﻿iv 
  

  

  fly 
  and 
  gradually 
  dragged 
  it 
  up 
  the 
  tree 
  and 
  I 
  then 
  took 
  

   both. 
  

  

  " 
  Further 
  on 
  I 
  came 
  on 
  a 
  nest 
  of 
  the 
  small 
  Ponerines 
  3946 
  

   [1 
  ^ 
  L. 
  pavesii, 
  4 
  ^ 
  Lepiogenys 
  {sensu 
  stricto) 
  stuhlmanni, 
  Mayr.], 
  

   similar 
  to 
  3936, 
  which, 
  at 
  about 
  10 
  a.m., 
  were 
  bringing 
  up 
  

   from 
  their 
  nest 
  in 
  a 
  rotten 
  stump 
  their 
  pupae 
  [three 
  examples] 
  

   and 
  callows 
  for 
  a 
  sunning. 
  Near 
  by 
  were 
  some 
  Bengalias 
  

   [three 
  (^ 
  of 
  same 
  sp.], 
  one 
  of 
  which 
  from 
  time 
  to 
  time 
  seized 
  

   a 
  callow 
  [one 
  of 
  these 
  is 
  evidently 
  indicated 
  by 
  394c, 
  the 
  

   label 
  on 
  an 
  immature 
  ^ 
  probably 
  of 
  L. 
  sivhhnamd], 
  as 
  it 
  was 
  

   carried 
  up 
  by 
  one 
  of 
  the 
  ants 
  and 
  frequently 
  got 
  away 
  with 
  

   it 
  to 
  a 
  distance, 
  when, 
  the 
  ant 
  having 
  relaxed 
  its 
  hold, 
  the 
  

   fly 
  was 
  able 
  to 
  suck 
  out 
  the 
  juices 
  at 
  its 
  leisure. 
  

  

  " 
  On 
  14th 
  January 
  a 
  further 
  series 
  of 
  Bengalia 
  [2 
  (^ 
  of 
  

   same 
  sp.] 
  were 
  taken, 
  each 
  detected 
  in 
  the 
  act 
  of 
  robbing 
  

   more 
  of 
  the 
  ants, 
  such 
  as 
  395rt 
  [several 
  ^ 
  of 
  Pheidole 
  liengmei], 
  

   but 
  belonging 
  to 
  a 
  different 
  nest, 
  returning 
  food-laden. 
  On 
  

   the 
  ground 
  and 
  near 
  to 
  the 
  nest 
  was 
  a 
  pupa 
  of 
  Catopsilia 
  

   florella, 
  too 
  heavy 
  for 
  the 
  ants 
  [several 
  ^ 
  of 
  same 
  sp.] 
  to 
  drag 
  

   away, 
  but 
  in 
  which 
  they 
  were 
  gnawing 
  a 
  small 
  hole 
  behind 
  

   the 
  left 
  wing 
  case. 
  A 
  Bengalia 
  3956 
  [a 
  ^ 
  and 
  $ 
  of 
  same 
  sp. 
  

   bear 
  this 
  number], 
  walked 
  round 
  the 
  pupa 
  several 
  times 
  

   and 
  eventually 
  approaching 
  warily 
  suddenly 
  thrust 
  its 
  pro- 
  

   boscis 
  into 
  the 
  breach, 
  momentarily 
  however, 
  for 
  it 
  was 
  

   extremely 
  suspicious 
  of 
  the 
  ants 
  ; 
  but 
  it 
  repeated 
  the 
  manoeuvre 
  

   many 
  times. 
  

  

  " 
  The 
  possibility 
  of 
  the 
  fly 
  feeding 
  on 
  dead 
  animal 
  matter 
  

   not 
  found 
  by 
  ants, 
  or 
  attacking 
  living 
  insects, 
  had 
  been 
  

   constantly 
  borne 
  in 
  mind, 
  but 
  though 
  dead 
  insects, 
  such 
  as 
  

   it 
  had 
  been 
  seen 
  to 
  hanker 
  after 
  when 
  taken 
  by 
  ants- 
  — 
  the 
  

   larvae 
  and 
  pupae 
  of 
  C. 
  fioreUa 
  for 
  instance 
  were 
  fairly 
  numer- 
  

   ous, 
  there 
  being 
  an 
  absolute 
  plague 
  of 
  these 
  butterflies 
  at 
  

   Lindi 
  — 
  yet 
  Bengalia 
  was 
  never 
  seen 
  to 
  approach 
  any 
  un- 
  

   touched 
  by 
  ants, 
  perhaps 
  because 
  it 
  is 
  unable 
  to 
  efl'ect 
  by 
  

   itself 
  a 
  breach 
  of 
  surface 
  in 
  cuticle 
  hardened 
  by 
  exjJosure 
  

   postmortem. 
  However, 
  on 
  several 
  occasions 
  the 
  fly 
  was 
  seen 
  

   to 
  attack 
  living 
  larvae 
  of 
  CafopsiUa, 
  in 
  each 
  case 
  sickly 
  and 
  

   rather 
  shrivelled 
  specimens 
  ; 
  for, 
  owing 
  to 
  the 
  voracity 
  and 
  

   abundance 
  of 
  earlier 
  broods, 
  the 
  leaves 
  of 
  the 
  food 
  plant, 
  a 
  

  

  