Force  in  the  Contact  of  Metals.  83 
if  we  admit  the  presence  of  an  electromotive  force  at  the  point 
of  contact  between  the  two  metals.  If  the  galvanic  current 
passes  in  the  same  direction  as  the  current  produced  by  the 
electromotive  force,  a  cooling  results  at  the  point  of  contact, 
which  is  changed  into  a  rise  of  temperature  in  the  opposite  case. 
The  theory  further  demonstrates  that  the  quantities  of  heat  pro- 
duced or  absorbed  are  proportional  to  the  electromotive  forces, 
and  consequently  are  found  to  be  in  full  conformity  to  that 
which  experiment  requires*.  In  the  present  state  of  science  it 
is  impossible  to  account  for  the  modifications  of  temperature 
noted  at  the  surface  of  contact  without  admitting  the  presence 
of  an  electromotive  force  at  that  surface.  Experiments  have  de- 
cisively proved  that  a  quantity  of  heat  actually  disappears  at  the 
surface  of  contact,  and  not  merely  that  a  less  production  of  heat 
takes  place  there  than  in  other  parts  of  the  circuit ;  Lenz's  ex- 
periments have  given  evident  proof  of  this  fact.  Now  it  is  im- 
possible that  heat  should  disappear  without  producing  mecha- 
nical work,  external  or  internal,  which  is  then  the  equivalent  of 
the  vanished  heat,  or  without  its  changing  into  another  form  of 
motion.  The  mechanical  work  which  ought  to  arise  as  the  equi- 
valent of  the  heat  which  has  disappeared  could  only  consist  of 
an  increase  of  disgregation ;  but  there  is  no  sufficient  reason  for 
the  existence  of  such  a  modification ;  far  from  that,  we  possess 
valid  proofs  of  the  contrary,  namely  that  there  exists  no  disgre- 
*  M.  Clausius,  to  whom  science  is  much  indebted  for  the  most  impor- 
tant discoveries  in  the  mechanical  theory  of  heat,  has  already,  before  me, 
treated  Peltier's  phenomena  after  the  principles  of  the  above-mentioned 
theory  (Pogg.  Ann.  vol.  xc.  p.  513),  and  called  attention  to  this  circum- 
stance in  consequence  of  my  work  above  indicated  (Pogg.  Ann.  vol.  cxxxix. 
p.  280).  After  having  noted  that  M.  Helmholtz's  supposition,  that  all  the 
electrical  phenomena  which  take  place  in  metallic  conductors  are  easily 
explained  if  we  attribute  to  the  different  chemical  substances  a  power  of 
attraction  different  for  the  two  electricities,  is  not  correct,  for  this  is  not 
sufficient  to  explain  the  whole  of  those  phenomena,  M.  Clausius  continues 
as  follows  : — "For  the  explanation  of  thermoelectric  currents,  and  of  the 
phenomena  discovered  by  Peltier,  viz.  the  heating  and  cooling  occasioned 
at  the  point  of  contact  of  two  substances  by  an  electric  current,  this  assump- 
tion does  not  suffice ;  for  that  purpose  another  assumption  is  necessary, 
namely  that  the  heat  itself  is  operative  in  the  production  and  preservation 
of  the  electrical  difference  at  the  point  of  contact,  in  that  the  molecular 
motion  which  we  call  heat  tends  to  expel  the  electricity  from  one  substance 
to  the  other,  and  can  only  be  prevented  from  doing  so  by  the  counteracting 
force  of  the  two  layers  of  electricity  thereby  formed,  when  these  have 
attained  a  certain  density."  Now,  if  we  admit  the  correctness  of  this  hy- 
pothesis, Peltier's  phenomena  can,  as  M.  Clausius  has  shown,  at  once  be 
explained  by  it ;  if,  on  the  contrary,  the  validity  of  it  can  be  called  in  ques- 
tion, those  phenomena  remain  still  unexplained.  In  order  to  show  the 
connexion  of  the  phenomena  with  the  electromotive  forces  at  the  surface 
of  contact,  I  have  thought  it  necessary  to  start  from  no  sort  of  hypothesis, 
but  solelv  and  exclusively  from  the  known  facts  and  circumstances. 
G2 
