﻿OXYPTILIN.E. 
  409 
  

  

  are 
  only 
  poorly-developed 
  and 
  the 
  dorsal 
  thorns 
  altogether 
  absent 
  ; 
  the 
  tubercles 
  

   of 
  the 
  abdominal 
  segments 
  are 
  small, 
  the 
  hooked 
  bristles 
  of 
  the 
  cremaster 
  are 
  only 
  

   scattered. 
  A 
  position 
  intermediate 
  between 
  the 
  two 
  forms 
  is 
  taken 
  by 
  the 
  pupa 
  of 
  

   O. 
  parvidactylus, 
  which 
  is 
  distinguished 
  by 
  the 
  remarkable 
  form 
  of 
  the 
  cremaster. 
  

  

  This 
  subfamily 
  agrees 
  in 
  its 
  general 
  characters 
  with 
  the 
  other 
  

   subfamilies 
  on 
  the 
  Platpytiliid 
  side 
  of 
  the 
  " 
  plume" 
  stirps. 
  It 
  has, 
  

   for 
  example, 
  the 
  characteristic 
  structure 
  of 
  the 
  discoidal 
  cell 
  of 
  the 
  

   forewings, 
  the 
  single 
  spina 
  to 
  the 
  £ 
  frenulum, 
  the 
  black 
  scale-tufts 
  

   on 
  the 
  third 
  plumule 
  of 
  the 
  hindwings, 
  and 
  other 
  general 
  characters. 
  

   It 
  is, 
  however, 
  in 
  certain 
  characters, 
  somewhat 
  removed 
  from 
  the 
  

   Platyptiliinae, 
  and 
  its 
  nearest 
  relatives 
  appear 
  to 
  be 
  found 
  in 
  the 
  Auiblyp- 
  

   tiliinae 
  and 
  Marasmarchinae. 
  In 
  the 
  Oxyptilines, 
  the 
  lobes 
  of 
  the 
  

   forewings 
  differ 
  in 
  shape, 
  the 
  upper 
  one 
  being 
  pointed, 
  and 
  the 
  lower 
  

   one 
  showing, 
  at 
  most, 
  an 
  ill-developed 
  obtuse 
  anal 
  angle, 
  whilst 
  the 
  

   plumules 
  of 
  the 
  hindwings 
  are 
  somewhat 
  similarly 
  shaped. 
  Hofmann 
  

   states 
  that, 
  normally, 
  as 
  in 
  the 
  Stenoptiliids 
  and 
  Platyptiliids, 
  there 
  

   are 
  five 
  branches 
  to 
  nervure 
  II 
  of 
  the 
  forewings, 
  although, 
  occasionally, 
  

   there 
  are 
  only 
  four, 
  a 
  specialisation 
  that 
  has 
  become 
  characteristic 
  

   of 
  Marasmarcha, 
  in 
  which 
  II 
  2 
  is 
  the 
  nervure 
  always 
  absent. 
  In 
  

   Buckleria 
  it 
  is 
  1^ 
  that 
  is 
  wanting. 
  In 
  the 
  $ 
  genital 
  organs 
  there 
  is 
  

   considerable 
  variation, 
  and 
  the 
  main 
  points 
  of 
  difference 
  appear 
  to 
  

   characterise 
  our 
  two 
  larger 
  sections, 
  " 
  Capperiidi 
  " 
  and 
  " 
  Oxyptilidi" 
  

   although, 
  in 
  this 
  particular, 
  Geina 
  is 
  distinctly 
  Oxyptilid 
  rather 
  than 
  

   Capperiid. 
  . 
  

  

  The 
  Oxyptiline 
  pupse 
  are 
  very 
  remarkable 
  in 
  their 
  structure, 
  and 
  

   show 
  not 
  only 
  marked 
  variation 
  within 
  the 
  limits 
  of 
  the 
  subfamily, 
  

   but 
  also 
  suggest 
  alliance 
  with 
  other 
  subfamilies 
  on 
  both 
  sides 
  of 
  the 
  

   " 
  plume 
  " 
  stirps, 
  with 
  the 
  Amblyptiliines 
  and 
  Marasmarchines 
  on 
  the 
  

   Platyptiliid 
  side, 
  and 
  with 
  the 
  Leioptilines 
  on 
  the 
  Alucitid 
  side. 
  

   Thus 
  the 
  pupa 
  of 
  Oxyptilus 
  is 
  rather 
  inclined 
  to 
  the 
  Platyptiliine 
  side, 
  

   whilst 
  that 
  of 
  Capperia 
  is 
  inclined 
  to 
  the 
  Leioptiline. 
  Comparing 
  the 
  

   pupal 
  structure 
  of 
  Capperia 
  with 
  those 
  with 
  which 
  it 
  has 
  real 
  and 
  

   apparent 
  alliance, 
  Chapman 
  notes 
  that 
  "the 
  pupa 
  of 
  Capperia 
  (hetero- 
  

   dactyla) 
  has, 
  on 
  the 
  abdominal 
  segments, 
  a 
  double 
  spine 
  on 
  either 
  

   side, 
  internal 
  to 
  i 
  and 
  ii 
  ; 
  single 
  spines 
  occur 
  in 
  the 
  same 
  positions 
  

   in 
  the 
  pupa3of 
  Alucita 
  (pentad 
  actyla) 
  and 
  Oidaematophorus 
  (lithodactyla). 
  

   In 
  Crombrugghia 
  (distans), 
  there 
  is 
  a 
  spine 
  in 
  this 
  position, 
  not, 
  however, 
  

   on 
  each 
  side, 
  but 
  a 
  central 
  unpaired 
  one. 
  Again, 
  in 
  its 
  general 
  

   appearance 
  the 
  pupa 
  of 
  Capperia 
  (heterodactyla) 
  bears 
  a 
  great 
  super- 
  

   ficial 
  resemblance 
  to 
  that 
  of 
  Ovendenia 
  (septodactyla), 
  yet 
  the 
  structure 
  

   of 
  the 
  former 
  agrees 
  essentially 
  with 
  that 
  of 
  the 
  pupa? 
  of 
  the 
  other 
  

   Oxyptilid 
  genera. 
  Without 
  pressing 
  the 
  definition 
  too 
  far, 
  the 
  pupa 
  

   of 
  Ovendenia 
  (septodactyla) 
  may 
  bo 
  described 
  as 
  an 
  Alucitine 
  pupa, 
  

   tending 
  to 
  have, 
  like 
  the 
  Alucitine 
  larva, 
  many 
  secondary 
  hairs 
  that 
  

   are 
  practically 
  indistinguishable 
  from 
  the 
  primaries. 
  the 
  pupa 
  of 
  

   Capperia 
  (heterodactyla) 
  is 
  of 
  the 
  Platyptiliine 
  type, 
  with 
  no 
  secondary 
  

   hairs, 
  whilst 
  the 
  secondary 
  hairs 
  of 
  the 
  larva 
  are 
  always 
  obviously 
  

   secondary. 
  What 
  definite 
  distinction 
  can 
  bo 
  drawn 
  between 
  the 
  

   hairs 
  along 
  the 
  wing-nervures 
  in 
  the 
  pupa 
  of 
  Ovendenia 
  (septodactyla) 
  

   and 
  those 
  of 
  Capperia 
  (heterodactyla) 
  is 
  uncertain; 
  perhaps 
  there 
  is 
  

   none. 
  The 
  lepidopterous 
  pupa, 
  as 
  a 
  rule, 
  refuses 
  to 
  have 
  hairs 
  on 
  any 
  

   of 
  the 
  appendages 
  (antenna, 
  Legs, 
  wings, 
  etc.). 
  In 
  the 
  pupa 
  of 
  

   Capperia, 
  the 
  hairs 
  on 
  the 
  wings, 
  when 
  carefully 
  examined, 
  appear 
  

  

  