﻿90 
  BUPEET 
  W. 
  JACK. 
  

  

  directions, 
  a 
  portion 
  of 
  the 
  flies 
  going 
  back 
  to 
  the 
  previous 
  dry 
  season 
  haunt 
  and 
  a 
  

   portion 
  to 
  the 
  new, 
  probably 
  whichever 
  is 
  the 
  nearer. 
  The 
  term 
  " 
  wet 
  season 
  " 
  is 
  

   not 
  used 
  loosely 
  in 
  this 
  connection, 
  notwithstanding 
  the 
  fact 
  that 
  the 
  forest 
  retains 
  it 
  3 
  

   foliage 
  for 
  some 
  months 
  after 
  the 
  cessation 
  of 
  the 
  rains. 
  The 
  wet 
  season 
  lasts 
  until 
  

   April, 
  and 
  the 
  three 
  months 
  following 
  are, 
  with 
  the 
  occasional 
  exception 
  of 
  August, 
  

   the 
  coldest 
  in 
  the 
  year, 
  when 
  the 
  breeding 
  rate 
  appears 
  to 
  be 
  at 
  its 
  lowest 
  ebb 
  and 
  the 
  

   tendency 
  to 
  extension 
  is 
  apparently 
  checked 
  by 
  this 
  fact. 
  It 
  is 
  to 
  be 
  remarked 
  that 
  

   whereas 
  advance 
  along 
  a 
  well 
  shaded 
  river 
  may 
  be 
  a 
  more 
  or 
  less 
  continuous 
  or 
  re- 
  

   gular 
  process, 
  advance 
  across 
  interfluvial 
  areas 
  tends 
  to 
  be 
  irregularly 
  spasmodic. 
  

   One 
  dry 
  season 
  haunt 
  having 
  been 
  attained, 
  a 
  pause 
  of 
  several 
  years 
  sometimes 
  inter- 
  

   venes 
  before 
  the 
  next 
  step, 
  and 
  observations 
  indicate 
  that 
  the 
  next 
  step, 
  if 
  it 
  be 
  a 
  

   fairly 
  long 
  one, 
  say 
  five 
  or 
  six 
  miles, 
  is 
  only 
  possible 
  when 
  the 
  fly 
  has 
  become 
  really 
  

   abundant 
  at 
  its 
  previous 
  halting 
  place. 
  As 
  an 
  instance, 
  the 
  passage 
  from 
  the 
  Mzola 
  

   to 
  the 
  Kana 
  River 
  in 
  the 
  Sebungwe 
  district, 
  a 
  distance 
  of 
  about 
  six 
  miles, 
  took 
  

   place 
  in 
  the 
  wet 
  season 
  of 
  1915-1916, 
  but 
  the 
  fly 
  was 
  present 
  on 
  the 
  Mzola 
  in 
  small 
  

   numbers 
  in 
  1912-13. 
  Once 
  the 
  pest 
  obtained 
  a 
  footing 
  on 
  the 
  Kana, 
  it 
  increased 
  

   rapidly 
  in 
  the 
  presence 
  of 
  large 
  quantities 
  of 
  game 
  and 
  late 
  in 
  1916 
  had 
  become 
  fairly 
  

   abundant, 
  showing 
  that 
  it 
  was 
  no 
  lack 
  of 
  favourable 
  conditions 
  ahead 
  that 
  caused 
  

   the 
  delay. 
  The 
  necessity 
  for 
  great 
  numbers 
  would 
  appear 
  to 
  be 
  explicable 
  as 
  fol- 
  

   lows 
  :• 
  — 
  At 
  the 
  beginning 
  of 
  the 
  rains 
  the 
  flies 
  scatter 
  at 
  once 
  for 
  a 
  short 
  distance 
  in 
  

   all 
  directions 
  from 
  their 
  narrow 
  dry 
  season 
  haunts. 
  A 
  certain 
  proportion 
  of 
  the 
  

   females 
  deposit 
  larvae 
  near 
  the 
  limit 
  of 
  their 
  spreading 
  range 
  towards 
  the 
  new 
  objec- 
  

   tive. 
  Flies 
  emerging 
  from 
  these 
  also 
  scatter 
  in 
  all 
  directions, 
  a 
  proportion 
  of 
  the 
  

   females 
  again 
  larvipositing 
  in 
  the 
  required 
  direction. 
  The 
  repetition 
  of 
  this 
  process 
  

   may 
  result 
  in 
  the 
  neighbourhood 
  of 
  the 
  next 
  river 
  or 
  vlei 
  being 
  reached 
  before 
  the 
  

   next 
  season 
  of 
  concentration 
  ; 
  but 
  it 
  is 
  clear 
  that, 
  the 
  rate 
  of 
  reproduction 
  being 
  so 
  

   low, 
  only 
  a 
  small 
  proportion 
  of 
  the 
  numbers 
  that 
  originally 
  scattered 
  would 
  extend 
  

   so 
  far 
  in 
  a 
  few 
  months, 
  hence 
  the 
  necessity 
  of 
  large 
  numbers 
  to 
  start 
  with. 
  

  

  These 
  remarks 
  may 
  seem 
  somewhat 
  academical, 
  but 
  they 
  have 
  a 
  distinctly 
  practical 
  

   side. 
  If 
  the 
  writer's 
  deductions 
  are 
  correct, 
  it 
  would 
  appear 
  that 
  the 
  clearing 
  of 
  a 
  

   comparatively 
  narrow 
  strip 
  of 
  forest 
  might 
  serve 
  to 
  check 
  the 
  advance 
  of 
  the 
  fly, 
  

   notwithstanding 
  the 
  fact 
  that 
  the 
  males 
  would 
  constantly 
  be 
  carried 
  across 
  to 
  the 
  

   other 
  side. 
  The 
  cost 
  and 
  maintenance 
  of 
  such 
  a 
  clearing 
  would 
  be 
  prohibitive 
  on 
  

   a 
  great 
  scale, 
  but 
  by 
  taking 
  advantage 
  of 
  local 
  conditions 
  it 
  might 
  prove 
  practicable 
  

   to 
  check 
  the 
  advance 
  of 
  a 
  belt 
  in 
  some 
  particular 
  direction. 
  An 
  experiment 
  of 
  this 
  

   nature 
  was 
  actually 
  commenced 
  at 
  the 
  south-western 
  extremity 
  of 
  the 
  Sebungwe 
  belt 
  

   in 
  October 
  1918, 
  but 
  has 
  had 
  temporarily 
  to 
  be 
  abandoned 
  owing 
  to 
  the 
  influenza 
  

   epidemic. 
  

  

  