THE ECOLOGY OF AN AFRICAN SAVANNA FRUIT BAT COMMUNITY:
RESOURCE PARTITIONING AND ROLE IN SEED DISPERSAL
by
DONALD W. THOMAS
Thesis submi ed o he deg ee
o Doc o o Philosophy o he
Uni e si y o Abe deen
FEBRUARY, 1982
I decla e ha he wo k p esen ed in his hesis was unde aken and
comple ed by mysel . I has no been submi ed in any p e ious
applica ion o a deg ee. All e ba im quo a ions a e iden i ied wi h
quo a ion ma ks and he sou ces o in o ma ion ha e been ully
acknowledged.
Donald W. Thomas
11 Feb ua y, 1982
ACKNOWLEDGEMENTS
Whe e does one begin o g a e ully acknowledge
all he help
and suppo ,
bo h ma e ial and mo al, ha has
been so eadily o e ed?
Fi s , I wish o exp ess my g a i ude
o he inanima e; o A ica, which
cap u ed my imagina ion, inspi ed his s udy, and spawned a people so
wa m and hospi able.
I wish o hank D . Ad ian Ma shall, who ollowed he p og ess o his
wo k wi h in e es , who o e ed many aluable sugges ions, and who
showed pa ience when necessa y.
I hank Roge Vua oux, di ec o o LAMTO, o making a ailable all he
esou ces o he ield s a ion uns in ingly and o making my s ay a
ewa ding expe ience. I also g a e ully acknowledge he help and
iendship o Konan N'd i, Konan Ge main, Kouadio Mix, and all he
es o he s a a LAMTO.
Thanks a e due o M. Jean-Luc Tou nie , di ec o o he S a ion de
G6ophysique, o always making he esou ces o his labo a o y eely
a ailable.
Thinks go o Jean-Louis and F ancoise Ti e o d who en iched my s ay wi h
wa m h and hospi ali y and many he long e ening o Ta o and bee .
D . Paul Racey dese es special men ion. His en husiasm o biology and
s imula ing con e sa ions we e e eshing o encoun e .
The smoo h p oduc ion o his hesis was made possible h ough he kind
help o D . R. Ralph, who pe mi ed me o ha e access o he acili ies
o he O sho e Ma ine S udies uni and hei wo d p ocesso .
Finally, bu ce ainly no o lesse (much!) me i , I wish o make
special acknowledgemen o Ma in Nicoll and And ew McWilliam, wi h whom
I passed many he hou in s imula ing disag eemen .
This s udy was suppo ed h ough a Na u al Sciences and Enginee ing
Resea ch Council o Canada Doc o al Fellowship and h ough
a
Na ional
Geog aphic Socie y esea ch g an o 1980/81.
ABSTRACT
This s udy was unde aken o examine he s uc u e o a ui ba
communi y, i s seasonal a ia ion in composi ion, he mechanisms o
pa i ioning o ui esou ces, and he ole o he ba s as seed
dispe se s in sou he n Guinea sa anna in A ica.
The communi y is s uc u ed on wo le els. A g oup o ou species
emain esiden in he sa anna- o es mosaic zone yea ound and h ee
species in ade he communi y om he sou he n o es zone seasonally.
Die s indica e ha he esiden species (Epomops bue iko e i,
Hypsigna hus mons osus, Lissonyc e is angolensis, and Mie op e opus
pusillus) o age in h ee mu ually exclusi e zones and he die o e laps
be ween hese species a e low, excep in he case o E. bue iko e i and
M. pusillus. The la e wo species o e lap hea ily-in bo h die
selec ion and in o aging zone, and e idence is p esen ed showing hem
o be in se e e compe i ion when esou ces a e limi ing. They co-exis
only in high ui densi y pa ches.
The mig an species (Eidolon hel um, Myonyc e is o qua a, and
Nanonyc e is eldkampi) in ade he communi y a he onse o he ains
when he di e si y o ui esou ces inc eases o e he d y season
low. Each uses one o he h ee o aging zones and co-exis ence is
media ed h ough: 1) he use o di e en ui sizes and 2) o e lap
wi h he esiden species only on ui s ha a e supe -abundan . D y
season limi a ion o he esiden species' popula ions appea s o be
impo an in pe mi ing he in asion o his communi y.
-The seed ain gene a ed by ba s and bi ds was measu ed wi h g ound-based
collec ing shee s on ansec s h ough he sa anna. Seed ains we e
imp essi ely high, wi h each squa e me e ecei ing on he a e age 3.3
seed loads annually. Ba s accoun ed o 95.7-98.2% o he measu ed
ain. The ba dominance appea s due o he highe p opo ion o ime
spen in ligh compa ed wi h bi ds. •
TABLE OF CONTENTS
SECTION
PAGE
1. INTRODUCTION
1
2.
STUDY AREAS
7
2.1
LABORATOIRE D'ECOLOCIE TROPICALE DE LAMTO
7
2.1.1 INTRODUCTION
7
2.1.2 CLIMATE
8
2.1.3
VEGETATION
9
2.2
WANGO FITINI
11
2.3
OTHER SITES
12
3.
MATERIALS AND METHODS
14
3.1
MIST NETTING 14
3.2
REPRODUCTION AND GROWTH
16
3.3
DIET
17
3.4
MOVEMENTS
19
3.5
POPULATION SIZE
20
3.6
FRUIT CONSUMPTION AND SEED DISPERSAL
21
3.7
SEED PREDATION AND GERMINATION
22
3.8 FRUIT TREE ABUNDANCE AND FRUIT PRODUCTION
22
3.9 CONTROLLED FEEDING EXPERIMENTS
23
3.10 STATISTICS
25
3.11
NOMENCLATURE
25
4. RESULTS
3.0
4.1 COMMUNITY COMPOSITION
30
4.1.1 SEASONAL COMMUNITY COMPOSITION:
RESIDENTS AND MIGRANTS
31
4..1.2 MIGRATORY SPECIES
32
4.2
REPRODUCTION AND GROWTH OF RESIDENT SPECIES AT LAMTO
37
4.2.1
BIRTH PERIODS
37
4.2.2 OESTRUS
39
4.2.3
MATING PERI:
I
DS
40
4.2.4 114_LE CALLTIG BEBA7IOUR
43
SECTION
PAGE
4.2.5
RECRUITMENT
AND
GROWTH: E. BUETTIKOFERI
42
4.2.6
RECRUITMENT AND GROWTH: M. PUSILLUS
46
4.3
REPRODUCTION OF MIGRANT SPECIES
50
4.4
REPRODUCTION AT WANG° FIIINI
52
4.5
ECOLOGY OF THE SPECIES AT LAMTO
53
4.5.1
ROOST SITES
53
4.5.2
ACTIVITY PERIODS
54
4.5.3
HABITAT USE
55
4.5.4
DIET AND RESOURCE USE
57
4.5.5
ABUNDANCE OF BAT EXPLOITED FRUIT SPECIES
77
4.5.6
FRUIT PRODUCTION OF F. CAPENSIS
79
4.5.7
MOVEMENTS AND POPULATIONS:
MICROPTEROPUS PUSILLUS
82
4.5.8
EPOMOPS BUETTIKOFERI:
MOVEMENTS AND POPULATIONS
87
4.5.9
RADIO TRACKING
96
4.6
FRUIT CONSUMPTION AND SEED DISPERSAL
103
4.6.1
F. CAPENSIS IN THE SNB
103
4.6.2
F. CAPENSIS ALONG GALLERY FOREST EDGE
106
4.6.3
ADENIA CISSAMPELOIDES
107
4.6.4
FAECAL AND SEED RAIN
107
4.6.5
SEED PREDATION
110
4.6.6
GERMINATION OF FICUS CAPENSIS
112
4.7
FRUIT COMPOSITION
114
4.8
CONTROLLED FEEDING EXPERIMENTS
114
4.9
ECOLOGY OF THE SPECIES AT WANG° FITINI
120
4.9.1
ROOST SITES AND CALLING BEHAVIOUR
120
4.9.2
HABITAT USE
121
4.9.3
DIET AND FOOD RESOURCE USE
123
5.
DISCUSSION
128
5.1 THE WEST AFRICAN PTER0i0DIDAE
128
5.2
THE WEST AFRICAN MIGRANTS
129
5.2.1
EIDOLON HELVUM
130
5.2.2
MYONYCTFRIS TOROUATA
133
5.2.3
NANONYCTERIS VELDKAMPI
136
5.3
COMMUNITY STRUCTURE AND RESOURCE PARTITIONING
137
5.3.1
THE DATA BASE
137
5.3.2
DIET OVERLAP AND SPECIES PACKIEG
138
5.3.3 NORTHWARD MIGRATIONS
141
SECTION
PAGE
5.3.4
RESOURCE PARTITIONING
144
5.3.5
E. BUETTIKOFERI AND M. PUSILLUS
148
5.4
THE TIMING OF REPRODUCTION
155
5.5
FRUITS AND FRUGIVORES
161
5.5.1
FRUITS AS A DIET
161.
5.5.2
PROTEIN AND ENERGY BUDGETS
164
5.5.3
PLANT STRATEGIES
178
5.5.4
BAT FRUITS
179
5.5.5
SEED RAIN
183
6. SUMMARY
188
7. LITERATURE CITED
190
-1-
1.
INTRODUCTION
Ea ly in hei his o y angiospe ms e ol ed a illa ed seeds and animal
dispe sed (zoocho ic) ui s. The de elopmen o hese ea u es, i no
causa i e, was closely associa ed wi h hei ea ly adia ion and
subsequen domina ion o mos plan communi ies (Co ne 1949). Now, in
Neo opical deciduous and we o es s 53-79%, and in A ican we o es s
up o 80%, o he ee and sh ub species p oduce leshy ui s consumed
p incipally by bi ds and mammals (Jones 1956, Daubenmi e 1972, F ankie
e
al.
1974, Hil y 1981).
Since he de elopmen o he p imi i e a illa ed seed, he co-e olu ion
o plan s and hei animal seed dispe se s has gi en ise o a as
a ay o ui o ms, sizes, nu ien con en s, display posi ions, and
ui ing phenologies. Since all o hese ea u es di ec ly a ec he
a ailabili y and a ac i eness o ui s o dispe se s and hence he
ep oduc i e success o he pa en plan s, hey a e unde s ong
selec i e p essu e. This has bes been demons a ed by He e a (1981)
who showed ha a ia ion in he a io o ewa d (pulp) o ballas
(seed) in he ui s o di e en Smilax aspe a (Liliaceae) popula ions
was ela ed o compe i ion o dispe se s. In popula ions whe e
compe i ion o a limi ed numbe o dispe se s was se e e and whe e he
ui ing season was es ic ed by clima e, selec ion bad a ou ed an
Inc eased in es men in he pulp and less in he seed (g ea e ewa d o
ballas a io) in o de o make ui s mo e a ac i e o he a ian
.dispe se s. Along a simila line, Snow (1965) a gued ha compe i ion
be ween sympa ic Miconia species o dispe se s in a opical
en i onmen had led o he s agge ing o ui ing seasons.
Snow (1971) poin ed ou ha ui a e p oduced o a ac ugi o es
and so a e usually conspicuous and some imes nu i ious. Howe e , some
ui s may be aken by a wide ange o ugi o es while o he s may be
used by only a ew o he po en ial ugi o es.. Ex eme examples o
his a e Ficus suma ana (Mo aceae) in Malaysia which is isi ed by up
o 25 species o bi ds and eigh species o mammals (no men ioning
noc u nal mammals; McClu e 1966) and Lo an hus species (Lo an haceae) in
he Neo opics which a e used p ima ily by he mis le oe h ush (Dlcaeum
-2—
sanguinolen um (Howe and Es ab ook 1977). F ugi o es di e in hei
ea men o seeds and mo emen s a ound ui ing ees (K e ing and Roe
1949, Howe and P imack 1975, Howe and Vande Ke ckho e 1981) and plan s
ha e di e ing equi emen s o "sa e si es" ( hose o e ing he minimum
equi emen s o seed su i al, ge mina ion, and g ow h; Janzen 1971,
Ha pe 1977)
.
. Selec ion, hen, should ac o modi y ui
cha ac e is ics in such a way as o make ui s mos a ac i e and
accessible o ugi o es o e ing he highes "quali y" o seed
dispe sal cha ac e is ics (sensu Howe and Es ab ook 1977). Thus plan s
o en a ac some subse o he ugi o es a ailable. The b ead h o
his subse is ela ed o he spa ial and empo al p edic abili y o
"sa e si es". I , o example, "sa e si es" a e unp edic able in space
and ime, hen plan s Maximize he dis ibu ion o seeds in he
en i onmen by using he g ea es ange o ugi o es possible (a la ge
subse ; Ficus s a egy). I "sa e si es" a e p edic able in space
and/o ime, selec ion a o s ui cha ac e s which es ic he subse
o species equen ing he app op ia e si es (Lo an hus s a egy).
McKey (1975) and Howe and Es ab ook (1977) ha e deal in de ail wi h he
po en ial e ec s o a ying nu ien alue (suga y, low p o ein ui s
s lipid— and p o ein— ich ui s) and ui ing schedules (mass ui e s
s ex ended ui e s) on he ypes and numbe s o ugi o es a ac ed.
The ne esul o he a ying ui ing cha ac e s (size, shape, nu ien
con en , and ui ing schedule among o he s) is o make di e en ui s
a ailable o di e en ugi o es. Supe imposed on his, compe i ion
be ween ugi o es ac s o educe die o e lap e en mo e. Fleming
(1979), e iewing he li e a u e on ugi o y, concluded ha he e was
low die a y o e lap be ween e en closely ela ed ugi o es and ha
compe i ion among ugi o es was uncommon. Lack (1976), a ionalizing
he appa en ly high o e lap in. he die s o ugi o ous bi ds in
T inidad, also s essed he unde lying di e ences in die s. The
di e ences in die s o ugi o es mean ha knowledge o he
pa i ioning o ui esou ces wi hin ugi o e guilds is no only o
in e es o zoologis s, bu is also c i ical o he unde s anding o
plan communi y composi ion, seed mo emen s, and he abili y o plan s
o ecolonize dis u bed si es.
-8-
2.1.2. CLIMATE
The clima e o Wes A ica is go e ned by he seasonal no hwa d and
sou hwa d mo emen s o he in e - opical con e gence zone, sepa a ing
he mois equa o ial ma i ime ai mass om he d y con inen al ai
mass, be ween app oxima ely 5°N in Janua y and 17-21°N in July (I eland
1962; Figu e 2). A ain bel ollows app oxima ely 320-480 km behind
he on , so a a pa icula si e he ains begin when he on is
h ee o i e deg ees la i ude o he no h. Since he ain bel is
ela i ely na ow, a sou he n si es he ains may pass o he no h in
July/Augus c ea ing a sho d y season. This mid- ainy season hia us is
less ma ked o absen a mo e no he n si es.
As a esul o he in e - opical con e gence zone mo emen s, he
clima e a LAMTO shows ou mo e o less de ined seasons: a majo d y
season (No embe o Feb ua y), a majo ainy season (Ma ch o July), a
mino d y season (July o Sep embe ), and a mino ainy season
(Sep embe o No embe ; Figu e 3). While ain all g aphs and
hy he g aphs show ou measu eable seasons (Figu es 3 and 4), i is
unclea whe he plan communi ies pe cei e he July/Augus mino d y
pe iod. Nei he o es no sa anna species show any no iceable inc ease
in lea all du ing his pe iod (De ineau 1976a, Menaul and Cesa 1978)
and clima og aphs (Figu e 5) show ei he no po en ial wa e de ici (by
he me hod o Wal e and Muelle -Dambois 1975) o only a b ie de ici
In Augus ( om he da a o Eldin in De ineau 1976a).- Fo plan s, his
d y pe iod may well be bu e ed by soil e en ion. Wha e e he
si ua ion, he mino d y season is b ie . Fo his eason, unless
o he wise speci ied, I will o con enience e e o only wo impo an
seasons: he we season las ing om Ma ch un il Oc obe and he d y
season las ing om No embe un il Feb ua y.
A LAMTO, a e age daily maximum empe a u es aken o e 18 yea s ange
om 30.1°C in Augus o 35.3°C in Feb ua y. Daily ela i e humidi y,
a e aged om eadings aken a 0600h, 1200h, and 1800h, anges om 71%
in Janua y o 84% in June. The mean annual, p ecipi a ion o e
he
pas
15 yea s was 1246
251mm (da a om J-L. Tou nie , pe s. comm. and
Leco die 1974).
20-
-
_
,
,
.1
,
1
J FMAM
.
JJ AS OND
MONTH
FIGURE 2: The app oxima e posi ion o he in e — opical con e gence
zone h oughou he yea in Wes A ica. The s ippled a eas
indica e he mon hs wi h ain a LAMTO (6°13'N) and Wango
Fi ini (9°50'N). (Da a om Hopkins 1965).
FIGURE 3: The mon hly ain all a LAMTO h ough 1979 and 1980 (da a
cou esy o J—L. Tou nie , S a ion de Geophysique de LAMT0).
The inse shows he 15 yea a e age ain all o each mon h
(mean ± s anda d e o ).
I
I
I
cS
cl,
c)
c)
(NI
,--
P
A
(-
1
-9 11VdNIV2:1 30V2EAV
1
II
-
0
0
Z
0
( )
<
2
<
1
H
2
(I)
<
--)
2
<
0
0
U-)
2
U-
--)
i•I
I
I
I
I
1
T
I
I
o
Q
D
o
N%--
o
c3
.....
)
(
11
-
1u
)
11 iNIVd
—
cD
_
aLl
Co
5
p 4
X—)
X
—)
X
0.11.01.
1
I
I.
I
U
"
)
c•4
• co
co
(Y)
(J0)
XVN
1NV1A1
FIGURE 5: Two ypes o clima e diag ams o IAMTO.
(A) shows he mean mon hly ain all
) and he
e apo anspi a ion po en ial (+--- ) [da a om De ineau
1976a].
(B) shows he mean mon hly ain all
and he mean
maximum daily empe a u e
whe e 10°C is equa ed
wi h 20mm o ain all on he Y axis. See Wal e and Muelle -
Dambois (1975) o a jus i ica ion.
The s ip
n
led a eas indica e mon hs whe e e apo anspi a ion
exceeds ain all, leading o a po en ial soil wa e de ici .
A
200-
160-
120-
80-
_
40-
_
E
E
200-
-J
< 160-
LL
Z
<
cc 120-
_
40- 80-
— 30-
o
0
—
20- 40-
o
_
m 10-
Lu
1---
_
-
j
F M AM J
i
A
b i
D
MONTH
-9-
2.1.3. VEGETATION
The ege a ion a LAMTO is a mosaic o open o Wooded sa anna bu ned
once yea ly (73.6% o he su ace a ea), an 80 ha plo o expe imen ally
i e-p o ec ed and egene a ing sa anna (3.2%), and closed o b oken
canopy o es (23.2%).
Sa anna is essen ially a i e-main ained ege a ion associa ion (Lamo e
1978) and as a esul is no ably poo in bo h densi y o woody plan s
and in species di e si y. Th ee heigh s a a a e ecognizable: a he b
s a um wi h ew woody species (0-2 m), a sh ub s a um (2-8m), and a
ee s a um (o e 8m). The 0-2 me e he b s a um domina es in
biomass, accoun ing o om 100% o a minimum o 79.9% o he o al
plan biomass a all sa anna si es (4enaul and Cesa 1979). The o e 8
me e s a um is gene ally ep esen ed only by Bo asus ae hiopum. In
he 2-8 me e s a um some 30 species o woody plan s a e ep esen ed.
O hese, 16 a e uncommon, being ep esen ed in only 33% o censused
plo s (n=6), 10 species a e ound in 66% o censused plo s, and only
ou species (B idelia e uginea, Cussonia ba e !, C ossop e y
eb i ugum, and Pilios igma honningi) a e ound in o e 70% o plo s
(calcula ed om Menaul 1971). These la e ou species accoun o
o e 90% o bo h he biomass and s ems pe hec a e o woody plan s a
LAMTO. The densi y o sh ubs and ees o e wo me es in heigh a ies
om 0 o 1500 s ems/ha (x= 250 s ems/ha; Menaul and Cesa 1979).
Fo es s co e nea ly 25% o he su ace a ea a LAMTO and con ain
app oxima ely 400 species o ees and sh ubs (J-L. De ineau, pe s.
comm.). These o m disc e e blocks a ying om 20 o 35 me es in
heigh and ha ing sha ply de ined bounda ies wi hsa anna. Fo
con enience o es s a e usually, di ided in o wo ypes, galle y and
i e ine, bu in eali y hey ep esen a con inuum o plan communi ies
om d ie inland (galle y o es ) o mois e ipa ian ( i e ine o es )
si es (De ineau 1976b). Fo excellen e iews o he a chi ec u e,
densi ies, and species composi ion o o es s a LAMTO see De ineau
(1976a, 1976b).
Galle y o es s p ecisely ollow hyd og aphic ea u es o he landscape
(dep essions a ound seasonal s eams leading in o he Bandama Ri e ; see
Figu e 6) and as a esul a e no mally no mo e han 15 o 20 me es
wide. This associa ion wi h dep essions is due in pa o he g ea e
soil mois u e and in pa o he i e-p o ec ion o e ed by he
b oad-lea ed pe e iials ha es ablish he e (eg. Thauma ococcus
danielli). Galle y o es s a e domina ed by Cola gigan ea, Mallo us
opposi i olius, Diallium guineense, Malacan ha alni olia, and Teclea
e doo niana (among o he species), bu also include such impo an
species as Elaeis guineese, An ia is a icana, Chlo opho a excelsa, and
Diospy os mespili o mis. These la e wo species and Cola gigan ea
p o ed o be impo an in his s udy.
The d ie inges o he i e ine o es s included many o he species
common o he galle y o es s, bu C. gigan ea, C. eXcelsa, and
D. mespili o mis ypically a e less common o absen . A a ie y o
species including Baphia pubescens, Nesogo donia papa e i e a,
Lasiodiscus milb aedii, and Panco ia bijuga each hei highes
densi ies a hese si es. On he mois e i e banks and islands hese
species a e eplaced by C o on sca ciessi, P e oca pus san alinoides,
Cola lau i olia, Cynome a magalophylla, Manilka a obo a a, and Pa ina i
congensis; all species ha a e a ely ound on he d ie si es.
The edge o galle y and i e ine o es s ha e hei own speci ic plan
associa ions domina ed by many o he mo e o less i e- ole an , as
g owing, and shade in ole an woody species and climbe s. In his wo
o ou me e wide zone, Ficus capensis, F. allis-choudae, Smea hmania
pubescens, An hocleis a nobilis, Nauclea la i olia, Mimusops kummel,
.01ax subsco piodea, Napoleanea ogelii, Adenia cissampeloides, A.
meigei, and A. loba a a e abundan . Fo es s a e con inually ep-nding
in o sa anna as is indica ed by he p esence o Bo assus ae hionam
inside he galle y o es s and by he obse ed ad ance
o
he edge in o
mapped si es (R. Vua oux pe s. comm.). I is h ough he ad ance o
hese i e- ole an species ha he g ound is p epa ed (i.e. be g ass
co e educed and he soils main ained mois e by lea co,.e ani
shading) and he mic oclima e made a o able o he ge mina ioa -nd
g ow h o he less i e- ole an species (Me aul 1977).
‘ 4 q
la i e l!
aead
% .
a 0,
e o
ono
cssz'
Oo
4
5441'''4
'eye
g°
°
0
0-4301.
0
6
p.
N4k
1
471
SCALE
4
1 km
0
b
CZ5
0
q3
(4);
%
a .
00
0
oo
0
:.•.
SAVANNA
GALLERY FOREST
RIVERINE FOREST
••
•
FIGURE 6: A map o LAMTO showing he majo ege a ion blocks wi hin he
2500 ha. ese e and he si es whe e mos o he wo k was
ca ied ou . The main ne ing si es we e he SNB, he
i e ine o es (A), and nine galle y o es edge si es
(B-J). The aecal collec ing-shee ansec s we e se ou in
he SNB and along h ee galle y o es edges (1-3).
a
-16-
3.2.
REPRODUCTION AND CROWTH
Females we e eco ded in six ep oduc i e classes:
1)
imma u e: smalle han adul size, no ha ing p e iously gi en bi h
as e idenced by small and unsuckled nipples (nullipa ous).
2)
nullipa ous, no -p egnan : adul in p opo ions bu no p egnan .
3)
nullipa ous, p egnan : p egnan o he i s ime.
4)
pa ous, non- ep oducing: ha ing p e iously gi en bi h as e idenced
by la ge and suckled nipples, bu nei he p egnan no lac a ing.
5)
pa ous, p egnan : ha ing p e iously gi en bi h and palpably
p egnan .
6)
pa ous, lac a ing: lac a ing (de e mined by he exp ession o milk
om he nipples) a he ime o handling. In cases whe e emales we e
simul aneously p egnan and lac a ing, hey we e classed as
'pa ous, lac a ing' unless o he wise speci ied.
I o en e e o emales as imma u e o adul . Adul s
comp ised
all pa ous emales and nullipa ous, p egnan emales.
•
Since p egnancy was de e mined by palpa ion only, ea ly s ages o
emb yonic de elopmen we e unde ec able. Due o he na ow abdomen and
he luid in es inal con en s o ui ba s, cho ionic sacs
app oxima ely 3mm in diame e o la ge we e de ec able and p egnancies
we e p obably missed only o a b ie pe iod ea ly in ges a ion.
Males we e classi ied in wo ca ego ies:
1)
imma u e: ha ing abdominal es es o small sc o al es es and no
ha ing ye de eloped seconda y sexual cha ac e s (epaule es o
E. buc iko e i, E. gambianus, M. pusillus, and N. eldkampi; nasal
In la ions o H. mons osus; h oa u s o longe coa se hai s o
L. angolensis and M. o qua a). This ca ego y also included pube al
males showing signs o inc easing es es size and de eloping seconda y
sexual cha ac e s.
2)
adul : ha ing ully de eloped es es and seconda y sexual cha ac e s.
To analyse he g ow h o coho s o young as g aphs o weigh agains
Lime i was necessa y o ha e a common ime
alds
o all indi iduals o
-17-
each coho . Since pa u i ion da es canno be de e mined p ecisely,I
a bi a ily measu ed ime as days pas 15 Ma ch o coho 1, o 15
Sep embe o coho 2 o each yea . These da es app oxima ed he
middle o he peak pa u i ion mon hs. I s ess ha his measu e is
a bi a y and is no in ended o ep esen he days elapsed since
bi h. Since coho s could ha e had modal pa u i ion da es be o e o
a e hese da es, he ele a ions o di e en cu es canno be
compa ed. Slopes, ep esen ing he g ow h a es, can be compa ed since
hey depend only on changes in weigh wi h ime.
Equa ions o all g aphed cu es we e machine calcula ed using he leas
squa es me hod (Sokal and Rohl , 1969). The cu es bes desc ibing da a
we e hose ha ing he highes eg ession coe icien (
2
). Weigh da a
o coho s we e also i ed o he logis ic g ow h equa ion ollowing
Rickle s' (1967) me hod, in ol ing he con e sion o indi idual weigh s
o pe cen s o asymp o ic (=adul ) weigh s. The asymp o ic weigh s we e
aken as 190g o male E. bne iko e i, 120g o emales, and 31g o
bo h sexes o M. pusillus. The slope o he s aigh line plo ob ained
om his me hod was used o calcula e he logis ic g ow h cons an (IC)
o compa ison wi h p e iously published da a.
3.3.
DIET
Any aeces oided in he ne o du ing handling we e collec ed o la e
iden i ica ion o he seeds (i p esen ) o pulp, as well as o
mic oscopic examina ion o de e mine pollen and/o plan ege a i e
issue con en . Iden i ica ion o he ui species om he seeds was
done by quali a i e compa ison wi h a pe sonal e e ence seed collec ion
om known ui s om he LAMTO o Wango Fi ini egion. Any
uniden i ied seeds we e plan ed a LAMTO o a he Depa men o Bo any,
Uni e si y o Abe deen, o la eT ecogni ion o he seedlings. I also
plan ed samples o iden i ied seeds o e i y my abili y o ecognize
seeds. Pulp aeces con aining no seeds usually had cha ac e is ic
colou s and ex u es (and in he case o Vi ex doniana a cha ac e is ic
smell). These we e iden i ied by compa ison wi h aeces p oduced by
cap i e ba s ed suspec ed ui s. When coupled wi h egula ou s made
o assess he ui ing species a ailable and obse a ions o ba s
eeding, aeces iden i ica ion was in mos cases ema kably simple.
-is-
The analyses o he die o he esiden and mig an ba species a
LAMTO a e based on 589 aeces (ne aeces) collec ed om indi iduals
ne ed a all si es o e he wo yea s. I also collec ed 1236 aeces
om below he E. hel um oos ( oos aeces). These ha e been ea ed
sepa a ely since hey swamp he smalle E. hel um samples om o he
imes o he yea . I ha e also included sca e ed obse a ions o ba s
eeding on pa icula ui species o supplemen he aeces da a.
The use o die i ems by he ba species we e analysed om he e-cal
da a pooled o he en i e yea in o de o maximize sample sizes.
While his may unde - ep esen impo an bu highly seasonal ui
species, his was una oidable. In his way seasonal changes in species
use was masked, bu I ha e d awn a en ion o seasonal a ia ions
whe e e his was in o ma i e. The ela i e specializa ion o he die s
o he ba s was analysed by calcula ing:
EVENNESS= [ p
ij
log
i
op
ij
]/ [logioN]
whe e p
ij
is he p opo ion o die i em j in he die o ba
species i and N is he o al numbe o ui s used by he species. Die
o e lap be ween species pai s was calcula ed by:
OVERLAP = 1
whe e
Pi
j
and P
hi a e he p opo ion o i em j in in he die s o ba
species i and h.
As an independen measu e o die species selec ion by E. bue iko e i,
I loca ed 32 eeding oos s and eco ded he ui emains below hem
o e 366 oos nigh s. The species we e coun ed as used i p esen
unde a oos ega dless o hei ela i e abundance.
.I also collec ed hai samples om he head and shoulde egions o ba s
o de e mine whe he hey had ecen ly isi ed lowe s based on aces
o pollen on he u . This p esupposes ha he ba s ac ed as
pollina o s and no nec a hei es, gaining access o nec a by pie cing
he ca olla and hus a oiding Con ac wi h he s amens as has been no ed
wi h Nanonyc e is eldkampi eeding on Spa hodea campanula a (Ayensu
1974). This la e eeding s a egy would be unde ec able by my
echniques. Hai samples we e immedia ely placed in clean pape pouches
and kep sealed un il la e moun ed on slides and examined unde a
compound mic oscope. Pollen was eco ded as p esen o absen in a
-19-
sample and iden i ied o species i possible by
.
compa ison wi h pollen
samples om known lowe s. I collec ed hai samples om 336 ba s a
LAMTO du ing Ma ch/Ap il, June/July, Sep embe /Oc obe , and No embe o
Janua y pe iods and om 113 ba s a Wango Fi ini in July/Augus and
No embe .
3.4
MOVEMENTS
To assess he mo emen s and si e ideli y o ba s a LAMTO I elied
hea ily on ecap u ing ma ked indi iduals wi h mis ne s. I ha e
ea ed each si e along galle y o es edges, in he SNB, and in he
i S ine o es as a poin and I did no conside ba s ecap u ed in
di e en ne s a he same si e as ha ing mo ed. In his way mo emen s •
we e only among galle y o es edge si es o be ween hese and he wo
o he si es ( he SNB and i e ine o es ). T ea ing hese mo emen s in
sec ion 4.5.7 necessi a ed making a leas one implici assump ion;
ha , in he absence o any habi a p e e ences, indi iduals would be as
likely o mo e among galle y o es si es as be ween hese
si es and he
SUB. S uc u al di e ences be ween he galle y o es edge (a na ow
"co ido " o habi a ) and he SNB (a plana habi a ) may ha e a ec ed
he alidi y o his assump ion, bu since he e was no way o es his
p oblem I ha e ea ed mo emen s di ec ed owa ds one habi a as
indica ing a eal p e e ence o ha habi a .
In addi ion, I a ached adio ansmi e s and acked
E.
bue iko e i
o p o ide mo e de ail on he ac i i y pa e ns and ange o
indi iduals. Ini ially I had hoped ha acking would be
a
majo
'componen o his s udy, bu due o ime con lic s and he
di icul ies
o ollowing mobile ba s in long g ass sa anna a nigh his p o ed o
be o only seconda y impo ance.
I loca ed indi iduals 'ca ying ansmi e s using a hand held ou
elemen Yagi an enna and AVM LA-12 ecei e . I a emp ed o s ay wi hin
. 100m o a ba 's posi ion and main ained a log o i s mo emen s and he
amoun io ime spen lying o hanging in. ees. Flying was easily
de e mined due o a ia ions in signal s eng h caused by he wa ing o
he an enna in ligh . This
has
been used success ully in p e ious
s udies on NeO opical ba s (Hei haus and Flem ing, 1978, Mo ison
1978).
-20--
The ansmi e s we e buil ollowing he ci cui y o Thomas (1979)
modi ied o inco po a e a ci cui boa d o speed cons uc ion, a mo e
e icien
-
1
4- wa eleng h an enna, and a mo e powe ul 2.8V li hium
ba e y. The inal ci cui y was encapsula ed in silicon ubbe (Dow
Co ning ATV 735 SILASTIC) and had inal weigh s be ween ou and i e
g ams (o be ween 2.5% and 4% o he ba s' body weigh s). The
ansmi e s had a heo e ical li e o 100 days; howe e in mos cases
ha achei ed was much less. The mos common cause o ailu e was due
o he shea ing o he an enna a he ansmi e /po ing junc ion. The
ansmi e , Yagi an enna, and AVM ecei e p o ided a wo king ange o
one o i e kilome es depending on he heigh o he ecei e o
ansmi e abo e he su ounding e ain. Ini ially he ansmi e s
we e glued o he do sal u wi h Silas ic (see B adbu y 1976, Hei haus
and Flemming 1978), bu ecap u es o wo indi iduals wi h la ge
abcesses o ced me o abandon his me hod wi h E. bue iko e i.
Subsequen ly ansmi e s we e a ached wi h a colla assembly
which
main ained he ansmi e in a mid-do sal posi ion. In no case did I
ind ha colla s caused any damage o he ba s.
3.5.
POPULATION SIZE
Popula ion sizes o E. bue iko e i and M. pusillus we e es ima ed
using a modi ied Jolly Sebe ma k- elease- ecap u e model a ailable on
compu e a he Uni e si y o Abe deen (D . S. Buckland, pe s. comm.).
This model was chosen since i accep s ec ui men and losses o
popula ions h ough immig a ion/bi h and emig a ion/dea h and also
p o ides con idence limi s o es ima ed pa ame e s. Since he Jolly
Sabe model has he weakness o occasionally gi ing nonsensical
es ima es o su i al (g ea e han uni y) o nega i e ec ui men (as
opposed o dea h) he model was 'modi ied o es ic es ima es o
pa ame e s wi hin he ze o- o-one ange. Buckland (1980) p ese s he
ma hema ical a ionale and p ocedu e o his modi ica ion.
To c ea e inpu da a o he model, cap u es we e pooled o one mon h
in e als. Indi iduals cap u ed once du ing any one mon h sample we e
igno ed o any subsequen cap u es.
-21--
3.6.
FRUIT CONSUMPTION AND SEED DISPERSAL -
To assess he impo ance o Megachi op e an ba s as ui consume s and
seed dispe se s, I concen a ed on he galle y edge zone and he SNB,
p ima ily because hese we e dynamic zones in a cons an s a e
o in asion
by
o es o sa anna species.
I selec ed Ficus capensis and Adenia cissampeloides o quan i a i e
measu es o ui consump ion. Using non— oxic la ex pain I
indi idually numbe ed ui s o he wo species. I hen no ed he ime
o disappea ance (=consump ion and dispe sal) o all (= ailu e o
dispe se)
by
means o dawn and dusk ou s. F ui s consumcd du ing he
nigh we e assumed o be aken by ba s and hose aken du ing he day
p ima ily by bi ds and squi els.
O he possible diu nal dispe se s
we e monkeys (Ce copi hecus ae hiops and C. pe au is a) and noc u nal
dispe se s we e po os (Pe odic icus po o). Howe e , hese species we e
uncommon a LAMTO due o hun ing p essu e and I ne e obse ed hem
aking F. capensis o A. cissampeloides ui s in he a eas whe e I
wo ked. Th ough his ma king p og am I ollowed he a e o 1049
F. capensis ui s on 38 ees in he SNB and 243 ui s on se en
ees along one galle y edge. A sample o 100 A. cissampeloides ui s
we e ma ked a his la e si e.
I measu ed he a e o dispe sal o seeds in
,
e
-
ces wi hin he SNB and
in o he i e me e wide bel o sa anna bo de ing h ee galle y o es
edges
(see
Figu e 6) using plas ic collec ing shee s. These shee s we e
se ou in he open (no unde ees) so he aeces ound we e only o
• lying bi d o ba o igin. Ini ially I se ou h ee ansec s (one
20m
2
in he SNB be ween 5 Janua y and 9 June 1980; one 20m
2
along a
galle y o es edge be ween 19 May and 14 Oc obe ; one 40m
2
along a
galle y o es edge be ween 5 Sep embe and 14 Oc obe ) ha I isi ed
a h ee day in e als o coun he all o aeces. These ansec s did
no enable me o iden i y he dispe se s (bi ds o ,ba s), bu did p o e
o me ha seed ain was measu eable.
F om 20 Oc obe o 5 Decembe I
moni o ed one 24m
2
ansec (SNB) and wo 110m
2
ansec s (galle y
o es edges) wi h dawn and dusk ou s. By no ing he ime o
appea ance o aeces (day o nigh ) he dispe se s we e iden i ied as
bi ds o ba s.
-22.—
3.7.
SEED PREDATION AND GERMINATION
I measu ed he p eda ion a e on seeds o F. capensis
associa ed
wi h
ui pa s (simula ing allen ui s o unconsumed ui pa s) and on
seeds collec ed om ba eces. In each o 20 plas ic pe i dishes
ha ing ou 3mm access holes, I placed 20 aecal seeds and o 10 o
hese dishes I also added pieces o ui . The dishes we e se ou ,
al e na ing dishes o seeds wi h ui wi h dishes o seeds alone, a
ou me e in e als along ansec s in sa anna. A e 24 hou s he
numbe o seeds emaining in each dish we e coun ed. Be o e subsequen
uns he dishes we e washed in concen a ed acid (H2SO4) o emo e any
possible scen ails le du ing p e ious isi a ions. No wo
expe imen s we e un along he same ansec .
The ge mina ion a e and inal pe cen ge mina ion o seeds collec ed
om F. capensis ui s (wi h and wi hou he su ounding o a y wall),
om ejec a pelle s, and om eces we e measu ed. Fi y seeds we e
placed on damp clo h s ips in each o 10 co e ed pe i dishes o each
expe imen al g oup. The dishes we e examined and dampened daily and any
ge mina ed seeds we e coun ed and emo ed. Ea ly in he expe imen i
became e iden ha seeds s ill co e ed wi h he o a y wall we e poo
ge mina o s. To es o he p esence o inhibi o y chemicals, I placed
aecal seeds unde wo expe imen al condi i ions. In one g oup (10
dishes o 10 seeds) seeds we e placed on damp clo h s ips. In ano he
g oup (10 dishes o 10 seeds) seeds we e se on clo h pouches con aining
pul e ized o a y walls and lowe Ta s collec ed om ig syconia. In
his way he la e g oup we e in chemical
bu
no physical con ac wi h
he lowe pa s.
3.8.
FRUIT TREE ABUNDANCE AND FRUIT PRODUCTION
To assess he abundance o he s
,
h ub and ee species ha I knew o be
exploi ed by ba s in he galle y edge zone, I made comple e coun s o
all he indi iduals o "ba plan s" along ansec s o aling
3600 me es
o edge a i e si es. In addi ion I coun ed he numbe o F. capensis
-23-
ees o e 5 cm in diame e a g ound le el along h ee 20 me e wide
ansec s in he SNB, o aling a leng h o 800 me es and an a ea o 1.6
ha. F om his sample I la e es ima ed he o al F. capensis popula ion
in he 80 ha SNB.
Ficus capensis p o ed o be one o he species o ui s mos commonly
exploi ed by ba s. To gain some insigh in o he p oduc ion and
a ailabili y o F. capensis ui s in he SNB and galle y edge I made
coun s o ipe ( ed o o ange) ui s wice weekly in he la e mo ning
o ea ly a e noon. In his way I ollowed he ui p oduc ion on 54
ees in he SNB be ween 8 No embe 1979 and 5 No embe 1980 and on 46
ees along 1.5 km o galle y edge be ween 21 Ma ch and 5 No embe 1980.
3.9.
CONTROLLED FEEDING EXPERIMENTS
I kep h ee E. bue iko e i and eigh
M.
pusillus in a la ge ligh
cage (8x4x4m) o 17 and 65 days espec i ely du ing Which I measu ed
hei in ake o F. capensis ui s. Each indi idual was weighed be o e
and a e con inemen o moni o weigh gain o loss du ing he
expe imen al pe iod, bu was no handled be ween hese imes. Each
nigh I p o ided weighed quan i ies o ui s on an ad libi um basis
(usually 150 g o M. pusillus and 500 g o E. bue iko e i). A
a ious imes h ough he nigh and/o he ollowing mo ning he
emaining unea en ui and mas ica ed pelle s ( ejec a) we e emo ed
and weighed. This pe mi ed he calcula ion o in ake by he di e ence
. [INTAKE = weigh gi en - (unea en
4-
ejec a)]. Samples o ejec a
pelle s we e o en d ied o cons an weigh a 80°C o de e mine he
mois u e con en and a d y o we weigh con e sion ac o . The we
weigh s o ejec a pelle s colle.c ed be o e 0800h we e used di ec ly in
calcula ions since e apo a ion was shown o be negligible in he ca.
100% nigh - ime humidi y. Rejec a pelleLs collec ed a e 08001i we e
o en d ied and he we weigh was calcula ed using he con e sion
ac o .
P o ein and ene gy alues we e de e mined o he ui s o F. capensis
as well as o F. allis-choudae, F. o a a, Chlo opho a excelsa,
B idelia e uginea, Nauclea la i olia, and Vi ex deniana. Only hose
pa s o he ui s no mally inges ed wele analysed. Fo he Ficus
-24-
B. e uginea, and V. doniana he syconium o pe ica p was sepa a ed
om he seeds, bu o N. la i olia he seeds and pulp p o ed
insepa able and we e used oge he in he analyses. Fo C. excelsa, he
sponge-like ui s we e p essed in a lexible ine meshed sc een o
sepa a e he seeds and ib e om he luid. Only he luids we e used
in analyses.
Fo ni ogen de e mina ion, pulp o luids we e o en-d ied o cons an
empe a u e a 80°C. Fi e g am samples o he d ied ma e ial we e
analysed a he O ice de Reche ches Scien i ique e Technologique
Ou e-Me (ORSTOM, Adiopoudoum6, I o y Coas ) using a s anda d Kjeldahl
diges ion and colo ime ic echnique. The ni ogen alue so ob ained
was mul iplied by 6.25 o es ima e he o al p o ein con en (Lloyd e
al. 1978).
Soluble ca bohyd a es in he pulp o luids we e measu ed using a
empe a u e-compensa ed e ac ome e (Bellingham and S anley, Tunb idge
Wells, England). Ca bohyd a e measu ed in his way is exp essed as
g ams o suc ose equi alen s pe 100g o solu ion. Since monosacca ide
suga s wi h hal he calo ic alue o suc ose pe g am ha e hal he
e ac i e index he e ac ome e eadings can be ansla ed o
ene ge ic alues (Kcal) ega dless o he ac ual suga composi ion o
he ui . I will e m all eadings as suc ose equi alen s. The ene gy
con en o ui pulp was calcula ed by:
Kcal= [4.20][%1120 in pulp][measu ed % suc ose]
1- [measu ed % suc ose]
whe e 4.20 is he calo ic alue (Kcal) o one g am o suc ose.
Ene gy inges ed is only a ailable o me abolic use a e assimila ion.
To de e mine he e iciency o assimila ion o soluble ca bohyd a es by
ba s I ained wo
E.
bue iko e i o accep 1cm
3
pieces o sponge
soaked in 5, 10, and 15% suc ose solu ions. Each ba , a e eaching
wa e equilib ium (de ined as no ha ing u ina ed du ing he p e ious 30
minu es), was ed a pa icula suc ose solu ion un il sa ia ion. The
suc ose concen a ion o he eces/u ine p oduced du ing o wi hin 10
minu es o he expe imen was measu ed wi h he e ac ome e .
-25--
3.10.
STATISTICS
Pa ame ic s a is ics ollow Sokal and Rohl (1969) and non—pa ame ic
s a is ics ollow Seigal (1956). The ejec ion egion in all es s was
p‘ 0.05. Whe e e means a e ea ed s a is ically hey a e p esen ed as
mean ± s anda d de ia ion unless o he wise s a ed.
3.11.
NOMENCLATURE
Nomencla u e o he ba species ollows Rose ea (1965) and Be gmans
e al: (1974) and ha o plan s ollows Hu chinson and Dalzeil (1957).
The ba and plan species along wi h au ho i ies a e lis ed in Tables 1
and 12.
Hm
n=92
I
I
I
I
I
Mp
n=414
Eb
n=996
0.15-
0.05-
=
uJ
:".
); 0 15-
i-
I l
0.05-
LIJ
I-
<
Li
•=
LJ
0.02-
JFMAMJ JASONDJFMAMJJ ASOND
1979
1980
MONTH
FIGURE 9: The ca ch a es and sample sizes o E. bue iko e i (Eb),
M. pusillus (Mp), and H. mons osus (Hm) a LAMTO h ough
1979 and 1980.
0.005
J FM AMJ J ASONDJ FMAMJ J ASOND
1979
1980
-
MONTH
FIGURE 10: Ca ch a es and sample sizes o L. angolensis (La),
M. o qua a (M ), E. hel um (Eh), and N. eldkampi (N )
a LAMTO h ough 1979 and 1980.
Eg
n=223
•
Mp
•
n=230
•
•
0.15
0.05
N
n=13
M
n=12
0.06
0,02
•41-- —•— -1
-1--,--1
,
J FMA MJ JASON D
•
MONTH
FIGURE 11: Ca ch a es and sample sizes o E. gambianus (Eg),
M. pusillus (Mp), M. o qua a (M ), and N. eldkampi (N )
a Wango Fi ini.
-31-
On his basis, he LAMTO communi y was comp ised o se en co e species;
E. bue iko e i, E. hel um
s
H. mons osus, L. angolensis, M. pusillus,
M. o qua a, and N. eldkampi. I is p ima ily wi h hese species ha
I conce n mysel in his s udy.
The Wango Fi ini co e communi y consis ed o ou species; E. gambianus,
M. pusillus, N. eldkampi, and M. o qua a. E. hel um was p esen in
he egion seasonally (see sec ion 4.1.2), bu was ne e cap u ed a
Wango Fi ini.
4.1.1 SEASONAL COMMUNITY COMPOSITION: RESIDENTS ND MIGRANTS
Figu es 9, 10, and 11 p esen he mon hly ca ch a es o he co e
species a bo h LAMTO and Wango Fi ini. A bo h si es all species
showed conside able luc ua ions, al hough M. o qua a, E. hel um, N.
eldkampi, and L. angolensis a ied mos , cycling om absence o
ela i e abundance seasonally. The luc ua ions a ose om wo
p incipal ac o s: 1) seasonal bi h pe iods and he subsequen
ec ui men o imma u e indi iduals in o he lying popula ion (see
sec ion 4.2), and 2) clea ly de ined seasonal mig a o y mo emen s. The
coincidence o he ec ui men pe iods o he species p esen
h oughou he yea wi h he mo emen s o he mig a o y species in o and
h ough each communi y, howe e , ends o mask he in e -speci ic
di e ences.
Figu e 12 p esen s he species a LAMTO and Wango Fi ini in e ms o
hei ep esen a ion in ne samples and hei ecap u e a es. The
species in each communi y clus e in o wo g oups: 1) hose p esen in
o e 80% o sampling pe iods and ha ing high ecap u e a es, and
2) hose p esen in ewe sampling pe iods and ha ing low ecap u e
a es. Assuming ha indi iduals o species ha emain in he a ea
h oughou he yea a e a ailable o ecap u e, while indi iduals o
species ha mo e seasonally h ough he communi ies a e no a ailable
o ecap u e, hese wo g oups co espond wi h he esiden and miglan
species espec i ely.
'100
80
60
cj
40
20
UJ
CC
(.4
F-
Z
100
LL
80
60
4
20
10 12 14 16 18
50 52
10
12
A
% RECAPTURED
FIGURE 12: The p opo ion o ma ked indi iduals ha we e ecap u ed
a some ime plo ed agains he pe cen o sampling mon hs
when each species was caugh a (A) LAMTO and (B) Wango
Fi ini. Residen species had high ecap u e a es and we e
caugh in o e 80% o sampling pe iods, while mig an
species we e ne e ecap u ed and we e caugh in ewe han
80% o sampling pe iods.Ba species designa ions a e as in
Table 1.
•
0.08 -
0.04
A.
•
J FMAMJ J AS OND
B.
.
•
J FMAMJ J ASO ND J FMAMJ J A SON D
1979
MONTH
1980
C.
0/0
100]
FIGURE 13: The ca ch a e o M. o qua a a (A) Wango Fi ini and
(B) LAMTO du ing 1979 and 1980. Fo LAMTO, he uppe line
ep esen s he o al ca ch a e and he lowe s ippled line
ep esen s he ca ch a e o emales only. (C) shows he
p opo ion o emales (s ippled) in he ca ch a he Tai
Fo es si e in No embe .
•m,
T77'
Thus a LAMTO E. bue ko e i, H. mons osus, M. pusilus, and (possibly)
L.
angolensis, and a Wango Fi ini E. gambianus and M. pusilus a e
esiden while bo h communi ies sha e M. o qua a, E. hel um (a leas .
egionally),and N. eldkampi as mig a o y species. The seasonal
pa e ns o hese la e h ee species wa an u he elabo a ion.
4.1.2. MIGRATORY SPECIES
Myonyc e is o qua a
MOVEMENTS: Figu e 13 and Table 2 show a clea ly bimodal dis ibu ion o
M.
o qua a annually a LAMTO wi h co esponding absences and peaks in
he ca ch a e a Wango Fi ini. M. o qua a was absen o g ea ly
educed in numbe s a bo h si es be ween Decembe and Ap il,
co esponding wi h he majo d y season. Associa ed wi h he no hwa d
mo emen o he in e - opical on and he onse o he ains a
LAMTO, M. o qua a i s appea ed in ne ing samples a his si e in
Ap il (1979) and
May
(1980) and became he mos nume ically abundan
species in June (1979) and May (1980). Nei he du ing his annual peak
no be ween yea s we e any indi iduals e e ecap u ed. A Wango Fi ini,
M. o qua a was i s aken in May (1980), only 12 days a e he i s
indi idual was aken o he sou h. Associa ed wi h he passage o he
in e - opical on o i s mos no he ly posi ion in Augus , ca ch
a es declined a bo h si es and subsequen ly inc eased in Sep embe o
No embe . I is impossible o de e mine when he second peak occu ed'
In he no h since no ne ing was ca ied ou be ween Augus and
No embe .
Du ing No embe , when ca ch a es we e declining a LAMTO, M. o qua a
was he mos abundan species a he Tai Fo es si e, comp ising 53.2%
o he cap u es (n=111).
SEX RATIO: A bo h LAMTO and Wango Fi ini males domina ed in all mon hly
samples. The a ia ion in he p opo ion o males in samples om he
wo peaks and be ween he si es indica ed ha he e we e sexual.
di e ences in he mo emen s. Adul and imma u e enales oge he
comp ised an a e age o 24.6% (n=134) and 23.7% ( e
.
--59) o he i s
peaks in 1979 and 1980 espec i ely. The maximum p opo ion o emales
-33-
occu ed in May 1979 (33.3%, n=24) and in June 1980 (27.8%, n
=
18). In
all mon hs he sex a io depa ed signi ican ly om uni y (p<0.05).
Du ing he i s peak, he ca ch a Wango Fi ini was en i ely o males
(n=10) and a bo h si es only males we e aken du ing he second annual
peak (LAMT0=107; Wango Ti ini=2).
While be ween Sep embe and Decembe only males we e cap u ed a sa anna
si es, a he Tai Fo es si e in No embe he si ua ion was e e sed.
Females cons i u ed 64.4% o he sample (n=59), ep esen ing a
signi ican ly unequal sex a io (X
2
=4.9, d =1, p<0.05).
Eidolon hel um
MOVEMENTS: Obse a ions o he sequen ial build-up and decline o
E. hel um colonies a Abidjan (5'12' ;4°00'W), LAMTO (6°13'N;5°02'W),
Fe kessedougou (9°30'N;4°30'W), and San, Mali (13°15'N;5°00'W)
conside ed oge he wi h he ne ing samples om LAMTO showed ha his
species exhibi ed a pa e n o mo emen s ela i e o he ains, simila
o ha o M. o qua a (Figu e 14).
Th ough he majo d y season (Oc obe o Ma ch) E. hel um oos ed in a
leas one la ge colony in he o es zone, loca ed in he cen al pa k
o Abidjan (Pa c du Pla eau). Beginning in Oc obe , he Abidjan colony
inc eased om se e al hund ed indi iduals o an es ima ed peak o
200,000-500,000 indi iduals in Janua y/Feb ua y when emales ga e
bi h. In Ma ch, his colony declined apidly un il i numbe ed only a
ew housands and he ea e declined slowly o a low o se e al
hund eds in July/Augus .
Beginning in la e Feb ua y (19. Feb ua y 1979, 29 Feb ua y 1980), I i s
obse ed E. hel um oos ing in small widely sca e ed g oups o 8-15
Indi iduals in he open sa anna Bo assus palms (B. ae hiopum) a LIMO.
Wi hin wo weeks o hese i s sigh ings, an es ima ed 100,000
indi iduals a i ed o e nigh (5 Ma ch 1979, 12 Ma ch 1980) and o med a
consolida ed colony in he Bo assus palms, co e ing an a ea o
app oxima ely 50 ha. In 1979 he colony a i ed one mon h p io o he
i s signi ican ains o he yea and in 1980 du ing he i s ainy
13°15'
0.3
z z
..111
n
1.0
0
9°30'
3
z,11:211
100
5°12'
J FMAMJJ ASOND
FIGURE 14: A schema ic diag am o showing he app oxima e iming o
a i al and depa u e o E. hel um colonies a Abidjan
(5°12'), LAMTO (6°13'), Fe kessedougou (9°50') and San
(13°15'). Es ima ed colony sizes a e gi en in housands.
XXXX indica es he mon hs wi h he highes ca ch a es o
E. hel um a LAMTO (see also Figu e 10).
-34-
mon h. The p esence o E. hel um in he a ea was no e lec ed in he
ne ing samples. While indi iduals could be seen o aging in he canopy
o he galle y o es s, hese did no descend o ne heigh s. The
colony emained a LAMTO o 45 days (1979) and 37 days (1980) and
depa ed as a g oup in a single nigh ; 20 Ap il 1979 and 19
Ap il
1980.
Th ee days and eigh days p io o he de ini i e mo emen o E. hel um
om he egion, he colony was ex emely agi a ed du ing he day and
o e a pe iod o se e al hou s shi ed he oos o a new si e 1.5km
dis an . Based on he numbe s o aeces ound unde he new oos
daily, i appea ed ha o aging success was educed du ing hese las
days, al hough his was impossible o quan i y.
I is no ewo hy ha no E. hel um we e e e seen oos ing a LAMTO
be ween 1960 and 1978, no in 1981 (R. Vua oux, pe s. comm.).
A e he depa u e o E. hel um om LAMTO, I loca ed ano he colony
a Fe kessedougou (la i ude 9°30'N), nea ly 380km o he no h. I
es ima ed his colony o numbe be ween 2,000 and 5,000 indi iduals on
25 Ap il 1980. I could no de e mine when his colony a i ed, bu he
illage s indica ed ha i had appea ed " ecen ly". The colony
emained un il a leas 31 May 1980, bu was no p esen be ween 26 July
and 21 No embe 1980.
On 12 July I also loca ed a colony numbe ing no mo e han se e al
hund eds a San, Mali (la i ude 13°15'). I ha e no indica ion when his
colony a i ed o depa ed.
Following he p og ession o colonies o he no h wi h some eaching a
leas la i ude 13°15'N by July, he e was an inc ease in E. hel um in
he ne ing samples a LAMTO, peaking in Sep embe 1979 and Oc obe
1980 (see Figu e 10). Du ing - his pe iod I could loca e no colonies
ei he in he icini y o Wango Fi ini and Fe kessedougou o nea LAMTO
[no we e any E. hel um o sale in he local ma ke s, which is a good
Indica ion o he absence o colonies om he egion]. The peak a
LAMTO and subsequen decline o a minimum in Decembe coincided wi h he
build-up o he Abidjan colony.
-39-
igge ed by ain all. In 1979 he i s bi h pe iod began and he
peak lac a ion mon h occu ed p io o any measu able inc ease In
ain all o e he p e ious wo mon hs, bu in 1980 his coincided wi h
an inc emen in ain all o e he Janua y low. In bo h yea s he second
bi h pe iod began du ing he
mon h
wi h he lowes ain all o he
mino ainy season (Augus ).
Bi h and lac a ion pe iods we e highly synch onized wi hin each
species. No emales o ei he E. bue iko e i (n=71) o M. pusillus
(n=49) caugh be ween June and July o Decembe and Feb ua y we e e e
ound o be lac a ing.
The exac du a ion o he lac a ion pe iods o indi iduals is
impossible o de e mine om he ield da a, bu he maximum possible
leng h can be es ima ed by he sp ead o lac a ion (ea lies o la es
da es o lac a ing emales) o he en i e popula ion. No single
emale could lac a e longe . Fo he i s lac a ion pe iod
E. bue iko e i and M. pusillus had sp eads o 59 and 58 days
espec i ely and o he second lac a ion pe iod sp eads o 93 and 79
days espec i ely (Table 4). In bo h cases he second lac a ion pe iod
had a g ea e sp ead han he i s . This could be due ei he o a eal
di e ence in he leng h o indi idual lac a ion pe iods be ween he
seasons o me ely o less synch onous bi hs.
Du ing he i s lac a ion pe iod one E. bue iko e i and wo
M. pusillus we e cap u ed ca ying newly bo n young (weigh s:
E. bue iko e i =20g; M. pusillus =4g). These indi iduals we e
ecap u ed s ill lac a ing 42 days la e (E. bue iko e i) and 39 and 43
days la e (M. pusillus), gi ing a minimum du a ion o lac a ion
be ween 39 and 43 days o bo h species. These da a indica e ha i s
lac a ion pe iodslas be ween.se en and eigh weeks and ha he second
lac a ion pe iods may las up o 13 weeks.
4.2.2. OESTRUS
Ea ly in he lac a ion pe iods emales we e a ely palpably p eguan ,
bu a e Ma ch and Sep embe o E. bue iko e i and Ap il and
-40-
mid-Oc obe o M. pusillus, all lac a ing emales we e also palpably
p egnan (E. bue iko e i =21; IC pusillus =7). One lac a ing emale
E. bue iko e i killed on 15 Ap il 1980 had an implan ed emb yo wi h a
cho ion measu ing 4mm in diame e .
Mos emales o p o en ep oduc i e capaci y (i.e. pa ous) we e ei he
p egnan o lac a ing when handled, indica ing ha ew ailed o
ep oduce du ing a gi en pe iod. O e all 11.4% o pa ous
E. bue iko e i (n=166) amd 8.8% o pa ous M. pusillus (n=114) we e non-
e
p
oduc i e. While ea ly p egnancies may ha e been o e looked
(o e -es ima ing he equency o ep oduc i e ailu es), all p egnan
emales should he e been de ec able by palpa ion be ween June and Augus
and Decembe and Feb ua y ( he second hal o he ges a ion pe iods).
This enables a compa ison o he ailu e a e o emales ca ying
p egnancies h ough he we and d y seasons (June o Augus and Decembe
o Feb ua y espec i ely). Fo nei he species was he e a signi ican
di e ence in he ailu e a e be ween seasons (E. bue iko e i: n=88,
X
2
=1.25, d =1, p 0.05; M. pusillus: n=51, X
2
=1.70, d =1,p 0.05), no was
he e a signi ican di e ence be ween he species in he p opo ion o
emales ailing o ep oduce (X
2
=0.21, d =1, p 0.05). The majo d y
season appa en ly had no e ec on he p obabili y o emales ini ia ing
o main aining p egnancies.
Six pa ous E. bue iko e i and 10 pa ous M. pusillus we e ecap u ed
du ing wo o h ee successi e b eeding pe iods. All emales we e
ei he lac a ing o p egnan du ing each pe iod.
4.2.3. MATING PERIODS
The seasonal pa u i ion coupled wi h he e iden pos -pa um oes us
indica e ha ma ing mus be seasonal as well, occu ing ea ly in he
obse ed lac a ion pe iods (Feb ua y o Ap il and Augus o Oc obe ).
Howe e , adul male E. bue iko e i and M. pusillus ecap u ed be ween
successi e ma ing and non-ma ing seasons showed no signi ican a ia ion
in es es leng h be ween he seasons (E. bue iko e i: xma ing=7.6mm,
x
non-ma ing
=
7.6mm, =0.04, d =66, p 0.05; M. pusillus: xma ing=6.0mm,
xnon-ma ing
=
5
.6mm, =1.56, d =34, p 0.05). Only ecap u ed males known
•
•
6'6
6'ö
6
co
°
CNI
0
CO
CO•d•
I
I
co
'6'
Cd
•
0
=
-
z
0
•
c )
•
•
-2
4.)
C
1
/
4
1
0
Cd
4-1
C.)
44
1-1
O
1
4-1
•
.O
• -1
O
0 CN1
C.)
b0
•
n-1
e-1/4 CO 1/40
0.)
14
W
X
u
N
c•i
C )
ON
▪
•--I
•
-
u
Cd
Cd
•
•
o
•
b8
4
0
0 •0
C.)
•
-4
%.0
•
.01
•••
•:
•
%0
44
O
CO
X
CO
O
a) CN
-
•
C
▪
d
b0 130
0
II
4•1
•
•-4
>4
4 4
b0 bo • •
•
4
PC)
a)
a)
:3:3
W
co
%.0
CU
.1-1
1-4
C )
1/40
CO
0.)
bo
b0 0
>4
•
•-I
c
1.4
CO
C•1
In•
W
•
•
a)
II
4-1
>4
bo
O
44 ••4
.
g
co
••
O
I-1
a)
4
a)
44
co
cd
(
6
) 11-1013M A008
•
•
gee
0
4
'
01'
JD
•
1-1-1
. •
..
•
•°
•
111
• b .•
•
a
I
*
.
0
(1)
..-1
4-1
• 4
$.4
co
co
•
CO 0 0 a)
O
4
0
•:
•
4-I
0
'0
•,-i
$4
C.)
Cl)
•--I
O
4-)
•
cj
/1
4-1
Cl)
•
0 4-1 •
0
n
-•
n
4-1 4J
—)
Cl)
I-1
0
/-1
0
I-1
ca
0)4-4
3-
4-1
..
cn
--)
CO oN
co
0
1--•
Cl)
0
▪
m 0 -H
Cl)
—i
4-1
4-1
4-1
C'1-1
C1
2
.1.)
C
0
a) CN1
•
c
O
En Gil
4-1
a)
•
H
;-1
Hi
4
$-1
o
cO
•
oi
•
El
1- )
1
4-1
0
Cl)
0
O
c) 4-1
..-
0
2
. .1 e•-
n
a)
V %.0
4-)
4
$4
P.
. ..'
N../
0
•
a)
4-1
4-1
ON
O
-
ca
al
--
I
-
LL
,0 cn 0
N-
04-
-F
•
o
•
N
- -1
il
—I
al >4
CO
C)
CA
U)
.-
$4
4-1
CNI
—.)
-
.--I
CO
4-1
•
Fl 4-1
4
-
1
71
C
•
CO
0
=
al
I—
a)
$-1
b0 H
9-1
0 "0 •-1
O
4 0 Cl)
='4-40 CO
• •
O
c)
Cll Ci
•P
CO
.
**
-1
1:4
al
••
z
a)
<4
0 3 --
-i
9-1
.--1
a
4-I
-I
•
FI
4-1
al
,--1
VD
0
0
4.)
al
L )
O
CO
0
0 CO
^0 +
0
•
H
1--I
0
•
H
al
ca
›C
b0 co
cn
co
a)
bi)
C
n
1
(/)
4-) 4-1
$-1 0
8
4 4
ozl
.--1
CO b1:1 0 -4
•
T-1
4-1
0"
H
U)
a)
co
0 >4
•<
3 3
al
0
.0
• •
>-,
a)
a)
4-)
PO
'0 >,
P.
o
•
H
0
co
••
.1:
21
co
CO
,--I
—)
CO
.
cu c )
4-1a)
4-
1
Hi
O 1-4
NI al
co
Ln
O
0 0
a)
--)
o
1.4
bo ,---1
+
4-1
la.
a)
al
.i.)
3-1
El
X
•
a)
ca.
a) ..o
4-1
4
4-1 CN1
W
4-1
4-1
•
4
-
i
0 a) 0
Cl) boa) 14
•
H
0 co 0 H
-al
•
-1
CU
4-)
>-1
4
3 $
4
0
a)
0
(
e
..
4 4 Cl) 64
E-
1
in
14
.1-1
2
O ' 6
O
6
06
O
7
1
pc
sj
2
co
(
D
)
11-1913M 1000
701.
Z01.
001.
96
96
76
Z6
06 =
99
LD
99
w
06
99
99
79
Ze
09
2Z.
24
6
1:=3,
co
si annioNi
•()N
FIGURE 19: The dis ibu ion o o ea m leng hs in samples o male and
emale E. bue iko e i cap u ed a LAMTO du ing 1979 and
1980. The dip in he his og am a 96-98mm o ea m leng h
o males co esponds wi h he size a pube y.
•
/Mb
NNO
CD
-
CD
-
-C)
•
•
(ID
CD
.111M
11.
•
•
•
•
•
•
•
••
0
X•
•,..:••
•ou
•
.....
•
e
~
o
,,
o
•1:66.:.
e.%*:
••
—1
.14
Iwo
•
•
•
n
•••
NI
i
I
I
I
1
O
CD
CD
CD
CD
0
0
CT)
00
CD
CT)
00
(
ww
)
HiON31 W8V380d
FIGURE 20: The eg ession o o ea m leng h on weigh o male and
emale E. bue iko e i. Reg ession lines we e calcula ed
using indi iduals weighing less han 160g o males and
110g o emales.
-41-
o be adul we e used in o de o exclude pube al males-o nea adul
size om he analyses.
4.2.4. MALE CALLING BEHAVIOUR
Male epomopho ine ba s o he gene a Epomopho us, Epomops, Hypsigna hus,
Mic op e opus, and possibly Nanonyc e is signal o emales by means o
conspicuous ocal, isual, and ol ac o y displays. Excep ing
H. mons osus, males o all species ha e glandula in agina ions on he
shoulde s (epaule es) which a e endowed wi h u s o whi e hai .
Males display by emi ing sho loud calls a app oxima ely one second
in e als, simul aneously lapping he hal -opened wings and e e ing
he epaule es. Fo desc ip ions o he displays and hei unc ional
signi icance see Wickle and Seib (1976) and B adbu y (1976). A
de ailed s udy o male calling beha iou s was beyond he scope o his
s udy, bu I will p esen some da a ele an o la e sec ions.
E. bue iko e i: Adul males o his species main ained adjacen bu
loosely packed e i o ies (es ima ed in e -indi idual dis ances we e
30-40m) sca e ed h ough he i e ine o es along he Bandama Ri e
and a ely in he inland galle y o es s. Males displayed a hese
si es o up o eigh hou s on mos nigh s and hey main ained hese
e i o ies h oughou he yea . Six e i o ies, on which 1 moni c ed
occupancy (a male p esen and calling) in each mon h o he s udy, we e
occupied a some ime du ing all mon hs.
O e 101 nigh s be ween Janua y and No embe 1980,1 eco ded whe he o
•
no males called, how many we e calling ( ull, hal , none), and wha
ime he calling began. I also no ed he moon condi ion as "b igh "
when he moon was abo e he ho izon be ween 1900h and 2400h and hal
phase o g ea e , and as "da k" when he moon was below he ho izon a
hese imes, less han hal phase, o co e ed by hea y cloud. Males
showed clea di e ences in hei esponse o moon condi ion be ween he
ma ing and non-ma ing seasons. Du ing Feb ua y/Ma ch and Augus , all
males called on all nigh s ega dless o he moon condi ion. The e was
no signi ican di e ence be ween he s a ing imes on "b igh " and
"da k" nigh s ( =0.79, d =53, p>0.05). The mean s a ing ime was
1942h.
-42-
In con as , du ing he non-ma ing season calling was s ongly a ec ed
by he moon condi ion. On only 7% o "b igh " nigh s (n=30) we e all
males calling and on 33% o he nigh s no males called a all. On 75%
o "da k" nigh s (n=16) all males called, and on he emaining 25% hal
he males we e judged o be calling. S a ing imes on "b igh " nigh s
when males did call (X=2204h, n=20) we e signi ican ly la e han on
"da k" nigh s ( =6.21, d =34, p<0.05) when males began calling a 2011h.
M. pusillus: I hea d males o M. pusillus call only b ie ly du ing Ap il
1979 and ne e he ea e a LAMTO, al hough I did hea males call a
si es. o
he no h and I could ecognize his species' call.
4.2.5. RECRUITMENT AND GROWTH: E. EUETTIKOFERI
Imma u e ba s i s appea ed in he lying popula ion in May/June and
No embe /Decembe o coho s 1 and 2 each yea . Male and emale
coho s could be ollowed as hey g ew o adul p opo ions (Figu es 17
and 18), bu became di icul o dis inguish om olde adul s abo e he
weigh s o 160g o males and 110g o emales. In he ollowing
analyses hese weigh s a e used as he uppe limi s o disce nable
coho s (see sec ion 3.2.).
The dis ibu ion o o ea m leng hs in he o al cap u es and he
eg ession o o ea m leng h on weigh a e p esen ed in Figu es 19
and 20. Bo h males and emales i s lew and we e cap u ed a o ea m
leng hs o 74-76mm and weigh s o 45-65g. I is impossible o de e mine
. he exac age o ecen ly weaned indi iduals caugh in May/June and
No embe /Decembe ; howe e , assuming ha a e age indi iduals we e bo n
a he s a o he peak lac a ioa mon h (1 Ma ch, 1 Sep embe ) wi h a
mean weigh equal o ha o newly-bo n ju eniles ca ied by cap u ea
emales (R=23.0g, n=4), a ough es ima e o he a e age weigh inc eman
pe day h ough he lac a ion pe iod can be made. Fi e indi iduals,
caugh weighing unde 65g ( ange =47-64g) and es ima ed o ange be ween
42 and 83 days o age, we e calcula ed o ha e gained an a e age o
0.64g/day ( ange=0.32-0.88g/day).
• -43-
ANALYSIS OF
COVARIANCE
COMPARISON
COHORT
d
F a io
P obabili y
Imm o s Imm ?
1,
1979
IA112
0.48
ns
Imm c s Imm
2,
1979
1&121
0.42
ns.
Lmn
d' s imm
1,
1980
1&129
0.53 ns
Lmn o (coho 1,
s
1979)
1&179
2.12
ns
Imm o (coho 2,
Imm e(coho 1,
s
1980)
1979)
1&206
1.39
ns
Imm
e(coho 1,
Imm 7.
(coho 1,
s
1980)
1979)
104
0.35
ns
Imm
(coho 2,
Imm
g
(coho 1,
s
1979)
1979)
1&52
0.02 ns
Imm
(coho 1,
1980)
TABLE 5: Compa isons o
he g ow h a es (slopes o he eg ession
lines) o male and emale coho s o E. bue iko e i a
LAMTO. ns=no signi ican a
p.10.05.
--44—
A.
COHORT
SEX‘'
GROWTH RATE
(g/day)
95% CL
uppe
lowe
1,
1979
0.292 0.314
0.270
1,
1979
0.260
0.319
0.201
2,
1979
0.266
0.294 0.237
2,
1979
0.232
0.313
0.151
1,
1980
e
0.318 0.356 0.279
1,
1980
0.288 0.386 0.190
All coho s x=0.276
B.
95% CL
COHORT
SEX
ASYMPTOTE (g)
K
uppe
lowe
1,
1979
190
0.0072
0.0088 0.0056
1,
1979 120
0.0128 0.0200 0.0056
2,
1979
190
0.0068
0.0084
0.0052
.2,
1979 120
0.0108 0.0172 0.0036
1,
1980
190
0.0076 0.0094
0.0060
1,
1980
.
120
0.0116
0.0156 0.0675
Table 6:
(A) G ow h a es o male and
emale coho s o
.
E. bue iko e i
calcula ed om i ed eg ession lines (see
ex ) and (B) logis ic
g ow h cons an s
calcula ed acco ding
o Rickle s (1967).
Bo h g ow h a es and K
a e p esen ed
wi h hei uppe and lowe 95%
con idence limi s.
-48--
A.
•
GROWTH
COHORT
EQUATION
N
R2
GROWTH RATE (g/day)
dl
1,
1979
log10Y=0.3161ogi0X+0.701
14
0.64
0.123
g:
1,
1979
logl0Y=0.238logioX+0.872
8
0.39
0.152
o :
2,
1979
log
ia=0.30510
g
10X+0.741
18
0.49
0.069
g:
2,1979
log10Y=0.420logioX+0.510
15
0.43
0.097
e:
1,
1980
log
10
Y=0.487log10X+0.428
17
0.78
0.139
g:
1,
1980
INSUFFICIENT DATA
B.
95% CL
COHORT
ASYMPTOTE (g)
uppe
lowe
1,
1979 32
0.010
0.028
0.000
2,
1979 32
0.014 0.021 0.007
TABLE 7:
(A) G ow h equa ions and g ow h a es o male and emale
coho s o
M. pusillus
a LAMTO and (B) logis ic g ow h.
cu e cons an s (K) calcula ed acco ding o Rickle _s
(1967). G ow h a es we e calcula ed as a e age a es o e
he 20-25g weigh in e al.
-49--
ANALYSIS OF COVARIANCE
COMPARISON
COHORT
d
F a io
P obabili y
Imm o s Imm
1, 1979
1&18
0.37
ns-
Imm ?' s Imm
7
2, 1979
1&29
0.46
ns
Imm ?(coho 1, 1979)
s
1&27
3.22
ns
Imm e(coho 2, 1980)
Imm e(coho 1, 1979)
s
1&28
0.01
ns
Imm e(coho 1, 1980)
Imm
(coho 1, 1979)
s
1&19
0.78
ns
Imm ? (coho 2, 1979)
TABLE 8: Compa isons o he g ow h a es (slopes o he eg ession
lines) o male and emale coho s o
M. pusillus
a LAMTO.
ns= no signi ican a p,5 0.05.
-50-
Male coho s, as wi h emales, eached app oxima ely adul weigh s by
he age o six mon hs (Figu e 21). Males unde six mon hs had es es
unde 3mm in leng h and unde eloped epaule es. O 13 males ecap u ed
a ages o six o se en mon hs, 31% we e beginning es icula
de elopmen ( es es=3-4mm) and in agina ion o he epaule es. All o
ou males caugh a eigh mon hs had pa ially de eloped es es
(31=4.5mm) and 100% o 12 males ecap u ed a nine o 12 mon hs o age
we e indis inguishable om olde adul s. Fo he pu poses o
popula ion analyses I assume ha males became adul a nine mon hs o
age.
4.3.
REPRODUCTION OF MIGRANT SPECIES
Nei he E. hel um no M. o qua a ep oduced a LAMTO; howe e , cap u es
and obse a ions he e as well as a Abidjan and he Tai Fo es si e
p o ided some insigh in o hei ep oduc ion. The da a o
N. eldkampi we e inconclusi e.
E. hel um: Dead new-bo n ju eniles we e ound unde he Abidjan oos
in Feb ua y 1979 ( he i s obse a ion a emp ed) and in Janua y and
Feb ua y 1980.
When he colony a i ed a LAMTO in Ma ch o bo h yea s, a la ge
p opo ion o he emales ca ied ju eniles. Du ing he i s wo
weeks, he ju eniles emained a he oos si e h oughou he nigh and
-could be hea d calling a i s , hen la e seen lying be ween oos
ees. Towa ds he end o Ma ch and in ea ly Ap il he numbe s o
ju eniles seen a he oos a nigh declined un il by he las week
none emained. A his ime he ju eniles we e appa en ly su icienily
old o o age wi h he adul s. When he colony mo ed o he new si e
1.5km dis an , I could see no emales ca ying young.
Du ing he e u n mig a ion (No embe o Janua y) I caugh . nine pa ous
emales o which 55.6% we e p egnan wi h la ge oe uses and one (caugh
In Janua y) was lac a ing.
-51-
H. o qua a: A LAMTO emale M. o qua a we e only caugh du ing he
no hwa d mig a ion (May o July) mo emen and ne e du ing he e u n.
Mos o he pa ous emales (86.5%, n=37) we e p egnan , ca ying small
palpable oe uses ha , on he basis o equi alen sizes wi h
E. bue iko e i and M. pusillus, indica ed an Augus /Sep embe bi h
pe iod. In No embe in he Tai Fo es , 100% (n=26) o he adul emales
had small palpable oe uses, sugges ing ano he bi h pe iod in
Feb ua y/Ma ch.
Du ing he i s annual peaks a LAMTO, 24.8% o males (n=137) did no
ha e he h oa u o long dense hai s cha ac e is ic o adul males,
bu du ing he second annual peaks only 2.9% o males (n=107) lacked
he u . Body weigh s o males wi h u s caugh du ing he i s
peaks did no di e signi ican ly om hose o males caugh du ing he
second peaks ( =0.07, d =62, p>0.05); howe e , es es leng hs and h oa
gland sec e ions o hese males did di e be ween he wo imes o
yea . Males wi h de eloped h oa u s had signi ican ly smalle
es es (5=5.6mm, n=36) han hose caugh in Sep embe o Decembe
samples (R=6.9mm, n=24; =3.57, d =58, 1)40.05). The u s o males
caugh be ween Ap il and July we e clean, d y, and g ey, while hose o
males caugh in Oc obe o Decembe we e co e ed wi h a yellow oily
exuda e.
N. eldkampi: A LAMTO he e we e insu icien cap u es o N. eldkampi
o de ine any clea ep oduc i e pa e n. Du ing he i s ca ch a e
peak wo pa ous, non— ep oduc i e emales we e caugh . Du ing he
second ca ch a e peak, h ee emales we e p egnan , wo we e lac a ing
(Oc obe ), and one was pa ous, non— ep oduc i e.
In he Tai Fo es in No embe , ou adul emales we e cap u ed. O
hese, h ee ca ied small palpable emb yos and had ecen ly been
lac a ing, and one was bo h p egnan and lac a ing.
The combined da a o LAMTO and he Tai Fo es sugges a bi h pe iod
occu ing du ing Oc obe and No embe .
5
.•
••••
J'F
'
M'A
M1J
12
77;
Eg n=20
11
0 N
ID
Mp n=42
20
100-
A'S
m
N n=5
100-
PREGNANT
LACTATING
FIGURE 24: The iming o p egnancies and lac a ion o E. gambianus
(Eg), M. pusillus (Mp), and N. eldkampi (N ) a Wango
Fi ini. Figu es abo e each his og am a e he sample sizes
o each sampling pe iod.
-52--
4.4.
REPRODUCTION AT WANGO FITINI
Samples o pa ous emale E. gambianus amd M. pusillus om Ap il, May,
July, Augus , and No embe we e small (20 E. gambianus, 42 M. pusillus),
bu p o ided an indica ion o he iming o ep oduc ion.
The e we e wo lac a ion pe iods o bo h species, one occu ing in
Ap il and May and he o he a ound No embe (Figu e 24). None o he
lac a ing emales caugh in Ap il (4 E. gambianus, 8 M. pusillus) we e
de ec ably p egnan , indica ing ha his mon h ell nea he beginning
o he lac a ion pe iod, be o e oe uses we e la ge. All o he pa ous
p egnan emales caugh be ween 27 and 30 May (1 E. gambianus, 2
M. pusillus) had been lac a ing ecen ly and ca ied small palpable
emb yos. In No embe all pa ous emales (11 E. gambianus, 20
M. pusillus) we e ei he lac a ing o had ecen ly inished and we e
palpably p egnan . These da a indica e ha Ma ch o Ap il and Oc obe
o No embe we e he peak lac a ion mon hs a Wango Fi ini.
Two lac a ing N. eldkampi we e caugh in Ap il and May, coinciding wi h
he lac a ion pe iods o he abo e species.
-53-
4.5.
ECOLOGY OF THE SPECIES AT LAMTO
4.5.1. ROOST SITES
E. bue iko e i: I loca ed 14 di e en oos si es, i e used by
adio- agged indi iduals which I moni o ed o a o al o 82 oos days
(2 imma u e males o 1 day each, 1 adul male o 29 days. l'adul male
o 6 days, 1 lac a ing emale o 45 days) and nine which I loca ed on
walks.
Wi h he excep ion o he lac a ing emale, all indi iduals oos ed in
he ipa ian o es along he Bandama Ri e , ei he on islands (3 o 4
adio- agged males) o on he main bank (1 adio- agged male, nine
uniden i ied indi iduals). All indi iduals oos ed alone in dense
oliage a heigh s o 2-5m and 11 o he 13 we e in ees o e hanging
and i ually enclosing side channels o he i e . The wo indi iduals
no oos ing di ec ly o e he wa e we e wi hin one o ou me es o
he edge. Each adio- agged adul male used he same oos each con ac
day, one o 29 consecu i e days.
E. bue iko e i appea ed o be ex emely sensi i e o dis u bance. None
o he nine indi iduals ha I igh ened om hei oos s e u ned on
subsequen days.
The single lac a ing emale occupied he same oos wi h he cu en
o sp ing o e 45 consecu i e days. This si e was loca ed 15m abo e he
g ound in a b oad-lea ed, dense canopied ee (Cola gigan ea), along he
edge o a galle y o es 2.5km inland om he Bandama Ri e .
The local Baoul6 people indica ed ha oos s o E. bue iko e i we e
mos commonly loca ed along he Eandama Ri e and ha his species
usually oos ed alone.
H. mons osus: I loca ed i e di e en oos s, ou di ec ly on he
edge o galle y o es s and one in he cen e o an isola ed o es
pa ch. All had a common o m; ha o an open-Q ded enclo
q
u e, 4-5m in
La I n=108 xx
M n=235
)*
CD
0 0 0
0
0 0 0
L()
0) , C )
0
.---
CNI (N
TIME PERIOD
0
0
CD
CD
15
100
EID
E b "
n=515
n=219
5
o-
Mp n=284
Hm n=58
150
E3 1 0 0
LU
Li-I,_
E--
L_)
<C
so
LU
Li
cc
0_
x
I— Li-
200
Li
CD
—
150
100
50
c)
0
0
0
04
0
0
0
al
CN CV
0
FIGURE 25: Ac i i y pe iods o i e p e opodid species a LAMTO.
Ac i i y was measu ed by ne cap u es du ing wo hou
pe iods h ough he nigh and is exp essed as a pe cen
o he cap u es ha would be expec ed, gi en an e en
dis ibu ion o ac i i y. ** indica es a signi ican
depa u e om an e en dis ibu ion (1)('0.05).
-54-
heigh c ea ed by a dense o e hang o climbe s. One oos was occupied
by a single adul male. The o he s held om wo o 15 emales and/o
imma u e males and a leas one had one adul male. Only one oos was
occupied o mo e han one day; howe e , all hose abandoned we e oos s
whe e I had dis u bed he ba s. The single successi ely used oos held
a g oup luc ua ing om eigh o 15 indi iduals on he six occasions
when I checked o e a h ee week pe iod. Once abandoned, he oos
was
ne e occupied again o e he ollowing 15 mon hs.
E. hel um: The oos o E. hel um a LAMTO does no wa an any u he
desc ip ion. Fo o he de ailed desc ip ions o oos s in Wes A ica
see Huggel-Wol 1965, Okon 1974, and Funmilayo 1976.
OTHER SPECIES: No oos s we e e e loca ed o M. pusillus,
N. eldkampi, L. angolensis, o M. o qua a a LAMTO o elsewhe e.
4.5.2. ACTIVITY PERIODS
The i s cap u es in ne s placed nea ui ing ees (whe e ba s should
a i e i s ) we e usually be ween 1845h and 1900h. F om obse a ions
i was clea ha ba s began o aging a his ime. The las ba
• cap u ed was a 0530h, and in 280 ne hou s be ween 0500h and 0600h only
wo indi iduals we e aken. I hus de ine nigh ly ac i i y as beginning
a 1900h and e oaina ing a 0500h.
No all species dis ibu ed hei lying ac i i y e enly h ougn he 10
po en ial o aging hou s o nigh (Figu e 25). Adul male
E. bue iko e i, L. angolensis., and M. o qua a all had signi ican ly
highe cap u es han expec ed in he middle o he nigh (2300-0300h).
In con as E. bue iko e i emales and imma u e males (g ouped),
H. mons osus, and M. pusillus showed no signi ican peaks in ca ch a e
a any ime o nigh .
The mid-nigh ac i i y peak o adul male E. bua iko e i was clea ly
linked wi h hei calling pe iods in hc i e ine o e. Males, when
-55-
hey called, began be ween 1915h and 1955h (x=1942h) on "b igh " and
"da k" nigh s du ing he ma ing season and be ween 1945h and 2230h
du ing non-ma ing seasons ("da k" nigh s R=2011h; "b igh " nigh s
R=2204h; see
sec ion 4.2.4.). This e ec i ely emo ed adul males om
he popula ion o lying ba s suscep ible o ne ing be o e 2400h on
mos nigh s. Males s opped calling o 1.5 o 2.5 hou s a e 2400h,
du ing which ime hey o aged. This esul ed in he obse ed peak in
cap u es a his ime. Fo mo e de ails o he ac i i y o adio- agged
adul males see sec ion 4.5.9.
4.5.3. HABITAT USE
The di e en ial use o he
six
ne ed habi a s o zones a LAMTO
(sa anna, SNB, i e ine o es , galle y o es edge, galle y o es a
g ound and unde -canopy le els) could only be es ed o he esiden
species. The apid, high ampli ude luc ua ions in he ca ch a e o
he mig a o y species mean ha o assess habi a use all si es would
ha e had o ha e been sampled simul aneously. Sampling one si e on a
high ca ch a e nigh would ha e a i icially biassed i ela i e o
o he si es sampled on low ca ch a e nigh s.
Wi h he excep ion o L. angolensis, all species used sa anna and he
wo le els inside galle y o es s signi ican ly less han expec ed.
When I no ed he low ne yields ea ly in he s udy I abandon
,
ed sampling
a hese si es. Thus, he main sampling e o ocused on he SNB,
galle y o es edge, and he i e ine o es . Among he la e h ee
.si es, imma u e emale E. bue iko e i, H. mons osus, and
L. angolensis showed no signi ican p e e ence (Table 9). M. pusillus
used he SNB mo e han expec ed and all o he si es less han expec ed.
E. bue iko e i males and adul : emales we e cap u ed signi ican ly mo e
o en han expec ed in he i e ine o es . The high ac i i y o
E. bue iko e i in he i e ine o es was c ea ed by he calling
agg ega ions o males and (appa en ly) he isi a ions o emales, no
by o aging ac i i y. None o he species o ui s p esen in he
aeces o E. bue iko e i we e e e ound in he i e i e o es (eee
sec ion 4.5.4.) and he p opo ion o indi iduals
oidi-Ig
aeces when
caugh was signi ican ly lowe a his si e han a o he s
-62--
SPECIES
MEANS OF RECOGNITION
ANACARDIACEAE
Mangi e a indica Linn.
CUCURBITACEAE
Momo dica spp?
EBENACEAE
Diospy os mespili o mis Hochs .
EUPHORBIACEAE
B idelia e uginea Ben h.
LOGANIACEAE
An hocleis a nobilis G. Don
MORACEAE
Chlo opho a excelsa (Welw.) Ben h.
Ficus capensis Thumb.
F. lep ieu i Miq.
F. ly a a Wa b.
F. o a a Vahl.
F. poli a Vahl.
F. sco -ellio i Mildb . & Bu e
F. honningii Blume
F. allis-choudae Del.
F. ogelii (Miq.) Miq.
OBSERVATIONS, PULP, ROOST
SEEDS
ROOST, SEEDS
OBSERVATIONS, PULP
ROOST, SEEDS
OBSERVATIONS, SEEDS
GERMINATED, OBSERVATIONS, ROOST
OBSERVATIONS
GERMINATED, OBSERVATIONS
GERMINATED, OBSERVATIONS, ROOST
OBSERVATIONS
OBSERVATIONS
OBSERVATIONS
GERMINATED, OBSERVATIONS, ROOST
GERMINATED, OBSERVATIONS
PAS SIFLORACEAE
Adenia cissampeloides Ha ms.
GERMINATED, OBSERVATIONS
A. meigei Ake Assi
GERMINATED, OBSERVATIONS, ROOST
Smea hmannia pubescens Soland.
GERMINATED, OBSERVATIONS
RUBIACEAE
Nauclea la i olia
SAPINDACEAE
Panco ia bijuga Willd.
SOLANACEAE
Solanum o um Swa z
S. e basci olium Linn.
STERCULIACEAE
Cola gigan ea A. Che .
VERBENACEAE
Vi ex doniana Bake
MYRTACEAE
Psidium guaja a Linn.
ROOST, SEEDS
OBSERVATIONS
GERMINATED
GERMINATED
OBSERVATIONS, PULP, ROOST
OBSERVATIONS, PULP, ROOST
OBSERVATIONS, ROOST, SEEDS
TABLE
13: (A) Cha ac e is ics o some o he plan s commonly ea en by
he P e opodid ba s a LAMTO. F ui posi ions we e judged
subjec i ely and (F) and (C) s and o ee o clu e ed
ui posi ions espec i ely. All measu emen s a ein mm.
(B) The numbe s and pe cen s (in pa en heses) o species wi h
gi en cha ac e is ics. Unde ui colou s, b own and blue
we e classed as da k.
CTI
3
0
,--1
,
-
4
W
a)
w
P
H
CD
A
o
0
H
?-1
H
0
a)
U
›-1
0
)-4
.*.1
w
w
0
P
.
H
0
44
0
•-1
b0
"
N4
I
•
'1J
1.11
<C
4
cc,
H
Cl)
c
C
1
3
Tzi
H
cU
0
cj
4-1
<
co
• 4
00)
X
0
c
4.,
01
0
4-1
H
-I
c
ca
!
4J 1
4
J
4-1
-I
1
o
H
bC u
o I
- -i .
. i
H
p
4
W
D
0
[ I
P
›,
<4
C2,
Z
P1
4H
2 4
74
.)
zo
Ps]
0
.0
S
.
1
<g
0
Cl)
7a
—:
P
4
-
1
o
'
H
4-1
..
-I
()
0
C.)
1
q
0
.64
0
CD E
) 0
c)
CIO
0
H
Cm)
za
',4"4
za
c
..D
0
0
C)
44
•
44
e
44
a
l
co
b
•
4414
•
T.
C ;
U• '
Q)
4-1
—64—
c
s
a0
VD
-
.—I •—I
1-1
--I
)
Cl)
PI
un
CD
C)
CD cD
0
0
00 CD c) CD c)
In CD
u
l
CD CD c) CD ) CD CD
UI
,4
,4 ,4
4.4
4-1
1-I
4-.1
4-.)
4-1
4-
I
4-
I
4-1
44
4..)
4
-)
4-)
4
-
I
4-3
a)
cu W
ww
a)
w
a)
a)
0)
a
l
a)
W
a)
w
w
n
•••
w
3
w
3
w
3
CU
(11
a.)
0.
1
a)
CD
1)
cl)
a)
0)
3 3 3 3 3 3 3
3
3
_3_
a)
3
U)
V)
un
cil
(1)
CD
En
C/) co
V) V)
c ) c/) Cl) c ) C )
un
cD CD CD
u
1
N. CD 01
Cl)
0-
C.1
CV
n
4
0
/
, -4.1 n1
D
ol
n
-
n n
-
▪
•
e-,
▪
x4
P
-
i C
x
4
x4
Ci
•
C14
$
-
1
•
P
PPPPPP
mummom aimmo
H
H
H -1
,4
4
.
1
1
i
•
-1
4
—
i *1-4 °T4
(144-U44.
3
3
3
c•-•
c•-•
g
0
a)
g
0
0
o00
0o
0
w
-I
W
0
0 -I
1--1
Q)1:1 1-1
CL)
Q)'
CU
-4 1:1
1-1
P
PP
,4
PI
P
pa
P0)
()
1-i
P
CO
P
a)
0)
(1)
0
›-i 0 0
0 P-
1
';,-,
W
R
PI
Z
,..1
0
4-)
c )
4
-
1
4-
1
4
-
1
4-
1
4-
I
4-
1
En
4-1
Cl)
0
m
.--,MMWMWM"---. M
WW
M
W
0.)
a)
a)
W
0.)
Q)
w
(1.1
co
W
a0
Ti
P
o
P.
M
b0
,li
P IO P P 3-1 P P P O P
0 lia. o o o 0 0 o-o o
P4
44
cn
4.1
.
LT4
Fx4
IL'
C
x
4
44
F -i
L .'
41
z4
Q.)
al
a
)
0)0)
W
P
a)
4-1
4-)
4-
1
4-1
4-1
Li
!•,-,
>-.
-.0
0
4-4
o
ww
w
cu
CI)
41
c1-
,s
4
..0
43)
a)
wP. cc
. W
a
w
W 4-4 4-1 4-I
W4
4-
4-1
W
4-1
H
PP
P
P P (P, c,
(P
C, G4 P
a.
CD
H
H
H
H [-I x1 41 1'4 H 4.1 CL1 H PL1
•••••1 •-1
in
C>
c
s
4
C•1
0
L )
Pzi
0
A
p
; 1
n
••••
n
CO C ;
c's1
l )
Cssi
n
C,
CO c'n
csj
C
) Cs)
01
a)
LI
C.7
0
Pe.
LI
a)
I
C.,
a)
4.)
9-1
a,
LI
-65-
•4"
i
n
In
Cl)
Cl)
cl
N
P:1
in
0
CN.1
9-4 Les
i (NJ
0
w
4
-
)
4-1 4J
)4 /-I
4-1
H
a)
Q) Q)
4-4
4..)
CV
(
1
)
a)
Cl)
a) cl) a)
a)
a)
4..1 4-)
a)
•
4
3 3 3
aJ
a)
1.1 4-1
E-1
VD Cl) Ci)
Cl)
Cl)
cn
Pa
PC)
/3
0 0 0
N
,--I
LI)
04
0
0
0 0
0
H
l )
H
—4-4
CY)
CID
•-• ..",..
>4
ia
,•-
n
......, ,.....
p
4
••-•
/-•
••••.
e-s
"....,
1.4
0
x4 I:14
U) U)
...-.
..."
..... ......
Z
0 0 ›..
CD
0 0
I-4
>1
1-4 -I
H
H H
il3
P
$.4
CI 0
H
0 0
I-I
Cd
23
CI
0
H
-0 "i H
H
H
4-4 • 4
Cl)
0 0
•
-S
-i
1--1
151
4
a
O
a) a)
N
4-I
•,--
n
44
P-
n
P-I
•d•Q
4
4
a1 a)
H Ei
g
3
0 o
g
OW,--(
.-I
a) ,--c
0
O
-i a) P
0
0 ›-
1
pa
3
3
--)
0
0
0
ij
H
a) )-I
H
a)
1-1
a)
H
0
1-4
a)
4
m
0
-0 -0
O
0
O
0
O
0
1-4 1.4
Z
CU
1:)0 b )
Cu
o
4
-
)
5
H
^C7/
-0
co ,
.-,
H
0) 0) CUCU
a)
0)0)
a)
al
<
4
la0 ).0
b.°
b0
P 1-I
$4
1>
C.,
10
71 MI
C
-ii
CV
0
al
o
w
1:4
[4
PO
Pa Pa
44
Cl)
14
$4 S.4
0)0)
.0 X)
,
3
..0
.A3
X) Xi
'4
0
5 0
•
•
H
I-I
H
0
0
Li
0 0
1
...
1
)-4
a)
W
a)
I)
O -1 -I -0
,.0
-0
4 4
14
s..1
c.4
0 0
cn
co
Cl)
Cl') co
H
E-1
co
Ca'0
a)
a)
-0
c.)
•,-)
co
0
(1)
as
l
H
XI
4-1
a)
I
-
I
(1)
•-4
Vi . 04
04
0
bD
E --1
'',
1E1
4-I
0
7-) 0
1
1-)Cu
2/
• -i
'--..
>
L
I-4
1
Cd
Q
4.4
C.) CO
•,-)
ca
0,
4-1
Po
-i
W
•= ..0
m.
') 0c)
0 ,_,
-,
.,
7-4
•
,-4
x1 4-!
▪
.
'c.'
•,-1
•
,-1
•
,-1 0
10 i-'.
x-2
-,z
4) Cl)
<4 cO
14 0
E.=.1
cO
-4b
c....)
0
I.-)
io 5
4.1 cc
i
-< -4
u 5 ..,-
...
. 1 .,--,
,. ..,j
44 co -c .=
00)
0x
---4 0 s-c
o
H
.
.-1 ci) 4J
.
s: .-+
..-_
7
_,* o
'
,2-
.
0 0
0
1-7-1 ::<
c ") 0 Ei 03
HC)
I--1 ci
-c 03 ::,
0
a'
WO)
al
ill Z..-3H
[=
-
1 ,--
CJ
-
,
-- :i -d • Fi
CLI
!Id 29
00 •
HO
PA .1-I
04
cn
ç:2',
in P6(
co in W
cn c...")
-66--
a ie y o cha ac e is ics, such as: smalle o la ge seeds, o en
p onounced oughening o pi ing on he seed coa , o en'a mild o
p onounced longi udinal idge, o en,an asyme ical ape ing o one end,
and one species had a gela inous seed coa ing.
Since pa o he ollowing analyses depended on my abili y o co ec ly
classi y he seeds ha I encoun e ed and also on a knowledge o he
cha ac e is ics o he pa en plan s, I plan ed and g ew 50 Ficus A
and 35 Ficus B seed g oups.
The 50 ge mina ed Ficus A we e ep esen ed by 40 F. capensis and 10
F. allis-choudae. This con i med my p e ious Imp ession gained om
isual compa isons o he seeds. F. capensis and F. allis-choudae a e
bo h ee-s anding ee species commonly ound along galle y o es
edges and bo h a e in asi e species in he SNB.
O he 35 Ficus B seed g oups plan ed, 20 had a gela inous seed coa ing
and 15 did no . The o me g oup consis ed o a single ye uniden i ied
species ha was clea ly epiphy ic in o m. The la e g oup o 15
Ficus B consis ed o h ee species: 10 F. o a a, ou F. ogellii, and
_
one F. ly a a. All hese species a e common epiphy es in he galle y
o es s a LAMTO. F. ly a a and F. o a a p oduce la ge ui s (o e 2cm
diame e ) and we e collec ed only om H. mons osus aeces.
F. ogellii p oduces small ui s (less han lcm) and was ound in
E.
hel um aeces. I is impossible o de e mine he en i e complemem o
Ficus B species, bu om obse a ions o ba h eeding, i ce ainly
also included F. lep ieu i, F. poli e, F. sco -ellio i, and
F.
honningi; all o which a e also epiphy es.
Based on he da a, I conside he g oups o species comp ising Ficus A
and Ficus B o ep esen wo con as ing li e- o ms; he o me being
ee-s anding ees o he galle y o es edge and SNB and he la e
being epiphy es ound in he o es canopy. Fo simplici y, I will
e e o hese g oups as one species each al hough hey clea ly
ep esen mo e.
Cu es ep esen ing he inc emen o ui species ound in he die
wi h inc easing sample size (Figu e 26) o E. b e iko e i,
30-
•
20-
Li
LU
30 60 '
90 120 10 180 210 240 270 300
Mp
M
Eb
8-
,/ •
.*
.
.'
..
../
.'
.
.
.
.
./
10
'
1
1
15 20 25 30 '
35 40
'
'
45 50 55 60 65 70 715
No. OF FAECAL SAMPLES
FIGURE 26: The cumula i e numbe o ui species ound in he die s o
E. bue iko e i (Eb), M. pusillus (Mp), and M. o qua a (M )
a LAMTO, wi h inc easing sample size.
4-
-67--
M. pusillus, and M. o qua a all pla eau be o e he end o sampling.
This indica es ha ew new species a e likely o be ound in hei
die s a LAMTO. Fo he emaining ba species, con inued wo k would
likely add new ui species o hei die s; howe e , he dominance o a
small numbe o ui s means ha he addi ion o a e species would no
subs an ially change he conclusions.
Seasonal F ui A ailabili y
Only 18 o he 34 ui species we e ound in mo e han h ee aecal
samples (Table 11), and hese ga e some indica ion o he seasonal
abundance o ui s. Thi een o hese species p oduced ui s du ing
six o ewe mon hs o he yea (seasonal ui e s), wo species
p oduced ui s du ing se en o eigh mon hs (ex ended ui e s), and
h ee species p oduced ui s yea ound (con inous ui e s). Among
he 13 seasonal ui e s, se en species p oduced hei ui c ops only
du ing he we season, ou species p oduced ui s la e in he we
season and in o he d y season, and only wo species ipened ui
du ing he d y season (Table 14).
The ne esul o he a ious seasonal, ex ended, and con inuous
ui ing s a egies was o c ea e a yea - ound ui supply. Only a
single wo mon h pe iod had as ew as six species o ui s a ailable
(Janua y/Feb ua y) and he es had eigh o 12 species. This, howe e ,
o e -es ima es he abundance o ui s du ing he heigh o he d y
season (la e Janua y o mid-Feb ua y). B idelia e uginea ui s we e
deple ed by ea ly Janua y o bo h yea s, Chlo onho a excelsa ne e
ipened ui s be o e he las week o Feb ua y. and D. mespili o mis
ui s we e only ound in ea ly Janua y. Du ing he heigh o he d y
season only h ee species con inued o p oduce ui ;
he
ee-s anding
and epiphy ic Ficus species and Solanum e hasci olium. The la e
species only p oduced a ew sca e ed ui s du ing his pe iod
(al hough I could always ind some plan s wi h ui s) and had i s peak
ui ing season do ing he we season.
+ + + +
+
+4
++
6 8-
>-;
P4
>4
P4
>4
i.
>-1
P4
A
--4
1--1
-.....
A
•.--..
A
....,
A
.......
(.7
Z
H
Z
c 3
<4
E-
I
o
---
c a
CO
<-4
-
.
4
43
H
zl
A
za
A
Z
Txl
H
4
H A
44
xl
C;)
Z
4.1
W
H
(/)
PEI
H
cia
Cl)
cn
<
4
P
P P
VI
1 .3
>-I
H
:4
A
Ej
H
H
>4
w
>-I
p4
A
E
-
4
E-I
E
-
4
V.-I
1
xj
,S
El
H
E
-
4
xj
4.1
cz)
++++
+4
0
.-I
CV
n
n
/4 U
.--1
Q -1
4-1
1-1
CL)
c.J
CJ 1.4 .0
4-3
(I)
4.3 o
P4 ,0
.c-
4) 0 " 4
Co) " -1 0 0
44
•
71 4-1
'0 0
144
w o
"0
04J 0
0
11
44
CO
CO
(I)
4.) H CO
0.
C.)
-
i
0) 4
CU
0
C.)
..;
'S
C.)
4)
9-4
d
4J
C )
4-1 4-1
F:).
0
'4-.
..1
C.)
CO
G)
3)
0 C.)
Ca
.1
-
-
)
C-I
'
0
1
4-4
co
• 4
" i
.) (A
1-1
"1-1
CD a) '0 14
4-3,11
0 (1)
4-4 4-
1
s
( 4
-
)
O
.-4
4-4
-I
l)
i
4-I
F_E1
C.,
0 44
4-1 4-1
O
'0 ..
.
'0 0
0
co
0
(x)
co
a) 0
Ck
4-I
0
4-
.1
a)
0
4
.
4
Cl)
Eil
CU
o
a. co o 4.a
co
co
;-.1
LI)
L'.4
'..
CO
El
.--1 C )
0 cd
4.)
4-1
.
H -
.
1 '0
O
0 0
(J)
4) .0
4
CU
0,
(1)
C.) co
• -0
O
43 ..-1
Cl.)
cd
CO
0)
00
0 4-1
.., 0
p
C!,
.)-) 0
P4 Z
..1
4..,
l'..3
0 . i
4)
co ,--1
O
41 0 1
Q
4-4 ,0 Ul
Cl)
Cd
--1
0
EC))
0
"1-1
0 U
CO 'd 7.4 Pc,
CI '
4
-
I
Q)
W
-4
.n.
$.4
H
+ 00
<4
+4+
++
4+
a.
+ + +
+4
0
-4
1-4
P-1
++ +
+4
CO
C4
PI
+ + +
+
n
C)
<4
04
N
1-.1
ii
(15
Cl)
I
• -
0
4-4
a)
a
4-4
0
ca
a)
ci
1
cii
c )
C.)
'0
-1;
0
Z
00
{.-4
•,-(
H
•
-1
c 1
0
CZ/
Cii
C )
Cl!
C)
(1)
n ,
,c.
-
H
-11
›•
<4
VD
14
H
cia
F.4
a-1
Cl)
Cl
Ci
x4
CC
Op
a)
n
-•
E
,
ci
CO 1
0
-1
Cli
.H
0
P- -
5
ci
-1
0
•
Cl)
(xi
1
C.)
1
4
13
•
n
-+
CL)
03
4-
n
i)
---1
0
ci
ci
;-4
0
4
O ,
71
CU
0
_0
0
C.,
ciiJ
0)
cii
ci
0
be
o
d
-1
0
(0
cD
J./
cc,
Q.)
0 5
CC
'A
4-1
0
-1-1
1-1
C-
a
0
. 4
0
.H
cii
1-4
1L4
11:
a
:=V1
0
14.4
Cii
ci
cii
›
ca
• --)
e-,'
5
17)
C
:1
4
ci
i
5
c .
ca
T.)
4
J
o -{
1.
en
C.-)
N
P4
C':
cx.,
0
E9,
c N
0
co
-69-
.0
i-)
co
1"--
).0
C ) C )
csi
0 0 0 4: 0 0
cN) c
q
-X
(NI cNI
in 0 co *
n
L )
4:
CV CV
O. CV 01 * CV CV
C()
CV
C
)
*
C
..
1
C'.1
0 *0 0 0 0
* C
) -4' CV CV
N- CY. )0 CV
.0 CO
U
.
)
CN
0 IC C
)
C )
IC
Ls-)
*
0
0
a)
1-1
CD
a)
0
g
H
:
0.
co
0
c.)
c .)
a) 4:
co
Cl)
•1-
(/)
Ca
E-4
4-
0
$-1
cc
•
•
H
1-1
Cl)
1-1
0
co
0
0
4./
•
A.
Eb
Mp
Hm-
La
M
Eh(N)
Eh(S)
EVENNESS
0.67
0.75
0.79
0.77
0.29
0.25
0.65
B.
Eb
Mp
Tim
La
M
Eh(N)
Eh(S)
Eb
0.71
0.38
0.42
0.12
0.16
0.26
Mp
0.28
0.38
0.09
0.09
0.18
Tim
0.33
0.08
0.01
0.13
La
. 0.48
0.14
0.09
M
0.05
0.11
Table 16: (A) Die e enness and (B) o e lap alues o he p e opodid
ba species a LAMTO. Ba species designa ions a e as in
Table 1. Eh(N) and Eh(S) ep esen E. hel um on he
no hwa d and
.
sou hwa d mig a ions espec i ely.
-75--
The E. hel um/H. mons osus Pai
Al hough E. hel um and H. mons osus o aged in he same o es canopy
zone ea ly in he yea (no hwa d mig a ion pe iod), hey selec ed
di e en ui s and had he lowes die o e lap o any' speciespai
(1.1%). Pa i ioning wi hin he zone appea ed o occu on he basis o
ui size and was media ed by dis inc mo phological di e ences.
The conspicuous ea u e o he die o E. hel um was he small size o
he ui s ha i exploi ed. On he no hwa d mig a ion E. hel um
elied p ima ily on C. excelsa, a ee p oducing ui s app oxima ely
25x8mm in dimensions. On he e u n mig a ion his species used
B. e uginea, a ee p oducing ui s only 6mm in diame e , o 50.0%
o i s die . The limi ed ge mina ions o Ficus B seeds om he aeces
(4) we e o all o F. ogellii. On 12 nigh s I obse ed E. heJ um
eeding on F. lep ieu i (5 nigh s), F. honningii (5 nigh s), and
F. ogellii (2 nigh s). All hese Ficus species p oduce small ui s
unde 15mm in diame e . When eeding in hese ees E. hel um did no
emo e he ui s om he ee, bu me ely ed by c awling along he
b anches and consuming he ui s in place. Thei ema kably long legs
and humb a e well adap ed o climbing and make E. hel um ex emely
agile in ees (see also Kulze 1971). I appea ed o be his
mo phological adap a ion o climbing ha made his species capable o
e icien ly exploi ing small, dispe sed ui packages.
Ele en Ficus B seeds om H. mons osus aeces, when ge mina ed, p o ed
o be F. o a a (10) and F. ly a a (1), bo h o which a e species wi h
la ge 30-40mm diame e ui s. In addi ion, on o e 20 occasions I
obse ed his la ge (and eadily iden i iable) species ly in o and
emo e ui s om F. o a a, F. sco -ellio i, and F. poli a ees
eme gen abo e he galle y o es and i e ine o es canopy. All hese
species p oduce la ge ui s 30-40mm in diame e . Du ing obse a ions
a ui ing F. lep ieu i and F. honningi ees which we e being isi ed
by E. hel um, I ne e obse ed H. mons osus. All e idence indica ed
ha H. mons osus used only he la ge ui ed species and I suspec
ha his ela i ely clumsy climbe could no ha e e ec i ely handled
he smalle ui s.
-76-
The L. angolensis/M. o qua a Pai
L. angolensis and M. o qua a had he second highes o e lap (47.7%)
among he species pai s. S. e basci olium cons i u ed he single mos
impo an ui species o bo h ba s, wi h L. angolensis using i o
43.8% and M. o qua a o 89.5% o he die . Da ing he d y season e
LAMTO M. o qua a e ea ed sou h o he o es zone and he wo species
we e geog aphically seg ega ed. Du ing he Janua y o Ap il d y season
pe iod L. angolensis con inued o use S. e basci olium, bu sp ead i s
die among o he non-Ficus species (S. e basci olium=20%,
o he s=80%, n=10). When M. o qua a mo ed h ough he communi y wi h
he onse o he ains and he peak ui ing pe iod o
S. e basci olium, i is unlikely ha ood was a limi ing esou ce and
ha he wo species we e in se ious compe i ion.
ALTERNATE FOOD SOURCES
None o he 16 un ecognizable ne aeces and 39 un ecqnizeable oos
aeces had any agmen s o plan walls o insec exoskele ons o
indica e ha ba s had been eeding on buds, lea es, o insec s o
supplemen hei nec a , pollen, o ui die s. In 90 aeces
con aining seeds I did ind numbe s o lepidop e an scales. Su ace
swabs om ipe ui s, howe e , p oduced la ge numbe s o scales.
Con olled eeding o such ui s o cap i e ba s demons a ed ha
hese scales u ned up in aeces in app oxima ely he same p opo ions
as was obse ed in he ne aeces. I is unlikely ha he ba s
ac i ely hun ed and caugh mo hs.
Among 85 Ficus aeces ha I examined, 35.3% con ained some emains o
ig wasps (Agaonidae). These we e po en ially an al e na e sou ce o
p o ein. I examined 50 F. capensis ui s a he s age o ipeness when
hey we e emo ed by ba s ( i s o second day a e so ening and
u ning o ange/ ed, easily de ached) o de e mine he incidence o
in es a ion. Six een pe cen o he ig syconia we e s ill in es ed,
ha ing a mean o 6.4mg esh weigh o wasps.
A
a le el o 18% p o ein
by esh weigh (Mo on 1973), his ep esen s 1.2mg o insec p o ein
In an in es ed ig.
-77-
4.5.5 ABUNDANCE OF BAT-EXPLOITED FRUIT SPECIES
GALLERY FOREST EDGE
Table 17 shows he numbe s o F. capensis and o he ui species known
o be used by ba s along i e ansec s o alling 3.6km o galle y
o es edge. Along hese ansec s I ound nine species: Adenia
cissampeloides, A. meigei, An hocleis a nobilis, F. capensis, F. o a a,
F. allis-choudae, Nauclea la i olia, Smea hmania pubescens, and Vi ex
doniana.
The numbe s o he la e wo species we e unde -es ima ed
since hey o en occu ed in he sa anna nea he edge, bu no o ming
pa o his communi y and so we e no coun ed.
The census shows ha ba -exploi ed ui species a e common. A ba
o aging along he galle y o es edge would ha e o mo e a mean
dis ance o only 18.4m o encoun e ano he po en ial ui sou ce.
Howe e , all he species we e seasonal ui e s wi h he excep ion o
he Ficus species and his unde -es ima es he eal in e - ui ing ee
dis ance. Only F. capensis p oduced abundan ui s on a con inuous
basis and his was he mos eliable ui ing species. The mean
in e - ee dis ance o F. capensis was 54.9m and he e we e a mean o
21.5 ees/km o edge. F om la ge scale maps o he a ea I es ima e
. he e o be 140±6.7km o edge wi hin he 2500ha ese e. On his basis
he o al numbe o F. capensis can be es ima ed a 3010 ees along he
galle y o es edge.
SNB
Table 17 shows he esul s o h ee 20m wide census ansec s aken in
he SNB a dis ances o 50, 100, and 150m om he adjoining galle y
o es edge. The e was no signi ican di e ence in he numbe o
F. capensis ees in he h ee ansec s (X
2
=1.7, d =2, 00.05), gi ing
a mean o 51.3 ees/ha. On his basis in he 80ha SNB he e we e
app oxima ely 4104 F. capensis ees o 1.36 imes he numbe in he
en i e galle y o es edge zone.
The s uc u al di e ences be ween he galle y o es edge (linea ) and
he SNB (plana ) cge he wi h he di e en densi ies esul s in
Fz.4
1-4
000
In c.)
0-1
0-1
.-4
c
‘
i
e )
Pc1
Fa pa
Z •
Cl)
i)
In
n
•
•
•
•
.
—
1
0000
•
CV
C
)
*
4"
In
4-)
Cl)
a)
x1
O
0
›N (-1
4-1 Cl)
H
1-1
110
QS 4
4-)
Cl)
4.)
C)
o 0
0
co 0 .H
0
4-1
.
•
c
4s.'
n-
cd
U)
(I)
a)
1.)
4-1
a)
"0
4-i
Cl)
)-1
O
0
•
CL)
O
4-1
( s
co
a)
CD
Cl)
a I
1-1
CI)
cu
P. 4.1
c i
i
o
4-1
1-1
•
4-
di
U)
c.)
O
0
a)
cd
CU
P4 4-I
.
H
4-
i
. 4
co
O
-78-
00 cn
0 CD
•
• • • •
0
c-, cn
.
4
.
o
....1-
.-.1
—+
.--1
cn
••-
n
)-4
CO
CD 0
•
•
•
•
%
.0 0 CO
('4
1`. C ) CV
in
0 N. 0 1`,
•
•
•
•
.0-
'00
cn
CV 0 1/40 I l
c )
0 in CO N.
%.0
N. 0000
0
•
•
•
•
4"
Cn
0 'l-
i
)
‘.0 CO CI
--
I-.
'
e-% ••-
n
CV C
)
In
•
•
•
C
)
VD ON 1/40
0.1
CV
CV
In
C
4
1 Csl "
4
•-•
•
•
•
•
00
C4 COO CNI
-
-4
ce)
N4 (NJ 0
In
CV
CO C
)
In
)-4
ce)
•
O
CV CV
e-1
‘C)
in
n
0 1/40 -)
C
)
C
V
CV
c )
•
•
•
•
00
0
000
CV
04 c
000
0 Ln
c ) 04 04
•
•
•
—1
•
•
6
CN
lindd HUM S33d1 dO
FIGURE 29: The pe cen o F. capensis ees in he SNB (n=54) and along
one galle y o es edge (n=46) ha had some ipe ui on a
gi en census ou .
• -79--
g ea ly di e en in e - ee dis ances. I ees we e dis ibu ed
e enly h ough he SNB, hen each ee would ha e had six equidis an
neighbou s, each 15.8m away. The ees app oached a andom dispe sion
(no quan i ied) in he SNB, bu he mean dis ance o he six nea es
neighbou s emains 15.8m (al hough he a iance on his mean would be
high). In con as , he mean dis ance be ween a ee and
i s six
nea es neighbou s in he galle y o es edge was 109.8m o 6.9 imes
g ea e .
4.5.6 FRUIT PRODUCTION OF F. CAPENSIS
Al hough indi idual ees did no p oduce ui s cons an ly, he
popula ion o ees moni o ed egula ly along he galle y o es edge
(n=46 ees) and in he SNB (n=54 ees) p oduced ui s yea - ound
(Figu e 29). On 74.5% o he wice weekly ou s along galle y o es
edge (n=47) and on 98.8% o ou s in he SNB (n=8I) he e we e some ipe
ui .
The e we e signi ican di e ences in he a e o ui p oduc ion
be ween he wo habi a s. Signi ican ly ewe edge ees had ipe ui
on any ou (x=6.6%) han SNB ees (x=23.4%; sign es , p<0.01). This
was due o he ac ha indi idual edge ees had ewe cycles o ui
p oduc ion han did SMB ees (Table 18). Fo analyses I ha e de ined a
p oduc ion cycle as he p esence o one o mo e ui on a ee on wo
consecu i e ou s (o h ee o mo e ipe ui on only one ou ) bounded
by ou ou s ( wo weeks) wi hou any ipe ui s. In he SNB only
16.7% o he ees (n=54) ne e cycled and 53.7% o he ees cycled wo
o ewe imes. Along he galle y o es edge 47.8% o he ees ne e
cycled and 91.3% cycled wo o ewe imes du ing he same pe iod.
The o m o ees also di e ed be ween he wo si es. Along galle y
o es edge only 4.3% o he ees had mo e han one s em, while in he
SNB 68.5% o ees consis ed o clus e s o wo o eigh s ems. I ne e
de e mined whe he hese s ems sha ed a common oo sys ems o consis ed
o sepa a e indi iduals. Howe e , he e was no signi ican co ela ion
be ween he numbe o s ems and ui cycles (F
=
0.15, n-51, p>0.05),
indica ing ha s ems did no cycle independen ly.
% OF TREES CYCLING X TIMES
# OF CYCLES
GALLERY FOREST EDGE
SNB
0
47.8
16.7
1
21.7 14.8
2
21.7 22.2
3
8.7
14.8
4 0
25.9
5
0
3.7
6
CONTINUOUS
0
1.9
Table 18: The equency o ui p oduc ion cycles among 46 F. capensis
ees along one galle y o es edge and 54 ees in he SNB a
LAMTO be ween Ma ch and No embe 1980.
-81—
The e was no clea ela ionship be ween ui 'cycling and. s em
diame e . Along galle y
,
o es edge he numbe o cycles was
signi ican ly co ela ed wi h s em diame e (F=6.0, n=39, p40.05), bu
diame e explained only 14% o he o al a ia ion in ui p oduc ion.
In he SNB he e was no signi ican co ela ion be ween s em diame e
and ui cycling (F=0,07, n=17, p>0.05) o hose ees wi h only one
s em.
FRUIT BIOMASS
While he o al biomass o ui s p oduced mon hly o yea ly canno be
es ima ed om he census ou s (some ui s we e ipened and emo ed
be ween ou s), he numbe and biomass o ui s a ailable o ugi o es
on each census day and nigh can be es ima ed.
GALLERY FOREST: Along he galle y o es a mean o 6.6% o ees bo e
ui a any gi en ime. T ees wi h some ui had a mean c op o 10.4
ui s (n=146 ee days wi h ui ). Thus o he 3010 ees in his
zone, 199 bo e a mean o 10.4 ui s on a gi en day, o a daily c op o
2070 ui s. A a mean weigh o 21.1g/ ui (n=58) his ep esen s
43.7kg ui /day in he en i e zone. The p oduc ion, howe e , o en
d opped well below his le el.
SNB: The p opo ion o ees bea ing' ui s on a gi en census ou
a ied g ea ly, since he ees appea ed o be oughly synch onized in
hei ui p oduc ion (Figu e 29). O e he yea a mean o 23.4% o
ees p oduced some ui s on a gi en day. Each ui ing ee bo e a
mean o 8.3 ui s (n=1024 ee days wi h ui ). Thus, in he SNB an
a e age o 960 ees bo e 8.3 ui s o a mean daily c op o 7968
ui s. This ep esen s 168.1kg ui /day. Du ing low pe iods in ui
p oduc ion (Janua y, Augus /Sep embe on Figu e 29) a mean o 8% o
ees had ui s. A hese imes he daily c op was 2725 ui s and he
daily biomass was 54.8kg. This dec eased a ailabili y, especially
du ing he Janua y pe iod when ew al e na e ui s we e a ailable,
appea ed o be o key impo ance in go e ning M. pusillus popula ions in
he SNB.
-82-
4.5.7 MOVEMENTS AND POPULATIONS: MICROPTEROPUS PUSILLUS
MOVEMENTS
Du ing he s udy I made 411 cap u es o M. pusillus (Table 19). O
hese, 178 we e o indi iduals ha I ma ked, 41 we e o indi iduals
ha I eleased unma ked, and 192 we e ecap u es o p e iously ma ked
indi iduals.
Nine y-se en (54.5%) o he ma ked indi iduals we e ecap u ed one o
mo e imes (Table 19). Ba s o iginally ma ked in he SNB we e no mo e
o less likely o be ecap u ed han hose ma ked elsewhe e (X2=0.0,
d =1, p>0.05). All sex-age classes had he same ecap u e a e (X2=0.1,
d =3, p>0.05), indica ing ha imma u e and adul ba s died o
emig a ed om he s udy a ea a he same a e.
The mo emen s o M. pusillus we e s ongly a ec ed by he SNB. The
si e ideli y (measu ed as he numbe o ecap u es a he o iginal
banding si e s a a new si e) o indi iduals o iginally ma ked in he
SNB was signi ican ly highe han o indi iduals ma ked along he
galle y o es edge (X
2
=87.5, d =1, p<0.001; Table 19). O 151
ecap u es o indi iduals o iginally ma ked in he SNB, 91.4% we e a
he same si e and only 8.6% we e ou side his habi a . In con as only
19.5% o ecap u es o indi iduals ma ked a galle y o es edge si es
(n=41) we e a he same si e. Thi y- h ee o hese ecap u es we e a
new loca ions and 87.9% o he mo emen s we e in o he SNB.
These mo emen s sugges ha he popula ion o M. pusillus was based in
he SNB wi h only a ew indi iduals making empo a y excu sions in o he
su ounding a ea. This app oxima es a closed popula ion occupying wha
could be conside ed essen ially an island habi a - he SNB. This
jus i ies a mo e comple e ma k- elease- ecap u e analysis o ec ui men
and popula ion size.
POPULATION STRUCTURE
When bands we e in sho supply, I p e e en ially ma ked imma u e males
and emales, so he igu es in Table 1913 unde - ep esen he ue
A.
MARKING SITE
INSIDE SNB
OUTSIDE SNB
COMPARISON
# Ma ked
124
54
X270.0
# Recap u ed
71
26
'
p>0.05
% Recap u ed
.57.3
48.1
B.
SEX—AGE GROUP
ADULT
di
s
IMMATURE O ADULT
IMMATURE
Y
COMPARISON
# Ma ked
44
47
45
42
X2=0.1
#Recap u ed
25
25
25
22
p>0.05
%
.
Recap u ed
56.8
53.2
55.6
52.4
C.
LOCATION OF RECAPTURES
ORIGINAL
AT SAME SITE
AT NEW SITE
BANDING SITE # RECAPTURES
N
N
%
COMPARISON
SNB
151
138
91.4
13
8.6
X2=87.5
OUTSIDE SNB
41
8
19.5
33
80.5
p<0.001
D.
MOVEMENT 10:
SNB
A
OTHER SITE
N
%
N
%
29
87.9
4
12.1
TABLE 23: The ecap u e and mo emen cha ac e is ics o
M.
pusillus
a LAMTO, showing:
(A)
The ecap u e a es o ba s o iginally ma ked in he SNB
o elsewhe e,
(B)
he ecap u e a es o adul and imma u e males and
emales,
(C)
and (D) he loca ions o ecap u es ela i e o he
o iginal banding si e.
-89-
TABLE 21: The ecap u e and mo emen cha ac e is ics o
E. bue iko e i a LAMTO, showing:
(A)
he ecap u e a es o adul and imma u e males and
emales,
(B)
he ecap u e a es o males and emales o iginally
banded in he SNB o elsewhe e,
(C)
and (D) he loca ions o ecap u es ela i e o he
o iginal banding si es o males and emales.
(no) indica es a non-signi ican X
2
compa ison.
-90--
A.
MALES
FEMALES
ADULT TMM. COMPARISON
ADULT INN. COMPARISON
# MARKED
155
290
104
110
# RECAPTURED
41
70
X2-0.3
12
10
X2=0.4
% RECAPTURED
26.5
24.1
p>0.05 (ns)
11.5
9.1
p>0.05 (ns)
B.
BANDING SITE (4)
BANDING SITE ()
SNB
OTHER COMPARISON
SNB
OTHER COMPARISON
# MARKED
182
263
X2=15.4
70
144
X2=7.8
# RECAPTURED
63
48
p<0.01
13
9
p<0.05
% RECAPTURED
34.6
18.3
18.6
6.3
Si!.
MALES:
LOCATION OF RECAPTURES
ORIGINAL
AT SAME SITE AT NEW SITE
BANDING SITE
# RECAPTURES
COMPARISON
SNB
121
99
81.8
22
18.2
X2=46.6
OUTSIDE SNB
68
21
30.9
47
69.1
p(0.001
MOVEMENT TO:
SNB ..][..NEW SITE
N
%
N
%
20 42.6
27
57.5
D. FEMALES:
LOCATION OF RECAPTURES
ORIGINAL
AT SAME SITE AT NEW SITE
BANDING SITE
# RECAPTURES
COMPARISON
SNB
20
13
65.0
7
35.0
X2=2.2
OUTSIDE SNB
13
5
38.5
8
61.5
p>0.5 (ns)
MOVEMENT TO:
SNB
NEW SITE
N
%
N
%
4 50.0
4
50.0
-91 -
Males appea ed o be less mobile han emales. The high p opo ion o
ecap u es ha we e in he SN13 (81.8%) indica es ha males spen mos
o hei ime in his habi a (see also sec ion 4.5.9). This jus i ies
he applica ion o
a
ma k- elease- ecap u e analysis o male .
popula ions.
' POPULATION STRUCTURE
When bands we e in sho supply I p e e en ially ma ked imma u e
Indi iduals. Fo his eason he igu es in Table 21A uncle - e es.ea
he ue numbe s o adul s o bo h sexes. I ha e used he o al
cap u es o 285 adul males, 402 imma u e males, 166 adul emales, and
140 imma u e emales as he da a base o he ollowing analyses.
SEX RATIO: The sex a io among adul s de ia ed signi ican ly om uni y
(X
2
=31.4 d =1, 1)(0.001; Table 22), ha ing an unde - ep esen a ion o
adul emales ela i e o adul males.
Among imma u es, he o iginal da a canno be compa ed di ec ly since he
sexes ma u ed a di e en ages. Imma u e emales became p egnan a
he age o six mon hs and we e classed as adul s he ea e . Imma u e
males achie ed pube y and we e classed as adul a 15 mon hs o age.
The a io o males o emales among imma u es should 2.5:1 (15/6 mon hs)
o show an equal sex a io. The a io o males o emales among
imma u es did no de ia e signi ican ly om ha expec ed (Table 22).
Imma u e
W :
Adul
7
Ra io: Each emale p oduced
0.5
emale o sp ing
•
e e y six mon hs. These o sp ing ma u ed and we e classed as adul s by
he ime o ec ui men o he ollowing coho . Thus he a io o
imma u e emales o adul emales should be 0.5:1. The p opo ions in
he
'
cap u es de ia ed signi ican ly om hose expec ed (Table 22).
Adul emales we e unde - ep esen ed ela i e o imma u e emales.
Imma u e o e:Adul ?? Ra io: Wi h an a e age p oduc ion o 0.5 male
o sp ing pe adul emale e e y six mon hs and wi h males
.
ma u ing a
15 mon hs, he e should be 1.25 imma u e males pe adul emale in he
cap u es. The obse ed p opo ions de ia ed signi ican ly om hose
expec ed (Table 22). Adul emales we e unde - ep esen ed ela i e o
Imma u e males.
-92-
A.
ADULT SEX RATIO
-
TOTAL
IMMATURE SEX RATIO
TOTAL
OBSERVED
EXPECTED
X
2
B.
285
225.5
15.7
166
225.5
15.7
451
31.4*
402
387.1
0.6
140
154.9
1.4
542
2.0 (no)
ADULT FEMALE
ADULT
4
imm. A
s
:
OFFSPRING RATIO
TOTAL
ADULT .
pp
imm.
TOTAL
OBSERVED
EXPECTED
X
2
C.
166
204
7.1
140
102
14.2
306
21.3*
166
252.4
29.6
402
315.6
23.7
568
53•3*
ADULT MALE : OFFSPRING RATIO
ADULT
R
IMM.
a
TOTAL
ADULT
ee
IMM.
gp
TOTAL
OBSERVED
285 140
425
285
402 687
EXPECTED
283.5
141.7
305.3 381.7
X
2
0.0
0.0
0.0 (no)
1.4
1.1
2.5
(ns)
TABLE 22: (A) Sex a ios among adul and imma u e
E. bue iko e i
and
(B) and (C) he a io o imma u es o bo h sexes o adul
emales and adul males. * designa es a X
2
compa ison
signi ican a p< 0.01 and (no) designa es a non-signi ican X2.
-93-
The sex a io among adul s and he a io o adul emales o imma u es
o bo h sexes all show adul emales o be unde - ep esen ed. Adul
emales appea o be less equen ly caugh a e pa u i ion han
expec ed. This could a ise om h ee causes: 1) pos -pa u i ion
emig a ion, 2) pos -pa u i ion mo ali y, o 3) a low a4bili y o
adul emales ela i e o o he sex-age g oups. The i s wo possible
causes would ha e educed he ecap u e a e o adul emales ela i e
o imma u e emales. This was no obse ed. This sugges s ha o
some eason adul emales we e p esen , bu me ely no cap u ed. I
he e was in ac an equal sex a io among adul s, hen he numbe s o
adul males would be good p edic o s o he imma u e popula ion size.
Table 24 shows his o be he case. Nei he imma u e emale o adul
male no imma u e male o adul male a ios de ia ed signi ican ly om
ha expec ed. This lends suppo o he conclusion ha adul emales
we e p esen , bu me ely less appable.
I his is ue, hen he eal popula ion can be es ima ed by equa ing
adul emale numbe s o hose o adul males, gi ing he sex-age classes
In he p opo ions o : 285 adul males: 402 imma u e males: 285 adul
emales: 140 imma u e emales. All emales ep esen 38% o his
popula ion and adul emales ep esen 26%. Females do no comp ise 50%
o his hypo he ical popula ion (as an equal sex a io a bi h would
gene a e) because by equa ing he numbe s o adul males and emales I
ha e assumed ha hey ha e equal li e expec ancies a e eaching
pube y. Since males emain imma u e nine mon hs longe han emales
he su plus o males o e emales in his imma u e ca ego y (1.5
ec ui men s) would boos he o e all male numbe s. While I canno
suppo his assump ion (equal pos -pube y longe i y) o
E. bue iko e i, his has been indica ed o a numbe o insec i o ous
ba s (B adbu y and Emmons 1974, Tannenbaum 1975, B adbu y and Veh encamp
1976a). Tha his holds ue o E. bue iko e i is sugges ed by he
ac ha he hypo he ical popula ions so gene a ed "wo k" empi ically,
explaining he obse ed p opo ions o imma u es in he popula ion and
(as we shall see) he ec ui men .
The, high ecap u e a e and low mo emen s om he SNB indica e ha
males which we e o iginal1y caugh he e spen mos o hei ime in he
con ines o his 80 hec a e habi a . Since Tulles and emales di e in
hei mo emen s I ha e calcula ed Jolly-Sebe popula ion es ima es o
MONTH (1979/1980)
MALE
POPULATION SIZE h SE
ESTIMATED TOTAL
POPULATION SIZE
JUNE 160 ± 200
221
.
JULY 177 * 107
244
AUGUST
178 *
79
246
SEPTEMBER
167 *
64
231
OCTOBER
NOVEMBER
223 *
66
308
DECEMBER
203 ±
44
280
JANUARY
168 *
40
232
FEBRUARY
153 *
44
211
MARCH
148 *
56
204
APRIL
160
h
79
221
MAY
167 *
76
231
JUNE
266 ±
82
267
JULY
AUGUST
275 *
98
380
TABLE 23: Mon hly Jolly Sebe popula ion es ima es o male
E.
bue iko e i
in he SNB a LAMTO, as well as he es ima ed
o al popula ion including emales. SE = s anda d e o .
^
/
'N
...
-
-
•
-
--
/
-
Ni
400-
-
300-
-
.........41
1.1.1
NI 200-
i"
.7)
a:
CD
a.
100-
IIIIIIIIFIIIIIIIIii,
MJ J ASONDJ FMAMJ J ASOND
MONTH
TABLE 31: Mon hly Jolly Sebe popula ion es ima es o male
E. bue iko e i in he SNB a LAMTO. Es ima es a e p esen ed
± one s anda d e o .
-95-
The high ecap u e a e and low mo emen s om he SNB indica e ha
males which we e o iginally caugh he e spen mos o hei ime in he
con ines o his 80 hec a e habi a . Since males and emales di e in
hei mo emen s I ha e calcula ed Jolly-Sebe popula ion es ima es o
males only. I hen calcula ed he o al popula ion based on he numbe s
o emales ha mus ha e been p esen a he sampling imes o c ea e
he sex-age a ios ou lined abo e.
Table 23 and Figu e, 31 show he mon hly popula ion es ima es o
E. bue iko e i males and o he calcula ed numbe s o bo h sexes in
he SNB. Be ween June and Sep embe 1979 (Figu e 31) he popula ion in
he SNB emained ela i ely s able numbe ing be ween 160 and 178 males
o 221 and 246 indi iduals o bo h sexes. The popula ion peaked be ween
Sep embe and No embe , co esponding wi h he ec ui men o coho 2,
1979 in o he lying popula ion. The popula ion declined h ough he
d y
season o a low o 148 o 160 males (204-221 indi iduals) in Ma ch
and Ap il. The ollowing inc ease be ween May and June coincided wi h
he ec ui men o coho 1, 1980. Due o low sample sizes no
popula ion es ima es could be calcula ed a e Augus 1980.
No men ion has been made o he s anda d e o s o he popula ion
es ima es. As wi h M. pusillus, he accu acy o he mon hly popula ion
es ima es can be e i ied by compa ing each obse ed popula ion
inc eases a e ec ui men wi h he expec ed inc ease based on he
numbe o adul emales es ima ed o be p esen du ing he mon hs p io
pos - ec ui men popula ion sizes. In bo h possible compa isons, he e
was no signi ican di e ence be ween he wo. The numbe o adul
emales es ima ed o be p esen p edic ed accu a ely he obse ed
.inc ease in he popula ions wi h each ec ui men , indica ing ha he
o iginal popula ion es ima es we e accu a e.
PRE-RECRUITMENT
MONTH
POP. SIZE
• PREDICTED POST-
RECRUITMENT SIZE
OBSERVED POST-
RECRUITMENT SIZE
X2
JULY 1979
FEBRUARY 1980
244
211
308
266
308
267
0.0
0.0
X
2
= 0.0
-.96—
Assuming a mean weigh o 150g o E. bue iko e i indi iduals (ca. 120g
o adul emales, 100g o imma u e emales, 140g o imma u e males,
200g o adul males), hen he biomass o his species in he SNB is as
ollows:
POP. SIZE
INDIVS./Ha
BIOMASS/Ha
(Kg)
mean
256..9
3.21
0.48
minimum
204
2.55
0.38
maximum
380
4.75
0.71.
The mean o 3.2 indi iduals/Ha in he SNB ga e ise o a ca ch a e o
0.10 ba s/ne hou . I indi iduals beha e in a simila ashion while
o aging along he galle y o es edge, hen he ca ch a e in his
la e zone (0.05 ba s/ne hou ) should es ima e he densi y o
E. bue iko e i he e. On his basis, he e we e an es ima ed 1.6
Indi iduals/Ha o galle y o es edge o a biomass o 0.24Kg/Ha.
4.5.9 RADIO TRACKING
Th ough he s udy I a ached adio ansmi e s o 27 E. bue iko e i (9
adul males, 12 imma u e males, 1 lac a ing emale, 5 imma u e
emales). Fo six indi iduals he ags we e as ened by gluing o he
do sal u . The subsequen ecap u e o one o hese indi iduals wi h a
sizeable
ab
A
ess unde he ag posi ion sugges ed ha his means o
a achmen caused a bes discom o and almos ce ainly se e e
beha iou al dis u bance. As a esul I was obliged o disca d all da a
o hese ba s. Fo he emaining 21 ba s I a ached he ansmi e s
wi h nylon colla s. Fou ecap u es showed no e idence o ab asions o
damage o he ba s, bu all ou ansmi e s had los hei an ennae.
This mean ha he pe manen loss o con ac wi h an indi idual could
no be cons ued as a pe manen mo emen om he a ea. The ollowing
obse a ions we e made on he only six indi iduals (2 adul males, 3
imma u e males, 1 lac a ing emale) ha we e loca ed on mo e han one
nigh .
—91--
ADULT MALES
The wo adul males we e adio— agged ollowing hei cap u e in he SNB
du ing Feb ua y (male A) and Oc obe (male B) 1980. Bo h pe iods we e
du ing ma ing seasons. Males A and B we e ollowed o 10 and 15 days .
espec i ely, he o me du ing a waning moon (days 19 o 1 in he luna
cycle) and he la e du ing a waxing moon (days 1 o 15). Bo h males
called a si es in he i e ine o es du ing he g ea e pa o mos
nigh s, bu male B abandoned calling du ing he las h ee days
app oaching a ull moon when mos o he males had also s opped calling.
The oos s o bo h males we e some dis ance up he Bandama Ri e ou o
acking ange and I ne e success ully loca ed ei he . On all calling
nigh s o male B and i e calling nigh s o male A, hey mo ed
di ec ly o he same p e iously used calling si es whe e hey emained
quie un il o he males began o call be ween 1933h and 1945h. On i e
nigh s male A i s mo ed in o he SNB and spen a mean ime o 22.1
minu es eeding be o e going o he calling si e. A e a i ing in he
SNB, male A loca ed an F. capensis ui and emo ed i o one o h ee
eeding oos s ha i used h ough he acking pe iod. I p ocessed
he ui , aking a mean o 15.2 minu es, hen ei he mo ed di ec ly o
he calling si e o ound ano he ui , p ocessed i , and hen mo ed o
he calling si e. On wo nigh s male A ed on wo ui s and on h ee
nigh s on one ui du ing his ea ly eeding pe iod.
Males called cons an ly un il a e 2400h, in e up ing calling a mean
imes o 0054h (male A) and 0159h (male B). A his ime each male
mo ed o he SNB and ed o a mean o 114 minu es (male A) and 82
minu es (male B). Feeding ac i i y was es ic ed o sho (ca. 30
second) sea ch ligh s and longe handling pe iods (male A: R=22.1
minu es; male B: 7:=19.5 minu es) which I e m eeding bou s. The
longes sea ch ligh was only h ee minu es and bo h males oge he
accumula ed 123 minu es o lying ime du ing he o al con ac ime o
2124 minu es away om he calling si es. This ep esen s 5.8% o he
o aging ime spen in ligh . Male A used h ee di e en eeding
oos s o e he 10 con ac nigh s, always using mo e han one on a gi en
nigh . Male B used wo eeding oos s, bo h on all nigh s. I only
loca ed one eeding oos o each male and so canno be ce ain o all
he die i ems aken; howe e , all he ui emains a he wo oos s
A.
ANT CATEGORY
FICUS CAPENSIS
% FALLEN
# TREES
NIGHT DAY
SNB
NO ANTS
77.5
(32.1-100)
14.5
(0-42.9)
8.6
(0-46.4)
19
LIGHT ANTS
74.4 (46.4-100)
13.6
(0-53.6) 7.5 (0-17.2)
9
HEAVY ANTS
62.0
(44.0-75.4)
7.9
(0-17.6) 33.8
(19.3-50)
10
GALLERY EDGE
NO ANTS
30.7
(0-57.7)
52.1
(30.7-100)
17.2
(0-27.5)
7
ADENIA CISSAMPELOIDES
LOCATION NIGHT DAY
% FALLEN
# STEMS
GALLERY EDGE
81
9
11
1
B.
NIGHT 1
FICUS CAPENSIS
NIGHT 4 # FRUITS
NIGHT 2
NIGHT 3
% CONSUMED
54•5
83.1
91.8
92.6
214
% FALLEN
1.7
5.6 7.4 7.4
17
% REMAINING
43.8
11.3
0.8
T0TAL---231
TABLE 25: (A): The p opo ion o F. capensis and A. cissampeloides i ui s
emo ed by diu nal and noc u nal ugi o es o alling
unconsumed in he SNB o galle y o es edge. Ranges a e in
pa en heses
(B):
The a e o emo c.1 o all o F. capensis ui s in he
SNB.
-105-
n
1
=9, n2=21, p>0.05). Thus he e was no e idence o a pa i ioning o
he wo heigh s a a be ween ba s and diu nal ugi o es.
To de e mine he a e o emo al o ipe ui s, a sample o 231 ui s
we e ma ked on he i s day o ipening (indica ed by he change in
colou om g een o deep o ange/ ed and he so ening o he pulp). O
hese, 56.2% had been emo ed o had allen ollowing he i s nigh
and only 0.8% emained on he ees a e h ee nigh s (Table 25).
"HEAVY ANT" TREES: On ees bea ing la ge an popula ions, a
signi ican ly g ea e p opo ion o ui s ell han on ees wi h ew
o no an s. Diu nal ugi o es emo ed he same p opo ion o ui s on
"hea y an " ees as on "ligh " o "no an " ees (U es , z=0.60,
n1
=
10, n
2
=28, p>0.05), bu ba s emo ed a signi ican ly snlalle
p opo ion o ui s on "hea y an " ees (U es , z=3.33, n1=10, n9=28,
p<0.001). The inc eased numbe o allen ui s we e hose no aken by
ba s. On "hea y an " ees ba s emo ed only 62.0% o he ma ked ui s
compa ed wi h 78.5% on "no an " and "ligh an " ees. I is
in e es ing o no e ha he diu nal ugi o es did no espond o he
inc eased a ailabili y o ui s by inc easing hei consump ion.
Despi e he educed impac o ba s on he ui c ops o "hea y an "
ees hey s ill emo ed signi ican ly mo e ui s han did diu nal
ugi o es (U es , U
=
0.0,.n1
=
n2=10, p<0.001).
I an s hemsel es we e esponsible o he change in dispe sal
cha ac e is ics (and we e no me ely co ela ed wi h some o he ac o
commonly a ec ing bo h, eg. loca ion) hen manipula ing an abundance
.should p oduce p edic able shi s in he p opo ions o he ui s
emo ed by ba s. To es his I el,,imina ed an s on wo "hea y an "
ees by emo ing he colonies and I also loca ed one p e iously
measu ed "ligh an " ee on which an popula ions had inc eased. T ees
wi h dec eased an numbe s inc eased he p opo ion o ui s emo ed by
ba s and he ee wi h inc eased an popula ions dec eased he
p opo ion o ui s aken by ba s. Bo h shi s we e in he expec ed
di ec ion.
-106-
Obse a ions o hese diu nally ac i e an s sugges ed ha a pa icula
noc u nal beha iou was esponsible o he changes in ui
consump ion. Du ing he day he an s we e dis ibu ed h oughou he
ees and on he g ound below. Some p opo ion o indi iduals o aged
o insec s while o he s collec ed "honeydew" om coccoideans (Insec a:
Coccoidea). I did no quan i y hei dispe sion in he ees, bu i
was my imp ession ha he coccoideans we e mos abundan on he
ui ing b anches and ui s han on o he pa s o he ees. An s
mo ed on o hese b anches, collec ed he "honeydew", and mo ed o
owa ds he nes dis ended wi h luid. This con inuous a ic esul ed
in he e being ela i ely ew an s on he ui ing b anches a any one
ime.. A nigh he an s did no o age and la ge numbe s emained on
and.a ound he ui s. On "hea y an " ees many o he ui s we e
comple ely co e ed and any noc u nal ugi o e (ba ) would ha e been
o ced in o con ac wi h la ge numbe s o hese agg essi e an s. Tha
he ba s had no e ec i e means o a oiding he an s i hey a emp ed
o eed on an -in es ed ees was e idenced by occasional cap u es o
ba s wi h O. longinoda a ached o he muzzle and eyes.
4.6.2 F. CAPENSIS ALONG GALLERY FOREST EDGE
Du ing Sep embe 1980 I ma ked 243 ui s on se en "no an " ees along
one galle y o es edge. The dispe sal cha ac e is ics o he ui
c ops on hese ees was ma kedly di e en om hose o equi alen
ees in he SNB (Table 25).
A signi ican ly g ea e p opo ion o ui s ell unconsumed along he
•
galle y o es edge (k=17.2%) han in he SNB (R=8.6%; U es , U=31.4,
n1=7, n2=19, p<0.05). Along he galle y o es edge ba s emo ed a
signi ican ly smalle p opo ion o ui s (R=30.7%) han in he SNB
(R=77.5%; U es , U=5, n
1
=7, n
2
=19, p<0.01) and diu nal ugi o es
emo ed a signi ican ly g ea e p opo ion (R=52.1%, U=13, n1
=
7, n2=19,
p<0.01). Ba s and diu nal ugi o es did no di e signi ican ly in
hei consump ion a his si e (U es , U=I3, n
1
=n
2
=7, p)0.05).
The lowe impac o ba s on F. capensis ui c ops along he galle y
o es edge in Sep embe ela i e o ha measu ed in he SNB du ing .
-107-
o he mon hs may in pa be due o a educed emphasis on F. capensis in
he ba s die du ing his pe iod. Al hough he o al aecal seed ain
measu ed in his zone did no decline o e ha o p e ious mon hs (see
Figu e 32), he p opo ion o F. capensis in he aeces declined om
100% in May/June and July/Augus samples o only 23.0% in Sep embe .
This was he lowes le el o any sampling pe iod (see Table 26).
Sep embe co esponded wi h he peak ui ing mon h o A. meigei along
' he galle y o es edge. Faecal samples showed ha he ba s ed
hea ily on his species in his mon h, sugges ing ha decline in ba
impac on he ui c op o F. capensis was due o a shi in he die
and p obably was only ansien .
4.6.3 ADENIA CISSAMPELOIDES ALONG GULERY FOREST EDGE
ma ked 100 ui s on a single A. cissampeloides plan in Sep embe ,
when he low ba impac on F. capensis was measu ed. On his species
ba s emo ed 81% o he ui s, diu nal ugi o es only 9%, and 11% o
he ui s ell unconsumed (Table 25).
4.6.4 FAECAL AND SEED RAIN
On ansec s along h ee galle y o es edges, collec ing shee s se ou
o a o al sampling e o o 9828 m
2
days yielded 121 aeces. In he
SNB an e o o 2220 m
2
days yielded 307 aeces (Table 26). In his
aecal ain, nine ui species we e iden i ied; namely A meigei,
B. e uginea, C. excelsa, F. capensis (which p obably included some
F. allis-choudae), N. la i olia, S. e basci olium, S. o um, and
V. doniana. All we e species known o be used by ba s. Only
F. capensis was ound a all si es du ing all he sampling pe iods. The
emaining species we e mo e seasonal. The ui species mos commonly
used by he mig an ba s, E. hel um and M. o qua a, we e only ound
du ing he mon hs ha hese ba s we e p esen . B. e uginea (lis ed
unde "seedless" in Table 26) and C. exceisa we e only ound in Ma ch
and No embe espec i ely, when E. hel um mo ed no h and sou h h ough
he LAMTO communi y. S. e basci olium was only ound in May/June and
Sep embe o No embe samples when M. o qua a was mos abundan a
cel
co cn c
p
•-1
I
I
I
I
I
I I
I
0000 0000CDC0-
•
—1 .--1
0
0
c
•
„
^cs
Q)
o
•
co
0
CO
u)
.•-
s
s ..--.
co cn
al ON
,..../ .....,
CO CO
I
I
-0
0 G
N
I G
N
I
.
•
CC
..--. ...-...
C')
cn
-
..
.
--.
5./ 41.„,./
IlliiIIII
I
I
I
I
I
I
I
I
I
I
I
III
11111
III
CO
U)
-"*".-.
'17
(1.1
Cl)
G)
'' '-
n-4
QJ
(1)
n
,. ..,
0-1 e
..,
c.)
c)
cn
......• .-..
co
a)
a)
›..,
ca,
Cl)
4-4
e)
En qi
a)
c ) a-1
c)
cd
.
-i
a) C..) CV
O
4-1
z.1 co
0
Cu
s
•,-4
ca
il -
.....
9-i L.)
Cl) Cl)
44
O
0
a)
H
O
2) -1
n
-1
.
-4 04
'0 'Ti 'V --
-4
.1
:
4
.
0
.1-1
Cl)
F-4
E-I
.....,
•
_-: 1.1 0 0 GA
CO E
A
E-
4
iz-1
cs.)
c
I
L(
L )OD
.--,
(11
•.../
..--.
U1
,...../
.
0
N.
•-4
1
,
..
CD
%0
0
CD
CD
c
y
)
CDOC.0
000000
.--.
--
••
•
CD
0
0
0
•
•
CC
p.so"'".
00 00
5.
CP%
C %
u")
CD
0
H 1-1
•
•
CC
z z
n
-•
CO
0,3
CN
)
Cl)
4.1
c1.1
Ci
E-4
!LI
<1
P=,
0
0
(I)
o)
E-,
0 0
C ,
P4
(.0
Ci
0
Cl:
!
H
cd N
cd
0
L
4
-
I
21
p
.
''
--
'
H
-I
....."
S..../
'-''--S
co
c N
s.../
%...,
...1.
Ci
4-1
0
a)
P.
4-1
4)
%.0
0
.-1
-I
CV
1
1
0
Ln
0
03
00
111
CO
CV
1/40
0 0
0
0
C 0
*--i '-'
4
C
s
i
•1.-4
CV
0
CN CO 1
,
4
0 0 0 0 0 CD 0 0 0 0 0 %.0
00
0 0 0 0 0 In
•
•
•
•
•
•
•
.../-
0
•
0
•
•
•
0 0 0
0 0
0
0
0 0 0 0
.1-1
cG
.
0
e
n
e
•
-
n
CD
01
C
)
...-.,
• ..I.
....
...
.
m
m
0
....
e.'s
e-s.
C
)
Lo,.,
•
nnn
•
••
nn
•
.
m
-•
..4.
m co
co
I-I
CJ
0.)
cO
1:14
•
C
o
o
C
s
4
-I
Cn
1/40
0
0
0 0 0
n
C)
0.)
--I
0
0
C.
)
0
,
-
.4
,-1
GNI
000000(000 OCD
n
O
d)
•
.
•
•
•
••
•
•
•
•
,--4
7:7
0
0 0 0 0 0
0
0
0 0 0
e•-
n
CI)
E-I
VD 0
C
n
1
$41
0
co
-I
—I
csi
cn
0
4-i
0 cp o
0 cp 0
0 -1 s--1
P.1
III
I
I
II
I
0 CD 0 0 0 0 CD 0 CD 0 0 ce)
II
II
I
III
•
•
•
•
•
c
n
!
Q)
0 0 0
0 0 0
C 0 0
bl)
/1 11
%.10
bO
a a
c."
<4
I
1
I
I
I
I
I
I
Cn
V)
c
)
cN1
es) 0 0 0
G
0
0
114
CO
in
<4
x-4
ag
H 0 C) C)
1•
4
-1
0
0
0
•
Ci
-4
0
CO
14-4
• n
•
H
Cl)
Ci
0.1
a)
14-1
c j
z P.
$-1
Cu
0
C.)
•
•
Cl)
C=4
FIGURE 32:" (A) The o al aecal ain, (B) he ain o F. capensis
aeces, and (C) he pe cen o he o al aecal ain ha
was F. capensis in he SNB and along galle y o es edge
h ough he sampling mon hs.
-
6
L.()
''.. .
...
-
El)
(N
0
LI)
(.
1
c
E! , D
n G
6
i
G
m
G d ,9 P
8
-
do
A a
zi m-oN
,
IC)
i l)
,---.
G
G
d
AVCI
zIAVoN
>--
C uj
UJ
Lo
< W
CO
LO
V)
.
6
c
p
o 6
6
,-
-
c,
• 1- )
6
c) N
,---
0
..._,
-.
,
‘--•
.a
-
....
-,"
-
;
-
n
Z
0
( )
_ <
--)
--)
2
<
2
Li.
-
--)
=
1.—
Z
0
1
IC)
5
d
,
a)
G
G
0
(4)
0
0
0
I )
NI
0
I
In
,---;
G
Li)
q
CD
7-
,
..... .
... - .-
s.,
-109—
LAMTO. Al hough S. e basci olium was ound o be ui ing in all
mon hs o he yea , seeds o his species we e no ound on he
collec ing shee s du ing mon hs when M. o qua a was no abundan .
On he galle y o es edge ansec s he aecal ain luc ua ed om
0.007 aeces/m
2
day in May/June o 0.015 aeces/m
2
day in No embe (Figu e
30), wi h a mean ain o 0.011*0.004 aeces/m
2
day. The aeces ell a
andom on he collec ing shee s o e he pe iod which hey we e
moni o ed (X
2
=2.1, d =3, p> 0.05) and he e was no e idence o a eas
wi h consi: en ly high aecal ain. In he SNB he aecal ain
luc ua ed om 0.032 aeces /m
2
day in No embe o 0.258 aeces/m
2
day in
May/June samples (Figu e 30), wi h a mean ain o 0.148*0.080
aeces/m
2
day. I did no eco d he dis ibu ion o aeces among
collec ing shee s in he SNB o acili a e a dispe sion analysis.
No all he aeces con ained seeds. On galle y o es edge ansec s
only 74.4% and in he SNB 97.1% o he aeces had some seed con en .
The seed ain a hese wo si es was 0.009±0.004 aecal seed
g oups/m
2
day and 0.142±0.083 aecal
seed
g oups/m
2
day espec i ely, i
is likely ha hese alues sligh ly unde es ima e he seed ain since
some seeds we e emo ed by an s. howe e , he collec ing shee s, aised
abo e he g ound su ace on he suppo ing g asses, did no appea o be
o en ised by an s and i was my imp ession ha seed p eda ion on he
shee s was minimal.
O he nine ni species ep esen ed in he aeces only F. capensis was
p esen in all sampling mon hs and a all si es. O e all,on he galle y
o es edge ansec s 53.3% (n=90) and in he SNB 94.0% (n=298) o he
seed—bea ing aeces we 2 om his single species. The p ope ion o .
F. capensis a ied seasonally om 100% a bo h si es o a low o 23.1%
along he galle y o es edge in Sep embe and 77.8% in he SNB in
Oc obe . The mean. F. capensis seed ain on galle y o es edge
ansec s was 0.005+0.003 aecal seed g oups/m
2
day and in he SNB was
0.132*0.075 aecal seed g oups/m
2
day. Gi en ha seed ain was andom,
a he measu ed
seed
ain e e y squa e me e o sa anna in he zone
bo de ing galle y o es s would ecei e 1.8 aecal seed g oups each
yea . I he ain was app oxima ely a -Qom in he SNB, each squa e
me e ecei ed 48.1 F. caceasis scea g oul.s
2ach
yea .
-.110-
BAT s BIRD INPUTS
Du ing Oc obe and No embe I isi ed he collec ing shee s a dawn and
, dusk in o de o quan i y he a e o diu nal and noc u nal inpu s.
Since he shee s we e se ou only unde open sky, inpu s we e om bi d
and ba o igins only. Table 27 shows he esul s o hese samples.
A all si es ba s accoun ed o he majo i y o he o al aecal ain
(wi h and wi hou seeds). In he SNB 92.7% (n=55) and on he wo
galle y o es edge ansec s 95.5% (n=44) and 92.1% (n=38) o he
aeces we e deposi ed by ba s. Conside ing he seed-bea ing aeces
only, he impo ance o ba s was e en g ea e . In he SNB 95.7% (n=46)
and along he galle y o es edge ansec s 100% (n=31) and 95.3% (n=24)
o he seed-bea ing aeces we e o ba o igin.
A signi ican ly smalle p opo ion o bi d aeces (33.3%, n=9) han ba
aeces (76.6%, n=128) ca ied seeds (X
2
=8.1, d
=
1, p<0.01).
4.6.5 SEED PREDATION
Faecal seeds se ou alone (simula ing seeds dispe sed in aeces) o
associa ed wi h ui pa s (simula ing aecal seeds and seeds in
ejec a pelle s le below eeding oos s) su e ed hea y p eda ion
du ing 24 hou p esen a ion expe imen s. The main seed p eda o s ha I
no ed we e an s, bu I did no iden i y he axonomic g oups in ol ed.
I o en in e up ed an s in he p ocess o emo ing seeds om he
p esen a ion dishes and i seemed likely ha gi en an addi ional ew
minu es all hese seeds would ha e been emo ed. Fo his eason I used
he emo al o mo e han one seed om a dish o indica e disco e y and
assumed ha once disco e ed all he seeds would he' p eyed upon.
. In 12 ials a signi ican ly g ea e p opo ion o "seeds plus ui "
dishes we e disco e ed han "seeds alone" dishes (Sign es ; p<0.01). A
mean o 91.7% o he "seeds plus ui " dishes we e disco e ed in 24
hou s, whe eas only 71.8% o "seeds alone" dishes we e disco e ed in he
same pe iod.
NOCTURNAL AND DIURNAL FAECAL RAIN
OCTOBER
NIGHT
DAY
NOVEMBER
NIGHT
DAY
OVERALL
NIGHT
DAY
SNB
ALL FAECES
35
1
16
3
51
(92.7)
4
(7.3)
FAECES WITH SEEDS
29
1
15
1
44 (95.7)
2 (4.3)
F. capensis FAECES
28
0
15
1
43 (97.8)
1
(2.2)
GALLERY FOREST EDGE
(SITE 1)
ALL FAECES
4
1
38
1
42 (95.5)
2 (4.5)
FAECES WITH SEEDS
3
028
0
31 (100)
0
F. capensis FAECES
2
0
13
0
15.(100)
0
GALLERY FOREST EDGE
(SITE 2)
ALL FAECES
7
1
28
2
35 (92.1)
3
(7.9)
FAECES WITH SEEDS
3
1
20
0
23 (95.8)
1
(4.2)
F. capensis FAECES
2
1
14
0
16
(94.1)
1
(5.9)
TABLE 27: The numbe o aeces alling on g ound-based collec ing-shee s
du ing he nigh and day in he SNB and along wo galle y.
o es edges du ing Oc obe and No embe . Pe cen s o he
o al ain a ein pa en heses.
Wol on, R.J., P.A. A ak,.H.C.J. God ay, & R.P. Wilson. 1982.
Ecological and beha iou al s udies o he Megachi op e a a Moun
Nimba, Libe ia, wi h no es on he Mic ochi op e a. Mammalia, in
p ess.
Wunde , B.A. 1975. A model o es ima ing me abolic a es o ac i e o
es ing mammals. J. Theo . Biol. 49:345-354.
Leck, C.F. 1969. Obse a ions o bi ds exploi ing a Cen al
•1
Ame ican ui ing ee. Wilson Bull. 81:264-269.
