﻿378 
  W. 
  B. 
  JOHNSON 
  AND 
  LLEWELLYN 
  LLOYD. 
  

  

  (a) 
  Breeding 
  Haunts. 
  

  

  Over 
  8,000 
  pupae 
  and 
  empty 
  cases 
  were 
  collected 
  and 
  the 
  breeding 
  haunts 
  appear 
  

   to 
  differ 
  in 
  some 
  points 
  from 
  those 
  of 
  palpalis. 
  The 
  fly 
  exhibits 
  a 
  preference 
  for 
  sand, 
  

   the 
  pupae 
  being 
  distributed 
  as 
  follows 
  : 
  — 
  7,540 
  in 
  coarse 
  sand, 
  525 
  in 
  fine 
  sand, 
  

   113 
  in 
  wood 
  ash, 
  80 
  in 
  cracks 
  in 
  baked 
  mud, 
  8 
  in 
  fibrous 
  loam, 
  and 
  2 
  in 
  gravel. 
  

   Selection 
  was 
  very 
  marked 
  once 
  in 
  the 
  case 
  of 
  the 
  buttresses 
  of 
  a 
  fig 
  tree 
  ; 
  one 
  angle 
  

   containing 
  sand 
  yielded 
  about 
  300 
  pupae, 
  and 
  the 
  other 
  angles, 
  floored 
  by 
  hard 
  

   mud 
  and 
  filled 
  in 
  with 
  leaves, 
  none 
  at 
  all. 
  The 
  great 
  majority, 
  7,970, 
  were 
  collected 
  

   where 
  the 
  only 
  shade 
  was 
  that 
  of 
  high 
  trees, 
  with 
  no 
  low 
  growth 
  ; 
  176 
  were 
  collected 
  

   in 
  14 
  positions 
  from 
  under 
  shrubs 
  or 
  fallen 
  trees 
  overshadowed 
  by 
  high 
  foliage 
  ; 
  

   124 
  were 
  taken 
  in 
  5 
  positions 
  where 
  there 
  was 
  only 
  the 
  shade 
  of 
  low 
  shrubs. 
  There 
  

   seems 
  to 
  be 
  a 
  well 
  marked 
  difference 
  between 
  tachinoides 
  and 
  palpalis 
  in 
  this, 
  the 
  

   latter 
  having 
  usually 
  low 
  shade 
  above, 
  or 
  close 
  to, 
  its 
  pupa 
  sites. 
  There 
  was 
  

   practically 
  always 
  some 
  vegetable 
  debris 
  on 
  the 
  surface 
  of 
  the 
  ground 
  above 
  the 
  

   pupae, 
  but 
  it 
  would 
  be 
  hard 
  to 
  find 
  a 
  spot 
  in 
  the 
  tachinoides 
  haunts 
  where 
  there 
  

   was 
  no 
  such 
  accumulation. 
  There 
  was 
  no 
  particular 
  association 
  with 
  animal 
  paths, 
  

   but 
  the 
  situation 
  of 
  the 
  breeding 
  spots 
  in 
  open 
  spaces 
  in 
  the 
  forest 
  amounts 
  almost 
  

   to 
  the 
  same 
  thing. 
  Mainly 
  they 
  were 
  very 
  close 
  to 
  water, 
  but 
  given 
  satisfactory 
  

   forest 
  they 
  may 
  be 
  very 
  distant 
  from 
  it, 
  since 
  at 
  Mashiwashi 
  great 
  numbers 
  of 
  pupae 
  

   were 
  taken 
  out 
  of 
  a 
  mile 
  of 
  stream-bed 
  in 
  which 
  there 
  were 
  only 
  two 
  small 
  water- 
  

   holes 
  close 
  together. 
  A 
  few 
  of 
  the 
  breeding 
  sites 
  deserve 
  special 
  mention. 
  

  

  1 
  . 
  High 
  flat 
  bank 
  of 
  Benue 
  River 
  : 
  type 
  (c) 
  forest 
  to 
  edge 
  of 
  bank, 
  with 
  large 
  

   open 
  spaces 
  ; 
  heavy 
  forest 
  about 
  100 
  yards 
  wide 
  backed 
  by 
  type 
  (d) 
  forest 
  ; 
  moderate 
  

   amount 
  of 
  game, 
  many 
  baboons 
  and 
  warthog, 
  together 
  with 
  the 
  river 
  fauna 
  available 
  

   as 
  food 
  supply 
  ; 
  ground 
  mainly 
  a 
  dry 
  cracked 
  mud 
  mosaic 
  ; 
  mid 
  dry 
  season. 
  Wood- 
  

   cutters 
  had 
  pulverised 
  about 
  25 
  sq. 
  ft. 
  of 
  the 
  mud 
  into 
  fine 
  sand, 
  very 
  shallow 
  with 
  

   small 
  cut 
  branches 
  lying 
  over 
  ; 
  moderate 
  shade 
  20 
  ft. 
  above 
  the 
  site 
  and 
  no 
  low 
  

   bushes 
  near 
  ; 
  yielded 
  59 
  pupae 
  and 
  281 
  empty 
  case's. 
  Twenty 
  feet 
  away 
  from 
  this 
  

   position 
  in 
  the 
  same 
  open 
  space 
  was 
  an 
  old 
  camp 
  fire 
  with 
  ash 
  and 
  charred 
  sticks 
  

   covering 
  an 
  area 
  of 
  9 
  sq. 
  ft., 
  very 
  shallow 
  fine 
  sand 
  below 
  the 
  ash 
  (mud 
  pulverised 
  

   by 
  heat) 
  ; 
  16 
  pupae 
  and 
  97 
  cases 
  were 
  taken 
  from 
  the 
  ash. 
  Much 
  of 
  the 
  mud 
  mosaic 
  

   was 
  pulled 
  up, 
  but 
  no 
  pupae 
  could 
  be 
  found 
  in 
  the 
  cracks 
  in 
  this 
  place. 
  In 
  the 
  

   same 
  locality 
  in 
  shallow 
  sandy 
  depressions 
  along 
  baboon 
  paths 
  10 
  pupae 
  and 
  8 
  cases 
  

   were 
  found. 
  

  

  2. 
  At 
  Patta, 
  a 
  town 
  near 
  the 
  Gongola 
  River, 
  in 
  Gombe 
  Division 
  : 
  town 
  watering 
  

   place 
  in 
  a 
  stream-bed 
  ; 
  type 
  (c) 
  forest 
  in 
  rocky 
  gorge, 
  the 
  shaded 
  part 
  being 
  about 
  

   200 
  yards 
  long 
  and 
  10 
  yards 
  wide 
  ; 
  dry 
  stony 
  ground 
  around 
  ; 
  two 
  small 
  deep 
  pools 
  

   forming 
  the 
  town 
  watering 
  place 
  ; 
  food 
  supply 
  practically 
  only 
  man 
  and 
  Varanus, 
  

   and 
  the 
  latter 
  could 
  not 
  be 
  numerous 
  ; 
  late 
  dry 
  season. 
  In 
  a 
  small 
  patch 
  of 
  coarse 
  

   sand 
  in 
  hollow 
  of 
  rock, 
  overshadowed 
  by 
  larger 
  rock, 
  10 
  pupae 
  and 
  1 
  case. 
  This 
  

   position 
  is 
  under 
  the 
  big 
  rock 
  in 
  the 
  background 
  in 
  Plate 
  xxii, 
  fig. 
  1. 
  A 
  coarse 
  sand 
  

   accumulation 
  in 
  angle 
  of 
  buttresses 
  of 
  a 
  fig 
  tree 
  (the 
  straight 
  white 
  trunk 
  in 
  the 
  

   illustration) 
  and 
  a 
  strip 
  of 
  coarse 
  sand 
  between 
  rock 
  and 
  bank 
  together 
  yielded 
  

   312 
  pupae 
  and 
  406 
  cases. 
  In 
  a 
  patch 
  of 
  sand, 
  2 
  sq. 
  yards, 
  among 
  rocks 
  over- 
  

   shadowed 
  bj? 
  a 
  large 
  fig 
  tree, 
  5 
  pupae 
  and 
  251 
  cases 
  were 
  taken. 
  This 
  last 
  breeding 
  

   site 
  was 
  apparently 
  being 
  abandoned 
  with 
  the 
  shedding 
  of 
  the 
  foliage 
  of 
  the 
  tree 
  

   which 
  overshadowed 
  it 
  (cf. 
  Fiske's 
  account 
  of 
  the 
  destruction 
  of 
  palpalis 
  pupae 
  by 
  

   sun 
  when 
  caterpillars 
  defoliated 
  the 
  shade 
  above 
  them). 
  

  

  3. 
  At 
  Mashiwashi 
  in 
  type 
  (b) 
  forest, 
  an 
  area 
  about 
  200 
  yards 
  wide 
  in 
  its 
  broadest 
  

   part 
  and 
  about 
  a 
  mile 
  long, 
  narrowing 
  away 
  at 
  its 
  ends 
  into 
  fadamas 
  ; 
  bordered 
  by 
  

   type 
  (d) 
  forest 
  ; 
  dense 
  thickets 
  in 
  parts, 
  but 
  with 
  considerable 
  open 
  spaces 
  and 
  

   traversed 
  by 
  a 
  sandy 
  stream-bed 
  containing 
  water 
  only 
  in 
  pools 
  for 
  most 
  of 
  the 
  

   dry 
  season. 
  From 
  24.iv. 
  to 
  15.v., 
  in 
  the 
  early 
  rains 
  before 
  the 
  stream 
  had 
  com- 
  

   menced 
  to 
  flow, 
  6,217 
  cases 
  and 
  124 
  pupae, 
  of 
  which 
  88 
  contained 
  dead 
  flies, 
  were 
  

  

  