2go 
One  settler  in  Xyasaland,  for  instance,  who  kept  cattle,  had  to  abandon  one 
locality  owing  to  its  invasion  b}'  fly,  and  is  now  finding  that  his  second  estate, 
which  was  free  of  flies  when  he  took  it,  and  on  which  he  has  been  settled  lor 
several  years,  is  very  quickly  being  rendered  useless  to  him  in  the  same  way. 
There  arc  large  herds  of  cattle,  belonging  lo  Muropeans,  at  Fort  John.ston,  at 
the  south  end  of  Lake  Nyasa,  and  numbers  of  these  used  to  he  driven  with  impunity  to 
Zoniba  some  years  ago.  .Since  that  lime,  however,  the  fly  has  spread  to  such  an  extent 
that  it  is  no  longer  possible  to  do  this. 
In  one  district  in  N.  K.  Rhodesia,  actual  fly  maps  were  made  by  a  competent 
observer,  and  these  show  that  the  same  thing  is  happening  there. 
This  increase  cannot  be  ascribed  altogether  to  a  corresponding 
increase  in  the  amount  of  game,  for  the  fly  is  found  in  many  districts 
where  this  is  very  small,  and  in  other  cases  the  reverse  holds  good. 
The  Native  Commissioner  of  the  Serenje  district  of  N.  E.  Rhodesia  told  us  that 
tliere  is  practicallv  no  game  in  the  area  close  to  Chitambo  s  xiliage,  y'et  the  fly  are 
very  plentiful  there. 
Along  a  road,  in  the  Congo  Free  State,  running  parallel  with  the  Luapula 
river,  67.  morsilans  was  present  in  large  numbers,  and  yet  there  were  no  traces  of  game 
in  the  neighbourhood. 
Near  Abercorn,  in  N.  E.  Rhodesia,  there  are  large  herd.s  of  eland,  hartebeest 
and  roan  antelopes,  while  tsetse  flies  are  absolutely  unknown.  Other  examples 
might  be  given. 
It  has  been  recently  suggested  that  the  occurrence  of  the  flies  may 
be  cyclical,  and,  if  so,  it  may  account  to  a  great  degree  foi  these  fads. 
In  Nyasaland,  on  the  South  Rukuru  river,  where  tsetse  flies  are  said 
to  have  been  abundant  twenty  years  ago,  there  are  none  at  the  present 
time,  while  they  are  retuniing  to  the  Chobi  river  and  A\  ankies,  in 
.Southern  Rhodesia,  from  which  they  have  been  absent  for  some  years. 
There  is  a  native  rumour  that  they  are  also  returning  to  a  portion  of 
the  Transvaal. 
As  in  the  case  of  GL  palfalis,  the  food  of  Gl.  morsiiam  will 
depend  on  the  local  fauna,  and  it  does  not  follow,  as  has  often  been 
said,  that  wherever  buffalo  are  found  there  will  also  be  this  fly. 
On  one  occasion,  on  the  west  shore  of  Lake  Uangweulu,  we  covered  a  piece  of 
country  in  which  a  herd  of  from  fifteen  lo  Iwenly  bufTalo  habitually  lived,  without  meetmi; 
with  the  fly  anywhere,  even  close  to  the  herd. 
If  buffalo  are  found  in  country  where  GL  nwrsitans  exists,  it  is 
altogether  natural  that  the  flies  should  be  more  abundant  ui  the 
neighbourhood  of  the  animals,  as  they  do  not  travel  quickly,  are  local 
in  their  habits,  remain  practically  always  under  cover,  and  afford  an 
abundant  supply  of  food.  Farther  than  this,  we  do  not  think  there 
is  any  closer  connection  between  this  particular  or  any  other  species 
