ArticlePDF Available

Abstract and Figures

Bats of the family Pteropodidae, also known as megabats or Old World fruit bats, are widely distributed in tropical areas of Africa, Asia, and Oceania. Of 45 genera in the family, 12 are endemic to the Afro-tropical region and two others have representative species on the African continent. African megabats inhabit wooded habitats and are nearly ubiquitous on the mainland and nearby islands with the exception of desert areas. Some species have been implicated as possible reservoirs of the Ebola Zaire virus. We studied the phylogenetic relationships of mainland African megabats using both mitochondrial and nuclear loci in separate and combined analyses. The phylogenetic trees obtained showed four main African clades: Eidolon, Scotonycterini (including two genera), African Rousettus (three species), and the previously identified 'endemic African clade' (nine genera). The latter three lineages form a clade that also includes the Asian species of Rousettus and the Asian genus Eonycteris; Eidolon does not show close relationships to other African genera, instead nesting elsewhere in the megabat tree. Although our results confirm many of the conclusions of previous studies, they challenge the taxonomic status and placement of Epomops dobsonii and Micropteropus, and provide evidence indicating that a new classification at subfamilial and tribal levels is highly desirable. The principal clades we detected represent four independent colonizations of Africa from most probably Asian ancestors. Estimates of divergence dates suggest that these events occurred in different periods and that although local diversification appears to have started in the late Miocene, the more extensive diversification that produced the modern fauna occurred much later, in the Pleistocene.
Content may be subject to copyright.
Megabats (Mammalia: Chiroptera: Pteropodi -
dae) represent a monophyletic group of mostly non-
echolocating, phytophagous bats specialized for
consumption of fruit and flower products (nectar
and pollen — Kunz and Pierson, 1994). Pteropodids
are distributed along the Old World tropics from
western Africa to eastern Polynesia (Kunz and Pie r -
son, 1994), and within this range Sub-Saharan Afri -
ca represents the largest land area where they occur.
The African megabat fauna includes species from
several clades that may have arrived, or evolved, in-
dependently on the continent and adjacent islands.
Epomophorinae sensu Bergmans (1997) is the
largest group of African Pteropodidae, comprising
11 genera almost exclusively found in continental
Africa. Bergmans’ (1997) Epomo phorinae included
species previously widely separated in the tradi-
tional taxonomy, i.e., taxa included by Ander sen
(1912) in the subfamilies Cynopterinae (Myo -
nycteris), Rousettinae (Lissonycteris) and Macro -
glossinae (Megaloglossus). Bergmans (1997) sub -
divided the epomophorines into four tribes:
Epo mophorini (genera Epomophorus, Epomops,
Hypsignathus, Micropteropus, and Nanonycteris),
Myonycterini (Myonycteris, Lissonycteris, and
Megaloglossus), Plerotini (with the monotypic
Plerotes), and Scotonycterini (Scotonycteris and
Ca si nycteris). Early molecular phylogenetic analy-
ses were in agreement with the existence of an en-
demic African clade that included the former two
Acta Chiropterologica, 18(1): 73–90, 2016
PL ISSN 1508-1109 © Museum and Institute of Zoology PAS
doi: 10.3161/15081109ACC2016.18.1.003
The evolutionary history of the African fruit bats (Chiroptera: Pteropodidae)
1Department of Mammalogy, Division of Vertebrate Zoology, American Museum of Natural History,
Central Park West at 79th Street, New York, NY 10024, USA
2Facultad de Ciencias Naturales e Instituto Miguel Lillo, Universidad Nacional de Tucumán, Consejo Nacional de
Investigaciones Científicas y Tecnológicas (CONICET),Unidad Ejecutora Lillo, Miguel Lillo, 205,
San Miguel de Tucumán, 4000, Argentina
3Current address: Departamento de Ecología, Genética y Evolución, Universidad de Buenos Aires, Consejo Nacional de
Investigaciones Científicas y Tecnológicas (CONICET), Intendente Güiraldes y Costanera Norte s/n,
Pabellón II - Ciudad Universitaria, 1428, Capital Federal, Argentina
4Corresponding author: E-mail:
Bats of the family Pteropodidae, also known as megabats or Old World fruit bats, are widely distributed in tropical areas of Africa,
Asia, and Oceania. Of 45 genera in the family, 12 are endemic to the Afro-tropical region and two others have representative species
on the African continent. African megabats inhabit wooded habitats and are nearly ubiquitous on the mainland and nearby islands
with the exception of desert areas. Some species have been implicated as possible reservoirs of the Ebola Zaire virus. We studied
the phylogenetic relationships of mainland African megabats using both mitochondrial and nuclear loci in separate and combined
analyses. The phylogenetic trees obtained showed four main African clades: Eidolon, Scotonycterini (including two genera), African
Rousettus (three species), and the previously identified ‘endemic African clade’ (nine genera). The latter three lineages form a clade
that also includes the Asian species of Rousettus and the Asian genus Eonycteris; Eidolon does not show close relationships to other
African genera, instead nesting elsewhere in the megabat tree. Although our results confirm many of the conclusions of previous
studies, they challenge the taxonomic status and placement of Epomops dobsonii and Micropteropus, and provide evidence
indicating that a new classification at subfamilial and tribal levels is highly desirable. The principal clades we detected represent
four independent colonizations of Africa from most probably Asian ancestors. Estimates of divergence dates suggest that these
events occurred in different periods and that although local diversification appears to have started in the late Miocene, the more
extensive diversification that produced the modern fauna occurred much later, in the Pleistocene.
Key words: phylogenetic analysis, Africa, Epomophorinae, molecular systematics, molecular clock, pteropodids, Rousettus,
74 F. C. Almeida, N. P. Giannini, and N. B. Simmons
tribes, although the other tribes were not sampled in
those studies (Hollar and Springer, 1997; Juste et al.,
1999; Álvarez et al., 1999). Subsequent phyloge-
nies, with more species and additional character data
(e.g., Giannini and Simmons, 2003, 2005), found
support for the endemic African clade both in mo-
lecular and morphological data sets, but anticipated
a more complex scenario of relationships, later
confirmed with the recovery of Scotonycterini as
a branch separated from other epomophorines
(Almeida et al., 2011). In the phylogeny recovered
by Almeida et al. (2011), typical rousettine mega-
bats (Eonycteris, Rousettus, and Stenonycteris) ap-
peared nested between Scotonycterini and the re-
mainder of epomophorines, a result confirmed in
subsequent analyses (e.g., Hassanin, 2014). Indeed,
Nesi et al. (2013) proposed that Stenonycteris is best
placed in a separate tribe, related to the endemic
African clade. Therefore, rousettines and eonycter-
ines are integral to the understanding of the evolu-
tion of African megabats, even though some of them
are distributed in the Australasian tropics.
Other lineages of megabats are distributed in
Africa in addition to epomophorines. Among the
eight species of Rousettus, three have African-
Malagasy distributions (R. aegyptiacus, R. mada-
gascariensis, and R. oblivious), with R. aegyptiacus
restricted to the mainland. Another African lineage
is represented by the genus Eidolon, which includes
two species and three subspecies (Simmons, 2005).
This genus has proven extremely difficult to place
within the pteropodid family tree. It seems likely
that Eidolon represents an independent lineage
not closely related to any other megabat clades
(Al meida et al., 2011), although associations with
rousettines (Andersen, 1912; Bergmans, 1997) and
pteropodines (Giannini and Simmons, 2005) have
been suggested. The latter comprise the flying foxes
(Pteropus), a speciose, chiefly Australasian genus,
that includes a few species distributed on islands of
the western Indian Ocean and offshore African
islands; nevertheless, no Pteropus species is found
in continental Africa. O’Brien et al. (2009) and
Almeida et al. (2014) have shown that Pteropus
species from islands of the western Indian Ocean,
including Madagascar, Aldabra, Seychelles, Como -
ros, Mascarene, and islands off the Tanzanian coast,
belong to several clades mainly associated to the
P. vampyrus species groups (sensu Almeida et al.,
2014) and are deeply nested within pteropodines.
Since none of these lineages have been linked nei-
ther to the ‘endemic African clade’ nor to the scoto -
nycterines in recent phylogenies, they represent yet
other independent instances of pteropodid coloniza-
tion of Afro-Malagasy region.
In recent years, members of Pteropodidae have
been identified as natural reservoirs of a number
of viruses that cause zoonotic diseases including
the Nipah, Hendra, Marburg, and Ebola viruses
(Epstein et al., 2008; Halpin et al., 2011; Drexler et
al., 2012; Olival et al., 2013; Schaer et al., 2013;
Wynne and Wang, 2013; Alexander et al., 2015;
Raza na jatovo et al., 2015). More specifically, in
Africa, five pteropodids (Epomops franqueti, Hypsi -
gnathus monstrosus, Myonycteris torquata, Rouset -
tus aegyptiacus, and Eidolon helvum) have shown
signs of infection (viral RNA and specific antibod-
ies) with the Ebola subtype Zaire virus, which
caused the 2014 epidemics in West Africa (Pourrut
et al., 2009; Hayman et al., 2010; Leroy et al., 2014;
Vogel, 2014). These bats have been shown to be
seropositive without manifesting the disease. In at
least one case, the long-term survival along a typical
migration cycle was recorded for a female of
E. helvum with antibodies for both Ebola subtype
Zaire and Lagos bat viruses (Hayman et al., 2010).
Nearby in Madagascar, novel coronaviruses have
been detected in congeneric E. dupreanum and also
in the endemic Pteropus rufus (Razanajatovo et al.,
2015). Better understanding of the diversity and
transmission of these viruses depends on not only
studies of the viruses themselves, but also on gain-
ing a better understanding of the biology of the bats
including their evolutionary history.
The focus in the present study is determining
phylogenetic relationships among taxa within the
well-supported endemic African clade (Epomopho -
rini and Myonycterini) and their closely-related al-
lies (Scotonycterini, Rousettini, and Stenonycterini)
as recovered in Almeida et al. (2011). The study by
Almeida et al. (2011) focused on relationships at the
generic level and, in terms of breadth of diversity
sampled, is the most comprehensive molecular phy-
logenetic study of pteropodids to date. However,
that study lacked breadth of sampling within many
African genera and some of the recovered relation-
ships lacked statistical support. Some subsequent
studies have sampled much more densely within
selected African taxa (e.g., Nesi et al., 2013; Hassa-
nin, 2014), but their samples have never been com-
prehensively analyzed in concert with broader taxo-
nomic and locus sampling, nor have some key
taxa (e.g., Plerotes anchietae) been included in prior
analyses. To address these issues, we conducted
a comprehensive set of analyses based on sequence
data from eight nuclear and mitochondrial markers,
A phylogeny of African pteropodids 75
and generated a robust phylogeny of the pteropodids
that occur in continental Africa. Our sample was
designed to include all the species that have been
identified as potential reservoirs of hemorrhagic-
fever viruses that infect humans and other primates.
We sampled members of all genera and most species in the
scotonycterine, rousettine, and epomophorine clades (the latter
inclusive of Epomo pho rini, Plerotini, and Myonycterini). In
addition to taxa studied in Almeida et al. (2011), we sampled
16 extra species of African genera by sequencing new tissue
samples available to us and gathering published sequences
available in GenBank. Among the new data that we generated
are sequences of P. anchietae (Plerotini, historically placed
with in Epomophorinae — Andersen, 1912), which has never be-
fore been included in a molecular phylogeny. Addition al ly, we
included sequences from another African genus, Eidolon, which
does not appear to be closely related to the remaining ingroup
genera. In Almeida et al. (2011), Eidolon was recovered as an
independent and perhaps basal lineage without close affinities to
any other pteropodid lineage from inside or outside of Africa.
Although determining the phylogenetic position of Eidolon
within Pteropodidae as a whole was not a focus of this study
(as that would require very broad sampling of Ptero podidae as
a whole and more nuclear markers than employed in the current
study; Almeida et al., 2011) inclusion of this genus completes
our matrix with sampling all megabat genera that occur in con-
tinental Africa. For each genus, we included all species for
which sequences of at least one of the eight chosen loci were
available, totaling 31 species (but possibly 32 — see Dis cussion
and Appendix). Three other pteropodid species were included in
the matrix as outgroups: Pteropus medius (= giganteus;
Pteropodinae), Cynopterus sphinx (Cyno pterinae), and Nycti -
mene albiventer (Nyctimeninae). This choice of outgroups was
based on the phylogeny presented by Almeida et al. (2011), who
convincingly demonstrated that these species are not closely re-
lated to any of the ingroup taxa sampled here. When ever possi-
ble we included up to three sequences per spe cies to make sure
that samples had been correctly identified. In case of doubt,
extra sequences were obtained from the GenBank for compari-
son, although some were not included in the trees shown here.
The number of loci sequenced per species was variable (be-
tween one and seven; Supplementary Table S1). For some spe -
cies for which we could not obtain tissue samples, we acquire
sequences from GenBank, usually represented by one or two
genes (Supplementary Table S1). For Epomops dobsonii, a frag-
ment of the 12S gene was sequenced from a skin biopsy of
a museum specimen collected in 1938. The procedures for col-
lecting the skin biopsy, DNA extraction, sequencing, and se-
quence curation were carried out as described in Almeida et
al. (2014). All tissue samples used herein were preserved and
deposited in museum collections, and, therefore, no animals
were collected or sacrificed specifically for this study.
Molecular Data
Four nuclear (RAG1, RAG2, vWF, and BRCA1) and four
mitochondrial (Cytb, 12S-rRNA, tRNA-val, and 16S-rRNA)
markers were included in the matrix as in Almeida et al. (2011).
The aligned nuclear matrix had 4410 bp, while the mitochondr-
ial matrix comprised of 3673 bp. Sequences from 22 specimens
were newly generated for this study (Appendix). Primers, mo-
lecular methods, and sequence editing followed those previ-
ously described in our prior papers on pteropodid relationships
(Almeida et al., 2009, 2011; Giannini et al., 2009).
Phylogenetic Analysis
Alignments were done with MAFFT version 7 (Katoh and
Standley, 2013), although coding fragments, with the exception
of BRCA1, could be aligned by eye since they did not include
indels. Each gene was first analyzed individually to check for
unexpected results (e.g., individuals of the same species that did
not cluster together) that could point to misidentifications or
contamination (Supplementary Figs. S1–S6). Three concate-
nated matrices (mitochondrial loci, nuclear loci, and all loci)
were then constructed for analysis. In all analyses, the datasets
were parti tioned by gene and codon position (first + second and
third) for the coding genes, a method previously suggested to
represent a good compromise between likelihood gain and par-
tition size (Almeida et al., 2011). The contiguous fragment con-
taining the 12S-rRNA, tRNA-val, and 16S-rRNA genes was
treated as a single partition (12S16S).
Before concatenating the nuclear and the mitochondrial
data sets, we ran an Incongruence Length Difference test (ILD
— Farris et al., 1994) in PAUP* (Swofford, 2002) to test for
phylogenetic incongruence between mitochondrial and nuclear
genes. The ILD test was run 500 times, with 10 random species
addition per search. We additionally used the Shimodaira-Hase -
gawa (SH — Shimodaira and Hasegawa, 1999) test to compare
tree likelihoods given different datasets (mitochondrial, nuclear,
all loci combined). In this test, incongruence is detected when
a tree obtained with a dataset has a significantly higher likeli-
hood than alternative trees obtained with other datasets. The SH
test was run on a reduced dataset of 26 taxa where missing data
was minimized as follows: we built chimeric sequences to com-
plete the data for each species combining sequences of two or
more individuals when possible and necessary, and removed
all taxa for which only a few sequences were available (Sup -
plementary Table S1). The test was done with the RAxML pro-
gram version 8.0.0. (option -f H — Stamatakis, 2014)
Maximum likelihood (ML) tree searches were also carried
out with the program RAxML. The analyses employed the
GTRGAMMA substitution model and were run 10 times inde-
pendently to obtain the best tree. Statistical support for clades
was measured with 200 standard bootstrap replicates (com-
mands: -f d –b 1234 -# 200). Maximum Parsimony (MP) tree
searches with the three datasets were carried out with TNT
(Goloboff et al., 2008). Bayesian inference (BI) trees were ob-
tained with MrBayes (Ronquist and Huelsenbeck, 2003) in two
runs with 106generations and six chains each. Trees were sam-
pled every 2000 generations. All data partitions were analyzed
under the GTRGAMMA model with unlinked parameter esti-
mation. Convergence was evaluated based on ESS values
(> 200) as visualized with Tracer v1.6 (Rambaut et al., 2014).
Tree illustrations were made with FigTree (Rambaut, 2009) and
rooted at the midpoint.
Divergence Time Estimates
Species divergence times were estimated with the pro-
gram BEAST2 (Bouckaert et al., 2014), which uses a Bayesian
frame work and allows for the use of relaxed clocks. Due to
a lack of ingroup fossils, we based this analysis on the substitu-
tion rate estimated for the Cytbgene in bats (Ruedi and Mayer,
2001; Hulva et al., 2004). The Cytbalignment was trimmed so
that only one individual per species was included. The sites
were partitioned by codon position, which were unlinked for
the estimation of the parameters of the GTR+Γ+I model.
We assumed a relaxed lognormal clock and the Yule speciation
model with a Gamma distribution prior. For the mean substitu-
tion rate, the prior was set as a lognormal distribution with mean
of 0.023 subs/site/My and a standard deviation of 0.3 to match
the rate estimated based on fossil data (Ruedi and Mayer, 2001;
Hulva et al., 2004). The chain was run for 107generations and
sampled every 10,000 generations. The convergence of param-
eters was checked with Tracer (Rambaut et al., 2014).
Previous studies have suggested that pteropodids had an
Australasian origin (Butler, 1984; Aguilar et al., 1986) and have
colonized the African continent on several independent occa-
sions (Juste et al., 1999; Almeida et al., 2011). To further inves-
tigate the history of the African colonization by pteropodids, we
analyzed our data using DIVA (Ronquist, 1997) as implemented
in RASP v 3.1 (Yu et al., 2015). For this analysis, we increased
the number of non-African (Australasian) pteropodids in the
data set, included two non-pteropodid taxa (Hippo sideros vitta-
tus and Megaderma lyra) as additional outgroups (Sup ple men -
tary Table S1), and decreased the number of ingroup terminals
to one sample per species. The ML tree used in the DIVA was
obtained as described above.
Phylogenetic analyses of mitochondrial and
nuclear datasets agreed in most relationships within
and among the main clades and each recovered
Scotonycterini, Eonycteris, Rousettus, and the en-
demic African clade as monophyletic groups (Figs.
1 and 2, Supplementary Figs. S1–S6). The two par-
titions disagreed, however, on the positions of
Rousettus and Eonycteris, with the nuclear dataset
favoring Rousettus as the second ingroup clade
to split, while the mitochondrial dataset favoring
Eonycteris in this position instead (Figs. 1 and 2,
Supplementary Figs. S1–S6). In neither case was
there statistical support for the respective place-
ments. Another difference between results based
on the two datasets involved basal relationships
within the endemic African clade, though again
in neither case did alternative relationships receive
statistical support (Figs. 1 and 2). This result mir-
rors the fact that there was disagreement (without
statistical support) between different mitochondr-
ial genes (Cytband 12S16S) with respect to basal
re lationships of the endemic African clade (Sup -
plementary Figs. S5 and S6). Finally, there was
lower resolution in the nuclear tree with respect
to relationships among the Epomophorus/Micro -
pteropus and Myonycteris species. This seems to be
due to lack of phylogenetic information in slowly-
evolving nuclear genes; very few substitutions were
observed within these clades and even fewer were
phylogenetically informative (e.g., only 15 out of
4,410 sites were variable among Myonycteris spe -
cies, and only three of these were phylogenetically
No significant incongruence between datasets
was detected with the ILD test. The SH test did not
reject congruence between either datasets (mito-
chondrial or nuclear) and the tree obtained with the
all-loci matrix (nuclear: D = 4.07; mitochondrial:
D = 6.27). The SH test also did not detect incongru-
ence between the nuclear dataset and the mito-
chondrial tree (D = -11.46). When considering the
mitochondrial dataset, however, the mitochondrial
tree was significantly better than the nuclear tree
(D = -80.46). These results point to a higher likeli-
hood of the mitochondrial tree, which is probably an
effect of the lower number of variable and informa-
tive characters of the nuclear dataset (5.7% of the
nuclear sites were informative, in comparison with
13% of the mitochondrial sites). In fact, the topol-
ogy of the ML tree based on the concatenated matrix
is more similar to the one based on the mitochondr-
ial dataset than it is to the tree based only on nuclear
genes. It is possible that the mitochondrial genes
(maternally inherited) and nuclear genes (bipa ren -
tally inherited) have different evolutionary histories
as seems to be the case in some bat groups (e.g.,
Larsen et al., 2010; Almeida et al., 2014) but we did
not detect evidence of this.
A tree summarizing the analyses done with the
all-loci matrix (including both mitochondrial and
nuclear loci) is shown in Fig. 3. The three recon-
struction methods employed yielded highly congru-
ent trees with high statistical support for all major
clades and most subclades. Eidolon, as previously
observed (Almeida et al., 2011), is not closely re-
lated to the other African taxa (ingroup taxon set).
The first ingroup clade to split from the others was
Scotonycterini. The second clade to split within the
ingroup in all analyses was Eonycteris, but only the
BI tree provided statistical support for this relation-
ship. In this way, the genus Rousettus was found to
be sister to the endemic African clade (Fig. 3). Rela -
tionships within the endemic African clade were in-
consistent across analytical methods (BI and MP
trees are illustrated in Supplementary Figs. S7 and
S8). Most higher-level clades within the ingroup
received high statistical support in all analyses.
76 F. C. Almeida, N. P. Giannini, and N. B. Simmons
The main exceptions were the sister relationship
between Rousettus and the endemic African clade
and the basal relationships within the endemic
African clade. Similar results were obtained in the
Almeida et al. (2011) analyses, which included
a subset of the ingroup species sampled in the pres-
ent study.
To rule out missing data as an explanation for
some of the low support values observed, we reran
the ML analysis on matrices with minimal missing
data (Supplementary Fig. S9). These matrices were
built by forming chimeras of terminals belonging to
the same species when not all gene sequences were
available for a particular sample (e.g., Epo mophorus
A phylogeny of African pteropodids 77
FIG. 1. Maximum likelihood phylogeny based on mitochondrial DNA data (Cytb). ML bootstrap support values are shown
above branches
FIG. 2. Maximum likelihood phylogeny based on nuclear data (RAG1, RAG2, vWF, and BRCA1 genes). ML bootstrap support
values are shown above branches
78 F. C. Almeida, N. P. Giannini, and N. B. Simmons
wahlbergi) and excluding all terminals missing
more than two of the seven genes analyzed here.
An exception was made for Plerotes anchietae,
which was included in one of the low-missing-
data matrices. We did not find significant differ-
ences either in topology or levels of perceived
support between trees obtained with these reduced
matrices and those obtained our complete matrix
(Fig. 3). This finding is in accordance with the lack
of empirical evidence that missing data have
a strong influence in phylogenetic inference and/or
statistical support of clades (Wiens, 2006).
Another piece of evidence suggesting that miss-
ing data is not a major issue across our matrix as
a whole was the high support found for the position
of Casinycteris ophiodon in all analyses (Fig. 3),
even though this species was represented in the
combined matrix by only the Cytbgene and a frag-
ment of the 12S gene (ca. 1,000 bp). However, miss-
ing data may have produced uncertainty in other
parts of the tree. Low support values were obtained
for some intraspecific relationships in the genera
Rousettus and Myonycteris. In both cases, some of
the involved species had few sequences available
FIG. 3. Maximum likelihood phylogeny based on total evidence (mitochondrial + nuclear genes concatenated). Above branches are
shown maximum likelihood bootstrap support values / Bayesian posterior probabilities / maximum parsimony bootstrap values
A phylogeny of African pteropodids 79
(e.g., for R. obliviosus and M. leptodon only the Cytb
gene was available). In these particular cases, which
appear to involve short branches, it is possible that
sequences from more loci could help improving sta-
tistical support for phylogenetic relationships.
Divergence time estimates based on the sub-
stitution rates of the Cytbgene are illustrated in
Fig. 4. Biogeographic reconstruction using DIVA
revealed four independent colonizations of Africa/
western Indian Ocean islands (maroon color in
circles of Fig. 5). Changes in the dataset for
the DIVA analysis (see methods) did not change
the main phylogenetic results (Supplementary Fig.
FIG. 4. Time-calibrated phylogeny bars on nodes represent the 95% confidence intervals of the estimated split dates
80 F. C. Almeida, N. P. Giannini, and N. B. Simmons
This tribe was included in the Epomophorinae
subfamily by Bergmans (1997), but cumulative
evidence from phylogenetic analyses (Almeida et
al., 2011; Hassanin, 2014) has shown that such
inclusion makes Epomophorinae a non-monophy-
letic grouping. Scotonycterini sensu Bergmans
(1997) is composed of two genera, Casinycteris and
Scoto nycteris, of small to medium-sized pteropodids
(18–95 g) endemic to either one or both the Guinean
and Congo Basin rainforests blocks (Nowak, 1994).
These bats have characteristic small patches of
white hair on the nose and behind the eyes. Until
recently, two species of Scotonycteris and one of
Casinycteris were known to science (Simmons,
2005). An analysis of the cytochrome bgene re-
vealed a closer relationship of S. ophiodon to C. ar -
gyn nis, rather than to S. zenkeri, prompting the
transfer of ophiodon to the genus Casinycteris
(Has sanin, 2014). In the same study a new species,
C. campomaanensis was described from Cameroon.
In all our analysis, Scotonycterini was recovered as
monophyletic with high statistical support, but did
not share an immediate common ancestor with
the endemic African clade. More recently, based on
ge netic discontinuity in both mitochondrial and nu-
clear loci, S. zenkeri has been split into three spe -
cies: S. zenkeri, S. occidentalis, and S. bergmansi
(Has sanin et al., 2015).
Bergmans (1997) grouped Rousettus and Eonyc -
teris together with Eidolon in the tribe Rousettini;
likely, this arrangement reflected the morphological
resemblance of Eonycteris to Rousettus, as well as
the traditional association of Eidolon with Rousettus
as described in Andersen (1912). In none of our
analyses, however, did Rousettus clustered with
Eonycteris, which suggests that Rousettus and
Eonycteris should each be in their own tribe (see
This is a megabat genus that includes medium-
sized, apparently unspecialized pteropodids. Several
other relatively unspecialized pteropodid taxa
(Boneia, Lissonycteris, and Stenonycteris) were pre-
viously included in Rousettus (e.g., Andersen, 1912;
Bergmans and Rozendaal, 1988) but these were sub-
sequently removed (Bergmans, 1997; Giannini and
Simmons, 2003; Giannini et al., 2009; Nesi et al.,
FIG. 5. Biogeographic reconstruction obtained with DIVA
A phylogeny of African pteropodids 81
82 F. C. Almeida, N. P. Giannini, and N. B. Simmons
2013). Remarkably, this is the only genus of Pte ro -
podidae found both in mainland Africa and eastern
Asia. In our analysis, we uncovered an interesting
problem with the identification of Rousettus speci-
mens from Indochina, where the distributions of
R. leschenaultii and R. amplexicaudatus overlap
(see Supplementary Text and Supplementary Fig.
S4). Two Asian Rousettus species are unfortunately
missing from our analyses (R. celebensis and R. lin-
duensis, both from Sulawesi) because neither sam-
ples nor published sequence were available.
In the trees presented herein (Figs. 1–3), a basal
split within Rousettus separates R. amplexicaudatus
from the remaining species in the late Miocene. The
second species to split is R. spinalatus (recorded in
Borneo and Sumatra, Simmons, 2005). Rousettus
leschenaultii appears as sister of a monophyletic
clade that contained all the African species. This is
a new and interesting result, since previous stud-
ies based on the cytochrome bgene alone were not
able to resolve the relationships between African
Rou settus and R. leschenaultii (Goodman et al.,
2010). Our results indicate that Rousettus colonized
Africa only once, in the late Pliocene or early
Pleistocene, and shortly thereafter diversified into
three species: one on the continent (R. aegyptiacus)
and two on islands (R. madagascariensis in Mada -
gascar and R. obliviosus in the Comoros archi -
pelago). Topology of our trees suggests that the first
split was between R. aegyptiacus and the lineage
leading to the island forms. A second split, occurring
in the island lineage, subsequently resulted in the
differentiation of R. madagascariensis and R. obliv-
iosus, also in the Early Pleistocene.
The Endemic African Clade
This clade includes a diverse set of species, clas-
sified into nine genera (Stenonycteris, Plerotes,
Myo nycteris, Megaloglossus, Hypsignathus, Nano -
nyc teris, Epomops, Epomophorus, and Micropte ro -
pus). According to our results, the clade has a basal
split into four main lineages corresponding to Steno -
nycteris,Plerotes, Myonycterini, and Epomopho ri -
ni. The relationships among these lineages were
variously resolved by our sequence data, with dif -
ferences across methods and a general lack of
statistical support for any single branching pattern
among the four lineages.
Plerotes and Stenonycteris are each monotypic;
while the former inhabits dry forests and savannas,
the latter is found in montane forests. Plerotes is
known from only about a dozen specimens, and the
holotype has been lost (a neotype has been desig-
nated since then, Bergmans, 1989). Examination of
a recently-captured specimen from Malawi (Sen -
cken berg Museum SMF 85.744), included in our
molecular data set and so far the only adult male in
collections, revealed interesting morphological fea-
tures that apparently link Plerotes to myonyc-
terines. Particularly, the specimen exhibits a ruff of
enlarged, clustered glandular hairs much longer
(> 10 mm) than the surrounding pelage and slightly
paler in coloration. This ruff appears in a few dis-
tantly related megabats but it is typical of myo-
nycterines; thus, locally it represents a potential
morphological synapomorphy supporting a close
rel a tion ship of Plerotes and myonycterines. The lat-
ter includes two genera: Megaloglossus and Myo -
nycteris. Recently, in a comprehensive molecular
anal ysis of this tribe, Lissonycteris angolensis (pre-
viously included in Rousettus) was included in
an expanded Myonycteris (Nesi et al., 2013). In
the same study, genetic discontinuity correlated
with allopatric distribution prompted the erection
of a new species, Megaloglossus azagnyi, and the
recognition of Myonycteris leptodon Andersen,
1908 as a distinctive species, both from West Africa
(Nesi et al., 2013).
The species of Epomophorini are characterized
by conspicuous white hair patches at the base of the
ears. With the exception of Hypsignathus monstro-
sus, males also possess ‘epaulets’, i.e. tufts of white
hair on the shoulders that are used in display (Berg -
mans, 1988). Hypsignathus monstrosus and Nano -
nycteris veldkampii inhabit the rainforests of West
and Central African, while the other epomophorine
species inhabit the adjacent deciduous forests, sa-
vanna woodlands, and montane forests. Within
Epomophorini, we found the only polytypic genera
included in our study that were not monophyletic:
Epomophorus and Epomops. The genus Epomops
has three currently recognized species (Simmons,
2005): E. buettikoferi, E. dobsonii and E. franqueti.
The inclusion of dobsonii in Epomops, however, has
been questioned (Bergmans, 1989). Although we
could not obtain tissue samples for DNA extraction
of the latter species, we were able to sequence a 776
bp fragment of the 12S gene from a museum skin
of E. dobsonii. In our analyses (Figs. 1 and 3), the
E. dob sonii sample clustered with Epomophorus
wahlbergi samples with high statistical support. The
identification of all three samples based on morph-
ology and measurements was double-checked upon
these results and confirmed. Bergmans (1989) no-
ticed the distinctiveness of E. dobsonii as compared
Bergmans (1997) New classification
Harpyionycterinae Harpyionycterinae
Dobsonia, Aproteles, Harpyionycteris Dobsonia,Aproteles,Boneia,Harpyionycteris
Rousettinae Rousettinae
Dobsoniini – Dobsonia, Aproteles Rousettini – Rousettus
Rousettini – Eidolon, Eonycteris, Rousettus1Eonycterini – Eonycteris
Epomophorinae Scotonycterini – Scotonycteris,Casinycteris
Scotonycterini – Scotonycteris, Casinycteris Epomophorini – Epomophorus,Epomops,Hypsignathus,
Epomophorini – Epomophorus, Epomops,Hypsignathus,Nanonycteris,Micropteropus
Nanonycteris,Micropteropus Stenonycterini – Stenonycteris
Myonycterini – Lissonycteris,Myonycteris,Megaloglossus Myonycterini – Myonycteris2, Megaloglossus
Plerotini – Plerotes Plerotini – Plerotes
1— inclusive of Boneia and Stenonycteris; 2 — inclusive of Lissonycteris
TABLE. 1. Bergmans’ (1997) classification (left) and a new classification (right) based on the phylogenetic results presented here for
the African megabats and their closely related allies
A phylogeny of African pteropodids 83
to other ‘typical’ Epomops. While he suggested that
Nanonycteris veldkampii might be its closest rela-
tive, he also observed the similarities between
E. dobsonii and Epomophorus species in several
characters (postdental palate concavity, palatal
ridges, and the morphology of the pterygoid bone).
Our molecular phylogeny is in agreement with the
latter observations, recovering E. dobsonii within
the genus Epomophorus. Accordingly, we remove
dobsonii from Epomops and transfer the species to
Epo mo phorus. A third species of Epomops, E. buet-
tikoferi, is missing from our analysis.
The genus Epomophorus was revealed to be pa-
raphyletic not only with respect to E. dobsonii, but
also due to inclusion of Micropteropus pusillus
within Epomophorus in our trees. These results are
in agreement with a lack of differentiation between
Micropteropus pusillus and Epomophorus gam-
bianus at the Dloop region of the mitochondrial
DNA as reported by Nesi et al. (2011). Our ML and
MP combined trees (but not the BI) showed each of
the Epomophorus species as monophyletic. How -
ever, we noticed that both samples of each spe cies
have come from the same localities (Appendix),
while Nesi et al. (2011) had samples of M. pusillus
and E. gambianus from several localities across
a wide geographic range. The clade formed by the
typical species of Epomophorus, E. dobsonii, and
Micropteropus apparently diversified very recent-
ly, in the Pleistocene (Fig. 4). Within this clade,
M. pu sillus, E. gambianus, and E. minor apparently
split less than 0.5 Mya. Although morphological
differentiation has evolved within this group, given
the recent diversification of the clade, incomplete
lineage sorting may blur its phylogenetic relation-
ships. Another possible explanation for the observed
results is the introgression of mitochondrial DNA
from an Epomophorus species into M. pusillus fol-
lowing hybridization (Nesi et al., 2011). A better
sampling of rapidly evolving nuclear loci will be
necessary to distinguish between the two hypotheses
and decide whether Micro pteropus should be syn-
onymized with Epomo phorus.
The genus Epomophorus has five additional
species that could not be included in the present
analysis: E. angolensis, E. crypturus, E. grandis,
E. labiatus, E. minimus, and the recently described
E. anselli (Bergmans and van Strien, 2004; Sim -
mons, 2005). Epomophorous grandis was originally
described as a Micropteropus but was transferred to
Epomophorus by Bergmans (1988). The other spe -
cies have been linked to E. gambianus, although the
validity of some of them (e.g., E. crypturus) as dis-
tinct species has been a matter of controversy (Berg -
mans, 1988; Boulay and Robbins, 1989; Cla essen
and De Vree, 1991). Also missing from our analysis
is Micropteropus intermedius, another reason why
the possible synonymization of Micropte ro pus with
Epomophorus given their molecular affinities seems
The phylogenetic trees obtained here challenge
the subfamilial classification proposed by Bergmans
(1997). Neither Epomophorinae nor Rousettinae
sen su Bergmans (1997) are monophyletic, and Berg -
mans’ (1997) treatment of these taxa therefore needs
to be replaced by a new classification that takes into
account our new understanding of their phyloge-
netic relationships (Table 1 and Fig. 5; see Sup ple -
mentary Table S2 for a species level classification
84 F. C. Almeida, N. P. Giannini, and N. B. Simmons
of subfamily Rousettinae). We propose retaining the
name Rousettinae but revising its membership to in-
clude the entire clade composed of Scotonycterini
(including Scotonycteris and Casinycteris), Eonyc -
teris, Rousettus, Myonycterini (Megaloglossus and
Myonycteris), Plerotini (Plerotes), Stenonycterini
(Stenonycteris), and Epomophorini (Epomophorus,
Epomops, Nanonycteris, Hypsignathus, and Micro -
pteropus). This clade has received high statistical
support both in the analyses described in this contri-
bution and prior family-level phylogenetic analyses
published by our group (Almeida et al., 2011). The
genera Dobsonia and Aproteles had been previously
excluded from Rousettinae and placed in Har -
pyionycterinae together with Boneia and Harpyio -
nycteris (Giannini et al., 2006, 2009). The genus
Eidolon, also placed in Rousettinae by Bergmans
(1997), is explicitly excluded from this subfamily in
our new classification. Because Eidolon did not
show close relationships to any other pteropo-
did genus in Almeida et al. (2011), we propose the
erection of a new subfamily of its own, Eidolinae.
Eidolinae,New Subfamily
Type genus: Eidolon Rafinesque, 1815
Presently known to contain one genus and two
species: E. helvum (Kerr, 1792) and E. dupreanum
(Pollen, 1866).
A large pteropodid (FA 105–135) with a dental
formula of i2/2, c1/1, p3/3, m2/3 = 34. Length of
rostrum much greater than width across lacrimals;
anterior rim of orbit located above upper first molar;
palate much broader posteriorly than between ca-
nines; gap present anteriorly between right and left
premaxillae; basicranial axis moderately deflected
in relation to palate; tympanic elongated to form
a short, bony auditory meatus; occiput not elongated
and tubular; first upper premolar much larger in
cross section than upper incisors; first lower molar
equal in length to length of second and third molars
combined; claw present on wing digit II; short exter-
nal tail present; pelage sexually dimorphic with
males possessing a neck tuft and females often
conspicuously larger and paler than males. Also
based on results shown herein, we propose the erec-
tion of two new exclusive tribes for the genera Eo -
nyc teris and Rousettus: Eonycterini and Rou settini
respectively (Fig. 5). In this manner, the tribal clas-
sification for pteropodids will include exclusively
monophyletic groups.
The DIVA analysis produced results in accor-
dance with an Australasian (including the Pacific)
origin of Pteropodidae as previously proposed
(Butler, 1984; Aguilar et al., 1986; Hollar and
Springer, 1997; Juste et al., 1999; Almeida et al.,
2011) and indicated at least four independent colo-
nization events of Africa, represented by the line-
ages comprising (1) Eidolon, (2) African Rousettus,
(3) Scotonycterini, and (4) the endemic African
clade including Plerotes (Fig. 5). This result is in ac-
cordance with and adds to Juste et al.’s (1999) hy-
pothesis of at least three independent colonizations
of Africa; the fourth colonization we inferred is due
to the inclusion of scotonycterines, which were
missing from their study. Our divergence time esti-
mates suggest that these colonization events hap-
pened at different times (Fig. 4). The Scotonycterini
and the endemic African clade were apparently pres-
ent in Africa by the Late Miocene, when they started
to diversify locally. The time of their arrival in
Africa, however, is not so clear. Scotonycterini ap-
parently split from its sister group in the Early
Miocene, and the endemic African clade appeared
during the Middle Miocene. Whether at the time of
the split their ancestors were in Africa or elsewhere
(and these clades reached Africa after cladogenesis)
cannot be determined, although all their extant di-
versity is in Africa. Scotonycterini and members of
the endemic African clade are mostly forest/wood-
land dwellers. If their migration route were through
the Middle East, it would be more likely that it
took place before increases in aridity and decreases
in forest cover that occurred at end of the Mio-
cene (Za chos et al., 2001; Bonnefille, 2010). A land
bridge connecting Africa/Arabia and Eurasia, the
Gomphotherium landbridge, was apparently present
in this region since the Early Miocene (Harzhauser
et al., 2007).
A migration route involving island-hopping
through the Indian Ocean, as has been proposed in
the case of Malagasy Pteropus (O’Brien et al., 2009),
seems less realistic for most African pteropodids
given the smaller sizes of the scotonycterines and
the epomophorines and the distances involved (sev-
eral hundred kilometers separate some consecutive
islands), although their ancestors could have been
larger and had different flight capabilities. Cur rent -
ly, with the exception of Myonycteris brachycephala
A phylogeny of African pteropodids 85
from São Tomé and Príncipe, epomophorine species
are not found on islands off the coast of Africa.
Although an early African colonization is likely
in these clades, rapid diversification during the
Pliocene and Pleistocene apparently produced most
of the extant diversity. This period of rapid diversi-
fication appears to be correlated with climatic oscil-
lations during these periods (Bonne fille, 2010),
similar to those associated with the Northern Hemi -
sphere glaciations. This relatively recent diversifica-
tion had been previously noticed in Myonycterini
and Scotonycterini (Nesi et al., 2013; Hassanin,
2014); here we extend the observation to Epomo -
phorini. In particular, the genus Epomopho rus +
Micro pteropus apparently diversified into 12 spe-
cies in the last 2.5 million years — a remarkable
The African Rousettus clade originated in the late
Pliocene or early Pleistocene and diversified during
the Pleistocene (Fig. 4). The phylogenetic arrange-
ment recovered herein could have resulted from an
initial split between R. leschenaultii and the African
species’ ancestor somewhere in Central Asia, fol-
lowed by the colonization of mainland Africa by
a continental route through the Middle East, and
finally the origin of the island forms by island hop-
ping from continental Africa through Comoros and
into Madagascar (Goodman et al., 2010). However,
our topology is not totally incompatible with an
oceanic, island-hopping route from South Asia.
Interestingly, colonization of Africa by the Rou -
settus clade occurred during the same time period as
the independent colonization of the western Indian
Ocean islands from Asia by Pteropus species, which
occurred in at least three separate events (O’Brien et
al., 2009; Almeida et al., 2014). The ancestor of
African Rousettus might have used a similar route,
especially if environmental conditions at that time
were favorable for island hopping, oceanic dispersal
through the Indian Ocean. However, as already
mentioned, Rousettus species are not as large as
some Pteropus species, which makes oceanic migra-
tions less likely in Rousettus.
The continental R. aegyptiacus has a widespread
distribution from South Africa to Turkey and from
Guinea to western Pakistan (Bergmans, 1994). Its
current distribution in the Middle East could be re-
cent and due to anthropogenically facilitated colo-
nization from African stocks (Benda et al., 2012).
Apparently, the main required condition for the pres -
ence of this generalist species is a constant, year-
around fruit availability (Benda et al., 2012). Due
to anthropogenic fruit cultivation, this condition is
met in most areas of the Middle East where the
species now occurs. A recent colonization scenario
for the Middle East is in agreement with low levels
of population genetic differentiation in the
region (Benda et al., 2012). Besides R. aegyptiacus,
E. helvum and E. labiatus also inhabit the Arabic
peninsula, but they are restricted to its Afro-tropical
region, surrounding the Mandeb strait.
The time of the arrival of the genus Eidolon in
Africa is less clear. This genus appears to have split
from its sister clade early in pteropodid history
(Almeida et al., 2011), which would place its origin
in the Miocene or Oligocene. Eidolon includes two
species: E. dupreanum in Madagascar and E. helvum
in continental Africa. Eidolon helvum has a wide-
spread distribution in sub-Saharan Africa, reaching
the Red Sea coast of Yemen and Saudi Arabia, and
is a strong flier (Nowak, 1994). Eidolon helvum
col onies are known to make massive seasonal
migration, covering large distances (Thomas, 1983).
The low level of differentiation in E. helvum and
R. aegy ptiacus over a large geographic range sug-
gest ongoing gene flow between distant local popu-
lations of these species (Benda et al., 2012; Peel et
al., 2013; Shi et al., 2014).
Implications for Zoonotic Diseases
African fruit bats have recently been in the news
for their implication as most likely natural reservoir
of the Ebola Zaire virus. The Ebola virus is part of
the Filoviridae, which also includes the Marburg
viruses; these are among the deadliest human
pathogens (Wynne and Wang, 2013). Extrapolation
of our phylogenetic results strongly suggest hori-
zontal transfer of filoviruses since signs of infection
have been found in species that are not closely
related such as E. helvum and R. aegyptiacus. Most
Ebola outbreaks in Africa have occurred in equato-
rial forests (one exceptional out break happened
in South Sudan — Gatherer, 2014). Nevertheless,
some of the bats implicated as Ebola Zaire reser-
voirs, such as E. helvum and R. aegyptiacus, are also
found in areas outside the limits of African rain-
forests. This points to a potential risk of these bats
spreading the disease out of Equatorial Africa —
assuming that they are actually the reservoirs for
Ebola. Moreover, several species of African fruit
bats undergo annual migrations to savanna areas
in the wet season when fruits are available in abun-
dance (Thomas, 1983; Richter and Cumming,
2008). Serological studies have shown a low preva-
lence of viruses in bats; studies using large samples
86 F. C. Almeida, N. P. Giannini, and N. B. Simmons
showed that less than 10% have detectable virus or
antibodies (e.g., Leroy et al., 2005; Pourrut et al.,
2009; Olival et al., 2013). It is possible that out-
breaks correlate with periods of increased preva-
lence/virulence due to ecological factors such as
food scarcity and pregnancy (Leroy et al., 2005).
Moreover, food scarcity may increase the contact
between different animal species and even the
human ingestion of bats (Leroy et al., 2005, 2014).
Another important factor is the seasonal migration
of fruit bats or stochastic dispersal, which also in-
crease bat consumption by people in some areas
(Leroy et al., 2009). These factors need to be taken
into consideration in feature policies to control
Ebola and other viral diseases outbreaks in Africa
and elsewhere.
Contents: Supplementary Text. Rousettus specimen identifi-
cation problem; Table S1. GenBank accession numbers of the
samples analyzed in this study; Table S2. New classification
of subfamily Rousettinae at species level. Changes since the
last mammal list (Simmons, 2005) are underlined; Fig. S1.
Maximum-likelihood tree obtained with the RAG1 gene; Fig.
S2. Maximum likelihood tree obtained with the RAG2 gene;
Fig. S3. Maximum likelihood tree obtained with the vWF gene;
Fig. S4. Maximum likelihood tree obtained with the BRCA1
gene. A Rousettus amplexicaudatus sequence from GenBank
(accession no. AY057829) and a Myo nyc teris angolensis unpub-
lished sequence (voucher no. 177097, housed at the Field Mu -
seum of Natural History) were included in this analysis, but not
in the combined analyses; Fig. S5. Maximum likelihood tree
obtained with the Cytbgene; Fig. S6. Maximum likelihood tree
obtained with the mitochondrial DNA fragment containing the
12S-rRNA and 16S-rRNA concatenated genes; Fig. S7. Bay es -
ian Inference tree based on the combined dataset. Numbers on
nodes are posterior probabilities; Fig. S8. Maximum parsimony
tree based on the combined dataset. Numbers on nodes are boot-
strap values; Fig. S9. Maximum likelihood trees based on re-
duced dataset to minimize missing data, with A) inclusions and
B) exclusion of Plerotes anchietae. Numbers on nodes are boot-
strap values; Fig. S10. Maximum likelihood tree obtained for
the biogeographic analysis (DIVA) with bootstrap values; Fig.
S11. Maximum likelihood tree obtained with a fragment con-
taining the 12S-rRNA and 16S-rRNA genes and additional Rou -
set tus amplexicaudatus samples from Malaysia (FJ529130–
FJ529133), obtained from the GenBank (Guan et al., 2006). The
other R. amplexicaudatus sample is from Papua New Guinea
(Teeling et al., 2002). Supplementary Information is available
exclusively on BioOne.
For loan of tissues we thank J. F. Meads, B. Lim, T. Martin,
J. Patton, B. Patterson, and J. Wible. We are indebted to Rob
DeSalle for kindly allowing us to use the facilities of the molec-
ular laboratory of the Sackler Institute of Comparative Ge -
nomics (AMNH). This contribution was based upon work
supported in part by the National Science Foundation under
Grant No. DEB-9873663 and by the National Institutes of
Health under Grant No. R21 AI105050 to N. B. Simmons, and
a Vernay Postdoctoral Fellow ship to F. C. Almeida (AMNH). F.
C. Al mei da and N. P. Giannini were supported by the CONICET
(Na tional Scientific and Technical Research Council, Argen -
tina) during the writing of this manuscript.
CHAUX, and B. SIGÉ. 1986. Première occurrence d’un mé-
gachiroptère ptéropodidé dans le Miocène moyen d’Europe
(gisement de Lo Fournas-II. Pyrénées-Orientales, France).
Palaeovertebrata, 16(3): 173–184.
V. M. DATO , M. C. EISENBERG, and S. EUBANK. 2015. What
factors might have led to the emergence of Ebola in West
Af rica? PLoS Neglected Tropical Diseases, 9: e0003652.
MONS. 2009. The phylogenetic relationships of cynopterine
fruit bats (Chiroptera: Pteropodidae: Cynopterinae). Molec -
ular Phylogenetics and Evolution, 53: 772–783.
MONS. 2011. Evolutionary relationships of the Old World
fruit bats (Chiroptera, Pteropodidae): another star phylo -
geny? BMC Evolutionary Biology, 11(281): 1–17.
HELGEN. 2014. Each flying fox on its own branch: A phylo-
genetic tree for Pteropus and related genera (Chiroptera:
Pteropodidae). Molecular Phylogenetics and Evolution, 77:
C. IBÁÑEZ, and J. M. BAUTISTA. 1999. Molecular phylogeny
and morphological homoplasy in fruitbats. Molecular Biol -
ogy and Evolution, 16: 1061–1067.
ANDERSEN, K. 1912. Catalogue of the Chiroptera in the col-
lection of the British Museum, 2nd edition. Volume 1:
Mega chiroptera. Trustees British Museum, London, ci +
854 pp.
BENDA, P., P. VALLO, P. HULVA, and I. HORÁČEK. 2012. The
Egyp tian fruit bat Rousettus aegyptiacus (Chiroptera: Pte -
ropodidae) in the Palaearctic: Geographical variation and
taxonomic status. Biologia (Bratislava), 67: 1230–1244.
BERGMANS, W. 1988. Taxonomy and biogeography of African
fruit bats (Mammalia, Megachiroptera) 1: General introduc-
tion; material and methods; results: the genus Epomo pho rus
Bennett, 1836. Beaufortia, 38(5): 75–146.
BERGMANS, W. 1989. Taxonomy and biogeography of African
fruit bats (Mammalia, Megachiroptera) 2: The genera Mi -
cro pteropus Matschie, 1899, Epomops Gray, 1870, Hypsi -
gna thus H Allen, 1861, Nanonycteris Matschie, 1899, and
Plerotes Andersen, 1910. Beaufortia, 39(4): 89–153.
BERGMANS, W. 1994. Taxonomy and biogeography of African
fruitbats (Mammalia, Megachiroptera) 4: The genus Rou set -
tus Gray, 1821. Beaufortia, 44: 79–126.
BERGMANS, W. 1997. Taxonomy and biogeography of African
fruit bats (Mammalia, Megachiroptera) 5: The genera Lisso -
nycteris Andersen, 1912, Myonycteris Matschie, 1899 and
Me galoglossus Pagenstecher, 1885; General remarks and
conclusions; Annex: key to all species. Beaufortia, 47(2):
A phylogeny of African pteropodids 87
BERGMANS, W., and F. G. ROZENDAAL. 1988. Notes on collec-
tions of fruit bats from Sulawesi and some off-lying islands
(Mammalia, Megachiroptera). Zoologische Verhandelingen,
248: 1–74.
BERGMANS, W., and N. J. VAN STRIEN. 2004. Systematic notes on
a collection of bats from Malawi, I. Megachiroptera: Epo -
mophorinae and Rousettinae (Mammalia, Chiroptera). Acta
Chiropterologica, 6: 249–268.
BONNEFILLE, R. 2010. Cenozoic vegetation, climate changes and
hominid evolution in tropical Africa. Global and Planetary
Change, 72: 390–411.
MOND. 2014. BEAST 2: a software platform for Bayesian
evolutionary analysis. PLoS Computational Biology, 10:
BOULAY, M. C., and C. B. ROBBINS. 1989. Epomophorus gam-
bianus. Mammalian Species, 344: 1–5.
BUTLER, P. M. 1984. Macroscelidea, Insectivora and Chiroptera
from the Miocene of East Africa. Paleovertebrata, 14(3):
CLAESSEN, C. J., and F. DEVREE. 1991. Systematic and taxo-
nomic notes on the Epomophorus anurus-labiatus-minor
complex with the description of a new species (Mammalia:
Chiroptera: Pteropodidae). Senckenbergiana Biologica, 71
(4/6): 209–238. [for 1990].
DANOV, A. SEEBENS, et al. 2012. Bats host major mammalian
paramyxoviruses. Nature Communications, 3: 796.
NINGHAM. 2008. Henipavirus infection in fruit bats (Pte -
ropus giganteus), India. Emerging Infectious Diseases, 14:
FARRIS, J. S., M. KÄLLERSJÖ, A. G. KLUGE, and C. BULT. 1994.
Testing significance of incongruence. Cladistics, 10:
GATHERER, D. 2014. The 2014 Ebola virus disease outbreak in
West Africa. Journal of General Virology, 95: 1619–1624.
GIANNINI, N. P., and N. B. SIMMONS. 2003. A phylogeny of
megachiropteran bats (Mammalia: Chiroptera: Pteropo di -
dae) based on direct optimization analysis of one nuclear
and four mitochondrial genes. Cladistics, 19: 496–511.
GIANNINI, N. P., and N. B. SIMMONS. 2005 Conflict and congru-
ence in a combined DNA-morphology analysis of megachi-
ropteran bat relationships (Mammalia: Chiroptera: Pteropo -
didae). Cladistics, 21: 411–437.
SAL LE. 2006. Phylogenetic relationships of the enigmatic
harpy fruit bat, Harpyionycteris (Mammalia: Chiroptera:
Pteropo di dae). American Museum Novitates, 3533: 1–12.
GIANNINI, N. P., F. C. ALMEIDA, and N. B. SIMMONS. 2009. Phy -
logenetic relationships of Harpyionycterine megabats (Chi -
roptera: Pteropodidae). In Systematic mam malogy: contri-
butions in honor of Guy G Musser (R. S. VOSS and M. D.
CARLETON, eds.). Bulletin of the Amer ican Museum of
Natural History, 331: 183–204.
GOLOBOFF, P. A., J. S. FARRIS, and K. C. NIXON. 2008. TNT,
a free program for phylogenetic analysis. Cladistics, 24:
GOODMAN, S. M., C. M. LAUREN, M. D. NOWAK, andA. D. YO -
DER. 2010. Phylogeny and biogeography of western In dian
Ocean Rousettus (Chiroptera: Pteropodidae). Journal of
Mam malogy, 91: 593–606.
GUAN, A. K. H., Y. ESA, A. A. SALLEHIN, J. R. RYA N, A. M.
JULAIHI, J. V. KUMARAN, and M. T. ABDULLAH. 2006. Phylo -
genetic relationships of fruit bats (Family: Pteropodidae) in
Malaysia inferred from partial mtDNA Cytochrome bgene.
Proceedings of the 1st International Conference on Natural
Resources Engineering & Technology, 2006: 77–97.
E. FIELD, et al. 2011. Pteropodid bats are confirmed as the
reservoir hosts of Henipaviruses: a comprehensive experi-
mental study of virus transmission. American Journal of
Tropical Medicine and Hygene, 85: 946–951.
LICH, M. REUTER, and B. BERNING. 2007. Biogeographic
responses to geodynamics: a key study all around the Oligo-
Miocene Tethyan Seaway. Zoologischer Anzeiger, 246:
HASSANIN, A. 2014. Description of a new bat species of the tribe
Scotonycterini (Chiroptera, Pteropodidae) from Southwest -
ern Cameroon. Comptes Rendus Biologies, 337: 134–142.
2015. The comparative phylogeography of fruit bats of the
tribe Scotonycterini (Chiroptera, Pteropodidae) reveals
cryptic species diversity related to African Pleistocene for-
est refugia. Comptes Rendus Biologies, 338: 197–211.
2010. Long-term survival of an urban fruit bat seropositive
for Ebola and Lagos bat viruses. PLoS ONE, 5: e11978.
HOLLAR, L. J., and M. S. SPRINGER. 1997. Old World fruit bat
phylogeny: evidence for convergent evolution and an en-
demic African clade. Proceedings of the National Academy
of Science of the USA, 94: 5716–5721.
Molecular architecture of Pipistrellus pipistrellus/Pipi strel -
lus pygmaeus complex (Chiroptera: Vespertilionidae):
further cryptic species and Mediterranean origin of the
divergen ce. Molecular Phylogenetics and Evolution, 32:
C. IBÁÑEZ, and J. M. BAUTISTA. 1999. Phylogeography of
African fruit bats (Megachiroptera). Molecular Phylogenet -
ics and Evolution, 13: 596–604.
KATO H , K., and D. M. STANDLEY. 2013. MAFFT multiple se-
quence alignment software version 7: improvements in per-
formance and usability. Molecular Biology and Evolution,
30: 772–780.
KUNZ, T. H., and E. D. PIERSON. 1994. Bats of the World: an in-
troduction. Pp. 1–46, in Walker’s bats of the World (R. M.
NOWAK, ed.). The Johns Hopkins University Press, Balti -
more, Maryland, vi + 288 pp.
2010. Natural hybridization generates mammalian lineage
with species characteristics. Proceedings of the National
Academy of Sciences of the USA, 107: 11447–11452.
GON ZALEZ, and R. SWANEPOEL. 2005. Fruit bats as reservoirs
of Ebola virus. Nature, 438(7068): 575–576.
2009. Human Ebola outbreak resulting from direct exposure
to fruit bats in Luebo, Democratic Republic of Congo, 2007.
Vector Borne Zoonotic Diseases, 9: 723–728.
2014. Ebola in West Africa: The outbreak able to change
many things. Clinical Microbiology and Infection, 20(10):
LUNDE, D. P., G. G. MUSSER, and N. T. SON. 2003. A survey of
small mammals from Mt Tay Con Linh II, Vietnam, with the
description of a new species of Chodsigoa (Insectivora: So -
ri cidae). Mammal Study, 28: 31–46.
DNA barcoding of African fruit bats (Mammalia, Ptero po -
didae): the mitochondrial genome does not provide a reli-
able discrimination between Epomophorus gambianus and
Micropteropus pusillus. Comptes Rendus Biologies, 334:
SHON GO, C. CRUAUD, and A. HASSANIN. 2013. Molecular
systematics and phylogeography of the tribe Myonycterini
(Mammalia, Pteropodidae) inferred from mitochondrial and
nuclear markers. Molecular Phylogenetics and Evolution,
66: 126–137.
NOWAK, R. M. 1994. Walker’s bats of the World. The Johns
Hopk ins University Press, Baltimore, vi + 288 pp.
D. YODER, and T. J. HAYDEN. 2009. Multiple colonisations
of the western Indian Ocean by Pteropus fruit bats (Me -
gachiroptera: Pteropodidae): the furthest islands were
colonised first. Molecular Phylogenetics and Evolution, 51:
KIN, et al. 2013. Ebola virus antibodies in fruit bats, Ban-
g ladesh. Emerging Infectious Dis eases, 19: 270–273.
L.-F. WANG, et al. 2013. Continent-wide panmixia of an
African fruit bat facilitates transmission of potentially
zoonotic viruses. Nature Communications, 4: 3770.
NICHOL, J. P. GONZALEZ, and E. LEROY. 2009. Large serolog-
ical survey showing cocirculation of Ebola and Marburg
viruses in Gabonese bat populations, and a high seropreva-
lence of both viruses in Rousettus aegyptiacus. BMC In -
fectious Diseases, 9: 159.
RAMBAUT, A. 2009. FigTree. Available at
2014. Tracer v 1.6. Available at
KINS, J. P. G. JONES, and J.-M. HERAUD. 2015. Detection of
new genetic variants of Betacoronaviruses in endemic fru-
givorous bats of Madagascar. Virology Journal, 12: 42.
RICHTER, H. V., and G. S. CUMMING. 2008. First application of
satellite telemetry to track African straw-colored fruit bat
migration. Journal of Zoology (London), 275: 172–176.
RONQUIST, F. 1997. Dispersal-vicariance analysis: a new ap-
proach to the quantification of historical biogeography. Sys -
tematic Biology, 46: 195–203.
RONQUIST, F., and J. P. HUELSENBECK. 2003. MrBayes 3: Bayes -
ian phylogenetic inference under mixed models. Bioinfor -
matics, 19: 1572–1574.
RUEDI, M., and F. MAYER. 2001. Molecular systematics of bats
of the genus Myotis (Vespertilionidae) suggests determinis-
tic ecomorphological convergences. Molecular Phylogenet -
ics and Evolution, 21: 436–448.
N. WEBER, and K. MATUSCHEWSKI. 2013. High diversity of
West African bat malaria parasites and a tight link with ro-
dent Plasmodium taxa. Proceedings of the National Acad -
emy of Sciences of the USA, 110: 17415–17419.
SHI, J. J., L. M. CHAN, A. J. PEEL, R. LAI, A. D. YODER, and
S. GOODMAN. 2014. A deep divergence time between sister
species of Eidolon (Pteropodidae) with evidence for wide-
spread panmixia. Acta Chiropterologica, 16: 279–292.
SHIMODAIRA, H., and M. HASEGAWA. 1999. Multiple com-
parisons of log-likelihoods with applications to phyloge-
netic inference. Molecular Biology and Evolution, 16:
SIMMONS, N. B. 2005. Order Chiroptera. Pp. 312–529, in Mam -
mal species of the World: a taxonomic and geographic ref-
erence (D. E. WILSON and D. M. REEDER, eds.). The Johns
Hopkins University Press, Baltimore, MD, 1242 pp.
STAMATAKIS, A. 2014. RAxML version 8: a tool for phyloge-
netic analysis and post-analysis of large phylogenies. Bio -
informatics, 30: 1312–1313.
SWOFFORD, D. L. 2002. PAUP*: Phylogenetic Analysis Using
Parsimony (* and other methods) Version 4. Sinauer Asso -
ciates, Sunderland, Massachusetts.
JONG, M. J. STANHOPE, and M. S. SPRINGER. 2002. Microbat
paraphyly and the convergent evolution of a key innovation
in Old World rhinolophoid microbats. Proceeding of the
National Academy of Science of the USA, 99: 1431–1436.
THOMAS, D. W. 1983. The annual migrations of three species of
West African fruit bats (Chiroptera: Pteropodidae). Cana -
dian Journal of Zoology, 61: 2266–2272.
VOGEL, G. 2014. Are bats spreading Ebola across Sub-Saharan
Africa? Science, 344(6180): 140.
WIENS, J. J. 2006. Missing data and the design of phylogenetic
analyses. Journal Biomedical Informatics, 39: 34–42.
WYNNE, J. W., and L.-F. WANG. 2013. Bats and viruses: friend
or foe? PLoS Pathogens, 9: e1003651.
YU, Y., A. J. HARRIS, C. BLAIR, and X. J. HE. 2015. RASP (Re -
con struct Ancestral State in Phylogenies): a tool for histori-
cal biogeography. Molecular Phylogenetics and Evolution,
87: 46–49.
2001. Trends, rhythms, and aberrations in global climate 65
Ma to present. Science, 292: 686–693.
88 F. C. Almeida, N. P. Giannini, and N. B. Simmons
Received 22 August 2015, accepted 19 October 2015
A phylogeny of African pteropodids 89
African and related taxa samples used in this study, with their voucher numbers, source, locality, and bibliographic reference in the case of previously published sequences.
Numbers after species names are references for the terminals in the illustrated phylogenies (Figs. 1–3 and Supplementary Figs. S1 –S11)
Species Voucher SourceaLocality Reference
Casinycteris argynnis 269915 AMNH Central Africa Republic, Dzanga Sangha Forest Reserve Almeida et al. (2011)
Casinycteris campomaanensis 2011-637 MNHN Cameroon, Campo Ma’an area, Village of Nkoe ́lon-Mvini Hassanin (2014)
Eidolon helvum 102021 CMNH Kenya, Western Province, Kakamega Dist Giannini et al. (2008, 2009)
Eonycteris major TK152179 ? Malaysia, Sarawak, Kubah National Park unpublished
Eonycteris robusta 178362 FMNH Philippines, Luzon, Mt Banahaw, Barangay Lalo Almeida et al. (2011)
Eonycteris spelaea 192353 CMNH India, Andhra Pradesh, Boora Cave, This study
Eonycteris spelaea 2176480 MVZ China, Yunnan Prov., Yunnan Institute of Tropical Botany This study
Eonycteris spelaea 3 176487 MVZ China, Yunnan Prov., Yunnan Institute of Tropical Botany Almeida et al. (2011)
Epomophorus gambianus 1113550 CMNH Ghana, Northern Region This study
Epomophorus gambianus 2113553 CMNH Ghana, Northern Region This study
Epomophorus minor 1102026 CMNH Kenya, Rift Valey Prov., West Pokot Dist., Sigor This study
Epomophorus minor 2102027 CMNH Kenya, Rift Valey Prov., West Pokot Dist., Sigor This study
Epomophorus wahlbergi 1 177209 FMNH Mozambique, Zambezia, Murabue Almeida et al. (2011)
Epomophorus wahlbergi 2177089 FMNH Mozambique, Zambezia, Murabue This study
Epomops dobsonii 115821 AMNH Zambia, Balovale This study
Epomops franqueti 1 269902 AMNH Central Africa Republic, Dzanga Sangha Forest Reserve Almeida et al. (2011)
Epomops franqueti 2268356 AMNH Central Africa Republic, Dzanga Sangha Forest Reserve This study
Hypsignathus monstrosus 116499 AMCC ? Almeida et al. (2011)
Megaloglossus azagnyi 1 2011-1001 MNHN Ivory Coast, Gboyo village Nesi et al (2013)
Megaloglossus azagnyi 2 G09175 n.a. Liberia, Gangra Nesi et al. (2013)
Megaloglossus woermanni 1 268358 AMNH Central Africa Republic, Dzanga Sangha Forest Reserve Almeida et al. (2011)
Megaloglossus woermanni 2268360 AMNH Central Africa Republic, Dzanga Sangha Forest Reserve This study
Micropteropus pusillus 113563 CMNH Ghana, Northern Region Almeida et al (2011)
Myonycteris angolensis 1 102135 CMNH Kenya, Western Province, Kakamega Dist Almeida et al (2011)
Myonycteris angolensis 2102136 CMNH Kenya, Western Province, Kakamega Dist This study
Myonycteris brachycephala 1 104 ST São Tomé, Monte Belo Nesi et al. (2013)
Myonycteris brachycephala 2 105 ST São Tomé, Monte Belo Nesi et al. (2013)
Myonycteris brachycephala 2 ? ? ? Juste et al. (1999)
Myonycteris relicta 171299 FMNH Tanzania, Mts Usumbara Nesi et al. (2013)
Myonycteris relicta ? ? ? Juste et al. (1999)
Myonycteris leptodon 1 G09196 n.a. Cote d’Ivoire, Besso Nesi et al. (2013)
Myonycteris leptodon 2 G09189 n.a. Liberia, East Nimba Nesi et al. (2013)
Myonycteris torquata 1 268362 AMNH Central Africa Republic, Dzanga Sangha Forest Reserve Almeida et al. (2011)
Myonycteris torquata 2268363 AMNH Central Africa Republic, Dzanga Sangha Forest Reserve This study
Nanonycteris veldkampii 1 F34376 ROM Ivory Coast, Parc National de Mont Peko Almeida et al. (2011)
Nanonycteris veldkampii 2F34377 ROM Ivory Coast, Parc National de Mont Peko This study
Plerotes anchietae 185744 SMF Malawi, 1,700 m altitude This study
Plerotes anchietae 285745 SMF Malawi, 1,700 m altitude This study
90 F. C. Almeida, N. P. Giannini, and N. B. Simmons
APPENDIX. Continued
Species Voucher SourceaLocality Reference
Rousettus sp. 1 274231 AMNH Vietnam, Ha Giang, Mount Tay Con Linh 2 This study
Rousettus sp. 2 CMF980124 ROM Laos, Khammouane, 2 km N of Ban Mouangkai This study
Rousettus amplexicaudatus 23045/6 AM Papua New Guinea Teeling et al. (2002)
Rousettus aegyptiacus 1177211 FMNH Mozambique, Zambezia, Murabu This study
Rousettus aegyptiacus 2 ? ? ? unpublished
Rousettus leschenaultii 1 176490 MVZ China, Yunnan Prov., 1 km E Menglung Almeida et al. (2011)
Rousettus leschenaultii 2 A8 ? China, Yunnan Li et al. (2007)
Rousettus leschenaultii 2 ? ? ? unpublished
Rousettus madagascariensis 1448922 NMNH Madagascar This study
Rousettus madagascariensis 2 449206 NMNH Madagascar Almeida et al. (2011)
Rousettus obliviosus 1 194543 FMNH Comoros Is. Goodman et al. (2010)
Rousettus obliviosus 2 194459 FMNH Comoros Is. Goodman et al. (2010)
Rousettus spinalatus 1 16 ? Malaysia, Liam Village, Sarawak Guan et al. (2006)
Rousettus spinalatus 2 SNP41 ? Malaysia, Similajau N.P., Sarawak Guan et al. (2006)
Casinycteris ophiodon 256534 AMNH Liberia, Gran Gedeh, Dugbe River This study
Casinycteris ophiodon 50001 ZMB Cameroon, Bipindi Hassanin, (2014)
Scotonycteris zenkeri 1 107997 CMNH Cameroon, Southwest Prov., Bake River Bridge Almeida et al. (2011)
Scotonycteris zenkeri 2107999 CMNH Cameroon, Southwest Prov., Korup Natl Park This study
Stenonycteris lanosus 1 102148 CMNH Kenya, Western Province, Kakamega Dist Almeida et al. (2011)
Stenonycteris lanosus 2102149 CMNH Kenya, Western Province, Kakamega Dist This study
a. — AMNH: American Museum of Natural History, MNHN: Muséum National d’Histoire Naturelle, CMNH: Carnegie Museum of Natural History, FMNH: Field Museum of Natural
History, MVZ: Museum of Vertebrate Zoology, AMCC: Ambrose Monell Cryo Collection, ST: Collection of Pr. Javier Juste Ballesta, Estación Biológica de Doñana, ROM: Royal Ontario
Museum, SMF: Senckenberg Forschungsinstitut und Naturmuseum, AM: Australian Museum, NMNH: National Museum of Natural History, ZMB: Zoologisches Museum Berlin, n.a.:
sample was not deposited in a museum collection
... Pteropodidae are older than Phyllostomidae, dating from approximately 39 Ma (Teeling et al. 2005). Some lineages of fruit bats from the Indo-Pacific region date from the Oligocene at approximately 31 Ma, with diversification events from the Miocene to the Pleistocene (Almeida et al. 2009(Almeida et al. , 2016. Recent phylogenetic studies indicate that Pteropodidae probably repeatedly colonized Africa from Asian ancestors (Almeida et al. 2016). ...
... Some lineages of fruit bats from the Indo-Pacific region date from the Oligocene at approximately 31 Ma, with diversification events from the Miocene to the Pleistocene (Almeida et al. 2009(Almeida et al. , 2016. Recent phylogenetic studies indicate that Pteropodidae probably repeatedly colonized Africa from Asian ancestors (Almeida et al. 2016). Estimates of divergence dates suggest that these events occurred in different periods and that although local diversification appears to have started in the late Miocene, the more extensive diversification that produced the modern fauna occurred much later, in the Pleistocene (Almeida et al. 2016). ...
... Recent phylogenetic studies indicate that Pteropodidae probably repeatedly colonized Africa from Asian ancestors (Almeida et al. 2016). Estimates of divergence dates suggest that these events occurred in different periods and that although local diversification appears to have started in the late Miocene, the more extensive diversification that produced the modern fauna occurred much later, in the Pleistocene (Almeida et al. 2016). ...
Full-text available
Background and aims – Pollination systems often reflect adaptations to specific groups of pollinators, and these morphological specialisations have been important in the diversification of the angiosperms. Here, we study the evolution of the capitulum and pollination system in the pantropical genus Parkia, which comprises 35 species of trees distributed largely in the forests of South and Central America, Africa, Madagascar, and the Indo-Pacific. The flowers are grouped into capitula that are composed of one, two, or three distinct morphological types, and are principally pollinated either by insects or by bats. Material and methods – Using BEAST, we estimated the ages of nodes in a phylogeny based on four chloroplast regions (matK, trnL, psbA-trnH, and rps16-trnQ) and the nuclear region ITS/18S/26S. This analysis also enabled us to reconstruct the ancestral state of the capitulum and hence infer the ancestral pollination system. Euclidean distance-based cluster analysis was performed to determine which characters are consistently related to a specific pollination system.Key results – Our results indicate that the ancestral capitulum in the genus had three types of flowers and a morphology associated with bat-pollination in both the Paleotropics and Neotropics. In one derived Neotropical clade, the number of floral types in each capitulum was reduced to two (capitulum also bat-pollinated) or one (insect-pollinated). Thus, entomophily, as seen in some Neotropical species of Parkia, has been derived from a bat-pollinated ancestor. Cluster analysis showed that the floral characters were mostly consistent with pollination systems.Conclusion – Chiropterophily is not an evolutionary dead end in Parkia because during the evolutionary history of the genus there has been at least one transition to entomophily. Parkia provides a unique example of evolutionary transitions from chiropterophily to entomophily in a pantropical genus of trees.
... The African straw-coloured fruit bat, Eidolon helvum, is a megabat that is widely distributed in sub-Saharan Africa [3]. The bats are nocturnal, live in large colonies, and can be found roosting on trees close to human habitation [27]. ...
... Their frugivorous, arboreal and migratory nature enables them to function in plant pollination and plant geographical distribution [30]. The species has been identified as a natural reservoir of a number of zoonotic viral diseases such as those caused by the Ebola subtype Zaire and Lagos bat viruses [3]. Their close proximity and association with humans create a need for a comprehensive body of knowledge on their biology especially since they have been associated with some epidemics. ...
... Its numerous projections, the choroidal papillae, projected markedly perpendicularly or obliquely towards the retina in the direction of the pupil (Figs. [3][4][5]. These papillae caused undulations in the retinal tissues. ...
Full-text available
Background: This work was designed to provide a morphologic, morphometric and histochemical description of the eye of the African straw-coloured fruit bat (Eidolon helvum). An explanation of the optical role of the choroidal papillae in the vision of megachiropteran bats was provided. Materials and methods: Enucleated eyes of captured fruit bats were measured and processed for light microscopy. Results: Typical gross features of the mammalian eye including an anterior transparent cornea, posterior whitish sclera and a golden-brown iris surrounding a round pupil were observed in the eye. Presence of undulating retina typically found in megachiropterans were also seen. The ratio of mean corneal diameter to mean axial eye diameter was 0.58 ± 0.08. The histochemical investigation of the eye indicated the presence of mucins, proteoglycans, hyaluronic acid, glycogen and/or glycoproteins in the corneal, scleral, choroidal and retinal tissues. Conclusions: The presence of reflective materials of the tapetum lucidum on the undulating retina was shown to be a morphological adaptation for increased light sensitivity as each parabolic surface of the choroidal papillae served as a convex mirror, reflecting the light rays to the adjacent parabolic surface, thus sensitizing photoreceptors in affected regions. This phenomenon thus empowers megachiropteran bats with improved scotopic visual capability and could explain why most of them are reliant on their vison without the need for echolocation.
... Rousettus amplexicaudatus is the most basal species of the genus and R. spinalatus is the most closed sister species of R. amplexicaudatus. Almeida et al. (2016) showed that R. spinalatus was the first species splitting from R. amplexicaudatus, and this can explain the basal place of R. spinalatus with R. amplexicaudatus. Recent published information (Almeida et al., 2016;Hassanin et al., 2019;Vogeler and Tschapka, 2021) indicates that R. lanosus falls outside the Rousettus clade and best placed in another genus; and here, mitochondrial sequence data supports this proposition. ...
... Almeida et al. (2016) showed that R. spinalatus was the first species splitting from R. amplexicaudatus, and this can explain the basal place of R. spinalatus with R. amplexicaudatus. Recent published information (Almeida et al., 2016;Hassanin et al., 2019;Vogeler and Tschapka, 2021) indicates that R. lanosus falls outside the Rousettus clade and best placed in another genus; and here, mitochondrial sequence data supports this proposition. There is no available sequence data for R. celebensis and R. linduensis. ...
Full-text available
The Old World tropical and subtropical frugivorous bat genus Rousettus (Pteropodidae) contains species with broad distributions, as well as those occurring in restricted geographical areas, particularly islands. Herein we review the role of Rousettus as a keystone species from a global “One Health” approach and related to ecosystem functioning, zoonotic disease and public health. Rousettus are efficient at dispersing seeds and pollinating flowers; their role in forest regeneration is related to their ability to fly considerable distances during nightly foraging bouts and their relatively small body size, which allows them to access fruits in forested areas with closed vegetation. Rousettus are also reservoirs for various groups of pathogens (viruses, bacteria, fungi, protozoa), which, by definition, are infectious agents causing disease. The study of day roosts of different species of Rousettus and the successful establishment of captive breeding colonies have provided important details related to the infection dynamics of their associated pathogens. Large-scale conversion of forested areas into agricultural landscapes has increased contact between humans and Rousettus, therefore augmenting the chances of infectious agent spillover. Many crucial scientific details are still lacking related to members of this genus, which have direct bearing on the prevention of emerging disease outbreaks, as well as the conservation of these bats. The public should be better informed on the capacity of fruit bats as keystone species for large scale forest regeneration and in spreading pathogens. Precise details on the transmission of zoonotic diseases of public health importance associated with Rousettus should be given high priority.
... Cheiromeles torquatus, which did not appear on this tree, was represented by Otomops c.f. formosus, the tip within family Molossidae (node 1027) closest to the mean branch length for that clade. Similarly, the missing taxon Eidolon helvum is a sister to the Cynopterus clade (Almeida, Giannini, & Simmons, 2016) and was represented by Dyacopterus spadiceus, the tip closest to mean branch length within this clade (node 671). After trimming the tree to our study taxa, package phytools (Revell, 2012) was employed to make the tree ultrametric (all tips have the same total length to the root) ( Figure 2). ...
Full-text available
Bats are the only mammals to have achieved powered flight. A key innovation allowing for bats to conquer the skies was a forelimb modified into a flexible wing. The wing bones of bats are exceptionally long and dynamically bend with wingbeats. Bone microarchitectural features supporting these novel performance attributes are still largely unknown. The humeri and femora of bats are typically avascular, with the exception of large‐bodied taxa (e.g., pteropodid flying foxes). No thorough investigation of vascular canal regionalization and morphology has been undertaken as historically it has been difficult to reconstruct the 3D architecture of these canals. This study augments our understanding of the vascular networks supporting the bone matrix of a sample of bats (n=24) of variable body mass, representing three families (Pteropodidae (large‐bodied, Species=6), Phyllostomidae (medium‐bodied, Species=2), and Molossidae (medium‐bodied, Species=1)). We employed Synchrotron Radiation‐based micro‐Computed Tomography (SRμCT) to allow for a detailed comparison of canal morphology within humeri and femora. Results indicated that across selected bats, canal number per unit volume is similar independent of body size. Differences in canal morphometry based on body size and bone type appear primarily related to a broader distribution of the canal network as cortical volume increases. Heavier bats display a relatively rich vascular network of mostly longitudinally‐oriented canals that are localized mainly to the mid‐cortical and endosteal bone envelopes. Taken together, our results suggest that relative vascularity of the limb bones of heavier bats forms support for nutrient exchange in a regional pattern. This article is protected by copyright. All rights reserved.
... The range of henipaviruses including Hendra [23], Cedar [24], and others [25][26][27] extends throughout the geographic range of pteropodid bats to Australia, Indian Ocean islands, and sub-Saharan Africa [28]. These data, combined with limited evidence of pathology in henipavirus-infected bats [29,30], suggest that henipaviruses have had a long association with their bat reservoirs that spans the dispersal of pteropodid bats out of Southeast Asia to other regions [31][32][33][34][35]. ...
Full-text available
Nipah virus is a bat-borne paramyxovirus that produces yearly outbreaks of fatal encephalitis in Bangladesh. Understanding the ecological conditions that lead to spillover from bats to humans can assist in designing effective interventions. To investigate the current and historical processes that drive Nipah spillover in Bangladesh, we analyzed the relationship among spillover events and climatic conditions, the spatial distribution and size of Pteropus medius roosts, and patterns of land-use change in Bangladesh over the last 300 years. We found that 53% of annual variation in winter spillovers is explained by winter temperature, which may affect bat behavior, physiology, and human risk behaviors. We infer from changes in forest cover that a progressive shift in bat roosting behavior occurred over hundreds of years, producing the current system where a majority of P. medius populations are small (median of 150 bats), occupy roost sites for 10 years or more, live in areas of high human population density, and opportunistically feed on cultivated food resources—conditions that promote viral spillover. Without interventions, continuing anthropogenic pressure on bat populations similar to what has occurred in Bangladesh could result in more regular spillovers of other bat viruses, including Hendra and Ebola viruses.
... Based on genetic records, it is suspected that the megachiropterans evolved from Australia or Melanesian Island. The lineage eventually dispersed to mainland Asia, the Mediterranean, and Africa (Almeida et al. 2016). Today, you can find megachiropterans in the subtropical parts of Eurasia, Africa, and Oceania (Almeida et al. 2011). ...
... S1 identity is lower, but globally S1 appears to be less constrained than the remaining segments across PRVs. That no segments of PRV16K appear as an outlier to an Asian clade of PRVs ( Figure 2) suggests that the virus did not co-diverge with African fruit bats from their ancestral Asian lineages in the Miocene [39], but rather spread between African and Asian bats more recently. The Angolan soft-furred bats are split into five subspecies. ...
Full-text available
Pteropine orthoreovirus (PRV; Reoviridae: Spinareovirinae) is an emerging bat-borne zoonotic virus that causes influenza-like illness (ILI). PRV has thus far been found only in Australia and Asia, where diverse old-world fruit bats (Pteropodidae) serve as hosts. In this study, we report the discovery of PRV in Africa, in an Angolan soft-furred fruit bat (Lissonycteris angolensis ruwenzorii) from Bundibugyo District, Uganda. Metagenomic characterization of a rectal swab yielded 10 dsRNA genome segments, revealing this virus to cluster within the known diversity of PRV variants detected in bats and humans in Southeast Asia. Phylogeographic analyses revealed a correlation between geographic distance and genetic divergence of PRVs globally, which suggests a geographic continuum of PRV diversity spanning Southeast Asia to sub-Saharan Africa. The discovery of PRV in an African bat dramatically expands the geographic range of this zoonotic virus and warrants further surveillance for PRVs outside of Southeast Asia.
Full-text available
The family Pteropodidae (Old World fruit bats) comprises >200 species distributed across the Old World tropics and subtropics. Most pteropodids feed on fruit, suggesting an early origin of frugivory, although several lineages have shifted to nectar-based diets. Pteropodids are of exceptional conservation concern with >50% of species considered threatened, yet the systematics of this group has long been debated, with uncertainty surrounding early splits attributed to an ancient rapid diversification. Resolving the relationships among the main pteropodid lineages is essential if we are to fully understand their evolutionary distinctiveness, and the extent to which these bats have transitioned to nectar-feeding. Here we generated orthologous sequences for >1400 nuclear protein-coding genes (2.8 million base-pairs) across 114 species from 43 genera of Old World fruit bats (57% and 96% of extant species- and genus-level diversity, respectively), and combined phylogenomic inference with filtering by information content to resolve systematic relationships among the major lineages. Concatenation and coalescent-based methods recovered three distinct backbone topologies that were not able to be reconciled by filtering via phylogenetic information content. Concordance analysis and gene genealogy interrogation shows that one topology is consistently the best supported, and that observed phylogenetic conflicts arise from both gene tree error and deep incomplete lineage sorting. In addition to resolving long-standing inconsistencies in the reported relationships among major lineages, we show that Old World fruit bats have likely undergone at least seven independent dietary transitions from frugivory to nectarivory. Finally, we use this phylogeny to identify and describe one new genus.
Full-text available
The North American Society for Bat Research (NASBR) was born in 1970 with the first Symposium on Bat Research, with 42 attendees and 26 presentations in Tucson, Arizona. The topics discussed in the earlier NASBR meetings were focused mostly on behavior, ecology, physiology, and taxonomy. It was not until the fourth annual symposium that Dr. Vincent M. Sarich presented the first talk on the use of molecular biology to infer phylogeny in bats. During the last 50 years, this subject has expanded rapidly with innovative techniques. Nowadays, it is widely used to understand the evolution of bat species. In this chapter, we discuss how molecular biology has contributed to bat systematics and evolutionary biology from immunological assays performed by Sarich to the use of Sanger sequencing and next generation sequencing that has allowed the assembly of whole genomes, transcriptomes, and viromes. We provide a chronology of how research in molecular biology has gained importance in the study of bat biology from one single presentation in 1973 to whole sessions in more recent NASBR meetings.
Full-text available
Tropical Africa is home to an astonishing biodiversity occurring in a variety of ecosystems. Past climatic change and geological events have impacted the evolution and diversification of this biodiversity. During the last two decades, around 90 dated molecular phylogenies of different clades across animals and plants have been published leading to an increased understanding of the diversification and speciation processes generating tropical African biodiversity. In parallel, extended geological and palaeoclimatic records together with detailed numerical simulations have refined our understanding of past geological and climatic changes in Africa. To date, these important advances have not been reviewed within a common framework. Here, we critically review and synthesize African climate, tectonics and terrestrial biodiversity evolution throughout the Cenozoic to the mid-Pleistocene, drawing on recent advances in Earth and life sciences. We first review six major geo-climatic periods defining tropical African biodiversity diversification by synthesizing 89 dated molecular phylogeny studies. Two major geo-climatic factors impacting the diversification of the sub-Saharan biota are highlighted. First, Africa underwent numerous climatic fluctuations at ancient and more recent timescales, with tectonic, greenhouse gas, and orbital forcing stimulating diversification. Second, increased aridification since the Late Eocene led to important extinction events, but also provided unique diversification opportunities shaping the current tropical African biodiversity landscape. We then review diversification studies of tropical terrestrial animal and plant clades and discuss three major models of speciation: (i) geographic speciation via vicariance (allopatry); (ii) ecological speciation impacted by climate and geological changes, and (iii) genomic speciation via genome duplication. Geographic speciation has been the most widely documented to date and is a common speciation model across tropical Africa. We conclude with four important challenges faced by tropical African biodiversity research: (i) to increase knowledge by gathering basic and fundamental biodiversity information; (ii) to improve modelling of African geophysical evolution throughout the Cenozoic via better constraints and downscaling approaches; (iii) to increase the precision of phylogenetic reconstruction and molecular dating of tropical African clades by using next generation sequencing approaches together with better fossil calibrations; (iv) finally, as done here, to integrate data better from Earth and life sciences by focusing on the interdisciplinary study of the evolution of tropical African biodiversity in a wider geodiversity context.
Full-text available
An Ebola outbreak of unprecedented scope emerged in West Africa in December 2013 and presently continues unabated in the countries of Guinea, Sierra Leone, and Liberia. Ebola is not new to Africa, and outbreaks have been confirmed as far back as 1976. The current West African Ebola outbreak is the largest ever recorded and differs dramatically from prior outbreaks in its duration, number of people affected, and geographic extent. The emergence of this deadly disease in West Africa invites many questions, foremost among these: why now, and why in West Africa? Here, we review the sociological, ecological, and environmental drivers that might have influenced the emergence of Ebola in this region of Africa and its spread throughout the region. Containment of the West African Ebola outbreak is the most pressing, immediate need. A comprehensive assessment of the drivers of Ebola emergence and sustained human-to-human transmission is also needed in order to prepare other countries for importation or emergence of this disease. Such assessment includes identification of country-level protocols and interagency policies for outbreak detection and rapid response, increased understanding of cultural and traditional risk factors within and between nations, delivery of culturally embedded public health education, and regional coordination and collaboration, particularly with governments and health ministries throughout Africa. Public health education is also urgently needed in countries outside of Africa in order to ensure that risk is properly understood and public concerns do not escalate unnecessarily. To prevent future outbreaks, coordinated, multiscale, early warning systems should be developed that make full use of these integrated assessments, partner with local communities in high-risk areas, and provide clearly defined response recommendations specific to the needs of each community.
Full-text available
Bats are amongst the natural reservoirs of many coronaviruses (CoVs) of which some can lead to severe infection in human. African bats are known to harbor a range of pathogens (e.g., Ebola and Marburg viruses) that can infect humans and cause disease outbreaks. A recent study in South Africa isolated a genetic variant closely related to MERS-CoV from an insectivorous bat. Though Madagascar is home to 44 bat species (41 insectivorous and 3 frugivorous) of which 34 are endemic, no data exists concerning the circulation of CoVs in the island's chiropteran fauna. Certain Malagasy bats can be frequently found in close contact with humans and frugivorous bats feed in the same trees where people collect and consume fruits and are hunted and consumed as bush meat. The purpose of our study is to detect and identify CoVs from frugivorous bats in Madagascar to evaluate the risk of human infection from infected bats. Frugivorous bats belonging to three species were captured in four different regions of Madagascar. We analyzed fecal and throat swabs to detect the presence of virus through amplification of the RNA-dependent RNA polymerase (RdRp) gene, which is highly conserved in all known coronaviruses. Phylogenetic analyses were performed from positive specimens. From 351 frugivorous bats, we detected 14 coronaviruses from two endemic bats species, of which 13 viruses were identified from Pteropus rufus and one from Eidolon dupreanum, giving an overall prevalence of 4.5%. Phylogenetic analysis revealed that the Malagasy strains belong to the genus Betacoronavirus but form three distinct clusters, which seem to represent previously undescribed genetic lineages. Our findings suggest that CoVs circulate in frugivorous bats of Madagascar, demonstrating the needs to evaluate spillover risk to human populations especially for individuals that hunt and consume infected bats. Possible dispersal mechanisms as to how coronaviruses arrived on Madagascar are discussed.
The seasonal movements of Eidolon helvum, Myonycteris torquata and Nanonycteris veldkampi were studied. During the dry season, E. helvum roosts in at least one large colony (c500 000 individuals) in the southern forest zone of Ivory Coast. Following the birth of young in February, males and females move into the savanna zones, and the progressive establishment and decline of colonies along a S-N axis indicates that E. helvum migrates at least to the Niger River basin by the middle of the wet season (July). During the dry season, both M. torquata and N. veldkampi are absent from savanna sites, but are common in the forest zone. With the onset of the rainy season in March, catch rates of both species increase first at a southern Guinea savanna site and subsequently at a savanna site 400km to the north, indicating that both species migrate at least to the southern Sudanese savanna zone during this season. Both sexes of N. veldkampi migrate, but the migration of M. torquata is restricted to the immature male cohorts. The high amplitude of the seasonal fluctuations in fruit abundance at savanna sites creates a wet-season surplus of food which results in low intra- and inter-specific competition levels at these sites relative to the forest zone. This may provide the conditions leading to the observed annual migrations.-Author