Flames  containing  Salt  Vapours  for  Alternating  Currents.     487 
The  conductivity  between  the  electrodes  was  determined 
by  means  o£  a  Wheatstone-bridge  arrangement,  of  which 
the  electrodes  formed  one  arm,  and  the  other  three  arms 
consisted  of  small  air-condensers,  the  capacity  of  one  of 
which  was  adjustable.     The  arrangement  is  shown  in  fig.  2. 
Fiir.  2. 
E.  Flame-Electrodes. 
C\  C.,.  Air-condensers. 
C,.  Adjustable  Air-condenser. 
I).  Detector. 
B.  Cell. 
Ct.  Galvanometer. 
T.  Tesla  Coil. 
J\  J2.  Ley  den  jars. 
S.  Spark-gap. 
I.  Induction-coil. 
An  induction-coil,  I,  charged  up  two  Leyden-jars,  Jl  J2, 
and  these  discharged  at  a  spark-gap  S.  The  outside  coatings 
of  the  jars  were  connected  through  the  primary  of  a  Tesla 
coil,  T.  The  primary  of  this  coil  consisted  of  33  turns  wound 
into  a  spiral,  29  cms.  long  and  19*2  cms.  in  diameter,  on  a 
large  glass  cylinder.  The  secondary  coil  had  three  turns,  and 
was  placed  inside  the  glass  cylinder,  halfway  up  it.  It  was 
connected  to  the  bridge  arrangement  at  A  and  B,  as  shown. 
An  "  electrolytic  detector/''  D,  was  eonnected  to  the  points 
M  and  N  of  the  bridge.  This  detector  consisted  of  two 
platinum  electrodes  dipping  into  20  per  cent,  sulphuric  acid. 
One  electrode  was  a  platinum  cylinder  3  cms.  in  diameter 
and  4  cms.  high,  while  the  other  was  a  platinum  wire 
fooo  inch  in  diameter  sealed  into  a  glass  tube  and  cut  off 
close  to  the  surface  of  the  glass.  The  large  electrode  was 
contained  in  a  test-tube,  which  it  just  fitted,  and  the  glass 
tube  carrying  the  small  electrode  was  supported  by  a  cork 
fitting  the  test-tube.  The  small  electrode  was  just  below  the 
surface  of  the  acid  at  the  middle  of  the  tube.  The  two  elec- 
trodes   were   connected    io    a    silver-chloride    cell    P>   ami    a 
