﻿12 
  DR. 
  CHA.ELES 
  DAVISON 
  ON 
  THE 
  [Feb. 
  1 
  89 
  7, 
  

  

  chief 
  seat 
  of 
  earthquakes 
  is 
  in 
  the 
  rectangle 
  D 
  in 
  1885 
  and 
  E 
  in 
  

   1886, 
  1888, 
  and 
  1889 
  ; 
  the 
  latter 
  being 
  the 
  second 
  in 
  numerical 
  

   importance, 
  and 
  the 
  most 
  persistent, 
  in 
  the 
  district 
  throughout 
  the 
  

   eight 
  years 
  considered. 
  

  

  Kemembering 
  that 
  the 
  rectangles 
  A-P 
  form 
  only 
  13 
  per 
  cent, 
  of 
  

   the 
  whole 
  area, 
  it 
  is 
  thus 
  clear 
  that 
  the 
  frequency 
  of 
  earthquakes 
  

   was 
  relatively 
  greater 
  within 
  the 
  fault-region, 
  1 
  and 
  that 
  the 
  seismic 
  

   importance 
  of 
  the 
  region 
  increased 
  considerably 
  within 
  about 
  two 
  

   years 
  before 
  the 
  great 
  earthquake. 
  

  

  12. 
  The 
  distribution 
  of 
  epicentres 
  during 
  the 
  interval, 
  January 
  

   1st, 
  1890, 
  to 
  October 
  27th, 
  1891, 
  is 
  illustrated 
  in 
  fig. 
  2 
  (p. 
  4). 
  The 
  

   total 
  number 
  is, 
  however, 
  so 
  small 
  comparatively 
  that 
  the 
  essential 
  

   features 
  of 
  the 
  distribution 
  are 
  only 
  brought 
  out 
  by 
  adding 
  curves 
  1 
  

   and 
  2. 
  While 
  the 
  rectangle 
  E 
  is 
  still 
  the 
  chief 
  seat 
  of 
  seismic 
  activity, 
  

   it 
  will 
  be 
  seen 
  at 
  a 
  glance 
  how 
  the 
  curves 
  follow 
  the 
  lines 
  of 
  the 
  main 
  

   and 
  secondary 
  faults. 
  In 
  this 
  as 
  in 
  other 
  maps, 
  there 
  is 
  a 
  discon- 
  

   tinuity 
  in 
  the 
  curves 
  in 
  part 
  of 
  the 
  main 
  fault 
  between 
  the 
  points 
  

   B 
  and 
  C. 
  Otherwise, 
  earthquakes 
  seem 
  to 
  have 
  occurred 
  along 
  the 
  

   whole 
  fault-system, 
  especially 
  in 
  the 
  secondary 
  fault 
  and 
  the 
  sup- 
  

   posed 
  continuation 
  of 
  the 
  main 
  fault 
  towards 
  the 
  south-east. 
  A 
  

   noteworthy 
  feature 
  is 
  the 
  uniformity 
  in 
  the 
  distribution 
  of 
  epicentres 
  

   throughout 
  the 
  fault-region 
  ; 
  the 
  marked 
  concentration 
  of 
  effort 
  

   which 
  characterizes 
  the 
  after-shocks 
  is 
  hardly 
  perceptible 
  here. 
  

   And 
  this 
  uniformity 
  (as 
  will 
  be 
  seen 
  from 
  Table 
  I., 
  p. 
  9) 
  is 
  still 
  more 
  

   noticeable 
  in 
  the 
  earthquakes 
  of 
  the 
  year 
  1890 
  alone. 
  

  

  13. 
  An 
  important 
  point 
  to 
  be 
  decided 
  is 
  how 
  far 
  this 
  remarkable 
  

   distribution 
  is 
  a 
  result 
  of 
  the 
  strong 
  earthquake 
  of 
  May 
  12th, 
  1889. 
  

   The 
  number 
  of 
  shocks 
  in 
  the 
  district 
  in 
  1890 
  was 
  52*5, 
  but 
  in 
  1889 
  

   was 
  only 
  27*5, 
  of 
  which 
  13 
  occurred 
  before 
  May 
  12th, 
  and 
  13*5 
  after. 
  

   Indeed, 
  the 
  earthquake 
  of 
  May 
  12th 
  seems 
  to 
  have 
  been 
  almost 
  

   unattended 
  by 
  after-shocks, 
  the 
  number 
  of 
  earthquakes 
  felt 
  in 
  the 
  

   district 
  during 
  the 
  last 
  eight 
  months 
  of 
  1889 
  being 
  respectively 
  1, 
  1, 
  

   1, 
  2, 
  3*5, 
  4, 
  0, 
  and 
  1. 
  I 
  think, 
  then, 
  that 
  we 
  may 
  conclude 
  that 
  the 
  

   distribution 
  of 
  earthquakes 
  in 
  1890 
  and 
  1891 
  was 
  in 
  very 
  small 
  

   measure, 
  if 
  at 
  all, 
  the 
  result 
  of 
  the 
  shock 
  of 
  May 
  12th, 
  1889, 
  but 
  is 
  

   rather 
  to 
  be 
  regarded 
  as 
  a 
  preparation 
  for 
  the 
  great 
  earthquake, 
  the 
  

   consequent 
  relief 
  at 
  numerous 
  and 
  widely-distributed 
  points 
  equal- 
  

   izing 
  the 
  effective 
  stress 
  along 
  the 
  whole 
  fault-system, 
  and 
  so 
  clearing 
  

   the 
  way 
  for 
  one 
  or 
  more 
  almost 
  instantaneous 
  slips 
  throughout 
  its 
  

   entire 
  length. 
  

  

  This 
  outlining 
  of 
  the 
  fault-system 
  by 
  the 
  frequency-curves 
  of 
  

   1890-91 
  points, 
  therefore, 
  to 
  the 
  previous 
  existence 
  of 
  the 
  faults, 
  

   and 
  implies 
  that 
  the 
  great 
  earthquake 
  was 
  due, 
  not 
  to 
  the 
  rupturing 
  

   of 
  strata, 
  but 
  probably 
  to 
  the 
  intense 
  friction 
  called 
  into 
  action 
  by 
  

   the 
  sudden 
  displacement. 
  2 
  

  

  1 
  As 
  53 
  epicentres 
  were 
  situated 
  during 
  1885-89 
  within 
  these 
  13 
  rectangles, 
  

   and 
  72 
  in 
  the 
  remaining 
  87 
  rectangles, 
  it 
  follows 
  that 
  the 
  average 
  frequency 
  in 
  

   one 
  of 
  the 
  rectangles 
  of 
  the 
  fault-region 
  was 
  five 
  times 
  as 
  great 
  as 
  in 
  one 
  of 
  those 
  

   outside 
  it. 
  

  

  2 
  It 
  would 
  be 
  interesting 
  to 
  determine 
  in 
  future 
  studies 
  whether 
  this 
  uniformity 
  

   in 
  distribution 
  along 
  a 
  fault 
  is 
  a 
  common 
  feature 
  of 
  preparatory 
  shocks. 
  If 
  so, 
  

   it 
  would 
  form 
  some 
  rough 
  indication 
  of 
  the 
  approach 
  of 
  a 
  great 
  earthquake. 
  

  

  