﻿390 
  W. 
  F. 
  FISKE. 
  

  

  colony 
  centre, 
  or 
  centres 
  of 
  infestation, 
  by 
  more 
  attractive 
  conditions 
  elsewhere. 
  

   In 
  the 
  colony 
  at 
  Lutoboka 
  Bay 
  (fig. 
  4) 
  or 
  at 
  Mujuzi 
  Creek 
  (Sect. 
  Ill 
  (e) 
  ) 
  dispersion 
  

   is, 
  on 
  the 
  other 
  hand, 
  forced, 
  and 
  the 
  flies, 
  instead 
  of 
  being 
  drawn 
  outward 
  by 
  a 
  superior 
  

   attraction, 
  are 
  forced 
  outwards 
  into 
  a 
  relatively 
  repulsive 
  or 
  intolerable 
  environment 
  

   by 
  the 
  stimulus 
  of 
  hunger. 
  In 
  this 
  latter 
  case 
  female 
  percentage 
  tends 
  to 
  soar 
  

   as 
  male 
  density 
  falls 
  ; 
  in 
  the 
  former 
  male 
  density 
  and 
  female 
  percentage 
  fall 
  

   together, 
  as 
  shown 
  in 
  Table 
  XX. 
  And 
  it 
  is 
  certainly 
  superior 
  food 
  in 
  combination 
  

   with 
  equally 
  attractive 
  shelter 
  which 
  explains 
  inland 
  dispersion 
  in 
  all 
  the 
  cases 
  

   that 
  have 
  come 
  under 
  observation, 
  except 
  inland 
  from 
  the 
  lake 
  along 
  the 
  banks 
  of 
  the 
  

   Mujuzi 
  and 
  a 
  few 
  other 
  creeks, 
  which 
  is 
  comparable 
  to 
  dispersion 
  along 
  the 
  shore 
  of 
  

   the 
  lake. 
  

  

  VI. 
  Hosts 
  and 
  Host 
  Preferences 
  of 
  Glossina 
  Palpalis. 
  

  

  VI 
  (a). 
  Methods 
  of 
  Studying 
  Hosts 
  and 
  Host 
  Preferences. 
  

  

  These 
  methods 
  were 
  used 
  for 
  the 
  study 
  of 
  hosts 
  and 
  host 
  preferences 
  of 
  Glossina 
  

   palpalis 
  : 
  — 
  

  

  (1) 
  Experiment. 
  The 
  animals 
  were 
  tethered 
  along 
  fly-infested 
  shore 
  (see 
  

  

  Sect. 
  VI 
  (c) 
  ). 
  

  

  (2) 
  Approach 
  and 
  observation. 
  Wild 
  and 
  domestic 
  animals 
  were 
  approached 
  

  

  and 
  the 
  behaviour 
  of 
  fly 
  toward 
  them 
  was 
  observed 
  under 
  entirely 
  natural 
  

   conditions 
  (see 
  Sect. 
  VI 
  (d) 
  ). 
  

  

  (3) 
  Shooting. 
  When 
  possible, 
  wild 
  animals 
  were 
  shot 
  in 
  the 
  fly 
  belt 
  at 
  hours 
  

  

  when 
  flies 
  were 
  active. 
  If 
  the 
  animal 
  falls 
  without 
  running 
  the 
  flies 
  which 
  

   have 
  been 
  following 
  may 
  linger 
  by 
  it 
  for 
  a 
  time, 
  and 
  some 
  of 
  them 
  will 
  

   usually 
  feed 
  (see 
  Sect. 
  VI 
  (e) 
  ). 
  

  

  In 
  addition 
  to 
  the 
  above 
  the 
  study 
  of 
  sex 
  ratio, 
  as 
  affected 
  by 
  the 
  density 
  of 
  favoured 
  

   hosts, 
  affords 
  an 
  opportunity 
  for 
  securing 
  supplementary 
  or 
  confirmatory 
  data 
  

   (see 
  Sect. 
  V 
  (g) 
  ). 
  

  

  A 
  list 
  of 
  the 
  principal 
  animals 
  that 
  occur 
  within 
  the 
  fly 
  belt, 
  together 
  with 
  methods 
  

   by 
  which 
  they 
  were 
  studied 
  as 
  hosts 
  of 
  fly, 
  and 
  a 
  summary 
  of 
  conclusions 
  reached 
  

   is 
  presented 
  in 
  Table 
  XXI. 
  The 
  list 
  excludes 
  many 
  small 
  mammals, 
  such 
  as 
  rats, 
  

   bats, 
  etc., 
  which 
  though 
  common 
  are 
  certainly 
  of 
  no 
  importance 
  as 
  hosts 
  of 
  fly, 
  and 
  

   it 
  also 
  excludes 
  others 
  such 
  as 
  Hyrax, 
  the 
  " 
  edible 
  rat," 
  and 
  others 
  which 
  are 
  rare 
  or 
  

   nocturnal 
  and 
  of 
  no 
  importance, 
  in 
  this 
  region 
  at 
  least. 
  

  

  All 
  animals 
  which 
  are 
  fed 
  upon 
  by 
  Glossina, 
  and 
  thus 
  technically 
  included 
  amongst 
  

   its 
  hosts, 
  are 
  not 
  equally 
  favoured 
  or 
  preferred 
  by 
  it, 
  and 
  a 
  broad 
  line 
  of 
  distinction 
  

   may 
  be 
  drawn 
  between 
  the 
  two 
  following 
  categories 
  : 
  — 
  

  

  (A). 
  Hosts 
  which 
  are 
  positively 
  attractive 
  and 
  which 
  are 
  favoured 
  or 
  preferred 
  

   to 
  such 
  an 
  extent 
  as 
  to 
  cause 
  concentration 
  of 
  fly 
  in 
  localities 
  that 
  they 
  

   frequent. 
  

   (B). 
  Hosts 
  which 
  are 
  rather 
  repulsive 
  than 
  attractive 
  and 
  which 
  are 
  avoided 
  

   to 
  such 
  an 
  extent 
  that 
  the 
  flies 
  tend 
  to 
  scatter 
  and 
  disperse 
  from 
  localities- 
  

   in 
  which 
  no 
  more 
  favoured 
  hosts 
  occur. 
  

  

  