﻿1 
  

  

  rime. 
  

  

  

  1896, 
  

  

  Feb. 
  

  

  IT 
  

  

  

  Feb. 
  

  

  23 
  

  

  

  Mar. 
  

  

  15 
  

  

  

  Apri 
  

  

  119 
  

  

  

  Aug. 
  

  

  5 
  

  

  1897, 
  

  

  Feb. 
  

  

  6 
  

  

  1899, 
  

  

  Dec. 
  

  

  12 
  

  

  No. 
  III. 
  

  

  Temperature. 
  

  

  — 
  -05 
  cm 
  

  

  16° 
  C. 
  

  

  — 
  •05 
  

  

  15 
  

  

  — 
  •03 
  

  

  17-5 
  

  

  ±•00 
  

  

  19 
  

  

  + 
  -U 
  

  

  22-7 
  

  

  •31 
  

  

  15-8 
  

  

  '70 
  

  

  19-0 
  

  

  142 
  C. 
  Barus 
  — 
  Motion 
  of 
  a 
  Submerged 
  Index 
  Thread, 
  etc. 
  

  

  Observations 
  of 
  the 
  lower 
  and 
  upper 
  meniscus 
  of 
  each 
  tube 
  

   were 
  taken 
  from 
  time 
  to 
  time 
  in 
  the 
  course 
  of 
  about 
  4 
  years. 
  

   Though 
  placed 
  in 
  a 
  vault, 
  the 
  seasonable 
  variations 
  of 
  tem- 
  

   perature 
  could 
  not 
  be" 
  excluded. 
  Change 
  of 
  temperature, 
  how- 
  

   ever, 
  was 
  very 
  gradual, 
  and 
  in 
  view 
  of 
  the 
  small 
  bulbs 
  and 
  

   wide 
  stems, 
  the 
  mercury 
  thread 
  was 
  but 
  slightly 
  shifted. 
  The 
  

   data 
  are 
  recorded 
  in 
  the 
  following 
  table, 
  and 
  for 
  brevity 
  the 
  

   mean 
  positions 
  of 
  the 
  threads 
  (point 
  midway 
  between 
  the 
  

   upper 
  and 
  lower 
  meniscus) 
  need 
  only 
  be 
  given, 
  as 
  observed 
  

   before 
  and 
  after 
  April 
  19, 
  1896. 
  

  

  No. 
  II. 
  

  

  — 
  -13 
  cm 
  

  

  — 
  •10 
  

  

  — 
  •03 
  

   ±'00 
  

   + 
  •07 
  

  

  •24 
  

   •92 
  

  

  The 
  mean 
  positions 
  are 
  referred 
  to 
  an 
  arbitrary 
  cm. 
  scale 
  

   reading 
  downward. 
  Since 
  the 
  effect 
  of 
  temperature 
  is 
  slight, 
  

   both 
  series 
  may 
  be 
  plotted 
  to 
  show 
  their 
  variations 
  in 
  the 
  lapse 
  

   of 
  time. 
  The 
  data 
  are 
  found 
  to 
  be 
  regular 
  and 
  consistent 
  

   throughout 
  the 
  whole 
  time 
  of 
  nearly 
  4 
  years. 
  The 
  larger 
  

   thinner 
  thread 
  II 
  shows 
  more 
  rapid 
  motion 
  than 
  the 
  shorter 
  

   thicker 
  thread 
  III. 
  Both 
  move 
  at 
  a 
  regularly 
  retarded 
  rate 
  

   through 
  infinite 
  time, 
  and 
  there 
  can 
  be 
  no 
  further 
  doubt 
  that 
  

   the 
  cause 
  of 
  motion 
  is 
  the 
  volume 
  viscosity 
  of 
  the 
  bulb, 
  which 
  

   is 
  thinner 
  and 
  sustains 
  slightly 
  more 
  pressure 
  in 
  case 
  of 
  II 
  

   than 
  in 
  case 
  of 
  III. 
  Taken 
  together, 
  the 
  present 
  rates 
  under 
  

   like 
  conditions 
  are 
  strikingly 
  less 
  than 
  the 
  rate 
  of 
  motion 
  for 
  

   the 
  above 
  large 
  piezometer 
  bulb, 
  which 
  is 
  again 
  in 
  accord 
  with 
  

   the 
  more 
  effective 
  conditions 
  for 
  exhibiting 
  volume 
  viscosity 
  

   in 
  tbe 
  latter 
  case. 
  Between 
  April, 
  1896, 
  and 
  December, 
  1899, 
  

   at 
  19° 
  C, 
  the 
  mean 
  motions 
  of 
  the 
  threads 
  were 
  but 
  29x10"* 
  

   per 
  hour 
  for 
  II 
  and 
  22x10"" 
  per 
  hour 
  for 
  III, 
  as 
  compared 
  

   with 
  the 
  rates 
  17000 
  XlO" 
  6 
  per 
  hour 
  and 
  1400 
  XlO" 
  6 
  per 
  hour 
  

   (above, 
  § 
  2), 
  for 
  the 
  large 
  bulb. 
  

  

  5. 
  I 
  was 
  not 
  prepared 
  for 
  the 
  above 
  results, 
  not 
  supposing 
  

   that 
  the 
  viscous 
  volume 
  expansion 
  of 
  a 
  totally 
  submerged 
  cold 
  

   glass 
  bulb, 
  resulting 
  from 
  the 
  pressure 
  excess 
  of 
  a 
  thread 
  of 
  

   mercury 
  but 
  2 
  - 
  6 
  cm 
  long, 
  would 
  be 
  of 
  such 
  marked 
  value. 
  More- 
  

   over 
  this 
  thread 
  is 
  filled 
  in 
  under 
  water 
  and 
  in 
  all 
  cases 
  retains 
  

   a 
  moist 
  appearance, 
  which 
  may, 
  however, 
  only 
  mean 
  a 
  clean 
  

   surface 
  of 
  contact. 
  Obviously, 
  however, 
  the 
  surface 
  viscosity 
  

   of 
  water, 
  i. 
  e. 
  the 
  change 
  of 
  viscosity 
  with 
  diminishing 
  thick- 
  

   ness 
  of 
  film 
  can 
  not 
  be 
  observed 
  in 
  this 
  way. 
  For 
  the 
  case 
  of 
  

   thin 
  oils 
  with 
  small 
  surface 
  tension 
  I 
  am 
  by 
  no 
  means 
  sure 
  that 
  

   the 
  case 
  is 
  hopeless 
  and 
  I 
  have 
  accordingly 
  arranged 
  a 
  series 
  of 
  

   tubes 
  with 
  this 
  end 
  in 
  view. 
  

  

  Brown 
  University, 
  Providence, 
  R. 
  I. 
  

  

  