526  Dr.  F.  Horton  on  the  Electrical 
was  then  removed  and  the  strip  was  heated  to  a  lower  tem- 
perature by  means  of  the  direct  current  from  a  battery  of 
10  E.P.S.  motor  cells.  The  cathode  was  also  connected  to 
the  negative  pole  of  a  battery  of  small  accumulator-cells,  the 
positive  pole  of  which  was  connected  to  the  anode  through  a 
current-measuring  instrument — a  Thomson  galvanometer  for 
small  currents  and  a  milliammeter  for  larger  ones. 
In  making  the  experiments,  the  required  temperature  of 
the  cathode  (indicated  by  the  thermocouple)  was  first  obtained 
by  adjusting  the  resistance  in  the  heating  circuit.  The  pressure 
of  gas  in  the  apparatus  was  then  measured  by  the  McLeod 
gauge,  and  the  apparatus  left  for  15  minutes,  after  which  the 
pressure  was  again  read  and  the  discharge  started  through 
the  tube.  The  current  was  allowed  to  pass  for  15  minutes. 
It  was  measured  at  intervals  during  this  time  and  the  mean 
value  was  calculated.  The  pressure  was  then  read,  and  the 
apparatus  was  left  to  itself  for  another  1 5  minutes.  In  this 
way  any  liberation  of  gas  due  merely  to  the  heating  of  the 
cathode  could  be  allowed  for,  the  temperature  of  the  cathode 
being  maintained  the  same  throughout  the  ex]:>eriinents.  It 
was  found  that  the  passage  of  the  discharge  slightly  increased 
the  temperature  of  the  cathode,  and  at  very  high  temperatures 
(1 100°  C.)  the  arrangement  was  unstable,  for  at  the  higher 
temperature  the  lime  gives  a  larger  supply  of  corpuscles  for 
carrying  the  current,  which  is  increased,  and  thus  still  further 
increases  the  temperature  of  the  cathode.  In  order  that  the 
current  passing  through  the  tube  should  not  become  sufficient 
to  raise  the  temperature  of  the  cathode  above  the  melting- 
point  of  platinum,  a  small  fuse  cut  from  a  sheet  of  thin  tin- 
toil  was  placed  in  the  circuit. 
The  results  obtained  varied  with  the  length  of  time  during 
which  the  oxide  had  been  heated  before  the  observations  were 
made.  It  was  found  in  one  case,  when  there  was  still  a  fair 
amount  of  gas  being  given  off  on  merely  heating  the  cathode, 
that  the  extra  liberation  of  gas  when  a  current  of  2" 66  X  10-4 
ampere  was  sent  through  the  tube  was  within  8  per  cent,  of 
that  required  by  Faraday's  law  on  the  supposition  that  the 
conduction  through  the  lime  was  entirely  electrolytic*.  On 
heating  the  cathode  to  about  1400°  C.  for  several  hours  longer, 
there  was  much  more  gas  given  off  from  it,  and  on  pumping 
this  out  and  repeating  the  experiment  it  was  found  that  the 
evolution  of  gas  had  practically  ceased,,  and  that  on  now 
sending  a  discharge  through  the  tube  the  pressure  was  altered 
*  It  is  interesting  to  note  in  this  connexion  that  Skinner  (Phys.  Rev. 
July  1905)  has  found  that  the  evolution  of  hydrogen  from  a  fresh  cathode 
oheys  Faraday's  Law. 
