﻿A 
  MONOGEAPH 
  OF 
  THE 
  VICTORIAN 
  SPONGES. 
  43 
  

  

  of 
  Bowerbank 
  as 
  the 
  sole 
  genus 
  of 
  the 
  group. 
  In 
  this 
  I 
  agree 
  with 
  Polejaeff, 
  but 
  at 
  

   the 
  same 
  time, 
  as 
  a 
  mere 
  matter 
  of 
  convenience 
  and 
  not 
  necessarily 
  as 
  indicating 
  

   genetic 
  relationship, 
  I 
  prefer 
  to 
  subdivide 
  the 
  genus 
  into 
  sections 
  and 
  subsections 
  

   according 
  to 
  the 
  nature 
  of 
  the 
  canal 
  system 
  of 
  the 
  colony. 
  I 
  prefer 
  to 
  make 
  use 
  of 
  

   the 
  canal 
  system 
  for 
  this 
  purpose 
  rather 
  than 
  the 
  spiculation, 
  for 
  I 
  have 
  frequently 
  

   found 
  the 
  spiculation 
  varying 
  greatly 
  in 
  different 
  [)arts 
  of 
  the 
  same 
  colony 
  but 
  never 
  

   so 
  the 
  canal 
  system 
  (except 
  in 
  cases 
  where 
  a 
  later-formed 
  portion 
  of 
  the 
  colony 
  

   shows 
  an 
  advance 
  in 
  organisation 
  as 
  compared 
  with 
  an 
  earlier-formed 
  portion, 
  as, 
  for 
  

   example, 
  in 
  Leiicosolenia 
  cavata 
  ; 
  but 
  this 
  is 
  simply 
  a 
  case 
  of 
  different 
  ontogenetic 
  

   stages 
  being 
  present 
  simultaneously. 
  May 
  not 
  a 
  similar 
  state 
  of 
  affairs 
  have 
  given 
  

   rise 
  to 
  Miklucho-Macleay's* 
  and 
  Ha^ckel's 
  statements 
  as 
  to 
  the 
  co-existence 
  of 
  

   different 
  types 
  of 
  canal 
  system 
  in 
  the 
  same 
  colony, 
  as 
  in 
  Hteckel's 
  genus 
  

   Ascometra 
  ?). 
  On 
  the 
  other 
  hand 
  I 
  have 
  excellent 
  series 
  of 
  specimens 
  of 
  obviously 
  

   the 
  same 
  species 
  in 
  which 
  the 
  canal 
  system 
  is 
  remarkably 
  constant 
  {e.g. 
  , 
  Leiicosolenia 
  

   cavata, 
  L. 
  tripodifera, 
  L. 
  pelliculata) 
  . 
  Moreover 
  the 
  canal 
  system 
  offers 
  a 
  wider 
  

   range 
  of 
  characters 
  to 
  choose 
  from 
  than 
  do 
  the 
  combinations 
  of 
  spicules, 
  of 
  which 
  

   seven 
  only 
  are 
  possible. 
  

  

  The 
  classification 
  which 
  I 
  therefore 
  propose 
  to 
  make 
  use 
  of 
  for 
  the 
  Calcarea 
  

   Homocoela 
  is 
  as 
  follows 
  : 
  — 
  

  

  Order 
  Honioccela 
  (Polejaeff). 
  

   Calcareous 
  sponges 
  in 
  which 
  the 
  endoderm 
  consists 
  throughout 
  of 
  collared 
  cells. 
  

   Genus, 
  Leiicosolenia 
  (Bowerbank). 
  

  

  With 
  the 
  characters 
  of 
  the 
  order. 
  

  

  Section 
  1. 
  Homuccvla 
  Siinplicia. 
  

  

  Honioccela 
  in 
  which 
  the 
  Ascon-persoiis 
  either 
  remain 
  solitary 
  and 
  do 
  not 
  form 
  

   colonies 
  or 
  they 
  form 
  simple 
  colonies 
  in 
  which 
  the 
  comi)Oiient 
  Ascon-[)ersons 
  may 
  

   branch 
  but 
  never 
  form 
  complex 
  anastomoses 
  nor 
  give 
  off 
  radial 
  tubes, 
  so 
  that 
  the 
  

   individuality 
  of 
  the 
  different 
  members 
  of 
  the 
  colony 
  is 
  easily 
  recognisable. 
  

  

  Section 
  2. 
  Hoinocccla 
  Reticulata. 
  \ 
  

  

  Homocttila 
  in 
  wliich 
  the 
  sponge-colony 
  forms 
  a 
  more 
  or 
  less 
  complex 
  network 
  of 
  

   branching 
  and 
  anastomosing 
  tubes, 
  so 
  that 
  it 
  is 
  no 
  longer 
  possible 
  to 
  distinguish 
  the 
  

   individual 
  Ascon-persons 
  of 
  which 
  the 
  colony 
  is 
  com[)osed. 
  

  

  * 
  Jenaische 
  Zeitschrift, 
  Vol. 
  IV., 
  18G8. 
  

  

  \ 
  This 
  section 
  is 
  practically 
  synonjmous 
  with 
  Gray's 
  genus 
  Clathrina, 
  used 
  by 
  Carter 
  in 
  his 
  desoriptions 
  o( 
  the 
  

   Victorian 
  sponges 
  {he. 
  cit.). 
  

  

  