.^21 
or  forty  yards  in  width,  accompanying  the  streams,  and  like  these 
unite  together  and  sometimes  open  out  as  the  large  grass  plains  which 
accompany  certain  of  the  rivers.  The  timber  for  the  most  part  is 
small,  rarely  exceeding  thirty  feet  m  height,  and  relatively  open. 
The  shade  furnished  by  these  trees  is  not  always  intense,  but  it  is 
sufficient,  when  in  full  foliage,  to  afford  comfort  after  ’the  open. 
Owing  to  the  almost  annual  fires,  the  branches  are  small  and  stumpy, 
and  the  undergrowth,  excepting  in  a  few  areas,  does  not  assume  any 
great  luxuriance  or  density  and  can  be  traversed  with  little  or  no 
discomfort.  In  the  more  densely  inhabited  regions  (three  to  four 
inhabitants  per  square  mile  is  the  average)  large  tracts  of  this  bush 
have  been  cut  down  about  four  feet  from  the  ground  and  the  land 
then  dug  for  gardens.  After  about  three  years  these  are  deserted, 
so  that  the  vegetation  of  considerable  areas  on  the  inter-dambo  ridges 
is  of  a  more  stunted  character,  but  is  apparently  equally  suited  to  the 
fly.  The  nearest  approach  to  the  dense  tropical  foliage  which  we 
met  is  found  in  the  oases,  termed  msitu^  surrounding  some  springs; 
these  are  small  areas  rarely  exceeding  a  few  acres  in  extent,  where 
the  trees  assume  forest  proportions,  and  are  clothed  with  rubber 
vines  and  other  cieepers,  and  are  interset  with  a  luxuriant  under¬ 
growth  ;  the  soil  is  soft  and  richly  vegetable,  and  the  water  is  close 
to  the  surface.  Although  on  arrival  in  the  territory  we  were  informed 
that  these  7n' situs  were  tsetse  areas  par  excellence,  we  have  never 
seen  Gl.  morsitans  in  any  of  them,  and  it  is  possible  that  small 
Tabanus  and  certain  Haanatapota,  which  are  numerous  in  most,  may 
have  been  mistaken  for  Glossina  by  our  informants.  Regarding  the 
nature  of  the  soil  we  can  say  but  little ;  the  opinion  is  held  by  most 
Europeans  and  natives  that  Glossina  morsitans  avoids  clays  and 
swampy  surfaces,  and  favours  those  of  a  friable  or  sandy  nature.  Our 
observations  tend  to  bear  this  out,  but  the  depth  of  the  soil  is  very 
variable,  being  so  shallow  in  many  places  as  to  permit  of  extensive 
out-crops  of  the  subjacent  rock. 
There  would  appear  to  be  no  special  desire  for  water  on  the  part 
of  the  fly.  In  most  parts  of  the  central  region  with  which  we  deal, 
many  of  the  water-courses  are  dry  for  two  or  three  months  (August  to 
October)  at  the  end  of  the  dry  season,  and  one  may  have  to  travel  in 
a  direct  line  for  over  twenty  miles  in  order  to  meet  a  permanent 
stream,  which  itself  is  little  more  than  a  trickle  ;  and  in  instances  could 
v 
