The Foraging Ecology of Fruit Bats in the Seasonal Environment of Central Zambia

Home , Casinycteris, Epomophorus, Epomops, Micropteropus, Myonycteris, Rousettus

THE FORAGING ECOLOGY OF FRUIT BATS IN THE SEASONAL ENVIRONMENT OF CENTRAL ZAMBIA

HEIDI V. RICHTER

A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE

UNIVERSITY OF FLORIDA

2004

Heidi V. Richter

ACKNOWLEDGMENTS

Many people and organizations contributed financially, logistically, or personally to the success of this project. The Lubee Foundation provided funding and Dr. Allyson

Walsh was fundamental in securing those research funds. This research was also supported in part by a David H. Smith Fellowship awarded by The Nature Conservancy to Dr. Graeme Cumming. I thank John Seyjaget for initially sparking my interest in fruit bats, and helping me get the funding to follow through on my project.

In Zambia, I would like to thank the Kasanka Trust as well as Edmund and Kim

Farmer for allowing me to live, work and conduct research at Kasanka National Park.

The Zambian Wildlife Authority was extremely helpful in issuing the necessary permits

for this research. I thank Julian and Kelly Green for their hospitality when I was in

Lusaka and Tadg Wixted and his office for logistical support.

Changwe Kenneth Yotamu and Clifford Kandonga were of particular value in

catching bats, identifying trees and keeping me safe from harm. I thank Christopher

Miselo, Christopher Mwape, Nyambe, Kennedy, Benson, Robert Chisenga, Paul, Friday

Bwanga, Boas Mwape, Brighton, BV, Timothy Ndashe, Able, Shebbie, Kebby, Edson

Chipale, Chibesa Mwaba, J. Mwewa, and Haggai Mwape for their help in collecting data.

I thank Kennedy and George for their early morning help when I returned from a long

night of mist-netting.

Robin Lentz, Charlie Bear and Donna Bear–Hull were invaluable assistants in the

field, and contributed an amazing amount during their stay in Zambia. I cannot thank

iii them enough for traveling to Africa to assist with this project. I thank Charlie for

providing me with his photographs from Zambia and allowing me to use them in my

presentations. Alastair provided entertainment on some long mist netting nights.

Cristina Evangelista was very supportive of my research, traveling to Zambia to help with fieldwork and assisting with editing in the U.S. Without her initial influence I never would have ended up in Africa. My husband, John Howard, has been extremely supportive of my working in far away places.

I thank my labmates, especially Jeremy Dixon, Elina Garrison, Arpat Ozgul, and

Melissa Moyer, for their help, understanding, and guidance whenever I had questions and for their daily support and friendship.

Dr. Katie Sieving, Dr. Colin Chapman and Dr. Tom Kunz also deserve thanks for serving on my committee and providing comments on my methodology and manuscripts.

I thank Dr. Madan Oli for not only serving as a de factor committee member but also for spending his free time teaching statistics. I thank my advisor and chair, Dr. Graeme

Cumming, for supporting me with funding so that I could pursue this project. I would also like to thank him for his advice and guidance throughout this process.

TABLE OF CONTENTS

page

ACKNOWLEDGMENTS ...... iii

LIST OF TABLES...... viii

LIST OF FIGURES ...... ix

ABSTRACT...... xi

CHAPTER

1 INTRODUCTION ...... 1

Megachiropteran Assemblages...... 2 Effects of Seasonality in Resources...... 3 Migration Strategies in Fruit Bats...... 3 Resident and Migratory Species Interactions ...... 4 Project Objectives...... 5

2 FOOD AVAILABILITY AND ANIMAL MIGRATIONS: THE BEHAVIOR OF THE STRAW–COLORED FRUIT BAT AT KASANKA NATIONAL PARK, ZAMBIA...... 7

Methods ...... 10 Study Species...... 10 Study Site...... 10 Vegetation Monitoring ...... 13 Results...... 15 Species Results ...... 15 Vegetation Results...... 16 Discussion...... 23 Summary of Results ...... 23 Alternative Hypotheses ...... 24 Conservation Implications...... 28

3 DISTINGUISHING MEGACHIROPTERAN SPECIES USING MORPHOLOGICAL CHARACTERISTICS IN KASANKA NATIONAL PARK, ZAMBIA...... 30

Methods ...... 32 Study Site...... 32 Mist Netting Methods...... 32 Analysis of Species Data...... 34 Results...... 35 Capture Data...... 35 Megachiroptera Species Identification...... 37 Species assignments ...... 37 Summary of group analysis...... 38 Palate Data...... 40 Discussion...... 43 Identification of Sympatric Species...... 43 Mathematical Methods in Species Discrimination...... 45 Research Needs ...... 45

4 THE RESPONSE OF FRUIT BATS TO CHANGING RESOURCE AVAILABILITY IN KASANKA NATIONAL PARK, ZAMBIA...... 48

Methods ...... 50 Study Site...... 50 Mist Netting Methods...... 50 Bat Species Data...... 51 Vegetation Methods...... 51 Capture Rates...... 53 Results...... 54 Bat Species Data...... 54 Vegetation Analysis...... 54 Analysis of Capture Rates ...... 57 Capture rates and food availability...... 57 Analysis of variance...... 59 Mantel test results...... 59 Discussion...... 63 Additional Influences on Assemblage Structure ...... 64 Research Needs ...... 66 Conservation and Management Implications ...... 68

5 CONCLUSIONS, RESEARCH AND MANAGEMENT RECOMMENDATIONS 70

Conclusions...... 70 Recommendations for Further Research ...... 71 Management Recommendations...... 74

APPENDIX

A MORPHOLOGICAL DATA...... 77

B TREE SPECIES ENCOUNTERED IN THIS STUDY...... 79

C CAPTURE DATA...... 81

D MIST NETTING LOCATIONS...... 88

E VEGETATION TRANSECT LOCATIONS ...... 91

F FRUIT TREE LOCATIONS ...... 93

G HABITAT ANALYSIS DATA...... 102

H MAP OF VEGETATION TRANSECT AND MIST NETTING LOCATIONS .....123

I PHOTOGRAPHS OF EPAULETED AND DWARF EPAULETED FRUIT BATS NETTED IN THIS STUDY ...... 125

LIST OF REFERENCES...... 134

BIOGRAPHICAL SKETCH ...... 143

vii

LIST OF TABLES

Table page

2–1 Measurements on Eidolon helvum specimens collected at Kasanka...... 16

2–2 Records of feeding observations and ejecta pellets at fruiting trees ...... 18

2–3 E. helvum food sources, their fruiting times, and mean species diameter and height in 2003 ...... 21

3–1 Description of morphological measurement used in this study ...... 34

3–2 Reproductive classes used in this study ...... 34

3–3 Range of measurements for fruit bats trapped in this study...... 42

4–1 Summary statistics for the canopy tree species in 24 transects...... 56

4–2 Results of non–parametric one–way ANOVA on capture data ...... 60

4–3 Results of Wilcoxon rank test in non–parametric one–way ANOVA ...... 63

4–4 Results of Mantel Tests...... 63

A–1 Published and unpublished measurements of some sympatric African fruit bat species ...... 77

B–1 List of tree species found at Kasanka National Park...... 79

C–1 Data for individual Megachiroptera netted at Kasanka National Park in 2003...... 82

C–2 Data collected on Microchiroptera netted at Kasanka National Park in 2003 ...... 87

D–1 Dates, locations, and coordinates for mist netting sites ...... 89

E–1 Vegetation transect locations ...... 91

F–1 Coordinates of trees monitored for phenology in 2003...... 93

G–1 Location, size, and species of trees used in habitat analysis ...... 102

viii

LIST OF FIGURES

Figure page

2–1 Location of study site, Kasanka National Park, in Zambia...... 11

2–2 Map of Kasanka National Park ...... 14

2–3 Index of fruit availability at Kasanka as a function of time ...... 17

2–4 Food availability at Kasanka during the study period...... 19

2–5 Availability of fruit trees known to be food sources for E. helvum at Kasanka...... 20

2–6 Trends in fruit availability in transects from 7 September 2003 to 3 January 2004 ...... 22

2–7 Known E. helvum colonies in Africa and dates in residence at each site...... 23

3–1 Dendrogram produced from Ward’s cluster analysis on species data...... 39

3–2 Photographs of palatal ridge configurations identified in this study...... 40

3–3 Distribution of animals based on size and palatal characteristics ...... 41

3–4 Mean body mass and one standard deviation for two palatal configurations ...... 43

4–1 The number of fruiting trees as a function of distance from the E. helvum roost site ...... 58

4–2 Capture rates as a function of distance from the main E. helvum roost site...... 58

4–3 Capture rates as a function of the date and the number of fruiting trees in a transect ...... 61

4–4 Capture rates of fruit bats for four transects netted three times ...... 62

H–1 Map showing locations of vegetation transects and mist netting sites...... 124

I–1 Photo of Epomophorus gambianus crypturus showing facial detail and epaulets.125

I–2 Photo of Epomophorus gambianus crypturus...... 126

I–3 Full body photo of Epomophorus gambianus crypturus...... 127

I–4 Photo of Epomophorus labiatus or Epomophorus wahlbergi...... 128

I–5 Photo of juvenile Epomophorus labiatus or Epomophorus wahlbergi...... 128

I–6 Photo of Epomophorus labiatus or Epomophorus wahlbergi showing facial detail...... 129

I–7 Photo of Epomophorus labiatus or Epomophorus wahlbergi...... 130

I–8 Photo of Micropteropus pusillus...... 130

I–9 Photo of Micropteropus pusillus illustrating the small size and short rostrum...... 131

I–10 Photo of a small Micropteropus pusillus...... 131

I–11 Photo of Epomophorus minor illustrating relatively short rostrum length ...... 132

I–12 Photo of Epomophorus minor ...... 132

I–13 Photo of Epomophorus minor ...... 133

Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science

THE FORAGING ECOLOGY OF FRUIT BATS IN THE SEASONAL ENVIRONMENT OF CENTRAL ZAMBIA

Heidi V. Richter

December 2004

Chair: Graeme S. Cumming Major Department: Wildlife Ecology and Conservation

One of the most conspicuous examples of long distance movements in animals is the dramatic migration of large herds of African ungulates. Less recognized, but no less impressive, is the large–scale movement of huge numbers of African fruit bats. When the food availability in a region declines, multiple factors at the individual, population, and species level ultimately lead to migration. Although for some species the optimal strategy may be to move to areas with a greater food supply, others are limited in their ability to move and will utilize available local resources. The spatial scale of decision making depends on the energetic, physical, and environmental constraints on each species.

This study examined the causes of the annual migration of straw–colored fruit bats through a Zambian national park, and the effects of the migration on resident fruit bat species. From August 2003 through January 2004, I monitored woodland phenology and mist netted bats along 46 transects. I used these data to compare the timing of straw–

colored fruit bat migration to changes in local fruit availability and variation in resident

fruit bat assemblages.

The timing of migration of the straw-colored fruit bat coincided with the seasonal peak in fruit production in Central Zambia. Six fruit bat species were identified at the

site; cluster analysis and palate records were used to divide five similar epauleted fruit bat

species into four species groups. The distribution of these species was not significantly

correlated with the date, canopy height, habitat type or the amount of food at a site.

While indices of fruit availability decreased over time, capture rates increased. Fruit bats

were increasingly captured in habitats without food sources and that had not previously

been successful netting sites. Epauleted species resort to increasing commute and search

times when food availability decreases in order to meet their daily food requirements

while larger fruit bats migrate to new areas with a more abundant food supply.

African fruit bats have high economic and ecological importance, yet they remain

largely unstudied. In today’s world, with concerns of climate change and habitat

destruction, it is important to understand both the role of fruit bats in maintaining

functional ecosystems and the ecosystem requirements for healthy fruit bat populations to

reduce the impacts of human beings on megachiropteran species.

xii

CHAPTER 1 INTRODUCTION

Megachiroptera, or flying foxes, are known to feed on at least 188 plant genera in

16 families in Asia and Africa (Marshall, 1983). Around the world at least 289 plant species, producing more than 448 economically valuable goods, rely on fruit bats to some degree (Fujita and Tuttle, 1991). Africa is home to 12 families of pteropodid bats

(Nowak, 1997), but the lack of even basic knowledge about them constrains conservation efforts (Fenton and Rautenbach, 1998; Racey and Entwistle, 2003). Furthermore, the high human population in Africa, combined with poverty, minimum education, and pervading stigmas about bats, makes the conservation of bat species a significant challenge (Fenton and Rautenbach, 1998).

Central Zambia is unique in that it has both numerous bat species that are not sympatric in other areas and a migratory fruit bat colony that is unparalleled in its size.

No study has explicitly examined the factors that influence migratory timing in African fruit bats or the effects migratory bats have on resident fruit bat populations. Although we have some information on similar bat fauna (Fenton, 1975), virtually nothing is known about the fruit bat assemblages in Zambia, their seasonality, roost sites, migration habits, food sources, foraging patterns, or timing of parturition. The most recent comprehensive record of bat species for Zambia is from 1978 (Ansell, 1978), and no extensive sampling at the study site, Kasanka National Park, has been conducted.

My research addresses this lack of information about Zambian fruit bat assemblages and their ecology. I explored the spatial and temporal distribution of both

1 2

the fruit bat species and their food supply, and discuss differences in the scale of foraging movements in response to changes in food availability. I examine the relationship between seasonal food availability and migratory fruit bat species at Kasanka National

Park. Furthermore, I describe the resident fruit bat species at Kasanka, identify their food sources, and investigate the effects of food availability and habitat structure on foraging behavior.

Megachiropteran Assemblages

To understand the ecology of megachiropteran assemblages, we must examine the extent of overlap or exclusivity in resource use by the different species (Findley, 1976).

Ansell (1978) provides only basic records for fruit bat species in Zambia, and most studies from nearby regions are heavily biased towards insectivorous Microchiroptera

(Fenton, 1975; Fenton and Thomas, 1980; Findley and Black, 1983; Fenton et al., 1998).

This bias may exist because the identification of sympatric fruit bat species can be complicated (Ansell, 1960; Bergmans, 1988; 1989), and few publications offer direct comparisons of multiple species (Bergmans, 1988; 1989).

Bat ecology and behavior are shaped by the spatial and temporal distribution of foraging and roosting sites, and are limited by predation and competition (Patterson,

Willig and Stevens, 2003). It has been speculated that bat populations may be limited by feeding or roost sites, but no direct evidence for this is available (Fenton and Rautenbach,

1998), and roosting information for Southern African bat species is incomplete (Fenton,

1983; Fenton et al., 1985). Since chiropteran foraging patterns can be influenced by resource distribution (Heithaus and Fleming, 1978), temporal variability in the bat fauna and a highly seasonal fruit crop may influence the interactions among species (Thomas,

1983; Fenton et al., 1992).

Effects of Seasonality in Resources

Shifts in phenology and changing environmental conditions can affect primary consumers, including fruit bats, who respond to these changes by diet switching, seasonal breeding or migration (Schoener, 1971; van Schaik, Terborgh and Wright, 1993; Parrish,

1997; 2000; Racey and Entwistle, 2000). While some bat species selectively forage

(Bonaccorso and Gush, 1987), other bats are known to change food habits and preferences seasonally (Fleming, 1982; Thomas, 1982; Bergmans, 1990). Some species show "sequential specialization," concentrating on a limited number of prey items during different times of the year (Fleming, 1982), and these specialists can migrate to take advantage of seasonal fluctuations in different fruit and flower resources (Fleming, 1982;

Fleming and Eby, 2003). When animal species exploit seasonal asymmetries in resource availability by migrating, they generally do so through a need for the population to increase food availability, increase reproductive success, decrease competition, reduce exposure to parasites, or reduce predation (Elgood, Fry and Downsett, 1973; Lack, 1983;

Thomas, 1983; Fryxell, Greever and Sinclair, 1988; Berthold and Terrill, 1991;

Lindstrom, 1995; Loehle, 1995; Weber and Houston, 1997; Chesser and Levey, 1998;

Whitney and Smith, 1998; Bakun and Broad, 2003; Fleming and Eby, 2003).

Migration Strategies in Fruit Bats

Since movement is energetically expensive, migratory populations must weigh the benefit of a possible increase in food supply against the cost of traveling to a new location (Schoener, 1971; Pyke, Pullman and Charnov, 1977). Megachiroptera in

Australia and Africa exploit seasonal changes in fruit production by migrating long distances (Thomas, 1983; Eby, 1991; Spencer, Palmer and Parry-Jones, 1991; Palmer and

Woinarski, 1999; Tidemann and Nelson, 2004). What makes movement beneficial for

African megachiroptera is that seasonal rainfall patterns lead to short, but high volume,

bursts in fruit and flower production that make savanna habitats richer in food resources,

on a per capita basis, than forest habitats (Morrison, 1978a; van Schaik et al., 1993;

Hepburn and Radloff, 1995; Fleming and Eby, 2003).

In Zambia, Eidolon helvum is the most conspicuous migratory megachiropteran

species, yet little is known about its migratory ecology either locally or throughout Africa

(Fenton and Rautenbach, 1998). E. helvum is highly mobile and capable of migrating

long distances, but the forces driving its migration are unstudied (Thomas, 1983;

Kingdon, 1984; DeFrees and Wilson, 1988; Nowak, 1994; Cosson, Tranier and Colas,

1996). It has been hypothesized that E. helvum is an opportunist, migrating to take

advantage of high regional food supplies and to increase reproductive success (Baker,

1973; Fenton, 1975; Thomas, 1983; DeFrees and Wilson, 1988; Taylor and Kankam,

1999). E. helvum could also migrate to avoid unfavorable climate or food shortages in

the breeding territory (Cohen, 1967), to seek limiting nutrients (Thomas, 1984), to reduce

the effect of disease or parasites on the population (Loehle, 1995), or teach young how to

forage and feed (Kingdon, 1984).

Resident and Migratory Species Interactions

No studies have been conducted in Africa to investigate the interactions between

resident and migratory species competing for highly seasonal resources and the

complicated spatial and temporal interactions that may result. At Kasanka, three major

interactions should be investigated. The first is the colonial behavior of E. helvum, and

whether the large colony is used for information sharing, to optimize the group use of

clumped resources, reduce predation, or as a mechanism for dispersal to reduce

intraspecific competition (Horn, 1968; Brown and Orians, 1970; Hamilton and Watt,

1970; Ward and Zahavi, 1973; Heithaus and Fleming, 1978; Fleming, 1982; Thomas,

1982; Fryxell et al., 1988; Fleming and Eby, 2003). Each explanation for refuging behavior can have effects on, or be affected by, resident fruit bat species.

Secondly, resident fruit bat colonies are smaller in size than the E. helvum colony at

Kasanka and are probably more widespread. Due to the smaller body size of resident fruit bat species and their energetic constraints, resident species may be restricted to a smaller foraging area, and could be affected by the introduction of more individuals into feeding sites. However no information is available about roost sites, colony sizes, predation, or foraging behavior for resident bat species. A better general understanding of the biology of resident species is needed to help establish how the introduction of

migratory species may affect the behavior of resident species.

Finally, it is important to understand the interactions between the resident and

migratory fruit bat species and how the behavior of one may affect the other. For

example, if food resources are limited, species may geographically, temporally or

vertically partition the available resources. The E. helvum colony may help the other bat

species by satiating predators. It is unknown how a large refuging colony may interact

with smaller and scattered fruit bat colonies. E. helvum could disperse large distances to

reduce intraspecific competition (Heithaus and Fleming, 1978; Bonaccorso et al., 2002),

or they may remain close to the roost, decrease commuting time, and potentially face

increased inter– and intraspecific competition.

Project Objectives

The goal of my research was to collect basic biological information about the bat

fauna at Kasanka National Park, Zambia, focusing on the different foraging strategies

fruit bats use when faced with a highly seasonal environment. Chapter 2 examines the

hypothesis that local and regional fruit availability governs the timing of migration by

Eidolon helvum. The objective of Chapter 3 is to identify the resident bat species that

may be interacting with migratory species. Chapter 4 explores the relationship between both resident and migratory fruit bat species and their environment, with emphasis on the habitat variables and changes in food availability that may influence foraging behavior.

To accomplish these goals, I first needed to identify the primary plant species used as food sources by E. helvum and other fruit bat species, monitor phenology at the study site, and examine the relationship between fruit availability and the timing of migration.

Second, I needed to identify the species composition of the fruit bat assemblage at

Kasanka, collect fruit availability and habitat structure data in foraging areas, and then

search for factors that could predict the spatial and temporal distribution of fruit bats at

foraging sites.

CHAPTER 2 FOOD AVAILABILITY AND ANIMAL MIGRATIONS: THE BEHAVIOR OF THE STRAW–COLORED FRUIT BAT AT KASANKA NATIONAL PARK, ZAMBIA

Many animal populations in variable environments have developed the ability to

exploit temporal and spatial changes in resource availability. Exploiting variability in

complex ecosystems can confer advantages to organisms, such as high reproductive

success, that are unexplained by trophodynamics (Bakun and Broad, 2003). Some

fluctuations in biological controls are stochastic, like the El Niño Southern Oscillation,

whereas others are predictable and more regularly exploited. Seasonal fluctuations in the

spatial distribution of resources can drive large–scale movements of animals that in turn

can have ecologically important local and landscape level consequences(Polis, Anderson

and Holt, 1997).

One group that has evolved to exploit seasonal fluctuations in food supply is the

Megachiroptera. In Australia and Africa, fruit bats respond to seasonal changes in fruit

production by migrating as far as 1500 km (Thomas, 1983; Eby, 1991; Spencer et al.,

1991; Cosson et al., 1996; Palmer and Woinarski, 1999; Fleming and Eby, 2003;

Tidemann and Nelson, 2004). In this paper I focus on Eidolon helvum [Kerr, 1792], an

African species known to undertake seasonal migrations. It has been hypothesized that

E. helvum is an opportunist, migrating to take advantage of variations in regional food supplies and to increase its reproductive success (Jones, 1972; Baker, 1973; Fenton,

1975; Thomas, 1983; Kingdon, 1984; DeFrees and Wilson, 1988; Taylor and Kankam,

1999). E. helvum may shift its diet and track changes in fruit availability, thereby

7 8 reducing inter– and intraspecific competition by exploiting more abundant food supplies

(Thomas, 1982; Cox, 1985; Bergmans, 1990; Levey and Stiles, 1992).

Several other factors may also be driving these movements. E. helvum could migrate to avoid unfavorable climate or food shortages in the breeding territory (Cohen,

1967), to seek limiting nutrients (Thomas, 1984), or to reduce the effect of disease or parasites on the population (Reeve, 1988; Loehle, 1995). Migratory colonies may exploit low predator levels or swamp local predators by aggregating in large numbers (Fryxell et al., 1988).

Another possible explanation is that the movement of E. helvum is not driven by food availability, but by the socio–sexual behavior of the bats in the colony (Kingdon,

1984). Breeding may hold large colonies together, and the break up of these colonies may indicate a decline in the males’ sexual drive (Kingdon, 1984). Female E. helvum may fulfill parental or neonatal demands by migrating to areas with a short–term superabundance of food (Cumming and Bernard, 1997; Racey, 2002). The survival of neonatal bats may be an important factor in the timing of migration, with weaning coinciding with the period of highest food availability (Fayenuwo and Halstead, 1974;

Cumming and Bernard, 1997). Adults may use the period of high food availability to more effectively teach young bats how to forage and find food (Kingdon, 1984).

In this paper I consider the first hypothesis, that E. helvum migrates in large numbers to opportunistically exploit seasonal variations in food availability. While food availability may be important at multiple scales, in this study I focus explicitly on the role of local food availability and the way in which it changes during a stopover by a migratory colony. This study was conducted from August 2003 through January 2004 at

Kasanka National Park in Zambia. Kasanka is an appropriate site for examining the

foraging behavior of migratory fruit bats because it serves as a seasonal roost site, from

October through January, for an estimated eight to ten million migratory E. helvum.

If food availability is an important driver of the E. helvum migration, the colony should arrive at Kasanka when food is most abundant and depart when food availability starts to decline (Katz, 1974; Charnov, 1976; Pyke et al., 1977). To reduce commuting time and conserve energy, individuals from the colony should forage close to the roost and deplete those resources before exploring more distant food resources (Hamilton and

Watt, 1970; Morrison, 1978b; Aronson and Givnish, 1983). Alternatively, if food resources have a patchy distribution, foraging activity should follow the spatial distribution of resources, and aggregations of animals should be found where food availability is higher than some threshold value (Arditi and Dacorogna, 1988). If E. helvum encounters limited food resources it may use foraging strategies to reduce

competition with resident bat species, such as resource stratification or the use of feeding

roosts (Thomas, 1982; Putman, 1994).

I compared local phenology with fruit bat behavior to test the food availability hypothesis. The results provide the first quantitative evidence to support the hypothesis

that the migration of E. helvum in Zambia is driven by food supply, although this

approach does not allow alternative explanations for this species’ migratory pattern to be

ruled out.

Methods

Study Species

Eidolon helvum, the straw–colored fruit bat, is the second largest fruit bat on the

African continent. E. helvum weighs approximately 250–310 grams and has an average wingspan of 80 cm (Bergmans, 1990; Taylor, 2000). While its primary habitat is equatorial Africa, its migratory range extends from Sub–Saharan Africa to South Africa

(Kingdon, 1984). E. helvum is a strong flyer built for endurance rather than mobility, and its body structure supports long migrations while restricting much of its foraging to the upper canopy layer (Thomas, 1983).

Five bats were captured by hand in the roost and used to collect morphological, sex and reproductive condition data. These data were collected to establish the reproductive and physical condition of individuals in the colony. Dispersal and foraging data are observational because even though E. helvum were easily grabbed by hand in the roost

site, mist netting (discussed in Chapter 3) failed to trap any bats of this species.

Study Site

This study was conducted at Kasanka National Park in Central Zambia, a privately

run national park that relies on a large migratory E. helvum colony to draw tourists and

their associated revenue. The park is managed by the Kasanka Trust, a private UK based

organization, under an agreement with the Zambian Wildlife Authority. Kasanka is one

of the smallest of Zambia’s 19 national parks with an area of 420 km2. It is located in

northern Central Province (12º30’S 30º14’E) and measures approximately 35 km from

east to west and 15 km from north to south (Figure 2–1). The region has undulating low

hills with an altitude of around 1200 m. Kasanka experiences unimodal rainfall totaling

approximately 1200 cm a year, with rains usually falling between November and April.

Kilometers

0 1200 1600

Kilometers 0200100

Figure 2–1. Location of study site, Kasanka National Park, in Zambia.

The landscape mosaic is composed of a matrix of miombo woodland intermixed with wide, grassy dambos and small stands of chipya forest. Miombo woodland is dominant both in Zambia, where it covers about 80 percent of the country, and in the general Zambezian region. The dominant woodland subtype, northern wetter miombo, is widespread in the northern areas of Zambia including northern Central and Copperbelt provinces, as well as Northwestern, Luapula and most of Northern Province (Chidumayo,

1987). It is dominated by leguminous Brachystegia, Isoberlinea, and Julbernardia species interspersed with other widespread trees including Uapaca, Protea, and Faurea.

The character of miombo can vary based on soil characteristics, and at Kasanka it ranges from a single story scrub to a tall woodland with a canopy of 20 m or more (Smith and

Fisher, 2001).

Associated with miombo woodland, and interspersed throughout the park, are

seasonally waterlogged, large grassy depressions called dambos (Smith and Fisher,

2001). Dambos and riverine grassland provide important grazing habitat for mammals in the park, but contain few trees. The Kasanka, Luwombwa, and Mulembo Rivers and their associated floodplains provide water to the area throughout the year.

Chipya forest is found in small patches in mosaic with miombo woodland throughout Kasanka. It is a wooded grassland dominated by fire resistant tree species such as Terminalia mollis, Erythrophleum africanum, and Combretum spp. Chipya is notable because of the absence of Uapaca and other common miombo fruiting genera.

Mateshi forest is a dry evergreen forest that in Kasanka exists only as small relics (Smith and Fisher, 2001).

Another small, but key, forest type at Kasanka is represented by only two small patches covering an area of about 0.4 km2. The “mushitu” evergreen swamp forest has

dominant tree species including Khaya nyasica, Parkia filicoidea, and Diospyros

mespilformis (Smith and Fisher, 2001). Swamp forests occur throughout the higher rainfall areas from Mwinilunga to Luapula and Northern Provinces (Storrs, 1995). They grow along streams and rivers as well as in swampy areas where the water table is at or near the surface throughout the year. Mushitu has three canopy layers with a closed evergreen canopy up to 27 m, an understory between 10–18 m and below this a dense thicket consisting mainly of ferns and climbers (Storrs, 1995; Smith and Fisher, 2001).

Syzygium cordatum (Waterberry) and Khaya nyasica (African mahogany) dominate the taller canopy layer, while Ficus trichopoda (Swamp fig) dominates the understory.

The mushitu forest is important because it is the only known roost site in the region for a large migratory Eidolon helvum colony. Large portions of the colony can be found roosting in either canopy layer at any given time and bats also shift roost sites throughout

the forest area. From late October through late December the colony disperses at sunset

into the surrounding miombo woodland to feed, with bats flying 30 km or more to

feeding areas.

Vegetation Monitoring

Vegetation transects were established and monitored starting 7 September 2003.

Transects were placed starting at the eastern park boundary at the Mulaushi Stream and

extended to the western edge, where the Luwombwa river forms a natural boundary

(Figure 2–2). Transects were located about one kilometer apart on primary dirt roads and

crossed approximately 45 km of park land. Road transects were used to increase

efficiency of monitoring and for personal safety. Every fifth transect was paired with an

offset transect 100 m in the bush to allow statistical comparison of roadside vegetation with off–road vegetation to test for road effects. Road transects were 100 m long and included five meters on either side of the road. Off–road transects were located perpendicular to the road, 100 m into the bush and were 10 m wide by 100 m long.

All fruiting trees and transect locations were marked with identification numbers using wire mounted aluminum tags (Corstor Ltd, South Africa). A fruiting tree was considered to be any tree with fruit, where the fruit bearing branches were in a transect, even if the base of the tree was not. In every transect I recorded the location (using a

Garmin GPS12 receiver), diameter at breast height (dbh), and species of each fruiting tree

and shrub taller than one meter.

N Luwombwa River

M u le m b o

Bwalyabemba Luwombwa Hill Mulembo River Musande

Chikufwe Wasa

Pontoon Mulaushi er Kafubashi r iv e R iv Wasa II a R Mpululwe w b a M Fibwe k Hill om u w n s u a o L s la a r K R e iv iv e R r i Shiwila h s 010 u Kilometers la u M

Figure 2–2. Map of Kasanka National Park. The star shows the location of the E. helvum roost. Dotted lines show southern and western park boundaries. The Luwombwa and Mulembo Rivers form the northern boundary while the Mulaushi River forms the Eastern boundary.

I measured the presence and availability of fruit in each transect using two methods. The first was by noting the presence or absence of fruit on each tree. The second method involved assigning each tree an index of fruit availability, ranging from a score of zero, for trees with no fruit, to a score of four, for trees with very abundant fruit

(Chapman, Wrangham and Chapman, 1994). While scores are subjective fruit indices, the results were reliable and consistent for the two observers. After the disappearance of

all fruits, trees were monitored for two weeks to ensure that the first ‘0’ was not due

solely to observer error. I also noted which fruiting trees showed signs of feeding by fruit bats, and marked and monitored feeding roosts.

Transects were monitored from 7 September 2003 until 5 January 2004. Each transect was checked every 7–10 days, for a total of 16 consecutive measures. A total of

350 trees were tagged and monitored for the presence of fruit in 45 transects. Trees and shrubs were identified by Zambian game scouts using local names, and identifications were confirmed using field guides (Fanshawe, 1984; Storrs, 1995; Coates Palgrave,

2002).

Food sources for E. helvum were compiled from observations collected in 2000 and

2003. Game scouts were provided with notebooks to record encounters with fruit bats.

The scouts noted where and when they saw fruit bats, how many there were, whether the bats were feeding or flying, and if the bats were feeding the scouts recorded what fruits the bats were feeding on. These records were combined with personal observations of feeding bats, the presence of ejecta pellets under fruiting trees in the vegetation transects,

and seeds and ejecta pellets found under feeding roosts. I collected samples from

unidentified food sources in 2000 and identified the species using reference samples at

the University of Zambia herbarium.

Results

Species Results

E. helvum were first sighted at Kasanka on 18 October 2003, and all bats had left

the park by 24 December 2003. The first storm arrived on 17 October, although the area

did not receive significant rainfall again until mid–November. The arrival and departure

dates of the colony conform to trends displayed in six previous years. In 1997, 1998,

1999, and 2001, E. helvum were first sighted on 21 October, while in 2000 the first bats

were spotted on 19 October. In the 2000 season the colony departed 9 January (2001),

while in 2001 all bats had departed by 23 December.

Five E. helvum were captured by hand in the roost site on 16 November 2003 – two

pregnant females and three adult males (Table 2–1). In 2000, all 10 female bats collected

from the roost site were pregnant (Stuart and Stuart, 2001). In 2002, Paul Racey

collected 18 female bats, of which 2 were lactating and 14 were pregnant (Racey, 2002).

Table 2–1. Measurements on Eidolon helvum specimens collected at Kasanka. Reproductive Weight Tail Body Wingspan Forearm Foot Ear Sex Condition (g) (mm) (mm) (mm) (mm) (mm) (mm) M Reproductive 255 14 195 720 117 33 29 F Pregnant 265 17 205 700 121 30 29 F Pregnant 295 10 205 660 123 33 29 M Reproductive 280 13 205 700 126 35 26 M Reproductive 260 15 200 740 119 33 28

Mist netting 24 sites on 23 nights, for a netting effort of 1970 net meter hours (one

net meter hour equals one meter of net open for one hour), yielded no E. helvum within

15 km of the roost. This result was interesting and unexpected since E. helvum were

observed in large numbers at netting sites and in total over 120 other fruit bats, from five species, were captured. (Results from mist–netting are presented in Chapter 3.)

Vegetation Results

Comparison of road and off–road transects using the Wilcoxon rank sum test showed no significant difference in the mean number of fruiting trees (n=14, Z=0.261, p=0.795). Fruit availability, whether measured using presence / absence data or the sum of scores for all trees, showed an identical response over time (Figures 2–3 and 2–4). For

clarity only the graphs illustrating the presence or absence of fruit on trees are shown in

the rest of the text.

600

500

400

300

Fruit Index Fruit 200

100

3 3 3 3 3 3 3 3 3 /0 /0 /0 /0 /0 /0 /0 /0 /0 /7 1 /5 9 /2 6 0 4 8 9 /2 0 /1 1 /1 /3 /1 /2 9 1 0 1 1 1 2 2 1 1 1 1 1 Date

Figure 2–3. Index of fruit availability at Kasanka as a function of time. The fruit index equals the sum of scores for all fruiting trees at each time period. The E. helvum colony arrived on 18 October 2003 and departed 23 December 2003; these dates are indicated by open squares.

Fruit availability increased starting in early October until it plateaued, remaining constant throughout November and decreasing at the end of that month. Food availability rapidly decreased throughout December until almost no fruiting trees were found at the beginning of January. The colony arrived immediately before peak fruit production, and left when fruit availability was very low, but before it was exhausted (Figures 2–3 and 2–

4).

Observations of feeding bats and ejecta pellets at fruiting trees yielded 297 records

that were used to compile the list of food sources for E. helvum (Table 2–2). Main food

sources included Ficus spp., Musa spp., Magnistipula butayeii, Parinari curatellifolia,

Syzygium spp., and Uapaca spp. The majority of both tree species and individual fruiting

trees that were fruiting from October through December are food sources for E. helvum

(Figure 2–4).

Table 2–2. Records of feeding observations and ejecta pellets at fruiting trees. Data are from feeding records collected by game scouts (Scout Records), presence of ejecta pellets under fruiting trees and at feeding roosts (Ejecta Pellets), or from personal observation of bats feeding at trees (Personal Obs). Data include the number of observations of bats at each tree species (# Obs), as well as the percent of all feeding records for the tree species each year (%). Scout Records Ejecta Pellets Personal Obs 2000 2003 2003 2000 2003 Tree Species # Obs % # Obs % # Obs % Ficus spp. 0 0.0 1 2.0 0 0.0 x x Magnistipula butayeii 15 10.0 7 16.0 0 0.0 Mangifera indica 1 <1.0 0 0.0 0 0.0 Musa spp. 0 0.0 2 5.0 0 0.0 x Parinari curatellifolia 18 12.0 3 7.0 0 0.0 x Syzygium cordatum 0 0.0 0 0.0 4 4.0 x Syzygium guineense guineense 4 3.0 6 14.0 59 59.0 x Syzygium guineense huillense 1 <1.0 15 34.0 13 13.0 x x Uapaca kirkiana 40 26.0 9 20.0 1 1.0 Uapaca banguelensis 32 21.0 1 2.0 18 18.0 Uapaca sansibarica 42 27.0 0 0.0 5 5.0 x

Four E. helvum food sources were relatively abundant while the colony was at

Kasanka: Syzygium guineense guineense, Syzygium guineense huillense, Uapaca

kirkiana, and Uapaca banguelensis. Uapaca and Syzygium species showed a strong

seasonal response, and the departure of the colony coincided with a marked decrease in

the availability of these food sources (Figure 2–5).

Fruit bats at Kasanka feed at trees covering a wide range of sizes, from the shorter

4 m Syzygium to expansive Parinari that were, on average, 11 m tall (Table 2–3). The

most abundant food source, Syzygium guineense guineense, is a small shrub and while

abundant in number, each plant typically carries less fruit biomass than the other species.

300 All Fruiting Trees

250 Food Sources Only

200

150

100

50 Number of Fruiting Trees

3 3 3 /0 03 0 /03 /7 6/ 4/ 9 /21/0 19/03 1 1 28 9 10/5/03 0/ 11/2/03 1/ 2/ 2/ 1 1 11/30/03 1 1 Date

Figure 2–4. Food availability at Kasanka during the study period. Solid curve indicates the total number of fruiting trees and the dotted curve includes only those trees known as food sources for E. helvum. The E. helvum colony arrived on 18 October 2003 and departed 23 December 2003; these dates are indicated by open symbols.

While ejecta pellets were found at many fruiting trees beginning in October, the first feeding roosts were not found until 24 November 2003, coinciding with the initial decrease in food availability in the area (Figure 2–4). Eight feeding roosts were found in the vegetation transects, and seeds and ejecta pellets were found from Uapaca kirkiana and Syzygium guineense guineense under these trees. Feeding roosts were located in

Brachystegia spiciformis, Isoberlinia angolensis, and Julbernardia paniculata trees.

100

40 20 30 Number of Fruiting Trees 20

0 9/7/03 9/21/03 10/5/03 10/19/03 11/2/03 11/16/03 11/30/03 12/14/03 12/28/03 Date Syzygium guineense huillense Syzygium guineense guineense Parinari curatellifolia Uapaca nitida Uapaca sansibarica Uapaca kirkiana Uapaca banguelensis Syzygium cordatum

Figure 2–5. Availability of fruit trees known to be food sources for E. helvum at Kasanka. The Uapaca kirkiana and Uapaca banguelensis lines exclude trees bearing unripe fruits.

E. helvum exhibit refuging behavior at Kasanka, and disperse from a central place

along radial lines to their feeding grounds. If E. helvum selectively forage to minimize

travel time and maximize energy gain, resources close to the roost site should be depleted

before the bats travel farther to feed (Pyke et al., 1977; Morrison, 1978b; Aronson and

Givnish, 1983). At the distances measured in this study, transects close to the roost did

not show decreases in the indices of fruit availability before those transects further away

(Figure 2–6). The majority of the colony each evening was still dispersing at distances

greater than 15 km from the roost site and showing high fidelity to radial flight lines. I

found no evidence that E. helvum at Kasanka is minimizing travel time by foraging close

to the roost before commuting longer distances.

Table 2–3. E. helvum food sources, their fruiting times, and mean species diameter and height in 2003. Values are mean and one standard deviation, DBH is diameter at breast height, HT is height. Tree Species Local Name Begin Fruiting End Fruiting DBH (cm) HT (m) Parinari curatellifolia Mupundu August Early November 64.4 ± 32.0 11 ± 5 Uapaca kirkiana Masuku Late August Early to mid Jan. 10.5 ± 4.0 6 ± 1 Uapaca nitida Nsokolobe Late August Early December 10.7 ± 3.0 6 ± 1 Uapaca banguelensis Makonko Late August Late December 9.5 ± 3.1 6 ± 1 Uapaca sansibarica Swebya Late August Early to mid Dec. 16.2 ± 7.4 7 ± 2 Syzygium cordatum Mufinsa Mid October Early to mid Dec. 22.9 ± 7.9 10 ± 2 Syzygium guineense Insafwa Late September Late December 7.2 ± 3.1 4 ± 1 guineense Syzygium guineense Mufinsa Mid October Early to mid Jan. 18.1 ± 11.7 5 ± 1 huillense

Many other E. helvum colonies disperse to form smaller colonies during the wet

season; however the timing of the arrival of E. helvum in Central Zambia does not

coincide with the dispersal of other large, well–known colonies (Figure 2–7). The annual

disappearance of an E. helvum colony in Mauritania closely precedes the arrival of the

Kasanka colony (Cosson et al., 1996), but because of the large distances involved it is unlikely that this is the same colony.

7 Sept 03 3 Oct 03

60 60

40 40

20 20

0 0 051015 0 5 10 15

26 Oct 03 18 Nov 03 Trees

g 60 60

40 40

20 20

0 0 0 5 10 15 0 5 10 15 Number of Fruitin

10 Dec 03 3 Jan 04

60 60

40 40

20 20

0 0 051015 051015

Distance to E. helvum Roost (km)

Figure 2–6. Trends in fruit availability in transects from 7 September 2003 to 3 January 2004. X–axis is km from the E. helvum roost; Y–axis is the number of fruiting trees of known food sources. Since transects were sampled over multiple dates within a short range, date is midpoint of the sampling period.

June July Aug. Sept. Oct. Nov. Dec. Jan. Feb. March April May Zambia

Malawi

West Africa

Uganda

Ivory Coast

Mauritania

Kasanka National Park

Figure 2–7. Known E. helvum colonies in Africa and dates in residence at each site (Mutere, 1967; Ansell, 1978; Thomas, 1983; DeFrees and Wilson, 1988; Bergmans, 1990; Cosson et al., 1996).

Discussion

Summary of Results

Migratory E. helvum may be tracking the fruits of the genus Syzygium and the

genus Uapaca on their annual migration to Central Zambia. The arrival of the colony at

Kasanka coincided with a dramatic increase in the number of fruiting Syzygium and

Uapaca; these trees accounted for greater than 80 % of all fruiting trees at that time. The colony departed immediately before the Syzygium and Uapaca fruit crops were depleted.

Syzygium is a known food source for E. helvum in Sub–Saharan Africa; however,

foraging on Uapaca may be a new example of diet switching in the region (Marshall,

1985). Correlation does not equal causation, and the actual relationship between the

colony’s movements and food availability is not apparent.

The arrival of the E. helvum colony at Kasanka is historically more predictable than the onset of rains and the associated rise in indices of fruit availability. The departure of

the E. helvum colony on 23 December 2003 coincided with decreasing trends in local fruit indices, but in previous years the colony has left as late as the second week of

January. This range in departure dates may be a function of varying resource depletion rates depending on the year’s fruit crop quantity and quality; however, no phenology data for Kasanka is available for previous years.

The lack of E. helvum in mist nets could be due to the bats flying further than 15

km before descending to feed, or to their feeding only in the upper canopy when staying close to the roost. E. helvum has been found to feed primarily in the upper canopy in some African regions (Thomas, 1982). Colonial roosting bats may make long commuting flights to reduce competition with conspecifics (Bonaccorso et al., 2002), and it may be that flying is energetically inexpensive for E. helvum, hence they can afford to fly further away to forage (Horner, Fleming and Sahley, 1998). However reducing competition with conspecifics by increasing commuting time should not necessarily decrease competition from other fruit bat species. It may be that as with some other bat species foraging decisions are made without the explanations for their behavior being

obviously apparent (Ayensu, 1974; Horner et al., 1998).

Alternative Hypotheses

The capture of both adult males and pregnant females in 2000 (Stuart and Stuart,

2001), 2002 (Racey, 2002), and 2003 clearly shows that the E. helvum colony at Kasanka

is not composed of only non–breeding 1–2 year–old individuals as previously

hypothesized (Sorensen and Halberg, 2001). While mating has been observed at

Kasanka, the main purpose of the large colony may be to increase parental or neonatal

nutrition (Cumming and Bernard, 1997), not to serve as a mating colony (Kingdon,

1984). Although the sample size in this study was small, data collected in previous years

supports this conclusion (Stuart and Stuart, 2001; Racey, 2002). If the adults are

migrating to teach their young to forage and find food (as hypothesized by Kingdon

(1984)), we would expect to find young at Kasanka. Some young were observed in the

Kasanka colony, but sightings were rare. Combined, these results suggest that E. helvum migrate through Kasanka to exploit the seasonal fruit resources to fulfill the energetic demands of pregnancy and lactation, and to support the growth of juveniles that may be weaned at that time (Thomas and Marshall, 1984; Racey, 2002; Fleming and Eby, 2003).

Further alternative hypotheses remain to be tested, including whether E. helvum is reducing inter– and intraspecific competition by exploiting a surplus in food supplies

(Thomas, 1982; Bergmans, 1990). The appearance of feeding roosts, coinciding with decreasing food availability, may indicate that competition for food resources is occurring at Kasanka. Animals may be removing fruits from trees and moving to feeding roosts to decrease interference competition (Putman, 1994; Morin, 1999).

By studying roost sites at locations other than Kasanka we may find that E. helvum is migrating to avoid unfavorable climate or food shortages in the breeding territory

(Cohen, 1967). It is also possible that the pregnant females in the colony are migrating to obtain limiting nutrients (Thomas, 1984; Dumont, 2003; Nelson, 2003). Analysis and comparison of the nutritional content of E. helvum food sources at Kasanka with other known food items could help answer this question.

It is surprising that with the substantial size of this E. helvum colony it has gone unnoticed elsewhere on the African continent. This roost cannot easily be explained by the vacating of other breeding colonies (Sorensen and Halberg, 2001) since its timing does not coincide with other large groups. The only colony that is recorded to migrate

during this time is in Mauritania, about 6000 km away, and is composed of only several

hundred individuals (Cosson et al., 1996). The migratory colony at Kasanka may have simply been undetected until now on the rest of its migration route, although I believe a

second explanation is more likely. The degree of asynchrony in reproductive timing is

evidence that the colony at Kasanka may be composed of smaller satellite colonies from

disparate African regions. E. helvum frequently disperses when rains arrive in an area

(Kingdon, 1984), and if multiple satellite colonies are congregating at Kasanka the

explanation for this behavior is unclear.

If the colony is migrating to reduce the effect of disease or parasites on the

population, we would expect the bats to form smaller colonies that would reduce, rather

than increase, probabilities of transmission (Loehle, 1995). Large aggregations of bats

may confer other advantages, including serving as an ‘information center’ where animals

congregate to transfer information about feeding sites (Ward and Zahavi, 1973). This

hypothesis could be tested by following individual bats and monitoring whether those

bats that are unsuccessful in foraging one night follow other bats to new foraging areas.

Coloniality can confer other advantages as well, including group defense against

predators, group defense of feeding areas, or to more efficiently exploit resources (Brown

and Orians, 1970).

It has been suggested that predation on E. helvum in general is relatively infrequent

and seems to pose no threat to the population (Jones, 1972; DeFrees and Wilson, 1988).

When foraging, E. helvum did not exhibit common predator avoidance strategies such as

lunarphobia. The purpose of migration in E. helvum therefore may not be to exploit low

predator levels or swamp local predators.

I did not consider the influence of temporal and spatial heterogeneity of fruit production on the colony’s foraging behavior. It would be worthwhile to examine the foraging behavior of the colony more closely, and investigate whether individual bats spend more time searching for food when they first arrive and whether they visit the same foraging areas each year. While local small–scale phenology is mostly in phase, within vegetation types large–scale phenology is variable (Fuller, 1999), and foraging areas should change accordingly on a yearly basis. Furthermore, the E. helvum colony appears well before both regular rainfall and peaks in fruit availability. Consequently they may be using other cues in conjunction with food availability to time their migration. In

Zambia, for example, pre–rain increases in the canopy foliage (Fuller, 1999) could serve as one cue used by migratory bats.

While closer monitoring of individual bats could provide valuable biological data, the logistics involved are imposing. Radio–telemetry studies are problematic since bats may fly over 30 km to feed at night and many areas around Kasanka (such as the

Democratic Republic of Congo) are difficult or dangerous to access. The bats fly too high to be netted in mist nets, and the forest structure does not provide many sites suitable for canopy nets. The use of megaharp traps may increase megachiropteran capture rates (Tidemann and Loughland, 1993). Alternatively, satellite tracking could offer data on both daily and long–term movements, but the equipment is currently too heavy to use on E. helvum and is prohibitively expensive for a pilot study. From viewing the evening dispersal it is clear that observations and netting are needed at sites further than 15 km from the roost. The use of satellite tracking equipment to gather detailed

28 foraging, migration, and roosting information would give researchers insight into habitat requirements and fruit bat movements.

Conservation Implications

We currently have no information regarding the movements of this colony except for the ten weeks that they spend at Kasanka. When migration is driven by the need to increase food availability, movement is always in the direction of increased resource abundance, usually along a north–south axis (Elgood et al., 1973). If E. helvum is following large–scale trends in food availability then the colony should arrive after depleting wet–season food supplies to the north (Fleming and Eby, 2003).

A key issue these results highlight is the potential effect of deforestation on foraging and roosting sites. Since preserving only one site on their migration route will not protect the colony (Fleming and Eby, 2003), identification of roost sites and migration routes before and after Kasanka should be a priority. In Australia, some flying foxes showing seasonal movements were found to have only three percent of their habitat protected in reserves (Palmer and Woinarski, 1999).

Megachiroptera frequently overlap in food sources with man (Marshall, 1985), and in Zambia this is the case with both Uapaca kirkiana and Parinari curatellifolia. Since most of Zambia is regenerated forest (Chidumayo, 1987), if Uapaca species are the first trees to regenerate in deforested areas food availability may be higher outside of the national park. Farmers, in this system of slash and burn farming, may selectively leave both tree species when clearing their fields. Vegetation analysis from nearby farming communities may explain why most E. helvum disperse outside of the national park to feed.

The potential roles of a community this size in ecosystem function are enormous.

Measuring relatedness and gene flow between colonies could facilitate conservation efforts as well as provide information about the degree of mixing (Fleming and Eby,

2003). Interactions between African fruit bat colonies are an area of special concern as their role in disease transmission, including Ebola, Lyssavirus, and Nipah virus, are unclear. A colony this size can move significant amounts of energy and nutrients across habitats, and can have an important influence on the dynamics of those recipient food webs (Polis et al., 1997; Huxel, McCann and Polis, 2002). It is not known what role E. helvum plays in pollination or seed dispersal in the region, nor whether other species could assume this role should the movements of the E. helvum colony change. As the environment changes around us, having a knowledge of basic bat biology and behavior as well as identifying and preserving key habitat can have important conservation implications for both the bats and the species that rely upon them (Fujita and Tuttle,

1991).

CHAPTER 3 DISTINGUISHING MEGACHIROPTERAN SPECIES USING MORPHOLOGICAL CHARACTERISTICS IN KASANKA NATIONAL PARK, ZAMBIA

Africa is home to 12 families of Megachiroptera and 35 families of Microchiroptera

(Nowak, 1997). These families and the species that comprise them have been examined in Eastern, Western, and Southern Africa (Fenton, 1975; Ansell, 1978; O'Shea and

Vaughan, 1980; Thomas, 1982; Findley and Wilson, 1983; Kingdon, 1984; Bergmans,

1988; 1989; 1990; Nowak, 1994; 1997; Taylor, 2000; Fahr and Ebigbo, 2003). Central

Africa is not well represented in the published literature (Ansell, 1978), and throughout

Africa available information about regional bat fauna is by no means comprehensive.

Large regions of the continent have not been sampled, and very few areas have been extensively studied (Jones, 1972; Fenton, 1975; Ansell, 1978; O'Shea and Vaughan,

1980; Marshall and McWilliam, 1982; Thomas, 1982).

The Megachiroptera in particular have been neglected as a group, with few attempts at direct comparison and discrimination of multiple species (Bergmans, 1988;

1989). Sympatric species of epauleted and dwarf epauleted fruit bats can be nearly indistinguishable (Bergmans, 1988; 1989; 1990; Taylor, 2000), and this may contribute to the absence of comparisons in the literature. Comparisons are needed because species designations have been contested and changed multiple times over the last 100 years, with species often being misclassified and reclassified or some species demoted to subspecies status (Bergmans, 1988; 1989). Additionally, a single sampling period may not fully describe a site’s species composition since the presence of several migratory

30 31

species may cause temporal variability in bat assemblages (Thomas, 1983; Cosson et al.,

1996).

This study addressed the lack of information about species discrimination and fruit bat assemblages in Zambia. The last comprehensive mammal list for Zambia was published by Ansell (1978), and provides only sparse records of species occurrences for most chiropteran species in the country. No systematic or extensive sampling at the study site, Kasanka National Park, has ever been conducted. The number of species found in the surrounding area, the large ranges of those species, the migratory behavior of

African Megachiroptera, and the large overlap in morphological characteristics suggests

that the diversity of bat species in Central Zambia may be higher than previously

recorded. The site may also be home to previously unidentified subspecies, since several

subspecies have already been identified in other parts of Africa (Bergmans, 1988; 1989).

I conducted mist netting in Kasanka National Park, Zambia, from November to

December 2003 to collect and identify fruit bat species. Data collected on epauleted and dwarf epauleted fruit bats were examined to test whether morphological measurements

and the postdental palate are sufficient for identifying the species in the region. I used a

combination of published data, field data, and multivariate statistics to address this goal.

Statistical methods for comparing species are needed because the number of

megachiropteran species, and their resemblance to one another, makes it extremely

difficult to conclusively identify some species in the field, and geographical variations in

species types only increases this uncertainty (Bergmans, 1988). In many regions the

main characteristic separating sympatric species that otherwise share size and color characteristics is the configuration of their postdental palatal ridges. When pregnant and

juvenile individuals are present, the species measurements can form a continuum, and postdental palatal ridge configuration may be the same for different species that overlap in size. This study is the first to statistically analyze morphological data from these species as an alternative approach for distinguishing them. My results demonstrate a need for further research into species’ characteristics in the region.

Methods

Study Site

This study was conducted at Kasanka National Park in the Central Province of

Zambia (12º 33’South 30º 09’East). Kasanka is a 420 km2 privately managed national

park measuring approximately 15 km north to south and 35 km east to west. The park

has an average elevation of 1050 m with the highest elevation in the park at just less than

1300 m above sea level. The vegetation of the region is mainly composed of miombo

woodland, which contains the dominant tree species Brachystegia, Isoberlinia, and

Julbernardia. The area is interspersed with seasonally wet grasslands and rivers as well

as some small patches of chipya, mushitu, and mateshe forests (Smith and Fisher, 2001).

The canopy height of these forests averages from five to twelve meters high. Seasonal

rainfall from November through April supports a large seasonal fruit crop of food

resources including Uapaca spp., Syzygium spp., and Parinari curatellifolia.

Mist Netting Methods

Mist netting data were collected from 4 November 2003 through 30 December

2003 at 26 sites throughout the park. I used two types of 4–shelf mist nets: a 2.6 x 4 m,

110 denier/2–ply, 60 mm mesh and a 2.6 x 12 m, 50 denier/2–ply, 38 mm mesh (Avinet,

Dryden, NY). Bats were netted over 23 nights in preset vegetation transects and at two

33 camp sites, Wasa and Shiwila. (See Chapter 1 for a description of transect placement methods.) Each transect was netted from one to four times over the two month period.

When sites were resampled the interval between sampling occasions ranged from one week to one month. Sites were netted with some combination of one 2.6 x 12 m and two

2.6 x 4 m mist nets set at ground level. Nets were placed within five meters of a road and at most sites between one to three nets were placed across the road.

Nets were opened beginning around 1815 (dusk) and closed at about 0445

(sunrise). A site was netted for approximately three hours before nets were moved to a new location, and in this manner I was able to cover two to three sites a night. Nets were checked every five to thirty minutes and bats were removed immediately upon discovery.

Lactating females caught near sunrise were released without processing while all other bats were released after processing.

Body mass was measured to the nearest five grams using an Avinet spring scale

(500 g, 5 g units), and the approximate time of capture was recorded. Standard measurements were taken on the bats and included ear, forearm, half wingspan, body, foot, and tail (Table 3–1) (Kunz, 1988). I also recorded the sex, relative age, reproductive condition, number of postdental palatal ridges, and species. Palate data were collected on epauleted species from 26 November through 30 December 2003; palate descriptions included the number and spacing of the postdental palatal ridges.

Epauleted fruit bats were marked with a permanent marker on an ear tuft to monitor recaptures over one night.

Females were assigned to one of six reproductive classes, while males were assigned to one of three reproductive classes (Table 3–2). Juveniles were identified by

backlighting their wings and recording the development of their epiphyses. Adult male

fruit bats were identified by their secondary sexual characteristics, including development

of shoulder tufts, a prominent larynx, and enlarged testes. Pregnant females represent a

reasonably advanced stage of pregnancy since palpation only detects embryos above a

certain size.

Table 3–1. Description of morphological measurement used in this study. Measurement Method used Ear From notch to tip Forearm Measured with the wing folded, included joints Half–wingspan Measured flat with the wing extended out by hand Body Head and body with head extended; does not include the tail Foot From joint to end of claws Tail From base to tip (when present) Palate Number of postdental ridges

Table 3–2. Reproductive classes used in this study. Sex Reproductive Class Description Female Juvenile Nulliparous: visible epiphyses and rudimentary nipples Adult non– Nulliparous: fused epiphyses, rudimentary nipples, reproductive and not pregnant Adult reproductive Parous: nipple cornification, but neither pregnant nor lactating Reproductive, pregnant Parous: nipple cornification and palpably pregnant Reproductive, lactating Having enlarged mammary glands expressing milk Pregnant, lactating Palpably pregnant with engorged mammary glands Male Juvenile Visible epiphyses and undeveloped secondary sexual characteristics (small testes, smaller larynx, and undeveloped epaulets) Adult non– Fused epiphyses and undeveloped secondary sexual reproductive characteristics Adult reproductive Large testes, prominent larynx, and well developed epaulets

Analysis of Species Data

Other researchers have experienced difficulty in identifying the species of

epauleted fruit bats captured in this study (Bergmans, 1988; 1989), thus I used

multivariate techniques to identify post hoc groups in the data. To begin, I tested the

capabilities of multivariate statistics to find a morphological measurement that would

reliably distinguish the species without using the palate data. All statistical analyses were conducted in SAS Version 9.0 (SAS Institute Inc., 2004).

A Spearman correlation matrix for the variables body mass, forearm, half–

wingspan, foot, ear, and body length showed high correlation among all the variables

(p<0.0001). I standardized the variables around a mean of zero and standard deviation of

one. This removed the problem of morphological measurements being a function of

overall size and provided a better measure of relative size. The data were examined for

outliers, and I removed the suspect measurements, but not the individuals, from future

analyses.

Body mass, length of body, forearm, foot, ear, and half–wingspan were entered as

standardized variables in a cluster analysis using a Ward’s minimum variance approach.

Since cluster analysis is sensitive to missing data and outliers, individuals with missing

data and two pregnant females were removed from the analyses.

Results

Capture Data

The total netting effort was 1970 net meter hours (where one net meter hour equals

one meter of net open for one hour) over 23 nights. A total of 149 bats were netted over

the duration of the study. The 26 netted insectivorous bats represented nine species:

Nycteris thebaica, Nycticeus schleiffeni, Pipistrellis kuhlii, Pipistrellis somalicus,

Scotoecus hirundo, Miniopterus sp., Myotis welwitschii, Eptesicus somalicus, and

Eptesicus capensis1. The 125 Megachiroptera netted represented five species of epauleted fruit bats. These included Epomophorus minor, Epomophorus gambianus crypturus, Epomophorus wahlbergi, Epomophorus labiatus, and Micropteropus pusillus.

Notably absent from the mist nets were Eidolon helvum, despite the fact that numerous

individuals were seen flying overhead. Epomophorus labiatus, Micropteropus pusillus,

Pipistrellis kuhlii, Pipistrellis somalicus, Nycticeus schleiffeni, Scotoecus hirundo,

Miniopterus sp., Eptesicus somalicus, and Eptesicus capensis are all new additions to the

species list for Kasanka National Park.

On 26 November, I trapped E. capensis pregnant females, and on 6 and 8

December I trapped juvenile S. hirundo and juvenile and pregnant E. capensis. I netted

juvenile Megachiroptera of various species starting 11 November and continued catching

juveniles throughout the netting period. Lactating Megachiroptera were trapped from 10

November through 30 December, and pregnant females were trapped 4 and 21 December.

Both pregnant females were also lactating, indicating they were probably close to their

parturition date. Sex ratios for the Megachiroptera were 50 adult females to 18 adult

males and 24 juvenile females to 32 juvenile males.

The first insectivorous bats were caught at approximately 1830 and the first fruit

bats at 1900. Fruit bats were caught throughout the night with the final animals netted at

about 0430, or immediately before sunrise. Only one fruit bat was recaptured in the same

night throughout the study period. I calculated capture rates as the number of

Megachiroptera netted per net meter hour (nmh), and rates ranged from 0 to 0.429 bats

per net meter hour.

1 Nycticeus schleiffeni and Pipistrellis somalicus were trapped and identified by Fiona Reid while netting with a group from Bat Conservation International.

Megachiroptera Species Identification

The morphological data could be used to separate the captured bats into four groups based on the cluster analysis results (Figure 3–1). The decision to trim the cluster tree at four groups was made by examining the behavior of the Pseudo T2 statistic, the Pseudo F statistic, and the R–squared value (SAS Institute Inc., 1999). With four clusters 75.4 % of the variability in the data was explained. Four clusters are also biologically meaningful, since in the field the bats appeared to be from five species, and two of those species are known to be indistinguishable based on external appearance alone (Bergmans,

1988).

Species assignments

Post hoc species assignments were based on cluster analysis group assignment in conjunction with published distribution maps, morphological measurements, and field notes. The smallest group, composed of Micropteropus pusillus, was the easiest cluster to identify statistically. The largest species, Epomophorus gambianus crypturus, was distinguishable both statistically and based on facial characteristics. The other two groups in cluster analysis represent E. minor and an indistinguishable (except for palate configuration) Epomophorus wahlbergi and Epomophorus labiatus group.

When captured in the field, the fruit bats clearly represented five different species, although definitive assignment of each individual to a species was frequently difficult.

Species were distinguishable based on their palatal configuration and also showed distinctive rostrum lengths and pouchiness of the lips. M. pusillus characteristically had a small, pale body, a short nose and was usually extremely docile. E. minor was the next larger species and had generally a shorter nose than both E. wahlbergi and E. labiatus but overall was larger than M. pusillus. E. wahlbergi and E. labiatus are similar in

appearance but could be differentiated by the configuration of their palatal ridges. The

distinction of E. minor, E. wahlbergi, and E. labiatus was complicated however by the fact that E. wahlbergi had a similar configuration of postdental palatal ridges as E. minor

and overlapped in size with that species as well. E. g. crypturus had a longer rostrum

than the other groups and had jowl–like lips. (Photographs of the species are presented in

Appendix I.)

All five species had similar reproductive timing, and juveniles and lactating

females were found in all four groups. Timing of parturition and reproductive status

were therefore not sufficient as supplemental data for distinguishing fruit bat species.

Summary of group analysis

A summary of body measurements for the different species is presented in Table 3–

3. (Original data are presented in Appendix C.) The size ranges of the species in this

study are similar to, but not the same as, those previously published. Published ranges of

body mass for M. pusillus were 10.5–35 g (FA 46–65 mm), whereas in this study they

were found to be larger, ranging from 24–45 g (FA 56–64 mm). The E. wahlbergi and E.

labiatus group has a published range of body masses from 38–125 g (FA 55–98 mm) and

in this study both body mass and forearm measurements were within this range at 52–93

g (FA 72–89 mm). The largest individuals, E. g. crypturus, could be expected to fall

within the range of published measurements of 56–140 g (FA 76–88.4 mm) but in this

study they were somewhat larger with body mass ranging from 79–147 g (FA 80–91

mm). The final group, E. minor, had a published body mass range of 25–65 g (FA 56–68

mm) and in this study body mass fell within this range at 40–61 g (FA 60–74 mm).

0.6

0.4 Squared – 39

0.2 Semi – partial R

0.0

Individual Bat ID Number

Figure 3–1. Dendrogram produced from Ward’s cluster analysis on species data.

The three mid–sized species overlap in body weight as well as forearm length, although E. minor measurements were smaller as expected (Bergmans, 1988). E.

wahlbergi and E. labiatus were indistinguishable based on body size alone, but can be distinguished by the number of postdental palatal ridges (Figure 3–2). Cluster analysis did separate E. minor and E. wahlbergi, but they had similar palatal configurations and

overlapped in size. The E. g. crypturus group showed two different palate

configurations; this indicates that the cluster analysis did not separate out a pure group.

A) B)

Figure 3–2. Photographs of palatal ridge configurations identified in this study. A) Configuration 1 showing two postdental palatal ridges with space between them. B) Configuration 2 showing two postdental palatal ridges with a gap between the last molar and the first postdental ridge.

For two of the five species the maximum body masses were higher than previously

reported. This could be a result of animals gaining weight from an abundant food supply.

The data collected here were not at the lower end of the range for the region since all

captured animals appeared in good condition.

Palate Data

Figure 3–3 illustrates the size distribution for all individuals with distinctive facial

features or for which palatal configuration is known. Configuration 3 is E. g. crypturus,

Configuration 2 is E. labiatus, and Configuration 1 is E. minor or E. wahlbergi. Not only

do groups overlap in body mass and linear measurements, but it is also clear that

postdental palate data alone do not separate the species.

The range of body masses for the two different palatal configurations established

that individuals with Configuration 2 have an average weight plus or minus one standard

deviation that is totally encompassed within the size range of Configuration 1 (Figure 3–

4). Since Configuration 1 encompasses both a smaller E. minor and a larger E. wahlbergi

this result is not surprising. It does illustrate again, however, the difficulty encountered when attempting to identify species through morphology alone.

75 Configuration 1 70 Configuration 2

Forearm Length (mm) 65 Configuration 3

60 20 40 60 80 100 120 140 160 Body Mass (g)

Figure 3–3. Distribution of animals based on size and palatal characteristics. Configuration 1 includes all bats with two postdental palatal ridges separated by a gap. Configuration 2 includes all bats with two postdental palatal ridges that are not separated by a gap but do have a gap between the last molar and first palatal ridge. Configuration 3 includes all bats recorded as having a long nose and large, distended lips in the field.

Table 3–3. Range of measurements for fruit bats trapped in this study. Species groups are based on cluster analysis results. N is the sample size, FA is forearm length, Halfwing is half of the total wingspan, Juv is juvenile. Palate indicates the number and configuration of postdental ridges found in individuals in the group, ‘s’ indicates a space between them. Weight Halfwing Wingspan Body Foot Ear Tail Species N Sex Age (g) FA (mm) (mm) (mm) (mm) (mm) (mm) (mm) Palate Eidolon 2 F Adult 255–280 117–126 350–370 700–740 195–205 33–35 26–29 13–15 N/A helvum 3 M Adult 265–295 121–123 330–350 660–700 205 30–33 29 10–17 N/A Group 1 16 F Adult 88–143 80–91 230–280 460–560 120–165 18–25 21–26 0–5 1s1 or s2 E. gambianus 9 M Adult 87–147 84–90 230–275 460–550 132–177 20–23 21–29 0–9 1s1 or s2 crypturus 1 F Juv 87 84 250 500 125 22 25 0 1s1 3 M Juv 79–98 81–87 250–265 500–530 135–145 20–21 23–28 0–5 s2 Group 2 13 F Adult 71–93 77–89 210–259 420–518 110–145 14–22 18–24 0–4 1s1 E. wahlbergi/ 4 M Adult 67–79 75–80 225–245 450–490 120–145 17–24 19–22 0–4 No data E. labiatus 13 F Juv 52–89 72–88 205–245 410–490 97–135 18–24 16–26 0–7 1s1 or s2 20 M Juv 57–84 72–88 198–250 396–500 110–155 18–25 21–27 0–6 1s1 or s2 Group 3 6 F Adult 41–45 60–64 163–188 326–376 94–106 13–14 16–18 0–4 1s1 42 M. pusillus 0 M Adult – – – – – – – – – 7 F Juv 25–38 56–61 140–176 280–352 84–105 11–17 14–20 0 No data 5 M Juv 24–37 56–63 145–180 290–260 80–105 12–19 11–20 0 No data Group 4 4 F Adult 40–53 60–65 180–200 360–400 95–112 15–18 18–21 0 No data E. minor 5 M Adult 50–61 63–69 177–197 154–394 98–130 15–17 18–21 0 1s1 1 F Juv 49 71 210 420 115 17 21 – No data 4 M Juv 54–60 67–74 205–221 410–442 105–125 15–18 18–23 0–4 1s1

120

100

20 Mean Body Mass (g) 0 Configuration 1 Configuration 2 Palate Configuration

Figure 3–4. Mean body mass and one standard deviation for two palatal configurations. Configuration 1 includes all bats with two postdental palatal ridges separated by a gap. Configuration 2 includes all bats with two postdental palatal ridges that are not separated by a gap but do have a gap between the last molar and first palatal ridge.

Discussion

This study demonstrates that Kasanka National Park has high megachiropteran diversity and that further work is needed to conclusively identify the species, or subspecies, that occur at Kasanka. Sympatric epauleted species could not consistently and definitively be identified in the field solely by examination of morphology.

Statistical analyses of the data indicate that morphological measurements and postdental palate are insufficient indicators of species, especially when the situation is confounded by age and reproductive condition.

Identification of Sympatric Species

The most recent comprehensive record of bat species for Zambia is found in Ansell

(1978). No definitive guide for bats of the region has been published and the bats at

Kasanka have not been thoroughly catalogued or identified. Based on distribution maps and species accounts, Epomophorus wahlbergi, Epomophorus minor, Epomophorus labiatus, Epomophorus gambianus crypturus, Epomops dobsonii, Eidolon helvum and

Micropteropus pusillus all have ranges that could encompass Central Zambia (Ansell,

1978; Bergmans, 1988; Nowak, 1997). Out of these species, only E. wahlbergi, E. gambianus (Racey, 2002), E. minor, and E. helvum had been previously captured and identified at this site.

Epomops dobsonii and Eidolon helvum have distinguishing characteristics that make them comparatively easy to identify in the field. However, the discrimination and identification of epauleted and dwarf epauleted fruit bats when they inhabit the same region can be difficult (Bergmans, 1988; 1989; Taylor, 2000). The first difficulty in identifying the species comes from the fact that not only do they overlap in geographic ranges and morphology, but they also frequently utilize the same habitat types (Ansell,

1960; Bergmans, 1988).

When multiple epomophorines are sympatric, descriptions and methods for identifying species are often conflicting and confusing. For example, E. minor and E. wahlbergi may be morphologically indistinguishable (Apps, 1996), or E. wahlbergi should always be larger than E. minor (Bergmans, 1988). Where the ranges of E. labiatus and E. minor overlap, “typical” E. labiatus is larger than “typical” E. minor

(Bergmans, 1988). E. wahlbergi can also overlap in size with E. labiatus and E. gambianus crypturus, but it can be distinguished from those species by its one postdental palatal ridge, whereas E. labiatus and E. gambianus crypturus have two postdental palatal ridges (Ansell, 1960). Although their palatal configuration is similar, E. labiatus

averages smaller in all measurements than E. gambianus (Ansell, 1960; Bergmans,

1988). (For a thorough review of the species measurements and identifying characteristics see Bergmans (1988; 1989; 1990).) See Appendix A for detailed published species data.

Mathematical Methods in Species Discrimination

Bergmans’ reviews of the species (Bergmans, 1988; 1989), while addressing the identification and discrimination of species, does not approach the question from a multivariate perspective. Using strong mathematical methods for discriminating species may yield new insights into characteristic morphology of the species if sufficient sample sizes are used and morphological measurements are precisely measured.

Body mass was measured with a spring scale marked in five gram increments and wing length was largely determined by how far the wing was stretched. Body length can

be error prone depending on the angle of a bat’s head relative to the rest of the body.

Pregnancy, breeding condition, and sexual dimorphism can all contribute to the degree of

overlap in species morphology and seasonally expand size ranges within the same

species. Juvenile morphological measurements may not change proportionately in

different growth stages which could lead to misclassification of individuals.

Research Needs

Future research should concentrate on collection and analysis of more extensive

data including cranial measurements and predental and postdental palate data. The

abundance of epauleted fruit bats indicates that more intensive sampling could yield a

large sample size for further palatal surveys and morphological analyses, and potentially

allow identification of new subspecies among those bats at Kasanka. Call structure, or

male vocalizations, and genetic data, such as nuclear mitochondrial markers, are other

methods that can be used to identify the different species of epomophorine bats.

I found that the palatal configuration of E. wahlbergi more closely resembles two

postdental palatal ridges, not one clear postdental palatal ridge as previously reported.

The first ridge was close to, but behind, the last molar, followed by a space and a second

postdental palatal ridge (Configuration 1 in this study). This may be a geographic morph of the species and should be further explored.

To more accurately survey the species at Kasanka mist netting should occur in the dry season and later in the rainy season as well as in additional habitats. Some of the species captured in this study are known to be migratory in other regions of Africa, and we do not know how many species migrate through Kasanka, other than Eidolon helvum, or the timing of their migrations. E. walhbergi is migratory in other African regions

(Thomas, 1983; Nowak, 1997), including in northern Zambia, and it is therefore possible that it too may be a transient species at Kasanka2.

Identifying species of mammals in a national park serves many roles. The first is to

increase our knowledge of regional biodiversity and identify what species our preserves

are protecting. Secondly, studies of assemblage dynamics cannot occur unless we know

what species are in the assemblage and understand the interactions among resident and

migratory species. Knowledge of the species and their interactions can help researchers

identify and protect their role in important ecological processes. This region shows the

potential for a host of interesting foraging studies, with seasonal shifts in food supply,

2 A migratory colony of thousands of an epauleted species, probably E. wahlbergi, visits a Northern Zambia site near Mpika, Shiwa N’gandu, in January. This account was substantiated through multiple sources.

47 variable responses in foraging behavior and interactions with migratory species. The interactions among species can have consequences for forest regeneration and food supplies should the bat species composition be changed (Fujita and Tuttle, 1991; Duncan and Chapman, 1999; Taylor and Kankam, 1999).

CHAPTER 4 THE RESPONSE OF FRUIT BATS TO CHANGING RESOURCE AVAILABILITY IN KASANKA NATIONAL PARK, ZAMBIA

The complexity and dynamic character of ecosystems makes the prediction of individual, population, and community level reactions to spatial and temporal variation in food resources difficult. Species do not exist in isolation, and within each community basic inter and intraspecific interactions, such as competition, predation, and mutualism, all contribute to foraging decisions at each level (Putman, 1994; Morin, 1999). Before studying complex foraging interactions, it is important and necessary to understand the salient processes and motivations influencing the foraging decisions of each species.

Zambian fruit bat assemblages are unique in species composition, and differences in habitat, flora, fauna, and climate make chiropteran research results from other regions of Africa not directly applicable to Central Zambia. The dominant habitat in Zambia, miombo woodland, is different from what is found in Africa at other megachiropteran research sites. Furthermore, not only does the region have at least five related sympatric species, it also experiences a large influx of migratory fruit bats. Component species have been studied in other areas (Jones, 1972; Fenton, 1975; Thomas and Fenton, 1978;

O'Shea and Vaughan, 1980; Marshall and McWilliam, 1982; Thomas, 1983; Fenton et al., 1985; Taylor and Kankam, 1999), but nothing is known about assemblage level species structure. Fruit abundance in Zambia is highly seasonal, and if resident and migratory species are competing for these resources their interactions may lead to complicated spatial and temporal foraging patterns.

48 49

Limited resources can influence inter and intraspecific relationships, and if the fruit bat species in Zambia are competing for roost sites or food resources, this may lead to spatial and temporal variation in competition. This competition could manifest itself in a number of ways, including partitioning of resources, territoriality, or aggression (Morin,

1999). If food resources are limited, the prevalence of competition can be affected by both the size of migratory and resident colonies as well as their spatial distribution. On the other hand, if food supplies are superabundant, even the appearance of a large migratory colony may not lead to competition. Specialization of sympatric fruit bats on non–overlapping food supplies may be the result of past competition.

Foraging distances, food preferences, and feeding strategies may reflect the morphological and energetic constraints of each species. Species composition could be a function of habitat selection, with each species finding an optimal trade–off between food availability and predation (Morin, 1999). When fruit bat species share common predators, large numbers of migratory bats may satiate the local predator population, and therefore favorably influence resident species by reducing their predation rates.

This study was conducted at Kasanka National Park, Zambia, to establish what fruit bat species are present in the area and identify factors that may be important in determining assemblage structure. Additionally, I explored the interactions between migratory E. helvum and other fruit bat species. My results suggest that the migratory and resident fruit bat species used different foraging and roosting strategies to exploit local food resources.

Methods

Study Site

This study was conducted at Kasanka National Park, in the Central Province of

Zambia (12º 33’South 30º 09’East). Kasanka is a 420 km2 national park measuring

approximately 15 km north to south and 35 km east to west. The park is bisected by the

Kasanka River, and has numerous other rivers and floodplains scattered throughout. The

region receives unimodal seasonal rainfall totaling about 1200 cm, usually between the

months of November and April, which leads to high seasonality in regional fruit

production. The park has an average elevation of 1050 meters with the highest elevation

in the park at just less than 1300 meters above sea level. The region is mainly composed

of Brachystegia, Julbernardia, and Isoberlinia dominated miombo woodland,

interspersed with small patches of chipya, mushitu, and mateshe forests (Smith and

Fisher, 2001).

Mist Netting Methods

Mist netting methods are described in detail in Chapter 3. From 4 November 2003 through 30 December 2003 I sampled 24 transects over 23 nights with mist nets. Nets were placed in vegetation transects associated with the phenology monitoring study described in Chapter 2. Nets were placed within five meters of roads and netting was conducted on clear nights and in light to moderate rainfall.

Netting started near dusk (1815), finished at about 0445 (sunrise), and nights were divided into three netting events occurring at three different sites. I conducted 46 total netting events for a netting effort of 1970 net meter hours (where one net meter hour equaled one meter of net open for one hour). Nets were checked every 5 to 30 minutes

51 and bats were removed immediately upon discovery. Species, body mass, body measurements, sex, and reproductive condition were recorded (Kunz, 1988).

Mist netting bats near feeding sites should not affect their foraging behavior

(Heithaus and Fleming, 1978), and it has been successfully used to establish relative abundances of species in feeding areas in both bird and bat studies (Heideman and

Heaney, 1989; Godinez-Alvarez, Valiente-Banuet and Rojas-Martinez, 2002). However results can be biased by net placement because fruit bats have been found to selectively forage along trails (Palmeirim and Etheridge, 1985). Fruit bats may furthermore use flight lines, with characteristics indistinguishable from the surrounding habitat (Marshall and McWilliam, 1982), making it difficult to control for net placement when mist netting.

All statistical analyses were conducted in Excel, PC–ORD Version 4.20 (McCune and Mefford, 1999), TRANSFOR (Legendre, 2001), and SAS Version 9.0 (SAS Institute

Inc., 2004).

Bat Species Data

Capture data were used to assign species to four groups using Ward’s minimum variance cluster analysis on the standardized morphological measurements (see Chapter

3). Since two species were difficult to discriminate, these four groups were used as variables in statistical analyses rather than the five species. The four groups correspond to the following species: Group 1 is Epomophorus gambianus crypturus, Group 2 is

Epomophorus wahlbergi and Epomophorus labiatus combined, Group 3 is

Micropteropus pusillus and Group 4 is Epomophorus minor.

Vegetation Methods

The methods used for the placement of vegetation transects and monitoring of phenology data are described in Chapter 2. Local knowledge and personal observation

52 were used to provide a working list of food sources for the fruit bats. While 46 transects were monitored for phenology data, additional structural data were collected in the 24 transects used as netting sites. I marked the tallest 40 trees in each transect and recorded their position (using a Garmin GPS12 receiver), diameter at breast height (dbh), and species. To estimate canopy height, I used a homemade clinometer to approximate the angle and a range finder (Bushnell Corp.) to measure the distance to the base of individual trees; with these two values I could calculate height. Tree species were identified initially by local Lala names, and a scout translated local to scientific names using field guides (Storrs, 1995; Coates Palgrave, 2002). Several publications were used to verify the translations and identify unknown species (Storrs, 1995; Smith, 2001; Smith and Fisher, 2001). I calculated Brillouin’s diversity index for non–random data for only those habitats in which I netted.

In the vegetation matrix the coefficient of variation (CV) of the row and column totals was small (<50%) for all values so relativization was not used. I removed rare species (those occurring in less than five percent of the samples) from the analyses

(McCune and Grace, 2002). I then used the program TRANSFOR to transform the species matrix using a chord distance measure (Legendre, 2001). The chord distance is the Euclidean distance computed after scaling the site vectors to length 1 (Legendre and

Gallagher, 2001). The chord distance is appropriate because it does not give disproportionate weight to the rare species and since it is a Euclidean distance, the transformed data were now appropriate for use in principal components analysis (PCA) or analysis of variance (ANOVA) (Legendre and Anderson, 1999). PCA was performed on the covariance matrix using PC–ORD (McCune and Mefford, 1999). Transects were

53 subdivided into habitat types for ANOVA using the results of the PCA combined with field based classifications. The four tree species that composed the first three principal components were also used as predictor variables after performing an arcsin transformation on the percent composition.

Capture Rates

First, the Spearman correlations for non–parametric data and tests of normality were examined for all the variables. These included the response variables capture rate, number of bats captured from Group 1, Group 2, Group 3, and Group 4 as well as the predictor variables date, diversity, habitat, number of fruiting trees, and the sum of the tree ranks each week (a relative measure of fruit availability). Capture rates were bimodal and did not fit any standard distribution. The correlation matrix was examined to identify cases of multicollinearity or singularity and exclude redundant variables.

I performed non–parametric ANOVA on the class variables to group the data and test for differences in the mean responses. I used Kruskal–Wallis non–parametric

ANOVA to test if overall capture rate means and the number of each species of fruit bat were independent of the time of night that netting was conducted in, habitat type, transect number, and week. Response variables included capture rate (untransformed), number of bats in Group 1, number of bats in Group 2, number of bats in Group 3, and number of bats in Group 4. I used the non–parametric Wilcoxon Mann–Whitney test to determine if the mean capture rates or the number of bats netted were equal with or without the presence of the moon. Since capture rates could be affected by net placement (Palmeirim and Etheridge, 1985), this variable was tested for its effect on capture rates with the

Wilcoxon test as well.

I used a Mantel test on the Euclidean distances to test for a relationship between the

capture matrix and a matrix composed of environmental variables. The strength of the

relationship was measured using randomization and 9999 runs of the data. The capture

matrix contained four variables, the number of each type of bat captured at each time and

site. I repeated the test for six matrices composed of different combinations of

environmental variables to ensure robust results. Although diversity values could not be

calculated for three transects, for one of these transects (Transect 9), I averaged the

diversity of two nearby transects and substituted this value for the missing data.

Transects 1 and 10 were distinct in composition, and extrapolation of diversity indices was not appropriate.

Results

Bat Species Data

Between 0 and 18 fruit bats were captured each night at the netting sites. I netted

121 total Megachiroptera over the duration of the study representing the species

Micropteropus pusillus, Epomophorus gambianus crypturus, Epomophorus labiatus,

Epomophorus minor, and Epomophorus wahlbergi. Cluster analysis of the species data yielded four groups for assemblage analysis because large morphological overlap made it difficult to separate E. labiatus and E. wahlbergi. Although Eidolon helvum were seen in large numbers, they were never trapped in a mist net, and therefore were not included in these analyses.

Vegetation Analysis

Overall, I recorded 40 tree species that comprised the majority of the canopy in the transects. Of these 40 species only 28 occurred in more than one transect and were used

in PCA. In PCA the first three eigenvectors explained approximately 61 % of the

variation. Additional eigenvectors added between three and ten percent to the proportion of variance explained. The first four eigenvectors were largely comprised of Isoberlinia

angolensis, Julbernardia paniculata, Brachystegia boehmii, and Combretum molle. The

first three species are considered characteristic of miombo woodland, while Combretum

molle is characteristic of chipya.

Examination of the graph of the first and third principal components showed three

apparent groups in the data. These groups represented the woodland habitats miombo,

chipya, and Brachystegia dominated miombo. While the Brachystegia dominated group

is usually considered miombo woodland, the results of the PCA indicated distinct

compositional differences. Three transects were missing species data, and I classified

two of these transects as miombo woodland based on nearby transect assignment and

other field data. I categorized the third transect as riverine because this was the only

transect that crossed a river, and this additional factor could influence capture rates

(Fenton, 1975). Since tree species composition is highly correlated with habitat type, I

used either the classification variable habitat or the species composition of the canopy in

future analyses.

The vegetation data are summarized in Table 4–1. When analyzed by habitat type,

the trees in the single riverine transect had a mean dbh of 34.5 cm and canopy height of

11.0 m. Miombo woodland transects had a mean diversity of 1.58, mean dbh of trees was

17.4 cm and canopy height averaged 8.1 m. Chipya forest showed a mean diversity index

of 1.60, a mean tree dbh of 18.7 cm, and the mean canopy height was 7.8 m. The

Brachystegia dominated woodland had a diversity of 1.14, a mean tree dbh of 14.6 cm, and the mean canopy height was 7.5 m.

Table 4–1. Summary statistics for the canopy tree species in 24 transects. Brachystegia refers to Brachystegia dominated miombo woodland. Transect Trees DBH Height Brillouin Habitat Number N Mean (m) StDev Mean (m) StDev Diversity Index Type 1 20 34.5 21.4 11 2 --- Riverine 3 40 22.9 8.1 12 3 1.64 Miombo 5 40 15.7 10.3 7 2 1.72 Miombo 7 40 19.9 11.3 9 3 1.84 Miombo 8 40 18.3 8.3 9 2 1.36 Miombo 9 40 14.4 7.7 11 2 --- Miombo 10 40 16.9 9.5 10 3 --- Miombo 15 40 15.6 8.5 8 1 1.29 Miombo 19 40 12.2 6.4 5 1 1.47 Miombo 20 40 16.4 14.5 7 3 1.45 Chipya 21 40 15.4 5.3 9 1 1.03 Brachystegia 23 40 16.5 9.3 7 1 0.78 Miombo 26 40 18.6 13.8 8 3 1.41 Miombo 28 40 14.7 7.8 7 2 1.52 Miombo 31 40 15.5 8.5 6 2 2.01 Miombo 33 40 18.1 12.1 8 3 1.95 Miombo 35 40 16.8 10.8 7 2 1.63 Miombo 36 40 16.2 11.3 7 2 1.59 Miombo 37 40 17.5 10.7 8 3 1.53 Chipya 38 40 28.9 26.8 11 5 1.69 Chipya 41 40 23.2 18.3 9 4 1.87 Miombo 42 40 19.5 8.3 8 3 1.62 Miombo 43 30 12.0 9.2 5 3 1.73 Chipya 45 40 13.8 9.0 6 1 1.25 Brachystegia

Canopy heights ranged from 5 to 12 m, and a Kruskal–Wallis test showed that the mean canopy heights for the four habitat classifications were not significantly different

(df=3, x2= 2.213, p=0.529). The Brillouin diversity index of a site was also not significantly different among habitat types (df=2, x2= 4.165, p=0.125).

Overall, I recorded seven species of fruiting trees in the transects that are known to be food sources for fruit bats in the area. Fruiting trees were more consistently found in miombo woodland, and showed a possible trend with numbers increasing from 6 to 12

km from the Eidolon helvum roost (Figure 4–1). Whether this result is biologically significant, due to landscape structure, or an artifact of sampling method is unclear.

Analysis of Capture Rates

The Spearman correlation coefficient indicated that capture rates were correlated with the sample week (n=46, r=0.321, p=0.030). The number of E. g. crypterus captured was correlated with habitat type (n=46, r=-0.311, p=0.035) and the number of fruiting trees in a transect was significantly correlated with the date (n=46, r=-0.372, p=0.011).

Capture rates of Megachiroptera were not significantly correlated with the placement of nets across the road (n=46, r=0.204, p=0.174) or the presence of the moon (n=45, r=0.062, p=0.684). The time of night in which netting occurred was not significantly correlated with capture rates (df=46, r=0.177, p=0.239). No clear relationship was found between fruit bat capture rates and the sampling distance relative to the E. helvum roost

(Figure 4–2).

Capture rates and food availability

The total number of fruiting trees, and the amount of fruit on each tree, decreased over time, but surprisingly capture rates increased. The highest capture rates were recorded in areas of low fruit abundance (Figure 4–3). Additionally, as time progressed,

the proportion of transects with zero capture rates decreased.

Seventeen of the 24 sites were netted more than once. Out of this subset, nine

showed an increase in capture rates over time, five decreased, and one showed no change.

Two sites showed increasing and then decreasing capture rates. Only four transects were netted three times, and these transects showed very different trends in capture rates

(Figure 4–4). While some sites showed marked increases in capture rates, others showed slowly declining rates.

5 Number of Fruiting Trees 0 0.00 5.00 10.00 15.00 Distance from Eidolon Roost (km)

Figure 4–1. The number of fruiting trees as a function of distance from the E. helvum roost site. The distance is non–directional, with sampling occurring to the northwest and northeast of the roost.

0.450 0.400 0.350 0.300 0.250 0.200 0.150 0.100 Capture Rate (nmh) 0.050 0.000 0.00 5.00 10.00 15.00 20.00 Distance to E. helvum Roost (km)

Figure 4–2. Capture rates as a function of distance from the main E. helvum roost site. Capture rates are for non–E. helvum fruit bats mist netted at sites around the E. helvum roost.

Analysis of variance

Statistical analysis of capture rates and the number of bats captured in relation to

environmental variables did not yield significant results (Table 4–2). I divided the data

into three groups using early (before 2200), middle (2200 to 0100), and late night (after

0100) netting periods. This variable had no effect on capture rates (df=2, χ2= 3.497, p

=0.174) and I excluded it from future analyses. Further non–parametric ANOVA showed

that habitat type, netting site, and week did not significantly affect the overall capture rate

or the number of bats in each group when examined without interaction effects.

The number of nets on the road was the only significant test, and the mean number of E. wahlbergi / E. labiatus netted was different depending on how many nets were placed across the road. Although the number of nets placed across a road when netting

may have had a significant effect on the number of E. wahlbergi and E. labiatus captured,

the simple presence or absence of nets on a road did not yield significant differences in

the mean netting result (Table 4-3). More than one net was placed across the road only in

five out of 46 netting events, which is not a large enough sample to draw a strong

inference.

The megachiropterans netted in this study did not show significant lunarphobia.

The mean capture rate and number of each size group of bats was independent of the

presence of a moon (Table 4–3).

Mantel test results

A Mantel test was run on combinations of matrices, using one of two species

matrices and six matrices containing different sets of environmental variables (Table 4–

4). No Mantel test yielded a significant result.

Table 4–2. Results of non–parametric one–way ANOVA on capture data. P–value is the result of testing if the group means are different between netting periods using the Kruskal–Wallis test. Capture rate was calculated based on all fruit bat captures, number is actual number of each species netted. Predictor variable Dependent variable df Chi–squared P–value Time of night Capture rate 2 1.435 0.488 Time of night Number of M. pusillus 2 5.356 0.069 Time of night Number of E. minor 2 0.659 0.719 Time of night Number of E. wahlbergi/E. labiatus 2 1.113 0.573 Time of night Number of E. g. crypturus 2 1.445 0.486 Number nets on road Capture rate 3 6.420 0.093 Number nets on road Number of M. pusillus 3 2.105 0.551 Number nets on road Number of E. minor 3 1.238 0.744 Number nets on road Number of E. wahlbergi/E. labiatus 3 12.082 0.007b Number nets on road Number of E. g. crypturus 3 5.515 0.134 Habitat Type Capture rate 3 2.450 0.485 Habitat Type Number of M. pusillus 3 4.317 0.229 Habitat Type Number of E. minor 3 1.605 0.658 Habitat Type Number of E. wahlbergi/E. labiatus 3 1.030 0.794 Habitat Type Number of E. g. crypturus 3 6.152 0.104 Transect Number Capture rate 23 23.310 0.443 Transect Number Number of M. pusillus 23 19.309 0.683 Transect Number Number of E. minor 23 22.095 0.515 Transect Number Number of E. wahlbergi/E. labiatus 23 17.758 0.771 Transect Number Number of E.g. crypturus 23 22.790 0.473 Weeka Capture rate 8 11.202 0.191 Weeka Number of M. pusillus 8 14.278 0.075 Weeka Number of E. minor 8 6.163 0.629 Weeka Number of E. wahlbergi/E. labiatus 8 7.563 0.477 Weeka Number of E.g. crypturus 8 12.704 0.123 aWeek dates are from 4 Nov. through 30 Dec. 2003, divided into equal intervals. bResults are significant at the p=0.05 level.

Number of Fruiting Trees Fruiting Number of 0 26-Oct 15-Nov 5-Dec 25-Dec 14-Jan -5 Date

Figure 4–3. Capture rates as a function of the date and the number of fruiting trees in a transect. The size of the bubble represents capture rate, with the smallest being zero and the largest 0.429 bats per net meter hour. The highest capture rates were recorded when fruit availability was at its lowest.

0.450 Transect Number 0.400 0.350 8 7 1 10 0.300 0.250 0.200 0.150 Capture Rate Capture 0.100 0.050 0.000 31-Oct 10-Nov 20-Nov 30-Nov 10-Dec 20-Dec 30-Dec Date

Figure 4–4. Capture rates of fruit bats for four transects netted three times. Individual sites have successive data points joined by a line to show trends. Some sites showed little change in capture rate over time while others showed large increases in catch rates.

Table 4–3. Results of Wilcoxon rank test in non–parametric one–way ANOVA. P– values are that mean capture rates and the mean number of each species trapped are independent of moonlight or placement of mist nets across a road. Dependent Variable Predictor Variable Z score Two–tailed p value df Capture rate Presence of Moon 0.402 0.687 1 Number of M. pusillus Presence of Moon -0.846 0.398 1 Number of E. minor Presence of Moon -0.957 0.338 1 Number of E. wahlbergi/E. labiatus Presence of Moon 0.672 0.502 1 Number of E. crypturus Presence of Moon 0.871 0.384 1 Capture rate Nets on Road 1.357 0.087 1 Number of M. pusillus Nets on Road 0.901 0.368 1 Number of E. minor Nets on Road -0.780 0.436 1 Number of E. wahlbergi/E. labiatus Nets on Road 1.208 0.227 1 Number of E. crypturus Nets on Road 0.578 0.563 1

Table 4–4. Results of Mantel Tests. Environmental variables are as follows: BDI = Brillouin’s Diversity Index for vegetation composition, CH = Canopy height, D = Distance to Eidolon helvum roost, H = Habitat type, NFT = Number of fruiting trees, M = Moonlight, C = Composition (arcsin transformed Julbernardia, Isoberlinia, Brachystegia, and Combretum), NFU = Number of fruiting Uapaca trees, NFS = Number of fruiting Syzygium trees. Species Matrix Variables in Environment Matrix r–value P–value All netting events CH, D, H, NFT 0.036 0.271 Excluding events from Week, BDI, CH, D, H, C, M, NFT -0.003 0.520 transects 1 and 10 Excluding events from BDI, CH, D, C, NFT -0.007 0.515 transects 1 and 10 Excluding events from BDI, CH, C, NFT -0.003 0.521 transects 1 and 10 Excluding events from BDI, CH, D, C, NFU, NFS 0.009 0.430 transects 1 and 10 Excluding events from BDI, CH, D, NFU, NFS 0.147 0.109 transects 1 and 10

Discussion

Capture rates of fruit bats have been shown to be higher along trails and other linear landscapes, where bats may more easily access understory fruits or utilize the open space as flyways (Palmeirim and Etheridge, 1985). Fruit bats at Kasanka were not statistically more likely to be captured when nets were place across possible flyways.

Fruit bats also did not show typical predator avoidance behaviors. Captures were equally

64 likely with or without moonlight, and E. helvum foraged in the upper canopy independent of visibility.

The capture of multiple sympatric epauleted fruit bat species, and the apparent overlap in their feeding areas and food sources, is unusual. The coexistence of multiple species in similar habitats implies low interspecific competition during the period of peak food availability. The lack of E. helvum in mist nets, the seemingly most abundant species, indicates that the fruit bats could be partitioning resources, a behavior frequently attributed to competition. The absence of E. helvum could also simply be due to their flying too high to be captured in mist nets.

The distribution of food resources could be a stronger driver of fruit bat captures than other environmental variables. Habitat type, canopy height, habitat diversity and the amount of food at the netting site did not significantly influence the number of fruit bats captured. Capture rates increased in variability over time while corresponding fruit availability decreased, and as time progressed fruit bat captures increased in transects without fruit trees known to be food sources for fruit bats.

As food availability decreased, the species in this study responded in dissimilar ways. E. helvum exploited large scale changes in resources by roosting at Kasanka during the period of peak fruit availability and migrating when food availability decreased. Epauleted species did not migrate and their capture rates increased towards the end of the sampling period. This response is more likely due to changes in foraging behavior than fluctuating population size.

Additional Influences on Assemblage Structure

If food availability drives the distribution of the species, we would expect to find a strong relationship between capture rates and food availability. Trends in this study

indicate that capture rates may increase as food availability decreases, rather than the

expected higher capture rates in areas with more food resources. The increase in capture

rates over the duration of the study may be due to bats spending more time commuting and searching for food as resource abundance decreased in the surrounding environment.

This suggests that the foraging strategies of epauleted fruit bats are a function of the

spatial distribution of food, and further investigation may identify a predictable relationship between the scale of food distribution and foraging decisions.

It is likely that the epauleted and dwarf epauleted fruit bat species at Kasanka are energetically and morphologically constrained to forage on a smaller scale than the E.

helvum colony. E. helvum is a strong flyer, and this factor greatly increases its capability

to follow large scale changes in resource abundance. Some E. helvum at Kasanka fed at

sites located more than 30 km away, while E. wahlbergi traveled only about 4 km to

feeding sites at a study site in South Africa (Fenton et al., 1985). However, the scale of

resource use and foraging decisions by fruit bats is likely to be affected by the

distribution of resources at a specific location and time, and it is difficult to draw

conclusions about foraging capabilities from studies in other regions.

I found evidence that the different fruit bat species at Kasanka may be partitioning

the available resources through vertical stratification. Vertical stratification is not

uncommon in areas where sympatric fruit bats share the same resources (Marshall and

McWilliam, 1982; Francis, 1994; Zubaid, 1994; Cosson, 1995; Kalko and Handley, 2001;

Dumont, 2003). E. helvum, not agile flyers, may be restricted to the top of the canopy

(Thomas, 1982), and at Kasanka this may explain why E. helvum were never captured in

the mist nets, even though they were seen flying overhead. Evidence of E. helvum

66 selectively foraging in the canopy was previously found in both Ivory Coast and

Equatorial Guinea (Jones, 1972; Thomas, 1982). Canopy nets and records of capture heights would provide the information needed to test for vertical stratification among species.

Fruit bats can also partition resources through other means. The different species could feed on fruits within a restricted size range (Heithaus, Fleming and Opler, 1975), but the six species in this study were observed feeding on many of the same fruit species.

If the bats are partitioning resources based on their body size, the three epauleted species overlapping in size would still be competing for the same sized fruits, even if those bats on either end of the spectrum were not. While in some regions bats decrease competition in this manner, in Africa and Asia there is some evidence that food size is not significantly correlated with bat size (Thomas, 1982; Utzurram, 1995).

Research Needs

Some E. helvum exhibited search behavior close to the day roost while others dispersed outside of the national park, traveling at least 25 km away from the roost to feed. If E. helvum flies long distances to its feeding sites, it is logical that at those sites some individuals should encounter the same competition from resident fruit bats as within the park. This cannot be tested without conducting additional sampling outside of the national park and comparing the species assemblage at each location.

Surprisingly, the number of fruiting trees peaked at about 10 km away from the roost site, and this trend was independent of the direction one traveled from the roost.

One explanation for this trend is that E. helvum is feeding at a distance away from the roost such that its feces fall when they are 6 to 12 km from the day roost. Due to the large size of many of the fruits in this region (with one notable exception being Ficus

spp.), the bats feed upon the fruits but do not swallow the seeds. Understanding the

distribution of other fruit bat species in the park and their use of these fruit trees may help explain these interesting results.

Potential directions for further studies include a comprehensive survey of fruit bat

species in Kasanka National Park and identifying other migratory species. Other avenues

for research at Kasanka include estimating population sizes of the many bat species,

identifying rare species, and investigating whether the E. helvum colony influences the

behavior of some species more than others.

The role of decreasing fruit availability in determining the foraging behavior of bat

populations would be another interesting research topic. This would involve creating a

comprehensive list of food sources and identifying whether resource switching occurs

between seasons. This study concentrated on fruit supply in Kasanka, but many

frugivorous bats also use a variety of flowers and leaves to supplement their diets.

Thomas (1982) found that several of the bat species in this study used flowers only as a

dry season, transient resource, so their exclusion in this rainy season study may have

limited consequences. Flowers could compose a larger proportion of the fruit bats’ diet during the rest of the year.

The duration of this study only included the time period when the E. helvum colony was at Kasanka and we are left with a large gap in our understanding concerning the bat assemblage structure during the rest of the year. Variations in resource distribution and

abundance throughout the year could affect the species assemblage, habitat use, and

roosting sites. Since the migratory habits of the other species in this study are uncertain,

it is possible that some epauleted fruit bats are migratory. E. wahlbergi has been shown

to be migratory in other areas of Africa (Thomas, 1982), and could be migratory at

Kasanka as well.

Conservation and Management Implications

Migratory and non–migratory Megachiroptera species have economically and ecologically important roles in ecosystems as seed dispersers, pollinators, and prey

(Fujita and Tuttle, 1991; Polis et al., 1997; Taylor and Kankam, 1999; Fleming and Eby,

2003). Yet we currently have little information about the structure of fruit bat assemblages in Africa and the interactions of these bats with their environment (Jones,

1972; Fenton, 1975; O'Shea and Vaughan, 1980; Marshall and McWilliam, 1982;

Thomas, 1982; Taylor and Kankam, 1999).

Deforestation is a disturbance of particular concern in Zambia (Smith and Fisher,

2001), but there is little basic information on regional bat species and their habitat use,

food sources, and roost sites. Foraging and roosting sites have the potential to limit the

bat fauna of an area, and these same factors are those most likely to be affected by habitat

destruction (Fenton et al., 1992; Fenton and Rautenbach, 1998). As forests diminish in

size, the reduction in food and roost availability could affect the sustainability of both bat

populations and bat dispersed or pollinated plant communities (Utzurram, 1995; Cosson,

Pons and Masson, 1999; Taylor and Kankam, 1999; Kalko and Handley, 2001).

This research provides us with some insight into fruit bat biology and behavior at

Kasanka National Park. The fruit bat species assemblage at Kasanka has unique features

and complex relationships that can contribute to large–scale ecosystem function.

Migratory bats cover long distances, and their movements can have important large–scale

seed dispersal and forest regeneration implications. To effectively conserve fruit bats,

main and seasonal roosting sites, foraging areas, and the connections between roosts and

69 feeding sites must be identified and protected at both the system and landscape scale

(Racey and Entwistle, 2003). The recognized ecological and economic importance of fruit bats (Fujita and Tuttle, 1991; Taylor and Kankam, 1999) should encourage the study of African fruit bat assemblages as a first step towards conserving the species and the habitats in which they live.

CHAPTER 5 CONCLUSIONS, RESEARCH AND MANAGEMENT RECOMMENDATIONS

Conclusions

This study contributes to the general fields of taxonomy, foraging ecology, and migratory ecology for multiple fruit bat species in a largely unstudied region. Chapter 2 provides strong evidence to support the hypothesis that Eidolon helvum migrates through

Kasanka National Park to exploit seasonally abundant resources, with the large colony arriving just prior to seasonal peaks in food availability, and departing when fruit availability decreases. Trees of the genera Uapaca and Syzygium may be key in timing this migration, and if so, Uapaca would be a new genus to be of such importance to E. helvum in Africa. The regular presence of pregnant females (Stuart and Stuart, 2001;

Racey, 2002), with a high degree of asynchrony, suggests that the colony at Kasanka is an aggregate of many smaller satellite colonies from widespread parts of Africa. Exact migratory cues for the species are unknown, including whether other migratory fruit bat species use the same cues, and no specific information exists on migratory bats’ movements in southern Africa.

Chapter 3 shows that Kasanka National Park is home to at least six species of

Megachiroptera. This study doubled the known number of bat species previously recorded at Kasanka, with insectivorous bats increasing by seven new species and the discovery of two new frugivorous species. I have no reason to believe that this list is exhaustive, and there is still much to learn about species in the region. I showed that the use of multivariate techniques to analyze measurement data does not clearly separate

70 71

species, and that mathematical methods need to be combined with complete palate

records to confidently identify sympatric epauleted and dwarf epauleted fruit bat species.

In Chapter 4 I addressed the influence of environmental variables on the

assemblage structure of fruit bats, and whether the large migratory E. helvum colony

noticeably interacts with the resident fruit bat species. I concluded that capture rates

were not significantly correlated with many common environmental variables. However

capture rates did increase over time, and although rates were expected to decrease as food

availability decreased, the highest capture rates were in fact achieved in transects when

food availability was at its lowest point. I also found evidence that E. helvum and the

resident epauleted fruit bat species may be vertically or geographically stratifying

resources.

Recommendations for Further Research

Our general knowledge about fruit bat biology and behavior would benefit from

both continued studies at Kasanka National Park and monitoring of large–scale fruit bat

movements. Closer monitoring of individual bats through the use of either radio– telemetry equipment or satellite transmitters could provide invaluable data. Satellite tracking would greatly expand our current knowledge of the movements of African fruit bats, but unfortunately the equipment is currently too large for African Megachiroptera as well as expensive. Development of lighter, cheaper satellite technology would greatly assist fruit bat research in Africa.

Effective conservation strategies for migratory bats will require a landscape–based

approach (Racey and Entwistle, 2003). Further information is needed on conditions at

additional stops on the migration path including food sources, resource abundance,

colony sizes, and roosting sites. It has been shown that migratory fruit bats may be

72 under–protected in current reserves (Palmer and Woinarski, 1999), yet we currently have no information on what sites may be vital to E. helvum and other migratory African bat species. Preserving only one stage in the migration route is inadequate to ensure the continued existence of migratory colonies, and identification of roost sites and migration routes before and after Kasanka should be a key priority for any migratory species.

Partnerships can be established with other wildlife organizations in Africa in an effort to facilitate information exchange about fruit bats and their movements. Compilation of simple records, including the dates of arrival or departure for known fruit bat colonies, may help us understand the timing and location of movements and identify vulnerable roost sites.

The degree of asynchrony in reproductive cycles indicates that the migratory E. helvum colony at Kasanka could be composed of smaller satellite colonies from a large geographic area in Africa. Learning about gene flow and the relatedness of satellite colonies could facilitate conservation efforts as well as provide information about the degree of mixing between colonies. Large congregations of bats coming together from potentially different areas of the continent may provide insight into diseases of concern, including Ebola or Nipah virus.

Multiple studies using less expensive radio–telemetry equipment could provide supplementary information about fruit bat ecology and behavior in Central Zambia. This region shows the potential for a host of interesting foraging studies, with seasonal shifts in food supply, responses in foraging behavior, and interactions with migratory species.

The temporal and spatial heterogeneity of fruit production may influence foraging behavior in novel and interesting ways, making a closer examination of the behavior of

both the migratory colony and resident species worthwhile. Monitoring migratory fruit

bats would provide information about whether these bats are naïve about resource distribution in the new area, leading to longer search times relative to resident species.

Resident fruit bats may have an advantage over visiting species through their a priori

knowledge of feeding sites.

Another research topic that has not been addressed in this area is the fidelity of both

resident and migratory fruit bats to their feeding sites. If the same fruit bats migrate to

Kasanka in consecutive years, do they revisit the same foraging areas? When the

migratory colony is at Kasanka, do individual bats return to the same feeding sites, or are

foraging decisions unpredictable? Monitoring resident bats throughout the year could

provide data to test whether the presence of the E. helvum colony affects resident species

in their use of traditional feeding sites.

Since this study only examined fruit bat behavior when the E. helvum colony was at

Kasanka, we are left with a large gap in our knowledge regarding the bat assemblage

structure during the rest of the year. We could expect assemblage changes if species

other than E. helvum are migratory, as well as behavioral adjustments because of a

changing food supply. More intensive mist netting, stratified both throughout the seasons

as well as in additional habitats, has the potential to identify more microchiropteran and

megachiropteran species, rare species, or other transient species. This would also provide

an opportunity to augment the current species data and allow more rigorous morphology

studies.

The size of the migratory E. helvum colony has led to concerns that it may destroy

the rare habitat it is using for a roosting site. Roost fidelity is variable among bat species

(Lewis, 1995), and it is unclear where the colony may go if this roost site becomes

uninhabitable. The colony has only been recorded at Kasanka beginning in the late

1980’s, and the length of time colonies use stopover sites is unknown. We also do not

know what effect the large colony has, either beneficial or detrimental, on the continued

existence of the mushitu. Although the colony may supplement the habitat through large

inputs of guano, it is also extremely destructive, and most of the large roost trees have

already been stripped of their branches. When bats rely on specific habitats or food

sources, habitat transformations may seriously affect the viability of bat populations

(Kalko, 1998; Fleming and Eby, 2003; Patterson et al., 2003).

Management Recommendations

The first recommended management step is to inventory bat species in the national

park, not only to increase our knowledge of regional biodiversity but also to identify what

species we are protecting. The dynamics of the bats in the region could be fragile, and

we do not know what the implications are for forests should this balance be disrupted.

Zambia, and southern Africa, is largely deforested and is becoming more so each day.

Disturbance levels may have important implications for the bat species in the area

(Fenton et al., 1992; Fenton and Rautenbach, 1998) and consequences for forest regeneration and food supplies should the bat species composition be changed (Fujita and

Tuttle, 1991; Duncan and Chapman, 1999; Taylor and Kankam, 1999). If bat populations

decline we could see a corresponding decrease in bat dispersed or pollinated plant

communities (Ayensu, 1974; Utzurram, 1995; Taylor and Kankam, 1999).

The most serious threat to fruit bat populations is man. While bats are preyed upon

in the wild, human hunting and habitat destruction are more prevalent (Marshall, 1983;

Fenton and Rautenbach, 1998). African fruit bats have been recorded using at least 38

families of plants as food sources, and many of these plant families are also used by man

(Marshall, 1985). With deforestation a growing problem in Africa, shifts from native

forests to agricultural fields can have serious implications for the resource availability for

the species in this study (Chidumayo, 1987; Utzurram, 1995). Megachiroptera play an important role in reforestation, and they can be responsible for dropping 90–98% of the

seeds that are potential forest regenerators (Thomas, 1982; Fujita and Tuttle, 1991). E.

helvum has already been shown to play a major role in seed dispersal of economically

important African trees (Fujita and Tuttle, 1991; Taylor and Kankam, 1999).

Although the local tribes around Kasanka do not eat fruit bats, education projects are needed to ensure that the local people recognize the value of bats and do not inadvertently cause harm through the careless destruction of vital habitat. An education partnership has already been initiated with the Jacksonville Zoo, the Bat Taxonomic

Advisory Group, and the Kasanka Trust. These groups are working together to develop education materials to be used by the Kasanka Community Project with local community conservation clubs. The project aims to educate the community, demystify bats, and teach that many of the stigmas about bats are unwarranted. Most importantly, the project aims to teach the community that bats have ecological value and educate about the importance of protecting habitats such as mushitu forest.

While the species in this study are not IUCN listed, many populations are reportedly in decline (Mickelburgh, Hutson and Racey, 1992). Future projects can use the results of this study to focus research efforts on the effects of disturbance on bat communities, identifying key bat–plant interactions, and exploring the role migratory species play in maintaining African ecosystems. Without knowing the assemblage

composition, food sources, movement patterns, and habitat requirements of the species

we are trying to protect it is difficult to develop effective conservation strategies (Fujita

and Tuttle, 1991). Effective conservation strategies for fruit bats will involve preserving

main and seasonal roosting sites, foraging areas, and the connections between roosting

and foraging areas (Racey and Entwistle, 2003).

Megachiroptera are ecologically and economically important species (Ayensu,

1974; Fujita and Tuttle, 1991; Taylor and Kankam, 1999), and their extensive movements following seasonal resources can have ecologically important local and landscape level consequences (Polis et al., 1997). A migratory colony of the size described in this study can move significant amounts of energy and nutrients across habitats and have an

important influence on the dynamics of recipient food webs (Huxel et al., 2002). Since migratory bats can cover thousands of kilometers during their migration phase, this loss of nutrients can have a significant effect on ecosystems across a significant portion of

Africa (Polis et al., 1997). Kasanka National Park is a system composed of unique features and complex relationships, and understanding the species in the region and their

interactions is the first step towards conserving the species and the habitats in which they live.

APPENDIX A MORPHOLOGICAL DATA

Table A–1. Published and unpublished measurements of some sympatric African fruit bat species. Data are from (Jones, 1972; Fenton, 1975; Marshall and McWilliam, 1982; Kingdon, 1984; Bergmans, 1988; 1989; Stuart and Stuart, 1989; Apps, 1996; Nowak, 1997; Taylor, 2000; Reid, 2003c; a; b). FA is forearm length, WS is wingspan, P is number of postdental palatal ridges. Species Common Author Year Sex Mass FA Body WS P Name Name (g) (mm) (mm) (mm) Epauleted Family Nowak 1997 40–120 60–100 125–250 508b Epomophorus Peters' Kingdon 1974 F 79–80 130–145 gambianus Epauleted Kingdon 1974 M 81–85 135–160 crypturus Fruit Bat Fenton 1975 F 85.9 79–80 2 Fenton 1975 M 81–85 2 Bergmans 1988 F 56–83 76.8–82.8 Bergmans 1988 Mc 99–136 82.5–88.4 Stuart 1989 80–140 83 150 560 Smithers 1996 F 76–88 76–82 110–125 450 2 Smithers 1996 M 105–140 80–86 130–170 500 2 Reid 2003 F 79–80 2 Reid 2003 M 81–85 2 Epomophorus Gambian Fenton 1975 F 100.9 gambianus Epauleted Fenton 1975 Fa 71.9 gambianus Fruit Bat Fenton 1975 F 81–86 2 Fenton 1975 M 87–93 2 Marshall 1982 M,F 62–84 74.8–80.5 Taylor 2000 F 56–100 75–84 2 Taylor 2000 M 91–140 80–88 2 Reid 2003 F 79–86 2 Reid 2003 M 75–93 2 Epomophorus Wahlberg's Kingdon 1974 F 54–97 72–86 110–150 wahlbergi Epauleted Kingdon 1974 M 60–120 77–98 115–175 Fruit Bat Fenton 1975 F 80.4 Fenton 1975 Fa 58.5 Fenton 1975 M 107.7 Fenton 1975 72–89 1 Bergmans 1988 F 54–125 67.7–87.5 1 Bergmans 1988 M 60–124 72.3–94.9 1 Stuart 1989 M 70–110 84 140 500 Smithers 1996 1 Taylor 2000 F 54–125 68–88 1 Taylor 2000 M 60–124 72–95 1 Reid 2003 72–89 1 a Denotes juvenile measurements, b are measurements for males only, c indicates Zambia specific data.

77 78

Table A–1. Continued. Species Common Author Year Sex Mass FA Body WS P Name Name (g) (mm) (mm) (mm) Epomophorus Ethiopian Kingdon 1974 F 38–70 55–65 95–105 labiatus Epauleted Kingdon 1974 M 48–65 62–70 105–125 Fruit Bat Bergmans 1988 F 51–81 64.8–78.3 2 Bergmans 1988 M 54–99 66.7–80.3 2 Reid 2003 66–75 2 Epomophorus East Bergmans 1988 F 25–65 63.1–65.3c 2 minor African Bergmans 1988 M 32–58 66–67c 2 Fruit Bat Reid 2003 56–68 2 Micropteropus Peters’ Jones 1972 M 10.5–27.5 pusillus Dwarf Jones 1972 F 10.9–29.4 Epauleted Kingdon 1974 25–35 48–57 75–105 Fruit Bat Marshall 1982 F 16.5–25.6 48–52.8 Marshall 1982 M 19.2–26 47–51.8 Bergmans 1989 F 20–34 49.5–55.7 Bergmans 1989 M 24–35 46.4–54.7 Nowak 1997 F 22 46–65 67–105 Nowak 1997 M 20 Reid 2003 50–54 Epomops Dobson's Bergmans 1989 F 80.4–86.5 2 dobsonii Fruit Bat Bergmans 1989 M 84.2–87.7 2 Stuart 1989 85 160 Smithers 1996 Taylor 2000 F 80–88 Taylor 2000 M 84–92 Reid 2003 80–92 1 Epomops franqueti Reid 2003 85–95 2 a Denotes juvenile measurements, b are measurements for males only, c indicates Zambia specific data.

APPENDIX B TREE SPECIES ENCOUNTERED IN THIS STUDY

Table B–1. List of tree species found at Kasanka National Park. This table gives the local name in Bemba, Lala, or Nyanja as well as the common and scientific names. Scientific Name Common Name Local Name Acacia heteracantha Kafifi Albizia antunesiana Musase Anisophyllea boemii Mufungo Brachystegia boehmii Musamba Brachystegia spiciformis Muputu Cassia abbreviata Long–pod Cassia Musambafwa Combretum molle Bush Willow Mufuka, Mulama Dalbergia nitidula Akasansabwangu Diospyros batocana Muntufita Diplorhnchus condylocarpon Mulimbo, Mwenge Erythrina abyssinica Lucky Bean Mulunguti Erythrophleum africanum Kayimbi, Mukoso Faurea intermedia African Beech Saninga Isoberlinia angolensis Mutobo Julbernardia paniculata Mutundu, Mutondo Lonchocarpus capassa Lilac Tree Chiya a Magnistipula butayeii Bush Avocado Mkuwe, Imipande Maprounea africana Kafulamume Monotes africanus Chimpampa, Chipampa Ochna schweinfurthiana Icooni Ozoroa reticulata Mubelemabele a Parinari curatellifolia Mupundu Pericopsis angolensis Mubanga Phyllocosmus lemaireanus Unknown Protea spp. Mupapa Pseudolachnostylis maprounefolia Kuduberry Musola, Musolo Pterocarpus angolensis Mukwa, Mulombe Rothmannia englerana Mwinebala Strychnos cocculoides Bush Orange Kasongole Strychnos pungens Mukome Strychnos spinosa Elephant Orange, Monkey Ball Sansa a Syzygium cordatum Waterberry Mufinsa a Syzygium guineense guineense Waterberry Musafwa, Insafwa a Syzygium guineense huillense Waterberry Mufinsa Terminalia mollis Chibobo b Uapaca nitida Nsokolobe a Uapaca kirkiana Wild Loquat Musuku, Masuku a Uapaca sansibarica Swebya a Uapaca banguelensis Makonko Zahna africana Chibangalume Unknown Kapanga a Known food sources, b Possible food source for fruit bats at Kasanka National Park.

79 80

Table B–1. Continued Scientific Name Common Name Local Name Unknown Mufila Unknown Mukomfwa Unknown Mulembo Unknown Munye Unknown Umwende

APPENDIX C CAPTURE DATA

Table C–1. Data for individual Megachiroptera netted at Kasanka National Park in 2003. U is unknown cluster, RC is reproductive condition, FA is forearm length, R is reproductive, L is lactating, P is pregnant, NR is non–reproductive, A is adult, J is juvenile. Weight is in grams, while all other measurements are in millimeters. Palatal ridge configuration (PR) gives the number and arrangement of postdental palatal ridges; five individuals have full ridge configuration data (RC). Commas indicate a gap between ridges. ‘2’ alone indicates 2 postdental ridges, but the configuration of those ridges is unknown. ID Date Time Transect ClusterSex RC Age Weight Tail Body Half Wingspan FA Foot Ear Comments RC PR No. Location group Wingspan 86 3–Nov 20:04 Shiwila 1 F R A 88 135 252 504 86 21 25 88 3–Nov 20:50 Shiwila 1 M R A 130 154 232 464 87 23 27 87 3–Nov 19:15 Shiwila 2 F NR A 93 138 242 484 80 14 23 90 4–Nov 00:45 28 2 F R A 82 136 245 490 83 17 20 91 4–Nov 01:15 28 2 F R A 84 145 232 464 83 19 21 93 4–Nov 23:20 28 2 F R A 92 120 259 518 82 20 20 94 4–Nov 04:30 31 1 F NR A 102 149 240 480 85 20 23 89 4–Nov 04:30 31 2 M NR A 67 120 227 454 75 19 20 82 92 4–Nov 02:45 31 2 F R A 89 130 210 420 83 19 23 10 10–Nov 00:15 33 1 F R A 106 147 242 484 85 19 23 97 10–Nov 01:05 33 1 F L A 112 140 260 520 90 25 25 105 10–Nov 01:45 33 1 F NR A 143 165 280 560 91 21 25 Slack lips 9 10–Nov 01:05 33 3 F R A 42 94 163 326 61 14 17 95 10–Nov 01:05 33 4 F R A 48 112 185 370 62 15 19 96 10–Nov 04:30 45 2 F R A 91 134 237 474 81 20 18 101 11–Nov 20:45 41 1 F NR A 123 155 275 550 88 24 25 Slack lips 102 11–Nov 23:50 41 1 M R A 130 155 230 460 90 20 26 Slack lips 99 11–Nov 21:20 41 2 M J J 65 118 241 482 79 21 23 100 11–Nov 21:20 41 2 F L A 77 131 247 494 78 16 23 103 12–Nov 21:20 35 2 M J J 62 135 228 456 79 20 22 11 12–Nov 19:25 35 3 F R A 41 106 188 376 62 14 16 12 16–Nov 21:30 Wasa 4 F L A 45 110 193 386 61 15 21 13 18–Nov 21:10 10 4 M J J 60 105 221 442 67 18 18 14 21–Nov 03:30 1 4 M R A 55 0 98 194 388 64 16 18

Table C–1. Continued. ID Date Time Transect ClusterSex RC Age Weight Tail Body Half Wingspan FA Foot Ear Comments RC PR No. Location group Wingspan 104 22–Nov 21:00 15 2 F L A 82 125 225 450 80 20 22 108 26–Nov 02:20 35 1 M R A 129 177 245 490 88 21 21 Slack lips 3,1,1,1 1,s,1 111 27–Nov 01:30 41 1 F L A 117 155 270 540 90 25 21 110 27–Nov 21:30 42 1 M R A 108 150 265 530 85 22 26 Slack lips 3,1,1,1 1,1 109 27–Nov 20:30 42 2 M J J 73 125 225 450 79 20 26 1,2,2 2 15 27–Nov 21:00 42 4 M R A 50 117 177 354 65 15 19 3,1,1,1 1,s,1 19 4–Dec 04:30 28 U F J J 43 3 70 23 4–Dec 04:30 28 U F PL A 60 70 26 4–Dec 04:30 28 U F L A 76 77 28 4–Dec 04:30 28 U F L A 80 83 29 4–Dec 04:30 28 U F L A 87 80 68 4–Dec 04:30 28 U F J J 30 58 113 4–Dec 20:45 33 1 M J J 82 145 265 530 81 20 23 83 114 4–Dec 20:45 33 1 F L A 99 145 250 500 86 21 22 115 4–Dec 20:45 33 1 F L A 102 160 245 490 85 20 25 112 4–Dec 21:00 33 2 M NR A 79 135 225 450 80 22 19 16 4–Dec 20:45 33 3 F J J 36 92 176 352 58 16 16 17 4–Dec 19:15 33 3 F J J 38 84 166 332 58 15 14 Short nose 20 4–Dec 20:45 33 3 F L A 45 104 170 340 60 13 18 18 4–Dec 21:00 33 4 F R A 40 105 180 360 60 18 18 24 4–Dec 22:50 45 2 F J J 65 110 210 420 75 21 20 25 4–Dec 01:15 45 2 M R A 69 4 133 234 468 78 24 22 27 4–Dec 01:15 45 2 M J J 76 130 220 440 85 19 24 67 4–Dec 23:15 45 3 F J J 25 90 140 280 57 11 15 21 4–Dec 22:30 45 4 M J J 54 115 206 412 71 15 20 22 4–Dec 22:30 45 4 M J J 54 125 205 410 74 18 20 130 6–Dec 19:30 1 2 F J J 66 120 230 460 79 20 20 69 6–Dec 22:30 7 3 M J J 24 105 145 290 56 16 11

Table C–1. Continued. ID Date Time Transect ClusterSex RC Age Weight Tail Body Half Wingspan FA Foot Ear Comments RC PR No. Location group Wingspan 116 7–Dec 19:24 23 1 M R A 115 4 152 275 550 89 22 23 Slack lips 3,1,1,1 1,1 30 7–Dec 21:10 23 2 F J J 75 7 135 245 490 82 24 23 33 8–Dec 00:30 36 2 F J J 71 4 123 230 460 83 20 24 31 8–Dec 21:00 37 2 M J J 60 0 115 207 414 72 21 21 32 8–Dec 21:00 37 2 F J J 69 5 130 230 460 78 20 24 1 13–Dec 19:10 5 1 M R A 147 0 155 240 480 90 20 29 Slack lips 2 117 13–Dec 22:05 9 1 F R A 102 3 142 275 550 88 22 26 35 13–Dec 22:05 9 2 F NR A 71 4 110 235 470 83 18 24 36 13–Dec 00:45 9 2 F J J 74 4 115 235 470 88 23 26 70 13–Dec 22:05 9 3 F J J 30 0 92 175 350 61 16 19 71 13–Dec 22:05 9 3 F J J 33 0 87 170 340 57 14 18 34 13–Dec 23:00 9 4 M R A 56 0 120 180 360 67 17 18 73 13–Dec 22:30 9 4 F J J 49 0 115 210 420 71 17 21 84 74 13–Dec 00:55 9 4 F L A 53 0 95 200 400 65 16 18 72 14–Dec 20:50 8 3 F J J 38 0 105 168 336 60 15 17 118 14–Dec 20:50 8 U F L A 104 4 137 235 470 87 30 25 2 14–Dec 00:45 10 1 M R A 110 5 148 245 490 87 21 24 Bad teeth 2 37 14–Dec 23:35 10 2 M J J 65 0 155 212 424 81 25 22 38 14–Dec 22:55 10 2 M R A 79 145 245 490 77 17 19 75 14–Dec 23:35 10 3 M J J 37 0 87 180 360 63 19 20 76 14–Dec 23:30 10 4 M R A 51 0 130 197 394 63 15 20 2 77 19–Dec 20:45 21 3 F R A 44 95 170 340 64 13 18 47 20–Dec 21:30 15 1 F L A 88 133 240 480 84 19 25 Injured R wing 2 39 20–Dec 20:30 15 2 F J J 52 97 218 436 73 21 23 40 20–Dec 20:30 15 2 M J J 61 5 130 231 462 77 21 23 43 20–Dec 20:30 15 2 F R A 79 120 230 460 89 20 23 3? 44 20–Dec 21:15 15 2 F NR A 82 130 230 460 83 22 22 3 20–Dec 23:55 19 1 F L A 107 0 146 260 520 86 21 26 Brown nose 1,s,1

Table C–1. Continued. ID Date Time Transect ClusterSex RC Age Weight Tail Body Half Wingspan FA Foot Ear Comments RC PR No. Location group Wingspan 46 20–Dec 22:35 19 1 F J J 87 125 250 500 84 22 25 1,s,1 48 20–Dec 00:15 19 1 F R A 89 0 133 245 490 80 18 23 1,s,1 119 20–Dec 23:40 19 1 M R A 112 9 150 250 500 88 20 27 V. long nose 1,s,1 41 20–Dec 23:40 19 2 F J J 75 4 120 230 460 81 19 26 s,2 42 20–Dec 23:40 19 2 F L A 76 138 210 420 78 17 22 1,s,1 45 20–Dec 23:50 19 2 F L A 83 125 223 446 82 21 22 2 78 20–Dec 23:40 19 3 M J J 25 80 152 304 56 12 15 No epauletes 121 21–Dec 22:30 8 1 M NR A 87 4 132 260 520 84 22 25 Recaptured s,2 122 21–Dec 22:50 8 1 M J J 98 5 140 250 500 87 21 28 s,2 49 21–Dec 23:00 8 2 M J J 57 3 113 234 468 76 21 21 1,s,1 52 21–Dec 22:30 8 2 M J J 62 3 120 198 396 76 22 22 1,s,1 53 21–Dec 00:15 8 2 M J J 65 114 220 440 80 18 23 1,s,1

54 21–Dec 23:45 8 2 M J J 84 3 120 220 440 83 22 27 85 56 21–Dec 22:10 8 2 F J J 89 5 130 210 420 79 22 24 s,2 120 21–Dec 23:45 8 2 M J J 82 133 245 490 84 24 22 s,2 79 21–Dec 23:20 8 3 M J J 31 90 175 350 58 14 18 80 21–Dec 22:50 8 3 F L A 41 0 100 180 360 62 14 17 1,s,1 50 21–Dec 00:35 8 4 M J J 58 4 118 220 440 73 15 23 1,s,1 51 21–Dec 23:45 8 4 M R A 61 115 185 370 69 16 21 55 21–Dec 00:15 8 U F PL A 85 135 225 450 79 21 23 1,s,1 81 21–Dec 21:30 9 U F R A 44 0 65 17 19 57 29–Dec 20:30 20 2 M J J 65 4 120 220 440 79 21 23 1,s,1 58 29–Dec 21:30 20 2 M J J 75 6 110 240 480 81 23 25 s,2 65 30–Dec 23:00 1 1 F R A 96 4 135 245 490 80 21 22 Bad teeth 1,s,1 125 30–Dec 01:00 1 1 M J J 79 0 135 255 510 85 20 27 s,2 129 30–Dec 23:00 1 1 F R A 105 120 245 490 86 22 24 s,2 59 30–Dec 01:00 1 2 F J J 55 4 110 205 410 72 18 23 1,s,1 60 30–Dec 23:00 1 2 M J J 60 4 120 230 460 88 20 21 1,s,1

Table C–1. Continued. ID Date Time Transect ClusterSex RC Age Weight Tail Body Half Wingspan FA Foot Ear Comments RC PR No. Location group Wingspan 61 30–Dec 01:00 1 2 M J J 61 3 122 215 430 75 20 22 1,s,1 62 30–Dec 01:00 1 2 M J J 66 3 118 225 450 85 22 21 1,s,1 63 30–Dec 01:00 1 2 M J J 68 4 130 233 466 77 22 23 1,s,1 64 30–Dec 01:00 1 2 F R A 84 0 130 220 440 77 19 21 1,s,1 123 30–Dec 00:00 1 2 M J J 68 3 115 235 470 82 22 25 s,2 124 30–Dec 01:00 1 2 F J J 78 4 120 245 490 83 23 16 s,2 126 30–Dec 22:33 1 2 M J J 80 6 125 250 500 80 24 26 s,2 127 30–Dec 23:00 1 2 F J J 80 5 121 235 470 82 22 25 s,2 82 30–Dec 01:00 1 3 F J J 29 0 84 165 330 56 17 20 83 30–Dec 22:33 1 3 M J J 29 85 165 330 58 14 17 84 30–Dec 23:00 1 3 F L A 42 4 95 165 330 60 14 16 85 30–Dec 01:00 1 U F R A 47 0 185 190 380 64 15 17 128 30–Dec 19:45 7 1 F R A 105 5 142 230 460 82 21 26 s,2 86 66 31–Dec 01:00 1 2 F J J 61 5 112 215 430 75 19 22 1,s,1

Table C–2. Data collected on Microchiroptera netted at Kasanka National Park in 2003. RC is reproductive condition, FA is forearm length, R is reproductive, L is lactating, P is pregnant, NR is non–reproductive, A is adult, J is juvenile. Weight is in grams, while all other measurements are in millimeters. Record Date Time Site Species Sex RC Weight Tail Body Half Wing FA Foot Ear Tragus Notes No. Wingspan span 1 28–Oct 21:30 Wasa Nycteris thebaica M NR 17.5 53 53 131 262 55 9 25 2 3–Nov 00:00 Kapabi Pipistrellis sp. M U 5 32 43 95 190 31 5 6 2 3 11–Nov 21:20 41 Scotoecus hirundo F NR 8 41 60 120 240 37 5 15 5 4 16–Nov 18:30 Wasa Eptesicus somalicus M 5 33 5 16–Nov 20:30 Wasa Pipistrellis kuhlii F R about 4 31 11 6 16–Nov 23:05 Wasa Nycticeus schleiffeni F L 31 12 7 7 17–Nov 18:30 Wasa Pipistrellis somalicus F 31 8 17–Nov 20:10 Wasa Pipistrellis kuhlii F L 31 9 21–Nov 02:00 1 Miniopterus sp. F R 10 42 93 139 278 43 9 8 5 10 21–Nov 02:00 1 Myotis welwitschii F R 18.5 51 73 183 366 55 11 21 10 11 21–Nov 00:00 3 Scotoecus hirundo F L 9 38 47 123 246 35 6 13 6

12 22–Nov 21:00 15 Pipistrellis kuhlii M about 4 26 40 115 230 31 5 8 3 87 13 26–Nov 18:40 36 Eptesicus capensis F P?R 5 28 51 101 202 33 7 10 4 14 26–Nov 18:40 36 Eptesicus capensis M R? 4 28 48 105 210 31 7 10 4 Half of L ear gone 15 4–Dec 18:30 33 Eptesicus capensis M R? about 5 45 114 228 33 6 5 3 16 6–Dec 18:45 1 Eptesicus capensis F P? 4 29 52 92 184 32 6 9 3 17 6–Dec 19:40 1 Scotoecus hirundo M J 5 40 55 110 220 35 5 11 6 18 6–Dec 20:30 1 Eptesicus capensis M J 4 30 48 92 184 31 5 9 2 19 8–Dec 23:30 36 Scotoecus hirundo F J 5 29 47 106 212 32 5 7 3 20 8–Dec 19:10 37 Scotoecus hirundo M J 4 32 50 85 170 31 6 8 2 21 8–Dec 19:10 37 Eptesicus capensis F P 5 29 50 88 176 34 6 9 3 22 13–Dec 22:05 9 Nycteris thebaica M R 51 55 140 280 50 10 25 23 13–Dec 23:00 9 Pipistrellus kuhlii M R 4 25 42 87 174 32 4 10 5 24 19–Dec 19:05 21 Eptesicus capensis M R? about 5 29 43 108 216 33 5 9 4 Redder dorsal w/pale belly 25 21–Dec 19:00 9 Myotis welwitschii M R 17 52 72 165 330 61 12 23 12 26 24–Dec 19:00 10 Scotoecus hirundo M R? 28 47 92 184 31 7 8

APPENDIX D MIST NETTING LOCATIONS

Table D–1. Dates, locations, and coordinates for mist netting sites. Coordinates are datum Arc1950 in UTM zone 36 L. Date Transect Net 1 Net 1 Net 1 Net 1 Net 2 Net 2 Net 2 Net 2 Net 3 Net 3 Net 3 Net 3 Start E Start S End E End S Start E Start S End E End S Start E Start S End E End S 4–Nov 19 0202016 8607514 0202006 8607520 0202017 8607521 0202020 8607068 0202054 8607530 0202056 8607534 4–Nov 28 0196481 8611205 0196484 8611204 0196482 8611206 0196491 8611208 0196493 8611206 0196490 8611205 4–Nov 31 0194481 8611945 0194485 8611947 0194491 8611955 0194487 8611958 0194492 8611962 0194481 8611959 5–Nov 5 0210196 8610803 0210193 8610798 0210149 8610802 0210145 8610804 0210146 8610787 0210155 8610795 5–Nov 8 0207957 8611311 0207968 8611314 0207959 8611311 0207963 8611309 0207955 8611313 0207958 8611318 7–Nov 21 0201609 8609680 0201606 8609680 0201594 8609683 0201588 8609683 0201586 8609690 0201578 8609685 8–Nov 36 0188747 8616008 0188747 8616011 0188746 8616011 0188750 8616008 0188753 8616006 0188755 8615994 8–Nov 37 0188985 8615090 0188983 8615087 0188975 8615060 0188974 8615055 0188967 8615054 0188978 8615045 10–Nov 26 0197984 8609774 0197983 8609786 0197983 8609786 0197978 8609787 10–Nov 33 0192464 8612122 0192472 8612140 0194475 8612138 0192480 8612138 0192476 8612122 0192473 8612121 10–Nov 45 0193185 8610449 0193183 8610446 0193184 8610457 0193190 8610459 0193205 8610458 0193216 8610452

89 11–Nov 41 0190080 8611885 0190070 8611882 0190071 8611892 0190071 8611897 0190065 8611897 0190062 8611894 11–Nov 42 0191012 8611290 0190999 8610733 0191002 8611299 0191001 8611302 0190993 8611304 0190990 8611305 12–Nov 35 0188584 8616406 0188587 8616408 0188575 8616407 0188575 8616411 0188582 8616427 0188591 8616432 12–Nov 37 0189021 8614490 0189025 8614490 18–Nov 10 0205545 8610939 0205541 8610937 0205538 8610938 0205545 8610945 0205546 8610948 0205543 8610951 21–Nov 1 0214917 8610783 0214915 8610782 0214910 8610794 0214907 8610794 0214894 8610792 0214889 8610801 21–Nov 3 0212751 8610750 0212738 8610757 0212743 8610748 0212751 8610749 0212740 8610753 0212745 8610756 21–Nov 7 0209060 8610671 0209057 8610682 0209055 8610684 0209052 8610684 0209058 8610667 0209049 8610666 22–Nov 15 0204117 8607717 0204038 8607720 0204124 8607733 0204136 8607733 0204128 8607727 0204125 8607729 26–Nov 35 0188600 8616457 0188600 8616455 0188580 8616425 0188585 8616423 0188574 8616419 0188568 8616417 26–Nov 36 0188804 8615443 0188807 8615444 0188812 8615444 0188818 8615437 0188830 8615420 0188831 8615418 26–Nov 38 0188794 8613526 0188798 8613529 0188788 8613536 0188786 8613535 0188774 8613537 0188768 8613527 27–Nov 41 0190089 8611876 0190072 8611881 0190076 8611882 0190076 8611881 0190089 8611866 0190089 8611862 27–Nov 42 0190990 8611310 0190984 8611309 0191004 8611304 0191011 8611316 0191003 8611300 0191005 8611296 27–Nov 43 0191388 8610598 0191384 8610602 0191384 8610602 0191374 8610611 0191384 8610595 0191383 8610591 4–Dec 28 0196519 8611200 0196513 8611202 0196508 8611197 0196507 8611196 0196502 8611203 0196495 8611196

Table D–1. Continued. Date Transect Net 1 Net 1 Net 1 Net 1 Net 2 Net 2 Net 2 Net 2 Net 3 Net 3 Net 3 Net 3 Start E Start S End E End S Start E Start S End E End S Start E Start S End E End S 4–Dec 33 0192470 8612125 0192480 8612130 0192474 8612124 0192476 8612122 0192491 8612133 0192490 8612135 4–Dec 45 0193162 8610448 0193158 8610446 0193194 8610458 0193203 8610464 0193219 8610458 0193214 8610457 6–Dec 1 0214924 8610794 0214921 8610798 0214913 8610793 0214914 8610790 0214903 8610793 0214896 8610801 6–Dec 7 0209033 8610683 0209038 8610685 0209042 8610681 0209048 8610683 0209054 8610677 0209055 8610667 7–Dec 23 0199654 8608883 0199657 8608883 0199662 8608881 0199670 8608876 0199679 8608874 0199683 8608872 13–Dec 5 0210156 8610810 0210157 8610808 0210175 8610802 0210186 8610800 0210212 8610821 0210214 8610819 13–Dec 9 0206711 8611096 0206702 8611089 0206697 8611087 0206692 8611088 0206719 8611097 0206725 8611099 14–Dec 8 0207987 8611313 0207999 8611317 0208014 8611330 0208018 8611330 0208018 8611320 0208013 8611321 14–Dec 10 0205540 8610939 0205532 8610933 0205554 8610895 0205553 8610898 0205553 8610925 0205557 8610927 19–Dec 21 0201664 8609111 0201665 8609109 0201660 8609118 0201660 8609120 0201646 8609119 0201636 8609115 20–Dec 15 0204133 8607733 0204144 8607731 0204157 8607751 0204157 8607754 0204163 8607762 0204163 8607765

20–Dec 19 0202003 8607495 0202016 8607500 0202019 8607512 0202016 8607514 0202043 8607524 0202040 8607530 90 21–Dec 8 0208021 8611321 0208016 8611322 0208001 8611322 0208003 8611323 0207992 8611319 0207992 8611321 21–Dec 9 0206680 8611091 0206679 8611087 0206678 8611097 0206683 8611096 0206690 8611104 0206701 8611103 24–Dec 10 0205541 8610929 0205538 8610919 0205551 8610902 0205550 8610898 0205557 8610872 0205554 8610874 29–Dec 20 0202346 8608479 0202342 8608486 0202341 8608479 0202340 8608482 0202338 8608501 0202340 8608505 30–Dec 1 0214923 8610797 0214920 8610800 0214910 8610791 0214906 8610791 0214894 8610792 0214888 8610804 30–Dec 7 0209016 8610687 0209016 8610690 0209009 8610695 0209004 8610705 0209049 8610680 0209052 8610681

APPENDIX E VEGETATION TRANSECT LOCATIONS

Table E–1. Vegetation transect locations. Coordinates are datum Arc1950 in UTM zone 36 L. R is a transect on a road while O indicates a transect located in the bush, 100m off the road. Transect Habitat Type Start E Start S End E End S R/O T1 Riverine 0214923 8610790 0214845 8610769 R T2 Miombo woodland 0214009 8610766 0213920 8610766 R T3 Miombo woodland 0212785 8610759 0212689 8610747 R T4 Miombo woodland 0211536 8610892 0211436 8610901 R T5 Miombo woodland 0210227 8610821 0210138 8610793 R T6 Miombo woodland 0210216 8610925 0210218 8611018 O T7 Miombo woodland 0209069 8610656 0209002 8610721 R T8 Miombo woodland 0208037 8611334 0207952 8611316 R T9 Miombo woodland 0206740 8611096 0206655 8611092 R T10 Miombo woodland 0205540 8610954 0205563 8610866 R T11 Dambo margin 0204964 8610355 0204876 8610395 R T12 Dambo with termitaria 0204927 8610270 0204900 8610186 O T13 Dambo 0204236 8609698 0204285 8609585 R T14 Dambo margin 0204315 8608924 0204313 8608842 R T15 Miombo woodland 0204161 8607750 0204091 8607703 R T16 Mixed miombo/acacia 0202954 8607153 0203027 8607188 R T17 Mixed miombo/acacia 0201869 8606920 0201790 8606948 R T18 Mixed miombo/acacia 0201875 8607017 0201947 8607078 O T19 Short miombo 0202006 8607492 0202071 8607562 R T20 Chipya 0202337 8608539 0202348 8608444 R T21 Brachystegia dominated miombo 0201681 8609103 0201598 8609117 R T22 Dambo margin 0200633 8609098 0200576 8609016 R T23 Miombo woodland 0199712 8608860 0199631 8608895 R T24 Miombo woodland 0199708 8608765 0199696 8608677 O T25 Miombo woodland 0198815 8609066 0198739 8609125 R T26 Miombo woodland 0198010 8609758 0197945 8609809 R T27 Isoberlinia dominated miombo 0197149 8610382 0197102 8610456 R T28 Miombo woodland 0196549 8611198 0196461 8611206 R T29 Miombo woodland 0195525 8611465 0195455 8611517 R T30 Miombo woodland 0195597 8611525 0195679 8611555 O T31 Miombo woodland 0194557 8611949 0194467 8611946 R T32 Miombo woodland 0193510 8611972 0193436 8612016 R T33 Miombo woodland 0192509 8612148 0192431 8612117 R T34 Dambo margin 0192955 8611592 0193013 8611526 R T35 Miombo woodland 0188562 8616385 0188596 8616474 R T36 Miombo woodland 0188840 8615391 0188777 8615462 R

91 92

Table E–1. Continued. Transect Habitat Type Start E Start S End E End S R/O T37 Chipya 0188989 8614442 0189044 8614528 R T38 Chipya 0188816 8613524 0188736 8613550 R T39 Chipya 0189461 8612710 0189404 8612778 R T40 Chipya 0189324 8612736 0189224 8612722 O T41 Miombo woodland 0190098 8611855 0190042 8611922 R T42 Miombo woodland 0191030 8611287 0190953 8611334 R T43 Chipya 0191390 8610536 0191396 8610630 R T44 Dambo margin 0192223 8610254 0192141 8610248 R T45 Brachystegia dominated miombo 0193260 8610445 0193173 8610456 R T46 Dambo 0193171 8610546 0193169 8610636 O

APPENDIX F FRUIT TREE LOCATIONS

Table F–1. Coordinates of trees monitored for phenology in 2003. Coordinates are datum Arc1950 in UTM zone 36 L. R is a transect on a road while O indicates a transect located in the bush, 100m off the road. Tree ID Transect R/O Local Name Tree Species DBH (cm) 1–M–1 1 R Mufinsa Syzygium cordatum 35.0 1–M–2 1 R Mufinsa Syzygium cordatum 26.4 1–M–3 1 R Mufinsa Syzygium cordatum 17.2 1–M–4 1 R Mufinsa Syzygium cordatum 30.0 2–M–1 2 R Mufungo Anisophyllea boehmii 9.2 2–M–2 2 R Mufungo Anisophyllea boehmii 9.3 2–M–3 2 R Mufungo Anisophyllea boehmii 8.3 3–I–1 3 R Icisongole Strychnos coccoides 5.0 3–M–1 3 R Insafwa Syzygium guineense 4.2 3–R–1 3 R Unknown Rothmannia englerana 1.2 3–U–1 3 R Masuku Uapaca kirkiana 6.5 3–Z–1 3 R Musola Pseudolachnostylis maprounefolia 15.0 4–B–1 4 R Icooni Ochna schweinfurthaii 11.7 4–B–2 4 R Icooni Ochna schweinfurthaii 27.8 4–F–1 4 R Unknown Phyllocosmus lemaireanus 5.0 4–M–1 4 R Insafwa Syzygium guineense 4.6 4–M–2 4 R Insafwa Syzygium guineense 9.7 4–M–3 4 R Insafwa Syzygium guineense 10.2 4–M–4 4 R Insafwa Syzygium guineense 9.1 4–M–5 4 R Insafwa Syzygium guineense 8.9 4–M–6 4 R Insafwa Syzygium guineense 6.3 4–M–7 4 R Insafwa Syzygium guineense 4.9 4–M–8 4 R Mufungo Anisophyllea boehmii 4.8 4–N–1 4 R Nsokolobe Uapaca nitida 12.7 4–R–1 4 R Unknown Rothmannia englerana 2.4 4–R–2 4 R Unknown Rothmannia englerana 3.4 5–B–1 5 R Icooni Ochna schweinfurthaii 3.4 5–L–1 5 R Kafifi Acacia heteracantha 4.9 5–M–1A 5 R Insafwa Syzygium guineense 6.3 5–M–1B 5 R Insafwa Syzygium guineense 6.2 5–M–2 5 R Insafwa Syzygium guineense 4.7 5–M–3 5 R Insafwa Syzygium guineense 8.0

93 94

Table F–1. Continued. Tree ID Transect R/O Local Name Tree Species DBH (cm) 5–M–4 5 R Insafwa Syzygium guineense 4.8 5–N–1 5 R Nsokolobe Uapaca nitida 12.1 5–N–2 5 R Nsokolobe Uapaca nitida 15.2 5–U–1 5 R Masuku Uapaca kirkiana 15.4 5–U–2 5 R Masuku Uapaca kirkiana 13.3 5–U–3 5 R Masuku Uapaca kirkiana 13.7 5–U–4 5 R Masuku Uapaca kirkiana 18.1 5–U–5 5 R Masuku Uapaca kirkiana 14.5 5–U–6 5 R Masuku Uapaca kirkiana 12.2 6–N–1 6 O Nsokolobe Uapaca nitida 11.3 6–N–2 6 O Nsokolobe Uapaca nitida 13.3 6–U–2 6 O Masuku Uapaca kirkiana 9.0 6–U–3 6 O Masuku Uapaca kirkiana 10.6 6–U–4 6 O Masuku Uapaca kirkiana 24.6 6–U–5 6 O Masuku Uapaca kirkiana 10.1 6–U–6 6 O Makonko Uapaca banguelensis 12.9 6–U–7 6 O Makonko Uapaca banguelensis 11.5 7–C–1 7 R Kafulamume Maprounea africana 7.0 7–L–1 7 R Kafifi Acacia heteracantha 5.0 7–M–1 7 R Insafwa Syzygium guineense 14.7 7–M–10 7 R Insafwa Syzygium guineense 4.8 7–M–2 7 R Insafwa Syzygium guineense 5.3 7–M–3 7 R Insafwa Syzygium guineense 9.5 7–M–4 7 R Insafwa Syzygium guineense 11.3 7–M–5 7 R Insafwa Syzygium guineense 11.7 7–M–6 7 R Insafwa Syzygium guineense 8.3 7–M–7 7 R Insafwa Syzygium guineense 9.9 7–M–8 7 R Insafwa Syzygium guineense 9.4 7–M–9 7 R Insafwa Syzygium guineense 6.8 7–N–8 7 R Swebya Uapaca sansibarica 11.9 7–O–1 7 R Unknown Unknown 2.4 7–O–2 7 R Unknown Unknown 3.1 7–O–3 7 R Unknown Unknown 2.7 7–U–10 7 R Masuku Uapaca kirkiana 19.0 7–U–11 7 R Makonko Uapaca banguelensis 12.4 7–U–12 7 R Makonko Uapaca banguelensis 8.5 7–U–5 7 R Makonko Uapaca banguelensis 7.8 7–U–6 7 R Masuku Uapaca kirkiana 15.0 7–U–7A 7 R Makonko Uapaca banguelensis 14.0 7–U–7B 7 R Makonko Uapaca banguelensis 7.0 7–U–9 7 R Makonko Uapaca banguelensis 7.7

Table F–1. Continued. Tree ID Transect R/O Local Name Tree Species DBH (cm) 7–Z–1 7 R Musola Pseudolachnostylis maprounefolia 18.9 7–Z–2 7 R Musola Pseudolachnostylis maprounefolia 12.8 8–B–1 8 R Icooni Ochna schweinfurthaii 13.0 8–B–2 8 R Icooni Ochna schweinfurthaii 11.9 8–B–3 8 R Icooni Ochna schweinfurthaii 11.7 8–B–4 8 R Icooni Ochna schweinfurthaii 11.8 8–R–1 8 R Unknown Rothmannia englerana 3.6 8–R–2 8 R Unknown Rothmannia englerana 3.1 8–U–1 8 R Masuku Uapaca kirkiana 13.3 8–U–2 8 R Masuku Uapaca kirkiana 10.3 8–U–3 8 R Masuku Uapaca kirkiana 6.6 8–U–4 8 R Makonko Uapaca banguelensis 6.3 8–U–5 8 R Masuku Uapaca kirkiana 10.9 9–U–1 9 R Masuku Uapaca kirkiana 4.9 10–B–1 10 R Icooni Ochna schweinfurthaii 19.8 10–B–2 10 R Icooni Ochna schweinfurthaii 10.3 10–B–3 10 R Icooni Ochna schweinfurthaii 11.2 10–B–4 10 R Icooni Ochna schweinfurthaii 10.4 10–B–5 10 R Icooni Ochna schweinfurthaii 6.5 10–B–6 10 R Icooni Ochna schweinfurthaii 15.3 10–B–7 10 R Icooni Ochna schweinfurthaii 9.2 10–M–1 10 R Insafwa Syzygium guineense 9.1 10–M–2 10 R Insafwa Syzygium guineense 7.9 10–N–1 10 R Nsokolobe Uapaca nitida 11.8 11–M–1 11 R Mufinsa Syzygium guineense 18.3 11–V–1 11 R Unknown Unknown 5.1 12–M–1 12 O Mufinsa Syzygium guineense 7.6 12–M–2 12 O Mufinsa Syzygium guineense 27.1 12–M–3 12 O Mufinsa Syzygium guineense 17.4 12–M–4 12 O Mufinsa Syzygium guineense 18.4 12–M–5 12 O Mufinsa Syzygium guineense 45.8 12–N–1 12 O Swebya Uapaca sansibarica 25.1 15–M–1 15 R Insafwa Syzygium guineense 4.9 15–M–2 15 R Mufinsa Syzygium guineense 18.0 15–M–3 15 R Insafwa Syzygium guineense 1.7 15–M–4 15 R Mufinsa Syzygium guineense 17.8 16–d–1 16 R Mukomfwa Unknown 7.9 16–E–1 16 R Mukomfwa Unknown 7.2 17–I–1 17 R Icisongole Strychnos spinosa 17.2 17–I–2 17 R Icisongole Strychnos spinosa 19.4 17–I–3 17 R Icisongole Strychnos spinosa 13.6

Table F–1. Continued. Tree ID Transect R/O Local Name Tree Species DBH (cm) 19–Z–1 19 R Musola Pseudolachnostylis maprounefolia 14.8 20–M–1A 20 R Insafwa Syzygium guineense 11.1 20–M–1B 20 R Insafwa Syzygium guineense 14.9 20–M–1C 20 R Insafwa Syzygium guineense 16.1 20–M–2 20 R Insafwa Syzygium guineense 14.4 20–MU–1 20 R Mupundu Parinari curatellifolia 71.6 23–M–10 23 R Insafwa Syzygium guineense 4.0 23–M–11 23 R Insafwa Syzygium guineense 6.7 23–M–12A 23 R Insafwa Syzygium guineense 5.9 23–M–12B 23 R Insafwa Syzygium guineense 6.3 23–M–3 23 R Insafwa Syzygium guineense 7.3 23–M–4 23 R Insafwa Syzygium guineense 5.3 23–M–5 23 R Insafwa Syzygium guineense 9.9 23–M–6 23 R Insafwa Syzygium guineense 6.8 23–M–7 23 R Insafwa Syzygium guineense 13.3 23–M–8 23 R Insafwa Syzygium guineense 4.3 23–M–9 23 R Insafwa Syzygium guineense 4.9 23–Z–1 23 R Musola Pseudolachnostylis maprounefolia 14.0 24–B–1 24 O Icooni Ochna schweinfurthaii 8.4 24–M–1 24 O Insafwa Syzygium guineense 4.6 24–M–2 24 O Insafwa Syzygium guineense 2.3 24–M–3 24 O Insafwa Syzygium guineense 6.1 25–M–1 25 R Mufinsa Syzygium guineense 14.9 25–M–10 25 R Mufinsa Syzygium guineense 21.9 25–M–2 25 R Mufinsa Syzygium guineense 19.1 25–M–3 25 R Mufinsa Syzygium guineense 20.7 25–M–4 25 R Mufinsa Syzygium guineense 32.8 25–M–5 25 R Mufinsa Syzygium guineense 13.1 25–M–6 25 R Mufinsa Syzygium guineense 14.1 25–M–7 25 R Mufinsa Syzygium guineense 12.7 25–M–8 25 R Mufinsa Syzygium guineense 8.2 25–M–9 25 R Mufinsa Syzygium guineense 25.2 26–M–1 26 R Mufungo Anisophyllea boehmii 27.0 26–M–2 26 R Insafwa Syzygium guineense 3.6 26–M–3 26 R Insafwa Syzygium guineense 6.2 26–M–4 26 R Insafwa Syzygium guineense 6.7 28–U–1 28 R Makonko Uapaca banguelensis 9.0 28–U–2A 28 R Makonko Uapaca banguelensis 8.8 28–U–2B 28 R Makonko Uapaca banguelensis 7.2 28–U–3 28 R Makonko Uapaca banguelensis 7.1 28–U–4A 28 R Makonko Uapaca banguelensis 8.9

Table F–1. Continued. Tree ID Transect R/O Local Name Tree Species DBH (cm) 28–U–4B 28 R Makonko Uapaca banguelensis 10.2 28–U–5 28 R Masuku Uapaca kirkiana 7.8 28–U–6 28 R Masuku Uapaca kirkiana 6.2 28–U–7 28 R Masuku Uapaca kirkiana 8.3 28–U–8 28 R Masuku Uapaca kirkiana 7.2 28–U–9 28 R Masuku Uapaca kirkiana 9.7 29–B–1 29 R Icooni Ochna schweinfurthaii 8.9 29–B–2 29 R Icooni Ochna schweinfurthaii 8.6 29–F–1 29 R Unknown Phyllocosmus lemaireanus 5.5 29–M–1 29 R Insafwa Syzygium guineense 5.9 29–M–2 29 R Insafwa Syzygium guineense 3.0 29–N–1 29 R Nsokolobe Uapaca nitida 11.4 29–N–2 29 R Nsokolobe Uapaca nitida 9.9 29–U–1A 29 R Makonko Uapaca banguelensis 8.4 29–U–1B 29 R Makonko Uapaca banguelensis 6.8 29–U–2 29 R Masuku Uapaca kirkiana 11.1 29–U–3 29 R Masuku Uapaca kirkiana 7.6 29–U–4 29 R Makonko Uapaca banguelensis 6.4 29–U–5 29 R Masuku Uapaca kirkiana 10.9 29–U–6A 29 R Makonko Uapaca banguelensis 9.1 29–U–6B 29 R Makonko Uapaca banguelensis 7.7 29–U–7 29 R Makonko Uapaca banguelensis 6.6 30–M–1 30 O Insafwa Syzygium guineense 5.5 30–M–2 30 O Insafwa Syzygium guineense 4.5 30–N–1 30 O Swebya Uapaca sansibarica 13.3 30–U–1 30 O Masuku Uapaca kirkiana 8.2 30–U–10 30 O Masuku Uapaca kirkiana 6.1 30–U–11 30 O Makonko Uapaca banguelensis 10.0 30–U–12 30 O Makonko Uapaca banguelensis 9.2 30–U–13 30 O Masuku Uapaca kirkiana 6.5 30–U–14 30 O Makonko Uapaca banguelensis 5.7 30–U–2 30 O Masuku Uapaca kirkiana 8.0 30–U–3 30 O Masuku Uapaca kirkiana 7.6 30–U–4 30 O Masuku Uapaca kirkiana 17.4 30–U–5 30 O Makonko Uapaca banguelensis 6.4 30–U–6 30 O Makonko Uapaca banguelensis 13.2 30–U–7 30 O Makonko Uapaca banguelensis 10.7 30–U–8 30 O Makonko Uapaca banguelensis 6.0 30–U–9 30 O Makonko Uapaca banguelensis 14.1 31–M–1 31 R Insafwa Syzygium guineense 5.4 31–M–2 31 R Insafwa Syzygium guineense 5.9

Table F–1. Continued. Tree ID Transect R/O Local Name Tree Species DBH (cm) 31–M–3 31 R Insafwa Syzygium guineense 8.6 31–M–4 31 R Insafwa Syzygium guineense 8.2 31–M–5 31 R Insafwa Syzygium guineense 4.5 31–M–6 31 R Insafwa Syzygium guineense 5.9 31–M–7 31 R Insafwa Syzygium guineense 8.1 31–M–8 31 R Insafwa Syzygium guineense 5.0 31–N–1 31 R Swebya Uapaca sansibarica 9.1 31–N–2 31 R Nsokolobe Uapaca nitida 8.9 31–U–1 31 R Makonko Uapaca banguelensis 20.0 31–U–2 31 R Makonko Uapaca banguelensis 6.7 31–U–3 31 R Masuku Uapaca kirkiana 12.1 31–U–4 31 R Masuku Uapaca kirkiana 5.3 31–U–5 31 R Masuku Uapaca kirkiana 5.6 31–U–6 31 R Masuku Uapaca kirkiana 9.3 32–B–1 32 O Icooni Ochna schweinfurthaii 6.1 32–M–1 32 O Insafwa Syzygium guineense 6.5 32–N–1 32 O Nsokolobe Uapaca nitida 5.7 32–U–1 32 O Makonko Uapaca banguelensis 6.4 32–U–2 32 O Makonko Uapaca banguelensis 10.0 33–M–1 33 R Insafwa Syzygium guineense 4.9 33–M–2 33 R Insafwa Syzygium guineense 3.9 33–N–1 33 R Swebya Uapaca sansibarica 33.9 33–N–2A 33 R Swebya Uapaca sansibarica 14.3 33–N–2B 33 R Swebya Uapaca sansibarica 13.5 33–U–1 33 R Makonko Uapaca banguelensis 11.5 33–U–10 33 R Makonko Uapaca banguelensis 9.4 33–U–11 33 R Makonko Uapaca banguelensis 8.3 33–U–12 33 R Makonko Uapaca banguelensis 9.8 33–U–13 33 R Makonko Uapaca banguelensis 12.6 33–U–14A 33 R Makonko Uapaca banguelensis 10.6 33–U–14B 33 R Makonko Uapaca banguelensis 7.7 33–U–15 33 R Makonko Uapaca banguelensis 13.6 33–U–16 33 R Makonko Uapaca banguelensis 9.8 33–U–17 33 R Masuku Uapaca kirkiana 11.1 33–U–18 33 R Masuku Uapaca kirkiana 9.5 33–U–19 33 R Makonko Uapaca banguelensis 7.7 33–U–2 33 R Masuku Uapaca kirkiana 11.0 33–U–20 33 R Makonko Uapaca banguelensis 5.4 33–U–21 33 R Makonko Uapaca banguelensis 9.1 33–U–22 33 R Masuku Uapaca kirkiana 15.4 33–U–3 33 R Masuku Uapaca kirkiana 11.5

Table F–1. Continued. Tree ID Transect R/O Local Name Tree Species DBH (cm) 33–U–4A 33 R Masuku Uapaca kirkiana 10.5 33–U–4B 33 R Masuku Uapaca kirkiana 7.6 33–U–6A 33 R Masuku Uapaca kirkiana 9.9 33–U–6B 33 R Masuku Uapaca kirkiana 8.7 33–U–7 33 R Makonko Uapaca banguelensis 7.2 33–U–8A 33 R Makonko Uapaca banguelensis 10.5 33–U–8B 33 R Makonko Uapaca banguelensis 8.3 33–U–9 33 R Makonko Uapaca banguelensis 9.3 34–M–1 34 R Mufinsa Syzygium guineense 55.5 34–M–2 34 R Mufinsa Syzygium guineense 10.6 34–M–3 34 R Mufinsa Syzygium guineense 36.8 34–N–1 34 R Swebya Uapaca sansibarica 9.6 34–W–1 34 R Kafifi Acacia heteracantha 4.0 35–A–1 35 R Unknown Unknown shrub 35–M–1 35 R Insafwa Syzygium guineense 5.6 35–M–2 35 R Insafwa Syzygium guineense 4.7 35–M–3 35 R Insafwa Syzygium guineense 7.1 35–U–1 35 R Makonko Uapaca banguelensis 5.1 38–MU–1 38 R Mupundu Parinari curatellifolia 48.6 38–MU–2 38 R Mupundu Parinari curatellifolia 145.9 38–MU–4 38 R Mupundu Parinari curatellifolia 49.1 39–MU–1 39 R Mupundu Parinari curatellifolia 50.5 40–B–1 40 O Icooni Ochna schweinfurthaii 9.9 40–J–1 40 O Mufila Unknown 11.4 40–M–1 40 O Insafwa Syzygium guineense 7.7 40–M–2 40 O Insafwa Syzygium guineense 7.4 41–F–1 41 R Unknown Phyllocosmus lemaireanus 5.5 41–H–1 41 R Kafulamume Maprounea africana 7.9 41–MU–1 41 R Mupundu Parinari curatellifolia 67.3 41–MU–2 41 R Mupundu Parinari curatellifolia 40.7 41–Z–1 41 R Musola Pseudolachnostylis maprounefolia 12.2 42–B–1 42 R Icooni Ochna schweinfurthaii 15.8 42–H–1 42 R Kafulamume Maprounea africana 2.5 42–M–1 42 R Insafwa Syzygium guineense 7.8 42–M–10 42 R Insafwa Syzygium guineense 6.0 42–M–11 42 R Insafwa Syzygium guineense 12.1 42–M–12 42 R Insafwa Syzygium guineense 5.5 42–M–13 42 R Insafwa Syzygium guineense 3.8 42–M–14 42 R Insafwa Syzygium guineense 9.7 42–M–15 42 R Insafwa Syzygium guineense 4.4 42–M–16 42 R Insafwa Syzygium guineense 6.0

100

Table F–1. Continued. Tree ID Transect R/O Local Name Tree Species DBH (cm) 42–M–17 42 R Insafwa Syzygium guineense 7.6 42–M–18 42 R Insafwa Syzygium guineense 3.8 42–M–19 42 R Insafwa Syzygium guineense 2.2 42–M–2 42 R Insafwa Syzygium guineense 10.1 42–M–3 42 R Insafwa Syzygium guineense 8.5 42–M–4 42 R Insafwa Syzygium guineense 14.8 42–M–5 42 R Insafwa Syzygium guineense 9.7 42–M–6 42 R Insafwa Syzygium guineense 6.6 42–M–7 42 R Insafwa Syzygium guineense 6.9 42–M–8 42 R Insafwa Syzygium guineense 9.0 42–M–9 42 R Insafwa Syzygium guineense 6.0 42–N–1 42 R Nsokolobe Uapaca nitida 4.7 42–U–1 42 R Makonko Uapaca banguelensis 12.9 42–U–2 42 R Makonko Uapaca banguelensis 16.0 42–U–3 42 R Masuku Uapaca kirkiana 6.8 43–B–1 43 R Icooni Ochna schweinfurthaii 4.1 43–M–1 43 R Insafwa Syzygium guineense 10.9 43–M–10 43 R Insafwa Syzygium guineense 5.6 43–M–11 43 R Insafwa Syzygium guineense 4.9 43–M–12 43 R Mufinsa Syzygium guineense 5.4 43–M–13 43 R Mufinsa Syzygium guineense 6.5 43–M–2 43 R Insafwa Syzygium guineense 6.7 43–M–3 43 R Insafwa Syzygium guineense 3.9 43–M–4 43 R Insafwa Syzygium guineense 6.0 43–M–5 43 R Insafwa Syzygium guineense 5.3 43–M–6 43 R Insafwa Syzygium guineense 5.0 43–M–7 43 R Insafwa Syzygium guineense 12.8 43–M–8 43 R Insafwa Syzygium guineense 9.8 43–M–9 43 R Insafwa Syzygium guineense 10.2 43–MU–1 43 R Mupundu Parinari curatellifolia 51.4 43–N–1 43 R Swebya Uapaca sansibarica 19.6 43–U–1A 43 R Masuku Uapaca kirkiana 9.6 43–U–1B 43 R Masuku Uapaca kirkiana 7.6 44–H–1 44 R Kafulamume Maprounea africana 7.7 44–K–1 44 R Kafulamume Maprounea africana 9.0 44–K–2 44 R Kafulamume Maprounea africana 12.5 44–M–1 44 R Insafwa Syzygium guineense 4.9 44–M–2 44 R Insafwa Syzygium guineense 7.3 44–M–3 44 R Mufinsa Syzygium guineense 13.9 44–M–4 44 R Mufinsa Syzygium guineense 16.7

101

Table F–1. Continued. Tree ID Transect R/O Local Name Tree Species DBH (cm) 44–M–5 44 R Mufinsa Syzygium guineense 6.9 44–M–6 44 R Mufinsa Syzygium guineense 8.2 44–M–7 44 R Mufinsa Syzygium guineense 7.2 44–M–8 44 R Mufinsa Syzygium guineense 2.8 44–M–9 44 R Mufinsa Syzygium guineense 4.8 44–N–1 44 R Nsokolobe Uapaca nitida 9.2 44–N–2 44 R Nsokolobe Uapaca nitida 13.0 44–U–1 44 R Makonko Uapaca banguelensis 18.0 44–U–2 44 R Makonko Uapaca banguelensis 8.1 44–U–3 44 R Makonko Uapaca banguelensis 10.5 45–M–1 45 R Insafwa Syzygium guineense 6.4 45–M–2 45 R Insafwa Syzygium guineense 14.4 45–M–3 45 R Mufinsa Syzygium guineense 29.5 45–M–4 45 R Mufinsa Syzygium guineense 25.6 45–M–5 45 R Mufinsa Syzygium guineense 9.7 45–M–6 45 R Insafwa Syzygium guineense 4.4 45–M–7 45 R Mufinsa Syzygium guineense 26.0 45–M–8 45 R Mufinsa Syzygium guineense 12.8 45–MU–1 45 R Mupundu Parinari curatellifolia 54.7 45–N–1 45 R Swebya Uapaca sansibarica 14.8 45–N–2 45 R Swebya Uapaca sansibarica 13.5 46–M–1 46 O Mufinsa Syzygium cordatum 16.1 46–M–2 46 O Mufinsa Syzygium cordatum 21.7 46–M–3 46 O Mufinsa Syzygium cordatum 13.6

APPENDIX G HABITAT ANALYSIS DATA

Table G–1. Location, size, and species of trees used in habitat analysis. Coordinates are datum Arc1950 in UTM zone 36 L. Height is m, DBH is cm, NR is not recorded. Transect Tree ID East South Tree Type Latin Name Height DBH 1 0214894 8610804 NR NR 15 24.43 1 0214930 8610857 NR NR 13 60.00 1 0214925 8610803 NR NR 9 19.97 1 0214936 8610807 NR NR 13 50.76 1 0214884 8610797 NR NR 14 37.07 1 0214889 8610803 NR NR 10 22.39 1 0214904 8610802 NR NR 10 31.75 1 0214919 8610818 NR NR 9 24.62 1 0214924 8610799 NR NR 8 27.77 1 0214902 8610807 NR NR 8 19.68 1 1–M–3 0214854 8610766 Mufinsa Syzygium cordatum 7 17.20 1 0214890 8610789 NR NR 12 34.39 1 1–M–4 0214853 8610766 Mufinsa Syzygium cordatum 10 30.00 1 0214910 8610795 NR NR 11 31.94 1 1–M–1 0214872 8610797 Mufinsa Syzygium cordatum 9 35.00 1 0214892 8610798 NR NR 7 15.35 1 0214869 8610787 Mufinsa Syzygium cordatum 13 111.46 1 0214839 8610763 Mufinsa Syzygium cordatum 10 23.63 1 1–M–2 0214876 8610795 Mufinsa Syzygium cordatum 12 26.40 1 0214910 8610788 NR NR 11 45.22 3 0212744 8610753 Muputu Brachystegia spiciformis 11 25.22 3 0212727 8610754 Chimpampa Monotes africanus 14 23.73 3 0212780 8610756 Muputu Brachystegia spiciformis 9 14.14 3 0212774 8610758 Mubanga Pericopsis angolensis 10 17.58 3 0212773 8610760 Munye Unknown 8 16.43 3 0212694 8610741 Munye Unknown 19 40.32 3 0212770 8610756 Munye Unknown 13 27.80 3 0212745 8610764 Musamba Brachystegia boehmii 14 25.22 3 0212754 8610754 Mutondo Julbernardia paniculata 15 24.39 3 0212746 8610758 Mutobo Isoberlinia angolensis 14 39.14 3 0212779 8610757 Mubanga Pericopsis angolensis 11 42.10 3 0212757 8610761 Mutobo Isoberlinia angolensis 9 12.26 3 0212721 8610752 Mutobo Isoberlinia angolensis 12 23.47 3 0212776 8610759 Mutobo Isoberlinia angolensis 10 26.11 3 0212739 8610760 Muputu Brachystegia spiciformis 20 29.55 3 0212722 8610752 Mubanga Pericopsis angolensis 9 19.36 3 0212856 8610770 Munye Unknown 7 13.79

102 103

Table G–1. Continued. Transect Tree ID East South Tree Type Latin Name Height DBH 3 0212711 8610750 Saninga Faurea intermedia 10 27.93 3 0212701 8610759 Munye Unknown 12 24.04 3 0212690 8610738 Munye Unknown 10 26.82 3 0212756 8610753 Muputu Brachystegia spiciformis 14 22.17 3 0212738 8610748 Mutondo Julbernardia paniculata 10 34.52 3 0212746 8610753 Mutobo Isoberlinia angolensis 11 13.54 3 0212759 8610744 Mubanga Pericopsis angolensis 9 15.73 3 0212761 8610751 Munye Unknown 10 16.82 3 0212684 8610733 Mutondo Julbernardia paniculata 18 25.13 3 0212766 8610756 Unknown Unknown 9 15.10 3 0212776 8610743 Munye Unknown 13 30.51 3 0212722 8610748 Munye Unknown 10 13.66 3 0212768 8610738 Munye Unknown 15 20.35 3 0212779 8610751 Munye Unknown 11 19.46 3 3–Z–1 0212688 8610746 Musolo Pseudolachnostylis maprounefolia 10 15.00 3 0212790 8610754 Mutobo Isoberlinia angolensis 9 16.27 3 0212776 8610759 Unknown Unknown 8 8.38 3 0212784 8610760 Munye Unknown 13 27.13 3 0212747 8610755 Mutondo Julbernardia paniculata 16 36.78 3 0212760 8610751 Munye Unknown 17 22.42 3 0212699 8610743 Muputu Brachystegia spiciformis 10 26.27 3 0212712 8610757 Mutobo Isoberlinia angolensis 8 16.43 3 0212752 8610759 Mutondo Julbernardia paniculata 14 19.75 5 0210189 8610808 Chimpampa Monotes africanus 12 26.27 5 5–N–2 0210153 8610803 Nsokolobe Uapaca nitida 6 15.22 5 0210140 8610805 Mutobo Isoberlinia angolensis 5 7.55 5 0210159 8610807 Mutondo Julbernardia paniculata 5 7.64 5 0210217 8610823 Musolo Pseudolachnostylis maprounefolia 6 18.63 5 0210171 8610811 Mutobo Isoberlinia angolensis 9 27.55 5 0210213 8610819 Mutobo Isoberlinia angolensis 10 61.15 5 0210151 8610803 Mukwa Pterocarpus angolensis 6 10.80 5 0210208 8610814 Mutobo Isoberlinia angolensis 9 27.20 5 0210160 8610806 Mubanga Pericopsis angolensis 14 30.41 5 0210144 8610807 Mutondo Julbernardia paniculata 5 6.69 5 0210137 8610798 Mutobo Isoberlinia angolensis 7 17.20 5 0210169 8610809 Mutondo Julbernardia paniculata 7 7.83 5 0210164 8610806 Musamba Brachystegia boehmii 7 7.45 5 0210153 8610808 Nsokolobe Uapaca nitida 7 12.55 5 0210157 8610807 Saninga Faurea intermedia 6 13.85 5 0210205 8610815 Nsokolobe Uapaca nitida 7 12.74 5 0210191 8610817 Musuku Uapaca kirkiana 6 12.90 5 0210219 8610823 Mufuka Combretum molle 12 26.56 5 0210136 8610803 Mutobo Isoberlinia angolensis 7 11.78 5 0210163 8610803 Mutondo Julbernardia paniculata 8 7.01 5 0210161 8610794 Mutobo Isoberlinia angolensis 6 19.43

104

Table G–1. Continued. Transect Tree ID East South Tree Type Latin Name Height DBH 5 0210174 8610798 Mutondo Julbernardia paniculata 8 7.80 5 0210166 8610794 Mutondo Julbernardia paniculata 9 7.80 5 0210209 8610806 Mubanga Pericopsis angolensis 8 12.83 5 5–U–4 0210178 8610802 Musuku Uapaca kirkiana 6 18.12 5 0210204 8610807 Mutobo Isoberlinia angolensis 7 10.99 5 0210144 8610799 Mutondo Julbernardia paniculata 7 9.04 5 5–U–2 0210181 8610804 Musuku Uapaca kirkiana 5 13.31 5 0210170 8610799 Mutondo Julbernardia paniculata 5 8.28 5 0210156 8610798 Mutondo Julbernardia paniculata 8 8.76 5 0210207 8610804 Chimpampa Monotes africanus 10 24.01 5 0210148 8610790 Mutobo Isoberlinia angolensis 5 13.54 5 0210139 8610797 Mutobo Isoberlinia angolensis 6 9.49 5 0210148 8610795 Mutondo Julbernardia paniculata 8 12.10 5 0210174 8610794 Mutobo Isoberlinia angolensis 6 11.15 5 0210179 8610803 Muputu Brachystegia spiciformis 9 21.72 5 0210194 8610804 Muputu Brachystegia spiciformis 7 9.71 5 0210193 8610804 Mutobo Isoberlinia angolensis 10 32.39 5 0210162 8610802 Musamba Brachystegia boehmii 7 9.87 7 0209061 8610678 Kayimbi Erythrophleum africanum 10 15.76 7 7–Z–2 0209030 8610694 Musolo Pseudolachnostylis maprounefolia 6 12.80 7 0209073 8610670 Mutobo Isoberlinia angolensis 9 19.90 7 0209018 8610716 Musamba Brachystegia boehmii 14 63.82 7 0209042 8610681 Mutobo Isoberlinia angolensis 8 8.92 7 7–N–8 0209044 8610685 Swebya Uapaca sansibarica 5 9.43 7 0209045 8610681 Mutondo Julbernardia paniculata 10 27.55 7 0209059 8610678 Musolo Pseudolachnostylis maprounefolia 9 23.79 7 0209039 8610096 Nsokolobe Uapaca nitida 8 141.72 7 0209042 8610688 Nsokolobe Uapaca nitida 9 18.85 7 0209061 8610683 Mutondo Julbernardia paniculata 8 17.01 7 7–U–10 0209043 8610690 Musuku Uapaca kirkiana 8 19.01 7 7–Z–1 0209049 8610681 Musolo Pseudolachnostylis maprounefolia 7 18.85 7 0209044 8610683 Mutobo Isoberlinia angolensis 10 12.20 7 0209046 8610681 Mutobo Isoberlinia angolensis 10 17.36 7 0209044 8610691 Nsokolobe Uapaca nitida 8 8.92 7 0209071 8610671 Mubanga Pericopsis angolensis 12 17.20 7 0209024 8610705 Mubanga Pericopsis angolensis 13 31.53 7 0209054 8610683 Mutobo Isoberlinia angolensis 10 11.34 7 0209065 8610675 Mubanga Pericopsis angolensis 13 29.59 7 0209052 8610671 Kayimbi Erythrophleum africanum 14 23.47 7 0209033 8610694 Mutobo Isoberlinia angolensis 7 9.87 7 0209099 8610710 Mubanga Pericopsis angolensis 15 30.25 7 0209010 8610697 Musamba Brachystegia boehmii 12 41.02 7 7–FR–1 0209042 8610670 Mutobo Isoberlinia angolensis 10 23.25 7 0209026 8610683 Mutondo Julbernardia paniculata 9 4.14 7 0209061 8610661 Makonko Uapaca banguelensis 7 17.32

105

Table G–1. Continued. Transect Tree ID East South Tree Type Latin Name Height DBH 7 0209030 8610678 Makonko Uapaca banguelensis 10 21.11 7 0209048 8610668 Mutobo Isoberlinia angolensis 7 11.75 7 7–U–6 0209048 8610667 Musuku Uapaca kirkiana 8 14.97 7 0209038 8610685 Musamba Brachystegia boehmii 8 10.19 7 0209056 8610665 Musamba Brachystegia boehmii 7 10.67 7 0209993 8610713 Musamba Brachystegia boehmii 12 37.58 7 0209025 8610685 Mutondo Julbernardia paniculata 7 10.19 7 0209040 8610687 Musamba Brachystegia boehmii 8 11.31 7 7–U–7 0209039 8610674 Makonko Uapaca banguelensis 6 14.01 7 0209039 8610679 Mutobo Isoberlinia angolensis 9 14.78 7 0209008 8610695 Kayimbi Erythrophleum africanum 16 33.31 7 0209095 8610715 Mubanga Pericopsis angolensis 11 26.91 7 0209071 8610651 Makonko Uapaca banguelensis 8 30.96 8 8–U–2 0207986 8611329 Musuku Uapaca kirkiana 7 10.32 8 0208008 8611333 Mutondo Julbernardia paniculata 6 8.38 8 0207997 8611338 Mutobo Isoberlinia angolensis 12 32.48 8 0208026 8611336 Mutobo Isoberlinia angolensis 12 33.03 8 0208041 8611333 Mutondo Julbernardia paniculata 7 15.92 8 0208018 8611333 Mutondo Julbernardia paniculata 10 10.99 8 0208001 8611334 Munye Unknown 9 16.56 8 0208006 8611330 Mutondo Julbernardia paniculata 7 10.83 8 0208013 8611331 Munye Unknown 13 29.78 8 0207966 8611332 Mutobo Isoberlinia angolensis 10 15.29 8 0207974 8611324 Unknown Unknown 7 11.69 8 0208035 8611336 Unknown Unknown 7 8.06 8 0208038 8611336 Mutondo Julbernardia paniculata 9 15.45 8 0208028 8611330 Mutondo Julbernardia paniculata 12 24.68 8 0207966 8611327 Munye Unknown 8 9.39 8 0208039 8611331 Mutondo Julbernardia paniculata 7 11.46 8 0207991 8611338 Mutondo Julbernardia paniculata 8 10.61 8 0208030 8611331 Munye Unknown 13 25.64 8 0207967 8611326 Mutondo Julbernardia paniculata 9 12.58 8 0207960 8611320 Musamba Brachystegia boehmii 13 42.36 8 0207954 8611320 Mubanga Pericopsis angolensis 12 16.21 8 0207957 8611311 Mutondo Julbernardia paniculata 9 28.47 8 0207994 8611316 Mutondo Julbernardia paniculata 7 11.43 8 0207998 8611323 Mutondo Julbernardia paniculata 8 10.99 8 0208040 8611327 Mutobo Isoberlinia angolensis 9 15.76 8 0208011 8611323 Mutondo Julbernardia paniculata 10 10.19 8 0208037 8611331 Mutobo Isoberlinia angolensis 11 20.70 8 0208001 8611319 Munye Unknown 13 34.46 8 0207965 8611316 Mutondo Julbernardia paniculata 7 12.26 8 0207967 8611305 Mutobo Isoberlinia angolensis 8 21.97 8 0208031 8611320 Kayimbi Erythrophleum africanum 14 29.30 8 0207982 8611320 Mutobo Isoberlinia angolensis 10 18.85

106

Table G–1. Continued. Transect Tree ID East South Tree Type Latin Name Height DBH 8 0207982 8611323 Musamba Brachystegia boehmii 12 21.59 8 0207981 8611316 Munye Unknown 11 16.34 8 0207985 8611315 Mutobo Isoberlinia angolensis 9 22.17 8 0208045 8611321 Mutobo Isoberlinia angolensis 8 15.16 8 0207983 8611317 Mutondo Julbernardia paniculata 11 14.81 8 0207975 8611318 Mutobo Isoberlinia angolensis 8 19.14 8 0208005 8611322 Mutobo Isoberlinia angolensis 8 17.87 8 0207960 8611304 Mutobo Isoberlinia angolensis 6 17.55 9 0206700 8611093 NR NR 8 10.51 9 0206696 8611094 NR NR 9 24.20 9 0206756 8611105 NR NR 7 9.55 9 0206673 8611094 NR NR 10 11.15 9 0206696 8611096 NR NR 10 11.78 9 0206732 8611103 NR NR 8 15.29 9 0206715 8611099 NR NR 10 14.65 9 0206728 8611101 NR NR 10 17.83 9 0206682 8611097 NR NR 11 12.10 9 0206742 8611105 NR NR 10 23.57 9 0206693 8611097 NR NR 11 11.78 9 0206708 8611099 NR NR 8 13.06 9 0206670 8611096 NR NR 10 10.51 9 0206739 8611103 NR NR 11 9.55 9 0206719 8611103 NR NR 12 39.49 9 0206686 8611098 NR NR 11 12.74 9 0206723 8611101 NR NR 10 28.03 9 0206707 8611098 NR NR 8 10.19 9 0206750 8611106 NR NR 8 10.51 9 0206730 8611101 NR NR 7 12.42 9 0206690 8611085 NR NR 11 14.33 9 0206681 8611083 NR NR 9 14.01 9 0206717 8611086 NR NR 12 26.75 9 0206682 8611089 NR NR 10 9.24 9 0206689 8611088 NR NR 9 16.88 9 0206701 8611083 NR NR 11 13.06 9 0206702 8611086 NR NR 10 13.06 9 0206741 8611087 NR NR 15 42.68 9 0206677 8611085 NR NR 9 25.80 9 0206755 8611095 NR NR 10 18.47 9 0206670 8611085 NR NR 9 23.25 9 0206742 8611092 NR NR 11 27.23 9 0206663 8611087 NR NR 9 15.13 9 0206725 8611090 NR NR 9 13.69 9 0206744 8611098 NR NR 8 12.74 9 0206675 8611081 NR NR 10 16.56 9 0206697 8611085 NR NR 11 17.20

107

Table G–1. Continued. Transect Tree ID East South Tree Type Latin Name Height DBH 9 0206736 8611091 NR NR 10 17.52 9 0206705 8611092 NR NR 12 15.61 9 0206713 8611099 NR NR 9 15.61 10 0205559 8610920 NR NR 6 8.25 10 0205555 8610925 NR NR 7 10.35 10 0205571 8610865 NR NR 10 26.78 10 0205554 8610920 NR NR 4 10.13 10 0205564 8610908 NR NR 13 31.66 10 0205555 8610949 NR NR 9 10.57 10 0205571 8610883 NR NR 12 26.15 10 0205550 8610947 NR NR 9 16.78 10 0205552 8610951 NR NR 11 13.38 10 0205566 8610889 NR NR 6 9.62 10 0205560 8610929 NR NR 7 7.90 10 0205557 8610917 NR NR 7 7.52 10 0205567 8610892 NR NR 12 23.98 10 10–B–1 0205582 8610924 Icooni Ochna schweinfurthiana 5 19.84 10 0205559 8610910 NR NR 10 15.38 10 0205553 8610952 NR NR 7 14.33 10 0205555 8610924 NR NR 6 6.37 10 0205566 8610915 NR NR 6 14.30 10 0205549 8610941 NR NR 8 9.24 10 0205561 8610892 NR NR 11 9.27 10 0205542 8610948 NR NR 5 11.31 10 0205541 8610942 NR NR 5 5.25 10 10–B–2 0205555 8610896 Icooni Ochna schweinfurthiana 6 10.29 10 0205561 8610873 NR NR 7 8.92 10 0205533 8610931 NR NR 12 32.80 10 0205554 8610907 NR NR 18 38.95 10 0205540 8610929 NR NR 9 9.01 10 10–M–2 0205534 8610954 Insafwa Syzygium guineense guineense 5 7.90 10 10–M–1 0205531 8610955 Insafwa Syzygium guineense guineense 5 9.11 10 0205541 8610933 NR NR 5 7.71 10 0205557 8610871 NR NR 8 24.87 10 0205540 8610945 NR NR 7 10.19 10 0205553 8610896 NR NR 7 7.64 10 10–B–3 0205555 8610897 Icooni Ochna schweinfurthiana 5 11.18 10 0205543 8610921 NR NR 8 36.78 10 0205552 8610904 NR NR 9 33.60 10 0205554 8610889 NR NR 10 9.71 10 0205558 8610879 NR NR 6 10.92 10 0205558 8610868 NR NR 8 28.98 10 10–B–6 0205555 8610886 Icooni Ochna schweinfurthiana 8 15.29 15 0204145 8607739 Mulimbo Diplorhnchus condylocarpon 8 15.83 15 0204123 8607714 Mutobo Isoberlinia angolensis 5 11.50

108

Table G–1. Continued. Transect Tree ID East South Tree Type Latin Name Height DBH 15 0204157 8607749 Mulimbo Diplorhnchus condylocarpon 6 12.61 15 0204149 8607760 Musamba Brachystegia boehmii 5 9.68 15 0204099 8607714 Mutobo Isoberlinia angolensis 6 16.94 15 0204130 8607729 Mufuka Combretum molle 8 12.99 15 0204149 8607745 Mulimbo Diplorhnchus condylocarpon 7 10.35 15 0204096 8607716 Nsokolobe Uapaca nitida 6 11.05 15 0204153 8607763 Mufuka Combretum molle 7 17.20 15 15–M–4 0204140 8607754 Mufinsa Syzygium guineense huillense 7 17.77 15 0204150 8607757 Mulimbo Diplorhnchus condylocarpon 8 17.20 15 0204163 8607759 Musamba Brachystegia boehmii 5 10.48 15 0204157 8607762 Mulimbo Diplorhnchus condylocarpon 6 10.19 15 0204114 8607716 Mutobo Isoberlinia angolensis 7 16.91 15 0204115 8607714 Mutobo Isoberlinia angolensis 8 21.66 15 0204157 8607763 Musamba Brachystegia boehmii 8 46.59 15 0204151 8607754 Musamba Brachystegia boehmii 6 6.62 15 0204122 8607718 Mutobo Isoberlinia angolensis 7 29.27 15 0204133 8607728 Mufuka Combretum molle 6 11.27 15 0204122 8607721 Mutobo Isoberlinia angolensis 5 7.42 15 0204144 8607732 Mutobo Isoberlinia angolensis 7 13.41 15 0204156 8607748 Musamba Brachystegia boehmii 5 12.58 15 0204120 8607706 Mutobo Isoberlinia angolensis 8 21.08 15 0204138 8607731 Musamba Brachystegia boehmii 4 6.24 15 0204107 8607699 Mutobo Isoberlinia angolensis 5 8.44 15 0204116 8607712 Mutondo Julbernardia paniculata 5 6.15 15 0204127 8607725 Mutobo Isoberlinia angolensis 5 6.97 15 0204144 8607733 Mutobo Isoberlinia angolensis 8 17.58 15 0204137 8607716 Musamba Brachystegia boehmii 6 9.94 15 0204097 8607702 Musamba Brachystegia boehmii 5 9.52 15 0204144 8607727 Mutobo Isoberlinia angolensis 8 41.72 15 0204116 8607700 Mutondo Julbernardia paniculata 7 5.51 15 0204116 8607707 Mutobo Isoberlinia angolensis 4 15.73 15 0204135 8607727 Mutobo Isoberlinia angolensis 6 10.13 15 0204100 8607699 Mutobo Isoberlinia angolensis 5 9.17 15 0204119 8607695 Mutobo Isoberlinia angolensis 5 18.95 15 0204164 8607743 Mutobo Isoberlinia angolensis 7 15.19 15 0204156 8607749 Mutobo Isoberlinia angolensis 5 10.10 15 0204133 8607725 Mutobo Isoberlinia angolensis 5 9.90 15 0204148 8607725 Mufuka Combretum molle 7 14.24 19 0202012 8607493 Musolo Pseudolachnostylis maprounefolia 5 11.82 19 0202028 8607511 Mutobo Isoberlinia angolensis 5 14.68 19 0202074 8607550 Munye Unknown 8 12.77 19 0202016 8607496 Mutobo Isoberlinia angolensis 9 38.41 19 0202069 8607553 Mutobo Isoberlinia angolensis 6 10.89 19 0202036 8607518 Musamba Brachystegia boehmii 4 9.14 19 0202009 8607490 Mufuka Combretum molle 6 11.85

109

Table G–1. Continued. Transect Tree ID East South Tree Type Latin Name Height DBH 19 0202047 8607530 Musamba Brachystegia boehmii 3 7.20 19 0202029 8607514 Mufuka Combretum molle 5 7.17 19 0202041 8607522 Mulimbo Diplorhnchus condylocarpon 6 7.68 19 0202075 8607555 Mutobo Isoberlinia angolensis 5 8.31 19 0202044 8607527 Mutobo Isoberlinia angolensis 4 11.18 19 0202006 8607493 Mufuka Combretum molle 4 9.14 19 0202039 8607525 Musamba Brachystegia boehmii 4 9.90 19 0202038 8607523 Mubanga Pericopsis angolensis 6 13.15 19 0202023 8607498 Mutobo Isoberlinia angolensis 3 5.73 19 0202038 8607520 Munye Unknown 6 8.50 19 0202040 8607520 Mubanga Pericopsis angolensis 7 12.32 19 19–Z–1 0202012 8607499 Musolo Pseudolachnostylis maprounefolia 5 14.81 19 0202044 8607528 Mutobo Isoberlinia angolensis 4 8.41 19 0202004 8607490 Mutobo Isoberlinia angolensis 5 21.78 19 0202010 8607505 Musolo Pseudolachnostylis maprounefolia 4 8.09 19 0202068 8607561 Mutobo Isoberlinia angolensis 6 22.01 19 0202015 8607516 Musuku Uapaca kirkiana 4 6.72 19 0202070 8607561 Muputu Brachystegia spiciformis 7 14.84 19 0202041 8607529 Mutobo Isoberlinia angolensis 5 9.81 19 0202005 8607496 Mutobo Isoberlinia angolensis 4 14.81 19 0202049 8607543 Mutobo Isoberlinia angolensis 6 17.87 19 0202018 8607511 Mutobo Isoberlinia angolensis 4 7.36 19 0202021 8607514 Mulimbo Diplorhnchus condylocarpon 5 7.74 19 0202064 8607560 Mulimbo Diplorhnchus condylocarpon 6 9.59 19 0202018 8607513 Mutobo Isoberlinia angolensis 3 12.90 19 0202011 8607501 Mutobo Isoberlinia angolensis 6 5.19 19 0202035 8607531 Mutobo Isoberlinia angolensis 4 8.95 19 0202008 8607498 Mutobo Isoberlinia angolensis 6 16.05 19 0202021 8607518 Mutobo Isoberlinia angolensis 3 7.01 19 0202037 8607524 Muputu Brachystegia spiciformis 4 6.21 19 0202009 8607498 Musolo Pseudolachnostylis maprounefolia 4 21.46 19 0202012 8607506 Musolo Pseudolachnostylis maprounefolia 5 10.22 19 0202051 8607541 Mutobo Isoberlinia angolensis 8 25.35 20 20–M–1 0202335 8608509 Musafwa Syzygium guineense guineense 5 16.15 20 0202340 8608459 Mufuka Combretum molle 0 7.20 20 0202347 8608495 Musase Albizia antunesiana 4 5.29 20 0202343 8608483 Musafwa Syzygium guineense guineense 5 13.73 20 0202348 8608498 Chibobo Terminalia mollis 11 23.41 20 0202347 8608475 Chibobo Terminalia mollis 12 24.46 20 0202352 8608468 Mufuka Combretum molle 6 11.40 20 0202351 8608451 Mulimbo Diplorhnchus condylocarpon 5 7.45 20 0202347 8608487 Chibobo Terminalia mollis 9 23.79 20 0202340 8608535 Mufuka Combretum molle 5 13.38 20 0202347 8608465 Mufuka Combretum molle 6 10.64 20 0202343 8608479 Musafwa Syzygium guineense guineense 6 11.50

110

Table G–1. Continued. Transect Tree ID East South Tree Type Latin Name Height DBH 20 0202346 8608493 Chibobo Terminalia mollis 10 25.92 20 0202340 8608248 Mufuka Combretum molle 8 13.06 20 0202353 8608453 Mufuka Combretum molle 6 15.00 20 0202348 8608508 Chibobo Terminalia mollis 10 22.77 20 0202352 8608463 Mufuka Combretum molle 6 8.60 20 0202348 8608455 Mufuka Combretum molle 7 13.82 20 0202340 8608484 Chibobo Terminalia mollis 10 27.52 20 0202337 8608535 Chibobo Terminalia mollis 9 33.12 20 0202339 8608400 Mufuka Combretum molle 3 4.04 20 0202344 8608451 Mufuka Combretum molle 3 4.04 20 0202337 8608497 Mulimbo Diplorhnchus condylocarpon 8 7.80 20 20–MU–1 0202333 8608483 Mupundu Parinari curatellifolia 17 71.59 20 0202339 8608486 Chibobo Terminalia mollis 7 15.16 20 0202337 8608456 Mufuka Combretum molle 6 9.68 20 0202340 8608454 Mufuka Combretum molle 5 4.71 20 0202335 8608499 Musambafwa Cassia abbreviata 6 5.51 20 0202335 8608523 Mufuka Combretum molle 5 11.18 20 0202335 8608511 Mufuka Combretum molle 5 8.50 20 0202339 8608484 Mulimbo Diplorhnchus condylocarpon 5 5.92 20 0202327 8608498 Musambafwa Cassia abbreviata 6 5.54 20 0202339 8608452 Mufuka Combretum molle 4 7.39 20 0202339 8608481 Mukwa Pterocarpus angolensis 16 56.05 20 0202339 8608481 Mupundu Parinari curatellifolia 5 7.45 20 0202334 8608473 Mufuka Combretum molle 4 6.88 20 0202343 8608462 Mufuka Combretum molle 5 11.46 20 0202333 8608477 Chibobo Terminalia mollis 14 31.56 20 0202329 8608495 Musambafwa Cassia abbreviata 5 6.82 20 0202329 8608520 Musolo Pseudolachnostylis maprounefolia 10 45.57 21 0201675 8609111 Muputu Brachystegia spiciformis 9 17.93 21 0201658 8609115 Musamba Brachystegia boehmii 9 13.69 21 0201666 8609113 Muputu Brachystegia spiciformis 9 7.96 21 0201620 8609123 Musamba Brachystegia boehmii 9 14.01 21 0201666 8609106 Mutobo Isoberlinia angolensis 9 23.66 21 0201668 8609114 Muputu Brachystegia spiciformis 8 7.83 21 0201643 8609113 Musamba Brachystegia boehmii 11 19.08 21 0201643 8609123 Musamba Brachystegia boehmii 13 19.14 21 0201667 8609113 Musamba Brachystegia boehmii 8 10.83 21 0201609 8609130 Musamba Brachystegia boehmii 8 19.08 21 0201655 8609121 Muputu Brachystegia spiciformis 8 13.79 21 0201668 8609115 Musamba Brachystegia boehmii 9 19.94 21 0201602 8609122 Mutobo Isoberlinia angolensis 8 10.70 21 0201594 8609120 Mutobo Isoberlinia angolensis 10 25.51 21 0201669 8609115 Muputu Brachystegia spiciformis 8 11.59 21 0201625 8609124 Mulembo Unknown 8 14.14 21 0201643 8609115 Musamba Brachystegia boehmii 8 24.78

111

Table G–1. Continued. Transect Tree ID East South Tree Type Latin Name Height DBH 21 0201664 8609100 Muputu Brachystegia spiciformis 10 11.34 21 0201650 8609125 Muputu Brachystegia spiciformis 11 14.33 21 0201628 8609118 Musamba Brachystegia boehmii 9 16.46 21 0201674 8609104 Musamba Brachystegia boehmii 8 10.92 21 0201674 8609105 Musamba Brachystegia boehmii 8 12.48 21 0201614 8609115 Musamba Brachystegia boehmii 10 28.28 21 0201603 8609112 Muputu Brachystegia spiciformis 11 24.36 21 0201677 8609102 Musamba Brachystegia boehmii 7 10.35 21 0201600 8609118 Mutobo Isoberlinia angolensis 8 10.00 21 0201669 8609106 Musamba Brachystegia boehmii 8 15.29 21 0201667 8609103 Muputu Brachystegia spiciformis 10 11.75 21 0201645 8609104 Muputu Brachystegia spiciformis 11 14.20 21 0201603 8609115 Muputu Brachystegia spiciformis 11 12.23 21 0201661 8609102 Mutobo Isoberlinia angolensis 8 13.85 21 0201644 8609106 Musamba Brachystegia boehmii 9 16.69 21 0201683 8609102 Musamba Brachystegia boehmii 7 8.54 21 0201597 8609118 Musamba Brachystegia boehmii 8 11.53 21 0201655 8609108 Muputu Brachystegia spiciformis 10 16.11 21 0201660 8609109 Musamba Brachystegia boehmii 9 14.14 21 0201662 8609104 Musamba Brachystegia boehmii 9 13.06 21 0201663 8609110 Muputu Brachystegia spiciformis 13 14.27 21 0201644 8609108 Unknown Unknown 9 26.69 21 0201617 8609112 Muputu Brachystegia spiciformis 10 13.57 23 0199680 8608870 Mutobo Isoberlinia angolensis 8 14.39 23 0199651 8608883 Mutobo Isoberlinia angolensis 8 15.54 23 0199683 8608867 Mutobo Isoberlinia angolensis 8 21.97 23 0199685 8608871 Mutobo Isoberlinia angolensis 8 13.12 23 0199704 8608859 Mufuka Combretum molle 6 15.03 23 0199697 8608866 Chibobo Terminalia mollis 6 10.76 23 0199658 8608877 Mutobo Isoberlinia angolensis 8 14.39 23 0199674 8608880 Mutobo Isoberlinia angolensis 13 70.06 23 0199637 8608892 Mutobo Isoberlinia angolensis 8 14.65 23 0199714 8608856 Mufuka Combretum molle 6 11.02 23 0199656 8608885 Mufuka Combretum molle 8 11.27 23 0199688 8608870 Mutobo Isoberlinia angolensis 7 12.68 23 0199682 8608875 Mutobo Isoberlinia angolensis 5 16.31 23 0199673 8608878 Sansa Strychnos cocculoides 6 16.31 23 23–M–2 0199698 8608865 Mufuka Combretum molle 7 23.18 23 0199717 8608862 Chibobo Terminalia mollis 6 14.33 23 0199645 8608890 Mutobo Isoberlinia angolensis 8 13.38 23 0199677 8608874 Mutobo Isoberlinia angolensis 7 11.53 23 0199715 8608859 Chibobo Terminalia mollis 7 24.27 23 0199682 8608879 Mutobo Isoberlinia angolensis 8 17.45 23 0199673 8608867 Mutobo Isoberlinia angolensis 7 15.86 23 0199668 8608851 Mutobo Isoberlinia angolensis 6 11.91

112

Table G–1. Continued. Transect Tree ID East South Tree Type Latin Name Height DBH 23 0199645 8608881 Mutobo Isoberlinia angolensis 5 12.68 23 0199669 8608876 Mutobo Isoberlinia angolensis 8 15.48 23 0199675 8608867 Mufuka Combretum molle 8 20.13 23 0199693 8608854 Chibobo Terminalia mollis 6 12.87 23 0199653 8608877 Mutobo Isoberlinia angolensis 7 12.80 23 0199678 8608869 Mutobo Isoberlinia angolensis 8 15.54 23 0199641 8608879 Mutobo Isoberlinia angolensis 6 14.27 23 0199709 8608855 Chibobo Terminalia mollis 6 19.24 23 0199659 8608879 Mutobo Isoberlinia angolensis 5 18.09 23 0199684 8608866 Mutobo Isoberlinia angolensis 6 15.92 23 0199667 8608867 Mutobo Isoberlinia angolensis 8 16.94 23 0199665 8608875 Mutobo Isoberlinia angolensis 7 14.27 23 0199689 8608862 Chibobo Terminalia mollis 7 16.37 23 0199661 8608881 Mutobo Isoberlinia angolensis 7 11.02 23 0199652 8608880 Mutobo Isoberlinia angolensis 8 13.95 23 0199665 8608877 Mutobo Isoberlinia angolensis 7 16.05 23 0199658 8608877 Mutobo Isoberlinia angolensis 6 11.85 23 0199670 8608877 Mutobo Isoberlinia angolensis 7 12.17 26 0197981 8609797 Mutobo Isoberlinia angolensis 10 37.77 26 0198003 8609766 Mutondo Julbernardia paniculata 6 6.94 26 0198006 8609764 Muputu Brachystegia spiciformis 8 11.53 26 0197983 8609791 Mutobo Isoberlinia angolensis 7 8.79 26 0198008 8609761 Musamba Brachystegia boehmii 4 7.58 26 0197990 8609772 Mutobo Isoberlinia angolensis 13 53.69 26 0197967 8609795 Mubanga Pericopsis angolensis 17 38.34 26 0198005 8609736 Musamba Brachystegia boehmii 7 12.48 26 0198013 8609762 Munye Unknown 4 7.39 26 0198000 8609774 Mutobo Isoberlinia angolensis 13 36.88 26 0197956 8609803 Munye Unknown 11 15.99 26 0197945 8609815 Musamba Brachystegia boehmii 8 9.49 26 0198009 8609758 Musolo Pseudolachnostylis maprounefolia 6 13.76 26 0197958 8609800 Muputu Brachystegia spiciformis 6 15.03 26 0197956 8609794 Mutobo Isoberlinia angolensis 7 9.17 26 0197999 8609780 Muputu Brachystegia spiciformis 11 12.80 26 0198015 8609759 Mutobo Isoberlinia angolensis 10 47.83 26 0198001 8609773 Munye Unknown 7 12.48 26 0197949 8609814 Musamba Brachystegia boehmii 5 8.98 26 0197962 8609801 Munye Unknown 5 9.30 26 0197995 8609765 Musamba Brachystegia boehmii 8 10.25 26 0198002 8609760 Mutobo Isoberlinia angolensis 6 9.43 26 0198005 8609752 Munye Unknown 6 10.89 26 0197984 8609774 Mutobo Isoberlinia angolensis 7 13.57 26 0198004 8609755 Mutobo Isoberlinia angolensis 6 13.76 26 26–M–1 0197968 8609788 Mufungo Anisophyllea boemii 5 27.04 26 0197958 8609793 Mutobo Isoberlinia angolensis 12 12.29

113

Table G–1. Continued. Transect Tree ID East South Tree Type Latin Name Height DBH 26 0198007 8609753 Munye Unknown 8 9.30 26 0198005 8609756 Mutobo Isoberlinia angolensis 8 12.48 26 0198003 8609753 Munye Unknown 7 8.66 26 0197977 8609781 Mutondo Julbernardia paniculata 9 9.43 26 0197944 8609796 Mutobo Isoberlinia angolensis 12 40.19 26 0197996 8609765 Mutobo Isoberlinia angolensis 6 12.17 26 0197996 8609763 Mutobo Isoberlinia angolensis 8 14.78 26 0197964 8609791 Mukwa Pterocarpus angolensis 7 13.06 26 0197981 8609781 Musamba Brachystegia boehmii 8 13.44 26 0198002 8609755 Mutobo Isoberlinia angolensis 8 12.29 26 0197999 8609751 Mutobo Isoberlinia angolensis 7 26.56 26 0197964 8609785 Mutobo Isoberlinia angolensis 10 48.47 26 0197958 8609799 Mutobo Isoberlinia angolensis 13 49.68 28 0196514 8611205 Mutondo Julbernardia paniculata 6 10.25 28 0196477 8611205 Mutobo Isoberlinia angolensis 7 15.22 28 0196513 8611201 Mutobo Isoberlinia angolensis 6 18.54 28 0196465 8611208 Mutondo Julbernardia paniculata 7 8.34 28 0196512 8611201 Mufuka Combretum molle 8 22.68 28 28–FR–1 0196504 8611203 Mutondo Julbernardia paniculata 7 8.57 28 0196514 8611204 Musamba Brachystegia boehmii 5 11.21 28 0196486 8611205 Musamba Brachystegia boehmii 7 9.87 28 0196491 8611203 Mutobo Isoberlinia angolensis 6 12.42 28 0196495 8611205 Musase Albizia antunesiana 7 8.73 28 0196524 8611204 Mutobo Isoberlinia angolensis 7 11.21 28 0196499 8611205 Mutondo Julbernardia paniculata 7 9.30 28 0196532 8611204 Chimpampa Monotes africanus 6 13.82 28 0196494 8611206 Mutondo Julbernardia paniculata 8 9.17 28 0196529 8611206 Mutobo Isoberlinia angolensis 7 16.50 28 0196521 8611203 Mutondo Julbernardia paniculata 5 9.94 28 0196469 8611204 Musamba Brachystegia boehmii 7 13.12 28 0196497 8611206 Munye Unknown 7 12.29 28 0196535 8611205 Musuku Uapaca kirkiana 5 8.79 28 0196498 8611205 Mutondo Julbernardia paniculata 7 8.98 28 0196536 8611194 Muputu Brachystegia spiciformis 10 39.04 28 0196466 8611203 Muputu Brachystegia spiciformis 9 20.45 28 0196516 8611198 Mutobo Isoberlinia angolensis 5 13.38 28 0196519 8611189 Musamba Brachystegia boehmii 7 13.63 28 0196527 8611195 Mutondo Julbernardia paniculata 13 9.17 28 0196530 8611199 Mutobo Isoberlinia angolensis 9 38.09 28 0196520 8611193 Mutondo Julbernardia paniculata 7 9.04 28 0196455 8611198 Mutobo Isoberlinia angolensis 11 32.23 28 0196477 8611201 Mutondo Julbernardia paniculata 7 7.90 28 0196495 8611194 Mutondo Julbernardia paniculata 7 9.55 28 0196495 8611197 Musamba Brachystegia boehmii 8 16.18 28 0196489 8611195 Mutondo Julbernardia paniculata 8 9.30

114

Table G–1. Continued. Transect Tree ID East South Tree Type Latin Name Height DBH 28 28–FR–2 0196494 8611197 Muputu Brachystegia spiciformis 8 25.03 28 0196480 8611199 Musamba Brachystegia boehmii 8 12.48 28 0196514 8611198 Musamba Brachystegia boehmii 10 24.14 28 0196511 8611194 Mufuka Combretum molle 7 20.57 28 0196465 8611199 Mutondo Julbernardia paniculata 6 10.57 28 0196482 8611199 Munye Unknown 9 14.78 28 0196506 8611195 Mutondo Julbernardia paniculata 7 12.17 28 0196524 8611197 Musamba Brachystegia boehmii 6 9.43 31 0194531 8611951 Musuku Uapaca kirkiana 6 11.53 31 0194478 8611952 Chimpampa Monotes africanus 7 10.00 31 0194533 8611953 Musuku Uapaca kirkiana 6 10.13 31 0194490 8611959 Makonko Uapaca banguelensis 6 16.43 31 0194544 8611953 Musase Albizia antunesiana 7 14.39 31 0194509 8611952 Mutondo Julbernardia paniculata 4 7.71 31 0194521 8611954 Mutobo Isoberlinia angolensis 9 37.01 31 0194498 8611956 Mutondo Julbernardia paniculata 5 11.53 31 31–N–2 0194504 8611950 Nsokolobe Uapaca nitida 5 8.90 31 0194485 8611957 Mutobo Isoberlinia angolensis 7 15.22 31 0194549 8611958 Mubanga Pericopsis angolensis 9 21.72 31 0194536 8611949 Mutobo Isoberlinia angolensis 6 23.63 31 0194525 8611950 Musamba Brachystegia boehmii 6 22.42 31 31–U–1 0194470 8611950 Makonko Uapaca banguelensis 5 15.03 31 0194528 8611950 Chimpampa Monotes africanus 10 20.13 31 0194544 8611951 Musase Albizia antunesiana 7 10.06 31 0194505 8611950 Unknown Unknown 5 8.66 31 0194464 8611953 Mutondo Julbernardia paniculata 5 35.92 31 0194553 8611951 Mutondo Julbernardia paniculata 5 7.39 31 0194546 8611954 Munye Unknown 6 7.13 31 0194517 8611739 Chimpampa Monotes africanus 8 15.22 31 0194476 8611944 Mutondo Julbernardia paniculata 5 5.99 31 0194517 8611944 Mubanga Pericopsis angolensis 10 30.19 31 0194514 8611941 Mkuwe Magnistipula butayeii 7 6.11 31 0194538 8611943 Musuku Uapaca kirkiana 6 10.70 31 0194480 8611940 Unknown Unknown 7 12.99 31 0194510 8611942 Unknown Unknown 7 10.89 31 0194481 8611949 Muputu Brachystegia spiciformis 6 15.86 31 0194504 8611945 Makonko Uapaca banguelensis 5 13.95 31 0194499 8611942 Makonko Uapaca banguelensis 5 17.77 31 31–U–3 0194547 8611945 Musuku Uapaca kirkiana 7 11.85 31 0194509 8611938 Mutobo Isoberlinia angolensis 9 34.01 31 0194535 8611946 Musamba Brachystegia boehmii 6 6.94 31 0194501 8611946 Mutobo Isoberlinia angolensis 6 21.91 31 0194514 8611948 Kapanga Unknown 14 28.09 31 0194555 8611948 Chimpampa Monotes africanus 7 19.87 31 0194494 8611944 Chimpampa Monotes africanus 5 12.99

115

Table G–1. Continued. Transect Tree ID East South Tree Type Latin Name Height DBH 31 0194473 8611947 Chimpampa Monotes africanus 5 12.04 31 0194547 8611942 Mutobo Isoberlinia angolensis 7 15.54 31 0194533 8611945 Musuku Uapaca kirkiana 5 11.72 33 33–U–12 0192500 8612141 Makonko Uapaca banguelensis 6 9.80 33 0192461 8612128 Mubanga Pericopsis angolensis 11 24.62 33 0192466 8612127 Chimpampa Monotes africanus 6 16.78 33 0192473 8612134 Kayimbi Erythrophleum africanum 12 23.34 33 0192511 8612145 Mutondo Julbernardia paniculata 13 26.69 33 0192496 8612141 Mutobo Isoberlinia angolensis 8 12.74 33 33–U–13 0192496 8612139 Makonko Uapaca banguelensis 5 12.60 33 33–U–15 0192503 8612140 Makonko Uapaca banguelensis 5 13.60 33 33–U–6 0192467 8612135 Masuku Uapaca kirkiana 7 9.80 33 0192495 8612139 Musuku Uapaca kirkiana 6 15.76 33 0192453 8612127 Musamba Brachystegia boehmii 10 36.53 33 33–U–10 0192495 8612137 Makonko Uapaca banguelensis 6 9.40 33 33–N–1 0192490 8612134 Swebya Uapaca sansibarica 10 33.90 33 0192456 8612121 Mutondo Julbernardia paniculata 13 37.90 33 0192481 8612135 Mutondo Julbernardia paniculata 15 45.00 33 33–U–3 0192438 8612119 Musuku Uapaca kirkiana 7 0.00 33 0192455 8612128 Mabelemabele Ozoroa reticulata 6 14.81 33 0192476 8612133 Mutondo Julbernardia paniculata 13 31.85 33 0192509 8612154 Musuku Uapaca kirkiana 5 16.24 33 0192483 8612136 Mutondo Julbernardia paniculata 9 27.61 33 0192475 8612125 Musamba Brachystegia boehmii 5 9.52 33 0192453 8612110 Chimpampa Monotes africanus 8 21.66 33 33–U–22 0192444 8612125 Masuku Uapaca kirkiana 8 15.40 33 0192471 8612117 Musuku Uapaca kirkiana 7 13.85 33 0192511 8612138 Musolo Pseudolachnostylis maprounefolia 5 13.06 33 0192467 8612119 Musuku Uapaca kirkiana 7 13.98 33 33–N–2 0192497 8612135 Swebya Uapaca sansibarica 7 14.30 33 0192510 8612135 Mutobo Isoberlinia angolensis 6 10.51 33 0192438 8612110 Chimpampa Monotes africanus 8 14.01 33 0192497 8612130 Mutondo Julbernardia paniculata 12 20.86 33 0192434 8612101 Mupundu Parinari curatellifolia 14 55.89 33 0192486 8612131 Kayimbi Erythrophleum africanum 9 22.29 33 0192500 8612133 Musuku Uapaca kirkiana 5 13.06 33 0192436 8612102 Kayimbi Erythrophleum africanum 8 18.47 33 0192434 8612106 Mutobo Isoberlinia angolensis 8 14.01 33 0192502 8612135 Kafulamume Maprounea africana 7 9.65 33 0192435 8612110 Unknown Unknown 9 11.02 33 33–U–16 0192479 8612133 Makonko Uapaca banguelensis 6 9.80 33 0192494 8612132 Mutobo Isoberlinia angolensis 9 27.55 33 33–U–17 0192459 8612123 Musuku Uapaca kirkiana 6 11.10 35 0188579 8616422 Musamba Brachystegia boehmii 6 9.94

116

Table G–1. Continued. Transect Tree ID East South Tree Type Latin Name Height DBH 35 0188589 8616429 Mutobo Isoberlinia angolensis 6 14.04 35 0188593 8616442 Mutobo Isoberlinia angolensis 6 13.22 35 0188581 8616407 Musamba Brachystegia boehmii 7 22.10 35 0188589 8616421 Chimpampa Monotes africanus 6 9.11 35 0188569 8616394 Musamba Brachystegia boehmii 5 10.48 35 0188582 8616424 Mutondo Julbernardia paniculata 6 8.38 35 0188597 8616462 Mutobo Isoberlinia angolensis 7 31.37 35 0188601 8616461 Mufungo Anisophyllea boemii 5 11.37 35 0188577 8616410 Chimpampa Monotes africanus 5 8.89 35 0188604 8616468 Mutobo Isoberlinia angolensis 6 7.20 35 0188580 8616411 Chimpampa Monotes africanus 5 7.32 35 0188577 8616401 Mutobo Isoberlinia angolensis 8 14.24 35 0188573 8616399 Mutobo Isoberlinia angolensis 8 22.36 35 0188572 8616387 Mutobo Isoberlinia angolensis 12 35.54 35 0188603 8616472 Mutobo Isoberlinia angolensis 11 34.84 35 0188597 8616456 Mutobo Isoberlinia angolensis 11 38.89 35 0188589 8616429 Musamba Brachystegia boehmii 7 34.97 35 0188587 8616435 Mutobo Isoberlinia angolensis 9 42.90 35 0188585 8616419 Mutobo Isoberlinia angolensis 7 8.95 35 0188568 8616408 Chimpampa Monotes africanus 5 8.92 35 0188571 8616408 Chimpampa Monotes africanus 5 8.34 35 0188580 8616428 Musase Albizia antunesiana 7 10.51 35 0188570 8616418 Musamba Brachystegia boehmii 9 19.75 35 0188573 8616423 Mutobo Isoberlinia angolensis 7 21.88 35 0188569 8616404 Mutobo Isoberlinia angolensis 8 27.99 35 0188571 8616414 Chimpampa Monotes africanus 6 8.73 35 0188582 8616460 Mutondo Julbernardia paniculata 11 38.22 35 0188574 8616431 Mufungo Anisophyllea boemii 6 12.83 35 0188585 8616447 Mkuwe Magnistipula butayeii 5 8.63 35 0188559 8616409 Unknown Unknown 7 8.25 35 0188583 8616446 Kasansubwangu Dalbergia nitidula 7 13.06 35 0188588 8616452 Mulimbo Diplorhnchus condylocarpon 8 9.55 35 0188570 8616409 Mukwa Pterocarpus angolensis 7 14.11 35 0188563 8616401 Mutobo Isoberlinia angolensis 8 27.26 35 0188579 8616443 Mulimbo Diplorhnchus condylocarpon 7 7.29 35 0188581 8616436 Musase Albizia antunesiana 6 9.87 35 0188569 8616411 Chimpampa Monotes africanus 5 8.76 35 0188579 8616427 Mulimbo Diplorhnchus condylocarpon 7 8.50 35 0188595 8616472 Mutobo Isoberlinia angolensis 6 13.25 36 0188837 8615402 Mutobo Isoberlinia angolensis 5 8.69 36 0188802 8615426 Unknown Unknown 7 12.87 36 0188806 8615431 Mukwa Pterocarpus angolensis 7 8.12 36 0188839 8615395 Musamba Brachystegia boehmii 5 12.29 36 0188826 8615405 Mutobo Isoberlinia angolensis 10 25.67 36 0188834 8615412 Musase Albizia antunesiana 10 11.75

117

Table G–1. Continued. Transect Tree ID East South Tree Type Latin Name Height DBH 36 0188843 8615406 Mutobo Isoberlinia angolensis 9 22.48 36 0188819 8615424 Mutobo Isoberlinia angolensis 9 24.59 36 0188801 8615450 Mukwa Pterocarpus angolensis 7 7.87 36 0188791 8615453 Mulimbo Diplorhnchus condylocarpon 6 11.37 36 0188814 8615435 Musamba Brachystegia boehmii 4 10.32 36 0188799 8615440 Mulimbo Diplorhnchus condylocarpon 7 10.86 36 0188839 8615385 Mutobo Isoberlinia angolensis 9 29.20 36 0188826 8615421 Mutobo Isoberlinia angolensis 13 51.66 36 0188820 8615414 Mukwa Pterocarpus angolensis 5 12.01 36 0188830 8615403 Mukwa Pterocarpus angolensis 4 11.24 36 0188823 8615408 Mutobo Isoberlinia angolensis 9 27.90 36 0188818 8615426 Musamba Brachystegia boehmii 5 9.94 36 0188839 8615402 Mutobo Isoberlinia angolensis 5 7.87 36 0188807 8615430 Mupundu Parinari curatellifolia 7 9.24 36 0188828 8615394 Mutobo Isoberlinia angolensis 10 26.08 36 0188835 8615389 Mutobo Isoberlinia angolensis 8 24.08 36 0188778 8615453 Mufuka Combretum molle 9 45.76 36 0188803 8615416 Musamba Brachystegia boehmii 5 9.55 36 0188828 8615400 Mutobo Isoberlinia angolensis 8 23.89 36 0188819 8615403 Unknown Unknown 4 6.37 36 0188834 8615391 Mulimbo Diplorhnchus condylocarpon 5 5.70 36 0188811 8615411 Mulimbo Diplorhnchus condylocarpon 5 8.28 36 0188839 8615386 Mulunguti Erythrina abyssinica 9 10.51 36 0188808 8615414 Mukwa Pterocarpus angolensis 9 9.87 36 0188811 8615419 Chiya Lonchocarpus capassa 7 15.86 36 0188804 8615417 Mukwa Pterocarpus angolensis 6 9.27 36 0188841 8615300 Mulimbo Diplorhnchus condylocarpon 7 8.09 36 0188805 8615414 Mukwa Pterocarpus angolensis 9 12.61 36 0188805 8615418 Musamba Brachystegia boehmii 5 6.69 36 0188832 8615390 Mulimbo Diplorhnchus condylocarpon 4 7.80 36 0188804 8615425 Mulimbo Diplorhnchus condylocarpon 5 12.71 36 0188813 8615410 Mutobo Isoberlinia angolensis 6 38.54 36 0188810 8615408 Mutobo Isoberlinia angolensis 11 34.39 36 0188814 8615415 Musamba Brachystegia boehmii 4 6.69 37 0189020 8614478 Mulimbo Diplorhnchus condylocarpon 6 7.04 37 0189022 8614477 Mufuka Combretum molle 6 11.82 37 0189017 8614459 Mufuka Combretum molle 7 13.41 37 0189027 8614489 Unknown Unknown 8 20.70 37 0189017 8614485 Mufuka Combretum molle 6 15.70 37 0189022 8614471 Mufuka Combretum molle 6 8.44 37 0189027 8614485 Mulimbo Diplorhnchus condylocarpon 7 8.79 37 0189021 8614468 Mufuka Combretum molle 5 10.64 37 0189044 8614511 Mufuka Combretum molle 12 22.93 37 0189004 8614337 Mupundu Parinari curatellifolia 12 38.69 37 0189023 8614473 Mufuka Combretum molle 7 19.59

118

Table G–1. Continued. Transect Tree ID East South Tree Type Latin Name Height DBH 37 0189025 8614475 Mubanga Pericopsis angolensis 13 32.93 37 0189020 8614493 Mufuka Combretum molle 10 33.95 37 0189029 8614479 Mufuka Combretum molle 11 25.57 37 0189024 8614477 Mufuka Combretum molle 5 8.79 37 0189037 8614500 Mufuka Combretum molle 8 17.36 37 0189022 8614476 Mulimbo Diplorhnchus condylocarpon 5 9.90 37 0189990 8614436 Musolo Pseudolachnostylis maprounefolia 10 10.86 37 0189016 8614439 Mupundu Parinari curatellifolia 10 33.73 37 0189019 8614464 Mufuka Combretum molle 6 12.45 37 0189018 8614492 Musase Albizia antunesiana 4 7.17 37 0189995 8614458 Unknown Unknown 8 28.15 37 0189031 8614520 Chibobo Terminalia mollis 10 14.01 37 0189035 8614543 Mukome Strychnos pungens 8 14.01 37 0189021 8614497 Mufuka Combretum molle 6 9.90 37 0189018 8614482 Unknown Unknown 3 5.19 37 0189030 8614486 Unknown Unknown 7 11.56 37 0189043 8614524 Umwende Unknown 7 8.47 37 0189039 8614529 Mukome Strychnos pungens 7 16.75 37 0189022 8614497 Mulimbo Diplorhnchus condylocarpon 6 6.82 37 0189017 8614487 Mulimbo Diplorhnchus condylocarpon 8 9.90 37 0189024 8614487 Mufuka Combretum molle 7 34.78 37 0189045 8614504 Mufuka Combretum molle 10 18.31 37 0189025 8614491 Unknown Unknown 4 6.78 37 0189024 8614497 Unknown Unknown 8 14.30 37 0189041 8614496 Mufuka Combretum molle 10 36.15 37 0189037 8614520 Chibobo Terminalia mollis 9 25.16 37 0189980 8614450 Unknown Unknown 14 46.53 37 0189992 8614453 Unknown Unknown 5 6.15 37 0189036 8614531 Chibobo Terminalia mollis 8 16.53 38 0188780 8613535 Mulimbo Diplorhnchus condylocarpon 7 11.18 38 0188828 8613528 Mupundu Parinari curatellifolia 17 41.18 38 0188741 8613551 Mubanga Pericopsis angolensis 11 25.51 38 0188782 8613537 Kayimbi Erythrophleum africanum 14 30.89 38 0188750 8613548 Kayimbi Erythrophleum africanum 17 34.84 38 0188776 8613537 Chibobo Terminalia mollis 8 28.25 38 0188745 8613553 Mufuka Combretum molle 10 34.46 38 0188812 8613520 Mulimbo Diplorhnchus condylocarpon 5 11.27 38 0188745 8613556 Mubanga Pericopsis angolensis 19 31.85 38 0188776 8613535 Musolo Pseudolachnostylis maprounefolia 9 27.90 38 0188820 8613524 Mulimbo Diplorhnchus condylocarpon 6 12.29 38 0188775 8613537 Unknown Unknown 7 12.04 38 0188747 8613554 Mulimbo Diplorhnchus condylocarpon 6 8.76 38 0188784 8613535 Mulimbo Diplorhnchus condylocarpon 7 10.06 38 0188784 8613537 Unknown Unknown 6 8.44 38 0188746 8613558 Mubanga Pericopsis angolensis 17 55.41

119

Table G–1. Continued. Transect Tree ID East South Tree Type Latin Name Height DBH 38 38–MU–1 0188817 8613524 Mupundu Parinari curatellifolia 14 48.57 38 0188808 8613525 Mulimbo Diplorhnchus condylocarpon 8 13.06 38 0188745 8613553 Musolo Pseudolachnostylis maprounefolia 7 15.13 38 0188776 8615541 Mubanga Pericopsis angolensis 14 29.46 38 0188766 8613538 Mulimbo Diplorhnchus condylocarpon 6 5.41 38 0188808 8613516 Mupapa Protea spp. 19 103.44 38 0188751 8613531 Musolo Pseudolachnostylis maprounefolia 8 16.40 38 0188761 8613533 Musolo Pseudolachnostylis maprounefolia 11 22.64 38 0188750 8613542 Mupundu Parinari curatellifolia 15 35.64 38 0188768 8613526 Mulimbo Diplorhnchus condylocarpon 8 13.57 38 38–MU–4 0188738 8613546 Mupundu Parinari curatellifolia 15 49.08 38 38–MU–2 0188792 8613542 Mupundu Parinari curatellifolia 19 145.92 38 0188747 8613541 Unknown Unknown 6 8.09 38 0188766 8613526 Mulimbo Diplorhnchus condylocarpon 8 13.15 38 0188794 8313526 Musolo Pseudolachnostylis maprounefolia 7 22.52 38 0188766 8613540 Mulimbo Diplorhnchus condylocarpon 8 15.16 38 0188764 8613545 Mulimbo Diplorhnchus condylocarpon 8 12.36 38 38–MU–3 0188756 8613532 Mukwa Pterocarpus angolensis 18 61.75 38 0188772 8613526 Mulimbo Diplorhnchus condylocarpon 9 11.31 38 0188752 8613537 Kayimbi Erythrophleum africanum 18 35.48 38 0188741 8613550 Mufuka Combretum molle 15 18.06 38 0188734 8613539 Unknown Unknown 14 33.38 38 0188745 8613535 Mupundu Parinari curatellifolia 15 32.55 38 0188765 8613529 Mulimbo Diplorhnchus condylocarpon 5 8.18 41 0190094 8611856 Mukwa Pterocarpus angolensis 5 6.72 41 0190055 8611902 Mutobo Isoberlinia angolensis 13 43.18 41 0190066 8611875 Chibangalume Zahna africana 16 29.33 41 0190084 8611855 Muputu Brachystegia spiciformis 5 6.75 41 0190065 8611870 Mutobo Isoberlinia angolensis 11 37.96 41 0190053 8611890 Mutobo Isoberlinia angolensis 10 23.73 41 0190051 8611902 Muputu Brachystegia spiciformis 9 11.78 41 0190046 8611903 Mukwa Pterocarpus angolensis 8 9.87 41 0190080 8611868 Mutobo Isoberlinia angolensis 14 41.15 41 0190056 8611885 Mutobo Isoberlinia angolensis 10 33.82 41 0190052 8611897 Muputu Brachystegia spiciformis 9 14.78 41 0190047 8611904 Muntufita Diospyros batocana 6 32.04 41 0190070 8611869 Unknown Unknown 10 21.21 41 0190049 8611910 Kasansubwangu Dalbergia nitidula 7 8.92 41 0190032 8611930 Kayimbi Erythrophleum africanum 12 34.62 41 0190092 8611852 Mubanga Pericopsis angolensis 17 27.80 41 0190039 8611904 Mutobo Isoberlinia angolensis 12 40.45 41 0190079 8611865 Icooni Ochna schweinfurthiana 7 7.83 41 41–MU–1 0190041 8611914 Mupundu Parinari curatellifolia 17 67.32 41 0190051 8611903 Mukwa Pterocarpus angolensis 7 7.36 41 0190077 8611888 Mubanga Pericopsis angolensis 13 49.68

120

Table G–1. Continued. Transect Tree ID East South Tree Type Latin Name Height DBH 41 0190062 8611901 Mutobo Isoberlinia angolensis 6 7.42 41 0190097 8611867 Mutobo Isoberlinia angolensis 19 74.84 41 0190079 8611890 Mutobo Isoberlinia angolensis 12 44.20 41 0190091 8611875 Mulimbo Diplorhnchus condylocarpon 6 7.26 41 0190041 8611918 Mutobo Isoberlinia angolensis 9 21.37 41 0190047 8611914 Chimpampa Monotes africanus 5 4.71 41 0190061 8611906 Mulimbo Diplorhnchus condylocarpon 6 3.31 41 0190096 8611867 Mubanga Pericopsis angolensis 13 29.27 41 0190059 8611916 Kasansubwangu Dalbergia nitidula 7 12.61 41 0190068 8611909 Muputu Brachystegia spiciformis 5 7.04 41 0190046 8611911 Icooni Ochna schweinfurthiana 7 11.91 41 0190094 8611867 Mutobo Isoberlinia angolensis 15 37.52 41 0190061 8611908 Mulimbo Diplorhnchus condylocarpon 5 4.78 41 0190056 8611911 Muputu Brachystegia spiciformis 7 10.25 41 41–MU–2 0190068 8611889 Mupundu Parinari curatellifolia 15 40.70 41 0190071 8611897 Mulimbo Diplorhnchus condylocarpon 5 5.10 41 0190102 8611867 Muputu Brachystegia spiciformis 13 39.71 41 0190049 8611911 Icooni Ochna schweinfurthiana 4 5.25 41 0190042 8611916 Mulimbo Diplorhnchus condylocarpon 4 5.99 42 0191973 8611322 Unknown Unknown 7 10.54 42 0191015 8611298 Mutondo Julbernardia paniculata 14 31.69 42 42–M–2 0191998 8611297 Insafwa Syzygium guineense guineense 5 10.06 42 0181970 8611330 Mutondo Julbernardia paniculata 11 24.71 42 0191964 8611329 Mutondo Julbernardia paniculata 15 30.67 42 0191024 8611285 Mutondo Julbernardia paniculata 12 30.41 42 0191021 8611293 Mutobo Isoberlinia angolensis 10 25.29 42 0191004 8611308 Mutobo Isoberlinia angolensis 7 8.82 42 0191973 8611328 Musamba Brachystegia boehmii 5 10.57 42 0191999 8611306 Musuku Uapaca kirkiana 5 16.08 42 0191959 8611337 Mutondo Julbernardia paniculata 7 7.48 42 0191014 8611299 Mutondo Julbernardia paniculata 12 26.31 42 0191997 8611303 Mutondo Julbernardia paniculata 11 28.50 42 42–U–1 0191027 8600290 Makonko Uapaca banguelensis 6 12.87 42 0191004 8611297 Musuku Uapaca kirkiana 4 14.94 42 0191956 8611338 Mutondo Julbernardia paniculata 8 21.50 42 42–M–17 0191966 8611329 Insafwa Syzygium guineense guineense 5 7.58 42 42–M–5 0191978 8611321 Insafwa Syzygium guineense guineense 5 9.71 42 42–B–1 0191963 8611333 Icooni Ochna schweinfurthiana 6 15.76 42 0191002 8611302 Musuku Uapaca kirkiana 8 17.36 42 019111 8611292 Mutondo Julbernardia paniculata 10 26.94 42 0191959 8611326 Mutobo Isoberlinia angolensis 11 40.06 42 0191953 8611328 Musuku Uapaca kirkiana 5 16.53 42 0191972 8611319 Mutondo Julbernardia paniculata 12 24.81 42 42–M–4 0191972 8611320 Insafwa Syzygium guineense guineense 6 14.84 42 42–U–2 0191988 8611313 Musuku Uapaca kirkiana 4 16.02

121

Table G–1. Continued. Transect Tree ID East South Tree Type Latin Name Height DBH 42 0191985 8611309 Unknown Unknown 4 7.17 42 0191002 8611297 Mutondo Julbernardia paniculata 13 31.62 42 0191981 8611312 Kayimbi Erythrophleum africanum 10 21.78 42 0191028 8611286 Musuku Uapaca kirkiana 6 25.48 42 0191979 8611305 Mutondo Julbernardia paniculata 10 22.07 42 0191007 8611296 Mutondo Julbernardia paniculata 13 26.43 42 0191002 8611297 Unknown Unknown 6 7.90 42 0191953 8611333 Mutondo Julbernardia paniculata 13 21.15 42 0191026 8611282 Kayimbi Erythrophleum africanum 13 21.40 42 0191959 8611319 Mutondo Julbernardia paniculata 10 21.53 42 0191982 8611306 Musamba Brachystegia boehmii 5 19.27 42 0191015 8611295 Musolo Pseudolachnostylis maprounefolia 7 17.58 42 0191975 8611311 Insafwa Syzygium guineense guineense 4 9.68 42 0191017 8611284 Musamba Brachystegia boehmii 9 27.83 43 43–B–1 0191392 8610593 Icooni Ochna schweinfurthiana 2 4.08 43 43–U–1 0191393 8610564 Musuku Uapaca kirkiana 5 9.55 43 0191397 8610533 Unknown Unknown 2 3.76 43 43–M–8 0191393 8610631 Insafwa Syzygium guineense guineense 3 9.84 43 0191400 8610622 Insafwa Syzygium guineense guineense 4 7.64 43 0191393 8610626 Nsokolobe Uapaca nitida 3 3.50 43 0191398 8610604 Insafwa Syzygium guineense guineense 3 5.25 43 0191397 8610539 Musuku Uapaca kirkiana 4 10.99 43 0191394 8610611 Unknown Unknown 7 17.17 43 0191395 8610546 Swebya Uapaca sansibarica 4 8.06 43 0191383 8610566 Musuku Uapaca kirkiana 5 12.29 43 43–M–12 0191383 8610602 Mufinsa Syzygium guineense huillense 3 5.35 43 43–MU–1 0191388 8610548 Mupundu Parinari curatellifolia 15 51.40 43 43–N–1 0191388 8610547 Swebya Uapaca sansibarica 7 19.59 43 0191381 8610603 Mupundu Parinari curatellifolia 6 11.94 43 0191388 8610544 Swebya Uapaca sansibarica 7 21.27 43 43–M–9 0191388 8610568 Insafwa Syzygium guineense guineense 4 10.22 43 0191385 8610572 Kafulamume Maprounea africana 4 8.12 43 0191388 8610603 Nsokolobe Uapaca nitida 6 15.86 43 0191389 8610564 Mupundu Parinari curatellifolia 6 8.79 43 43–M–5 0191386 8610630 Insafwa Syzygium guineense guineense 3 5.32 43 0191386 8610530 Unknown Unknown 3 5.73 43 0191384 8610566 Musuku Uapaca kirkiana 6 13.54 43 43–M–7 0191381 8610632 Insafwa Syzygium guineense guineense 3 12.77 43 43–M–2 0191387 8610597 Insafwa Syzygium guineense guineense 3 6.66 43 0191385 8610548 Mubanga Pericopsis angolensis 5 13.85 43 0191378 8610622 Nsokolobe Uapaca nitida 6 14.33 43 0191384 8610543 Mubanga Pericopsis angolensis 6 10.83 43 0191385 8610550 Mubanga Pericopsis angolensis 10 25.16 43 43–M–10 0191383 8610591 Insafwa Syzygium guineense guineense 2 5.64 45 0193218 8610458 Musamba Brachystegia boehmii 7 15.29

122

Table G–1. Continued. Transect Tree ID East South Tree Type Latin Name Height DBH 45 0193214 8610452 Musamba Brachystegia boehmii 5 5.73 45 0193214 8610458 Musamba Brachystegia boehmii 6 14.11 45 0193193 8610456 Mupundu Parinari curatellifolia 5 8.44 45 0193212 8610456 Musamba Brachystegia boehmii 6 17.48 45 0193222 8610458 Musamba Brachystegia boehmii 7 15.92 45 0193173 8610457 Unknown Unknown 3 5.32 45 0193174 8610459 Musamba Brachystegia boehmii 6 11.72 45 0193227 8610455 Musamba Brachystegia boehmii 6 14.81 45 45–M–4 0193226 8610450 Mufinsa Syzygium guineense huillense 5 25.57 45 0193209 8610455 Musamba Brachystegia boehmii 6 15.92 45 0193186 8610458 Mupundu Parinari curatellifolia 5 7.04 45 0193193 8610458 Mupundu Parinari curatellifolia 4 6.05 45 0193216 8610453 Musamba Brachystegia boehmii 3 6.46 45 0193222 8610457 Musamba Brachystegia boehmii 6 11.31 45 0193208 8610459 Musamba Brachystegia boehmii 6 19.43 45 0193216 8610454 Musamba Brachystegia boehmii 6 6.62 45 0193225 8610451 Musamba Brachystegia boehmii 6 16.21 45 0193196 8610455 Mupundu Parinari curatellifolia 4 8.38 45 45–M–3 0193184 8610458 Mufinsa Syzygium guineense huillense 6 29.46 45 45–M–7 0193200 8610450 Mufinsa Syzygium guineense huillense 6 25.99 45 0193192 8610448 Musamba Brachystegia boehmii 7 14.17 45 0193179 8610453 Nsokolobe Uapaca nitida 7 11.62 45 0193241 8610442 Musamba Brachystegia boehmii 6 11.15 45 0193170 8610450 Mutondo Julbernardia paniculata 3 4.33 45 0193203 8610443 Musamba Brachystegia boehmii 3 4.46 45 0193254 8610445 Nsokolobe Uapaca nitida 5 9.55 45 0193254 8610444 Musamba Brachystegia boehmii 6 13.69 45 0193188 8610445 Musamba Brachystegia boehmii 6 11.78 45 45–M–5 0193230 8610447 Mufinsa Syzygium guineense huillense 4 9.75 45 45–N–1 0193188 8610445 Swebya Uapaca sansibarica 6 14.81 45 0193224 8610440 Mufinsa Syzygium guineense huillense 5 22.13 45 0193173 8610449 Mupundu Parinari curatellifolia 4 4.14 45 45–M–8 0193186 8610458 Mufinsa Syzygium guineense huillense 5 12.83 45 45–M–2 0193197 8610454 Insafwa Syzygium guineense guineense 5 14.39 45 45–MU–1 0193252 8610440 Mupundu Parinari curatellifolia 10 54.68 45 0193186 8610449 Musamba Brachystegia boehmii 8 12.01 45 0193180 8610452 Musamba Brachystegia boehmii 7 21.05 45 0193262 8610446 Nsokolobe Uapaca nitida 6 9.81 45 0193263 8610443 Musamba Brachystegia boehmii 4 7.17

APPENDIX H MAP OF VEGETATION TRANSECT AND MIST NETTING LOCATIONS

N Luwombwa River

M ule m bo

Bwalyabemba Luwombwa Hill Mulembo River Musande 124 Chikufwe Wasa Pontoon Mulaushi r Kafubashi ive Wasa II R r Mpululwe a e w v M Fibwe b i u om R s Hill w a o u k l L n a R a i s ve a r r K e iv Shiwila i R sh 010 u Kilometers la u M Figure H–1. Map showing locations of vegetation transects and mist netting sites. Open markers indicate vegetation transects mist netted in, diamonds represent vegetation transects not netted in, and the star indicates the central Eidolon helvum roost.

APPENDIX I PHOTOGRAPHS OF EPAULETED AND DWARF EPAULETED FRUIT BATS NETTED IN THIS STUDY

Figure I–1. Photo of Epomophorus gambianus crypturus showing facial detail and epaulets. (Photo by Charlie Bear 2003.)

125 126

Figure I–2. Photo of Epomophorus gambianus crypturus. (Photo by Charlie Bear 2003.)

127

Figure I–3. Full body photo of Epomophorus gambianus crypturus. (Photo by Charlie Bear 2003.)

128

Figure I–4. Photo of Epomophorus labiatus or Epomophorus wahlbergi. (Photo by Charlie Bear 2003.)

Figure I–5. Photo of juvenile Epomophorus labiatus or Epomophorus wahlbergi. (Photo by Charlie Bear 2003.)

129

Figure I–6. Photo of Epomophorus labiatus or Epomophorus wahlbergi showing facial detail. (Photo by Graeme Cumming 2003.)

130

Figure I–7. Photo of Epomophorus labiatus or Epomophorus wahlbergi. (Photo by Charlie Bear 2003.)

Figure I–8. Photo of Micropteropus pusillus. (Photo by Charlie Bear 2003.)

131

Figure I–9. Photo of Micropteropus pusillus illustrating the small size and short rostrum. (Photo by Charlie Bear 2003.)

Figure I–10. Photo of a small Micropteropus pusillus. (Photo by Charlie Bear 2003.)

132

Figure I–11. Photo of Epomophorus minor illustrating relatively short rostrum length. (Photo by Charlie Bear 2003.)

Figure I–12. Photo of Epomophorus minor. (Photo by Charlie Bear 2003.)

133

Figure I–13. Photo of Epomophorus minor. (Photo by Charlie Bear 2003.)

LIST OF REFERENCES

Ansell, W. F. H. (1960). Mammals of Northern Rhodesia. Lusaka, Zambia: The Government Printer.

Ansell, W. F. H. (1978). The mammals of Zambia. Chilanga, Zambia: The National Parks and Wildlife Service.

Apps, P., (ed.) (1996). Smithers' mammals of Southern Africa: a field guide. Cape Town, South Africa: Southern Book Publishers.

Arditi, R. and Dacorogna, B. (1988). Optimal foraging on arbitrary food distributions and the definition of habitat patches. Am. Nat. 131: 837-846.

Aronson, R. B. and Givnish, T. J. (1983). Optimal central-place foragers: a comparison with null hypotheses. Ecology 64: 395-399.

Ayensu, E. S. (1974). Plant and bat interactions in West Africa. Ann. MO Bot. Gard. 61: 702-727.

Baker, H. G. (1973). Evolutionary relationships between flowering plants and animals in American and African tropical forests. In Tropical forest ecosystems in Africa and South America: a comparative review: 350. Meggers, B. J., Ayensu, E. S. and Duckworth, W. D. (Eds.). Washington, D. C.: Smithsonian Institution Press.

Bakun, A. and Broad, K. (2003). Environmental 'loopholes' and fish population dynamics: comparative pattern recognition with focus on El Nino effects in the Pacific. Fisheries Oceanography 12: 458-473.

Bergmans, W. (1988). Taxonomy and biogeography of African fruit bats (Mammalia, Megachiroptera). 1. General introduction; material and methods; results: the genus Epomophorus Bennett, 1836. Beaufortia 38: 75-146.

Bergmans, W. (1989). Taxonomy and biogeography of African fruit bats (Mammalia, Megachiroptera). 2. The genera Micropteropus Matschie, 1899, Epomops Gray, 1870, Hypsignathus H. Allen, 1861, Nanonycteris Matschie, 1899, and Plerotes Andersen, 1910. Beaufortia 39: 89-153.

Bergmans, W. (1990). Taxonomy and biogeography of African fruit bats (Mammalia, Megachiroptera). 3. The genera Scotonycteris Matshcie, 1894, Casinycteris Thomas, 1910, Pteropus Brisson, 1762, and the fruit bat Eidolon Rafinesque, 1815. Beaufortia 40: 111-177.

134 135

Berthold, P. and Terrill, S. B. (1991). Recent advances in studies of bird migration. Annu. Rev. Ecol. Syst. 22: 357-378.

Bonaccorso, F. J. and Gush, T. J. (1987). Feeding behaviour and foraging strategies of captive phyllostomid fruit bats: an experimental study. J. Anim. Ecol. 56: 3907-3920.

Bonaccorso, F. J., Winklemann, J. R., Dumont, E. R. and Thibault, K. (2002). Home range of Dobsonia minor (Pteropodidae): a solitary foliage-roosting fruit bat in Papua New Guinea. Biotropica 34: 127-135.

Brown, J. L. and Orians, G. H. (1970). Spacing patterns in mobile animals. Annu. Rev. Ecol. Syst. 1: 239-262.

Chapman, C. A., Wrangham, R. and Chapman, L. J. (1994). Indices of habitat-wide fruit abundance in tropical forest. Biotropica 26: 160-171.

Charnov, E. L. (1976). Optimal foraging, the marginal value theorem. Theor. Popul. Biol. 9: 129-136.

Chesser, R. T. and Levey, D. J. (1998). Austral migrants and the evolution of migration in New World birds: diet, habitat, and migration revisited. Am. Nat. 152: 311-319.

Chidumayo, E. N. (1987). Species structure in Zambian miombo woodland. J. Trop. Ecol. 3: 109-118.

Coates Palgrave, K. (2002). Trees of Southern Africa. Cape Town, South Africa: Struik Publishers.

Cohen, D. (1967). Optimization of seasonal migratory behavior. Am. Nat. 101: 5-17.

Cosson, J.-F. (1995). Captures of Myonycteris torquata (Chiroptera: Pteropodidae) in forest canopy in South Cameroon. Biotropica 27: 395-396.

Cosson, J.-F., Pons, J.-M. and Masson, D. (1999). Effects of forest fragmentation on frugivorous and nectarivorous bats in French Guiana. J. Trop. Ecol. 15: 515-534.

Cosson, J.-F., Tranier, M. and Colas, F. (1996). On the occurrence and possible migratory behaviour of the fruit bat Eidolon helvum in Mauritania, Africa. J. Afr. Zool. 110: 369-371.

Cox, G. W. (1985). The evolution of avian migration systems between temperate and tropical regions of the New World. Am. Nat. 126: 451-474.

Cumming, G. S. and Bernard, R. T. F. (1997). Rainfall, food abundance and timing of parturition in African bats. Oecologia 111: 309-317.

DeFrees, S. and Wilson, D. E. (1988). Eidolon helvum. Mammalian Species 312: 1-5.

136

Dumont, E. R. (2003). Bats and fruit. In Bat ecology: 398-429. Kunz, T. H. and Fenton, M. B. (Eds.). Chicago: University of Chicago Press.

Duncan, R. S. and Chapman, C. A. (1999). Seed dispersal and potential forest succession in abandoned agriculture in tropical Africa. Ecol. Appl. 9: 998-1008.

Eby, P. (1991). Seasonal movements of gray-headed flying foxes, Pteropus poliocephalus (Chiroptera, Pteropodidae), from 2 maternity camps in northern New South Wales. Wildl. Res. 18: 547-559.

Elgood, J. H., Fry, C. H. and Downsett, R. J. (1973). African migrants in Nigeria. Ibis 115: 1-115.

Fahr, J. and Ebigbo, N. M. (2003). A conservation assessment of the bats of the Simandou Range, Guinea, with the first record of Myotis welwitschii (Gray, 1866) from West Africa. Acta Chiroptol. 5: 125-141.

Fanshawe, D. B. (1984). Fifty common trees of Zambia. Lusaka, Zambia: The Government Printer.

Fayenuwo, J. and Halstead, L. B. (1974). Breeding cycle of the straw-colored fruit bat, Eidolon helvum, at Ile-Ife, Nigeria. J. Mammal. 55: 453-454.

Fenton, M. B. (1975). Observations on the biology of some Rhodesian bats, including a key to the Chiroptera of Rhodesia. Ontario, Canada: Life Sciences contributions- Royal Ontario Museum.

Fenton, M. B. (1983). Roosts used by the African bat, Scotophilus leucogaster (Chiroptera: Vespertilionidae). Biotropica 15: 129-132.

Fenton, M. B., Acharya, L., Audet, D., Hickey, M. B. C., Merriman, C., Obrist, M. K., Syme, D. M. and Adkins, B. (1992). Phyllostomid bats (Chiroptera: Phyllostomidae) as indicators of habitat disruption in the neotropics. Biotropica 24: 440-446.

Fenton, M. B., Brigham, R. M., Mills, A. M. and Rautenbach, I. L. (1985). The roosting and foraging areas of Epomophorus wahlbergi (Pteropodidae) and Scotophilus viridis (Vespertilionidae) in Kruger National Park, South Africa. J. Mammal. 66: 461-468.

Fenton, M. B., Cumming, D. H. M., Rautenbach, I. L. N., Cumming, G. S., Cumming, M. S., Ford, G., Taylor, R. D., Dunlop, J., Hovorka, M. D., Johnston, D. S., Portfors, C. V., Kalcounis, M. C. and Mahlanga, Z. (1998). Bats and the loss of tree canopy in African woodlands. Conserv. Biol. 12: 399-407.

Fenton, M. B. and Rautenbach, I. L. (1998). Impacts of Ignorance and Human and Elephant Populations on the Conservation of Bats in African Woodlands. In Bat Biology and Conservation: 261-270. Racey, P. A. (Ed.) Washington, D.C.: Smithsonian Institution Press.

137

Fenton, M. B. and Thomas, D. W. (1980). Dry-season overlap in activity patterns, habitat use, and prey selection by sympatric African insectivorous bats. Biotropica 12: 81- 90.

Findley, J. S. (1976). The structure of bat communities. Am. Nat. 110: 129-139.

Findley, J. S. and Black, H. (1983). Morphological and dietary structuring on a Zambian insectivorous bat community. Ecology 64: 625-630.

Findley, J. S. and Wilson, D. E. (1983). Are bats rare in tropical Africa? Biotropica 15: 299-303.

Fleming, T. H. (1982). Foraging strategies of plant-visiting bats. In Ecology of Bats: 287- 325. Kunz, T. (Ed.) New York: Plenum Press.

Fleming, T. H. and Eby, P. (2003). Ecology of bat migration. In Bat ecology: 156-208. Kunz, T. H. and Fenton, M. B. (Eds.). Chicago: The University of Chicago Press.

Francis, C. M. (1994). Vertical stratification of fruit bats (Pteropodidae) in lowland dipterocarp rainforest in Malaysia. J. Trop. Ecol. 10: 523-530.

Fryxell, J. M., Greever, J. and Sinclair, A. R. E. (1988). Why are migratory ungulates so abundant? Am. Nat. 131: 781-798.

Fujita, M. S. and Tuttle, M. D. (1991). Flying foxes (Chiroptera:Pteropodidae): threatened animals of key ecological and economic importance. Conserv. Biol. 5: 455-463.

Fuller, D. O. (1999). Canopy phenology of some mopane and miombo woodlands in Eastern Zambia. Global Ecology and Biogeography 8: 199-209.

Godinez-Alvarez, H., Valiente-Banuet, A. and Rojas-Martinez, A. (2002). The role of seed dispersers in the population dynamics of the columnar cactus Neobuxbaumia tetetzo. Ecology 83: 2617-2629.

Hamilton, W. J., III and Watt, K. E. F. (1970). Refuging. Annu. Rev. Ecol. Syst. 1: 263- 286.

Heideman, P. D. and Heaney, L. R. (1989). Population biology and estimates of abundance of fruit bats (Pteropodidae) in Philippine submontane rainforest. J. Zool. (Lond.) 218: 565-586.

Heithaus, E. R. and Fleming, T. H. (1978). Foraging movements of a frugivorous bat, Carollia perspicillata (Phyllostomatidae). Ecol. Monogr. 48: 127-143.

Heithaus, E. R., Fleming, T. H. and Opler, P. A. (1975). Foraging patterns and resource utilization in seven species of bats in a seasonal tropical forest. Ecology 56: 841-854.

138

Hepburn, H. R. and Radloff, S. E. (1995). First approximation to a phenology of the honeybees (Apis mellifera) and flora of Africa. Oecologia 101: 265-273.

Horn, H. S. (1968). The adaptive significance of colonial nesting in the Brewer's blackbird (Eiphagus cyanocephalus). Ecology 49: 682-694.

Horner, M. A., Fleming, T. H. and Sahley, C. T. (1998). Foraging behaviour and energetics of a nectar-feeding bat, Leptoycteris curasoae (Chiroptera: Phyllostomidae). J. Zool. (Lond.) 244: 575-586.

Huxel, G. R., McCann, K. and Polis, G. A. (2002). Effects of partitioning allochthonous and autochthonous resources on food web stability. Ecol. Res. 17: 419-432.

Jones, C. (1972). Comparative ecology of three pteropid bats in Rio Muni, West Africa. J. Zool. (Lond.) 167: 353-370.

Kalko, E. K. V. (1998). Organisation and diversity of tropical bat communities through space and time. Zoology 101: 281-297.

Kalko, E. K. V. and Handley, C. O. J. (2001). Neotropical bats in the canopy: diversity, community structure, and implications for conservation. Plant Ecol. 153: 319-333.

Katz, P. L. (1974). A long-term approach to foraging optimization. Am. Nat. 108: 758- 782.

Kingdon, J. (1984). East African mammals: an atlas of evolution in Africa: Vol. 2 Part A, insectivores and bats. Chicago: University of Chicago Press.

Kunz, T., (ed.) (1988). Ecological and behavioral methods for the study of bats. Washington, D. C.: Smithsonian Institution Press.

Lack, P. C. (1983). The movements of Palaearctic landbird migrants in Tsavo East National Park, Kenya. J. Anim. Ecol. 52: 513-524.

Legendre, P. (2001). TRANSFOR. http://www.fas.umontreal.ca/biol/casgrain/en/telecharger/index.html#Transformation sSpeciesScores. Last accessed 5 July 2004.

Legendre, P. and Anderson, M. J. (1999). Distance-based redundancy analysis: testing multispecies responses in multifactorial ecological experiments. Ecol. Monogr. 69: 1-24.

Legendre, P. and Gallagher, E. D. (2001). Ecologically meaningful transformations for ordination of species data. Oecologia 129: 271-280.

Levey, D. J. and Stiles, F. G. (1992). Evolutionary precursors of long-distance migration - resource availability and movement patterns in neotropical landbirds. Am. Nat. 140: 447-476.

139

Lewis, S. E. (1995). Roost fidelity of bats: a review. J. Mammal. 76: 481-496.

Lindstrom, A. (1995). Stopover ecology of migrating birds - some unsolved questions. Isr. J. Zool. 41: 407-416.

Loehle, C. (1995). Social barriers to pathogen transmission in wild animal populations. Ecology 76: 326-335.

Marshall, A. G. (1983). Bats, flowers and fruit: evolutionary relationships in the Old World. Biol. J. Linn. Soc. 20: 115-135.

Marshall, A. G. (1985). Old World phytophagous bats (Megachiroptera) and their food plants: a survey. Zool. J. Linn. Soc. 83: 351-369.

Marshall, A. G. and McWilliam, A. N. (1982). Ecological observations on epomorphine fruit-bats (Megachiroptera) in West African savanna woodland. J. Zool. (Lond.) 198: 53-67.

McCune, B. and Grace, J. B. (2002). Analysis of ecological communities. Gleneden Beach, OR: MjM Software Design.

McCune, B. and Mefford, J. (1999). Multivariate analysis of ecological data. (4.20 edn). Gleneden Beach, OR: MjM Software.

Mickelburgh, S. P., Hutson, A. M. and Racey, P. A. (1992). Old World fruit bats: an action plan for their conservation. Gland, Switzerland: IUCN, IUCN/SSC Chiroptera Specialist Group.

Morin, P. J. (1999). Community ecology. Malden, MA: Blackwell Science, Inc.

Morrison, D. W. (1978a). Foraging ecology and energetics of the frugivorous bat Artibeus jamaicensis. Ecology 59: 716-723.

Morrison, D. W. (1978b). On the optimal searching strategy for refuging predators. Am. Nat. 112: 925-934.

Mutere, F. A. (1967). The breeding biology of equatorial vertebrates: reproduction in the fruit bat, Eidolon helvum, at latitude 0º 20'N. J. Zool. 153: 153-161.

Nelson, S. L. 2003. Nutritional ecology of Old World fruit bats: a test of the calcium- constraint hypothesis. PhD Thesis, University of Florida, Gainesville.

Nowak, R. M. (1994). Walker's bats of the world. Baltimore: Johns Hopkins University Press.

Nowak, R. M., (ed.) (1997). Walker's mammals of the world. Baltimore: Johns Hopkins University Press.

140

O'Shea, T. J. and Vaughan, T. A. (1980). Ecological observations on an East African bat community. Mammalia 44: 485-496.

Palmeirim, J. and Etheridge, K. (1985). The influence of man-made trails on foraging by tropical frugivorous bats. Biotropica 17: 82-83.

Palmer, C. and Woinarski, J. C. Z. (1999). Seasonal roosts and foraging movements of the black flying fox (Pteropus alecto) in the Northern Territory: resource tracking in a landscape mosaic. Wildl. Res. 26: 823-838.

Parrish, J. D. (1997). Patterns of frugivory and energetic condition in Nearctic landbirds during autumn migration. Condor 99: 681-697.

Parrish, J. D. (2000). Behavioral, energetic, and conservation implications of foraging plasticity during migration. Stud. Avian Bio. 20: 53-70.

Patterson, B. D., Willig, M. R. and Stevens, R. D. (2003). Trophic strategies, niche partitioning, and patterns of ecological organization. In Bat ecology: 536-579. Kunz, T. H. and Fenton, M. B. (Eds.). Chicago: University of Chicago Press.

Polis, G. A., Anderson, W. B. and Holt, R. D. (1997). Toward an integration of landscape and food web ecology: the dynamics of spatially subsidized food webs. Annu. Rev. Ecol. Syst. 28: 289-316.

Putman, R. J. (1994). Community ecology. London, England: Chapman & Hall.

Pyke, G. H., Pullman, H. R. and Charnov, E. L. (1977). Optimal foraging: a selective review of theory and tests. Q. Rev. Biol. 52: 137-154.

Racey, P. A. (2002). Visit to Kasanka 21-27 November 2002. Aberdeen, Scotland: University of Aberdeen.

Racey, P. A. and Entwistle, A. C. (2000). Life-history and reproductive strategies of bats. In Reproductive biology of bats: 363-414. Crichton, E. G. and Krutzsch, P. H. (Eds.). New York: Academic Press.

Racey, P. A. and Entwistle, A. C. (2003). Conservation ecology of bats. In Bat ecology: 680-743. Kunz, T. and Fenton, M. B. (Eds.). Chicago: University of Chicago Press.

Reeve, J. D. (1988). Environmental variability, migration, and persistence in host- parasitoid systems. Am. Nat. 132: 810-836.

Reid, F. (2003a). Key to the bats at Kasanka National Park, Zambia. Toronto, Canada: Royal Ontario Museum.

Reid, F. (2003b). Key to the bats of Mozambique. Toronto, Canada: Royal Ontario Museum.

141

Reid, F. (2003c). Key to the bats of Zimbabwe. Toronto, Canada: Royal Ontario Museum.

SAS Institute Inc. (1999). SAS/STAT® user's guide, version 8. Cary, NC: SAS Institute Inc.

SAS Institute Inc. (2004). SAS Version 9.0. (9.0 edn). Cary, NC: SAS Institute Inc.

Schoener, T. W. (1971). Theory of feeding strategies. Annu. Rev. Ecol. Syst. 2: 369-404.

Smith, P., (ed.) (2001). Ecological survey of Zambia: the traverse records of C.G. Trapnell, 1932-43. Kew, Great Britain: The Board of Trustees of the Royal Botanic Gardens.

Smith, P. P. and Fisher, R. (2001). Chipya in Kasanka National Park, Zambia: floristics, soils and dynamics. In Plant systematics and phytogeography for the understanding of African biodiversity: 923-934. Robbrecht, E., K., D. and Friis, I. (Eds.). Belgium: National Botanic Garden of Belgium.

Sorensen, U. G. and Halberg, K. (2001). Mammoth roost of nonbreeding straw-coloured fruit bat Eidolon helvum (Kerr, 1792) in Zambia. Afr. J. Ecol. 39: 213-215.

Spencer, H. J., Palmer, C. and Parry-Jones, K. (1991). Movements of fruit-bats in Eastern Australia, determined by using radio-tracking. Wildl. Res. 18: 463-468.

Storrs, A. E. G. (1995). Know your trees: some common trees found in Zambia. Lusaka, Zambia: Regional Soil Conservation Unit.

Stuart, C. and Stuart, T. (1989). Chris and Tilde Stuart's field guide to the mammals of Southern Africa. Cape Town, South Africa: Struik Publishers (Pty) Ltd.

Stuart, C. and Stuart, T. (2001). Bats en masse. In Africa - environment & wildlife: 72-76.

Taylor, D. A. R. and Kankam, B. O. (1999). The role of the straw-colored fruit bat, Eidolon helvum, in seed dispersal, survival and germination in Milicia excelsa, a threatened West African hardwood. Flagstaff, Arizona; Kumasi, Ghana: Northern Arizona University; The Forestry Research Institute of Ghana.

Taylor, P. J. (2000). Bats of Southern Africa. Pietermaritzburg, South Africa: University of Natal Press.

Thomas, D. W. 1982. The ecology of an African savanna fruit bat community: resource partitioning and role in seed dispersal. PhD, University of Aberdeen, Aberdeen, Scotland.

Thomas, D. W. (1983). The annual migrations of 3 species of West-African fruit bats (Chiroptera, Pteropodidae). Can. J. Zool. 61: 2266-2272.

142

Thomas, D. W. (1984). Fruit intake and energy budgets of frugivorous bats. Physiol. Zool. 57: 457-467.

Thomas, D. W. and Fenton, M. B. (1978). Notes on dry season roosting and foraging behavior of Epomophorus gambianus and Rousettus aegyptiacus (Chiroptera- Pteropodidae). Journal of Zoology 186: 403-406.

Thomas, D. W. and Marshall, A. G. (1984). Reproduction and growth in three species of West Africa fruit bats. J. Zool. (Lond.) 202: 265-281.

Tidemann, C. R. and Loughland, R. A. (1993). A harp trap for large megachiropterans. Wildl. Res. 20: 607-611.

Tidemann, C. R. and Nelson, J. E. (2004). Long-distance movements of the grey-headed flying fox (Pteropus poliocephalus). J. Zool. (Lond.) 263: 141-146.

Utzurram, R. C. B. (1995). Feeding ecology of Philippine fruit bats: patterns of resource use and seed dispersal. Symp. Zool. Soc. Lond. 67: 63-77. van Schaik, P. C., Terborgh, J. W. and Wright, S. J. (1993). The phenology of tropical forests: adaptive significance and consequences for primary consumers. Annu. Rev. Ecol. Syst. 24: 353-377.

Ward, P. and Zahavi, A. (1973). Importance of certain assemblages of birds as information-centers for food-finding. Ibis 115: 517-534.

Weber, T. P. and Houston, A. I. (1997). Flight costs, flight range and the stopover ecology of migrating birds. J. Anim. Ecol. 66: 297-306.

Whitney, K. D. and Smith, T. B. (1998). Habitat use and resource tracking by African Ceratogymna hornbills: implications for seed dispersal and forest conservation. Animal Conservation 1: 107-117.

Zubaid, A. (1994). Vertical stratification of pteropodid bats in a Malaysian lowland rainforest. Mammalia 58: 309-311.

BIOGRAPHICAL SKETCH

Heidi V. Richter was born on February 5, 1975, in Putnam Valley, New York.

Daughter of a construction consultant and innovator father and a graphic artist turned sheep farmer mother, she spent most of her formative years on a sheep farm in Glen, New

York. She attended high school at Fonda–Fultonville Central School and graduated in

June 1993. In August 1993 she began her university career at Cornell University in

Ithaca, New York, from which she earned the degree of Bachelor of Science in biological sciences, with a concentration in ecology and evolution, in May 1997. After a three–year stint with Peace Corps Zambia, first as an Aquaculture Extension Agent and then as

Volunteer Leader, she began her true wildlife career volunteering as a safari guide for

Kasanka National Park in Zambia. Upon her return to the United States she worked in wetland rehabilitation and salmon habitat restoration for the state of Washington. Her continuing interest in wildlife and the environment brought her to the University of

Florida in August 2002. She graduated with a Master of Science in wildlife ecology and conservation in December 2004. She lives with her husband John Howard in Seattle,

WA.

143