﻿390 
  OF 
  AN 
  IMPROVED 
  

  

  being 
  closed. 
  The 
  sources 
  of 
  inaccuracy, 
  if 
  any 
  exist, 
  must 
  lessen 
  in 
  

   proportion 
  to 
  the 
  result, 
  as 
  the 
  quantity 
  acted 
  upon 
  is 
  augmented. 
  It 
  

   would, 
  of 
  course, 
  be 
  safer 
  to 
  extend 
  the 
  cylinder 
  in 
  length 
  than 
  in 
  dia- 
  

   meter. 
  

  

  Description 
  of 
  the 
  Gage. 
  

  

  It 
  is 
  well 
  known, 
  that 
  if 
  a 
  vertical 
  glass 
  tube 
  communicate, 
  through 
  

   its 
  upper 
  orifice, 
  with 
  a 
  receiver, 
  while 
  its 
  lower 
  orifice 
  is 
  situated 
  be- 
  

   neath 
  the 
  surface 
  of 
  an 
  adequate 
  quantity 
  of 
  mercury, 
  in 
  any 
  convenient 
  

   receptacle 
  ; 
  on 
  exhausting 
  the 
  receiver, 
  the 
  metal 
  will 
  rise 
  in 
  the 
  bore 
  

   of 
  the 
  tube 
  in 
  proportion 
  to 
  the 
  quantity 
  of 
  air 
  removed. 
  Hence, 
  if 
  

   zero 
  of 
  the 
  ascending 
  column 
  of 
  degrees, 
  counting 
  upwards 
  from 
  one 
  

   to 
  ten, 
  be 
  placed 
  on 
  a 
  level 
  with 
  the 
  surface 
  of 
  the 
  mercury 
  in 
  the 
  re- 
  

   ceptacle 
  at 
  the 
  foot 
  of 
  the 
  gage 
  tube 
  G, 
  the 
  quantity 
  of 
  gas 
  condensed 
  

   or 
  withdrawn 
  will 
  be 
  as 
  the 
  number 
  of 
  degrees 
  opposite 
  the 
  surface 
  of 
  

   the 
  column 
  of 
  the 
  mercury 
  in 
  the 
  gage 
  tube. 
  

  

  Again, 
  supposing 
  it 
  were 
  possible 
  to 
  exhaust 
  the 
  vessel 
  perfectly, 
  the 
  

   column 
  of 
  mercury 
  in 
  the 
  gage, 
  would 
  attain 
  the 
  height 
  of 
  a 
  well 
  

   filled 
  Torricellian 
  tube. 
  By 
  having 
  such 
  a 
  tube 
  by 
  the 
  side 
  of 
  the 
  gage 
  

   tube, 
  as 
  represented 
  at 
  B 
  in 
  the 
  figure, 
  its 
  orifice 
  communicating 
  with 
  

   the 
  mercury 
  of 
  the 
  same 
  receptacle, 
  and 
  placing 
  zero 
  of 
  the 
  descending 
  

   column 
  of 
  graduations 
  on 
  a 
  level 
  with 
  the 
  surface 
  of 
  the 
  mercury 
  in 
  the 
  

   Torricellian 
  tube, 
  the 
  quantity 
  of 
  air 
  in 
  the 
  receiver 
  will 
  always 
  be 
  as 
  

   the 
  number 
  of 
  degrees, 
  between 
  the 
  surface 
  of 
  the 
  mercury 
  in 
  the 
  gage 
  

   and 
  the 
  surface 
  of 
  the 
  same 
  metal 
  in 
  the 
  Torricellian 
  tube. 
  

  

  The 
  scale 
  comprises 
  ten 
  divisions, 
  each 
  containing 
  ten 
  subdivisions. 
  

   The 
  whole 
  scale 
  may 
  therefore 
  be 
  estimated 
  to 
  divide 
  the 
  capacity 
  of 
  

   a 
  receiver 
  into 
  ten 
  volumes, 
  or 
  into 
  one 
  hundred, 
  whenever 
  the 
  zeros 
  of 
  

   the 
  right 
  and 
  left 
  hand 
  columns 
  of 
  degrees 
  coincide 
  simultaneously, 
  

   the 
  one 
  with 
  the 
  surface 
  of 
  the 
  mercury 
  in 
  the 
  receptacle, 
  and 
  the 
  

   other 
  with 
  that 
  of 
  the 
  Torricellian 
  column. 
  But 
  on 
  this 
  it 
  were 
  vain 
  

   to 
  rely, 
  since 
  the 
  altitude 
  of 
  the 
  Torricellian 
  column 
  is 
  liable 
  to 
  vary 
  

   while 
  the 
  scale 
  remains 
  unchanged. 
  This 
  difficulty 
  is, 
  however, 
  easily 
  

   surmounted 
  by 
  restricting 
  the 
  length 
  of 
  the 
  graduated 
  part 
  of 
  the 
  scale 
  

   to 
  the 
  minimum 
  height 
  of 
  the 
  mercurial 
  column, 
  or 
  twenty-seven 
  

  

  