UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Decline of the Mara woodlands : the role of fire and elephants Dublin, Holly T. 1986

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
[if-you-see-this-DO-NOT-CLICK]
UBC_1987_A1 D82.pdf [ 9.6MB ]
Metadata
JSON: 1.0097282.json
JSON-LD: 1.0097282+ld.json
RDF/XML (Pretty): 1.0097282.xml
RDF/JSON: 1.0097282+rdf.json
Turtle: 1.0097282+rdf-turtle.txt
N-Triples: 1.0097282+rdf-ntriples.txt
Original Record: 1.0097282 +original-record.json
Full Text
1.0097282.txt
Citation
1.0097282.ris

Full Text

DECLINE OF THE MARA WOODLANDS: THE ROLE OF F I R E AND  ELEPHANTS  by HOLLY T. DUBLIN B.A., A n t i o c h C o l l e g e , Y e l l o w S p r i n g s , O h i o , 1976, M . S c , U n i v e r s i t y o f W a s h i n g t o n , S e a t t l e , W a s h i n g t o n , 1980 A THESIS SUBMITTED I N PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Department of Zoology) We a c c e p t t h i s t h e s i s a s c o n f o r m i n g t o t h e required standard  THE UNIVERSITY OF B R I T I S H COLUMBIA November 1986 ( c j H o l l y T. D u b l i n ,  1986  In  presenting  degree  at  this  the  thesis in  University of  partial  fulfilment  of  British Columbia, I agree  freely available for reference and study. I further copying  of  department  this or  publication of  thesis for by  his  or  her  representatives.  DE-6(3/81)  that the  for  an advanced  Library shall make  it  It  is  granted  by the  understood  that  head of copying  my or  this thesis for financial gain shall not be allowed without my written  ZOOLOGY  The University of British Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date  requirements  agree that permission for extensive  scholarly purposes may be  permission.  Department of  the  23 •hecvmWor  ISSlo  Supervisor:  Dr. A . R . E .  Sinclair  ABSTRACT The  Masai  northernmost  M a r a Game R e s e r v e extension  the c r i t i c a l d r y season migratory  of southwestern  Kenya forms t h e  o f the S e r e n g e t i ecosystem  and provides  range  1.5  wildebeest.  f o r approximately  Over  the past  e c o l o g i c a l changes have o c c u r r e d .  100  million  years,  major  The a r e a h a s e x p e r i e n c e d  a  t r a n s f o r m a t i o n f r o m open g r a s s l a n d t o d e n s e w o o d l a n d s a n d b a c k . This  study  addressed  from b o t h an h i s t o r i c a l study  focused  on  thet r a n s i t i o n  and a  i n t h e Mara woodlands  contemporary  two c e n t r a l  perspective.  questions:  1) what f a c t o r ( o r  f a c t o r s ) were r e s p o n s i b l e f o r t h e d e c l i n e o f the  1960s?  And  2)  what  factor  The  t h e woodlands  (or f a c t o r s ) are  in  currently  r e s p o n s i b l e f o r i n h i b i t i n g woodland r e g e n e r a t i o n ? In wild  t h e 1880s a n i n t r o d u c e d d i s e a s e ,  and domestic  ungulates  rinderpest,  i n the Serengeti-Mara  L o c a l p a s t o r a l i s t s , d e p e n d e n t on t h e i r  livestock  succumbed t o d i s e a s e a n d s t a r v a t i o n .  Elephant  been  greatly  reduced  decimated  by i n d i g e n o u s h u n t e r s .  region.  for survival, numbers h a d a l s o Explorers, slave  t r a d e r s , a n d h u n t e r s d e s c r i b e d t h e a r e a a s a n open g r a s s l a n d 1900.  In  t h e f o l l o w i n g decades,  the e s t a b l i s h m e n t o f elephant browsing By  woodlands;  c o n d i t i o n s were c o n d u c i v e t o  burning  t h e 1930s,  high  t h e area  rainfall,  f i r e s c h a r a c t e r i z e d the 1967).  rates  were  low and  was n e g l i g i b l e . was  c o v e r e d by d e n s e w o o d l a n d .  T h e s e w o o d l a n d s began a s t e a d y d e c l i n e Unusually  by  Although  high  period  woodland  several  decades  later.  grass p r o d u c t i v i t y , and severe of  losses  greatest were  d e c l i n e (1961  intially  viewed  as  "elephant was  the  problems", primary  elephants merely Elephants  findings  factor  in  from  the  the  open  rate  in  elephants  s e l e c t e d wooded h a b i t a t s .  Mara  grasses  and  season.  In  females  ate  matter.  of  the  fit  elephant herbs  in  selective  rainy  of  trees  feeding  diets  fire  woodlands,  while  in  the  Average  group  than  dry.  the  patterns  Elephants season  males browsed  diverse  class  season  that  swamps,  same f e e d i n g  the  Elephants u t i l i z e d height  wet  the  suggest  decline.  However,  populations.  general, more  season.  in  elephants  many A f r i c a n  other  higher  of  grasslands,  thickets  significantly  wet  study  d i s a p p e a r a n c e of  a c c e l e r a t e d the  preferred  this  relict  dry  season,  size  for  concentrated  on  than  containing  s e e d l i n g s under the  in  the  females, more  1m  dry  while  herbaceous  more  year.  significantly  was  reported  and browse  more  throughout  and  than  This  reduced  any  pattern seedling  survivorship. Large-scale treatments: browsed  browsed  nor  intensities,  of  burned  damaged  experiments  only,  burned.  and  removed a  field  browsed  Although  resprouted  significant others.  trampling,  thrashing,  seedlings  protected  and from  height.  Woodland dynamics  affected  by  elephants,  total  of  in  burning the  wildebeest,  plants  to  and  neither  current  fuel  loadings  the  majority  topkill, six  months.  Elephants  and  severely  s e e d l i n g growth browsing.  are  and o t h e r  through  Only  and browsing Mara  three  burned,  seedlings  inhibited  accidental both  at  within  proportion  Wildebeest  and  fire,  p r o d u c e d an a l m o s t  individuals  subjected  those  increased  currently browsers  in  more  than  by  iv  fire.  Elephants  can  be c o n s i d e r e d  a  "keystone"  species  in  this  system. I the  concluded  woodland  However,  tree  perturbations in  findings  did  phase,  woodland  or  necessary  alone, another, there.  the  move  apparently but  once  the  may  it  is  low  the  have  move the  of  "stable  holding  two  between  the  the  system  a of  decline.  My  limit  stable  two.  in  a  pattern states,  such as  fire  Elephants,  f r o m one they  cycle"  Mara This  factor  phase,  previous  frequency  rates.  an e x t e r n a l  ago.  following  the  locally  grassland  by  rates  rate  initiating decades  reduced  increased  burning  system  in  are  of  two  burning  Caughley's  and t h a t  cannot  over  and an  elephants  Mara  capable  been  accelerated  support  grassland, to  had  numbers  despite  not  started  increased  elephants  Today,  that  is  as  not  hypothesis.  suggests  which  wildebeest  fires),  were  densities  (such  man-made  grassland  elephants  declines  once  reduction  that  state  to  can hold  it  V  TABLE OF CONTENTS  ABSTRACT  ii  L I S T OF TABLES  x  L I S T OF FIGURES  xii  ACKNOWLEDGEMENTS  XV  CHAPTER  1.  GENERAL INTRODUCTION  CHAPTER 2.  THE  SERENGETI-MARA  1 WOODLANDS:  AN  HISTORICAL  PERSPECTIVE  10  Introduction  10  Methods  12  The  Q u a l i t a t i v e and E y e w i t n e s s  E v i d e n c e of  Vegetation  Changes The  12  D i r e c t E v i d e n c e of  Vegetation  Changes  13  The  Acacia  woodlands  16  The  Croton  thickets  17  Results  19  Vegetation  Changes  from 1880  to  1980:  the  qualitative  evidence Vegetation  19 Changes  from  1950  to  1982:  the  direct  evidence  30  The  Acac i a woodlands  30  The  Croton  34  thickets  Discussion  39  Summary  47  CHAPTER 3 .  HABITAT  SELECTION AND GROUP S I Z E OF ELEPHANTS  IN  vi  THE MASAI  MARA GAME RESERVE  49  Introduction  49  Study Area  50  Climate  51  Plant  53  Communities  Geology  and H y d r o l o g y  58  Methods  59  Wet and D r y S e a s o n T o t a l Monthly Census  C o u n t s of  Elephants  60  Circuits  62  Results  63  Habitat  Selection  63  Group S i z e  67  Discussion Habitat  69 Selection  69  Group S i z e  74  Summary CHAPTER 4.  77 FEEDING  ECOLOGY OF ELEPHANTS  IN THE  MASAI  MARA  GAME RESERVE  79  Introduction  79  Methods  81  Results  85  Feeding  differences  Diet  composition  85  Diet  diversity  90  Lengths  of  feeding  Characteristics Female  feeding  between m a l e s and f e m a l e s  of  bouts plants  patterns  85  99 in  the  diet  99 102  Diet composition  102  Diet diversity  103  Lengths of feeding bouts  106  Characteristics  106  of p l a n t s i n the d i e t  Male feeding p a t t e r n s  108  Diet composition  108  Diet diversity  111  Lengths of f e e d i n g bouts  111  Characteristics  112  of p l a n t s i n the d i e t  Discussion  113  Effects  of food q u a l i t y  114  Effects  of the w i l d e b e e s t  Effects  of seasonal  Effects  of woodland disappearance  migration  change  Summary CHAPTER  117 122 125 127  5.  THE EFFECTS OF ELEPHANTS, F I R E , WILDEBEEST AND  BROWSERS ON THE ACACIA WOODLANDS  129  Introduction  129  Background Information  130  Methods  133  Measuring Browsing E f f e c t s  136  Measuring Burning  137  Effects  Data A n a l y s i s  141  Results The  142 Large-scale  Mortality  F i e l d Experiments  of T r e e s and S e e d l i n g s  Seedling Inhibition  142 142 143  vi i i  Seedling  Growth  150  The  M u l t i p l e - b u r n Experiments  156  The  Seed G e r m i n a t i o n E x p e r i m e n t s  158  Discussion  158  Elephants  160  Fire  164  Wildebeest  169  Other  170  Browsers  Summary CHAPTER 6.  171 DYNAMICS OF THE MARA WOODLANDS:  PAST,  PRESENT,  AND FUTURE  173  Introduction  173  The  176  Model The  Assumptions  176  Elephants  176  Fire  177  Wildebeest  178  Resident  178  Constants The  Test  Browsers  and V a r i a b l e s  i n the Model  Conditions  179 182  Results  184  Woodland D e c l i n e s  i n the  1 960s  Woodland D y n a m i c s  i n the  1980s  189  Woodland Dynamics  i n the  1890s  191  Escapement  Rates  of T r e e s  Discussion The  1 960s and 1 970s  184  193 195 1 95  ix  C u r r e n t Dynamics  196  The F u t u r e  1 99  Management  Implications  202  LITERATURE CITED  204  APPENDIX A . THE WOODLAND DYNAMICS MODEL  224  X  L I S T OF TABLES  Table  2.1.  D e s c r i p t i o n of  Table  3.1.  P e r c e n t of  Table  3.2.  the  aerial  habitat  Average herd  types  size  photography  14  i n t h e Mara R e s e r v e  by h a b i t a t  ...  t y p e and s e a s o n  in  t h e Mara R e s e r v e Table  3.3.  68  Average herd  size  for  other  African  elephant  populations Table  4.1.  75  Number  of  focal animals,  mean m o n t h l y r a i n f a l l 4.2.  C o m p a r i s o n of  female  Table  4.3.  Comparison of  absolute  Table  4.4.  females Table  4.5.  types Table  Table  4.7.  4.8.  females Table Table  5.1. 5.2.  fires  of  and male d i e t diet  82  composition  diversity  .  94  relative  diet  diversity  between 97  f o r a g i n g bouts a c r o s s  C o m p a r i s o n of  all  forage 100  diet  composition for  females  and  seasons  Comparison and m a l e s  a n d males  105  of  length  between  Comparison  88  between  by s e a s o n  Mean l e n g t h o f  between  females Table  and  by s e a s o n  Comparison and m a l e s  rainfall,  by month and s e a s o n  4.6.  males  and males  total  for each sampling p e r i o d  Table  females  61  of  between  Seedling densities Characteristics  of  foraging  bouts  for  seasons feeding  107 height  preferences  for  seasons  109  for  five  sample p l o t s  of  large-scale  134  experimental 140  xi  Table  5.3.  ANOVA  results  f o r the  l o s s of stems a c r o s s  seasons, s i t e s , and treatments T a b l e 5.4. ANOVA r e s u l t s  f o r stem damage a c r o s s  145 sites  and  treatments T a b l e 6.1. C o n s t a n t s  148 i n t h e model  180  T a b l e 6.2. P a r a m e t e r s i n t h e model  181  T a b l e 6.3. T e s t c o n d i t i o n s f o r t h e 1 960s a n d 1980s  183  T a b l e 6.4. W o o d l a n d r e c r u i t m e n t r a t e s w i t h f i r e , and  elephants,  b r o w s e r s i n t h e 1960s  T a b l e 6.5. W o o d l a n d r e c r u i t m e n t r a t e s w i t h f i r e , b r o w s e r s , a n d w i l d e b e e s t i n t h e 1980s  185 elephants, 190  xi i  L I S T OF FIGURES  Figure  1.1.  Map of  the  Figure  1.2.  Map of  t h e M a s a i Mara Game R e s e r v e  Figure  2.1.  Narok  Figure  2.2.  Figure  rainfall  Changes  woodlands 2.3.  from  S e r e n g e t i - M a r a ecosystem  in  1950  to  Absolute rate  -  five-year  2 3  averages  27  t h e mean c o v e r d e n s i t y  of  Acac i a  1982  31  of  cover  loss  in  the  Acac i a  woodlands Figure  2.4.  32 Relative  rate  of  cover  loss  i n the Acac i a  woodlands Figure  2.5.  Cover d e n s i t y  through Figure  33 of  Croton  thickets  from  1950  1982  2.6.  35  Absolute  rate  of  cover  loss  i n the  Croton  thickets Figure  2.7.  36 Relative  rate  of  cover  loss  in  the  Croton  thickets Figure Figure  2.8. 2.9.  37 Mean a r e a o f Trends  in  Croton  thickets  in  1950  and 1982  human p o p u l a t i o n d e n s i t y  near  .  the  Mara Game R e s e r v e Figure  3.1.  Monthly  41 rainfall  totals  for  1982,  1983,  and  1984 Figure Figure  52 3.2. 3.3.  aerial Figure  38  Mean m o n t h l y maximum and minimum t e m p e r a t u r e s Wet  and  d r y season h a b i t a t  preferences  .  from  survey data  3.4.  Habitat  preferences  54  64 for  females  and  males  in  xi i i  the  dry season  Figure  3.5.  the  wet  Figure in  in  Habitat  preferences  for  females  and males  in  season  4.1.  66  Seasonal d i e t  1982,  Figure  65  1983,  4.2.  c o m p o s i t i o n of  females  and  males  and 1984  86  Seasonal d i e t  c o m p o s i t i o n of  females  and m a l e s  1983-84  87  Figure  4.3.  Absolute diet  Figure  4.4.  L e n g t h of  species Figure  4.5.  species Figure  eaten  observation  females  time  and males  versus  number  .  Length  of  92  o b s e r v a t i o n time v e r s u s  number o f  by m a l e s  Relative diet  93 d i v e r s i t y of  females  a n d m a l e s by  month Figure  96  4.7.  Relative  browsing  diversity  of  females  and  m a l e s by month Figure  4.8.  monthly 4.9.  Figure  4.10.  Figure the  five Figure  Percent  browse  in elephant  diets  v e r s u s mean 104  S t a n d i n g c r o p biomass t h r o u g h the Crude p r o t e i n l e v e l s  in long  dry season  grasses  ..118  through  dry season 4.11.  119  Grams of  d r y season  grazing Figure  98  rainfall  Figure  the  91  of  by f e m a l e s  eaten  4.6.  d i v e r s i t y for  crude p r o t e i n in long grasses  and t h e change  in elephant  and b r o w s i n g b e f o r e and a f t e r  5.1.  Height d i s t r i b u t i o n s  of  diets  through between  the d r y season  Acacia  gerrardi i  ...121 in  sample p l o t s 5.2.  P e r c e n t of  135 marked stems removed by  elephants,  xiv  wildebeest,  and o t h e r  browsers  in  burned  and  unburned  plots Figure not Figure  144 5.3.  Percent  damaged 5.4.  Frequency  elephants, 5.5.  Figure  fire,  trampled,  and  of  149  return  originally  times  for  removed  by  and w i l d e b e e s t in  average  151  stem  heights  in  burned,  and e x c l o s e d p l o t s  5.6.  Change  unburned, Figure  distribution  "regenerates"  Change  unburned,  marked stems b r o w s e d ,  i n b u r n e d and u n b u r n e d p l o t s  resprouting  Figure  of  in  152  average  stem  numbers  in  burned,  and e x c l o s e d p l o t s  5.7.  Seedling  155  survivorship  in  multiple-burn  experiments Figure  5.8.  Average  treatment Figure  5.9.  and  1984  Figure  Figure  -  Height  surviving  five  fire159  structure  of  Acac i a  woodlands  in  1974 162  Elephant  occupancy  of  the  Mara Game  Reserve,  6.2.  burning 6.3.  165  Woodland r e c r u i t m e n t  population  Figure  seeds  1983  6.1.  Figure  number of  intensities  5.10.  1980  1 57  r a t e s at  varying  elephant  sizes Woodland  187 recruitment  rates  under  varying  rates Escapement  188 rates  for  trees  by  height  and  age  class Figure  194 6.4.  elephant  Woodland  population  recruitment sizes  rates  under  and b u r n i n g r a t e s  varying 200  XV  ACKNOWLEDGEMENTS Research cooperation  of  an  and  exception.  In  approval  the  of  international  assistance  National  Conservation  and Management  organizations.  Program  for  funding  S i g m a X i , New Y o r k Society, balance was the  work  and of  the  E.  Don  to  am  the  grateful  to of  Zoological  Society,  the  Graduate  by.World  Mara.  For  Train,  Jorie  this  Judy  in  my  Kenya.  Wildlife  Kent  year  faith  and R i c k  I  Blachford,  Susan  am  especially  grateful.  the  University  throughout  b u o y e d my m o r a l e My  my e x t e n d e d  Wildlife  covering of  the  writing  kept  stay  in  of to  Robin  e a c h r e a d my  criticisms  Ertis,  my c o m p u t e r  Monica Thain  also  am i n d e b t e d  Roy S t r a n g  Alistair  of  Scholars  Sinclair,  comments.  aspects  of  Weyerhaeuser.  A.R.E.  and  support  F u n d - U . S and F r i e n d s  of  members,  Myers,  My  for  many u s e f u l  all  Mwango,  East African  and p r o v i d e d  in  Kenya  fieldwork,  thoroughly  helped  the  County  C.A.  Fulbright  dissertation  Graystone  no  Wildlife  Narok  thank,  Fellowship  gesture  Butler  committee  Ludwig,  the  the  the  portion  entirely  was  through  financial  major  My s u p e r v i s o r y Liley,  and  like  requires  my r e s e a r c h e x p e n s e s  Masai  study  and Simon M a k a l l a h .  a University  supported  Russell  would  also I  Project,  Department I  David Mbuvi,  International many  Monitoring  particular  David Andere,  This  the  S c i e n c e and Research C o u n c i l , the  Ecological  In  many p e o p l e .  requires  K e n y a , my r e s e a r c h w a s made p o s s i b l e  Rangelands  Council.  of  nature  and  work  for  and  Shirley which  I  me l e g i t i m a t e  with  Kenya.  Lee  Gass  fieldwork  both  o n many o c c a s i o n s .  assistant,  Joseph  Kibrech,  made  the  xvi  possible  and e n j o y a b l e .  Lamprey,  and  intellectual  Robert  thank  and,  sincere  Although  the  more  field  My necessary  part  motivation  back  into  me t o the  highest  his  siblings  the  gratitude  idea  a n d on t h e  questions,  a  scope of  go  turn  reality.  my  for  I  David  to  into  when  kept  humour  to  my l i f e  which  undertaking.  Harvey  Hinsz  who  first  could  imagination.  took  answer  I  me t o  accomplish  to  took  support  the  question,  since  Sinclair. he  to  those  to  support  Tony  criticism  for  like  ever  me t r y i n g  my l o n g i n g s  Western  Mara.  unfaltering  His  Mike  ecological  moral  supervisor,  his  do  my l i f e  the  and  while  kept  Without  my p a r e n t s what  directed  my  homefront.  any major  that  chance  in  Richard  especially  who g a v e me t h e  to  marks  field, and  would  times  Frank,  companionship  the  w h a t h e was g e t t i n g  trust  to  in  first  largely  difficult  added  of  to  believe  has  thanks  I  me t h a t  Myles Turner  h e h a d no  he d e s e r v e s  my y e a r s  Douglas-Hamilton,  giving  my m o s t  Laurence  provided  from N a i r o b i .  Kenya which  on d u r i n g My  in  for  posthumously,  carry  on,  in  over  Iain  good a d v i c e  Ted Goss  research  Sapolsky  stimulation  Norton-Griffiths, supplied  My c o l l e a g u e s ,  me  both asking  them. is  open  owe  my  Africa  such a  gave  wide  was o n l y  me  and  me  plains deepest raised  limited  by  1  CHAPTER  1.  GENERAL INTRODUCTION  T h e M a s a i M a r a Game R e s e r v e of  the  Serengeti-Mara  Mara's  extensive  recent  years  Lamprey  1985).  for  (Glover  approximately  taurinus  1913,  in  White  replaced  et  the  1914,  by  1915,  1935,  Darling  late  1950s  these  of  trees  through within  The in  the  Mara.  of  woodland  Mara  Game  quantified Serengeti the  the  the  the  1927), over  began  1960s and  1970s  of  these  a wealth  Serengeti little  Reserve.  of  season  range  century  following  pers.  comm.).  few  was n o t  years  In  the which  stands  biological Park  information  information over  was  until  well-documented  the  had  three  available  declines  in  the  neighbouring  was d r a w n  to  the  status  similar  30  today.  woodland  many  were  intact  was n o t  attention  (Woosnam  decline  It  that  were  grasslands  steady  declines  National  Connochaetes  grasslands  this  leaving  in  dry  the  a  The  1984,  these  Reserve boundaries  change,  rapid  of  1.2).  Dublin  wildebeest,  Downey,  woodlands  While  in  S.  and  1973,  critical  turn  portion  significantly  Although  and t h i c k e t s  1960,  progression  collected  Buxton  woodlands  the  1985). the  1.1  Taiti  migratory  Mara around  (Lewis  continued  al.  northernmost  expanded  1970,  provide  million  the  (Figures  have  and Trump they  1.5  (Sinclair  extensive  ecosystem  grasslands  Today  forms  decades for  Norton-Griffiths  of  been  the (1979)  northern  woodlands  in  Mara. Like  east,  central,  woodland "elephant 1967,  losses  and in  Russell  1968,  southern  the  problems"  declines  in  parks  Africa  Serengeti-Mara (Darling  Glover  and r e s e r v e s  (Laws were  1964,  Glover  and Trump  1970).  1970,  Caughley  initially 1965,  throughout 1976),  viewed  Lamprey  Ecological  et  as al.  doctrine  Figure  1.1.  A map  of the Serengeti-Mara  ecosystem.  F i g u r e 1.2.  A map of the s t u d y a r e a , the Masai Mara Game R e s e r v e .  4  of  this  period  densities were  in  argued  parks  that  and  reserves,  becoming a r t i f i c i a l l y  grew  and  concentrated the  lands  areas elephants  mortality  1972,  in  the  al.  et  1967,  al.  Goddard  1974, and  Agnew  elephants (Cumming  Owen-Smith  was  1981,  1968,  Barnes  1975, in  Hanks Kenya  1968,  Laws  Pienaar  1983),  in  solution  elephant/woodland  on  park  to  pursue  managers similar  Lamprey into they  the  the  et  in  dilemmas.  Glover  and  Controlled  1983),  parks  South  et  al. al.  Barnes  1983)  P r e s s u r e was woodland  of  Africa 1983,  (Laws et  areas experiencing  and  Hall-Martin  1963,  1976,  Africa  Caughley  1981,  de V o s  and Uganda  1971,  Tanzania  national  (Bell  1969,  Field  1970).  the  ( R o s s e_t a l .  other  1975) as  a  placed  declines  options.  al.  (1967)  Serengeti-Mara  had not  in  Malawi  and recommended to  others  1969,  Dawkins  southern  1974, al.  effect  Douglas-Hamilton  1983),  (Glover  instituted 1983),  et  and  1978),  1970,  1982,  destroyed  This  1971,  Malpas  Laws  of  elephant-related  Spence and Angus  v a n Wyk a n d F a i r a l l  1983,  elephants  (Buechner  1976,  Once  capacity  regeneration.  Uganda  pressures  areas.  and  Anderson and Walker  parks  culling  1969,  b,  Thomson  1964,  (Pienaar  1970,  &  Sheldrick  Zimbabwe  of  Savidge  1966,  other  protected  Consequently,  F i e l d and Ross  1974a  (Pienaar  of  parks  Serengeti-Mara,  exceeded the  was e x c e e d i n g n a t u r a l  1975,  et  into  africana)  settlement  over-utilization,  1961, Laws  Croze  1984)  them.  through  was d o c u m e n t e d  (Lamprey  these  support  tree  a_l.  human  in  habitats  Laws e t  as  elephants  to  Buss  high  (Loxodonta  such as the  forced  their  1961,  elephant  been p r e s e n t  in in  described the  late  an  influx  of  1950s and e a r l y  significant  numbers  for  elephants 1960s  where  at  least  5  the  previous  40  h i s t o r y of t h e s e immigrating conducted  years.  Dublin  movements  elephants.  the  and  the  a  review  In  i n the Serengeti-Mara,  p r e s s ) document t h i s g r a d u a l  late  1950s and  southern  This increase in elephant  anecdotal  origins  of a l l e l e p h a n t  D u b l i n and  the l a t e  the  geographical  and  by  (in  elephants  northern  of  censuses  Douglas-Hamilton  re-colonization  e v e n t u a l merging of t h e i r  between the  (1984) r e v i e w e d  and  populations  1970s.  numbers and  densities  coincided  c l o s e l y w i t h the r e p o r t e d d e c l i n e of woodlands i n the S e r e n g e t i M a r a and  was  elephants  were  grassland. decided their  c i t e d as e v i d e n c e responsible  impact  with l i t t l e  elephants on  success  & b) c o n c l u d e d  for  was  killed  by  and  from  around  the observed  there  local  and  The  culling  his  studies,  in  the  that elephants  woodland d e c l i n e s .  His  Croze  were n o t results  regeneration to offset  to  responsible showed  that  t h e number o f  trees  researchers  in  the  Serengeti  1930,  Fire  1965,  Buss the  1961,  West 1 9 6 5 ) .  detrimental  woodland r e g e n e r a t i o n i n the S e r e n g e t i . investigated  these  effects  had  Africa  Glover  effects  of  a  (1965,  fire  Norton-Griffiths in  to  bush encroachment  i n the r e d u c t i o n of n a t u r a l woodlands i n  studied i n i t i a l l y  further  (1974a  elephants.  a s an a g e n t  1972)  met  Serengeti  f u r t h e r f o r p o s s i b l e c a u s e s of woodland d e c l i n e .  (Phillips  reduce  operations  central  l o n g been r e c o g n i z e d a s a means o f c o n t r o l l i n g and  researchers  woodland l o s s e s c o n t i n u e d .  These f i n d i n g s i n s p i r e d look  authorities  woodlands.  that  r e v e r s i o n t o open  e x p e r i m e n t a l l y i n an a t t e m p t  Seronera,  sufficient  of the h y p o t h e s i s  f o r the observed  In the mid-1970s, park  to c u l l  woodlands  i n support  detailed  on  (1979) study  6  correlating canopy  studied  and  the  dry  were  over  study  factor  factors)  woodlands  are  in  currently  experimental  1)  factors,  woodland  cover  trees,  fire,  alone,  woodlands  was  or  severity,  and  reduced recruitment  fires  experienced during Currently,  the  the  in  responsible  and t h i c k e t s .  the  rates time  of  by did  hypothesized  central  that  for  the  what  changes  Viewing  these  1)  what  observed decline  factor  in  perspective,  questions:  of  (or  factors)  of  woodland  inhibition  answered  hypotheses  for  in  the  mortality a  through  the below the  to  younger,  1960s  to  other  losses  in  decline fire  smaller  adult  in  mature  a level  decline;  rapid  increase  the  addition  rates  the  the  in  By k i l l i n g  population for  that  observed  Mara Reserve.  T h r o u g h an  destroyed  area  investigation  years.  mortality,  was r e s p o n s i b l e  resulting  the  decade  findings.  increased tree  rates,  of  investigate  were  f o c u s e d on t h e  in  to  the  questions  density  elephants  recruitment  for  elephant-related  mortality  the  were  well.  30  2)  woodland  Serengeti  and a contemporary  And,  and o b s e r v a t i o n a l  1970s:  as  past  responsible  1960s?  These  I  (1979)  a d d r e s s i n g two  were  the  Specifically, and  the  his  of  over  use  Although  there  the  fire  the  initiated  responsible  regeneration?  with  loss  in  i n c i d e n c e of  season.  was  the  losses  an h i s t o r i c a l  was a i m e d a t  (or  the  operating  Mara woodlands both  that  with  Norton-Griffiths  study  from  fires  secondarily  1982 t h i s  changes  the  with  Mara,  factors  In  the  the  of  found  only  during  include  similar  He  correlated  elephants  the  frequency  cover.  primarily  not  the  above  and  2)  of  the  frequency  size  classes  mortality  rates  declines.  M a r a w o o d l a n d s do n o t  appear  to  be  returning  7  to  their  former  woodlands,  I  addition  to  status.  hypothesized that: other  top-killing  growing  more  to  Mara today  successful no  mature  feed  which  individuals  and  recruitment;  a n d 5)  population  is  the  recruitment  raises  By  below  for  and  increase  in  a  inhibitory  major  adult  woodland  the  small  recovery  by  mortality  them  elephants  in  their  they  are kill  seed-producing above  migratory  current  wildebeest  effect  on  woodland  wildebeest  rates,  and  from  elephants  rates  trees,  recovery  in  where  of  mortality  the  regimes,  prevent  classes,  number  adult  4)  Furthermore,  the  damaging  rates  potential  an  the  classes;  height  fires.  fire  preventing  seedlings  larger  reduces  day  dynamics of  woodland  and  height  on  to  having  regeneration.  the  heavily  current  present  seedlings  susceptible  trees  the  prevent  fire-tolerant  escape into  longer  3)  factors,  continually  the  To s t u d y  thereby  keep  reducing  furthering  woodland  declines. I  begin  in  quantitative  Chapter  documentation  and t h i c k e t s .  Using the  interviews  with  the  history  of  The  loss  analyzed  and q u a n t i f i e d  historical  documented  time  overview  boundaries  and in  Chapter  the  the  writing  decline  of  cover, from  of  of  the  over  which  a  both  explorers  residents  Mara's vegetation  woodland  the  of  long-time  the  of  spanning  2 w i t h a summary o f  the  began  series  the  Mara  and area,  I  aerial  of  woodland d e c l i n e s w i t h i n  an  analysis  and  years.  1950's,  is  photographs  The c h a p t e r  provides the  an  Reserve  woodland changes which have  Serengeti National  3 presents  100  the  1948 -  also discusses  woodlands  reconstruct  past in  and  hunters  period the  1982.  of  qualitative  been  Park. of  habitat  selection  by  8  elephants  in  techniques, seasonal are  the M a r a .  elephants  bases.  compared t o  over  the  habitat  U s i n g b o t h a e r i a l and g r o u n d c e n s u s i n g  are p l o t t e d  These the  same  a v a i l a b i l i t y of  area.  describes  the  Reserve at In  for  elephant the  wet  d i f f e r e n t i a l use  Chapter  times  4,  the  I  of  on  different  habitat  habitats  the  the  foraging  dynamics.  I  feed.  The f e e d i n g  elephants on  was  a seasonal  examine what p l a n t  patterns context  woodlands factors  to  exposed, trampling, wildebeest,  they  Mara  what g r o w i n g c o n d i t i o n elephants  on  select,  prefer  b o t h a d u l t male and a d u l t  the to  female  g r a z i n g on g r a s s e s and  The b a l a n c e is  an  habitat  in this  5  and  which  types.  therefore,  aimed a t  The  feeding  discussed  seedlings  browsing  intermittent  and  trees  by e l e p h a n t s  burning  by  are  and o t h e r  harvesting  in  the  Reserve.  e x p e r i m e n t a l work i n t h e  t h r a s h i n g , and i n a d v e r t e n t and  elephant  decisions  d y n a m i c s i n t h e Mara the  of  T h i s c h a p t e r p r o v i d e s an o v e r v i e w  Acac i a  including:  plant  summarized h e r e  aspect  t o management  study a r e ,  summarizes  the Mara.  between important  and c r i t i c a l  current vegetation  of  of  basis.  observed  Chapter  the  and i s  specific  of  in  r e c o r d e d on a m o n t h l y b a s i s  ecology  preserving  are  influence  species  what h e i g h t s  b e h a v i o u r of  b r o w s i n g on woody s p e c i e s foraging  are eaten,  i n , and a t  the  analysis  habits  woodland  are  This  by e l e p h a n t s  towards u n d e r s t a n d i n g t h e i r  plants  with  preferences  w i t h a view  preferred  types  year.  examine  these p l a n t s  and  habitats  habitat  and d r y s e a s o n s .  of  by  sightings  elephants  what p a r t s o f  monthly  elephants  By c o m p a r i n g e l e p h a n t  both  different  habitat  d i s t r i b u t i o n s of  availabilities,  established  by  by  wildfires  Acacia of  the  currently browsers, migratory in  the  9  Reserve. scale  I  and  burning, fire the  multiple  and the  Mara  is  Chapter model  6,  central for  and the  explain  the  model  which  wildebeest the  then  burning  the  experiments exclusion  rates  of  light  and  seeds  for of  dramatic  inhibition Chapter  patterns  of  of 6.  of  present  a  in  simple  that  Several hypotheses change.  the  impacts  the  scenarios might  factors  in  under  management  of  of  are  this  the  in  Mara  put  the today  forth  to  simple  simulation  of  fire,  elephants,  test  predictions  to  1960s and t o d a y . various  realistically  findings  the  were  A  the  generate  or  recovery  in  for  used  woodland  measured  implications  are  observed  conditions  and  in  Results  declines  constructed  Lastly,  woodlands  woodland  is  future.  of  and  varying  findings.  factor  browsers  the  to  studies  and other  regimes  large-  browsers  exposed  these  what  woodland  incorporates  used to  both  recovery  controlled  both  data.  questions  the  on  of  s u m m a r i z e my r e s e a r c h a n d  these  addressed in  under  I  of  potential  discussed in  where  responsible 1960s  The  experiments  utilizing The  is  germination  then  these  results  burns,  intensities.  from  are  d i s c u s s the  Reserve  model  browsing  occur  study  The  in are  are  the  Mara  and in  summarized  discussed.  10  CHAPTER 2.  THE SERENGETI-MARA WOODLANDS:  AN HISTORICAL  PERSPECTIVE  Introduction  Over  the  taken place central,  past  30  i n many of and a  grasslands  (Caughley  tree tree  and  that  mortality is  The Kenya  Praet  of  (Glover  1973,  Tanzania  (Laws  elephants 1963,  Parker  (Lamprey e t  1969,  Croze  1974  Barnes  1983,  1985,  ajL.  1966, (Buss  1962,  Buechner  Parker  1968,  al.  1975,  (Caughley  1961,  the  Malawi  Field  and  Goddard  ( A n d e r s o n and W a l k e r  explain  these  inhibited  increased  of  This  adult  overall problem'  elephants  1980,  1982,  1985),  into  1981,  1963,  1974,  Watson  1976,  1970,  Caughley  Thomson 1975,  van  1985), Bell  Weyerhaeuser  1982,  (Pienaar  Hall-Martin Brooks  Field  Smart e t  and  and  Buss and S a v i d g e  Wing and Buss Ross  1961,  studied  and B e l l  South A f r i c a  aJL  been  Western  (Jachmann 1968,  Hanks e t  1974,  open  'elephant  Savidge  Barnes  al.  as  Laws 1969,  B u e c h n e r and Dawkins  et  well  concentration  1968,  Weyerhaeuser  Laws 1970,  and  as  to  actively  i n w o o d l a n d d y n a m i c s has  1967,  1969,  to  have  1970).  1983),  a&b,  Pienaar  Uganda  the  Agnew  al.  as  east,  areas  and b a r k - s t r i p p i n g .  r e f e r r e d to  1969,  has  have  of  woodland  hypothesis  browsing  to  affected  dense  recruitment  attributed  role  One  through f e l l i n g  areas  The  from  1976).  and  p a r k s and r e s e r v e s  Africa.  reversion  frequently  largely  d r a m a t i c woodland d e c l i n e s  national  elephant  regeneration  protected  in  rapid  states  phenomenon  the  southern  experienced  changes  years,  al.  1976), Guy 1976,  et  1984),  and  Buss  1966,  Laws and  1971,  Laws  1985), and 1981,  et  Zambia  Zimbabwe Cumming  11  1981).  Many  of  elephant  foraging  these was t h e  Norton-Griffiths to  explain  the  Serengeti  National  rinderpest  epidemic  grazing  ungulates  unused burned  grass the  The northern in  In  review  both  Serengeti-Mara  area  events  and p r o v i d e  aerial  paper, to  I  compare to  floral place  over  loss  photographs  the of  of  the  pattern  woodland  the  long-run due  of  which  of  dynamics  eyewitness 1880s  to  mature  to  that  major  of  in  canopy cover area  taken  losses not  hypotheses  and  the  Mara.  accounts the  early of  of  the  1980s.  historical in  the  woodlands  addition, from  I  of  changes  in  the  evidence  and e x t e n t In  borders  woodland  reconstruction  century.  Reserve  Tree  elephants.  structure,  woodland  annually  some  loss  alternative  evidence  past  the  these  late  This  recruitment  new l i n e s  and  diversity,  in  an  1960s.  In  examine  a step-wise  strong  fire.  Game R e s e r v e ,  woodland  from the  allow  offset  general  qualitative  accounts  the  the  relevance  These  analyzed  by  of  in  which  and  reduction  National  this  available  their  taken  destroyed  immigrating  evaluate  have  inhibited  of  fits  wildfires  1950s and e a r l y  activities  Serengeti,  and  grasslands.  for  the  population  productive  longer  Mara  that  resulted  late  because the  northern  Serengeti  fuel  the  hypothesis  argued the  that  losses.  Tanzania's  reduced  no  Africa.  faunal  in  in  could  Masai  previously  highly  eventually  burning the  the  the  hypothesis  woodland  Norton-Griffiths  therefore  declined  extensive to  were  woodlands  in  area  in  the  f o r t h an a l t e r n a t i v e  profoundly  the  was  woodlands  trees  that  provided  recruitment communities  of  supported  factor  put  Park.  of  entire  major  (1979)  loss  underutilization  studies  five 1950,  I  have  sets  of  1961,  12  1967,  1974,  taken  and  1982.  together  The q u a l i t a t i v e  reveals  experienced  a transition  and back  less  in  than  that  the  from open  100  and q u a n t i t a t i v e  evidence  Serengeti-Mara ecosystem grassland  to  dense  has  woodland  years.  Methods  The Q u a l i t a t i v e and E y e w i t n e s s ' E v i d e n c e of The records  evidence for  century  personal sources the  several  from  explorers,  comes  journals  of  were f o u n d  collection  president,  of  of  Joseph  Photographs  slave  these  Kenya N a t i o n a l A r c h i v e s ,  mainly  from  first  vice-  donated  from t h e  Johnson i n the  Johnsons.  The  relocated  using  of  vegetation  present Mr.  Mara the  Mara  M.I.M. 1972,  Museum  the  i n the Turner,  Kenya's  landscape permanent  warden of  provided detailed  the  sources.  in  the  This  photographs  allowed  recollections Some of were  Serengeti  accounts  taken  The  1920s and 1930s by  his  of  the were  a comparison  r e c o r d e d 50-60 y e a r s  1940s and 1950s,  the  Chanute, Kansas, provided  landmarks.  1940s.  period covering  from s e v e r a l  i n the  panoramas  with that  early  time  in  Serengeti  S y d Downey t o l d me h i s  since  by  hunters,  Murumbi.  Martin  taken  the  A l l of  visiting  reports  of  Commissioners.  Africana  traders,  turn  the  of  1950s were c o m p i l e d  photographs  the  obtained  and  and a c c o u n t s  Osa  I  early  1930s t h r o u g h t h e and  sources.  and a f t e r  Narok D i s t r i c t  i n the  Changes  of  descriptions  records  numerous  decades before  the  the  from  Vegetation  ago.  the  Serengeti-  photographs,  taken  in  subsequently  relocated.  N a t i o n a l Park  from  from h i s  1956-  daily diaries  of  13  the  period.  1950s, his  Rodney  Eliott,  s e n t me d e s c r i p t i o n s  field  For  and  p e r i o d from 1950  1967,  and 1982  1974,  Norton-Griffiths  Lamprey  scale.  to  of  but  changes  time,  the  in  taken  both  the from  Acac i a  Mara Game R e s e r v e ,  I used a s e r i e s  (Table 2.1).  were  of  flown  Vegetation  was m o d i f i e d  10  to  during  complete in  1961,  analysis  fit  aerial  1950,  the  followed available  of  on t h e  smallest  1985).  across  100  less  'dot-grid'  were made f o r  the  each  differences  described  set  in  analysis.  of  r e g u l a r l y - s p a c e d d o t s was c o n s t r u c t e d the  1:50,000  analyses.  scale  Each dot  photography.  photography  years. at  it  was  than  aerial  photographic  was  (1974, 2.5m  was  Standardizing  1:68,000),  on  The g r i d s  the  were t h e n  20x-magnification. necessary 2.5m,  i.e.  to  ignore  ground  those  Woodland c a n o p y c o v e r  smaller (canopy  the trees  from  (Lamprey size  with a  density)  scale canopy  chosen was  for  strong  larger  than the  in  t h e minimum  a n a l y z e d under  For all  in a  0.05mm  p r o v i d e d a " l o w e s t common d e n o m i n a t o r " d o t  all  photography,  size.  for  for  scale  illumination  diameter  Dot g r i d s  canopy d i a m e t e r  This  by a  were p r e p a r e d p h o t o g r a p h i c a l l y a s  pattern  detectable  were e s t i m a t e d  account  A grid  diameter  dot  of  Five surveys  changes  (1985).  photography  use  1982,  (1979)  The m e a s u r i n g g r i d s  the  that  in  coverage.  Woodland  x  thickets  to  surveys.  10  area at  Mara  Changes  analysis  Croton  photographic  by  the  Vegetation  quantitative  woodlands  aerial  of  the  notes.  The D i r e c t E v i d e n c e of  the  a game warden of  grid  measured  14  Table  2.1.  The a e r i a l p h o t o g r a p h i c s e r i e s u s e d i n t h e a n a l y s e s o f woodland cover l o s s i n the Masai Mara Game R e s e r v e . Each survey p r o v i d e d t o t a l coverage o f the study area.  FLOWN BY  SOURCE  YEAR  SCALE  1950  1:30,000  Royal A i r Force  1961  1:50,000  Royal A i r Force  Survey o f  Kenya  1967  1:57,000  Royal A i r Force  Survey o f  Kenya  1974  1:68,000  Meridian  Survey o f  Kenya  1982  1:48,000  Kenya R a n g e l a n d s Ecological Monitoring Unit  Airways  Directorate of Overseas Survey  W i l d l i f e Planning Unit W.C.M.D.  1 5  as  a p r o p o r t i o n of  vegetation This  out  analyzed  adjacent  to  When  the  the  the  total  the  density size  of  bias  u s e d was  direction In were  to of  the  changes  dot-grid of  for  the  grid.  vegetation  area  size  cover  immediately  on  small  scale  i n a way r e l a t e d  Norton-Griffiths  technique.  2.5m,  error.  photographic  between t h e  He  analyze  concluded  Since  sets,  different  was assumed  to  be  and C r o t o n c o m m u n i t i e s  to  1979).  would o v e r e s t i m a t e  100% b i a s  all  detected  Acac i a  dot  1964,  two o r a  same  an  woody  sample  F) u s e d computer m o d e l l i n g t o  size,  of  with  or  time p e r i o d .  overestimates  be c o n s t a n t  both  finite  in t h i s  by a f a c t o r  in  same  trees  d o t s on t h e  (1985)  (Kershaw  dot  touching  comparisons  the  a  Appendix  chosen  considered  over  with  diameter  (1985,  degree  that  Reserve  dots  number of  Lamprey  one a l w a y s  dot's  Lamprey  by  working  photography,  the  of  number o f  procedure allowed d i r e c t  trends  to  the  this sets  cover  the  dot  bias  was  and  the  unbiased.  three  measurements  calculated: 1)  The mean c o v e r  density  p h o t o g r a p h y was d e r i v e d u s i n g 2)  From  these  data,  c a l c u l a t e d by t a k i n g t h e successive  years  if  mean  cover  absolute  change  loss  year.  per 3)  cover  30%  would be  data.  each  year's  i n mean  absolute cover  and d i v i d i n g t h i s  the  two m e a s u r e m e n t s .  in  1950  change  aerial  methods.  annual  and 20% i n  10% d i v i d e d by  The a n n u a l r e l a t i v e density  for  above  the  series  separating was  the  difference  photographic  number of  value  was  between  two  by t h e For  1961,  11 y e a r s  was c a l c u l a t e d  T h i s measure c o r r e s p o n d s  change  to  actual  example, the  annual  o r -0.91%  cover  from t h e  mean  the  percentage  16  of  the  p r e c e d i n g time p e r i o d ' s cover  following  time p e r i o d .  was  30% and i n  the  entire  1961  rate  change  cover  lost  measuring  earlier  lost  by  the  t h e mean c o v e r  in  1950  change  for  If,  for example,  the  annual r e l a t i v e  been  rate  of  t i m e p e r i o d w o u l d be -33.0% o r -3.0% p e r  Absolute percent  20%,  w h i c h has  provides  each y e a r ,  what  time p e r i o d i s  statistical  analyses  a  measure  whereas  the  percentage  of  lost  next  by t h e  were  of  the  relative  the  period on  species  cannot  mean a n n u a l  change  is  density  i n an  measured.  All  cover  performed  year.  MINITAB  (Ryan e t  a  al.  1985).  The A c a c i a  woodlands  Although aerial that  photographs at  the  Acac i a  individual  communities with  a  these  degree  Trump 1972,  Taiti  1973,  identification  analysis,  as  to  a  al.  was  not  were  dominated  species  1967,  Glover  and Trump  &  Lamprey  deemed  b,  from verify  other  1976a  in t o t a l  rainfall  (Norton-Griffiths  by  of  identified  ground r e c o r d s  analysis  Herlocker  decline  from 800mm p e r y e a r  divided  used as  variety  be  occurring  by to a  1970, 1985).  necessary  for  the  density  was  the  woody c o v e r  focus.  Due  was  scales,  for  (Lamprey et  Species  primary  small  chosen  complex  much l e s s e r  the  tree  into  a baseline  flight  line,  i n the  et  al.  i n t h e Mara R e s e r v e ,  east,  1200mm p e r y e a r  1975,  east  to  E p p and A g a t s i v a  and w e s t .  for a l l then  gradient  The 1950  analyses.  transects  ranging  i n the the  area  a e r i a l photos  were  They were  1980),  west  first  were drawn down t h e  assembled center  of  17  e a c h of of  the  parallel  sampling p o i n t s  were  then  density  flight  on s u c c e s s i v e  randomly  selected  was m e a s u r e d .  photographs measured.  of  1:50,000  all  of  Overseas  scale  This  later of  in  sample  the  mean c o v e r All All  in  estimate  sample p o i n t s woodland  before  was  re-  was g r e a t l y a i d e d by use  of  and  of  density  of  density  Kenya,  same a e r i a l were  photographs.  be  for  Tanzania  photographs. made  This  each  independent  Directorate  and t h e  at  of  one  e s t i m a t e s were a r c s i n e  each  provided a  sample  c o m p a r i s o n s were p e r f o r m e d ( Z a r  Croton areas  changes  cover  in  thickets of  was  therefore,  on a l l  measured  s e t s were  occur  thickets  density  was e s t i m a t e d  T h e s e two  cover  counts  and s t a n d a r d e r r o r  cover  measured t h e  area  identified  point. another.  transformed  1984).  thickets  Because  density  series  grid  were assumed t o  statistical  The C r o t o n  in  each  and c o v e r the  Survey  r a n d o m l y - o r i e n t e d dot  points  on  series points  woodlands  were  overlap  Fifty  t o p o g r a p h i c a l map c o m p i l e d by t h e  Surveys,  point  lines.  Acacia  S u r v e y and Mapping D i v i s i o n from t h e s e Five  guarded a g a i n s t  flight  The same p o i n t s  Identification  the  paths.  only  and t h e i r  within five  outlines  thickets  are  changes  individual  photographic  on t h e  photographed  thicket  in d i s c r e t e ,  at  clumped u n i t s ,  I  in a d d i t i o n  to  thickets. series.  1950  and 1982  the  largest  Cover Change  photographs. scale  were c l e a r l y d e l i n e a t e d  and,  and  easily  east  side.  compared. Croton Therefore,  to  select  f a r more a b u n d a n t  an e q u a l number of  on  the  sample p o i n t s  from  each  18  side,  sample  limited east  sizes  occurrence  and  west  sample  points  across  the  woodlands. thickets  that  together,  to  the  loss  as  groups  w h i c h were  a  1950  series.  an a r e a  times to  dot  reduce  directly  the  Area  fragmented.  by  the  considering to  select  in Croton cover  above  the  clearly  50  loss  the  of in  thicket  these square  was  U s i n g a Bausch  15 base  s a m p l e s were  photos.  from the  1950  in  Thicket  reand then area  base t r a c i n g s  and  grid.  test,  as  then m e a s u r e d ,  1982  thicket  A r e a measurements  inaccurate  individual  outlines  E a c h sample  the  15  Acac i a  photographed  the  variance.  of  for  by s e l e c t i n g  was  grid.  each  over  area dot  comparative  a digitizer.  increasingly  trend  T r a c i n g s were made of  r e m a i n e d was drawn d i r e c t l y  As  By  I was a b l e  described  determined  Transferscope,  with  west s i d e .  general  was  1982.  fit  reduced  Reserve.  using  five  Zoom  measured  using  thickets  estimated  on the  kilometers,  adjusted  side  Area  and  thickets  Lomb  on t h e  or t h i c k e t  1950  measured  thickets  and examine  was  been n e c e s s a r i l y  of  entire  Cover  both  would have  thickets  a r e a s were a l s o  using became  the  measured  digitizer  smaller  and  became more  19  Results  V e g e t a t i o n Changes from The  1880  Serengeti-Mara  ecosystem  d e s c r i b e d as M a a s a i l a n d . Maasai  who  system.  have  had  Maasai  pastoralists However,  (Sandford  the  lore  the  In  the  tribes,  significant  relates  latter  Fosbrooke  t h i s time t h a t the Maasai continued well  l i e s w i t h i n an a r e a b r o a d l y  into the early  1948,  part  these  Jacobs  1965). of  struck  t h e a r e a and  around which  domestic  succumbed t o  o f t h e d i s e a s e was  tremendous l o s s e s of l i v e s t o c k  ungulates  -  rinderpest  P l o w r i g h t a n d M c C u l l o c h 1967, A t a n g a n d  bovine  n o n - e x i s t e n t and  disease  -  (Thomson  an  exotic  (Plowright  Plowright  1963,  1969).  The  i n t r o d u c t o r y l o c u s of t h i s endemic A s i a n d i s e a s e remains  u n c e r t a i n but s e v e r a l t h e o r i e s e x i s t . introduced  moved r a p i d l y  Some s a y t h e d i s e a s e  i n t o E g y p t by t h e N i l e e x p e d i t i o n  south from t h e r e .  transmitted  been b r o u g h t  1889  not  century  setbacks  viral  that  1980).  I t was  W i t h i n a few y e a r s , t h e n e x t d i s a s t e r h i t  was  this  nomadic  19th  1885).  of  the  1900s.  Control  Maasai herdsmen s u f f e r e d  the  suffered a series  1880s, d r o u g h t  pleuro-pneumonia.  first  is  i n t h e a r e a do  of  s t o c k , a l r e a d y weakened by a r i d c o n d i t i o n s ,  actual  i t  i n f l u e n c e on  that  r e c o r d s of Maasai h i s t o r y  until  1919,  most  evidence  came t o t h e a r e a i n t h e m i d - 1 7 0 0 s ( M o l  detailed  a c t u a l l y begin  the q u a l i t a t i v e  Of a l l t h e l o c a l  tribal  first  t o 1980;  by  cattle  i n t o Khartoum.  i n 1884  Others suggest that  held  i t entered through E t h i o p i a during the I t a l i a n ( B r a n a g a n and Hammond 1965, Mack 1970,  and  rinderpest  f r o m t h e B l a c k Sea a r e a , w h i c h The most commonly  was  belief  had is  i n v a s i o n of  F o r d 1971,  Plowright  20  1982).  According  introductory this  foreign  September within  (Sandford  their  total  only  Mallett  at  extinction.  famine  in the  on l i v e s t o c k .  now the  the  Serengeti  where  had become  beggars  scarcely  for  enough  as  a  to  Maasai  as,  Arab  of  the  his  cries  of  characteristic  hide,  would not  their  eat.".  Bernsten  brink  barrier  a pitiful of  o l d men. even  nature,  Maasailand. eyes  more  like  c r a w l on  all  These people  ate  the  horns  They had f l e d the  of  through  from whose h o l l o w  refuse.  fellow  traders  trek  c h i l d r e n who were  yes,  the  self-reliant  slave  on  the  in  themselves around the  languishing  to of  enkidaaroto,  " w a r r i o r s " who c o u l d s c a r c e l y  the  loss  travelling  people  uttering  due  1948).  recalled  gathered  the  Fosbrooke  of  hunger had d e p o p u l a t e d to  dead  1919,  the  they  were  once p r o u d and  intrusion  hunger,  dazed,  beast  cattle  and  human  of  dead a s s e s ,  By  devastating  skeletons,  were  numbers.  to  for  reduced to  fours  slaughtered  the  (1894a)  meantime  t h a n men,  anything:  the  earlier,  naked f r o g s and  (Sandford  distinguishing  madness  immunity  The c a t a s t r o p h i c  conditions  people  women had been  gleamed  the  inevitable  described  years  of  The  of  Europeans  time  the  Serengeti-Mara region  (Baumann 1894a & b ) .  "The  is  great  the Maasai  Maasailand,  which  was  the  Baumann  o u r camp,  in  as  1923).  who had r e s i s t e d few  90%  served  c a r r y i n g no  known amongst  this  a  over  i n c l u d i n g smallpox  destruction  Maasai,  months  became  cattle,  had a r r i v e d i n t h e  dependence  1890s  cattle  d i e d r a p i d l y and  created  epidemics,  area  short  ensuing  livestock  The  it  sources,  African  disease,  1919,  The  all  hosts.  1890  a few  to  entire  of  from  district  a the and  c o u n t r y m e n who t h e m s e l v e s had (1979)  recounts  how  Maasai  21  tried  to  join  eked  out  Baumann  Baumann's caravan  a  Dorobo-like  (1894b)  Serengeti-Mara area  was l e f t  century  and  Wild rinderpest  for  many y e a r s  outbreak.  As  cattle  numerous  caffer)  and  gregarious, was  quick  Although  the  actual  buffalo  in  by  (1914)  East  to  highlights  of  1911  lost  a loss  of  a  recruitment  of  age.  The d i s e a s e t o o k  During area  was  pressure  young a n i m a l s  Hence,  it  this also  was  its  up t o  to  the  after  as  a  lively  in  animals However,  must  have  been  calves,  who  six  yearling  Heavy  trade,  months  disease".  population  ivory  herds  Serengeti-  wild  about  size.  of  major  disease.  "the  them.  the  northern  at  viral  White  of  on  the  fact,  the  toll  to  a few  one  elephant  (Syncerus  accounts  as  weaning, to  giraffe  amongst  In  many  the  virus  the  only  populations  experiencing a reduction by  90%.  the  greatest  the  were  to  by  Due  unknown,  that  became r e f e r r e d  same p e r i o d ,  spread  show  into  immunity  generated  the  the  buffalo  species,  there  safari  immunity  of  the  aethiopicus),  1979).  buffalo  hunting  affected  1977,  are  show t h a t  developed  acquired  the  of  including  s p e c i e s were  and  gradually  their  turn  (Phacochoerus  hold  records  lost  species  catch  Veterinary  low.  in Thus,  dwindled,  these  Mara a r e a .  very  1890s. the  others  existence.  people  also  of  shooting  his  by  nature  Africa  described  early  numbers  warthog  numbers  1900 a n d s u g g e s t  the  were  (Sinclair  herd-forming  of  the  area  affected  wildebeest  disease  by  wildlife  camelopardalis), Two h e a v i l y  while  thereafter. the  in  alive,  hunter-gatherer)  influence  of  and o t h e r s .  left  human  stay  75%  died  ungulates  persisted (Giraffa  that  had  d e v o i d of  to  (nomadic  estimated region  just  of  the  hunting and  the  22  flourishing b a c k s of  slave  slaves,  (Wakefield of  trade helped  to  1882,  the  coast  1870,  themselves  as  "the  these very elephant Baumann  (1894,  a means  t o me t h a t  they  nine  cattle  figures first  of  is  international  ivory  trade  destitution.".  (1973)  the  long. is  were  now  one.".  years  as  p.67)  wrote,  every  kind  swarmed i n p l a c e s T h i s absence  their  of  as  elephants  I  i n the  the  among i t s  elephants, course  Serengeti  of  is  that  an  where  and  these for  the  in  the  do so by by  their  Spinage elephant  area  without  "...ten  years  now you w i l l  interpreted area.  to  stated  of  widespread  was  in  not  later  Buxton  (1927,  we c o u l d have  shot  nearly  our w a n d e r i n g  [through  does not  p.176) and has  stated, never  the  "...so numbered  inhabitants.".  of  the  reduced  by  hunting,  wild  the  Maasai  people  and  whom  long,  involved  known as  (1938,  turn  rinderpest/  of  this,  men  feet  forced to  elephants  a r e a ] " and M e l l a n d  know,  By t h e greatly  game  6-7  reported that,  like  to  Some of of  accuracy  h a v i n g o c c u r r e d i n the  "...excepting  Serengeti-Mara far  never  of  cases,  been  reports  Grogan and S h a r p (1900)  journals  referred  records compiled  elephants.  the  abroad  1968).  important  never  The S e r e n g e t i - M a r a became  find  The  had  these  the  Maasai  In a few  slaughter.  ago e l e p h a n t  on  They a p p a r e n t l y t u r n e d  for a tusk  Ivory export  substantiate  itinerant  alive.  What  who  to  (Fosbrooke  "...  feet  questionable.  t i m e many M a a s a i  sheer  were  30 c a t t l e  2-4  According  people"  staying  for a tusk  1882).  wrote,  received  ivory,  l o c a l Dorobo h u n t e r s  elephant  p.323)  h u n t i n g as  the  hunters  transport this  f o r m a r k e t i n g and s h i p m e n t  Farler  e a r l y caravan leaders,  to  century  elephant and by  numbers  domestic famine  had  been  ungulates and  by  disease.  23  Explorers  and  Serengeti-Mara expanses Buxton  of  hunters  grasslands,  the  early  w h i c h l o o k e d much as grassland  (1927,  of  p.67)  and l i g h t l y  remarked,  lovely  it  came  Near us was  a  running  and  (presumably  south  every  other  direction  some  areas,  these  variety  of  assigned River, the  grasslands  determine  gave s i m i l a r  of  northern 1913, few  yellow-barked  a hunter  described  scattered  rolling  the  thorn  country.  Croton  ]  thickets  march a r o u n d t h e m . " . near  the  meanders with  are  o r 40  River,  We c a n  m.p.h.".  land,  with  nowhere  in  level,  the  the  p.57)  early  1910s  with  White  mostly  in  The p l a i n  run our c a r s  is  rocky,  and one  grassy circular  can  [  always  the  area  "...the  river  plain,  s p r i n k l e d with at  to  with a  The  described  anywhere  the  from 1910  thickets...  smooth  the  (1915,  open,  where  in  was a b r o a d  studded  period  a  Mara  river,  "It  low,  green  In  warden  extensively  continuous,  (1927, the  along  "...an  are  in  on.".  game  And,  the  (p.162)  there  Eastman  of  Reserve.  trees."  but  " . . . a high grass plateau  and  and  rounded h i l l s .  size...  side  travelled  as  crowned  through a v a s t ,  low,  moderate 30  Sand  trees"  a  west  who  hills  w i t h A c a c i a s of  the  grazing  of  and  p.275),  and Mara t h r o u g h o u t  Here  outcropping h i l l o c k s  dotted  present  almost  Kuka H i l l s ) ,  distributions  fine  same a r e a  range  on  open  undulating  seemed  fly  Acacia  and w r i t e r  Serengeti  of  the  park-like country,  occasional p.158),  of  the  (1913,  tsetse  accounts  wooded  a  predominated.  onto  which  stretching were  Woosnam  Mara T r i a n g l e a r e a  plain  to  were p l a i n s  species. to  thickly  Broad,  savanna out  country  uninhabited... north  encountered  does t o d a y .  wooded  "...we  fertile  1900s  any  bordered trees speed  of up  24  By  the  1930s t h e  years  earlier  Lewis  (1935,  belts  of  River,  a r e a had c h a n g e d .  had been  open,  p.445)  as  it  was  necessary  the  (1976)  Maasai]  and M o l ( 1 9 8 0 ) ,  species  of  records 1930s, Air  Acac i a  hunters  leopards area  In  which  unlikely  is  habitat  Colonial  several  a name  Syd  i n the in  northern  Downey r e c a l l e d  'taking'  Tanganyika Serengeti,  i n what was confident  National  Mara status.  perceived that  any c h a n g e s  they  area,  Royal  hilltop  Serengeti up t o  five  Triangle,  an  open g r a s s l a n d s ,  protection  followed  inherent of  reserve.  Just  suit,  and  The g o a l  was  preserve  as  "pristine"  its  had s u c c e e d e d  (p.55),  and even  the  which c o m p r i s e d the  to  that,  today  it  the  an  area.  over  this  stated,  the  of In  m a j o r i t y of a  decade gained  wooded  condition.  "The t s e t s e is  value  Serengeti  in securing  o c c u r r e d , Park a u t h o r i t i e s  Parks Report  this  for  was d e c l a r e d a game  Kenya's park  wide,  the  leopards.  the  national  its  a d m i n i s t r a t o r s , who r e a l i z e d t h e  ecosystem,  first  the  for  other  in  showed w i d e s p r e a d  Mara  known  photographic  and t h e  the  pass  Huntingford  and Osa J o h n s o n  throughout  d e n s e bush of  known  Early  1940s,  1950,  species  for  Mara a r e a .  s u c h as M a r t i n  were q u i c k t o move  later  over  used  is  Mara  to  this  region,  1937  a path for v e h i c l e s  to  i n the  for  the  According  Syd Downey  now  reach  '01-jerai'.".  fact,  i n a day  to  by  and narrow  and a n o t h e r  found i n the  like  described  seyal  and Acac i a woodlands  Mara.  out  was  intersecting  In o r d e r  cut  Force a e r i a l photography  and  the  to  ].  Acac i a  from n a t u r a l i s t s  thickets  the  as  plains  "...numerous  thorn-brush [ Croton  through densely-growing [to  grassy  The same a r e a w h i c h 30  area  Feeling  area  before  in a  1946  Royal  fly  stands  guard  v i r t u a l l y a glimpse  into  25  Africa  as  it  was b e f o r e  These  early  years  earlier,  of  open  these  and woodlands  assemblage of sites  pallidipes, 1955,  Langri'dge  vertebrate  for  tsetse  (Lewis  e_t a l .  1970, tsetse  known  as  trypanosomiasis  man.  Because the hosts,  domestic,  and most  and the  by  that,  the  Reserve cattle  the for these  "The  of  than  50  expanses  prime  and  blood  late  Narok  by  in  District  watering  (and  it  many  human, (Ford The  organisms  wrote  possibly  holes,  in  1985).  [trypanosomiasis]  toads  in  its  other  a  disease  1800s  Rogers and Randolph persistence  1970,  sickness"  l o s s e s of the  1944,  inflicting  c a n be m a i n t a i n e d  in  and  Trump  to  a  shade  Beaumont  "sleeping  the  for  in  a  1925  was  known  lung  fish)  would  poison  blood.". 1950's and  the  smaller A  by  stock 1955  this  Maasai  western  seasonal grazing  of  1936,  a parasitic  hosts  disease  at  habitats  swynnertoni  addition  and  for  be c a r r i e d  central  whole  In  survived  its  strange  the  trypanosomiasis. that,  shores.".  less  provided  Glover  parasite  1982,  by c a t t l e  its  Throughout currently  aware  to  trampled with  cattle  wildlife  Molyneux  "...the  1968,  vector its  afforded  also  transmit  Commissioner  Maasai  water  of  keenly  District  a n d when the  that  Glossina  1971).  flies  the  1978,  were  report  its  had been wide  Sywnnerton  Trypanasoma  secondary  Bourn  Ford  in  which  fauna  F o r d and C l i f f o r d  bite,  Maasai  crossed  realized  flies,  1935,  painful  1971,  had not  same d e n s e w o o d l a n d s  thickets  resting  Buxton  man e v e r  grassland.  diverse  G.  white  conservationists  These  and  the  portion their  were  Kenya  section  used the  sheep and largely  National is  of  dominated  the  area  that  Mara  goats.  Game Unlike  unaffected  Parks report by  the  is  tsetse  by stated fly  26  and p o s s i b l y is  of  for  absolutely  small  stock...".  north  of  the  that  the  reason  no u s e  to  the  As l a t e  as  1960,  remained  Whereas  earlier  inhospitable  to  wild  rinderpest  in  ungulates  (Taylor  hosts  1979).  Early  attempts  if  Despite  of any,  programs tsetse  of  were  to to  at  time,  be  tsetse  several money, in  response  ecological  factors  It their  Aitong  (well  and  south  to  1930s  vegetation  was  and  1940s  the  Also,  many  developed  immunity  Stewart  1968)  trypanosomiasis  vector  and  to now  (Sinclair  bush-clearing  and  the  were  largely  unsuccessful  having  fly  numbers  (Langridge  1960).  to  eliminate  decades.  1960s  this  lead  flies  after  tsetse  (Ford  official which  the  Finally,  effort,  around  to  game.  grazing  proliferation.  1967,  and  the  of  1985). the  time,  efforts  began d e c l i n i n g in  the  mechanical  failures, for  by  for  River,  century  this  hosts  on  abandoned  flies  believed rather  impact  continued  expenditure  by  the  wildlife  repeated  Maasailand  for  Talek  their  and and Watson  prime  elimination  had,  quota  a r e a between  (Lamprey  occupation,  became  little,  the  the  had a l l o w e d  full  Maasai except  uninhabited  tsetse  increase  resident  its  M a r a Game R e s e r v e ) , t h e  Serengeti  woodland  retains  Although  same t i m e ,  to  great  eradication  1971).  control  from  it  was  programs  a reduction  in  not but  their  habitats. Two o f in  fire  heavy  Gwynne Tanzania  important  frequency  Unusually in  the  high  ecological  and  rainfall  an in  grass production 1961,  Kenya  National  increase  the  early  the  region  Report  the  elephant (Figure  Department  Annual  in  were  1960s  throughout  Game  Parks  factors  densities.  2.1)  resulted  (Glover  Annual Reports 1962-1964,  increase  and  1961-1964, Talbot  and  1100 n  600-  —I  1920  1  1930  1  1940  1  1950  1  1960  1  1970  YEAR  F i g u r e 2.1.  F i v e - y e a r r a i n f a l l a v e r a g e s f o r Narok, Kenya over the p e r i o d 1914 - 1980.  1  1980  28  Talbot  1963,  Glover  swards above fires  the  which  ungulate  and Trump 1 9 7 0 ) .  bonnet  of  prevailed  populations  his  Turner  recalled  L a n d R o v e r and t h e  during  were  M.  this  unable  to  time.  fiercely  The g r e a t l y  remove t h e s e  grass hot  reduced  tremendous  fuel  described  this  loads. Langridge phenomenon rainfall  in  was  luxuriant  they  after  is  t r e e s and  of  occurring  elephant  ecosystem.  the  and  district  and  produced  the  the  times  The v e r y  vegetation  bi  such that  1963  and t r i - a n n u a l  per y e a r ) .  plains  burning within  are  As  fired  soon by  new g r o w t h of  the  same  fires  season.  caused  w i t h marked c h a n g e s  the  grass Thus  w i t h i n a s e a s o n o r even fierce  three  accelerated  i n the  numbers  shrubs...". the  late  1950s  and t h e  Serengeti-Mara noted  and  subsequent  1970),  It  was  the  areas  This  Lambwe V a l l e y  communication).  to  the  These  woody  not  altogether  immediately  Isuria)  on  Watson and  It  is  researchers  numbers  were  vegetation  Bell  most  1969,  clear  was  Glover  where  these  that  they  likely  s u r r o u n d i n g the S e r e n g e t i - M a r a such  as  Loliondo,  E s c a r p m e n t , the  n o r t h and west areas  1960s,  elephant  1967,  included areas  O l o o l o l o or  early  that  browsing  immigrants o r i g i n a t e d .  from  (also  for  1961  grassy  are  208)  "...between  allowed  the  conditions  (Lamprey e_t a l .  Trump  and  and t h r e e  enough,  ready  the  i n the  increasing  came  twice  a year...  Between  and  which  a r e a may be b u r n e d t w i c e  during  working  throughout  cover  dry  a fire  p.204  Mara Game R e s e r v e :  (meaning  destruction of  (1970,  v e r y heavy  Sometimes  a single times  the  are  Maasai.  al.  grass  burning... as  et  were  the  Siria  C h e p a l u n g a F o r e s t and  (M.  Turner,  experiencing  personal  rapid  human  29  settlement,  increasing  cultivation, This in  the  (1955)  and t h e r e f o r e  led  years  to  number  of  of  damaging  drive  these  a  certain  These p l a c e s are exists". p.31) the  This  these  results portion  in  the  that  Serengeti  increasing and B e l l  "...  of of  within  our  1969,  Viewing  the  Glover  woodlands pressure  was  on a v a r i e t y et  An  al.  analysis  Griffiths  (1979)  Serengeti  could  Trump  browsing  elephants.  and  elephant  histories  of  two  Burn  species  the  press),  species  1974a &  fires  the  figured  1960s  the  to  shamba  old  in  densities  lands  (Watson 1979,  change  heavy  followed  in  the  browsing  (Glover  1965,  b).  woodland  data  cover  primarily secondly  to to  by  loss the the  scars  prominently  in  unpublished  Nortonin  the  increased impacts  found  Mara suggested t h a t  (Dublin,  common  Norton-Griffiths  monitoring  and  haunts.  (1962-1964,  elephant  and  to  cultivation  became more  subsequent  a  whole,  has been  no  and r e s e r v e  of  [farm]  on t h e  their  where  for  shooting  attack  and b a r k - s t r i p p i n g in  problems  been,  back  1970,  that  of  browsing  during  park  attributed  severity  Narok  With  long-term  be  and  cores  tree  showed  C o m m i s s i o n e r of  against  elephant  out. areas  Ring-barking  Croze  of  frequency  tree  of  1967,  of  cases,  protected  Parks Annual Report  protected  inevitable.  some  serious  Reserve  in  into  persistent  Park".  and  forced  has  elephants  and D o u g l a s - H a m i l t o n ,  Lamprey  areas  Maasai  National  being  necessitated  The c a m p a i g n  The T a n z a n i a N a t i o n a l  noted  Dublin  in  in  elephants  " Elephants created  elephants.  The  were  and,  The D i s t r i c t  areas.  elephants  satisfactory.  elephants  followed.  that,  these  numbers  a c o m p r e s s i o n of  that  reported  settlement  livestock  their data).  both  in  of the fire  individual  30  Vegetation  Changes from  The A c a c i a  woodlands  Results show  that  Acac i a  of  the  in  woodland  stands  (Figure  Fifty  sample p o i n t s  since  56.8%  sides.  the  cover  way  showed  year  P<.01).  but  were  1% l e v e l .  This  same f o r  the  side  west  fires  subsequently  after  1967. The  and west  hypotheses in  different (Figures rates east  of  absolute time  2.3  and 2 . 4 ) .  change  for  the  significantly losses  periods  were  not  greater  ranged  estimates the  west.  on b o t h  the  east  time  and  a  woodland  significant  s i d e s of  and west in  the the  more  there  west  trees  were  relative were  also  1950 losses  significantly  to  of  grass in  the  1974 a n d higher  absolute  rate  1961.  From  on  between  dry  season  the  east  between  change  two-way tests  1967 west. east  east  for  the  ANOVAS  on  was h i g h e r  the  1967  rainfall  than  comparison  cover  the  west  using  different  3.91,  1982 a t  for  tested  two-  and  fuel  of  in  A  (F =  1961,  rates  occurred  highly  showed t h a t  no d i f f e r e n c e s  Tukey m u l t i p l e the  in  loss  interaction  Mara  1950,  result more  the  test in  in  50.4% i n  continual  comparison  showed t h a t period  Cover 10.7% t o  was  produced  or  30%  period.  east  that  Reserve  was a p p r o x i m a t e l y  through  and west  killing  Mara  the  each successive time  likely  which  for  showed  lower  most  evidence  west  multiple  is  and  east  there  east  was s i g n i f i c a n t l y  cover  suffered  Both  direct  analysis  followed  that  and the  was t h e  east  losses in  Tukey's  density  the  1950s.  significant  between  in  the  2.2).  The M a r a has  early  ANOVA  density  mean woody  8%  west  cover  1982;  1950,  from  and  to  1950 t o  to  mean in  the  1974, Absolute  and  west  31  30  -I  (H  1  1950  1  1960  1  1970  ~  ~  i  "  1980  YEAR  e 2.2.  Mean c o v e r d e n s i t y of A c a c i a woodlands i n t h e Masai Mara Game R e s e r v e . P e r c e n t cover d e n s i t i e s a r e d e r i v e d from d o t - g r i d a n a l y s e s of f i v e s e t s of a e r i a l photographs.  32  3.0  n.s.  [] -  EAST  g  WEST  -  CO CO  o  2.0-  o  % w  1.0-  **  3 O  CO PQ <  0  1  n.s.  1950  1961  1967  1974  to  to  to  to  1961  1967  1974  1982  TIME PERIODS  F i g u r e 2.3.  Absolute r a t e (percent/year) of cover l o s s f o r A c a c i a woodlands i n t h e Mara Mara Game R e s e r v e . The Reserve was broken i n t o e a s t and west s i d e s f o r a n a l y s i s , (* = P<.05, ** = P<.01),  33  12.0  n.s.  11.0  IS  10.0  GO  9.0  o  8.0  o  ***  7.0 6.0 5.0  >  % w  4.0 3.0 2.0 1.0  0  ***  Ii 1967 to 1974  1950 1961 TIME PERIODS to to 1961 1967  F i g u r e 2.4.  1 1974 to 1982  R e l a t i v e r a t e ( p e r c e n t / y e a r ) of c o v e r l o s s f o r A c a c i a woodlands i n t h e Mara Mara Game Reserve. The Reserve was broken i n t o e a s t and west s i d e s f o r a n a l y s i s , (* = P<.05, *** «= P<.001).  34  for of  the  periods  change  to  were h i g h e s t  Relative side  1961  rates  losses  of  1967  to  change  followed  similar  west e x c e e d e d  those  and  1974  and e a s t  to  1967.  Looking at  Reserve, to  i n the  65% d e c l i n e  the of  first the  thickets  1950,  past  individual  32  to  1961,  periods  rates  1967  to  equal  from  of  across  the  to  change,  1967  alone,  In t h e  and t h e  there  last  in  in 1974  1961 entire  was a 60%  32 y e a r s , 1982,  i n two ways o v e r  average  the  followed  km  km  2.80  the  between o v e r 95%  of  thicket some  thinning  (i.  2.6,  Croton  time,  of  Both a  total The  area  analyses  significant  &  2.7,  over  all  P-values  1961-1967  (-2.33%  relative thickets  loss). or  The  thicket  d r o p p e d from a mean o f  4.12  v a r i a b i l i t y was e s p e c i a l l y  high  2.8).  area-loss  thickets e.  highly  and -3.12% p e r y e a r  through  time.  taken p l a c e .  o c c u r r e d from  15 i n d i v i d u a l  (Figure  Croton  have  2.5,  s h o u l d be m e n t i o n e d t h a t  perimeters internal  2  declines  loss  groupings,  It  (Figures  The l a r g e s t  a r e a of  thickets  and a l o s s  l o s s e s have been  years  year absolute  to  East  s i d e s were  density.  have c h a n g e d  both these  <.0001).  in  1961  cover in  rates  trend.  1967.  1950  the  rates  thickets  by  show t h a t  2  for  and west  p e r i o d from  from  i n canopy cover w i t h i n t h i c k e t s  covered  per  east  a  Absolute  A c a c i a woodlands had v a n i s h e d .  Croton  the  i n the  aerial series  The C r o t o n  loss  t h a n west  relative  in tree  1982.  p e r i o d 1961  were g r e a t e r  1982,  to  i n the  the  to  and 1974  loss  did of  measurements. not  cover  change  density)  at  The  actual  all,  though  occurred in  all  35  80 - i  —I  1950  1  :  1960  1  1970  1  1980  YEAR  F i g u r e 2.5.  Mean c o v e r d e n s i t y of C r o t o n t h i c k e t s i n the Masai Mara Game R e s e r v e . P e r c e n t c o v e r d e n s i t i e s a r e d e r i v e d from d o t - g r i d a n a l y s e s of f i v e s e t s of a e r i a l p h o t o g r a p h s .  1950 to 1961  1961 to 1967  1967 to 1974  1974 to 1982  TIME PERIODS  F i g u r e 2.6.  A b s o l u t e r a t e ( p e r c e n t / y e a r ) of cover l o s s f o r C r o t o n t h i c k e t s i n t h e Masai Mara Game Reserve. A l l time p e r i o d s a r e h i g h l y s i g n i f i c a n t a t t h e P<.0001 l e v e l .  1950 to 1961  1961 to 1967  1967 to 1974  1974 to 1982  TIME PERIODS  F i g u r e 2.7.  R e l a t i v e r a t e ( p e r c e n t / y e a r ) of c o v e r l o s s f o r C r o t o n t h i c k e t s i n t h e Masai Mara Game R e s e r v e . A l l time p e r i o d s a r e h i g h l y s i g n i f i c a n t a t t h e P<.0001 l e v e l .  38  Paired t =  -5.98  p-value < 0.00001  4H C\}  3H  < <  2-  0 1950  1982  YEAR  F i g u r e 2.8.  Mean a r e a of f i f t e e n C r o t o n t h i c k e t s sampled i n t h e 1950 and 1982 a e r i a l photography (± one s . d . ) .  39  t h i c k e t s sampled. u n c l e a r b u t may thickets.  The  reason  for this s t a b i l i t y  h a v e t o do w i t h t h e o r i g i n a l  Thickets  or  had  bottlenecks  fragmented  and  the  over  The  time.  probably  within  subsequent  agents  acted both  shape of  time  together  were  more  and  for  area  and  1950 and  readily  segments d i s a p p e a r e d  responsible  in  T h o s e t h a t were l o n g them  is  individual  w h i c h were r e c t a n g u l a r o r c i r c u l a r  t e n d e d t o be more r e s i s t a n t t o c h a n g e . narrow  through  entirely,  cover  loss  individually.  Discussion A  synthesis  the Serengeti-Mara grassland  to  years.  1900  By  hunting, due  e c o s y s t e m has  dense  woodland  elephant  wildebeest,  and  stage  thickets  disease.  for  the  and  evidence  a change from  been  and  greatly  annual  formation  hot  c a t t l e p o p u l a t i o n s were  low  had  These e v e n t s  i n the  late of  been  extensive  f o r m e r l y been open  progressed  which  30  years  dense  woodlands  set  by  unimpeded.  earlier and  had  g a z e t t e d as a n a t i o n a l r e s e r v e . authorities area  vowed  to maintain  i n what t h e y m i s t a k e n l y  the By  decimated  by  l a s t century  set  woodlands  In  and  grasslands.  Maasai, the  and  grazing  b u s h and  mid-l930s,  been open g r a s s l a n d now  thickets.  a hundred by  the Maasai people  fires  open  reduced  S h i e l d e d from the e f f e c t s of heavy b r o w s i n g the  shows t h a t  back i n l e s s t h a n  numbers had  establishment  i n what had  direct  experienced  buffalo,  t o r i n d e r p e s t , and  famine the  o f q u a l i t a t i v e and  1937  the  and  woodland the  area  comprised  Serengeti  was  E a r l y c o n s e r v a t i o n i s t s and  park  t h i s d i v e r s e , woodland  v i e w e d as  dominated  i t s "pristine" condition.  40  These despite in  the  woodlands and t h i c k e t s  the late  medicines  desire  to  1950s and e a r l y which  protected  advanced v e t e r i n a r y  and other  and c a t t l e  populations  Serengeti-Mara  increases  which  return  the  to  Two  It  was f i r e .  of  rinderpest,  standing  crop  rainfall fires set  in  the  the  neighbouring overall  fire  wandering Under  early  Maasai  subsequent served  to  about  honey  some  high high  normal  clear  this  rainfall fuel the  of  other were  a r e a of  human the  Kurji  effects  1976,  of  increase  woodland still  these and  a  the  human  declines.  The  from  were  grazing pastures  set  woodland  increase  conditions these  intentionally and to  of  authorities  by E u r o p e a n  in  the  fires  bush and a t t r a c t  fire  were game  by  under  lit  by  hunters.  regeneration  early  clear  hunting  inadvertently  rainfall,  high  progressed,  facilitated by p a r k  the  unusually  season  fires  burns  result  significantly  dry  these  in  sparse as a  reduce  others  production  against  declines  from the  and s t i l l  sudden  with  today.  resulting As  modern  coupled  1966,  indirect  s c h e m e s , some w e r e  hunters of  began  surrounding  woodland  to  improve  bush, while  management  declines  increases in  areas  populations,  Some o f  to  b e e n h a m p e r e d by the  and  initiated  1960s.  tribes,  conditions  Under  in  stemming  grass  became w i d e s p r e a d . by  direct  unable  dry  but,  livestock  Morgan and S h a f f e r  have  were  time  advent  unprecedented  brought  Ungulate  of  tsetse-infested  have  2.9,  pathways to  the  vaccinated  open g r a s s l a n d s seen  appear  some  precipitate  With  had o c c u r r e d  finally  first the  which  was t h e  primary  populations  1960s.  diseases,  (Figure  1985).  them,  for  a n d e x t e n d e d human l i f e ,  care,  rinderpest  Lamprey  preserve  persisted  would  occurrence. 1960s and  the  devastating  and  to  the  lush  41  8-  or  PLH  O  6-  4-  Oh  2-  —I  1960  1  1970  1  1980  I  1990  YEAR  F i g u r e 2.9. Human p o p u l a t i o n d e n s i t y t r e n d s f o r t h e a r e a near the Masai Mara Game R e s e r v e . The d o t t e d line ( ) represents the trend calculated from s e t t l e m e n t numbers i n t h e a r e a a d j a c e n t t o t h e Game Reserve (Lamprey 1985). The dashed l i n e ( ) represents the trend c a l c u l a t e d from n a t i o n a l c e n s u s d a t a which included the densely populated Ololunga area. (Morgan and S h a f f e r 1966). T h i s a r e a i s not a d j a c e n t t o t h e R e s e r v e . Graph adapted from Lamprey (1985).  42  grazing  "lawns"  T h e human loss of  in  created  fires.  increase also  protected  elephants  notoriety  by  areas,  in  opened a second pathway  the  c o m p r e s s i o n of  dramatic  and a t t e n t i o n  woodland  throughout  Africa  1962,  Simon  1962,  1966,  Field  1971, Anderson and Walker  et  al.  Glover  1981).  Curry-Lindahl no  (Buechner  other  In  (1968,  animal  drastically Africa,  the  increased  elephant  the  the  period  major  occurred et  two  longer,  in  died.  The  could  have  existing trees  the  of  exact to  timing  Buss  and  time, Buss  Savidge  elephants  and woodland  loss,  "...except  for  is  alter  able  Like  Caughley  to  man  there  felt  is  a habitat  s o many o t h e r  high  of  per  the  so  areas  of  effects  of  impacts,  such  of  as branch  or  them more  tree  this  effects  of  to  highest  in  even  and bark  fire,  fire.  by  a  (Langridge  withstood  elephants, density  to  which  unbrowsed  by c a u s i n g p h y s i o l o g i c a l vulnerable  better  fires  time  may h a v e  breakage  a  significantly  e v e n many of  which  The  subsequent  trees  influx  provided  losses.  during  fires  declines  loads  these  year  these  the  photography,  decline,  and  healthy  such hot  exacerbated  of  rainfall  times  aerial  corresponded  drove  and making  same  Croze  of  heavy  fuel  much  1961, Brooks and  1974,  woodland  1967,  While  face  had g a i n e d this  1964,  woodlands  additional  period,  Elephant  period  three  1970).  role  1976,  elephant".  analysis  1961  or  The  densities.  the  unusually  al.  time  of  from of  that  Serengeti-Mara  understanding losses,  stated,  Africa  The q u a n t i t a t i v e covered  Darling  discussing  in  as does  1963,  p.26)  at  and Dawkins  woodland  elephants.  declines  among e c o l o g i s t s ,  to  fire trees  this  same  further. stripping, adding  stress  to  Elephants  to the and  43  fires  acted  together,  easily  distinguished.  Separate Reserve  one  all  other or  greater  the  reverse  time  period  that  part  of  prefer  the  The g r e a t e r  the  1960,  Glover  tsetse  therefore,  elephants  a n d may h a v e  the  rate  earlier  would  would  losses  also  1974  have  the  early  from  woodland  Elephants these  are  Laws  1969,  1970,  cover Croton  tend  Laws  first  elephant  et  because  woodlands  colonized  to  available  humans  The e a s t e r n  to  Report,  woodlands,  of  1982  elephants  when  been  to  emigration  devoid  on  experienced  their  throughout.  the  and by  depredation  years. frequency  of  from  1967  east-west be e x p e c t e d  rainfall  between  in  greater  of  density  of  Acacia  1968,  Mara  influx  experienced greater  fire  loss  of  infestations.  Greater  strong  with  were  tree  east  losses  not  pattern  Commissioner, Annual  an a r e a  Agnew  the  1974 a n d  eastern  the  relative  the  to  and b u s h l a n d s  1963,  fly  1967  in  greatest  1961,  are  of  different  following  west,  These woodlands  high  of  a larger  District  woodlands  thickets,  in  to  was a l w a y s  to  differences  to  from  Reserve,  east  sides  showed g r e a t e r  1950  and  (Narok  west  period  and Combretum woodlands  1975).  of  areas  dense  (Darling  trees  the  relative  thickets,  From  may be a t t r i b u t e d  The  density  al.  other.  contributions  significantly  periods  was t r u e .  neighbouring 1955).  the  was n o t  of  and  interesting  time  losses  relative  east  Although  1961-1967,  side  the  some  loss.  decline, two,  a n a l y s e s of  uncovered  woodland  their  and the  to  1982,  rainfall to  and  intensity on t h e  gradient,  be p o s i t i v e l y  subsequent  in  may e x p l a i n  west  side.  normal  correlated  biomass production  Due  years to  the  the  fire  higher to  the impact  amount  of  (Norton-Griffiths  44  1979).  In  the  period  declines occurred  from  1961  t o 1967, when t h e g r e a t e s t  and t h e l e a s t d i f f e r e n c e s  e a s t and west, u n u s u a l l y h i g h r a i n f a l l This  high r a i n f a l l  swamped.  have a f f e c t e d both Trends Acacia  in  area  fire  the  separating  these  germination in light  under  other  over  the  in  past  both  thickets  trees  or  data).  l i m i t e d and area  could  east  and  peripheries,  of the  Although west, on  some  d i d not  average  T h e r e i s g e n e r a l l y an  from  open  the  abrupt  o r no e c o t o n e  grasslands.  Croton  growing  grass  sites  fires  and  and  regeneration  losses generally area  the  Lamprey  (1985).  Thicket they  loss  boundaries  are  t r a m p l i n g and  fire  Dublin,  i n c i d e n c e and  thickets  found  in  the  are contrary to the  findings  and a r e a  of  shade  is  very  result.  losses  sampled i n t h i s study  Reserve  the  because  Under c o n d i t i o n s o f h i g h  His  in  ( N o r t o n - G r i f f i t h s 1979,  thickets  p a t t e r n of cover  rarely  shrubs (personal o b s e r v a t i o n ) .  to  significant  the  years.  gaps w i t h i n t h e t h i c k e t s but  h e r b i v o r e use, growth  within  fire  those  types with l i t t l e  g r a z i n g e f f e c t s of h e r b i v o r e s  The  fuel,  gradient  i s most s u c c e s s f u l on t h e edge o f t h i c k e t s a n d l e s s  subjected  unpublished  throughout  rainfall  paralleled  30  the  vegetation  edges, however, a r e h i g h r i s k frequently  normal  thickets  l o s s e s h a v e been s i g n i f i c a n t . between  so  Croton  effects to their  transition  throughout the area.  W i t h abundant d r y g r a s s  thickets,  experience  the  between  t h e e a s t and west e q u a l l y .  woodlands,  individual  fell  seen  l e d t o such h i g h grass p r o d u c t i o n  the Reserve that the e f f e c t s of were p r o b a b l y  were  d i d not found  in  in this  reflect the  study.  m a j o r i t y of findings  of  the general  Croton  thickets  P r e s u m a b l y , he  45  found  no c h a n g e s  location outside areas  outside the  heavy  the  by  of  used  limited  this  he  not  the  domestic  limited  of  due  to  to  their  same  reduced the  Areas  pressures  little fuel  power  points  loads  latter  to  for  his  as  influence after  due  to  a v o i d human c o n t a c t .  sampling  his  the  and  of  boundaries.  may h a v e h a d  stock  actively  because  Reserve  experienced  former  number  studied  Game  and e l e p h a n t s  elephants  may h a v e  Mara  have  Reserve,  tendency  thickets  the  Fire  grazing  a  the  Reserve  inside.  outside  in  the  Lamprey  analysis  discriminate  and  significant  losses. All with  thickets  those  reported  Serengeti. to  In  trees  thickets and  types, time is  by  thickets  these  in  thickets  are  animals,  they  that  thickets. the  to  high  e_t a _ l .  gaps  spend  of  (Langridge  fire  now  body  due  of  the  in  incursion,  loss  the  This  the  soil  primarily  damage  habitat  when  3m w e r e  of  data).  move  in  these  burning frequently  into  to  to  However, and  the  recently  experiments killed  fire  undergrowth  elephants  susceptible  their  3).  sparse  unpublished actions  of  seem r e s i s t a n t  and  the  preference  elephants  (Chapter  to  woodland  selective  moisture  My  other  proportion  grasses growing  bushes under  structural of  browsing  inordinate  increasingly  1979).  and  thickets  the  in  movements  loss  forage  Dublin,  through  c a n be a t t r i b u t e d  season  Croton  (1979)  consistent northern  general an  thinning, the  extensive  shade and  do become  (Norton-Griffiths Croton  internal  Their  dry  1970,  opened,  through  in  With  s e a r c h of  The main  this  results  elephants  accentuated  Mara,  elephants.  shrubs.  within  suffered  by N o r t o n - G r i f f i t h s  the  utilization  within  have  in  hot  as  other spread opened showed fires  46  and those  over  little  no  or  activity  3m e x p e r i e n c e d a r e d u c t i o n  seed crop  and  the  subsequent  rapid  loss  of  from  the  main  small  significantly their  in  fire  thicket  body  lower  of  in  "structural  next  in  fruiting  the  which  integrity",  as  for  become  Area  thickets  Elephant  accounts  have  thickets.  produced  season.  damage p r o b a b l y  patches  those  canopy and  isolated  losses  which  have  rectangular  the  have  been  maintained or  circular  conf i g u r a t i o n s . Further thickets  studies  began  photographic cattle is  is  (Lamprey  in  1985).  prime  habitat  areas  adjacent  This  Inside numbers  migratory grasses season that  the  have  rinderpest  is  in  early  under  re-establish  there  is  M a r a Game  Reserve.  a  the  1973).  reduce  the  1976)  frequency  Norton-Griffiths  conditions, as  no e v i d e n c e  the  woodlands  they  have  of  woodland  provides  increasing  et  al.  prevail.  since  burning  incidence  (Rossiter  by  re-established  are  high  significantly  themselves  Currently  numbers  aerial  no  encroachment"  (Sinclair  and the  or  becoming  conditions  five-fold  reduced.  these  overgrazing  so l i t t l e  last  and  Reserve,  and  cattle  woodlands the  are  their  these  following  "bush  similar  1960s  of  crop  plants  Reserve  increased  are  standing  occurring, in  (McNaughton  the  flies,  the  wildebeest  fires  of  widespread  Reserve,  the  immediately  woody  tsetse  to  trypanosomiasis  the  and  for  dynamics  Outside  removing  place  the  1982,  survey.  now  taking  into  of  1983).  Wildebeest  disappearance Like  cattle,  standing  (1979)  recovery  in  of  of dry  predicted  and t h i c k e t s the  of the  crop  and s e v e r i t y  outside  in  would  Reserve. the  Masai  47  Summary The  Serengeti-Mara  ecosystem  has e x p e r i e n c e d  from open g r a s s l a n d t o  d e n s e w o o d l a n d and back  years.  reconstructs  This  chapter  through both q u a l i t a t i v e in  vegetation  implicated At  in these  the  explorers  patterns  a t t r i b u t e d to  the  late  wildlife  by t h e  viral  on t h e s e a n i m a l s ) scale.  eliminated  by l o c a l  on t h e  hunters  coast.  were  With  ideal  for  the  factors  However, This  by  the  transition  and humans  had a  (dependent  been  on  herbivore populations of  an  virtually  flourishing  establishment  In  greatly  s t a r v a t i o n and d i s e a s e  supplied  large  changes  numbers had been  populations  who  The  i n man-made f i r e s .  rinderpest,  Elephant  changes  a r e a was d e s c r i b e d by  wooded.  reduction  and l i v e s t o c k  disease,  100  these  biological  grassland.  had succumbed t o  epidemic  the  densely  a concomitant  1800s,  conditions  an open  than  discussed.  century,  same a r e a had become  is  trade  the  and t r a d e r s as  1930s t h i s  reduced  of  are  of  evidence.  a r e d e s c r i b e d and t h e  transition  in less  history  and q u a n t i t a t i v e  transitions  turn  the  a  at  ivory a  low,  woodlands  and  thickets. The Masai  areas  Mara  vegetation woodlands aerial showed  set  Game  aside  Reserve  i n the  photography  absolute Reserve continued  steady and  were into  the  were  Serengeti  However,  by  flown of  in  1950,  cover  An  analysis  rates  of  highest  from  1961 These  and  and t h i c k e t s .  woodland cover to  1974,  1967, declines  loss though  i n the  the  woody  1950s  1967,  i n woodlands  relative  1980s.  1961,  dense  by t h e  were a l r e a d y d e c l i n i n g .  loss  the  N a t i o n a l Park and  characterized  1930s and 1940s.  and t h i c k e t s  the  as  these of 1982 The Mara  declines  c o i n c i d e d w i t h an  48  increase fire  in  local  frequency  mean c o v e r  of  the  Croton  Although recovery  has  currently  no  Mara  is  and  still  densities  severity.  of  Mara  as w e l l  During  this  a s an 30-year  Acacia  woodlands  dropped  Reserve  as d i d  mean c o v e r  the  increase  in  period,  the  significantly density  and  thickets. burning  rates  been  predicted  sign  Game R e s e r v e . elephants  and  density  throughout area  elephant  The  of  a return  role  of  migratory not  are  fully  presently  under to  these  woodland  unprecedented wildebeest  understood.  in  low  and  conditions,  vegetation numbers  woodland  in of  woodland there the  is  Mara  resident  dynamics  of  the  49  CHAPTER 3 .  H A B I T A T S E L E C T I O N AND GROUP S I Z E OF E L E P H A N T S  I N THE  M A S A I MARA GAME R E S E R V E  Introduction The  differential  significantly  alter  (Curry-Lindahl  1968,  use the  of  structure  Laws  patterns  are  availability  (Leuthold  and S a l e  well  as  conditions (Lamprey Lindsay Over  by  et  al.  1985),  the  concentration their  protected  25 y e a r s ,  of  range.  1967,  elephants  Today  parks to  order  understand  parks  in  most  know  the  and r e s e r v e s , and to  1970,  Maasai of  on t h r e e people  Tanzania.  In  the  the  late  preference  and  of  weather  cultivation  1985, Western  (Douglas-Hamilton factors refuges  have in  impact  on t h e  of  led  they  make d e c i s i o n s on l o c a l  fourth  by  1970s t h e  a  settlements  of  Serengeti  Serengeti  the  provided  has  been  pers.  reversed  and  comm.).  S i n c e about  poaching  pressure  1980,  in  in the  Park  elephants  the  Mara Reserve  in  area  National  and to  Douglas-Hamilton,  is  pastoral  time  the  in  It  small  f r o m p o a c h i n g p r e s s u r e w h i c h was h i g h of  of  management.  is  the  the  occupy  refuge  north  to  survive  with a safe inside  1979).  elephants  habitats  Mara)  and  many p a r t s  elephants  utilization  and  1985)  (Cumming a n d J a c k s o n 1 9 8 4 ) .  s i d e s by p e r m a n e n t  a n d on  1975).  extreme  Lamprey  external  T h e M a s a i M a r a Game R e s e r v e ( t h e bordered  as  settlement  Africa's  patterns their  can  communities  Thomson  forage  such  protected  of  plant  1971, by  activity  and r e s e r v e s  necessary to  Laws  these  elephants  1973, Western and L i n d s a y  human  and p o a c h i n g  past  Field  factors  1973),  by  affected  influenced  external  (Corfield  of  1970,  Utilization  as  habitats  at  Kenya  that (I.  situation  has mounted  on  the  50  Serengeti  s i d e , thus c u t t i n g o f f t h i s escape  route t o the south.  In t h e mid-1980s, d u r i n g t h i s s t u d y , elephant i n and o u t of t h e Reserve  were s e v e r e l y c u r t a i l e d ,  800-1100 e l e p h a n t s u t i l i z e d residency  produced  vegetation,  This  browsing  influencing  i n t h e Reserve  paper  selection  t h e Mara y e a r - r o u n d .  noticeable  significantly  p l a n t communities  s e a s o n a l movements  describes  approximately This  effects  between  the  foraging are analyzed functions  served  (Chapter 5 ) .  the  seasonal  changes  seasons, further  while  in  i n Chapter  4.  gazetted  the northernmost  and  ecosystem.  extension In  f o r t h e Game R e s e r v e  modifications,  only  west,  large-scale t h e Mara  Escarpment, Park.  and  feeding  possible  1,000  i s bordered  of  but, following km  agricultural  availability.  border w i t h  1 9 7 4 , 1,530  2  remains.  e a s t o f t h e M a r a i s now p e r m a n e n t l y  and  The  (34° 45' t o 35° 30' E,  45' S) l i e s on K e n y a ' s s o u t h w e s t e r n  Serengeti-Mara  in  Area  M a s a i M a r a Game R e s e r v e  forms  Habitat  by e l e p h a n t a g g r e g a t i o n s a r e e x a m i n e d a s w e l l  Study  and  habitat  the d e t a i l s of elephant  a s t h e c o n s t r a i n t s p l a c e d on g r o u p s i z e by f o o d  1°  the  o f many  and group s i z e of e l e p h a n t s w i t h i n t h e Mara.  patterns  to  on  the structure  p r e f e r e n c e s a r e d i s c u s s e d as they r e l a t e t o changes  The  full-time  schemes by  km  2  15'  Tanzania  25,000 were  km  2  formally  recent  boundary  The a r e a t o t h e n o r t h  s e t t l e d by  pastoralists  (Lamprey 1 9 8 5 ) .  the S i r i a  t o the south extends  the  1°  (Esoit  To t h e  Oloololo)  to the Serengeti National  51  This dry  small,  season  migrants hundred  which  these  include  thousand  wildebeest the  area presently  (June-October)  These migrants  to  protected  move  At  migrants  the  leave  by  the  many  1979,  the of  the  of  short  essential  the  wildebeest  Serengeti  and over  et  al.  one  1985).  Thomson's g a z e l l e s  Mara from  the  of  Sinclair  thousands  into  start  for  one m i l l i o n  (Maddock  joined  which  north.  over  zebras  are  range  provides  the  dry,  rains,  in  Loita  and  plains  late  October,  rains  falling  Mara.  Climate Rainfall in  the  Mara  November-December  (Figure  3.1).  Convergence from  This Zone  Zimbabwe,  Griffiths follows about area  the 5  southern  with  the  (ITCZ) the  is  south,  Driven  to  extreme  in  Mara twice  Sudan,  movement  of  trade  its  side  averaging  west  side  approximately  1975,  Epp and A g a t s i v a  the  the  April-May  Intertropical  East African  of  thus  (Norton-  pressure  sun w i t h  fronts  a  lag  converge rain  limit  region  north  low  which  northern  in  in  and late  of this  cloudy July  and  affecting  weather  rainfall  gradient  year. east-west  approximately  1 200 mm/year  1980,  the in  in  the  system brings  Mara a l s o has a pronounced east  by  winds  January,  each  short  occurring  affects  band  the  late  rains  This  by  reaches  with  influenced  which  1975).  ITCZ  the  bimodal, long  hemispheres,  The  The  and  n o r t h and south  both  in  is  pattern  a_l.  weeks.  from  patterns  in  et  weather. its  in  Stelfox  et  800 mm/year  and  (Norton-Griffiths al.  1986).  The  the e_t  al.  west  52  1  I  I  I  I  1  1  1  1  1  1  1  J F M A M J J A S O N D MONTH  F i g u r e 3.1.  M o n t h l y r a i n f a l l t o t a l s from Keekorok Lodge M e t e o r o l o g i c a l S t a t i o n l o c a t e d i n t h e study a r e a . The t h r e e s t u d y y e a r s 1982, 1983, and 1984 a r e p r e s e n t e d .  53  side  experiences  by t h e  convergence  systems This the  much h e a v i e r  and  rainfall west  side  Onyeanusi,  of  the  locally  by  gradient  results  of  the  pers.  rainfall  Lake V i c t o r i a the  and  to  west  Agatsiva  the  year  Escarpment.  on t h e  in  (Figure 3.2).  the  differences  sides  1980,  and h e r b s  years  east  the  (A.  Mara  The a v e r a g e  1982-1984 was  28.1°C  and  degraded  are daily  and  the  to  the  some  woodlands  on t h e in  rocky,  the  Mara  whereas the  west  vegetation  along  trees  and s h r u b s  (1961)  and a l l  higher merge  by  were  grasses  Combretum  elevations, into  relict  - Commiphora  exposed  deeper  and  the  are covered  Acac i a  Epp  (1951).  woodlands  t o as  the  at  1973,  p o r t i o n of  woody  All  far east,  These  (Taiti  west,  Edwards and Bogdan  (referred  these h i l l s ,  the  D a l e and Greenway  slopes.  (1985))  than  with  patterns,  a r e a l s o c h a r a c t e r i z e d by  The e a s t e r n  hilltops.  Heeria reticulata  Lamprey  base o f  1985).  on  according to  thickets  Mara  communities  according to  steeper  the  in annual r a i n f a l l  community c o m p o s i t i o n  Lamprey  The Ngama H i l l s , -  of  open g r a s s l a n d  courses  classified  by  weather  p r o d u c t i v i t y on  that  temperatures  the  in plant  primarily  river  the  Range  Siria  grass  compared to  for  has more woody p l a n t  molle  the  influenced  Communities  differences  is  Mau  is  14.8°C.  In a d d i t i o n east  it  comm.).  maximum t e m p e r a t u r e  Plant  of  in a higher  Mara R i v e r  throughout  minimum was  and  effects  Minimum and maximum d a i l y constant  because  slopes  soils,  or  woodlands  below.  Acac i a  on  At  the  qerrardi i  54  36 - i O o 30-  H  W  24-  s  ac  E-  1  O < w  •  -  MAXIMUM  O  -MINIMUM  18-  o—cr  ,o  O  -o-  -o-  -o-  -O  Or  -O  o  o  JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC MONTH  F i g u r e 3.2.  Mean monthly maximum and minimum t e m p e r a t u r e s from Keekdrok Lodge M e t e o r o l o g i c a l S t a t i o n l o c a t e d i n t h e s t u d y a r e a . These maximum and minimum t e m p e r a t u r e s a r e averaged over t h e t h r e e s t u d y y e a r s 1982, 1983, and 1984.  55  woodlands  form a m o s a i c  Relict seedlings  thickets  and  with are  coppicing  rootstocks  Acac i a b r e v i s p i c a ,  Albizia  amara,  africana, Grewia  trichocarpum, here  are  seen to  be  community.  derived  described Acac i a northern  it  from  as  degradation  in this  The cohort  Acac i a  stands  thickets  community  past  time  (1972)  Taiti by  1973,  be  similar  new  as  this the  of  A variety  coppicing  s p e c i e s a r e commonly f o u n d  of  the  of  the  a  1985).  distinct  increase  of  over  the  attributed and  to  its  Guinness  persist  underground  in this  relict  Lamprey  Mara  Spinage  of  clumped  the  rapid  Individual plants  b u r n i n g or b r o w s i n g .  thicker  highly  as  the  largely  from  (1985)  composition  with  1976a & b ,  1970,  following  community  classes.  occur  i n many a r e a s  shoots  commonly  phase  species  documented  is  listed  Large areas  distinguish  Trump  this  i n an e a r l i e r  (Herlocker  Chapter 5).  sprouting  to  proliferation  ( G l o v e r and  plants  w h i c h had been  past.  woodlands  (1985)  woodlands This  areas  in the  d i d not  cinerea,  whereas Lamprey  found i n a s s o c i a t i o n  above  Lamprey  gerrardii  fire-tolerance 1972,  frequently  all  considered  larger height  gerrardi i  type,  decade.  i n the  Commiphora  a m e r i c a n u m , Ormocarpum  Virtually  of  These Senegal,  Dichrostachys  thickets,  community w i t h  mentioned  A l t h o u g h Trump  Acacia  Croton  now a p p e a r  b u t many more p l a n t s  A.  S t a n d i n g dead t r e e s a r e  a degraded phase  Serengeti  containing  species.  A. hocki i ,  Ocimum  Trump (1972)  Commiphora b u s h l a n d s  diverse,  many  ovalis,  high.  communities.  Boscia a n g u s t i f o l i a ,  javanica,  1 meter  and of  and Solanum i n c a n u m .  below  in t h i s  Cordia  Lippia  thicket  A. gerrardi i ,  petersiana,  C. trothae,  spp.,  relict  widespread  include:  A.  the  through  rootstocks  s e e d l i n g s and community.  rootThese  56  include: and  Acac i a  where  burning is  individual relict been  trees  thickets,  or is  suppressed  indeterminate Ridge  by  both the  species very  africana, heights  up  to.  Strychnos  frequently  never  found  inside as  Mara.  like  those  in  growth  has  fires  for  drainage  lines  Grewia  In  are  15 m e t e r s  found.  the  4 - 5  meter  foliolosum,  lower  This  use  layer  Acac i a  natalensis, is  jovis-tonantis edge and a r e  This  community,  (1960),  is  currently a  and c o v e r  and r e p e a t e d  canopy c o v e r  attain  thicket  proper.  browse,  Plea  community  Gardenia  from t h e  thickets  heavy  still  trichocarpa  the  the  clumps are  t r i c h o c a r p a , Rhus  o r more a d u l t  This  dominates  the  Teclea  covered  fires  and a r e a  loss  for  many  have in  led  recent  2).  Balanites  aegyptiaca  section  Balanites  and  meters.  the  t h i n n i n g of  of  Mara a r e  Haplocelum  dry season  i n the  western  as  below  shade,  animals  (Chapter  fall  the  thicket  bush" by D a r l i n g  of  age  and  Croton  "lion  source  internal  species  7  10 -  primary  in  true  when t h e i r  of  Though  such  marked by one  to  portions  discrete  henningsi i  occur  referred  these  ovalis,  which  community,  determine  thickets.  and  trees  The  and many s e a s o n a l  graveolens,  Cordia  severe.  trampling,  and w e s t e r n  6  cinerea,  time.  species  Tarenna  brevispica,  years  of  W h i l e most  height,  to  Dichrostachys  r e m a i n u n d e r one meter  in this  small h i l l s ,  eastern  diverse.  and  seedlings  composition,  Croton  as  spp.,  or  browsing,  periods  tops,  frequent  difficult  by C r o t o n d i c h o g a m u s  to  Commiphora  Ormocarpum t r i c h o c a r p u m w h i c h a l l  areas  in  Senegal,  of -  the  woodlands  Mara.  Acac i a  This  seyal  occur  community,  woodland  only  in  also  referred  (Taiti  1973,  the  far to  Herlocker  57  1976b, like  Lamprey  1985),  e x p a n s e s of  a decline Glover  i n the  and  the  giraffes  Trump  adult  and b r o w s e r s Also the  occurs  in  the  created  by t h e  in t h i s  (1970)  and P e l l e w  (1981)  attributed  still  apparent  today.  neither  believed  these  of  large  Syzyqium  and t h e has  edulis,  an  heavy  majority  of  r e a c h of  limited  both  by  1 984)  the  and  river.  past  changed,  its  formerly  large for  light  these  disease  effectively  20  gaps  losses  However,  removed by  to  those  north  the  The c a u s e s  been  which  over  understood.  leaving  forest  south along  now marked w i t h  it  is  elephants, or  merely  thinned  the  s p e c i e s which are  not  The p r i m a r y c a n o p y  trees are  Albizia  E u c l e a d i v i n o r u m , and W a r b u r g i a Cassine  cordatum,  equally  to  browsing  riverine  scarcely  also  include:  which  and S o u t h g a t e  altered  succumbed  have  occasional  the  been  nor  Diospyros abyssinca,  rueppelina,  Pellew  to  trees.  have  may be  reported  area  from Mara S e r e n a  t r e e s may have  food.  Others  Carissa  has  is  the  a band of  patches  forests  uqandensis.  understory  is  The  park-  5).  Mara R i v e r  canopy  Elephants of  1974,  b o r d e r s have  smaller  p r e f e r r e d as  reclinata  side  forest  t r e e s may  senesced.  above  Chapter  well-documented  taller  petersiana,  a_l.  discrete  its  well  and r e g e n e r a t i o n  the  meter  that  understory  are  1984,  of  loss  (1973)  stands  and d i v e r s e  15-20  of  west  course  Though  highly  Taiti  grassy,  Balanites  (Belsky  closed,  while  Mara T r i a n g l e a r e a .  trees  small,  thick  years.  are  open,  (Lamprey e t  on  follows  This  on t h e  and e l e p h a n t s  seed p r e d a t o r s  prevalent  density  b r o w s i n g by g i r a f f e , remaining  is  buchanani i ,  Pappea  P o d o c a r p u s and diverse  Chaetacme a r i s t a t a ,  Ekebergia  . capensis, Ficus  trees.  s h r u b component Croton  Phoenix The  including:  dichoqamus,  Grewia  58  trichocarpa, Ziziphus  Phyllanthus  derived  and  is  comprised  Themeda t r i a n d r a ,  rains  other  tall  Hyparrhenia  also  productivity  is  flower  (Sinclair  1975,  wildebeest  may remove  (McNaughton  1976,  permit  racemosa,  -  as  et  Aristida  of  Mara  and f l o w s  by  the  The the  al.  and  -  plains  adoensis,  grass  kg/ha/yr  migratory  before  of  Setaria  8,000  standing  1986)  zebra  crop  they  other,  leave less  Eraqrost i s  and  biomass  returning  and S p o r o b o l u s  the  plains, the  Mara was  River,  into  major Talek  the  Lemek  the  drains  described largest  the  Lake V i c t o r i a  River originates  several  include:  drain  90 % o f  Harpachne s c h i m p e r i ,  The  Mara  Loita  1979).  long  macroblephara,  The o v e r a l l  7,000  g r o w t h and f l o w e r i n g  Serengeti-Mara ecosystem,  fed  grasslands.  the  south behind  competitve tenui f o l i a ,  stapfianus.  and H y d r o l o g y  (1966).  The  mezianum,  The s h o r t - c r o p p e d  The h y d r o l o g y  region  Pennisetum  from  perennial  Following  Digitaria  ranging  Stelfox  the  such  Geology  and  edaphically-  by t h e  as  in these  80  of  dominated  grass".  such  McNaughton  Serengeti.  grasses E.  trichocarpa,  combination  "red oat  grasses  high,  a  grasslands  filipendula,  sphacelata  then  of  fire-induced  grass,  the  Teclea  mucronata.  The Mara  to  s e p i a l i s,  i n the  River  valley  northern some  to  the  100  along  which a r i s e s  O l a r e Orok and  perennial  Mau Range  tributaries  in d e t a i l  river  Serengeti km to  its i n the  Jagartiek  by G l o v e r  to  the  the  west.  n o r t h and  course.  is  These  Siana h i l l s  the  the  and Mara  watercourses  n o r t h and j o i n  in  and  which  Talek River  59  close  to  which  its  confluence  originates  in  with the  the  Loita  Mara R i v e r , Hills  Mara a l o n g  the  Kenya-Tanzania border.  year-round  but  in  Williams  dry  (1964)  extensive  peneplain  sediments  which  have  and v o l c a n i c  volcanic  origin  and  soils.  Below  this  basement  system of  forms  the  above  western  rising  to  1900  of  boundary  layer  area  of  below.  the  in  of  the  small  Mara and  by  area  are  the  plains  as  an  to  faulting, mainly  black tuffs  of silt  lies  a  The  most  Escarpment  which  rises  elevation  flow  pools.  time  Siria  the  pre-Cambrian  volcanic  the  joins  region  loams  River  often  and q u a r t z i t e s .  is  Mean  on  over  sandy  schists,  the  1600 m e t e r s  Soils  from brown,  gneisses,  and  metamorphosed  modified  shallow  plains  approximately  Mara  the  Sand  rivers  described  activity.  in  These  east  of  been  range  relief  the  has  the  become a s e r i e s  comprised  erosion,  significant  years  to  and the  100-300  in  with  the the  meters Mara  is  escarpment  meters.  Methods Habitat independent used basis,  to  selection techniques.  distinguish  whereas  analysis  of  the  habitat  by  elephants  The  first,  selection  second, monthly selection  was d e t e r m i n e d  total  only  on  aerial a wet  counts, and dry  census c i r c u i t s ,  b o t h by  using  s e a s o n a n d by  was  season  allowed sex.  two  an  60  Wet and D r y Season  dry  and  Counts of  Two wet  season  (late  April  season  total  counts  of  conducted the  Total  in the  entire  Elephants  1984  elephants  northern Serengeti  M a s a i Mara Game R e s e r v e ,  mapped  on  recorded only wooded  areas  correct  for  all  three  on t h e were  the  flights  October  1984  (late  October  1984)  were  Tanzania,  Kenya.  were  but  Herds  and  counted  o c c u r r e n c e by h a b i t a t  intensively  visibility  and one  National Park,  and May 1985  s e a r c h e d more  relative  and e a r l y May 1985)  of  counts.  Densely  t h a n open a r e a s  elephants  was  to  in  different  were  randomly  habitats. To  test  distributed elephants  i n the  in  the  area  habitat  relocate  the  random  in e a r l y  Unit.  areas.  Thirty  for  (Table  expected  distributions  compare  the  1982  was compared t o expected  by t h e  on t h e  line  3.1).  size.  observed  elephants For  the  if  Kenya  were the  purpose  were c o n v e r t e d Chi-square versus  the  they  to  Rangelands  were  at  a  t y p e s on  the  them. checked  Reserve  drawn  on  was  to the of  theoretical  showed  no  habitat  analysis  numbers,  expected  scale  proportions  of  analysis  an  frequency  p h o t o g r a p h s were  for  of  p r o d u c e d from an  habitat  transects  analyzed  number  These p r o p o r t i o n s p r o v i d e d the  selection  sample  This  distinguish  and  distribution  observed  type  1:50,000 t o p o g r a p h i c map o f  mosaic  habitats.  total  These photographs  types d i s t i n g u i s h e d  photographic different  habitat  the  from a p h o t o m o s a i c  enough t o  ground u s i n g a  elephants  habitats,  each  derived  large  that  distribution.  Monitoring  1:50,000,  on t h e  was  survey of  Ecological  All  available  frequency  distribution  of  hypothesis  observed  expected  aerial  the  these  b a s e d on  then  the  used  to  numbers a c r o s s  all  Table 3.1.  Percent o f each h a b i t a t type measured from a e r i a l photography o f the Masai Mara Game Reserve flown i n 1982 and monthly census c i r c u i t s . These percentages were used to c a l c u l a t e the number o f e l e p h a n t s t o be expected i n each h a b i t a t type under a random d i s t r i b u t i o n .  HABITAT TYPE  GRASSLAND  RELICT THICKET  Aerial Photography (%)  41  29  Monthly Circuits (%)  43  34  ACACIA WOODLAND  CROTON THICKET  BALANITES WOODLAND  SWAMP  OTHER  62  habitat  types.  Monthly Census  A 152  km. c i r c u i t  Initially, which km.  Circuits  the  entire  intersected  (Table 3 . 1 ) .  then  this  as  by  sizes  elephants,  converted to  within  the  and t h e  habitat  c i r c u i t were mapped t o  a  elephants  observed  c i r c u i t was d r i v e n  length  proportion  proportions  of  established  The c u m u l a t i v e  expressed  From t h e s e  was  a  habitat  random  of  the  total  frequency  of  males  and f e m a l e s  type  circuit  U s i n g the  observed  distribution  was  for  sample  was  then  expected  in each h a b i t a t  0.1  length.  distribution  numbers w h i c h a l l o w e d a c o m p a r i s o n of  numbers  types  nearest  each h a b i t a t  frequency  was p r o d u c e d . this  of  the  Reserve.  type  with for  each month. This through season  May  1985.  and t h e  vehicle were  c i r c u i t was d r i v e n  were  seen.  This  1984  r e c o r d e d by a g e ,  of  to  Chi-square  observed expected the  numbers  elephants  of  numbers f o r  sexes w i t h i n each  analysis  both the season.  and t h e  different  1983  1983-84  observed  from  i n which  excellent habitats  and  was  by h a b i t a t  type  wet the they the  assumed  was a g a i n u s e d t o compare  and f e m a l e s wet  the  habitat  c i r c u i t was  i n the  males  encompassed  A l l elephants  sex,  from t h e  sightability  e a c h month from November  period  dry season.  Visibility  be e q u a l .  once  to  the the  and d r y s e a s o n s and t o compare  63  Results  Habitat  Selection  Seven hundred dry  s e a s o n a n d 912  the  aerial  habitat dry  counts  type  = 347.7,  preferences  the  were  wet  not  in  either  d.f.  = 6,  In  All  randomly  season (seasonal x P<.001). in  the  dry  They  the  habitats  counted  with  in  the  during  regard  values:  2  showed  wet  wet  =  to  126.3,  distinct  habitat  season elephants  selected  thickets  season  were  Elephants surveyed  distributed  3.3):  other  elephants  season.  swamps, and C r o t o n  use.  selected.  in  (Figure  grasslands, random  and e i g h t y - f i v e  more  only  than  expected  Croton  contained  fewer  thickets  by were  elephants  than  expected. These monthly  patterns  census c i r c u i t s  showed non-random males: n  wet  = 25,  = 247,  =  11.5,  P<.025,  P<.025,  dry  grasslands  whereas  males  and  were  and  season (x  in 2  (x  2  and  both  (n)  values  Croton  more  = 58.7,  not  either d.f.  for d.f.  the  seasons (x 111, P<.05,  for  females:  P<.001,  d.f.  differ  = 5,  males = 5,  chose  (x  (x  2  the  thickets.  = 33.2,  P<.001).  d.f.  Females season In  in  = 5,  the and  their n.s.)  or  differences  = 5,  Figures  n  distribution.  Seasonal d.f.  14.3,  thickets  random  = 8.0,  =  for  = 13.9,  wet  significantly  n.s.). 2  wet  Croton by  dry  the  sexes  values  2  = 5).  in  on  Both  =  expected  wet  recorded  3.5).  thickets  both  than  did  = 4.9,  significant  females  in  n = 116,  females  females  distributions dry  = 38.9,  2  those  3.4  males x  to  s e l e c t e d g r a s s l a n d s and r e l i c t  woodlands  Males  of  = 5,  season males and  Acac i a  habitats  number  d.f.  similar  (Figures  use of  selected  dry  were  P<.001)  3.4  and  and 3.5  GR = grassland 1-1 RT = relict thicket AW = Acacia woodland CT = Croton thicket BW = Balanites woodland SW = swamp OT = other habitat types  GR  T l T n r r  WL J  RT AW CT BW SW OT  DRY  -l-  -2  J  GR  RT AW CT BW SW OT HABITAT TYPE  The h a b i t a t p r e f e r e n c e s o f e l e p h a n t s i n t h e Masai Mara Game Reserve i n t h e wet and d r y s e a s o n s . The e x p e c t e d v a l u e s a r e c a l c u l a t e d from a breakdown o f t h e Reserve by h a b i t a t type u s i n g t h e 1982 t o t a l a e r i a l photography. The observed numbers were t a k e n from a e r i a l counts flown i n both seasons. Notes : 1) V a l u e s g r e a t e r than z e r o i n d i c a t e h a b i t a t s used more than e x p e c t e d i f e l e p h a n t s were randomly d i s t r i b u t e d , w h i l e those v a l u e s l e s s than z e r o a r e h a b i t a t s used l e s s than e x p e c t e d and 2) * = those h a b i t a t t y p e s i n which e l e p h a n t s were e x p e c t e d b u t were n o t seen.  65  In  GR = grassland RT = relict thicket AW = Acacia woodland CT = Croton thicket BW = Balanites woodland OT = other habitat types  FEMALES  o  feJU  O  ,  -1  GR  RT  AW  CT  BW OT  MALES  In W X  O  g£ W CO CD,  ,0 CuD O -1  J  GR  RT AW CT HABITAT TYPE  BW OT  F i g u r e 3.4. Dry season h a b i t a t p r e f e r e n c e s o f male and female e l e p h a n t s i n t h e Masai Mara Game R e s e r v e . The e x p e c t e d v a l u e s were c a l c u l a t e d from a breakdown o f t h e monthly c i r c u i t by h a b i t a t t y p e . The observed numbers were d e r i v e d from censuses d r i v e n on t h i s c i r c u i t each month. Note : V a l u e s g r e a t e r than z e r o h a b i t a t s used more than e x p e c t e d were randomly d i s t r i b u t e d , w h i l e l e s s than z e r o a r e h a b i t a t s used expected.  indicate i f elephants those v a l u e s l e s s than  66  , Q  v  1 -i  GR = grassland RT = relict thicket AW = Acacia woodland CT = Croton thicket BW = Balanites woodland OT = other habitat types  FEMALES  o  OjO O  4  -1  GR  RT  AW  CT  In  BW OT  MALES  0  GR Figure 3 . 5 .  RT AW CT HABITAT TYPE  BW OT  Wet season h a b i t a t p r e f e r e n c e s o f male and female e l e p h a n t s i n t h e Masai Mara Game Reserve. The e x p e c t e d v a l u e s were c a l c u l a t e d from a breakdown o f t h e monthly c i r c u i t by h a b i t a t t y p e . The o b s e r v e d numbers were d e r i v e d from censuses d r i v e n on t h i s c i r c u i t each month. Note : V a l u e s g r e a t e r than z e r o h a b i t a t s used more than e x p e c t e d were randomly d i s t r i b u t e d , w h i l e l e s s than z e r o a r e h a b i t a t s used expected.  indicate i f elephants those v a l u e s l e s s than  67  illustrate and  the  females  general preferred  Acac i a woodlands  Group  data.  these d i f f e r e n c e s .  grasslands  and C r o t o n  variation  Average  during  thickets  the  i n the  and  individuals/group. 2.84,  d.f.  herds  i n the  = 102,  Amboseli  These P<.01).  dry  Both  wet  males  s e a s o n and  season.  by a d u l t females,  subadults,  averaged  1.9  herds  season,  whereas  within  a  average  herd size  either  the  season.  wet  n.s.).  in  even the  Tsavo  was  (Laws  14.8  158  Euclea  13.2  averaged  recorded  the The  and t h e  w h i c h were  400  in  largest largest  accompanied  Those h e r d s 7.5  season  700 or  (t =  with  and a l l - b u l l  only  groups  basis.  in  average  The l a r g e s t  the  1969)  animals  individuals.  all  difference  (r = 0 . 4 3 , s  was  National Parks.  Cow-calf herds  T h e r e was and  22.5  l a r g e r wet  larger  little  woodlands,  it  d i d v a r y between s e a s o n s  were  season.  season  was  However,  on a y e a r - r o u n d  general  season  count  significantly  and c a l v e s  Mean g r o u p s i z e  wet  r e a c h e d numbers s u c h as  40.  averaged  the  from  averages d i f f e r e d  season a g g r e g a t i o n  males  was c a l c u l a t e d  dry  ( W e s t e r n and L i n d s a y 1985)  season group only  Acac i a  in  the  aggregations  r e c o r d e d wet  n = 8,  size in  Mara n e v e r  seasonal  i n group s i z e  herd  individuals/group  dry  of  Size Seasonal  in  pattern  habitat occurred  no s i g n i f i c a n t numbers  n = 8,  n.s.)  per  types  and  3.2). i n the  In wet  between  habitats  correlation  between  habitat  or dry season  g r o u p s were seen thickets  (Table  in r e l i c t grasslands  type (r = s  for 0.54,  thickets, i n the  wet  Table 3.2  Average herd s i z e s f o r each h a b i t a t type i n the wet and d r y seasons. Number i n parentheses i s the number o f herds observed i n each h a b i t a t type during the monthly census c i r c u i t s summed over the e n t i r e season.  HABITAT TYPE  GRASSLAND  RELICT THICKET  ACACIA WOODLAND  CROTON THICKET  Wet Season  11.6 (38)  11.8 (94)  14.5 (15)  10.2 (40)  Dry Season  8.9 (38)  7.4 (46)  6.0 (25)  9.6 (40)  SWAMP  9.2 (13)  7.8  (22)  69  Discussion  Habitat  Selection  The M a s a i M a r a changes these  in  the  past  changes  has  woodlands time  Mara  surviving  is  stands and  Resident  elephants  which  habitats  are  use  conditions  in  of  the  of  Mara in  thickets region  river  past  are  loss  three  of  available  shade, and a l s o  thrive  in  other  the  Consequently,  to  moist,  damage  to  of  Croton  Mara  River.  largely  prevented  from  using  routes  are  on t h e  now  living  remaining  s p e c i e s of  and  all  within to  of  an  in  the  woodland  sorts,  the  the  across  the  Serengeti  and  on h i l l t o p s  and  the  now p r o v i d e Here  on woody  internal  in  within  structure  With  are  the  these the  the last  able  and h e r b s  of  the  Mara over  one of  they  species,  have  thickets.  Reserve.  habitats  elephants.  elephants  northern  dry  reduced  Croton  bush s i t t i n g  and  when  already  shade t r e e s ,  shady c o n d i t i o n s the  few  the  thickets  forage  patches  Facing  woodland  to  present  along  4).  unique  scattered  the  increased significantly  islands  decades, these  wooded r e f u g e s find  discrete  Acac i a  5).  time  are  At  in  of  forests  woody  forage  their  which  drainages  significant  (Chapter  and  impacts  and t r e e s  browse  begun c o n c e n t r a t i n g These  been  (Chapter  reduction  result  open g r a s s l a n d c o n t a i n i n g  riverine have  vegetation  The m a j o r  forests.  woodland  Mara ate  shrubs  prevailed  availability  a wide,  and t h e i r  pronounced  1984).  riverine  seasonal migration  year-round  experienced  significant  Acacia  canopy  Reserve  Elephants  of  has  (Dublin  a  largely  high  historical  their  been  of  thicket  Reserve  30 y e a r s  and a t h i n n i n g  the  their  Game  to  which  thickets. Croton  70  thickets  has become  extensive.  their  mainstems  within  these t h i c k e t s ,  within  10 cm of  limit  of  the  are  the  rainy  large  use  these  (Norton-Griffiths predators and  Reserve  these and  As t h e  years  pies  the  the  progress  Acac i a  groups  well.  trees  unusually  feed  on t h e  approximately  had  km. m a t u r e Acac i a left  34%  of  6  forests  during was  this  Other  all  low-growing  has  of  grass  of  hidden to  occur  dry  in  burn  the very  boundaries.  fragmented  -  like  pieces. to  for lot  trees  very dry p e r i o d .  similar,  height  left  along t h e i r  i n the  a  bull  risk  thickets  similar  occasionally  t r e e s dead or  22% w i t h many b r o k e n b r a n c h e s . damage  the  pressures,  from  cow-calf  Mara a r e  comprised  shade. of  However,  time  in  ( L i n d s a y 1982,  In t h e  d r o u g h t of  elephants  s t a n d near E m a r t i .  the  to  subsequent  swards  When f i r e s  become  4 and 5 ) .  In  frequently  be u s e d  heavily  herd  lower  f o o d and shade  a v o i d the  smaller  but  spend  Chapter  for  subjected  bulls  1985,  sq.  thickets  small to  and L i n d s a y of  through  bulls,  too  dry seasons, and  is  Most Acac i a w o o d l a n d s  which are  woodlands  grass  been  the  thick  t r e e s and b u s h e s  have  from e l e p h a n t  of  grow  and removed i n e v e r  woodlands  is  species  a r e browsed  vegetation.  herb l a y e r .  these  primarily as  the  Most g r a z e r s  pathways  destroy  thickets  gaps  this  shredded,  ability).  through  so  in  grass  being cut  Croton  1979).  litter  hotly  of  light  in thickets  form  foraging  are  More p r e f e r r e d  ground (presumably t h i s  pathways  seasons,  and b r o k e n .  s u c h as A c a c i a b r e v i s p i c a ,  elephants'  This constant opened  split  Many C r o t o n b u s h e s  Within  fatally stands  these Western  1984,  visited 24  in  hours  a  a  2  they  damaged and a n o t h e r experienced  E l e p h a n t damage vegetation  to  was  similar riverine removed.  71  Now  these  forests  whose l e a v e s  are well  are beginning certain  they  effects  of  the  to  would will  l i k e l y due t o  of  provide  soil  a  good  probably suffer  areas,  however,  present  been  during  Cynodon a r e a s .  the  additional factor  of  elephants  availability  feed  by e l e p h a n t s the  of  determined  may  switch quality  to  and  distribution  of  these  of  Their  fine  surface  The  used  in  Mara  by  standing of  in  triandra. the crop  wildebeest  elephants  from  extensive  use  damage.  wildebeest  (Chapter  4).  based  on  the  that  they  rains  and  occurs.  different  the  habits,  d r y s e a s o n when a d e c l i n e  types  but  w h i c h grows  indicate  species  trees  Their  foraging  following  herbaceous forage  in  extensive  habitats  observation,  quantity  implicated  Themeda  situation  through  i n the  the  migratory  types.  a browse d i e t  But  water.  can l e a d to  immediately  not  Reserve.  number  p r e f e r r e d forage  on g r a s s e s and h e r b s  is  browsing.  initial  competing  direct  losses  canopy  a grass  the  select  trees  and a r e a l s o  sheer  the  Many  regeneration  i n the  and s u b s e q u e n t  d r y season e x a c e r b a t e s  Mara  forest  composed of  alone  trees  of  fallen  surface  and z e b r a a f t e r The  be  from e l e p h a n t  in extent  Dry c o n d i t i o n s  The  the  by  grasslands  canopy  or o l d a g e .  dry season p r o b a b l y d i s p l a c e s  woodland a r e a s  as  limited  removed.  of  during  pressure  i n u n d a t e d by  wildebeest  Themeda has  left  are p r i m a r i l y  are  these  destruction  Cynodon d a c t y l o n ,  often  Cynodon a r e a s  migratory  is  of  disease  the  gaps  the  tall  elephants.  cause  opportunity for  do u t i l i z e  of  mammals may a l s o  and  heavy  species  grasslands,  r e a c h of  a r e most  Newly-formed l i g h t  low-lying  the  t r a m p l i n g by l a r g e  preferred  of  of  A l t h o u g h the  Elephants  The  out  die.  compaction  roots.  are comprised p r i m a r i l y  in  the The  habitats  72  probably  dictates  seasons.  This  Sale who  (1973), found  that  and  is  of  forage  other  factors  During Mara  elephants  were  grasslands, This  is  (Field  sprouting  seedlings  nutritious.  Although  protein  when  interfere  with  part,  habitats  of  feeding  What is  of  by  the  elephants is  their  with  the  in not  whether  correlated  all  forage  two  during  that  factors,  in  in  1976,  forage  these  abundant,  those  other Barnes  selected  the  c h o i c e because the  wet  open  season.  quality  is  the  new g r a s s e s  and  are  highly  l e a v e s may h a v e h i g h e r  absolute  new g r a s s e s  areas  (Weinmann  1964,  Dougall  et  a_l.  1985),  they  compounds (Coley  like  Mara e l e p h a n t s  idea  new b r o w s e  was  1 9 7 6 , Guy  thickets  the  than  types  grazers,  1985).  growing  1976, McNaughton et  the  forage  habitat  Ross  secondary  for  and  (1985)  determined  Kenya.  some c o m b i n a t i o n  with  levels  Leuthold  Lindsay  Mara elephants  primarily  D o u g a l l and G l o v e r  of  in  between  disturbance.  1963,  levels  Parks  swamps, and C r o t o n  in  of  and  largely  significantly  and L i n d s a y  factor  were  1971, F i e l d and Ross  consistent  determining  crude  such as rains  Western  most  findings  and q u a n t i t y  for  quality,  the  populations 1982,  National  is  the  selection  and Western  quality  established  habitat  with  (1977a),  forage  habitats  quantity, or  consistent  Amboseli  of  of  s e a s o n a l movements  quantitatively choice  patterns  Leuthold  distribution Tsavo  the  such  1983,  selected during  al.  as  Sukumar the  wet  1959,  1964,  Dougall  Field  and  may a l s o c o n t a i n  high  tannins 1985).  which For  season were  the  may most  dominated  by g r a s s e s . Unlike elephants  Amboseli elephants  followed  a dry  (Lindsay  season foraging  1982,  Poole  pattern  more  1982),  Mara  like  that  73  reported  for  Elizabeth  and K i d e p o V a l l e y N a t i o n a l P a r k s , Uganda ( F i e l d  Field (Guy  and  out  et  relative quality  to  (1964),  Croton  in of  thickets  of  biomass  habitat  From that  significantly this  evidence  elephants  quality  of  both  levels  The n u t r i t i o n a l  begin  to  age  in  sexes  the these  were  observed  elephant  habitat  woodlands. that  was  f u r t h e r by  the  critical  to  sources  which  during  the  idea.  In  the  are  dry season  (quality)  but  elephants  was than  lowest swamps  and my f i n d i n g s primarily in their  the  in  the  were forage  bushed-  dry season.  Mara, the  Amboseli  swamps  used  in available  choice  Amboseli  highest  in a v a i l a b l e  on  in  These  i n the  i n the  (Corfield  dry season,  heavily.  a s i m i l a r extent  but may be l i m i t e d  pointed  and may be c o n s t r a i n e d water  may s e l e c t  season.  the  swamps most  higher  dry  crude p r o t e i n  W e s t e r n and L i n d s a y (1985)  habitat  1982).  d r y s e a s o n may r e f l e c t  Park  However, to  Zimbabwe  (1982)  times.  they  1971,  l i m i t e d by  in crude p r o t e i n  bushed-grassland but  the  i n the  Queen  mostly  this  (quantity).  grassland  diet  suggested  particularly  utilized  lowest  i n the  (1973)  of  such as  (Barnes  and B a r n e s  r a p i d l y as  National  supported  elephants  Tanzania  and Acac i a  permanent  The f i n d i n g s  partially  the  Tsavo  survival,  1973).  (1971)  Elephants  food a v a i l a b l e  distribution their  Field  Habitat choices  L e u t h o l d and S a l e  quantity  Park,  species maintained higher  preferences.  areas  Sengwa W i l d l i f e R e s e a r c h A r e a ,  grasses declines  selection  seasonal  grasses during water-limited  season.  utilizing  other  a p r e d o m i n a n t l y browse  woody  of  forage  1976),  to  al.  that  in  a n d Ruaha N a t i o n a l  switched  Dougall  dry  Ross  1976),  They  elephants  forage  biomass  crude  protein.  it  basis  by t h e  This  seems  likely  of  forage  amount of  food  74  available  within  importance elephant more  of  quality  restricted of  habitat  in  densities habitats than  In  the  may b e  are  is  much  to  appear  In to  of the  a greater  extent  to  be  of  Mara, high,  in  may  available  type  be  local  elephants  of  the  by  their by  food,  forage  where  eaten  elephant choice  forage  Amboseli,  be  of  quality  where  local  as a  direct  higher.  possible  consequence support sizes  of  the  higher are  that  numbers  specific  other  than  (Table  not to  general, season,  fully  large  (Laws et  fact,  there  habitats,  Group s i z e  in  mean  3.3),  but  preferred  therefore,  data  do  not  and average group  b a s e d on h a b i t a t  herd the  size  actual In  herd  herds  form  size in  does the  wet  support  this  by  determinants  not  seem  elephant in  these factors  preference.  has been r e p o r t e d  some a r e a s ,  which did  herd  size  be d e t e r m i n e d  habitats  larger  between  of  in  other  group  size  poaching a c t i v i t y  show  seasonal  1975, D o u g l a s - H a m i l t o n and H i l l m a n  average larger  (i.e.  and,  appeared to  understood.  aggregations  al.  the  aggregations  was no c o r r e l a t i o n  random a g g r e g a t i o n s  Variation studies  a habitat  However,  In  same h a b i t a t s .  in  form  numbers and d e n s i t i e s  found).  in  elephants  numbers  hypothesis.  led  found.  influenced  density,  relative  Size It  are  is  The  may b e m e d i a t e d  quantity  value  considered  as they  movement.  high  absolute  it  not  of  a r e a s of  nutritional  which  quantity  Group  the  are  densities  by  range  versus quantity  densities.  regardless or  their  to  1981).  has  changes But,  be c o r r e l a t e d  s e a s o n when a v a i l a b i l i t y  in  with of  Table 3.3. Average herd s i z e s r e p o r t e d f o r a v a r i e t y o f A f r i c a n e l e p h a n t p o p u l a t i o n s . Table adapted from Laws (1969).  LOCATION  AVERAGE GROUP SIZE  Tsavo N a t i o n a l Park, Kenya  13.5 13.0  G l o v e r (1963) Watson and B e l l  Amboseli N a t i o n a l Park, Kenya  25.0  Western and L i n d s a y (1985  Murchison F a l l s N a t i o n a l Park, Uganda  12.1 12.0  Buechner e t a l . (1963) Laws e t a l . (1975)  Queen E l i z a b e t h N a t i o n a l Park, Uganda  5.9  Laws and Parker (1968)  Luangwa V a l l e y ,  3.9  Laws and Parker (1968)  S e r e n g e t i N a t i o n a l Park, Tanzania (north)  52.0 54.5  Watson and B e l l (1969) D u b l i n (unpubl. data)  S e r e n g e t i N a t i o n a l Park, Tanzania (south)  19.0  Watson and B e l l  Masai Mara Game Reserve, Kenya (wet season)  22.5  D u b l i n (unpubl. data)  Masai Mara Game Reserve, Kenya (dry season)  13.2  D u b l i n (unpubl. data)  Zambia  SOURCE (1969)  (1969)  76  preferred sizes  forage  recorded  therefore, on  be t h e  mating  occurs (Hanks  1985).  gestation  pattern;  of  mating  to  estrous  and occur  in  different  herd could,  seasons  different  larger  average  size  group  or  levels  of  been a t t r i b u t e d  to  and  example,  means of  elephants  S. Andelman, data)  with certain of  suggest  that  behavioral  aggregations  comm.). the  would  on  low  wet  provide  suggested  their  long  Poole that  b a s i s as a establishing  merely  among  lives  (Moss  findings sounds  by K.  Payne  associated  may a c t u a l l y  previously an  the  formation.  or or  levels  was  and  and  herd  among e l e p h a n t s  similar  season.  groups  frequency  1969,  males  Moss  for  to  22-month  groups,  have a l l  Recent  herds  (Hanks  adult  long  Lindsay  a  rains  bonds  throughout  patterns  information  kin  the  and  on a p e r i o d i c  strengthening  pers.  the  basis  hence,  may s h o w a  (1981),  (1985)  of  cow-calf  cow-calf in  Moss  a strong  interact  the  containing  was h i g h e r  within  peak  follow  herds  may a g g r e g a t e  and  who may  then  exclusively  may p r o v i d e  join  during  herds  (1972),  individuals  Periodic  of  overall,  hierarchies  exchanges  Births or  the  and,  1983, Western  elephant  than  dominance  (unpubl.  before  size  maintaining  after  temporarily  Western and L i n d s a y  benefits  peaks. . Estrus  1983, Moss  Mara  Douglas-Hamilton  has a l s o  slightly  females.  1982).  considerably  size  and b i r t h  males  just  average  1981,  done  average  populations  experiencing  herd  or  time  the  (1983),  in  during  this  Weyerhaeuser  For  in  elephant  counts  elephants  1969, D u b l i n  At  access  social  of  variation  seasonality  gain  variation  different  result of  The  pressure.  Seasonal  rains  greater. for  populations  hunting  the  is  be  unrecognized.  opportunity  for  such  77  exchanges cited  between  predation  Western  and  benefits  which  information  the  on  no  relative  (1985)  larger  groups  through hard  is  desirable  for  local  supplies are  an  food  elephant,  large  groups  difficult  this to  rainy  for  the  various  of  these  Larger  r e a s o n s but In  could  season  (1975)  aggregation.  possible the  foraging  exchange  ideas.  factors  limited.  al.  emphasized that  support  condition  the  basis  but  task.  not  Laws et  may a c c r u e t h r o u g h  of  variety  a  discussed  evidence to  and tenure  a  as  facilitation  contributions a  and h e r d s .  elephants  Lindsay  or  currently  individuals  to  are  there  formation  seem  to  be  possible only  an a n i m a l  restrict  the  the  is  Untangling  group  herds  of  when  size  of  formation  of  pattern  of  only.  Summary In  summary,  Mara  habitat  selection  elephant  populations.  produced season, browse size than  large  result  of  were more herds food  the in  dry  Seasonal  likely breed not  hierarchies, Before  an  limited, and to any  chose  other  In  the to  result rains  of  re-establish  herd  which  in  bulls  to  which the  the  wet  provided  average  size  preferred  and herds  interact,  types  addition,  in in  African  grasses during  habitats,  forming  other  habitat  habitats  season. all  general  observed in  differences  indirect  during  the  nutritious  virtually  random a g g r e g a t i o n s  to was  dry.  sexes  to  larger,  the  Both  switched in  fit  formation  of  vegetation  in  and group  quantities  and both  was  elephants  wet  were  group season  not  the  habitats,  but  joining  cow-calf  congregating,  determine  when  dominance  bonds.  p r o g r e s s c a n b e made o n a s i t e - s p e c i f i c  course  78  of  management,  it  is  essential  that  elephant  habitat  utilization  and  elephants  on  habitats  fully  magnitude  these  of  their  such as weather, seasonal  overutilization must  be t a k e n  management protected  of into  of areas.  effects  poaching  migratory  be  account  local  to In  or  by  the  allow  impacts  of  The  local  dispersal  conditions  and  these and o t h e r  and  actual  accessibility  formulating  populations  of  understood.  future,  when  patterns  potential  may b e m o d i f i e d  routes  elephant  the  activity,  habitats.  the  plant  plans  of  avoid factors  for  communities  the in  79  CHAPTER 4.  FEEDING  ECOLOGY OF ELEPHANTS  IN THE MASAI  MARA GAME  RESERVE  Introduction The  past  woodland  30 y e a r s  habitats  have  of  east,  many p a r k s and r e s e r v e s , explanations controversy Caughley  for and  exponential linked  to  elephants  human-induced Both  Buss  Griffiths primary  converse.  vegetation  in  deal  Laws  during this  increase  in  of  1970,  that  the  century  is  expansion  of  in both c o n c e n t r a t i o n  have  a  frequency  are  widely  Laws 1969,  Angus  1971,  1975,  Guy  1961,  of of  Harrington 1976,  and Ross  Field  The p o t e n t i a l  fires  and Ross  synergistic  fire  to  Field  1974,  Laws e t Barnes  effects  of  in  Norton-  played  the  primary  (Eggeling  al. 1982, fire  Spence 1975,  al. and  Thomson  Smart e t  and  the  1947,  Lamprey e t  1971,  the the  to m a i n t a i n  1962,  1970,  is  supports be  acting  thereafter  1976,  documented  and  research  B r o o k s and Buss  Wing and Buss  that  thought  with  phase  (1970),  idea  but most  i n the  contribution  Gough  s u p p o r t e d the  grassland  and Dawkins  relative  and  woodland change,  1967,  1985).  a great  agree  played roles  their  Lawton  Elephants of  Buechner  have  i n woodland l o s s ,  initiators  and an  In  Theoretical  subsequent  resulted  areas,  though  (1961),  (1979)  role  The  the  Africa.  1969,  researchers  loss. has  to  (Pienaar  many  and e l e p h a n t s  losses,  debate.  led  in  fires.  fire  woodland  have  human p o p u l a t i o n s  protected  and s o u t h e r n  changes  and s e t t l e m e n t  into  m a j o r change  declined.  speculation  woodland  cultivation  central,  of  have  However,  g r o w t h of  a time  woodlands  these much  1976).  been  al.  elephants  80  have  not  yet  Over  been  the  past  Serengeti-Mara taken  place  example Ruaha  (Savidge  and  1970,  Field  the  disappearing.  eastern,  1968,  1982)  1961,  southern, Kenya  Barnes  and t h i c k e t s Similar  1969,  1983)  1971,  Sengwa 1976,  1966,  1981).  Queen  The s e p a r a t e elephants until  In  this  conjunction exclosures these  press).  of  relative  study,  At the  Kurji  data  Ross  woodland  1963, and Laws  1976)  with  and  1974,  intensive  concentration  1976,  same t i m e ,  and  unable  Dublin  fire  to and  frequency  climatic  factors  N o r t o n - G r i f f i t h s 1979).  of  man-induced  fire  and  in these h a b i t a t  changes,  but,  has  an  prevented  accurate  importance.  controlled  fires  al.  habitats  Guy  1976,  effects  elephant  and v a r y i n g f i r e  of  1973,  implicated  lack  with  into  b o t h human a c t i v i t y  been  their  areas  (Anderson  1970,  Buechner  Zimbabwe ( A n d e r s o n and W a l k e r  peripheral  in  experimental  structuring  and  and s u b s e q u e n t  and s y n e r g i s t i c  of  Field  1983),  Elizabeth  Field  boundaries  recently,  assessment  1975,  1968,  on p a r k  al.  have  and P a r k e r  coincided  due t o  et  Buechner et  for  Manyara  1961,  c h a n g e s have  Douglas-Hamilton,  (Langridge  Area,  numbers  increased  al.  1962,  Africa,  Lake  These  from  their  Laws  Laws e t  Research  elephants  support  B r o o k s and Buss  the have  Parker  and  Tanzania,  in  events  and c e n t r a l  (Laws  1982,  National Parks,  Savidge  human s e t t l e m e n t  has  been  woodlands  M u r c h i s o n F a l l s N a t i o n a l P a r k s , Uganda (Buss  Buss  of  years,  i n Tsavo N a t i o n a l Park,  and Dawkins  Guy  30  have over  (Weyerhaeuser and  analysed.  feeding  experiments  regimes. and  b e h a v i o u r was using  I consider  mammalian  communities  observed  the  browsing effects  of  in  the  herbivores (Chapter  in  5).  Direct  81  observation  of  free-ranging  years  (1982-1985)  i n the  were  accustomed  to  male  and  female  activities. resident Reserve  elephant  length in  of  the  in the  describes:  composition;  b)  and v e h i c l e s . without  investigated  current  a)  seasonal  where  half  elephants  I collected  data  their  normal the  impacting vegetation  in  the  feeding  patterns  are  woodland r e g e n e r a t i o n .  the  on  in  and s e x u a l  and d)  a  increases  and r e l a t i v e  bouts;  and  disturbing  elephant  i n h i b i t i o n of  absolute  foraging  two  how r e c e n t  p o p u l a t i o n may be  whether  playing a role chapter  humans  study  covered  M a s a i Mara Game R e s e r v e  elephants  This  and,  elephants  diet  differences  diversity;  characteristics  This  in  diet  c)  average  of  plants  -  October  diet.  Methods Observations 1982,  1983,  covered  and 1984)  and two wet  83  and  dry  s e a s o n a r e more c o m p l e t e  the  loss  periods during of  1983-84).  three  an  the  some  the  seasons wet  than the  original  seasons  season  previous  data.  and more t h a n  1,000  the  study  (Table 4.1).  Each o b s e r v a t i o n  By c h o o s i n g rather  than  potential  patterns. individuals  block  a different  following problem  be o b s e r v e d  one of  Short observation to  of  seasons  More t h a n of  15-30  500  females  to  observation  consisted  with each  individual a l l  day,  focal  observation I  avoided  elephant  feeding  also allowed month.  due  period  minutes  idiosyncratic  1984  recorded  each  i n any g i v e n  1982-  were  i n d i v i d u a l for  periods  - May  1983-84 and t h e  bulls  uninterrupted  (June  (November  of  animal. period  of  Data f o r  dry  more  different  82  T a b l e 4.1. Study p e r i o d s , number o f f o c a l a n i m a l s , s e a s o n a l and mean monthly r a i n f a l l .  SEASON  FOCAL ANIMALS M  F  TOTAL RAINFALL  MEAN MONTHLY  (mm)  (mm)  Dry 1982  75  156  196.6  39.4  Dry 1983  104  221  378.1  75.6  Dry 1984  82  175  221.3  44.3  Wet 1982-83  152  289  775.7  110.8  Wet 1983-84  127  230  540.9  77.3  83  Focal  elephants  were  observed  25 m e t e r s .  When n e c e s s a r y  for  species  forage  vehicle  in open,  exposed  were  individuals  were c h o s e n  driven were  from  in  high  through the  were  were  mobile  more  disappeared in  the  possible females The  i n the  observed  on  or  woodland  areas.  or  the  adult  the  E a c h month  sex  I  This  Serengeti-Mara Diet  choice  each  season,  differences  was  between  seasons  diet  average  characteristics  of  were  on  Bulls  individuals remained  to  sample  was not skewed  always towards  constitutes  other  and  analyses  between  each  sex  the  the  dry  for  the  Norton-Griffiths  within  of  1979).  sexes and  within overall  f o r a g i n g were a n a l y z e d :  diversity,  duration  plants  r e f e r r e d to as  year.  components  absolute  performed  with  (Sinclair  throughout the  evenness,  October  examined f o r d i f f e r e n c e s  following  composition,  through  consistent  system  resident  year.  June is  transects  others  this  p e r i o d from November t h r o u g h May i s while  some  r a t i o was h i g h l y  were  year.  attempted  however,  Focal  many  throughout the  while  when  No i n d i v i d u a l s  though  a time,  from a  that  random  and  used  foot  herds  along  herds  f o r months a t  sexes,  from  month  cow-calf  both  b i n o c u l a r s were  e a r l y morning.  per  exceeding  I  points  once  than  of  throughout  season,  The  but  observed repeatedly  because  season.  diet  Reserve  not  I observed elephants  bush  vantage  entirely  numbers  Leitz  on a d a i l y b a s i s  Reserve y e a r - r o u n d .  equal  wet  areas,  thick  s a m p l e d more t h a n  individuals  x 40'  identification.  elephants  located  10'  from a d i s t a n c e  i n the  percentage  of  relative foraging  diet. data  diet  where  diversity  bouts,  Arcsine  diet  and  or the  transformations  necessary.  All  84  statistical by  Zar  analyses  were p e r f o r m e d  (1984).  Absolute dietary d i v e r s i t y species  included  season.  Relative diversity  expresses  in  the  the observed  possible diversity the  using techniques described  diet  i s the  observed of  diet  total diets or  number  i n a g i v e n month o r diet  devoted  to  each  forage  species u t i l i z e d ,  f o r a g e s p e c i e s consumed.  a v a r i a t i o n on t h e S h a n n o n - W i e n e r d i v e r s i t y  and  is referred  t o as r e l a t i v e d i v e r s i t y  This relative diversity  a similar  way  J  values  of  =  J  index  "evenness".  general  (Zar by  i s d e f i n e d by  In t h i s Zar  1984) Pielou (1982)  study I  (1984) a s :  indicate  more  even  foraging  B e c a u s e k i s g e n e r a l l y an  However,  foraging patterns  i t i s a useful  of  males  and  index  among  the  underestimate  t h e a c t u a l number of s p e c i e s i n t h e d i e t , J w i l l  diet  index  u s e d by B a r n e s  diets.  of  -|pi*log(pt)/log(k)  selected plant species. of  J,  k i s the  This  or evenness  i n d e x was  for a n a l y z i n g elephant  h a v e u s e d t h e same f o r m u l a w h i c h  Higher  "evenness",  i n a g i v e n data s e t , pt i s the p r o p o r t i o n  is  in  plant  d i v e r s i t y a s a p r o p o r t i o n of t h e maximum  number o f d i s t i n g u i s h a b l e  (1966).  of  overestimate f o r comparing  females  in  the  two  seasons. The animal was  l e n g t h o f a f e e d i n g b o u t was  focal  ate a given forage type before s w i t c h i n g to another.  assumed t h a t t h e l e n g t h o f a  correlated bout.  t h e amount o f t i m e a  to  the  total  Average feeding  feeding  bout  was  It  positively  amount o f f o r a g e consumed i n a g i v e n  bouts  were  calculated  for  males  and  85  females  in  b o u t s were of  each  bouts  Forage use, condition,  part  eaten,  were  recorded.  leafless,  dry  with  flowers,  into  the c a t e g o r i e s :  r e c o r d e d as  t y p e and d i v i d e d  leaves  food  and t h e  leaves,  roots,  : 0 -  the  feeding  by t h e number  green  fruits.  leaves,  which f o r a g i n g  2m, 2 -  conditions  budding,  Plant parts  l e a v e s and r o o t s ,  1 -  at  phenological  phenological  o n l y and bark  1m,  plant's  height  Plant  and l e a v e s w i t h  b r a n c h e s and t w i g s , were  Then a v e r a g e  observed.  with regard to  plant  place  included:  and e a c h s e a s o n .  summed f o r e a c h f o r a g e  individual  took  month  were  broken  and  shoots,  Foraging  heights  leaves  only.  leaves  3m, and above 3m.  Results  Feeding d i f f e r e n c e s  Diet  c o m p o s i t i o n by f o r a g e Seasonal  females. the  five  and  forage  shrubs,  (grasses,  in  females  differed  herbs,  a l l d r y and wet  between  bull  males  and  and cow d i e t s  over  shrubs,  trees,  seasons  and  sampled  tree  (Figures  4.2).  when  of  females  these  seasonal  were compared f o r t h e season,  and  type  composition  types  varied  The s o u r c e s apparent  diet  males  The f r e q u e n c y d i s t r i b u t i o n s o f  seedlings) 4.1  between  seasonal  means  w h i l e males ate  differences  for each forage  1983-84 d a t a  devoted  diet  (Table  type  4.2).  s i g n i f i c a n t l y more t i m e  more t r e e  seedlings  and  (Figure In  to  became  the  4.2) wet  f o r a g i n g on  trees.  Males  86  DRY males  WET females  males i females  1982 1982-83  1983  1983-84  SHRUBS  ^^TREES^X ^-SEEDLINGS-^  7\  ^GRASSES  1984  F i g u r e 4.1.  HERBS  Mean p e r c e n t a g e d i e t c o m p o s i t i o n by f o r a g e type f o r females and males. The d a t a c o v e r s t h r e e d r y seasons (1982, 1983, 1984) and two wet seasons (1982-83 and 1983-84). The s t a r s compare f e e d i n g p a t t e r n s b o t h between sexes and s e a s o n s .  87  50-|  O  E-  40-  1  30 H  o  20 H  X  + - DRY SEASON, FEMALES  +  O - WET SEASON, FEMALES X - DRY SEASON, MALES  O  •  •  o •  + X  O  - WET SEASON, MALES  + o  • + X  X  o S H R U B GRASS SEEDL HERBS TREES FORAGE T Y P E  F i g u r e 4.2.  Mean p e r c e n t a g e d i e t c o m p o s i t i o n by f o r a g e type f o r females and males i n t h e 1983-84 wet season and t h e 1984 d r y season.  88  Table 4.2.  FORAGE TYPE GRASSES  HERBS  SHRUBS  TREES  SEEDLINGS  Mean d i e t composition comparisons between sexes f o r a l l forage t y p e s . Tukey's m u l t i p l e comparison t e s t s (with unequal sample s i z e s ) were performed. Significance l e v e l s a r e p r e s e n t e d . (Dry season f o c a l "n" = 257; wet season = 357).  SEASON  MEAN % FEMALES  MEAN % MALES  SIGNIFICANCE LEVEL  dry  22  15  n.s.  wet  33  28  n.s.  dry  12  7  0.01  wet  15  15  n.s.  dry  45  47  n.s.  wet  38  30  0.01  dry  4  9  0.05  wet  1  5  0.05  dry  17  22  0.05  wet  13  22  0.01  89  and  females  the  wet  season  d i d not  season,  April  sexes.  However,  on  the  significantly dry  season  Females  i n the  while  than  also  a whole,  At t h e  early  dry season,  greater  did  (t  s p e n t more t i m e  males  devoted  =  more  time  of  the  tree  dry long both  females  time  feeding  = 99,  P<.02).  i n the  eating  in  late  p r e f e r r e d by  d.f.  on h e r b s  to  the  their  2.37,  feeding  or  June and J u l y ,  p r o p o r t i o n of  males  on g r a s s e s  height  and May, new g r a s s e s were h i g h l y  a significantly  grasses  differently  as  from A u g u s t t h r o u g h O c t o b e r .  rains,  spent  feed  dry  season  s e e d l i n g s and  trees. In browse  general, components  overall  female  difference P<.05)  was  and  P<.02)  d.f.  =  seen (t  in  the  diet  in a highly  wildebeest  were not  rains  April the  wet  d.f. (t  the  Grasses  i n the  P<.025),  this  the  manner  Themeda i n the  before  355,  seasons.  ( g r a s s e s and  (t  d.f. =  3.03,  c o n t r i b u t e d more  year. -  was  was  the  most  to  the  important  Themeda t r i a n d r a  during  and  e a t e n more f r e q u e n t l y  Reserve,  that  migrants  is  arrived.  p r e f e r r e d h e r b b u t more t h a n a d o z e n  other  was  immediately  during  November and December and d u r i n g t h e  and May, j u s t  =  = 2.35,  dry season diet  than  P<.01)  (t  in  trees)  d.f.  species  year  higher  herbs.  throughout  present  255,  herbaceous  Cynodon d a c t y l o n  seasonal  rains.  of  than  2.09,  =  the  and  = 612,  =  d.f.  more t h r o u g h o u t  grasses,  the  was  utilized  seedlings,  2.45,  2.67,  P<.01).  following  of  =  difference  Among t h e  short  =  were s i g n i f i c a n t l y  and p a r t i c u l a r l y d u r i n g t h e  255,  species  (t  tree  i n both the  females  observed  chosen  diets  significantly  612,  this  (shrubs,  dry  Conversely, herbs)  o v e r a l l male d i e t s  long  when the rains  Beeiurn s p . species  were  90  eaten  during  shrub  species  dichogamus, seedlings  short  of  were  divornum,  Acac i a  the  shrubs,  diets trees,  s e a s o n s but the  brevi spica,  Solanum i n c a n u m . Dichrostachys  were  dry  diversity changed  important  and t r e e  grasses could only  season.  Figure  Croton T r e e s and  c inerea,  in  and  elephant  diets  this  is  underestimate  an  differences  in elephant  To t e s t  if  diet  number  compared w i t h significant  the  correlation, 0.07,  diet  n = 12,  diversity  not  a function  in  a  given  was of  (Table  g  across 4.3).  of  the  to  genera  absolute  forage  dietary  was  was  eaten 0.22,  females  n = 12,  of in  observation the  each  of  (Figure  n.s.)  From t h i s ,  amount  significant  diets  month.  f o u n d between o b s e r v a t i o n  for  I  time  the  4.4,  or males assumed  measure  of  diet  elephants  year  and i n  both  the  time  that  dry  No and  4.5,  absolute  preference were  was  Spearman  (Figure  Females m a i n t a i n e d a s i g n i f i c a n t l y the  during  detectable.  recorded for  all  Although  components,  time  in  number of  types.  independent  species  an i n d e p e n d e n t the  Identification  s p e c i e s was p o s s i b l e  all  observation  n.s.).  month.  than males  seasons  r =  year.  presents  of  plant  total  species  to  identified  across  diversity  of  contributing  f o o d c h o i c e s were s t i l l  correlation  number of  rank  diets,  species  the  to  be  4.3  in the  the  plant  seedlings  species  time  of  throughout  plant  diet  and  Among t h e  diversity  elephant  g  availability.  year.  The a b s o l u t e  r =  of  Acac i a  gerrardi i ,  trichocarpum  throughout  the  periods  preferred  Euclea  Ormocarpum  Diet  their  and  observed broader and  wet  35 n t—>i Q  W  30  H  > - females I o - males  Ei—i  CO  w o w  25  CO  20 H  o w m S  15  \  I—I  10  H  o  •  o—o  H — i — i — i — i — i — i — i — i — i — i — i — | —  ND J F M A M J J A S0 MONTH  F i g u r e 4.3.  A b s o l u t e d i e t d i v e r s i t y (number o f s p e c i e s i n the d i e t each month) f o r females and males i n 1983-84.  92  FEMALES  35 -i  r 30  H  25  H  20  H  s  = 0.22,  n.s.  o w CO  o W  PQ 15  — i 300 400  1 500  r—i 600  700  i  i  800  900  TOTAL FEEDING TIME (min)  Figure 4 . 4 .  The number of s p e c i e s eaten v e r s u s t h e t o t a l f e e d i n g time o b s e r v e d i n each month f o r female e l e p h a n t s .  93  28  MALES  26-  CO W  o w PU  r  c  n.s.  = 0.07,  24-  r—I  22 20-  CO  O w PQ S &  1816-  14  i 150  | 200  i  i 250  -  i 300  •i 350  1  I 400  TOTAL FEEDING TIME (min)  Figure 4.5.  The number of s p e c i e s eaten v e r s u s t h e t o t a l f e e d i n g time observed i n each month f o r male e l e p h a n t s .  Table 4.3.  A b s o l u t e d i e t d i v e r s i t y comparison between males and females by season.  d.f.  FEMALES  MALES  t-value  p-value  22  25.17  19.75  3.75  0.01  Dry season  8  28.20  20.60  3.43  0.01  Wet season  12  23.00  19.14  2.59  0.05  Annual mean  95  Relative distribution available between  of  diet  (Figure  (Table time  However,  c h o i c e s to  4.6).  4.4).  in  among  the  months  of  fell  below  .65  came a t  the  end of  was  low  quantity  diet  "evenness" indicates a  few  dietary were  the  "evenness"  numbers  No s i g n i f i c a n t  and  on  or  On a v e r a g e  evenly  diversity months  diversity  males and females a c r o s s the  seasons their  diet  preferred  diversity  all  all  and  spells  differences  found  wet  and  types.  relative  These  forage  quality  c h o i c e was r e d u c e d .  those  fed  two  The  low  disproportionately  months.  browse  diet  females.  when  dry  distributed forage  April .50 -for  elephants  was a n a l y z e d f o r  significant  in  the  choices  were  the  species  and f o r a g e  species in  in  males  dry  that  or  f e m a l e s and  males and  long  different  differences  year  November for  of  relates  When  species  relative  alone  there  between males and females  (Figure  4.7) . Averaged across the  year,  were  distributed  d.f.  = 22,  when  males chose a s i g n i f i c a n t l y  = 3.51,  d.f.  more  relative  P<.02).  This  = 12,  had  similar  n.s.).  The  lower the  continually  spent  time  on  however, their  a  preferred  browsing  in  the  number  demonstrated In  species to  browse.  of  occurred  the  dry  patterns  diets  of  of  of  females in  the  (t  than  season, (t  = 0.12,  of  indicated  amount  of  changes  in  season,  they  However,  in  the  =  2.55, season  females males d.f.  browse  preferred  males  wet  females  highly  dry  diets  even d i e t  diversity  seasonal the  those  more  a disproportionate  small  browsing.  In  relative  year  than  difference  P<.01).  females  throughout  evenly  browse  (t and  =  8,  species that  their  they  browsing  species.  Males,  the  "evenness"  of  also  selected a  few  wet  season,  when  F i g u r e 4.6.  R e l a t i v e d i e t d i v e r s i t y or d i e t "evenness" f o r female and male e l e p h a n t s i n 1983-84.  Table 4 . 4 .  Seasonal comparison o f r e l a t i v e d i e t between the sexes.  diversity  d.f.  FEMALES  MALES  22  .73  .80  1.20  n.s.  Dry season  8  .70  .79  0.57  n.s.  Wet season  12  .75  .80  1.73  n.s.  Annual mean  t-value  p-value  98  1.0 n Et — I  CO  DH  0.9 H  W  > i—i  Q O  CO  0.8 H  0.7 H  o DH  PQ  0.6 H  > i—i  0.5  •  -  f e m a l e s  O  -  m a l e s  DH "1  I  l  I  l  I  I—l—I—i—i—i—i  ND J F M A M J J A S 0 MONTH  F i g u r e 4.7. R e l a t i v e b r o w s i n g d i v e r s i t y o r "evenness" f o r female and male e l e p h a n t s i n 1983-84.  99  their  diet  apparently  was  largely  comprised  sampled  small  amounts  feeding  bouts  on  of  of  a  herbaceous wide  species,  variety  of  males  available  species.  Lengths  one  of  Males  and  forage  type  This  applied & b).  4.5a no  females before  an  significant  between  the  see  analysis  sexes  across  of  plants  whether  the  c r i t e r i a ,  u t i l i z a t i o n  the  plant,  at  chosen which  found  feeding  between  properties,  could  not  being  made  other  any  and  forage  rule  out  which  I  For  leaf  stage  was  was  selected  type, the  to  plant  to  of  bouts).  year  types,  length  (Table  there  were  foraging  bouts  either  in  food  terms  parts  season.  using  of  for  any  any  plant  same  condition  and  these  season.  more  the  the  of  height  differences  of  parts  that  the  eaten,  the  green  phenology  shrubs,  entire  on  were  conditional Although  used subtle  were  no  found,  choices  I  were  detect.  in  exclusively.  according  forage  their  in  unable  of  the  significant  females  forage  feeding  (foraging  within  plant  was  herbs,  used  chose  No  in  time  diet  p o s s i b i l i t y  categories  available.  or  the  u t i l i z e d  across  examined  place.  differences  Elephants a l l  took  of  another  average  the  sexes  amounts  a l l  year  specific  males  for  significant  the  and  in  in  was  to  across  the  types  similar  seasons  differences  Characteristics  To  spent  switching  between  In  forage  trees, For  and  leaf (see  So,  more  Methods),  seedlings,  grasses,  a v a i l a b i l i t y .  stage  leaf more  the  than when green  condition dry  leaves  Table 4.5.  (a)  Average l e n g t h o f f o r a g i n g bouts across a l l f o r a g e types f o r males and females throughout 1983-84 (a); and between sex comparisons ( b ) .  Average l e n g t h o f f o r a g i n g bouts i n minutes  N  D  J  F  M  A  M  J  J  A  S  O  Females  5.0  3.6  2.5  2.8  3.7  3.0  3.5  7.7  6.8  3.9  5.7  3.4  Males  3.7  4.4  2.4  2.9  3.8  3.0  4.5  6.4  6.0  4.1  5.0  4.2  (b)  Between sex comparisons o f mean l e n g t h o f f o r a g i n g bouts i n each season.  d.f.  Females  Males  t-value  p-value  Annual mean  22  4.3  4.2  0.17  n.s.  Wet season  12  3.4  3.5  0.22  n.s.  5.5  5.1  0.39  n.s.  Dry season  101  were consumed d u r i n g wet  season.  leaf 1,  the  P<.005)  and wet  difference  bulk  are better  able  differences  females  season  supports  of  Male  the  to  and  and  female all  dry  Feeding height  feeding  the  time  1m h e i g h t  seedlings hour.  fed almost  patterns  Their  Both sexes  season presented  i n the  1971), These  on  occasional  use.  in small  the use  i n any s e a s o n ,  p a t t e r n of  = 80%,  (males  activity  of  female in  five  the  the  = 63%,  but  Shrub amounts,  In a s i n g l e removed  a rate  differences  of  of  their  i n the  = 60%)  removing  h i g h e r and h i g h e r  sections.  and s h r u b s .  = 79%)  females  selectively  minutes,  trees  majority  females  elephants  Seasonal  following  to  spent  (males  foraged at  progressed.  (Sikes  exclusively  significant  apply only  elephants  and 1m)  greater  grasses.  with only  same g e n e r a l  and on t r e e s  (below  This  below.  types  was n o t  dry  d.f.  their  demands  the dry  P<.001).  with  consumed by b o t h s e x e s  on s h r u b s  male  = 1,  quality,  forage  the  d.f.  in  i n the  = 8.59,  2  energy  f o r a g i n g was t e r m e d " w e e d i n g " .  individual  leaves  season.  year,  class  2m c l a s s .  while  poor  d r y (x  males,  elephants  Root f e e d i n g  i n the  Throughout  that  relative  utilize  b a r k was a l s o  primarily  1 -  idea  and cows f o l l o w e d tree  = 46.64,  2  a r e a n a l y z e d more f u l l y  roots.  bulls  in both the  (x  and l o w e r  l e a v e s and s h o o t s of of  and more g r e e n  M a l e s consumed s i g n i f i c a n t l y more g r a s s  c o n d i t i o n than  intake  d r y season  in  bout,  a s many as over  within  the  seedlings  foraging  levels  0 -  400 as  the  sexes  35 per dry are  102  Female  Diet  feeding  patterns  c o m p o s i t i o n by f o r a g e  type  Females changed t h e i r d i e t two-way  analysis  showed t h a t both  the  season  P<.01).  variance  diets  of  =  5.59,  (F  T h e r e was a l s o  and f o r a g e the  of  type  of  season  between  females  P<.05). on  season  4.2).  and f o r a g e  types  (F =  interaction is  relative  type with 3.44,  i n t e r a c t i o n between  This  A  significantly  and f o r a g e  a significant  season  (Figure  differed  P<.025)  (F = 2 . 4 1 ,  influence  with  season due  q u a l i t y between  to  forage  types. Dry  season  diet  between  1982  decrease  in browsing ( t r e e s ,  increase due t o  in  and  composition  grazing  unseasonably  Wet  season  This  difference  rainfall wet  i n the  season  therefore, the  also  was  1982-83 wet when  prevailed.  probably  data,  desirable.  higher pressure  in to  and  between to  season  the The  the  it  been  that  period.  two s e a s o n s  sampled.  significantly versus  largely  In t h e  was a p p a r e n t t h a t other  1983-84 wet due  1983 may have  wet  to  hand,  season a  season  of  and  of  feeding  dry be  in  1982-83  rainfall  the  females  of  1983-84  appeared to  when f e m a l e s  shortage  higher  the  failed  more a b u n d a n t due t o h i g h e r  On t h e  browse  the  main d i f f e r e n c e s  types.  The  and  during  (775.7mm)  rains  pattern  (Figure 4.1).  in  the  the  similar  seedlings)  (378.1mm)  constituted a larger proportion  1983-84  highly  varied  attributed  (540.9mm)  were  shrubs,  a  1983  ( g r a s s e s and h e r b s )  h e r b and s h r u b f o r a g e  herbs  but d i f f e r e d i n  high r a i n f a l l  diets  conditions the  1984  showed  diet.  and, From  f o u n d new h e r b s on  shrubs  was  were u n d e r more  green,  herbaceous  103  forage. C o m p a r i s o n of shows  a general  higher  rainfall  season,  the  tendency  grasses  dry  in  and l a t e  than  P<.001)) 173,  and  and  wet  trees  and  (32.9%  more  4.2  the  = 3.39,  proportion (June and  In the  late  significantly d.f.  (50.8% (t  Table  differed  early  (3.7%) was  =  = 2.83, the  335,  d.f.  wet  season  s e e d l i n g s c o n t r i b u t e d more t o  a significant  of dry  and 4 . 8 , also  the  in  season  d u r i n g the  diets  important  dry  in times  dry season.  (t  dry season  of  females  expanded  s e a s o n and n a r r o w e d i n t h e  4.3,  t  diets  = 3.05, of  types  of  diet  into  composition  dry  (t  = 1.14, d.f.  diversity  t h e s e months  quality  d.f.  = 10,  females d i d not  = 0.47,  diet  diet  To f u r t h e r a n a l y z e ,  tree  to  1984  dry  degree.  diversity The d i e t s  (t  to  through October)  early  the  (Figures  grasses  s e a s o n of  but not  i n the  was d i v i d e d  H e r b s were  season d i e t s ,  Diet  of  grass  i n the  P<.01)).  diets,  the  i n the  browse shrubs  diet  (August  season to  a more h e r b a c e o u s  seasons.  the  season d i e t s ,  less  from  The c o n t r i b u t i o n  between d r y and wet  July)  for  and i n c r e a s e d  particularly  4.6a).  of  1983-84 wet  d.f. = 10,  for  Females  travelled  shrubs  and  was  trees  to  The  n.s.)  n.s.).  the  permanent  or  (Table  for  and  water's  edge.  the  and F i g u r e  all  lowest  alone  relative  November.  where  was  they  The low d i e t  Each  when  activity  of  forage  species  long dry s p e l l s  sources  in  diversity  across  their  browsing  water  4.3  browse  i n M a r c h and  end o f  diversity  relative  Cows showed  species  scarce  by t h e  season  v a r y by s e a s o n e i t h e r  represented  forage  wet  P<.02).  = 10,  browse  in absolute  high high.  fed  on  "evenness"  104  90 n CO E1  r = -0.96 s  p < .001  O O  80  • - females O - males  70 CO  o PS  60 O  PQ E-<  50O  W O PS W 4030  — I — i — | — i — |  20  40  60  80  100 120 140  MEAN MONTHLY RAINFALL (mm)  F i g u r e 4.8.  The r e l a t i o n s h i p of percent browse i n the d i e t t o the mean monthly r a i n f a l l f o r female and male e l e p h a n t s . Note : * = both males and females showed equal percentages of browse i n the d i e t s when mean monthly r a i n f a l l was 75.6mm.  105  T a b l e 4.6.  (a)  Between season comparison o f d i e t s by forage type w i t h i n females and males. Tukey's m u l t i p l e comparion t e s t s (with unequal sample s i z e s ) were performed. Significance levels are p r e s e n t e d .  FEMALES (dry season = 175; wet season = 230) FORAGE TYPE  DRY  WET  SIGNIFICANCE LEVEL  GRASSES  22  33  0.01  HERBS  12  15  n.s.  SHRUBS  45  38  0.05  4  1  n.s.  17  13  n.s.  TREES SEEDLINGS  (b)  MALES (dry season = 82; wet season = 175) FORAGE TYPE GRASSES HERBS SHRUBS TREES SEEDLINGS  DRY  WET  SIGNIFICANCE LEVEL  15  28  0.001  7  15  0.01  47  30  0.01  9  5  n.s.  22  22  n.s.  106 indicates  that  females  chose  a few,  f a v o u r e d browse  species  and  f e d d i s p r o p o r t i o n a t e l y on them.  L e n g t h s of  feeding  Females particular bouts  of  4.7a).  bouts  did  not  forage  types  differences.  bouts  longer  i n the  seasons, between  a  closer  feeding  early  bout  on g r a s s e s ,  of  minutes  the  diet  during  bouts  = 0.42,  d.f.  Characteristics All only  when  differed choice  forage in  of  wet  wet  seasonal  Although between  feeding  the  wet  important  10.2  they  (t  those  dry  differences  per  In  foraging only  = 6.33,  major  H o w e v e r , by t h e to  and  averaged  difference were a  bouts  seasons.  minutes  season,  were  d.f.  component  late  dry  season  i n the  wet  season  n.s.).  plants  green  all,  t r e e s and s e e d l i n g s  grasses  period.  (Table  but n o t  and e a r l y d r y  a significant  foraging  d r y season  significant  showed  s i m i l a r in length  types,  between  these  d u r a t i o n on  fact,  on s e v e r a l ,  significantly  d r y season  = 8,  i n the  dry season.  i n the  this  were  In  cows a v e r a g e d  while  Early  seasons.  on h e r b s ,  i n the  season  per bout,  P<.001).  feeding (t  dry  bout  patterns  examination  patterns  the  = 7,  differ  constant  longer  for  Foraging  on g r a s s e s d i d not  a  significantly  accounted  significantly  types  in d i f f e r e n t  V a r i a t i o n s in feeding  forage  4.34  maintain  types  cows were  on f o r a g e  in the  w i t h the leaf.  seasons  diet exception  The  grasses,  condition  (x = 835.6, 2  of  d.f.  was d e t e r m i n e d by what was a v a i l a b l e .  of  were  grasses =  When  eaten eaten  1, P<.001) green  and  grass  107  Table 4.7.  (a)  (b)  Between season comparisons o f the l e n g t h o f f o r a g i n g bouts on d i f f e r e n t forage types w i t h i n females and males. Tukey's m u l t i p l e comparison t e s t s were performed. Significance l e v e l s a r e p r e s e n t e d . Mean l e n g t h s a r e measured i n minutes.  FEMALES FORAGE TYPE  DRY  WET  SIGNIFICANCE LEVEL  GRASSES  6.3  4.3  n.s.  HERBS  2.9  2.0  0.01  SHRUBS  4.6  3.7  n.s.  TREES  4.5  1.6  0.01  SEEDLINGS  3.5  2.1  0.05  OVERALL  5.5  3.4  0.05  FORAGE TYPE  DRY  WET  SIGNIFICANCE LEVEL  GRASSES  6.9  3.4  n.s.  HERBS  2.0  2.8  n.s.  SHRUBS  4.6  2.6  0.01  TREES  6.9  3.6  0.02  SEEDLINGS  3.7  3.3  n.s.  OVERALL  5.1  3.5  0.01  MALES  108  leaves  were  green  leaf.  made  up  However,  87% o f  weather, the  the  females  diet  where  u b i q u i t o u s , over  as  grasses  green  leaf,  some g r e e n g r a s s e s  new  stems  throughout  of the  There  seasons  (Table  shrubs their  the  primarily  at  particular  In  the  to  season,  proportions  preferred  newer  patterns  Diet  c o m p o s i t i o n by f o r a g e  two-way diets seasons P<.01).  s h o o t s n e a r e r the  analysis of  diet  were  used  in  female  leaves.  whether  different elephants  sprouted cows  of  This  from  expanded  other  of  and  the  test  heights  was p r o b a b l y a d i r e c t c o n s e q u e n c e  feeding  3.52,  less  Leaves  In t h e d r y s e a s o n ,  include greater  in  in  which f r e q u e n t l y  level.  component  and woody b r a n c h e s  to  1m  dry  p e r e n n i a l water  predominated  wet  stems  severe  equally.  shrubs  feed  in  round.  on  Male  between  bark  of  data  of  the  grass  sufficient  availability  that  the  season.  therefore,  dry season  A  types  and  i n t h e most  dry  the  and,  10% o f  c h o s e n were  leaves  i n the  4.8a).  f o r a g i n g to  Even  a l l plant parts  new s h o o t s a n d l e a v e s  below  classes  eat  Roots,  preferred  dry  c o u l d be f o u n d y e a r  year.  were  grasses  by m o v i n g t o a r e a s  forage  females  preferred  eaten.  all  s m a l l amounts,  the  dry season,  m a i n t a i n e d over  Females d i d not  in  i n the  80% o f  height  change  the  in  reduced  ground.  type  of  bulls  (F = 1 5 . 7 2 ,  v a r i a n c e of were  diet  composition  significantly  P<.0005)  different  showed both  and between  forage  types  (F  The f r e q u e n c y d i s t r i b u t i o n o f  forage  types  in  Table 4.8.  (a)  Feeding h e i g h t p r e f e r e n c e s on shrubs and a t e s t o f the d i s t r i b u t i o n o f f o r a g i n g h e i g h t s between seasons f o r females and males.  FEMALES TOTAL MINUTES  HEIGHT CLASSES 0-lm  l-2m  2m+  Dry season  855  249  83  1187  Wet season  1142  188  13  1343  X* = 91.53 d.f.  = 2  p < .001  (b)  MALES TOTAL MINUTES  HEIGHT CLASSES 0-lm  l-2m  2m+  Dry season  387  115  8  Wet season  538  55  546 611  X2 = 53.25 d.f. p  = 2  < .001  110  the  diet  varied  (Figure of  4.1).  within  The l o w e r  1 9 8 3 may h a v e  than  in  higher  either  change  quality  the  time  late  dry  diets  which  season  (t  in  and  eating  a  in  d.f. d.f. the  reduced a v a i l a b i l i t y  In  but  dry  of  the  4.6b).  not  diet  Trees  comprised  throughout  the in  year,  one hour  to  This  were  more  The  season and  the failed  pattern  of  the  in  increased  for  only  4.4%  32.7% i n  the  and  in  depth.  females.  27.8%  of  to  the  were  also  so.  the  wet  in  decline of  was  However,  and males  a  season  it  no is  quarter frequently  feeding.  more  the  of  their  to  up  and  season  change note  overall  consumed  higher  4.2 dry  seasonal important  and  changes.  (Figure in  its  wildebeest  phenological  utilized  There  almost  wet  male dry  The s i g n i f i c a n t  of  In  early  s h r u b s made u p a s i g n i f i c a n t l y the  rains  more  seen  arrival  due  the  relative  s e a s o n was b o t h a p r o d u c t  in  of  4.1).  that  P<.02).  than  u s e among m a l e s .  seedlings  seedlings  dry  following  season,  significantly  seedling  late  period  activity  analyzed  P<.001)  = 174,  that  4.1).  l e s s than  reduced p a l a t a b i l i t y  the  proportion  = 80,  season  dry  reflected  types.  grasses accounted  was s i g n i f i c a n t l y  = 2.52,  its  (Figure  to  sampled  dry  which  than  a general  pattern  the  during  browsing  forage to  in  conditions  g r a s s was  similar  season,  grass eaten  Table  these  years  seasons (Table  use merely  season led  spent  use  conditions  herb  between  = 3.44,  (t  perhaps  forage  season  grass  followed  season  local  and reduced g r a z i n g  The  the  to  1 9 8 3 - 8 4 wet  browsing  1982 o r  the  rainfall  The o b s e r v e d d e c r e a s e i n  in  Males  1984 d r y  rainy  increase  of  the  to  in  s e e d l i n g and t r e e greater  led  conditions.  between  been due t o  rainfall  similar  each season  in that  diets, to  400  111  Diet  diversity  Among diet  bulls,  there  d i v e r s i t y between  Figure plant  4.3,  t  = 0.76,  individual  Bulls diversity (Table spent  did or  4.4  the  wet  d.f.  species during  between  was no s i g n i f i c a n t  the  not  year,  any s e a s o n a l  higher 4.7,  and F i g u r e  4.6,  t  = 0.21,  =  i n the  2.44,  wet  d.f.  season  = 10,  on browse  higher  suggests  browse  species  season,  when  but  d.f.  dry  50  switched  in r e l a t i v e  the  = 10,  n.s.).  Males  and  less  had  indicates that  t o a few p r e f e r r e d  bulls  time.  amounts. of  (Figure  that  b u l l s used a g r e a t e r  75%  of  significantly  d r y season  species during  over  time  browse c a t e g o r i e s  This  individual  diet types  season  for  P<.05).  comprised  a p p a r e n t l y d e v o t e d more t i m e females  and  forage  than i n the  that  in smaller  browse  4.3  all  combined, b u l l s  f o r a g e d more e v e n l y "evenness"  the  However,  and s e e d l i n g s  J-values t  change  when a n a l y z e d a c r o s s  trees,  (Table  seasonally.  show  4.8).  absolute  Males u t i l i z e d over  "evenness"  (Figure  shrubs,  n.s.).  in  though these p r e f e r e n c e s  more t i m e b r o w s i n g d u r i n g  grazing  number o f  In the  This  the  diet,  species,  dry males  just  as  d i d when b r o w s i n g .  L e n g t h s of  feeding  The a v e r a g e types  and d r y s e a s o n s  = 10,  species  change  were  difference foraging females,  bouts  lengths  higher  on f o r a g e of  i n the  was e x p l a i n e d by bouts  males  on  trees  types  foraging dry season the and  bouts  across  (Table 7b).  significantly shrubs i n the  showed d i s t i n c t l y d i f f e r e n t  all  forage  This  seasonal  higher  average  dry season.  feeding  patterns  Like on  1 12  t h e g r a s s component o f t h e d i e t , when b r o k e n dry  season  and  wet s e a s o n .  longer  ( t = 3.00, d . f .  (mean  =  i n t o e a r l y and l a t e  F o r a g i n g b o u t s were  = 7, P<.02)  in  the  significantly  early  dry  7.87 m i n u t e s / b o u t ) t h a n i n t h e wet s e a s o n  minutes/bout). significantly  The l e n g t h o f between  the  foraging  bouts  season  (mean = 3.43  d i d not  wet a n d l a t e d r y s e a s o n s  differ  ( t = 0.95,  d.f.=8,n.s.).  C h a r a c t e r i s t i c s of p l a n t s Like leaves d.f.  i n the diet  females, males used  in  the  wet  = 1, P < . 0 0 1 ) .  significantly  more  green  season than i n the d r y season  (x  2  = 256.6,  I n t h e wet s e a s o n , 7 0 % o f t h e g r a s s  leaves  e a t e n by b u l l s were g r e e n , w h e r e a s i n t h e d r y s e a s o n t h i s dropped  to  6%.  Green  d r y when a v a i l a b l e . were  over  I n a l l other forage types o n l y green l e a v e s  utilized.  forage  types.  branches  were  significant  seen  shrubs,  that  seen  and  in  the  i n small  dry  season  amounts. diets  Bark  but  at different heights  these  preferences  in  of  and  not t o a  different  predominantly  seasons.  c h a n g e d i n a manner s i m i l a r t o  f o r females ( T a b l e 8 b ) . Whereas, i n t h e  wet  f e d on t h e 0 - 1m h e i g h t c l a s s ,  season, foraging a t other h e i g h t s increased. distribution  shoots  degree.  Males browsed For  on t h e l e a v e s  R o o t s o f some s h r u b s , s e e d l i n g s , h e r b s , a n d  grasses a l l occurred i n the diets  males  figure  g r a s s l e a v e s were a l w a y s p r e f e r r e d  Male e l e p h a n t s f e d p r i m a r i l y all  grass  season,  i n thedry  T h i s change i n t h e  of feeding h e i g h t s r e f l e c t e d a p r e f e r e n c e f o r . t h e 0  - 1m c l a s s d u r i n g p e r i o d s when  new  growth  was  stimulated  by  1 13  rainfall.  Many  the  of  onset  shrubs sent the  rains.  following  hot  wildebeest  and o t h e r s .  no  longer  include enough t o the  fires  or  general  was  same.  the  forage  when at  expanded the  forests  i n the  forage  i n each season,  however,  found f o r  of  at  feeding  at  in  this  tree the  crowns.  shrub  greater  still  greater  heights  open,  Acac i a  use.  heights  heights,  2m, y e t  Fallen  to  extensive  that  c l a s s e s above  were  foraging  in the  T h i s was p r o b a b l y due t o a s c a r c i t y of  i n the and  elephants,  when s p r o u t s  d a t a were  f e e d more e a s i l y their  apparent  by  their  at  not  was a v a i l a b l e and o c c a s i o n a l l y  leaves  extent  particularly  dry season,  s i m i l a r to  Males d i d u t i l i z e it  c o p p i c i n g shoots  mainstems  preferences  d i s t r i b u t i o n of  species  reach.  the  For t r e e s ,  height  A l t h o u g h males can the  was to  elephants  p a t t e r n was  females,  damage  levels.  establish  extensive,  This  During  available,  higher  out  Mara,  preferred  within than  pushed t r e e s  their females  over  t r e e s w i t h i n the  than  to  get  riverine  woodlands d e m o n s t r a t e d  the  behaviour.  Pi scussion Mara e l e p h a n t s observed  in  many A f r i c a n  Buss and Wing Parker  and  Barnes  1982,  Ross  Laws e t 1976,  al.  1982,  and  and  1977a  & b,  Hansen e t diet  1963,  Sale  Thomson  Laws  1973,  1975,  T i e s z e n et al.  patterns  (Nicholson  Sheldrick  1975,  seasonal  foraging  populations  Leuthold  Leuthold  similar  same g e n e r a l  elephant  1971,  Weyerhaeuser  showed  the  Napier-Bax  Field  1974,  Field  females  1970,  1968,  Eltringham  followed  1985).  changes.  1954, and  Wyatt and Guy  1976,  al.  1979,  M a l e s and Both  sexes  1 14  concentrated  on g r a s s e s and h e r b s  species  the  while  in  female  throughout a r e most  diets the  of  of  a  season. that  a  branches)  than  However,  woody  In  of  manner  as  Barnes may  proportion  of  diets  ones  forage i n any  the  be  season  levels  in  structural problems  for  the  males diets  relative  of  elephants  both  sexes  but  i n the  wet  some  net  debilitating (Rhoades that  available  defenses,  females  Barnes'  such  grazers  Mara.  (1982)  to  loads  1983).  (Olivier  a  theory,  forage  of  in  Olivier  certain  (1978)  in elephant  grasses  into and diets  compound  may  McNaughton  a  1986).  be worked  which c r e a t e  1978,  been  of  forage.  by s e c o n d a r y  While  and  season.  (Belovsky  must  changes  silica,  dry  foraging  intake  and c o n s t r a i n e d  as  (twigs  component  and a s s i m i l a t i o n  seasonal  finding  T h i s may have  should select  toxin  forage.  a large  optimal  energy  1979,  in the  p r e f e r r e d woody  recent  acquisition  controlled  the  of  herbivores,  speculated  of  comprise  i n the  their  of  support  by  more woody components  scarcity  generalist  both  to  not  with  equation  (1982)  both  ( l e a v e s and s h o o t s ) did  local  containing  overall  than  in elephant  ability  year  significantly  The a d d i t i o n a l c o s t s  this  and t h e  the  which maximizes  forage  of  throughout  accordance  elephants,  reflection  no e v i d e n c e  leafy  diets  changes  females,  levels.  higher  females ate  reflection  seasonal  matter  browse  quality  T h e r e was  elephant  These  direct  S h r u b s were u s e d comprised  M a l e s browsed more t h a n  p r e f e r r e d forage  food  r a i n y s e a s o n and  c o n t a i n e d more h e r b a c e o u s  crude p r o t e i n  Effects  season.  year.  likely  quantities track  dry  i n the  have  digestion et  al.  1 15  1985),  they  toxin  present  levels  considered Barnes  in  a  no  problem  grasses  constraint  are  for  elephants.  negligible  on e l e p h a n t  diet  and  not  choice  Mbi  (1978),  Milton  (1979),  al.  (1980),  Macauley and Fox  (1980)  found  that  vulnerable  parts  shrubs  the  species,  seasonally higher  comprise a large  than  contradict  of  older  those  Rhoades and Cates foliage  plant  such as young  amounts  toxins),  of  (Olivier  portion  (1980),  of  of  of  Feeny  1978,  (1976)  who trees loads  investigators  did  contain  concentrations  than  speculate,  phenols  McKey  new  qualitative  elephant  elephants  ate  forage  less  species,  browse  browse  species in  lends  support  to  manner  aimed at  minimizing  assume  the  latter  look  for  Coley  (1980) plant  other  the  wet  the  their  explanations  defense  out  that  the  that  toxic  be t r u e for the  strategies,  the  were  was  well-  more largely  that  bouts  on  abundant, forage  However, case,  as  findings  feeding  observed diet  latter,  be  former  elephants  this  such  observations  intake. in  same  l e a v e s may  still  when new g r o w t h  hypothesis  to  the  and had s h o r t e n e d  season,  findings  pointed  then  greater  toxins,  from  and  mature  These  young  defended in  more  growth.  we a s s u m e t h a t  1979),  contained  and c y a n o g e n i c g l y c o s i d e s and t h e r e f o r e If  findings  (1974,  that  (which  (quantitative  older,  shrubs  and  contain  These  alkaloids  herbivory.  trees  diets)  however, of  of  et  (1985)  elephant  that  and  al.  and l e a v e s  1976),  found  Oates  stems,  and t o t a l  (1970,  et  a variety  l e a v e s a n d woody m a t e r i a l s .  chemical  higher  Coley  and B r y a n t  shoots,  terpenoids  perennial  quantitative  of  generally  1982).  However,  apply  Furthermore,  in  a  if  we  we  must  choices.  But  then  classical developed  views for  116  temperate  s y s t e m s and needed  tropical  systems,  case.  is  in  of  the  on  patterns  of  and t h e  particularly  have  tended to  coevolution  temperate  plant defense  e v o l v e d w i t h a b r o a d r a n g e of  in  be  has  the been  zones,  more work  among A f r i c a n  species,  herbivores  for  millions  years. Olivier  elephants  While  it  toxins, daily  (1978)  may be t r u e  that  it  the  to  elephant  more  is  also  requirements  meet diets  these  that  this  case  that  (Sikes  needs  may be c h o s e n  toxins  needs.  may  diversify  Although  they  1971,  primarily  while  Laws e_t a_l. of  s e c o n d a r i l y to  are  seriously  affected  by t o x i c i t y  Elephants  are,  however,  affected  by d e c l i n i n g f o o d  overall  body  levels,  elephants  condition  (Albl  acute  cases,  while  in chronic situations  rates,  t h e y may even  deferred  sexual  initially 1971,  starve they  to  shortage,  their  death  energy  unlikely  that  i n the  diet.  quality  forage  Guy 1976,  this,  with  is  show  intake  interfere  levels  suitable  In  avoid a buildup  elephants  subsistence  absolute  1 975).  meeting  it  plant  achieve  and, thus,  assimilation,  than  to  food  To  of  choice.  to maximize energy  still  a v a i l a b i l i t y of  diet  have v e r y h i g h  processes  When t h e  toxins  their  digestive  quantity.  system  sensitive  ingestion.  may a c t  and e n e r g y  plant  times  their diets  toxins  to  are  during  from d e f e n d e d p l a n t s  digestive  influences  elephants  and s e c o n d a r i l y t o m i n i m i z e t o x i n elephants  the  sensitive  r u m i n a n t s and t h a t  energy  order  argued  r e n d e r s them  sympatric  of  re-examined,  plant-herbivore  neotropics  r e q u i r e d on p a t t e r n s  w h i c h have  be  where o p p o s i t e  As most work  conducted  to  a  falls  and below  decline  Barnes  1982),  (Corfield  in in  1973),  may show d e c l i n i n g r e p r o d u c t i v e  maturity,  lengthened  interbirth  11 7  intervals,  and i n c r e a s e d  1981b, W e y e r h a e u s e r  the  first  dry  the  (grasses  season),  f o r m e r became  patterns  the  to  the  remove  over  obvious  90% of  of  the  r a i n s and browse  In  the  arrival  of  occurred.  the  of  new g r a s s  for  the  grasses  this  height  both  northernmost every  of  as  foraging intake  (Figure a  feeding  wildebeest  distinct  time,  occupancy,  down t o  g r a s s e s may have  local  (Figure  change  despite storms.  the  choice  feeding in the  on  latter  p r o d u c t i o n of  Wildebeest  kept  (McNaughton 1 9 8 4 ) .  difficult  for  an  their  upon  in diet  to  newly  peak of  their  the  showed the  However,  altogether  lawn h e i g h t become  the  a  migrants  stay  elephants  decreased  on g r a s s  intermittent  These  over  r a i n s and p r i o r  this  4.10).  extension  dry season,  the  s e x e s of At  immediately  grazed  in  only  maximizing energy  the  cessation  wildebeest,  following  species  forage  crop during their  grasses.  levels  stopped  the  nutritious  elephant  were v e r y a b u n d a n t and a t  Elephants  and t h e y  most  up r e s i d e n c e .  standing  wildebeest,  protein  the  general,  is  take  following  grasses  digestible  stages  the  preference  germinated  grass  pick  S e r e n g e t i - M a r a ecosystem where,  arrival  Laws  migration  migratory wildebeest  the  1975,  to  to poorer q u a l i t y  Mara Game R e s e r v e  Immediately  al.  loads.  million  4.9).  (Laws et  be aimed more a t  wildebeest  The M a s a i of  following  unavailable.  appeared  of  tended  and r e s o r t e d  than m i n i m i z i n g t o x i n  Effects  mortality  1982).  M a s a i Mara e l e p h a n t s species  calf  elephants  At to  eat. This  change  in  feeding  patterns  probably  reflected  a  118  350 00  s  co  < o I — I  300 250200-  m  o DS  150  o  100  I—I  Q  50  CO  M  J  J  A  S  0  MONTH  Figure  4.9.  Standing c r o p biomass through the dry season (during the time the migratory w i l d e b e e s t are present i n the study a r e a ) .  119  12-i  11^ »—i W  10-  EH  9H  O  8-  DH DH W Q P DH O  w U  DH  w  DU  76543 2 H n  M  Figure  4.10.  1  1  1  J J A MONTH  i  i  S  0  Percent crude p r o t e i n of long g r a s s e s over the course of the dry season (redrawn from S i n c l a i r 1975).  1 20  decline  i n both  remain  relatively  s p e c i e s as that A.  the  some  grass  quantity  higher  browse  and Commiphora  levels  above  Dougall  et  leaves  of  13%  al.  even  the  (1964)  Acacia  triandra  dry  1964,  season,  on t h e  agents of Field  1971,  elephants  Throughout  the  Africa change  (Laws  suggested  that  habitats  significantly,  resources.  suggests into  the  annual  that  the  The sheer  Mara e a c h y e a r , stay  that  elephants  late  activity  increases  rainfall exacerbate  i n the  (Figure this  by  they  by Mara  giraffe  no o t h e r to  numbers o f so  pattern  as  5% and 8%  Dougall height  their  feeding  considered  and  of  the  1975).  may compete  the time  primary Although  with  elephant  species  is  compete  with elephants  the  known t o  above,  habitat  direct  Although  alter  however, coming  during  their  competition elephant  for  with  browsing  during  any  period  of  low  with  the  wildebeest  may  competition and  elephants  migratory wildebeest,  into  significantly  (33%),  By t h e  provided  alter  come  mature  such  low as  1963,  Thomson  evidence  the  and  4.11).  are  dry season.  4.8),  1975).  1970,  or  (1963)  elephants,  (Dougall  (Figure  elephants  circumstances,  grazing  used  m a j o r i t y of  species  under c e r t a i n as  for  protein  Boscia a n g u s t i f o l i a  eaten  Sinclair  spent  browse  habitat  (1971)  values  w h i c h were a l s o  dry season  reported  crude  Dougall  and Cynodon d a c t y l o n d r o p as  Field  available  (17%),  herbaceous  Acac i a g e r r a r d i i ,  maintained leaves.  species  (1981)  such as  f o u n d even h i g h e r  crude p r o t e i n d u r i n g the Glover  than  Pellew  older  The g r a s s e s  Woody  protein  trothae,  brevispica  dry season.  Themeda  crude  species,  in  and Solanum incanum (30%) in  quality.  dry season p r o g r e s s e s .  preferred  Senegal,  in  and  result  in elephants  placing  even  121  Figure 4 . 1 1 .  Grams crude p r o t e i n per m of l o n g g r a s s e s d u r i n g the dry season ( l e f t a x i s , l i n e graph) and a percentage breakdown of elephant d i e t s between g r a z i n g (G) and browsing (B) i n the wet and dry seasons ( r i g h t a x i s , bar graphs). 2  1 22  greater Croze an  p r e s s u r e on a v a i l a b l e  (1974a)  area  studied  of  present  the  in  the  dry  70% o f  of  in  Between  time the  feeding spent  two  their  their  is  behaviour  He  time  vegetation  This  species in  Serengeti  season.  habitats,  time.  70% o f  feeding  central  approximately their  the  browse  in  the  along  time  to on  dry  bull  season.  elephants  wildebeest that  Acac i a  in  were  bulls  not  spent  w o o d l a n d s a n d 30%  permanent  took  contrast  foraging  where found  in  browsing  of  the  water  u p 34% o f elephants browse  the in  sources. elephants'  the  during  M a r a who the  dry  season.  Effects  of  s e a s o n a l change  Female  elephants  environmental elephants their  relative in  changes  are  adult  ways  items 1971, diet  year,  therefore,  with  This the  should  prediction  when more d e s i r a b l e  diversity.  intake  rate  items  Males  With  on a  few  female  to  meet  their  most  from  males  differences.  diet  foraging  diet  choice  was  theory  that  (Schoener  s e a s o n a l change  n e e d s by  species during  in  less-preferred  available  energetic  of  high  absolute  include  become l e s s  relative  Cow  became e v e n g r e a t e r  showed no p a r a l l e l  lower to  differed  a higher  optimal  diets  to  demands a r e  energetic  difference  their  throughout  energy  patterns  in  sensitive  counterparts.  lactating  these  from  more  male  their  expansion  expand  1984).  be  females maintained  m a l e s may h a v e b e e n a b l e bulk  reflect  males and t h i s  season.  Pyke  or  may  than  foragers  pregnant  which  diversity  consistent  their  Female f o r a g i n g  the  dry  and,  to  males.  Throughout  the  than  either  lives  to  seemed  in  requirements  increasing  energy-limited  their times,  1 23  rather  t h a n by d i v e r s i f y i n g  Barnes  1982)  f e m a l e s must This  speculated  and  that  diets. due  be even more c a u t i o u s  proposed  partially  their  explain  reduced  to  of  in  why f e m a l e s  elephants  risk  of  their  toxin  difference  the  Olivier  content to  levels  body  than  by size  bulls.  toxins  could  their  diets  i n any  single  diversified  toxin  cited  smaller  sensitivity  dangerous  (as  species. Relative change  diet  diversity  seasonally  compared. among  Elephant  all  seedlings species  in  bulls  of  throughout  the  showed  dietary  time  grasses, year.  distributed  herbs,  shrubs,  However,  significant  season  whereas  females  findings  suggest  among  more  "evenness"  browse i n the  simultaneous of  onset  to  distribute  seasonal  the  patterns  was  positively though the currently  wet  their  comparable data It  species during  This  for  diet  of  time  Barnes evenness  were  trees, of  females  their  and  browse  sexes and  i n the  change.  time  more  evenly  season.  Greater  been  largely  due  of  seedlings  a  broad  reported  bull  diet  to  the  following  new g r o w t h a l l o w e d among  wet These  wet  (1982) for  types  bulls  range  these  elephants  of  same  but had no  cows.  assumed  correlated  that to  quantitative being  flush  not  between  seasonal  a variety  foraging  items. of  the  s e a s o n may have  rains.  browse  no  distributed  s p r o u t i n g peak of  the  nutritious  showed  than  did  similarly  an a n a l y s i s  differences  M a l e s f e d more e v e n l y  males  forage  was  between s e a s o n s .  that  "evenness"  o r cows when a l l  foraging  species  only  or  the  the  length  absolute  expression  investigated  of  amount of  by L i n d s a y  foraging of  this  forage  bouts  consumed,  relationship  (in p r e p . ) .  was  is  B u l l s and  124  cows d i d n o t of  their  differ  significantly  f o r a g i n g bouts  throughout the  b o t h s e x e s had s i g n i f i c a n t l y all  forage  types.  found,  elephants  type.  These  accounted types bouts  on  length  of  to  foraging  trees  in  longer  season  in  than  seasonal  feeding  the  difference  interpreted  as  a  late in  of  t o a new were  individual  forage longer  increased  This  was  is  the  similar  browsing p a t t e r n s  of  bulls  Park,  Tanzania.  Both  sexes  had  bouts  on g r a s s e s  i n the  early  dry  dry  season  or  wet  bouts  on  grasses  foraging of  the  immediately  just  to  the  of  seasons.  elephants'  q u a l i t y and q u a n t i t y arrival  type  season  w h i l e males  on t h e  reflection  the  across  significantly  high grass prior  analyzed  dry  on s h r u b s and t r e e s .  National  significantly  use  spent  and s e e d l i n g s ,  season,  switching  the  length  dry  good f o r a g e  before in  i n the  In t h e  when a  i n the  observations  Ruaha  time  Females  f o r a g i n g bouts (1982)  bouts  bouts  differences  year.  once  for a longer  and c o w s .  herbs,  Barnes'  season  fed  by  by b u l l s  longer  Presumably,  longer  for  from one a n o t h e r  has  been  preference  following  wildebeest  This  the  for  rains,  migration  (see  above). The  characteristics  remained c o n s i s t e n t forage  types  as  in their  occurred  twigs, i n the  (1971),  Field  Barnes  (1982),  localities.  bark,  Ross  year.  used  by  condition,  always  eaten  more t h a n o t h e r  or b r a n c h e s . This  (1976),  (1971)  Some use  when  of  (1983) explains  (1976),  this  as  plant  from the  all  parts,  and b a r k  s e e n by  Olivier  in elephants  ate  available.  roots  same p a t t e r n was Guy  elephants  Cows and b u l l s  leaf  and I s h w a r a n  Field  plants  green  dry season.  and  food  t h r o u g h o u t the  L e a v e s and s h o o t s were such  of  Field  (1978), different  result  of  1 25  elephants levels  constantly  in their  Effects  of  fed  during  Because hour,  primarily  the  dry  annual  seedlings  were  effects  "weeding" o u t  took  range.  These  findings  Buss  (1961),  Guy ( 1 9 7 6 ) , of  Croze  woodlands  i n the  largely except  in  or  result  of  other, this  height  diets  to  are  the  early  possibly  proportion populations  of  to  Croze  a rate  in  relative  lost  seedlings  the  recruitment  potential  of  Brooks and Ross  Either  which are is  for a in  fed  classes  eaten.  of  to  Serengeti  size  classes.  may a c c o u n t  1m h e i g h t  elephants  reasons  per  forage  contrary  under one meter  height  400  other  Field  all  rarely  diets.  began  central  for  meter  herds  0 -  but  availability  preferred, of  of  one  over  those  found t h a t  w h i c h were  forage  the  (1970),  availability  of  of  In a l l  (1985)  height  browse  entire  in  have c h a n g e d f o r  use  below  the  with  for elephants  1970's.  more  the  of  Buss  range.  not  a direct  forage Either  in way,  significant local  tree  today.  know  taken place  and  1m  include other  when  Weyerhaeuser  (1974b)  change  80%  consistent  Wing  high  of  -  feeding  exclusively  under one m e t e r ,  this  their  consumed a t  place  and  d i f f e r e n t i a l use  We  0  i n open g r a s s l a n d s .  food p r e f e r e n c e s  clear  the  noticeable  proportion  those  elephant  mean often  seedlings  feeding  those  crude p r o t e i n  season,  were  types,  (1976),  available  in  expanded t h e i r  f o r an  the  (1957),  highest  woodland d i s a p p e a r a n c e  Although elephants  accounted  the  forage.  Elephants  classes  seeking  that  over  the  major past  changes 25 y e a r s .  i n woodland communities These  include  a  loss  have of  126  tree  cover  i n the  larger  Reserve.  Some a r e a s  loss  1950  since  result  of  and t h e i r today  fire  (Belsky  factors  2).  negative These  as  in  supports  are  0.75  -  Amboseli  much a s  the  Mara Game  a 95%  was  on t h e  trees  i n the  the past,  regeneration,  exacerbated  small,  relative  largely  on s e e d l i n g  were  and  certainly  1.00/km )  activity  by  mammalian  or  to  influence.  and  woodland  dynamics  is  inhibitor  (Chapter  5).  The r e l a t i v e t r e e s or  the  dependent  other  browsers  increased  elephants' of  been  relative  local  1985).  e y e s of  casual  elephants  are  day r o l e not  of  so  elephants virtually  to  Park  not  eliminated,  to  while  signs  subtle in  one  of  be  Dry y e a r s , species  as  loss  may  of  bark-  elephants  the  to  today's  more  destroyer  to  up  in  a  of  adult highly  like  1984,  b e c a u s e of  during dry p e r i o d s .  switch  1982),  uniformly  Mara  of  elephants  conditions.  on browse  National  observers  Masai  b r o w s i n g on woody p l a n t  dependence  parks  Few o b v i o u s  woodland r e g e n e r a t i o n  weather  other  densities  exerting  much one  biological densities  to  (Weyerhaeuser  2  has  c o n t r i b u t i o n of  local  low  primary  elephant  Standing t r e e s are  i n h i b i t i o n of  on  Park  Lindsay  meet t h e  The p r e s e n t  the  Mara  6.00/km  National  toppled,  of  T a n z a n i a ' s L a k e Manyara  M a s a i Mara Game R e s e r v e . stripped  one  are  2  up  (Western  2  elephant  have  influence  changes,  in A f r i c a .  densities  Kenya's  ability  as  decline  impacts  impacts  woodland  and r e s e r v e s  to  This  wildebeest  elephants  (approximately  lead  experienced  throughout  1984).  While  10.00/km  classes  and e l e p h a n t  (Chapter 5). such  have  (Chapter  continued  factors,  height  the  The  innate  g r a s s e s when browse  species  makes  them  a  unique,  and  127  potentially browsers, closely  like  1979,  to  browse  it  may be t o o  if  the  late  woodlands  numbers  1983).  habitat  tree  until  browsers,  woodland  Norton-  loss  and  more  populations  decline  fires,  inhibit  are  through  al.  populations. a  strict  1971,  s u g g e s t e d by Laws e t  survive  effectively  dynamics  elephant  woodland  save a f f e c t e d  other  might  to  those  (Field  Although  r a t e s as  could  occurs,  wildebeest,  to  than  population  availability  respond  reproductive  herbivore  whose  Pellew  eventually  decreased  detrimental  giraffes,  linked  Griffiths might  more  (1975)  A n d , even  in the  elephant migratory  regeneration.  Summary The f e e d i n g Mara  Game R e s e r v e  by d i r e c t of  grasses,  impacts  of  shrubs,  male  i n the  wet  consistently herbaceous  the  most  the  migrants.  The s t u d y herbs,  an  wildebeest standing Females  in  species  females,  t i m e more e v e n l y  grass  was  among  the  a  year.  season. the  the  and  period use  by  both  overall  thickets.  on g r a s s e s  the  dry.  females  upon t h e  and  Males  used  more  arrival  of  from  grazing  to  rapidly  reduced  by  switching  maintained  dry  while  changed  herds,  c r o p of  in  year  trees  on  concentrated  Masai  seasonal  and a d u l t  emphasis  elephants  throughout the  the  r e m a i n i n g woodlands  Elephant d i e t s  i n the  a two and a h a l f  seedlings,  with  on t h e  elephants  investigated  s e a s o n and woody  than males  pronounced  • foraging  over  browsed more t h a n  migratory  diversity  free-ranging  studied  and f e m a l e  forage.  b r o w s i n g as  of  elephants,  elephants  Both herbs  was  observation.  male and f e m a l e  the  ecology  higher  absolute  diet  This  difference  Males d i s t r i b u t e d  available  browse  was their  species  128  during  the  wet  season  showed no s e a s o n a l season,  a  change.  few  forage  bouts  type  new t y p e . 0  -  and s h o o t s of  in  the  the  arrival  former  dry  class  the dry  and a l m o s t  forage  height  of  before  feeding  exclusively  ate  dry  fed  had  i n d i c a t i n g that longer  the  were  sexes  females  upon longer  once  a good  switching  activity the  to a  in  green  the  leaves  types. i n t h e Mara  season.  further exacerbated  of  the  These  leaving  elephants season  no a l t e r n a t i v e c u r r e n t l y have  dispersal  development,  factor  i m p a c t s were  migratory wildebeest  poaching a c t i v i t y .  a primary  the  Both  dry season fed  d r y , whereas  species  sexes.  elephant  the  i m p a c t s on woody v e g e t a t i o n  grass  agricultural severe  browse  both  the  found,  all  Elephant  As w e l l ,  However, at  Elephants concentrated t h e i r  1m h e i g h t  available  by  in  was  during  favoured  disproportionately foraging  than  i n the  t h e Mara Game R e s e r v e .  areas  intensive  who removed forage  almost  the  by all  elephants.  r e s t r i c t e d access  to  their  w h i c h a r e now e x p e r i e n c i n g livestock  As a r e s u l t ,  i n h i b i t i o n of  for  increased  grazing,  elephants  woodland  and  are c u r r e n t l y  regeneration  in  129  CHAPTER 5.  THE EFFECTS OF ELEPHANTS, BROWSERS ON THE ACACIA  FIRE,  WILDEBEEST AND  WOODLANDS  Introduct ion  The  loss  Kenya o v e r  of  the  grassland.  woodlands  past  These  protection  30  f o r many s p e c i e s  afforded other  1970,  Dublin  regeneration ecological be  maturity.  has  the  wildlife.  Today,  1985). is  of  t h e s e woodlands  is  that  i n h i b i t e d by a c o m b i n a t i o n of  factors;  each  one a f f e c t i n g  different  times.  woodlands  work,  t h e s e a g e n t s of  m o r t a l i t y and i n h i b i t i o n as  giraffe.  She  added  wildebeest  to  impacts effects  on of  woodland (Glover  this  list  woodlands. these  factors  dynamics 1965,  Griffiths  the but  Pellew  did  of not  this  al.  of  1967,  1983),  on  the to  process  is  biological  (1984)  ways  at  summarized  elephants,  and  s m a l l b r o w s i n g mammals and quantify  their been  on mature t r e e s and t h e i r areas  to  grow  several  A l t h o u g h r e s e a r c h has  in l o c a l  Lamprey e t  1979,  effects  seems  Reserve.  and  fire,  for an  in different  Belsky  the  both  dependent  being  B a s e d on p a s t  from  i n the  currently  the  of  potential  themselves  hypothesized  little  ( G l o v e r and Trump  Regeneration  now  open  is  much  interest  establish  been  there  The  perspective.  an  of  b o t h f o o d and  and  biological conditions  to  area  provided  Lamprey  seedlings It  left  browsing s p e c i e s  and a management  of  has  formerly  these woodlands  r e c o v e r y of  ability  to  1984,  of  of  M a s a i Mara Game R e s e r v e  s p e c i e s has d w i n d l e d away  l i m i t e d by e x i s t i n g The  years  woodlands  woody m a t e r i a l a v a i l a b l e cover  i n the  the  Serengeti  Croze  1974a  no q u a n t i t a t i v e  relative  done  on  the  influence  on  National &  b,  Park  Norton-  i n f o r m a t i o n was  130  available in  on t h e  process  woodland r e g e n e r a t i o n The e x p e r i m e n t s  analyze  the  chapter or  the  establishment  of  hypotheses  and t h e  fire,  Using  are  to  elephants,  experimental  that  these  role  separate  and  and  wildebeest  evidence,  factors,  effectively  r e c o v e r y of  and i t s  today.  s t u d y were d e s i g n e d  effects  combination,  establishment  i n t h e Mara  populations.  examines  in  seedling  in this  relative  on s e e d l i n g  of  acting  inhibiting  Acac i a woodlands  this alone  seedling  i n the  Masai  Mara Game R e s e r v e .  Background I n f o r m a t i o n A  s t r o n g c o r r e l a t i o n has been e s t a b l i s h e d  woodland l o s s 1979, has  Lamprey  l o n g been  and  the  i n the 1985,  S e r e n g e t i - M a r a ecosystem Dublin,  of  seedling  B u e c h n e r and Dawkins 1961, Obeid  1971,  Chapter  2).  r e c o g n i z e d as a p r i m a r y  inhibition  Spinage and G u i n n e s s  factor  1965, 1972,  1968,  Harrington  a  t o combat b u s h e n c r o a c h m e n t and p r o t e c t  (West  Strang  1973,  ecosystems, have  1958,  Thomas and P r a t t  1974,  like  the  Trollope  al.  fire  l o o k e d f o r ways t o m i n i m i z e t h e  Pratt  1982b).  Serengeti-Mara,  mortality  El  1960,  Din  and Ross Trollope  and 1974,  1982a,  has been w i d e l y u s e d  1967,  1972,  1980,  fire  and Keay  Seif  1986).  reasons,  tree  (Charter  Sabiiti  For these  et  in  1975,  lands  Brookman-Amissah  and  Throughout A f r i c a  growth  Glover  fire  (Norton-Griffiths  Thomson  tool  between  domestic  grazing  and K n i g h t  However,  ecologists  in  and  as  1971, natural  managers  detrimental effects  of  fire  on w o o d l a n d s and g r a s s l a n d s . Norton-Griffiths  (1979)  estimated  that  over  a  ten-year  131  period  (1962-1972)  Serengeti  the  had d e c l i n e d  northernmost  areas.  effects  of  increases  However,  he d i d not  the  of  loss  reported period.  activities Croze elephants  on  the the  loss  that  of  related  primarily  on  the  restricted  area  wildebeest  the  impact  of  the  to  of  trees  of  20  population  has  in his  and  years,  speculated  that  the  area,  d i d not  pose  disagreed,  arguing  and w o o d l a n d  on  re-  concentrated  mature the  trees  in a  effect  reduce  standing  Serengeti's by  a  factor  of  thereby  migratory of  five  N o r t o n - G r i f f i t h s (1979)  these g e n e r a l i z e d  grass  study  per  (1965),  on  the  N o r t o n - G r i f f i t h s 1979).  c r o p of  6.0%  of  general.  increased  of  a rate  Glover  studies  less  of  woodlands.  at  elephants  elephants  and  increase  local  role  h i g h as  woodlands.  (Sinclair  the  Serengeti  l o s s e s as  survivorship in  past  time  browsing  the  elephants  These  system  in  for he  same  their  the  fires.  system,  this  detail  was more e x t e n s i v e  of  the  i m p a c t of  and H e r l o c k e r (1976a)  limited.  to  of  increase  central  local  future  damage  on s e e d l i n g the  an  the  responsible  7% o v e r  in greater  structure  the  (1967),  more  Over  with  in  stands.  mature t r e e s  age  to  establishment  elephants  of to  was  within  under t h e s e c o n d i t i o n s ,  threat  elephant  of  50%  and s e v e r i t y  areas  subsequent  northern  losses primarily  alone  loss  described  on a v e r a g e ,  al.  fire  woodland  dynamics  B a s e d upon t h e  Lamprey e t  that  in the  and up t o  frequency  In o t h e r  and t h e  (1974a & b) in  woodlands  these  were a t t r i b u t e d  individual  per year  severe  the  woodland  losses  he c o n c l u d e d t h a t a  in  conclude  densities  He e s t i m a t e d  year.  He a t t r i b u t e d  overall  These  of  by 26% on a v e r a g e  canopy c o v e r .  an  elephant  2.5%  canopy cover  grazers  reducing f i r e  would severity  1 32  and  enhancing  other  directly  Wildebeest  rut.  These  northern  have  the  woodlands, woodland  the  extended  (1981,  the  be most  w i t h woody  have  vegetation.  numbers t h r o u g h  and Mara woodlands  contact  woody  and s p a r r i n g a c t i v i t i e s  are p r e d i c t e d to  1983),  investigated dynamics.  from one  extent  growth  of  He  trees  of  their during  pronounced  where  the  species  The  fire  existing due  prefer. (1979)  into  overall  and e l e p h a n t s the  to  the  the  loss  losses  in  wildebeest  during  northern many  there  a consequent  within  been  the  Reserve  their  jeopardy.  unprecedented  trees  numbers,  fires  added t r a m p l i n g , t h r a s h i n g ,  browsing  of  in  local  growth  of  dependent  on  retarded  period  merely  of  of  the  those  not  fire-  minimal.  favourable classes  (Dublin,  Mara  to  Chapter  to  in t h e i r  currently  continue  the  for they  found by N o r t o n - G r i f f i t h s  K.R.E.M.U.  the  the In  are  height  emigrated  decrease  the  compounded  giraffe  are  in  have  Elephants  rate  their  Serengeti  in  the  played  Serengeti  on w o o d l a n d r e g e n e r a t i o n .  (Lamprey 1985,  Woodland r e g e n e r a t i o n serious  This  effects  giraffes  central  n e x t was h e a v i l y  effect  exceed  Consequently, has  the  that  communities of  the  which g i r a f f e s  prolonged  woodland  These in  role  in  concluded  and  M a s a i Mara Game R e s e r v e ,  giraffes  working  b r o w s i n g by g i r a f f e .  of  impacts  the  size-class  susceptibility.  The  on  wildebeest  dry season m i g r a t i o n .  Pellew  trees  effects  However,  in s i g n i f i c a n t  browsing,  Serengeti  most  effects  seedlings  inadvertent  the  annual  survivorship.  detrimental  damage  trampling, the  seedling  other  areas  population unpubl.  is,  burn,  data). in  year-round in  and t h e r e  and b r o w s i n g e f f e c t s  and  numbers  nonetheless,  reside  2).  of  an  are  the  increased  133  wildebeest months  p o p u l a t i o n which v i s i t s  each  the  area  for  three  to  four  year.  Methods Five  study  gerrardi i  woodland  homogeneity seedlings  of  (see  difference  amongst  n.s.).  stems  Also,  frequency  show  a size  1m h e i g h t small  site  The l a t t e r  Treatments  treatment against  plot  to  (c)  invasion  all  by  sites n.s.;  of  conducted  on  Figure  in  height  no  the  the  classes  F  (b)  three  (a)  and  plots large  surrounded wildfires.  (b)  15 x  browsing (a) by  and a  under a  numbers. plots:  neither  burned  mammals.  10m  height  treatment  (c)  but  and  only  low  i n 25 x  15m f e n c e d  (b)  1.86,  the  comprising  were c o n d u c t e d  were  or  distributions  towards  separate  browsed o n l y and  =  different  t r e e s and o c c u r r i n g i n v e r y into  height  densities  These biased  the  significant  on e i t h e r  of  5.1).  heavily  e a c h of  stems:  i n terms  Acacias  treatments  was  five  was  for  and d e n s i t y  were  There  2.04,  of  was d i v i d e d  excluded  access  F =  with other  Treatment  the  which i s  b u r n e d and b r o w s e d ,  plots.  of  5.1,  all  ANOVAS  Acac i a  selected  distributions  began.  were d e s c r i b e d  (Table  p r o p o r t i o n of  nor browsed.  free  any  structure  class,  Each (a)  study  distributions  c l a s s e s measured  height  were  the  h e i g h t s and stem numbers of  (heights:  sites  within  Sites  One-way  below)  when t h e  established  communities.  sites.  sites  number o f  were  composition,  between  transformed five  sites  100m  exclosures. Grazers  not  firebreak  (c). to  had Each guard  134  Table 5.1.  S e e d l i n g d e n s i t i e s p e r h e c t a r e by h e i g h t c l a s s . Mean d e n s i t i e s (± one s.d.) a r e a l s o p r e s e n t e d .  HEIGHT CLASS (m) 1-2  2-3  3+  850  25  38  36  982  17  20  32  SITE NO.  0-1  1  <  5 Mean D e n s i t y :  1105  0  29  24  720  19  0  32  0  51  37  29 ± 17  32 ± 5  595 850 ± 202  12 ± 1 1  135  SITE 1  SITE 2  100-  2  CL AS  d  80 80-  80-  60H  2  60-  ER CE NT  B<  lOO-i CO  40-  £  20H  CL  4020-  a. 0-1 1-2 2-3 3+ HEIGHT (m)  SITE 4  lOO-i  lOO-i  80-  80-  60-  60-  40  u u  40-  e  20 H 0  -JZ3-  0-1 1-2 2-3 3+ HE1GHT (m)  SITE 3  c_>  1 1  w  200  0-1 1-2 2-3 3+ HEIGHT (m)  0-1 1-2 2-3 3+ HEIGHT (m)  SITE 5  lOO-i  8 o 2  80804020 0  I — i  -r  <=P-  -1 1-2 2-3 3+ HEIGHT (m)  F i g u r e 5.1. H e i g h t d i s t r i b u t i o n s o f A c a c i a i n f i v e sample s i t e s .  gerrardii  1 36  Measuring  the  The Acac i a s  term less  without versus  Effects  1m,  destructive  the  systems  Guinness  Sabiiti  Due t o  have  every  this  seedlings mapped  were  months and t r e e s  to  impossible  Leuthold  of  were as  May-June  follows:  May  them  both the  burning  divided in h a l f ,  The  t h r e e months on one clipping  effects  treatment  experienced  browsed o n l y realistic  (b)  treatment  control. the  growth  neighbouring  manipulation  only  side most  possible  1973,  of  than  Spinage  1984,  Lamprey  1982  and  1985,  the  measured  each p l o t  and  treatments.  browsing  treatments  and November-December  study, with  the  the  treatment grass  closely  simulated  competitive grasses. the  It  the  was  was  also  inhibition  for  (a)  the  and most  performed to of  s h o u l d be n o t e d  potential  other. grazing  b u r n e d and browsed  and, therefore,  (c)  clipped  u n d i p p e d on t h e  i n the  all  accurately  following  and  manipulation  addressed  rather  and l e f t  plots  This  investigate by  into  by p l a n t s  classes  tagged  (pre-wildebeest)  year  were  fire-derived  Within  again  exclosure every  many  of  individually  A  were  in  1985.  (post-wildebeest). plots  studies  coppicing,  of  1985).  until  for  seed  of  Belsky  i n June  Sampling p e r i o d s  distinguish  g r o w i n g from  Vesey-Fitzgerald 1977b,  all  most  trees  established  finding  to  for  revealed that  been b a s e d on h e i g h t  were  insure  areas,  African  text  However, e x c a v a t i o n s  process  Jachmann and B e l l  plots  six  was  ( G l o v e r 1965,  1972,  1986,  All  it  in adjacent  shoots.  distributions  and  used throughout the  s a m p l i n g t h o s e w h i c h were  p o p u l a t i o n d y n a m i c s of  savanna age  as  is  s p r o u t i n g from r o o t s t o c k s .  many s u c h s e e d l i n g s , coppicing  Browsing  "seedling"  than  those  of  seedling that  above  this  ground  137  competition. ground this  for  The  water  study  or o t h e r  competition  resources  was  between p l a n t s  not  within  the  below  scope  of  research. Initially,  tagged.  were  the  In e a c h  marked s t e m ,  total  plants. stem  In t h e  the  remaining  cause  of  of  if  each  periods,  their  original  were  followed  three  years.  October,  at  arrived  in  d i d not  to  a  minimum  Effects  first the the  of  year,  end o f study  every  plant  of  b r o w s i n g by  trampling  and  responsible  seedling  with missing  r e a p p e a r by t h e  longest original  and  were later  end of  the  determined  by  "regenerates",  of  still  altogether  which r e s p r o u t e d as  was  the  for  When a  r e c o r d e d and t h e  i n d i v i d u a l was  identification  over  the  of  sampling p e r i o d  in  subsequent  were  retagged  numbers.  All  six  and a maximum of  New s e e d l i n g s were t a g g e d  the  In t h e  plants  in that  Plants  referred  was  tagged  "missing"  when p o s s i b l e .  of  readily distinguished.  then  they  of  factor  individual  Those  height  any s i g n s  effects  the  death  "missing"  "dead"  The f a t e  sampling  was  number.  as  inspection  the  the  on l e a v e s and stems of marked  cases,  stem  stem  n o t e d as  classified  but  was m e a s u r e d and  and c o n d i t i o n  the  o r damage c o u l d be  identification  Measuring  period  stems,  species plus  m a j o r i t y of  live  initially  each p l a n t  C o n d i t i o n r e f e r r e d to  marked stem had d i e d  study.  sample  by m i g r a t o r y w i l d e b e e s t  removal  alive,  stem o f  number o f  and o t h e r  thrashing  longest  following  measured.  elephants  with  of  months as  they  individuals  were  found.  Burning  burning experiments the area  dry season. each y e a r  were  conducted  The m i g r a t o r y in late  in  wildebeest  June o r e a r l y  July  138  and  i n some a r e a s  grass  before  October all  removed up t o  leaving  proved d i f f i c u l t  subsequent  burns  ( F e b r u a r y and M a r c h ) . and t h e r e For  was  burn  air  temperatures  in  fuel  plots  fuel  treatment  short  the  wildebeest  (1977),  fuel  moisture,  As h e a v y ,  h e a t y i e l d of  approximately  estimated  using  oven-dried grass were c a l c u l a t e d  a  18,000  I = fire  from B y r a m ' s (1959)  consumed  intensity (kg/m ), 2  Rothermel  and  in experimental  applied  in  this  study.  s p r e a d and  to  actual  and  grass and  Sneeuwjagt  to  produce a  Fuel  loadings  calibrated  Fire  from  intensities  = Hwr  (1980)  burning  yield  (kJ/kg),  situations  in character  w = dry  (m/sec).  recommended t h e  F i r e s were s e t  h o u r s and were v e r y c o n s i s t e n t  relative  formula:  and r = r a t e - o f - s p r e a d  equation  4).  triandra  kg.  analysis  (kW/m), H = h e a t  Deeming  present,  were uncommon  were assumed  (McNaughton 1 9 7 9 ) .  I  of  amount  (1972)  per  point-frame  clippings  rate  were Themeda  kj  and  season  b u r n s were e x c l u s i v e l y  fuels  of  below-surface  wood f u e l s  insignificant  grass  Chapter  and  B a s e d on R o t h e r m e l these  dry  p r e v a i l i n g wind s p e e d ,  experimental  mezianum.  fuel  were not  b u r n i n g (see  surface  and c o n t r i b u t e d an  and F r a n d s e n  fuel  i n the  The d o m i n a n t g r a s s e s p r e s e n t  Pennisetum  where  conducted time  crop  Therefore burning in grass  for  burns,  the  of  standing  l a c k of  temperature,  loadings,  fires.  were  were  h e i g h t s were r e c o r d e d .  the  the  the  At t h i s  during  of  early October.  because  sufficient  each  humidity,  flame  in  95%  use  similar  between both  1200  of  to and  between  this those 1500 sites  139  and b u r n i n g  seasons  (Table 5.2).  2000 kW/m or k j / m / s e c  A l l fires  1000  -  hot"  f o l l o w i n g M c A r t h u r and Cheney In  addition  multiple-burn wild  to  seedlings.  effects  This  of  repeated  seedlings.  Three  recorded  under  experiments:  range and c l a s s i f i e d  these  experiments  on t h e  different  natural  conditions 300g/m ,  in  the  surface  s e e d s were  to  simulate  layer  collected  five  randomly  selected  loadings: This  sets  200  the  2  one set of  of  ten  g/m ,  of  fuel  the  2  entire  Seeds were  one of  s e e d s not  groups  were  Germination  groups  400  sown  square metre p l o t s . at  the  exposed then rates  (0  five to put were  to  600  Fuel  g/m , 2  recorded  of  200  wildfires  these were  on  seeds  Acac i a g e r r a r d i i  seeds  five  800 was  field  g/m , 2  From  fuel  and 1000  g/m .  soil  dishes  to  be  found  in  were  then  A c o n t r o l group  different and  2  on c l e a r e d ,  on t h e s e p l o t s  five  were  different  conditions  loadings.  to  these  i n each)  considered  t o p 2cm of  loadings  petri  in  These p l a n t s  2cm).  of  b u r n i n g and t h e in  been  i n t h e Mara and s o r t e d  one  experimental  had  different  (with  the  established  specimens.  range of  the  c o n d u c t e d under  -  loadings  into  simulate  heights.  effects  soil  and s u b j e c t e d  distribution  Mara.  the  100  original  insect-infested  of  g/m ,  representative  the  from mature t r e e s  remove any damaged o r seeds,  of  were  g r o u p s of  used  reached t h e i r  Seed g e r m i n a t i o n e x p e r i m e n t s conditions  were 2  times.  burned ten  (1982b).  which  and 6 0 0 g / m .  2  "moderately  of  loadings  the  experiments,  to  reserves  fuel  All  burning  was d e s i g n e d  root  they  were  burning  were c o n d u c t e d on t h r e e  fires  2  as  and T r o l l o p e  large-scale  w a t e r e d and r e - b u r n e d once plants  (1966)  experiment  150g/m ,  were m e a s u r e d i n  kept  on a d a i l y b a s i s .  treatment moist. A one-way  140  Table 5.2.  C h a r a c t e r i s t i c s o f l a r g e - s c a l e e x p e r i m e n t a l burns.  Fuel loads  =  200 - 400  Fuel moisture  <  35%  Air  =  27°C - 30°C  R e l a t i v e humidity  <  30%  Wind speed  <  10 km/hr  Fire  =  1000 - 2000  temperature  intensity  g/m  2  kW/m  141  ANOVA was p e r f o r m e d on t h e s e d a t a t o measure loadings  the  effects  of  fuel  on s e e d g e r m i n a t i o n .  Data A n a l y s i s  Square  roots  transformations of  all  allow  plant  heights,  logarithmic  stem n u m b e r s , and a r c s i n e  data the  problem  (Hurlbert marked  1984)  use  (i.  e.  percent  of  parametric  of  transformations  stems removed)  statistical  treatment  patterns animal  of  "pseudoreplication"  addressed.  plots because  elephant,  entered  seedling,  the  similarly  and  an  not  of  nature of  the  experimental of  would  a  be  were  analyses  and  the  plots  performed  on  were MINITAB  weighted (Ryan  s e t s and BMDP ( D i x o n  by et  1985)  that  (a)  plot  individual  ate  neighbouring For the  and movement If  seedling  being  reason,  plots  i n the  from  size.  an  o r damaged one  this  measurements  only  independent  browsers.  and  sample  analysis  and browsed  feeding  other  increase.  ANOVAS on s e e d l i n g  data  considered  were a v e r a g e d w i t h i n e a c h of all  in  concluded  could  probability  treatment  models.  I  wildebeest,  affected  measurements sites  was  of  s e e d l i n g s w i t h i n b u r n e d and browsed  observations  data  all  1984). The  (b)  all  percentage  performed to (Zar  of  of  all five  large-scale ANOVAS  were  al.  1985)  for  small,  balanced  for  weighted,  mixed  effects  142  Results  Large-scale  Mortality For three  F i e l d Experiments  of  T r e e s and  the  p u r p o s e of  hundred t r e e s  throughout  the  throughout  the  Reserve, study.  s.e.=  another  1%  (n  However,  under e x t r e m e  For  elephants Within the  very dry  Other  sq.  five  killed  166  (22%)  753  or of  the  experienced  natural  the  areas.  Acac i a  bull  stand.  damaged 256 had  and  1984,  approximately 6  trees  of  causes.  localized  mature  fatally  damage  drought of in  a h e r d of  followed  mortality rates  elephant  the  selected  fates  from o t h e r  like  of  had k i l l e d  stands  on  five  0.5%)  treatment  of  seedlings  sites.  based  the =  rates  study  were  amongst (s.e.  km.  randomly  rates,  (34%)  many  of  broken  s i m i l a r damage d u r i n g  this  period.  Mortality the  to  were much h i g h e r 1984,  mortality  their  suffered  0.6%)  conditions,  a 2  t r e e s and a n o t h e r  branches.  s.e.=  trees  24 h o u r s t h e y  and  p e r y e a r due  i n O c t o b e r of  visited  3m h i g h were  marked  300,  of  than  Mature t r e e s  1.5%) =  rates  example,  c a l c u l a t i n g mature t r e e  greater  7% (n = 300,  mortality  Seedlings  plots  Estimates the  sites.  of (a)  = 0.6%)  by w i l d e b e e s t  (s.e.  = 0.3%)  the  6000  and (b)  of  averages Within  the and  plots  (a),  killed  natural fire  each  of  standard  marked  deviations  by e l e p h a n t s ,  causes.  killed  sites,  initially  and In t h e  of  seedlings  these experimental  seedlings were  for  percent  t r a m p l i n g and t h r a s h i n g ,  through other  browsed t r e a t m e n t  were c a l c u l a t e d  1%  another  4% in  (s.e. 1%  b u r n e d and  an a d d i t i o n a l 4%  (s.e.  143  =  0.8%)  of  mortality  all  marked  r a t e s averaged  annual m o r t a l i t y g r e a t l y replacement  of  2%  seedlings 10% ( s . e .  s e e d l i n g s per h a / y r )  across  Seedling  the  The g r e a t e s t wildebeest under 60%  effect  t r a m p l i n g was  1m i n h e i g h t of  all  =  0.6%)  the  five  effects of the  (Figure  5.2).  between  s e e d l i n g s were not by  this  below  extensive  1m t h r o u g h  elephants  and o t h e r  The f r a c t i o n other for  browsers the  wet  and n e i t h e r three-way, effects  of  on  stems  fire  and  burned  of  seedling  In e a c h  fire,  other  growth  in  and  seedlings 30% -  p e r i o d were removed 100%  of  marked  elephants,  and t h e y  in  stems  wildebeest  and  Although i n d i v i d u a l  g r o w t h was  effects  browsers,  sampling p e r i o d ,  previous 90 -  17 new  sites.  severely  inhibited  were e f f e c t i v e l y of  fire,  held  wildebeest,  ANOVA.  i n the than  greater  the  wildebeest, five  i n b u r n e d and b r o w s e d ,  Season, (Table  plots  site, 5.3).  unburned  plots  level).  losses  than  the  sites  browsed  only,  a  and Stem wet,  (Tukey's  Two o f other  and  study  using  dry season than the  5% s i g n i f i c a n c e  significantly  This  animals  removed by e l e p h a n t s ,  significant  greater  the  rate  experimental  c o m b i n e d was a n a l y z e d a c r o s s  all  year.  browsers.  mixed e f f e c t s  were  seedling  p e r annum ( a p p r o x .  b u r n e d nor b r o w s e d t r e a t m e n t  comparison at showed  combined  and d r y s e a s o n  significantly greater  their  utilization  the  estimated  sampling p e r i o d s .  killed,  Total  per  elephants,  and i n b u r n e d p l o t s  browsers  1.2%)  i n h i b i t i o n of  were removed by a c o m b i n a t i o n of other  of  fire,  stems marked i n t h e  unburned p l o t s  =  e x c e e d e d the  (s.e.  Inhibition:  each y e a r .  the  weighted, treatment loss  was  and  was  multiple five  three.  sites  144  ELEPHANTS  WILDEBEEST  OTHER BROWSERS  40  • — burned o - unburned o -j  30 H  S  CJ ECO 1  O E-  80  o 10 H  DRY  WET  DRY  DRY  WET  SEASONS  F i g u r e 5.2.  Percent of marked stems removed i n each season by e l e p h a n t s , w i l d e b e e s t , and other browsers (± one s.e. showing the v a r i a b i l i t y between s i t e s ) .  145  Table  5.3.  EFFECT  ANOVA r e s u l t s f o r t h e f r a c t i o n o f stems removed by: (a) a l l c a u s e s o f m o r t a l i t y c o m b i n e d ; (b) e l e p h a n t - r e l a t e d m o r t a l i t y o n l y ; a n d (c) w i l d e b e e s t - r e l a t e d m o r t a l i t y o n l y a c r o s s s e a s o n s , s i t e s , and t r e a t m e n t s . A weighted t h r e e - w a y m i x e d e f f e c t s ANOVA m o d e l was u s e d ( s i t e was c o n s i d e r e d a random e f f e c t and t h e o t h e r s were f i x e d ) . F - r a t i o s are reported below, * * = P<.01, * * * = P<.001.  D.F.  A L L CAUSES COMBINED  ELEPHANT ONLY  63.60***  SEASON  1  74.30***  SITE  4  4.76**  TREATMENT  2  REMAINDER  22  118.41***  n.s. 965.58***  WILDEBEEST ONLY  138.61*** n.s.  146  A  weighted  two-way  wildebeest-related  stem  were a n a l y z e d b e c a u s e season.  Migratory  marked s t e m s , to  four  These  by  migrants.  85%  over  were  not  there  plots  separately  Only d r y season  a r e not p r e s e n t  removed  residency killed  The r a n g e of  on  average their  i n the  by  study  years.  three  browsing,  and t r a m p l i n g by c o n c e n t r a t e d was  10  higher  i n burned than unburned p l o t s  n a t u r e of  areas  were  only  virtually all f o r by t h e  concentrate  owing  damage by w i l d e b e e s t .  were  dry season,  effect  lightly  damaged.  sites.  herds  significantly  site  These  movement p a t t e r n s  to  losses  the  seedlings  but  highly  in  differences  of m i g r a t o r y  wildebeest  follow  local  in areas  recent  rainfall  of  Stem  Whereas s e e d l i n g s  damaged,  in  other  may  be  wildebeest.  thunderstorms for  all  three  stem r e m o v a l by w i l d e b e e s t  significant  wet  annual  five  a  data  29% o f  unintentional  treatment  on  i n the  on t h e  locations  then  (Table 5.3).  combined, during  r u t t i n g males  some  the  performed  years  was  accounted  was  wildebeest  dry-season  three  localized  In  the  stems were p r i m a r i l y  thrashing of  loss  wildebeest  in a l l  month  ANOVA  three  and  to  four  alone  was  days. The  percent  of  stems  analyzed  across  seasons,  three-way  ANOVA  (Table 5.3).  patterns  of  higher  stem  use.  p r o p o r t i o n of  unburned  plots  significance  level).  marked stems e v e r y unburned  plots  removed  sites,  Elephants  resprouting  elephants  and t r e a t m e n t s  First,  (Tukey's  by  they stems  multiple  with a  showed removed a on  months  (Figure  5.2).  in  plots  comparison  burned  plots  Secondly,  important  significantly  burned  E l e p h a n t s removed an a v e r a g e six  two  weighted  at of  the  5%  33% o f  all  and they  than  20%  in  removed  147  significantly season  more stems  (Table 5.3,  effect  was  found.  melampus), thomsoni)  season Other  Grant's  of  14%  seasons.  such  granti)  (Madoqua  rhinoceros  burned  Elephants  plots removed  season  than  P<.001).  browsers,  dikdik  and  dry  effect,  (Gazella  gazelles,  camelopardalis), average  i n the  the  wet  No s i g n i f i c a n t  site  as  in  impala  and  (Aepyceros  Thomson's  (Gazella  giraffe  (Giraffa  kirkii),  (Diceros bicornis)  and  5%  unburned  plots  more stems t h a n o t h e r  b o t h b u r n e d and u n b u r n e d p l o t s  during  both  removed an  the  between  browsers  wet  and  on dry  season.  A  weighted  two-way  ANOVA was p e r f o r m e d on t h e  stems b r o w s e d ,  t r a m p l e d , o r not  plots  e a c h of  and  across treatment  seedlings versus  in  were  unburned  plots  16% i n b u r n e d p l o t s damage,  highly  significant  plots.  Browsing  only,  unburned p l o t s the  plots  had been due  had  been  no  differences  the  visible  some  and u n b u r n e d browsed  site  38%  of  on  differences  but  and  unburned  h i g h e r on b u r n e d p l o t s level).  browsed,  This  difference  majority  While  78% of  seedlings.  the  site  analysis  burned  extent.  a qualitative  Both  b r o w s i n g damage  A second  between  5% s i g n i f i c a n c e were  unburned  Overall,  showed no s i g n i f i c a n t  browsed t o  burned s e e d l i n g s  had  and  (Table 5.4).  significant.  was s i g n i f i c a n t l y  (at  to  sites  (Figure 5.3).  unburned s e e d l i n g s  probably  which  5 study  effects  browse  of  the  damaged i n b u r n e d  f r a c t i o n of  only  those  in  57%  burned  preference  between Of  than  was  resprouting,  those  seedlings  had damage t o  both  the  alone,  showed  no  in  percent  of  l e a v e s and s t e m s . A third significant  analysis  for  difference  trampling between  damage,  treatments  the  14U  Table  5.4.  ANOVA r e s u l t s f o r t h e f r a c t i o n o f stems damaged by: (a) a l l c a u s e s c o m b i n e d ; (b) b r o w s i n g damage o n l y ; a n d (c) t r a m p l i n g damage o n l y a c r o s s s i t e s and t r e a t m e n t s . A w e i g h t e d two-way m i x e d e f f e c t s ANOVA m o d e l was u s e d ( s i t e was c o n s i d e r e d a random e f f e c t and t r e a t m e n t f i x e d ) . F - r a t i o s are reported below, ** = P<.005 * * * = P<.001. /  EFFECT  D.F.  33.08**  SITE TREATMENT REMAINDER  A L L CAUSES COMBINED  13615.13*** 4  BROWSING ONLY  TRAMPLING ONLY  n.s.  54.62***  9459.93***  n. s.  149  100 90  •  -  BURNED  O  -  UNBURNED  80 70 60  O PS OH  50  4>  40  4  30 20 10  4 •  0 BROWSED  F i g u r e 5.3.  NOT DAMAGED  TRAMPLED  Percent of marked stems browsed," not damaged and trampled over the e n t i r e study p e r i o d (± one s.e. showing the v a r i a b i l i t y between s i t e s ) .  150  individuals plots).  t r a m p l e d (5% on u n b u r n e d  There  multiple  was,  comparison t e s t  had a s i g n i f i c a n t l y other  three  (at  e x p l a i n e d by t h e described  at  later  seedlings  another.  the  p e r i o d of  majority  few  two o u t  browsed to  on  site  of  burned  effect.  the  five  A  sites  level).  This  of  the  is probably  wildebeest  impacts  recorded  The most time 90%)  however,  were  the  for  18  months,  much l o n g e r  of  seedlings  to  5.5  period  Removals  of  these  (Figure  months.  was  5.4).  in  54% t o o k  r e m a i n i n g 7% r e q u i r e d remove  but  from  The  or  fire,  T h o s e removed by  delays  another  by  removals  removed by w i l d e b e e s t  ground l e v e l  their  distinguished  resprouting  plants  of  these  over  recovery. a y e a r , 9% 2 years  the  to  greatest  seedlings  took  recover.  S e e d l i n g Growth:  Figure  detail  only d i d elephants at  each  resprouting  cause  possible.  e x p e r i e n c e d much g r e a t e r  T h e r e f o r e , not  proportion  The  easily  returned within six  and t h e  killed  ground l e v e l ,  whenever  interesting  required of  were a c t u a l l y  was t e r m e d " r e v e r s a l " and t h e s e  31% r e s p r o u t e d w i t h i n 6 m o n t h s ,  weighted  6%  stems t r a m p l e d t h a n  "regenerates".  and w i l d e b e e s t  did resprout,  regrow.  of  seedlings  This process  called was  (over  elephants,  study  that  localized patterns  repeatedly  fire,  one  took  sigificant  5% s i g n i f i c a n c e  highly  were  elephant,  Only  the  date.  disappearance  that  a  higher percentage  relatively  many were  a  showed  and  above.  Although year,  however,  plots  stem numbers and  shows t h e averaged  two-way  change over  heights  i n stem h e i g h t s  all  sites  ANOVA model was u s e d t o  for  over  the  entire  each treatment.  examine  stem  A  heights  151  FIRE 100 -i  CY  80-  2  60-  E or  40-  w  E  200-  WILDEBEEST 100 60  g D  35  6 0  Sf  40 20 0  ELEPHANT 100 80 Z  60 40  35  20 0  F i g u r e 5.4.  rV\l 0-6 6-12 1 2-18 18-24 24+ NUMBER OF MONTHS UNTIL RESPROUTING  Frequency d i s t r i b u t i o n , of r e t u r n times f o r r e s p r o u t i n g "regenerates" o r i g i n a l l y removed by e l e p h a n t s , w i l d e b e e s t , and fire.  152  60  -I  • 50-  40-  30a "Q  20• - CLIPPED EXCLOSURE  10-  o  -  UNCLIPPED EXCLOSURE BROWSED ONLY BROWSED AND BURNED  i  J  0  1982  1983  \O  1984  1985  TIME OF YEAR  Figure  5.5.  Change i n the average h e i g h t s of marked stems over the study p e r i o d . Arrows i n d i c a t e experimental burns.  153  amongst  sites  significant (F = 1 . 9 6 ,  and t r e a t m e n t s  differences n.s.)  A weighted overall  two-way  differences  the  sites  (F = 1.66,  change  i n stem  588.59, the  study.  seedlings  w h i c h were both  n.s.),  burned  However,  stem  heights  only  declined  three  from  divided  in  differences  those  38cm,  (a)  26cm,  that  the  mean  stem  w i t h i n the i n the  undipped  treatment  that grew  (b)  study.  plots  did  treatment browsed  whereas  in exclosures  =  16cm grew  those over  between  year.  seedlings the  seedlings  (F  only  Stems w h i c h were to  the  level).  browsed  throughout  average,  in  and"those  browsed and b u r n e d  level).  between  browsed  significance and  the  showed  browsed o n l y t r e a t m e n t  i n the  on  nor  of  at  browsed o n l y  5%  burned  treatments  burned  declined  i n the  In c o n t r a s t ,  The g r a p h of  those  significant  b u r n e d and browsed d r o p p e d from 37.5cm t o  10cm and 15cm p e r  those  found  between  the  and an a v e r a g e  were  the  sites  between  differences  (at  No  examine  (c)  browsed  5% s i g n i f i c a n c e  years.  suggested  to  plots  found  heights  as much as  the  exclosure  neither  and  stem  (at  (b),  more t h a n t h o s e w h i c h were  plots  w h i c h were  were  were  treatment  plots  performed  of  study.  between  Tukey's m u l t i p l e comparison t e s t s  which  the  the  n.s.).  i n stem h e i g h t s  however,  browsed and b u r n e d  not d e c l i n e  (F = 3 . 0 1 ,  browsed o n l y p l o t s  heights  significantly  (c)  changes  No s i g n i f i c a n t  P<.001).  b e g i n n i n g of  was  c l i p p e d and u n d i p p e d h a l v e s  end of  In  (a),  plots  ANOVA  in  the  f o u n d i n stem h e i g h t s  or t r e a t m e n t  and b r o w s e d p l o t s the  were  at  (c)  into  heights neither  clipped half.  over  time  (Figure  b u r n e d nor b r o w s e d half So,  two s e p a r a t e  grew in  even  5.5)  treatment more  than  a further analysis  treatments  -  I  c l i p p e d and  154  undipped.  I then performed a weighted  change  stem  in  plots: or  heights  between  b u r n e d and b r o w s e d ,  two-way  study s i t e s  browsed o n l y ,  ANOVA  i n four  on  treatment  c l i p p e d w i t h no  b r o w s i n g a n d u n d i p p e d w i t h no b u r n i n g o r b r o w s i n g .  no d i f f e r e n c e s was  still  with  were  f o u n d between  a significant  the  reduced  treatment  sample s i z e  sites  (F = 1 . 8 9 ,  the  burning Although  n.s.),  effect  (F = 3 9 . 6 1 ,  resulting  from t h e  there  P<.01)  even  s e p a r a t i o n of  c l i p p e d and u n d i p p e d t r e a t m e n t s . The clipped  change  than u n d i p p e d p l o t s  significantly both  more  at  the  clipped  declines  half  started  grew t o  burning  enhance  height  while  study.  showed which  clipped  36cm,  negative effect  the  undipped half  The r e m o v a l of  the  swamp t h i s  the  of  in  comparison  A l t h o u g h both the  g r o w t h , but  and b r o w s i n g a p p e a r t o  treatments  multiple  from an a v e r a g e  the  greater  two t r e a t m e n t s  (Tukey's  level).  end o f  seedling  these  other  55cm on a v e r a g e ,  grew t o o n l y 47cm by t h e appears to  significantly  than the  5% s i g n i f i c a n c e  undipped plots  was  and e a c h o f  growth  showed stem h e i g h t  tests and  i n stem h e i g h t s  grass  effects  of  in unprotected  plots. No per  significant  plant  treatments (Figure  were  of  5.6).  In  of  A  amongst  n.sl)  weighted  entire  study  n.s.).  plots, stems  there per  at  i n the sites the  number  (F  3.11,  =  beginning  two-way still  average  ANOVA  by t r e a t m e n t  of  in e i t h e r  the  stems  n.s.) this  i n the  or study  in  significant  stem site  a v e r a g e number  (F = 4 1 0 . 8 5 ,  was no a p p a r e n t change  plant  of  on c h a n g e s  showed no  However, c h a n g e s  per p l a n t d i d vary  exclosure  number  the  (F = 3 . 9 3 ,  stems  found  (F = 2 . 4 9 ,  numbers o v e r effect  differences  i n the  P<.001). average  c l i p p e d or u n d i p p e d  155  7-1  E< CO  4-  w CO  3-  o b  2-  .  • - CLIPPED EXCLOSURE  1-  o 0  UNCLIPPED EXCLOSURE BROWSED ONLY BROWSED AND BURNED  J  i  1982  1983  1 1984  1985  TIME OF Y E A R  F i g u r e 5.6.  Change i n the average number of stems per marked s e e d l i n g over the study p e r i o d . Arrows i n d i c a t e e x p e r i m e n t a l burns.  156  sides.  Those p l o t s  steady  decline  to  1.7.  loss  in average  stems  on b r o w s e d p l o t s  exclosures  significance number of  or  By  (5.5  which  significance  level).  stimulated  an  the  but  were  due t o  rose  of  that  (at  a 3.6  this  greater  than  the  5%  study,  the  average  greater  in  burned  in exclosure  plots  b u r n e d nor b r o w s e d  (at  (3.5  t h e 5%  experiments  apparently  i n stem n u m b e r s ,  stem numbers  browsing i n  again  from  the  than  burning  increase  individual  plots  significantly  While  showed  significantly  end  neither  burned  revealed  browsed  stems/plant)  immediate  rapidly  was  and  stems p e r p l a n t was  stems/plant)  declined  burned  level).  and b r o w s e d p l o t s  burns  stem numbers p e r  Tukey's m u l t i p l e comparison t e s t s  of  either  w h i c h were b r o w s e d but n o t  following  the  time  each  periods  subsequent  between  burn  (Figure  5.6) .  M u l t i p l e - b u r n Experiments Figure  5.7  multiple-burn  shows t h e  experiments.  plotted against group.  p a t t e r n of  the  Plants  in  The p e r c e n t  number of the  seedling  burning  lightest  of  mortality  plants  trials  for  burning  over  the  (300g/m ) 2  entire had o v e r  survivorship the  end of  showed  a  Those i n the  the  study.  similar  pattern  i n the  survivorship  The  had  the  leaving only most  of  sequence fallen  plants  severe  10% o f  (I50g/m ) 2  being  the  treatment  treatments:  below  treatment  seventh  survivorship, of  each  second b u r n i n g  80% s u r v i v o r s h i p u n t i l  dropped r a p i d l y ,  occurred e a r l i e r burn  period.  the  s u r v i v i n g are  treatment  showed v e r y h i g h s u r v i v o r s h i p w i t h o n l y e i g h t  in  treatment fire  60% and by t h e  when  plants  by  (600g/m ) 2  b u t heavy by  lost  losses  the final  fifth burn  157  100  -i  90  OH  co  80  DH  70  >  60  o >  50 CO EH  iz;  w o  DH W DH  40 30 20  A V O  10  - 150 g/m - 300 g/m - - 600 g/m  2 2  O  2  200 - 400 g/iri  \  0 0  Figure  5.7.  1  2  3  4  5  6  7  8  BURN NUMBER  9  -9  10  Seedling survivorship f o r multiple-burn e x p e r i m e n t s under t h r e e b u r n i n g t r e a t m e n t s . Note : * = s e e d l i n g s u r v i v o r s h i p f o r three, sequential, large-scale experimental burns where f u e l l o a d i n g s were 200 - 400 g/m . 2  1 58  there  were  no  survivorship fuel  loads  from  between  survivorship first  survivors  three  the  at  first  patterns  were  of  Figure  large-scale  the  trials  all.  the  5.7  also  burn experiments  and s e c o n d in l i n e  treatment  shows  w h i c h had  levels.  The  w i t h t h o s e p r o d u c e d by  the  multiple-burn experiments.  Seed G e r m i n a t i o n E x p e r i m e n t s The  fire-treated  significant  effect  5.8,  F = 1175.5,  64%)  in those  the  cooler  the  P<.001).  d e c l i n e d as  highly  on g e r m i n a t i o n r a t e s  (Figure  Germination  at  to  400  rates  g/m  g/m  than  2  200  were  higher  (mean =  i n those exposed  g/m  2  (mean = 29%),  to or  A l l three  of  produced higher germination rates  than those  of  loadings  600  demonstrated  (mean = 40%).  (mean = 18%).  fuel  a  e x p e r i e n c e d at  temperatures  seeds  loadings  seeds exposed  treatments  control  experiments  fuel  temperatures  hotter  these  of  seed  2  The r a t e  rose  to  of  800  successful  and 1000  germination  g/m . 2  Discussion If  woodland  existing  seedling  would be a s s u r e d . still  high  pressure  but  regeneration densities,  were  the  recovery  Seedling densities may  be  steadily  and  Acac i a  woodlands  predation,  wildfires.  and  in  While the  declining  Croze  Serengeti  Norton-Griffiths  of  dependent Mara  throughout the  from t r a m p l i n g by w i l d e b e e s t ,  others,  strictly  woodlands Reserve  under  could  (1979)  concluded  withstand  predicted  are  continued  b r o w s i n g by e l e p h a n t s (1974b)  on  and that  elephant  that  the  159  O O  125-  C\2  OS OH  100-  Q 75-  s OS o co Q  CO  50-  25-  d  0  CON 200 400 600 800 F U E L LOADINGS ( g / m )  1000  2  F i g u r e 5.8.  The average number of seeds g e r m i n a t i n g i n t e n s e t s (200 seeds each) under each of f i v e d i f f e r e n t f u e l l o a d i n g t r e a t m e n t s (± one s . d . ) . F u e l l o a d i n g s of 200, 400, and 600 g/m a l l s t i m u l a t e d g e r m i n a t i o n beyond c o n t r o l l e v e l s . 2  160  increase a  w h i c h has  occurred in wildebeest  reduction in f i r e s  this  recovery  regeneration elephants,  and a g r e a t e r  has  is  not  fire,  chance  occurred  severely  numbers w o u l d l e a d  yet  of in  i n h i b i t e d by t h e  and w i l d e b e e s t  woodland  (Chapter  recovery,  the Mara. combined  to  Woodland  effects  of  6).  Elephants Under require  good  15-20  years  vulnerable  to  this  height  they  fire  and  Belsky  r e a c h 3m, a t t h e most  above  (1974b)  this  ( C r o z e 1974a and P e l l e w  demonstrate  that  factor  in seedling  mortality  growth  i n the  findings.  elephants  avoided  First,  & b,  are  are  longer  they  reach  trampling  1983).  "largely  and  i n f l u e n c e d by Based  Norton-Griffiths  little  the  Mara may no  Until  still  Pellew  on  results  on  (1979)  ignored"  affect  browsing is  and  late  possible Croze  trees  seedling from  this  c u r r e n t l y a primary  inhibition  up  under to  studied  explanations  (1974a & b)  1960s and e a r l y  proportion  trees  comprising woodlands  several  i n the  approximate  they  In c o n t r a s t ,  elephant  the  of  seedling  Mara.  There are in  fires.  elephants  Serengeti.  study  which they  (1983),  had  in  from b r o w s i n g ,  height  therefore,  i n the  trees  severe  damage  assumed t h a t  and,  survivorship  conditions,  experience  (1984)  1m  to  but  and g i r a f f e  work by C r o z e  under  all  even  elephants  and  growing  60%  to 1m. of  (Lamprey e t  1970s  height  Trees  above  al.  total  this  concluded that  their  the  for  browsed  difference Serengeti trees  availabilities, 1m  were  Glover  but  abundant,  p o p u l a t i o n i n the  1967,  in  1968,  Acac i a Croze  161  1974b,  and N o r t o n - G r i f f i t h s  major  change  progressive Unlike  the  current  loss  the of  height  height  towards 4)  in  the  the  1979).  height  taller classes  s t r u c t u r e of  showed t h a t  d i s t r i b u t i o n of  distribution  0 -  of  (Dublin  trees  class  there  has  following a  1984,  Chapter  2).  1960s,  the  the is  (Figure 5.9).  heavily Dublin  the  browse d i e t s  of  Mara came from t r e e s  less  than  1m, w h i c h i s  in proportion  current  distribution  woodland s t a n d s . feeding change  i n the  appear  to  highest  From  a  densities  often  through  it  i n t h e Mara  individual  studies  animals  on  elephants after  the  forage  (Dublin,  seedling  to mistake  indirect  is  methods  seedlings, completely  findings  from o t h e r  elephant  populations.  and  Bell  (1985),  the  a n d Ross  O k u l a and S i s e  can is  removed.  in  Field  4),  there  recorded  (1961),  the  from  this  occur  in  Mara.  feeding  on  selection  observations  whereas of  earlier  elephants  and t h e i r d i s t r i b u t i o n among  foraging  Buss  Acac i a  the  their  preferences  elephant  (1957),  the  mistakenly  of  close-up  Chapter  forage  to  in  Seedlings  areas  through d i r e c t ,  on  may  I determined forage  i m p a c t on v e g e t a t i o n  These l a t t e r ,  seedlings  possible  the  trees.  i n open g r a s s l a n d is  in  resulted  nearby g r a s s e s .  determined the  their  habitats.  left  feeding  in  difference  may have  b r o w s i n g on  elephants  heights  observed  f o r g r a z i n g on g r a s s e s .  elephants  When  elephants  distance,  seedlings  tree  p o p u l a t i o n s t r u c t u r e of  be  their  of  the  b e h a v i o u r o f Mara e l e p h a n t s  Secondly,  by  Therefore,  of  biased  (Chapter  80% o f  frequency  been a  trees  in  t h e Mara w o o d l a n d s  1m h e i g h t  over  In t h e Mara  Mara  (1976),  (1986),  be  misleading.  no o b v i o u s The p a t t e r n s  are consistent Studies  by  Guy ( 1 9 7 6 ) ,  sign of with  Brooks Jachmann  and W e y e r h a e u s e r  (1985)  162  8  > 3  s  s  2-3  CROZE 1974  l-J  o  £  o  ii§illllllll  1-2  w <  DUBLIN 1984  j  1  1  90  1  •  1  60  i  '  30  .  0  I  30  I  I  I  60  90  PERCENT  F i g u r e 5.9.  D i s t r i b u t i o n by h e i g h t c l a s s f o r A c a c i a s i n the Seronera woodlands (Croze 1974b) and i n the Mara woodlands ( D u b l i n 1984).  163  all  reported  where  they  earlier  that were  studies  failed  to  elephants  utilized  available.  of  elephant  notice  the  It  is,  feeding  extent  regenerating therefore,  in  to  the  which  seedlings  possible  Serengeti elephants  that  simply utilized  seedlings. Elephants seedling  were  height  back t o  consistently  higher  proportion  of  these  taken  elephants  by  the  removed by f i r e  in  the  took  hence  the  the  period).  During  seedlings  were  affecting  s u r v i v o r s h i p and  Mara  local  habitat. number o f  the In  Mara, the  elephants  elephants intensive  in  plants  for  a  elephants  is  an  1960s  required  press)  recovery  on t h e  are being driven poaching  seasonal  those  elephants or  did  trampling  longer  recovery  from e l e p h a n t  browsing,  time  by t h e  extensive  to  other  factors  of  woodlands  area  were out  Dublin  resident  the  Because  elephants  prime  all  northern of  this  on t h e  out  to  of  and year.  large  the  Reserve  Mara  DouglasToday,  Serengeti influx  the  elephant  area supported a  1969,  not of  of  just  t r a v e l l e d n o r t h to  al.  in  c u r r e n t d i s t r i b u t i o n of  and 1970s t h i s  pressure.  m i g r a t i o n of  those  regeneration  s o u t h and i n c r e a s i n g human s e t t l e m e n t the  resprout,  The n o r t h e r n S e r e n g e t i ,  Watson et but  large  that  a  was  Although a  than d i d f i r e  which s e a s o n a l l y  1969,  effect  t o come back t h a n  suggests  longer  reducing  establishment.  populations.  (Watson and B e l l Hamilton,  plants  p e r i o d of  exposed  eventually  This  in This  plots.  longer  further exacerbated  elephant  s o u t h of  this  i m p a c t of is  much  or w i l d e b e e s t .  effects  the  did  cause  each y e a r .  burned  seedlings  damage t o  The  important  ground l e v e l  more e x t e n s i v e (and  most  from  by the  boundaries,  t h e Mara has  been  cut  164  off  almost  entirely.  concentrated  within  elephants  are  in  northern  Serengeti  into  area,  on  that  woodlands  Currently, the  Reserve  residence  i n the  (Figure  year-round.  could  thereby  elephants  be c u r b e d ,  are  5.10) If  and o v e r  poaching  elephants  temporarily easing  heavily  the  1000  in  the  might move back  browsing  pressure  Mara.  Fire Fire  has  woodlands Mara  and  (Dublin  (Buechner Pellew  played the  1984,  an  important  i n h i b i t i o n of Chapter  2)  and Dawkins 1961,  1983,  Trollope  role  in  regeneration  1984).  But t h e  changed  i n t h e Mara o v e r  the  past  5-10  1970s a  greater  of  the  migratory  population dry  number  numbers c l o s e  season  i n the M a r a .  residence,  the  c r o p of  past,  d r y season  vegetation, Most  fires  season  wildebeest abundant fuels. produce  burned  a r e not  i n t h e Mara now  in  February are  not to  However,  the  the  same  take  and  and  remove short high  as  their  At t h i s resident,  grass  of  fire  the m i d (whose  3 - 4  4).  in  destroying  time  the of  the the  woody  do o c c u r . short  the  p o r t i o n of  the  the  month  Whereas,  herbivores  as  has  spending  November-December  biomass  1968,  p r o p o r t i o n of  during  a significant  rains  Glover  damaging when t h e y  place  Africa  wildebeest  annually,  March.  present  enough  of  (Dublin Chapter  fires they  course  of  Since  have been  of  Serengeti-  i m p a c t of  years.  million)  the  areas  1967,  remove a s i g n i f i c a n t  dry grass  today  1.5  Over  wildebeest  standing hot  to  decline  i n the  and i n many o t h e r Thomas and P r a t t  1982a,  the  dry  year  the  are  not  the do  grass not  long r a i n s and,  165  F i g u r e 5.10.  E l e p h a n t occupance w i t h i n t h e Mara Game Reserve from 12 a e r i a l s u r v e y s f l o w n by the Kenya Rangelands E c o l o g i c a l M o n i t o r i n g U n i t (K.R.E.M.U.) between 1980 and 1983. The b l o c k s r e p r e s e n t t h e number of f l i g h t s i n which e l e p h a n t s were seen i n each of t h e 10km x 10km g r i d s q u a r e s . Note : The s u r v e y s were f l o w n over a much l a r g e r a r e a (approx. 6,400 sq.km.), however, t h e o n l y e l e p h a n t s seen were i n o r c l o s e t o t h e R e s e r v e , as shown.  166  therefore, in  the  short  late  dry season  dry  The  characteristics  of  hot"  Acac i a g e r r a r d i i 5%. wide  The  Presently,  Trump  showed large  and  reported 1959,  merely  fires  to  1971,  repeated  experiments  had not  reduced the  this  of  through  the  the  effects  seedling  the  primarily  commonly  in A f r i c a  (Pienaar  1971,  Trollope  that  number of  repeated was  watering  after  most  woody p l a n t s  sped  of but  But,  indefinitely?  b u r n s on  and,  1982a,  eleven years  m u l t i p l e - b u r n experiments of  of  browsers  woodland development.  recovery  artificial  regrowth  limited  b u r n i n g and r e s p r o u t i n g go on question,  1972,  their  be  reported  and  i n d i c a t e d by the  species  to  Mara  fact,  Knight  (1974)  of  in  reappeared s h o r t l y  in  woody  Mara.  (Glover  g r a d u a l l y r e d u c e d by  is,  and  the  a  the  stimulated  seedlings  3  apply to  seems l i m i t e d  seedlings  R e s p r o u t i n g seemed t o  b u r n i n g of  in  of  only  and G u i n n e s s  b u r n i n g s as  in fact,  here  gerrardi i  Spinage  Acac i a g e r r a r d i i  Pratt  100% t o p k i l l  species  Acac i a  the these  m o r t a l i t y was  i n t h e Mara  Strang  simulate  process  fires  Although  in almost  tree  c o p p i c i n g and d e l a y e d  pattern  To answer  The  to  1971,  to  as  mimicked  impact d i s c u s s e d  as  Coppicing  response  led  designed  such  m a j o r i t y of  rains.  hot  this  fire  regeneration  Fire,  1986).  regular  damaging  burns  seedling  abundant  of  burns.  and the  Kennan  Sabiiti  of  total  and K n i g h t  history  next  as  well.  resulted  new stems w h i c h were t h e n  by w a t e r  can  Pratt  repeated  the  species  rootstocks.  multiple  fires  less  1976a & b ) .  a  wildfires  seedlings,  successful  1970,  Herlocker  these  grass  of  fire-resistant  not  experimental  patterns  variety  are  season.  large-scale  "moderately  fires  up  were  seedlings. in  therefore,  these the  167  results  may  have  annual  wildfires.  perspective, and  that  the  declined  crop a  to  burns  would  their  former  to  the  (Maddock 1 9 7 9 ) ,  which  is  Chapter  reduce  effects the  -  If  in  an  1979)  dry  process  finite  eventual populations and t h e y  grass  of  become  season.  woodland  (Norton-Griffiths  no  standing  could again  o r even t w i c e e a c h  the  by  conservative  wildebeest  the  caused  reserves are  (Sinclair  remove  once  a  result  3m.  taking place  favourable  actual  of  not  may  temperatures g/m ), 2  control  to  This  decline  1979,  Dublin,  be  true  the  were  soil  to  rates  burns  (200  However,  hotter  burns  significantly  gerrardii  (in  press)  germination ( 1982a)  that  burns  surface  cooler  Acacia  (1952),  are  a  followed  function  g/m ) 2  during  to of  germinate those  also  in  the the  exposed.  well  of by  of  average  of  argued  stimulation  which seeds a r e  somewhat  of  and  also  Story  Experimental findings  stimulated  The g e r m i n a t i o n  the  Trollope  r a t e s may be  fire  r e c o r d e d on t h e  levels.  when  conditions.  of  seeds  However, so  bed may  and Wein  stimulated  necessarily  soil  recovery.  and S a b i i t i  fires  germination  intensity  i n the  woodland  seeds.  only  that  on s e e d s  (1986),  Acac i a  moisture  suggest  for  exposure  was  germination  wildfires  Sabiiti  that  this  of  potential  (1959),  some s p e c i e s  rates  root  dry season w i l d f i r e s  accelerate  already  reported  that  hot  from  that  low numbers to  mortality  2).  The  Harker  under  Mara  occurrence  would p r e s u m a b l y  (400  suggest  trees  r a t e s of  even  results  of  migrated  common  Mara  However,  repeated  disappearance  longer  e x a g g e r a t e d the  At burns  beyond  seeds exposed  to  exceeded  controls.  r e t a r d e d the  germination  s e e d s compared t o  the  controls.  Such  168  fuel  loadings  conditions The  i n the Mara  proportion of  life  probably those  and  woodlands mature,  for  seeds  the  Reserve  conditions,  or  may e v e n t u a l l y  seeds  In t h e (1984) this  the  role soil  of  they  manipulating  elephant  browsed even more likely  actually fire  intensively  et  al.  1964,  Pellew  in  supply  i n the  Mara.  The (Figure (fuel  results 5.7)  loading  being depleted continued  of  that 200  by s h o r t  browsing  by  -  from  Under  current  movements by  potential  i n the of  this  1981)  Mara  existing rates  problem. Bell  and  and r e d u c i n g t h e in t h i s  seedlings  Jachmann  study, This  of  From  tool  impact  were h i g h  and browse  by  local  avoided burned a r e a s .  from my c o n t r o l l e d ,  suggest  elephants,  on b u r n e d p l o t s .  resprouting  (Dougall  like  Malawi,  However,  is are  c o u l d be an e f f e c t i v e  distributions  b e c a u s e newly  short  of  it  available  in reducing germination  woodlands  b r o w s i n g on w o o d l a n d s .  length  seeds  seed source  exacerbates  that  small  unknown,  inhabited  regeneration  elephants  is  long-distance  areas  fires  very  A l t h o u g h the  within.  The l i m i t e d  further  concluded  other  only  to  a  soil  herbivores,  not  hot  Brachystegia  the  deposited  travel  limit  found t h a t  their  and  poachers.  the  i n the  The o n l y  requires  also  pastoralists  f o u n d under n a t u r a l  trees.  within  transient  this  but  are  contain  seed-producing  indefinite. by  woodlands;  currently  already  carried  carriers  intensities  Mara.  of  not  outside  of  fire  all  in of  elephants was in  most  protein  kinds  was  multiple-burn  experiments  even u n d e r c u r r e n t b u r n i n g  conditions  400g/m ),  dry season elephants,  2  root  fires.  reserves This,  diminishes  the  are probably coupled  with  likelihood  of  169  woodland r e c o v e r y . regimes  will  grasslands  A  be  is  to  change  necessary be s l o w e d  in if  both the  or h a l t e d  burning  and  browsing  c u r r e n t t r e n d towards (see  Chapter  open  6).  Wildebeest Though t h e effectively Mara, in  c o m b i n a t i o n of  inhibiting  there  is  a third  removing the  fires  has  second,  been  discussed  less  obvious  Reserve  impact When  cessation The  of  wildebeest  Inadvertently, seedlings others. are  dry  bite  enter the  finally into  the  The r o l e  of  wildebeest  intensity  however,  wildebeest  the  Under  i n the  leave  following  tall  a dry wasteland,  of  grass,  Mara  feeding.  and  g r a z i n g as  stems  tall  the  significant  Mara,  these a r e a s , many  the  and  are  their  with  a  present  inhabiting  just  of  have  trampling  grasses  off  are hidden  When t h e y  in  season are having a  through  they  which  transformed  rains,  move  Acac i a w o o d l a n d s  l i m i t i n g the  through  first  long  be  Wildebeest,  numbers of  woody v e g e t a t i o n  the  to  i n woodland d y n a m i c s .  the  wildebeest  appears  factor.  above.  role  and f i r e  of  c r o p and t h u s  sheer  during  on t h e the  the  recovery  important  standing  day c o n d i t i o n s , Mara  the  elephants  the  green. they  go.  regenerating  and t h e y  trample  grazing  grounds  little  sign  of  live  seedlings. In  addition  effects,  male  vegetation. rutting  to  these  wildebeest  The S e r e n g e t i  p e r i o d when t h e y  wildebeest  ( f r o m the  trampling have  a  and further  wildebeest first  arrive  Mara R e s e r v e  and  accidental  are at  browsing  i m p a c t on t h e  woody  the  their  end  i n the  Mara,  Loita  Plains  of  and  resident  area)  are  170  still  in  rutting trees  the  ungulates,  i n the  Mara)  (1984)  it  of  seedlings  large  Other  other  that  i n the  reduce  numbers of  (which  bull  range  of  stems  as  is  damage  1m, a s  30-150cm. the  most in  Belsky  tops  and  Such t h r a s h i n g  plants  Mara e a c h  well.  broke  t r a m p l i n g and  i n the  is  underrepresented  under  of  other  s e e d l i n g s and s m a l l  wildebeest  vigour  factor  L i k e many  Though t h i s  in trees  the  a mortality  spar with  class  young  season.  but  is  not  browsing  of as  which  year.  Browsers In  addition  to  sufficient  habitat  for  mammals.  Among  rhinoceroses gazelles al.  breeding  males.  obvious  Acacias  significant kill  to  is  may  their  1-2m h e i g h t  reported  branches  of  male w i l d e b e e s t  in addition  prevalent the  peak  1979,  been  pers.  comm.).  did  a measureable each  ground  level  growth  by  seedlings  are  were  year  not  the  i n the  is  Mara t o  even feed  the  still  provides  of  other  browsing  Thomson's  of  of  a significant reach  This  factor  1m l e t  e_t  these  but  they  study  has  were r e d u c e d  to  inhibition  of  i n the  alone  few  (Tieszen  mortality, As t h i s  a  Grant's  suggested,  seedlings  species.  and  species  (1984)  major a g e n t s  browsing  some g i r a f f e ,  on woody  Belsky  majority  ever  Mara  diversity  i m p a c t on s e e d l i n g s .  by s e v e r a l browsing  As  the  impala, dikdik,  shown t o  probably  shown,  rich  seasonally  species have  a  these  and  have  elephants,  failure  grow beyond  of it.  171  Summary  The  number of  over  the  past  being  lost  at  In  study, by  Mara,  all  other  natural  Wildebeest, all  similar  in  under  Browsing  significantly  higher  m a j o r i t y of fire  suggests  than d i d f i r e  or  Seedlings decreases than  seedlings  of  these  potential  for  of  seedlings  and  In  this  a n n u a l l y by e l e p h a n t s ,  and o t h e r  4%  experienced browsers  browsers  i n unburned p l o t s While  for  in  than  did  in  and  unburned  ground l e v e l  were  browsers  burned  accordingly  to  greater  impacts  seedlings  months.  greater  and an even  and o t h e r  removed a t six  removed up  wildebeest  varied  the  i n the d r y  elephants  i n burned p l o t s  seedlings  plots. were plots.  by  wildebeest  T h o s e t a k e n by  elephants,  delays  more s e v e r e  in  recovery.  damage t o  the  This plants  wildebeest.  exposed  in height  those  killed  seedlings  elephants  reduced  inhibited.  and o t h e r  e x p e r i e n c e d much that  severely  elephants  resprouted within  however,  the  potential  treatments,  on  on  Seedlings  preference  effects  The f u t u r e  causes.  plots.  both  showed a d i s t i n c t  were  elephants,  burned  been  t r a m p l i n g and t h r a s h i n g and a n o t h e r 1%  s u r v i v i n g stems  proportion  or  from  has  seedlings.  being  seedlings  impacts  season.  dependent  growth  is  Mara  8% p e r y e a r .  is  the  rootstocks  greatest  The  over  1% by w i l d e b e e s t  through  60% of  of  i n the  The r e m a i n i n g m a t u r e t r e e s a r e now  from r e g e n e r a t i n g  4% of  fire,  years.  woodlands  the  coppicing  30  a rate  disappearing replacement  mature t r e e s  over  seedlings  w h i c h were  to  browsing the  study  animals but were  w h i c h had been  b r o w s e d but  not  showed  significant  significantly  taller  b u r n e d and b r o w s e d .  Those  burned  showed  a  steady  172  decline  in  browsing,  s t i m u l a t e d an i n c r e a s e  seedlings.  stem  In  numbers.  exclosure  b u r n e d nor b r o w s e d , while  the  they  average  However, i n the  plots,  burning,  stem numbers  where  grew a maximum of  number  of  coupled  with  of  marked  s e e d l i n g s were  neither  10 -  15cm  per  year  stems r e m a i n e d more o r l e s s  the  same. Multiple-burn survivorship  experiments  was  inversely  intensity.  Seedling  cool  (fuel  burns  after  repeated  germination of  200  -  600g/m  Under will  the  s u r v i v o r s h i p remained h i g h  after  repeated  2  were  burns  150g/m ) (fuel  or even  higher it  the  is  be r e p l a c e d and t h a t persist.  Reserve  to  >  300g/m ).  successfully  fuel  Seed  2  by  fires  mature  trees  i n t h e Mara  will  loadings.  u n l i k e l y that woodlands  A significant  prevailing inhibitory factors  within  loads  s t i m u l a t e d beyond c o n t r o l l e v e l s  but d e c l i n e d a t  fire  but d r o p p e d s i g n i f i c a n t l y  2  current conditions,  eventually  expand  of  hot  to  seedling  level  up t o  related  that  the  loads  rates  demonstrated  change  i n one o r more o f  must t a k e p l a c e regenerate.  for  woodlands  173  CHAPTER 6.  DYNAMICS OF THE MARA WOODLANDS: PAST,  PRESENT, AND  FUTURE  Introduction The  role  of  much c o n t r o v e r s y In  the  cause  throughout  Ross 1982,  1962,  Field  al.  Laws e t  1975,  Tsavo  no e x c e p t i o n  the  1960s,  were  the  to  Park,  point  for  study  1964,  Laws 1969,  1970,  c o n t r o l l e d c r o p p i n g of  forefront. ecologists halt  Throughout  Pienaar  practiced  (Glover  1965,  1969).  (Pienaar  And, today,  in  i n a number of  Vos e t  al.  1983,  these  cases,  if  countries  the  O w e n - S m i t h 1983, any,  has  it  Pienaar  ever  been  1968,  Lamprey  in  the  controversy.  had become  1983)  1966,  (Bell  woodlands  a  focal  ( G l o v e r and S h e l d r i c k  Africa,  culling  Barnes  managers  populations  e_t a l .  1976,  and  1970).  Park  Parker  southern  reserves  Serengeti-Mara  problems"  elephant  the  Harrington  Guy  is  and t h e was much  park  had a l r e a d y begun c r o p p i n g e l e p h a n t s  woodland l o s s e s  1969,  1971,  1975,  and  M y e r s 1973,  as  1970,  involved  "elephant  cited  Wing and Buss  in neighbouring Kenya,  of  been  and  years.  1966,  G l o v e r and Trump  deeply  the  1961,  The  researchers  stimulated  over  parks  and Angus  case  has  B u e c h n e r and Dawkins  Thomson  this  1968,  already  National  al.  Spence  have  in  1947,  1985).  Russell  Serengeti  of  al.  1967,  By  losses  P i e n a a r et 1971,  and managers  elephants  (Eggeling  Weyerhaeuser  presented et  cases  woodland  Africa  1970, 1974,  of  of  B r o o k s and Buss  i n woodland d e c l i n e s  among e c o l o g i s t s  majority  primary  Laws  elephants  in  the  managers  and  i n an a t t e m p t  van Wyk and still  1983,  In  to  Fairall  advocated  Cumming  1983). fully  question  and  1983,  de  few  of  established  that  174  elephants trends  are,  in  fact,  the  p r i m a r y agent  from d e n s e w o o d l a n d t o  My  study  of  the  open  decline  driving  and d y n a m i c s of address  deductive  were c e n t r a l  First, over  what  the  the  factor(s)  past  currently  30  the that  hypothesis the  And,  any  for  this  the  second,  significant  elephants,  that  1960s,  during  effect rates  on to  Mara  issue to  in  this  woodland  what  a  study.  declines  factor(s)  r e c o v e r y of  rates  Griffiths  are  woodlands  1979),  i n the  states  seedlings  regeneration. of  of  in  1967,  states  they  The  fire  the  early  that  the  that  could  damage measured most  extreme  1974a  & b),  had been  fire,  hypothesis 1960s  Alternatively,  alone,  offset predicts  (Sinclair  through  its  burning Norton-  documented  the  through i t s  capable  The h y p o t h e s i s  predicts Reserve  that  tree  loss  and M a r a .  question,  alone,  adult  1975,  c a u s i n g the  is  w i t h i n the  This  the  not  and s m a l l t r e e s  burning,  the  under 3m, r e d u c e d r e c r u i t m e n t  second  fire,  of  for  trees.  losses.  northern Serengeti  to  elephant  Croze  observed  hypothesis  responsible  period  had been c a p a b l e of  answer  hypothesis  rates  in  were  rates  the  the  elephant"  i m p a c t s on mature  seedlings  where  rates.  woodlands In  and  point  measured  the  al. for  "1960s  alone,  hypothesis  trees  the  mortality  et  to account  "1960s f i r e "  the  through t h e i r  (Lamprey  sufficient  question,  predicts  early  declines  on  years?  first  observed d e c l i n e s  of  responsible  the Masai  Reserve?  suggests  the  questions  were  preventing  Under  in  Two  observed  grasslands.  w o o d l a n d s p r o v i d e d an o p p o r t u n i t y t o manner.  the  of that  "1980s  repeated  fire" effect  i n h i b i t i n g woodland the  boundaries,  present could halt  day the  175  recovery today. in  of  Using  the  the  Mara  hypothesis both  t h e Mara w o o d l a n d s a n d  over  argues  seedlings  regeneration established the Mara, that  that  fire the  is  under (over  fully  To  3m),  levels  burning  trees,  I recorded  "1980s  are  elephant" on  preventing  From r e c e n t  the  census work, I  unprecedented highs  areas  elephants  of  their  taking into b o t h on t h e  woodlands w i l l  to  resident  range,  these  four  individual  and  the  on  consideration seedling  the  n o t be a b l e  primary and  (other  i n each time  escaped  in  to  class  effects recover  of with  elephants. hypotheses,  synergistic  than  I  also  effects  elephants)  at  of  their  scenario. plausibility  of  these  different  I have d e v e l o p e d a model w h i c h i n c o r p o r a t e s a number factors  dynamics. under  rates,  a  involved  in  The model i s variety  elephant  browsing r a t e s , the  past  impact  determined  from a c t u a l measurements  i n the  present the  above  Combinations  wildebeest of  and  t o examine  conditions.  effects,  model a n d r e s u l t s  and  of  both  designed  thrashing  the  the  that  seen  through t h e i r e f f e c t s  numbers were a t  browsers  measureable  hypotheses  mature  of  investigate  woodland  alone,  I predicted that,  the  the and  hypotheses,  years,  1m) and t h o s e w h i c h have  addition  wildebeest  patterns  now e x p e r i e n c i n g heavy b r o w s i n g by e l e p h a n t s  effect  investigated  of  elephants  c u r r e n t numbers o f  estimated  four  t h e Mara w o o d l a n d s .  basis.  observed  In  past  elephant  the  b r o w s i n g by e l e p h a n t s  due t o p o a c h i n g i n o t h e r  t h e Mara  (trees  of  the  that  and of  a year-round the  rates  produce  resident field,  of  t r a m p l i n g and browsers, are  used  a r e t h e n compared w i t h p r e d i c t i o n s .  as in  176  The  1.  The A s s u m p t i o n s  This and  model  dynamic  is  necessarily limited  a first  natural  simultaneously.  is  Model  system  For  necessary  simulating  which  many  reason,  but,  nonetheless,  In d e s i g n i n g  a qualitative  to  make  a  conditions  under w h i c h t h e  The  assumptions  stated  in  at  this  simplistic  way.  attempt  number  this  of  a  complicated  factors  initial  operate  effort  informative model o f  this  assumptions  model c a n be e x p e c t e d  type  about  to  s h o u l d be b o t h b i o l o g i c a l l y  in  be  is its it the  accurate.  realistic  and  conservative. The  factors  included within impact  the  to  the  model a r e  the  other  dynamics  annual rates adult  trees,  i n the  of: b)  a)  wildebeest,  than  and  d)  Mara and elephant  b u r n i n g , c)  t r a m p l i n g , t h r a s h i n g and i n a d v e r t e n t  migratory  browsers  A.  the  woodland  on s e e d l i n g s and m a t u r e ,  damage due by  affecting  browsing  the  browsing  by  resident  elephants.  Elephants Elephants affect  woodlands treated  of as  determined question 1970s,  the  Mara.  from a e r i a l  have  the  of  to  heights of  counts).  found i n  population In a n s w e r i n g  woodland d e c l i n e s  that  the  Acac i a  b r o w s i n g by e l e p h a n t s  elephant  total  c a u s e s of  assumed  all  The r a t e s  proportional  about I  trees  elephants  fed  in the  sizes the  are (as  initial  1960s  a c c o r d i n g to  and the  1 77  patterns  stated  by C r o z e  ignored  seedlings  (1974a  &  1m  during  under  were more a b u n d a n t .  In  answering  current  woodland d y n a m i c s ,  direct  observations  measured e f f e c t s height  B.  of  i n the area  trees  also  all  present  elephant have  largely  t i m e when l a r g e r second  feeding had  burned  over  rates  are  question  trees about  on  from  b e h a v i o u r and from  the  marked  of  all  The b u r n i n g  rate  lit  therefore,  trees  that  at  is  d e t e r m i n e d by t h e  (Norton-Griffiths the  percentage  of  correlated  only with  people  fires  the  resident  in t h i s  turn  and t h e r e  not  of  also  fires,  relatively  few  human p o p u l a t i o n s  i n the  a r e a were v e r y  that  loadings  fires small.  but  assumption  are  believed I  assumed,  although there  were p r o b a b l y s u f f i c i e n t  severe  assumed  by l i g h t n i n g .  century,  is  (when  This  Africa  which  burned  fuel  nearby.  p a r t of  started the  is  percentage  1979),  trees  It  by humans and not  wildebeest support  most  to  t h a n 3m h i g h .  distributed).  number of  made b e c a u s e  less  time  be e q u i v a l e n t  w i t h the  be  trees  calculations  are u n i f o r m l y  burning  to  the  they  I have d e t e r m i n e d b r o w s i n g r a t e s  elephants  affects  assumed t o  few  the  that  Fire  used  to  of  and  classes.  Fire  is  b)  fuel  occurred  were  loadings because  1 78  C.  Wildebeest In  trees  the  under  Mara, 1m.  wildebeest  The  seedlings  due  during  p e r i o d of  this  the  period,  recovery did  to  not  measureable reached  them t o  animals.  D.  Resident Like  in  the  Browsers  Mara,  trees  this  and  Grant's gazelles,  censuses with  involved  that  significantly assumed  that  size  of  visit  movements (Pennycuick to  when  have they  a r o u n d 1.5 the  their  a had  million  Mara  during  1m.  the  in  Giraffe a  census  work  woodland  browser  over  past  browsing  rates  of  year,  population  15 -  forage. dikdik,  are  primary  the  Recent  data)  s i z e s have  20 y e a r s .  by a n i m a l s  other  low  impala,  dynamics.  unpublished  affect  relatively  desirable  such as  the  to  as  aerial  compared  ( S i n c l a i r and N o r t o n - G r i f f i t h s  resident the  numbers a r e  species,  some t i m e s of the  Mara a r e assumed  shortage  smaller  at  elephants)  i n the  ( S i n c l a i r and D u b l i n ,  earlier  indicate  that  only  During  following  numbers  on  negligible  1960s.  assumed  dynamics  than  browsers  under  So,  browsers  (other  p r o b a b l y due t o  means  be  those  loss  migratory  were  these animals  to  low  in s i g n i f i c a n t  population  stem  i n the  annual  only  year.  wildebeest,  those  and t h e i r  woodland  The m a j o r i t y of  dry season each  were c o n s i d e r e d  Wildebeest  current  affect  and  numbers were s t i l l  Mara  1979).  the  only  the  i m p a c t on  their  mortality  r a p i d woodland l o s s  wildebeest  Maddock  of  wildebeest  from r i n d e r p e s t , take  1975,  rates  significantly  not  Therefore,  1979)  changed I  than e l e p h a n t s  have were  179  the in  same the  i n the  Mara  Only  1960s as  fire  and e l e p h a n t s  and r e v e r s i o n  whereas,  wildebeest  1979,  inhibition  (see  Constants  several the  agents  &  Pellew  are considered  trees as  into  to  them t o  be a g e n t s of  smaller  reversion  Pellew  be  1983),  height  and  classes,  inhibition  and b r o w s e r s  both  act  agents  only  as  Appendix A ) .  population  constants.  (1974a  shown  and V a r i a b l e s  tree  results  of  act  (Norton-Griffiths  The  have  today.  mortality  2.  my measurements  of  al.  Herlocker 1983),  we're c o n v e r t e d  in  These c o n s t a n t s  Lamprey e t  b),  (1981,  used  to  10  variables  are  model  (Table 6.1)  (1967),  (1976a),  in addition a  the  is are  Glover  described  d e r i v e d from  (1968),  Croze  N o r t o n - G r i f f i t h s (1979),  t o my own d a t a .  hectare  basis  for  by  Tree the  and  densities  purpose  of  6.2).  Six  analysis. Seven of  these v a r i a b l e s  "inhibition" seedlings given  rates.  of  adult  "Reversal"  s e e d l i n g s which are resprout  burning  some rate,  browsed o r  burned each year  refer  refer  have  to  the  the  to  percentage  of  ground l e v e l  removed  the  to  of the  but  still  variable,  woodland percent  of  percentage  though  The s e v e n t h  proportion  be e q u a l  or  proportion  p e r t a i n to  stems  "reversion",  k i l l e d a n n u a l l y by a  to  ground l e v e l .  from t h e  and assumed  (Table  t r a m p l e d back t o  and " i n h i b i t i o n " r a t e s  derived  model  "mortality",  rates  rates  stems above is  as  the  t r e e s which are  s e e d l i n g s which r e p e a t e d l y  maintain  into  expressed  Mortality  or mature,  factor.  later  are  built  of  area trees  180  Table 6.1.  Constants which are b u i l t  i n t o the model.  FOR ALL PLANTS; 1) S e e d l i n g 2) A d u l t  d e n s i t i e s : 850/ha  tree  (>3m) d e n s i t i e s : 32/ha  3) New s e e d l i n g s :  17/ha/year  4) Growth o f s e e d l i n g s :  15cm/year  5) F i r e escapement h e i g h t :  3m  6) N a t u r a l m o r t a l i t y r a t e o f a d u l t t r e e s and s e e d l i n g s :  1%/year  FOR BURNED PLANTS: 1) Number o f p l a n t s under lm which a r e k i l l e d by f i r e : 5% 2) Number o f p l a n t s i n the 0-lm c l a s s which r e v e r t t o the 0-lm c l a s s : 95% 3) Number o f p l a n t s i n the l-2m c l a s s which r e v e r t t o the 0-lm c l a s s : 90% 4) Number o f p l a n t s i n the 2-3m c l a s s which r e v e r t t o the 0-lm c l a s s : 5% 5) Number o f p l a n t s i n the 2-3m c l a s s which r e v e r t t o the l-2m c l a s s : 29%  181  Table 6.2.  The parameters v a r i e d i n the model.  1) Burning r a t e 2) E l e p h a n t m o r t a l i t y r a t e on 3) E l e p h a n t r e v e r s a l r a t e on  seedlings seedlings  4) Wildebeest i n h i b i t i o n r a t e on 5) W i l d e b e e s t r e v e r s a l r a t e on 6) Browser i n h i b i t i o n r a t e on  seedlings  seedlings seedlings  7) M o r t a l i t y r a t e o f mature, a d u l t  trees  182  exposed  to  fires  hypotheses, actual  values  woodlands the  the  input values  reported  d u r i n g the  1980s.  time  scenario  (i.  using  appropriate  time  Test  the  two  (Table  "elephant  and  woodland  of  data  are  the  pertinent  value  of  answering  my two  set  at  Serengeti  i n the  1980s)  Mara for  in  each  and  each v a r i a b l e  then for  Recruitment  factor  than then  natural  t h o s e of (R)  the  tree  log  of is  was  synergistic  effects  of  elephants the  3m t o m a t u r e  When l n ( R )  adult  is  to  the  r a t i o of trees  population  of  fire,  were  effects  held within  of  these  browsers  and  fire.  trees  entering If  growing.  used  so  that  greater  than  0,  adult  tree  of  varied  R is  m o r t a l i t y and t h e  the  inhibition  dying. is  the  the  other and  I  explain  rates  interest  the  the  other  for  from  individual effects  all  of  To d e t e r m i n e  taller  the  factors,  rate  than u n i t y  and  To i s o l a t e  were added t o  exceeding  generated  current  on woodland d y n a m i c s ,  zero.  were  questions,  1960s and one  the  ranges.  the  the  1960s  realistic  class  for  i n the  the  wildebeest  one  central  forth  while  is  the  put  constant,  is  and  hypotheses  other  figures,  in  "fire"  and  value  1960s  Predictions  recovery.  elephants,  greater  studies  from my own s t u d i e s  sets,  6.3).  woodland l o s s e s  height  seven v a r i a b l e s  period.  purpose  established  factors  four working  Conditions  For  1980s  these  the  a r e made from e a c h h y p o t h e s i s  the  the  examining  field  1960s o r  e.  In of  from  Predictions  considered  3.  annually.  the  population  the  R  is  For  the  threshold  recruitment is  growing,  Table  6.3.  The t e s t c o n d i t i o n s f o r the 1960s and 198 which were used i n the s i m u l a t i o n model.  THE 1960s  450 62.0% 2.5% 6.0% 1.0% 26.0%  approximate e l e p h a n t p o p u l a t i o n s i z e burning rate o v e r a l l elephant-related adult tree mortality l o c a l i z e d elephant-related adult tree mortality other natural m o r t a l i t y of adult trees browser i n h i b i t i o n r a t e on s e e d l i n g s  THE 1980s  1100 5.0% 7.0% 4.0% 1.0% 8.0% 1.0% 29.0% 26.0%  approximate elephant p o p u l a t i o n s i z e burning rate elephant-related adult tree mortality e l e p h a n t m o r t a l i t y r a t e on s e e d l i n g s other natural m o r t a l i t y of adult trees e l e p h a n t r e v e r s a l r a t e on s e e d l i n g s w i l d e b e e s t r e v e r s a l r a t e on s e e d l i n g s w i l d e b e e s t i n h i b i t i o n r a t e on s e e d l i n g s browser i n h i b i t i o n r a t e on s e e d l i n g s  184  and v i c e  versa.  Results  Woodland D e c l i n e s The  i n the  values  of  variables  using data c o l l e c t e d time.  It  ignored  seedlings  1960s  i n the  were  used  I used the  mortality rates  central  (1974b),  Serengeti  conservative  estimates  mortality  of  rates  were a l s o  set  rate  ln(R) adult  elephants,  woodlands  factors  were of  of  set  less  were  have  at  alone,  were  than  0  mortality rates  of  then  at  an  this  Burning  period,  time.  the  more  the by  tall species  wildebeest  time. for  the  hypothesis), overall  3.5% p e r y e a r  the  100% b u r n i n g  26% p e r y e a r and  should occur.  of  the  Browsing r a t e s ,  responsible  elephant"  zero,  model.  throughout  this  (1967)  C r o z e ' s more  r e p o r t e d a 75 -  1975). set  al.  elephant-related  measured  burning  tree First,  l o c a l areas  of  that  adult  Second,  rate  those  no i m p a c t a t  ("1960s at  62%  of  et  run t h r o u g h the  (1979)  (Sinclair  elephants,  were c o n s i d e r e d t o  the  overall  a c c o r d i n g to  average  was u s e d  than  If  the  6%).  a r e a b o r d e r i n g t h e Mara d u r i n g  conservative grasslands  affected  (approx.  were  Norton-Griffiths  i n the  other  2.5%  two l e v e l s  set  largely  in these c a l c u l a t i o n s .  severely  for  period elephants  r e p o r t e d by Lamprey  woodlands,  tree  Although  for  1960s s c e n a r i o were  N a t i o n a l Park d u r i n g  this  However,  by e l e p h a n t s  Croze  during  1m.  mortality  and  the  Serengeti  was assumed t h a t under  in  Table  decline  and i f  all  recruitment 6.4  (2.5% due  other value  shows t h a t to  in  at  elephants,  Table  6.4.  The In(Recruitment r a t e s ) f o r f i r e , e l e p h a n t s , and browsers a c t i n g alone i n the 1960s s c e n a r i o . Those v a l u e s l e s s than zero i n d i c a t e woodland l o s s , whereas those v a l u e s g r e a t e r than zero i n d i c a t e woodland i n c r e a s e .  FACTOR Fire  In(Recruitment -2.91  Elephants (at 2.5% e l e p h a n t - r e l a t e d adult tree mortality)  2.74  Elephants (at 6.0% e l e p h a n t - r e l a t e d adult tree mortality)  2.05  Browsers  2.46  rate)  186  plus,  1.0% due t o  At  highest  the  to  plus,  v a l u e was  still  predicted  by t h e  then  Table  6.4  shows  different current  addition  related  ln(R)  addition  Figure  1.0%  rates  1.16  due  the  ln(R)  contrary  to  that  the  1960s  were h e l d  other  factors  s h o u l d be l e s s  to  fire  be  -2.91  reduce  6.1  shows  elephants  under  the  the  (as  ln(R)  ln(R) a  without  8.23.  When a d u l t  tree  (2.5%  due t o e l e p h a n t s ,  for  percentage  of  other  browsers or  estimate  mortality  6.1).  were  at  set  (2.5%/year)  (1.0%/year)  (Figure  and  Even when  their  highest  above z e r o  (0.47).  to  negative  7.0%/year  other  elephantreported  fire,  alone,  (Figure  6.2).  When  due the  were  1.0% due t o  in  t h o s e of  (6.0%  causes),  mortality rates plus  value  A t C r o z e ' s (1974b)  mortality  natural  these  values  and b r o w s e r s a r e a l l a c t i n g t o g e t h e r  to other  0.  coupled  browsing e f f e c t s  at  than  and  tree  more  at  hypothesis.  to  l n ( R ) was s t i l l  are set  due  causes)  b r o w s e r s b u t no w i l d e b e e s t ,  ln(R) value  elephants,  mortality  of  natural  of  (6.0%  all  (ln(R))  with the  mortality rates  (6.0%/year),  7.0%  2.74.  of  were u n a b l e  was  equals  woodland d e c l i n e s  numbers i n t h e Mara)  other  of  2.05,  value  elephant-related  o n l y made t h e  ln(R) value  natural  at  browsers and w i l d e b e e s t .  tree  The  plus  ln(R)  with other  to  browsers,  fire,  alone,  elephant  overall  levels  the  population sizes  with other for  and i f  (Table 6.4).  wildebeest,  0,  caused the  i n agreement  zero  other  recruitment value  Browsers, below  the  hypothesis.  hypothesis),  the  conditions,  than  elephant alone,  fire"  zero,  1.0% due t o  greater  fire,  ("1960s  causes),  reported adult mortality rates  elephants,  If  natural  to  and a d u l t elephants,  l n ( R ) v a l u e was set  other  at  natural  -  3.5%/year causes),  187  —6.0  '—•—i—|—i—i—i—j—i—i—i—|—i—i—i—|—i—i—i—j—i  0  20  40  60  80  100  % OF C U R R E N T E L E P H A N T POPULATION  F i g u r e 6.1.  The n a t u r a l l o g of r e c r u i t m e n t r a t e s (R) f o r v a r y i n g e l e p h a n t p o p u l a t i o n s : a) w i t h o u t w i l d e b e e s t or browsers; b) w i t h browsers but no w i l d e b e e s t ; and c) w i t h both w i l d e b e e s t and b r o w s e r s . Those v a l u e s l e s s than z e r o i n d i c a t e woodland l o s s , w h i l e those v a l u e s g r e a t e r than z e r o i n d i c a t e woodland i n c r e a s e s . Note : The d o t t e d l i n e a t 40% r e p r e s e n t s e l e p h a n t p o p u l a t i o n s i z e s of the 1960s. The d o t t e d l i n e a t 100% r e p r e s e n t s the c u r r e n t e l e p h a n t p o p u l a t i o n s i z e i n the Mara. B u r n i n g rates are held at zero.  188  0  20  40  60  80  100  B U R N I N G RATE (%)  F i g u r e 6.2.  The n a t u r a l l o g of r e c r u i t m e n t r a t e s (R) f o r v a r y i n g b u r n i n g r a t e s : a) w i t h o u t w i l d e b e e s t or browsers; b) w i t h browsers but no w i l d e b e e s t ; and c ) w i t h both w i l d e b e e s t and browsers. Those v a l u e s l e s s than z e r o i n d i c a t e woodland l o s s , w h i l e those v a l u e s g r e a t e r than z e r o i n d i c a t e woodland i n c r e a s e s . Note : The d o t t e d l i n e a t 62% r e p r e s e n t s t h e b u r n i n g r a t e s of t h e 1960s. The d o t t e d l i n e a t 5% r e p r e s e n t s t h e c u r r e n t b u r n i n g r a t e s i n t h e Mara. E l e p h a n t impacts a r e h e l d a t z e r o .  189  the  ln(R)  the  actual  occurred  was  -7.95.  1960s  (Chapter  This  scenario  those  r e c o r d e d i n the  been  very  have moved t o For  the  day  burning  highest counted and  i n the  Hamilton, 8.0%  Browsing as  rates  recent  these  of  in recent  years,  The  Talbot  year.  responsible  al.  numbers,  tree  elephants the  (1986)  responsible,  then at  fire,  6.5  of  Talbot  that  of  hypothesis  today's zero,  and F i g u r e  acting alone,  6.2 is  current  at  their  times  those (Stewart  D u b l i n and D o u g l a s were  i n the  change  set  natural  at  causes).  1960s  and S t e w a r t  the  scenario  (1964)  i n the  migratory  states  and  population  ln(R)  that  show t h a t 4.66,  rates, value the  in  wildebeest.  model. fire,  woodland r e c o v e r y .  burning the  set  important a d d i t i o n a l f a c t o r  i n h i b i t i o n of  being held at  Table  those used  have  season.  t h e Mara  rates  incorporated in this  fire" the  mortality  The most  been  1964,  from  wildebeest  are  2.4  1970s f o r  set  dry  I have  over  shows l i t t l e  is  the  and 1.0% due t o  same a s  species.  for  losses  rates  the  Elephants  and S t e w a r t  Adult  to  since  numbers d u r i n g  1960s and e a r l y  were  "1980s  woodland  Burning  modelling exercise,  population  i m p a c t s have  factors  study.  per  c u r r e n t day s i t u a t i o n  Their  describes  1980s s c e n a r i o were  this  c e n s u s d a t a and t h a t  sizes  for  5%  late  due  et  0.  at  in press).  Stelfox  the  this  1962,  (7.0%  of  in large  of  recorded  Talbot  course  t h e Mara  rates  significant  under the  reduced  purposes  when  accurately  1980s  variable values  much  c a s e most  2).  Woodland D y n a m i c s i n t h e All  latter  with  alone, If  fire  all  ln(R)  a number c o n s i d e r a b l y  is  other  s h o u l d be l e s s  current  is  than value  greater  Table  6.5.  The In(Recruitment r a t e s ) f o r f i r e , e l e p h a n t s , browsers, and w i l d e b e e s t a c t i n g alone i n the 1980s s c e n a r i o . Those v a l u e s l e s s than zero i n d i c a t e woodland l o s s , whereas those v a l u e g r e a t e r than zero i n d i c a t e woodland i n c r e a s e .  FACTOR Fire Elephants (at 7.0% e l e p h a n t - r e l a t e d adult tree mortality)  In(Recruitment 4.66  •0.14  Browsers  2.46  Wildebeest  2.58  rate)  191  than  zero.  seedling  Fire,  "1980s  the  absence  the  Mara  of  conditions,  value  in a ln(R) ln(R)  hypothesis  factors, In  regeneration  browsers  of  all  main  value value  states  can prevent  this  case,  of  inhibitor  of  that  elephants,  in  the  expansion  of  elephant  less  was,  under  than  in fact, are  impact,  alone,  Under  these  0.  -0.14  capable  current  the  current  ( T a b l e 6.5 of  conditions  burning  browsers  even  conditions  factors,  and  preventing but not  the  to  of  a  (Figure  for elephants,  fire,  wildebeest,  value  woodlands  are  still  declining.  Woodland D y n a m i c s i n t h e challenge  the  equals  6.1).  -5.64,  and  early  model  1900s  g r o w t h would be g e n e r a t e d , in  similar  negative added  still  remain  wildebeest drives  Under  and o t h e r indicating  ln(R)  current browsers, that  the  1890s under  a  variety  I went back and r e c o n s t r u c t e d  1890s  the  are  of  elephants  zero  ln(R)  operating  addition  in  to a  values  f u r t h e r below  the  conditions,  ln(R)  effects  ln(R)  and b r o w s e r s  O b v i o u s l y , the to  considered  reduced the  effects,  combined,  To  when  Even when w i l d e b e e s t  (Figure 6.2).  other  values  nor w i l d e b e e s t ,  other  (Table 6.5).  positive  late  the  degree.  absence  and  not  Therefore, elephants  Neither  to  any o t h e r  the  6.1).  seedling great  elephant"  woodlands.  result  Figure  is  regeneration.  The  should  therefore,  to  assuming  ways  then  see that as  the  what  of  plausible  conditions p a t t e r n of  of  woodland  woodland dynamics they  are  the  today.  were It  is  1 92  important this  to  note t h a t  assumption.  possible  from  (Chapter  2),  rainfall,  Test  my  burning  rates,  the  where  decades to  no  be m i n i m a l  And,  I assumed  different  their  numbers, of  all  underestimate  well  turn the  of  If  of  Under above  populations. expansion  of  elephant  the  amounts.  impact  was  (though  rinderpest  did affect  not  this  to  scenario,  be  of  browsers the  browsers  This  which took  is  the  set  burning effects  at  the  indicating consistent  place  be  an  reduced  and  the  and,  Natural  an with  between t h e  rate  were s e t  current  century conditions,  (3.08),  nearby.  reduce  combined  therefore, mortality  of  was  set  at  zero,  and  1.0%.  and w i l d e b e e s t  other  presumed  also  this rates.  the  several  could  had  period  and  significantly  four  recruitment  for  in areas  this  rinderpest  during  to  F i r e s were  if  factors  for  Likewise,  population  century.  today  impact  as  period  wildebeest  may o v e r e s t i m a t e  turn  zero  epidemic,  t h e n my c a l c u l a t i o n  elephant  impacts  the  were r e p o r t e d by h u n t e r s  of  their  accurately  records a v a i l a b l e  negligible  the  browsing  that  as of  low human p o p u l a t i o n s  t r e e s was c o n s i d e r e d  5%/year,  year.  reduced to  the  tree  To examine  the  that  numbers).  effects  adult  due t o  from  overestimation their  the  set  accompany  numbers.  rinderpest  reduced  which  reconstruction  or a n i m a l  elephants  following  were  were no a c t u a l  great  p o a c h i n g had g r e a t l y  limitations  conditions  there  human numbers had been  point  are  historical  however,  Following  there  the  rate  late  of  ln(R)  increase the  at  pattern  in of  26% p e r  value  was  woodland woodland  1890s and t h e  1940s.  193  Escapement Rates  of  T r e e s under D i f f e r e n t  In a d d i t i o n t o height  class,  variety  of  I also  classes as  to  these  Figure  6.3  as  loglO  class  relative  make  these  assumed t h a t the  and t h e  turn  of  the  r e a c h the  in  class  calculations  the  turn  to  of  of  the  adult  recruitment  into a  escaping the  each  rates  approximately  one  2m c l a s s  in Figure  existing  i n t h e Mara  (except  for  escapement  lowest.  This  is  whereas into  consistent  reported for  the  the  11  the  conditions  two t i m e  1)  conditions  greatest  patterns  years,  different  of  periods.  the  of  the  today.  The  escapement  rates  of  the  15cm and 1m c l a s s e s )  with the  these  in  6.3:  those c o n d i t i o n s  the  To have  2)  c e n t u r y p e r i o d showed  in  I  century,  classes,  class.  I have c o n s i d e r e d t h r e e presented  I  height  by a g e ,  the  height  class.  are presented  from  in  the  3m  zero height  survival  15cm c l a s s  into  the  and t h e y  in 6 years,  15 y e a r s .  into a l l  and d e c l i n e  rates  100,000 s t a r t i n g a t  plants  at  rates  reaching  number  comparable  i n the  and 3)  the  the to  escapement of  1m h e i g h t  conditions  prior  rates  3m c l a s s  scenarios  1960s,  to  recruitment rates  looked at  height  refer  year,  l o o k i n g at  Scenarios  1960s  were  the  woodland growth  194  F i g u r e 6.3.  Escapement v a l u e s f o r A c a c i a q e r r a r d i i t o 0.15m, 1.0m, 2.0m, and 3.0m h e i g h t c l a s s e s p l o t t e d as t h e l o g ( a c o h o r t of 100,000 i n d i v i d u a l s i n the s t a r t i n g p o p u l a t i o n ) . I assume t h a t 0.15m i s a c h i e v e d a t 1 y e a r , 1.0m a t 6 y e a r s , 2.0m a t 11 y e a r s , and 3.0m a t a p p r o x i m a t e l y 15 y e a r s . Three d i f f e r e n t time s c e n a r i o s a r e p r e s e n t e d .  195  Discussion  The  1960s and The  the  1970s  hypothesis  observed  unlikely. related  that  woodland  Even under t h e tree  rates  fire  was e a s i l y  alone  mortality  rates  I  Therefore, responsible lower  for  the  Norton-Griffiths  northern  own.  the  1962,  Dublin  and  blamed  for  These prevailing  and loss  on  moving t h e  Talbot  Mara  Talbot  which  to  6.4).  were  reduce  However, well  below  conservative  ( N o r t o n - G r i f f i t h s 1979).  explanation  were  the  that  fire  was  Browsers served o n l y  not  capable  contention  others of  that  woodland  region.  local  woodlands  tree  of  to  reducing  canopy  stands,  boundaries during 1964,  Croze  elephants  i n t o a phase  and S t e w a r t  of  Despite  p o p u l a t i o n s were  Douglas-Hamilton, the  elephant-  able  (Table  time  the  but  and Mara  effects  and  rates  p e r i o d of  support  Although elephant  Serengeti  not  of  own.  Serengeti  c a p a b l e of  were  for  and 1970s seems  to hold recruitment rates  reject  (1979)  for  significant  elephants  responsible  conditions  woodland d e c l i n e s .  findings  responsible  extreme  rates  on t h e i r  These  1960s  most  burning  that  recruitment  woodlands  the  in  able  c o u l d not for  were  below m o r t a l i t y r a t e s  even a t  per year)  alone,  losses  mortality,  recruitment  (62%  elephants,  cover  decline  period  Watson  in p r e s s ) ,  and  they  not  in  the  sometimes  elephants  increasing this  were  their  of  (1974b),  were on  not  their  within  the  (Stewart  and  Bell  cannot  1969,  fairly  be  loss.  results  support  burning conditions  the of  argument the  time,  that fire,  under  alone,  was  the able  196  to  halt  trend  the  succession  towards  uncommonly  the  being  rinderpest,  held  significantly the  early  northern  Current  Dynamics  Though from t h o s e  is  effects  of  impact  measured that  rates  in  seems l i k e l y .  With  wildebeest,  the  woodlands  to  Mara. Mara  fires  of  are  adult  at  said of  (Table exceed for  1980s,  rates the  the  the  woodland  recruitment  and  fire,  elephant  preventing  mortality  of are  that  early  holding  browsers  be  rates  the  to  factor(s)  burning rates  G i v e n the  decreasing  different  no s i g n  hypothesis  during  balance  is  the  Dublin,  continue  there  same c a n n o t  alone,  in  1970,  what  current the  fuels  seems i m p l a u s i b l e  Elephants are  addition are  the  recovery  However,  the  of  disease,  Consequently,  a_l.  present  were  grass  woodlands  question  elephants,  below those n e c e s s a r y  and a t  under  in the  the  and s e v e r e  the  was  productivity  viral  1979).  et  period  Mara a r e c u r r e n t l y v e r y  regeneration,  rates.  elephants  hypothesis  6.5).  the  reduce  frequent  observed  populations  by t h e  (Sinclair  ago,  i n h i b i t i n g woodland  Recruitment  recovery  i n the  to  woodland  mortality  wildebeest  (Langridge  Chapter 2),  In answer  preventing  adult  season  r e c o r d e d 20 y e a r s  recovery.  high grass  t h e y d i d not  and Mara  conditions  (Dublin  6.5).  dry  the  during this  unusually  In a d d i t i o n ,  therefore,  Serengeti  decline  Rainfall  1960s was a p e r i o d of  2).  and p r o d u c e  below c a r r y i n g c a p a c i t y  each  Chapter  alone,  promoting  area.  and,  woodlands  grasslands.  high,  throughout still  open  of  rates (Table  migratory  an even more r a p i d  rate  197  (Figure  6.1).  Wildebeest  and  coupled with current burning drive  recruitment  rates  other  rates  below  browsers  (Figure  6.2),  mortality  in  together, still the  or  cannot  absence  of  elephants. Due  to  migratory  the  present  wildebeest  high  (Chapter 4),  no l o n g e r  sufficient  season.  Today  burning  season,  when f u e l  loads  and  fires  are  conditions concern there  to  less  persist,  in  the  once  of  is are  lower,  then  fire  inhibition  loadings  cannot  be  end o f  dry  the  in  a g a i n become a major e l e m e n t  is  considered  wildebeest i n the  the  short dry  do o c c u r .  woodland r e c o v e r y .  change  the are  moisture content they  by  i n t h e Mara  r e s t r i c t e d to  when  of  offtake  b u r n s by t h e  largely  damaging  grass  fuel  support hot  were a s i g n i f i c a n t  could  rate  higher, If  these  a  major  However,  if  numbers,  fire  c o n t r o l of  woody  regenerat i o n . Elephants,  alone,  now  woodland  populations  Serengeti  and  the  settlement  have  led  to  Reserve  distribution  in  appear the  a  in  numbers the  press).  recorded in the  available  season in  the  (Chapters  Croze  efforts (1974b)  in past  1,000  on t h e  land  the few  years  t h e Mara  particularly there  elephants  "weeding"  reported that  patterns  elephants,  Because  out  i n the  to  hold  of  down in  surrounding  These e l e p h a n t s  material,  classes,  able  Poaching a c t i v i t y  most  Today  1960s.  3 & 4).  l a r g e r height  browsing 5).  browse  of  the  y e a r - r o u n d by a p p r o x i m a t e l y  be  Mara.  change  and numbers i n  Douglas-Hamilton,  to  of  the the  elephant  (Dublin  and  is  utilized  all  over  twice  the  rely heavily during  the  on dry  is  a s c a r c i t y of  trees  now  concentrate  their  seedlings  1960s  and  (Chapter 4 & early  1970s  198  elephants today  in  trees  elephant  the  Serengeti  under  1m  diets.  largely  comprise  Although  a  this  significant change  availability  rather  repercussions  f o r woodland d y n a m i c s .  concluded, sensitive  the to  impacts  trees.  cycle"  hypothesis  in  the  have  Mara  In  of  However,  the  previous  adult  decline.  Today,  a  grassland  phase.  stable  fires)  elephants This  states  as  states. state  fire  is  necessary  tree  elephants  another, can h o l d  but it  once  there.  be  than to  f a r more those  "stable work  browsing.  on  limit unless  Seedlings  elephants  (i.  e.  are  numbers  and  suggests sense  the  rates  increased the  the  Mara has  in  the  in two  factor  s y s t e m between t h e s e system  rate  (1973)),  an e x t e r n a l  move t h e is  burning  Holling  and t h a t the  decades  Mara e c o s y s t e m  that of  not  r e d u c e d by  did accelerate  h o l d i n g the  move  it  had been  increased  elephants  were  r e p o r t e d two  densities  Elephants a p p a r e n t l y cannot to  important  to  cannot  woodlands,  the  to  of  immunity.  pattern (in  of  (1976)  from e l e p h a n t  p r e d o m i n a n t l y woodland or g r a s s l a n d , such  seedlings  recovery  in wildebeest  man-made  proportion  has  seem  woodland d e c l i n e s  perturbation  of  locally  Mara  the  once t h e  a reduction of  no s u c h  the  initiating  frequency  refuge  but  As N o r t o n - G r i f f i t h s (1979)  Caughley's  woodland  a safe  c a s e of  ago.  following  of  t o d a y have  the  capable  Furthermore,  it  woodlands  on r e g e n e r a t i n g  seedlings  may be a r e f l e c t i o n  preference,  Serengeti-Mara  mature  seedlings  than  ignored these  from  grassland  two one  phase,  199  The  Future These  findings  the  future  of  the  1)  Would  the  due  to  helps  to  g r a p h of  not  included  postulated be  a total highly  Given present  stabilize  population current  in  further  low. if  question  impacts of  declined  and  elephants  6.4).  on  Mara  an  the of  of  per  would have  to  a balance  of  in elephant  current  if  we  b u r n i n g each  year.  the  50% l o w e r  tree  even  but  rates  elephant than  their  recruitment  i n woodlands  numbers  levels  recovery  year,  be  scenarios  Even  woodland  rate  5%  increase  The  wildebeest.  elephants,  rates  achieve  and  would r e q u i r e a  if  burning  rates  dry season b u r n i n g r a t e s w i l l  not  numbers r e m a i n h i g h .  2)  What w o u l d happen  if  to  the  reduced  and  (Figure  of  wildebeest  about  numbers  fire  the  Realistically,  system  of  browsing at  attain  reduction  remained increase  To  elephant  interaction  dependent  numbers t o  mortality.  the  decline  the  if  graph i n c l u d e  burning  questions  factors?  this  by t h i s  some i n t e r e s t i n g  woodlands:  woodlands  answer  represented  will  Mara  p o a c h i n g or o t h e r  A simple  have  stimulate  a virulent  strain  wildebeest  of  rinderpest  numbers  to  returned pre-1960s  levels? At similar low  and  the to  this  fire  human  of  the  scenario;  woodlands  difference, because  turn  century  wildebeest  increased.  conditions and e l e p h a n t  However,  p r o b a b l y had a l i m i t e d populations  in  the  were,  there  role  in  fact,  numbers  were  was  in the  a r e a had a l s o  one  notable  early been  1900s  reduced.  200  % OF C U R R E N T POPULATION  F i g u r e 6.4.  The n a t u r a l l o g of r e c r u i t m e n t r a t e s (R) f o r both v a r y i n g elephant p o p u l a t i o n s i z e s and burning r a t e s . In a l l cases, the impacts of wildebeest are h e l d at zero and those of other browsers are h e l d at t h e i r c u r r e n t r a t e . Burning r a t e s are i n d i c a t e d on the r i g h t , v e r t i c a l a x i s . Those v a l u e s l e s s than zero i n d i c a t e woodland l o s s , while those v a l u e s g r e a t e r than zero i n d i c a t e woodland i n c r e a s e s .  201  Today,  the  area adjacent  year-round basis always for  used  their  fires  rates  fire  if  might  year)  as  At  well  what  recruitment  is  wildebeest  and f i r e  Mara  is  i n the  the  would,  to  populated  cultivation  of  limiting  amount of  numbers  return  densely  These p a s t o r a l i s t  The main f a c t o r  woodland d y n a m i c s .  3)  a tool  by man t o d a y  Therefore,  the  (Lamprey 1 9 8 5 ) .  livestock.  set  to  fuel  were  t h o s e of  the  have  pasture  land  spread  available  to  today  1960s  burning  (above 60% p e r  become  an i m p o r t a n t  factor  would then  become  important to  ask:  burning  rate  would  and m o r t a l i t y  rates  6.4  to  we a c h i e v e  if  elephants  of  burn.  once a g a i n ,  It  a  peoples  the  reduced  on  a balance  in  between  and w i l d e b e e s t  were  reduced? Figure the  combined  allows  us  effects  of  impacts.  Even  if  Reserve,  the  woodlands  r a t e s were h e l d a t unlikely normal  i n the  (Norton-Griffiths  varying  elephants  less  Mara  were  than  40%  Given  the  would  significant  wildebeest  returned  their  1960s  size),  woodlands  to  population annual  burning rate,  For are  held  elephant  tree at  recruitment their  to  current  numbers had t o  be  recovery  elephant  be  likely  seems  of  numbers  an  of  over  75%  absence in the  the  sustained  of  Mara  current  u n d e r a 10%  possibility.  mortality  when  simulations  r e d u c e d by 40%  the  produced in a  i n the  40%  elephant from  This  grass  expected  at  unless burning  burning rates  levels,  and  eliminated  (approx.  exceed  by l o o k i n g  fire  year.  could only  an even l e s s  of  amount of  If  levels  levels  per  be  offtake.  question  begin  ( C h a p t e r 4)  1979)  this  totally  would not  possibility.  year  answer  or  wildebeest  revealed  more,  even  that with  202  burning  rates  assume t h a t but,  given  Africa this  held  fire the  c o u l d be t o t a l l y  are  not  to  reverse  this  wildebeest of  elephants held The  population  to  ecosystem  i n many  parts  of  reductions  of  alternative.  Mara seems t o  Nothing short  trend.  Fire  the  of  dramatic  c o n t r o l of  virtually  habitat  devoid  exist.  from f r e e l y numbers  It  utilizing  could  be h e l p e d  to  s t o p the  In  Africa  (Douglas-Hamilton  available  for  to  it  History  age  reasons them t o for has  of  of  is  if  extensive  Serengeti  elephant  1983),  is  it  historically  not  Park  poaching  When our  §_1.  Mara  which  cropping  unacceptable.  roam f r e e l y ,  the  the  poaching e f f o r t s  Mara  w h i c h keeps  Watson e t  into  the  of  alone,  1969,  them?  although areas  a r e a w h i c h has  elephants  be r e l i e v e d  north.  unnecessary  this  this  remove  which b o r d e r s  poaching,  able  the  to  low  seem  in  how c a n we a l l e v i a t e  elephants,  (Watson and B e l l  concentration  but  is  of  reductions  if  having p h y s i c a l l y  northern Serengeti,  locked  levels  without  the  but  be  burning  may be m a i n t a i n e d a t  elephants  is  desirable  the  high  resulting  secure  elephants  1979),  unreasonable  from t h i s  remain  large  could  it  numbers  Currently,  suitable  of  day c o n d i t i o n s ,  numbers c o m b i n e d w i t h  Reserve,  eliminated  a desirable  a grassland phase.  impacts  status  is  Implications  Under p r e s e n t  elephant  Not o n l y  (Douglas-Hamilton  magnitude  into  zero.  precarious  today  Management  at  1969). is  authorities plague  the  throughout  elephants there  not  is  for habitat  responsibility  them? repeatedly  shown  that  the  Serengeti-Mara is  by  203  no means a s t a b l e made,  system.  p l a y a major r o l e  Norton-Griffiths trends  i n the  1979).  in vegetation  closely  P e r t u r b a t i o n s , both n a t u r a l  For t h i s  and  monitored over  d y n a m i c s of  animal  time  so  occur.  From a management  track  changes  as  outcome change  they  when c h a n g e s as  planning  take  grassland,  the  future.  whether  it  should  local  its  own  now  the  pressures  f u t u r e of  rest  with  responsibility the them  essential  in  the  is  that  Mara  be  far better  than  as to  confront  the  Managers s h o u l d always  view  essential  consideration  should  be  in  s h o u l d r e m a i n an open  be c o n s c i o u s l y  without  (Sinclair &  managed t o  allowed  intervention  is  to  allow  proceed  a decision  for  authorities. Political  to  course  it  is  it  rather  an  area  man-  c a n be documented  perspective,  Whether t h e Mara  woodland r e c o v e r y or whether along  changes  place  thus,  it  populations  go u n n o t i c e d .  i n t e g r a l and, for  reason,  that  they  this  and  needs o f with  management  the  Kenyan  on  which  here  t h e Mara  in  may  central  for  the  Reserve  will  be  their  p l a n s which w i l l  address  government.  can  must base  have  future.  It  and t h e  scientists they  be  Conservation goals  l o c a l peoples  task,  presented of  realities  f o r m u l a t e management  both the  information findings  the Mara.  to  this  and economic  will  wildlife. p r o v i d e the  their  To  help  necessary  decisions.  important i m p l i c a t i o n s  The for  204  LITERATURE CITED Agnew,  A.D.Q. 1968. Observations Tsavo N a t i o n a l Park ( E a s t ) .  Albl,  P. the  1971. S t u d i e s on a s s e s s m e n t of p h y s i c a l c o n d i t i o n African elephant. Biol. Cons. 3:134-140.  Anderson, G.D. 1973. Vegetation damage i n t h e major h a b i t a t R e s e a r c h A r e a of R h o d e s i a . Rhodesia. A n d e r s o n , G . D . and B . H . W a l k e r . and e l e p h a n t damage i n t h e Rhodesia. J. Sth. Afr. Atang,  on t h e c h a n g i n g v e g e t a t i o n in E. Afr. Wild. J. 6:75-80. in  c o m p o s i t i o n and e l e p h a n t t y p e s of t h e Sengwa W i l d l i f e M . S c . t h e s i s , U n i v . of 1974. Vegetation composition Sengwa W i l d l i f e R e s e a r c h A r e a , Wildl. Mgmt. Assc. 4:1-14.  P . G . and W. P l o w r i g h t . 1969. E x t e n s i o n of t h e J P - 1 5 r i n d e r p e s t c o n t r o l campaign to E a s t e r n A f r i c a : the e p i z o o t i o l o g i c a l background. Bull. Epizoot. Pis. Afr. 17:161-170.  Barnes, R.F.W. National  1980. The d e c l i n e of Park, Tanzania. Afr.  the J.  baobab t r e e i n Ruaha Ecol. 18:243-252.  Barnes, R.F.W. National  1982. E l e p h a n t f e e d i n g b e h a v i o u r i n Ruaha Park, Tanzania. Afr. J. Ecol. 20:123-126.  Barnes, R.F.W. 1983. E f f e c t s o f e l e p h a n t b r o w s i n g on w o o d l a n d s i n a T a n z a n i a n N a t i o n a l P a r k : m e a s u r e m e n t s , m o d e l s , and management. J. Appl. Ecol. 20:521-540. Barnes, R.F.W. 1985. Woodland c h a n g e s ( T a n z a n i a ) between 1976 and 1982. 23:215-221 .  i n Ruaha N a t i o n a l Afr. J. Ecol.  Baumann, 0 . 1894a. Durch M a s a i l a n d zur N i l q u e l l e . Reimer, B e r l i n .  Park  Dietrich  Baumann, 0 . 1894b. D i e k a r t o g r a p h i s c h e n e r g e b n i s s e des E x p e d i t i o n des d e u t s c h e n a n t i - s k l a v a r e i - k o m i t e s . Petermann's, M i t t e i l u n g e n .  Masai.  Beaumont, E . 1944. C o n t r o l of e a s t c o a s t f e v e r i n t h e M a s a i D i s t r i c t - a g e n e r a l s u r v e y of t h e Lemek V a l l e y , Narok District. Kenya A r c h i v e s Document , D C / N R K . 1 / 1 2 / 2 . Bell,  R.H.V. use of Ecol.  and H . J a c h m a n n . 1984. I n f l u e n c e of f i r e B r a c h y s t e g i a w o o d l a n d by e l e p h a n t s . Afr. 22:157-163.  on J.  the  205  Bell,  R.H.V. 1983. D e c i s i o n - m a k i n g i n w i l d l i f e management w i t h r e f e r e n c e t o p r o b l e m s of o v e r p o p u l a t i o n . I n : Management of L a r g e Mammals i n A f r i c a n C o n s e r v a t i o n A r e a s (OwenSmith, R . N . , ed.) pp.145-172. Haum E d u c . P u b l . , P r e t o r i a .  B e l o v k s y , G . E . 1986. Generalist herbivore foraging role in competitive i n t e r a c t i o n s . Amer. Z o o l .  and i t s 26:51-69.  Belsky, A . J . 1984. R o l e of s m a l l b r o w s i n g mammals i n p r e v e n t i n g w o o d l a n d r e g e n e r a t i o n i n the S e r e n g e t i Park, Tanzania. Afr. J. Ecol. 22:271-279.  National  Bernsten, J . L . 1979. P a s t o r a l i s m , r a i d i n g , and p r o p h e t s : M a s a i l a n d in the n i n e t e e n t h c e n t u r y . Ph.D. d i s s e r t a t i o n , U n i v . of W i s e . 384pp. Bourn,  D. Arid  1978. Env.  Cattle, 1:49~61 .  rainfall,  and t s e t s e  B r a n a g a n , D . and J . A . Hammond. 1965. a review. Bull. Epizoot. Pis.  in A f r i c a .  J.  Rinderpest in Tanganyika: Afr. 13:225-246.  B r o o k m a n - A m i s s a h , J . , H a l l , J . B . , S w a i n e , M . P . , and J . Y . Attakorah. 1980. A r e - a s s e s s m e n t of a f i r e p r o t e c t i o n e x p e r i m e n t i n n o r t h - e a s t e r n Ghana s a v a n n a . J. Appl. Ecol. 17:85-99. B r o o k s , A . C . 1957. N o t e s on some e c o l o g i c a l s t u d i e s i n t h e Queen E l i z a b e t h N a t i o n a l Park w i t h p a r t i c u l a r r e f e r e n c e grasses. Game and F i s h e r i e s P e p t . , E n t e b b e , U g a n d a , 24pp.(mimeo). B r o o k s , A . C . and I . O . B u s s . 1962. P a s t and p r e s e n t s t a t u s elephant i n Uganda. J. Wild. Manage. 26:38-50. B r y a n t , J . P . , W i e l a n d , G . P . , C l a u s e n , T . , and P . K u r o p a t . I n t e r a c t i o n s of snowshoe h a r e and f e l t l e a f w i l l o w i n Alaska. Ecology 66:1564-1573.  to  of 1985.  B u e c h n e r , H . K . and H . C . P a w k i n s . 1961. V e g e t a t i o n change i n d u c e d by e l e p h a n t s and f i r e i n M u r c h i s o n F a l l s N a t i o n a l P a r k , Uganda. Ecology 42:752-766. B u e c h n e r , H . K . , B u s s , I . O . , L o n g h u r s t , W . M . And A . C . B r o o k s . 1963. Numbers and m i g r a t i o n o f e l e p h a n t s i n M u r c h i s o n F a l l s N a t i o n a l P a r k , Uganda. J. Wild. Manage. 27:3653. Buss,  I.O. 1961. b e h a v i o u r of 25: 1 3 1 - 1 4 8 .  Some o b s e r v a t i o n s on f o o d h a b i t s the A f r i c a n e l e p h a n t . J. Wild.  and Manage.  206  Buss,  I . O . and J . M . S a v i d g e . 1966. Change i n p o p u l a t i o n number and r e p r o d u c t i v e d a t e of e l e p h a n t s i n U g a n d a . J. Wild. Manage. 30:791-809.  Buxton, A . M . 242pp.  1927.  Kenya D a y s .  Edward A r n o l d and C o . , L o n d o n .  Buxton, P . A . 1955. The N a t u r a l H i s t o r y of H.K. Lewis, London. Byram,  Tsetse  G . M . 1959. C o m b u s t i o n of f o r e s t f u e l s . F i r e : C o n t r o l and Use ( D a v i s , K . P . , e d . ) , M c G r a w - H i l l , New Y o r k .  Flies.  In: Forest pp.61-89.  Caughley, G. 1976. The e l e p h a n t p r o b l e m - an a l t e r n a t i v e hypothesis. E. Afr. Wild. J. 14:265-283. C a u g h l e y , G . and J . G o d d a r d . 1974. Abundance and d i s t r i b u t i o n of e l e p h a n t s i n t h e Luangwa V a l l e y , Z a m b i a . E. Afr. Wild. J. 13:39-48. Charter, J . R . And R . W . J . K e a y . 1960. A s s e s s m e n t of t h e O l o k e m e j i f i r e - c o n t r o l e x p e r i m e n t 28 y e a r s a f t e r institution. Nig. For. Inf. Bull. , New S e r i e s 32pp. Coley,  P.D. 1980. E f f e c t s of l e a f age and p l a n t p a t t e r n s on h e r b i v o r y . Nature 284:545-546.  Coley,  P.D. 1983. tree species 53:209-233.  Corfield, T. Kenya.  life  3,  history  H e r b i v o r y and d e f e n s i v e c h a r a c t e r i s t i c s of in a lowland t r o p i c a l f o r e s t . Ecol. Monogr.  1973. Elephant m o r t a l i t y in Tsavo N a t i o n a l E. Afr. Wild. J. 11:339-368. problem.  I.  Croze,  H. 1974a. The S e r o n e r a b u l l E. Afr. Wild. J. 12:1-27.  Croze,  H. 1974b. The S e r o n e r a b u l l p r o b l e m . E. Afr. Wild. J. 12:29-47.  Croze,  H . , H i l l m a n , A . K . K . , and E . M . L a n g . 1981. E l e p h a n t s and t h e i r h a b i t a t s : how do t h e y t o l e r a t e e a c h o t h e r ? In: D y n a m i c s of L a r g e A n i m a l P o p u l a t i o n s ( F o w l e r , C . W . and T.D. Smith, e d s . ) , pp.297-316. John W i l e y & S o n s , New York.  II.  The  Park,  The  bulls. trees.  Cumming, D . H . M . 1981. The management of e l e p h a n t and o t h e r l a r g e mammals i n Zimbabwe. I n : P r o b l e m s i n Management o f L o c a l l y Abundant W i l d Mamma1s ( J e w e l l , P . A . and S. H o l t , e d s . ) , pp.91-118. A c a d e m i c P r e s s , New Y o r k .  207  Cumming, D . H . M . 1982. The i n f l u e n c e of l a r g e h e r b i v o r e s on savanna s t r u c t u r e i n A f r i c a . I n : E c o l o g y of T r o p i c a l Savannas. E c o l o g i c a l S t u d i e s 42, ( H u n t l e y , B . J . and B . H . Walker, e d s . ) , pp.217-245. Springer-Verlag, Berlin. Cumming, D . H . M . 1983. The d e c i s i o n - m a k i n g framework w i t h r e g a r d t o c u l l i n g of l a r g e mammals i n Zimbabwe. In: Management of L a r g e Mamma1s i n A f r i c a n C o n s e r v a t i o n A r e a s (Owen-Smith, R . N . , e d . ) , p p . 1 7 3 - 1 8 6 . Haum E d u c . P u b l . , Pretoria. Cumming, D . H . M . and P . J a c k s o n ( E d s . ) . 1984. The s t a t u s and c o n s e r v a t i o n of A f r i c a ' s e l e p h a n t s and r h i n o s . P r o c e e d i n g s of t h e j o i n t m e e t i n g of IUCN/SSC A f r i c a n e l e p h a n t and A f r i c a n r h i n o s p e c i a l i s t g r o u p s , Zimbabwe, 1981 . C u r r y - L i n d a h l , K. 1968. Z o o l o g i c a l a s p e c t s of c o n s e r v a t i o n of vegetation in t r o p i c a l A f r i c a . I n : C o n s e r v a t i o n of v e g e t a t i o n i n A f r i c a s o u t h of t h e S a h a r a , A c t a P h y t o g e o g r a p h i c a S u e c i c a 54, ( H e d b e r g , I . and 0 . , eds.). U p p s a l a , Sweden. 3 20pp. Dale,  I . R . and P . J . G r e e n w a y . 1961. Kenya T r e e s and S h r u b s . B u c h a n a n ' s Kenya E s t a t e s , L t d . , N a i r o b i .  Darling, F . F . 1960. An e c o l o g i c a l r e c o n n a i s s a n c e of t h e Mara p l a i n s i n Kenya c o l o n y . Wildl. Monogr. N o . 5. The Wildlife Society. Darling, F . F . 1964. C o n s e r v a t i o n and e c o l o g i c a l Ecol. 52(Suppl.):36-45. Daubenmire, Ecol. De  R. 1968. E c o l o g y of Res. 5:209-266.  fire  theory.  in grasslands.  J.  Adv.  Vos, V . , Bengis, R . G . , and, H . J . Coetzee. 1983. Population c o n t r o l of l a r g e mammals i n t h e K r u g e r N a t i o n a l P a r k . In: Management of L a r g e Mamma1s i n A f r i c a n C o n s e r v a t i o n A r e a s (Owen-Smith, R . N . , e d . ) , p p . 2 1 3 - 2 3 1 . Haum E d u c . P u b l . , Pr etoria.  Dixon,  W.J. (Ed.). 1985. BMDP S t a t i s t i c a l California Press. Berkeley.  Software.  D o u g a l l , H.W. 1963. A v e r a g e c h e m i c a l c o m p o s i t i o n of g r a s s e s , l e g u m e s , and b r o w s e . E. Afr. Wildl. 1:120-121 . D o u g a l l , H . W . and P . E . G l o v e r . 1964. c o m p o s i t i o n of Themeda t r i a n d r a Afr. Wildl. J. 2:67-70.  Univ.  of  Kenya J.  On t h e c h e m i c a l and Cynodon d a c t y l o n  .  E.  208  D o u g a l l , H . W . , D r y s d a l e , V . M . , and P . E . G l o v e r . 1964. c h e m i c a l c o m p o s i t i o n of Kenya browse and p a s t u r e E. Afr. Wild. J. 2:86-121.  The herbage.  Douglas-Hamilton, I. 1972. On t h e e c o l o g y and b e h a v i o u r of the African elephant. Ph.D. dissertation, U n i v . of O x f o r d . Douglas-Hamilton, I. 1979. The A f r i c a r e p o r t f o r IUCN/WWF/NYZS E l e p h a n t Programme.  elephant action p l a n . A S u r v e y and C o n s e r v a t i o n  Douglas-Hamilton, I. 1983. E l e p h a n t s h i t by arms r a c e : r e c e n t factors a f f e c t i n g elephant populations. A f r i c a n Elephant and R h i n o S p e c i a l i s t G r o u p N e w s l e t t e r 2 : 1 1 - 1 3 . D o u g l a s - H a m i l t o n , I . and A . K . K . H i l l m a n . 1981. Elephant c a r c a s s e s and s k e l e t o n s as i n d i c a t o r s of p o p u l a t i o n trends. I n : Low L e v e l S u r v e y T e c h n i q u e s . ILCA Monograph No.4, pp.113-129. Dublin, H.T. 1983. C o o p e r a t i o n and r e p r o d u c t i v e competition among f e m a l e A f r i c a n e l e p h a n t s . I n : Soc i a l B e h a v i o r of Female V e r t e b r a t e s . (Wasser, S . K . , e d . ) , pp.291-313. Academic P r e s s , New Y o r k . Dublin, H.T. 13.  1984.  The S e r e n g e t i - M a r a  ecosystem.  Swara  7:8-  D u b l i n , H . T . and I . D o u g l a s - H a m i l t o n . (in press). Status t r e n d s of e l e p h a n t s i n t h e S e r e n g e t i - M a r a e c o s y s t e m . J. Ecol. 24. Eastman, G . E . printed.  1927. Chronicles 87pp.  Edwards, D . C . 1942. 10:219-231.  of  Grass burning.  an A f r i c a n Empire J .  Trip. Exp.  and Afr.  Privately Agr.  E d w a r d s , D . C . and A . V . B o g d a n . 1951. Important grassland p l a n t s of Kenya . S i r Isaac Pitman & Sons, London. Eggeling, W.J. 1947. Observations Budongo r a i n f o r e s t , U g a n d a . Epp,  on t h e e c o l o g y of the J. Ecol. 34:20-87.  H . and J . A g a t s i v a . 1980. H a b i t a t t y p e s of t h e M a r a - N a r o k a r e a , western Kenya. T e c h n i c a l R e p o r t S e r i e s N o . 20, K.R.E.M.U., Nairobi.  Farler, J.P. 1882. N a t i v e r o u t e s i n E a s t A f r i c a from P a n g a n i t o t h e M a s a i c o u n t r y and t h e V i c t o r i a N y a n j a . Proc. Roy. Geog. Soc. 4:730-742.  209  Feeny,  P.P. 1970. S e a s o n a l change i n oak l e a f t a n n i n s and n u t r i e n t s as a c a u s e of s p r i n g f e e d i n g by w i n t e r - m o t h caterpillars. Ecology 51:656-681.  Feeny,  P.P. 1976. P l a n t a p p a r e n c y and c h e m i c a l d e f e n s e . In: B i o c h e m i c a l I n t e r a c t i o n s between P l a n t s and I n s e c t s ( W a l l a c e , J . and R . L . M a n d e l l , e d s . ) , p p . 1 - 4 0 . Recent A d v a n c e s i n P h y t o c h e m i s t r y , V o l . 10. P l e n u m , New Y o r k .  Field,  C.R. 1971. Elephant ecology P a r k , Uganda. E. Afr. Wild.  Field,  C . R . and R . M . L a w s . 1970. Animal p o p u l a t i o n s in Queen E l i z a b e t h N a t i o n a l P a r k , U g a n d a . J. Appl. 7:273-294.  Field,  C . R . and I . C . R o s s . 1976. The savanna e c o l o g y of V a l l e y Park. II. F e e d i n g e c o l o g y of e l e p h a n t and giraffe. E. Afr. Wild. J. 14:1-15.  Ford,  J. 1971. Ecology.  Ford,  J . and H . R . C l i f f o r d . 1968. Changes i n t h e d i s t r i b u t i o n of c a t t l e and of b o v i n e t r y p a n o s o m i a s i s a s s o c i a t e d w i t h t h e s p r e a d of t s e t s e f l i e s ( G l o s s i n a ) i n s o u t h - w e s t Uganda. J. Appl. Ecol. 5:301-337.  The R o l e of t h e Clarendon Press,  i n Queen E l i z a b e t h N a t i o n a l J. 9:99-123.  Trypanosomiases Oxford.  in  the Ecol. Kidepo  African  Fosbrooke, H . A . 1948. An a d m i n i s t r a t i v e s u r v e y of t h e s o c i a l system. T a n g a n y i k a N o t e s R e c o r d 26: 1-50.  Masai  Fosbrooke, H . A . 1968. E l e p h a n t s i n the S e r e n g e t i P a r k : an e a r l y r e c o r d . E. Afr. Wildl. J.  National 6:150-152.  Glover, J . Wild.  E.  Glover, P . E . 13.  1963. The e l e p h a n t J. 1:30-39. 1965.  Research  problem at  i n the  Tsavo.  Serengeti.  Afr.  Africana  2:10-  Glover, P . E . 1966. An e c o l o g i c a l s u r v e y of t h e N a r o k D i s t r i c t of Kenya M a a s a i l a n d , 1 9 6 1 - 1 9 6 5 . P a r t 1: g e o l o g y , s o i l s , c l i m a t e and h y d r o l o g y . Kenya N a t i o n a l P a r k s , (typescript). Glover, P . E . 1967. The i m p o r t a n c e of c o n t r o l of t s e t s e f l i e s . Bull. 37:581-614. Glover, P . E . 1968. The r o l e of the savannah h a b i t a t , w i t h research. E. Afr. Wild.  ecological studies Wrld. Hlth. Org.  f i r e and o t h e r i n f l u e n c e s suggestions for further J. 6:131-137.  in  on  the  210  Glover, P . E . 1972. N a t u r e , w i l d l i f e , and h a b i t a t w i t h a d i s c u s s i o n on f i r e and o t h e r i n f l u e n c e s . Proc. Tall Timbers F i r e E c o l . Conf. 11:319-336. G l o v e r , P . E . and M . D . Gwynne. 1961. The d e s t r u c t i o n Masailand. New S c i . 11:450-453.  of  G l o v e r , P . E . and D . S h e l d r i c k . 1964. An u r g e n t research p r o b l e m on t h e e l e p h a n t and r h i n o p o p u l a t i o n s of t h e N a t i o n a l Park i n K e n y a . Bull. Epizoot. Pis. Afr. 12:33-38 G l o v e r , P . E . and E . C . Trump. 1970. An e c o l o g i c a l Narok P i s t r i c t of Kenya M a s a i l a n d . Part II: vegetation. Kenya N a t i o n a l P a r k s , N a i r o b i . G r o g a n , E . S . and A . H . S h a r p . 1900. H u r s t and B l a c k e t t , L o n d o n .  From t h e  survey the  Cape t o  Tsavo  of  the  Cairo.  Guy,  P.R. 1976. The f e e d i n g b e h a v i o u r of e l e p h a n t ( L o x o d o n t a a f r i c a n a ) i n t h e Sengwa A r e a , R h o d e s i a . S. Afr. J. Wildl. Res. 6:55-63.  Guy,  P.R. 1981. Changes i n b i o m a s s and p r o d u c t i v i t y of w o o d l a n d s i n t h e Sengwa R e s e a r c h A r e a , Zimbabwe. J. Appl. Ecol. 18:507-519.  H a l l - M a r t i n , A. 1984. C o n s e r v a t i o n and management of e l e p h a n t s i n the Kruger N a t i o n a l P a r k , South A f r i c a . I n : The S t a t u s and C o n s e r v a t i o n of A f r i c a ' s E l e p h a n t s and R h i n o s (Cumming, P . H . M . and P . J a c k s o n , e d s . ) , p p . 1 0 4 - 1 1 5 . I . U . C . N . , G l and, S w i t z e r l a n d . Hanks,  J. 1969. Seasonal breeding Zambia. E. Afr. Wild. J.  of the A f r i c a n 7:167.  elephant  in  Hanks,  J . , Pensham, W . D . , Smuts, G . L . , J o o s t e , S . C . J . , l e Roux, P . , and P . l e S . M i l s t e i n . 1981. Managment of locally a b u n d a n t mammals - t h e S o u t h A f r i c a n e x p e r i e n c e . In: . P r o b l e m s i n Management of L o c a l l y Abundant W i l d Mammals ( J e w e l l , P . A . and S . H o l t , e d s . ) , p p . 2 1 - 5 5 . Academic P r e s s , New Y o r k .  H a n s e n , R . M . , Mugambi, M . M . , and S . M . B a u n i . 1985. t r o p h i c r a n k i n g of u n g u l a t e s of t h e n o r t h e r n J. Wild. Manage. 49:823-829. Harker, K.W. Agric.  1959. An A c a c i a weed of T r i n i d a d 36:45-51.  P i e t s and Serengeti.  Uganda g r a s s l a n d s .  Trop.  21 1  H a r r i n g t o n , G . N . and I . C . R o s s . 1974. The s a v a n n a h e c o l o g y o f Kidepo V a l l e y N a t i o n a l p a r k , Uganda. I. E f f e c t s of b u r n i n g and b r o w s i n g on t h e v e g e t a t i o n . E. Afr. Wildl. J. 12:93-105. Herlocker, D . J . 1976a. S t r u c t u r e , c o m p o s i t i o n , and e n v i r o n m e n t of some woodland v e g e t a t i o n t y p e s of t h e S e r e n g e t i National Park, Tanzania. P h . D . d i s s e r t a t i o n , Texas A & M Univ. Herlocker, D . J . 1976b. Woody v e g e t a t i o n of t h e National Park. Kleberg Studies in Natural Texas A & M U n i v . 31pp. Holling, C.S. systems.  1973. Ann.  Huntingford, G.W.B. Maasai, Nandi, 71:423-440.  R e s i l i e n c e and s t a b i l i t y of Rev. Ecol. Syst. 4:1-23.  Serengeti Resources, ecological  1976. Some names of t r e e s and p l a n t s and D o r o b o ( E a s t A f r i c a ) . Anthropos  Hurlbert, S.H. 1984. ecological field  in  P s e u d o r e p l i c a t i o n and t h e d e s i g n of experiments. Ecol. Monogr. 54:187-211.  Ishwaran, N. 1983. E l e p h a n t and w o o d y - p l a n t r e l a t i o n s h i p s Gal Oya, S r i Lanka. Biol. Cons. 26:255-270.  in  J a c h m a n n , H . and R . H . V . B e l l . 1985. U t i l i z a t i o n by e l e p h a n t s of B r a c h y s t e g i a w o o d l a n d s o f t h e Kasungu N a t i o n a l P a r k , Malawi. Afr. J. Ecol. 23:245-248. J a c o b s , A . H . 1965. The t r a d i t i o n a l p o l i t i c a l o r g a n i z a t i o n of the p a s t o r a l M a a s a i . P h . D . d i s s e r t a t i o n , U n i v . of O x f o r d . Kennan, T . C . D . 1971. The e f f e c t s of f i r e on two v e g e t a t i o n t y p e s of M a t o p o s . Proc. T a l l Timbers F i r e E c o l . Conf. 11:53-98. Kenya Game D e p a r t m e n t A n n u a l Department, N a i r o b i .  Report.  1953.  Kenya Game  Kenya Game D e p a r t m e n t A n n u a l Department, N a i r o b i .  Report.  1954.  Kenya Game  Kenya Game D e p a r t m e n t A n n u a l Department, N a i r o b i .  Report.  1962.  Kenya Game  Kenya Game D e p a r t m e n t A n n u a l Department, N a i r o b i .  Report.  1963.  Kenya Game  212  Kenya Game D e p a r t m e n t A n n u a l R e p o r t . Department, N a i r o b i .  1964.  Kenya Game  Kenya Game D e p a r t m e n t A n n u a l R e p o r t . Department, N a i r o b i .  1965.  Kenya Game  Kershaw, K . A . 1964. Quantitative A r n o l d , London. Kurji,  and Dynamic  Ecology.  Edward  F. 1976. Human e c o l o g y . Serengeti Research I n s t i t u t e A n n u a l R e p o r t , 1974-1975, p p . 1 2 - 3 1 . Tanzania National Parks, Arusha.  L a m p r e y , H . F . , G l o v e r , P . E . , T u r n e r , M . I . M . , and R . H . V . B e l l . 1967. The i n v a s i o n of t h e S e r e n g e t i N a t i o n a l P a r k by elephants. E. Afr. Wild. J. 5:151-166. L a m p r e y , H . F . , H a l e v y , G . , and S. M a k a c h a . 1974. Interaction between Acac i a , b r u c h i d s e e d b e e t l e s and l a r g e h e r b i v o r e s . E. Afr. Wild. J. 12:81-85. Lamprey, R . H . 1985. M a s a i i m p a c t on Kenya s a v a n n a v e g e t a t i o n : A remote s e n s i n g a p p r o a c h . P h . D . d i s s e r t a t i o n , U n i v . of B i rmingham. Langridge, W.P. outdated.  1960. In t s e t s e c o n t r o l , W i l d l i f e 2:13-17.  game s l a u g h t e r  is  L a n g r i d g e , W . P . , S m i t h , J . A . , and L . E . D . W a t e r i d g e . 1970. Some o b s e r v a t i o n s on t h e e c o l o g y of G l o s s i n a s y w n n e r t o n i A u s t e n i n t h e Mara r e g i o n of K e n y a . Proc. 12th M e e t i n g i f the I n t e r n a t i o n a l S c i e n t i f i c C o u n c i l f o r T r y p a n o s o m i a s i s Resea rch, Bangui, Central A f r i c a n Republic. OAU/Sci. & Techn. Res. Comm. Publ. 102:203-211. Laws,  R.M. Fert.  1969. The T s a v o r e s e a r c h p r o j e c t . Suppl. 6:495-531.  J.  Laws,  R.M. 1970. Elephants change i n E a s t A f r i c a .  Laws,  R.M. 1981. L a r g e mammal f e e d i n g s t r a t e g i e s and r e l a t e d overabundance problems. I n : Problems i n Managing L o c a l l y Abundant W i l d Mamma1s ( J e w e l l , P . A . and S. H o l t , e d s . ) , pp.217-232. Academic P r e s s , New Y o r k .  Laws,  R . M . and I . S . C . P a r k e r . 1968. R e c e n t s t u d i e s on e l e p h a n t populations in East A f r i c a . Symp. Zool. Soc. Lond. 21 :319"359.  as a g e n t s of h a b i t a t Oikos 21:1-15.  Reprod. and  landscape  213  Laws,  R . M . , P a r k e r , I . S . C . , and R . C . B . J o h n s t o n e . 1975. E l e p h a n t s and t h e i r H a b i t a t s . Clarendon P r e s s , London.  L a w t o n , R . M . and M . G o u g h . 1970. t o blame? Oryx 1 0 : 2 4 4 - 2 4 8 .  E l e p h a n t s or  fire  -  which  L e u t h o l d , W. 1977a. S p a t i a l o r g a n i z a t i o n of h a b i t a t u t i l i z a t i o n of e l e p h a n t s i n Tsavo N a t i o n a l P a r k , Z. Sauqertierkunde 42:358-379. L e u t h o l d , W. 1977b. Changes i n t r e e National Park,Kenya. Afr. J.  is  Kenya.  p o p u l a t i o n s of Tsavo E a s t Ecol. 15:61-69.  L e u t h o l d , W. and J . B . S a l e . 1973. Movements and p a t t e r n s of h a b i t a t u t i l i z a t i o n in Tsavo N a t i o n a l Park, Kenya. E. Afr. Wild. J. 11:369-384. Lewis,  E.A. 1935. Tsetse f l i e s colony. Bull. Ent. Res.  i n the Maasai 25:439-455.  Reserve,  Kenya  L i n d s a y , W . K . 1982. H a b i t a t s e l e c t i o n and s o c i a l g r o u p d y n a m i c s of A f r i c a n e l e p h a n t s i n A m b o s e l i , K e n y a . M.Sc. t h e s i s , Univ. B r i t i s h Columbia, Vancouver. M a c a u l e y , B . J . and L . R . F o x . 1980. V a r i a t i o n in t o t a l phenols and c o n d e n s e d t a n n i n s i n E u c a l y p t u s : l e a f p h e n o l o g y and insect grazing. Austr. J. Ecol. 5:31-35. Mack,  R.N. 1970. The A f r i c a n c a t t l e p l a g u e e p i d e m i c 1890's. Trop. Anim. Hlth. Prod. 2:210-219.  in  the  Maddock, L . 1979. The " m i g r a t i o n " and g r a z i n g s u c c e s s i o n . In: S e r e n g e t i : d y n a m i c s of an e c o s y s t e m ( S i n c l a i r , A . R . E . and M. N o r t o n - G r i f f i t h s , e d s . ) , p p . 1 0 4 - 1 2 9 . U n i v . of C h i c a g o Press, Chicago. M a l l e t t , M. 1923. A W h i t e Woman among the M a s a i . & C o . , New Y o r k . 288pp. Malpas, R . C . Uganda. Mbi,  E . P . Dutton  1978. The e c o l o g y of t h e A f r i c a n e l e p h a n t P h . D . d i s s e r t a t i o n , U n i v . of C a m b r i d g e .  in  C.N. 1978. C o m p a r a t i v e b i o c h e m i s t r y of t h e f l o r a of two A f r i c a n r a i n f o r e s t s w i t h s p e c i a l r e f e r e n c e t o t a n n i n s and napthoquinones. P h . D . d i s s e r t a t i o n , U n i v . of S t r a t h c l y d e .  M c A r t h u r , A . G . and N . P . C h e n e y . 1966. The c h a r a c t e r i s a t i o n f i r e s in r e l a t i o n to e c o l o g i c a l s t u d i e s . Aust. For. Res. 2:36-45.  of  214  McKey,  D. 1974. Amer. Nat.  Adaptive patterns 108:305-320.  in a l k a l o i d  physiology.  McKey,  D. 1979. The d i s t r i b u t i o n of s e c o n d a r y compounds w i t h i n plants. In: H e r b i v o r e s : t h e i r i n t e r a c t i o n with secondary plant metabolites ( R o s e n t h a l , G . A . and D . H . J a n z e n , e d s . ) , pp.55-133. A c a d e m i c P r e s s , New Y o r k .  McNaughton, S . J . 1976. Serengeti migratory wildebeest: f a c i l i t a t i o n of e n e r g y f l o w by g r a z i n g . S c i e n c e 191:9294. McNaughton, S . J . 1979 G r a s s l a n d - h e r b i v o r e d y n a m i c s . In: S e r e n g e t i : d y n a m i c s of an e c o s y s t e m ( S i n c l a i r , A . R . E . and M. N o r t o n - G r i f f i t h s , e d s . ) , p p . 4 6 - 8 1 . U n i v . of C h i c a g o Press, Chicago. McNaughton, form,  S.J. 1984. Grazing lawns: animals in h e r d s , and c o e v o l u t i o n . Amer. Nat. 124:863-886.  McNaughton, S . J . Serengeti.  1985. Ecol.  E c o l o g y of a g r a z i n g e c o s y s t e m : Monogr. 55:259-294.  plant the  M c N a u g h t o n , S . J . , T a r r a n t s , J . L . , M c N a u g h t o n , M . M . , and R.H. Davis. 1985. S i l i c a as a d e f e n s e a g a i n s t h e r b i v o r y a n d a growth p r o m o t e r i n A f r i c a n g r a s s e s . Ecology 66:528535. Melland, F. 1938. Elephants London. 186pp.  in A f r i c a  .  Country L i f e L t d ,  M i l t o n , K. 1979. F a c t o r s i n f l u e n c i n g l e a f c h o i c e by h o w l e r monkeys: a t e s t o f some h y p o t h e s e s of f o o d s e l e c t i o n by generalist herbivores. Amer. Nat. 114:362-378. Mol,  F.  1980.  Maasai Mara.  Prudential Printers,  Nairobi.  Molyneux, D . H . 1982. T r y p a n a s o m e s , t r y p a n o s o m i a s i s , and t s e t s e c o n t r o l : i m p a c t on w i l d l i f e and i t s c o n s e r v a t i o n . In: Animal Disease in R e l a t i o n to Animal C o n s e r v a t i o n ( E d w a r d s , M . A . and U . M c D o n n e l l , e d s . T ! Symp. Zool. Lond. 50: 2 9 - 5 6 . M o r g a n , W . T . W . and N . M . S h a f f e r . d e n s i t y and d i s t r i b u t i o n . 1981.  1966. P o p u l a t i o n of Kenya: Oxford Univ. Press, N a i r o b i .  Moss,  C.J.  Social circles.  W i l d l i f e News  Moss,  C.J. 1983. O e s t r u s b e h a v i o u r and f e m a l e African elephant. Behaviour 86:167-196.  16:2-7.  choice  in  the  215  Moss,  C . J . and J . H . P o o l e . 1983. R e l a t i o n s h i p s and s o c i a l s t r u c t u r e of A f r i c a n e l e p h a n t s . In: Primate S o c i a l R e l a t i o n s (Hinde, R . A . , e d . ) . Blackwell S c i . Publ., Oxford.  Myers,  N. 1973. T s a v o N a t i o n a l P a r k , K e n y a , and i t s an i n t e r i m a p p r a i s a l . Biol. Cons. 5:123-132.  Napier  B a x , P . and D . L . W . S h e l d r i c k . 1963. Some p r e l i m i n a r y o b s e r v a t i o n s on t h e f o o d o f t h e e l e p h a n t i n t h e R o y a l T s a v o N a t i o n a l P a r k ( E a s t ) of K e n y a . E. A f r . Wild. J. 1:40-53.  elephants:  Narok D i s t r i c t C o m m i s s i o n e r ' s R e p o r t s . 1914 A r c h i v e s Documents, D C / N R K . 1 / 1 / 1 - 7 .  -1961.  Kenya  Nicholson, B.D. 1954. The A f r i c a n e l e p h a n t africana ). Afr. Wildl. 8:313-322.  ( Loxodonta  N o r t o n - G r i f f i t h s , M. 1979. The i n f l u e n c e of g r a z i n g , b r o w s i n g , and f i r e on t h e v e g e t a t i o n d y n a m i c s of t h e S e r e n g e t i . In: S e r e n g e t i : d y n a m i c s of an e c o s y s t e m ( S i n c l a i r , A . R . E . and M. N o r t o n - G r i f f i t h s , e d s . ) , p p . 3 1 0 - 3 5 2 . U n i v . of C h i c a g o Press, Chicago. N o r t o n - G r i f f i t h s , M . , H e r l o c k e r , D . , and L . P e n n y c u i c k . 1975. The p a t t e r n s of r a i n f a l l i n t h e S e r e n g e t i E c o s y s t e m , Tanzania. E. A f r . Wild. J. 13:347-374. Oates,  J . F . , Waterman, P . G . , and G . M . C h o o . 1980. Food s e l e c t i o n by t h e s o u t h I n d i a n l e a f - m o n k e y , Presbytis j o h n i i , in r e l a t i o n to l e a f c h e m i s t r y . Oecologia (Berlin) 45:45-56.  Okula,  J . P . and W . R . S i s e . 1986. E f f e c t s of elephant browsing on A c a c i a s e y a l i n Waza N a t i o n a l P a r k , C a m e r o o n . Afr. J. Ecol. 24:1-6.  Olivier, R.C.D. 1978. On t h e e c o l o g y of t h e A s i a n P h . D . d i s s e r t a t i o n , U n i v . of C a m b r i d g e .  elephant.  Owen-Smith, R . N . 1983. D i s p e r s a l and t h e d y n a m i c s of l a r g e h e r b i v o r e s i n e n c l o s e d a r e a s : I m p l i c a t i o n s f o r management. I n : Management of L a r g e Mammals i n A f r i c a n C o n s e r v a t i o n Areas (Owen-Smith, R . N . , e d . ) , pp.127-143. Haum Educ. P u b l . , P r e t o r i a . Parker, I . S . C . 1983. The T s a v o s t o r y : An e c o l o g i c a l c a s e history. I n : Management o f L a r g e Mamma1s i n A f r i c a n C o n s e r v a t i o n Areas (Owen-Smith, R . N . , e d . ) , pp.37-49. Haum E d u c . P u b l . , P r e t o r i a .  216  Pellew, R . A . P . 1981. The g i r a f f e ( G i r a f f a c a m e l o p a r d a l i s t i p p e l s k i r c h i M a t s c h i e ) and i t s A c a c i a f o o d r e s o u r c e i n the S e r e n g e t i N a t i o n a l Park. P h . D . d i s s e r t a t i o n , U n i v . of London. Pellew, R.A.P. 1983. The i m p a c t s of e l e p h a n t , g i r a f f e , and f i r e upon t h e A c a c i a t o r t i l i s w o o d l a n d s of t h e S e r e n g e t i . Afr. J. Ecol. 21:41-74. P e l l e w , R . A . and B . J . S o u t h g a t e . 1984. The p a r a s i t i s m A c a c i a t o r t i l i s s e e d s i n the S e r e n g e t i . Afr. J. 22:73-75.  of Ecol.  Pennycuick, L . 1975. Movements of t h e m i g r a t o r y w i l d e b e e s t p o p u l a t i o n i n t h e S e r e n g e t i a r e a between 1960 and 1973. E. A f r . Wild. J. 13:65-87. Phillips, J.F.V. 1930. F i r e : i t s i n f l u e n c e on b i o t i c c o m m u n i t i e s and p h y s i c a l f a c t o r s i n S o u t h and E a s t S. Afr. J. Sci. 27:352-367.  Africa.  Phillips, J.F.V. 1965. F i r e as m a s t e r and s e r v a n t : its i n f l u e n c e i n the b i o c l i m a t i c r e g i o n s of t r a n s - s a h a r a n Africa. Proc. T a l l Timbers F i r e E c o l . Conf. 4:7-109. Phillips, J.F.V. 1974. E f f e c t s of f i r e e c o s y s t e m s of s u b - S a h a r a n A f r i c a . ( K o z l o w s k i , T . T . and C . E . A h l g r e n , A c a d e m i c P r e s s , New Y o r k . P i e l o u , E . C . 1966. The measurement o f t y p e s of b i o l o g i c a l c o l l e c t i o n s . 13:131-144. P i e n a a r , A . J . 1959. burning alone. Pienaar, U. De V . Afr. Wildl.  i n f o r e s t and s a v a n n a I n : F i r e and E c o s y s t e m s eds.), pp.435-481. d i v e r s i t y in different J. Theor. Biol.  Bush e n c r o a c h m e n t n o t c o n t r o l l e d by Farm. S. Afr. 35:16-17. 1969. Why e l e p h a n t 23:181-195.  culling  is  veld  necessary.  Pienaar, U. De V . 1983. Management by i n t e r v e n t i o n : t h e pragmatic/economic o p t i o n . I n : Management of L a r g e Mammals i n A f r i c a n C o n s e r v a t i o n A r e a s ( O w e n - S m i t h , R . N . , e d . ) , pp.23-36. Haum E d u c . P u b l . , P r e t o r i a . Pienaar, U. De V . , van Wyk, P . , and N . F a i r a l l . 1966. An a e r i a l c e n s u s o f e l e p h a n t and b u f f a l o i n K r u g e r N a t i o n a l P a r k and i m p l i c a t i o n s t h e r e o f on i n t e n d e d management schemes. Koedoe 9 : 4 0 - 1 0 7 .  217  P l o w r i g h t , W. 1963. The r o l e o f game a n i m a l s i n t h e e p i z o o t i o l o g y of r i n d e r p e s t and m a l i g n a n t c a t a r r h a l f e v e r in East A f r i c a . Bull. Epizoot. Pis. Afr. 11:149-162. P l o w r i g h t , W. 1982. The e f f e c t s o f r i n d e r p e s t and r i n d e r p e s t c o n t r o l on w i l d l i f e i n A f r i c a . In: Animal Disease in R e l a t i o n t o A n i m a l C o n s e r v a t i o n ( E d w a r d s , M . A . and U. McPonnell, e d s . ) . Symp. Zool. Soc. Lond. 50:1-28. P l o w r i g h t , W. and B . M c C u l l o c h . 1967. I n v e s t i g a t i o n s on t h e i n c i d e n c e of r i n d e r p e s t v i r u s i n f e c t i o n i n game a n i m a l s of N o r t h e r n T a n g a n y i k a and S o u t h e r n Kenya 1 9 6 0 / 6 3 . J. Hyg. 65:343-358. Poole,  J.H. 1982. Musth and m a l e - m a l e c o m p e t i t i o n i n t h e African elephant. P h . P . d i s s e r a t i o n , U n i v . of C a m b r i d g e .  Pratt,  P . J . and J . K n i g h t . 1971. Bush c o n t r o l s t u d i e s i n t h e d r i e r areas of Kenya. V. E f f e c t s of c o n t r o l l e d b u r n i n g and g r a z i n g management on T a r c o n a n t h u s / A c a c i a t h i c k e t . J. Appl. Ecol. 8:217-237.  Pyke,  G.H. Ann.  1984. Rev.  Optimal foraging theory: a c r i t i c a l Ecol. Syst. 15:523-575.  review.  Rhoades, P . F . 1979. E v o l u t i o n of p l a n t c h e m i c a l d e f e n s e against herbivores. In: Herbivores: their interaction w i t h s e c o n d a r y p l a n t m e t a b o l i t e s ( R o s e n t h a l , G . A . and P.H. Janzen, e d s . ) , pp.3-54. Academic P r e s s , New Y o r k . Rhoades, P . F . 1983. H e r b i v o r e p o p u l a t i o n d y n a m i c s and p l a n t chemistry. I n : V a r i a b l e P l a n t s and H e r b i v o r e s i n N a t u r a l and Managed Systems ( P e n n o , R . F . and M . S . M c C l u r e , e d s . ) , pp.155-220. A c a d e m i c P r e s s , New Y o r k . R h o a d e s , P . F . and R . G . C a t e s . 1976. Toward a g e n e r a l t h e o r y of plant antiherbivore chemistry. In: Biochemical I n t e r a c t i o n s between P l a n t and I n s e c t s ( W a l l a c e , J . and R . L . Mandell, e d s . ) , pp.168-213. Recent Advances i n Phytochemistry, V o l . 10. P l e n u m , New Y o r k . R o g e r s , P . J . and S . E . R a n d o l p h . 1985. Population ecology tsetse. Ann. Rev. Entom. 30:197-216. Ross,  of  I . C . , F i e l d , C . R . , and G . N . H a r r i n g t o n . 1976. The s a v a n n a e c o l o g y of K i d e p o V a l l e y N a t i o n a l P a r k . III. A n i m a l p o p u l a t i o n s and management r e c o m m e n d a t i o n s . E. Afr. Wild. J. 14:35-48.  R o s s i t e r , P . B . , J e s s e t t , P . M . , Wafwa, J . S . , K a r s t a d , L . , Chema, S . , T a y l o r , W . P . , Rowe, L . W . , N y a n g e , J . C . , O t a r u , M . , Mumbala, M . , and G . R . S c o t t . 1983. The r e - e m e r g e n c e of r i n d e r p e s t as a t h r e a t i n E a s t A f r i c a , 1979-1982. Vet. Rec. 113:459-461.  218  Rothermel, R . C . 1972. A m a t h e m a t i c a l model spread in w i l d l a n d f i r e s . USDA F o r e s t T e c h n i c a l Report INT-115.  for p r e d i c t i n g fire Service. General  R o t h e r m e l , R . C . and J . E . D e e m i n g . 1980. M e a s u r i n g and i n t e r p r e t i n g f i r e behaviour for c o r r e l a t i o n with f i r e effects. USDA F o r e s t S e r v i c e . General T e c h n i c a l Report INT-93. Royal National  P a r k s of  Kenya.  Annual Reports  R u s s e l l , E.W. 1968. The e l e p h a n t Oryx 9 : 4 0 4 - 4 0 6 . Ryan,  problem  A . R . , J o i n e r , B . L . , and B . F . R y a n . Handbook. Duxbury P r e s s , Boston.  1946  i n the 1985.  -  1955.  Serengeti. MINITAB  Student  Sabiiti, E.N. 1986. B u r n i n g e f f e c t s on Acac i a regeneration. Ph.D. dissertation, U n i v . of New B r u n s w i c k . S a b i i t i , E . N . and R . W . W e i n . (in p r e s s ) . F i r e and A c a c i a - a h y p o t h e s i s of c o l o n i z a t i o n s u c c e s s . J. Ecol. .  seeds  Sandford, G.R. 1919. An A d m i n i s t r a t i v e and P o l i t i c a l H i s t o r y of t h e M a s a i R e s e r v e . W a t e r l o w and S o n s , L o n d o n . S a v i d g e , J . M . 1968. E l e p h a n t s i n t h e Ruaha N a t i o n a l P a r k , T a n z a n i a - Management p r o b l e m . E. Afr. Agric. For. 33:191-196. Schoener, T.W. 1971. T h e o r y of Ecol. Syst. 2:369-404. Seif  feeding  strategies.  Ann.  J. Rev.  E l - D i n , A . and M . O b e i d . 1971. E c o l o g i c a l s t u d i e s of t h e v e g e t a t i o n of t h e S u d a n . II. The g e r m i n a t i o n of s e e d s and e s t a b l i s h m e n t of s e e d l i n g s of A c a c i a Senegal (L) Wildl. Under c o n t r o l l e d c o n d i t i o n s i n the S u d a n . J. Appl. Ecol. 8:191-201.  Sikes,  S.K. 1971. The N a t u r a l H i s t o r y of W e i d e n f e l d and N i c h o l s o n , L o n d o n .  Simon,  N. 1962. w i l d l i f e of  Sinclair, A.R.E. wildebeest 107.  the  African  Elephant.  Between the s u n l i g h t and t h e t h u n d e r : Kenya. C o l l i n s , London. 384pp. 1973. i n the  Sinclair, A.R.E. 1975. l e v e l s in t r o p i c a l 44:497-520.  Population Serengeti.  i n c r e a s e s of b u f f a l o E. Afr. Wild. J.  The r e s o u r c e l i m i t a t i o n of grassland ecosystems. J.  the and 11:93-  trophic Anim. Ecol.  219  Sinclair, A.R.E. 1977. Press, Chicago.  The A f r i c a n  Buffalo.  Univ.  of  Chicago  Sinclair, A.R.E. 1979. D y n a m i c s of t h e S e r e n g e t i e c o s y s t e m . I n : S e r e n g e t i : d y n a m i c s of an E c o s y s t e m ( S i n c l a i r , A . R . E . and M . N o r t o n - G r i f f i t h s , e d s . ) , p p . 1 - 3 0 . U n i v . of Chicago Press, Chicago. S i n c l a i r , A . R . E . and M . N o r t o n - G r i f f i t h s ( E d s . ) . S e r e n g e t i : d y n a m i c s of an E c o s y s t e m . Univ. Press, Chicago.  1979. of C h i c a g o  S i n c l a i r , A . R . E . and H . T . D u b l i n . 1985a. The c o l l a p s e northern Serengeti buffalo population. Report to T a n z a n i a N a t i o n a l P a r k s , 12pp.  of t h e the  S i n c l a i r , A . R . E . and H . T . D u b l i n . 1985b. D r y s e a s o n c e n s u s of the n o r t h e r n S e r e n g e t i b u f f a l o p o p u l a t i o n . Report to the T a n z a n i a N a t i o n a l P a r k s , 15pp. S i n c l a i r , A . R . E . , D u b l i n , H . T . , and M . B o r n e r . 1985. P o p u l a t i o n r e g u l a t i o n of S e r e n g e t i w i l d e b e e s t : a t e s t the food h y p o t h e s i s . Oecologia ( B e r l i n ) 65:266-268. Smart,  of  N . O . E . , H a t t o n , J . C , and D . H . N . S p e n c e . 1985. The e f f e c t s of l o n g - t e r m e x c l u s i o n of l a r g e h e r b i v o r e s on v e g e t a t i o n in Murchison F a l l s N a t i o n a l P a r k , Uganda. Biol. Cons. 33:229-245.  S n e e u w j a g t , R . J . and W . H . F r a n d s e n . 1977. B e h a v i o r of e x p e r i m e n t a l g r a s s f i r e s v s . p r e d i c t i o n s b a s e d on Rothermel's f i r e model. Can. J . For. Res. 7:357-367. S p e n c e , D . H . N , and A . A n g u s . 1971. African grassland management - b u r n i n g and g r a z i n g i n M u r c h i s o n F a l l s N a t i o n a l Park, Uganda. I n : The S c i e n t i f i c Management of A n i m a l and P l a n t C o m m u n i t i e s f o r C o n s e r v a t i o n ( D u f f e y , E . and A . S . W a t t , e d s . ) , p p . 3 1 9 - 3 3 1 . Blackwell Sci. Press, Oxford. Spinage, C . A . 1973. A r e v i e w of i v o r y e x p l o i t a t i o n and elephant population trends in A f r i c a . E. A f r . Wild. 11:28l-289.  J.  S p i n a g e , C . A . and F . E . G u i n n e s s . 1972. E f f e c t s of f i r e i n t h e A k a g e r a N a t i o n a l P a r k , Rwanda. Revue Z o o l . Bot. Afr. 86:302-336. S t e l f o x , J . G . , P e d e n , D . G . , E p p . H . , H u d s o n , R . J . , Mbugua, S . W . , A g a t s i v a , J . L . , and C . L . Amuyunzu. 1986. Herbivore dynamics i n southern Narok, Kenya. J. Wild. Manage. 50:339-347.  220  S t e w a r t , D . R . M . and L . M . T a l b o t . 1962. Census of w i l d l i f e t h e S e r e n g e t i , Mara and L o i t a p l a i n s . E. A. Agric. For. J. 28:58-60. Stewart, D.R.M. 1963-1966. Story,  1968. Bull.  Rinderpest Epizoot.  on  among w i l d a n i m a l s i n K e n y a , Pis. Afr. 16:139-140.  R. 1952. A b o t a n i c a l s u r v e y of the Keiskammahoek Pistrict. Memoirs of t h e B o t a n i c a l S u r v e y o f S o u t h 27:112-120.  Africa  S t r a n g , R . M . 1973. Bush e n c r o a c h m e n t and v e l d management i n s o u t h - c e n t r a l A f r i c a : t h e need f o r a r e a p p r a i s a l . Biol. Cons. 5:96-104. S t r a n g , R . M . 1974. t r e n d s on t h e 11:249-263.  Some man-made c h a n g e s i n s u c c e s s i o n a l Rhodesian h i g h v e l d . J. Appl. Ecol.  Sukumar, R . 1985. E c o l o g y of t h e A s i a n e l e p h a n t ( E l e p h a s maximus ) and i t s i n t e r a c t i o n s w i t h man i n S o u t h I n d i a . Phd. t h e s i s . Volumes 1 and 2. I n d i a n I n s t , of S c i e n c e , B a n g a l o r e , I n d i a , 542pp. Swynnerton, Ent. Taiti,  C.F.M. 1936. The t s e t s e Soc. Lond. 84:1-579.  flies  S.W. 1973. A vegetation survey N a t i o n a l R e s e r v e , Narok P i s t r i c t , U n i v . of N a i r o b i .  T a l b o t , L . M . and M . H . T a l b o t . 1963. Masailand. Wild. Monogr. No.  of  East  Africa.  of t h e M a a s a i Mara Kenya. M.Sc. thesis,  The w i l d e b e e s t i n 12. The W i l d l i f e  T a l b o t , L . M . and P . R . M . S t e w a r t . 1964. First wildlife of the e n t i r e S e r e n g e t i - M a r a R e g i o n , E a s t A f r i c a . Wildl. Manage. 28:815-827. T a n z a n i a N a t i o n a l Parks Report Arusha, Tanzania.  1962  -  R.  1964.  western Society. census J.  A n n u a l R e p o r t TNP,  T a y l o r , W . P . and R . M . W a t s o n . 1967. S t u d i e s on t h e e p i z o o t i o l o g y of r i n d e r p e s t i n b l u e w i l d e b e e s t and o t h e r game s p e c i e s of N o r t h e r n T a n z a n i a and S o u t h e r n K e n y a , 1965-1967. J. Hy_£. (Camb. ) 6 5 : 5 3 7 - 5 4 5 . Thomas, P . B . and P . J . P r a t t . 1967. B u s h - c o n t r o l s t u d i e s in the d r i e r a r e a s of K e n y a . I V . E f f e c t s of c o n t r o l l e d b u r n i n g on s e c o n d a r y t h i c k e t i n u p l a n d Acac i a w o o d l a n d . J. Appl. Ecol. 4:325-335.  221  Thomson, J . 583pp.  1885.  Through Masai  Land.  Sampson Low,  London.  Thomson, P . J . 1975. The r o l e of e l e p h a n t s , f i r e , and o t h e r a g e n t s i n t h e d e c l i n e of a B r a c h y s t e g i a b o e h m i i w o o d l a n d . J. Sth. Afr. Wildl. Manage. 5:11-18. T i e s z e n , L . L . , H e i n , D . , Q v o r t r u p , S . A . , T r o u g h t o n , J . H . , and S . K . Imbaamba. 1979. Use of d e l t a - C 1 3 v a l u e s t o determine v e g e t a t i o n s e l e c t i v i t y in East A f r i c a n herbivores. Oecoloqia ( B e r l i n ) 37:351-359. T r o l l o p e , W.S.W. 1972. F i r e as a method o f v e g e t a t i o n i n t h e A m a t o l e M o u n t a i n s of e x p e r i m e n t a l and f i e l d s c a l e r e s u l t s . Fire Ecol. Conf. 11:99-120. T r o l l o p e , W.S.W. 1982a. A f r i c a n savannas. Ecological Studies e d s . ) , pp.292-306.  e r a d i c a t i n g Macchia South A f r i c a Proc. T a l l Timbers  E c o l o g i c a l e f f e c t s of f i r e i n S o u t h I n : E c o l o g y of T r o p i c a l S a v a n n a s . 42, ( H u n t l e y , B . J . and B . H . W a l k e r , Springer-Verlag, Berlin.  T r o l l o p e , W.S.W. 1982b. V e l d r e h a b i l i t a t i o n and management s t u d i e s i n t h e C i s k e i : c o n t r o l of b u s h e n c r o a c h m e n t w i t h f i r e and g o a t s . F i n a l R e p o r t on t h e M a s t e r R e s e a r c h P r o g r a m , U n i v e r s i t y of F o r t H a r e . T r o l l o p e , W.S.W. 1984. F i r e in savanna. In: Ecological E f f e c t s of F i r e i n S o u t h A f r i c a n E c o s y s t e m s (de V . B o o y s e n , P . and N . M . T a i n t o n , e d s . ) , pp.151 - 1 7 5 . Springer-Verlag, Berlin. Trump,  E . C . 1972. Vegetation District. Working paper  and l a n d use s u r v e y of N a r o k 10, UNDP/FAO, N a i r o b i .  Van Wyk, P . and N . F a i r a l l . 1969. e l e p h a n t on t h e v e g e t a t i o n of Koedoe 1 2 : 5 7 - 8 9 .  The i n f l u e n c e of t h e A f r i c a n Kruger N a t i o n a l Park.  V e s e y - F i t z G e r a l d , D. 1972. F i r e and a n i m a l vegetation in Tanzania National Parks. Timbers F i r e E c o l . Conf. 11:297-317.  impact Proc.  in Tall  V e s e y - F i t z G e r a l d , D. 1973. The d y n a m i c s a s p e c t s of t h e secondary v e g e t a t i o n in Arusha N a t i o n a l Park, T a n z a n i a . E. A. Aqric. For. J. 38:314-327. Wakefield, T. 1870. to the i n t e r i o r 40:303-338.  R o u t e s of n a t i v e of E a s t A f r i c a .  caravans J. Roy.  from t h e c o a s t Geog. Soc.  222  Wakefield, Roy.  T. 1882. New r o u t e s t h r o u g h M a s a i c o u n t r y . Geog. Soc. 4:742-747.  Proc.  W a t s o n , R . M . and R . H . V . B e l l . 1969. The d i s t r i b u t i o n , a b u n d a n c e , and s t a t u s of e l e p h a n t i n t h e S e r e n g e t i r e g i o n of n o r t h e r n T a n z a n i a . J. Appl. Ecol. 6:115-132. W a t s o n , R . M . , Graham, A . D . , and I . S . C . P a r k e r . 1969. A census of t h e l a r g e mammals of L o l i o n d o C o n t r o l l e d A r e a , n o r t h e r n Tanzania. E. Afr. Wildl. J. 7:43-59. Weinmann, H . 1959. The c h e m i s t r y and p h y s i o l o g y of g r a s s e s . I n : The G r a s s and P a s t u r e s of S o u t h A f r i c a ( D . M e r e d i t h , ed.). C e n t r a l News A g e n c y , S o u t h A f r i c a . West,  O. 1958. Bush e n c r o a c h m e n t , v e l d b u r n i n g , and management. Rhod. Agric. J. 55:407-425.  grazing  West,  0. 1965. F i r e i n v e g e t a t i o n and i t s use i n p a s t u r e management w i t h s p e c i a l r e f e r e n c e t o t r o p i c a l and sub tropical Africa. Mimeograph N o . 1/1965. Commonwealth A g r i c u l t u r a l Bureau.  W e s t e r n , D . and C . van P r a e t . 1973. h a b i t a t and c l i m a t e of an E a s t (Lond.) 241:104-106.  C y c l i c a l c h a n g e s i n the A f r i c a n ecosystem. Nature  W e s t e r n , D . and W . K . L i n d s a y . 1984. a savanna e l e p h a n t p o p u l a t i o n . 244.  Seasonal herd dynamics of Afr. J. Ecol. 22:229-  Weyerhaeuser, F . J . 1982. On t h e e c o l o g y of elephants. M.Sc. t h e s i s , Yale Univ. Weyerhaeuser, F . J . in Tanzania's 23:235-243.  the  L a k e Manyara  1985. S u r v e y of e l e p h a n t damage t o baobabs L a k e Manyara N a t i o n a l P a r k . Afr. J. Ecol.  White,  S.E. 1914. African London. 415pp.  Camp F i r e s .  White,  S.E. 1915. The R e d i s c o v e r e d London. 358pp.  Thomas N e l s o n  Country.  & Sons,  Hodden & S t o t t e n ,  Williams, L . A . J . 1964. G e o l o g y of t h e Mara R i v e r - S i a n a a r e a . R e p o r t N o . 66, G e o l o g i c a l S u r v e y of K e n y a , M i n i s t r y of Natural Resources, N a i r o b i . Wing,  L . and I . O . B u s s . Monogr. N o . 19.  1970. E l e p h a n t s and f o r e s t s . The W i l d l i f e S o c i e t y .  Wild.  223  Woosnam, R . B . 1913. R e p o r t on a s e a r c h f o r G l o s s i n a on t h e Amala ( E n g a b e i ) R i v e r , s o u t h e r n M a s a i R e s e r v e , E a s t A f r i c a protectorate. Bull. E n t . Res. 4:271-278. Wyatt,  Zar,  J . R . and S . K . E l t r i n g h a m . 1974. The d a i l y a c t i v i t y of the e l e p h a n t i n the Rwenzori N a t i o n a l P a r k , Uganda. E. Afr. Wild. J. 12:273-290.  J . H . 1984. B i o s t a t i s t i c a l A n a l y s i s , 2nd E d i t i o n . P r e n t i c e - H a l l , Engelwood C l i f f s , N . J .  224  APPENDIX A . This  appendix  mathematical model  of  Therefore,  or it  6.  is  and  the  logic  of  the  "Reversed"  and  resprouting  but  those  s e e d l i n g s which l o s e kept  and  as  fire  act  and  other  only  wildebeest  some s c e n a r i o s )  each  adult  height  year.  mortality  seedlings  agents. all  rate  (R)  class  (3m)  totally  in  the  order  to  in  "awk"  as  to  do  refers  not  below  and  Only  are  elephants  wildebeest,  reversion  ground  describes  stems  1m.  to  return.  reduced to  all  ( w h i c h may i n c l u d e  is  seedlings:  "Inhibited"  not  Fire,  act  a g e n t s of  are  class  on  "Killed"  and  date.  but  height  and b r o w s e r s  The d a t a  used  in  effects  removed  some  can  as  expression  "inhibiting".  some l a t e r  i n the  act  Recruitment the  at  browsers  calculations.  terminology  different  are  those  level  effectively  the  further  dynamics  particular  The p r o g r a m was w r i t t e n  three  which to  in  a  and  computer.  have  trees  refers  define  model.  "reversing", or  to  woodland  expresses  used  each mathematical  A n i m a l s and f i r e  seedlings  is  definitions  i n the  equation  which  r u n on a Vax 11/750  "killing",  parameter  incorporated  important of  the  Each  number  word d e s c r i p t i o n s follow  provides  expressions  Chapter  proportion  THE WOODLAND DYNAMICS MODEL  elephants  agents,  whereas,  elephants  under  inhibition. the  the  presented  p r o p o r t i o n of number of as  the  trees  adult  natural  entering  trees log  (In)  dying of  values. Definitions $1  of  the  parameters  entered  into  =  p r o p o r t i o n of  s e e d l i n g s and t r e e s  $2 =  p r o p o r t i o n of  seedlings  "reversed"  the  model:  burned by  R-  225  wildebeest $3 =  p r o p o r t i o n of  seedlings  " k i l l e d " by  seedlings  "reversed"  seedlings  "inhibited"  elephants $4 =  p r o p o r t i o n of  by  elephants $5 =  p r o p o r t i o n of elephants plus  $6 =  no.  of  other $7 =  adult natural  p r o p o r t i o n of  other  trees  by  browsers  " k i l l e d " by e l e p h a n t s  and  causes seedlings  "inhibited"  by  wildebeest  Word d e s c r i p t i o n s  of  equations  q5 =  no.  of  seedling  q6 =  no.  of  seedlings  by q7 =  q7l  =  wildebeest  no.  q9l  =  of  by  seedlings  population after  annually  "reversal"  regenerating  after  regenerating  after  "reversal"  regenerating  after  "reversal"  wildebeest of  seedlings  elephants  starting no.  model:  fire  "reversal" no.  the  regenerating  seedlings  by q9 =  entering  p r o p o r t i o n of  by q8 =  i n the  number of  surviving  fire,  regenerating elephant,  seedlings  and  wildebeest  effects qlO  =  no.  of  seedlings  "inhibited"  by  browsers  q101  =  no.  of  seedlings  "inhibited"  by  wildebeest  226  q11  =  no.  of  recruits  q12  =  no.  escaping  fire  q13  =  no.  escaping  (F)  q14 =  no.  escaping  F , W, and e l e p h a n t  "reversal" q15 = q151  =  in f i r s t  height  class  (<l5cm)  (F) and w i l d e b e e s t  " r e v e r s a l " (W) "killing"  and  (E)  no.  escaping  F , W, E , and o t h e r  browsers  no.  escaping  F , W, E , and B = n o .  (B)  escaping  year 1 q16 =  proportion escaping  after  ql7  =  no.  escaping  at  year 6  ql8  =  no.  escaping  at  year 7  ql9  =  proportion escaping  at  q20 =  no.  escaping  at  year  11  q21  =  no.  escaping  at  year  12  q22  =  proportion escaping  q23 =  no.  q24 =  recruitment rate  q25 =  natural  Mathematical q5 =  escaping  log  at  year  at  year 7  year  (R)  of R  expressions  i n c o r p o r a t e d i n the  170.  q7 = $ 2 * ( 1 - $ 1 ) = q7*8500  q8 = $4*8500 q9 = (q5+q6+q71+q8) q9l  12  15  q6 = 0.95*$1*8500  q71  year 1  = q9*(1-$1)*(1-$3-$4)*(1-q7)  model:  at  227  q10 qlOl q11  q91*$5 = q9l*(1-$5)*$7 = q9+q10+101  q1 2 = ( q 1 1 * ( 1 - $ 1 ) ) + ( 0 . 0 5 * $ 1 * q 1 1 ) q1 3 = ( l - q 7 ) * q 1 2 q1 4 = ( 1 - $ 3 - $ 4 ) * q 1 3 q15  = (1-$5)*q14  q1 51  = (1-$7)*q15  q16  = q151/q11  q36  = Iog10(l00000*q16)  q17  = q151*(exp(5*log(q16)))  q37  = Iogl0(l00000*(q17/q11))  ql8  = (q17*(1-$1)) + (q17*$1*0. 1 )  q1 9 = q 1 8 / q 1 7 q20  = ql8*(exp(4*log(q19)))  q47  = log10(100000*(q20/q11))  q21  = (q20*(1-$1))+(q20*$1*0.66)  q22 = q 2 1 / q 2 0 q23  = q21*(exp(3*log(q22)))  q57  = Iogl0(l00000*(q23/q11))  q24 q25  q23/$6 = ln(q24)  

Cite

Citation Scheme:

    

Usage Statistics

Country Views Downloads
United States 20 3
China 15 12
Germany 7 2
United Kingdom 5 3
Kenya 5 4
Unknown 3 2
Ivory Coast 2 0
Japan 2 1
Brazil 1 1
Norway 1 0
City Views Downloads
Unknown 16 13
Ashburn 14 0
Beijing 10 0
Shenzhen 5 12
Guildford 4 0
Mountain View 3 0
Hamburg 3 0
Tokyo 2 0
Redwood City 1 0
Kansas City 1 0
Redmond 1 0
Newry 1 0

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}
Download Stats

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0097282/manifest

Comment

Related Items