Fields  by  means  of  an  Oscillating  Electric  Needle.       407 
The  lines  of  force  are  shown  as  full  lines,  the  equipotentials 
as  broken  lines.  The  equipotentials  near  the  needle  are  much 
curved,  but  the  effect  diminishes  rapidly  on  going  away  from 
the  needle,  so  that  the  equipotentials  PP  and  P'P'  are 
practically  straight.  Beyond  these  the  equipotential  surfaces 
are  a  series  of  parallel  planes,  and  the  field  is  uniform  and  of 
the  same  value  as  if  the  needle  were  not  present. 
The  result  suggests  the  possibility  of  determining  by  the 
use  of  an  electric  needle  the  specific  inductive  capacity  K  of 
a  dielectric.  For  if  a  pair  of  parallel  plates  kept  at  fixed 
potential-difference  be  used  to  establish  a  uniform  field,  their 
distance  apart  well  exceeding  the  limiting  distance  at  which 
the  disturbing  effect  of  the  needle  is  appreciable  ;  and  if 
between  one  of  the  plates  and  the  critical  plane  a  parallel 
slab  of  the  dielectric  be  introduced,  the  effect  is  to  increase 
the  field  in  the  air  space  by  an  amount  involving  K  in  a 
simple  and  well-known  way.  Of  this  nothing  further  will 
be  said  in  this  paper,  as  careful  experiments  are  being  under- 
taken dealing  with  the  determination  of  specific  inductive 
capacity  by  such  a  method. 
In  regard  to  the  disturbance  introduced  in  other  cases,  it 
is  important  to  notice  that  for  a  given  arrangement  of 
conductors  and  given  position  of  the  needle,  variations  of 
field  due  to  alteration  of  potential  will  be  measured  by  the 
frequency  of  the  needle,  the  ratio  of  frequencies  being 
unaffected  by  the  distortion  of  the  field.  For  example,  in  the 
case  of  the  field  between  a  pair  of  parallel  plates  at  a  distance 
apart  only  slightly  exceeding  the  length  of  the  needle ;  though 
the  frequency  of  the  needle  is  greater,  as  shown  above,  than 
in  an  infinite  field  of  the  same  strength,  yet  it  increases 
exactly  in  proportion  to  the  difference  of  potential  between 
the  bounding  plates. 
The  choice  of  Needles. 
It  is  desirable  that  the  needle  should  be  as  small  as  possible, 
to  reduce  the  disturbing  effect;  and  sensitive,  for  the  measure- 
ment of  as  small  fields  as  possible. 
To  investigate  this  latter  condition  six  cylindrical  needles 
were  made  of  aluminium,  with  dimensions  as  follows  : — 
Three  needles  of  diam.  of  1  mm. ;  lengths  li  cms.,  1  cm.,  -5  em. 
,,  ,,  „  **)mm.;       ,,       1^    ,,      1    ,,     \>    ., 
A  similar  set  was  afterwards  made  in  brass. 
These  needles  were  drilled  centrally  at  right-angles  to  their 
length  with  a  hole  of  about  '3  mm.  in  diameter,  into  which 
was  fitted  a  suspending  wire  of  the  same  diameter,  and  of 
