506  Dr.  F.  Horton  on  the  Electrical 
metals  is  not  due  to  this  cause,  for  we  cannot  conceive  that 
a  molecule  of  an  elementary  substance  could  be  split  up  into 
two  dissimilar  atoms,  or  groups  of  atoms,  having  opposite 
charges  ;  and  experiments  have  shown  that  the  conduction  of 
electricity  through  metals  is  not  accompanied  by  any  detectable 
transportation  of  matter*. 
A  theory  of  metallic  conduction  not  involving  the  trans- 
portation of  matter  in  the  atomic  state  has  been  put  forward 
by  Professor  Thomson  f*  On  this  theory  metallic  conductors 
owe  their  conductivity  to  the  dissociation  of  the  atoms  of  the 
metal,  the  products  of  dissociation  arising  from  a  single 
atom  being  an  extremely  small  negatively  charged  corpuscle 
and  the  remaining  larger  portion  of  the  atom  which  bears 
an  equal  positive  charge.  These  negative  corpuscles  are 
identically  the  same  in  mass  and  charge  from  whatever  atom 
they  are  produced,  and  they  can  move  about  freely  in  the 
mass  of  the  metal.  When  an  electric  force  acts  upon  them, 
they  travel  in  a  direction  opposite  to  that  of  the  force,  and 
this  convection  of  charges  constitutes  the  electric  current. 
The  conductivity  of  compound  bodies  may  belong  to  either 
the  metallic  or  the  electrolytic  class.  The  distinction  between 
these  two  classes  is  generally  based  on  the  sign  of  the  tem- 
perature coefficient  of  the  conductivity  of  the  conductor  in 
question.  If  the  conductivity  decreases  with  rise  of  tempera- 
ture, the  conduction  is  taken  to  be  metallic  :  if  it  increases, 
the  conduction  is  considered  to  be  electrolytic.  This,  how- 
ever, cannot  be  taken  as  an  infallible  rule,  for,  although  the 
conductivity  of  liquid  electrolytes  generally  increases  with 
the  temperature,  a  series  of  electrolytic  liquids  is  known 
which  behave  in  the  opposite  manner.  Such  acids  as 
sulphuric  and  phosphoric,  the  hydrolytic  dissociation  of 
which  takes  place  exothermically,  belong  to  this  series.  In 
these  cases  an  increase  of  temperature  decreases  the  electro- 
lytic dissociation,  causing  a  diminution  in  the  conductivity 
which  more  than  compensates  for  the  increased  mobility  of 
the  ions.  On  the  other  hand,  the  conductivity  of  all  the 
metalloids,  which  cannot  be  electrolytic  on  account  of  the 
elementary  nature  of  the  conductors,  increases  with  rise  of 
temperature,  the  increase  being  probably  due  to  changes 
in  the  crystalline  structure  when  heated  %. 
It  has  long  been  known  that  many  substances  which,  at 
the  ordinary  temperature  of  the  laboratory,  may  be  regarded 
*  E.  Riecke,  P/n/s.  Zeitschr.  ii.  p.  G39  (1901). 
t  J.  X  Thomson,  Rapports  prcsentes  au  Conr/res  International  de 
Physique,  vol.  iii.  p.  138  (Paris,.  1900). 
\  Siemens.  Pogg.  Ann.  clix,  p.  11 7  (1876). 
