﻿PYCNODONTID^. 
  195 
  

  

  to 
  the 
  presence 
  of 
  a 
  bony 
  base 
  in 
  the 
  one 
  case 
  and 
  its 
  absence 
  in 
  

   the 
  other. 
  Moreover, 
  in 
  comparing 
  the 
  detached 
  vomers 
  and 
  

   splenials 
  of 
  Pycnodonts, 
  the 
  smaller 
  specimens 
  cannot 
  be 
  compared 
  

   as 
  a 
  whole 
  with 
  the 
  larger 
  ones 
  — 
  they 
  must 
  be 
  compared 
  only 
  with 
  

   that 
  portion 
  of 
  the 
  latter 
  which 
  happens 
  to 
  be 
  of 
  corresponding 
  

   size. 
  

  

  The 
  notochord 
  is 
  always 
  persistent, 
  and 
  no 
  ossifications 
  which 
  

   can 
  be 
  ascribed 
  to 
  its 
  sheath 
  have 
  hitherto 
  been 
  observed. 
  The 
  

   neural 
  and 
  haemal 
  arches, 
  however, 
  are 
  invariably 
  well-ossified, 
  and 
  

   their 
  bases 
  are 
  so 
  much 
  expanded 
  that 
  they 
  sometimes 
  {Palceo- 
  

   balistum, 
  Pycnodus) 
  completely 
  encircle 
  the 
  notochord. 
  Ribs 
  are 
  

   present 
  ; 
  and 
  those 
  of 
  Gyrodus 
  and 
  Palceobalistum 
  are 
  distinctly 
  

   observed 
  to 
  exhibit 
  the 
  laminar 
  expansion 
  so 
  characteristic 
  of 
  the 
  

   neural 
  spines 
  throughout 
  the 
  trunk 
  and 
  of 
  the 
  haemal 
  spines 
  in 
  the 
  

   caudal 
  region. 
  This 
  expansion 
  is 
  commonly 
  confined 
  to 
  the 
  anterior 
  

   edge 
  of 
  each 
  spine, 
  but 
  in 
  Gyrodus 
  and 
  Palceobalistum 
  it 
  occurs 
  

   both 
  anteriorly 
  and 
  posteriorly 
  ; 
  and 
  in 
  the 
  last-mentioned 
  genus 
  

   the 
  lamina? 
  of 
  adjacent 
  spines 
  are 
  actually 
  united 
  for 
  a 
  considerable 
  

   length 
  in 
  a 
  jagged 
  suture. 
  The 
  neural 
  spines 
  in 
  the 
  abdominal 
  

   region 
  extend 
  to 
  the 
  dorsal 
  margin 
  of 
  the 
  fish, 
  and 
  are 
  often 
  

   thickened 
  apparently 
  for 
  the 
  support 
  of 
  the 
  ridge-scales. 
  The 
  

   haemal 
  spines 
  supporting 
  the 
  caudal 
  fin 
  are 
  also 
  more 
  or 
  less 
  

   expanded, 
  but 
  never 
  fused 
  together. 
  No 
  intermuscular 
  bones 
  have 
  

   been 
  seen. 
  

  

  The 
  precise 
  mode 
  of 
  suspension 
  of 
  the 
  pectoral 
  arch 
  to 
  the 
  

   cranium 
  still 
  remains 
  to 
  be 
  discovered; 
  but 
  the 
  clavicle 
  is 
  well 
  

   known, 
  and 
  some 
  specimens 
  of 
  Gyrodus 
  appear 
  to 
  exhibit 
  a 
  supra- 
  

   clavicle 
  extending 
  towards 
  the 
  back 
  of 
  the 
  cranial 
  roof. 
  The 
  

   clavicle 
  is 
  much 
  vertically 
  elongated, 
  tapering 
  above, 
  and 
  with 
  a 
  

   spatulate 
  expansion 
  below 
  ; 
  a 
  very 
  small 
  surface 
  is 
  exposed 
  on 
  the' 
  

   flank 
  of 
  the 
  fish. 
  There 
  is 
  no 
  infra-clavicle. 
  The 
  scapula 
  and 
  

   coracoid 
  cannot 
  yet 
  be 
  described, 
  but 
  the 
  pectoral 
  fin 
  is 
  attached 
  

   just 
  above 
  the 
  inferior 
  expansion 
  of 
  the 
  clavicle. 
  The 
  base 
  of 
  this 
  

   fin 
  forms 
  a 
  little 
  lobe, 
  distinctly 
  covered 
  with 
  thin 
  rounded 
  scales 
  

   in 
  Gyrodus 
  ; 
  and 
  a 
  single 
  series 
  of 
  seven 
  or 
  eight 
  basals 
  can 
  be 
  

   readily 
  counted 
  in 
  specimens 
  both 
  of 
  Gyrodus 
  and 
  Palceobalistum, 
  

   while 
  some 
  vacant 
  space 
  may 
  have 
  been 
  occupied 
  by 
  still 
  more 
  of 
  

   these 
  elements. 
  The 
  pelvic 
  fins, 
  though 
  often 
  quite 
  rudimentary, 
  

   are 
  exhibited 
  in 
  all 
  the 
  genera 
  of 
  which 
  the 
  trunk 
  is 
  satisfactorily 
  

   known. 
  The 
  rays 
  of 
  the 
  dorsal 
  and 
  anal 
  fins 
  are 
  equal 
  in 
  number 
  

   to 
  their 
  supports, 
  which 
  are 
  more 
  numerous 
  than 
  the 
  vertebral 
  

   arches 
  in 
  the 
  same 
  region 
  ; 
  but 
  each 
  fin-ray 
  is 
  sometimes 
  liable 
  to 
  

   be 
  subdivided 
  into 
  its 
  right 
  and 
  left 
  halves 
  by 
  crushing, 
  and 
  there 
  

  

  o2 
  

  

  