﻿968 
  Mr. 
  R. 
  W. 
  Lawson 
  on 
  the 
  Spectra 
  of 
  

  

  of 
  short 
  lengths 
  of 
  one-inch 
  glass 
  tubing 
  made 
  to 
  grip 
  the 
  

   tube 
  by 
  means 
  of 
  a 
  tightly 
  packed 
  annulus 
  of 
  asbestos 
  string, 
  

   the 
  resulting 
  cups 
  being 
  filled 
  with 
  mercury. 
  The 
  apparatus 
  

   was 
  in 
  permanent 
  connexion 
  with 
  a 
  U-tube 
  mercury 
  mano- 
  

   meter, 
  and 
  by 
  means 
  of 
  suitable 
  stopcocks 
  it 
  could 
  be 
  connected 
  

   respectively 
  with 
  a 
  Toepler 
  pump 
  and 
  a 
  gas-collecting 
  tube. 
  

   This 
  last 
  contained 
  phosphorus 
  pentoxide, 
  and 
  the 
  gases 
  were 
  

   allowed 
  to 
  stand 
  in 
  contact 
  with 
  this 
  drying 
  agent 
  during 
  

   several 
  days 
  before 
  use. 
  

  

  For 
  the 
  photography 
  of 
  the 
  spectra 
  obtained, 
  a 
  Hilger 
  

   fixed-arm 
  spectrograph 
  was 
  used. 
  Owing 
  to 
  the 
  prism 
  being 
  

   of 
  dense 
  glass, 
  the 
  limits 
  of 
  spectral 
  lines 
  obtainable 
  were 
  

   about 
  X 
  6700 
  and 
  A, 
  4000. 
  By 
  the 
  use 
  of 
  Panchromatic 
  plates, 
  

   as 
  sold 
  by 
  Wratten 
  and 
  Wainwright, 
  the 
  less 
  refrangible 
  

   parts 
  of 
  the 
  spectrum 
  were 
  included 
  on 
  the 
  photographs. 
  A 
  

   wave-length 
  chart 
  was 
  constructed 
  by 
  measuring 
  up 
  standard 
  

   lines 
  in 
  an 
  iron 
  arc 
  spectrum, 
  and 
  plotting 
  the 
  wave-lengths 
  

   of 
  these 
  against 
  the 
  reading 
  as 
  given 
  by 
  a 
  dividing 
  engine. 
  

   For 
  accurate 
  measurements 
  of 
  spectral 
  lines, 
  the 
  spectrum 
  in 
  

   question 
  was 
  measured 
  up 
  and 
  the 
  wave-lengths 
  found 
  from 
  

   the 
  chart. 
  For 
  purposes 
  of 
  identification 
  only, 
  it 
  was 
  found 
  

   to 
  be 
  most 
  convenient 
  to 
  use 
  an 
  iron 
  arc 
  plate 
  on 
  which 
  the 
  

   wave-lengths 
  of 
  the 
  principal 
  lines 
  had 
  been 
  marked. 
  By 
  

   superposition 
  of 
  this 
  plate 
  upon 
  the 
  one 
  to 
  be 
  examined, 
  

   a 
  tolerably 
  accurate 
  estimate 
  of 
  wave-lengths 
  could 
  be 
  

   obtained. 
  

  

  For 
  the 
  measurement 
  of 
  voltage 
  across 
  the 
  discharge 
  tube 
  

   a 
  Kelvin 
  electrostatic 
  voltmeter 
  was 
  used 
  in 
  a 
  manner 
  sub- 
  

   sequently 
  to 
  be 
  figured. 
  The 
  discharge 
  current 
  was 
  measured 
  

   by 
  passing 
  it 
  through 
  a 
  fine 
  platinum 
  wire, 
  which 
  heated 
  a 
  

   copper-constantan 
  thermo-couple, 
  and 
  the 
  resulting 
  electro- 
  

   motive 
  force 
  was 
  registered 
  on 
  a 
  Paul 
  single-pivot 
  galva- 
  

   nometer. 
  Another 
  form 
  of 
  current-measurer 
  used 
  consisted 
  

   of 
  a 
  fine 
  copper 
  wire, 
  through 
  which 
  the 
  current 
  passed. 
  

   This 
  was 
  enclosed 
  in 
  one 
  bulb 
  of 
  a 
  differential 
  air 
  thermometer, 
  

   and 
  the 
  current 
  measured 
  by 
  noting 
  the 
  increased 
  pressure 
  in 
  

   the 
  bulb 
  due 
  to 
  the 
  heating 
  wire. 
  Previous 
  calibration 
  of 
  

   these 
  ammeters 
  by 
  means 
  of 
  direct 
  current 
  was 
  necessary 
  to 
  

   obtain 
  the 
  current-reading 
  in 
  amperes. 
  

  

  An 
  induction 
  coil 
  was 
  kept 
  conveniently 
  near 
  the 
  discharge 
  

   tubes, 
  so 
  that 
  the 
  discharge 
  as 
  obtained 
  by 
  the 
  Poulsen 
  

   generator 
  could 
  be 
  compared 
  with 
  either 
  the 
  condensed 
  or 
  

   theuncondensed 
  induction-coil 
  discharges. 
  Direct 
  comparison 
  

   could 
  therefore 
  be 
  made 
  under 
  the 
  same 
  conditions 
  of 
  pressure 
  

   and 
  mean 
  current 
  density 
  in 
  the 
  discharge 
  tube. 
  

  

  In 
  addition 
  to 
  the 
  measure 
  of 
  current, 
  voltage, 
  &c, 
  of 
  the 
  

  

  