﻿330 
  C. 
  F. 
  M. 
  SWYNNERTON. 
  

  

  VIII. 
  — 
  The 
  Seasonal 
  Distribution 
  and 
  Preferences 
  in 
  

   Vegetation 
  of 
  the 
  Fly. 
  

  

  When 
  I 
  left 
  the 
  belt 
  in 
  mid- 
  June, 
  the 
  fly 
  was 
  still 
  dispersed 
  through 
  the 
  bush. 
  

   " 
  Tinji 
  " 
  (Odina 
  sp.) 
  was 
  the 
  only 
  tree 
  that 
  had 
  everywhere 
  lost 
  its 
  leaves, 
  and 
  in 
  

   patches 
  of 
  this 
  I 
  failed 
  to 
  find 
  fly, 
  though 
  searching 
  carefully. 
  This 
  observation 
  

   and 
  the 
  consideration 
  that 
  the 
  acacias 
  generally 
  are 
  not 
  merely 
  themselves 
  deciduous, 
  

   but, 
  being 
  for 
  the 
  most 
  part 
  low, 
  must 
  inevitably 
  be 
  defoliated 
  by 
  the 
  fires, 
  suggest 
  

   that 
  the 
  distribution 
  of 
  the 
  fly 
  in 
  August 
  must 
  be 
  far 
  patchier 
  than 
  it 
  was 
  as 
  we 
  found 
  it. 
  

  

  The 
  choice 
  of 
  a 
  retreat 
  would 
  appear 
  to 
  lie 
  between 
  the 
  granite 
  kopjes, 
  the 
  thickets, 
  

   and 
  the 
  borders 
  of 
  damp 
  mbugas, 
  streams 
  and 
  water-holes. 
  It 
  may 
  be 
  of 
  interest 
  

   to 
  state 
  my 
  observations 
  in 
  this 
  connection. 
  The 
  period 
  covered 
  is 
  10th 
  May 
  to 
  

   16th 
  June. 
  

  

  Granite 
  Kopjes 
  (PI. 
  xv, 
  fig. 
  2). 
  — 
  It 
  was 
  difficult 
  to 
  find 
  tsetse 
  in 
  the 
  bush 
  of 
  the 
  

   granite 
  kopjes 
  even 
  when 
  they 
  were 
  abundant 
  in 
  the 
  wooding 
  immediately 
  round, 
  

   yet 
  puparia, 
  a 
  small 
  proportion 
  of 
  them 
  full, 
  could 
  be 
  found 
  under 
  most 
  rocks, 
  even 
  

   to 
  the 
  summits 
  of 
  some 
  of 
  the 
  kopjes 
  searched, 
  and 
  natives 
  frequently 
  referred 
  to 
  the 
  

   fondness 
  of 
  the 
  flies 
  for 
  the 
  rocks, 
  saying 
  : 
  " 
  There 
  are 
  few 
  here, 
  but 
  many 
  at 
  the 
  

   kopjes 
  yonder." 
  

  

  The 
  natives 
  also 
  stated 
  that 
  the 
  kopjes 
  are 
  less 
  scorched 
  by 
  the 
  grass-fires 
  than 
  

   the 
  country 
  round 
  them 
  ; 
  and 
  it 
  was 
  obvious 
  from 
  the 
  relative 
  lack 
  of 
  grass 
  on 
  most 
  

   of 
  them, 
  that 
  resulted 
  from 
  the 
  fact 
  that 
  scrub 
  filled 
  the 
  interstices 
  between 
  the 
  

   rocks, 
  that 
  this 
  must 
  be 
  the 
  case, 
  and 
  probable 
  that 
  (as 
  the 
  natives 
  stated 
  further) 
  

   the 
  kopjes 
  would 
  tend 
  to 
  retain 
  leaf 
  when 
  their 
  surroundings 
  were 
  already 
  leafless. 
  

  

  Thickets 
  (PL 
  xiii, 
  fig. 
  2). 
  — 
  In 
  general 
  the 
  fly 
  was 
  found 
  in 
  its 
  largest 
  numbers 
  

   in 
  mgongwa 
  acacias 
  interspersed 
  with 
  considerable 
  but 
  broken 
  thicket 
  undergrowth, 
  

   and 
  in 
  mhali 
  wooding, 
  chiefly 
  where 
  there 
  were 
  thickets 
  of 
  a 
  type 
  that 
  would 
  have 
  

   attracted 
  pallidipes. 
  The 
  ihusi, 
  mkwata 
  and 
  mhali 
  (the 
  first 
  two 
  unidentified 
  

   botanically) 
  were 
  stated 
  by 
  the 
  natives 
  not 
  to 
  lose 
  leaf 
  readily, 
  the 
  mgongwa 
  and 
  

   ilula 
  coming 
  next. 
  The 
  bridge 
  of 
  bush 
  between 
  the 
  Sansui 
  mbuga 
  and 
  the 
  cleared 
  

   country 
  of 
  Luguru, 
  composed 
  largely 
  of 
  ilula 
  and 
  other 
  Mimoseae 
  with 
  overhead 
  

   mhali 
  (A. 
  spirocarpa), 
  was 
  all 
  very 
  leafless 
  on 
  1st 
  June 
  with 
  the 
  exception 
  of 
  the 
  

   ilula, 
  and 
  in 
  the 
  latter, 
  light 
  though 
  it 
  was, 
  occasional 
  tsetses, 
  mostly 
  males, 
  were 
  

   taken. 
  

  

  Not 
  all 
  thickets 
  or 
  all 
  mhali 
  were 
  equally 
  suitable. 
  To 
  judge 
  both 
  from 
  native 
  

   accounts 
  and 
  from 
  my 
  own 
  results, 
  the 
  fly 
  was 
  nearly 
  absent 
  from 
  the 
  strip 
  of 
  unusually 
  

   fine 
  spreading 
  mhali 
  acacias 
  (with 
  and 
  without 
  heavy 
  thickets 
  that 
  sheltered 
  some 
  

   brevipalpis) 
  that 
  follows 
  the 
  lake 
  submarginally 
  between 
  the 
  Mbarangeti 
  and 
  the 
  

   Rowana, 
  though 
  it 
  occurs 
  in 
  drier, 
  thinner 
  bush 
  on 
  the 
  mbuga 
  side 
  of 
  the 
  strip. 
  

  

  On 
  the 
  other 
  hand, 
  the 
  worst 
  place 
  in 
  Itilima 
  was 
  stated 
  to 
  be 
  near 
  Zagayu 
  on 
  

   the 
  right 
  of 
  the 
  road 
  to 
  Luguru 
  — 
  a 
  jumble 
  of 
  granite 
  kopjes, 
  small 
  vleis 
  and 
  

   thickish 
  bush. 
  

  

  The 
  following- 
  extract 
  is 
  from 
  my 
  road 
  notes 
  of 
  22nd 
  May 
  : 
  — 
  

   " 
  4 
  p.m. 
  Entering 
  broad-leaved 
  bush 
  dominated 
  by 
  mgongwa 
  and 
  numer- 
  

   ous 
  thickets 
  — 
  Commiphora, 
  Combretum, 
  Albizzia, 
  Markhamia, 
  some 
  Odina. 
  

  

  Numerous 
  tsetses. 
  

   " 
  Further, 
  on 
  down 
  slope, 
  drier 
  bush— 
  scrub 
  patches 
  with 
  kinumburi, 
  some 
  ilula, 
  

  

  a 
  great 
  tendency 
  to 
  leaflessness, 
  especially 
  in 
  Commiphora. 
  No 
  bigger 
  trees 
  

  

  or 
  mgongwa. 
  No 
  tsetses. 
  

   " 
  Then 
  much 
  Odina, 
  practically 
  leafless. 
  Two 
  flies 
  only, 
  taken 
  at 
  thickets 
  in 
  

  

  between 
  with 
  dense 
  Euphorbia. 
  

   " 
  4.30. 
  Narrow 
  green 
  vlei 
  and 
  pool 
  with 
  lilies, 
  surrounded 
  with 
  tall 
  green 
  grass 
  

  

  and 
  overhung 
  by 
  dense 
  mwotobarasi 
  bushes. 
  Beyond 
  it 
  simply 
  Odina 
  and 
  

  

  light 
  ilula. 
  Took 
  great 
  numbers 
  at 
  the 
  pool." 
  

  

  