[     535     ] 
XLV.  Genesis  of  Ions  by  Collision  and  Sparking -Potentials 
in  Carbon  dioxide  and  Nitrogen.  By  H.  E.  Hurst,  B.A., 
B.Sc,  Hertford  College,  Oxford  *. 
AN  explanation  of  the  theory  of  ionization  by  collision.^ 
of  positive  and  negative  ions  with  molecules  of  a  gas 
in  a  uniform  field  of  force  was  given  in  a  paper  on  the 
"  Genesis  of  Ions  in  a  Gas "  (J.  S.  Townsend,  Phil.  Mag. 
Nov.  1903)  and  the  experiments  were  continued  in  a  further 
paper  (Dec.  190-1).  The  apparatus  previously  used  for  these 
experiments  with  air  and  hydrogen  has  recently  been  used 
to  verify  the  theory  in  the  case  of  nitrogen  and  carbon 
dioxide,  and  the  results  of  these  experiments  are  recorded  in 
the  present  paper.  A  detailed  description  of  the  apparatus 
and  of  the  method  of  conducting  experiments  will  be  found 
in  the  first-mentioned  paper.  Briefly,  the  apparatus  con- 
sisted of  two  parallel  plate  electrodes,  one  of  which  was  zinc. 
In  the  first  experiments  on  nitrogen  and  those  on  carbon 
dioxide,  the  other  electrode  was  a  quartz  plate  coated  on  one 
side  with  silver.  Fine  lines  were  ruled  on  this  to  allow 
ultra-violet  light,  produced  by  a  spark  between  aluminium 
terminals,  to  fall  on  the  zinc  electrode.  In  the  second  set  of 
experiments  on  nitrogen,  a  zinc  plate  with  fine  slits  was 
substituted  for  the  silvered  quartz.  These  plates  can  be 
arranged  at  distances  from  '1  to  1*2  cm.  apart.  By  means 
of  an  electrometer  the  currents  between  these  plates  were 
measured  for  different  values  of  the  electric  intensity  between 
them,  their  distance  apart,  and  the  pressure  of  the  gas  in 
which  they  wTere  immersed.  In  the  previous  experiments 
the  electrometer  was  kept  at  zero  potential  while  the  current 
was  passing,  by  means  of  a  condenser  and  potentiometer 
arranged  as  an  induction  balance.  In  the  experiments  here 
recorded  the  same  effect  was  obtained  by  finding  the  rise  of 
potential  during  half  the  time  of  an  experiment,  and  then 
starting  with  the  potential  of  the  electrometer  and  electrode 
connected  to  it  below  zero  by  this  amount.  Hence  during 
half  the  time  of  an  experiment  the  potential-difference 
between  the  electrodes  is  slightly  too  large,  and  during  the 
other  half  too  small  by  nearly  equal  amounts.  The  same 
precautions  with  regard  to  insulation  were  taken  as  were 
described  in  the  previous  paper. 
It  has  been   shown  that  if  ??0  negative  ions  are  liberated 
from    the    zinc    plate  by  the   ultra-violet  light,  and  travel 
through  the  gas  under  the  action  of  the  electric  force  Ik 
the  plates,  the  number  n  which  arrive  at  the  positive  electrode 
*  Communicated  by  Prof.  J.  S.  Townsend,  F.E.S. 
