508  Mr.  t>E  lug  on 
It  is  evident  then,  that  there  is  no  longer  any  fenfible 
error  arifing  from  the  differences  of  fpirituofity,  which 
is  already  a capital  advantage  in  the  cafe  propofed  as'  an 
example;  in  which,  fince  what  one  wants  to  know  is  the 
degree  of  fpirituofity  of  a liquor,  one  cannot  fuppofe  it 
known  before  hand,  in  order  to  have  regard  to  it  in 
feekihg  for  it.  One  likewife  fees  in  general,  that  there  is 
fcarce  any  error  to  apprehend;  even  in  confidering  the 
augmentation  of  the  bulk  of  thefe  liquors,  or  the  dimi- 
nution of  their  fpecific  gravities,  as  being  proportional  to 
the  indication  of  the  quickfilver  Thermometer. 
Here  is  then  tne  method,  in  which,  according  to  this 
principle,  I would  conftrudt  the  comparative  Areometer , 
that  is  fuch  a one  as  might  be  made  the  fame  every 
where.  I chufe  this  example  becaufe  it  will  afford  me 
other  applications  of.  the  general  rule. 
Project 
On  volt  done  qu’il  n’y  a plus  d’erreur  fenfible  refultante  des  differences  de 
fpirituofite;  et  e’eft  d’abord  un  avantage  capital  dans  le  cas  propofe'  pour  exera- 
ple,  ou,  cherchant  a connoitre  le  degre  de  fpirituofite  d’une  liqueur,  on  ne  peut 
pas  la  fuppofer  d’avance  pour  y avoir  egard  en  la  mefurant.  On  voit  auffi  en 
general,  qu’il  n’y  a prefque  plus  d’erreur  a craindre,  meme  en  regardant  l’aug- 
mentation  de  volume  de  ces  liqueuns,  ouleur  diminution  depefanteur  ipecifique, 
comme  proportionnelles  a l’indication  du  TLermometre  de  mercure. 
Yoici  d’apres  ce  principe,  comment  je  conftruirois  l’Areometre  coiftparable; 
e’eft-a-dire  celui  qu’on  pourroit  faire  de  memepartout.  Je  choifis  cct  exemple, 
parce  qu’il  me  fournira  encore  d’autres  applications  de  la  regie. 
Ji 
