accuracy 
  of 
  the 
  obser- 
  

   vations, 
  but 
  the 
  errors 
  are 
  practically 
  entirely 
  due 
  to 
  the 
  

   observer, 
  as 
  the 
  instrument 
  was 
  of 
  exceptionally 
  good 
  

   quality, 
  and 
  most 
  carefully 
  adjusted. 
  Readings 
  were 
  most 
  

   difficult 
  when 
  the 
  light 
  was 
  of 
  high 
  intensity, 
  and 
  in 
  

   such 
  cases 
  their 
  number 
  was 
  increased 
  so 
  as 
  to 
  give 
  a 
  

   fairly 
  reliable 
  mean. 
  It 
  was 
  also 
  difficult 
  to 
  match 
  the 
  

   intensities 
  in 
  the 
  red, 
  probably 
  because 
  in 
  this 
  region 
  the 
  

   dispersion 
  is 
  less, 
  and 
  the 
  colour 
  seen 
  not 
  so 
  pure 
  as 
  in 
  

   other 
  parts. 
  

  

  Some 
  diagrams 
  (pp. 
  732-33) 
  are 
  given 
  showing 
  the 
  nature 
  

   of 
  the 
  absorption. 
  In 
  them 
  the 
  fraction 
  of 
  the 
  total 
  light 
  re- 
  

   flected 
  by 
  the 
  paint 
  is 
  plotted 
  against 
  the 
  wave-length. 
  They 
  

   show 
  at 
  a 
  glance 
  the 
  true 
  nature 
  of 
  the 
  reflected 
  light, 
  and 
  

   also 
  the 
  percentage 
  of 
  white 
  light, 
  this 
  being 
  given 
  by 
  the 
  

   ordinate 
  of 
  the 
  lowest 
  point 
  of 
  the 
  curve. 
  It 
  is 
  also 
  worthy 
  

   of 
  note 
  that 
  in 
  all 
  cases 
  the 
  region 
  of 
  maximum 
  intensity 
  

   extends 
  over 
  a 
  considerable 
  range 
  of 
  wave-length, 
  that 
  is 
  to 
  

   say, 
  the 
  apparent 
  colour 
  of 
  a 
  paint 
  is 
  never 
  even 
  approxi- 
  

   mately 
  monochromatic, 
  although 
  to 
  the 
  eye 
  it 
  may 
  appear 
  

   to 
  be 
  so. 
  

  

  In 
  addition 
  to 
  this, 
  the 
  effect 
  of 
  mixing 
  one 
  or 
  more 
  paints 
  

   may 
  be 
  approximately 
  obtained 
  from 
  the 
  curves. 
  Thus 
  in 
  

   the 
  simplest 
  case 
  of 
  a 
  mixture 
  of 
  equal 
  quantities 
  of 
  two 
  

   paints, 
  the 
  absorption-curve 
  of 
  the 
  resultant 
  colour 
  will 
  be 
  

   obtained 
  by 
  multiplying 
  together 
  the 
  ordinates 
  of 
  the 
  two 
  

   curves 
  for 
  each 
  wave-length, 
  and 
  joining 
  the 
  resultant 
  points. 
  

   An 
  example 
  of 
  this 
  is 
  given 
  for 
  a 
  mixture 
  of 
  emerald 
  

   green 
  and 
  madder 
  orange 
  ; 
  these, 
  although 
  both 
  very 
  

   bright 
  colours, 
  will 
  only 
  combine 
  to 
  give 
  a 
  brownish- 
  

   yellow 
  of 
  low 
  intensity. 
  This 
  is 
  of 
  course 
  only 
  an 
  ap- 
  

   proximation, 
  but 
  it 
  helps 
  to 
  illustrate 
  the 
  fact 
  that 
  in 
  

   order 
  to 
  obtain 
  a 
  bright 
  colour 
  from 
  a 
  mixture 
  of 
  others, 
  

   it 
  is 
  necessary 
  that 
  these 
  should 
  have 
  some 
  part 
  of 
  the 
  

   brightest 
  portion 
  of 
  their 
  spectra 
  in 
  common, 
  and 
  conse- 
  

   quently 
  the 
  most 
  efficient 
  paints 
  for 
  producing 
  bright 
  colours, 
  

   when 
  mixed 
  with 
  others, 
  are 
  those 
  possessing 
  broad 
  maxima 
  

   of 
  intensity. 
  

  

  In 
  conclusion, 
  I 
  should 
  like 
  to 
  express 
  my 
  best 
  thanks 
  to 
  

   Sir 
  William 
  Ramsay 
  for 
  suggesting 
  this 
  work 
  to 
  me, 
  and 
  

   for 
  his 
  frequent 
  help 
  in 
  it 
  ; 
  and 
  also 
  to 
  Dr. 
  Mees 
  for 
  the 
  loan 
  

   of 
  his 
  spectrophotometer. 
  

   University 
  College, 
  London, 
  January 
  1908. 
  

  

  