Force  in  the  Contact  of  Metals.  87 
quantities  of  heat  produced  or  lost  at  the  surfaces  of  contact, 
and  not  merely  the  variations  of  temperature  which  have  been 
produced  there.  In  order  to  find  the  former  by  the  help  of  the 
latter,  it  is  necessary  to  know,  under  the  control  of  ordinary 
circumstances,  the  calorific  capacity  of  the  body  in  which  the 
change  of  temperature  is  effected.  But  always  to  endeavour  to 
measure  the  capacity  for  heat  of  the  metals  submitted  to  experi- 
ment would  take  too  long,  and  would  with  difficulty  lead  to  per- 
fectly sure  results.  The  experiments  will  therefore  be  regulated 
so  that  the  calorific  capacity  shall  not  exercise  any  sensible  in- 
fluence on  the  exactness  of  the  results.  Iu  order  to  fulfil  these 
conditions  as  far  as  possible,  I  contrived  a  kind  of  air-thermo- 
meter, of  the  following  peculiar  construction  : — 
■KUaA 
Fig.  1  represents  the  apparatus  viewed  from  above ;  and 
iig.  2  is  a  side  view,  a  and  b  are  two  perfectly  equal  cylinders 
of  thin  sheet  copper,  125  millions,  long,  and  80  in  diameter, 
with  their  outer  surface  silvered.  To  the  centre  of  their  circular 
ends  the  brass  tubes  c,  d  are  soldered.  Through  these  tubes, 
which  are  placed  opposite  to  one  another  and  make  a  right  angle 
with  the  ends  of  the  cylinder,  are  introduced  the  wires  intended 
for  the  experiments,  in  such  a  manner  that  the  surface  of  soldering 
between  the  two  wTires  is  nearly  in  the  middle  of  the  cylinder. 
The  cylinders  are  supported  by  a  mahogany  board  dd,  which, 
