﻿GILLMERIA 
  OCHRODACTYLA. 
  231 
  

  

  the 
  hollows 
  between 
  the 
  pectinations. 
  The 
  tubercles 
  can 
  be 
  detected 
  on 
  

   the 
  usual 
  sites, 
  they 
  carry 
  very 
  minute 
  hairs, 
  thick 
  and 
  stubby, 
  barely 
  

   0-03mm. 
  long. 
  The 
  dorsal 
  ones 
  (i 
  and 
  ii) 
  are 
  very 
  obvious 
  on 
  the 
  

   mesothorax 
  on 
  the 
  flange, 
  one 
  towards 
  each 
  end 
  ; 
  similarly, 
  on 
  the 
  1st 
  

   and 
  2nd 
  abdominal 
  segments, 
  the 
  flange 
  (with 
  the 
  ribbing) 
  occupies 
  

   about 
  two-thirds 
  of 
  the 
  segments, 
  and 
  the 
  tubercles 
  are 
  at 
  each 
  end 
  

   of 
  the 
  flange 
  ; 
  the 
  posterior 
  third 
  of 
  the 
  segment 
  is 
  unribbed, 
  and 
  

   corresponds 
  with 
  what 
  is 
  the 
  intersegmental 
  membrane 
  in 
  movable 
  

   segments. 
  At 
  the 
  spiracular 
  level, 
  the 
  ribbing 
  becomes 
  wider 
  below, 
  

   so 
  that, 
  in 
  the 
  movable 
  incisions, 
  there 
  is 
  more 
  intersegmental 
  mem- 
  

   brane 
  dorsally 
  than 
  ventrally, 
  corresponding 
  with 
  the 
  great 
  curve 
  

   backwards 
  that 
  these 
  pupae 
  make. 
  This 
  curious 
  feature 
  seems 
  to 
  occur 
  in 
  

   all 
  plume 
  pupge, 
  though 
  also 
  curiously 
  liable 
  to 
  escape 
  observation. 
  

   On 
  the 
  3rd 
  abdominal 
  segment, 
  i 
  is 
  at 
  the 
  middle 
  of 
  the 
  flange 
  ; 
  on 
  the 
  

   following 
  segments, 
  i 
  is 
  at 
  the 
  middle 
  of 
  the 
  ribbed 
  piece, 
  ii 
  at 
  the 
  

   posterior 
  margin 
  of 
  the 
  ribbed 
  piece, 
  the 
  two 
  tubercles 
  dividing 
  the 
  

   segment 
  into 
  three 
  almost 
  equal 
  pieces 
  ; 
  these 
  tubercles 
  are 
  closer 
  

   together 
  on 
  the 
  posterior 
  segments, 
  i 
  passing 
  backwards 
  towards 
  ii, 
  iii 
  

   being 
  generally 
  on 
  the 
  same 
  subsegment 
  or 
  ridge 
  as 
  i. 
  Structurally 
  the 
  

   pupa 
  may 
  be 
  said 
  to 
  be 
  very 
  much 
  like 
  those 
  of 
  the 
  other 
  Platyptilias. 
  

   The 
  nose-horn 
  is 
  long 
  and 
  thick 
  (vertically), 
  with 
  sharp 
  dorsal 
  edge 
  

   and 
  fairly 
  sharp 
  point, 
  and 
  rather 
  compressed 
  laterally 
  above 
  the 
  brown 
  

   edge, 
  and, 
  therefore, 
  with 
  somewhat 
  triangular 
  section 
  ; 
  it 
  is 
  rough 
  

   and 
  ridged, 
  but 
  not 
  extremely 
  so 
  ; 
  length 
  08rnm., 
  or 
  a 
  little 
  less 
  if 
  

   measured 
  along 
  the 
  lower 
  angle. 
  The 
  cover 
  of 
  the 
  1st 
  (prothoracic) 
  

   spiracle 
  is 
  similar 
  to 
  that 
  in 
  the 
  pupa 
  of 
  Platyptilia 
  isodactylas, 
  but 
  is 
  

   only 
  of 
  about 
  half 
  the 
  length, 
  although 
  similarly 
  covered 
  with 
  waved 
  

   lines 
  and 
  spiculas. 
  The 
  wing- 
  venation 
  is 
  as 
  in 
  the 
  others 
  ; 
  tubercle 
  iii 
  

   is 
  apt 
  to 
  be 
  obsolete 
  on 
  the 
  forward 
  abdominal 
  segments. 
  The 
  sub- 
  

   spiracular 
  flange 
  is 
  very 
  marked, 
  as 
  is 
  the 
  dorsal 
  one, 
  which 
  is 
  very 
  high 
  

   over 
  the 
  meso-and 
  metathorax, 
  so 
  that 
  over 
  the 
  metathorax 
  and 
  posterior 
  

   part 
  of 
  mesothorax 
  those 
  of 
  each 
  side 
  approach 
  each 
  other, 
  but 
  keep 
  

   well 
  separate, 
  and, 
  diverging 
  again, 
  terminate 
  behind 
  the 
  middle 
  of 
  the 
  3rd 
  

   abdominal 
  segment, 
  reappearing 
  again 
  on 
  the 
  9th 
  and 
  10th 
  abdominals, 
  

   giving 
  four 
  marked 
  angles 
  to 
  the 
  cremastral 
  spike 
  ; 
  the 
  cremastral 
  

   hooks 
  are 
  about 
  the 
  same 
  length 
  as 
  in 
  the 
  pupa 
  of 
  P. 
  isodactylus, 
  but 
  are 
  

   much 
  finer 
  and 
  vastly 
  more 
  numerous; 
  so 
  much 
  so, 
  that 
  to 
  count 
  

   them 
  seems 
  useless, 
  and 
  one 
  says 
  100 
  on 
  either 
  side 
  of 
  the 
  terminal 
  

   set, 
  as 
  a 
  round 
  number, 
  which 
  cannot 
  be 
  much 
  too 
  large, 
  and 
  may 
  

   be 
  a 
  great 
  many 
  too 
  few. 
  The 
  forward 
  set 
  are 
  of 
  the 
  same 
  size 
  and 
  

   structure, 
  and 
  also 
  very 
  numerous, 
  each 
  with 
  the 
  end 
  sharply 
  bent 
  back, 
  

   and 
  ending 
  in 
  a 
  very 
  fish-hook 
  like 
  point; 
  these 
  count 
  about 
  a 
  hundred 
  

   on 
  either 
  side, 
  and 
  make 
  that 
  figure, 
  as 
  an 
  estimate 
  of 
  the 
  terminal 
  set, 
  

   seem 
  insufficient. 
  They 
  arise 
  from 
  the 
  large 
  genital 
  double 
  projection 
  

   of 
  the 
  9th 
  abdominal 
  segment, 
  and 
  I 
  gather 
  that 
  I 
  must 
  have 
  been 
  

   wrong 
  in 
  attributing 
  them 
  to 
  the 
  8th 
  abdominal 
  segment 
  (supposing, 
  

   as 
  seems 
  probable, 
  their 
  position 
  to 
  be 
  uniform 
  throughout 
  the 
  family); 
  

   they 
  are 
  certainly 
  here 
  on 
  the 
  9th. 
  [The 
  error 
  is 
  an 
  easy 
  one 
  to 
  make. 
  

   as 
  the 
  8th 
  abdominal 
  segment 
  is 
  here 
  exceedingly 
  narrow, 
  and, 
  in 
  the 
  

   Hying 
  pupa, 
  this 
  ventral 
  portion 
  of 
  the 
  9th 
  abdominal 
  segment 
  is 
  m 
  

   alignment 
  with 
  the 
  8th 
  abdominal 
  segment, 
  regarded 
  laterally 
  ami 
  

   dorsally. 
  | 
  The 
  transverse 
  ribbing 
  of 
  the 
  abdominal 
  segments 
  is 
  pro- 
  

   nounced, 
  the 
  ribs 
  count 
  about 
  fifteen. 
  Probably 
  the 
  structure 
  is 
  the 
  

  

  