ELECTEICAL  DISCHAEGES  AS  OBSEEVED  IN  TOEEICELLIAN  VACUA.  7 
in  a film  on  some  portions  of  the  vessel  when  it  was  exhausted.  In  this  condition  of 
the  apparatus  the  discharge  was  similar  to  what  I had  often  obtained  (with  Mr.  Geove’s 
experiment)  in  the  vapour  of  phosphorus,  the  stratifications  being  very  distinct  near  the 
upper  or  positive  wire,  and  spreading  towards  the  negative  in  a confused  manner.  If 
the  mercury  was  permitted  to  rise  above  the  negative  wire,  as  in  the  former  experiment 
(16.),  the  surface  of  the  mercury  would  become  covered  with  the  bright  red  glow,  sur- 
mounted by  a thin  film  of  white  flame,  the  entire  surface  being  thus  covered  until  it  had 
risen  to  within  4 inches  of  the  positive  wire : from  this  point,  as  the  mercury  rose,  the 
diameter  of  the  red  spot  gradually  reduced  in  size,  until  the  mercury  was  within  ‘1  or  '2 
of  an  inch  of  the  wire,  when  the  glow  contracted  into  a minute  spot.  In  this  manner 
the  diameter  of  the  base  of  the  conical  discharge  could  be  measured,  showing  its 
increase  in  proportion  to  the  length  of  the  discharge  when  it  is  not  compressed  by  a 
narrow  tube. 
25.  Although  great  care  had  been  taken  in  boiling  the  mercury  in  this  apparatus,  the 
results  obtained  were  so  unsatisfactory,  in  respect  to  the  stratification,  that  I had  the  mer- 
cury withdrawn  and  the  glass  thoroughly  cleaned : for  this  purpose  it  became  necessary 
to  remove  the  upper  wire ; this  afforded  me  an  opportunity  of  examining  it.  This  wire 
had  been  generally  used  as  the  negative  terminal ; it  was  black  and  rough,  as  if  much 
corroded,  e\idently  from  the  abstraction  of  the  minute  particles  deposited  on  the  surface 
of  the  glass  tube  (14.) ; the  wire  being  resealed  in  the  tube,  which  was  subsequently  filled 
mth  entirely  new  mercury,  a minute  bubble  of  air  remained,  but  the  result  obtained 
was  so  satisfactory,  that  I have  not  attempted  to  remove  it.  In  Plate  I.  fig.  5,  I have 
endeavoured  to  give  a faint  representation  of  the  beautiful  appearance  of  the  stratifica- 
tion, as  exhibited  when  the  upper  wire  is  positive  and  the  lower  negative.  If  the  mer- 
cury is  permitted  to  ascend,  immediately  it  covers  the  negative  wire  the  stratifications 
flisappear,  and  the  inteiior  of  the  globe  is  filled  with  bluish  light ; a bright  spot  of  light 
is  visible  on  the  end  of  the  positive  wire,  but  the  negative  mercury  no  longer  exhibits 
the  red  glow  of  the  former  experiment,  its  entu’e  surface,  until  it  ascends  to  within  an 
inch  of  the  positive  wire,  being  covered  with  a brilliant  white  phosphorescent  film  about 
jth  of  an  inch  in  thickness. 
26.  Fig.  6 represents  an  apparatus  having  for  terminals  platinum  balls  ^th  of  an  inch 
in  diameter  instead  of  the  wires,  the  distance  between  each  being  about  5 inches ; with 
this  apparatus  the  stratifications  at  first  were  nearly  the  same  as  in  the  previous  experi- 
ment with  points  (12.) ; but  when  the  mercury  had  been  permitted  to  rise  and  fall  in 
the  tube  several  times,  a very  marked  difference  in  the  character  of  the  stratification 
was  observed ; instead  of  exhibiting  eight  or  ten  divisions,  they  were  reduced  to  two  or 
three,  of  the  shape  as  in  fig.  6. 
These  stratifications  did  not  take  place  directly  from  the  end  of  the  positive  ball,  but 
from  that  portion  of  the  wire  that  was  immediately  in  connexion  with  the  glass ; by 
carefully  manipulating  with  the  contact  breaker,  the  discharge  could  be  observed  to 
travel  down  the  wire,  overlap  the  baU  and  descend  into  the  tube,  followed  by  another. 
