468     M.  F.  Zollner  on  the  Origin  of  the  Earth's  Magnetism. 
Suppose  we  had  a  magnet  at  the  mean  distance  of  the  earth 
from  the  sun,  the  length  of  which  was  equal  to  the  earth's  diame- 
ter. In  the  most  favourable  condition  a  mass  able  to  be  mag- 
netized on  the  sun's  surface  could  only  be  y-j-J-g^-  nearer  to  one 
end  of  the  needle  than  to  the  other.  The  variation  of  this  frac- 
tion in  consequence  of  the  sun's  rotation  could  not  be  perceptible, 
even  without  regard  to  the  fact  that  any  local  magnetic  pheno- 
menon on  the  sun  must  be  accompanied  by  both  magnetisms, 
which  would  destroy  each  other  in  consequence  of  the  small  dif- 
ference in  distance  at  which  they  are  from  the  earth. 
The  conditions  will  be  found  to  be  quite  different  if  we  con- 
sider the  sun  itself  to  be  a  large  magnet  which,  as  it  does  not 
coincide  with  the  axis  of  rotation,  is  subjected  to  a  mutation,  just 
as  the  magnetic  axis  of  the  earth,  in  consequence  of  the  rotation 
of  the  heavenly  body.  Supposing,  for  instance,  the  magnetic 
poles  of  the  sun  had  heliographic  latitudes  of  70°,  similar  to  that 
of  the  magnetic  north  pole  of  the  earth,  the  variation  produced 
by  their  rotation  would  amount  to  about  one  per  cent  of  the  total 
magnetic  force  exerted  by  the  sun  upon  the  earth.  At  the  same 
time  the  constancy  in  the  situation  of  the  magnetically  active 
poles,  which  is  discussed  by  Hornstein,  is  found  to  be  a  physical 
necessity  under  this  supposition  (which,  according  to  my  theory, 
is  a  consequence  of  the  premises  made). 
If  the  sun,  in  consequence  of  the  continually  unequal  distances 
of  its  two  magnetic  poles  from  the  earth,  acts  magnetically  on 
the  earth,  it  is  clear  that  the  mean  inclination  of  the  magnetic 
axis  of  the  earth  must  have  a  considerable  influence  over  the 
strength  of  the  induction.  It  will  therefore  depend  upon  the 
position  as  well  as  relative  intensity  of  the  earth's  poles  (perhaps 
also  upon  the  secondary  influences  of  heating  and  cooling  in 
summer  and  winter)  what  modifications  the  periodical  variations 
will  undergo  in  the  course  of  the  day  or  the  year. 
It  is  self-evident  that,  according  to  my  theory,  the  distribution 
of  heat  and  cold  on  the  earth's  surface  must  have  an  influence 
on  the  glowing  streams  within  the  earth.  Whether  we  can 
perceive  this  influence  in  the  magnetic  variations  caused  by  it, 
depends  on  its  intensity,  and  cannot  be  decided  a  priori,  but 
only  empirically.  There  are,  however,  some  observations  which 
I  think  can  only  be  explained  by  an  influence  of  the  distribution 
of  temperature,  or  of  the  inner  configuration  of  the  earth's  crust. 
Sabine*,  for  instance,  has  found  in  his  magnetic  observations  at 
Spitzbergen  that  the  influence  of  the  sun's  position  upon  the 
daily  variations  of  declination  is  much  more  decided  and  regular 
than  at  other  places.  On  this  subject  the  reporter  in  Gehler's 
Worterbuch  (vol.  vi.  p.  1097)  says  : — 
*  An  account  of  Experiments  to  determine  the  Figure  of  the  Earth  (4th 
ed.,  London,  1825),  p.  500. 
