RELATION  OF  HYDROGEN  TO  PALLADIUM. 
without  any  change  of  volume,  the  alloy  may  therefore  be  said  to 
be  composed  of — 
By  volume. 
Palladium,  100         or  95-32 
Hydrogenium,   4-908    or  4-68 
104-908  100 
The  expansion  which  the  palladium  undergoes  appears  enormous 
if  viewed  as  a  change  of  bulk  in  the  metal  only,  due  to  any  con- 
ceivable physical  force,  amounting  as  it  does  to  sixteen  times  the 
dilatation  of  palladium  when  heated  from  0°  to  100°  C.  The  den- 
sity of  the  charged  wire  is  reduced  by  calculation  from  12-3  to 
11-79.  Again,  as  100  is  to  4-91,  so  the  volume  of  the  palladium, 
0-1358  c.c.  is  to  the  volume  of  the  hydrogenium  0*006714  c.G. 
Finally  dividing  the  weight  of  the  hydrogenium,  0*01147  grm  by 
its  volume  in  the  alloy,  0*006714  c.c.  we  find 
Density  of  hydrogenium  1*708 
The  density  of  hydrogenium,  then,  appears  to  approach  that  of 
magnesium,  1*743,  by  this  first  experiment. 
Further,  the  expulsion  of  hydrogen  from  the  wire,  however 
caused,  is  attended  with  an  extraordinary  contraction  of  the  lat- 
ter. On  expelling  the  hydrogen  by  a  moderate  heat,  the  wire 
not  only  receded  to  its  original  length,  but  fell  as  much  below  that 
zero  as  it  had  previously  risen  above  it.  The  palladium  wire 
first  measuring  609-144  m.m.,  and  which  increased  9*77  m.m., 
was  ultimately  reduced  to  599-444  m.m.,  and  contracted  9*7  m.m. 
The  wire  is  permanently  shortened.  The  density  of  the  pal- 
ladium did  not  increase,  but  fell  slightly  at  the  same  time,  namely, 
from  12-38  to  12-12  ;  proving  that  this  contraction  of  the  wire  is 
in  length  only.  The  result  is  the  converse  of  extension  by  wire- 
drawing. The  retraction  of  the  wire  is  possibly  due  to  an  effect 
of  wire-drawing,  in  leaving  the  particles  of  metal  in  a  state  of 
unequal  tension,  a  tension  which  is  excessive  in  the  direction  of 
the  length  of  the  wire.  The  metallic  particles  would  seem  to  be- 
come mobile,  and  to  right  themselves  in  proportion  as  the  hydro- 
gen escapes  ;  and  the  wire  contracts  in  length,  expanding,  as 
appears  by  its  final  density,  in  other  directions  at  the  same  time. 
A  wire  so  charged  with  hydrogen,  if  rubbed  with  the  powder 
