Conductivity  of  Metallic  Oxides.  527 
only  very  slightly,  the  alteration  being  in  most  cases  an 
increase,  but  in  a  few  cases  a  decrease.  The  following 
numbers  serve  to  illustrate  the  results  obtained  : — 
Temperature  of  cathode  =1100°  C. 
Potential-difference  between  the  electrodes  =40  volts. 
Time.  Pressure  in  mm.  of  Kg.     Current  in  amperes. 
3-54  p.m.  -0018 
4-3       ,,  -00185  n°»e 
4'21     »  '0021  none 
4-31     „  -00224  n0ne 
mm. 
Increase  of  pressure  per  minute  while  discharge  was  passing  =-000014 
Mean  increase  per  minute  when  no  discharge  was  passing  .  .    =  -000010 
If  the  conductivity  of  the  lime  was  entirely  electrolytic, 
the  passage  of  *03  ampere  would  have  liberated  oxygen  at 
a  rate  sufficient  to  increase  the  pressure  of  gas  in  the  apparatus 
by  '0803  mm.  per  minute.  So  the  actual  increase  of  pressure 
while  the  discharge  is  passing  is  only  .f  *  of  what  would 
be  expected  if  the  conduction  were  entirely  electrolytic.  The 
experiment  may  therefore  be  taken  as  showing  that  in  the 
conduction  of  electricity  through  heated  lime  only  a  very 
small  fraction  of  the  total  current  is  carried  by  means  of 
electrolytic  ions. 
A  result  similar  to  this  was  obtained  by  using  barium  oxide 
on  the  cathode  in  place  of  lime.  Some  experiments  were  also 
made  with  a  cupric-oxide  cathode,  and  no  increase  of  gas- 
pressure  in  the  apparatus  could  be  observed  when  a  discharge 
was  allowed  to  pass. 
Conclusion. 
The  point  of  most  interest  in  connexion  with  the  con- 
duction of  electricity  through  the  metallic  oxides  is  the 
method  by  which  the  current  is  carried.  Nernst  *  has  given 
reasons  for  supposing  the  conductivity  to  be  electrolytic,  the 
chief  of  these  being  (a)  that  definite  traces  of  electrolysis 
were  found  in  some  cases  ;  the  chemical  composition  at  the 
cathode  after  the  current  had  been  allowed  to.  flow  through 
the  oxide  for  a  long  time  being  found  to  be  different  from 
that  at  the  anode ;  (U)  the  conductivity  of  a  mixture  of 
oxides  was  found  to  be  much  greater  than  that  of  each  oxide 
separately.  Nernst  states  that  the  products  of  electrolysis 
are  continually  recombining,  and  so  the  supply  of  electrolytic 
ions  is  kept  up. 
*  Nernst,  Zeitschr.f.  Elcktroch.  vi.  pp.  41-43  (1899). 
