﻿ROTATION 
  OF 
  PILLARS. 
  215 
  

  

  if 
  the 
  observer 
  look 
  due 
  south 
  at 
  a 
  square 
  pyramid, 
  for 
  example, 
  whose 
  sides 
  

   stood 
  cardinal, 
  and 
  it 
  be 
  tilted 
  by 
  the 
  first 
  semiphase 
  of 
  a 
  shock 
  from 
  east 
  to 
  

   west, 
  the 
  pyramid 
  will 
  tilt 
  or 
  rise 
  upon 
  the 
  eastern 
  edge 
  of 
  its 
  base 
  ; 
  and 
  if, 
  before 
  

   it 
  has 
  had 
  time 
  to 
  fall 
  back, 
  it 
  be 
  acted 
  on 
  by 
  another 
  shock 
  from 
  north 
  to 
  south, 
  

   the 
  pyramid 
  will 
  rotate, 
  upon 
  the 
  bisection 
  or 
  on 
  some 
  other 
  point, 
  of 
  the 
  edge 
  

   on 
  which 
  it 
  momentarily 
  rested, 
  and 
  will 
  hence 
  come 
  to 
  repose, 
  after 
  having 
  twisted 
  

   from 
  left 
  to 
  right, 
  or 
  with 
  the 
  hands 
  of 
  a 
  watch. 
  

  

  If 
  the 
  tilting 
  up, 
  had 
  been 
  produced 
  by 
  the 
  second 
  semiphase, 
  of 
  the 
  same 
  shock 
  

   from 
  east 
  to 
  west, 
  then 
  the 
  pyramid 
  would 
  have 
  risen 
  upon 
  the 
  western 
  edge 
  of 
  its 
  

   base, 
  and 
  the 
  same 
  direction 
  (north 
  to 
  south) 
  of 
  second 
  shock, 
  would 
  have 
  produced 
  

   rotation 
  upon 
  that 
  edge, 
  but 
  in 
  a 
  contrary 
  direction 
  to 
  the 
  proceeding 
  or 
  from 
  right 
  

   to 
  left 
  or 
  against 
  the 
  hands 
  of 
  a 
  watch. 
  

  

  Again, 
  if, 
  on 
  the 
  first 
  supposition, 
  the 
  first 
  semiphase 
  of 
  the 
  east 
  to 
  west 
  shock 
  

   had 
  tilted 
  the 
  pyramid 
  upon 
  its 
  eastern 
  edge 
  of 
  base, 
  but 
  the 
  second 
  shock 
  had 
  

   been 
  from 
  south 
  to 
  north, 
  in 
  place 
  of 
  the 
  reverse 
  as 
  before, 
  then 
  the 
  rotation 
  would 
  

   have 
  been 
  from 
  right 
  to 
  left; 
  and 
  if 
  tilted 
  by 
  the 
  second 
  semiphase 
  on 
  the 
  western 
  

   edge, 
  the 
  second 
  shock, 
  south 
  to 
  north, 
  would 
  produce 
  rotation 
  left 
  to 
  right. 
  

  

  It 
  would, 
  therefore, 
  appear 
  at 
  first 
  impossible, 
  to 
  determine 
  the 
  direction 
  of 
  

   motion 
  in 
  transit, 
  of 
  either 
  shock, 
  from 
  such 
  an 
  observation 
  : 
  we 
  can, 
  however, 
  

   generally 
  discover 
  upon 
  which 
  edge 
  of 
  the 
  base 
  any 
  heavy 
  body 
  of 
  stone 
  or 
  

   masonry 
  has 
  tilted, 
  by 
  the 
  abrasion 
  or 
  splintering 
  of 
  the 
  arris, 
  and 
  the 
  rotation 
  

   must 
  have 
  taken 
  place 
  round 
  some 
  point 
  in 
  that 
  edge. 
  If, 
  therefore, 
  we 
  know 
  

   the 
  direction 
  of 
  either 
  one 
  of 
  the 
  two 
  shocks, 
  we 
  can 
  always 
  discover 
  that 
  of 
  the 
  

   other, 
  by 
  the 
  rotation 
  observed, 
  and 
  if 
  the 
  time 
  of 
  oscillation 
  of 
  the 
  body 
  be 
  ascer- 
  

   tainable, 
  we 
  are 
  enabled 
  to 
  calculate 
  a 
  major 
  limit, 
  for 
  the 
  interval 
  of 
  time 
  that 
  

   must 
  have 
  elapsed, 
  between 
  the 
  arrival 
  at 
  the 
  twisted 
  body 
  of 
  the 
  first 
  and 
  of 
  the 
  

   second 
  shock, 
  when 
  both 
  the 
  wave-paths 
  are 
  known. 
  

  

  With 
  a 
  single 
  instance 
  of 
  such 
  twisting, 
  it 
  may 
  be 
  impossible 
  to 
  decide 
  

   whether 
  the 
  twist 
  has 
  been 
  due 
  to 
  one 
  shock 
  (1st 
  case), 
  or 
  to 
  two 
  shocks 
  in 
  suc- 
  

   cession 
  (2nd 
  case) 
  ; 
  but 
  when 
  several 
  bodies 
  alike 
  or 
  dissimilar, 
  at 
  the 
  same 
  

   locality, 
  are 
  all 
  found 
  twisted 
  in 
  one 
  direction, 
  it 
  is 
  certain 
  to 
  have 
  been 
  the 
  work 
  

   of 
  two 
  distinct 
  shocks, 
  for 
  it 
  is 
  beyond 
  the 
  reach 
  of 
  probability 
  that 
  several 
  

   bodies 
  should 
  all 
  happen 
  to 
  have 
  their 
  respective 
  centres 
  of 
  adherence, 
  at 
  the 
  

   same 
  side 
  of 
  their 
  respective 
  centres 
  of 
  gravity, 
  and 
  unless 
  they 
  have, 
  some 
  will 
  

   rotate 
  in 
  one, 
  some 
  in 
  the 
  other 
  direction 
  by 
  any 
  single 
  shock 
  ; 
  rotation 
  thus 
  pro- 
  

   duced, 
  being 
  always 
  by 
  the 
  centre 
  of 
  gravity 
  moving 
  contrary 
  to 
  the 
  first 
  or 
  

   second 
  semiphase 
  of 
  the 
  wave 
  and 
  carried 
  round 
  the 
  centre 
  of 
  adherence, 
  by 
  the 
  

   line 
  joining 
  them 
  as 
  a 
  radius 
  vector 
  ; 
  the 
  inertia 
  of 
  motion 
  at 
  the 
  centre 
  of 
  gravity, 
  

   and 
  the 
  resistance 
  of 
  the 
  point 
  of 
  rotation 
  in 
  the 
  edge 
  of 
  the 
  base, 
  or 
  of 
  the 
  centre 
  

   of 
  adherence, 
  forming 
  in 
  every 
  case, 
  the 
  extremities 
  of 
  the 
  dynamic 
  couple. 
  " 
  

  

  There 
  remains 
  one 
  more 
  explanation 
  to 
  be 
  mentioned 
  which 
  on 
  

   account 
  of 
  its 
  simplicity, 
  has 
  been 
  very 
  generally 
  adopted. 
  It 
  is 
  that 
  

   of 
  Mr. 
  Gray 
  and 
  contained 
  in 
  Professor 
  J. 
  Milne's 
  account 
  of 
  the 
  

   Japanese 
  Earthquake 
  of 
  22nd 
  February, 
  1880. 
  After 
  mentioning 
  a 
  

  

  ( 
  215 
  ) 
  

  

  