﻿Heat 
  produced 
  by 
  Emission 
  of 
  Ions 
  from 
  Hot 
  Bodies. 
  473 
  

  

  thermal 
  changes 
  which 
  take 
  place 
  when 
  the 
  thermionic 
  

   current 
  is 
  turned 
  on 
  or 
  off 
  all 
  occur 
  within 
  a 
  few 
  seconds. 
  

   In 
  fact 
  in 
  our 
  experiments 
  the 
  time 
  within 
  which 
  these 
  

   changes 
  were 
  completed 
  was 
  comparable 
  with 
  the 
  period 
  of 
  

   the 
  galvanometer 
  used 
  in 
  the 
  Wheatstone's 
  bridge. 
  With 
  

   the 
  lime-coated 
  platinum 
  wires, 
  under 
  otherwise 
  identical 
  

   circumstances, 
  it 
  was 
  questionable 
  whether 
  the 
  changes 
  were 
  

   complete 
  in 
  a 
  time 
  comparable 
  with 
  ten 
  minutes. 
  The 
  

   precise 
  nature 
  of 
  the 
  changes 
  also 
  depended 
  rather 
  definitely 
  

   on 
  the 
  mode 
  of 
  preparation 
  of 
  the 
  lime-coated 
  surface. 
  

   Different 
  wires 
  presented 
  different 
  peculiarities 
  of 
  behaviour, 
  

   but 
  there 
  were 
  certain 
  features 
  common 
  to 
  the 
  behaviour 
  of 
  

   all 
  the 
  wires 
  examined, 
  which 
  were 
  briefly 
  as 
  follows 
  : 
  — 
  

   Upon 
  allowing 
  conditions 
  to 
  become 
  steady, 
  with 
  the 
  Wheat- 
  

   stone's 
  bridge 
  balanced 
  it 
  was 
  found 
  that 
  turning 
  on 
  the 
  

   thermionic 
  current 
  caused 
  the 
  galvanometer 
  to 
  indicate 
  a 
  

   decrease 
  in 
  the 
  resistance 
  of 
  the 
  wire 
  under 
  examination. 
  

   The 
  nature 
  of 
  this 
  change 
  was 
  not, 
  however, 
  similar 
  to 
  that 
  

   observed 
  in 
  the 
  case 
  of 
  osmium 
  and 
  tungsten 
  wires, 
  but 
  

   rather 
  indicated 
  a 
  fairly 
  steady 
  rate 
  of 
  decrease 
  in 
  the 
  

   resistance 
  with 
  the 
  time. 
  In 
  the 
  case 
  of 
  some 
  wires 
  

   examined 
  this 
  rate 
  of 
  decrease 
  of 
  resistance 
  due 
  to 
  the 
  flow 
  

   of 
  the 
  thermionic 
  current 
  showed 
  no 
  clear 
  signs 
  of 
  abatement 
  

   at 
  the 
  end 
  of 
  an 
  interval 
  of 
  fifteen 
  minutes. 
  This 
  peculiarity 
  

   of 
  behaviour 
  could 
  not 
  be 
  ascribed 
  to 
  a 
  natural 
  sluggishness 
  

   of 
  the 
  wires 
  to 
  adjust 
  themselves 
  to 
  alternations 
  in 
  the 
  

   thermal 
  or 
  energy 
  conditions, 
  for 
  when 
  the 
  rate 
  of 
  supply 
  of 
  

   electrical 
  energy 
  was 
  suddenly 
  altered 
  by 
  increasing 
  the 
  

   heating 
  current 
  in 
  the 
  manner 
  described 
  in 
  our 
  previous 
  

   paper, 
  the 
  galvanometer 
  indicated 
  that 
  the 
  corresponding 
  

   resistance 
  changes 
  were 
  complete 
  in 
  considerably 
  less 
  than 
  

   thirty 
  seconds, 
  as 
  in 
  the 
  case 
  of 
  the 
  osmium 
  and 
  tungsten 
  

   wires. 
  With 
  regard 
  to 
  the 
  magnitude 
  of 
  the 
  steady 
  rate 
  of 
  

   decrease 
  of 
  resistance 
  observed 
  when 
  the 
  thermionic 
  current 
  

   was 
  flowing, 
  it 
  was 
  found 
  that 
  the 
  deflexion 
  of 
  the 
  galva- 
  

   nometer 
  produced 
  by 
  ten 
  minutes 
  flow 
  of 
  thermionic 
  current 
  

   yielded 
  a 
  value 
  of 
  the 
  apparent 
  loss 
  of 
  energy 
  per 
  unit 
  

   current 
  of 
  considerably 
  over 
  one 
  hundred 
  equivalent 
  volts, 
  

   the 
  calculations 
  being 
  made 
  on 
  the 
  assumption 
  that 
  the 
  

   galvanometer 
  deflexion 
  was 
  due 
  to 
  a 
  true 
  cooling 
  effect 
  

   brought 
  about 
  by 
  the 
  thermionic 
  emission, 
  as 
  demanded 
  by 
  

   the 
  theory 
  of 
  the 
  experiments. 
  It 
  seems 
  evident 
  that 
  this 
  

   change 
  in 
  resistance 
  is 
  not 
  to 
  be 
  ascribed 
  to 
  the 
  cooling- 
  

   effect 
  sought 
  for 
  in 
  this 
  investigation. 
  This 
  conclusion 
  is 
  

   supported 
  by 
  the 
  fact 
  that 
  upon 
  stopping 
  the 
  thermionic 
  

  

  