﻿Chemistry 
  and 
  Physics. 
  149 
  

  

  were: 
  petroleum 
  ether, 
  carbon 
  tetrachloride, 
  benzol, 
  carbon 
  

   bisulphide, 
  and 
  bromonaphthalene. 
  

  

  The 
  arrangement 
  of 
  the 
  parts 
  of 
  the 
  apparatus 
  may 
  be 
  very 
  

   briefly 
  described 
  as 
  follows. 
  The 
  long 
  axis 
  of 
  the 
  trough 
  (a 
  

   rectangular 
  parallelepiped) 
  is 
  parallel 
  to 
  the 
  optic 
  axis 
  of 
  the 
  

   collimator, 
  so 
  that 
  the 
  end 
  windows 
  of 
  the 
  vessel 
  are 
  approxi- 
  

   mately 
  normal 
  to 
  this 
  axis. 
  The 
  standard 
  prism 
  is 
  placed 
  in 
  

   the 
  liquid 
  with 
  its 
  hypothenuse 
  face 
  in 
  contact 
  with 
  a 
  side 
  wall 
  

   of 
  the 
  trough, 
  hence 
  its 
  refracting 
  edge 
  and 
  its 
  hypothenuse 
  

   surface 
  are 
  parallel 
  respectively 
  to 
  the 
  slit 
  and 
  to 
  the 
  optic 
  axis 
  

   of 
  the 
  collimator. 
  The 
  specimen 
  to 
  be 
  investigated 
  is 
  placed 
  

   in 
  the 
  liquid 
  in 
  the 
  path 
  of 
  the 
  beam 
  of 
  light, 
  either 
  in 
  front 
  

   of, 
  or 
  behind, 
  the 
  prism 
  of 
  reference. 
  A 
  suitable 
  aperture 
  in 
  

   an 
  opaque 
  screen 
  so 
  limits 
  the 
  cross-section 
  of 
  the 
  beam 
  that 
  

   only 
  light 
  which 
  has 
  traversed 
  the 
  specimen 
  may 
  enter 
  the 
  

   telescope. 
  

  

  The 
  practical 
  application 
  of 
  the 
  new 
  development 
  to 
  the 
  

   determination 
  of 
  the 
  index 
  of 
  refraction 
  of 
  a 
  given 
  inviolable 
  

   lens 
  will 
  now 
  be 
  outlined. 
  By 
  preliminary 
  trials 
  with 
  both 
  a 
  

   standard 
  prism 
  and 
  the 
  lens 
  in 
  question 
  immersed 
  in 
  a 
  mixture 
  

   of 
  two 
  liquids, 
  — 
  one 
  of 
  high 
  and 
  one 
  of 
  low 
  index, 
  — 
  it 
  is 
  neither 
  

   difficult 
  nor 
  tedious 
  to 
  select 
  that 
  one 
  of 
  the 
  calibrated 
  prisms 
  

   whose 
  index 
  is 
  nearest 
  to 
  that 
  of 
  the 
  lens, 
  and 
  to 
  obtain 
  a 
  mix- 
  

   ture 
  bearing 
  the 
  same 
  refractive 
  relation 
  to 
  the 
  lens. 
  The 
  lens 
  

   being 
  removed, 
  the 
  telescope 
  is 
  carefully 
  focussed 
  on 
  each 
  

   chosen 
  spectral 
  line 
  (C, 
  D, 
  and 
  F, 
  say) 
  and 
  readings 
  are 
  made 
  

   both 
  of 
  the 
  position 
  of 
  the 
  ocular 
  with 
  respect 
  to 
  its 
  linear 
  scale 
  

   and 
  of 
  the 
  angular 
  displacement 
  of 
  the 
  telescope 
  with 
  reference 
  

   to 
  the 
  direction 
  of 
  the 
  beam 
  emergent 
  from 
  the 
  trough 
  when 
  

   the 
  prism 
  is 
  not 
  immersed. 
  This 
  gives 
  approximate 
  values 
  of 
  

   the 
  quantities 
  which 
  require 
  slight 
  correction. 
  The 
  lens 
  is 
  then 
  

   replaced 
  in 
  the 
  liquid 
  with 
  the 
  result 
  that 
  the 
  telescope 
  is 
  

   thrown 
  out 
  of 
  focus 
  slightly 
  because 
  of 
  the 
  lack 
  of 
  absolute 
  

   agreement 
  between 
  the 
  indices 
  of 
  refraction 
  of 
  the 
  liquid 
  and 
  

   lens 
  material. 
  By 
  adding 
  a 
  drop 
  or 
  two 
  of 
  the 
  appropriate 
  

   component 
  of 
  the 
  liquid 
  mixture, 
  and 
  repointing 
  and 
  refocus- 
  

   sing 
  the 
  telescope, 
  it 
  is 
  a 
  simple 
  matter 
  to 
  obtain 
  two 
  or 
  more 
  

   pairs 
  of 
  values 
  of 
  the 
  change 
  in 
  the 
  deviation, 
  AD, 
  and 
  of 
  the 
  

   change 
  in 
  the 
  linear 
  displacement 
  of 
  the 
  ocular, 
  Ax. 
  Obviously, 
  

   the 
  range 
  of 
  liquid 
  mixtures 
  must 
  be 
  so 
  regulated 
  as 
  to 
  

   "bracket" 
  the 
  value 
  Ax 
  = 
  0, 
  thus 
  enabling 
  a 
  graph 
  to 
  be 
  

   plotted 
  for 
  the 
  observed 
  coordinates 
  (Ax, 
  AD) 
  and 
  the 
  value 
  of 
  

   AD 
  corresponding 
  to 
  Ax 
  = 
  to 
  be 
  obtained 
  by 
  (usually 
  linear) 
  

   interpolation. 
  The 
  value 
  of 
  the 
  deviation, 
  D, 
  which 
  would 
  have 
  

   been 
  observed 
  if 
  the 
  agreement 
  between 
  the 
  index 
  of 
  refrac- 
  

   tion 
  of 
  the 
  liquid 
  and 
  that 
  of 
  the 
  lens 
  had 
  been 
  perfect, 
  is 
  now 
  

   known. 
  A 
  fairly 
  simple 
  and 
  rapidly 
  converging 
  series 
  formula 
  

   is 
  given 
  (without 
  proof) 
  in 
  the 
  paper 
  which 
  expresses 
  the 
  

  

  