Yol.  49. ] 
PvISE  AXD  FALL  OF  LAKE  TAXGAXYIKA. 
5S1 
have  very  soon  cleared  the  entrance  of  the  Lukuga  of  those  bars,  or 
at  least  bowed  beside  and  over  the  masses  of  vegetation  which 
blocked  its  course. '  If,  on  the  contrary,  the  lake  was  sinking,  the 
diminishing  current  is  more  likely  to  have  been  too  feeble  to  remove 
such  a  hindrance.  Therefore  the  accumulation  of  sindi  may  have 
been  a  consequence  of  the  subsidence  of  the  lake,  which  prevented  any 
considerable  action  being  accomplished  by  the  dwindling  Lukuga. 
Perhaps  also  the  formation  of  a  bar  or  bank  at  the  entrance 
of  the  river,  described  by  the  explorers  Cameron  and  Thomson,  is  a 
mere  consequence  of  sinking  water-level.  The  old  embankment  of 
20  feet  in  height,  described  and  sketched  by  Mr.  Carson,  was  no 
doubt  formed  while  the  lake  was  high.  But  we  must  suppose  that 
here  there  were  not  the  same  conditions  for  sedimentation  as  at  the 
bar  of  the  Lukuga.  The  sediments  in  question  are  deposited  on  the 
point  where  a  Hat  valley  or  former  bay  debouches  into  the  lake ; 
they  could  be  carried  there  by  rain  and  small  streams,  and  accu¬ 
mulated  by  wind,  wave-action,  etc.,  the  more  easily  if  the  lake  was 
rising,  because  there  was  no  current  strong  enough  to  prevent  their 
deposition.  But  how  could  the  outlet  of  the  lake  be  closed  so  long  by 
such  a  bar,  considering  that  it  is  undiminished  in  breadth  and  in 
volume  and  swiftness  of  current  ?  In  the  struggle  between  sedi¬ 
mentation  and  river-erosion  by  a  stream  enlarging  and  deepening 
its  bed,-  victory  is  clearly  on  the  side  of  the  latter.  If  the  level  of 
the  lake  rises,  its  outlet  must  all  the  more  easily  overcome  the 
obstacle  existing  from  former  times.  The  formation  of  a  new  bar, 
however,  is  probable  only  in  so  far  as  the  river-current  becomes 
more  and  more  enfeebled:  that  is,  if  the  level  of  the  lake  be  sinking. 
Only  in  this  case  the  annual  variation  of  rainfall  and  water-level 
may  produce  the  effects  described  by  Mr.  Carson  as  a  continuous 
struggle  between  the  choked  bed  and  the  flood,  while  the  filling-up 
01  the  separated  mud-pool,  the  accumulation  of  sindi ,  tire  growth 
of  aquatic  vegetation,  etc.,  accelerate  the  closure  of  the  lake’s  outlet. 
But  it  is  at  least  an  equally  probable  hypothesis  that  tlie  for¬ 
mation  of  the  bar  commenced  while  the  lake  had  no  outlet  and  a 
comparatively  small  river,  also  called  the  Lukuga,  discharged  into 
the  lake  at  the  same  locality  (of.  Stanley,  Storms,  and  others).  In 
this  case  the  deposition  of  sediments  might  take  place  just  in  the 
same  way  as  in  the  locality  mentioned  above,  and  it  might  go  on 
until  the  rising  lake  regained  at  least  a  temporary  outlet.  The 
small  channel  mentioned  by  Commander  Cameron  may  be  considered 
as  the  work  of  the  current  in  its  initial  period,  eroding  perhaps 
only  during  the  wet  season.  It  would  be  very  interesting  to  study 
the  present  condition  of  the  bar,  its  increase  or  decrease,  and  its 
relation  to  the  rise  or  fall  of  the  lake.  At  all  events,  I  believe 
that  the  dwindling  of  the  lake,  due  to  climatic  oscillation,  is  the 
real  cause  of  all  the  alterations  of  its  outlet. 
The  above  hypothesis  agrees  with  the  views  of  Herr  Carl 
Ochsenius,1  of  Dr.  Paul  Bernhard,2  and  others,  who  likewise  regard 
1  Zeitsckr.  d.  Deutsck.  geol.  Gesellsch.  vol.  xliv.  (1S92)  p.  89. 
2  ‘  Deutsch.  Ostafrika,’  1892,  pp.  8S7  e,t  scyq. 
