the  wall.  With  one  or  two  exceptions  this  stage  was  recognisable 
after  eighty-four  hours.  In  these,  and  in  those  after  ninety-six  hours, 
the  gut  contents  were  fluid,  only  slightly  pigmented  and  largely 
crystalline,  but  the  cytoplasm  of  the  lining  cells  is  still  deeply 
pigmented.  In  unfed  flies — for  present  purposes  we  assume  as  unfed 
those  showing  no  pigmentation— the  cell  contents  are  dull,  finely 
granular,  and  free  from  pigmentation  and  vacuolisation ;  in  those 
in  which  digestion  is  proceeding  the  cytoplasm  becomes  more 
refractile,  less  granular,  vacuoles  of  varying  sizes  make  their 
appearance,  and  the  nucleus  is  more  readily  discernible. 
It  is  consequently  possible  to  state  with  certainty  whether  a  given 
fly  had  obtained  a  meal  of  blood  within  four  days  of  examination  ; 
it  is  probable  that  evidence  of  the  meal  remains  for  at  least  six  days, 
and  it  is  fair  to  suppose  that  a  fly  showing  no  signs  of  blood  in  its 
gut  has  not  fed  in  this  manner  for  five  days. 
On  Lake  Tanganyika  {Gl.  fal falls)  and  at  Kambole  {Gl. 
morsitans)  we  dissected  nearly  400  freshly-caught  flies,  of  which 
approximately  66  per  cent,  showed  no  signs  of  blood.  Near  Mpika 
82  out  of  a  total  of  112  freshly-caught  Gl.  morsitans  (79  per  cent.) 
were  free  from  all  traces  of  haemoglobin,  and  of  the  twenty-three 
which  showed  such  traces  five  had  but  recently  fed— in  three  the 
sucking  stomach  contained  large  quantities — most  probably  from  our 
caravan. 
In  three  Gl.  falpalis  of  which  we  have  notes — prior  to  the  first 
observation  it  possibly  passed  unnoticed — the  cells  lining  the 
intestine  from  proventriculus  to  proctodaeum,  were  in  the  state  we 
associate  with  active  digestion,  that  is  to  say,  there  was  refractility 
and  vacuolisation  of  the  cytoplasm,  and  the  lumen  was  occupied  b}' 
a  clear  fluid.  It  is  impossible  to  state  whether  this  was  due  to  a 
pathological  condition  or  to  the  digestion  of  a  fluid  which  was  not 
blood. 
These  observations,  though  unquestionably  limited,  would 
certainly  point  to  Glossina  as  being  capable  of  existence  for 
considerable  periods  without  blood,  and  possibly  to  their  ability  to 
feed  on  vegetable  juices  in  its  absence.  We  cannot  neglect  the 
additional  evidence  that  45  per  cent,  of  our  Gl.  fal falls  .showed  an 
intestinal  infection  with  flagellates,  and  it  is  possible  that  some  of 
