﻿oSS 
  Mr. 
  J. 
  Satterly 
  on 
  the 
  Amount 
  oj 
  

  

  experiments. 
  The 
  insulation 
  leak 
  was 
  often 
  tested. 
  When 
  

   it 
  became 
  serious 
  enough 
  to 
  gradually 
  slow 
  down 
  the 
  motion 
  

   of 
  the 
  spot 
  of 
  light 
  when 
  taking 
  the 
  normal 
  air 
  leak 
  the 
  

   insulation 
  was 
  attended 
  to 
  and 
  put 
  right. 
  In 
  taking 
  the 
  air 
  

   leak 
  and 
  other 
  leaks 
  of 
  about 
  the 
  same 
  order 
  I 
  observed 
  that 
  

   the 
  leaks 
  for 
  successive 
  minutes 
  often 
  differed 
  largely 
  from 
  

   each 
  other, 
  while 
  leaks 
  for 
  longer 
  periods, 
  say 
  five 
  minutes, 
  

   agreed 
  very 
  closely. 
  From 
  this 
  it 
  seems 
  that 
  small 
  ionizations 
  

   are 
  rather 
  irregular, 
  the 
  irregularities, 
  however, 
  smoothing 
  

   out 
  if 
  the 
  leaks 
  are 
  taken 
  over 
  sufficiently 
  long 
  periods. 
  

   When 
  the 
  leak 
  is 
  large 
  the 
  values 
  for 
  successive 
  minutes 
  

   agree 
  much 
  more 
  closely 
  than 
  when 
  the 
  leak 
  is 
  small. 
  

  

  Throughout 
  the 
  experiments 
  air 
  was 
  never 
  passed 
  through 
  

   the 
  testing 
  vessel. 
  The 
  method 
  of 
  renewing 
  the 
  air 
  was 
  

   always 
  that 
  of 
  exhaustion 
  and 
  refilling 
  through 
  the 
  lower 
  of 
  

   the 
  two 
  side 
  tubes. 
  The 
  exhaustions 
  were 
  performed 
  by 
  

   means 
  of 
  a 
  hand 
  Fleuss 
  pump. 
  This 
  pump 
  exhausted 
  down 
  to 
  

   within 
  about 
  9 
  mms. 
  of 
  vacuo. 
  Exhaustion 
  took 
  a 
  very 
  long- 
  

   time 
  if 
  a 
  calcium 
  chloride 
  tube 
  was 
  inserted 
  between 
  the 
  

   vessel 
  and 
  the 
  pump, 
  so 
  that 
  usually 
  the 
  vessel 
  was 
  exhausted 
  

   without 
  the 
  intermediary 
  of 
  a 
  drying-tube, 
  and 
  the 
  drying- 
  

   tube 
  exhausted 
  separately 
  and 
  then 
  connected 
  to 
  the 
  vessel. 
  

  

  (2) 
  The 
  Charcoal 
  Tubes. 
  — 
  Preliminary 
  experiments 
  were 
  

   made 
  with 
  charcoal 
  in 
  glass 
  combustion-tubes 
  and 
  iron 
  tubes; 
  

   but 
  the 
  glass 
  tubes 
  did 
  not 
  stand 
  heating 
  and 
  the 
  iron 
  tubes 
  

   took 
  too 
  long 
  to 
  get 
  hot. 
  Finally, 
  porcelain 
  tubes 
  were 
  

   tried. 
  Three 
  tubes 
  were 
  obtained 
  about 
  50 
  cms. 
  long 
  and 
  

   I'd 
  em. 
  in 
  internal 
  diameter. 
  The 
  coconut 
  charcoal 
  was 
  

   made 
  by 
  strongly 
  heating 
  fragments 
  of 
  coconut 
  shells 
  under 
  

   sand 
  in 
  a 
  fire-clay 
  crucible. 
  The 
  fragments 
  were 
  then 
  ground 
  

   fine 
  by 
  means 
  of 
  a 
  pestle 
  and 
  mortar. 
  The 
  central 
  12 
  inches 
  

   of 
  the 
  tubes 
  was 
  filled 
  with 
  the 
  charcoal, 
  the 
  column 
  of 
  

   charcoal 
  being 
  kept 
  in 
  position 
  by 
  wads 
  of 
  asbestos 
  fibre. 
  

   The 
  ends 
  of 
  the 
  tube 
  were 
  closed 
  by 
  indiarubber 
  stoppers 
  

   provided 
  with 
  glass 
  tubes 
  and 
  indiarubber 
  caps. 
  Thirty-nine 
  

   grams 
  of 
  charcoal 
  were 
  placed 
  in 
  each 
  tube. 
  The 
  three 
  

   tubes 
  made 
  were 
  denoted 
  by 
  A, 
  B, 
  and 
  C 
  for 
  purposes 
  of 
  

   identification. 
  The 
  furnace 
  used 
  was 
  made 
  of 
  two 
  6-inch 
  

   sections 
  of 
  a 
  Fletcher's 
  tube-furnace, 
  and 
  was 
  supplied 
  with 
  

   twelve 
  burners, 
  and 
  the 
  tubes 
  were 
  always 
  placed 
  on 
  the 
  

   furnace 
  so 
  that 
  the 
  12 
  inches 
  of 
  charcoal 
  got 
  to 
  a 
  bright 
  

   red 
  heat. 
  

  

  (3) 
  The 
  Asjnrators. 
  — 
  Two 
  aspirators 
  were 
  made 
  up 
  to 
  

   collect 
  the 
  gases 
  drawn 
  through 
  the 
  heated 
  charcoal. 
  Each 
  

   aspirator 
  consisted 
  of 
  two 
  large 
  glass 
  bottles 
  of 
  slightly 
  larger 
  

   volume 
  than 
  the 
  testing 
  vessel. 
  When 
  collecting 
  the 
  gas 
  

  

  