University of Texas at El Paso DigitalCommons@UTEP

Open Access Theses & Dissertations

2016-01-01 Phylogeography Of The aN tal Puddle Natalensis (anura: Phrynobatrachidae) In Sub-Saharan Africa Joshua Adan Lara University of Texas at El Paso, [email protected]

Follow this and additional works at: https://digitalcommons.utep.edu/open_etd Part of the Biology Commons

Recommended Citation Lara, Joshua Adan, "Phylogeography Of The aN tal Puddle Frog Phrynobatrachus Natalensis (anura: Phrynobatrachidae) In Sub- Saharan Africa" (2016). Open Access Theses & Dissertations. 876. https://digitalcommons.utep.edu/open_etd/876

This is brought to you for free and open access by DigitalCommons@UTEP. It has been accepted for inclusion in Open Access Theses & Dissertations by an authorized administrator of DigitalCommons@UTEP. For more information, please contact [email protected]. PHYLOGEOGRAPHY OF THE NATAL PUDDLE FROG

PHRYNOBATRACHUS NATALENSIS (ANURA: PHRYNOBATRACHIDAE) IN

SUB-SAHARAN AFRICA

JOSHUA ADAN LARA

Master’s Program in Biology

APPROVED:

Eli Greenbaum, Ph.D

Michael Moody, Ph.D

Edward Castañeda, Ph.D

Charles Ambler, Ph.D. Dean of the Graduate School

Copyright ©

by Joshua Adan Lara 2016

PHYLOGEOGRAPHY OF THE NATAL PUDDLE FROG

PHRYNOBATRACHUS NATALENSIS (ANURA: PHRYNOBATRACHIDAE) IN

SUB-SAHARAN AFRICA

by JOSHUA ADAN LARA, B.S.

THESIS

Presented to the Faculty of the Graduate School of The University of Texas at El Paso in Partial Fulfillment of the Requirements for the Degree of

MASTER OF SCIENCE

Department of Biological Sciences THE UNIVERSITY OF TEXAS AT EL PASO May 2016 ACKNOWLEDGMENTS

First, I would like to thank my graduate committee, Dr. Eli Greenbaum, Dr. Michael

Moody, and Dr. Edward Castañeda. I would like to give special thanks to my thesis advisor Dr.

Eli Greenbaum, who took me on as a graduate student on short notice. I would like to thank the entire faculty at UTEP for providing support. I thank Breda Zimkus, Ph.D. (Harvard University’s

Museum of Comparative Zoology), Werner Conradie (Port Elizabeth Museum, South Africa),

Stephane Boissinot, Ph.D. (New York University, Abu Dhabi), and Mark-Oliver Rödel, Ph.D.

(Museum für Naturkunde, Berlin, Germany) for access to specimens. Moreover, I would like to thank the UTEP Border Biomedical Research Center (BBRC) Genomics Analysis Core Facility, especially Ana Betancourt. I am also thankful to all my lab members Frank Portillo, Danny

Hughes, Maria “Fernie” Medina, and Thornton Larson for all of their assistance.

iv ABSTRACT

The Family Phrynobatrachidae (Laurent, 1941) has one (Phrynobatrachus

Günther, 1862) containing 89 . Commonly known as puddle , they comprise one of the most species rich sub-Saharan groups. The objective of this study was to investigate Phrynobatrachus natalensis from the Albertine Rift (AR) with systematic and phylogeographic methods. One mitochondrial DNA gene (16S = 555 base pairs [bp]) and one nuclear gene (RAG1 = 774 bp) were analyzed for 61 representatives of P. natalensis. Nine cryptic lineages were identified from , , Côte d’Ivoire, Democratic Republic of the Congo, , Ghana, , Namibia, , , and . Several of these lineages are likely new species. Genetic diversity among populations of P. natalensis for

16S was moderate to high (4–13%). Low genetic divergence was recovered in RAG1 (0–1%).

Morphological characters were used to examine species boundaries in distinct lineages identified from the molecular phylogeny. No clear pattern was evident in these data. The results of this study suggest the diversity of P. natalensis has been severely underestimated. Additional research efforts should focus on the biogeographic patterns, reproductive isolation, acoustic calls, reproductive behavior, and ecology to determine species boundaries. Furthermore, it is imperative to sample more African sites, especially in the AR, to get a better understanding of

African biodiversity.

v TABLE OF CONTENTS

ACKNOWLEDGMENTS ...... iv

ABSTRACT ...... v

TABLE OF CONTENTS ...... vi

LIST OF TABLES ...... viii

LIST OF FIGURES ...... ix

CHAPTER 1: INTRODUCTION ...... 1 1.1 Background ...... 1 1.2 Natural history ...... 5 1.3 Reproduction ...... …...... 6 1.4 Voice ...... …. 7 1.5 and systematics ...... ….7 1.6 Study site ...... 9 1.7 Research objectives and questions ...... 14

Chapter 2: MATERIALS AND METHODS ...... 15 2.1 Specimen acquisiton ...... 15 2.2 DNA extraction ...... 15 2.3 DNA amplification, sequencing and alignment ...... 16 2.4 Phylogenetic analyses ...... 19 2.5 Analyses of genetic distance ...... 19 2.6 Morphological analyses of adult frogs ...... 20

Chapter 3: RESULTS ...... 22 3.1 Phylogenetic analyses ...... 22 3.2 Analyses of genetic distance ...... 23 3.3 Albertine Rift P. natalensis morphology ...... 23 Chapter 4: DISCUSSION ...... 31 4.1 Taxonomic status and phylogeography of the Albertine Rift ...... 31 4.2 Conservation ...... 35 4.3 Conclusion and future directions ...... 38

vi

RERFERNCES ...... …..39

APPENDIX I ...... 49

APPENDIX II ...... 52

VITA … ...... 53

vii LIST OF TABLES

Table 2.3.1 Primers used for sequencing genes…………………………………………………18

Table 3.1.1 Model parameters specified by PartitionFinder and alternate models implemented in MrBayes ...... 22

Table 3.2.1 Uncorrected p-distances (16S) among selected individuals from the P. natalensis phylogeny…………..…………………………………………………………………………….27

Table 3.2.2 Uncorrected p-distances (RAG1) among selected individuals from the P. natalensis phylogeny……………...…………………………………………………………………………28

Table 3.3.1 Morphometric data of adult male voucher specimens of P. cf. natalensis from selected clades from Fig. 3.1.1 (in mm). Data are presented as mean + one standard deviation, followed by range……………………………………………………………………....…………29

Table 3.3.2 Principal components analysis comparing adult male specimens from clades of Albertine Rift P. cf. natalensis with raw mensural data regressed against snout–vent length. Eigenvalues, percent variance, cumulative variance, and loadings are shown for the first three principal components……………………………………………………………………………30

viii LIST OF FIGURES

Figure 1.4 Map of sub-Saharan Africa showing the disjunct distribution of Phrynobatrachus natalensis. Sampled localities for the genetic analyses are shown with circles. Circle colors correspond to the clades in Fig. 3.1.1. Map was modified from Sarye (2013). A black polygon indicates the type locality of Phrynobatrachus natalensis (KwaZulu-Natal, South Africa)…..………………………………………………...... 17

Figure 3.1.1 MrBayes phylogenetic tree (16S and RAG1 genes) of Phrynobatrachus natalensis samples. Support values represent posterior probability (> 0.95)/bootstrap support (> 70). Colors in the phylogenetic tree correspond to Figure 1.4…………………...... 25

Figure 3.3.1 Score plots of first and second principal components obtained from a PCA with SVL as the covariate for adult male specimens of P. cf. natalensis clades from the Albertine Rift ……………...…………………………………………………………………………………….26

ix CHAPTER 1: INTRODUCTION

1.1 Background The identification of many diverse vertebrate groups has been made possible with the advent of molecular phylogenetics. Little to no baseline data and widespread extinctions are growing concerns for many vertebrate groups globally (Zimkus et al., 2010). Therefore, thorough taxonomic revisions are needed for lineages with longstanding confusion resulting from geographic and intraspecific variation (Rödel, 2000; Zimkus et al., 2010).

One of the most ubiquitous groups of vertebrates is the Order Anura (frogs and toads), which is composed of 55 families with 445 genera and 7,517 species (as of March 10, 2016)

(AmphibiaWeb, 2016). This list continues to be updated on a daily basis. are reliable environmental indicators because of their response to changes in the whole environment, rather than just a few or locations (Collins and Halliday, 2005). This response to their environment has been one of the contributing factors to amphibian declines worldwide. The

International Union for Conservation of Nature and Natural Resources (IUCN) Red List of threatened species website (http://www.iucnredlist.org/) has been highlighting the status of globally threatened biodiversity in order to provide the world with the most objective, scientifically based information. According to this website, nearly one third (32%) of amphibians are threatened, representing 1,856 species. It is thought that nearly 168 species have gone extinct, and 2,469 (43%) species have populations that are declining (IUCN, 2014). Data deficient (DD) status in amphibians is also an important factor to consider. In 2008, (24.5%) of amphibians were considered DD, particularly in tropical ecosystems (IUCN, 2014; Stuart et al., 2004).

Assessments of amphibian populations in tropical ecosystems are lacking, especially with regard to anuran biodiversity (Vieites et al., 2009). Several factors are contributing to population

1 declines in amphibians. Biologists around the globe agree that loss, environmental alteration, fragmentation, and a pathogen (Batrachochytrium dendrobatidis) associated with amphibian declines, are the immediate causes (Boyle et al., 2004; Collins and Halliday, 2005).

An outbreak of amphibian chytridiomycosis, caused by chytrid fungus

(Batrachochytrium dendrobatidis) (Bd), has been responsible for massive amphibian declines worldwide (Rödder et al., 2009; Soto-Azat et al., 2010). In terms of numbers of species impacted, Bd has been described as “the worst infectious disease recorded, with propensity to drive them [amphibians] to extinction” (Gascon et al., 2007). Batrachochytrium dendrobatidis causes rapid declines and species extinctions, even in pristine and protected areas, thus playing a decisive role in the global biodiversity crisis (Rödder et al., 2009). The spread of this fungus has extended worldwide except in Antarctica, where no amphibians can survive. The earliest case of

Bd infection was found in the Natural History Museum, London in six specimens of Xenopus fraseri (Family ) collected from Cameroon in 1933 (Soto-Azat et al., 2010). The first description of amphibian chytridiomycosis was in 1998, but studies have provided evidence of

Bd emerging as a global threat to amphibians as early as the 1970s (Burrowes et al., 2004;

Voyles et al., 2007). Chytrid fungus infection causes a light-brown discoloration, granular texture, and prominent lesions on the ventral abdomen, pelvis, and legs of its amphibian hosts

(Nichols et al., 2001). The infection directly infects the superficial, keratin-containing layers of amphibian skin, leading to low osmotic regulation, and a drop in electrolyte blood levels, which results in death from cardiac arrest (Berger and Hyatt, 1999; Voyles et al., 2009). The Aquatic

Animal Health Code of the World Organization for Health (OIE) has listed Bd as a notifiable disease. A popular hypothesis that Bd originated in Africa and spread by global trade in Xenopus sp., because of its use in a pregnancy assay, is in dispute (Farrer et al., 2011).

2 Xenopus have been known to be an asymptomatic carrier of the Bd infection (Weldon et al.,

2004). In a recent analysis, Farrer et al. (2011) found multiple genetically diverse hypervirulent recombinant lineages by using comparative population genomics. These authors proposed that anthropogenic mixing of allopatric lineages of Bd has led to the generation of the hypervirulent global panzootic lineage (BdGPL). The global trade in amphibians resulted in cross-transmission of Bd among previously naive host species, causing the large frequency of recombinant genotypes (Farrer et al., 2011). The largely unknown manner of Bd’s geographic origin is still under debate.

The potential for Bd to cause catastrophic die-offs in species-rich areas where endemism and the number of threatened amphibians are among the highest in the world (such as Central

Africa) is a major concern (Burgess et al., 2004; IUCN, 2014). Most areas in Central Africa are poorly surveyed with very little data regarding the status of threatened amphibians (Broadley and

Cotterill, 2004; Channing and Howell, 2006; Laurent, 1983; Plumptre et al., 2003, 2007). An infection in understudied areas could potentially cause extinctions, denying scientists the ability to assess amphibian biodiversity and species richness. A preliminary survey of chytrid infections in amphibians was conducted in Kahuzi-Biega National Park, South Kivu Province, Democratic

Republic of Congo (DRC) in 2007. A total of 24 specimens were tested, and two showed evidence of chytridiomycosis. The positive specimens included a subadult kivuensis with no evidence of lesions, and a juvenile H. langi (originally reported as H. kuligae) with small white lesions on the venter of the hind limbs and abdomen (Greenbaum et al., 2008). A more recent survey in DRC evaluated the threat of Bd in multiple habitats of the Albertine Rift (AR).

A total of 166 skin swabs were taken from 45 sites in the AR, and 58 frogs tested positive for Bd

(34.9% prevalence) (Greenbaum et al., 2015). When historical specimens (collected from 1925–

3 1994) from museum collections were tested, only one sample (Phrynobatrachus asper) out of

232 skin swabs tested positive for Bd (Seimon et al., 2015). Further testing for additional species and localities considered biodiversity hotspots, including the AR, are of the utmost importance in order to assess potential threats (Greenbaum et al., 2008; Plumptre et al., 2007). The AR is one of the most important regions for biodiversity in Africa, containing more vertebrate and endemic species than any other region in continental Africa (Plumptre et al., 2007).

Worldwide, tropical ecosystems contain 80% of the world’s amphibian species.

Historically, amphibian studies have been biased towards temperate regions, resulting in poor sampling for natural history data, taxonomic research, and conservation status of the species in tropical ecosystems. In 1986, 4,015 amphibians had been described and named, but as of March

10, 2016, 7,517 species had been described at the rate of one to two weekly (AmphibiaWeb,

2016; Halliday, 2008). Research in poorly explored areas can lead to more meaningful amphibian conservation measures, and identify ecosystems affected by environmental change.

Therefore, amphibian studies will lead to ecosystem protection and conservation. In order to organize conservation plans there must be sufficient knowledge of natural history, taxonomy, and new species descriptions (Mace, 2004).

The Family Phrynobatrachidae (first named by Laurent, 1941) is a natural group comprised of one genus: Phrynobatrachus Günther, 1862 (89 species) (Frost, 2016). The research herein is focused on Phrynobatrachus natalensis, the Natal Puddle Frog (Smith, 1849), with a focus on populations in the AR. Phrynobatrachus natalensis is found throughout the savannas of sub-Saharan Africa (AmphibiaWeb, 2016). Because of its widespread distribution, and morphological diversity of its eggs, larvae, and adults, P. natalensis populations have been

4 regarded as a complex of multiple cryptic species for several years (Channing, 2001; Channing and Howell, 2006; Largen, 2001; Rödel, 2000; Zimkus et al., 2010).

1.2 Natural history

Phrynobatrachus natalensis is found throughout the savannas of sub-Saharan Africa

(RÖdel, 2000). This species ranges from Senegal in the west to Somalia in the east, and southward through East Africa. In the south, it inhabits northeastern Namibia, northern

Botswana, and the Eastern Cape Province of South Africa. Phrynobatrachus natalensis can be found in relatively dry areas, but it is common in savanna and biomes, where rainfall exceeds 500 mm. Typically, P. natalensis has a pointed and slightly rounded snout. Males reach

25–30 mm, and females reach 26–31 mm snout–vent length (SVL) (Du Preez et al., 2009).

Dorsal coloration is often varied, but for the most part, the species is brown with a light vertebral stripe. Webbing is not uniform, which can be attributed to the species’ history of taxonomic confusion. Toe tips lack disks or distinct swellings, and males bear a thenar tubercle that enlarges during the breeding season (Harper et al., 2010). This species mainly preys on terrestrial invertebrates such as beetles, termites, spiders, flies, cockroaches, orthopterans, and butterflies.

During the rainy season, termites are the main source of their diet (Wells, 2007). Known predators of this species include the black-necked spitting cobra (Naja nigricollis) and herald snake (Crotaphopeltis hotamboeia) (Channing, 2001; Loveridge, 1935; Noble, 1924). Field observations suggest P. natalensis is most active from late afternoon to late evening (Lamotte and Xavier, 1966; Stewart, 1974). In males, the throat is dark grey, and the vocal sac has longitudinal folds analogous to the jawline (Du Preez et al., 2009).

5 1.3 Reproduction Phrynobatrachus natalensis eggs are laid in clutches of approximately 500–800 eggs in a single layer on the surface of water in ponds and temporary pools. Oviposited eggs are about one millimeter in diameter. It is not uncommon to find P. natalensis near human habitations, because of its tolerance to human disturbance (Harper et al., 2010; Passmore and Carruthers, 1995).

Tadpoles grow to 35 mm, the tail is 1.5x body length, and the nostrils are nearer to the eye than to the snout tip. Eyes are high on the head, and closer to the snout tip than the spiracle opening.

The spiracle is situated closer to the posterior end of the body than the eye. The dorsal fin starts at the beginning of the tail, and the body is more than twice the height of the ventral fin. The tail tip is obtuse. Fins are transparent and the body is brown. The vent opening is positioned on a small tube on the dextral margin of the ventral fin. The labial tooth row formula is 2(2)/2, and row P is very short. Thin jaw sheaths bear minute serrations and the oral disk is small. The bordering papillae occur in two rows where most outer rows are elongated, whereas those papillae at the angle of the jaw are rounded and short (Channing et al., 2012). According to

Rödel (2000), hatching was observed within 3–4 days, whereas metamorphosis took 4–5 weeks.

Breeding in West Africa starts in spring after the first rain and continues until late summer.

By the late rainy season, P. natalensis calls regularly at dusk and continues throughout the night. Breeding ponds have vegetation or cover along their banks. Males will use the vegetation cover for calling sites. Both male and female frogs migrate to the ponds after rainfall. Males find mates by calling during the day, and well into the night. The call has been described as “a slow quiet snore” (Channing and Howell, 2006; Passmore and Carruthers, 1995). When males are calling they are usually in a concealed site. Encounters with other males are usually aggressive during the breeding season (Wager, 1965, 1986).

6 1.4 Voice

Amphibian acoustic sounds communicate a variety of social interactions, mediating reproduction, and determining vulnerability or risk of predation (Ziegler et al., 2011). According to Channing and Howell (2006), P. natalensis can be heard during the day as well as at night.

Calls are exclusively allocated to males (Wells, 2007). The call has been described as being similar to Sclerophrys, but not as loud and heavy. The call is similar to P. acridoides, though the repetition rate is extended (Channing and Howell, 2006). Recordings of P. natalensis calls from the Serengeti National Park, Tanzania were described as a series of deep buzzes in unmeasured rhythm (Schiøtz, 1964). The rate of pulses was 48–56, and the frequency intensity maximum was between 1100−1500 Hz. Each pulse consisted of 33–44 pulses (Van den elzen and Kreulen,

1979). Recordings of P. natalensis are available at www.AmphibiaWeb.org.

1.5 Taxonomy and systematics

The Family Phrynobatrachidae (Laurent, 1941) has one genus (Phrynobatrachus,

Günther, 1862) containing 89 species. Commonly known as puddle frogs, they comprise one of the most species rich sub-Saharan amphibian groups. Dubois (2004) considered

Phrynobatrachidae to be a subfamily of Ranidae. Frost et al. (2006) placed Phrynobatrachidae as the sister taxon to and , and rendered a partial revision of the former genus.

Rödel (2000) was among the first to discuss the taxonomy of the species in West African savannas, noting the genus Phrynobatrachus was in need of great revision. Channing (2001) provided accounts for multiple Phrynobatrachus in southern Africa, whereas Channing and

Howell (2006) had accounts for the species in East Africa.

7 Further research led to a morphological data matrix of the species in Zimkus and

Blackburn (2008), and these authors discussed which external characteristics separate

Phrynobatrachus from the easily confused, but distantly related genus Arthroleptis

(). Two external morphological characters (presence of a tarsal tubercle and outer metatarsal tubercle) were found to be unique to Phrynobatrachus. Zimkus (2009) discussed the intra- and interspecific variability of “Little Brown Frogs” (LBF) from East Africa. Results revealed a wide range of genetic differentiation among the species, including P. natalensis, demonstrating the necessity for rigorous additional taxonomic study. In Zimkus et al. (2010), mitochondrial (12S rRNA, Valine-tRNA, and 16S rRNA) and nuclear (RAG1) data were used to construct the first genus-level phylogeny of puddle frogs. These authors analyzed biogeographic data, including habitat preference, elevation, and geographic distribution. Phylogenetic analyses showed support for three major clades (A, B, C) of puddle frogs, all of which were well supported in the Phrynobatrachidae family.

In a later study conducted by Zimkus and Schick (2010) in East African puddle frogs, 23 samples of P. natalensis were analyzed using 16S rRNA data. Populations of P. natalensis had moderate levels of genetic differentiation, ranging from 0.2–6.4% in uncorrected p-distances.

Herpetologists have suspected that P. natalensis is a cryptic species complex, because of its large distribution, combined with considerable morphological diversity of its eggs, larvae, and adults

(Channing, 2001; Channing and Howell, 2006; Largen, 2001; Rödel, 2000). Therefore, it is evident the true diversity of P. natalensis is underestimated. A thorough phylogenetic analysis is of the utmost importance to understand the taxonomy and evolutionary history of this cryptic species complex.

8 1.6 Study site

This study focused on Phrynobatractus natalensis, primarily from the AR in Central

Africa, as well as specimens from other areas of sub-Saharan Africa. The AR is located in the eastern border region of DRC, and extends from the northern end of Lake Albert to the southern end of Lake Tanganyika. The AR includes parts of western Uganda, Rwanda, Burundi, and

Tanzania (Plumptre et al., 2007).

The geological makeup of Africa started in the Phanerozoic, reflecting both the assembly of Pangea and the subsequent breakup of the Gondwanan supercontinent. Major tectonic folding and movement of Africa took place during the Ordovician, 480 million years ago (mya). Around

250 mya in the Permian, Africa was included in a large land mass called Pangaea, when all the continents were joined. The northward movement of Pangaea led to the separation of Laurasia from Gondwanaland during the Triassic, around 248 mya. During the Jurassic about 206 mya,

Gondwanaland began to break apart. Massive eruptions of basalts led to the separation of Africa and the Antarctic. This created the Table Mountains in the Cape Region in South Africa. This process eventually led to the formation of present-day South America, Africa, Antarctica,

Madagascar, Indonesia, and Australia (Vande weghe, 2004).

Africa today has remained isolated from other continents, and its surface has barely changed in the last 400 million years. The exception has been unflagging erosion and sea level fluctuations in the Cretaceous and tension on the earth’s crust, causing slow lifting of the entire western margin of Central Africa. This happened because of flexing of the continental plate, most likely because of the breakup of Gondwanaland and formation of the Atlantic Rift. The

Atlantic Rift is a mid-ocean ridge, a divergent tectonic plate located along the floor of the

Atlantic Ocean, which formed in the Triassic as Africa and South America began to split apart.

9 During Africa’s isolation period, modern tropical and plants appeared, and mammals, birds, and squamates diversified. In the Miocene, Africa formed the Arabo-African Rift, causing the separation of Arabia 20 mya. A land bridge between Asia and Africa was the outcome.

During the formation of the Arabo-African Rift, East Africa was affected by rifting. Faults split the earth’s crust for about 6,000 kilometers from present-day Palestine to Mozambique (Vande weghe, 2004).

The AR is part of the African superswell, which has gone through tectonic uplifting, volcanism, and rifting. This area experienced complex patterns of mantel circulation and plume development around 3040 mya (Malhi et al., 2013). The AR is thought to have formed from the impact of a mantle plume around 30 mya. The western branch of the AR was initiated by stress transmission across the rigid cratonic lithosphere and the rift basins developed along the preexisting ductile shear belts (Klerkx et al., 1998; Nyblade and Brazier 2002). The AR is located in the East African Rift System (EARS). This rift system has sections superimposed on large shear zones and sutures in the Proterozoic mobile belts. These belts were reactivated during the Cretaceous period. The uplifting influenced large river systems such as the Nile, Congo, and

Zambezi. The result was a multifaceted landscape breakup. Eruptions in the AR took place about

12.6 mya, with the subsequent formation of Lake Tanganyika. The Virunga volcanic chain has greatly affected the landscape. Virunga’s lavas created a natural dam that formed Lake Kivu.

Virunga’s lava flows created small lateral valleys, causing multiple lakes to form. The consecutive up-thrusting set off intense erosion, clearing off Cretaceous surfaces and transforming hydrological networks (Vande weghe, 2004). Combined with climatic shifts and development of ecological corridors, the stage was set for Africa’s unique fauna (Roberts et al.,

2012).

10 The modern-day AR is adjacent to the Great Rift Valley, which has experienced a unique combination of climate change, geological activity, and environmental history (Malhi et al.,

2013; Plana, 2004). The deep tectonic basins of the valley are 32–64 kilometers wide. Still in the process of dividing, the Somali and Nubian Plates form this area. Habitats include the Rwenzori

Mountains (5100 m) downward through alpine moorland (3400–4500 m), Senecio and Lobelia vegetation (3100–3600 m), heather (3000–3500 m), raised bogs (3000–3600 m), bamboo

(2500–3000 m), montane forest (1500–2500 m), lowland rainforest (600–1500), savanna woodland (600–2500 m), and savanna grassland (600–2500 m). Furthermore, species richness in the AR is among the highest in continental Africa. The AR has more than 50% of Africa’s birds,

39% of its mammals, 19% of its amphibians, 14% of its reptiles, and 14% of its plants (Plumptre et al., 2003, 2007).

The AR contains more endemic species of vertebrates than any other region in continental Africa. The AR is considered a biodiversity hotspot, holding an extensive quantity of organisms, yet much of them remain poorly studied (Burgess et al., 2007; Myers et al., 2000;

Plumptre et al., 2003, 2007). The rift includes large numbers of endemic and threatened vertebrate, invertebrate, and plant species. Savanna parks in the AR once contained the highest biomass of large mammals recorded on earth in the 1960s (Cornet d’Elzius, 1996). Although major decreases in fauna have occurred because of war and poaching, species are still present and could recover, if properly managed (Plumptre et al., 2003, 2007). Furthermore, human population density, especially in Burundi and Rwanda, are among the highest in Africa. In many parts of these countries, human population density may reach up to 800 inhabitants/km2. Some highland forests have been completely destroyed. Virunga and Kahuzi-Biega National Parks are among the areas threatened by encroachment for land and human settlement. Agriculture is

11 another growing concern, resulting in fragmentation in the AR ecoregions. Hunting, poaching, and firewood collection in and adjacent to protected areas is also a serious problem. Lastly, wars and political unrest create problems for proper management. Recent wars in Rwanda, Burundi, and DRC have created further instability. These wars have also caused spillover of refugees to

Uganda and Tanzania. These wars prevent effective conservation (Kusamba and Behangana,

2013). Afrotropical amphibians are among the species negatively affected by these human activities.

The AR currently contains 175 species of amphibians, but as surveys, DNA analyses, and fieldwork progresses, new lineages are being identified (e.g., Portillo and Greenbaum, 2014 a,b).

Presently, this area consists of 45 endemic amphibians (25.7%) as well as 11 vulnerable and two endangered species (Kusamba and Behangana, 2013). In the study of Behangana et al. (2009), species richness was analyzed in the AR. Amphibians were sampled from 37 sites and 15 protected areas in the AR. The study showed species richness in AR amphibians and how they are influenced significantly by two environmental variables—percent canopy cover and altitude.

The data indicated closed forests (i.e., mature and primary forests) that are vital to the maintenance of ecologically and diversified amphibian biota in the entire AR in Central Africa.

Additionally, disturbances correlated with human activities showed a negative impact on amphibian abundance, providing evidence that microhabitat alterations may be detrimental for amphibians. Human developments continue to put pressure on these ecosystems. Although some conservation activities have been implemented, AR biodiversity is expected to decline

(Kashambuzi and Mugisha, 2003).

The global importance for biodiversity conservation in the AR is critical because of the region’s high levels of endemism (Brooks et al., 2006; Myers et al., 2000). The need for the AR

12 to have a strategic framework in order to deal with the change in socioeconomic, political, and environmental conditions needs to be addressed. A regional approach for protection of critical terrestrial and freshwater habitats, combined with consistent representative data to monitor biota, should be of the utmost importance (Kusamba and Behangana, 2013; Mace, 2004). Therefore, it is important now, more than ever, to identify cryptic lineages and conserve biodiversity in the

AR.

The high levels of endemism and species richness seen in the tropics can be at least partially explained with the Pleistocene forest refuge hypothesis. Edward Forbes first proposed this idea in 1846. The hypothesis was originally applied only to species living in temperate regions, but the Pleistocene also caused dramatic changes in vegetation distribution in the tropics

(Mayr and O’ Hara, 1986). The Pleistocene refuge hypothesis proposes that wide-ranging ancestral taxa were isolated into forest refugia during the last glacial maximum. That isolation provided them with the opportunity to speciate. Many authors have used the Pleistocene refuge hypothesis to account for discontinuities in animal and plant distributions (Booth, 1954; Morea,

1952; Selander et al., 1969). Colder Pleistocene climates reduced rainfall, thereby modifying landscapes of lowland tropical rain forests, causing islands of mosaic forests and savannas, which were surrounded by stretches of savanna. According to Plana (2004), rainforests are of varying age, and it is the contraction and expansion of forests throughout history, in response to global climatic fluctuations, that have driven their composition. Over the past two million years, ice-age cycles have caused dramatic shifts in Africa’s climate (Gasse, 2000; Burgess et al.,

2007). Glaciers in the northern hemisphere during the last ice age occurred around 18,000–

20,000 years ago. During that time, a large portion of Earth’s surface was covered (30%) in ice, compared to only 10% today. This resulted in the lowering of temperatures by 10–20° C in

13 temperate regions and 2–5° C in the tropics (Tallis, 1991). Both current and past climate changes are a useful tool in order to explain biogeographic patterns of African amphibian species, including Phrynobatrachus.

1.7 Research objectives and questions

The objective of this study is to investigate amphibian systematics in understudied environments. The diversity of Phrynobatrachus natalensis will be explored using phylogeographic methods. By conducting research in such areas, we will be contributing to the much-needed inventory of life on our planet, and to conserving our planet’s important biodiversity hotspots. The global importance of the AR is evident because of its endangered, rare, and endemic species of plants and (Greenbaum et al., 2011; Kusamba and

Behangana, 2013). The questions and objectives I will address are as follows:

1) Does current taxonomy represent true species richness of Phrynobatrachus natalensis in the

AR?

2) Determine the extent of genetic variation occurring in the Albertine Rift populations of P. natalensis, and identify lineages representing cryptic species.

3) Propose a phylogeny illustrating evolutionary relationships of P. natalensis from the Albertine

Rift with available comparative data form GenBank and tissue samples from other sites in sub-

Saharan Africa.

14 CHAPTER 2: MATERIAL AND METHODS

2.1 Specimen acquisition

My advisor Dr. Eli Greenbaum and Daniel F. Hughes (UTEP Ph.D. student) collected specimens included in this study from the AR in DRC, Uganda, and Burundi. Specimens in this study represent eight field seasons of collection. Field seasons 2007–2009 were during the dry season, and 2010–2012 were conducted in the wet season. The field season in 2014 was conducted in both the dry and wet seasons. The last season of collection (2015) was conducted in the dry season. Additional samples were obtained from colleagues and other museum collections, and previously published GenBank data were also included. Additional genetic data is also included from samples originating from Côte d’ Ivoire, Ethiopia, Ghana, Mozambique,

Namibia, South Africa, and Tanzania. The following taxa were used as outgroups for phylogenetic analyses: Amietia desaegeri, Amietia sp. 3 (sensu Larson et al., 2016),

Phrynobatrachus acridoides, P. asper, P. bullans, P. francisci, P. graueri, and Ptychadena cf. nilotica. Collection localities of samples used in the genetic analyses are shown in Fig 1.4.

2.2 DNA extraction

The Qiagen DNeasy Blood and Tissue Kit (Valencia, CA, USA) was used to extract

DNA from tissue samples. Small pieces of tissue, about 25 mg, were cut, soaked in deionized water, and placed in a refrigerator at 0.2C for two hours (h), allowing the ethanol to diffuse from the tissue. The samples were then dried at 0.2C for two h. Once the two-hour drying period was completed, 180 ul of Buffer ATL and 20 ul of Proteinase K were added to the samples. Samples were then heated for 30 minutes (min) at 55C. After the 30 min incubation period, samples were removed, vortexed for 20 seconds (s), and then left heating for one h at

55C. Next, 200 ul of Buffer AL was added to each sample. Samples were vortexed for 20 s and

15 heated for 10 min at 70C. Following the 10 min incubation, 200 ul of 100% ethanol was added, and each sample was vortexed for 20 s. The resulting solution was placed into spin columns and centrifuged at 8000 revolutions per min (rpm) for one min. Liquid remaining in the spin columns was discarded, 200 ul of Buffer AE was added and allowed to incubate at room temperature for one min, and samples were then centrifuged at 8000 rpm for one min. The remaining genomic

DNA solution was saved for polymerase chain reaction (PCR). The previous step was repeated to obtain a total of 500 ul of genomic DNA for PCR.

2.3 DNA amplification, sequencing and alignment

One mitochondrial (16S) and one nuclear (RAG1) gene were used in this study. Primers for this gene are specified in Table 2.3.1. Amplifications were completed using a negative control to ensure no contaminated DNA was amplified. Amplifications were completed using a denaturing temperature of 95C, annealing at 50C, and extension at 72C for 32 cycles

(mitochondrial gene) or 34 cycles (nuclear gene). The PCR products were visualized with a 1.5% agarose gel stained with SYBRsafe gel stain (Invitrogen, Carlsbad, CA, USA). Products from

PCR were purified with AMPure XP magnetic bead solution with the manufacturer’s protocol

(Beckman Coulter, Danvers, MA, USA). Sequencing of forward and reverse strands of amplified

PCR products were done on an ABI 3700x1 capillary DNA sequencer at the UTEP Border

Biomedical Research Center (BBRC) Genomics Analysis Core Facility.

16

Fig 1.4 Map of sub-Saharan Africa showing the disjunct distribution of Phrynobatrachus natalensis. Sampled localities for the genetic analyses are shown with circles. Circle colors correspond to the clades in Fig. 3.1.1. Map was modified from Sarye (2013). A black polygon indicates the type locality of Phrynobatrachus natalensis (KwaZulu-Natal, South Africa).

17 Chromatograph data were interpreted using the program SeqMan (Swindell and Plasterer,

1997). Sequences were aligned using the program MEGALIGN (DNASTAR, Madison, WI,

USA) with the ClustalW algorithm and adjusted by eye in MacClade v4.08 (Maddison and

Maddison, 2000). No hypervariable regions from the 16S gene data set were removed from the

final analysis.

Table 2.3.1 Primers used for sequencing genes.

Primer Name Primer Sequence Approximate Primer Source

Fragment Length

16Sa 5 –CGCCTGTTTATCAAAAACAT-3 600 bp Palumbi et al. (1991)

Palumbi et al. (1991) 16Sb 5-CCGGTCTGAACTCAGATCACGT-3 600 bp

RAG1MartF1 5-AGCTGCAGYCARTAYCAYAARATGTA-3 800 bp (Chiari et al., 2004;

Pramuk et al., 2008)

RAG1AmpR1 5-AACTCAGCTGCAATTKCCAATRTCA-3 800 bp (Chiari et al., 2004;

Pramuk et al., 2008)

For the 16S ribosomal gene, I removed a hypervariable region consisting of 13 bp,

resulting in a final size of 555 bp. The protein-coding nuclear gene RAG1 was checked for

accuracy by translation to amino acids in MacClade.

18 2.4 Phylogenetic analyses

One mitochondrial DNA gene (16S = 555) and one nuclear gene (RAG1 = 774 bp) were analyzed for 61 representatives of P. natalensis. Sequence data were analyzed with maximum likelihood (ML) criteria in RAxML v7.2.6 (Stamatakis, 2006) and Bayesian inference (BI) criteria in MrBayes v3.1.2 (Ronquist and Huelsenbeck, 2003). The ML anlaysis was conducted using the GTRGAMMA model in RAxML v7.2.6 (Stamatakis, 2006). Maximum-likelihood analyses were initiated with random starting trees, and utilized the rapid-hill climbing algorithm.

Support values for clades inferred by ML analyses were assessed with the rapid bootstrap algorithm with 1,000 replicates (Stamatakis et al., 2008). The program PartitionFinder v.1.1.1

(Lanfear et al., 2014) was used to find the model of evolution that best fit the data for subsequent

BI analyses. Bayesian inference (BI) analyses were conducted in MrBayes v.3.1.2 (Huelsenbeck and Ronquist, 2001; Ronquist and Huelsenbeck, 2003). MrBayes v. 3.2.3 was utilized to conduct BI analyses on the CIPRES Science Gateway v. 3.3 web portal (Miller et al., 2010). Our model included four data partitions: a single partition for 16S and independent partitions for each codon position of the protein-coding gene RAG1. Concatenated data sets were partitioned identically for ML and BI analyses. Bayesian analyses were conducted with random starting trees, run for 20,000,000 generations, and Markov chains were sampled every 1000 generations; 25% of the trees were discarded as ‘‘burn in.’’ Phylogenies were visualized using

FigTreev1.2.3 (Rambaut and Drummond, 2010).

2.5 Analyses of genetic distance

Levels of divergence between haplotypes were inferred using uncorrected p-distances in

MEGA v 7.0 (Kumar et al., in press). Specimens were grouped according to the well-supported

19 clades recovered in the phylogenetic analyses. Two separate comparisons were made: one from the combined mitochondrial dataset (16S) and the other from the nuclear (RAG1) dataset.

2.6 Morphological analyses

Specimens collected for this study were photographed in life, sampled for tissues (95% ethanol), fixed in 10% buffered formalin, and transferred to 70% ethanol for long-term storage in the herpetology collection at the University of Texas at El Paso Biodiversity Collections (UTEP).

Specimens examined for this study were collected from the AR, and additional specimens were obtained from the Field Museum of Natural History, Chicago (FMNH), Museu Tridentino di

Scienze Naturali, Trento (MTSN), and California Academy of Sciences, San Francisco (CAS).

All measurements were taken with digital calipers under a dissecting microscope and rounded to the nearest 0.1 mm. All body measurements were taken on the right side of the body.

Morphological data consisted of 12 mensural and 15 meristic chararacters: snout–vent length (SVL); head width (HW) measured at the widest point of head; head length (HL)  measured at angle of jaw, from posterior edge of mandible to tip of snout; interorbital distance measured by the minimum length between upper eyelids perpendicular to the body axis (IO)

(Matsui, 1984); eye diameter (ED); tympanum diameter (TD); eye–nostril distance (EN); nostril– nostril distance (NN); tibia length (TL); foot length (FL)  measured from proximal edge of inner metatarsal tubercle to tip of longest toe; and hand length (HaL). Key morphological features for P. natalensis are as follows: throat color; nuptial pads (males); gular folds (males); tympanum (indistinct, scarcely visible, or distinct); tympanum pigment (translucent, opaque, or stippled); annulus; bands or stripes behind the thigh; bands on upper and lower jaw; dorsal skin tubercles; minute spinules; dark bands on dorsal surface of limbs; dorsal bands; dorsal spots;

20 ventral spots or blotches; spots or pattern on the lateral surfaces; manual digit tips; pedal digit tips; number of phalanges of longest toe free of webbing; tarsal ridge between midtarsal tubercle/inner metatarsal tubercle; and hand and foot webbing formulas, which were assessed by counting the number of phalanges (recognizable by subarticular tubercles) free of webbing

(Glaw and Vences, 1994).

Statistical comparisons of selected measurements were conducted with paired t-tests. In order to eliminate the effect of size, analyses of covariance (ANCOVA) were conducted with snout–vent length as the covariate (Packard and Boardman, 1999). Principal components analyses (PCA) were conducted in Minitab 15 (Minitab ® Statistical Software, State College,

PA, USA), which was used to identify patterns of variation in the data. All the analyses used the covariance matrix. Details of localities of samples included in the molecular analyses and a list of examined specimens are provided in Appendices I–II, respectively.

21 CHAPTER 3: RESULTS

3.1 Phylogenetic analyses

Phylogenetic reconstruction utilizing the mtDNA (16S = 555 bp) and nuclear gene

(RAG1 = 774 bp) for P. natalensis showed multiple lineages from different biogeographic regions in Africa (Figure 3.1.1). The total alignment for the dataset included 1329 bp.

Nucleotide substitution models selected by PartitionFinder v.1.1.1 and alternative models implemented in MrBayes are presented in Table 3.1.1. Topology for the ML and BI analyses were identical, with similar, strong support values. The ML analysis likelihood score was –

8934.501133. There is strong support for a monophyletic group that includes all

Phrynobatrachus.

Table 3.1.1 Model parameters specified by PartitionFinder and alternate models implemented in MrBayes.

Gene and codon position Model selected by Model implemented in PartitionFinder MrBayes 16S GTR+ G GTR RAG1 codon 1 TIM + I + G HKY RAG1 codon 2 HKY + I HKY RAG1 codon 3 HKY + G HKY

Basal relationships within the Phrynobatrachus clade are well supported. Intrageneric relationships were also well supported, and nine cryptic lineages were identified from Angola,

Burundi, Côte d’Ivoire, DRC, Ethiopia, Ghana, Mozambique, Namibia, Tanzania, Rwanda, and

Uganda. Populations represented from the aforementioned locations are likely to be new species, although available synonyms for P. natalensis should be considered.

22 Several strongly supported clades, (> 0.95 posterior probability and > 70 bootstrap support) are (Fig 3.1.1) included in the following monophyletic clades: P. cf. natalensis 1 from

Namibia and Angola; P. cf. natalensis 2 from South Kivu DRC; P. cf. natalensis 3 from

Katanga, DRC; P. cf. natalensis 4 from Mozambique and Tanzania; P. cf. natalensis 5 from

Tanzania; P. cf. natalensis 6 from Burundi, Rwanda, and Uganda; P. cf. natalensis 7 from Ghana and Côte d’Ivoire; P. cf. natalensis 8 from DRC and Uganda, and P. cf. natalensis 9 from

Ethiopia. The evidence presented in this phylogeny supports the need for taxonomic revision in the P. natalensis species complex (Rödel, 2000; Zimkus, 2010).

3.2 Analyses of genetic distance In order to examine genetic diversity among clades of P. natalensis in the Albertine Rift, specimens were assigned to groups based on well-supported clades indicated in Fig. 3.1.1.

Uncorrected p-distances (Tables 3.2.1, 3.2.2) within and among clades of P. natalensis are provided for the 16S and RAG1 datasets. Sequence divergences for the 16S gene (Table 3.2.1) had high intraspecific sequence divergences (4–13%). Low genetic divergence was seen in

RAG1 (0–1%) (Table 3.2.2). The specimen belonging to P. cf. natalensis 1 from Namibia had the highest divergence seen among the nine analyzed clades, but this is not surprising because it was recovered in a clade with P. acridoides and P. bullans, whereas all other P. natalensis samples were recovered in a monophyletic clade together. The relatively high intraclade divergences among the latter P. natalensis populations suggest these samples represent a species complex.

3.3 Albertine Rift P. natalensis morphology

Morphological and statistical analyses were conducted for multiple specimens from the

AR. A total of 13 adult males were used for the statistical analyses. Mensural data for adult

23 specimens of P. natalensis are shown in Table 3.3.1. The data were categorized according to clades from Figure 3.1.1. For the analysis with raw morphometric data regressed against SVL, the first two principal component axes explained 64% of the total variance (Table 3.3.2 and

Figure 3.3.1). The first PC axis loaded positively for variable FL (Foot length) and F4 (Finger

4); specimens with larger values on the first axis were associated with longer legs, relative to body size. The second PC loaded positively for variable T3 and T4 (toes) relative to body size.

No clear pattern is evident in the data, and the analysis was limited by small sample sizes. In order to make an accurate morphological assessment, more samples will be needed from represented clades (Figure 3.1.1) and additional lineages represented in the phylogeny.

24

Figure 3.1.1 MrBayes phylogenetic tree (16S and RAG1 genes) of Phrynobatrachus natalensis samples. Support values represent posterior probability (> 0.95)/bootstrap support (> 70). Colors in the phylogenetic tree correspond to Figure 1.4.

25

Figure 3.3.1 Score plots of first and second principal components obtained from a PCA with SVL as the covariate for adult male specimens of P. cf. natalensis clades from the Albertine Rift.

26 Table 3.2.1 Uncorrected p-distances (16S) among selected individuals from the P. natalensis phylogeny.

27 Table 3.2.2 Uncorrected p-distances (RAG1) among selected individuals from the P. natalensis phylogeny.

28 Table 3.3.1 Morphometric data of adult male voucher specimens of P. cf. natalensis from selected clades from Fig. 3.1.1 (in mm). Data are presented as mean + one standard deviation, followed by range.

P. cf. P. cf. P. cf. P. cf. natalensis 2 natalensis 3 natalensis 6 natalensis 8 (N = 1) (N = 5) (N = 1) (N = 6)

SVL 28.49 29.84 27.5 27.5 ± 2.5 ± 1.5 29.3–33.3 26.2–28.5 HW 8.7 8.89 9.6 8.7 ± 0.6 ± 0.61 7.6–9.4 7.2–9.4 HL 7.76 8.3 6.14 6.3 ± 0.87 ± 1.1 7.6–9.4 4.5–7.8 SL 3.3 3.7 3.19 3.3 ± 0.54 ± 0.45 3.8–4.3 2.4–4.1 IO 2.36 3.2 3.1 2.5 ± 0.76 ± 0.62 2.3–4.5 1.1–3.4 HE 3.57 3.4 2.4 2.8 ± 0.31 ± 2.3 2.3–3.8 2.0–3.4 TD 2.42 2.63 2 2 ± 0.28 ± 0.38 2.0 1.4–2.7 EN 2.6 2.4 2.4 2.3 ± 0.36 ± 0.41 2.6–3.5 1.4–3.6 IN 2.48 2.7 1.8 2.1 ± 0.36 ± 0.53 2.7–3.1 1.4–2.0 HL 5.44 6.3 4.5 5.2 ± 0.38 ± 0.33 5.3–7.4 4.4–5.7 RU 6.21 6.7 5.2 5.5 ± 0.66 ± 0.95 5.2–7.9 4.1–7.0 FL 15.18 12.6 12.6 12.5 ± 0.82 ± 1.4 11.5–13.4 11.4–13.6 TL 15.55 14.7 13.7 78.3 ± 0.63 ± 0.91 13.3–15.6 12.1–14.4 FL 15.36 15.3 13.4 13.4 ± 3.2 ± 1.1 13.4–21.6 11.6–14.3 F1 3.17 3.9 3.2 4 ± 0.53 ± 0.85 3.3–4.4 2.2–5.4 F2 2.86 2.8 2.3 2.2 ± 0.36 ± 0.48 2.1–3.5 1.7–3.9 F3 3.94 3.5 4.5 3.6 ± 0.47 ± 0.66 2.9–4.2 2.2–4.9 F4 3.14 3.5 2.4 3.3 ± 3.1 ± 1.7 1.4–9.6 1.0–6.4 T1 2.48 2.4 2.2 2.1 ± 0.34 ± 0.36 1.1–2.6 1.3–2.8 T2 3.12 3.5 4.3 3 ± 0.43 ± 0.53 2.8–3.4 2.5–4.3 T3 6.34 6.1 6.5 5 ± 1.8 ± 0.37 3.7–8.1 4.3–5.8 T4 8.01 7 8 7.3 ± 3 ± 0.54 1.2–9.0 7.0–8.6 T5 5.73 5.3 3.6 4.5 ± 1.9 ± 0.63 2.6–8.4 3.3–5.4

29 Table 3.3.2 Principal components analysis comparing adult male specimens from clades of Albertine Rift P. cf. natalensis with raw mensural data regressed against snout–vent length. Eigenvalues, percent variance, cumulative variance, and loadings are shown for the first three principal components.

Variable PC1 PC2 PC3 HW 0.051 0.159 0.032 HL 0.214 -0.095 -0.299 SL 0.004 0.098 -0.063 IOD 0.025 0.023 0.116 HED -0.006 -0.036 -0.074 TD 0.064 -0.065 -0.016 END -0.012 0.050 -0.005 IND 0.047 0.025 -0.078 HL 0.044 -0.147 -0.120 RUL 0.071 -0.238 -0.310 FL 0.107 0.174 -0.279 TL 0.069 -0.021 0.333 FL 0.581 -0.231 -0.320 F1 0.046 -0.129 0.020 F2 0.061 0.005 -0.108 F3 -0.022 0.052 0.085 F4 0.562 -0.313 0.577 T1 0.014 -0.015 -0.168 T2 0.019 0.110 -0.010 T3 0.162 0.409 -0.253 T4 0.324 0.687 0.185 T5 0.361 0.131 -0.008 Eigenvalue 9.9828 4.0813 2.7799 Proportion 0.456 0.187 0.127 Cumulative 0.456 0.643 0.770

30 CHAPTER 4: DISCUSSION

4.1 Taxonomic status of Phrynobatrachus natalensis populations

According to the final reconstructed phylogeny (Figure 3.1.1), nine cryptic lineages were recovered—several are likely to be new species or represent scientific names that are not in current use. Phrynobatrachus natalensis has ten synonyms, mainly because of the species complex’s widespread distribution in disparate habitats.

According to the Amphibian Species of the World website (Frost, 2016), ten names are available for the cryptic lineages: Stenorhynchus natalensis Smith, 1849, described from “the country around Port Natal [= Durban],” KwaZulu-Natal, South Africa; Dicroglossus angustirostris Cope, 1862, presumed to be described from “Umvoti, Natal [= KwaZulu-Natal],”

South Africa; Arthroleptis natalensis var. irrorata Peters, 1875, described from “Accra,” Ghana;

Arthroleptis bottegi Boulenger, 1895, described from “Auta [River],” Bale Province, Ethiopia

(0.5º 05' N, 39º 08' E); Phrynobatrachus ranoides Boulenger, 1895, described from “near

Pietermaritzburg, Natal,” South Africa; Arthroleptis moorii Boulenger, 1898 described from

“Kinyamkolo, Lake Tanganyika” (= Nyamkolo, southeastern corner of Lake Tanganyika) northeastern ; Phrynobatrachus natalensis forma gracilis Andersson, 1904, described from “terminal Station Ghrab el Aish south of Kaka on the White Nile”; Phrynobatrachus maculatus FitzSimons, 1932 described from “Victoria Falls,” Zimbabwe; Arthroleptis-

Phrynobatrachus zavattarii Scortecci, 1943, described from “Caschei” (= Ciacche), Gemu-Gofa

Province, Ethiopia, 05º 00' N, 37º 33' E, ca. 1500 m; Arthroleptis-Phrynobatrachus sciangallarum Scortecci, 1943, described from “Murlè,” Gemu-Gofa Province, Ethiopia, 05º 10'

N, 36º 13' E, 400 m; and Phrynobatrachus duckeri Loveridge, 1953, described from “one of the cotton Growers Experimental Station dams at Chitala River, Nyasaland [Malawi].”

31 The results shown in the final reconstructed phylogeny (Fig 3.1.1) bolster the inferences made by previous herpetologists that P. natalensis is a complex of cryptic species (Channing,

2001; Channing and Howell, 2006; Largen, 2001; Rödel, 2000; Zimkus et al., 2010). Genomic data were recovered from specimens near the original type localities for four of the above synonyms, including Arthroleptis moorii Boulenger, 1895, Arthroleptis natalensis var. irrorata

Peters, 1875, Phrynobatrachus natalensis forma gracilis Andersson, 1904, and Arthroleptis bottegi Boulenger, 1895. Below I attempt to match clades to these four available names and identify clades that are likely to represent new species.

The problem with the original description by Andrew Smith (1849) was the lack of a detailed type locality description for P. natalensis, which was given as, “the country around Port

Natal.” However, genomic data from several South African samples near Port Natal (modern-day

Durban) were recovered in a well-supported clade in the phylogeny (Fig. 3.1.1). These specimens were from Mtunzini, Dwesa, Cwebe Wildlife Reserve, Coffee Bay, and Hluleka

Nature Reserve, South Africa, and range from 123–469 km south of Durban. Two other synonyms, Dicroglossus angustirostris (Cope, 1862) from Umvoti (50 km north of Durban) and

Phrynobatrachus ranoides Boulenger, 1895 from Pietermaritzburg (50 km west of Durban) are also likely represented by these South African samples. Thus, the latter two taxa should remain as synonyms of P. natalensis.

Phrynobatrachus cf. natalensis 1 from Namibia and Angola have no attributable to this clade, and it is likely to be a new species. One sample from Mozambique’s

Gorongosa National Park (391, Fig. 3.1.1) was embedded among samples of P. acridoides, suggesting it belongs to this species. Phrynobatrachus cf. natalensis 2 from Nyamibungu,

Mwenga, Kalundja and Luvungi, South Kivu, DRC is also likely a new species as well. Clades

32 P. cf. natalensis 3 from Pweto, Kakunko, Kabongo and Kimari, Katanga, DRC and P. cf. natalensis 4 from Niassa, Mozambique and Tanzania are both likely to be new species.

Phrynobatrachus cf. natalensis 5 includes samples near the synonym Arthroleptis moorii

Boulenger, 1895, described from Kinyamkolo, Zambia, near the southeastern corner of Lake

Tanganyika. The distance between Kinyamkolo and the nearest P. cf. natalensis 5 sample is ca.

500 km. The specimens belonging to P. cf. natalensis 6 from Gihofi and Muranvya, Burundi,

Mukungwe, Butare and Ruhengeri Rwanda, and Rukungiri, Uganda (Western region), are likely new species. Phrynobatrachus. cf. natalensis 7 is located in West Africa and includes three specimens—two from Ghana and one from Côte d’Ivoire. These specimens are likely attributable to Arthroleptis natalensis var. irrorata Peters, 1875, which was described from

Accra, Ghana. The total distance between the Kyabobo National Park samples and the type locality (Accra) is ca. 300 km. Phrynobatrachus cf. natalensis 8 from Banywani, Mambasa,

Virunga National Park and Kamango in Orientale and North Kivu, DRC, and Nagojje Beat, Pan

Upe and Nakapiripirit, Uganda (central and northern regions) are attributed to Phrynobatrachus natalensis forma gracilis Andersson, 1904 from Ghrab el Aish, south of Kaka, South on the White Nile. The White Nile runs south through South Sudan and Uganda to Lake Albert and then east to Lake Victoria. This path likely provided a dispersal path for Phrynobatrachus natalensis forma gracilis to spread. The Ethiopian (Dola-Mena, Jinka, and Doketu) clade at the base of the tree (P. cf. natalensis 9) is attributable to two possible synonyms when one considers the topography of Ethiopia. The Simien Mountain range divides Ethiopia from north to south, creating a large east-west barrier for amphibians (Largen, 2001). The samples from Dola-Mena

(A 2008203, A 2008205, and A 2008229) are likely attributable to Arthroleptis bottegi

Boulenger, 1895 from Yabelo Wildlife Sanctuary (Borena Zone of the Oromia Region), which is

33 190 km southwest of Dola-Mena. The samples from Jinka (750-11) and Doketu (694-11) are attributable to Arthroleptis-Phrynobatrachus zavattarii Scortecci, 1943 from Gemu-Gofa

Ethiopia. The distance between Gemu-Gofa and Jinka is 100 km. Samples attributable to the following available names were not available for this study: Phrynobatrachus maculatus

FitzSimons 1932 from Victoria Falls, Zimbabwe; Arthroleptis-Phrynobatrachus sciangallarum

Scortecci, 1943 from Murlé, Ethiopia; and Phrynobatrachus duckeri Loveridge, 1953 from

Experimental Station dams at the Chitala River, Malawi.

Multivariate statistical morphological assessments (PCA) did not show distinct clusters attributable to genetic clades, and sample sizes were low. The Phrynobatrachus genus has been a long-standing source of confusion to systematists, because of similar morphological features along with intra-and interspecific variability. Drawing the species boundary line has been difficult. On one hand, some suggest recognizing a species when lineages develop reproductive barriers, whereas others prefer drawing the line once genetic divergence is evident (de Queiroz,

1998, 2005, 2007). Therefore, in this study we utilized the General Lineage Concept (GLC). This concept accepts a species when the lineage evolves separately from others with its own evolutionary roles and tendencies (de Queiroz, 1998; Hey et al., 2003). Fouquet et al. (2007) mentions that poorly known species boundaries need to be examined with various lines of evidence (e.g., morphology and molecular data) in order to reveal true species richness. They proposed a 0.03% genetic difference between taxa for candidate species of Neotropical anurans

(16S uncorrected p distances). Implementing multiple lines of evidence (molecular, morphological, ecological, and behavioral) is needed in order to diagnose species. Various methods are used, because single lines of evidence can be subjective and vague (Greenbaum et al., 2013; Padial et al., 2009).

34 Results from my phylogenetic and morphometric analyses indicated unresolved taxonomic issues that require new species descriptions and additional sampling in the AR and other poorly known areas of sub-Saharan Africa where P. natalensis may occur. Multiple areas of the AR and other areas in Africa seem to have unique lineages, which are likely endemic and are genetically distinct. According to the reconstructed phylogeny (Fig. 3.1.1) five new species are in need of new descriptions and four current synonyms of P. natalensis need to be elevated to full species status. Genomic data for the putative P. natalensis type locality (Durban, South

Africa) is represented in this study (Mtunzini, Dwesa Cwebe Wildlife Reserve, Coffee Bay, and

Hluleka Nature Reserve in South Africa).

The vast genetic diversity represented in this study is not surprising. The AR is one of the most endemic-rich regions in sub-Saharan Africa, despite being one of the most poorly documented. The AR holds six mountain ranges, five major lakes, and a total of five different ecoregions (Plumptre et al., 2007) This dynamic environment allows for evolutionary diversification to happen, especially in amphibians. As herpetologists conduct research in remote locations, vast amounts of new species descriptions are published. In 2009 only 75 species of the

Phrynobatrachus genus were recognized, but as of 2016, there were 89 species in the genus. The current study of P. natalensis will likely involve at least one new species description as well.

Further investigations and larger sampling will be needed from all nine unique lineages to help clarify the unsettled taxonomy within the P. natalensis complex.

4.2 Conservation The rapid destruction of the world’s biosphere is happening at an unprecedented scale.

The loss of habitat and extinction rates in “hotspots” may be summed up into four general principles (Brooks et al., 2002). First, most species susceptible to extinction have small range

35 sizes, increasing their probability of extinction by chance alone (Kunin and Gaston, 1997).

Second, species that have a small range tend to be scarce within that range, thereby increasing extinction on two counts (Brown, 1984). Third, quite often, small range endangered species co- occur in the same location, further intensifying land management needs (Fjeldsaå and Lovett,

1997). Last, centers of endemism are disproportionately threatened by human activity (Cincottoa et al., 2000).

There are 35 biogeographically distinctive hotspots worldwide. Given these hotspots around the world hold 44% of the world’s plants and 35% of terrestrial vertebrates in just 1.4% of the land area, it is evident a conservation strategy is needed (Brooks et al., 2002). This loss of habitat is most evident in the African continent (Barnes, 1990). Africa is best known for the enormous diversity and richness of its wildlife. It has a great variety of large ungulates, or hoofed mammals, freshwater fish, amphibians, and reptiles, just to name a few (Du Preez et al.,

2009; Vitt and Caldwell, 2013).

Covering about one-fifth of the total land surface of the Earth, Africa is the second largest continent (after Asia). Africa’s total land area is approximately 11,724,000 square miles

(30,365,000 square km) (Burgess et al., 2004). Most of the biodiversity of Africa is located in the

Afrotropical realm (south of the Sahara Desert). According to Olson and Dinerstein (1998), the

Afrotropical realm is one of Earth’s most biologically valuable ecoregions. In the Afrotropical realm, 47% of its area is considered critical or endangered, 29% vulnerable, and 24% relatively stable or intact. The Afrotropical region itself is made up of 13 distinct terrestrial ecoregions with

10 being considered critical or endangered. One of the most critical areas for conservation is the

AR located in the western branch of the East African Rift.

36 The AR is one of Africa’s most species rich and endemic rich regions, despite being one of its most poorly documented. The AR is at the center of Africa’s montane rainforest circle.

Reaching from the northern end of Lake Albert to the southern end of Lake Tanganyika, the rift straddles the borders of five different nations: DRC (70% of the AR), Uganda (20% of the AR),

Rwanda (6% of the AR), Burundi (3% of the AR), and Tanzania (1% of the AR) (Burgess et al.,

2004; Dowsett, 1986; Kingdon, 1989). The AR has been mentioned in multiple studies as the most species rich area in Africa. Over the past decades, the AR has shown declines in all fauna, with no significant recent reductions in rate. On the contrary, increases were seen in invasive alien species, nitrogen pollution, overexploitation, and climate change impacts (Stuart et al.,

2010). One of the contributors to this decline is human population density. The AR can contain up to 1000 people per square kilometer in some areas (Plumptre et al., 2007). As Blamford et al.

(2001) revealed across Africa, species richness combined with dense human settlement is associated with high threats to biodiversity. As aforementioned, amphibian diversity and species richness has suffered continuously, with no end in sight.

Amphibians are declining at an exponential rate worldwide. A recent study conducted by

Alroy (2015) applied conservative Bayesian methods to estimate recent amphibian and squamate extinction rates in nine crucial tropical and subtropical regions. The data provided clear evidence of extinction pulses of frogs and suggested that about 200 frog extinctions have occurred, and hundreds more will be lost over the next century. Furthermore, surveys (166 frogs in 45 localities) in the AR for Bd in frogs found that 34.9% of tested animals were positive

(Greenbaum et al., 2015).

Socioeconomic factors in areas of conservation concern must be included in management plans in order to be allocated efficiently (Bode et al., 2008). When focused on conservation, one

37 must consider income and other benefits from land use, investment, and development opportunities. This is especially true for developing areas such as the AR. Developing a management plan is particularly hard in developing countries, where livelihood depends strictly on land use (Norton-Griffiths and Southey, 1995). A study conducted by Chiozza et al. (2010) in

Uganda focused on the cost of conserving amphibians and mammals. Their goal was to find how to conserve species at a minimum cost. They did this by using yearly net profit from agricultural activities along with the estimated area potentially occupied by species. Results indicated that in order to reach conservation targets for amphibians and mammals, 38% of the country would need to be conserved. With a rapidly increasing population, this is unrealistic.

4.3 Conclusions and future directions

The results of this study suggest that the diversity of P. natalensis has been severely underestimated. We recovered nine cryptic lineages from the AR. Clades that are attributable to synonyms of P. natalensis should be recognized as full species, and new species should be described. Additional research efforts should focus on the biogeographic patterns, reproductive isolation, acoustic calls, and reproductive ecology to determine species boundaries. An increase in sample sizes would help resolve geographic patterns and provide better resolution between clades. For example, additional sampling is still needed for populations in Burundi, DRC,

Malawi, Mozambique, Rwanda, Tanzania, Uganda, Zambia, and Zimbabwe for a more robust taxonomic conclusion for this complex. Furthermore, it is imperative to sample more African sites, especially in the AR, to get a better understanding of African biodiversity.

38 REFERENCES

AmphibiaWeb: Information on amphibian biology and conservation. [web application]. 2016. Berkeley, California: AmphibiaWeb. Available at: http://amphibiaweb.org/. (Accessed: Mar 14, 2016).

Barnes, R. F. W. (1990). Deforestation trends in tropical Africa. African Journal of Ecology 28: 161–173.

Berger, L., R. Speare, and A. Hyatt. (1999). Chytrid fungi and amphibian declines: Overview, implications and future directions. Pp. 23–33. In A. Campbell (Ed.), Declines and Disappearances of Australian Frogs. Environment Australia, Canberra, Australia.

Behangana, M., P. M. Kasoma, and L. Luiselli. (2009). Ecological correlates of species richness and population abundance patterns in the amphibian communities from the Albertine Rift, East Africa. Biodiversity and Conservation 18: 2855–2873.

Booth, A. H. (1954). The Dahomey Gap and the mammalian fauna of the West African forests. Revue de zoologie et de botanique africaines 1: 305–314.

Bossuyt, F., and M. C. Milinkovitch. (2000). Convergent adaptive radiations in Madagascan and Asian ranid frogs reveal covariation between larval and adult traits. Proceedings of the National Academy of Sciences, USA 97: 6585–6590.

Boyle, D. G., D. B. Boyle, V. Olsen, J. A. T. Morgan, and A.D. Hyatt. (2004). Rapid quantitative detection of chytridiomycosis (Batrachochytrium dendrobatidis) in amphibian samples using real-time Taqman PCR assay. Diseases of Aquatic Organisms 60: 141–148.

Bode, M., K. A. Wilson, T. M. Brooks, W. R. Turner, R. A. Mittermeier, M. F. McBride, and H. P. Possingham. (2008). Cost-effective global conservation spending is robust to taxonomic group. Proceedings of the National Academy of Sciences, USA 105: 6498– 6501.

Brandley, M. C., A. Schmitz, and T. W. Reeder. (2005). Partitioned Bayesian analyses, partition choice, and the phylogenetic relationships of scincid lizards. Systematic Biology 54: 373– 390.

Broadley, D. G., and F. P. D. Cotterill. (2004). The reptiles of southeast Katanga, an overlooked ‘hot spot.’ African Journal of Herpetology 53: 35–61.

Brooks, T. M., R. A. Mittermeier, G. A. da Fonseca, J. Gerlach, M. Hoffmann, J. F. Lamoreux, and A. S. Rodrigues. (2006). Global biodiversity conservation priorities. Science 313: 58– 61.

39 Brooks, T. M., R.A. Mittermeier, C.G. Mittermeier, G.A. Da Fonseca, A.B. Rylands, W. R. Konstant, and C. Hilton‐Taylor. (2002). Habitat loss and extinction in the hotspots of biodiversity. Conservation Biology 16: 909–923.

Brown, J. H. (1984). On the relationship between abundance and distribution of species. American Naturalist 24: 255–279.

Burrowes, P. A., R. L. Joglar, and D. E. Green. (2004). Potential causes for amphibian declines in Puerto Rico. Herpetologica 60: 141–154.

Burgess, N., J. D. Hales, E. Underwood, E. Dinerstein, D. Olson, I. Itoua, and E. Dinerstein. (2004). Terrestrial Ecoregions of Africa and Madagascar: A Conservation Assessment. Island Press, Washington, D.C.

Burgess, N. D., T. M. Butynski, N. J. Cordeiro, N. H. Doggart, J. Fjeldså, K. M. Howell, and S. N. Stuart. (2007). The biological importance of the Eastern Arc Mountains of Tanzania and . Biological Conservation 134: 209–231.

Crosskey, R. W., B.H. Cogan, P. Freeman, A.C. Pont, K.G.V. Smith, and H. Oldroyd. (1980). Catalogue of the Diptera of the Afrotropical Region. British Museum (Natural History), London, United Kingdom.

Cincotta, R. P., J. Wisnewski, and R. Engelman. (2000). Human population in the biodiversity hotspots. Nature 404: 990–992.

Chiozza, F., L. Boitani, and C. Rondinini. (2010). The opportunity cost of conserving amphibians and mammals in Uganda. Brazilian Journal for Nature Conservation 8: 177– 183. Channing, A., M. O. Rödel, and J. Channing. (2012). Tadpoles of Africa: The Biology and Identification of All Known Tadpoles in sub-Saharan Africa. Chimaira, Frankfurt am Main, Germany.

Channing, A., and K. M. Howell. (2006). Amphibians of East Africa. Cornell University Press, Ithaca, New York.

Channing, A. (2001). Amphibians of Central and Southern Africa. Cornell University Press, Ithaca, New York.

Chiari, Y., M. Vences, D. R. Vieites, F. Rabemananjara, P. Bora, O. Ramilijaona Ravoahangimalala, and A. Meyer. (2004). New evidence for parallel evolution of colour patterns in Malagasy poison frogs (Mantella). Molecular Ecology 13: 3763–3774.

Collins, J. P., and T. Halliday. (2005). Forecasting changes in amphibian biodiversity: Aiming at a moving target. Philosophical Transactions of the Royal Society B: Biological Sciences 360: 309–314.

40 Cornet d’Elzius, C. (1996). Ecologie, Structure et evolution des Populations des Grands Mammiferes du Secteur Central du Parc National des Virunga (Parc National Albert) Zaire (Congo Belge). Fondation pour Favoriser les Recherches Scientifiques en Afrique, Brussels, Belgium. de Queiroz, K. (1998). The general lineage concept of species, species criteria, and the process of speciation: A conceptual unification and terminological recommendations. Pp. 57–75. In D. J. Howard and S. H. Berlocher (Eds.), Endless Forms: Species and Speciation. Oxford University Press, New York, NY. de Queiroz, K. (2005). Different species problems and their resolution. BioEssays 27: 1263– 1269. de Queiroz, K. (2007). Species concepts and species delimitation. Systematic Biology 56: 879– 886.

Dowsett, R. J., (1986). Origins of the high-altitude avifaunas of tropical Africa. Pp. 557–585. In F. Vuilleumier and M. Monasterio (Eds.), High Altitude Tropical Biogeography. Oxford University Press, New York and Oxford.

Dubois, A. (2004). Developmental pathway, speciation and supraspecific taxonomy in amphibians: Why are there so many frog species in Sri Lanka? Alytes 22: 19–37.

Du Preez, L. H., V. Carruthers, and M. Burger. (2009). A Complete Guide to the Frogs of Southern Africa. Struik Nature, Cape Town, South Africa.

Farrer, R. A., L. A. Weinert, J. Bielby, T. W. Garner, F. Balloux, F. Clare, and M. C. Fisher. (2011). Multiple emergences of genetically diverse amphibian-infecting chytrids include a globalized hypervirulent recombinant lineage. Proceedings of the National Academy of Sciences, USA 108: 18732–18736.

Fouquet, A., A. Gilles, M. Vences, C. Marty, M. Blanc, and N. J. Gemmell. (2007). Underestimation of species richness in Neotropical frogs revealed by mtDNA analyses. PLoS ONE 10: e1109.

Frost, D. R., T. Grant, J. Faivovich, R. H. Bain, A. Haas., C. F. Haddad, O. Rafael, De Sá, A. Channing, M. Wilkinson, S. C. Donnellan, C. J. Raxworthy, J. A. Campbell, B. L. Blotto, P. E. Moler, R. C. Drewes, R. A. Nussbaum, J. D. Lynch, D. M. Green, and W. C. Wheeler. (2006). The amphibian tree of life. Bulletin of the American Museum of Natural History 297: 1–370.

Frost, D. R. (2016). Amphibian Species of the World: an Online Reference. Version 6.0 (Accessed 15 April 2016) Electronic Database accessible at: http://research.amnh.org/herpetology/amphibia/ American Museum of Natural History, New York.

41 Fjeldsaå, J., and Lovett, J. C. (1997). Geographical patterns of old and young species in African forest biota: The significance of specific montane areas as evolutionary centres. Biodiversity and Conservation 6: 325–346.

Gascon, C., Collins, J. P., Moore, R. D., Church, D. R., McKay, J. E. and Mendelson, J. R. III (eds). 2007. Conservation Action Plan. IUCN/SSC Amphibian Specialist Group. Gland, Switzerland and Cambridge, UK. 64 pp. Accessed January 23, 2016 at: http://www.amphibianark.org/pdf/ACAP.pdf

Gasse, F. (2000). Hydrological changes in the African tropics since the last glacial maximum. Quaternary Science Reviews 19: 189–211.

Glaw, F., and M. Vences. (1994). A Field Guide to the Amphibians and Reptiles of Madagascar, 2nd Ed., Including Mammals and Freshwater Fish. Moos Druck, Köln, Germany.

Greenbaum, E., C. Kusamba, M. M. Aristote, and K. Reed. (2008). Amphibian chytrid fungus infections in Hyperolius (Anura: ) from eastern Democratic Republic of Congo. Herpetological Review 39: 70–73.

Greenbaum, E., C. O. Villanueva, C. Kusamba, M. M. Aristote, and W. R. Branch. (2011). A molecular phylogeny of Equatorial African Lacertidae, with the description of a new genus and species from eastern Democratic Republic of the Congo. Zoological Journal of the Linnean Society 163: 913942.

Greenbaum, E., C, Kusamba. (2012). Conservation implications following the rediscovery of four frog species from the Itombwe Natural Reserve, eastern Democratic Republic of the Congo. Herpetological Review 43: 253–259.

Greenbaum, E., U. Sinsch, E. Lehr, F. Valdez, and C. Kusamba. (2013). Phylogeography of the reed frog Hyperolius castaneus (Anura: Hyperoliidae) from the Albertine Rift of Central Africa: Implications for taxonomy, biogeography and conservation. Zootaxa 3731: 473– 494.

Greenbaum, E., J. Meece, K. D. Reed, and C. Kusamba. (2015). Extensive occurrence of the amphibian chytrid fungus in the Albertine Rift, a Central African amphibian hotspot. The Herpetological Journal 25: 91–100.

Halliday, T. R. (2008). Why amphibians are important. International Zoo Yearbook 42: 7–14.

Harper, E. B., G. J. Measey, D. A. Patrick, M. Menegon, and J. R. Vonesh. (2010). Field Guide to Amphibians of the Eastern Arc Mountains and Coastal Forests of Tanzania and Kenya. Camerapix, Nairobi, Kenya.

Hey, J., R. S. Waples, M. L. Arnold, R. K. Butlin, and R. G. Harrison. (2003). Understanding and confronting species uncertainty in biology and conservation. Trends in Ecology & Evolution 18: 597–603.

42 Huelsenbeck, J. P., and F. Ronquist. (2001). MRBAYES: Bayesian inference of phylogeny. Bioinformatics 17: 754–755.

IUCN. (2015). 2014 IUCN Red List of Threatened Species. Version 2015–4. Accessible at: http//www.iucnredlist.org.

Kusamba, C., and M. Behangana. (2013). Status of Amphibians and Reptiles in the AR and Priorities for Conservation. Pp 46–51. In S. Kanyamibwa (Ed.), Albertine Rift Conservation Status Report. Albertine Rift Conservation Series No. 1. ARCOS Network. Accessible at: http://itfc.must.ac.ug/sites/default/files/publications/Albertine_Rift_Conservation_Status_ Report(ARCSR)_Final_Draft_March2013.pdf.

Kashambuzi, R., and F. Mugisha. (2003). Exploration history and petroleum potential of Uganda. Ministry of Energy and Mineral Development, Republic of Uganda, Accessible at: www.energy.eac.int/eapc2003/contents/22b.%pdf. Accessed on: January 23, 2016.

Kingdon, J. (1990). Island Africa: The Evolution of Africa's Rare Animals and Plants. Princeton University Press, London, United Kingdom.

Klerkx, J., K. Theunissen, and D. Delvaux. (1998). Persistent fault controlled basin formation since the Proterozoic along the Western Branch of the East African Rift. Journal of African Earth Sciences 26: 347–361

Kumar, S., G. Stecher, and K. Tamura. (In press). MEGA7: Molecular Evolutionary Genetics Analysis version 7.0 for bigger datasets. Molecular Biology and Evolution.

Kunin, W.E., and K. J. Gaston. (1997). The Biology of Rarity: Causes and Consequences of Rare-Common Differences. Chapman & Hall, London, United Kingdom.

Lamotte, M., and F. Xavier. (1966). Phrynobatrachus natalensis (Smith) et Phrynobatrachus francisci (Boulenger): Deux espèces de l'Ouest africain difficiles à distinguer. Bulletin de l'Institut Fondamental d’Afrique Noire (IFAN), sér. A 28: 343–361.

Lanfear, R., B. Calcott, D. Kainer, C. Mayer, and A. Stamatakis. (2014). Selecting optimal partitioning schemes for phylogenomic datasets. BMC Evolutionary Biology 14: 82.

Largen, M. J. (2001). Catalogue of the amphibians of Ethiopia, including a key for their identification. Tropical Zoology 2: 307–402.

Larson, T. R., D. Castro, M. Behangana, and E. Greenbaum (2016). Evolutionary history of the river frog genus Amietia (Anura: Pyxicephalidae) reveals extensive diversification in Central African highlands. Molecular Phylogenetics and Evolution 99: 168–181.

Laurent, R. F. (1983). About the herpetofauna of Central African montane forest. Pp. 350–358. In A. G. J. Rhodin and K. Miyata (Eds.), Advances in Herpetology and Evolutionary

43 Biology: Essays in Honor of Ernest E. Williams. Museum of Comparative Zoology (Harvard), Cambridge, Massachusetts.

Loveridge, A. (1935). Scientific results of an expedition to rain forest regions in Eastern Africa: New reptiles and amphibians from East Africa. Bulletin of the Museum of Comparative Zoology (Harvard) 79: 1–19.

Mace, G. (2004). The role of taxonomy in species conservation. Philosophical Transactions of the Royal Society B: Biological Sciences 359: 711–719.

Malhi, Y., S. Adu-Bredu, R. A. Asare, S. L. Lewis, and P. Mayaux. (2013). African rainforests: Past, present and future. Philosophical Transactions of the Royal Society B: Biological Sciences 368: 1–10.

Maddison, D.R., and W.P. Maddison. (2000). MacClade 4: Analysis of phylogeny and character evolution. Version 4.0. Sinauer Associates, Sunderland, Massachusetts.

Matsui, M. (1984). Morphometric variation analyses and revision of the Japanese toads (Genus , Bufonidae). Contributions from the Biological Laboratory, Kyoto University 26: 209–428. Mayr, E., and R. J. O’Hara. (1986). The biogeographic evidence supporting the Pleistocene forest refuge hypothesis. Evolution 40: 55–67.

Miller, M., W. Pfeiffer, and T. Schwartz. (2010). Creating the CIPRES Science Gateway for inference of large phylogenetic trees. In Proceedings of the Gateway Computing Environments Workshop (GCE), 14 Nov. 2010, New Orleans, LA, pp. 1–8. Accessible at: https://www.phylo.org/portal2/login!input.action.

Mittermeier, R. A., W. R. Turner, F. W. Larsen, T. M. Brooks, and C. Gascon. (2011). Biodiversity Hotspots. Pp 3–22. In F. E. Zachos and J. C. Habel (Eds.), Global Biodiversity Conservation: The Critical Role of Hotspots. Springer, Berlin and Heidelberg, Germany.

Moreau, R.E. (1952). Africa since the Mesozoic: With particular reference to certain biological problems. Proceedings of the Zoological Society of London 121: 869–913.

Myers, N., R. A. Mittermeier, C. B. Mittermeier, G. A. Da Fonseca, and J. Kent. (2000). Biodiversity hotspots for conservation priorities. Nature 403: 853–858.

Nichols, D. K., E. W. Lamirande, A. P. Pessier, and J. E. Longcore. (2001). Experimental transmission of cutaneous chytridiomycosis in dendrobatid frogs. Journal of Wildlife Diseases 37: 1–11.

Noble, G. K. (1924). Contributions to the herpetology of the Belgian Congo based on the collection of the American Museum Congo Expedition, 1909–1915, Part III. Amphibia. Bulletin of the American Museum of Natural History 49: 147–347.

44 Norton-Griffiths, M., and C. Southey. (1995). The opportunity costs of biodiversity conservation in Kenya. Ecological Economics 12: 125–139.

Nyblade, A.A., and R.A. Brazier. (2002). Precambrian lithosphere controls on the development of the East African rift system. Geology 30: 755–758.

Nylander, J. A., J. C. Wilgenbusch, D. L. Warren, and D. L. Swofford. (2008). AWTY (are we there yet?): A system for graphical exploration of MCMC convergence in Bayesian phylogenetics. Bioinformatics 24: 581–583.

Olson, D. M., and E. Dinerstein. (1998). The Global 200: A representation approach to conserving the Earth’s most biologically valuable ecoregions. Conservation Biology 12: 502–515.

Packard, G. C., and T. J. Boardman. (1999). The use of percentages and size-specific indices to normalize physiological data for variation in body size: Wasted time, wasted effort? Comparative Biochemistry and Physiology Part A 122: 37–44.

Padial, J. M., S. Castroviejo‐Fisher, J. Köhler, C. Vilà, J. C. Chaparro, and I. De la Riva. (2009). Deciphering the products of evolution at the species level: The need for an integrative taxonomy. Zoologica Scripta 38: 431–447.

Palumbi, S., A. Martin, S. Romano, W. O. McMillan, L. Stice, and G. Grabowski. (1991).The Simple Fool’s Guide to PCR. Version 2. The University of Hawaii, Honolulu, HI.

Pramuk, J. B., T. Robertson, J. W. Sites, and B. P. Noonan. (2008). Around the world in 10 million years: Biogeography of the nearly cosmopolitan true toads (Anura: Bufonidae). Global Ecology and Biogeography 17: 72–83.

Passmore, N. I., and V.C. Carruthers. (1995). South African Frogs: A Complete Guide. Southern Book Publishers, Johannesburg, South Africa.

Portillo, F., and E. Greenbaum. (2014a). A new species of the Leptopelis modestus complex (Anura: Arthroleptidae) from the Albertine Rift of Central Africa. Journal of Herpetology 48: 394–406.

Portillo, F., and E. Greenbaum. (2014b). At the edge of a species boundary: A new and relatively young species of Leptopelis (Anura: Arthroleptidae) from the Itombwe Plateau, Democratic Republic of the Congo. Herpetologica 70: 100–119.

Plana, V. (2004). Mechanisms and tempo of evolution in the African Guineo–Congolian rainforest. Philosophical Transactions of the Royal Society of London. Series B: Biological Sciences 359: 1585–1594.

Plumptre, A. J., M. Behangana, T. Davenport, C. Kahindo, R. Kityo, E. Ndomba, and G. Eilu. (2003). The biodiversity of the Albertine Rift. Albertine Rift Technical Reports, No. 3, 1– 112. Accessible at: https://portals.iucn.org/library/node/26890.

45

Plumptre, A. J., T. R. Davenport, M. Behangana, R. Kityo, G. Eilu, P. Ssegawa, C. Ewango, D. Meirte, C. Kahindo, M. Herrenmans, J. K. Peterhans, J. D. Pilgrim, M. Wilson, M. Languy and D. Moyer. (2007). The biodiversity of the Albertine Rift. Biological Conservation 134: 178–194.

Rambaut A., and A. Drummond. (2010). FigTree v1.3.1. Institute of Evolutionary Biology, University of Edinburgh, Edinburgh, United Kingdom. Accessible at: http://tree.bio.ed.ac.uk/software/figtree/ (Accessed: November 29, 2014).

Roberts, E. M., N. J. Stevens, P. M. O’Connor, G. M. Dirks, M. D. Gottfried, W.C. Clyde, and S. Hemming. (2012). Initiation of the western branch of the East African Rift coeval with the eastern branch. Nature Geoscience 5: 289–294.

Rödder, D., J. Kielgast, J. Bielby, S. Schmidtlein, J. Bosch, T. W. Garner, and S. Lötters. (2009). Global amphibian extinction risk assessment for the panzootic chytrid fungus. Diversity 1: 52–66. Rödel, M. O., (2000). Herpetofauna of West Africa, Vol. I. Amphibians of the West African Savanna. Edition Chimaira, Frankfurt am Main, Germany.

Ronquist, F., and J. P. Huelsenbeck. (2003). MRBAYES 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 19: 1572–1574.

Sayre, R., P. Comer, J. Hak, C. Josse, J. Bw, H. Warner, M. Larwanou, E. Kelbessa, T. Bekele, H. Kehl, R. Amena, R. Andriamasimanana, T. Ba, L. Benson, T. Boucher, M. Brown, J. Cress, O. Dassering, B. Friesen, F. Gachathi, S. Houcine, M. Keita, E. Khamala, D. Marangu, F. Mokua, B. Morou, L. Mucina, S. Mugisha, E. Mwavu, M. Rutherford, P. Sanou, S. Syampungani, B. Tomor, A. Vall, J. Vande Weghe, E. Wangui, and L. Waruingi. (2013). A new map of standardized terrestrial ecosystems of Africa. Association of American Geographers Washington, D.C. 24 pp. Accessible at: http://rmgsc.cr.usgs.gov/ecosystems/docs/Africa_Ecosystems_Booklet_final.pdf.

Schiøtz, A. (1964). A preliminary collected in Ghana. Videnskabelige Meddelelser fra Dansk Naturhistorisk Forening 127: 1–17.

Seimon, T. A., S. Ayebare, R. Sekisambu, E. Muhindo, G. Mitamba, E. Greenbaum, and D. Meirte. (2015). Assessing the threat of amphibian chytrid fungus in the Albertine Rift: Past, present and future. PLoS ONE 10: 12–24.

Selander, R. K., W. G. Hunt, and S. Y. Yang. (1969). Protein polymorphism and genic heterozygosity in two European of the house mouse. Evolution 23: 379–390.

Soto-Azat, C., B. T. Clarke, J.C. Poynton, and A.A. Cunningham. (2010). Widespread historical presence of Batrachochytrium dendrobatidis in African pipid frogs. Diversity and Distributions 16: 126–131.

Stamatakis, A. (2006). RAxML-VI-HPC: Maximum likelihood-based phylogenetic analyses 46 with thousands of taxa and mixed models. Bioinformatics 22: 2688–2690.

Stamatakis, A., P. Hoover, and J. Rougemont. (2008). A rapid bootstrap algorithm for the RAxML web servers. Systematic Biology 57: 758–771.

Stewart, M. M. (1974). Parallel pattern polymorphism in the genus Phrynobatrachus (Amphibia: Ranidae). Copeia 4: 823–832.

Stuart, S. N., J. S. Chanson, N. A. Cox, B. E. Young, A. S. Rodrigues, D. L. Fischman, and R. W. Waller. (2004). Status and trends of amphibian declines and extinctions worldwide. Science 306: 1783–1786.

Stuart, B.H.M., M. Walpole, B. Collen, A. van Strien, J. P. Scharlemann, R.E. Almond, and R. Watson. (2010). Global biodiversity: Indicators of recent declines. Science 328: 1164– 1168.

Swindel, S. R., and T. N. Plasterer. (1997). SEQMAN: Contig assembly. Methods in Molecular Biology 70: 75–89.

Tallis, J.H. (1991). Plant Community History. Chapman and Hall, London, United Kingdom.

Vande weghe, J. P. (2004). Forests of Central Africa. Protea Book House, Pretoria, South Africa.

Van den elzen, P., and H. Kreulen. (1979). Notes on the vocalizations of some amphibians from the Serengeti National Park, Tanzania. Bonner Zoologische Beitraege 303: 385–403.

Vieites, D. R., K. C. Wollenberg, F. Andreone, J. Köhler, F. Glaw, and M. Vences. (2009). Vast underestimation of Madagascar's biodiversity evidenced by an integrative amphibian inventory. Proceedings of the National Academy of Sciences, USA 106: 8267–8272.

Vitt, L. J., and J. P. Caldwell. (2013). Herpetology: An Introductory Biology of Amphibians and Reptiles. Academic Press, San Diego, California.

Voyles, J., S. Young, L. Berger, C. Campbell, W. F. Voyles, A. Dinudom, and R. Speare. (2009). Pathogenesis of chytridiomycosis, a cause of catastrophic amphibian declines. Science 326: 582–585.

Voyles, J., L. Berger, S. Young, R. Speare, R. Webb, J. Warner, and L. F. Skerratt. (2007). Electrolyte depletion and osmotic imbalance in amphibians with chytridiomycosis. Diseases of Aquatic Organisms 77: 113–118.

Wager, V. A. (1965). The Frogs of South Africa. Purnell and Son’s, Cape Town, South Africa.

Wager, V. A. (1986). Frogs of South Africa: Their Fascinating Life Stories. Delta Books, Johannesburg, South Africa.

47 Weldon, C., L. H. Du Preez, A. D. Hyatt, R. Muller, and R. Speare (2004). Origin of the amphibian chytrid fungus. Emerging Infectious Diseases 10: 2100–2105.

Wells, K. D. (2007). The Ecology and Behavior of Amphibians. University of Chicago Press, London, United Kingdom.

Ziegler, L., M. Arim, and P. M. Narins. (2011). Linking amphibian call structure to the environment: The interplay between phenotypic flexibility and individual attributes. Behavioral Ecology 22: 520–526.

Zimkus, B. M, and D. C. Blackburn. (2008). Distinguishing features of the sub-Saharan frog genera Arthroleptis and Phrynobatrachus: A short guide for field and museum researchers. Breviora 513: 1–12.

Zimkus, B. M. (2009). Biogeographical analysis of Cameroonian puddle frogs and description of a new species of Phrynobatrachus (Anura: Phrynobatrachidae) endemic to Mount Oku, Cameroon. Zoological Journal of the Linnean Society 157: 795–813.

Zimkus, B. M., and S. Schick. (2010). Light at the end of the tunnel: Insights into the molecular systematics of East African puddle frogs (Anura: Phrynobatrachidae). Systematics and Biodiversity 8: 39–47.

Zimkus, B. M., M. O. Rödel, and A. Hillers. (2010). Complex patterns of continental speciation: Molecular phylogenetics and biogeography of sub-Saharan puddle frogs (Phrynobatrachus). Molecular Phylogenetics and Evolution 55: 883–900.

48 APPENDIX I

Voucher specimens and localities of taxa included in molecular analyses. DRC = Democratic Republic of the Congo, NP = National Park.

Identification Catalog Number Locality Amietia sp. 3 UTEP 21214 DRC: South Kivu, near Irangi Amietia desaegeri UTEP 21232 DRC: North Kivu, Lwanoli River Ptychadena cf. nilotica EBG 1384 DRC: South Kivu, east of Lwiro Ptychadena cf. nilotica EBG 1163 DRC: South Kivu, Lwiro Phrynobatrachus cf. natalensis 1 ES 938 Namibia Phrynobatracus cf. natalensis 1 WC12-A176 Angola: Kuanado Kubango Phrynobatrachus cf. natalensis 2 ELI 516 DRC: South Kivu, Nyamibungu

Phrynobatrachus cf. natalensis 2 ELI 448 DRC: South Kivu, Mwenga Phrynobatrachus cf. natalensis 2 EBG 2155 DRC: South Kivu, Kalundja Phrynobatrachus cf. natalensis 2 EBG 1991 DRC: South Kivu, Luvungi Phrynobatrachus cf. natalensis 2 EBG 2156 DRC: South Kivu, Kalundja Phrynobatrachus natalensis ZFMK 73452 South Africa: Mtunzini Phrynobatracus natalensis WC-DNA-518 South Africa: Dwesa Phrynobatracus natalensis WC-DNA-638 South Africa: Dwesa Phrynobatracus natlensis WC-DNA-753 South Africa: Coffee Bay

Phrynobatracus natalensis WC-DNA-387 South Africa: Hluleka Nature Reserve Phrynobatracus natalensis WC-DNA-388 South Africa: Hluleka Nature Reserve Phrynobatrachus cf. natalensis 3 ELI 1458 DRC: Katanga, Kimari Village Phrynobatrachus cf. natalensis 3 ELI 1457 DRC: Katanga, Kimari Village Phrynobatrachus cf. natalensis 3 ELI 010 DRC: Katanga, Pweto Phrynobatrachus cf. natalensis 3 ELI 081 DRC: Katanga, Kabongo Phrynobatrachus cf. natalensis 3 ELI 234 DRC: Katanga, Kakunko Phrynobatrachus cf. natalensis 4 Ni 52 Mozambique: Niassa Phrynobatrachus cf. natalensis 4 Ni 73 Mozambique: Niassa Phrynobatrachus cf. natalensis 4 Ni 27 Mozambique: Niassa Phrynobatrachus cf. natalensis 4 MCZ A-138354 Tanzania: Nguru Mountain Phrynobatrachus cf. natalensis 5 BM 20051189 Tanzania: Lugugu Phrynobatrachus cf. natalensis 5 MCZ 138337 Tanzania: Sao Hill Phrynobatrachus cf. natalensis 5 ZMK 10.2 Tanzania Phrynobatrachus cf. natalensis 6 1036 Rwanda: Gisenyi Phrynobatrachus cf. natalensis 6 622 Rwanda: Nyabarongo Phrynobatrachus cf, natalensis 6 1055 Rwanda: Ruhengeri Phrynobatrachus cf. natalensis 6 UTA A-58262 Rwanda: Mukungwe Phrynobatrachus cf. natalensis 6 DFH 523 Uganda: Western Region, Rukungiri, Rutooma Village Phrynobatrachus cf. natalensis 6 DFH 524 Uganda: Western Region, 49 Rukungiri, Rutooma Village Phrynobatrachus cf. natalensis 6 ELI 852 Burundi: Muranvya Phrynobatrachus cf. natalensis 6 ELI 851 Burundi: Muranvya, Gatambo road Phrynobatrachus cf. natalensis 6 534 Rwanda: Butare Phrynobatrachus cf. natalensis 6 ELI 1042 Burundi: Gihofi, Mosso plantation Phrynobatrachus cf. natalensis 6 535 Rwanda: Butare Phrynobatrachus cf. natalensis 6 UTA A-58263 Rwanda: Mukungwe Phrynobatrachus cf. natalensis 7 MVZ 245161 Ghana: Kyabobo Park Phrynobatrachus cf. natalensis 7 MVZ 245160 Ghana: Kyabobo Park Phrynobatrachus cf. natalensis 7 ZMB 73717 Côte d'Ivoire

Phrynobatrachus cf. natalensis 8 DFH 135 Uganda: Northern Region, Nakapiripirit, Mount Kadam, Nawaye Puru Phrynobatrachus cf. natalensis 8 DFH 136 Uganda: Northern Region, Nakapiripirit, Mount Kadam, Nawaye Puru Phrynobatrachus cf. natalensis 8 DFH 118 Uganda: Northern Region, Pian Upe Wildlife Reserve, Okudud Village Phrynobatrachus cf. natalensis 8 EBG 2327 DRC: Orientale, Banywani Phrynobatrachus cf. natalensis 8 EBG 2328 DRC: Orientale, Banywani Phrynobatrachus cf. natalensis 8 EBG 1863 DRC: North Kivu, Kamango Phrynobatrachus cf. natalensis 8 EBG 1791 DRC: North Kivu, Virunga NP Phrynobatrachus cf. natalensis 8 EBG 1865 DRC: North Kivu, Kamango Phrynobatrachus cf. natalensis 8 EBG 2458 DRC: Orientale, Mambasa Phrynobatrachus cf. natalensis 8 DFH 7 Uganda: Central Region, Marbira Central Forest Reserve, Nagojje Beat Phrynobatrachus cf. natalensis 8 EBG 2457 DRC: Orientale, Mambasa Phrynobatrachus cf. natalensis 8 DFH 6 Uganda: Central Region, Marbira Central Forest Reserve, Nagojje Beat Phrynobatrachus cf. natalensis 8 MVZ 234059 Uganda: Kampala Phrynobatrachus cf. natalensis 9 A 2008229 Ethiopia: Dola-Mena Phrynobatrachus cf. natalensis 9 A 2002203 Ethiopia: Dola-Mena Phrynobatrachus cf. natalensis 9 A 2008205 Ethiopia: Dola-Mena Phrynobatrachus cf. natalensis 9 750-11 Ethiopia: Jinka Phrynobatrachus cf. natalensis 9 694-11 Ethiopia: Doketu Phrynobatracus acridoides BM 2002347 Tanzania Phrynobatracus acridoides MZCZ A-138214 Tanzania Phrynobatracus cf. natlensis 391 Mozambique: Gorongosa NP Phrynobatrachus asper UTEP 20426 DRC: South Kivu, Itombwe Plateau, Nakihomba River Phrynobatracus bullans MB 90 Uganda: Murchison Falls NP Phrynobatracus bullans AAUA 2009032 Ethiopia Phrynobatracus bullans BMNH 20051179 Tanzania

50 Phrynobatracus bullans MVZ 234151 Tanzania Phrynobatracus bullans MVZ 238716 Kenya Phrynobatrachus graueri ELI 806 DRC: South Kivu, Itombwe Plateau, Mitamba Phrynobatrachus francisci ZMB 73779 Benin Phrynobatrachus francisci MVZ 249545 Ghana Phrynobatrachus francisci MVZ 245148 Ghana

51 APPENDIX II

Other Specimens Examined

Phrynobatrachus cf. natalensis 2—Democratic Republic of the Congo: South Kivu

Province: Nyamibungu, S03.25732, E28.11797, 682 m (ELI 516); Phrynobatrachus cf. natalensis 3—Democratic Republic of the Congo: South Kivu Province: Kimari Village,

S04.38729, E28.35958, 737 m (ELI 1457–1458); Katanga Province: road ca. 4 km N of Pweto,

S08.43258, E28.92493, 960 m (ELI 010), Kakunko, S08.47412, E27.31540, 1570 m (ELI 234), vicinity of Kabongo, S08.73308, E28.20122, 1014 m (ELI 081); Phrynobatrachus cf. natalensis

6—Uganda: Western Region: Rukungiri Town, Rutooma Village, S00.75172, E29.98220, 1682 m (DFH 524); Phrynobatrachus cf. natalensis 8—Democratic Republic of the Congo: North

Kivu Province: Kamango, N00.66489, E29.87811, 841 m (EBG 1865); Orientale Province:

Banywani, N01.55056, E30.25288, 1248 m (EBG 2327–2328), near Mambasa, (EBG 2458);

Uganda: Northern Region: Nakapiripirit, Mount Kadam, Nawaye Puru, N01.80875, E34.74598,

2187 m (DFH 135–136).

52 VITA

Joshua Adan Lara began his college career in 2007, after completing a B.S. in biology at the University of Texas at El Paso (UTEP) in December 2012. During this time he has participated in three fellowships for undergraduate research, Bridges to the Baccalaureate Program (BRIDGES), Howard Hughes Medical Institute (HHMI), and Research Initiative for Scientific Enhancement (RISE). The outcome of his participation was publication in the scientific journal Vaccine, “Serna, C., J. A. Lara, S. P. Rodrigues, A. F. Marques, I. C. Almeida, and R. A. Maldonado. (2014). A synthetic peptide from Trypanosoma cruzi mucin-like associated surface protein as candidate for a vaccine against Chagas disease. Vaccine, 32: 3525—3532.” Joshua began his graduate degree at UTEP in August 2013. During his graduate stint, he has been a teaching assistant for Medical Parasitology (ZOOL 3364), Prokaryotic Molecular Genetics (MICR 3449), and General Biology Laboratory (BIOL 1107). He is a member of the Society for Advancement of Chicanos and Native Americans in Science (SACNAS). He plans on applying to the industrial scientific sector and teach part-time at a

Community College.

Contact Information:

This thesis was typed by Joshua Adan Lara

53