﻿308 
  BRITISH 
  LEPIDOPTERA. 
  

  

  Lonicera 
  involucrata 
  (Dyar), 
  Geranium, 
  pratense, 
  Er 
  odium 
  cicutarium 
  

   (Zeller), 
  Arthocarpus 
  sp. 
  (Walsingham). 
  

  

  Pupation. 
  — 
  The 
  larva 
  attaches 
  itself 
  to 
  the 
  spike 
  of 
  Stachys 
  

   sylvatica, 
  on 
  the 
  flowers 
  of 
  which 
  it 
  has 
  fed, 
  in 
  order 
  to 
  undergo 
  pupa- 
  

   tion 
  (Williams). 
  The 
  pupa 
  is 
  attached 
  to 
  the 
  foodplant 
  by 
  the 
  tail, 
  

   and 
  two 
  somewhat 
  curved 
  pointed 
  protuberances, 
  which 
  spring 
  from 
  

   the 
  back, 
  give 
  it 
  a 
  curious 
  appearance 
  (Porritt). 
  Attached 
  by 
  the 
  anal 
  

   hooks 
  to 
  the 
  foodplant, 
  hanging 
  like 
  a 
  little 
  slender 
  butterfly 
  pupa 
  

   (Barrett). 
  All 
  the 
  pupae 
  that 
  I 
  have 
  collected 
  have 
  been 
  found 
  on 
  the 
  

   stems 
  of 
  Stachys 
  sylvatica. 
  Their 
  ordinary 
  position 
  is 
  identical 
  with 
  that 
  of 
  

   A. 
  cosmodactyla, 
  Hb., 
  viz., 
  suspended 
  head 
  downwards 
  along 
  the 
  portion 
  

   of 
  the 
  stem 
  between 
  two 
  of 
  the 
  whorls 
  of 
  the 
  calyces 
  of 
  the 
  foodplant, 
  

   and 
  the 
  method 
  by 
  which 
  the 
  pupa 
  is 
  attached 
  to 
  the 
  stem 
  is 
  precisely 
  

   similar 
  (Bankes). 
  In 
  almost 
  all 
  cases 
  that 
  I 
  have 
  noted, 
  the 
  pupa 
  has 
  

   been 
  suspended 
  lengthwise 
  along 
  the 
  stem 
  (Riding). 
  The 
  fullfed 
  larva 
  

   usually 
  leaves 
  the 
  flowers 
  of 
  Stachys 
  sylvatica 
  on 
  which 
  it 
  has 
  been 
  

   feeding, 
  and 
  crawls 
  somewhat 
  lower 
  down 
  the 
  stem, 
  before 
  spinning 
  

   its 
  pad 
  of 
  silk, 
  to 
  which 
  it 
  attaches 
  itself 
  for 
  pupation 
  (Wocke). 
  The 
  

   pupa 
  hangs 
  free 
  on 
  the 
  flower-stalk, 
  or 
  on 
  a 
  seed-capsule 
  of 
  Aquilegia, 
  

   the 
  pupal 
  stage 
  lasting 
  from 
  14 
  to 
  21 
  days 
  (Kaltenbach). 
  

  

  Variation 
  of 
  pupa. 
  — 
  As 
  in 
  the 
  larvae, 
  there 
  are 
  two 
  distinct 
  

   varieties 
  — 
  (1) 
  purple, 
  (2) 
  green 
  ; 
  but 
  both 
  forms 
  have 
  oblique 
  dark 
  

   markings 
  (Porritt). 
  

  

  Pupa. 
  — 
  The 
  pupa 
  of 
  this 
  species 
  is 
  substantially 
  the 
  same 
  as 
  that 
  

   of 
  A. 
  cosmodactyla. 
  The 
  latter 
  varies 
  considerably 
  in 
  just 
  those 
  points 
  

   in 
  which 
  the 
  pupa 
  of 
  A. 
  punctidactyla 
  differs 
  from 
  it, 
  and 
  very 
  probably 
  

   the 
  last-named 
  does 
  the 
  same, 
  but 
  my 
  specimens 
  are 
  too 
  few 
  to 
  say 
  

   whether 
  this 
  be 
  so 
  or 
  not. 
  Four 
  or 
  five 
  points 
  may 
  be 
  noted: 
  (1) 
  The 
  

   great 
  halbert-shaped 
  process 
  of 
  the 
  3rd 
  abdominal 
  is, 
  in 
  cosmodactyla, 
  

   rough, 
  nodulated, 
  and 
  angulated 
  in 
  outline 
  ; 
  in 
  punctidactyla, 
  the 
  

   posterior 
  border 
  is 
  a 
  smooth, 
  regular, 
  curve, 
  and 
  the 
  front 
  margin 
  is 
  less 
  

   deeply 
  notched. 
  A 
  good 
  many 
  specimens, 
  however, 
  of 
  cosmodactyla 
  

   are 
  very 
  close 
  to 
  punctidactyla 
  in 
  this 
  respect. 
  (2) 
  The 
  small 
  process 
  

   of 
  this 
  spine, 
  which 
  carries 
  ii, 
  is 
  rather 
  smaller 
  in 
  punctidactyla, 
  and 
  

   is 
  less 
  a 
  mere 
  process 
  on 
  the 
  large 
  spine, 
  and 
  nearly, 
  but 
  not 
  quite, 
  

   separate. 
  (3) 
  The 
  remaining 
  processes 
  on 
  the 
  4th-8th 
  abdominal 
  

   segments, 
  carrying 
  tubercles 
  i 
  and 
  ii, 
  are 
  larger 
  in 
  cosmodactyla 
  than 
  in 
  

   punctidactyla, 
  and, 
  although 
  it 
  is 
  the 
  fact, 
  in 
  both 
  species, 
  that 
  the 
  two 
  

   processes 
  (one 
  carrying 
  i, 
  and 
  the 
  other 
  carrying 
  ii) 
  are 
  fused 
  together 
  

   as 
  one 
  process, 
  they 
  sometimes 
  have 
  the 
  appearance, 
  in 
  punctidactyla, 
  

   of 
  being 
  really 
  two 
  separate 
  processes. 
  (4) 
  In 
  punctidactyla, 
  the 
  

   posterior 
  portion 
  of 
  this 
  process 
  (carrying 
  ii) 
  is 
  very 
  distinctly 
  smaller 
  

   than 
  in 
  cosmodactyla. 
  (5) 
  The 
  hairs 
  of 
  ii, 
  iii, 
  iv, 
  v, 
  and 
  vi, 
  are 
  very 
  much 
  

   shorter 
  in 
  punctidactyla 
  than 
  in 
  cosmodactyla, 
  in 
  the 
  former, 
  being 
  of 
  

   a 
  length 
  about 
  equal 
  to 
  the 
  diameter 
  of 
  the 
  spiracle, 
  and, 
  in 
  cosmodactyla, 
  

   several 
  times 
  that 
  length, 
  nearly, 
  in 
  fact, 
  half 
  the 
  width 
  of 
  a 
  segment. 
  

   I 
  have 
  examined 
  something 
  like 
  200 
  pupae 
  of 
  cosmodactyla, 
  a 
  

   few 
  of 
  my 
  own 
  breeding, 
  but 
  the 
  mass 
  sent 
  me 
  by 
  Mr. 
  E. 
  R. 
  Bankes. 
  

   Amongst 
  these, 
  there 
  is 
  much 
  variation 
  in 
  all 
  these 
  points 
  of 
  difference, 
  

   but, 
  though, 
  in 
  many 
  instances, 
  one 
  or 
  other 
  of 
  these 
  characters 
  

   approaches 
  very 
  closely 
  to 
  the 
  form 
  belonging 
  to 
  punctidactyla, 
  so 
  

   closely 
  that 
  one 
  would 
  be 
  inclined 
  to 
  say 
  that, 
  in 
  that 
  character, 
  the 
  pupa 
  

   was 
  more 
  like 
  punctidactyla 
  than 
  cosmodactyla, 
  still, 
  it 
  does 
  not 
  quite 
  

  

  