and  the  Magnetic  Relations  of  the  Heavenly  Bodies.       465 
an  influence  of  the  above  kind  by  the  rotation  of  its  magnetic 
poles. 
Hornstein  justly  takes  for  the  time  of  rotation  of  the  sun  that 
of  the  equatorial  zone  as  found  by  the  observations  of  Carring- 
ton  and  Sporer.  For  I  have  shown,  in  my  paper  "  On  the  Law 
of  the  Sun's  Rotation/'  that  the  curious  difference  in  the  times 
of  rotation  for  different  heliographic  latitudes  is  only  generated 
by  the  friction  against  the  polar  undercurrents;  audit  is  there- 
fore clear  that  even  the  equatorial  part  of  the  sun  rotates  in  a 
time  slightly  different  from  that  of  the  inner  nucleus,  with  which 
time  the  magnetic  poles  are  connected. 
Hornstein  says,  speaking  of  the  difference  of  rotation  for  dif- 
ferent heliographic  latitudes  :  — 
"  If  we  are  unwilling  to  make  suppositions  which  are  in  con- 
tradiction with  the  fundamental  laws  of  mechanics,  we  must 
assume  the  time  of  rotation  of  the  sun-spots  at  the  equator 
(24* 541  days),  or  one  very  near  it,  to  approach  nearest  to  the  real 
time  of  rotation  of  the  sun." 
Then  Hornstein  explains  the  reasons  which  first  led  him  to  the 
idea  that  possibly  the  magnetic  variations  on  the  earth  might 
show  a  dependence  upon  the  time  of  rotation  of  the  sun. 
The  words  of  Hornstein  are  the  following  : — 
"  It  has  been,  as  is  known,  observed  several  times  during  the 
last  few  years  that  exceptional  changes  on  the  sun's  surface  hap- 
pened simultaneously  with  great  changes  of  the  direction  and 
force  of  the  earth's  magnetism.  At  the  same  time  it  has  been 
proved,  by  the  important  researches  of  Sabine,  Wolf,  Lamont, 
and  others,  that  the  mean  daily  variations  of  magnetic  declina- 
tion show  the  same  period  of  eleven  years  as  the  sun-spots, 
and  that  the  same  thing  holds  for  the  variation  of  horizontal  in- 
tensity. This  is  a  proof  that  great  changes  on  the  sun's  surface 
(which  are  most  likely  nothing  but  great  revolutions  in  the  sun's 
nucleus)  may  cause  changes  in  the  elements  of  the  magnetic 
force  of  the  earth. 
"  Different  conditions  on  the  surface  of  the  sun  do  not  only 
take  place  in  the  course  of  the  eleven-year  period  of  the  sun- 
spots,  they  are  present  at  the  same  time  and  by  the  side  of  each 
other,  if  we  look  at  regions  of  different  heliographical  longitudes 
of  the  sun-spot  zone.  As  by  the  rotation  of  the  sun  all  these 
regions  are  turned  one  after  the  other  towards  the  earth,  and  as 
during  this  period  every  point  of  the  zone  mentioned  changes  its 
distance  from  the  earth  nearly  by  an  entire  diameter  of  the  sun, 
I  came  to  the  idea  to  look  whether  I  could  not  find  periodical 
changes  in  the  elements  of  the  earth's  magnetism,  the  period  of 
which  is  equal  to  the  synodic*  time  of  rotation  of  the  sun,  or  auy 
aliquot  part  thereof. 
*  That  is,  with  regard  to  the  earth's  movement,  the  time  required  for 
Phil.  Mag.  S.  4.  Vol.  13.  No.  288.  June  1872.  2  K 
