Pyrornetry  and  Afeometry , See.  439 
fcales  of  the  two  inftruments,  which,  going  over  im- 
moveable fcales  of  the  fame  fort,  fhews  their  dif- 
ference of  height.  When  this  difference  is  no  longer 
conformable  to  the  indication  of  the  Thermometer,  it  is 
ealily  rectified  by  turning  fmall  pegs,  on  which  are 
twilled  the  thread  of  lilk-grafs  which  ferves  for  the  fuf- 
penfion  of  the  fcales. 
The  irregularity  of  which  I have  been  fpeaking  con- 
Jfifts  in  this,  that  when  the  heat,  after  having  varied, 
returns  to  the  fame  point  of  the  quickfilver  Thermome- 
ter, the  metallic  Thermometer  does  not  return  to  it 
exactly,  but  varies  nearly  in  the  following  manner. 
During  the  fummer  the  metallic  Thermometer  gains  con- 
ftantly  on  the  other  ; I mean,  that  amidft  its  variations, 
it  always  preferves  a fmall  part  of  the  lengthening, 
which 
echelles  mobiles  des  deux  inftrumens,  et  qui,  pafFant  fur  des  echelles  immobiles 
femblables,  fait  voir  ainfi  leur  difference  de  hauteur.  Si  cettfe  difference  ceffe 
d’etre  conforme  a l’indicati-on  du  Thermometre,  on  la  redlifie  aifement,  en  tour- 
nant  de  petites  chevilles  fur  lefquelles  s’enveloppent  les  fils  de  pite  qui  fervent  a 
la  fufpenfion  des  echelles. 
L’irregularite  dont  il  s’agit  confifte  en  ce  que,  Iorfque  la  chaleur,  apres  avoir 
varie,  revient  au  meme  point  fur  le  Thermometre  de  mercure,  le  Thermometre 
metallique  n’y  revient  pas  toujours  exacfement;  et  la  marehe  des  irregulari:e$ 
eft  celle-ci:  en  Ete,  le  thermometre  metallique  gagne  de  plus  en  plus  fur  f autre  ^ 
e’efl  a dire  que,.  dans  fes  variations,,  il  conferve  toujours  quelque  petite  partie  de 
Tallongs- 
