Force  in  the  Contact  of  Metals.  89 
vent  them  from  oxidizing  so  readily.  They  had  also  apertures 
for  the  passage  of  the  before-mentioned  brass  tubes,  of  the  brass 
collars  which  served  for  turning  the  cocks,  and  of  the  wires  to  be 
experimented  on  and  for  that  purpose  introduced  into  the  copper 
cylinders. 
In  the  brass  tubes,  c  c  and  c'  tJ,  through  which  the  wires  passed 
intended  for  the  experiments,  small  disks  of  wood  pierced  with 
one  hole  were  placed,  and  introduced  far  enough  to  be  near  the 
ends  of  the  copper  cylinders.  The  aperture  in  these  disks  was 
quite  large  enough  for  the  passage  of  the  thickest  wires.  They 
were  for  the  purpose  of  cutting  off  the  wires  from  conductive 
contact  with  the  copper  cylinders,  and  also  to  form  a  sort  of  bot- 
tom in  the  tubes.  To  make  an  hermetical  closure  around  the 
wires  the  following  process  was  adopted : — I  first  introduced, 
round  the  wire,  a  small  flock  of  cotton,  which  was  strongly 
pressed  against  the  wooden  disk ;  then  I  filled  the  tube  with  a 
melted  mixture  of  wax  and  rosin.  The  cotton  prevented  the 
fused  mixture  from  running  into  the  cylinder.  Of  all  the  means 
I  tried  for  hermetically  sealing  these  tubes,  this  proved  to  be 
the  best;  in  no  one  instance  has  the  purpose  not  been  attained. 
Before  the  union  of  the  glass  tube  with  the  two  brass  tubes,  I 
introduced  therein,  as  index,  an  alcoholic  mixture  forming  a 
column  of  a  few  centims.  length.  This  index  could  be  placed 
in  a  suitable  position  in  the  glass  tube  by  turning  the  cocks  so 
as  to  put  the  two  extremities  of  the  tube  in  communication  with 
the  external  air  and  properly  turning  the  apparatus  about  the 
pivots  e.  Before  the  commencement  of  each  series  of  observa- 
tions, I  always  ascertained  that  the  tubes  c  c'  and  the  caoutchouc 
tubes  were  hermetically  closed  by  the  mixture  of  wax  and  rosin  : 
the  examination  took  place  in  the  following  manner : — Before 
putting  the  glass  tube  in  the  place  it  was  to  occupy,  one  of  the 
brass  tubes  was,  by  means  of  a  caoutchouc  tube,  connected  with 
a  manometer  consisting  of  a  bent  glass  tube  placed  vertically 
and  partly  filled  with  water.  I  increased  or  diminished  by  some 
centimetres  the  height  of  the  column  of  water  in  the  open  leg  of 
the  tube,  so  that  the  pressure  of  the  air  in  the  copper  cylinder 
was  slightly  greater  or  less  than  that  of  the  outer  air.  If  the 
column  of  water  remained  the  same  during  a  sufficiently  long 
time,  it  wa3  a  proof  that  the  cylinder  was  hermetically  closed. 
To  ascertain  the  hermeticity  of  the  caoutchouc  tubes  between 
the  glass  tubes  and  the  two  brass  ones,  I  proceeded  as  follows  : — 
One  of  the  extremities  of  the  glass  tube  was  closed  by  means  of 
one  of  the  cocks,  while  the  other  extremity  was  put  in  commu- 
nication with  the  external  air  by  means  of  the  other  cock.  If, 
on  inclining  the  glass  tube,  the  index  remained  motionless  for  a 
certain  time,  it  wras  a  proof  of  the  hermeticity  of  the  binding 
