﻿358 
  PROCEEDINGS 
  OF 
  THE 
  GEOLOGICAL 
  SOCIETY. 
  [Apr. 
  27, 
  

  

  the 
  occasional 
  presence 
  of 
  undigested 
  remains 
  of 
  the 
  Palceonisci 
  be- 
  

   tween 
  the 
  scales 
  of 
  the 
  abdominal 
  region 
  of 
  the 
  Acrolepis 
  would 
  as 
  

   certainly 
  seem 
  to 
  prove 
  it. 
  The 
  occurrence 
  of 
  so 
  many 
  uninjured 
  

   individuals 
  of 
  the 
  Palceonisci 
  along 
  with 
  the 
  Acrolepis 
  would 
  further 
  

   indicate 
  that 
  both 
  the 
  pursuers 
  and 
  the 
  pursued 
  were 
  ultimately 
  

   overtaken 
  by 
  circumstances 
  that 
  rendered 
  powerless 
  their 
  instincts 
  

   in 
  one 
  common 
  catastrophe. 
  

  

  In 
  conclusion 
  I 
  would 
  remark, 
  that 
  though 
  these 
  fossils 
  form, 
  as 
  

   it 
  were, 
  a 
  distinct 
  group 
  in 
  the 
  Permian 
  life-system, 
  they 
  are 
  yet, 
  

   on 
  the 
  one 
  hand, 
  connected 
  with 
  those 
  of 
  earlier 
  Permian 
  date 
  by 
  

   Ulmannia 
  selaginoides, 
  and 
  probably 
  by 
  the 
  Acrolepis 
  and 
  the 
  

   Palceoniscus 
  I 
  have 
  referred 
  to 
  P. 
  angustus, 
  which 
  occur 
  either 
  in 
  

   the 
  Marl-slate 
  of 
  Durham 
  or 
  the 
  Brandschiefer 
  of 
  Saxony 
  ; 
  and, 
  on 
  

   the 
  other 
  hand, 
  they 
  would 
  appear 
  to 
  be 
  connected 
  with 
  the 
  Triassic 
  

   life-groups 
  by 
  the 
  Calamite 
  referred 
  to 
  C. 
  arenaceus. 
  Should 
  the 
  

   identification 
  in 
  this 
  case 
  prove 
  correct, 
  we 
  then 
  shall 
  have 
  at 
  last 
  a 
  

   connecting 
  link 
  of 
  species 
  between 
  the 
  life-systems 
  of 
  the 
  Palaeozoic 
  

   and 
  Mesozoic 
  eras. 
  It 
  would 
  be 
  premature, 
  however, 
  as 
  yet 
  to 
  dis- 
  

   tinctly 
  assert 
  that 
  such 
  a 
  connecting 
  link 
  exists 
  ; 
  for 
  though 
  the 
  

   tendency 
  of 
  the 
  evidence 
  that 
  may 
  be 
  deduced 
  from 
  the 
  fragments 
  

   of 
  the 
  Calamite 
  discovered 
  may 
  be 
  in 
  that 
  direction, 
  the 
  evidence 
  is, 
  

   nevertheless, 
  too 
  imperfect 
  to 
  allow 
  the 
  decision 
  to 
  be 
  clear 
  of 
  

   doubt. 
  

  

  EXPLANATION 
  OF 
  PLATE 
  XYIIL 
  

  

  Fig. 
  1. 
  Palaoniscus 
  alius, 
  Kirkby. 
  Natural 
  size. 
  Old 
  Quarry, 
  Fulwell 
  Hill. 
  

  

  2. 
  Palceoniscus 
  varians, 
  Kirkby. 
  Natural 
  size. 
  New 
  Quarry, 
  Fulwell 
  Hill. 
  

  

  3. 
  Palceoniscus 
  Abbsii, 
  Kirkby. 
  a, 
  anterior 
  portion 
  ; 
  b, 
  posterior 
  portion 
  of 
  

  

  the 
  same 
  fish. 
  Natural 
  size. 
  Old 
  Quarry, 
  Fulwell 
  Hill. 
  

  

  3. 
  On 
  the 
  Fossil 
  Corals 
  of 
  the 
  "West 
  Indian 
  Islands. 
  Part 
  III. 
  

   By 
  P. 
  Martin 
  Duncan, 
  M.B. 
  (Loud.), 
  Sec. 
  G.S. 
  

  

  Contents. 
  

  

  I. 
  Introduction. 
  

  

  II. 
  Changes 
  in 
  recently 
  dead 
  Corals. 
  

  

  1 
  . 
  By 
  the 
  action 
  of 
  the 
  sea. 
  

  

  2. 
  By 
  decomposition 
  of 
  the 
  soft 
  

  

  parts. 
  

  

  3. 
  By 
  fracture 
  from 
  the 
  parent 
  mass. 
  

  

  4. 
  By 
  prefossil 
  accidents 
  attending 
  

  

  the 
  imbedding 
  of 
  Corals. 
  

  

  III. 
  Table 
  of 
  the 
  forms 
  of 
  minerahza- 
  tive. 
  

  

  tion. 
  ' 
  8. 
  Calcareo-siliceous 
  cast 
  

  

  IV. 
  Description 
  of 
  the 
  forms 
  of 
  mine- 
  

   ralization. 
  

  

  1. 
  Calcareous. 
  

  

  2. 
  Siliceous. 
  

  

  3. 
  Siliceous 
  and 
  crystalline. 
  

  

  4. 
  Siliceous 
  and 
  destructive. 
  

  

  5. 
  Siliceous 
  cast. 
  

  

  6. 
  Calcareo-siliceous. 
  

  

  7. 
  Calcareo-siliceous 
  and 
  destruc- 
  

  

  V. 
  Conclusion. 
  

  

  § 
  I. 
  Introduction. 
  

  

  It 
  has 
  been 
  remarked 
  with 
  great 
  truth, 
  that 
  the 
  fossil 
  condition 
  

   of 
  the 
  Antiguan 
  corals 
  renders 
  their 
  specific 
  determination 
  very 
  

  

  