﻿OF 
  THE 
  WESTERN 
  ISLES 
  OF 
  SCOTLAND. 
  217 
  

  

  materials 
  found 
  their 
  way 
  from 
  below. 
  The 
  materials 
  filling 
  these 
  

   conduits, 
  I 
  argued, 
  would 
  have 
  consolidated 
  slowly 
  and 
  under 
  great 
  

   pressure, 
  and 
  I 
  therefore 
  inferred 
  that 
  while 
  agreeing 
  in 
  ultimate 
  

   chemical 
  composition 
  with 
  the 
  lavas, 
  they 
  would 
  exhibit 
  a 
  more 
  

   highly 
  crystalline 
  structure. 
  I 
  then 
  showed 
  that 
  in 
  the 
  great 
  bosses, 
  

   sheets, 
  and 
  dykes 
  composed 
  of 
  gabbro 
  and 
  granite 
  we 
  have 
  just 
  such 
  

   masses 
  of 
  rock 
  — 
  occupying 
  precisely 
  the 
  positions 
  they 
  might 
  be 
  

   expected 
  to 
  do, 
  if 
  the 
  interpretation 
  which 
  I 
  gave 
  of 
  the 
  structure 
  of 
  

   the 
  country 
  was 
  the 
  true 
  one. 
  On 
  the 
  other 
  hand, 
  I 
  maintain 
  that 
  

   the 
  long 
  narrow 
  dykes 
  are 
  altogether 
  inadequate 
  as 
  the 
  sole 
  channels 
  

   for 
  the 
  extrusion 
  of 
  these 
  enormous 
  volumes 
  of 
  lava. 
  

  

  The 
  hypothesis 
  of 
  "fissure-eruption 
  " 
  appears 
  to 
  me 
  now, 
  as 
  it 
  has 
  

   always 
  done 
  *, 
  as 
  unnecessary 
  as 
  it 
  is 
  vague. 
  Effusive 
  action, 
  the 
  

   welling-out 
  of 
  great 
  masses 
  of 
  lava, 
  is 
  as 
  much 
  a 
  characteristic 
  of 
  

   volcanoes 
  as 
  is 
  the 
  dispersal 
  of 
  scorise, 
  lapilli, 
  and 
  dust 
  by 
  explosive 
  

   action. 
  In 
  the 
  Sandwich 
  Islands 
  we 
  have 
  an 
  admirable 
  example 
  

   of 
  the 
  formation 
  of 
  plateaux 
  built 
  up 
  of 
  basaltic 
  lava-streams, 
  with 
  

   only 
  small 
  and 
  rare 
  intercalations 
  of 
  volcanic 
  tuffs. 
  In 
  the 
  great 
  

   tuff-cones 
  of 
  Java 
  and 
  their 
  remarkably 
  explosive 
  outbursts 
  we 
  have 
  

   the 
  other 
  extreme 
  type 
  of 
  volcanic 
  activity 
  admirably 
  exemplified. 
  

   The 
  study 
  of 
  the 
  recent 
  tremendous 
  catastrophe 
  at 
  Krakatoa 
  was 
  

   particularly 
  interesting, 
  as 
  affording 
  to 
  us 
  a 
  striking 
  instance 
  of 
  the 
  

   explosive 
  type 
  of 
  volcanic 
  activity, 
  so 
  strongly 
  contrasted 
  with 
  the 
  

   effusive 
  type, 
  as 
  illustrated 
  in 
  the 
  Sandwich 
  Isles. 
  But 
  I 
  am 
  quite 
  

   at 
  a 
  loss 
  to 
  understand 
  why 
  the 
  explosive 
  action 
  that 
  has 
  produced 
  

   the 
  great 
  tuff-cones 
  of 
  Java 
  should 
  be 
  regarded 
  as 
  " 
  ordinary 
  volcanic 
  

   activity," 
  w 
  T 
  hile 
  the 
  effusive 
  action 
  which 
  has 
  given 
  rise 
  to 
  the 
  basaltic 
  

   plateaux 
  of 
  the 
  Sandwich 
  Islands 
  should 
  be 
  denied 
  that 
  title. 
  

  

  Professor 
  Dana, 
  after 
  very 
  clearly 
  and 
  forcibly 
  pointing 
  out 
  the 
  

   distinction 
  between 
  the 
  two 
  types 
  of 
  volcanic 
  activity, 
  very 
  justly 
  

   adds 
  : 
  — 
  " 
  The 
  marked 
  contrast 
  between 
  volcanoes 
  of 
  the 
  Mount 
  Loa 
  

   and 
  Vesuvius 
  types, 
  based 
  on 
  the 
  liquidity 
  of 
  the 
  lava, 
  making 
  Mt. 
  

   Loa 
  discharges 
  to 
  be 
  almost 
  solely 
  outflows, 
  and 
  those 
  of 
  Vesuvius 
  

   both 
  upthrows 
  of 
  cinders 
  and 
  outflows 
  of 
  lava, 
  has 
  been 
  sufficiently 
  

   explained. 
  With 
  this 
  exception, 
  the 
  contrast 
  as 
  to 
  their 
  eruptions 
  

   as 
  well 
  as 
  to 
  their 
  ordinary 
  action 
  is 
  far 
  less 
  than 
  is 
  generally 
  

   supposed. 
  There 
  is 
  no 
  reason 
  to 
  regard 
  the 
  forces 
  as 
  different 
  in 
  

   kind 
  or 
  mode 
  of 
  action 
  "t. 
  

  

  The 
  volcanic 
  outbursts 
  of 
  Iceland 
  in 
  recent 
  times 
  — 
  and 
  there 
  is 
  

   every 
  reason 
  for 
  believing 
  that 
  those 
  of 
  the 
  British 
  Islands 
  during 
  

   the 
  Tertiary 
  periods 
  were 
  precisely 
  similar 
  to 
  them 
  — 
  combine 
  in 
  a 
  

   most 
  striking 
  and 
  instructive 
  manner 
  the 
  explosive 
  and 
  the 
  effusive 
  

   types 
  of 
  activity. 
  The 
  eruption 
  of 
  Varmadalr 
  (Skaptar 
  Jokull) 
  in 
  

   1783 
  was 
  not 
  more 
  remarkable 
  for 
  the 
  enormous 
  volume 
  and 
  great 
  

   liquidity 
  of 
  the 
  lavas, 
  than 
  for 
  the 
  violently 
  explosive 
  action 
  which 
  

   accompanied 
  their 
  extrusion 
  — 
  these 
  explosions 
  producing 
  effects 
  quite 
  

   comparable 
  to 
  those 
  which 
  resulted 
  from 
  the 
  Krakatoa 
  outburst, 
  just 
  

   a 
  century 
  later 
  ; 
  the 
  Icelandic 
  eruption 
  moreover 
  resulted 
  in 
  the 
  

  

  * 
  Volcanoes, 
  what 
  they 
  are 
  and 
  what 
  they 
  teach 
  (1881), 
  p. 
  188. 
  

   t 
  Amer. 
  Journ. 
  Sci. 
  ser. 
  3, 
  vol. 
  xxxvi. 
  pp. 
  172, 
  173. 
  

  

  