420 
Prof.  Challis  on  the  Hydrodynamical 
verging  streams  of  the  magnet  at  the  north  end  and  the  circular 
part  of  the  galvanic  current  to  be  resolved  in  a  horizontal  plane 
passing  through  the  axis  of  the  needle  and  perpendicularly  to 
that  axis,  the  two  portions  will  concur  on  the  east  side  of  the 
needle  and  be  opposed  to  each  other  on  the  west  side,  so  that 
the  needle  will  be  urged  eastward.  Similarly  at  the  south  end 
the  circular  part  of  the  galvanic  stream  and  the  resolved  parts  of 
the  divergent  magnetic  streams  concur  in  direction  on  the  west 
side  and  ai?e  in  opposite  directions  on  the  east  side,  and  that  end 
will  thus  be  urged  westward.  Hence  on  both  accounts  the  north 
end  deviates  towards  the  east.  But  according  to  Ganot  and 
Atkinson  the  deviation  in  this  case  of  the  experiment  is  towards 
the  west.  What,  then,  is  the  explanation  of  this  disagreement  ? 
Simply  that  the  experimental  result,  as  given  by  Ganot,  is  incon- 
sistent with  the  course  of  the  current  from  the  zinc  to  the  copper, 
as  indicated  by  the  signs  -f-  and  —  and  by  arrows.  The  result 
for  the  same  case  of  the  experiment,  as  stated  by  M.  de  la  Rive 
(Traite  de  VElectricite,  vol.  i.  p.  207),  and  by  Mr.  Airy  (Trea- 
tise on  Magnetism,  top  of  p.  210),  accords  with  the  above  deduc- 
tion from  the  theory  in  giving  an  eastward  deviation.  It  is, 
further,  to  be  remarked  that  Ganot's  statements  (in  art.  627)  of 
the  results  of  the  experiment  in  the  four  cases,  together  with  the 
indicated  directions  of  the  current,  are  in  agreement  with  Am- 
pere's well-known  rule  for  determining  the  direction  of  the  de- 
viation of  the  north  end  of  the  needle,  and  would  also  agree  with 
the  lrypothesis  of  spiral  motion,  if  that  motion  might  be  assumed 
to  be  sinistrorsum  when  the  direction  of  the  current  is  from  the 
zinc  to  the  copper.  But  this  supposition  is  inadmissible,  be- 
cause experiment  taken  in  conjunction  with  the  theory  shows 
conclusively  that  either  the  current  is  from  the  zinc  to  the  cop- 
per and  the  spiral  motion  dextrorsum,  or  from  the  copper  to  the 
zinc  and  the  spiral  motion  sinistrorsum;  but  neither  theory 
nor  experiment  appears  at  present  to  be  capable  of  deciding 
which  of  these  laws  is  the  true  one.  In  any  case  the  inconsistent 
statements  of  experimental  results  I  have  referred  to,  which  have 
caused  me  a  great  deal  of  perplexity,  require  to  be  rectified. 
All  the  other  cases  of  Oersted's  experiment  may  be  similarly 
explained  by  the  hydrodynamical  theory  on  the  same  hypotheses. 
35.  One  of  the  most  remarkable  phenomena  relating  to  mag- 
netism is  the  effect  which  a  mass  of  copper  in  the  neighbourhood 
of  a  magnetic  needle  has  upon  the  number  and  extent  of  its  vi- 
brations. The  hydrodynamical  theory,  combined  with  a  simple 
magneto-galvanic  law  established  experimentally  by  Faraday, 
offers  the  following  explanation  of  this  fact  and  of  others  of 
the  same  class.  Faraday  found  that  when  a  plate  of  copper 
(1J  inch  wide,  I  of  an  inch  thick,  and  12  inches  long)  was 
