Force  in  the  Contact  of  Met  ah.  85 
such  a  nature  as  to  produce,  under  certain  circumstances,  a  gal- 
vanic current.  But  it  is  clear  that  such  a  current  could  only- 
have  a  momentary  duration.  Immediately  after  contact  the 
particles  would  be  at  rest;  the  rise  of  a  galvanic  current  would 
be  impossible,  seeing  that  it  cannot  be  created  out  of  nothing. 
If,  on  the  contrary,  the  bodies  act  chemically  on  each  other,  the 
particles  enter,  after  the  contact,  into  chemical  combination,  new 
particles  are  attracted  and  lose  their  acquired  velocity,  and  thus 
there  exist  materials  for  the  formation  of  a  galvanic  current  as 
long  as  the  chemical  activity  continues."  If  the  vis  viva  lost  by 
the  contact  of  the  particles  after  their  approach  were  the  only 
element  suited  to  produce  a  galvanic  current,  the  proof  above  ex- 
hibited would  be  perfectly  valid,  and  it  would  be  absolutely  im- 
possible that  contact  between  bodies  chemically  indifferent  should 
bring  about  a  galvanic  current  of  a  certain  duration.  But  the 
elements  necessary  for  the  formation  of  the  galvanic  current  are 
not  composed  of  the  vis  viva  lost  in  the  collision  of  the  particles. 
The  explanation  of  Peltier's  experiments  demonstrates  that  it  is 
the  heat  which  is  transformed  into  electricity.  The  opinion 
above  quoted,  therefore,  proves  nothing  with  respect  to  the  pre- 
sence of  an  electromotive  force  in  the  surface  of  contact  between 
chemically  indifferent  bodies. 
It  is  proved  that,  for  one  and  the  same  force  of  the  current, 
the  quantities  of  heat  lost,  as  well  as  those  produced,  in  Peltier's 
experiments,  are  proportional  to  the  electromotive  forces  at  the 
surfaces  of  contact.  By  the  appreciation  of  these  quantities, 
then,  we  gain  a  relative  measure  of  the  electromotive  forces  in 
question.  I  give  below  the  account  of  the  experiments  under- 
taken by  me  for  the  purpose  of  measuring  the  electromotive 
forces  springing  from  the  contact  of  metals.  The  researches  I 
have  already  published  on  this  subject*  must  only  be  regarded 
as  preliminary.  Not  only  was  the  method  employed  not  suffi- 
ciently delicate  for  the  measurement  of  the  smallest  of  the  forces, 
but  certain  arrangements  did  not  permit  the  obtaining  of  results 
with  perfect  certainty  and  rigour.  I  have  done  my  best  to  get 
rid  of  these  imperfections  in  the  researches  which  are  described 
below  f. 
*  Ofcersigt  af  Vet.  Akademiens  Vbrhandl.  for  18/0,  p.  3.  Pogg.  Ann. 
vol.  cxl.  p.  435.     Phil.  Mag.  S.  4.  vol.  xli.  p.  18. 
t  On  this  subject  we  are  indebted  to  M.  le  Roux  for  some  very  merito- 
rious researches,  published  in  Annates  de  Chimie  et  de  Physique,  vol.  x. ; 
but  his  method  is,  from  beginning  to  end,  totally  different  from  that  em- 
ployed by  me.  To  measure  Peltier's  quantities  of  heat,  M.  le  Roux  used 
two  calorimeters,  each  containing  an  equal  volume  of  water,  and  placed 
side  by  side.  To  ascertain  the  quantities  of  heat  lost  or  gained  in  the  con- 
tact of  two  metals — for  example,  copper  and  bismuth — one  surface  of 
contact  was  placed  in  each  calorimeter,  and  the  current  was  conducted  so 
