Comparative phylogeography of the Somali-Masai region in ...

Comparative phylogeography of theSomali-Masai region in eastern Africausing selected rodent species as a model

Tatiana AghováPh.D. dissertation

Brno, 2018

MASARYK UNIVERSITY Faculty of Science

Department of Botany and Zoology

Comparative phylogeography of the

Somali-Masai region in eastern Africa

using selected rodent species as a model

Tatiana Aghová

Ph.D. dissertation

Supervisor: doc. Mgr. et Mgr. Josef Bryja, Ph.D. Co-supervisor: doc. Mgr. Radim Šumbera, Ph.D.

Brno 2018

Bibliographic Entry Author: Mgr. Tatiana Aghová Faculty of Science, Masaryk University Department of Botany and Zoology Title of Thesis: Comparative phylogeography of the Somali-Masai

region in eastern Africa using selected rodent species as a model

Degree programme: Biology Field of Study: Zoology Supervisor: doc. Mgr. et Mgr. Josef Bryja, Ph.D.

Institute of Vertebrate Biology AS CR, v.v.i Co-supervisor: doc. Mgr. Radim Šumbera, Ph.D.

Faculty of Science, University of South Bohemia in !"#$%&'()*+,-.! Department of zoology

Academic Year: 2017/2018 Number of Pages: 285 Keywords: Somali-Masai savanna, phylogeography, rodents,

Nannomys, Gerbilliscus, Saccostomus, Acomys

Bibliografický záznam Autor: Mgr. Tatiana Aghová !"#$%$&'%()*+,-.*/01.2,3.4.#5*$&.,/67&(#871. Ústav botaniky a zoologie Názov práce: Porovnávacia fylogeografia východoafrickej oblasti

Somali-Masai na príklade modelovej skupiny hlodavcov

91/%7:6;,<#$=#.>: Biológia 91/%7:6;,$?$#: Zoológia 9*$071(@: doc. Mgr. et Mgr. Josef Bryja, Ph.D. A41.&,?7$0$=7(,$?#.10$&)B,CD,EF2,&G&G7 Konzultant: %$)G,3=#G,F.%7>,9/>?(#.2, HGIG

J"#$%$&'%()*+,-.*/01.2,K7H$L(4*+,/67&(#871.,v E(4*;)H,M/%':$&7)")H

Katedra zoologie Akademický rok: 2017/2018 $L(1,41#+6: 285 N@OL$&P,40$&+: Somali-Masai savana, fylogeografia, hlodavce,

Nannomys, Gerbilliscus, Saccostomus, Acomys

ABSTRACT

The Somali-Masai region, and especially its non-forested savanna-like ecosystems, is one of

the least known ecoregions of Africa from biogeographic point of view. For example,

information on genetic structure within specialized species or among closely related species is

entirely missing; therefore making it difficult to assess which historical factors (e.g. Plio-

Pleistocene climate fluctuations, geomorphological barriers, etc.) have influenced the evolution

of savanna biota in this part of Eastern Africa.

Small mammals, and especially rodents, are suitable model for phylogeographic

reconstructions, because they are tightly linked to specific habitat type, have low dispersal

ability and relatively high substitution rate of mitochondrial DNA. In the present PhD study, I

have analyzed for the first time the genetic structure of four rodent genera preferring open

habitats and living (at least partly) in Somali-Masai region: pygmy mice (Mus), pouched mice

(Saccostomus), gerbils (Gerbilliscus), and spiny mice (Acomys).

For the first part of the thesis, I summarized the genetic diversity of model taxa and

showed that in each genus studied there are lineages endemic to Somali-Masai region. Some of

them might correspond to new species which should be further analysed by integrative

taxonomic revision. For second part, I provided a revised list of nine fossils that could be

considered as reference constraints for molecular clock calibrations for murid rodents. Finally,

for the third part I reconstructed the history of Somali-Masai savanna, including the

identification of the most important long-term geographical barriers of gene flow (e.g. mountain

chains, aridity belts, paleo-lakes), which have played an important role in the evolution of

savanna ecosystems in Eastern Africa especially during the Plio-Pleistocene. Results of the

comparative phylogeographic study can shed light also on evolutionary history of other

savanna-living organisms, including our own ancestors (as numerous crucial localities for

studies of human evolution are located in the Somali-Masai region).

ABSTRAKT

Región Somali-Masai (predovšetkým savanové ekosystémy) patria k najmenej preskúmaným oblastiam

Afriky z biogeografického hladiska; chýbajú základné genetické informácie o vnútrodruhovej

a !"#$"%&'()!*+ ,-%&.-/%!+ 01)12()34'+ (%512$# ()6+ 7%!-(+ *!+ )!8 $+ (9-$1:2!+ $"!2-$;$.()1<+ '$0-(%$4.=+

faktory (ako Plio-Pleistocéne klimatické zmeny, geomorfologické bariéry, a->6?@+ .-(%=+ ()ABC)2$B$+

evolúciu savanovej bioty vo východnej Afrike.

D%(92=+ 4$41)4!@+ A%!"(),!-.3 + 'B("1)4!+ 0/+ )'("23 + ("!B()3 + (%512$# ( + A%!+

;CB(5!(5%1;$4.=+%!.(2,-%&.4$!+# 2$!.(8.34'+"E)("()6+F/+)$1#12=+21+,A!4$;$4.3+-CA+'19$-1-&@+ 1*/+2G#.&+

04'(A2(0<+ $5%()1<+21+ )!8.=+)#"$1B!2(0-$+ 1 %!B1-G)2!+)C0(./+ &-1H2/+ %34'B(0<+ 21+ $-(4'(2"%$IB2!*+

DNA. V A%!".B1"12!*+"$#!%-1H2!*+A%I4$+0( +121BC#()1B1+5!2!-$4./+,-%&.-/%&+,-C%(4'+%("()+'B("1)4()@+

.-(%=+:$*/+A%!"(),!-.3 JH$10-(H2!+) otvorených habitatoch v Somali-Masai regióne: myš (Mus, podrod

Nannomys), Saccostomus, pieskomil (Gerbilliscus) a C,+9("B$21-I+KAcomys).

V A%)!*+ H10-$+ "$#!%-I4$!+ 01+ #1 !%$1)1 + 21+ 5!2!-$4./+ "$)!%#$-&+ ("!B()34'+ -1LM2()6+ N,!-.C+

,-&"()12=+%("C+ 1*/+!2"! $4.=+5!2!-$4.=+BG2$!+) Somali-Masai regióne, niektoré z nich predstavujú

nové druhy a 1B$+9C+9C<+A("%(9!2=+$2-!5%1-G)2!*+-1L(2( $4.!*+%!)G#$$6+N "%&'!*+H10-$+A%$2I,1 +#(#21 +

"!)$1-$4'+ ;(0GB$G@+ .-(%=+ E:&+9C<+A(&:$-=+1.(+ %!;!%!2H2=+.1B$9%1H2=+9("C+ A%!+ (B!.&BI%2!+ "1-()12$!+

hlodavcov. V tretej H10-$+ %!.(2,-%&&*! +'$0-M%$&+ 01)12C+F( 1B$-Masai a $"!2-$;$.&*! +21*"EB!:$-!*,$!+

"B'("(9=+ 5!(5%1;$4.=+ 91%$=%C+ A%!+ -(.+ 5=2()+ K21A%6+ '(%0.=+ 10G)C@+ A/,-!@+ A1B!(-jazerá), ktoré hrali

významnú rolu v !)(B/4$$+ 01)12()34'+ !.(0C0-= ()+ )(+ )34'("2!*+ O;%$.!+ A(H10+ 7Bio-Pleistocénu.

N30B!".C+ (*!*+A(%()2I)14!*+;CB(5!(5%1;$4.!*+,-/"$!+ E:&+A( E4<+(9*102$<+!)(B&H2/+'$0-M%$&+>1B,G4'+

01)12()34'+(%512$# ()@+)%I-12!+2I,'(+)B10-2='(+A%!"4'("4&+K%("+Homo?6+7%!-(:!+ 2(:0-)(+#I01"234'+

B(.1BG-+A%!+,-/"$& +8&"0.!*+!)(B/4$!+*!+A%I)e v Somali-Masai regióne.

ACKNOWLEDGEMENT

First, I would like to thank my wonderful supervisor, Josef Bryja. He trusted me from our first

meeting and guided me in the scientific world including editing my documents (many of which

had bad grammar), giving me the opportunity to travel for conferences and fieldworks,

introducing me to local and international collaborators, and giving me space to work on my

own, he showed me how to have fun, balance work and life, and he has tought me lot of non-

scientific skills as well (e.g. how travel through airports and how to tap a beer ). He is great

inspiration for me and an example for how to be a great scientist and kind human as well.

THANK YOU PEPA!

I acknowledge my great co-supervisor Radim Šumbera, who was my support in

difficult situations and for his humor, /012#3+ !"4*5 for his sence for perfect analysis (even

if it makes me crazy sometimes). Special thanks to 675+8-93%&' and :50"5+;%0&!<"%&', who

introduced me to the molecular lab and =5>?5-+@AB"%&' and C4D%$+E!'*#" for their help with

NGS sequencing. For the field expedition, I would like to say big thanks Vladimír Mazoch for

his patience (and lot of fun too ), Yonas Meheretu in Ethiopia, Jana Vrbová Komárková

and Christopher Sabuni for Tanzania fieldwork and all local collaborators. Additional thanks

goes out to our international colaborators, who significantly help me increase quality of my

research: Leonid Lavrenchenko for numerous Ethiopian samples, Molly M. McDonough, for

long discussions about our favourite genus Gerbilliscus and for correction of my English, Yuri

Kimura, who help me with fossils and support my scientific ideas. Big thanks also to my

supervisors at CBGP Montpellier: Laurent Granjon for taking care of me, Gauthier Dobigny

for motivation finishing things on time, Gael J. Kergoat from whom I learned a lot about

analyses, but also how important is to analyze and write with a focus on every detail. I thank

also Alexandre Dehne-Garcia for helping me with cluster and his optimism when I needed it.

At University of Antwerp I would like to thank Frederik Van De Perre, Erik Verheyen and

Herwig Leirs, for giving me their trust in the challenging molecular part of COBIMFO project

and Natalie van Houtte for her laboratory and personnel support. It was big pleasure to work

with you! Special thanks to my colleagues at Natural History Museum in Prague, they give

me their trust in sequencing museum specimens and at the same time give me time and space

to finish my PhD. For the administrative part during my whole PhD I acknowledge Andrea

!"#$"%&' and (!)%*#+,-!*!..

Here, I would like to thank all the people in Institute in Vertebrate Biology in

Studenec. During last six years you become my second family . I would like to thanks Stuart

Baird for correction of my English. I acknowledge also my friend and female scientific idol,

Joëlle Goüy de Bellocq, thank you for your support, it was my honor to share an office with

you. The biggest THANKS goes to my best friend, colleague and roommate Terka Králová

who shared with me all the victories and downs during my PhD. Thank you all, earning my

PhD was one of the best parts of my entire life!

And last, but not least I would like to thank my family and friends. They supported me

in my decisions, even if sometimes they did not understand why I would like to go somewhere

or do something or even if they were afraid for me lot of time. They give me the trust in myself,

when I didn’t see it. Thank you and I love you!

FUNDING

This PhD dissertation was supported by projects of the Czech Science Foundation, nos.

P506/10/0983 and 15-20229S, the Ministry of Culture of the Czech Republic (DKRVO

2017/15, National Museum, 00023272). Additionally the travel cost were covered from Projects

of specific research on Masaryk University (2012-2017), NextGen Project

(CZ.1.07/2.3./20.0303 !" #$%&'(&)" *+," -./+,*01.+" 23*,4+5" 6*78-*!" '*918" *" :,8;<="

Hlávkových”.

TABLE OF CONTENTS Preface ........................................................................................................ 19

Introduction................................................................................................. 21

Aims of the thesis ....................................................................................... 28

Methods ...................................................................................................... 29

Summary of main results ............................................................................ 34

Conclusions and future directions .............................................................. 48

References................................................................................................... 51

Paper I ......................................................................................................... 61

Paper II........................................................................................................ 83

Paper III ......................................................................................................109

Paper IV ......................................................................................................125

Paper V ....................................................................................................... 199

Appendix..................................................................................................... 271

19

21

28

29

34

48

51

61

83

109

125

199

271

19

Tatiana Aghová / Ph.D. dissertation (2018)

PREFACE

This thesis examined phylogeograpical patterns in the Somali-Masai savanna in East Africa

and its evolutionary history during Plio-Pleistocene, using selected groups of rodents as

models. Besides general introductory parts, this thesis is comprised of five papers. The results

concerning model genera Mus (Paper I), Saccostomus (Paper II), and Gerbilliscus (Paper III)

have been already published in the peer reviewed journals. The manuscript focused on

phylogeny and biogeography of a genus Acomys (Paper V) is presented in the form of

unpublished manuscript, but with small improvements the will be submitted in the near future.

Study dealing with wider biogeography patterns and optimalisation of calibration points

for molecular dating of family Muridae (Paper IV), is currently under review.

All data were produced and analyzed during my Ph.D. studies at Institute of

Vertebrate Biology of the Czech Academy of Sciences, Research Facility Studenec (Papers

I-V) and during my stay at the Center for Biology and Management of Populations in

Montferrier-sur-Lez, France (Paper IV). The general framework of the thesis also benefited

from my half-year stay at University of Antwerp, Belgium. Data for genetic analysis were

collected mainly during field expeditions in Ethiopia (2012, 2013, 2014), Kenya (2010, 2011)

and Tanzania (2013, 2015, 2016). I personally participated at the expeditions in Ethiopia 2014

and Tanzania 2015.

The introductory part of the thesis encompasses a summary of current knowledge

about diversity and history of savanna habitats since the Miocene, as well as justification

for selection of rodents as suitable model taxa. The introduction is followed by a short

description of aims of the thesis and brief description of the methods used. This part is

followed by short summary of main findings of the five studies, overall conclusions

concerning evolutionary history of Somali-Masai savanna based on the results obtained within

this thesis, and proposed directions in comparative phylogeography, molecular dating and

reconstruction of history of Somali-Masai savanna. All papers with the specifics of my

contributions are attached . The thesis is supplemented by a CV with a complete list of my

publications, contributions at international and domestic conferences, workshops and other

activities during my PhD study. Posters from conferences are attached as supplementary

materials to the end of the thesis.

20


List of papers included in the thesis:

I. Bryja, J., Mikula, O., Šumbera, R., Meheretu, Y., Aghová, T., Lavrenchenko, L. A.,

Mazoch, V., Oguge, N., Mbau, J.S., Welegerima, K., Amundala, N., Colyn, M., Leirs,

H., Verheyen, E., 2014. Pan-African phylogeny of Mus (subgenus Nannomys) reveals

one of the most successful mammal radiations in Africa. BMC Evolutionary Biology,

14(1), 256.

II. Mikula, O., Šumbera, R., Aghová, T., Mbau, J.S., Bryja, J., 2016. Evolutionary history

and species diversity of African Pouched Mice of the genus Saccostomus (Rodentia:

Nesomyidae). Zoologica Scripta, 45(6), 595-617.

III. Aghová, T., Šumbera, R., Piálek, L., Mikula, O., McDonough, M.M., Lavrenchenko,

L.A., Meheretu, Y., Mbau, J.S., Bryja, J., 2017. Multilocus phylogeny of East African

gerbils (Rodentia, Gerbilliscus) illuminates the history of the Somali-Masai savanna.

Journal of Biogeography, 44(10), 2295–2307.

IV. Aghová T., Kimura Y, Bryja J., Dobigny G., Granjon L., Kergoat G.J.: Fossils know it

best: using a new set of fossil calibrations to improve the temporal phylogenetic

framework of murid rodents (Rodentia: Myomorpha: Muroidea: Muridae).

(Molecular Phylogenetics and Evolution, Major revision)

V. Aghová, T. !"#$%&'()*+, !-. !Šumbera, R., Frynta, D, Lavrenchenko, L.A., Meheretu,

Y., Sádlová, J., Votýpka, J., Mbau, J.S., Modrý, D., Bryja, J.: Multiple radiations of

spiny mice (Rodentia: Acomys) in arid environments of Afro-Arabia: evidence from

multi-locus phylogeny. (Prepared for BMC Evolutionary Biology)

21


INTRODUCTION

Diversity of African ecosystems

Africa is a continent of amazing biodiversity with biota (fauna and flora) inhabiting various

environments (Fig. 1), such as equatorial rainforests, deserts, savannas and mountain habitats;

although individual biomes differ in species diversity and richness. Equatorial rain forests

(“Guinea-Congo forests”) are species-poor in global terms (Richards, 1973; Hamilton, 1976;

Beentje et al., 1994) in comparison to Afromontane forest, which are remarkably species-rich

(Burgess et al., 2007a; Cordeiro et al., 2007). Three main desert areas occur in Africa: the

worlds´ largest desert, the Sahara, with the most extensive species-poor area outside the Arctic

(Burges et al., 2004), the Horn of Africa (Somalia) and south-west Africa (Namibia). Between

these extremes are moist forest-savanna mosaics, woodlands, various wooded grasslands or

savannas, often dominated by Acacia bush or scrub, and verging on sub-desert or semi-desert

in places. The extensive savannas host (among others) the largest remaining mammalian

megafauna on the planet (Fjedså et al.,

2004). The East African Rift System

(EARS) and three widely separated

glaciated mountains close to the

equator (Rwenzori, Mt. Kenya, Mt.

Kilimanjaro) harbour a very distinct,

often endemic, biota (Friis et al.,

2010). The Cape Floristic Region of

southern Africa also deserves special

attention as this is a global

biodiversity hotspot (Myers et al.,

2000; Linder, 2005; Mittermeier et al.,

2005).

Fig. 1: Map of Africa’s biomes (www.10000birds.com)

Grouping specific biomes across Africa into a biogeographical system has challenged

biologists over the past 100 years. The earliest classifications, produced in the late 1800s,

already recognized the separation of the savannas, desert and rain forest (Wallace, 1876;

22


Engler, 1879–1882). A century later, White (1983) defined a set of phytochoria for Africa and

estimated the number of species and the degree of endemism in each phytochorion. The

phytochoria were fundamentally based on species-level distribution patterns and were derived

by intuitive methods. More recently, Linder et al. (2012; Fig. 2) analysed distribution patterns

of higher plants and vertebrates and divided sub-Saharan Africa into seven well-defined regions

that are partly concordant with previous classifications: Saharan (subregions: a, Sahara; b,

Sahel; c, Nubian Desert), Sudanian, Congolian (subregions: a, Guinea; b, Congo; c, Shaba),

Ethiopian (subregions a, Djibuti; b, Ethiopia), Somalian (subregions: a, Somalia; b, Horn),

Zambezian, and Southern African region (subregions: a, Cape; b, Natal; c, Kalahari; d,

Namib; e, south-west Angola).

Fig. 2: Biogeographical regionalization and phenogram and phenogram based on combine matrix (mammals, birds, reptiles, amphibian, plants). The map is based on a 1°grid (Linder et al., 2012)

The most widespread terrestrial habitats in Africa are tropical grasslands, savannas

and shrublands (Sayre et al., 2013). They can be loosely defined as ecosystems with

continuous and important grass/herbaceous stratum, a discontinuous layer of trees and shrubs

of variable height and density, where plant growth patterns are closely associated with

alternating rainy and dry seasons (Bourliere and Hadley, 1983). Unlike forest, for which the

United Nation’s Food and Agriculture Organization (FAO) provides a clear definition (5m in

height, 10% or more canopy cover, <0.5 ha, and not under agricultural or other non-forest land

use, FAO, 2010), savannas are more ambiguously defined from either a climatic or vegetation

point of view (Table 1). The savanna biome spans the tropical grasslands, scrublands and

23


wooded savannas of sub-Saharan Africa, encompassing the Sahel (the vast semi-arid region of

North Africa south of the Sahara), the Rift Valley formation in equatorial East Africa and most

of Southern Africa (Mayaux et al., 2004). Four biogeographical regions with predominating

savanna-like habitats were distinguished across African continent (e.g. Linder et al. 2012; Fig.

2): Zambezian, Somalian, Sudanian and South African savanna. Affected mainly by

climatic conditions, we can distinguish various savanna types (Fig. 3), ranging from savanna

grassland lacking woody vegetation through savanna parkland, low tree-shrub savanna to more

wooded savanna, e.g. savanna woodland and thicket and shrub with dense trees and shrubs

(Wynn, 2000).

This Ph.D. dissertation focuses on open habitats of Somali-Masai region in Eastern

Africa (Fig. 4). International Vegetation Classification (IVC; naturserve.org; Faber-

Langendoen et al., 2014) identifies Somali-Masai savanna as Eastern Africa xeric scrub and

grassland (1 701 057 km2; Dixon et al., 2014) with four ecoregions: (1) Somali Acacia-

Commiphora bushlands and thickets, (2) Southern Acacia-Commiphora bushlands and

thickets, (3) Northern Acacia-Commiphora bushlands and thickets and (4) Masai xeric

grasslands and shrubland (Fig. 4).

Fig. 3: Illustration of vegetation classification, arranged in order of relative degree of openness (Cole, 1963; Wynn, 2000)

24


Table 1: Definitions of savannas (based on Mistry, 2000; Mistry and Beradi, 2014)

Definition of savanna Authors

Climatic

Ecosystems which lie in the tropical savanna and monsoon climatic zones, largely between the latitudes of Cancer and Capricorn, where annual precipitation is between 250 and 2000 mm, most of which falls in the wet season

Köppen (1990, 1884)

Any formation or landscape within the region experiencing a winter dry season and summer rains is a savanna.

Jaeger (1945), Lauer (1952), Troll (1950),

Ecosystems bound by dry forests at higher rainfall (> 1000 mm), by thorn forests at lower rainfall (< 500 mm), and by thorn steppe and temperate savannas at lower temperatures (< 18°C)

Holdrige (1947)

Ecosystems with low to moderate rainfall (500-1300 mm) and high mean annual temperatures (18-30 °C)

Whittaker (1975)

Vegetation

A mixed physiognomy of grasses and woody plants in any geographical area

Dansereau (1957)

A mixed tropical formation of grasses and woody plants, expluding pure grasslands

Walter (1973)

Open formations dominated by grasses, in the lowland tropics, where trees and shrubs, if presnt, are of little physiognomic significance

Beard (1953), Whittaker (1975)

Fig. 4: Distribution of Somali-Masai savanna with four Terrestrial Ecoregions of the World (TEOW; modified from Dixon et al., 2014)

25


Ecosystems which lie in the tropical savanna and monsoon climatic

Any formation or landscape within the region experiencing a winter dry

Ecosystems bound by dry forests at higher rainfall (> 1000 mm), by

1300 mm) and high

of grasses and woody plants in any geographical

A mixed tropical formation of grasses and woody plants, expluding pure

Open formations dominated by grasses, in the lowland tropics, where , Whittaker

Distribution, diversity and history of the Somali-Masai savanna

The Somali-Masai region was originally proposed by White (1983) as a distinct dry

phytochorion, extending from Eritrea to Tanzania (Fig. 4). This region exhibits complex

climatic and topographical patterns resulting in different biologies of the taxa (Linder et al.,

2012). The major geomorphological feature in this region is the East African Rift System

(EARS, Fig. 5), which has a long-term influence on the climatic regime in Africa (Sepulchre et

al., 2006). While this region has relatively low floristic and faunistic species diversity, a high

level of endemism exists for plants (Thulin, 1993), reptiles (Burgess et al., 2007b) and rodents

(Varshavsky et al., 2007). This region is home to the Horn of Africa biodiversity hotspot,

which ranks among the oldest and most stable arid regions of Africa (Kingdon, 1990). Thulin

(1994) proposed this region as a refugium for arid-adapted plants, from there Socotra, Arabia

and southern Africa were

subsequently colonised. Importanty,

this region is one of the least known

African bioregions and possibly also

one of the most threatened due to

the rapid increase of human

populations and climatic changes

(Geist & Lambin, 2004).

Fig. 5: The East African Rift System with its rift segments. The basemap is a Space Shuttle radar topography image by NASA (authors Wood and Guth, https://geology.com/articles/east-africa-rift.shtml )

The current biological diversity of Somali-Masai savanna was influenced mainly by

global climate changes and by local tectonic processes, especially the development of the

EARS (Sepulchre et al., 2006). Volcanism associated with the Great Rift Valley began as early

as 45-33 Ma in the Ethiopian Rift (Maslin and Trauth, 2009). Major faulting in Ethiopia occured

between 20-14 Ma and was followed by faulting in northern Kenya between 12 and 7 Ma.

Central and southern Kenya Rift was formed between 9 and 6 Ma (Baker et al., 1988; Strecker

26


et al., 1990; Ebinger et al., 2000; Maslin and Trauth, 2009). In the Tanzanian sector of the rift,

sedimentation in isolated basins began at ~5 Ma (Foster et al., 1997). A major phase of rift

faulting occurred at 1.2 Ma and produced the present-day rift escarpments (Foster et al., 1997).

In combination with Plio-Pleistocene climatic changes promoting expansion of C4 grasses

(Cerling et al., 1998), Eastern Africa became a center of endemism for organisms living in

open habitats (Kingdon, 2015). Many evolutionary divergent clades of ungulates such as

warthog (Randi et al., 2002); lesser kudu, desert warhog, Grevy’s zebra, Grant’s gazelle

(Lorenzen et al., 2012), giraffe (Brown et al., 2007; Fennessy et al., 2016), evolved in Somali-

Masai savanna.

Rodents as suitable model group for phylogeographical studies

Murid rodents (Rodentia: Muridae) are excellent model organisms for biological research

for several reasons. It is the most diverse family of mammals with more than 155 genera and

816 recognized species (Denys et al., 2017). They colonized different ecological niches in a

wide array of environments ranging from warm (deserts or tropical forests) to cold habitats such

as high altitude mountain ranges (Vaughan et al., 2011). Life habits in murids are also diverse,

as the family encompasses amphibious, arboreal, fossorial or terrestrial taxa (Musser and

Carleton, 2005). Family Muridae also includes two key mammalian model organisms: the lab

mouse (Mus) and rat (Rattus). They are widely used in biomedicine, physiology, genetics and

evolutionary biology.

More than 60 zoonotic diseases (diseases transmitted from free-living animals to

humans or livestock) can be transmitted from rodents via direct and indirect ways (see the

comprehensive review of Meerburg et al., 2009). For example, African rodents are reservoirs

for arenaviruses, e.g. Lassa (Fichet-Calvet and Rogers, 2009), Mopeia, Luna, Mobala

(Meheretu et al., 2012); hantaviruses, e.g. Sangassou (Klempa et al., 2006) and Tigray (Goüy

de Bellocq et al., 2016). Most of these viruses were detected in the tribe Praomyini (e.g. host

rodent species Mastomys natalensis, M. awashensis, Stenocephalemys albipes, Hylomyscus

simus). Other genera within Muridae can function as reservoirs as well. For example, Lecompte

et al. (2007) detected a new arenavirus Kodoko from Mus minutoides. Kassahun et al. (2015)

identified parasitic protozoans from the genus Leishmania in gerbils (Gerbillinae:

Gerbilliscus nigricaudus, Gerbillus nannus), spiny mice (Deomyinae: Acomys sp.) and murines

(Murinae: Arvicanthis sp. and Mastomys erythroleucus).

27


Rodents are also very good candidates for inferring the biogeographic history for a

particular ecosystems by inferring contemporary patterns of genetic variation owing to their

short generation time and rapid mtDNA substitution rate, their relatively limited dispersal

ability, and their close associations with particular habitats (Fedorov et al., 2008). Their

outstanding diversity combined with their widespread and abundant distribution and rich fossil

record, makes them a excellent model for phylo(bio-)geographical studies. In this thesis four

model genera of African rodents were selected: three genera from the family Muridae (Murinae:

Mus, Gerbillinae: Gerbilliscus, Deomyinae: Acomys) and one genus from the family

Nesomyidae (Saccostomus; Fig. 6). All of them live in savanna ecosystems in Eastern Africa,

and analysis of their evolutionary history can thus help to understand which factors affected

evolution of these ecosystems.

a) b)

c) d)

Fig.6: Model taxa: a) Nannomys ( !"#$%&'(%)(*+&,-); b) Saccostomus . !"#$%&/!01(% &2-34&53&Gerbilliscus (author /!01(% &2-34&63&Acomys (http://www.hlasek.com/acomys_cahirinus1sk.html)

28


AIMS OF THE THESIS

Somali-Masai savanna in East Africa is one of the least known African biogeographic regions.

Information about itraspecific genetic structure or genetic structure among closely related

species is almost absent from this region. Thus, it is difficult to assess which factors (e.g. Plio-

Pleiostocene climate fluctuations, geomorphological barriers, etc.) have influenced the

evolution of savanna biota in this part of Africa. Understanding the drivers and mechanisms

of species diversification and persistence is of central interest to biogeography, evolutionary

biology and conservation genetics (Frankham et al., 2004, Höglund, 2009). Furthermore,

comparative phylogeography could provide robust conclusions regarding biogeographical

processes operating during the Plio-Pleistocene in Eastern Africa. In this Ph.D. dissertation

three main aims were defined:

1. Description of genetic diversity of four model rodent genera inhabiting Somali-Masai

savanna with suggestions for possible taxonomic implications (Paper I-III, V).

2. Optimalization of methods for divergence dating in phylo(bio-)geographic

reconstructions of murid rodents (Paper IV).

3. Reconstruction of evolutionary history of Somali-Masai savanna during Plio-

Pleistocene and identification of factors that have influenced current distribution of

genetic diversity in this ecosystem (Paper I-III, V).

29


METHODS

Data collection

Material collected during recent field expeditions to Ethiopia, Kenya and Tanzania was

supplemented by additonal material (especially tissue samples) from American Museum of

Natural History (AMNH, New York), Natural History Museum (BMNH, London), Magyar

Természettudományi Múzeum (HNHM, Budapest), Institut des Sciences de l’Evolution (ISEM,

Montpellier), Livingstone Museum (LM, Livingstone), Museum of Comparative Zoology

(MCZ, Boston), Museum of Vertebrate Zoology (MVZ, Berkeley), Muséum national d’Histoire

naturelle (MNHN, Paris), Institut royal des Sciences naturelles de Belgique (RBINS, Brussel),

Musée royal de l’Afrique centrale (RMCA, Tervuren), Senckenberg Museum (SMF, Fankfurt),

Staatliches Museum für Naturkunde (SMNS, Stuttgart), Ditsong National Museum of Natural

History (TM, Pretoria), Texas Tech University (TTU, Lubbock), Smithosonian Institution-

National Museum of Natural History (USNM, Washington, D.C.) and Museum für Naturkunde

(ZMB, Berlin).

Table 2: Number of individuals used for particular papers with total number of individuals (# individuals), number of new individuals sequenced (# new) and number of samples obtained from museums (# museum).

# individuals # new # museum # loci Matrix (bp)

Paper I 657 395 16 2 2416

Paper II 167 119 31 3 *

Paper III 219 158 34 4 4473

Paper IV 166 0 0 15 10482

Paper V 482 248 14 4 4005

* For Paper II we didn’t provide information about length of concatenate matrix, because three different genes were analyzed separately.

PCR amplification and sequencing

As a standard marker for molecular identification and study of phylogeographic structure we

used the mitochondrial cytochrome b (CYTB) gene in all studies (Paper I-V) or short fragment

of CYTB in mini-barcoding protocol (Paper I-III). Additionaly we sequenced another

mitochondrial marker Control region (D-loop) and four nuclear markers including: exons of

breast cancer susceptibility gene (BRCA1), interphotoreceptor retinoid binding (IRBP),

recombination activating gene 1 (RAG1) and intron 7 of the gene for -fibrinogen (FGB;

see Table 3).

30


Table 3: Overview of genetic markers used for phylogenetic analysis.

Marker Primers Sequences Amplicon

size (bp) Ta

(°C)

Author Used in

study CYTB L14723 ACC AAT GAC ATG AAA AAT CAT CGT T 1 140 55 Ducroz et

al. (1998) Paper I-

III,V H15915 TCT CCA TTT CTG GTT TAC AAG AC

CYTB

mini

L14411F GAY AAA RTY CCV TTY CAY CC 136 45 Galan et al. (2012)

Paper I-III H15546R AAR TAY CAY TCD GGY TTR AT

D-loop La1 ATAAAAATTACTCTGGTCTTGTAAAC 564 50 Nicolas et al. (2009)

Paper IV-V

Hb TGTCTTAATTTAGGGGAACG-

BRCA1

ext

BRCA1-AF TGC ACC TGA RAN GCA TCC AGA AAA 2 200 65 * Adkins et al. (2001)

Paper III BRCA1-3R TTW GGY CCT CTG TTT CTA YCT AG

BRCA1

int

BRCA1-Fseq

GAA AAA GAC TTC TTC AAA CC 1 130 58 Granjon et al. (2012)

Paper III

BRCA1-Rseq

CTT CCA GAT TTT RGA AAC C

FGB Fgb-I7U GGG GAG AAC AGA ACC ATG ACC ATC CAC

577 59 Wickliffe et al. (2003)

Paper III

Fgb-I7L ACC CCA GTA FTA TCT GCC ATT CG GAT T

IRBP IRBP217 ATG GCC AAG GTC CTC TTG GAT AAC TAC TGC TT

1 226 55 Stanhope et al. (1992)

Paper I-V

IRBP1531 CGC AGG TCC ATG ATG AGG TGC TCC GTG TCC TG

RAG1 RAG1F1705 GCT TTG ATG GAC ATG GAA GAA GAC AT

1246 65 * Teeling et al. (2000)

Paper II-V

RAG1R2951 GAG CCA TCC CTC TCA ATA ATT TCA GG

*Touchdown PCR

All museum tissue samples (usually from old voucher individuals) were handled in a

specialized laboratory of Institute of Vertebrate Biology ASCR in Studenec, designed for work

with rare DNA to prevent contamination by samples with high quantity of DNA or by PCR

products. Museum samples (taken mostly from dry skins) were analyzed using the CYTB mini-

barcode protocol (Galan et al., 2012), including pyrosequencing on a GS Junior (Roche,

Basel, Switzerland). This method allows for the separation of focal sequences in samples

contaminated by distantly related organisms (e.g. human DNA). The obtained sequences were

analysed using |S|E|S|AM|E| Barcode software (Piry et al., 2012). In Paper I, eleven

individuals were sequenced from Democratic Republic of Congo, including the type specimen

of Mus bufo (RMCA 17897) from locality Idjwi; and others from Angola, Central African

Republic and South Africa. In Paper II, the same approach was used for 31 samples from

Eastern Africa, especially valuable were the samples from Somalia and South Sudan. We

sucessfully sequenced the syntype of Saccostomus lapidarius (ZMB MAM-85437) and

paralectotype of Saccostomus fuscus (ZMB MAM-85450). Unfortunately, I was not able to

genotype the paratype of Saccostomus mashonae (BMNG 1895.8.27.9). In Paper III, museum

31


Marker Amplicon Ta Author Used

ACC AAT GAC ATG AAA AAT CAT CGT T 55 et Paper I-

CYTB 45 Galan et al. Paper I-

50 Nicolas et Paper

BRCA1 AF 65 * et Paper

3R

BRCA1 58 Granjon et Paper

GGG GAG AAC AGA ACC ATG ACC ATC 59 Wickliffe Paper

ACC CCA GTA FTA TCT GCC ATT CG GAT

ATG GCC AAG GTC CTC TTG GAT AAC 55 Stanhope Paper I-

CGC AGG TCC ATG ATG AGG TGC TCC

RAG1F1705 ATG GAA GAA GAC 65 * Teeling et Paper

RAG1R2951

specimens were important from more reasons. First, we sequenced syntype of Gerbilliscus

nigricaudus bayeri (RMCA 5183-M) from Maroon River (Kenya), and additional samples

from South Sudan. Second, we broadened the range of putatively new species Gerbilliscus sp.

n. (Babile) by records from two new localities in Ethiopia and Somalia. Finally, we confirmed

distribution of G. robustus from Ethiopia to Chad by genotyping museum vouchers from Sudan.

Paper V includes recently collected specimens from various museum, therefore, these samples

were processed using standard protocols (specifically it concerns the samples from Djibouti,

Egypt, Jordan, Niger, South Africa and Yemen). Most important are the tissue samples from

Djibouti (601474, 601476, 602643) which may represent a new species of Acomys, and material

of Acomys subspinosus (VV2001118), which represents the first nuclear data, therby resolving

the phylogenetic position of this very distinct taxon.

Phylogenetic analysis, species delimitation and divergence dating

Phylogenetic analyses were conducted using both Bayesian inference (BI; MrBayes v3.2.6,

Ronquist et al., 2012a) and maximum likelihood (ML; RAxML v8.2.8, Stamatakis, 2014).

Analyses were performed on the online computer cluster CIPRES Science Gateway (Miller et

al., 2010; www.phylo.org) and on the high performance computing (HPC) cluster hosted in

the Centre de Biologie pour la Gestion des Populations (CBGP) in Montferrier-sur-Lez, France

or METACENTRUM (https://metavo.metacentrum.cz). For both phylogenetic analytical

approaches we carried out partitioned analyses to improve phylogenetic accuracy (Nylander et

al., 2004). The best partitioning scheme and substitution models were determined with

PartitionFinder 1.1.1 (Lanfear et al., 2014).

For species delimitation we used several alternative approaches: (i) single locus

delimitation with Generalized Mixed Yule Coalescent (GMYC, Pons et al., 2006) in Papers I

and V; multi-rate Poisson Tree Process (mPTP, Kapli et al., 2016) in Paper V, and Automatic

Barcode Gap Discovery (ABGD, Puillandre et al., 2012) in Paper V; and (ii) multilocus

coalescent based model Bayesian Phylogenetics & Phylogeography (BP&P, Yang and

Rannala, 2014) in Paper IV and Species Tree And Classification Estimation, Yarely (STACEY,

Jones 2014) in Paper V. The species tree was calculated under the Bayesian framework

implemented in *BEAST package (Heled & Drummond, 2010), an extension of BEAST 1.8.2

(Drummond et al., 2012).

32


Bayesian relaxed-clock analyses were conducted with BEAST v1.8 (Drummond et al.,

2012) for Papers III-IV and BEAST v2.4 (Drummond et al., 2015) for Paper V. The priors for

fossil constraints (Papers I-V) were defined by using lognormal statistical distributions. The

final analyses were carried out by two independent runs each with 50 million generations and

trees sampled every 5,000 generations. We used a conservative burn-in of 12.5 million

generations per run. Post burn-in trees from both analyses were further combined using the

LogCombiner module of BEAST. Convergence of runs was assessed graphically under Tracer

v.1.6 and by examining the ESS of parameters.

Historical biogeography analysis

Ancestral biogeography was reconstructed using the R package BioGeoBEARS (Matzke,

2013). We constrained biogeographic areas depending on the aims of particular analysis: in

paper IV we followed Olson et al. (2001), and in Paper V we defined the regions using Linder

et al. (2012) and Holt et al. (2013). Six models of geographic range evolution were compared

in a likelihood framework: (i) Dispersal-Extinction Cladogenesis model (DEC) similar to

Lagrange (Ree and Smith, 2008), which parameterizes dispersal and extinction; (ii) DEC +J

model (Matzke, 2013; 2014), which adds founder-event speciation with long-distance dispersal

(cladogenesis, where daughter lineage is allowed to jump to a new range outside the range of

the ancestor; Matzke, 2013) to the DEC framework; (iii) Dispersal Vicariance Analysis (DIVA;

Ronquist, 1997); (iv) DIVA with long-distance dispersal (DIVA +J; Matzke, 2013); (v)

Bayesian inference of historical biogeography for discrete areas (BayArea; Landis et al., 2013);

and (vi) BayArea with long-distance dispersal (BayArea +J; Matzke, 2013). Model fit was

assessed using the Akaike information criterion (AIC) and likelihood-ratio tests (LRT).

Species distribution modelling

The present distribution of the East African Gerbilliscus, Saccostomus and Acomys was

predicted by maximum entropy modelling (MaxEnt, Phillips et al., 2006). Only CYTB-

barcoded specimens were included to avoid taxonomic confusion resulting in 66 unique

presence records for Gerbilliscus and 203 for Acomys. In Paper II, the georeferenced records

of Saccostomus were supplemented with records from public databases (e.g. African Rodentia,

Terryn et al., 2007; GBIF www.gbif.org, MaNIS www.manis.org). We used 19 BioClim

variables downloaded from the WorldClim website (Hijmans et al., 2005). We modelled

distribution of particular taxa in extant conditions as well as paleoclimatic projections for the

last glacial maximum (LGM, 21 ka; Braconnot et al. 2007) and the last interglacial (LIG,

33


120-140ka; Otto-Bliesner et al. 2006). The species distribution modelling (SDM) analysis was

performed using MaxEnt v. 3.3.3k (Phillips et al. 2004) or using interface with R computing

environment by packages ‘dismo’ (Hijmans et al., 2016) and ‘ENMeval’ (Muscarella et al.,

2014).

Beside SDM presented in Papers II, III, and V, we performed additional SDM for

comparative purpose at finer scale for five clades living in sympatry in Somali-Masai savanna:

Acomys cahirinus group (background data composed of bioclimatic variables from 35

localities), Acomys wilsoni group (55 localities), Gerbilliscus nigricaudus group (20

localities), Gerbilliscus robustus group (81 localities) and Saccostomus mearnsi group (27

localities). We followed the methods described above. The results were convertet in map using

QGIS with treshold sensitivity-specificity sum maximize (Manel et al., 2001; Jiménez-

Valvarde and Lobo, 2007).

34


SUMMARY OF MAIN RESULTS

Somali-Masai savanna harbours phylogenetically distinct lineages of rodents

Using comprehensive material collected by our team and collaborators, we analysed if and to

what extent the rodent diversity in the Somali-Masai savanna is distinct from neighbouring

biogeographical regions within savanna dominated ecosystems, especially compared to

Sudanian and Zambezian savannas. For this purpose we performed phylogenetic analyses of

four widespread rodent genera prefering open habitats. Three of them (Acomys, Mus, and

Gerbilliscus) are widespread across sub-Saharan Africa, while Saccostomus is limited to

eastern and southern Africa. Phylogenetic analyses unambiguously showed that the Somali-

Masai region harbours phylogenetically unique clades in all four genera and exhibit clades

that comprise species complexes, suitable for more detailed comparative phylogeographic

analysis. Although the presence of taxa specific to this region was expected, the occurrence of

endemic clades even in the most abundant rodent groups is surprising and predefines rodents

as suitable phylogeographic models for testing phylogeographic scenaria. Below, I summarize

the main results obtained by phylogenetic analysis of four rodent genera, with main focus on

genetic lineages living in Somali-Masai savanna.

Pygmy mice (genus Mus, subgenus Nannomys)

Rodents of the genus Mus represent one of the most valuable biological models for biomedical

and evolutionary research (Macholán et al., 2012). Currently, four subgenera of genus Mus are

recognized: Mus, Coelomys, Pyromys and Nannomys (Chevret et al., 2005). The African pygmy

mice, Nannomys (Fig. 6a), are small rodents (4-12 g in most taxa, but see Meheretu et al., 2014)

endemic to the sub-Saharan Africa. In Paper I, a comprehensive genetic analysis was

performed for 657 individuals of Nannomys collected at approximately 300 localities across

sub-Saharan Africa. Based on our analysis we provide evidence for African colonization from

Asia in end of Miocene. The radiation of the group started in Eastern Africa and resulted in

important cryptic diversity (27 molecular operational taxonomic units; "MOTUs"). The

subgenus is composed of three basal branches representing mountainous species (M. sp.

„Nyika“, M. imberbis and M. sp. „Harrena“) and five well supported species groups. We call

them herafter triton, setulosus, baoulei, sorella and minutoides groups, based on the

previous use of these names, representing the best known species within particular clades. The

subgenus Nannomys is distributed throughout sub-Saharan Africa (Fig. 7), but there are several

35


lineages endemic to Eastern African savanna. Typical savanna species belong to two species

groups (minutoides and setulosus). Sub-saharan pygmy mice (Mus minutoides Smith 1834) are

widespread in savanna-like habitats across all of sub-Saharan Africa, but with two clades

endemic to Somali-Masai savanna (ET and KE, Fig. 7). Two other taxa from the minutoides

group, Mus cf. gerbillus (not recognized as a valid species by the most recent reviewers, Musser

and Carleton, 2005; Denys et al., 2017) and the delicate mouse Mus cf. tenellus Thomas, 1903

represent distinct genetic lineages adapted to arid habitats of Somali-Masai savana. Finally,

Mus cf. procodon (a basal lineage of the setulosus group, currently synonymised with Mus

setulosus Peters, 1876; Musser and Carleton, 2005; Denys et al., 2017) is a typical inhabitant

of Somali-Masai savanna, with a distribution restricted to Ethiopian Rift Valley.

Fig. 7: Distribution of species an MOTUs of Nannomys in Eastern Africa. Species are represented with different

colours and MOTUs are indicated by different symbol shapes. Grey dots represent genetically confirmd records

of other Nannomys species or MOTUs.

36


The pouched mice (Saccostomus)

Pouched mice, of the genus Saccostomus (Fig. 6b) are typical inhabitants of savannas in

southern and eastern Africa. Saccostomus are small to medium sized, slow-moving rodents,

which are locally abundant in highly seasonal environments of grasslands, savannas, shrublands

and woodlands of southern and eastern Africa (Gaster et al., 2007). In Paper II, we performed

analyses of genetic and morphological variability across the geographic distribution of the

genus. Results confirm the distinction between two recognized species complexes, S.

campestris and S. mearnsi with parapatric distributions in the Zambezian and Somali-Masai

bioregions, respectively. Our molecular and morphological data suggest cryptic diversity.

Southern African pouched mouse S. campestris group consists of two subspecies S. campestris

campestris (Peters, 1846; comprising two CYTB clades) and S. campestris mashonae (de

Winton, 1897) that are moderately differentiated, albeit distinct at nuclear IRBP marker and

skull form. Saccostomus mearnsi group consists of two species distributed in Somali-Masai

savanna, East African pouched mouse S. mearnsi Heller, 1910 and gray-bellied pouched mouse

S. umbriventer Miller, 1910. These lineages are markedly differentiated in both nuclear

markers and skull form and may possibly co-occur in south-western Kenya and north-eastern

Tanzania. The distribution of S. umbriventer is confined to the narrow belt in northern Tanzania

and southwest Kenya, whereas S. mearnsi is distributed from central Kenya to southern

Ethiopia, reaching South Sudan and southern Somalia, where the limits of its distribution are

unresolved (Fig. 8).

37


Fig. 8: Distribution of Saccostomus mearnsi group in East Africa. Species are represented with different colours and shapes. Grey dots represent genetically confirmd records of Saccostomus campestris group.

Gerbils (Gerbilliscus)

Gerbils of the genus Gerbilliscus (Thomas, 1897; Fig. 6c) are widespread rodents living in

savannas, woodlands, grasslands and semi-deserts of sub-Saharan Africa. Currently, four main

clades of Gerbilliscus are recognized: eastern (Paper III), western (Granjon et al., 2012),

southern (McDonough et al., 2015) and the so-called Gerbillurus clade from south-western

Africa (Colangelo et al., 2007). The distribution of the eastern clade of Gerbilliscus (Fig.9) is

largely concordant with the borders of the Somali-Masai region and parapatric with respect to

western and southern clades of Gerbilliscus (Paper III). Multilocus genetic analyses of this

eastern clade provides existence of two main lineages - „robustus“ with four species (Fringe-

tailed gerbil G. robustus Cretzschmar, 1826, East African gerbil G. vicinus Peters, 1878,

Phillips’s gerbil G. phillipsi Winton, 1898 and G. sp. n. (Babile)) and „nigricaudus“ with two

38


species (Black-tailed gerbil G. nigricaudus Peters, 1878 and G. cf. bayeri). Further, we have

found significant phylogeographic structure within largely distributed species G. robustus (two

mitochondrial haplogroups), G. vicinus and G. nigricaudus (four haplogroups each). In addition

to the four valid species, we identified two lineages that may represent separate but

unrecognized species. The first one, Gerbilliscus sp. n. (Babile), was first reported as a

genetically distinct lineage (12.5-17.6% CYTB distances from other species) from Babile

Elephant Sanctuary in eastern Ethiopia (Lavrenchenko et al., 2010). We captured additional

individuals at the same site and increased our sampling by including museum specimens from

other localities in Ethiopia and Somalia. A coalescence-based species tree demonstrated that G.

sp. n. (Babile) is the sister lineage to G. robustus, from which it seems to be separated by a

mountain ridge in the eastern Ethiopian highlands. Present records of G. sp. n. (Babile) suggest

its occurence is asociated with transitional semi-evergreen bushland (van Breugel et al., 2016),

which is localized only in the narrow belt along mountain chains in Ehtiopia. We referred to

the second currently unrecognized species as Gerbilliscus cf. bayeri. We sequenced a syntype

of G. nigricaudus bayeri and it clustered within a clade distributed west of the Rift Valley, as

sister to G. nigricaudus. Based on the results of a coalescent species delimitation and high

CYTB genetic distances (13.4-15.1% from G. nigricaudus), this clade may deserve the status

of a separate species.

39


Fig. 9: Distribution of Gerbilliscus in East Africa. Species are represented with different colours and shapes. Grey dots represent genetically confirmd records of other Gerbilliscus species.

Spiny mice (Acomys)

Spiny mice of the genus Acomys (Fig. 6d) represent a speciose genus, widely distributed in arid

and semi-arid regions of Africa, Arabia and Middle East (Fig. 10). We compiled the largest

multilocus genetic dataset to date for this genus (482 genotyped individuals from more than

200 localities covering a majority of its distribution), reconstructed phylogenetic relationships

and estimated the spine mice diversity. Phylogenetic analysis revealed five major clades:

subspinosus, spinosissimus, russatus, wilsoni and cahirinus. We show the presence of 26

genetically distinct lineages (MOTUs) with the highest diversity in Somali-Masai savanna

(11 lineages). The highest diversity is reported from the cahirinus group with 7 lineages

endemic to Eastern Africa (in total 16). The wilsoni group is divided into four well supported

lineages distributed only in the Somali-Masai savanna (Paper V).

40


Fig. 10: Distribution Acomys MOTUs in East Africa. a) Distribution of Acomys cahirinus group; b) Distribution of Acomys wilsoni group. Grey dots represent genetically confirmd records of other Acomys MOTUs.

41


Optimalization of methods for divergence dating in phylo(bio-)geographic

reconstructions

Although divergence time dating is now a well-established cornerstone of evolutionary

biology, there is still no widely accepted objective methodology for converting data from the

fossil record to calibration information of use in molecular phylogenies (Drummond and

Bouckaert, 2015). In the last few years, several methodological approaches to better

implement fossil calibrations have been developed, for instance allowing one to directly

include fossil lineages in phylogenies (‘total-evidence dating’; Pyron, 2010; Ronquist et al.

2012b) or to account for information on the density of the fossil record (‘fossilized birth-death

(FBD) process’; Stadler, 2010; Heath et al., 2014). Nevertheless, there is still lack of reliable

fossil records for family Muridae. Until 2008 there were numerous studies based only on a

single calibration point represented by the split „Mus/Rattus“ (e.g. Adkins et al., 2001; Steppan

et al., 2004). This split represent separation of two fossil genera †Progonomys and †Karnimata.

†Progonomys was considered the ancestor of the Mus lineage and †Karnimata as the ancestor

of a lineage containing Rattus (Jacobs and Flynn, 2005). After detailed study of these fossils,

Kimura et al. (2015) showed, that †Karnimata is more related to current Arvicanthini tribe,

therefore this split represent divergence between Mus and Arvicanthis. This example shows

how important is to localize particular fossils on the phylogenetic tree correctly.

The uncertainity in using different fossil record (or different ages) leads to very distinct

estimation of divergence dating. In Paper IV, we examine previous studies to compare

different estimations of divergence times in the family Muridae. Most controversy is usually

in the calibration of molecular clock in the crown Murinae. In the past, the authors most often

calibrated it based on †Progonomys/†Karnimata split as shown above (with estimated age of

crown Murinae 10-14 Ma; Schenk et al., 2013, Fabre et al., 2013, Rowe et al., 2008, 2013,

Lecompte et al., 2008; Steppan et al., 2004; or 23 Ma in Adkins et al., 2001). Nevertheless,

there was no clear consensus, if this calibration point is on the node between Rattus and Mus,

or on the crown of Murinae (including Phloeomyini, the most basal lineage or not). The other

very controversal fossil calibration is for the tribe Apodemyini or Apodemus clade. In both

cases the authors used the fossil †Parapodemus, with different ages (4.89-11 Ma, Lecompte et

al., 2008; Fabre et al., 2013; Schenk et al., 2013, Bryja et al., 2014).

For correctly estimation of divergence date for family Muridae, the list of reliable fossil

records is needed. The main advantage of our Paper IV lies in detailed descriptions of each

42


fossil record, which can be re-evaluated if needed in the future (see Appendix B in Paper IV).

Each candidate fossil possesses the information about the collection site, unique identification

number, and the state of preservation along with justification for the age of the fossil (i.e., age

of fossil-bearing formation and stratigraphic level, preferably with an absolute age by

radiometric dating and/or reliable relative age estimates, for example, by magnetostratigraphy;

following Parham et al., 2012). Before assignation of fossil to specific node of the phylogeny,

all potential fossil constraints were carefully assessed. They were subjected to several cross-

validation analyses (following protocols described in Near and Sanderson, 2004; Near et al.,

2005). From 18 candidated fossil records, we select nine for final divergence time

estimation (Table 4, more details in Paper IV).

Table 4: Overview of fossils finally selected for the divergence dating of family Muridae (modified from Tables 3 and 4 in Paper IV).

Fossil Age (Ma) Locality References

1 †Antemus chinjiensis

13.8 Pakistan, Potwar Plateau, YGSP 491

Jacobs et al. (1989)

2 †cf. Karnimata sp. 11.2 Pakistan, Siwalik Group, Nagri Formation, YGSP 791, YGSP 797

Jacobs and Flynn (2005); Kimura et al. (2015)

3 †Parapodemus lugdunensis

9.6 France, Dionay Lungu (1981), Mein et al. (1993), Renaud et al. (1999)

4 †Karnimata

darwini

9.2 Pakistan, Siwalik Group, Dhok Pathan Formation, YGSP 182

Jacobs (1978); Kimura et al. (2015)

5 †Abudhabia pakistanensis

8.7 Pakistan, Siwalik Group, Dhok Pathan Formation, YGSP 387

Flynn and Jacobs (1999)

6 †Mus sp. 8.0 Pakistan, Siwalik Group, Dhok Pathan Formation, YGSP 547

Kimura et al. (2013, 2015)

7 †Aethomys sp. 6.1 Kenya, Lemudong'o, locality 1 Manthi (2007)

8 †Arvicanthis sp. 6.1 Kenya, Lemudong'o, locality 1 Manthi (2007)

9 †Gerbilliscus sp. 6.1 Kenya, Lemudong'o, locality 1 Manthi (2007)

43


Age (Ma)

1 Pakistan, Potwar Plateau,

2 sp. Group, Nagri nd Flynn (2005);

3 Lungu (1981), Mein et al.

4 Pakistan, Siwalik Group, Dhok Jacobs (1978); Kimura et

5 Pakistan, Siwalik Group, Dhok Flynn and Jacobs (1999)

6 Pakistan, Siwalik Group, Dhok 2015)

7 Kenya, Lemudong'o, locality 1

8 Kenya, Lemudong'o, locality 1

9 sp. Kenya, Lemudong'o, locality 1

Evolutionary history of Somali-Masai savanna in Plio-Pleistocene

The evolutionary history of African savanna ecosystems and their biota has started in Miocene.

During the Early Miocene (23.0-16.0 Ma), the reopening of the Mediterranean-Indo-Pacific

seaway (Rögl, 1999) separated Africa from Eurasia, thus giving rise (among others) to the main

clades of Afrotropic murid rodents. Three subfamilies, Deomyinae, Gerbillinae and

Lophiomyinae, then likely diversified primarily in the Afrotropics (Chevret and Dobigny,

2005; Ndiaye et al., 2016; Schenk et al., 2013; Paper IV). The subfamily Murinae started to

diversify in Indomalaya, most probably in Southeast Asia, where is the highest phylogenetic

diversity of this subfamily (Fabre et al., 2013). In comparison family Nesomyidae has origine

in Africa and Madagascar (Schenk et al., 2013).

During the Middle Miocene (16.0-11.6 Ma) a global cooling caused general

aridification („Parathethys Salinity Crisis“; Rögl, 1999) and vegetation shifts (from moisture-

adapted C3 plants to tropical arid-adapted C4 grasses; Cerling et al., 1997, 1998). This climatic

and faunal changes (savanna expansions) gave rise the current rodent genera in Africa, e.g.

Acomys split from sister genus Deomys (Papers IV,V), Gerbilliscus split from Desmodillus and

Tatera (Papers III, IV) and particular clades of Murinae colonized Africa and some of them

intensively radiated there (e.g. Praomyini or Arvicanthini; Paper IV).

The first splits in model genera studied in my dissertation occured in the Late Miocene

(11.6-5.3 Ma). For example, the Eastern Gerbilliscus clade split from the southern and western

clades around 7.03 Ma (HPD 6.17-10.13 Ma; Paper III). The similar pattern was found in genus

Acomys (Paper V). The first Acomys divergence i.e. between lineages from Zambezian region

(subspinosus+spinosissimus) and Somali-Masai region (russatus+cahirinus+wilsoni) was

estimated to occurred 8.69 Ma (HPD 8.51-9.29 Ma). At the same time, there was probably split

between the genus Saccostomus and its sister genera Cricetomys and Beamys (Paper II). These

separations were caused most probably by "coast-to-coast" tropical forest, which is evidence

for occurrence tropical forest is evidenced by phylogenetic analyses of plants and animals living

in nowadays fragmented forests of Congo basin and eastern African montane and coastal forests

(Couvreur et al., 2008; Bryja et al., 2017). This continuous forest was likely one of the most

important factors in early evolution of savanna inhabitants, because is separated northern (=

Somali-Masai) and southern (= Zambezian) savannas. Around 8 Ma, the maximum of the shift

from C3 forest and woodland to C4 savanna occurred (Cerling et al., 1998; rodents fossils

Kimura et al., 2013; bovid fossils Garrett et al., 2015). During short period the Somali-Masai

44


savanna was probably connected with the Mediterranean area in Egypt and Sinai –

probably during this time Acomys russatus split from the rest of the East African group and

dispersed to the north along the Nile river. Later (ca 7 Ma) the Sahara desert become

hyperarid (Pickford et al., 2006; Schuster et al., 2006; Carranza et al., 2008) and this migration

route was no longer possible (but this vegetation corridor was repeatedly used for dispersal of

savanna taxa much later (e.g. Arvicanthis niloticus, Dobigny et al. 2013; Crocidura olivieri,

Jacquet et al., 2015; Acomys cahirinus group, Paper V). The aridification of Africa was

promoted also by the Messinian salinity crisis at ca. 5.3 Ma (Krijgman et al., 1999; Maslin

and Chirstensen, 2007; Hodell et al., 2001). During this period of global sea level depression

(Haq et al., 1987), Africa and Arabia were reconnected through Neguev-Sinai landbridge

(„Levantine corridor“, Fernandes et al., 2006) and/or landbridge in the Bab-el-Mandeb

(Bosworth et al., 2005). In murids, evidence to support this faunal exchange can be found in

the Mus (Nannomys), which colonized Afrotropics and started there its radiation ca. 5.2 Ma

(Paper II).

During the Pliocene (5.3-2.5 Ma), futher diversification within the model genera

continued. This period gave rise to different groups of Acomys (estimation of the first

diversification in wilsoni group 4.67 Ma, split between „ignitus“ and „Djibouti“ clade 4.57 Ma;

Paper V), and split between major Saccostomus groups (mearnsi and campestris 3.9 Ma; Paper

II). With influence from intensification of the Northern Hemisphere Glaciation (iNHG, 3.2-

2.5 Ma; Haug and Tiedemann, 1988), Acomys dimidiatus diverged after colonisation of the

Arabian peninsula. At least two different colonization events occured for Acomys species from

Somali-Masai into Sudanian savanna, where they diverged into A. johannis complex (Cah2-

Cah4) and A. chudeaui (Cah11; Paper V). Connection between the Somali-Masai and Sudanian

savannas is further evidenced by westward dispersal of Gerbilliscus robustus (clade R2; Paper

III). There are also some other examples of bi-directional exchange of rodent fauna between

Somali-Masai and Sudanian savanna, e.g. Mastomys erythroleucus (Brouat et al., 2009),

Arvicanthins niloticus (Dobigny et al., 2013), or Gerbilliscus kempi in (Granjon et al., 2012,

our unpublished data). Our ecological niche modelling (Papers III and V) suggests that

conditions for east-west migration have been suitable especially during more humid periods

(represented by last interglacial in our models). Because ENM predicts only patchy suitable

habitats for Somali-Masai taxa in Sudanian savanna, is likely why the distribution of other

probably more specialized Somali-Masai taxa (e.g. Grant’s gazelle, Beisa oryx, Grevy’s zebra

45


or pouched mice) remained limited to the easternmost part of Africa (Lorenzen et al., 2012;

Kebede et al., 2014; Paper II).

The Pleistocene was a climatically very turbulent period that significantly affected

current distribution of genetic diversity of Somali-Masai fauna. Within the genus Gerbilliscus,

all six current species almost simultanously diversified around 1.7 Ma: split of G. robustus and

G. sp. n. (Babile) is estimated at 1.62 Ma (HPD 0.73-2.95 Ma), G. phillipsi and G. vicinus at

1.75 Ma (HPD 0.73-3.31) and G. nigricaudus adn G. cf. bayeri at 1.80 Ma (HPD 0.86-3.53

Ma). Most extant Acomys species occured already in Pleistocene, including the second

migration from Somali-Masai region to the north (Cah9 = A. cahirinus). This period was

important also for speciation in Mus (Nannomys), including the radiation within the minutoides

group in open habitats across whole sub-Saharan Africa and, later on, the spread of a single

species, M. minutoides, again across whole sub-Saharan Africa. Two species of Saccostomus

mearnsi group split at 2.0 Ma (HPD 1.3-2.9 Ma). S. umbriventer occupied (based on predicted

distribution modelling) southern refugium in Tanzania and S. mearnsi refugium north near Lake

Victoria and Lake Turkana (Paper II).

Species distribution modelling and the extent of Somali-Masai savanna

The distribution of model groups in my thesis was most significantly influenced by annual

precipitation (bio12) and precipitation of coldest quarter (bio19). There is not a substantial

difference between predicted distribution models for present, LGM, and LIG for taxa

primarily distributed in Somali-Masa savanna (Fig. 11). This is in contrast to rodents living in

Zambezian savanna, where suitable area shrank to very small refugia especially during LIG

period, and current populations show signs of recent population expansion (e.g. Gerbilliscus

leucogaster McDonough et al., 2016; Aethomys chrysophilus Mazoch et al., 2017; Saccostomus

campestris group, Paper II).

Different model groups had slightly different habitat preferences within Somali-

Masai savanna (e.g. Acomys generally prefers rocky habitats, Saccostomus grasslands, and

Gerbilliscus shrubs, etc.), but their predicted distribution shows coincident pattern (Fig. 11).

We can therefore use the results of ENMs for generalization concerning the extent of suitable

habitats of organisms typical for Somali-Masai savanna. The southern border of Somali-

Masai savanna is in northern and central Tanzania, which is concordant with the

distribution of most model species. The southern Kenya and northern Tanzania are

46


important centers of biodiversity of savanna-like ecosystems. Cuthbert et al. (2017) showed

that this region is water-rich (e.g. rivers: Tana, Galana, Tsavo, Pangani; springs and

groundwater). Groundwater, which acts to buffer climate variability, in the form of springs

and baseflow to perennial streams, created hydro-refugia that likely persisted through long dry

periods (Cuthbert et al., 2017). Important barriers to dispersal for savanna inhabitants (both

current and past) are mountains like Mt. Elgon or the Kenyan Highlands (Cherrangani, Mau,

Mt. Kenya), where more continuous forests are expected especially in humid periods of

Pleistocene. In contrast, the Tanzanian Highlands (Mt. Kilimanjaro, northern part of Eastern

Arc Mts) probably did not play as important of a role, even if they probably separated some

intraspecific lineages of savanna species (see below). Another important unsuitable habitat is

represented by extremly dry Masai xeric shrublands near lake Turkana (Fig. 4). Extremely

unsuitable conditions are also present in the Horn of Africa, usually considered as large part

of Somali-Masai region (e.g. Linder et al. 2012), where the only suitable conditions occur on

the sout-eastern part of Somalia. Suitable climatic conditions for some taxa are also predicted

in the south of Arabian peninsula (Fig. 11), but from the model groups only Acomys are known

to have succesfully colonized this area.

47


Fig. 11: MaxEnt predictions of the present distribution (green), last glacial maximum (ca 21 ka; blue) and last intreglacial period (120-130 ka, red) for a) Acomys cahirinus group, b) Acomys wilsoni group, c) Gerbilliscus

nigricaudus group, d) Gerbilliscus robustus group, e) Saccostomus mearnsi group.

48


CONCLUSIONS AND FUTURE DIRECTIONS

Genetic diversity

This Ph.D. thesis provides the first analysis of detailed genetic structure for rodents from

Somali-Masai savanna, including identification of the main factors responsible for their

evolutionary history since the Miocene. Using molecular methods, we uncovered cryptic

diversity in this region, where the basic faunistic data were largely missing.

Most studies in the thesis are based on initial DNA barcoding of collected material

using the mitochondrial CYTB gene, followed by analysis of several nuclear genes for each

mitochondrial clade. This approach has, of course, some weak points. Nuclear markers are often

not variable enough for reliable reconstruction of phylogeny of closely related species, and

multilocus phylogenetic trees are still biased by the CYTB phylogeny. In such cases,

mitochondrial introgression can lead to uncorrect species trees and/or species misidentification.

In the future, it can be advantagous to use genomic approaches to uncover unbiased

phylogenies and population structure, e.g. by full/representative genome sequencing (e.g.

anchored hybrid enrichment, Lemmon et al., 2012; RAD sequencing, Leaché et al., 2014; Kai

et al., 2014, etc.).

The advance of next-generation sequencing can be used also by genotyping of museum

specimens. The importance of museum specimens has increased in last years, because

fieldwork in some parts of Eastern Africa is not possible for political or other reasons (e.g.

South Sudan, Somalia). Additionaly, museums keep type material collections, which is

essential for further taxonomic investigation. Nevertheless, museum conditions were optimized

for expecially specimen longevity, not molecular stability, therefore work with museum DNA

is challenging (e.g. DNA is fragmented, degradated and/or contains DNA contaminations).

Nevertheless, in last decade several methods were designed to to overcome these complications:

a) amplicon sequencing (Hajibabaei et al., 2006; Galane et al., 2012); b) whole-genome

shotgun sequencing (Tilak et al., 2015); c) target enrichment (Mason et al., 2011); d)

sequence capture of ultraconserved elements (McCormack et al., 2016).

Optimalization of methods for divergence dating

The second part of this thesis is dealing with optimalization of fossil calibrations for estimation

of the dated phylogeny of muroid rodents. We have created a detailed, annotated fossil

49


database including 18 fossils with description, why each fossils should or shouldn’t be used

for calibration of ultrametic divergence trees.

We also see at least one weakness of our study (Paper IV). The fossil constrains are

assigned on phylogenetic tree based only on paleontological information. The detailed

information about teeth morphology were considered, but they were not always available.

Recent advances in divergence time estimation, e.g. „tip dating“ (Ronquist et al., 2012b) and

fossilized birth-death processes (Stadler, 2010; Heath et al., 2014; Drummond and Stadler et

al., 2016) allow integration of fossil taxa directly into phylogenetic analyses based on both

molecular (extant species) and morphological data (extinct and extant species). Therefore

allowing for better assignment of fossils onto the phylogenetic tree of current taxa, it should be

very useful to prepare morphometric matrix for extinct and extant taxa as well. Such dataset

then could be analysed by a „total evidence dating“ (Ronquist et al., 2012b; Pyron, 2011). This

methods allows one to use all available fossil records, not only the oldest representatives of

selected lineages (Gavryushkina et al., 2014).

Evolutionary history of the Somali-Masai savanna

In the third part of my Ph.D. thesis, I identified factors, which played role in evolution of

Somali-Masai rodent fauna during the Plio-Pleistocene. Our molecular dataset, with estimation

of divergence dates and modelling of present and past species distributions might facilitate our

understanding of evolutionary history of savanna in Eastern Africa.

Repeated expansions and contractions of savanna habitats were caused by global and local

factors. The main global climatic transitions, which influence Eastern Africa were Mesian

salinity crisis (6-5.3 Ma) and intesification of Norhern Hemisphere glaciation (3.2-2.5 Ma). The

local geomorphology play important role in evolution of the East African biota. The current

distribution of genetic diversity of Somali-Masai region seems to be influenced mainly by

combination of following factors:

(1) Mountain chains. The most important mountains are Kenyan Highlands (from Mt.

Elgon to Mt. Kenya), especially forests currently confined to their higher altitudes. In

Ethiopia, a similar role was played by Ethiopian highlands, especially in its Eastern part

where mountains separate Somalian and Afar lowlands.

50


(2) Regions of extreme aridity. Extremly dry environmens are not suitable even for

savanna-living organisms. Based on distribution maps and species distribution modelling

we have observed, that Masai xeric shrublands in northern Kenya and Horn of Africa

are too dry and very likely uninhabited by model organism studied in this thesis.

(3) Water bodies. One of the factors, which influenced distribution of model organisms is

accesibility to water resources as underground water acts as a buffer to climate

variability and promotes the long-term refugia of open habitats (e.g. in Tanzania/Kenya

border; Cuthbert et al. 2017). On the other hand, water bodies are important natural

barriers to gene flow for terrestrial organisms. Rifting generated numerous Rift basins,

either filled with water or were extremely dry with halophytic vegetation (White, 1983),

which could prevent the dispersal of small terrestrial animals (Trauth et al., 2010; Maslin

et al., 2014). Genetic structure of multiple Somali-Masai taxa studied in this thesis was

likely affected by rift lakes, separating populations west vs. east of the rift, or north vs.

south within the rift itself.

These factors, likely influencing the evolutionary history for taxa inhabiting the Somali-

Masai region, are selected on the basis of the studies of limited number rodent genera. To assess

their general role in evolution of savanna organisms, it will be necessary to perform also studies

on other savanna living organism (reptiles, invertebrates, plants). Comparative

phylogeographic approach will be important not only for general understanding of evolution

of wildlife in Eastern Africa (with numerous conservation implications), but at the same time

for better understanding of our own history, because similar factors in the same region very

likely influenced also the evolution of genus Homo.

51


REFERENCES

Aghová, T., Šumbera, R., Piálek, L., Mikula, O., Mcdonough, M.M., Lavrenchenko, L.A., Meheretu, Y., Mbau, J.S., Bryja, J., 2017. Multilocus phylogeny of East African gerbils (Rodentia, Gerbilliscus) illuminates the history of the Somali-Masai savanna. Journal of Biogeography, 44(10), 2295–2307.

Adkins, R.M., Gelke, E.L., Rowe, D., Honeycutt, R.L., 2001. Molecular phylogeny and divergence time estimates for major rodent groups: evidence from multiple genes. Mol. Biol. Evol. 18, 777–91.

Baker, B.H., Mitchell, J.G., Williams, L.A.J., 1988. Stratigraphy, geochronology and volcano-tectonic evolution of the Kedong-Naivasha-Kinangop region, Gregory Rift Valley, Kenya. Journal of the Geological Society of London 145, 107–116.

Beard, J.S., 1953. The Savanna Vegetation of Northern Tropical America. Ecol. Monogr. 23, 149–215. Beentje, H.J., Adams, B., Davis, S.D., Hamilton, A.C., 1994. Regional overview: Africa. Centres of plant

diversity (ed. by S.D. Davis, V.H. Heywood and A.C. Hamilton), pp. 101–148. IUCN Publication Unit, Cambridge. Beerling,

Bosworth, W., Huchon, P., McClay, K., 2005. The Red Sea and Gulf of Aden Basins. J. African Earth Sci. 43, 334–378. doi:10.1016/j.jafrearsci.2005.07.020

Bourliere, F., Hadley, M., 1983. Present-day savannas: an overview. Ecosystems of the world. Braconnot, P., Otto-Bliesner, B., Harrison, S., Joussaume, S., Peterchmitt, J.Y., Abe-Ouchi, A., Crucifix, M.,

Driesschaert, E., Fichefet, T., Hewitt, C.D., Kageyama, M., Kitoh, A., Laîné, A., Loutre, M.F., Marti, O., Merkel, U., Ramstein, G., Valdes, P., Weber, S.L., Yu, Y., Zhao, Y., 2007. Results of PMIP2 coupled simulations of the Mid-Holocene and Last Glacial Maximum - Part 1: experiments and large-scale features. Clim. Past 3, 261–277. doi:10.5194/cp-3-261-2007

Breugel, P., Friis, I., Demissew, S., 2016. The transitional semi-evergreen bushland in Ethiopia: characterization and mapping of its distribution using predictive modelling. Applied vegetation science, 19(2), pp.355-367.

Brouat, C., Tatard, C., Bâ, K., Cosson, J.F., Dobigny, G., Fichet-Calvet, E., Granjon, L., Lecompte, E., Loiseau, A., Mouline, K. and Piry, S., 2009. Phylogeography of the Guinea multimammate mouse (Mastomys erythroleucus): a case study for Sahelian species in West Africa. Journal of Biogeography, 36(12), pp.2237-2250.

Bryja, J., Mikula, O., Šumbera, R., Meheretu, Y., Aghová, T., Lavrenchenko, L. A., Mazoch, V., Oguge, N., Mbau, J.S., Welegerima, K., Amundala, N., Colyn, M., Leirs, H., Verheyen, E., 2014. Pan-African phylogeny of Mus (subgenus Nannomys) reveals one of the most successful mammal radiations in Africa. BMC Evolutionary Biology, 14(1), 256.

Brown, D.M., Brenneman, R.A., Koepfli, K.-P., Pollinger, J.P., Milá, B., Georgiadis, N.J., Louis, E.E., Grether, G.F., Jacobs, D.K., Wayne, R.K., 2007. Extensive population genetic structure in the giraffe. BMC Biol. 5, 57. doi:10.1186/1741-7007-5-57

Burgess, N.D., D’Amico Hales, J., Underwood, E., Dinerstein, E., Olson, D., Itoua, I., Schipper, J., Ricketts, T., Newman, K., 2004. Terrestrial ecoregions of Africa and Madagascar: a continental assessment. Island Press, Washington, DC.

Burgess, N.D., Balmford, A., Cordeiro, N.J., Fjeldså, J., Kuper, W., Rahbek, C., Sanderson, E.W., Scharlemann, J.P.W., Sommer, J.H., Williams, P.H., 2007a. Correlations among species distributions, human density and human infrastructure across the high biodiversity tropical mountains of Africa. Biological Conservation, 134, 164–177

Burgess, N.D., Butynski, T.M., Cordeiro, N.J., Doggart, N.H., Fjeldså, J., Howell, K.M., Kilahama, F.B., Loader, S.P., Lovett, J.C., Mbilinyi, B., Menegon, M., Moyer, D.C., Nashanda, E., Perkin, A., Rovero, F., Stanley, W.T., Stuart, S.N., 2007b. The biological importance of the Eastern Arc Mountains of Tanzania and Kenya. Biol. Conserv. 134, 209–231. doi:10.1016/j.biocon.2006.08.015

Carranza, S., Arnold, E.N., Geniez, P., Roca, J., Mateo, J.A., 2008. Radiation, multiple dispersal and parallelism in the skinks, Chalcides and Sphenops (Squamata: Scincidae), with comments on Scincus and Scincopus and the age of the Sahara Desert. Mol. Phylogenet. Evol. 46, 1071–1094. doi:10.1016/j.ympev.2007.11.018

Cerling, T.E., Harris, J.M., Macfadden, B.J., Leakey, M.G., Quadek, J., Eisenmann, V., Ehleringer, J.R., 1997. Global vegetation change through the Miocene / Pliocene boundary. Nature 389, 153–158. doi:10.1038/38229

Cerling, T.E., Ehleringer, J.R., Harris, J.M., 1998. Carbon dioxide starvation, the development of C4 ecosystems, and mammalian evolution. Philos. Trans. R. Soc. B Biol. Sci. 353, 159–171. doi:10.1098/rstb.1998.0198

52


Chevret, P., Dobigny, G., 2005. Systematics and evolution of the subfamily Gerbillinae (Mammalia, Rodentia, Muridae). Mol. Phylogenet. Evol. 35, 674–688. doi:10.1016/j.ympev.2005.01.001

Chevret, P., Veyrunes, F., Britton-Davidian, J., 2005. Molecular phylogeny of the genus Mus (Rodentia: Murinae) based on mitochondrial and nuclear data. Biological Journal of the Linnean Society, 84(3), pp.417-427.

Colangelo, P., Granjon, L., Taylor, P.J., Corti, M., 2007. Evolutionary systematics in African gerbilline rodents of the genus Gerbilliscus: inference from mitochondrial genes. Mol. Phylogenet. Evol. 42, 797–

806. doi:10.1016/j.ympev.2006.10.001 Cole, M. M., 1963. Vegetation nomenclature and classification with particular reference to the savannas.

South African Geogr. J. 55, 3–14. Cordeiro, N.J., Burgess, N.D., Dovie, D.B.K., Kaplin, B.A., Marrs, R., 2007. Conservation in areas of high

population density in sub-Saharan Africa. Biological Conservation, 134, 155–163 Couvreur, T.L., Chatrou, L.W., Sosef, M.S., Richardson, J.E., 2008. Molecular phylogenetics reveal multiple

tertiary vicariance origins of the African rain forest trees. BMC Biology, 6(1), p.54. Cuthbert, M.O., Gleeson, T., Reynolds, S.C., Bennett, M.R., Newton, A.C., McCormack, C.J., Ashley, G.M.,

2017. Modelling the role of groundwater hydro-refugia in East African hominin evolution and dispersal. Nat. Commun. 8, 15696. doi:10.1038/ncomms15696

Dansereau, P., 1957. Biogeography. An ecological perspective. The Ronald Press Co., New York. Denys, C., Taylor, P.J., Aplin, K.P., 2017. Family Muridae. In: Wilson et al. (Eds.) Hadbook of the Mammals

of the World, Volume 7: Rodents II. Lynx Edicions. ISBN-13: 9788416728046 Dixon, A.P., Faber-Langendoen, D., Josse, C., Morrison, J., Loucks, C.J., 2014. Distribution mapping of

world grassland types. J. Biogeogr. 41, 2003–2019. Dobigny, G., Tatard, C., Gauthier, P., Ba, K., Duplantier, J.M., Granjon, L. and Kergoat, G.J., 2013.

Mitochondrial and nuclear genes-based phylogeography of Arvicanthis niloticus (Murinae) and sub-Saharan open habitats Pleistocene history. PLoS One, 8(11), p.e77815.

Drummond, A. J., Bouckaert, R. R., 2015. Bayesian evolutionary analysis with BEAST. Cambridge University Press.

Drummond, A.J., Stadler, T., 2016. Bayesian phylogenetic estimation of fossil ages. Philos. Trans. R. Soc. London B 371, 20150129. doi:10.1098/rstb.2015.0129

Drummond, A.J., Suchard, M.A., Xie, D., Rambaut, A., 2012. Bayesian phylogenetics with BEAUti and the BEAST 1.7. Molecular biology and evolution, 29(8), pp.1969-1973.

Ducroz, J.F., Volobouev, V., Granjon, L., 1998. A molecular perspective on the systematics and evolution of the genus Arvicanthis (Rodentia, Muridae): inferences from complete cytochrome b gene sequences. Mol. Phylogenet. Evol. 10, 104–17. doi:10.1006/mpev.1997.0477

Ebinger, C.J., Temane, T., Harding, D., Tesfaye, S., Kelley, S., Rex, D., 2000. Rift deflection, migration, and propagation: linkage of the Ethiopian and Eastern rifts, Africa. Geological Society of America Bulletin 112, 163–176.

Engler, A., 1879-1882. Versuch einer Entwicklungsgeschichte der Pflanzenwelt. Engelman, Leipzig Faber-Langendoen, D., Keeler-Wolf, T., Meidinger, D., Tart, D., Hoagland, B., Josse, C., Navarro, G.,

Ponomarenko, S., Saucier, J.P., Weakley, A., Comer, P., 2014. EcoVeg: A new approach to vegetation description and classification. Ecol. Monogr. 84, 533–561. doi:10.1890/13-2334.1

Fabre, P.H., Pagès, M., Musser, G.G., Fitriana, Y.S., Fjeldså, J., Jennings, A., Jønsson, K.A.., Kennedy, J., Michaux, J., Semiadi, G., Supriatna, N., Helgen, K.M., 2013. A new genus of rodent from Wallacea (Rodentia: Muridae: Murinae: Rattini), and its implication for biogeography and Indo-Pacific Rattini systematics. Zool. J. Linn. Soc. 169, 408–447. doi:10.1111/zoj.12061

Fedorov, V.B., Goropashnaya, A. V., Boeskorov, G.G., Cook, J.A., 2008. Comparative phylogeography and demographic history of the wood lemming (Myopus schisticolor): Implications for late Quaternary history of the taiga species in Eurasia. Mol. Ecol. 17, 598–610. doi:10.1111/j.1365-294X.2007.03595.x

Fennessy, J., Bidon, T., Reuss, F., Kumar, V., Elkan, P., Nilsson, M.A., Vamberger, M., Fritz, U. and Janke, A., 2016. Multi-locus analyses reveal four giraffe species instead of one. Current Biology, 26(18), pp.2543-2549.

Fernandes, C.A., Rohling, E.J., Siddall, M., 2006. Absence of post-Miocene Red Sea land bridges: Biogeographic implications. J. Biogeogr. 33, 961–966. doi:10.1111/j.1365-2699.2006.01478.x

Fichet-Calvet, E., Rogers, D.J., 2009. Risk maps of lassa fever in West Africa. PLoS Negl. Trop. Dis. 3. doi:10.1371/journal.pntd.0000388

Fjeldsa, J., Burgess, N.D., Blyth, S., de Klerk, H.M., 2004. Where are the major gaps in te reserve network for Africa’s mammals? Oryx 38, 17–25.

53


Flynn, L.J., Jacobs, L.L., 1999. Late Miocene small mammal faunal dynamics: the crossroads of the Arabian peninsula. in: Whybrow, P.J., Hill, A., (Eds.), Fossil vertebrates of Arabia. Yale University Press, New Haven. pp. 412–419.

Foster, A., Ebinger, C., Mbede, E., Rex, D., 1997. Tectonic development of the northern Tanzanian sector of the East African Rift System. Journal of the Geological Society 154, 689–700.

Frankham, R., Ballou, J.D., Briscoe, D.A., 2004. A primer of conservation genetics. Cambridge University Press, Cambridge

Friis, I., Demissew, S., van Breugel, P. 2010. Atlas of the potential vegetation of Ethiopia. The Royal Danish Academy of Science and Letters, Copenhagen.

Galan, M., Page, M., Cosson, J.F., 2012. Next- !"!#$%&'"( )!*+!",&"-( .'#( /'0!"%( 1$#,'0&"-23( )4!,&!5(Identification from Fresh, Degraded and Environmental Samples. PLoS One 7. doi:10.5061/dryad.1j6v6

Galster, S., Burgess, N. D., Fjeldsa, J., Hansen, L. A., Rahbek, C. 2007. One Degree Resolution Databases of the Distribution of 1085 Species of Mammals in Sub-Saharan Africa. On-line data source-Version 1.00. Denmark: Zoological Museum, University of Copenhagen. Available via http://www.daim.snm.ku.dk/african-vertebrates-search

Garrett, N.D., Fox, D.L., McNulty, K.P., Faith, J.T., Peppe, D.J., Van Plantinga, A., Tryon, C.A., 2015. Stable isotope paleoecology of late Pleistocene middle stone age humans from the Lake Victoria Basin, Kenya. Journal of human evolution, 82, pp.1-14.

Gavryushkina, A., Welch, D., Stadler, T., Drummond, A.J., 2014. Bayesian Inference of Sampled Ancestor Trees for Epidemiology and Fossil Calibration. PLoS Comput. Biol. 10, e1003919. doi:10.1371/journal.pcbi.1003919

Geist, H.J., Lambin, E.F., 2004. Dynamic Causal Patterns of Desertification. Bioscience 54, 817. '67( 0!( 1!88',*9( :;9( <=>?@'AB9( :;9( C!D!#!%+9( E;9( F?G@'AB9( H;9( 1#7I'AB9( J;9( K!&#59( L;9( 1#7I$9( :;9( MNOP;(

Complete genome characterisation and phylogenetic position of Tigray hantavirus from the Ethiopian white-footed mouse, Stenocephalemys albipes. Infect. Genet. Evol. 45, 242–245. doi:10.1016/j.meegid.2016.09.009

Granjon, L., Colangelo, P., Tatard, C., Colyn, M., Dobigny, G., Nicolas, V., 2012. Intrageneric relationships within Gerbilliscus (Rodentia, Muridae, Gerbillinae), with characterization of an additional West African species. Zootaxa 25, 1–25.

Hajibabaei, M., Singer, G.A.C., Hebert, P.D.N., Hickey, D.A., 2007. DNA barcoding: how it complements taxonomy, molecular phylogenetics and population genetics. Trends Genet. 23, 167–72. doi:10.1016/j.tig.2007.02.001

Hamilton, A.C., 1976. The significance of patterns of distribution shown by forest plants and animals in tropical Africa for the reconstruction of upper Pleistocene palaeoenvironments: a review. Paleoecology of Africa, 9, 63–97

Haq, B.U., Hardenbol, J.A.N., Vail, P.R., 1987. Chronology of fluctuating sea levels since the Triassic. Science, 235(4793), pp.1156-1167.

Haug, G.H., Tiedemann, R., 1998. Effect of the formation of the Isthmus of Panama on Atlantic Ocean thermohaline circulation. Nature, 393(6686), p.673.

Heath, T.A., Huelsenbeck, J.P., Stadler, T., 2014. The fossilized birth–death process for coherent calibration of divergence-time estimates. Proceedings of the National Academy of Sciences, 111(29), pp.E2957-E2966.

Heled, J., Drummond, A.J., 2010. Bayesian inference of species trees from multilocus data. Mol. Biol. Evol.

27, 570–80. doi:10.1093/molbev/msp274 Hijmans, R.J., Cameron, S.E., Parra, J.L., Jones, P.G., Jarvis, A., 2005. Very high resolution interpolated

climate surfaces for global land areas. Int. J. Climatol. 25, 1965–1978. doi:10.1002/joc.1276 Hijmans, R.J., Phillips, S., Leathwick, J., Elith, J., 2016. dismo: Species distribution modeling. R package

version 1.0-15. Available at: https://CRAN.R-project.org/package=dismo Hodell, D.A., Curtis, J.H., Sierro, F.J., Raymo, M.E., 2001. Correlation of Late Miocene to Early Pliocene

sequences between the Mediterranean and North Atlantic. Paleoceanography 16, 164–178. doi:10.1029/1999PA000487

Holdridge, L. R., 1947. Determination of world plant formations from simple climatic data. Science 105(2727), 367-368.

Holt, B.G., Lessard, J.-P., Borregaard, M.K., Fritz, S.A., Araújo, M.B., Dimitrov, D., Fabre, P.-H., Graham, C.H., Graves, G.R., Jønsson, K.A., Nogués-Bravo, D., Wang, Z., Whittaker, R.J., Fjeldså, J., Rahbek, Q;9( MNOR;( J"( S40$%!( '.( T$88$,!U5( V''-!'-#$4D&,( /!-&'"5( '.( %D!( T'#80; Science. 339, 74–78.

54


doi:10.1126/science.1228282 Höglund, J., 2009. Evolutionary conservation genetics. Oxford University Press, Oxford. Jacobs, L. L., 1978. Fossil rodents (Rhizomyidae and Muridae) from Neogene Siwalk deposits, Pakistan.

Mus. North. Ariz. Bull. 52, 1–103. Jacobs, L. L., L. J. Flynn, Downs, W.R., 1989. Neogene Rodentia of Southern Asia. in: Black, C.C., Dawson,

M.R. (Eds.), Papers on fossil rodents in honor of Albert Elmer Wood. Natural History Museum, Los Angeles. pp. 157–177

Jacobs, L.L., Flynn, L.J., 2005. Of mice… again: the Siwalik rodent record, murine distribution, and

molecular clocks. Interpreting the past: essays on human, primate, and mammal evolution in honor of David Pilbeam, 63, p.80.

Jacquet, F., Denys, C., Verheyen, E., Bryja, J., Hutterer, R., Peterhans, J.C.K., Stanley, W.T., Goodman, S.M., Couloux, A., Colyn, M., Nicolas, V., 2015. Phylogeography and evolutionary history of the Crocidura olivieri complex (Mammalia, Soricomorpha): from a forest origin to broad ecological expansion across Africa. BMC Evol. Biol. 15, 1–15. doi:10.1186/s12862-015-0344-y

Jaeger, F., 1945. Zur Gliederung und Benenming des tropischen Graslandgurtels. Verh. Naturf. Ges. Basel 56: 509-520

Jiménez-Valverde, A., Lobo, J.M., 2007. Threshold criteria for conversion of probability of species presence to either-or presence-absence. Acta Oecologica 31, 361–369. doi:10.1016/j.actao.2007.02.001

Jones, G.R., 2014. STACEY: species delimitation and phylogeny estimation under the multispecies coalescent. BioRxiv, 010199.

Kai, W., Nomura, K., Fujiwara, A., Nakamura, Y., Yasuike, M., Ojima, N., Masaoka, T., Ozaki, A., Kazeto, Y., Gen, K., Nagao, J., Tanaka, H., Kobayashi, T., Ototake, M., 2014. A ddRAD-based genetic map and its integration with the genome assembly of Japanese eel (Anguilla japonica) provides insights into genome evolution after the teleost-specific genome duplication. BMC Genomics 15, 233. doi:10.1186/1471-2164-15-233

Kapli, P., Lutteropp, S., Zhang, J., Kobert, K., Pavlidis, P., Stamatakis, A., Flouri, T., 2017. Multi-rate Poisson tree processes for single-locus species delimitation under maximum likelihood and Markov chain Monte Carlo. Bioinformatics 33, 1630–1638. doi:10.1093/bioinformatics/btx025

Kassahun, A., Sadlova, J., Dvorak, V., Kostalova, T., Rohousova, I., Frynta, D., Aghova, T., Yasur-Landau, D., Lemma, W., Hailu, A., Baneth, G., Warburg, A., Volf, P., Votypka, J., 2015. Detection of Leishmania donovani and L. tropica in Ethiopian wild rodents. Acta Trop. 1–6. doi:10.1016/j.actatropica.2015.02.006

Kebede, F., Rosenbom, S., Khalatbari, L., Moehlman, P.D., Beja-Pereira, A., Bekele, A., 2016. Genetic diversity of the Ethiopian Grevy’s zebra (Equus grevyi) populations that includes a unique population of the Alledeghi Plain. Mitochondrial DNA Part A, 27(1), pp.397-400.

Kimura, Y., Hawkins, M.T., McDonough, M.M., Jacobs, L.L., Flynn, L.J., 2015. Corrected placement of Mus-Rattus fossil calibration forces precision in the molecular tree of rodents. Scientific reports, 5, p.14444.

Kimura, Y., Jacobs, L.L., Cerling, T.E., Uno, K.T., Ferguson, K.M., Flynn, L.J., Patnaik, R., 2013. Fossil Mice and Rats Show Isotopic Evidence of Niche Partitioning and Change in Dental Ecomorphology Related to Dietary Shift in Late Miocene of Pakistan. PLoS One 8, 1–10. doi:10.1371/journal.pone.0069308

Kingdon, J., 1990. Island Africa: the evolution of Africa’s rare animals and plants. Collins, London. Kingdon, J., 2015. The Kingdon field guide to African mammals. Bloomsbury Publishing Plc London. Klempa, B., Fichet-Calvet, E., Lecompte, E., Auste, B., Aniskin, V., Meisel, H., Denys, C., Koivogui, L., ter

Meulen, J., Krüger, D.H., 2006. Hantavirus in African wood mouse, Guinea. Emerging Infectious Diseases, 12(5), p.838.

Köppen, W., 1900. Versuch einer Klassifikation der Klimate, vorzugsweise nach ihren Beziehungen zur Pflanzenwelt. Geographische Zeitschrift, 6(11. H), pp.593-611.

Köppen, W., 1884. Die Wärmezonen der Erde, nach der Dauer der heissen, gemässigten und kalten Zeit und nach der Wirkung der Wärme auf die organische Welt betrachtet. Meteorologische Zeitschrift, 1(0), pp.215-226.

Krijgsman, W., Hilgen, F.J., Raffi, I., Sierro, F.J., Wilson, D.S., 1999. Chronology, causes and progression of the Messinian salinity crisis. Nature, 400(6745), p.652.

Landis, M.J., Matzke, N.J., Moore, B.R., Huelsenbeck, J.P., 2013. Bayesian analysis of biogeography when the number of areas is large. Syst. Biol. 62, 789–804. doi:10.1093/sysbio/syt040

Lanfear, R., Calcott, B., Kainer, D., Mayer, C., Stamatakis, A., 2014. Selecting optimal partitioning schemes

55


for phylogenomic datasets. BMC Evol. Biol. 14, 82. doi:10.1186/1471-2148-14-82 Lauer, W., 1952. Humide and aride Jahreszeiten in Afrika und Sudamerika und ihre Beziehungen zu den

Vegetationsgurteln. Banner Geogr. Abh. 9. Lavrenchenko, L.A., Kruskop, S. V, Bekele, A., Belay, G., Morozov, P.N., Ivlev, Y.F., Warshavsky, A.A.,

2010. Mammals of the Babille Elephant Sanctuary (Eastern Ethiopia). Russ. J. Theriol. 9, 47–60. Leaché, A.D., Fujita, M.K., Minin, V.N., Bouckaert, R.R., 2014. Species delimitation using genome-wide

SNP Data. Syst. Biol. 63, 534–542. doi:10.1093/sysbio/syu018 Lecompte, E., Aplin, K., Denys, C., Catzeflis, F., Chades, M., Chevret, P., 2008. Phylogeny and

biogeography of African Murinae based on mitochondrial and nuclear gene sequences, with a new tribal classification of the subfamily. BMC Evolutionary Biology, 8(1), p.199.

Lecompte, E., ter Meulen, J., Emonet, S., Daffis, S., Charrel, R.N., 2007. Genetic identification of Kodoko virus, a novel arenavirus of the African pigmy mouse (Mus Nannomys minutoides) in West Africa. Virology 364, 178–183. doi:10.1016/j.virol.2007.02.008

Lemmon, A.R., Emme, S.A., Lemmon, E.M., 2012. Anchored hybrid enrichment for massively high-throughput phylogenomics. Syst. Biol. 61, 727–744. doi:10.1093/sysbio/sys049

Linder, H.P., 2005. Evolution of diversity: the Cape flora. Trends in Plant Science, 10(11), 536–541. Linder, H.P., de Klerk, H.M., Born, J., Burgess, N.D., Fjeldså, J., Rahbek, C., 2012. The partitioning of

Africa: statistically defined biogeographical regions in sub-Saharan Africa. J. Biogeogr. 39, 1189–1205. doi:10.1111/j.1365-2699.2012.02728.x

Lorenzen, E.D., Heller, R., Siegismund, H.R., 2012. Comparative phylogeography of African savannah ungulates. Mol. Ecol. 21, 3656–3670. doi:10.1111/j.1365-294X.2012.05650.x

Lungu, A. N., 1981. Hipparion Fauna of Middle Sarmatian of Moldavia (Insectivora, Lagomorpha and Rodentia). State Tiraspol Institut Pedagogiki Publ. 1557, 1-137.

Macholan, M., Vyskocilova, M.M., Bejcek, V., Stastny, K., 2012. Mitochondrial DNA sequence variation and evolution of Old World house mice (Mus musculus). Folia Zool. 61, 284–307.

Manel, S., Williams, H.C., Ormerod, S.J., 2001. Evaluating presence absence models in ecology; the need to count for prevalence. J. Appied Ecol. 38, 921–931. doi:10.1046/j.1365-2664.2001.00647.x

Manthi, F.K., 2007. A preliminary review of the rodent fauna from Lemudong’o, southwestern Kenya, and its implication to the late Miocene paleoenvironments. Kirtlandia 56, 92–105.

Maslin, M.A., Trauth, M.H., 2009. Plio-Pleistocene East African Pulsed Climate Variability and Its Influence on Early Human Evolution, in: Leakey, R.E., Grine, F.E., Fleagle, J.G., Sponheimer, M., Lee-Thorp, J. a., Bobe, R., Leakey, M.G., Reed, K.E., Russak, S.M., Maslin, M. a., Trauth, M.H., Roche, H., Blumenschine, R.J., Shea, J.J., Ungar, P.S., Scott, R.S., Dean, M.C., Smith, B.H., Larson, S.G., Smith, E.J., Heesy, C.P., Smith, H.F., Lordkipanidze, D., Rightmire, G.P., Kimbel, W.H., Wood, B. a., Tobias, P. V., Jungers, W.L., Lieberman, D.E., Bramble, D.M., Raichlen, D.A. (Eds.), The First Humans: Origin and Early Evolution of the Genus Homo. Springer, pp. 151–158. doi:10.1007/978-1-4020-9980-9

Maslin, M. A., Brierley, C.M., Milner, A.M., Shultz, S., Trauth, M.H., Wilson, K.E., 2014. East african climate pulses and early human evolution. Quat. Sci. Rev. 101, 1–17. doi:10.1016/j.quascirev.2014.06.012

Maslin, M. A, Christensen, B., 2007. Tectonics, orbital forcing, global climate change, and human evolution in Africa: introduction to the African paleoclimate special volume. J. Hum. Evol. 53, 443–64. doi:10.1016/j.jhevol.2007.06.005

Mason, V.C., Li, G., Helgen, K.M., Murphy, W.J., 2011. Efficient cross-species capture hybridization and next-generation sequencing of mitochondrial genomes from noninvasively sampled museum specimens. Genome Res. 21, 1695–1704. doi:10.1101/gr.120196.111

Matzke, N.J., 2013. Probabilistic historical biogeography: new models for founder-event speciation, imperfect detection, and fossils allow improved accurancy and model-testing. Berkeley Plan. J. 26, 217–220. doi:10.5811/westjem.2011.5.6700

Matzke, N.J., 2014. Model selection in historical biogeography reveals that founder-event speciation is a crucial process in island clades. Syst. Biol. 63, 951–970. doi:10.1093/sysbio/syu056

Mayaux, P., Bartholomé, E., Fritz, S., Belward, A., 2004. A new land-cover map of Africa for the year 2000. Journal of Biogeography, 31(6), pp.861-877.

!"#$%&'()&' *+,-!&'.)&'/012!&'3)&'4#56*7+#68&'9., Russo, I.R., Verheyen, E., Šumbera, R., Phylogeography of a widespread sub-Saharan murid rodent Aethomys chrysophilus: the role of geographic barriers and paleoclimate in the Zambezian bioregion. Mammalia.

McCormack, J.E., Tsai, W.L.E., Faircloth, B.C., 2016. Sequence capture of ultraconserved elements from

56


bird museum specimens. Mol. Ecol. Resour. 16, 1189–1203. doi:10.1111/1755-0998.12466 McDonough, M.M., Šumbera, R., Mazoch, V., Ferguson, A.W., Phillips, C.D., Bryja, J., 2015. Multilocus

phylogeography of a widespread savanna-woodland-adapted rodent reveals the influence of Pleistocene geomorphology and climate change in Africa’s Zambezi region. Mol. Ecol. 24, 5248–5266. doi:10.1111/mec.13374

Meerburg, B.G., Singleton, G.R., Kijlstra, A., 2009. Rodent-borne diseases and their risks for public health. Critical reviews in microbiology, 35(3), pp.221-270.

Meheretu, Y., Cizkova, D., Tesikova, J., Welegerima, K., Tomas, Z., Kidane, D., Girmay, K., Schmidt-Chanasit, J., Bryja, J., Guenther, S., Bryjova, A., Leirs, H., de Bellocq, J.G., 2012. High Diversity of RNA Viruses in Rodents, Ethiopia. Emerg. Infect. Dis. 18, 2047–2050. doi:10.3201/eid1812.120596

Meheretu, Y., Šumbera, R., Bryja, J., 2015. Enigmatic Ethiopian endemic rodent Muriculus imberbis (Rüppell 1842) represents a separate lineage within genus Mus. Mammalia, 79(1), pp.15-23.

Mein, P., Martín-Suárez, E., Agustí, J., 1993. Progonomys Schaub, 1938 and Huerzelerimys gen. nov.(Rodentia); their evolution in Western Europe. Scripta Geologica, 103, pp.41-64.

Mikula, O., Šumbera, R., Aghová, T., Mbau, J.S., Bryja, J., 2016. Evolutionary history and species diversity of African Pouched Mice of the genus Saccostomus (Rodentia: Nesomyidae). Zoologica Scripta, 45(6), 595-617.

Miller, M.A., Pfeiffer, W., Schwartz, T., 2010. Creating the CIPRES Science Gateway for inference of large phylogenetic trees, in: 2010 Gateway Computing Environments Workshop, GCE 2010. pp. 1–8.

Mistry, J., 2000. Savannas. Prog. Phys. Geogr. 24, 601–608. Mistry, J., Beradi, A., 2014. World savannas: ecology and human use. Routledge. Mittermeier, R.A., Gil, P.R., Hoffman, M., Pilgrim, J., Brooks, T., Mittermeier, C.G., Lamoreux, J., da

Fonseca, G.A.B., 2005. Hotspots revisited: Earth’s biologically richest and most endangered terrestrial ecoregions. University of Chicago Press, Chicago.

Muscarella, R., Galante, P.J., Soley-Guardia, M., Boria, R.A., Kass, J.M., Uriarte, M., Anderson, R.P., 2014. ENMeval: An R package for conducting spatially independent evaluations and estimating optimal model complexity for Maxent ecological niche models. Methods Ecol. Evol. 5, 1198–1205. doi:10.1111/2041-210X.12261

Musser, G.G., Carleton M.D., 2005. Superfamily Muroidea, in: Wilson D.E., Reeder D.A.M. (Eds.), Mammal Species of the World. The Johns Hopkins University Press, Baltimore, pp. 894–1531.

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

Ndiaye, A., Chevret, P., Dobigny, G., Granjon, L., 2016. Evolutionary systematics and biogeography of the arid habitat-adapted rodent genus Gerbillus (Rodentia, Muridae): a mostly Plio-Pleistocene African history. Journal of Zoological Systematics and Evolutionary Research, 54(4), pp.299-317.

Near, T.J., Meylan, P.A.,Shaffer, H.B., 2004. Assessing concordance of fossil calibration points in molecular clock studies: an example using turtles. The American Naturalist, 165(2), pp.137-146.

Near, T.J., Sanderson, M.J., 2004. Assessing the quality of molecular divergence time estimates by fossil calibrations and fossil-based model selection. Philos. Trans. R. Soc. Lond. B. Biol. Sci. 359, 1477–1483. doi:10.1098/rstb.2004.1523

Nicolas, V., Granjon, L., Duplantier, J.M., Cruaud, C., Dobigny, G., 2009. Phylogeography of spiny mice (genus Acomys, Rodentia: Muridae) from the south-western margin of the sahara with taxonomic implications. Biol. J. Linn. Soc. 98, 29–46. doi:10.1111/j.1095-8312.2009.01273.x

Nylander, J.A.A., Ronquist, F., Huelsenbeck, J.P., Nieves-Aldrey, J.L., 2004. Bayesian Phylogenetic Analysis of Combined Data. Syst. Biol. 53, 47–67. doi:10.1080/10635150490264699

Olson, D.M., Dinerstein, E., Wikramanayake, E.D., Burgess, N.D., Powell, G.V.N., Underwood, E.C., D’amico, J.A., Itoua, I., Strand, H.E., Morrison, J.C., Loucks, C.J., Allnutt, T.F., Ricketts, T.H., Kura, Y., Lamoreux, J.F., Wettengel, W.W., Hedao, P., Kassem, K.R., 2001. Terrestrial Ecoregions of the World: A New Map of Life on Earth. Bioscience 51, 933. doi:10.1641/0006-3568(2001)051[0933:TEOTWA]2.0.CO;2

Otto-Bliesner, B.L., 2006. Simulating Arctic Climate Warmth and Icefield Retreat in the Last Interglaciation. Science (80-. ). 311, 1751–1753. doi:10.1126/science.1120808

Parham, J.F., Donoghue, P.C.J., Bell, C.J., Calway, T.D., Head, J.J., Holroyd, P. A., Inoue, J.G., Irmis, R.B., Joyce, W.G., Ksepka, D.T., Patané, J.S.L., Smith, N.D., Tarver, J.E., Van Tuinen, M., Yang, Z., Angielczyk, K.D., Greenwood, J.M., Hipsley, C.A., Jacobs, L., Makovicky, P.J., Müller, J., Smith, K.T., Theodor, J.M., Warnock, R.C.M., Benton, M.J., 2012. Best practices for justifying fossil calibrations. Syst. Biol. 61, 346–359. doi:10.1093/sysbio/syr107

57


Phillips, S.J., Anderson, R.P., Schapire, R.E., 2006. Maximum entropy modeling of species geographic distributions. Ecological modelling, 190(3-4), pp.231-259.

Phillips, S.J., Dudík, M., Schapire, R.E., 2004. A maximum entropy approach to species distribution modeling. In Proceedings of the twenty-first international conference on Machine learning (p. 83). ACM.

Pickford, M., Wanas, H., Soliman, H., 2006. Indications for a humid climate in the western desert of egypt 11–10 Myr ago: evidence from Galagidae (Primates, Mammalia). Comptes Rendus Palevol 5: 935–943

Piry, S., Guivier, E., Realini, A., Martin, J.F., 2012. |SE|S|AM|E| Barcode: NGS-oriented software for amplicon characterization–application to species and environmental barcoding. Molecular Ecology Resources, 12(6), pp.1151-1157.

Pons, J., Barraclough, T.G., Gomez-Zurita, J., Cardoso, A., Duran, D.P., Hazell, S., Kamoun, S., Sumlin, W.D., Vogler, A.P., 2006. Sequence-based species delimitation for the DNA taxonomy of undescribed insects. Systematic biology, 55(4), pp.595-609.

Puillandre, N., Lambert, A., Brouillet, S., Achaz, G., 2012. ABGD, Automatic Barcode Gap Discovery for primary species delimitation. Molecular ecology, 21(8), pp.1864-1877.

Pyron, R.A., 2011. Divergence time estimation using fossils as terminal taxa and the origins of Lissamphibia. Syst. Biol. 60, 466–481.

Pyron, R.A., 2010. A likelihood method for assessing molecular divergence time estimates and the placement of fossil calibrations. Syst. Biol. 59, 185–194. doi:10.1093/sysbio/syp090

Randi, E., D’Huart, J.P., Lucchini, V., Aman, R., 2002. Evidence of two genetically deeply divergent species of warhthog, Phacochoerus africanus and P. aethiopicus (Artiodactyla: Suiformes) in Eastern Africa. Mamm. Biol. 67, 91–96.

Ree, R.H., Smith, S.A., 2008. Maximum likelihood inference of geographic range evolution by dispersal, local extinction, and cladogenesis. Syst. Biol. 57, 4–14. doi:10.1080/10635150701883881

Renaud, S., Michaux, J., Mein, P., Aguilar, J.P., Auffray, J.C., 1999. Patterns of size and shape differentiation during the evolutionary radiation of the European Miocene murine rodents. Lethaia 32, 61–71. doi:10.1111/j.1502-3931.1999.tb00581.x

Richards, P.W. 1973. Africa, the ‘odd man out’. Tropical forest ecosystems in Africa and South America: a comparative review (ed. by B.J. Meggers, E.S. Ayensu and W.D. Duckworth), pp. 21–26. Smithsonian Institution Press, Washington, DC.

Rögl, F., 1999. Mediterranean and Paratethys. Facts and hypotheses of an Oligocene to Miocene paleogeography (short overview). Geol. Carpathica 50, 339–349.

Ronquist, F., 1997. Dispersal-Vicariance Analysis: A New Approach to the Quantification of Historical Biogeography. Syst. Biol. 46, 195. doi:10.2307/2413643

Ronquist, F., Teslenko, M., van der Mark, P., Ayres, D.L., Darling, A., Höhna, S., Larget, B., Liu, L., Suchard, M.A., Huelsenbeck, J.P., 2012a. MrBayes 3.2: efficient Bayesian phylogenetic inference and model choice across a large model space. Syst. Biol. 61, 539–42. doi:10.1093/sysbio/sys029

Ronquist, F., Klopfstein, S., Vilhelmsen, L., Schulmeister, S., Murray, D.L., Rasnitsyn, A.P., 2012b. A total-evidence approach to dating with fossils, applied to the early radiation of the hymenoptera. Syst. Biol. 61, 973–999. doi:10.1093/sysbio/sys058

Rowe, K.C., Reno, M.L., Richmond, D.M., Adkins, R.M., Steppan, S.J., 2008. Pliocene colonization and adaptive radiations in Australia and New Guinea (Sahul): Multilocus systematics of the old endemic rodents (Muroidea: Murinae). Mol. Phylogenet. Evol. 47, 84–101. doi:10.1016/j.ympev.2008.01.001

Sayre, R., Comer, P., Hak, J., Josse, C., Bow, J., Warner, H., Larwanou, M., Kelbessa, E., Bekele, T., Kehl, H., Amena, R., Andriamasimanana, R., Ba, T., Benson, L., Boucher, T., Brown, M., Cress, J., Dassering, O., Friesen, B., Gachathi, F., Houcine, S., Keita, M., Khamala, E., Marangu, D., Mokua, F., Morou, B., Mucina, L., Mugisha, S., Mwavu, E., Rutherford, M., Sanou, P., Syampungani, S., Tomor, B., Vall, A., Weghe, J. Vande, Wangui, E., Waruingi, L., 2013. A new map of standardized terrestrial

ecosystems of Africa. 1–24. doi:ISBN 978-0-89291-275-9 Schenk, J.J., Rowe, K.C., Steppan, S.J., 2013. Ecological opportunity and incumbency in the diversification

of repeated continental colonizations by muroid rodents. Syst. Biol. 62, 837–64. doi:10.1093/sysbio/syt050

Schuster, M., Duringer, P., Ghienne, J.-F., Vignaud, P., Mackaye, H.T., Likius, A., Brunet, M., 2006. The Age of the Sahara Desert. Science (80-. ). 312, 1138-1139-1139. doi:10.1126/science.312.5777.1138b

Sepulchre, P., Ramstein, G., Fluteau, F., Schuster, M., Tiercelin, J.-J., Brunet, M., 2006. Tectonic uplift and Eastern Africa aridification. Science 313, 1419–1423. doi:10.1126/science.1129158

Stadler, T., 2010. Sampling-through-time in birth-death trees. J. Theor. Biol. 267, 396–404.

58


Stamatakis, A., 2014. RAxML version 8: a tool for phylogenetic analysis and post-analysis of large phylogenies. Bioinformatics, 30(9), pp.1312-1313.

Stanhope, M.J., Czelusniak, J., Si, J.-S., Nickerson, J., Goodman, M., 1992. A molecular perspective on mammalian evolution from the gene encoding interphotoreceptor retinoid binding protein, with convincing evidence for bat monophyly. Mol. Phylogenet. Evol. 1, 148–160.

Steppan, S., Adkins, R., Anderson, J., 2004. Phylogeny and divergence-date estimates of rapid radiations in muroid rodents based on multiple nuclear genes. Syst. Biol. 53, 533–553. doi:10.1080/10635150490468701

Strecker, M.R., Blisniuk, P.M., Eisbacher, G.H., 1990. Rotation of extension direction in the central Kenya Rift. Geology 18, 299–302.

Teeling, E.C., Scally, M., Kao, D.J., Romagnoli, M.L., Springer, M.S., Stanhope, M.J., 2000. Molecular evidence regarding the origin of echolocation and flight in bats. Nature 403, 188–192.

Terryn, L., Wendelen, W., Leirs, H., Lenglet, G., Verheyen, E., 2007. African Rodentia. Web address: http://projects. biodiversity. be/africanrodentia.

Thulin, M., 1993. Flora of Somalia, Volume 1-Pteridophyta; Gymnospamae; angiospamae (Annonacae-Fabaceae). Flora of Somalia.

Thulin, M., 1994. Aspects of disjunct distributions and endemism in the arid parts of the Horn of Africa, particularly Somalia. Proceedings of the XIIIth plenary meeting of AETFAT, Zomba, Malawi, 2–11

April 1991 (ed. by J.H. Seyani and A.C. Chikuni), pp. 1105–1119. National Herbarium and Botanic Gardens of Malawi, Zomba.

Tilak, M.K., Justy, F., Debiais-Thibaud, M., Botero-Castro, F., Delsuc, F., Douzery, E.J.P., 2015. A cost-effective straightforward protocol for shotgun Illumina libraries designed to assemble complete mitogenomes from non-model species. Conserv. Genet. Resour. 7, 37–40. doi:10.1007/s12686-014-0338-x

Trauth, M.H., Maslin, M. a., Deino, A.L., Junginger, A., Lesoloyia, M., Odada, E.O., Olago, D.O., Olaka, L. a., Strecker, M.R., Tiedemann, R., 2010. Human evolution in a variable environment: the amplifier lakes of Eastern Africa. Quat. Sci. Rev. 29, 2981–2988. doi:10.1016/j.quascirev.2010.07.007

Troll, C., 1950. Savannentypen und das Problem der Primarsavannen. Proc. 7th Int. Bot. Congress, Stockholm, 670-674.

Varshavsky, A.A., Karimova, T.Y., Neronov, V.M., 2007. Specific features of the rodent fauna formation in Somali Peninsula. Zoologichesky Zhurnal, 86, 1498–1506.

Vaughan, T.A., Ryan, J.M., Czaplewski, N.J., 2011. Mammology. 5th ed. Jones and Bartlett Publishers, Sudbury.

Wallace, A.R., 1876. The Geographical Distribution of Animals with a Study of the Relations of Living and

Extinct Faunas, Harper & Brothers, London. Walter, H., 1973. Vegetation of the Earth in Relation to Climate and the Eco-physiological Conditions.

English Universities Press.. White, F., 1983. The vegetation of Africa: A descriptive memoir to accompany the

UNESCO/AETFAT/UNSO vegetation map of Africa (Natural Resources Research 20). UN

Educational, Scientific and Cultural Organization, Paris. White, F., 1987. Vegetation Map of Africa. The Vegetation of Africa: A Descriptive Memoir to Accompany

the Unesco/AETFAT/UNSO Vegetation map of Africa. J. Ecol. 75, 1214. doi:10.2307/2260340 Whittaker, R.H., 1975. Communities and ecosystems, 2nd edn, p. 385. Macmillan, New York Wickliffe, J.K., Hoffmann, F.G., Carroll, D.S., Dunina-Barkovskaya, Y.V., Bradley, R.D., Baker, R.J., 2003.

PCR and sequencing primers for intron 7 (Fgb-I7) of the fibrinogen, B beta polypeptide (Fgb) in mammals: a novel nuclear DNA marker. Occasional Papers, Museum of Texas Tech University, 219, p.i8.

Wynn, J.G., 2000. Paleosols, stable carbon isotopes, and paleoenvironmental interpretation of Kanapoi, Northern Kenya. Journal of Human Evolution, 39(4), pp.411-432.

Yang, Z., Rannala, B., 2014. Unguided species delimitation using DNA sequence data from multiple loci. Mol. Biol. Evol. 31, 3125–3135. doi:10.1093/molbev/msu279

59


Pan-African phylogeny of Mus

(subgenus Nannomys) reveals one of the

most successful mammal radiations in

Africa

Bryja, J., Mikula, O., umbera, R., Meheretu, Y., Aghová, T.,

Lavrenchenko, L. A., Mazoch, V., Oguge, N., Mbau, J.S.,

Welegerima, K., Amundala, N., Colyn, M., Leirs, H.,

Verheyen, E., 2014. BMC Evolutionary Biology, 14(1), 256.

https://doi.org/10.1186/s12862-014-0256-2

Study design: JB, R , EV

Collecting material: JB, R , VM, YM, LL, KW, NO, JM,

NA, MC, HL

Laboratory analysis: TA, YM

Data analysis: JB, OM

Writing: JB, OM

My contribution: 10%

Paper

I

Paper

I

60


61


RESEARCH ARTICLE Open Access

Pan-African phylogeny of Mus (subgenusNannomys) reveals one of the most successfulmammal radiations in AfricaJosef Bryja1,2,13*, Ondřej Mikula1,3, Radim Šumbera4, Yonas Meheretu5, Tatiana Aghová1,2, Leonid A Lavrenchenko6,

Vladimír Mazoch4, Nicholas Oguge7, Judith S Mbau8, Kiros Welegerima5, Nicaise Amundala9, Marc Colyn10,

Herwig Leirs11 and Erik Verheyen11,12

Abstract

Background: Rodents of the genus Mus represent one of the most valuable biological models for biomedical and

evolutionary research. Out of the four currently recognized subgenera, Nannomys (African pygmy mice, including

the smallest rodents in the world) comprises the only original African lineage. Species of this subgenus became

important models for the study of sex determination in mammals and they are also hosts of potentially dangerous

pathogens. Nannomys ancestors colonized Africa from Asia at the end of Miocene and Eastern Africa should be

considered as the place of their first radiation. In sharp contrast with this fact and despite the biological importance

of Nannomys, the specimens from Eastern Africa were obviously under-represented in previous studies and the

phylogenetic and distributional patterns were thus incomplete.

Results: We performed comprehensive genetic analysis of 657 individuals of Nannomys collected at approximately

300 localities across the whole sub-Saharan Africa. Phylogenetic reconstructions based on mitochondrial (CYTB) and

nuclear (IRBP) genes identified five species groups and three monotypic ancestral lineages. We provide evidence for

important cryptic diversity and we defined and mapped the distribution of 27 molecular operational taxonomic units

(MOTUs) that may correspond to presumable species. Biogeographical reconstructions based on data spanning all of

Africa modified the previous evolutionary scenarios. First divergences occurred in Eastern African mountains soon after

the colonization of the continent and the remnants of these old divergences still occur there, represented by

long basal branches of M. (previously Muriculus) imberbis and two undescribed species from Ethiopia and Malawi.

The radiation in drier lowland habitats associated with the decrease of body size is much younger, occurred mainly in

a single lineage (called the minutoides group, and especially within the species M. minutoides), and was probably linked

to aridification and climatic fluctuations in middle Pliocene/Pleistocene.

Conclusions: We discovered very high cryptic diversity in African pygmy mice making the genus Mus one of

the richest genera of African mammals. Our taxon sampling allowed reliable phylogenetic and biogeographic

reconstructions that (together with detailed distributional data of individual MOTUs) provide a solid basis for further

evolutionary, ecological and epidemiological studies of this important group of rodents.

Keywords: Biogeography, Tropical Africa, Molecular phylogeny, Pygmy mice, Plio-Pleistocene climatic fluctuations,

Divergence timing, Muridae (Murinae), Mus minutoides, Phylogeography, DNA barcoding

* Correspondence: [email protected] of Vertebrate Biology, Academy of Sciences of the Czech Republic,

Brno, Czech Republic2Department of Botany and Zoology, Faculty of Science, Masaryk University,

Brno, Czech Republic

Full list of author information is available at the end of the article

© 2014 Bryja et al.; licensee BioMed Central. This is an Open Access article distributed under the terms of the CreativeCommons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, andreproduction in any medium, provided the original work is properly credited. The Creative Commons Public DomainDedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article,unless otherwise stated.

Bryja et al. BMC Evolutionary Biology (2014) 14:256

DOI 10.1186/s12862-014-0256-2

RESEARCH ARTICLE Open Access

Pan-African phylogeny of Mus (subgenusNannomys) reveals one of the most successfulmammal radiations in AfricaJosef Bryja1,2,13*, Ondřej Mikula1,3, Radim Šumbera4, Yonas Meheretu5, Tatiana Aghová1,2, Leonid A Lavrenchenko6,

Vladimír Mazoch4, Nicholas Oguge7, Judith S Mbau8, Kiros Welegerima5, Nicaise Amundala9, Marc Colyn10,

Herwig Leirs11 and Erik Verheyen11,12

Abstract

Background: Rodents of the genus Mus represent one of the most valuable biological models for biomedical and

evolutionary research. Out of the four currently recognized subgenera, Nannomys (African pygmy mice, including

the smallest rodents in the world) comprises the only original African lineage. Species of this subgenus became

important models for the study of sex determination in mammals and they are also hosts of potentially dangerous

pathogens. Nannomys ancestors colonized Africa from Asia at the end of Miocene and Eastern Africa should be

considered as the place of their first radiation. In sharp contrast with this fact and despite the biological importance

of Nannomys, the specimens from Eastern Africa were obviously under-represented in previous studies and the

phylogenetic and distributional patterns were thus incomplete.

Results: We performed comprehensive genetic analysis of 657 individuals of Nannomys collected at approximately

300 localities across the whole sub-Saharan Africa. Phylogenetic reconstructions based on mitochondrial (CYTB) and

nuclear (IRBP) genes identified five species groups and three monotypic ancestral lineages. We provide evidence for

important cryptic diversity and we defined and mapped the distribution of 27 molecular operational taxonomic units

(MOTUs) that may correspond to presumable species. Biogeographical reconstructions based on data spanning all of

Africa modified the previous evolutionary scenarios. First divergences occurred in Eastern African mountains soon after

the colonization of the continent and the remnants of these old divergences still occur there, represented by

long basal branches of M. (previously Muriculus) imberbis and two undescribed species from Ethiopia and Malawi.

The radiation in drier lowland habitats associated with the decrease of body size is much younger, occurred mainly in

a single lineage (called the minutoides group, and especially within the species M. minutoides), and was probably linked

to aridification and climatic fluctuations in middle Pliocene/Pleistocene.

Conclusions: We discovered very high cryptic diversity in African pygmy mice making the genus Mus one of

the richest genera of African mammals. Our taxon sampling allowed reliable phylogenetic and biogeographic

reconstructions that (together with detailed distributional data of individual MOTUs) provide a solid basis for further

evolutionary, ecological and epidemiological studies of this important group of rodents.

Keywords: Biogeography, Tropical Africa, Molecular phylogeny, Pygmy mice, Plio-Pleistocene climatic fluctuations,

Divergence timing, Muridae (Murinae), Mus minutoides, Phylogeography, DNA barcoding

* Correspondence: [email protected] of Vertebrate Biology, Academy of Sciences of the Czech Republic,

Brno, Czech Republic2Department of Botany and Zoology, Faculty of Science, Masaryk University,

Brno, Czech Republic

Full list of author information is available at the end of the article

© 2014 Bryja et al.; licensee BioMed Central. This is an Open Access article distributed under the terms of the CreativeCommons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, andreproduction in any medium, provided the original work is properly credited. The Creative Commons Public DomainDedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article,unless otherwise stated.

Bryja et al. BMC Evolutionary Biology (2014) 14:256

DOI 10.1186/s12862-014-0256-2

62


Background

One of the main challenges of current nature conserva-

tion is the accelerating loss of biodiversity. Even if this

problem is generally recognized, there are several diffi-

culties in quantifying the loss of biodiversity at the level

of species. For example, there is a lack of traditional

taxonomic specialists for particular groups of organisms

and the real amount of biodiversity is therefore unknown

[1]. This is especially true for some tropical areas, where

the overall biodiversity level is the highest and its loss

is the most intensive. Another problem for practical

biodiversity conservation is the delimitation of species

(e.g. [2] vs. [3]). Traditional concepts of typological or

biological species are not universally applicable and

with accumulating knowledge in evolutionary biology it

is increasingly difficult to define generally what a species

is. Genetic approaches, like DNA barcoding, are now

routinely used to overcome some of these problems.

They provide a cheap and easily applicable approach for

discovering the taxa worth future taxonomical research

and areas with high phylogenetic diversity with special

conservation concern (e.g. [4]). For example, 175 new

extant taxa of mammals were described from African

mainland, Madagascar and all surrounding islands between

1988–2008 [5], and in the majority, the first consideration

for taxonomic delimitation was motivated by the use of

genetic data.

Rodents of the genus Mus represent one of the most

valuable biological models for biomedical and evolution-

ary research [6]. Out of the four currently recognized

subgenera, i.e. Mus, Coelomys, Pyromys and Nannomys,

the latter comprises the African pygmy mice [7]. These are

small rodents (4–12 g in most taxa, but see [8]), endemic

to the sub-Saharan Africa. The phylogenetic relationships,

species diversity, ecology and chromosomal evolution of

Nannomys were recently reviewed [9]. They represent the

most diverse lineage of the genus, with currently about 18

species recognized [9,10], comprising almost half of the

described Mus species [10]. While predominantly savannah

dwellers [11], several species have also been trapped in

forest, agricultural fields and rural areas [12-14].

Mainly due to their extensive chromosomal diversity

coupled with highly conserved morphology, African pygmy

mice have attracted the attention of evolutionary scientists

[9,11,15-17]. They exhibit chromosomal features that are

rarely recorded in other taxa, e.g. the greatest diversity

of sex-autosome translocations reported so far in any

mammalian lineage (e.g. [18]). Thus Nannomys became

an important biological model for the study of processes

of chromosomal speciation and mechanisms of sex deter-

mination in mammals [19]. Recent studies have also

shown that African pygmy mice are important hosts of

arena viruses [20-23], making them a target group for

epidemiological surveys.

Increasing numbers of molecular genetic data provide

evidence for high cryptic diversity in Nannomys and it is

highly probable that further integrative taxonomy research

will reveal new undescribed species [11,14]. Furthermore,

the inclusion of poorly known African Mus-related rodents

in molecular phylogenetic datasets may provide surprising

results changing the current view on the evolutionary sce-

narios of Nannomys. For example, the Ethiopian endemic

genus Muriculus was recently recognized to be an internal

lineage of Mus [8].

The genus Mus diverged in Asia approximately 6.7

to 7.8 Mya and shortly after this time the ancestor(s)

of Nannomys colonized Africa through the Arabian

Peninsula and Miocene land bridges [9]. The oldest

fossils of Mus in Africa are reported from Tugen Hills

(Kenya) about 4.5 Mya [24]. The highly heterogeneous

environment of Eastern Africa can thus be considered

as the place of first diversification of African Mus in

Early Pliocene, followed by a radiation caused by climatic

oscillations and habitat modification [9,11]. In this context

it is important to note that genetic data used so far for

molecular phylogenetic inference of the African pygmy

mice are strongly biased geographically in favour of mater-

ial collected from savannahs in the western and southern

part of the continent, while specimens from Eastern Africa

(including those from mountains and forests) are clearly

under-represented (Figure 1a).

More thorough geographical sampling is necessary

for obtaining the correct biogeographical scenario of

Nannomys evolution. Only a comprehensive and reliable

phylogenetic hypothesis can lead to meaningful inferences

on the evolution of sex-determination or virus-host

co-evolution. In this study, we provide the so far most

comprehensive geographic sampling of genetically charac-

terized African pygmy mice composed of 657 Nannomys

individuals from most parts of sub-Saharan Africa. First,

we use this pan-African dataset for the reconstruction

of phylogenetic relationships within Nannomys lineage.

Second, using the combination of species delimitation

methods, we aim to estimate the presumable species rich-

ness of Nannomys, highlighting groups and geographical

regions necessitating further taxonomical research. Finally,

the dating of divergences and biogeographical reconstruc-

tions allow us to modify previous scenarios that were

suggested to explain the Nannomys radiation in Africa.

Methods

Sampling

New genetic data were produced for 395 individuals of

subgenus Nannomys sampled in sub-Saharan Africa by

the authors and their collaborators. All fieldwork com-

plied with legal regulations in particular African countries

and sampling was in accordance with local legislation (see

more details on wildlife authorities that permitted the

Bryja et al. BMC Evolutionary Biology (2014) 14:256 Page 2 of 20

63


research in Acknowledgements). Each individual was iden-

tified to the genus by the external features and the tissue

sample (tail, toe, spleen, etc.) was stored in 96% ethanol

until DNA extraction. GPS coordinates of each locality

were recorded. New data were supplemented with 262

published records of genotyped and georeferenced Nan-

nomys, i.e. partial or complete sequences of mitochondrial

gene for cytochrome b (CYTB) were downloaded from

GenBank. Geographical coordinates of published data

were either retracted from original publications (if avail-

able) or approximately estimated from Google maps. In

total, the analysed dataset includes genetic information

of 657 individuals from approximately 300 localities in

30 African countries (Figure 1a). For more details on

analysed individuals see Additional file 1.

DNA sequencing of CYTB and IRBP

DNA was extracted using the DNeasy Blood & Tissue

Kit (Qiagen). The complete CYTB gene was amplified by

polymerase chain reaction (PCR) using primers L14723

and H15915 [25]. PCR mix contained 3 μl of genomic

DNA, 0.5 units of Taq polymerase (Fermentas), final

concentrations of 3 mM MgCl2, 1 x Taq buffer with

(NH4)2SO4 (Fermentas), 0.2 mM of each dNTPs, 0.2 μM

of each primers and ddH2O to final volume of 30 μl.

The thermal profile of the PCR started with an initial

denaturation at 94°C for 3 min, followed by 35 cycles

composed of 30 s of denaturation at 94°C, 30 s of

annealing at 50°C, and 3 min of extension at 72°C and

PCR was finished by a final extension at 72°C for 10 min.

The part of nuclear gene encoding the Interphotoreceptor

Binding Protein (IRBP) was amplified in selected individ-

uals (from each main clade identified previously by CYTB

marker) by the primers IRBP1531 and IRBP217 [26]. PCR

conditions were the same as above, except the final con-

centration of MgCl2 (2 mM). The thermal profile of the

PCR started with an initial denaturation of one step at

94°C for 3 min, followed by 30 cycles of 60 s at 94°C,

60 s at 55°C, 2 min at 72°C and finished by a final exten-

sion at 72°C for 10 min. The PCR products were purified

with Calf Intestine Alkaline Phosphatase (ThermoScienti-

fic) and Exonuclease I (ThermoScientific) and sequenced

along both strands commercially in Macrogen Europe

using the same primers as for the PCR. Both genetic

(b)(a)

(e)(d)

(c)

(f)

Mus minutoides

(MOTU 27)

minutoides group

(excl. M. minutoides) baoulei and

sorella groups

setulosus group

ancient lineages and

triton group

Min

Chin

?

?

?? ?

??

?

Figure 1 Distribution of genotyped specimens and individual MOTUs. (a) Distribution of analyzed material of Nannomys in Africa. Blue

dots indicate the geographical position of published sequences (downloaded from GenBank), red dots show the localities of newly sequenced

individuals. Geographical distribution of (b) MOTUs of the triton group (circles) and three ancient monotypic lineages (rhombuses); (c) MOTUs

of the setulosus group; (d) MOTUs of the baoulei (rhombuses) and the sorella (circles) groups; (e) MOTUs of the minutoides group (except M.

minutoides); (f) phylogeographic structure of MOTU 27, i.e. M. minutoides. In Figures 1e identical symbol shapes represent monophyletic groups.

In Figure 1f the clade abbreviations correspond to Figure 4. Question marks indicate doubtful records based on genotyping of old museum

material (see [64]). For more information on analysed material see Additional file 1.


64


markers have been previously used to successfully resolve

systematic relationships in a wide range of related murid

rodents (e.g. [7,25].

Genetic data from fresh material were complemented

by museum samples (mostly dry skins) from the Royal

Museum for Central Africa (Tervuren, Belgium), Muséum

National d'Histoire Naturelle (Paris, France), American

Museum of Natural History (New York, USA) and

Hungarian Natural History Museum (Budapest, Hungary)

(for more details see Additional file 1). Museum samples

comprised only minor part of analysed material and we

used especially those from geographical areas difficult to

be accessed today (e.g. Central African Republic, eastern

Democratic Republic of Congo) and the type material of

Mus bufo. All museum samples were handled in a special-

ized laboratory of Institute of Vertebrate Biology ASCR in

Studenec, designed for work with rare DNA to prevent

contamination by samples with high quantity of DNA or

PCR products. DNA was extracted using the JETQUICK

Tissue DNA Spin Kit (Genomed). PCR amplification and

pyrosequencing on GS Junior were performed according

to mini-barcode protocol described by [27]. The main

advantage of this approach in analysis of museum samples

is that it allows for separating individual sequences in

samples contaminated by distantly related organisms (e.g.

contamination by human DNA), which is not possible

through the Sanger sequencing.

Phylogenetic reconstructions

Sequences of CYTB and IRBP were edited and aligned

in SeqScape v2.5 (Applied Biosystems), producing a final

alignment of 1140 and 1276 bp, respectively. The Find-

model web application (http://www.hiv.lanl.gov/content/

sequence/findmodel/findmodel.html) was used to iden-

tify the most appropriate substitution model for each

gene. The Akaike information criterion (AIC), compared

among 12 substitution models, revealed that the model

best fitting the ingroup data was the General time

reversible model with a gamma-distributed rate variation

across sites (GTR +G) for both CYTB and IRBP. As out-

groups, we used sequences of four species from other

subgenera of the genus Mus, i.e. M. platythrix (CYTB

GenBank Acc. code AJ698880, IRBP GenBank Acc. code

AJ698895), M. pahari (AY057814, AJ698893), M. caroli

(AB033698, AJ698885) and M. musculus (V00711,

AF126968); two sister lineages of the genus Mus within

subfamily Murinae, i.e. Apodemus flavicollis (AB032853,

AB032860) and Ratus norvegicus (V01556, AJ429134);

and one species from the subfamily Acomyinae, Acomys

cahirinus (AJ233953, AJ698898) see also [7,9,11].

Phylogenetic relationships within Nannomys were in-

ferred by maximum likelihood (ML) and Bayesian (BI)

approaches. ML analysis was performed using RAxML

8.0 [28]. The GTR + G model (option -m GTRGAMMA)

was selected for the six partitions, i.e. 1140 bp of CYTB,

1276 bp of IRBP, and both genes were partitioned also by

the position of nucleotides in the codons (option -q). The

robustness of the nodes was evaluated by the default boot-

strap procedure with 1,000 replications (option -# 1000).

Bayesian analysis of evolutionary relationships was per-

formed by Markov chain Monte Carlo (MCMC) method

in MrBayes v. 3.2.1 [29]. Three heated and one cold chain

were employed in all analyses, and runs were initiated

from random trees. Two independent runs were con-

ducted with 5 million generations per run; and trees and

parameters were sampled every 1,000 generations. Con-

vergence was checked using TRACER v1.5 [30]. For each

run, the first 10% of sampled trees were discarded as

burn-in. Bayesian posterior probabilities were used to

assess branch support of the Bayesian tree.

The most widespread Nannomys species (= MOTU, see

below) is M. minutoides. For this species we performed

more detailed analysis of intraspecific genetic variability.

We selected 131 sequences belonging to this clade and

trimmed the final alignment to the length of 741 bp.

Haplotypes were generated using DNaSP software [31]

and a median-joining network of haplotypes was produced

in the software Network 4.6.1.2 (downloaded on 10.2.2014

from http://www.fluxus-engineering.com/sharenet.htm).

Delimitation of MOTUs

We estimated the possible number of putative species

(called here molecular operational taxonomic units,

MOTUs, until the thorough taxonomic evidence will be

provided) of Nannomys in the sampled dataset by using

two types of divergence thresholds and the CYTB data-

set. The first was the time threshold estimated by the

Generalized Mixed Yule Coalescent (GMYC) model [32]

which describes single-locus branching pattern as a suc-

cession of speciation events replaced at a fixed threshold

time by a succession of intraspecific coalescent events.

The two stages are modelled by Yule process and neutral

coalescent, respectively, which allows finding maximum

likelihood estimate of the threshold time and evaluating

statistical support for the delimited species [33,34]. In

this framework reliably delimited species are those whose

basal internal split occurred well after the speciation-

coalescence threshold and which diverged from sister

species well before it. We therefore calculated two kinds

of support: (1) for each intra-specific basal split we calcu-

lated relative likelihood that it represents coalescence

rather than speciation event by summing up Akaike

weights of all threshold times older or equal to its age;

(2) for each inter-specific split we calculated relative

likelihood that it represents speciation as a sum of Akaike

weights of threshold times younger to it. The ultrametric

tree required by GMYC was produced by BEAST 1.8.0 [35]

with uncorrelated lognormal distribution of substitution


65


rates and lognormal priors for node ages mimicking poste-

riors from the divergence dating (see below). We used

the Yule prior assuming no intra-specific divergences

(alternative analyses with a coalescent prior assuming

no speciation events lead to almost identical results of

GMYC analyses; not shown). The topology was con-

strained to match the branching order of main lineages

observed in the maximum likelihood phylogeny. The

GMYC analysis was performed using the R package

‘splits’ (http://r-forge.r-project.org/R/?group_id=333).

The second threshold was based on sequence divergence,

taken as a proxy for the amount of genetic difference

among distinct gene pools. We therefore analyzed the

distribution of Kimura-2 parameter (K2P) corrected

genetic distances on CYTB among GMYC-delimited

species (calculated in Mega 5.05; [36]) and merged the

lineages with less than 7.3% genetic distance, i.e. the

mean value between sister species of rodents [37]. The

resulting groups were designated as molecular oper-

ational taxonomic units (MOTUs) and provisional

names were assigned to them. It is important to note

that the aim of our MOTUs delimitation approach is

not to change the current taxonomy, but to highlight

the taxa and geographical areas worthy of further taxo-

nomic study, including morphological, ecological and

more detailed genetic approaches.

Divergence dating

Time to the most recent common ancestors (TMRCA) of

clades identified by phylogenetic analyses was estimated

using a relaxed clock model with substitution rates drawn

from an uncorrelated lognormal distribution in BEAST

1.8.0 [35] and three fossil-based calibration points: origin

of extant Mus, origin of extant Apodemus and the Arvi-

canthis/Otomys lineage split. To avoid disproportionate

impact of Nannomys we fitted the evolutionary model to

63 concatenated CYTB and IRBP sequences representing

main lineages of Nannomys and correspondingly deep

divergences across the tribes Apodemini, Arvicanthini,

Malacomyini, Murini, Otomyini and Praomyini (sensu

[38]). The data set is reported in detail in the Additional

file 2.

Following [39] we used lognormal calibration densities

with zero means whose 5% and 95% quantiles were

specified by appropriately chosen standard deviations

and offsets and corresponded to the fossil derived mini-

mum and maximum ages. In particular the parameters

(standard deviation, offset, 5% and 95% quantile) were:

(1) 0.74, 7.00, 7.30 and 10.38 for Mus, based on the earli-

est fossil Mus and a member of Progonomys considered

belonging to Mus stem lineage [40]; (2) 0.54, 4.89, 5.30

and 7.30 for Apodemus corresponding to 95% confidence

interval of first appearance as reported by [39], although

we applied it to the basal split of extant species rather

than to the origin of their stem lineage; (3) 0.80, 5.81,

6.08 and 9.54 for Arvicanthis/Otomys split which was

derived from the earliest records of Otomys (ca. 5 Mya;

[41]) and arvicanthine genera Aethomys, Arvicanthis and

Lemniscomys (6.08–6.12 Mya; [42]) and the next relevant

sample where these and related genera (except for a

tentative Aethomys) are absent (9.50-10.50 Mya; [43]).

Based on the previous studies [38,39,44] we constrained

the topology to include a basal split between Arvi-

canthini+Otomyini and the rest of the species.

The MCMC simulations were run twice with 25 million

iterations, with genealogies and model parameters sam-

pled every 1000 iterations. Trees were linked, models and

clocks were unlinked for two markers. Convergence was

checked using TRACER v1.5 [30], both runs being com-

bined in LOGCOMBINER 1.7.1 [35] and the maximum

clade credibility tree calculated by TREEANNOTATOR

1.7.1 [35], following the removal of 10% burn-in.

Biogeographical analysis

Ancestral habitat types were inferred by the Bayesian

analysis of discrete traits [45]. It models discrete states

of a trait at the end of each branch as a result of a con-

tinuous time Markov chain with infinitesimal transition

rates determined by an overall transition rate, pair-wise

transition probabilities and a base frequency of the states.

Following the current implementation in BEAST 1.8.0 we

used strict clock time-irreversible model so the overall

transition rate was assumed uniform across the tree

and transition probabilities were allowed to differ in the

opposite directions. Using the distribution data, we

coded the 27 MOTUs as inhabiting either (i) tropical

forests in the Congo Basin, Central and Western Africa;

(ii) mountains in Eastern Africa (various habitats), or

(iii) savannah habitats in sub-Saharan Africa. Some

MOTUs can inhabit more habitat types (e.g. MOTU

27, M. minutoides). The analysis in BEAST does not

allow more variants of the tip trait, so we assigned the

trait (habitat) that is the most widespread in a particular

MOTU (e.g. savannah in M. minutoides). The topology

was fixed to match relationships between MOTUs on the

ML tree and branch lengths were time-calibrated as in the

ultrametric tree for GMYC.

Alternatively, we identified ancestral habitat types and

rough geographic ranges by using the maximum likeli-

hood approach implemented in the software Lagrange

[46,47]. The implemented model estimates geographic

range evolution using a phylogenetic tree with branch

lengths scaled to time, geographic (habitat) areas for all

tips, and an adjacency matrix of plausibly connected

areas. We used the same tree and distribution data as in

the BEAST analysis described above. We allowed the

connection between all three habitats with equal prob-

abilities of each transition. The maximum number of


66


ancestral ranges was set to two. The resulting recon-

structions returned all models within two likelihood

units of the best model, which we summarized for each

daughter branch and plotted in the form of pie-charts

along the tree in R [48].

Results

Overview of collected data

For the phylogenetic analysis we retained 179 CYTB

sequences at least 700 bp (133 new sequences and 46

sequences from GenBank) representing as complete a

geographical distribution of each clade as possible

(Additional file 1). The remaining 478 sequences (usually

shorter and/or from the same or close neighbouring

localities), including 16 sequences obtained by 454 pyrose-

quencing of old museum samples, were unambiguously

assigned to particular MOTU by neighbour-joining ana-

lysis in MEGA 5.05 (bootstrap values higher than 95%)

and these data were used for mapping the geographical

distribution of phylogenetic clades.

We also selected 1–2 individuals from each of the main

significantly supported CYTB clades (if the tissues were

available) and sequenced them at IRBP gene. The final

phylogenetic analyses included 42 sequences of IRBP

(32 new sequences and 10 sequences from GenBank;

see Additional file 1) from all main species groups

except the baoulei group (see below). ML analyses were

performed separately for both genes, and because the

topology of trees was very similar (although the reso-

lution of IRBP was much lower; Additional file 3), we

finally performed both ML and BI reconstructions only

using a concatenated CYTB and IRBP dataset produced

in SEAVIEW [49].

Phylogeny of African Nannomys

Phylogenetic trees based on the concatenated dataset

were well resolved and with very similar topology of 179

ingroup sequences in both ML and BI analyses (Figure 2).

Subgenus Nannomys (including “Muriculus” imberbis; see

[8]) was strongly supported. There are three long branches

representing three ancient mountainous species with

unresolved relationships to other groups (M. sp.

“Nyika” =MOTU 1, M. imberbis =MOTU 2, and M. sp.

“Harenna” =MOTU 3) and five well supported species

groups. We call them hereafter triton, setulosus, baoulei,

sorella, and minutoides groups, based on the previous use

of these names, representing the best known species

within particular clades. Each group contains several

distinct lineages that may represent separate species; the

most diversified is the minutoides group. The relation-

ships among species groups are not well resolved, but

in most topologies the triton group is non-significantly

clustered with three ancient species, while all other

species groups cluster together. Within the latter, the

setulosus group separates the first, and the baoulei

group is the sister of the sorella group (Figure 2).

Number of potential species and their distribution

The application of the GMYC model provided the de-

limitation of 49 maximum likelihood entities (hereafter

GMYC-species; 95% CI = 42–62 entities) based on the

ML estimate of speciation-coalescence threshold at 0.46

(0.27–0.86) Mya. Figure 3a depicts support for the “in-

traspecific” basal splits as coalescences as well as support

for “interspecific” splits as speciation events. In both

cases white circles indicate support < 0.95 and black cir-

cles > 0.95. Low “intraspecific” support suggests there

may be more species present, whereas low “interspecific”

support suggests the two sister clades may be in fact

conspecific populations. Where two neighbouring “inter-

specific” and “intraspecific” supports are low, the

speciation-coalescence transition is blurred.

K2P distances among the GMYC-species (3.16-20.77%)

were not overlapping with “intraspecific” distances (0.12-

2,38%) (Additional file 4). The detailed analysis of

geographical distribution of GMYC-species showed that

many sister groups among them are parapatric, i.e. most

probably representing the results of allopatric differenti-

ation and secondary contacts. For example, in the clade

corresponding to M. minutoides in previous studies (e.g.

[9]), the GMYC method delimited 12 GMYC-species with

prevailing parapatric distribution pattern and with “inter-

specific” K2P distances 3.27-6.96%. Using the threshold

value of 7.3%, we grouped these lineages and considered

them as phylogeographical differentiation within the single

species M. minutoides (see Figure 1f for the distribution of

phylogeographical lineages that roughly correspond to

“species” identified by GMYC method). Using this com-

bined approach (i.e. analysis of geographic distribution

of GMYC-species and threshold of K2P distances), we

reduced 49 GMYC-species to 27 highly supported molecu-

lar operational taxonomic units (MOTUs, Figure 3a), which

are further discussed below. Genetic distances among 27

MOTUs were always significantly higher and did not over-

lap with those within MOTUs (Additional file 4).

There were 17 MOTUs that exactly matched a single

GMYC-species, 11 of them represented by more than

one sequence. 7 MOTUs comprised two GMYC-species,

2 MOTUs were composed of three GMYC-species and a

single MOTU, MOTU 27 =M. minutoides, comprised

12 GMYC-species (Figure 3a). In 12 cases, however,

there was strong support for the presence of multiple

species within a single MOTU (marked by black circles

left of the GMYC threshold in Figure 3a).

Below we follow the nomenclature of [10] that recognizes

18 valid species. Possible names for newly recognized

MOTUs are discussed in the text.


67


supported by both ML (>75) and BI (>0.95) analyses

supported by only one analysis

MOTU 3 („Harenna“)MOTU 2 (imberbis)MOTU 1 („Nyika“)outgroups

MOTU 6 (cf. callewaerti)

MOTU 5 („Kikwit“)

MOTU 4 (triton)

MOTU 16 (sorella)MOTU 15 („Koi River“)

MOTU 17 (neavei)

MOTU 14 („Dakawa“)MOTU 13 (cf. baoulei„West“)

MOTU 12 (baoulei)

MOTU 11 (mahomet)

MOTU 10 (bufo)

MOTU 9 (cf. setulosus„West“)

MOTU 8 (setulosus)

MOTU 7 (cf. proconodon)

MOTU 27 (minutoides)

MOTU 26 (musculoides)

MOTU 22 (indutus)

MOTU 21 (cf. kasaicus)

MOTU 20 (mattheyi)

MOTU 19 (haussa)MOTU 18 („Zakouma“)

MOTU 25 (cf. tenellus)

MOTU 24 (cf. gerbillus)

MOTU 23 (cf. gratus)

min

uto

ides g

roup

baoulei

group

sorella

group

setulosus

group

triton

group

Figure 2 (See legend on next page.)


68


(1)Ancient mountain lineages (Figure 1b):

A tri-phyletic group with very restricted distribution

ranges. They are known from only a few individuals

captured in the highest East African mountains. They

were not included in previous phylogenetic studies of Nan-

nomys and on the phylogenetic tree they form very long

branches, in most topologies they are related to the triton

group, but not always with significant nodal support.

(MOTU 1) Mus sp. “Nyika”

It is a very distinct ancient lineage of Nannomys,

known from a single, relatively large individual (14 g),

captured in the high plateau of Nyika Mts. in Malawi

(cca 2100 m a.s.l.). Albeit partially broken, the cranium

of this specimen clearly shows features that are typical

for insectivorous rodents, namely proodont (forward ori-

ented) incisors and slender mandibles. This lineage is

sympatric with MOTU 17 (M. neavei) and even syntopic

with MOTU 6 (M. cf. callewaerti).

(MOTU 2) Mus imberbis Rüppell, 1842

It is an easily distinguished taxon, very large (sequenced

individual weighted 25 g) and with a black dorsal stripe. It

has been considered as a separate genus Muriculus, but

genetic analysis of a recently captured individual clearly

shows that it is an internal lineage of Mus [8]. It is an

endemic species of the high plateaux of Ethiopia, known

only from a few of individuals (reviewed in [8]).

(MOTU 3) Mus sp. “Harenna”

It is a large species (cca 16 g), very probably endemic

to the moist Harenna forest in the Bale Mts. in Ethiopia,

Figure 3 Reconstruction of divergence dates and ancestral distributions of MOTUs. (a) Phylogenetic relationships among 49 GMYC-species and

definition of 27 MOTUs. The vertical line indicates the threshold where the speciation processes are replaced by coalescence. Black circles indicate

strong support (>95%) for either speciation (left of the threshold) or intraspecific coalescence (right of the threshold). White circles indicate weak

support (<95%) for these processes. The dating of divergences within Nannomys was assessed by BEAST using the previously estimated divergence

times (see Additional file 2) as priors for calibration of relaxed molecular clock. (b) Reconstruction of ancestral distribution areas (blue – mountains

in Eastern Africa, green – tropical forests of central and western Africa; red – open savannah-like habitats surrounding forests and mountains

in sub-Saharan Africa. The different colours on pie charts indicate the probability of a particular state of the trait for each node. The analysis of

ancestral traits was performed in BEAST (see text for more details).

(See figure on previous page.)

Figure 2 Inferred phylogenetic relationships within Nannomys. Maximum likelihood phylogenetic tree of Nannomys is based on the

combined dataset of mitochondrial (CYTB) and nuclear (IRBP) genes. Black circles indicate the support by both ML (bootstrap values > 75%) and BI

(posterior probabilities > 0.95) analyses; grey circles indicate nodes supported by only one analysis. MOTUs were identified by the combination of

GMYC approach and distribution of genetic distances on CYTB. Only outgroups from the genus Mus are shown. GenBank accession numbers

correspond to CYTB sequences, for IRBP numbers see Additional file 1. Abbreviations of countries: BE: Benin, BF: Burkina Faso, BOT: Botswana, BUR:

Burundi, CAM: Cameroon, CAR: Central African Republic, CI: Côte d’Ivoire, CON: Congo, DRC: Democratic Republic of Congo, ETH: Ethiopia, GAB:

Gabon, GUI: Guinea, KE: Kenya, MAL: Mali, MOZ: Mozambique, MW: Malawi, NIG: Niger, RWA: Rwanda, SA: South Africa, SEN: Senegal, TOG: Togo,

TCH: Tchad, TZ: Tanzania, ZA: Zambia.


69


a region with very pronounced endemicity [50,51]. Based

on morphometry this taxon was previously reported as

M. triton [50] and in most topologies it is also the sister

taxon to the triton group. Genetically it is a very distinct

lineage (13.5-14.4% K2P distance to taxa of the triton

group) with a remarkably different karyotype than M.

triton [52]. Earlier studies have already suggested that

this taxon represents a valid species [51]. It can be sym-

patric with M. mahomet, but differs in habitat preferences;

M. sp. “Harenna” lives mostly in the forests, while M.

mahomet inhabits more open grassy habitats [[53]; L.

Lavrenchenko, pers. obs.].

(2)The triton group (Figure 1b):

It is the group of MOTUs of relatively large body size,

distributed mostly south of the equator (largely parapa-

tric with the setulosus group - see Figure 1b vs. 1c).

Genetic data suggest important cryptic variability (K2P

distance among three MOTUs = 8.80-11.05%). Only

nominotypical MOTU has a clear valid name, remaining

lineages require further taxonomic studies.

(MOTU 4) Mus triton (Thomas, 1909)

This species was described from Mt. Elgon in Kenya

and we provide the sequence from the type locality. It is

distributed in the Kenyan highlands and northern part

of Albertine rift. The same species probably occurs in

southern Sudan also (described as M. imatongensis) [54]),

but this should be confirmed by barcoding Sudanese

specimens.

(MOTU 5) Mus sp. “Kikwit”

This distinct genetic lineage within the triton group

was detected in two localities in south-western Demo-

cratic Republic of Congo (DRC). It may represent a new

species, but more material and analyses are necessary to

substantiate this claim. This MOTU supports important

biogeographical distinctiveness of the Kikwit region in

DRC (see also MOTU 21 from the minutoides group).

The type locality of Mus callewaerti (Thomas, 1925)

(Kananga, Kasaï occidental, DRC) is relatively near, so it

is possible that they are conspecific, but a comparison

with the type material will be necessary before a final

conclusion can be reached (see also MOTU 6).

(MOTU 6) Mus cf. callewaerti

This taxon forms a well-supported separate lineage

within the triton group. Its distribution range comprises

a fairly important area situated between the Tanzanian

Eastern Arc Mountains, through Southern Rift Moun-

tains and northern Zambia till the Angolan highlands. In

miombo woodlands of north-western Tanzania, it may

have overlapping distribution ranges with M. triton, but

no locality with sympatric occurrence was found in our

study. The Angolan specimens were recently reported

as M. callewaerti (Thomas, 1925) [14]. It is therefore

possible that the whole clade should belong to M. calle-

waerti, but a comparison with type material will be neces-

sary. The taxon prefers the miombo woodland or

montane forest edges. There is important genetic variabil-

ity within this taxon, with animals from Eastern Arc

Mountains forming a distinct clade supported as a separ-

ate GMYC-species (Figure 3a).

(3)The setulosus group (Figure 1c):

We recognized five MOTUs within this highly supported

monophyletic lineage. It includes relatively large-bodied

species, with distribution ranges mostly north of the equa-

tor, i.e. largely parapatric with the triton group. Two of

these MOTUs were only recorded in Ethiopia.

(MOTU 7) Mus cf. proconodon

It represents a lineage probably endemic to Ethiopia,

where it mainly occurs in lowlands of the Rift Valley.

We suggest assigning this MOTU to the species M.

proconodon Rhoads, 1896, i.e. the Ethiopian taxon that

was synonymised with M. setulosus [10] even if genetic-

ally it represents the most distinct lineage of the whole

setulosus group.

(MOTU 8) Mus setulosus Peters, 1876

This highly supported MOTU from western-central

Africa (north-west of the Congo River) represents the

true M. setulosus (type locality is Victoria, Cameroon).

The western border of its distribution likely lies in the

dry region of the Dahomey gap. In the north-east (i.e.

southern Central African Republic (CAR)), it is probably

in contact with M. bufo (MOTU 10), and it is worthy of

further study to analyse the possible contact zone and

reproductive barriers between these two taxa in CAR.

(MOTU 9) Mus cf. setulosus “West”

MOTUs 8–11 form a monophyletic group of four

strongly supported lineages with roughly parapatric dis-

tribution (Figure 1c). Two of them (MOTUs 8 and 9) have

been previously named M. setulosus (e.g. [9]). MOTU 8 is

distributed in central African forests, while MOTU 9 in

western Africa (west of the Dahomey gap). MOTUs 10

and 11 represent valid species M. bufo (Thomas, 1906)

and M. mahomet Rhoads, 1896, respectively. The topology

and genetic distances (K2P distance = 8.1%) suggest that

MOTUs 8 and 9 should be given different names. Because

M. setulosus was described from Cameroon (i.e. distribu-

tion area of MOTU 8), we suggest that the West African

populations of M. cf. setulosus, i.e. MOTU 9, may repre-

sent a separate new species, but this claim needs to be

substantiated by further taxonomic work.

(MOTU 10) Mus bufo (Thomas, 1906)

The species was described from Ruwenzori Mts. in

Uganda and it was considered endemic to the Albertine

Rift. There are few sequences identified as M. bufo in Gen-

Bank. The first (Acc. no. DQ789905) from Bujumbura in


70


Burundi was reported by [9] as an incorrectly assigned

species. Recently, new sequences of M. bufo from

Kahuzi-Biega (DRC) were published [14] and all clearly

cluster with the new sequences from CAR, DRC and

Kenya reported in our study. Furthermore, we obtained

a short sequence from the paratype of M. bufo from

DRC (locality Idjwi) that also grouped with this clade.

Although the morphological comparison with additional

type material is necessary, we suggest that M. bufo has a

much larger distribution range than previously assumed.

This taxon may also involve additional populations of the

setulosus group from Eastern Africa, especially those

assigned toM. emesi Heller, 1911 (described from Uganda;

morphologically similar to M. mahomet, with which it was

synonymised [10]), and M. pasha Thomas, 1910

(East-African taxon that was synonymized first with M.

proconodon and later on with M. setulosus [10]).

(MOTU 11) Mus mahomet Rhoads, 1896

It is an abundant species with a distribution range

restricted to the Ethiopian Plateau. We provide the first

sequences of this taxon, confirming its position within

the setulosus group as a strongly supported monophy-

letic lineage. We therefore support the view of [55], who

considered M. mahomet as an Ethiopian endemic, con-

trary to previous opinions merging it with Kenyan and

Ugandan populations (i.e. most probably with MOTU

10, which is significantly supported sister group to M.

mahomet; Figure 2).

(4)The baoulei group (Figure 1d):

This is a West African clade, until now known as a

single species, but with very pronounced divergences

between two subclades (mean K2P distance on CYTB =

9.46%) that have partially overlapping distribution

ranges in Ghana and Ivory Coast. Only very limited

genetic data are available, because the species of the

baoulei group are probably rare or difficult to capture

[12,13,23]. The species of this group occur in the

forest-savannah ecotone and are generally larger than

other West African species (except M. setulosus) [12]. The

baoulei group is a sister lineage to the sorella group

(Figure 2), which is also reflected in morphology [56].

(MOTU 12) Mus baoulei (Vermeiren & Verheyen, 1980)

The species M. baoulei was described from Lamto in

the Ivory Coast [56]. Two individuals sequenced from

the type locality [12] belong to the genetic clade that is

distributed mainly in Ghana, Benin and western Nigeria

(i.e. the type locality represents the westernmost record

of this lineage).

(MOTU 13) Mus cf. baoulei “West”

Specimens from this lineage were found in Guinea and

single individuals were sequenced from the eastern Ivory

Coast [12] and Ghana [23]. Future more-detailed studies

(using more samples, morphology and nuclear markers)

are required to resolve whether MOTUs 12 and 13 rep-

resent separate species.

(5)The sorella group (Figure 1d):

It is a lineage of relatively large animals living in the

Congo Basin’s forest-savannah transit zones, but also

reported from south-eastern Africa (Mozambique and

Zimbabwe) [57]. While very limited genetic data are

available, our sampling shows very divergent sequences

that may represent up to four species, but more data are

required for taxonomic revision of this group.

(MOTU 14) Mus sp. “Dakawa”

Two sequences from Dakawa (Tanzania) belong to the

M. sorella group, but they are very distinct from other

lineages of the group (K2P distance = 8.74-9.75%). It is

possible that they represent a new species, but more

taxonomic research is necessary. There is an existing

name, M. wamae, that may be valid for this MOTU.

This taxon was described as a member of the sorella

group from the Kapiti Plains in southern Kenya [57].

(MOTU 15) Mus sp. “Koi River”

A single specimen from the moist savannah area near

Koi River in south-western Ethiopia clearly belongs to

the sorella group, but is very divergent at CYTB (K2P-dis-

tance between MOTU 15 and other lineages of the sorella

group are 9.72-9.83%). Further taxonomic work is neces-

sary to resolve the taxonomic rank of this lineage. This is

the first record of the sorella group in Ethiopia.

(MOTU 16) Mus sorella (Thomas, 1909)

The first sequence of this MOTU was published under

the name M. sorella by [58] from Sangba (CAR). The

species M. sorella was described from hills around Mt.

Elgon, an area which has clear biogeographical connec-

tions to CAR (see e.g. MOTU 10 or clade C of MOTU27;

Figure 1c and f). We obtained one additional short

sequence from this lineage by 454 pyrosequencing of a

museum specimen from the Garamba National Park in

north-eastern DRC, thus connecting Sangba with the type

locality. However, it is also possible that these sequences

represent another currently valid species described from

CAR, i.e. M. oubanguii Petter & Genest, 1970 or M. goun-

dae Petter & Genest, 1970. More samples and detailed

analyses are required to resolve this taxonomic problem.

(MOTU 17) Mus neavei (Thomas, 1910)

Even if more morphological comparisons are necessary,

hereafter we call this south-east African clade M. neavei

and we report the first sequences of this species. The type

locality of M. neavei (also morphologically belonging to

the sorella group; [57]) is Petauke, Zambia. In our mater-

ial, this taxon is distributed in hilly areas of southern

Tanzania, Malawi and one locality in Zambia (not far from

the type locality). It occurs in sympatry with MOTU 6


71


from the triton group [57] and in the Nyika Mountains

in Malawi also with MOTU 1. The records from South

African Republic (SAR) are not yet confirmed genetically;

the specimen mentioned by [14] was finally identified as

M. minutoides and no other sequences of M. neavei were

obtained despite intensive recent sampling efforts in SAR

(F. Veyrunes, pers. comm.)

(6)The minutoides group (Figures 1e-f ):

This is the most diversified group within Nannomys,

inhabiting various, mostly open habitats of sub-Saharan

Africa. It harbours the real “pygmy” mice, i.e. the rodents

with the smallest body size (some of them with body

mass < 5 g). Most previous published genetic studies of

Nannomys mainly targeted representatives of this group.

Our phylogenetic analysis reveals three clear subgroups:

subgroup 1 (MOTUs 18 to 20), subgroup 2 (MOTUs 21

and 22), and subgroup 3 (MOTUs 23 to 27).

(MOTU 18) Mus sp. “Zakouma”

A single specimen of this taxon was captured in the

Zakouma National Park in south-eastern Chad [11]. It

is genetically very distinct from its sister species, M.

mattheyi F. Petter, 1969 and M. haussa (Thomas &

Hinton, 1920), and further taxonomic work on more

material from southern Chad may confirm it as a new

distinct species. Together with M. mattheyi and M.

haussa, this species forms a monophyletic group that

diverged in West African savannahs.

(MOTU 19) Mus haussa (Thomas & Hinton, 1920)

It is a Sahelian taxon, recorded in the belt from Senegal

to western Chad [9]. Similarly as in M. mattheyi and other

West African savannah species of rodents [59-61], there

is also indication of longitudinal genetic structure in M.

haussa, but more detailed data are needed for more

conclusive phylogeographical inferences.

(MOTU 20) Mus mattheyi F. Petter, 1969

M. mattheyi is typical species of Guinean savannah-forest

mosaic from westernmost Africa (Senegal) to the Dahomey

gap, the relatively dry region separating Guinean and

Congolese forest blocks [9]. It is divided into western

and eastern phylogeographic subclades with a presum-

able contact zone in the Ivory Coast (not shown). It is

often the most abundant Nannomys in the rodent

assemblages [13,23].

(MOTU 21) Mus cf. kasaicus

Two sequenced individuals from the Kikwit region

(DRC) formed this genetically very distinct genetic MOTU.

There are also indications from other rodent groups that

the Kikwit area is a local centre of endemism (see e.g.

MOTU 5 or [62]). There is an existing name, M. kasaicus

(Cabrera, 1924), described from Kasaï Occidental Province,

Kananga, DRC, for the taxon belonging morphologically to

M. minutoides group [10], that may apply to this MOTU.

(MOTU 22) Mus indutus (Thomas, 1910)

M. indutus is a south African species, found in a rela-

tively large area from northern Botswana to southern

SAR [11,14,63,64]. Records from Zambia and Malawi are

based on genotyping of old museum material [64] and

should be taken with caution. It is probably sympatric

with M. minutoides Smith, 1834 (= MOTU 27) in most

of its distribution range.

(MOTU 23) Mus cf. gratus

Specimens from this taxon were typically captured in

forest clearings and the ecotone between forest and open

habitats in equatorial Africa. There are three distinct

clades with clear west–east geographical structure: (i) a

single specimen from lowland tropical forest in Congo

(K2P distance to two remaining clades is cca 7%); (ii) the

Kisangani region in DRC; and (iii) both montane and

lowland tropical forests in southern Kenya and northern

Tanzania. More taxonomic work is necessary to link this

clade to an existing species; possibly M. gratus (Thomas &

Wroughton, 1910), a taxon from the minutoides group

described from eastern Ruwenzori, “upper Congo” and

Virunga mountains. Again, the comparison with the types

will be required to verify this hypothesis.

(MOTU 24) Mus cf. gerbillus

This taxon is distributed in dry Somali-Maasai savannah

in Kenya and Tanzania. In all phylogenetic analyses, it

is a sister clade to the Ethiopian MOTU 25 (mean K2P

distance between these two clades is 8.87%). Further

taxonomic work is necessary, but M. gerbillus (G.M. Allen

& Loveridge, 1933) (currently the synonym for Tanzanian

populations of M. tenellus) is an available name that may

apply to this lineage.

(MOTU 25) Mus cf. tenellus

This lineage was found at two close localities in northern

Ethiopia - in Hagere Selam and in the Mekelle University

campus. It may represent true M. tenellus (Thomas, 1903)

described from Blue Nile in Sudan, but the comparison

with the type material is necessary. On the contrary, mor-

phological studies of museum material suggested that most

published Ethiopian records of M. tenellus were actually

M. minutoides [10].

(MOTU 26) Mus musculoides Temminck, 1853

It is a typical species of the Sudanian savannah belt. It

was previously reported from western Africa [11,12,17]

and northern Cameroon [65]. We provide a new very

distant record from western Ethiopia, representing the

easternmost genetically confirmed locality of the species.

Very probably it is also present in poorly sampled coun-

tries such as Chad, northern CAR and South Sudan.

(MOTU 27) Mus minutoides Smith, 1834

M. minutoides is a widely distributed species in most

of sub-Saharan Africa (probably except continuously for-

ested areas in the Congo Basin and deserts; Figure 1f ).

This MOTU also includes specimens from southern


72


Ethiopia; some of them were previously called M. tenellus

[9,10]. This species has a very strong intraspecific phylo-

geographical structure. Median-joining network analysis

of 131 sequences from this MOTU resulted in 84 haplo-

types that form 11 strongly delimited haplogroups

(Figure 4). The mean K2P-corrected distances among hap-

logroups ranged from 1.21% (TZw vs. KE) to 3.65% (ZA

vs. Chin). All haplogroups are connected in the form of a

star, suggesting multiple synchronous vicariance events.

Allopatric divergences with subsequent expansions are

further supported by current parapatric distribution of

most clades and frequent, but narrow, secondary contacts

among them (Figure 1f). The geographic structure within

individual haplogroups is relatively weak, except clade SE,

where it is possible to distinguish the separate sublineages

from South Africa (h79-h81), Mozambique (h15-h17) and

Tanzania (remaining haplotypes). Two haplogroups are

only represented by animals from single localities (Minziro

and Chingombe in Tanzania), but it is possible that they

are more widespread in neighbouring regions in eastern

DRC, where the relevant samples are missing.

Divergence dating

The basal split of the extant Nannomys was dated at

5.24 Mya with 95% of the highest posterior density

(HPD) between 4.58–5.96 Mya. Successive divergence of

the extant major species groups then took place through-

out Pliocene, with median estimates of divergence times

ranging from 4.9 Mya (split off of MOTU 1 “Nyika”) to

2.44 Mya, i.e. the divergence of MOTU 23 (cf. gratus) and

MOTUs 24–27 (i.e. four other species of the minutoides

group) (Additional file 2). Posterior estimates of diver-

gence dates at the calibration points are shifted towards

past in the case of Apodemus (prior median 5.89, posterior

median 7.38) and Arvicanthis-Otomys (6.81 vs. 8.13)

but towards the present in the case of Mus (8.00 vs.

7.44). Two other divergence dates are also worth not-

ing: the split-off of Myomyscus yemeni estimated at 6.21

(5.12–7.33) Mya, which is consistent with its migration

to Arabian peninsula across the land bridge during the

Messinian crisis, and the origin of modern Otomys 3.77

(2.83–4.81) Mya, first appearing in the fossil record

around 3 Mya [[66], p.290]. Complete results of the

h36h35

h33h34

h31

h28h71h32

h7

h26

h27

h84

h70

h5

h6h4

h2

h3

h1

h15h17

h16h18

h76h75

h21h77

h22h82

h74h29

h19h20

h79

h80

h81

h8

h25

h24

h37-h64

h78

h69

h67h68

h73h72

h83

h9

h65

h10

h11h13

h14h12

h66

h29

h30

S Africa

nord (SAn)

S Africa

south (SAs)

S Kenya

(KE)Zambia and

surrounding (ZA)

SE Africa

(SE)

„Chingombe“,

Tanzania (Chin)

Ethiopia,

N Kenya (ET)

Central

Africa (C)

W Africa

(W)

W Tanzania

(TZw)

„Minziro“,

Tanzania (Min)

Figure 4 Phylogeographical structure of Mus minutoides (MOTU 27). Haplotype network was constructed by the median-joining algorithm

from 131 CYTB sequences (84 haplotypes) in the program Network. The circle size is proportional to haplotype frequency and the connecting

lines are proportional to number of substitutions.


73


divergence dating analysis are reported in Additional

file 2.

The full set of branching times between 27 MOTUs is

given in Figure 3a. It is based on the secondary dating of

the ultrametric tree for GMYC, but the posterior esti-

mates of divergence dates are consistent with previous

analysis (compare Figure 3a and Additional file 2). Main

species lineages diverged in lower Pliocene (5.2-4.5 Mya)

and an intensive period of speciation is also visible in

the lower Pleistocene (2.1-1.6 Mya), when many extant

lineages within main species groups appeared.

Biogeographical analysis

Bayesian analysis of discrete traits in BEAST revealed that

the most ancestral distribution (98% support) of Nan-

nomys included mountains of Eastern Africa (Figure 3b).

This type of distribution is currently present in all three

ancient monotypic lineages (MOTUs 1–3), as well as in

numerous lineages of the triton and setulosus groups.

There are two major habitat shifts in the Nannomys

evolution. (1) The lineage leading to the baoulei group

colonized the forests (and forest-savannah mosaic) in

western Africa cca 4 Mya, where it split to western and

eastern sublineages later on; (2) the minutoides group

descended from mountains, adapted to more arid open

habitats, and started to radiate across the whole sub-

Saharan Africa cca 3.5 Mya. In the first radiation phase,

MOTUs 18–27 speciated in savannah-like habitats over

all of Africa (approx. 3.5-1.6 Mya). Geographically similar,

but much more recent (cca 1 Mya) radiation occurred

inside MOTU 27, i.e. M. minutoides (Figures 1f, 3a, and 4).

Very similar results were obtained by the maximum

likelihood approach in Lagrange (Additional file 5). Most

basal splits occurred with the highest probability in the

mountains of East Africa, also where most of the MOTUs

from the triton group diverged. The first clear shifts to

other habitats are visible in the ancestors of the baoulei

group (to the forests or forest edges, where both MOTUs

from this group occur until today) and in the ancestors

of the minutoides group (to the savannah). The most

intensive radiation in the latter took place in savannahs,

with one shift to the forest habitat detected in MOTU

23 (M. cf. gratus). The estimates of ancestral ranges are

less clear in the setulosus and the sorella groups. While

the former started to diverge most probably in mountains

(with subsequent spreading of two “setulosus” MOTUs to

forests of central and eastern Africa), the latter had ances-

tors occurring with similar support either in savannahs or

in hills of Eastern Africa.

Discussion

For the purpose of our study, we compiled new and

existing sequences into the largest genetic dataset to

date of the subgenus Nannomys and performed the first

phylogenetic analysis of the group that contains most of

the currently recognized valid species across the whole

sub-Saharan Africa. We detected a surprisingly high

amount of cryptic diversity, with numerous candidates for

new species. Wide geographical sampling also allowed

the first empirical definition of the distribution areas of

all the detected lineages based on physically present

genotyped individuals. Using several calibration points

and the current distributional data, we also carried out

biogeographical analysis and reconstructed the possible

evolutionary scenario of this highly successful group of

sub-Saharan murines.

Species concepts and estimation of the number of

Nannomys species

Species diversity crucially depends on the adopted species

concept. Widely used concepts of typological or biological

species are not always applicable for species delimitation

because of frequent convergent evolution, cryptic species,

and the impossibility of proving reproductive isolation

among allopatric populations. Together with the rapidly

increasing amount of genetic data from free-living popula-

tions, these concepts are often complemented by genetic

[37] or phylogenetic [3] species concepts, creating the so-

called integrative taxonomic approach. Although genetic

approaches can sometimes lead to an unjustified increase

in the number of species (so-called taxonomic inflation

[2,67]), they often detect cryptic diversity within evolu-

tionary lineages that can be generally important from

the taxonomic as well as conservation point of view. In

our study, we used the combination of maximum likeli-

hood delimitation of phylogenetic species and the

genetic distances to estimate the number of MOTUs

(= putative species) of Nannomys in Africa. We are

aware of the drawbacks of these approaches (e.g. the

use of only maternally inherited mtDNA), however,

our aim was not to perform the taxonomic changes

based solely on limited genetic data, but rather to

identify the taxa and regions of high cryptic diversity

requiring more detailed taxonomic studies.

The combination of different approaches revealed the

existence of 27 MOTUs. This is considerably more than

the 18 currently accepted Nannomys species [9,10], sug-

gesting that numerous putative species have so far

remained undetected, and therefore undescribed. Most of

the genetic data of Nannomys that have been collected to

date originate from Western and Southern Africa, where

the taxonomy of this group has been intensively explored

(reviewed by [9]). The number of candidates for potential

new species in western Africa revealed by our study is

therefore relatively low (only M. cf. setulosus “West” or M.

cf. baoulei “West”) and it is also possible that these

MOTUs just represent marked phylogeographical struc-

ture of within-species lineages with parapatric distribution


74


(but see [12] that already suggested M. setulosus as a

species complex).

The situation is completely different in Eastern Africa,

from where only fragmentary genetic data were available

prior to this study. Our results may lead to the description

of more than 10 new species that are already now

sufficiently delimited using the combination of genetic,

ecological and geographic data. Many of these so far

undescribed taxa occur in mountains or highland habitats,

but a few other potential new species (like M. cf. gerbillus)

are typical inhabitants of dry savannahs. The taxonomic

diversity of Nannomys is probably the highest in Ethiopia.

As for many other organisms, the Ethiopian highlands

represent an important hot-spot of African endemism for

Mus. We have revealed the presence of 8 MOTUs in this

country, and only two of them (M. minutoides and M.

musculoides) have also been recorded outside Ethiopia.

The six remaining species are probably endemic, mak-

ing Mus the genus with the second highest number of

Ethiopian mammal endemics (after Lophuromys with 9

endemic species; [68]).

Even if we have not sequenced the type material of

most currently valid taxa (except paratypes of M. bufo),

we have been able to assign the most probable species

names to 13–14 MOTUs based on previous genetic

studies (including karyotypes; [11,12]), geographical dis-

tribution (i.e. sequences from the type locality or close

neighbourhood) and external morphology. Therefore,

the genetic dataset from this study represents a solid

basis for future identification of morphologically similar

Nannomys species via DNA barcoding (using e.g. evolu-

tionary placement algorithm; [69]). Unfortunately, our

dataset lacks sequences of four valid species. M. oubanguii

Peter & Genest, 1970 and M. goundae Peter & Genest,

1970 represent two species from the sorella group known

only from few localities in the Central African Republic.

They were described mainly on the basis of external

morphology [57] and their specific status has been ques-

tioned previously ([10]; but see conspicuous differences in

karyotypes of these two species - reviewed in [9]). The

whole sorella group requires a profound revision including

new sampling in savannahs north of the Congo Basin and

additional genetic data. We found high genetic variation

within the sorella group, but most clades are represented

by only one or two localities (except M. neavei) and in

most cases it is not possible to assign the particular clades

to currently valid species names. M. setzeri Petter, 1978 is

a rare taxon with limited distribution in dry areas of

Namibia, Botswana and western Zambia [70,71]; it is

probably a valid species as it can be morphologically

distinguished from sympatric Nannomys species [63,72].

M. orangiae Roberts, 1926 is the fourth species that is

currently valid and missing from our dataset. It also is a

southern African species with unclear taxonomical status.

It was previously considered a subspecies of either M.

setzeri or M. minutoides [10] and may just represent one

of the cytotypes of the latter [9].

Phylogenetic estimate of species richness of Nannomys

in our study (25–30 MOTUs that may represent separ-

ate species) suggests that it is one of the most speciose

groups of African terrestrial mammals. Similarly well

studied species-rich genera of African rodents usually

have a lower number of monophyletic genetic lineages

considered as species, e.g. Praomys (16–20 species; [73])

or Hylomyscus (21 species, including undescribed and

recently described taxa; [74], J. Kennis et al., submitted).

The only genus with higher described species richness is

Lophuromys (29 species; [10,68,75]). However, this genus

is specialized to tropical forests and ecotones and it is

likely that intensive genetic drift in fragmented forest

habitats (especially in Eastern Africa) caused morpho-

logical distinctiveness allowing differentiation of a high

number of genetically similar morphospecies [68,75]. It

is also worth to note that in comparison with the above-

mentioned genera, Nannomys colonized a much wider

spectrum of habitats (from Afroalpine meadows and

mountain forests to very arid savannah).

Mus minutoides as a model for pan-African phylogeography

The MOTU with the largest distribution of all Nannomys

is M. minutoides (=MOTU 27). There are only a very few

such widespread savannah-forest mosaic species distrib-

uted across almost complete sub-Saharan Africa. Among

rodents, only the ubiquitous Mastomys natalensis had

held this habitat breadth, and it was considered the rodent

species with the largest distribution area in Africa [10].

Our genetic data confirm that M. minutoides has very

similar and probably even larger distribution than M.

natalensis. It can be argued that MOTU 27 does not rep-

resent a single species but rather a species complex, which

may be supported by the GMYC analysis revealing signifi-

cant support for additional speciation events within this

clade (see Figure 3a). However, in absence of more de-

tailed evidence, we prefer to maintain all genetic lineages

of MOTU 27 within the species M. minutoides. They do

not show visible external differences (although detailed

morphological analysis of genotyped material is still miss-

ing), they radiated relatively recently (last 1 Mya) and the

Tamura-Nei corrected genetic distances among clades

(1.21-3.65% on CYTB) are comparable with those among

clades of M. natalensis (2.1-3.8%; [76]), i.e. much lower

than usual genetic distances between sister species of

rodents [37]. Further detailed studies should focus on the

contact zones of divergent clades to reveal whether they

can interbreed or not.

Species with large distributions and strong affinities to

open habitats can serve as possible models for comparative

pan-African phylogeography of the savannah-like biomes.


75


Recent phylogeographic studies of M. natalensis showed

that populations were strongly influenced by Pleistocene

climate fluctuations [76]. The presence of genetically

divergent clades with parapatric distribution is congru-

ent with the scenario invoking allopatric fragmentation

and vicariance. Almost the exact same geographic pattern

of genetic differentiation is visible in M. minutoides (com-

pare Figure 1f in this study with Figure 1 in [76]). The

phylogeographic pattern suggests at least 11 different

savannah refugia approximately 1 Mya, i.e. in the period

of very strong climatic instability [77]). The genetic line-

ages evolved in allopatry and subsequently spread during

suitable periods of savannah expansion. Further research

should focus on precise localization of refugia by combin-

ing information from population genetics with modelling

of past ecological conditions [78]).

A new biogeographical scenario of Nannomys radiation in

Africa - from mountains to lowland forests, savannahs

and arid Sahelian environments

More complete taxon sampling from the whole sub-

Saharan Africa now allows significant modification and

increased precision of the previously proposed biogeo-

graphical scenario of Nannomys radiation in Africa [11].

Our molecular dating based on plausible paleontological

calibration and taxon-unbiased phylogenetic tree suggests

that the divergence of the genus Mus to the current sub-

genera occurred in Asia in the late Miocene (cca 6.8-7.4

Mya), which is in good agreement with previous studies

[7,11,44]. The colonization of Africa by Mus occurred very

probably in the Messinian period (7.3-5.3 Mya) when the

temporary land bridge connected Africa and southwest

Arabia. In this period, many faunal exchanges between

Africa and Asia are well documented [79-81]. It is there-

fore highly probable that Mus was already in Africa at the

beginning of the Pliocene. The basic split of the extant

Nannomys was dated at 5.24 Mya (95% HPD 4.58–5.96

Mya), i.e. very soon after a land bridge between Africa and

Southwest Arabia disappeared (5.3-6 Mya; [82,83]). The

oldest fossil evidence of the genus Mus in Africa was from

the early to middle Pliocene in Ethiopia (the Omo valley

in the south of the Ethiopian Rift Valley and Hadar in the

east, 5–2.5 Mya; [84,85]) and Kenya (4.5 Mya; [27]).

Due to incomplete sampling (mainly in eastern Africa)

previous studies could not adequately explore the evo-

lutionary history of Nannomys, especially since our

biogeographical reconstructions demonstrate that the

first divergence of Nannomys occurred in eastern Africa.

Paleoclimatic and paleoanthropological research in eastern

Africa suggested repeated association of critical events

in hominin evolution with the most prolonged intervals

of high climate variability. Potts (2013) [77] defined

eight intervals of predicted high climate variability in

the last 5 My and argued that most important events in

hominin evolution occurred within these periods. Three of

the most prolonged intervals of predicted high climate vari-

ability are 2.79-2.47 Ma, 1.89-1.69 Ma, and 1.12-0.92 Ma

and they largely overlap with the previously defined periods

of the occurrence of large lakes [86] as well as with inferred

aridity phases based on dust records, paleosol δ13C, and

the prevalence of grazing bovids [87].

Clear associations between periods of climatic instability

and divergence events are also visible in phylogenetic

reconstructions of Nannomys. The first splits leading

to ancestors of most current species groups probably

occurred in eastern Africa in the period 5.2-4.5 Mya

(Figure 3a), which corresponds to the longest era of

strong wet-dry variability [77]. Nothing is known about

the ecology of the extinct Mus taxa, but surviving

ancient lineages (MOTUs 1–3) may provide some clues.

They can be considered “living fossils”, i.e. monotypic rel-

ict taxa living in very restricted areas in Eastern African

mountains. The period 4–3.5 Mya is considered relatively

stable with few documented evolutionary events [77] and

we observed only two vicariance events in Nannomys

during this period. The first is the north–south split of

MOTU 3 (M. sp. “Harenna”) and the triton group, and

the second is the west–east split of the baoulei and the

sorella groups (see Figure 3a and compare it with distri-

butions at Figure 1). The most intensive radiation of

Nannomys is dated into the period 3.5-1.4 Mya (see

Figure 3a), when most current MOTUs (i.e. putative

species) appeared. The beginning of this period coin-

cides with the start of a cooling and aridification trend

[88]. The open savannah-like habitats were spreading

intensively and at the same time the climate was very

variable (four prolonged periods of strong wet-dry variabil-

ity are dated into this range; [77]). This variable climate

likely yielded environmental changes that increased the fre-

quency of evolutionary responses like adaptation, dispersal

(especially in open habitats), and ultimately, speciation (for

example it was also the period with the highest number

of hominin taxa; [89]). Our biogeographic analyses are

consistent with these findings because the most intensive

radiation occurred in the minutoides lineage in savannahs.

The presumed shift from mountains to more arid and

open habitats was clearly linked with the decrease of the

body size in the minutoides lineage. The ancient M.

imberbis (MOTU 2) has a body size of 25 g [8], MOTU 1

(M. sp. “Nyika”) has 14 g, MOTU 3 (M. sp. “Harenna”)

has cca 16 g (our unpublished data) and the members of

other non-minutoides groups weight 8–13 g [9]. In

contrast, all species of the minutoides clade have body size

3–8 g, making them one of the smallest mammals in the

world [9]. The last period of climatic instability is dated to

1.12-0.92 Mya, which coincides with the likely simultan-

eous split of the MOTU 27, i.e. Mus minutoides, into 11

distinct genetic lineages (see above).


76


Ecological constraints and multi-species sympatry

Previous studies revealed that at several sites more than

one species of Nannomys occurs in sympatry [12,13,23].

Their observations are in agreement with the distribution

ranges based on genotyped individuals (Figure 1) showing

largely overlapping distribution areas of many species.

However, if we exclude widely distributed M. minutoides

(MOTU 27), the distribution of individual species within

the same species group is predominantly parapatric (most

illustrative in Figure 1b, c, d), while sympatry is typical for

species from different species groups. This suggests that

the species groups might have evolved specific morpho-

logical adaptations that allow their sympatric occurrence

with the members of other Nannomys lineages. Although

detailed morphological analysis of genetically identified

specimens is still missing, preliminary data suggest clear

differences in the skull morphology among the species

groups ([12], E. Verheyen et al., unpublished data), with

possible functional consequences in separation of eco-

logical niches (for example dietary).

Even if the distribution areas of two or more species

from the same species group overlap, closer examination

of our data provide evidence for the preference of differ-

ent habitats. For example two Ethiopian endemics from

the setulosus group, M. cf. proconodon (MOTU 7) and

M. mahomet (MOTU 11), have never been captured at

the same locality; the former prefers lowland habitats in

the Rift Valley, while the latter is common species across

the Ethiopian highlands. Up to four species of the minu-

toides group can be found in western Africa, but their

ecological requirements are probably different. Based on

the data summarized at Figure 1 and published records,

it seems that M. minutoides is able to live in Western

Africa in relatively humid places, M. mattheyi prefers

dry Guinean savannah and the transition zone between

forest and savannah, M. musculoides is a typical inhabitant

of Sudanian savannah belt from Guinea to western

Ethiopia and M. haussa lives in arid Sahelian environment

([12,65], figure one in this study). Similarly M. minutoides

can occasionally be found in the same localities as M.

indutus in southern Africa, but the latter probably prefers

drier habitats ([63,64] and references therein).

Relevance to the understanding of karyotype evolution

and sex determination

The subgenus Nannomys has previously been used as a

suitable model for studies of karyotype evolution due to

very high variability of chromosomal rearrangements

[11,17,90,91]. The ancestral karyotype of the pygmy mice

was composed of 36 acrocentric chromosomes [17,92], but

the wide spectrum of mutational mechanisms modified

the chromosomal constitution. Besides relatively frequent

Robertsonian translocations, other chromosomal rearrange-

ments were described in Nannomys, including variable

sex-autosome translocations, pericentric inversions, tandem

fusions and WARTs (Whole-Arm-Reciprocal Transloca-

tion) [9]. Pan-African phylogeny based on more complete

taxon sampling presented in our study can help to under-

stand the karyotype evolution in general and sex determin-

ation mechanisms in particular. The mapping of karyotype

features on the phylogenetic tree can help to define spe-

cific predictions that can be further verified by sampling

focussed on particular species and geographical areas.

For example, tandem fusions - one of the rarest

chromosomal rearrangements - were evidenced in M.

triton (MOTU 4) and M. sp. “Harenna” (MOTU 3) that

in most phylogenies cluster together. Even if they were

suspected in two other species of the sorella group in

the CAR (M. goundae and M. oubangui, not sampled in

our study; [9]), further detailed studies of these rare

mutations should direct their focus on widely distributed

and common MOTU 6 belonging to the triton group. One

of the most conspicuous features of Nannomys karyotypes

is the fusions of autosomes and sex-chromosomes. These

fusions were most frequently studied in two terminal taxa

of the minutoides group (MOTU 26 - musculoides and

MOTU 27 - minutoides), but they were also observed in

M. goundae and M. oubangui (very probably belonging

to the sorella group) and in M. triton (MOTU 4) [9].

Since they appeared several times independently, it is

therefore clear that predispositions for translocations of

sex chromosomes exist in more lineages of Nannomys,

yet these translocations are not a general feature of the

whole subgenus, as, for example the setulosus group is

very conservative and all MOTUs karyotyped until

today have the ancestral karyotype (2n = 36, NF =36)

([78] and references in [9]). Future research on East

African species of the minutoides group (MOTUs 23–25,

i.e. M. cf. tenellus, M. cf. gerbillus, and M. cf. gratus), the

sorella group and the triton group could thus potentially

bring interesting new insights on the evolution and poly-

morphism of sex-autosome translocations. Finally, the

phylogeographic pattern described in our study for the

most karyotypically variable species, M. minutoides, can

help to design further sampling of chromosomal data in

lineages, where the karyotypes are not yet known. The

haplotype network suggests 11 main lineages that prob-

ably differentiated in small allopatric populations at the

same time, which could have led to establishment and fix-

ation of important karyotypic differences [64,90]), possibly

involving presently unknown means of sex determination

[91]. If such karyotype differences among genetic lineages

exist, it would be also extremely interesting to study the

possible contact zones among them (see Figure 1f).

Consequences for future epidemiological studies

Rodents are reservoir hosts of important human patho-

gens, of which some can cause serious diseases. Most


77


recent examples of emerging and re-emerging diseases

have been caused by RNA viruses [93] and understanding

of their evolution and epidemiology is essential for pre-

dicting future emergences and designing interventions (e.

g. vaccinations). Among RNA viruses hosted primarily by

African rodents, the Lassa arenavirus has received most

attention, because it is responsible for Lassa hemorrhagic

fever in West Africa, which causes thousands of human

deaths each year [94]. The host specificity of arenaviruses

is thought to be relatively strict with often a single species

described as the primary reservoir host [21]. A long-term

evolutionary history between arenaviruses and their hosts

(co-evolution) was originally suggested due to the almost

perfect sorting of arenavirus lineages into rodent clades

(e.g. [95]). Recent studies suggest that pygmy mice are

frequent hosts of arenaviruses (and probably also other

important parasites) and they often live close to human

habitations (e.g. [13]). The pan-African phylogeny of

Nannomys proposed in this paper can help to describe

the co-evolutionary patterns of arenaviruses and their hosts

and even provide the potential clues for understanding the

occasional disease outbreaks.

The first virus found in Nannomys was the virus Kodoko,

described from Mus minutoides in western Africa, and

belonging to the lineage of lymphocytic choriomeningitis

virus (LCMV) that is hosted by the house mouse [20]. This

finding (followed by description of new Kodoko strain in

Eastern African, [21]) thus supported a co-evolutionary sce-

nario, because all arenaviruses known from African murine

hosts at that time grouped according to taxonomic position

of theirs hosts (i.e. three main lineages of African arena-

viruses were hosted by rodents of three tribes, Praomyini,

Arvicanthini and Murini; Lecompte et al. 2007, Gouy de

Belocq et al. 2010). However, the next arenavirus, called

Gbagroube and described from Mus cf. setulosus (MOTU 9

in this study) from the Ivory Coast, does not belong to the

LCMV lineage (specific to Mus), but surprisingly clusters

with the Lassa virus strains [22]. Very recently, two other

arenaviruses were found in Nannomys in Ghana [23]. The

virus Natorduori is hosted by M. mattheyi (MOTU 20, the

minutoides group) and clusters clearly into Mus-specific

LCMV lineage. In contrast, the virus Jirandogo, the first

arenavirus reported fromM. baoulei, in various phylogenies

based on its different genome segments belongs to the

Lassa virus group (similarly as Gbagroube virus). African

pygmy mice are therefore the first group of African rodents

that host two very different lineages of arenaviruses; one of

them seems to be Mus-specific (in Africa now reported

from two species in the minutoides group), but the second

forms the sister lineage of the highly pathogenic Lassa virus

(hosted by species from the setulosus and the baoulei

groups). Further surveillance for new arenaviruses focussed

preferentially on Nannomys lineages where no viruses have

yet been found (e.g. the triton or sorella groups widely

distributed in central and eastern Africa) can increase un-

derstanding of the evolution of these pathogens and predict

the regions of possible epidemiological importance.

Conclusions

The known species diversity of tropical organisms is highly

underestimated even for relatively well known animals like

mammals. Here we performed a phylogenetic analysis of

the largest available set of genetic data collected from the

only indigenous African lineage of the genus Mus, called

Nannomys. A conservative definition of MOTUs suggests

that the number of species described to date represents

only approximately 60% of possible species diversity and

intensive taxonomic work is now required to allow the for-

mal description of genetically divergent lineages. We also

provide the first reliable genotype-based distribution ranges

of particular MOTUs that can aid in future species inven-

tories in different parts of Africa. The dating of divergences

and biogeographical analyses strongly suggest that ances-

tors of Nannomys colonized Africa at the end of Miocene

and diverged to ancestors of the main species groups in

mountains of Eastern Africa in lower Pliocene. The aridifi-

cation that started in Africa cca 3 Mya led to spreading of

open habitats and provided new ecological niches that were

fully utilized by Nannomys. In particular, the so-called

minutoides lineage underwent an exceptionally intensive

radiation in savannah-like habitats and occupied almost

whole sub-Saharan Africa in several colonization waves.

The combination of a detailed phylogeny based on an

almost complete taxon sampling combined with genotype-

based distributional data of lineages, taxa and valid species

provides a solid foundation to address specific ecologically-

explicit evolutionary hypotheses using Nannomys as a

model system, i.e. in evolution of sex determination and

host-virus co-evolution.

Availability of supporting data

The newly produced sequences were submitted to GenBank

under accession numbers KJ935741-KJ935873 (CYTB) and

KJ935874-KJ935905 (IRBP) (see Additional file 1 for more

details). The final alignment of concatenated sequences

used in phylogenetic analyses is in Additional file 6.

Additional files

Additional file 1: Details on collecting localities and genetic data

for all Nannomys specimens.

Additional file 2: List of sequences used for divergence dating and

resulting fossil-calibrated timetree.

Additional file 3: Maximum likelihood phylogeny of Nannomys

based on separate analyses of mitochondrial CYTB and nuclear IRBP

genes.

Additional file 4: Distribution of genetic distances at CYTB within

and among taxa delimited by different methods.


78


Additional file 5: Biogeographical reconstruction of ancestral

distribution of Nannomys lineages using maximum likelihood in

Lagrange.

Additional file 6: Alignment of 179 ingroup and 7 outgroup

concatenated sequences of CYTB and IRBP.

Abbreviations

CYTB: Mitochondrial gene for cytochrome b; IRBP: Gene for interphotoreceptor

binding protein; Mya: Million years ago; mtDNA: Mitochondrial DNA;

ML: Maximum likelihood; BI: Bayesian inference; MOTU: Molecular operational

taxonomic unit; GMYC: Generalized mixed Yule-coalescent model;

PCR: Polymerase chain reaction; AIC: Akaike information criterion;

LCMV: Lymphocytic choriomeningitis virus.

Competing interests

The authors declare that they have no competing interests.

Authors’ contributions

JB, RŠ and EV conceived and designed the study, JB, RŠ, VM, YM, LL, KW, NO,

JM, NA, MC, HL collected important part of samples, TA and YM performed

laboratory analyses, JB and OM analysed data (phylogenetic and

biogeographic analyses), and JB (with help of OM) wrote the first draft of the

manuscript. All authors contributed to the final version of the paper. All

authors read and approved the final manuscript.

Authors’ information

JB is head of molecular ecology group at Institute of Vertebrate Biology ASCR,

generally interested in factors affecting evolution of vertebrate populations. His

actual topics include phylogeography and speciation in Africa, conservation

genetics and mechanisms of host-parasite co-evolution. OM and TA are post-doc

and PhD student in JB's lab, respectively. RŠ, YM, LL, HL and EV are leaders of

research groups studying ecology and evolution of vertebrates, mainly mammals

and especially rodents, in different parts of Africa. VM, NO, JM, KW, NA and MC

are African and European collaborators of above-mentioned researchers, with

joint interest in biodiversity of African rodents.

Acknowledgements

This study was supported by the projects of the Czech Science Foundation, no.

P506-10-0983 (JB and RŠ), the Russian Foundation for Basic Research, no. 12-04-

01283 (LL), and the Belgian Science Policy project 'Evaluating the effect of Pleis-

tocene climate changes on speciation patterns in selected African vertebrates’

(EV). For help in the field, we acknowledge H. Patzenhauerová-Konvičková, J.

Šklíba, M. Lövy, C. Sabuni, G. Mhamphi, F. Sedláček, S. Šafarčíková, A. Konečný, S.

Gambalemoke Mbalitini, and all local collaborators. The assistance of R. Makundi,

W.N. Chitaukali, B. Dudu Akaibe and late W. Verheyen with project logistics and

sample collection is highly appreciated. H. Patzenhauerová-Konvičková and L.

Piálek helped with genotyping and E. Fichet-Calvet and S. Gryseels provided

unpublished sequences. For permission to carry out the research and to collect

specimens we are obliged to the National Research Council and Forestry

Department in Malawi, the National Council for Science and Technology, the

Kenyan Forest Service and the Kenyan Wildlife Service, the Ethiopian Wildlife

Conservation Authority, Zambian Wildlife Authority, Sokoine University of

Agriculture, and the ‘Centre de surveillance de la biodiversité’ in Kisangani, RD

Congo. We would also like to thank the curators of the museums (G. Csorba, N.

Duncan, V. Nicolas and W. Wendelen) for allowing us to study the material in

their care, and two anonymous reviewers and J.C. Winternitz for comments on

the previous version of the manuscript.

Author details1Institute of Vertebrate Biology, Academy of Sciences of the Czech Republic,

Brno, Czech Republic. 2Department of Botany and Zoology, Faculty of

Science, Masaryk University, Brno, Czech Republic. 3Institute of Animal

Physiology and Genetics, Academy of Sciences of the Czech Republic, Brno,

Czech Republic. 4Department of Zoology, Faculty of Science, University of

South Bohemia, České Budějovice, Czech Republic. 5Department of Biology,

College of Natural and Computational Sciences, Mekelle University, Tigray,

Ethiopia. 6A.N.Severtsov Institute of Ecology and Evolution RAS, Moscow,

Russia. 7Earth Watch Institute, Nairobi, Kenya. 8College of Agriculture and

Veterinary Sciences, University of Nairobi, Nairobi, Kenya. 9University of

Kisangani, Eastern Province, Kisangani, DR, Congo. 10CNRS UMR 6552/53,

Université de Rennes 1, Station Biologique, Paimpont, France. 11Evolutionary

Ecology Group, Biology Department, University of Antwerp, Antwerpen,

Belgium. 12Royal Belgian Institute for Natural Sciences, Operational Direction

Taxonomy and Phylogeny, Brussels, Belgium. 13Institute of Vertebrate Biology,

Academy of Sciences of the Czech Republic, Research Facility Studenec,

Studenec 122, 675 02 Koněšín, Czech Republic.

Received: 30 May 2014 Accepted: 27 November 2014

References

1. Mora C, Tittensor DP, Adl S, Simpson AGB, Worm B: How many species are

there on earth and in the ocean? PLoS Biol 2011, 9(8):e1001127.

2. Zachos FE, Apollonio M, Bärmann EV, Festa-Bianchet M, Göhlich U, Habel JC,

Haring E, Kruckenhauser L, Lovari S, McDevitt AD, Pertoldi C, Rössner GE,

Sánchez-Villagra MR, Scandura M, Suchentrunk F: Species inflation and

taxonomic artefacts - A critical comment on recent trends in mammalian

classification. Mamm Biol 2013, 78:1–6.

3. Groves CP: The nature of species: A rejoinder to Zachos et al. Mamm Biol

2013, 78:7–9.

4. Krishna Krishnamurthy P, Francis RA: A critical review on the utility of DNA

barcoding in biodiversity conservation. Biodivers Conserv 2012, 21:1907–1919.

5. Hoffmann M, Grubb P, Groves CP, Hutterer R, Van der Straeten E, Simmons

N, Bergmans W: A synthesis of African and western Indian Ocean Island

mammal taxa (Class: Mammalia) described between 1988 and 2008: an

update to Allen (1939) and Ansell (1989). Zootaxa 2009, 2205:1–36.

6. Macholán M, Baird SJE, Munclinger P, Piálek J: Evolution of the house mouse.

New York: Cambridge University Press; 2012.

7. Chevret P, Veyrunes F, Britton-Davidian J: Molecular phylogeny of the

genus Mus (Rodentia: Murinae) based on mitochondrial and nuclear

data. Biol J Linnean Soc 2005, 84:417–427.

8. Meheretu Y, Šumbera R, Bryja J: Enigmatic Ethiopian endemic rodent

Muriculus imberbis (Rüppell 1842) represents a separate lineage within

genus Mus. Mammalia 2014, in press (doi:10.1515/mammalia-2013-0119).

9. Britton-Davidian J, Robinson TJ, Veyrunes F: Systematics and evolution of

the African pygmy mice, subgenus Nannomys: A review. Acta Oecol 2012,

2012(42):41–49.

10. Musser GG, Carleton MD: Superfamily Muroidea. In Mammal Species of the

World: A Taxonomic and Geographic Reference. Edited by Wilson DE, Reeder

DM. Baltimore: Johns Hopkins University Press; 2005:894–1531.

11. Veyrunes F, Britton-Davidian J, Robinson TJ, Calvet E, Denys C, Chevret P:

Molecular phylogeny of the African pygmy mice, subgenus Nannomys

(Rodentia, Murinae, Mus): implications for chromosomal evolution. Mol

Phylogenet Evol 2005, 36:358–369.

12. Kan Kouassi S, Nicolas V, Aniskine V, Lalis A, Cruaud C, Couloux A, Colyn M,

Dosso M, Koivogui L, Verheyen E, Akoua-Koffi C, Denys C: Taxonomy and

biogeography of the African Pygmy mice, Subgenus Nannomys (Roden-

tia, Murinae, Mus) in Ivory Coast and Guinea (West Africa). Mammalia

2008, 72:237–252.

13. Fichet-Calvet E, Audenaert L, Barrière P, Verheyen E: Diversity, dynamics

and reproduction in a community of small mammals in Upper Guinea,

with emphasis on pygmy mice ecology. Afr J Ecol 2009, 48:600–614.

14. Lamb J, Downs S, Eiseb S, Taylor PJ: Increased geographic sampling

reveals considerable new genetic diversity in morphologically

conservative African Pygmy Mice (Genus Mus; Subgenus Nannomys).

Mamm Biol 2014, 79:24–35.

15. Matthey R: L'eventail robertsonien chez les Mus (Leggada) africains du

groupe minutoides-musculoides. Rev Suisse Zool 1970, 77(3):625–629.

16. Jotterand M: Un nouveau système polymorphe chez une nouvelle

espèce de Leggada (Mus goundae Petter). Experientia 1970, 26:1360–1361.

17. Veyrunes F, Catalan J, Sicard B, Robinson TJ, Duplantier JM, Granjon L,

Dobigny G, Britton-Davidian J: Autosome and sex chromosome diversity

among the African pygmy mice, subgenus Nannomys. Chromosome Res

2004, 12:369–382.

18. Veyrunes F, Watson J, Robinson TJ, Britton-Davidian J: Accumulation of rare

sex chromosome rearrangements in the African pygmy mouse, Mus

(Nannomys) minutoides: a whole-arm reciprocal translocation (WART)

involving a X-autosome fusion. Chromosome Res 2007, 15:223–230.

19. Veyrunes F, Chevret P, Catalan J, Castiglia R, Watson J, Dobigny G, Robinson

TJ, Britton-Davidian J: A novel sex determination system in a close relative

of the house mouse. Proc R Soc B 2010, 277:1049–1056.


79


20. Lecompte E, ter Meulen J, Emonet S, Daffis S, Charrel RN: Genetic identificationof Kodoko virus, a novel arenavirus of the African pigmy mouse (Mus

Nannomys minutoides) in West Africa. Virology 2007, 364:178–183.21. de Bellocq JG, Borremans B, Katakweba A, Makundi R, Baird SJE, Becker-Ziaja

B, Günther S, Leirs L: Sympatric occurrence of 3 arenaviruses, Tanzania.

Emerg Infect Dis 2010, 16(4):692–695.22. Coulibaly-N’Golo D, Allali B, Kan Kouassi S, Fichet-Calvet E, Becker-Ziaja B,

Rieger T, Ölschläger S, Dosso H, Denys C, ter Meulen J, Akoua-Koffi C,Günther S: Novel arenavirus sequences in Hylomyscus sp. and Mus

(Nannomys) setulosus from Cote d'Ivoire: Implications for evolution of

arenaviruses in Africa. PLoS One 2011, 6:e20893.23. Kronmann KC, Nimo-Paintsil S, Guirguis F, Kronmann LC, Bonney K, Obiri-Danso

K, Ampofo W, Fichet-Calvet E: Two novel arenaviruses detected in Pygmy

mice, Ghana. Emerg Infect Dis 2013, 19(11):1832–1835.24. Winkler AJ: Neogene paleobiogeography and East African

paleoenvironments: contributions from the Tugen Hills rodents and

lagomorphs. J Hum Evol 2002, 42:237–256.25. Lecompte E, Granjon L, Peterhans JK, Denys C: Cytochrome b-based

phylogeny of the Praomys group (Rodentia, Murinae): a new African

radiation? C R Biol 2002, 325(7):827–840.26. Stanhope MJ, Czelusniak J, Si JS, Nickeson J, Goodman M: A molecular

perspective on mammalian evolution from the gene encoding

interphotoreceptor retinoid binding protein, with convincing evidence

for bat monophyly. Mol Phylogenet Evol 1992, 1:148–160.27. Galan M, Pagès M, Cosson JF: Next-generation sequencing for rodent

barcoding: species identification from fresh, degraded and

environmental samples. PLoS One 2012, 7(11):e48374.28. Stamatakis A: RAxML Version 8: A tool for phylogenetic analysis and post-

analysis of large phylogenies. Bioinformatics 2014, 30(9):1312–1313.29. Ronquist F, Huelsenbeck JP: MrBayes 3: Bayesian phylogenetic inference

under mixed models. Bioinformatics 2003, 19:1572–1574.30. Rambaut A, Drummond AJ: Tracer v1.5. 2007 [http://beast.bio.ed.ac.uk/Tracer]31. Librado P, Rozas J: DnaSP v5: a software for comprehensive analysis of

DNA polymorphism data. Bioinformatics 2009, 25:1451–1452.32. Pons J, Barraclough TG, Gomez-Zurita J, Cardoso A, Duran DP, Hazell S,

Kamoun S, Sumlin WD, Vogler AP: Sequence-based species delimitation for

the DNA taxonomy of undescribed insects. Syst Biol 2006, 55(4):595–609.33. Powel JR: Accounting for uncertainty in species delineation during the

analysis of environmental DNA sequence data. Methods Ecol Evol 2012,3:1–11.

34. Fujisawa T, Barraclough TG: Delimiting species using single-locus data and

the Generalized Mixed Yule Coalescent approach: A revised method and

evaluation on simulated data sets. Syst Biol 2013, 62(5):707–724.35. Drummond AJ, Suchard MA, Xie D, Rambaut A: Bayesian phylogenetics

with BEAUti and the BEAST 1.7. Mol Biol Evol 2012, 29:1969–1973.36. Tamura K, Peterson D, Peterson N, Stecher G, Nei M, Kumar S: MEGA5:

Molecular evolutionary genetics analysis using maximum likelihood,

evolutionary distance, and maximum parsimony methods. Mol Biol Evol

2011, 28:2731–2739.37. Baker RJ, Bradley RD: Speciation in mammals and the Genetic Species

Concept. J Mammal 2006, 87:643–662.38. Lecompte E, Aplin K, Denys C, Catzeflis F, Chades M, Chevret P: Phylogeny

and biogeography of African Murinae based on mitochondrial and

nuclear gene sequences with a new tribal classification of the subfamily.

BMC Evol Biol 2008, 8:e199.39. Schenk JJ, Rowe KC, Steppan SJ: Ecological opportunity and incumbency

in the diversification of repeated continental colonizations by muroid

rodents. Syst Biol 2013, 62:837–864.40. Jacobs LL, Flynn LJ: Of mice… again: the Siwalik rodent record, murine

distribution, and molecular clocks. In Interpreting the past: essays on

human, primate and mammal evolution. Edited by Lieberman D, Smith R,Kelley J. Leiden: Brill Academic Publishers; 2005:63–80.

41. Denys C: Deux nouvelles espèces d’Aethomys (Rodentia, Muridae) à

Langebaanweg (Pliocène, Afrique du Sud): Implications phylogénétiques.

Annales de Paléontologie 1990, 76:41–69.42. Ambrose SH, Bell CJ, Bernor RL, Boisserie J-R, Darwent CM, Degusta D, Deino

A, Garcia N, Haile-Selassie Y, Head JJ, Howell FC, Kyule MD, Manthi FK,Mathu EM, Nyamai CM, Saegusa H, Stidham TA, Williams MAJ, Hlusko LJ:The paleoecology and paleogeographic context of Lemudong'o locality

1, a Late Miocene terrestrial fossil site in southern Kenya. Kirtlandia 2007,56:38–52.

43. Mein P, Pickford M, Senut B: Late Miocene micromammals from the

Harasib karst deposits, Namibia: Part 2b. Cricetomyidae, Dendromuridae

and Muridae, with an addendum on the Myocricetodontinae. Commun

Geological Surv Namibia 2004, 13:43–61.44. Fabre PH, Hautier L, Dimitrov D, Douzery EJP: A glimpse on the pattern

of rodent diversification: a phylogenetic approach. BMC Evol Biol 2012,12:e88.

45. Lemey P, Rambaut A, Drummond AJ, Suchard MA: Bayesianphylogeography finds its roots. PLoS Comput Biol 2009, 5:e1000520.

46. Ree RH, Smith SA: Maximum likelihood inference of geographic range

evolution by dispersal, local extinction, and cladogenesis. Syst Biol 2008,57(1):4–14.

47. Ree RH, Moore BR, Webb CO, Donoghue MJ: A likelihood framework for

inferring the evolution of geographic range in phylogenetic trees.

Evolution 2005, 59(11):2299–2311.48. R Core Team: R: A language and environment for statistical computing. Vienna,

Austria: R Foundation for Statistical Computing; 2013 [http://www.R-project.org/]49. Galtier N, Gouy M, Gautier C: SEAVIEW and PHYLO_WIN: two graphic tools

for sequence alignment and molecular phylogeny. Comput Appl Biosci

1996, 12:543–548.50. Yalden DW: Small mammals of the Bale Mountains, Ethiopia. Afr J Ecol

1988, 26:281–294.51. Lavrenchenko LA: The mammals of the isolated Harenna Forest (southern

Ethiopia): structure and history of the fauna. In Isolated Vertebrate

Communities in the Tropics, Proc. 4th Int. Symp. Bonn, Bonn Zool Monogr.Edited by Rheinwald G. ; 2000. 46:223-231.

52. Aniskin VM, Lavrenchenko LA, Varshavskii AA, Milishnikov AN: Karyotypesand cytogenetic differentiation of two African mouse species of genus

Mus (Rodentia, Muridae). Rus J Genet 1998, 34(1):80–85.53. Yalden DW, Largen MJ, Kock D, Hillman JC: Catalogue of the mammals of

Ethiopia and Eritrea. 7. Revised checklist, zoogeography and

conservation. Trop Zool 1996, 9:73–164.54. Dieterlen F, Happold DCD: Mus triton Grey-bellied Pygmy Mouse. In

Mammals of Africa, Volume III Rodents, hares and rabbits. Edited by HappoldDCD. London and New York: Bloomsbury Publishing; 2013:497–499.

55. Yalden DW, Largen MJ: The endemic mammals of Ethiopia. Mamm Rev

1992, 22:115–150.56. Vermeiren L, Verheyen WN: Notes sur les Leggada de Lamto, Cote d'Ivoire,

avec la description de Leggada baoulei sp. n. (Mammalia, Muridae). Revue

de Zoologie Africaine 1980, 94:570–590.57. Petter F: Les souris africaines du groupe sorella (Rongeurs, Muridés).

Mammalia 1981, 45(3):313–320.58. Mboumba JF, Deleporte P, Colyn M, Nicolas V: Phylogeography of Mus

(Nannomys) minutoides (Rodentia, Muridae) in West Central African

savannahs: singular vicariance in neighbouring populations. J Zool Syst

Evol Res 2011, 49(1):77–85.59. Brouat C, Tatard C, Bâ K, Cosson JF, Dobigny G, Fichet-Calvet E, Granjon L,

Lecompte E, Loiseau A, Mouline K, Duplantier JM: Phylogeography of the

Guinea Multimammate Mouse (Mastomys erythroleucus): a case study for

Sahelian species in West Africa. J Biogeogr 2009, 36:2237–2250.60. Bryja J, Patzenhauerová H, Granjon L, Dobigny G, Konečný A, Duplantier JM,

Gauthier P, Colyn M, Durnez L, Lalis A, Nicolas V: Plio-Pleistocene history of

West African Sudanian savanna and the phylogeography of the Praomys

daltoni complex (Rodentia): the environment/geography/genetic

interplay. Mol Ecol 2010, 19:4783–4799.61. Dobigny G, Tatard C, Gauthier P, Ba K, Duplantier JM, Granjon L, Kergoat GJ:

Mitochondrial and nuclear genes-based phylogeography of Arvicanthis

niloticus (Murinae) and sub-Saharan open habitats Pleistocene history.

PLoS One 2013, 8(11):e77815.62. Colangelo P, Corti M, Verheyen E, Annesi F, Oguge N, Makundi RH,

Verheyen W: Mitochondrial phylogeny reveals differential modes of

chromosomal evolution in the genus Tatera (Rodentia: Gerbillinae) in

Africa. Mol Phylogenet Evol 2005, 35:556–568.63. McDonough MM, Sotero-Caio CG, Ferguson AW, Lewis PJ, Tswiio M,

Thies ML: Mitochondrial DNA and karyotypic data confirm the presence

of Mus indutus and Mus minutoides (Mammalia, Rodentia, Muridae,

Nannomys) in Botswana. Zookeys 2013, 359:35–51.64. Chevret P, Robinson TJ, Perez J, Veyrunes F, Britton-Davidian J: A

phylogeographic survey of the Pygmy mouse Mus minutoides in

South Africa: Taxonomic and karyotypic inference from cytochrome

b sequences of museum specimens. PLoS One 2014, 9(6):e98499.


80


65. Dobigny G, Tatard C, Kane M, Gauthier P, Brouat C, Ba K, Duplantier JM: A

cytotaxonomic and DNA-based survey of rodents from Northern

Cameroon and Western Chad. Mammal Biol 2011, 76(4):417–427.

66. Winkler AJ, Denys C, Avery MD: Rodentia. In Cenozoic Mammals of Africa.

Edited by Werdelin L, Sanders WJ. Berkeley: University of California Press;

2010:263–304.

67. Isaac JB, Mallet J, Mace GM: Taxonomic inflation: its influence on

macroecology and conservation. Trends Ecol Evol 2004, 19:464–469.

68. Lavrenchenko LA, Verheyen WN, Verheyen E, Hulselmans J, Leirs H:

Morphometric and genetic study of Ethiopian Lophuromys

flavopunctatus Thomas, 1888 species complex with description of three

new 70-chromosomal species (Muridae, Rodentia). Bulletin de l’Insitut

Royal des Sciences Naturelles de Belgique, Biologie 2007, 77:77–117.

69. Berger SA, Krompass D, Stamatakis A: Performance, accuracy, and web

server for evolutionary placement of short sequence reads under

maximum likelihood. Syst Biol 2011, 60:291–302.

70. Skinner JD, Chimimba CT: The Mammals of the Southern African Subregion.

3rd edition. Cambridge, UK: Cambridge University Press; 2005.

71. Monadjem A: Mus setzeri Setzer’s Pygmy Mouse. In Mammals of Africa,

Volume III Rodents, hares and rabbits. Edited by Happold DCD. London and

New York: Bloomsbury Publishing; 2013:493–494.

72. Vermeiren L, Verheyen WN: Additional data on Mus setzeri Petter

(Mammalia, Muridae). Annales Musée Royal de l'Afrique Centrale, ser 8

(Sciences Zoologiques) 1983, 237:137–141.

73. Missoup AD, Nicolas V, Wendelen W, Keming E, Bilong Bilong CF, Couloux A,

Atanga E, Hutterer R, Denys C: Systematics and diversification of Praomys

species (Rodentia: Muridae) endemic to the Cameroon Volcanic Line (West

Central Africa). Zool Scr 2012, 41:327–345.

74. Demos TC, Agwanda B, Hickerson MJ: Integrative taxonomy within the

Hylomyscus denniae complex (Rodentia: Muridae) and a new species

from Kenya. J Mamm 2014, 95(1):1–15.

75. Verheyen WN, Hulselmans JLJ, Dierckx T, Mulungu L, Verheyen E, Corti M,

Kerbis JP, Leirs H: Craniometric,cytogenetic and genetic characterization

of the Kilimanjaro Lophuromys aquilus True 1892 population, its

implications for the taxonomy of the Lophuromys flavopunctatus species

complex and description of five new taxa. (Rodentia-Muridae-Africa).

Bulletin de l’Insitut Royal des Sciences Naturelles de Belgique, Biologie 2007,

77:23–75.

76. Colangelo P, Verheyen E, Leirs H, Tatard C, Denys C, Dobigny G, Duplantier

JM, Brouat C, Granjon L, Lecompte E: A mitochondrial phylogeographic

scenario for the most widespread African rodent, Mastomys natalensis.

Biol J Linn Soc 2013, 108(4):901–916.

77. Potts R: Hominin evolution in settings of strong environmental

variability. Quat Sci Rev 2013, 73:1–13.

78. Alvarado-Serrano DF, Knowles LL: Ecological niche models in

phylogeographic studies: applications, advances and precautions. Mol

Ecol Resour 2014, 14:233–248.

79. Šmíd J, Carranza S, Kratochvíl L, Gvoždík V, Nasher AK, Moravec J: Out of

Arabia: A complex biogeographic history of multiple vicariance and

dispersal events in the gecko genus Hemidactylus (Reptilia: Gekkonidae).

PLoS One 2013, 8(5):e64018.

80. Portik DM, Papenfuss TJ: Monitors cross the Red Sea: The biogeographic

history of Varanus yemenensis. Mol Phylogenet Evol 2012, 62:561–565.

81. Winney BJ, Hammond RL, Macasero W, Flores B, Boug A, Biquand V, Biquand S,

Bruford MW: Crossing the Red Sea: phylogeography of the hamadryas

baboon, Papio hamadryas hamadryas. Mol Ecol 2004, 13:2819–2927.

82. Bosworth W, Huchon P, McClay K: The red sea and gulf of aden basins.

J Afr Earth Sci 2005, 43:334–378.

83. Fernandes CA, Rohling EJ, Siddall M: Absence of post-Miocene Red Sea

land bridges: biogeographic implications. J Biogeogr 2006, 33:961–966.

84. Catzeflis FM, Denys C: The African Nannomys (Muridae): an early offshoot

from the Mus lineage - Evidence from scnDNA hybridization experiments

and compared morphology. Isr J Zool 1992, 38:219–231.

85. Denys C: Of mice and men: Evolution in East and South Africa during

Plio-Pleistocene times. In African Biogeography, Climate Change, and

Human Evolution. Edited by Bromage TG, Schrenk F. Oxford: Oxford

University Press; 1999:226–252.

86. Trauth MH, Maslin MA, Deino AL, Strecker MR, Bergner AGN, Dühnforth M:

High- and low-latitude forcing of Plio-Pleistocene East African climate

and human evolution. J Hum Evol 2007, 53:475–486.

87. de Menocal PB: African climate change and faunal evolution during the

Pliocene-Pleistocene. Earth Planet Sc Lett 2004, 220:3–24.

88. Zachos J, Pagani M, Sloan L, Thomas E, Billups K: Trends, rhythms, and

abberations in global climate 65 Ma to present. Science 2001, 292:686–693.

89. Schultz S, Maslin M: Early human speciation, brain expansion and dispersal

influenced by African climate pulses. PLoS One 2013, 8(10):e76750.

90. Veyrunes F, Catalan J, Tatard C, Cellier-Holzem E, Watson J, Chevret P,

Robinson TJ, Britton-Davidian J: Mitochondrial and chromosomal insights

into karyotypic evolution of the pygmy mouse, Mus minutoides, in South

Africa. Chromosome Res 2010, 18:563–574.

91. Veyrunes F, Perez J, Paintsil SNC, Fichet-Calvet E, Britton-Davidian J: Insights

into the evolutionary history of the X-linked sex reversal mutation in

Mus minutoides: clues from sequence analyses of the Y-linked Sry gene.

Sex Dev 2013, 7(5):244–252.

92. Matthey R: Nouvelles contributions à la cytogénétique des Mus africaines

du sous-genre Leggada. Experientia 1966, 22(6):400–401.

93. Nichol ST, Arikawa J, Kawaoka Y: Emerging viral diseases. Proc Natl Acad Sci

U S A 2000, 2000(97):12411–12412.

94. Günther S, Lenz O: Lassa virus. Crit Rev Clin Lab Sci 2004, 41:339–390.

95. Hugot JP, Gonzales JP, Denys C: Evolution of the Old World Arenaviridae and

their rodent hosts: generalized host-transfer or association by descent?

Infect Genet Evol 2001, 1:13–20.

Submit your next manuscript to BioMed Centraland take full advantage of:

• Convenient online submission

• Thorough peer review

• No space constraints or color figure charges

• Immediate publication on acceptance

• Inclusion in PubMed, CAS, Scopus and Google Scholar

• Research which is freely available for redistribution

Submit your manuscript at www.biomedcentral.com/submit


81


Evolutionary history and species

diversity of African Pouched Mice of the

genus Saccostomus (Rodentia:

Nesomyidae)

Mikula, O., umbera, R., Aghová, T., Mbau, J.S., Bryja, J.,

2016. Zoologica Scripta, 45(6), 595-617.

https://doi.org/10.1111/zsc.12179

Study design: OM, JB

Collecting material: RS, TA, JM, JB

Laboratory analysis: TA

Data analysis: OM

Writing: OM, JB


Paper

II

Paper

II

83


Evolutionary history and species diversity of African pouched

mice (Rodentia: Nesomyidae: Saccostomus)OND�REJ MIKULA, RADIM �SUMBERA, TATIANA AGHOV A, JUDITH S. MBAU, ABDUL S. KATAKWEBA,CHRISTOPHER A. SABUNI & JOSEF BRYJA

Submitted: 21 October 2015Accepted: 11 February 2016doi:10.1111/zsc.12179

Mikula, O., �Sumbera, R., Aghov a, T., Mbau, J.S., Katakweba, A.S., Sabuni, C.A., Bryja, J.

(2016). Evolutionary history and species diversity of African pouched mice (Rodentia:

Nesomyidae: Saccostomus). — Zoologica Scripta, 45, 595–617.

We explore diversity of African pouched mice, genus Saccostomus (Rodentia, Nesomyidae),

by sampling molecular and morphological variation across their continental-scale distribu-

tion in southern and eastern African savannahs and woodlands. Both mitochondrial (cy-

tochrome b) and nuclear DNA (IRBP, RAG1) as well as skull morphology confirm the

distinction between two recognized species, S. campestris and S. mearnsi, with disjunct distri-

bution in the Zambezian and Somali–Maasai bioregions, respectively. Molecular dating sug-

gests the divergence of these taxa occurred in the Early Pliocene, 3.9 Ma before present,

whereas the deepest divergences within each of them are only as old as 2.0 Ma for

S. mearnsi and 1.4 Ma for S. campestris. Based on cytochrome b phylogeny, we defined five

clades (three within S. campestris, two in S. mearnsi) whose species status was considered in

the light of nuclear DNA markers and morphology. We conclude that S. campestris group

consists of two subspecies S. campestris campestris (Peters, 1846; comprising two cytochrome

b clades) and S. campestris mashonae (de Winton, 1897) that are moderately differentiated,

albeit distinct in IRBP and skull form. They likely hybridize to a limited extent along the

Kafue–Zambezi Rivers. Saccostomus mearnsi group consists of two species, S. mearnsi (Heller,

1910) and S. umbriventer (Miller, 1910), that are markedly differentiated in both nuclear

markers and skull form and may possibly co-occur in south-western Kenya and north-east-

ern Tanzania. Analysis of historical demography suggests both subspecies of S. campestris

experienced population expansion dated to the Last Glacial. In the present range of S. cam-

pestris group, the distribution modelling suggests a moderate fragmentation of suitable habi-

tats during the last glacial cycle, whereas in the range of S. mearnsi group it predicts

substantial shifts of its occurrence in the same period.

Corresponding author: Ond�rej Mikula, Institute of Animal Physiology and Genetics, Academy of

Sciences of the Czech Republic, Veve�r ı 97, 60200 Brno, Czech Republic. E-mail: onmikula@gmail.

com

Ond�rej Mikula, Institute of Vertebrate Biology, Academy of Sciences of the Czech Republic, Brno,

Czech Republic and Institute of Animal Physiology and Genetics, Academy of Sciences of the Czech

Republic, Brno, Czech Republic. E-mail: [email protected]

Radim �Sumbera, Department of Zoology, Faculty of Science, University of South Bohemia, �Cesk e

Bud�ejovice, Czech Republic. E-mail: [email protected]

Tatiana Aghov a, Institute of Vertebrate Biology, Academy of Sciences of the Czech Republic, Brno,

Czech Republic and Department of Botany and Zoology, Faculty of Science, Masaryk University,

Brno, Czech Republic. E-mail: [email protected]

Judith S. Mbau, College of Agriculture and Veterinary Sciences, University of Nairobi, Nairobi,

Kenya.

Abdul S. Katakweba, Pest Management Center, Sokoine University of Agriculture, Morogoro,

Tanzania. E-mail: [email protected]

Christopher A. Sabuni, Pest Management Center, Sokoine University of Agriculture, Morogoro,

Tanzania. E-mail: [email protected]

Josef Bryja, Institute of Vertebrate Biology, Academy of Sciences of the Czech Republic, Brno, Czech

Republic and Department of Botany and Zoology, Faculty of Science, Masaryk University, Brno,

Czech Republic. E-mail: [email protected]

ª 2016 Royal Swedish Academy of Sciences, 45, 6, November 2016, pp 595–617 595

Zoologica Scripta

84


Introduction

Following the spread of molecular-genetic toolkit in the

last two decades, there has been a dramatic increase in

number of systematic surveys of African rodents, shrews

and bats often uncovering a high degree of spatially struc-

tured genetic variation (Corti et al. 1999; Fadda et al. 2001;

Faulkes et al. 2004; Colangelo et al. 2007; Nicolas et al.

2009; Taylor et al. 2012; Demos et al. 2014; Bryja et al.

2014 and others). Nevertheless, many taxa are still left

underexplored in spite of their wide distributions and eco-

logical importance.

African pouched mice of the genus Saccostomus are

small- to medium-sized slowly moving rodents, which are

locally abundant in highly seasonal environments of grass-

lands, savannahs, scrublands and woodland of southern

and eastern Africa (Galster et al. 2007). They are solitary

burrowing rodents with nocturnal activity. These rodents

are ecological opportunists, primarily granivorous, with

high-turnover population dynamics. During suitable peri-

ods with abundant food supply, they accumulate fat; when

food is scarce during dry season, they enter torpor up to

five hours long (Ellison 1993; Skinner & Chimimba

2005). The genus is morphologically well defined by

heavily built bodies with short limbs and short tails as

well as by the presence of cheek pouches (Carleton &

Musser 1984; Ryan 1989).

The genus Saccostomus belongs to the family Nesomyidae

and within it to subfamily Cricetomyinae together with its

sister clade comprising genera Beamys and Cricetomys, which

is supported by analysis of four nuclear genes (Schenk et al.

2013). These three genera also have the same molar cusp

configuration lacking t1 cusp of the first upper molar and

t1 and t3 cusps of the second upper molar (Roberts 1951;

Petter 1966; de Graaf 1981). The genus Saccostomus

appeared in an African rodent radiation during the second

half of Miocene, ca. 5–12 Ma (Winkler et al. 2010; Schenk

et al. 2013), with the oldest well-dated fossil record at

6 Ma (Mein & Pickford 2006), where Ma is for a million

of years (mega-annum). Its monophyly was also unambigu-

ously supported also by cytochrome b data (Corti et al.

2004). Hubert (1978) reviewed the taxonomy of the genus

using morphological and karyological data and assigned all

available specimens to two species: Saccostomus campestris,

which is small (body weight 40–85 g) with white ventral

pelage and occurs in southern Africa, and Saccostomus

mearnsi, a larger species (body weight 48–121 g) having

grey belly and being distributed in the Somali-Maasai

region (Elison et al. 1993; Keesing 1998b). This distinction

was later confirmed by an extensive morphological revision

by Denys (1988) and by the first phylogenetic analysis of

the genus (Corti et al. 2004). The latter two studies,

however, documented remarkable morphological, chromo-

somal and molecular variation within both species of

Hubert (1978), which led Corti et al. (2004) to consider

them as species complexes. Nevertheless, large karyotypic

differences (Gordon & Rautenbach 1980; Corti et al. 2004)

are not necessarily linked to reproductive isolation. Mapu-

tla et al. (2011) cross-bred two South African populations

of S. campestris with extreme diploid numbers (2n = 32 vs.

2n = 46) and did not find any indication of impaired viabil-

ity or fertility up to the third generation of hybrids. These

authors thus challenged the idea that S. campestris may rep-

resent several valid biological species.

From a taxonomic point of view, there is no doubt that

the valid names of Hubert’s species are S. campestris and

S. mearnsi, but if we accepted the idea that both of them

include two or more species, the situation would become

unclear. Saccostomus campestris has ten junior synonyms, and

S. mearnsi has three synonyms (Musser & Carleton 2005).

However, the complete type material has never been criti-

cally examined. Most comprehensive comments on the type

material, although without detailed measurements, were

provided by Corti et al. (2004), who speculated about a new

species from the S. mearnsi group in northern Tanzania.

For the S. campestris group, the situation is much more

complex and although Corti et al. (2004) provided new

karyotypes, first sequence data and examined some type

material, they did not provide any clear taxonomic conclu-

sion. Therefore, exact delimitation of species is still lacking.

From a biogeographical point of view, the genus Saccosto-

mus represents an excellent model to study long-term

impacts of contractions and expansions of African open dry

habitats on the evolution of small mammals since the Mio-

cene. African pouched mice are found in various types of

dry habitats over a large part of the continent and have

quite a rich fossil record (Winkler et al. 2010). Further-

more, because of its conspicuousness and relatively easy

trapability, specimens of Saccostomus are well represented in

natural history collections. However, testing any biogeo-

graphical hypothesis requires a reliable phylogenetic analy-

sis, ideally based on the combination of nuclear and

mitochondrial markers, and detailed distributional data of

main genetic lineages. Unfortunately, our current phyloge-

netic understanding of this genus is based only on the sin-

gle molecular-genetic study of Corti et al. (2004) that

analysed limited sample of 12 localities.

In this study, we conducted a comprehensive revision of

the genus based on sampling covering the whole distribu-

tion range and combining both genetic and morphological

data. The current distribution and phylogenetic diversity of

the genus is interpreted in terms of past changes in the

distribution of African biomes.

596 ª 2016 Royal Swedish Academy of Sciences, 45, 6, November 2016, pp 595–617

Diversity in rodent genus Saccostomus � O. Mikula et al.

85


Material and methods

Sampling

Molecular and morphological data were collected from most

of the known distributional range of the genus. Maps of our

sampling localities are provided in the Results section, and

complete list of material is available in Appendix S1.

Material collected recently by our team in the field was

supplemented by material from American Museum of Natu-

ral History (AMNH, New York), Natural History Museum

(BMNH, London), Magyar Term�eszettudom�anyi M�uzeum

(HNHM, Budapest), Institut des Sciences de l’Evolution

(ISEM, Montpellier), Livingstone Museum (LM, Living-

stone), Museum of Comparative Zoology (MCZ, Boston),

Museum of Vertebrate Zoology (MVZ, Berkeley), Mus�eum

national d’Histoire naturelle (MNHN, Paris), Institut royal

des Sciences naturelles de Belgique (RBINS, Brussel),

Mus�ee royal de l’Afrique centrale (RMCA, Tervuren),

Senckenberg Museum (SMF, Frankfurt), Staatliches

Museum f€ur Naturkunde (SMNS, Stuttgart), Ditsong

National Museum of Natural History – former Transvaal

Museum (TM, Pretoria), Texas Tech University (TTU,

Lubbock), Smithsonian Institution – National Museum of

Natural History (USNM, Washington, D.C.) and Museum

f€ur Naturkunde (ZMB, Berlin).

Molecular data

Fresh tissues were stored in 96% ethanol, and DNA was

extracted using the DNeasy Blood & Tissue Kit (Qiagen).

The complete mitochondrial cytochrome b gene was

amplified by polymerase chain reaction (PCR) using pri-

mers L14723 and H15915 (Lecompte et al. 2002). Parts of

nuclear genes encoding the interphotoreceptor-binding

protein (IRBP) and recombination activating gene (RAG1)

were amplified in a subset of individuals by the primers

IRBP1531 and IRBP217 (Stanhope et al. 1992) and RAG1-

F1705 and RAG1-R2951 (Teeling et al. 2000). For PCR

conditions, see Bryja et al. (2014). Bidirectional Sanger

sequencing of PCR products was conducted using Big-

DyeTM chemistry (Applied Biosystems, Foster City, CA,

USA). Genetic data from fresh material were comple-

mented by partial cytochrome b sequences from museum

samples (mostly dry skins; for more details, see

Appendix S1). All museum samples were handled in a spe-

cialized laboratory of Institute of Vertebrate Biology ASCR

in Studenec, designed for work with degraded DNA to

prevent contamination by samples with high quantity of

DNA or PCR products. Partial cytochrome b sequences

were obtained by 454 pyrosequencing on GS Junior

according to mini-barcode protocol described by Galan

et al. (2012). The main advantage of this approach in anal-

ysis of museum samples is that it allows separation of indi-

vidual sequences in samples contaminated by distantly

related organisms (e.g. contamination by human DNA),

which is not possible through Sanger sequencing. The

obtained sequences were analysed using the |SE|S|AM|E|

BARCODE software (Piry et al. 2012).

For cytochrome b, we got 114 long (>800 bp) and 37

shorter sequences from Sanger sequencing and 20 short

sequences from pyrosequencing. In total, we used 171

sequences, 120 newly obtained and 51 taken from Gen-

Bank. Ultimately, phylogenetic inference was based on 97

unique haplotypes found among 114 long sequences,

whereas the rest of sequence data was used only for assign-

ment of individuals to major clades. For IRBP, we had 31

sequences (30 new) including seven heterozygous individu-

als phased out into different parental haplotypes using the

PHASE algorithm (Stephens et al. 2001) as implemented in

DNAsp 5.10.1 (Librado & Rozas 2009). Among the result-

ing 38 sequences, we found 22 unique haplotypes. For

RAG1, we obtained 10 sequences, 4 of them heterozygous

that were processed as for IRBP. The resulting 14

sequences were all unique. Newly obtained sequences used

in phylogenetic reconstructions were entered into GenBank

(accession numbers KM347745–KM347882, KP050321,

KT873816–KT873842, KT894107; see also Appendix S1).

Phylogeny and candidate species

First, we inferred phylogenetic trees separately for each

gene using Bayesian inference as implemented in MrBayes

3.2.2 (Ronquist et al. 2012). Substitution models were

selected via FindModel web application (http://hiv.

lanl.gov/content/sequence/findmodel/findmodel.html) using

algorithms described in Posada & Crandall (1998). In cyto-

chrome b, the best supported model was general time

reversible model (GTR; Tavar�e 1986), in the nuclear genes

Hasegawa–Kishino–Yano model (HKY; Hasegawa et al.

1985), both with gamma-distributed variation in substitu-

tion rates, discretized here into four categories. Variation

in substitution rates among branches was unconstrained,

and the root position was inferred by including up to three

outgroup taxa (Beamys hindei, Cricetomys ansorgei and Crice-

tomys emini, see Appendix S1 for details). Metropolis cou-

pling with three heated and one cold chain was applied to

speed-up Markov chain Monte Carlo (MCMC) conver-

gence. Each chain consisted of 2 million generations, and

the cold chain was sampled every 500 generations after dis-

carding the first 25% as burn-in. Two independent runs

were performed to check the convergence on common pos-

terior probability maximum.

Following overall phylogenetic reconstruction, cyto-

chrome b tree was partitioned into five major clades to

establish a working hypothesis about species boundaries

within the genus (see the Results). We pinpointed those

that were reciprocally monophyletic and whose average


O. Mikula et al. � Diversity in rodent genus Saccostomus

86


sequence divergence from the sister lineage was at least

5%, as measured by the sum of branch lengths joining any

two individuals in the estimated phylogenetic tree. Short

cytochrome b sequences, either from Sanger sequencing or

from 454 pyrosequencing, were placed into the estimated

phylogenetic tree using evolutionary placement algorithm

(EPA; Berger et al. 2011), run under RAXML version 8.0.9

(Stamatakis 2014). Support for placement into particular

branches was expressed by Akaike weights (∑Aw; e.g. John-

son & Omland 2004), and the sum of these weights across

branches within particular clade represented probability of

belonging to it.

Divergence dating and historical demography

To date putative speciation events, we employed Bayesian

analysis based on relaxed molecular clock with uncorrelated

lognormal distribution of evolutionary rates as imple-

mented in the BEAST 2.2.1 software (Bouckaert et al. 2014).

Here, we analysed a data set consisting of cytochrome b,

IRBP and RAG1 sequences representing five major clades

of Saccostomus as well as nine species from genera Beamys,

Cricetomys, Dendromus and Steatomys (Appendix S1), that

were consistently found as the closest relatives of Saccosto-

mus in previous analyses (Jansa & Weksler 2004; Steppan

et al. 2004; Fabre et al. 2012; Schenk et al. 2013). For all

markers, we assumed a single common tree (with birth–

death prior; Gernhard 2008) but different evolutionary

rates and substitution model parameters. The cytochrome

b alignment used here did not allow estimation of GTR

parameters (lack of convergence in spite of large number

of MCMC generations), and thus, we used less complex

HKY model. We also used empirical, instead of estimated,

nucleotide frequencies. The MCMC was run twice with

10 million iterations with trees and model parameters sam-

pled every 5000 iterations. XML file used as an input for

the analysis is available as Appendix S2. Convergence was

checked using Tracer 1.6 (Rambaut et al. 2014), and both

runs were combined in LogCombiner 2.2.1. The maximum

clade credibility tree representing the best consensus of the

inferred posterior distribution was calculated in TreeAnno-

tator 2.2.1 together with medians and 95% highest poste-

rior density (HPD) limits of model parameters.

Time calibration of the tree was based on two fossil con-

straints, namely time to most recent common ancestors

(TMRCAs) of Saccostomus and of the subfamilies Cricetomyi-

nae and Dendromurinae. The subfamilies were constrained

to be reciprocally monophyletic and comprising the genera

Beamys+Cricetomys+Saccostomus and Dendromus+Steatomys,

respectively, in accordance with Musser & Carleton (2005)

as well as the most recent molecular phylogenies (Schenk

et al. 2013). Prior calibration density of Saccostomus

TMRCA was lognormal with l = 1.22 and r = 0.28 which

implies 2.5%, 50% and 97.5% quantiles equal to 2.0, 3.4

and 5.8, respectively, all values in Ma. The minimum

bound is motivated by the earliest fossil record assigned to

S. mearnsi dating from 1.7 to 1.9 Ma (Olduvai Gorge, Tan-

zania; Denys 1992). Back in time, there are numerous

Saccostomus fossils including Early Pliocene specimens with

dentition very similar to the extant species (Denys 1987,

2011; Winkler 1997) and, on the contrary, the Late Mio-

cene (5.8–6.1 Ma) record (Lukeino formation, Kenya; Mein

& Pickford 2006) which is similar to even earlier fossils

(Geraads 2001; Mein et al. 2004) rather than to the extant

Saccostomus. TMRCA of Cricetomyinae and Dendromurinae

was constrained by lognormal distribution with l = 2.40

and r = 0.10, implying 2.5%, 50% and 97.5% quantiles

equal to 9.1, 11.0 and 13.4, respectively. This is based on

fossil records of both subfamilies from Ethiopia (Geraads

2001) and Namibia (Mein et al. 2004) dated by faunal simi-

larities to about 8.7–10.5 Ma. Although the dating is not

firmly based and the proposed identity of the specimens

with extant genera is not universally accepted (Winkler

et al. 2010), it may be assumed that roughly at that time

both subfamilies were already present. On the contrary, at

12.5 Ma (Ngorora formation; Winkler 2002), a single

molar of only general resemblance to these two subfamilies

was found (A. Winkler, pers. comm.)

In addition, we analysed past population dynamics of

two clades with sufficient samples of cytochrome b. For

both of them, we calculated Tajima’s D statistic (Tajima

1989) indicating whether the observed molecular variation

can be explained by neutral evolution under constant popu-

lation size. P values for Tajima’s D were based on beta dis-

tribution approximation (Tajima 1989). Finally, we

modelled the past population dynamics using Bayesian sky-

line plots (BSP; Drummond et al. 2005) as implemented in

BEAST 2.2.1. Here, we used strict molecular clock with prior

distributions of evolutionary rates mimicking posteriors on

the particular terminal branch from the divergence dating

analysis.

The plots of phylogenetic trees and distribution patterns

were created in R 3.2.1 (R Core Team 2015) with a help

of contributed packages APE (Paradis et al. 2004), FIELDS

(Nychka et al. 2015) and MAPTOOLS (Bivand & Lewin-Koh

2014).

Species distributions modelling

To extrapolate the observed distributions to non-sampled

areas as well as to the past, we created species distribution

models separately for S. campestris and S. mearnsi groups.

As a base for the models, we used one degree resolution

grids from the Copenhagen databases of African vertebrates

(Galster et al. 2007) supplemented with all georeferenced

records from the present data set and public databases



87


African Rodentia (Terryn et al. 2007), GBIF (www.

gbif.org) and MaNIS (www.manisnet.org) that were also

projected to the grids. Following this data collection, the

grid cells were considered as either ‘occupied’ (with at least

one presence record) or ‘query’ with no knowledge of

presence/absence and we assumed the particular species

group to be present in all query cells climatically similar to

the occupied ones.

The climate was characterized by 19 bioclimatic vari-

ables (data set BIOCLIM, Table 1) from the WorldClim

database (Hijmans et al. 2005) subsampled to 1° resolution.

We considered current conditions as well as two available

paleoclimatic reconstructions, one for the last glacial

(21 ka; Braconnot et al. 2007) and one for the last inter-

glacial (120–140 ka; Otto-Bliesner et al. 2006). Climatic

variables were restricted in geographical space using a

bounded square that was large enough to include all

peripheral locations of specimen records. Climatic varia-

tion was summarized using principal component analysis

(PCA) with the number of retained components deter-

mined by the scree test optimal coordinate criterion

(Raıche et al. 2013). In both S. campestris and S. mearnsi

group, five components were retained and each cell was

therefore represented by a point in the resulting five-

dimensional climatic space. Regions of this space populated

by the occupied cells were estimated using algorithm

introduced by Blonder et al. (2014). In a query cell, the

presence was predicted if it was located in some of these

regions. The models were built separately for S. campestris

and S. mearnsi groups.

Species distribution modelling was performed with R

packages HYPERVOLUME (Blonder 2015), NFACTORS (Raıche

2010), RASTER (Hijmans 2014) and SPERICH (Lange et al.

2012).

Morphological data

Skull form variation was explored using geometric morpho-

metric analysis. In this framework, the form is described by

a configuration of anatomical landmarks that is alge-

braically decomposable into size and shape components.

Individual form is represented as a vector consisting of log-

arithm of size (estimated as so-called centroid size) and

Procrustes shape coordinates embracing all geometrical

properties of the configuration after standardization for

size, position and orientation (Dryden & Mardia 1998).

Standardization was achieved with generalized Procrustes

analysis (GPA; Rohlf & Slice 1990) by aligning individual

configurations to an iteratively estimated mean configura-

tion. In the present study, skull form was captured by three

landmark sets located on two-dimensional images taken

from dorsal, ventral and lateral views of the skull (Fig. 1).

Following the transformation into size and shape variables,

these three partial forms were combined into a single mul-

tivariate trait by their common PCA. Skull images and raw

morphometric data are accessible through MorphoBank

(O’Leary & Kaufman 2012), project number 7639.

Table 1 Original definitions of 19 bioclimatic variables used in the

species distribution modelling, downloaded from www.world-

clim.org (Hijmans et al. 2005)

BIO1: Annual mean temperature

BIO2: Mean diurnal range of temperatures (mean monthly difference between

maximum temperature and minimum temperature)

BIO3: Isothermality (BIO2/BIO7) (*100)

BIO4: Temperature seasonality (standard deviation*100)

BIO5: Maximum temperature of warmest month

BIO6: Minimum temperature of coldest month

BIO7: Temperature annual range (BIO5 � BIO6)

BIO8: Mean temperature of wettest quarter

BIO9: Mean temperature of driest quarter

BIO10: Mean temperature of warmest quarter

BIO11: Mean temperature of coldest quarter

BIO12: Annual precipitation

BIO13: Precipitation of wettest month

BIO14: Precipitation of driest month

BIO15: Precipitation seasonality (coefficient of variation)

BIO16: Precipitation of wettest quarter

BIO17: Precipitation of driest quarter

BIO18: Precipitation of warmest quarter

BIO19: Precipitation of coldest quarter

● ● ● ● ●

●●

●●

●

●

● ●●

●

●●

●●

●

●

● ●●

●

●●

●

●

●

●●

●● ● ●

●

●

●

●

●

●

●

●

●

●

●● ● ● ●

●

● ●

●

●

●

●●

●

●

●

●●

●

●

●

●●

●

● ●

●

●

●

●●

●

●

●

●●

●

●

●

●●

Fig. 1 Anatomical landmarks recorded on the skull images taken

from dorsal, lateral and ventral view. Sliding semilandmarks are

marked by open circles.


O. Mikula et al. Diversity in rodent genus Saccostomus

88


In total, 299 skulls were included in morphometric anal-

yses. Due to skull damage, some individual configurations

were incomplete. Their shape was imputed in Bayesian

PCA (Oba et al. 2003) by estimating relative positions of

missing landmarks. First, the data were preprocessed by

preliminary GPA of complete configurations and one-by-

one alignment of incomplete configurations to particular

subsets of landmarks of the GPA mean (Arbour & Brown

2014). Then Bayesian PCA was applied with number of

components determined by fivefold cross-validation accord-

ing to Krzanowski (1987). Overall, about 0.5% of landmark

coordinates was imputed this way. Logarithm of centroid

size was calculated for all complete configurations, other-

wise considered missing and imputed by linear model with

log size of partial and whole configurations as a predictor

and response, respectively. To control for ontogenetic vari-

ation, we discarded juvenile specimens (the first of six tooth

wear classes of Denys 1988). The sexes were pooled, as our

sampling was not sufficient to disentangle intersexual and

eco-geographical variation. This is also justified by non-

significant intersexual differences reported in previous stud-

ies, namely in samples from Botswana (Smithers 1971), the

former Transvaal province of South Africa (Rautenbach

1982; Ellison et al. 1993) and the current KwaZulu-Natal

province (Maputla 2008).

Morphological differentiation

To reduce the dimension of data while retaining high sig-

nal-to-noise ratio, we applied between-group PCA, that is

axis rotation based on the covariance matrix of species

means. With five major clades, there is up to four infor-

mative components and visual inspection of the corre-

sponding eigenvalues via scree plot suggested to retain

them all. In this analysis, individuals were classified

according to their cytochrome b sequences, but the strong

spatial pattern of genetic variability allowed us to further

increase sample size by including additional non-

sequenced individuals, if they were surrounded by locali-

ties of the same clade or if they were separated by locali-

ties of single clade from the rest of the particular species

group (see Appendix S3 for details). We avoided, however,

classifying non-sequenced individuals from close vicinity

of assumed contact zones.

Morphological differentiation of five major clades was

assessed with the Gaussian mixture modelling, where each

clade was represented by a single multivariate normal dis-

tribution. As a part of the fitting procedure a discrete set

of constraints imposed on within-group covariance matri-

ces is considered (Bensmail & Celeux 1996). The whole

set consists of 14 different combinations of shape, volume

and orientation constraints, and as special cases, it includes

linear discriminant analysis (equal but otherwise uncon-

strained covariance matrices) and quadratic discriminant

analysis (no constraints at all). Bayesian information crite-

rion (BIC; Schwarz 1978) was used to select the best sup-

ported set of constraints with difference (∆BIC) higher

than six being considered as strong evidence (Kass & Raf-

tery 1995). Fivefold cross-validation was used to estimate

classification success of the estimated model. To correct

for the effect of random partition into folds, the procedure

was repeated 100 times and each individual was classified

according to posterior probabilities averaged across the

repetitions. Distinctiveness of clades was then inferred

from their mutual misclassification rates. Type specimens

were not included in the model fitting even if they could

be classified into some clade. They were classified post hoc

using the selected Gaussian mixture model together with

other skulls of uncertain affinity, that were of interest to

infer distributions of clades in genetically non-sampled

areas.

Finally, we estimated skull form features maximally dis-

tinguishing groups defined by successive splits of the phy-

logenetic tree. For this purpose, we used generalized

partial least-squares (GPLS) regression (Bastien et al.

2005), where predictors were principal components and

binomial response was posterior probability of belonging

to groups defined by the particular split (e.g. S. campestris

vs. S. mearnsi). If two or more clades were included in

any such group (e.g. three in S. campestris group), their

posterior probabilities were summed. The number of

GPLS components (linear combinations of predictors)

optimal for the prediction of posterior probabilities was

evaluated by BIC based on degrees of freedom as a gener-

alized measure of model complexity (Kraemer & Sugiyama

2011).

All morphometric analyses were performed in R using

packages geomorph (Adams & Otarola-Castillo 2013),

PCAMETHODS (Stacklies et al. 2007), MCLUST (Fraley et al.

2012) and PLSRGLM (Bertrand et al. 2014).

Type material

To achieve the most robust taxonomic conclusions, com-

plete type material of the genus was examined. It com-

prised 20 holotypes or syntypes of 15 formally named

species and subspecies whose skulls are sufficiently pre-

served and also 11 paratypes and topotypes (Appendix S1).

Seven of them were juveniles and thus omitted from the

analysis, but fortunately none of these was of crucial

importance from taxonomic point of view.

In addition to the morphometric analysis, DNA was iso-

lated from dry tissue samples of the syntype of S. lapidarius

(ZMB MAM-85437) and paralectotype of S. fuscus (ZMB

MAM-85450) and PCR products of partial cytochrome b

were included in the pyrosequencing run. We attempted



89


the same also with paratype of S. mashonae (BMNH

1895.8.27.9), but with no success.

Results

Phylogeny, delimitation and distribution of candidate species

Maximum clade credibility trees of three analysed genes

are shown in Fig. 2. Cytochrome b tree shows a clear

structure with long branches supporting clusters of much

more closely related sequences. Five major clades were

defined as candidate species and denoted by letters A–E.

They were all monophyletic with posterior probability

(PP) equal to unity and separated from each other by

more than 5% sequence divergence (minimum 8.66%

between B and C). Relationships between them were also

fully resolved (PP = 1.00) with three clades (A–C) corre-

sponding phenotypically to S. campestris group being dee-

ply divergent from two clades (D, E) corresponding to

S. mearnsi group.

In the nuclear gene trees, similar structure is apparent,

although differentiation is more pronounced within

S. mearnsi group. While monophyly of clades D and E

from S. mearnsi group was unambiguously supported

(PP = 1.00) in both IRBP and RAG1 phylogenies, situation

within S. campestris was more complex. In IRBP tree, indi-

viduals from clades B and C are intermixed forming a com-

mon BC clade. A and BC clades are monophyletic with

reasonable support (PP = 97 and 80, respectively). Another

remarkable point here is the position of phased haplotypes

of specimen RS1023 from the left bank of Kafue River

(14.62°S, 26.19°E) found in different clades, A and BC,

suggesting hybrid origin of the individual. In RAG1 tree,

differentiation within S. campestris group is even weaker.

Clades A and combined BC are also monophyletic in the

maximum clade credibility tree, but with very low support

(PP = 0.39 and 0.65, respectively). B and C individuals are

again mixed within their common clade.

All short cytochrome b sequences as well as some mini-

barcodes from 454-pyrosequencing were associated with one

of the clades with high support (∑Aw ≥ 0.95). Six mini-bar-

codes were placed with varying support into different clades

as well as to branches ancestral to more than one clade and

were excluded from the analysis. Figure 3A shows geograph-

ical distribution of clades based on the complete cytochrome

b data set including also short sequences. It is clear that

S. campestris and S. mearnsi groups are allopatric, separated

by about 500-km-wide gap in central Tanzania, while within

both groups clades designated as candidate species are parap-

atric.

In S. campestris group, clade A is distributed over a vast

area in southern Africa, whereas B was recorded only from

central and north-eastern half of Zambia, south-western

Tanzania and northern Malawi and clade C is confined to

Fig. 2 Phylogeny of Saccostomus as estimated by Bayesian inference from sequences of cytochrome b, IRBP and RAG1 genes. Bayesian

posterior probabilities are indicated for the main clades. Vertical lines delimit major cytochrome b clades and their counterparts in IRBP

and RAG1 trees. Tips leading to individuals from cytochrome b clade C are always painted in violet to show their position within nuclear

gene trees. Phased haplotypes from the A9 BC heterozygote and the maximum likelihood placement of S. fuscus paralectotype are

indicated.



90


southern Malawi and the adjacent part of Mozambique.

Clades A and BC come into contact at the Kafue River and

Middle to Lower Zambezi, where A is generally present on

their right (western to southern) banks, whereas BC was

found on the left banks (eastern to northern). One individ-

ual having both A and BC haplotypes of IRBP was there-

fore found exactly at the contact line of these clades. It

should be also noted that in spite of being captured on the

left bank of the Kafue River, it possessed A haplotype of

cytochrome b, which might be also interpreted as an

instance of introgression. A single major outlier from this

parapatric pattern is paralectotype of S. fuscus from south-

ern Mozambique whose minibarcode is nested within B

(Fig. 3A). Clade C was found mostly eastward of Lake

Malawi and Shire River (Fig. 3A).

All three clades of S. campestris group show rather shal-

low internal branching. In clade A, however, there is still a

structure of three well defined (PP ≥ 0.79), although not

deeply divergent, subclades (A1–A3). While A1 (two haplo-

types recorded from the coasts of South African KwaZulu-

Natal) and A2 (six haplotypes in western part of Zambia)

are geographically very restricted, the third subclade (A3) is

far most widespread and comprises 51 haplotypes (Figs 2

and 3A). Within B no geographically distinct, close-to-

basal subclade was observed and clade C is a local cluster

of closely related sequences.

In S. mearnsi group, clade D is confined to the narrow

belt in northern Tanzania and south-west of Kenya,

whereas clade E is distributed across the Kenya to southern

Ethiopia and reaching South Sudan and southern Somalia,

where the limits of its distribution are unresolved (Fig. 3A).

The major geographical feature separating the known dis-

tributions of D and E is a mountain chain consisting of

Kilimanjaro, Pare and Usambara mountains. Although

being much less sampled, both D and E contain diver-

gences deeper than observed within clades of S. campestris

group. For example, two basal subclades are geographically

distinct in clade D, one (D1) including specimens from

southern Kenya and Serengeti region and the other one

(D2) from north-east Tanzania (Fig. 3A).

A B

Fig. 3 (A) Geographical distribution of major cytochrome b clades. Distribution of subclades within A and D is marked by thin links to

their labels, except for subclade A3 which comprises the rest of A haplotypes. Type locality of Saccostomus fuscus and sampling site of the

putative A9BC hybrid are also indicated. (B) Geographical distribution of the clades A, BC, D and E as estimated from morphometric data

and smoothed using thin-plate spline. Each point corresponds to a sampling site and colour intensity indicates strength of evidence for the

presence of particular clades (white colour corresponds to entirely ambiguous assignment). Type localities (yellow stars) and records from

central Tanzania (black arrows) are also shown (see more details in the text). Names of central Tanzanian localities are coloured according

to the likely assignment of the specimens.



91


Divergence dating and demographic history

Divergence dating analysis estimated TMRCA of Criceto-

myinae and Dendromurinae at 10.2 (95% HPD 8.5–12.2)

Ma, that is the posterior was shifted slightly towards the

present (prior median = 11.0 Ma). TMRCA of stem Saccos-

tomus was estimated at 8.8 (6.9–10.8) Ma and TMRCA of

crown (in the sense extant) Saccostomus at 3.9 (2.9–4.9) Ma,

that is it was shifted towards the past (prior

median = 3.4 Ma). TMRCA of S. mearnsi group was esti-

mated at 2.0 (1.3–2.9) Ma and that of S. campestris group at

1.4 (0.9–1.9) Ma. TMRCA of clade BC was estimated to

0.8 (0.5–1.2) Ma (Fig. 4).

Sampling of clades A and B (52 and 30 cytochrome b

haplotypes, respectively) was sufficient to examine their

demographic history in details. In agreement with shallow

phylogenetic structure observed in both these clades

(Fig. 2), Bayesian skyline plots suggest sustained and rela-

tively recent rise in effective population size (Fig. 5). In

clade A, expansion was particularly steep, beginning about

70 ka and peaking between 50 and 40 ka. Demographic

history of clade B shows two episodes of accelerated popu-

lation growth, one about 80 ka and the other at approxi-

mately 25 ka. Intriguingly, when population multiplication

rates (approximated by the first order differences) are plot-

ted (inset of Fig. 5), there is a temporal coincidence

between acceleration of growth in clade A and slowdown

of growth in clade B and vice versa. Tajima’s D statistic

also rejects hypothesis of neutral evolution of cytochrome

b sequence in both A (D = �3.09, P < 0.001) and B

(D = �3.67, P < 0.001). Large negative values of D

strongly suggest history of either population expansion as

suggested by Bayesian skyline plots or some kind of purify-

ing selection.

Species distribution models

Figure 6 summarizes the predicted distributions of two Sac-

costomus species groups in the present, and during last gla-

cial maximum (21 ka) and last interglacial (120–140 ka)

conditions. In both groups, the algorithm predicted exten-

sive changes in distribution patterns during the last glacial/

interglacial cycle. At present, S. campestris group is pre-

dicted to occur throughout the whole southern Africa,

except for driest areas in South Africa and Namibia, with

northern limits roughly coincident with margins of wide-

spread arid habitats in Central Tanzania and margins of

Congo basin forests. When going back in time, however,

its predicted distribution is getting progressively frag-

mented with the largest continuous area in today’s Bots-

wana and many isolated patches. The picture is different in

S. mearnsi group that is predicted to be present now across

the whole region considered, but the same is predicted also

for the last glacial maximum. Only in the last interglacial,

the predicted distribution changes substantially: although it

is largely preserved in today’s Tanzania, it is severely

reduced north of it with just two patches present north of

Lake Victoria and north-east of Lake Turkana.

Beamys

Cricetomys ansorgei

Cricetomys emini

Saccostomus A

Saccostomus B

Saccostomus C

Saccostomus D

Saccostomus E

Dendromus insignis

Dendromus mesomelas

Steatomys krebsii

Steatomys parvus

Steatomys pratensis

●

●

●

●

●

●

●

●

●

●

●

●

10.2

9.0

5.9

1.8

3.9

1.4

0.8

2.0

9.6

0.8

1.2

1.0

3 MYA

Fig. 4 Time calibrated tree from the BEAST analysis showing

estimated TMRCAs (in Ma); medians are displayed in circles and

95% HPD intervals are underlain as grey bars.

0 20 40 60 80 100

−3

−2

−1

01

23

Time [ka]

Log (

effective p

opula

tion s

ize *

genera

tion tim

e)

24 43 81

0.0

10

.05

0.0

9

Time [ka]

Mu

ltip

lica

tio

n r

ate

Fig. 5 Bayesian skyline plots of cytochrome b clades A (red; grey

[in B&W version]) and B (blue; black [in B&W version]) with

95% highest posterior density bands. In the inset approximated

and smoothed population, multiplication rates are shown.


O. Mikula et al. Diversity in rodent genus Saccostomus

92


To get a sense of climatic factors limiting the predicted

distributions, we inspected differences between 19 climatic

variables in cells where presence was either observed or

predicted and their nearest neighbours in the principal

component space with the absence predicted. This way we

assessed four variables as being most influential in S. cam-

pestris group, one involving temperature (BIO 10) and three

its covariation with precipitation (BIO8, BIO18, BIO19).

In summary, occurrence of these animals appears limited to

warm climate characteristic by temporal coincidence of rel-

atively warm and relatively wet parts of the year. In the

case of S. mearnsi group, the limiting factor seems to be

excessively large annual seasonality (BIO4 and BIO7).

Skull form differentiation

Four between-group principal components together

explained about 92% of size variation and 41% of shape

variation. In the Gaussian mixture modelling, the model

with arbitrary yet equal within-group covariance matrices

was supported as the best one (∆BIC > 12) and thus we

effectively performed linear discriminant analysis. Cross-

validation results were summarized in the form of confu-

sion matrix (Table 2) showing on diagonal classification

success and off diagonal degree of misclassification between

any two groups. Apparently, S. campestris and S. mearnsi

are entirely distinct with no confusion between them and

the same holds true for clades D and E within S. mearnsi

group. On the contrary, there was a small proportion of

misclassified specimens between any pair of clades within

S. campestris group suggesting their overlap in the PCA-

reduced form space. In particular, the classification success

was 96%, 90% and 71% in A, B and C, respectively, but

the misclassifications were not confined to any pair of

MOTUs nor to any direction. For instance, four B individ-

uals were classified as A, but other three as C and also one

A individual as B.

Another useful summary of skull form differences is den-

drogram of clade means constructed by average linkage

method (also called UPGMA) from their pairwise Maha-

lanobis distances, calculated with respect to the common

within-group covariance matrix. In Fig. 7, it is shown faced

to Saccostomus clade of time calibrated phylogeny, which

reveals general correspondence between morphological and

phylogenetic differentiation.

Figure 3B shows distribution map of clades A, BC, D

and E based on posterior probabilities of clade member-

ship obtained from the Gaussian mixture model. We

merged predictions made for a priori unclassified speci-

mens with cross-validated predictions used in the confu-

sion matrix and applied thin-plate spline smoothing to

obtain synthetic picture taking into account spatial

Campestris

Present Glacial Interglacial

Mearnsi

Fig. 6 Predicted distributions of

Saccostomus campestris (upper row) and

Saccostomus mearnsi (lower row) species

groups in the interglacial, glacial and

present-day climatic conditions (left,

middle and right columns, respectively).

Green points mark grid cells with actual

observations, yellow points mark cells

where the presence is predicted.

Table 2 Confusion matrix showing counts of correctly and incor-

rectly classified specimens. Row names indicate true membership,

and column names cross-validated classification. N indicates total

number of specimens included

A B C D E N

A 48 1 1 0 0 50

B 4 63 3 0 0 70

C 3 1 10 0 0 14

D 0 0 0 17 0 17

E 0 0 0 0 17 17



93


distribution of (mis-)classifications, including within-site

variation where multiple specimens from the same place

were present. The intensity of colour indicating strength

of evidence reveals a narrow belt along Kafue–Zambezi

where the most of ambiguous or contradicting classifica-

tions to either A or BC were located. The contact line

between A and BC is estimated to continue beyond

genetically investigated areas to Katanga region of Demo-

cratic Republic of Congo (DRC) and perhaps even to

Angola, although this is based on a single specimen

assigned to B. Distinction between D and E ranges is

pretty sharp predicting E to be present also in Uganda,

in the Coastal province of Kenya and in Mkomazi region

of Tanzania area on the northern side of Kilimanjaro–

Pare–Usambara mountain chain. In the Appendix S3, we

also present predicted membership to clades B and C.

Outside the known range of clade C, morphometric sup-

port for its presence was weak.

Figure 8 shows size variation within and between

MOTUs. Centroid sizes calculated on different skull sides

are highly correlated (r > 0.93) and just centroid size of

the dorsal side is therefore presented. Generally, skulls in

S. mearnsi group are larger than in S. campestris group, but

there is a huge variation within MOTUs, especially within

A, where the maximum size was recorded. On average,

skull size decreases from A to B to C in S. campestris and

from E to D in S. mearnsi group.

Skull shape features differentiated between successive

splits of major clades are depicted in Fig. 9. The two main

groups, S. campestris and S. mearnsi, differ in a number of

clearly distinct features. In particular, S. mearnsi group has

relatively narrower interorbital constriction, larger molars,

longer incisive foramens, as well as upward shifted zygo-

matic plate and differently shaped braincase, the latter

being evident on the posterior edge of squamosal and

occipital region of the skull. Clade A differed from BC by

several minor features – relatively smaller interparietal, dif-

ferently shaped auditory bullae and upward shifted skull

basis (seen on molars from the lateral view). Shape differ-

ences between B and C involved differently shaped zygo-

matic plate and relatively shorter incisive foramens in B.

Within S. mearnsi group clade E differed from D especially

by relative position of the posterior edge of squamosal,

skull basis and zygomatic arch.

Affinity of type specimens

Morphometric and where available also molecular classifi-

cation of type material is summarized in Table 3 and

Appendix S1, where further details about type localities are

provided. According to cytochrome b mini-barcodes,

S. lapidarius (Peters, 1852) was placed in clade A

(∑Aw = 1.00), whereas S. fuscus (Peters, 1852) in clade B

(∑Aw = 0.96).

Morphometric classifications are usually in accordance

with expectations based on location of type localities.

There are five exceptions from this: (i) Paralectotype of

S. fuscus (Peters, 1852) from southern Mozambique was

classified to C, which could relate to its small size, but it

could be also related to its genuine affinity to clade BC.

Overall, S. fuscus is special in all respect. Compared to the

Ma

xim

um

cla

de

cre

dib

ility

tre

e

De

nd

rog

ram

ba

se

d o

n M

aha

lan

ob

is d

ista

nce

sA

B

C

D

E

Fig. 7 Dendrogram summarizing Mahalanobis distances between

clade mean forms faced to Saccostomus clade from divergence

dating analysis.4

04

55

05

5

A B C D E

Ce

ntr

oid

siz

e [

mm

]

Fig. 8 Violin plots showing skull size variation within major

cytochrome b clades. Open black dots indicate median values,

violin shapes show probability densities obtained as kernel density

estimates and trimmed to the observed range of values.



94


main body of material, it is a very old record (mid-19th

century), it is the only outlier from otherwise clear phylo-

geographic pattern, and it has very dark fur colour (hence

the name fuscus meaning ‘dark’ in Latin). Image of its car-

cass preserved in Museum f€ur Naturkunde, Berlin, is in

Appendix S4. (ii) There was a contradiction in classifica-

tion of syntypes of S. lapidarius (Peters, 1852) as one of

them is classified to A (PP = 0.95) and other three to BC

(PP = 0.81, 0.85 and 1.00), which is perhaps not surpris-

ing given they were collected in a broader region mostly

on the left bank of lower Zambezi but including also Villa

de Senna on its right bank. The specimen classified to A

was the barcoded one (ZMB MAM-85437) so the molecu-

lar and morphometric classifications are in mutual agree-

ment. (iii) Holotype of S. elegans (Thomas, 1897) was

classified to A despite its type locality deep in the range

of B (Karonga, northern Malawi). It is, however, quite

large and very old individual with the most damaged skull

included so it might be misclassified due to its size, age or due

to error in missing data imputation. Furthermore, it was clas-

sified into A with PP = 0.69 only. (iv) Holotype of

S. mashonae was found almost equally likely to belong to A

(PP = 0.47) and BC (PP = 0.53), but its paratypes belong

unambiguously to A (PP = 0.83 and 1.00). (v) Holotype of

S. limpopoensis (Roberts, 1914) was classified into clade BC

but with low support (P = 0.55) and in spite of being deep in

the range of A as well as having paratype clearly belonging to

A (PP = 1.00).

Lectotype of S. fuscus (Peters, 1852) could not be

included due to skull damage and other types due to their

juvenile age, which is also the case for one syntype of

S. campestris (Peters, 1846), holotype of S. cricetulus (Allen

& Lawrence, 1936) and five topotypes of S. isiolae (Heller,

1912).

Taxonomic revision of Saccostomus species

In this study, we performed the comprehensive revision of

the genus Saccostomus from most of its known distribution

range using the combination of genetic and morphological

approaches. Based on the obtained evidence, we propose to

split S. mearnsi in two species and S. campestris in two sub-

species.

Saccostomus mearnsi group

Diagnosis. This group can be characterized by grey belly

and a number of cranial features including wide zygomatic

● ● ● ● ●

●

●

●

●

●

●

● ●

●

●

●

●

●

●

●

●

● ●

●

●

● ● ● ● ●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

● ● ● ●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●● ● ● ● ●●

●

●

●●

● ● ● ● ●

●

●

●

●

●

●●

●● ●

●

●

●

●

●

●

●

●

●

●

●

●

●

● ●

●● ● ● ● ●

●

●

●

●●

● ● ● ● ●

● ● ● ● ●

●

● ●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

● ●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

● ● ● ● ●

●

● ●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

● ●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

● ● ● ● ●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

● ● ● ● ●

●

●

●

●

●

●

● ●

●

●

●

●

●

●

●

●

● ●

●

●

●

●

●

●

●

●

●

●● ● ●

●

●

●

●

●

●

●

●

●

●

●

●

● ●●● ● ● ● ●

●

●

●

●●

●● ● ● ●

●

●

●

●

●

●●

● ● ● ●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●● ● ● ● ●●

●

●

●●

● ● ● ● ●

● ● ● ● ●

●

● ●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

● ●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

● ● ● ● ●

●

● ●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

● ●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

● ● ● ● ●

●

●

●

●

●

●

● ●

●

●

●

●

●

●

●

●

● ●

●

●

● ● ● ● ●

●

●

●

●

●

●

● ●

●

●

●

●

●

●

●

●

● ●

●

●

●

●

●

●

●

●●

● ● ● ●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●● ● ● ● ●●

●

●

●●

● ● ● ● ●

●

●

●

●

●

●

●

●● ●

●

●

●

●

●

●

●

●

●

●

●

●

●

● ●

●● ● ● ● ●

●●

●

●●

●● ● ● ●

● ● ● ● ●

●

● ●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

● ●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

● ● ● ● ●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

● ● ● ● ●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

● ● ● ● ●

●

●

●

●

●

●

● ●

●

●

●

●

●

●

●

●

● ●

●

●

●

●

●

●

●

●

●

●● ●

●

●

●

●

●

●

●

●

●

●

●

●

●

● ●

●● ● ● ● ● ●●

●

●●

● ● ● ● ●

●

●

●

●

●

●●

●● ● ●

●

●

●

●

●

●

●

●

●

●

●

●

● ●●● ● ● ● ●

●

●

●

●●

●● ● ● ●

● ● ● ● ●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

● ● ● ● ●

●

● ●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

● ●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

campestris

vs. mearnsi

A v

s.

BC

B v

s.

CD

vs.

E

Fig. 9 Skull shape differences between clades defined by successive splits in Saccostomus phylogeny, that is S. campestris group vs. S. mearnsi

group, A vs. BC, B vs. C and D vs. E. The mean shape of the latter from any pair is depicted in black.



95


bone, larger first upper molar, longer molar row and nar-

rower interorbital constriction (Hubert 1978; Denys 1988;

this study). It is monophyletic in both mitochondrial and

nuclear markers and distributed in Somali-Maasai biore-

gion (Corti et al. 2004; this study).

Contents. Two clades within this group were clearly dis-

tinct in both mitochondrial and nuclear DNA markers

(reciprocal monophyly, mean cytochrome b divergence

17.5%) as well as in skull form and karyotype in spite of

their geographical proximity. We therefore conclude they

are separate species, even though with present data it

was not possible to demonstrate their reproductive

isolation. Morphometric classification of type material

was unambiguous with posterior probabilities equal to

1.00. Saccostomus mearnsi (Heller, 1910) and S. isiolae

(Heller, 1912) were classified into the northern clade E,

whereas S. umbriventer (Miller, 1910) into the southern

clade D.

Saccostomus mearnsi Heller, 1910: 3

Saccostomus isiolae Heller, 1912: 14

Saccostomus cricetulus Allen and Lawrence, 1936: 100

Type locality. Changamwe, Kenya.

Type material. The holotype (male USNM 162882) and

topotype (female USNM 164854). Skin and skull preserved

for the holotype, but only skin for the topotype.

Emended diagnosis. On average larger than S. umbriventer,

with several distinguishing cranial features. Its skull is rel-

atively wider at the posterior end of zygomatic arch, rela-

tively higher (when measured form molars to skull roof)

and with differently shaped brain case (evident at the pos-

terior edge of squamosal). This is the species upon which

Hubert (1978) based its discrimination of S. mearnsi from

S. campestris. Both available karyotypes have at least 40

predominantly acrocentric chromosomes (Hubert 1978;

Corti et al. 2004). Distinct in mitochondrial and two

Table 3 List of Saccostomus type specimens and their molecular and morphometric classification, where available

Species group Original name Author Year Status Specimen Classification

Posterior

probability

Cytochrome b

mini-barcode

S. campestris S. campestris Peters 1846 Syntype BMNH 1858.6.18.19 B (BC) 0.78 (0.97) –

Syntype BMNH 1907.1.1.181 – – –

S. lapidarius Peters 1852 Syntype ZMB MAM-1712 C (BC) 0.84 (0.85) –

Syntype ZMB MAM-85435 B (BC) 0.87 (1.00) –

Syntype ZMB MAM-85436 B (BC) 0.71 (0.81) –

Syntype ZMB MAM-85437 A 0.94 A

S. fuscus Peters 1852 Lectotype ZMB MAM-85455 – – –

Paralectotype ZMB MAM-85450 C (BC) 0.76 (0.99) B

S. mashonae de Winton 1897 Holotype BMNH 1895.8.27.10 B (BC) 0.53 (0.53) –

Paratype BMNH 1895.8.27.9 A 0.83 –

Paratype BMNH 1904.12.1.17 A 1.00 –

S. elegans Thomas 1897 Holotype BMNH 1897.18.1.207 A 0.69 –

S. anderssoni de Winton 1898 Holotype BMNH 1869.8.11.4 A 1.00 –

S. hildae Schwann 1906 Holotype BMNH 1904.10.1.49 A 1.00 –

Topotype BMNH 1904.10.1.50 A 1.00 –

S. limpopoensis Roberts 1914 Holotype TM 1343 C (BC) 0.55 (0.55) –

Paratype TM 1344 A 1.00 –

S. streeteri Roberts 1914 Holotype TM 1356 A 1.00 –

S. pagei Thomas & Hinton 1923 Holotype BMNH 1910.6.3.54 A 1.00 –

S. anderssoni angolae Roberts 1938 Holotype TM 8023 A 1.00 –

S. mearnsi S. mearnsi Heller 1910 Holotype USNM 162882 E 1.00 –

S. umbriventer Miller 1910 Holotype USNM 162612 D 1.00 –

Topotype USNM 162610 D 1.00 –

Paratype USNM 162611 D 1.00 –

S. isiolae Heller 1912 Holotype USNM 181803 E 1.00 –

Topotype USNM 183688 E 1.00 –

Topotype USNM 183678 – – –





S. cricetulus Allen & Lawrence 1936 Holotype MCZ 31475 – – –



96


nuclear markers; it corresponds to genetic clade E in this

study.

Distribution. Somali-Maasai bioregion, from northern-

most Tanzania across Kenya to southern Ethiopia and

Somalia, South Sudan and eastern Uganda (Fig. 3). The

precise distribution in South Sudan, Uganda and Somalia

is unresolved. It has parapatric distribution with S. um-

briventer in north-eastern Tanzania and south-western

Kenya, where the two species can meet, but such evidence

is missing.

Notes. Since its first collection (28.11.1909 by E. A.

Mearns), it has never been recorded from the same area,

now Coast Province of Kenya. The holotype of S. cricetulus

(MCZ 31475) is juvenile and its skull is broken but accord-

ing to its type locality in Uganda it was classified to

S. mearnsi rather than S. umbriventer.

Ecological remarks. Dominant rodent species in wooded

grasslands of central Kenya (Mpala Research Center; Kees-

ing 1998a), but it was found by us also in more open grass-

lands in northern Kenya (near Marsabit) and southern

Ethiopia. Extensively studied with respect to its behaviour,

dietary preferences and determinants of its population den-

sity (Keesing 1998b; Metz & Keesing 2001; Keesing &

Young 2014).

Saccostomus umbriventer Miller, 1910: 1

Type locality. Njori Osolali, Guaso Nyiro (now Ewaso

Ngiro), Sotik, Kenya.

Type material. The holotype (female USNM 162612),

two paratypes (females USNM 162611 and 162613) and a

topotype (USNM 162610). Skin and skull preserved for all

of them except for the paratype USNM 162613 whose

skull is unaccounted for.

Emended diagnosis. For its discrimination from S. mearnsi,

see the account of that species. The single available kary-

otype has 2n = 32 with multiple metacentric chromosomes

(Corti et al. 2004). Distinct in mitochondrial and two

nuclear markers; it corresponds to genetic clade D in this

study.

Note. The populations from Massai Steppe in northern

Tanzania were proposed as separate species already by

Denys (1988) and Corti et al. (2004). However, it was sug-

gested by the latter authors to differ from the S. umbriven-

ter type (without providing more details). Geometric

morphometric analysis in our study clearly assigned S. um-

briventer types to genetic clade D that involves also

specimens mentioned by Corti et al. (2004). This is also

consistent with our recent records of specimens belonging

to clade D near the type locality of S. umbriventer in

south-western Kenya.

Distribution. Somali-Maasai bioregion, in a narrow belt in

northern Tanzania and south-western Kenya. Parapatric to

S. mearnsi – see the account of that species.

Ecological remarks. Its record is mostly anecdotal but a lar-

ger sample from Kijungu (=Kujungu, north-eastern Tanza-

nia) in AMNH collections suggests it may be locally

abundant. The other larger sample comes from pellets col-

lected by Reed (2007) in Serengeti, where S. umbriventer

represented 2–3% of small mammals. In our recent trap-

ping in north-western Tanzania, it was the only species

being captured together with Mastomys natalensis in a

mosaic of fields and Massai Steppe.

Saccostomus campestris group

Diagnosis. This group can be characterized by white belly

(apart from aberrant type of S. fuscus) and a number of cra-

nial features including narrow zygomatic bone, relatively

shorter molar row, larger interparietal and wider interor-

bital constriction (Hubert 1978; Denys 1988; this study). It

is monophyletic on both mitochondrial and nuclear mark-

ers and distributed in Zambezian and South African biore-

gions (Corti et al. 2004; this study).

Contents. The genetic clades A, B and C within S. cam-

pestris group were not as strikingly distinct as those

within S. mearnsi group. From the perspective of nuclear

markers, clades A and BC are reciprocally monophyletic

in maximum clade credibility trees but with low support

and clades B and C are not distinct at all. Morphologi-

cally, A, B and C were comparably, yet incompletely dif-

ferentiated from each other and two nuclear genes

supported only split between clades A and BC. On the

other hand, a degree of reproductive isolation is likely

present between A and BC, given coincidence of their

contact with Zambezi–Kafue water flows and congruence

of mitochondrial and nuclear DNA variation, both in face

of episodic evidence of hybridization. Therefore, we sug-

gest clades A and BC to be referred as subspecies unless

it will be convincingly demonstrated they deserve species

status. Morphometric discrimination, even though not

perfect, was still good enough to be useful for classifica-

tion of type material. The oldest available name for BC

clade is shown to be S. campestris (Peters, 1846), as its

syntype belongs to BC with PP = 0.97 (sum of PPs for

B and C) and the oldest name for A clade is S. mashonae

(de Winton, 1897), whose holotype belongs to A with



97


PP = 0.47 only, but two paratypes with PP = 0.83 and

1.00.

Saccostomus campestris campestris Peters, 1846: 258

Saccostomus lapidarius Peters, 1852: 167

Saccostomus fuscus Peters, 1852: 168

Saccostomus elegans Thomas, 1897: 431

Type locality. Tete, Zambezi River, Mozambique.

Type material. Two syntypes, BMNH 1858.6.18.19 (fe-

male) and 1907.1.1.181 (sex unknown).

Emended diagnosis. On average smaller than

S. c. mashonae, with relatively larger interparietal and differ-

ently shaped auditory bullae. Distinct in mitochondrial and

at least two nuclear markers; it comprises genetic clades B

and C in this study.

Distribution. Zambezian bioregion, including Malawi,

Tanzania as well as Zambia and Mozambique north to

Zambezi and Kafue Rivers. Likely present also in the

Katanga province of Democratic Republic of Congo and

perhaps further west.

Notes. Two specimens of S. fuscus are recorded as lecto-

type (ZMB MAM-85455, alcoholic specimen and frag-

mented skull) and paralectotype (ZMB MAM-85450, skull

and skin) in the catalogue of Museum f€ur Naturkunde Ber-

lin but no information is given who designated them as

such. They represent the only record of this subspecies

south of the Zambezi River.

Saccostomus campestris mashonae de Winton, 1897: 804

Saccostomus anderssoni de Winton, 1898: 6

Saccostomus hildae Schwann, 1906: 110

Saccostomus limpopoensis Roberts, 1914: 183

Saccostomus streeteri Roberts, 1914: 183

Saccostomus pagei Thomas and Hinton, 1923: 495

Saccostomus anderssoni angolae Roberts, 1938: 240

Type locality. Mazowe, Mashonaland Central, Zimbabwe.

Type material. The holotype (male BMNH 1895.8.27.10)

and two paratypes (male BMNH 1895.8.27.9 and female

BMNH 1904.12.1.17).

Emended diagnosis. For its discrimination from S. c. cam-

pestris, see the account of that species. Distinct in mito-

chondrial and at least two nuclear markers; it corresponds

to genetic clade A in this study.

Distribution. Zambezian and South African regions, south

of the Zambezi and Kafue Rivers. Its distribution limits in

Angola are unresolved.

Ecological remarks. The subspecies is known to inhabit

remarkable variety of habitats from periodically flooded

shrublands in southern Mozambique to semideserts in

Namibia. Its ability to enter spontaneous torpor (Ellison

1993) and size variation (Ellison et al. 1993) were both

linked to climatic variation.

Note. Corti et al. (2004: 423) mistakenly mentioned type

locality of ‘mashonae’ as lying ‘north of the Zambezi’. In

the original description (de Winton 1897) it is cited as

‘Mazoe, Mashunaland’ and on the museum tag of the holo-

type it is referred to as ‘Mazoe, Mashonaland’ in Zim-

babwe. The spelling ‘Mazowe’ is preferred, for example by

Google Maps.

Discussion

In this study, we analysed the largest series of Saccostomus

examined to date. We confirmed the two main lineages of

the genus, currently known under names S. campestris and

S. mearnsi (Hubert 1978; Corti et al. 2004), which are recip-

rocally monophyletic and deeply divergent in both mito-

chondrial and nuclear markers. The two species groups also

clearly differ in cranial form (Hubert 1978; Denys 1988),

involving both size (S. campestris much smaller on average)

and shape. Numerous skull shape differences were observed

between these groups, but they were complex and therefore

geometric morphometric approaches were required to com-

prehend them fully. This complexity explains why there were

historically very few diagnostic differences in the early analy-

sis of skull measurements (Denys 1988).

The S. campestris group is widely distributed in southern

part of the continent from Namibia eastwards to Mozam-

bique and from south-western Tanzania southwards to the

Cape (Figs 3 and 6). The taxon is known also from

Angola and southern DRC, but the general lack of sam-

pling in these areas prevents detailed characterization of

its distribution and taxonomy there. The S. mearnsi group

harbours populations that occur in Somali-Maasai biore-

gion, that is from northern Tanzania through Kenya to

Uganda, South Sudan, southern Ethiopia and Somalia

(Fig. 6). Most probably the distribution of the two main

groups is effectively allopatric with little, if any, contem-

porary contact across ca. 500-km-wide gap in central Tan-

zania. The southernmost record of S. mearnsi group in

this area is from Berega (6.17 S, 37.13 E; Makundi et al.

2010), but the authors report trapping it here is rare. The

only confirmed record of S. campestris group in this region

is from the other side of Rubeho massif (Kichangani;

6.49 S, 37.41 E), just about 50 km away. There is also a

historical record of S. campestris from Mpwapwa (near

Dodoma, 6.35 S, 36.49 E; Swynnerton & Hayman 1951).

Nevertheless, these populations on the border of continu-




98


ous distribution (cf. Fig. 3B) have probably low density,

which is supported by their uniqueness on the background

of otherwise rich collections from this region (e.g. Field

Museum, Chicago; Mus�ee royal de l’Afrique centrale,

Tervuren).

Numerous Saccostomus fossils exists from East African

localities that are dated to the Early Pliocene, 2.5–3.7 Ma

(Denys 1987, 2011) or even earlier, 4–5 Ma (Winkler

1997). These specimens belong to the extinct species Sac-

costomus major, described by Denys (1987) from the Laetolil

Beds. Another fossil species, Saccostomus geraadsi, was

described by Mein et al. (2004) from the Harasib karst in

Namibia. The authors claimed it was identical to specimens

collected from Ethiopian Chorora (Geraads 2001) and sug-

gested an age 8.7–10.5 Ma for both places. Also, the speci-

mens from the Lukeino formation in Kenya (first reliably

dated record of Saccostomus, 5.8–6.1 Ma; Mein & Pickford

2006) are similar to S. geraadsi. The first record of the

crown S. mearnsi group is from 1.7 to 1.9 Ma (Denys

1992). Older fossils of S. campestris group are unknown to

date. Given the abundant fossil record, it seems unlikely

that TMRCA of crown Saccostomus occurred before

6.1 Ma, and our divergence dating by relaxed molecular

clock restricted it to 3.9 (2.9–4.9) Ma. This is in line with

observation that many extant African rodent genera have

their basic splits dated to Late Miocene or Early Pliocene

boundary (Montgelard & Matthee 2012; Schenk et al.

2013). In this framework, S. major might belong to the

crown Saccostomus, perhaps as a sister lineage of S. mearnsi

group, whereas S. geraadsi is considered a stem lineage.

The origin of stem Saccostomus was estimated at 8.8 (6.9–

10.8) Ma, a wide time span covering both proposed ages of

the Harasib and Chorora localities and one of the ages sug-

gested by Corti et al. (2004, 9.3 Ma).

Following their divergence in Early Pliocene, further

diversification of S. campestris and S. mearnsi groups took

place in different biogeographical regions, Somali-Maasai

and Zambezian, respectively, whose faunas are recognizable

in fossil record since upper Pliocene (Denys 1999, p. 246).

This situation is similar to gerbils of the genus Gerbilliscus,

where the final radiation of two subgeneric lineages

occurred during Early Pliocene in Zambezian and South

African regions (‘Southern’ group sensu Colangelo et al.

2007) and in Somali-Maasai region (‘Eastern’ group of

Colangelo et al. 2007).

Diversity of African pouched mice

In accord with Corti et al. (2004), we detected remarkable

variation within the two main Saccostomus groups. In addi-

tion to earlier studies combining classical morphometrics

with karyotypic and mitochondrial DNA variation, we pro-

vide here the first analysis of nuclear DNA sequences and

geometric morphometric analysis of Saccostomus rodents.

Based on cytochrome b phylogeny, we proposed up to five

clades within the genus and assessed their species status.

In the S. campestris group, we recognized two clades, A

and BC, and proposed their subspecies status. Incomplete

skull form differentiation, low support of their reciprocal

monophyly especially in RAG1 tree and lack of knowledge

about the extent of their reproductive isolation (with evi-

dence of occasional hybridization at their contact zone)

prevented us from describing them as distinct species, yet

this may change in future with improved knowledge of

their differentiation and interaction. Namely, denser sam-

pling, population genomic data, analysis of their functional

morphology and behaviour including possible assortative

mating experiments might show them as well separated,

even though occasionally hybridizing species. In fact, lack

of differentiation at RAG1 may be due to slow evolution of

this gene and hybridization may be confined to a very nar-

row space around the contact line with little impact on

gene pools further apart. Alternatively, it may be proved in

future that S. campestris campestris and S. campestris masho-

nae produce somewhat less fit hybrids, but otherwise they

are ecologically equivalent, showing no mating preference

and significant gene flow. In such cases, category of sub-

species is a meaningful and should be retained for them.

The prime areas of interest for further research are along

the proposed contact line on Kafue–Zambezi as well as

near the Kafue Flats at the upper reaches of the river and

further north-west in the Katanga province of DRC, where

the two subspecies should also meet.

Within S. campestris campestris, there is a deep divergence

of mitochondrial cytochrome b dividing it in two phy-

logroups, called B and C in this paper. This divergence is

absent, however, in the nuclear markers used. In morpho-

metric analysis, the clade C was the least distinct of all

clades with just 71% correctly classified individuals. Again,

without any additional data, opposite interpretations of this

observation are possible. Either too conservative nuclear

markers were chosen to detect recent and perhaps unfin-

ished speciation or the spatial pattern observed on cyto-

chrome b is confined to mitochondrial DNA as a result of

stochastic demographic process. Nevertheless, at present,

we have no evidence supporting reproductive isolation or,

more broadly, evolutionary distinctiveness of B and C and

thus we treat them as belonging to a single subspecies.

More sampling, especially in broad vicinity of the Shire

River would be beneficial to test whether this conclusion is

appropriate.

In the S. mearnsi group, two distinct species S. mearnsi

and S. umbriventer were recognized, corresponding to

genetic clades E and D, respectively. This distinction was

unambiguously supported by all types of data. In all gene



99


trees, they were reciprocally monophyletic with PP = 1.00

and their skull form is so distinct that all specimens were

correctly classified. At the same time, they occur in a close

geographical proximity and so they most likely represent

well-separated species. Nevertheless, it would be still desir-

able and interesting to study them in direct contact that

should take place in south-western Kenya and possibly also

in north-eastern Tanzania.

In both main groups of Saccostomus, there is large kary-

otypic variability that was earlier proposed to serve as a

basis for species delimitations (Gordon & Watson 1986).

In the S. campestris group, diploid number of chromosomes

(2n) varies from 32 to 46 (reviewed by Corti et al. 2004;

Maputla 2008), but most of this variability was due to

Robertsonian fan in S. campestris mashonae (Gordon &

Rautenbach 1980). However, Corti et al. (2004, 2005) also

described conservative G- and C-banded karyotypes

(2n = 44) from southern Tanzania and northern Zambia

(S. campestris campestris). It is therefore evident that in this

group there is no clear relationship between DNA-based

phylogenetic clades and karyotypic variability. Further-

more, absence of evidence for hybrid disadvantage in

hybridization experiments in populations with 2n = 32 and

2n = 46 (Maputla et al. 2011) suggests that karyotypes are

unlikely to play important role in speciation of Saccostomus.

The first karyotypes, although low quality, for the

S. mearnsi group were provided by Hubert (1978) from

individuals in southern Ethiopia (2n = 40–42). More

recently, Corti et al. (2004) provided two very distinct kary-

otypes from Somalia (2n = 44) and northern Tanzania

(2n = 32). Both have NFa = 44 (incorrectly reported in

their Table 1 as 42 or 48), and even if the banding pattern

was not provided for the Somali individual, the changes

between the karyotypes can be explained by multiple

Robertsonian fusions. Therefore, variability of karyotypes

is consistent with our phylogenetic reconstructions. The

northern species S. mearnsi has high numbers of predomi-

nantly acrocentric chromosomes and the southern species

S. umbriventer has 2n = 32 with seven pairs of biarmed

chromosomes (Corti et al. 2004). It does not imply, how-

ever, that chromosomal evolution is driving speciation here.

Robertsonian fusions may have no direct impact on repro-

ductive success (Maputla et al. 2011; Horn et al. 2012), and

we are also unaware of actual karyotypic variability within

either of the species.

Phylogeographic scenarios for S. campestris group

Representatives of the S. campestris group are widespread,

closely associated with drier habitat types (woodlands,

grassland and scrubland savannah), abundant and easy to

capture providing an excellent opportunity to use them as

phylogeographic models for analyses of historical processes

affecting evolution of biota in the Zambezian bioregion.

This region provides rich information about fossil Mam-

malian fauna (Werdelin & Sanders 2010), paleoclimatic

variables for the last 5 Ma (de Menocal 2004; Trauth et al.

2010; Potts 2013; etc.), and drainage evolution data (Stan-

kiewicz & de Wit 2006; Moore et al. 2007, 2012). The

Pleistocene splits within the S. campestris group are dated

to intervals 1.9–0.9 Ma (S. c. campestris vs. S. c. mashonae)

and 1.2–0.5 Ma (clades B and C within S. c. campestris). In

this time span, two periods of high climatic variability took

place (Potts 2013), promoting diversification and speciation

by rapid and repeated shifts, merger and fragmentation of

suitable habitats. Also emerging and disappearing geomor-

phological features like paleolake Deception in the area of

recent Okavango delta and Chambeshi–Kafue–Zambezi

River system (Moore et al. 2012) might play a role in frag-

mentation of ancestral distribution area of S. campestris

group.

Today, S. c. campestris and S. c. mashonae are separated

by the Zambezi and Kafue Rivers. These two rivers are

obvious barriers to gene flow and have been implicated in

shaping genetic diversity in organisms as diverse as killi-

fishes (Bart�akov�a et al. 2015), gerbils (McDonough et al.

2015), baboons (Keller et al. 2010) and several species of

antelopes (Cotterill 2003). It does not necessarily mean that

Saccostomus is unable to cross these massive water flows, but

rather that rivers are likely to impede migration and thus

they can easily trap moving hybrid zones (Barton 1979).

Indeed, the contact of S. c. campestris and S. c. mashonae

at the Zambezi-Kafue line might be established relatively

recently. Shallow structures in cytochrome b genealogies

within both subspecies and distribution of nucleotide varia-

tion suggest recent population expansion, which was dated

to the last glacial. Moreover, distribution of suitable habi-

tats was presumably fragmented during the last interglacial

(Fig. 6). Large distribution gap is predicted for that period

in place of the current contact line, consistent with several

intervals of extreme aridity in the Lake Malawi region

between 135 and 90 ka (Cohen et al. 2007). Palynological

analyses provided the evidence of high pollen accumulation

of Poaceae grasses, especially in 105–90 ka (Beuning et al.

2011). This episode of a megadrought was so severe that

even dominant miombo trees like Brachystegia completely

disappeared from the pollen source area of Lake Malawi

(Beuning et al. 2011), which might have caused the disap-

pearance of Saccostomus from a large part of its current dis-

tribution. The next period is characterized by the

vegetation recovery and by 85–80 ka there was a resur-

gence of savannah woodland habitat documented in the

pollen core at Lake Malawi (Beuning et al. 2011), which

possibly promoted population expansion from relatively

small source populations.




100


It is also intriguing that according to Bayesian skyline

plots, the steepest population growth in S. c. mashonae took

place precisely when the growth slowed down in S. c. cam-

pestris and the growth in S. c. campestris was accelerated

again when it slowed down in S. c. mashonae. This might

be interpreted, for instance, as a signature of direct interac-

tion between the subspecies in a moving hybrid zone, but

with data in hand we cannot test this hypothesis. Similarly,

we do not know whether subclade A3 became dominant

within S. c. mashonae due to its selective superiority on

either gene or organismal level. Nevertheless, such

hypotheses might be tested in a future with genomic data

and more intensive sampling, for example by a method of

Peter & Slatkin (2013) that is tracking changes in frequen-

cies of numerous biallelic markers.

A significant exception from the pattern of parapatric

distributions is the paralectotype of S. fuscus collected near

Inhambane in southern Mozambique, which is placed to

S. c. campestris by its cytochrome b mini-barcode as well as

by its skull form. Occurrence of relict population of

S. c. campestris deep in the range of S. c. mashonae would

have important phylogeographic implications, but we are

hesitant to draw far-reaching conclusions from a single

observation, even though intriguing (e.g. dark belly of

S. fuscus). In any case, it would be beneficial to obtain more

molecular data from the paralectotype and/or lectotype as

well as to enlarge skull form data set and explicitly model

allometric (size-correlated) shape variation within both sub-

species to assess whether morphometric classification of the

paralectotype might be due to its subadult age. No Saccosto-

mus were captured recently near Inhambane town despite

our extensive trapping effort. It can be explained by the

fact that the environs in this area are currently highly

degraded and in many places changed to oil palm planta-

tions, so it is also possible that the original, possibly

unique, population went extinct.

More data are also needed to assess nature and origin of

B vs. C divergence of cytochrome b within S. c. campestris.

Future field work and employment of rapidly evolving

nuclear markers are necessary to assess how much mito-

chondrial gene pools are separated, whether this distinction

is apparent also in some nuclear loci and what is the role

of geomorphological features (most notably Lake Malawi

and Shire River) in shaping patterns of molecular variation

of Saccostomus in this region. Geographically distinct clade

C might arise in situ by a series of bottlenecks in repeat-

edly fragmented populations as well as by a founder event

following colonization either cross the Shire River or

through the migration corridor between lakes Malawi and

Rukwa called ‘Mbeya triple junction’. The latter area was

used for colonization of Tanzania, for example by mole-

rats of the genus Fukomys approximately 1 Ma (Faulkes

et al. 2010). Both population fragmentation and range shifts

might take place in a dry climatic period taking place 135–

90 ka in the region (Cohen et al. 2007).

Phylogeographic scenarios for S. mearnsi group

Saccostomus mearnsi and S. umbriventer are separated by the

mountains in northern Tanzania (e.g. Kilimanjaro, Pare

and Usambara), although they probably come together in

south-west of Kenya, where more detailed sampling is

required (Fig. 3B). This phylogeographic pattern has not

been described yet, which is related to lack of detailed

genetic data from Somali-Maasai bioregion. It is also

unclear whether observed separation is due to general lack

of suitable habitats along that mountain chain or due to

specialization to, for instance, different climate or micro-

habitats on its northern and southern side. Our species dis-

tribution model suggests fragmentation of the continuous

distribution of S. mearnsi group in the last interglacial with

conditions being unsuitable over the large portion of its

present range in Ethiopia and Kenya including the area of

presumed contemporary contact between S. mearnsi and

S. umbriventer (Fig. 6), which suggests possible separation

of the two species by climatic fluctuations.

In addition, a spatial substructure is visible within S. um-

briventer with different haplotype clusters on the western

(D1) and eastern (D2) side of the Rift Valley (Fig. 3A).

Similar pattern we observed also in a preliminary analysis

of cytochrome b variation in Gerbilliscus vicinus, another

rodent species living in similar habitats as Saccostomus (T.

Aghov�a et al. in prep.) Tendency to such differentiation

might be caused by amplifier lakes repeatedly appearing in

the Rift Valley during humid climatic periods (Trauth et al.

2010) and serving as an effective migration barrier. At pre-

sent, however, our sampling is too scarce to allow testing

of this hypothesis using Saccostomus data.

Genus Saccostomus and modern taxonomy

In our present revision of genus Saccostomus, we addressed

earlier suggestions (Denys 1988; Corti et al. 2004) that

each of the species described by Hubert (1978) may be in

fact a group of morphologically cryptic species. Based on

the joint molecular and morphometric analysis, we con-

clude there are cryptic taxa within the genus although there

are probably not as many of them as might be supposed.

Namely we recognized two species within S. mearnsi group

and two subspecies in S. campestris group that are cryptic in

the sense ‘not easily recognized’ yet showing signatures of

at least partial reproductive isolation and independent evo-

lution. This conclusion is perhaps typical for taxonomic

endeavour of the last few decades. The consensus is grow-

ing that even good species may be extremely similar in fea-

tures apparent to humans (‘cryptic species’) but on the



101


other hand there is also a growing recognition that even

large karyotypic changes and striking genetic divergences

do not automatically imply species status.

Acknowledgements

This study was supported by the projects of the GACR

Czech Science Foundation, nos. P506-10-0983 and 15-

20229S. For help in the field and with logistics, we

acknowledge W.N. Chitaukali, H. Konvi�ckov a-

Patzenhauerov a, J. Kr asov a, M. L€ovy, R. Makundi, V.

Mazoch, J. �Skl ıba, J. Vrbov a Kom arkov a and all local col-

laborators. A. Bryjov a, D. Chovanec and H. Konvi�ckov a-

Patzenhauerov a helped with genotyping. For permission to

carry out the research and to collect specimens, we are

obliged to the National Research Council and Forestry

Department in Malawi, the National Council for Science

and Technology, the Kenyan Forest Service and the Ken-

yan Wildlife Service, the Ethiopian Wildlife Conservation

Authority, Sokoine University in Morogoro, and Zambian

Wildlife Authority. We would also like to thank the cura-

tors of the collections, R. Baker (TTU), J. Britton-Davi-

dian (ISEM), J. Chupasko (MCZ), C. Conroy (MVZ), G.

Csorba (HNHM), N. Duncan (AMNH), P. Jenkins

(BMNH), T. Kerney (TM), D. Lunde (USNM), C.

Mateke (LM), F. Mayer (ZMB), D. Moerike (SMNS), V.

Nicolas (MNHN), E. Verheyen (RBINS), V. Volpato

(SMF) and W. Wendelen (RMCA), for providing us with

tissue samples and allowing us to study the skeletal material

in their care. T. Kearney (TM), N. Lange (ZMB) and M.

Omura (MCZ) contributed by taking images of type mate-

rial. We thank M. McDonough for language correction

and comments on the earlier version of the manuscript.

The paper was also significantly improved by comments of

M. Carleton, C. Denys and one anonymous reviewer.

ReferencesAdams, D. C. & Otarola-Castillo, E. (2013). geomorph: An R

package for the collection and analysis of geometric morphomet-

ric shape data. Methods in Ecology and Evolution, 4, 393–399.Allen, G. M. & Lawrence, B. (1936). Reports on the scientific

results of an expedition to rain forest regions in eastern Africa.

Bulletin of the Museum of Comparative Zoology at Harvard College,79, 100–102.

Arbour, J. H. & Brown, C. M. (2014). Incomplete specimens in

geometric morphometric analyses. Methods in Ecology and Evolu-tion, 5, 16–26.

Bart akov a, V., Reichard, M., Bla�zek, R., Pola�cik, M. & Bryja, J.

(2015). Terrestrial fishes: rivers are barriers to gene flow in

annual fishes from the African savanna. Journal of Biogeography,42, 1832–1844.

Barton, N. H. (1979). Gene flow past a cline. Heredity, 43, 333–339.Bastien, P., Esposito-Vinzi, V. & Tenenhaus, M. (2005). PLS gen-

eralised linear regression. Computational Statistics & Data Analysis,48, 17–46.

Bensmail, H. & Celeux, G. (1996). Regularized Gaussian discrimi-

nant analysis through eigenvalue decomposition. Journal of theAmerican Statistical Association, 91, 1743–1748.

Berger, S. A., Krompass, D. & Stamatakis, A. (2011). Performance,

accuracy, and web server for evolutionary placement of short

sequence reads under maximum likelihood. Systematic Biology, 60,291–302.

Bertrand, F., Meyer, N. & Maumy-Bertrand, M. (2014). Partial

least squares regression for generalized linear models. R package

version 1.1.1.

Beuning, K. R. M., Zimmerman, K. A., Ivory, S. J. & Cohen, A.

S. (2011). Vegetation response to glacial-interglacial climate vari-

ability near Lake Malawi in the southern African tropics. Palaeo-geography Palaeoclimatology Palaeoecology, 303, 81–92.

Bivand, R. & Lewin-Koh, N. (2014). maptools: Tools for reading

and handling spatial objects. R package version 0.8-29.

Blonder, B. (2015). hypervolume: High-dimensional kernel density

estimation and geometry operations. R package version 1.3.0.

Blonder, B., Lamanna, C., Violle, C. & Enquist, B. J. (2014). The

n-dimensional hypervolume. Global Ecology and Biogeography, 23,595–609.

Bouckaert, R., Heled, J., K€uhnert, D., Vaughan, T., Wu, C.-H.,

Xie, D., Suchard, M. A., Rambaut, A. & Drummond, A. J.

(2014). BEAST 2: A software platform for Bayesian evolutionary

analysis. PLoS Computational Biology, 10, e1003537.Braconnot, P., Otto-Bliesner, B., Harrison, S., Joussaume, S.,

Peterchmitt, J. Y., Abe-Ouchi, A., Crucifix, M., Driesschaert E,

Fichefet, T., Hewitt, C. D., Kageyama, M., Kitoh, A., Laine, A.,

Loutre, M.-F., Marti, O., Merkel, U., Ramstein, G., Valdes, P.,

Weber, S..L., Yu, Y. & Zhao, Y. (2007). Results of PMIP2 cou-

pled simulations of the Mid-Holocene and Last Glacial Maxi-

mum – Part 1: Experiments and large-scale features. Climate ofthe Past, 3, 261–277.

Bryja, J., Mikula, O., �Sumbera, R., Meheretu, Y., Aghov a, T., Lav-

renchenko, L. A., Mazoch, V., Oguge, N., Mbau, J. S., Weleger-

ima, K., Amundala, N., Colyn, M., Leirs, H. & Verheyen, E.

(2014). Pan-African phylogeny of Mus (subgenus Nannomys)reveals one of the most successful mammal radiations in Africa.

BMC Evolutionary Biology, 14, 256.Carleton, M. D. & Musser, G. G. (1984). Muroid rodents. In S.

Anderson & J. K. Jones Jr (Eds) Orders and Families of RecentMammals of the World (pp. 289–379). New York: John Wiley

and Sons.

Cohen, A. S., Stone, J. R., Beuning, K. R. M., Park, L. E., Rein-

thal, P. N., Dettman, D., Scholz, C. A., Johnson, T. C., King, J.

W., Talbot, M. R., Brown, E. T. & Ivory, S. J. (2007). Ecologi-

cal consequences of early Late Pleistocene megadroughts in

tropical Africa. Proceedings of the National Academy of Sciences ofthe United States of America, 104, 16422–16427.

Colangelo, P., Granjon, L., Taylor, P. J. & Corti, M. (2007). Evo-

lutionary systematics in African gerbilline rodents of the genus

Gerbilliscus: inference from mitochondrial genes. Molecular Phylo-genetics and Evolution, 42, 797–806.

Corti, M., Scanzani, A., Rossi, A. R., Civitelli, M. V., Bekele, A. &

Capanna, E. (1999). Karyotypic and genetic divergence in the

Ethiopian Myomys-Stenocephalemys complex (Mammalia, Roden-

tia). Italian Journal of Zoology, 66, 341–349.Corti, M., Castiglia, R., Annesi, F. & Verheyen, W. (2004). Mito-

chondrial sequences and karyotypes reveal hidden diversity in




102


African pouched mice (subfamily Cricetomyinae, genus Saccosto-mus). Journal of Zoology, 262, 413–424.

Corti, M., Castiglia, R., Colangelo, P., Capanna, E., Beolchini, F.,

Bekele, A., Oguge, N. O., Makundi, R. H., Sichilima, A. M.,

Leirs, H., Verheyen, W. & Verhagen, R. (2005). Cytotaxonomy

of rodent species from Ethiopia, Kenya, Tanzania and Zambia.

Belgian Journal Zoology, 135(Suppl.), 197–216.Cotterill, F. P. D. (2003). Geomorphological influences on vicar-

iant evolution in some African mammals in the Zambezi basin:

some lessons for conservation. In: Plowman A, ed. Ecology andConservation of Small Antelope. Proceedings of an International Sym-posium on Duiker and Dwarf Antelope in Africa. F€urth: FilanderVerlag, 11–58.

Demos, T. C., Peterhans, J. C. K., Agwanda, B. & Hickerson, M.

J. (2014). Uncovering cryptic diversity and refugial persistence

among small mammal lineages across the Eastern Afromontane

biodiversity hotspot. Molecular Phylogenetics and Evolution, 71,41–54.

Denys, C. (1987). Rodentia and Lagomorpha. 6.1. Fossil rodents

(other than Pedetidae) from Laetoli. In M. D. Leakey & J. M.

Harris (Eds) Laetoli, a Pliocene Site in Northern Tanzania (pp.

118–170). Oxford: Clarendon Press.

Denys, C. (1988). Apports de l’analyse morphometrique a la deter-

mination des especes actuelles et fossils du genre Saccostomus(Cricetomyinae, Rodentia). Mammalia, 52, 497–532.

Denys, C. (1992). Pr!esence de Saccostomus (Rodentia, Mammalia)

a Olduvai Bed-I (Tanzanie, Pl!eistoc ene inf!erieur). Implica-

tions phyl!etiques et pal!eobiog!eographiques. Geobios, 25, 145–

154.

Denys, C. (1999). Of mice and men. Evolution in East and South

Africa during Plio-Pleistocene times. In T. Bromage & F.

Schrenk (Eds) African Biogeography, Climate Change & HumanEvolution (pp. 226–252). Oxford: Oxford University Press.

Denys, C. (2011). Rodents. In T. Harrison (Ed.) Paleontology andGeology of Laetoli: Human Evolution in Context: Fossil Homininsand the Associated Fauna (vol. 2, pp. 15–53). Dordrecht: Springer.

Drummond, A. J., Rambaut, A., Shapiro, B. & Pybus, O. G.

(2005). Bayesian coalescent inference of past population dynam-

ics from molecular sequences. Molecular Biology and Evolution, 22,1185–1192.

Dryden, I. L. & Mardia, K. V. (1998). Statistical Shape Analysis.Chichester: Wiley.

Ellison, G. T. H. (1993). Evidence of climatic adaptation in spon-

taneous torpor among Pouched mice Saccostomus campestris fromsouthern Africa. Acta Theriologica, 38, 49–59.

Ellison, G. T. H., Taylor, P. J., Nix, H. A., Bronner, G. N. &

McMahon, J. P. (1993). Climatic adaptation of body size among

pouched mice (Saccostomus campestris, Cricetidae) in the southern

African subregion. Global Ecology and Biogeography Letters, 3, 41–47.

Fabre, P.-H., Hautier, L., Dimitrov, D. & Douzery, E. J. P.

(2012). A glimpse on the pattern of rodent diversification: a phy-

logenetic approach. BMC Evolutionary Biology, 12: 88.Fadda, C., Corti, M. & Verheyen, E. (2001). Molecular phylogeny

of Myomys/Stenocephalemys complex and its relationships with

related African genera. Biochemical Systematics and Ecology, 29,585–596.

Faulkes, C. G., Verheyen, E., Verheyen, W., Jarvis, J. U. M. &

Bennett, N. C. (2004). Phylogeographical patterns of genetic

divergence and speciation in African mole-rats (Family: Bathy-

ergidae). Molecular Ecology, 13, 613–629.Faulkes, C. G., Mgode, G. F., Le Comber, S. C. & Bennett, N. C.

(2010). Cladogenesis and endemism in Tanzanian mole-rats,

genus Fukomys: (Rodentia Bathyergidae): a role for tectonics?

Biological Journal of the Linnean Society, 100, 337–352.Fraley, C., Raftery, A. E., Murphy, T. B. & Scrucca, L. (2012)

mclust Version 4 for R: Normal Mixture Modeling for Model-

Based Clustering, Classification, and Density Estimation. Tech-

nical Report No. 597, Department of Statistics, University of

Washington.

Galan, M., Pages, M. & Cosson, J.-F. (2012). Next-generation

sequencing for rodent barcoding: species identification from

fresh, degraded and environmental samples. PLoS One, 7:e48374.

Galster, S., Burgess, N. D., Fjelds"a, J., Hansen, L. A. & Rahbek,

C. (2007). One Degree Resolution Databases of the Distribution of1085 Species of Mammals in Sub-Saharan Africa. On-line data

source-Version 1.00. Denmark: Zoological Museum, University

of Copenhagen. Available via http://www.daim.snm.ku.dk/afri-

can-vertebrates-search

Geraads, D. (2001). Rodents of the Upper Miocene of Chorora,

Ethiopia: Murinae, Dendromurinae and conclusions. Palaeoverte-brata (Montpellier), 30, 89–109.

Gernhard, T. (2008). The conditioned reconstructed process. Jour-nal of Theoretical Biology, 253, 769–778.

Gordon, D. H. & Rautenbach, I. L. (1980). Species complexes in

medically important rodents – chromosome studies of Aethomys,Tatera and Saccostomus (Rodentia, Muridae, Cricetidae). SouthAfrican Journal of Science, 76, 559–561.

Gordon, D. H. & Watson, C. R. B. (1986). Identification of cryptic

species of rodents (Mastomys, Aethomys, Saccostomus) in the Kruger

National Park. South African Journal of Zoology, 21, 95–99.de Graaff, G. (1981). The Rodents of Southern Africa. Durban: But-

terworths.

Hasegawa, M., Kishino, H. & Yano, T. A. (1985). Dating of the

Human-Ape splitting by a molecular clock of mitochondrial

DNA. Journal of Molecular Evolution, 22, 160–174.Heller, E. (1910). Five new rodents from British East Africa.

Smithsonian Miscellaneous Collections, 54, 3.Heller, E. (1912). New rodents from British East Africa. Smithso-nian Miscellaneous Collections, 59, 14–15.

Hijmans, R. J. (2014). raster: Geographic data analysis and model-

ing. R package version 2.2-31.

Hijmans, R. J., Cameron, S. E., Parra, J. L., Jones, P. G. & Jarvis, A.

(2005). Very high resolution interpolated climate surfaces for glo-

bal land areas. International Journal of Climatology, 25, 1965–1978.Horn, A., Basset, P., Yannic, G., Banaszek, A., Borodin, P. M.,

Bulatova, N. S., Jadwiszczak, K., Jones, R. M., Polyakov, A. V.,

Ratkiewicz, M., Searle, J. B., Shchipanov, N. A., Zima, J. &

Hausser, J. (2012). Chromosomal rearrangements do not seem

to affect the gene flow in hybrid zones between karyotypic races

of the Common Shrew (Sorex araneus). Evolution, 66, 882–889.Hubert, B. (1978). Revision of the genus Saccostomus (Rodentia,

Cricetomyinae), with new morphological and chromosomal data

from specimens from the lower Omo Valley, Ethiopia. Bulletinof Carnegie Museum of Natural History, 6, 48–52.

Jansa, S. A. & Weksler, M. (2004). Phylogeny of muroid rodents:

relationships within and among major lineages as determined by



103


IRBP gene sequences. Molecular Phylogenetics and Evolution, 31,256–276.

Johnson, J. B. & Omland, K. S. (2004). Model selection in ecology

and evolution. Trends in Ecology & Evolution, 19, 101–108.Kass, R. E. & Raftery, A. E. (1995). Bayes Factors. Journal of theAmerican Statistical Association, 90, 773–795.

Keesing, F. (1998a). Impacts of ungulates on the demography and

diversity of small mammals in central Kenya. Oecologia, 116,381–389.

Keesing, F. (1998b). Ecology and behavior of the pouched mouse,

Saccostomus mearnsi, in central Kenya. Journal of Mammalogy, 79,919–931.

Keesing, F. & Young, T. P. (2014). Cascading consequences of the

loss of large mammals in an African savanna. BioScience, 64, 487–495.

Keller, C., Roos, C., Groeneveld, L. F., Fischer, J. & Zinner, D.

(2010). Introgressive hybridization in southern African baboons

shapes patterns of mtDNA variation. American Journal of PhysicalAnthropology, 142, 125–136.

Kraemer, N. & Sugiyama, M. (2011). The degrees of freedom of

partial least squares regression. Journal of the American StatisticalAssociation, 106, 697–705.

Krzanowski, W. J. (1987). Cross-validation in principal component

analysis. Biometrics, 43, 575–584.Lange, M., Lautenbach, S. & Raedig, C. (2012). sperich: Auxiliary

functions to estimate centers of biodiversity. R package version

1.5-5.

Lecompte, E., Granjon, L., Peterhans, J. K. & Denys, C. (2002).

Cytochrome b-based phylogeny of the Praomys group (Rodentia,

Murinae): a new African radiation? Comptes Rendus Biologies, 325,827–840.

Librado, P. & Rozas, J. (2009). Dnasp v5: A software for compre-

hensive analysis of DNA polymorphism data. Bioinformatics, 25,1451–1452.

Makundi, R. H., Massawe, A. W., Mulungu, L. S. & Katakweba,

A. (2010). Species diversity and population dynamics of rodents

in a farm-fallow field mosaic system in central Tanzania. AfricanJournal of Ecology, 48, 313–320.

Maputla, N. W. (2008). Taxonomic status of Saccostomus cam-pestris (Rodentia: Cricetomyinae) from southern Africa: a multi-

disciplinary approach. MSc thesis, Pretoria: University of

Pretoria.

Maputla, N. W., Dempster, E. R., Raman, J. & Ferguson, J. W.

H. (2011). Strong hybrid viability between two widely divergent

chromosomal forms of the pouched mouse. Journal of Zoology,285, 180–187.

McDonough, M., �Sumbera, R., Mazoch, V., Ferguson, A. W.,

Phillips, C. D. & Bryja, J. (2015). Multilocus phylogeography

of a widespread savanna-woodland adapted rodent reveals the

influence of Pleistocene geomorphology and climate change in

Africa’s Zambezi region. Molecular Ecology. 24, 5248–5266.Mein, P. & Pickford, M. (2006). Late Miocene micromammals

from the Lukeino formation (6.1 to 5.8 Ma), Kenya. BulletinMensuel de la Societe Linneenne de Lyon, 75, 183–223.

Mein, P., Pickford, M. & Senut, B. (2004). Late Miocene micro-

mammals from the Harasib karst deposits, Namibia. Part 2b –

Cricetomyidae, Dendromuridae and Muridae, with an addendum

on the Myocricetodontidae. Communications of the GeologicalSurvey of Namibia, 13, 43–61.

de Menocal, P. B. (2004). African climate change and faunal evolu-

tion during the Pliocene-Pleistocene. Earth and Planetary ScienceLetters, 220, 3–24.

Metz, M. R. & Keesing, F. (2001). Dietary choices of the pouched

mouse (Saccostomus mearnsi) in central Kenya. Biotropica, 33, 182–187.

Miller, G. S. (1910). A new rodent of the genus Saccostomus from

British East Africa. Smithsonian Miscellaneous Collections, 54, 1–3.Montgelard, C. & Matthee, C. A. (2012). Tempo of genetic diver-

sification in southern African rodents: the role of Plio-Pleisto-

cene climatic oscillations as drivers for speciation. ActaOecologica-International Journal of Ecology, 42, 50–57.

Moore, A. E., Cotterill, F. P. D., Main, M. P. L. & Williams, H.

B. (2007). The Zambezi River. In A. Gupta (Ed.) Large Rivers:Geomorphology and Management (pp. 311–332). New-York:

Wiley.

Moore, A. E., Blenkinsop, T. & Cotterill, F. P. D. (2012).

Dynamic evolution of the Zambezi-Limpopo watershed, central

Zimbabwe. South African Journal of Geology, 115, 551–560.Musser, G. G. & Carleton, M. D. (2005). Superfamily Muroidea.

In D. E. Wilson & D.-A. M. Reeder (Eds) Mammal Species of theWorld. A Taxonomic and Geographic Reference (3rd edn, Vol. 2,

pp. 894–1531). Baltimore: John Hopkins University Press.

Nicolas, V., Granjon, L., Duplantier, J.-M., Cruaud, C. &

Dobigny, G. (2009). Phylogeography of spiny mice (genus

Acomys, Rodentia: Muridae) from the south-western margin of

the Sahara with taxonomic implications. Biological Journal of theLinnean Society, 98, 29–46.

Nychka, D., Furrer, R. & Sain, S. (2015). fields: Tools for spatial

data. R package version 8.2-1.

Oba, S., Sato, M., Takemasa, I., Monden, M., Matsubara, K. &

Ishii, S. (2003). A Bayesian missing value estimation method for

gene expression profile data. Bioinformatics, 19, 2088–2096.O’Leary, M. A. & Kaufman, S. G. (2012). MorphoBank 3.0: Web

application for morphological phylogenetics and taxonomy.

Available via http://www.morphobank.org.

Otto-Bliesner, B. L., Marsha, S. J., Overpeck, J. T., Miller, G. H.,

Hu, A. X. & Mem, C. L. I. P. (2006). Simulating Arctic climate

warmth and icefield retreat in the last interglaciation. Science,311, 1751–1753.

Paradis, E., Claude, J. & Strimmer, K. (2004). APE: analyses of

phylogenetics and evolution in R language. Bioinformatics, 20,289–290.

Peter, B. & Slatkin, M. (2013). Detecting range expansion from

genetic data. Evolution, 67, 3274–3289.Peters, W. C. H. (1846). Bericht €uber die zur BekanntmachungGeeigneten Verhandlungen (pp. 257–258). Berlin: K€oniglichen

Preussien Akademie der Wissenschaften.

Peters, W. C. H. (1852). Reise Nach Mossambique. I. S€augethiere(pp. 167–168). Berlin: Druck & Verlag.

Petter, F. (1966). Affinites des genres Beamys, Saccostomus et Crice-tomys (Rongeurs, Cricetomyinae). Annales Mus ee Royal de l’Afri-que Centrale, ser. 8 (Sciences Zoologiques), 144, 13–25.

Piry, S., Guivier, E., Realini, A. & Martin, J. F. (2012). |SE|S|AM|

E| Barcode: NGS-oriented software for amplicon characteriza-

tion – application to species and environmental barcoding.

Molecular Ecology Resources, 12, 1151–1157.Posada, D. & Crandall, K. A. (1998). MODELTEST: testing the

model of DNA substitution. Bioinformatics, 14, 817–818.




104


Potts, R. (2013). Hominin evolution in settings of strong environ-

mental variability. Quaternary Science Reviews, 73, 1–13.

R Core Team (2015). R: A Language and Environment for Statistical

Computing. Vienna, Austria: R Foundation for Statistical Com-

puting. Available via http://www.R-project.org

Raıche, G. (2010). nFactors: An R package for parallel analysis and

non graphical solutions to the Cattell scree test. R package

version 2.3.3.

Raıche, G., Walls, T. A., Magis, D., Riopel, M. & Blais, J.-G.

(2013). Non-graphical solutions for Cattell’s scree test. Methodol-

ogy-European Journal of Research Methods for the Behavioral and

Social Sciences, 9, 23–29.

Rambaut, A., Suchard, M. A., Xie, D. & Drummond, A. J. (2014)

Tracer v1.6. Available from http://beast.bio.ed.ac.uk/Tracer

Rautenbach, I. L. (1982). Mammals of the Transvaal. Pretoria: Eco-

plan.

Reed, D. N. (2007). Serengeti micromammals and their implica-

tions for Olduvai paleoenvironments. In R. Bobe, Z. Alemseged

& A. K. Behrensmeyer (Eds) Hominin Environments in the East

African Pliocene: An Assessment of the Faunal Evidence (pp. 217–

255). Book Series: Vertebrate Paleobiology and Paleoanthropol-

ogy. Dordrecht: Springer.

Roberts, A. (1914). Supplementary list of African mammals in the

collection of the Transvaal Museum, with descriptions of some

new species. Annals of the Transvaal Museum, 4, 183.

Roberts, A. (1938). Descriptions of new forms of mammals. Annals

of the Transvaal Museum, 19, 240.

Roberts, A. (1951). The Mammals of South Africa (700 pp.). Johan-

nesburg: Trustees of “The mammals of South Africa” Book

Fund.

Rohlf, F. J. & Slice, D. (1990). Extensions of the Procrustes

method for the optimal superimposition of landmarks. Systematic

Zoology, 39, 40–59.

Ronquist, F., Teslenko, M., van der Mark, P., Ayres, D. L., Dar-

ling, A., Hohna, S., Larget, B., Liu, L., Suchard, M. A. &

Huelsenbeck, J. P. (2012). MrBayes 3.2: Efficient Bayesian phy-

logenetic inference and model choice across a large model space.

Systematic Biology, 61, 539–542.

Ryan, J. M. (1989). Evolution of cheek pouches in African pouched

rats (Rodentia: Cricetomyinae). Journal of Mammalogy, 70, 267–

274.

Schenk, J. J., Rowe, K. C. & Steppan, S. J. (2013). Ecological

opportunity and incumbency in the diversification of repeated

continental colonizations by muroid rodents. Systematic Biology,

62, 837–864.

Schwann, H. (1906). A list of the mammals obtained by Messrs. R.

B. Woosnam and R. E. Dent in Bechuanaland. Proceedings of the

Zoological Society, London, 1906, 110.

Schwarz, G. E. (1978). Estimating the dimension of a model.

Annals of Statistics, 6, 461–464.

Skinner, J. D. & Chimimba, C. T. (2005). The Mammals of the

Southern African Subregion (3rd edn). Cambridge: Cambridge

University press.

Smithers, R. H. N. (1971). The mammals of Botswana. Museum

Memoir, National Museums of Rhodesia, Salisbury, 4, 1–340.

Stacklies, W., Redestig, H., Scholz, M., Walther, D. & Selbig, J.

(2007). pcaMethods – A Bioconductor package providing PCA

methods for incomplete data. Bioinformatics, 23, 1164–1167.

Stamatakis, A. (2014). RAxML version 8: A tool for phylogenetic

analysis and post-analysis of large phylogenies. Bioinformatics, 30,

1312–1313.

Stanhope, M. J., Czelusniak, J., Si, J.-S., Nickerson, J. & Good-

man, M. (1992). A molecular perspective on Mammalian evolu-

tion from the gene encoding interphotoreceptor retinoid

binding protein, with convincing evidence for bat monophyly.

Molecular Phylogenetics and Evolution, 1, 148–160.

Stankiewicz, J. & de Wit, M. J. (2006). A proposed drainage evolu-

tion model for Central Africa – did the Congo flow east? Journal

of African Earth Sciences, 44, 75–84.

Stephens, M., Smith, N. J. & Donnelly, P. (2001). A new statistical

method for haplotype reconstruction from population data.

American Journal of Human Genetics, 68, 978–989.

Steppan, S. J., Adkins, R. M. & Anderson, J. (2004). Phylogeny

and divergence-date estimates of rapid radiations in muroid

rodents based on multiple nuclear genes. Systematic Biology, 53,

533–553.

Swynnerton, G. H. & Hayman, R. W. (1951). A checklist of the

land mammals of the Tanganyika Territory and the Zanzibar

Protectorate. Journal of the East African Natural History Society,

20, 274–392.

Tajima, F. (1989). Statistical method for testing the neutral

mutation hypothesis by DNA polymorphism. Genetics, 123,

585–595.

Tavar!e, S. (1986). Some probabilistic and statistical problems in

the analysis of DNA sequences. Lectures on Mathematics in the

Life Sciences (American Mathematical Society), 17, 57–86.

Taylor, P. J., Stoffberg, S., Monadjem, A., Schoeman, M. C., Bay-

liss, J. & Cotterill, F. P. D. (2012). Four new bat species (Rhi-

nolophus hildebrandtii complex) reflect Plio-Pleistocene divergence

of dwarfs and giants across an afromontane archipelago. PLoS

One, 7, e41744.

Teeling, E. C., Scally, M., Kao, D. J., Romagnoli, M. L., Springer,

M. S. & Stanhope, M. J. (2000). Molecular evidence regarding

the origin of echolocation and flight in bats. Nature, 403, 188–

192.

Terryn, L., Wendelen, W., Leirs, H., Lenglet, G. & Verheyen, E.

(2007). African Rodentia, Available via http://projects.biodiver-

sity.be/africanrodentia

Thomas, O. (1897). Exhibition of small mammals collected by Mr.

Alexander Whyte during his expedition to the Nyika plateau and

Masuka Mountains, NR Nyasa. Proceedings of the Zoological Soci-

ety, London, 1897, 431.

Thomas, O. & Hinton, M. A. C. (1923). On mammals collected

by Captain Shortridge during the Percy Sladen and Kaffrarien

museum expedition to the Orange River. Proceedings of the

Zoological Society, London, 1923, 495–496.

Trauth, M. H., Maslin, M. A., Deino, A. L., Junginger, A., Leso-

loyia, M., Odada, E. O., Olagof, O., Olakaa, L. A., Streckera,

M. R. & Tiedemanng, R. (2010). Human evolution in a variable

environment: the amplifier lakes of Eastern Africa. Quaternary

Science Reviews, 29, 2981–2988.

Werdelin, L. & Sanders, W. J., Eds. (2010). Cenozoic Mammals of

Africa. Berkeley & London: University of California Press.

Winkler, A. J. (1997). Systematics, paleobiogeography, and pale-

oenvironmental significance of rodents from the Ibole member,

Manonga valley, Tanzania. Topics in Geobiology, 14, 311–332.



105


Winkler, A. J. (2002). Neogene paleobiogeography and east African

paleoenvironments: contributions from the Tugen hills rodents

and lagomorphs. Journal of Human Evolution, 42, 237–256.

Winkler, A. J., Denys, C. & Avery, D. M. (2010). Rodentia. In L.

Werdelin & W. J. Sanders (Eds) Cenozoic Mammals of Africa (pp.

263–304). Berkeley & London: University of California Press.

de Winton, W. E. (1897). On collections of rodents made by Mr.

J. ffolliott Darling in Mashunaland and Mr. F. C. Selous in

Matabeleland, with short field-notes by the collectors. Proceedings

of the Zoological Society, London, 1896, 804.

de Winton, W. E. (1898). On the nomenclature and distribution

of some of the rodents of South Africa, with descriptions of new

species. Annals and Magazine of Natural History, 7, 6.

Supporting Information

Additional Supporting Information may be found in the

online version of this article:

Appendix S1. Complete lists of material used in molecu-

lar and morphological analyses.

Appendix S2. Input file for divergence dating in

BEAST.

Appendix S3. A priori classification of morphometric

specimens.

Appendix S4. Lectotype of Saccostomus fuscus..




107


Multilocus phylogeny of East African

gerbils (Rodentia, Gerbilliscus)

illuminates the history of the Somali-

Masai savanna

Aghová, T., umbera, R., Piálek, L., Mikula, O., McDonough,

M.M., Lavrenchenko, L.A., Meheretu, Y., Mbau, J.S., Bryja,

J., 2017. Journal of Biogeography, 44(10), 2295–2307.

https://doi.org/10.1111/jbi.13017

Study design: JB, R

Collecting material: TA, R , OM, MM, LL, YM, JM, JB

Laboratory analysis: TA, LP, MM

Data analysis: TA, OM

Writing: TA, R , JB


Paper

III

109


ORIGINALARTICLE

Multilocus phylogeny of East Africangerbils (Rodentia, Gerbilliscus)illuminates the history of theSomali-Masai savanna

Tatiana Aghov�a1,2* , Radim Sumbera3, Lubom�ır Pi�alek3,

Ond rej Mikula1,4, Molly M. McDonough5,6, Leonid A. Lavrenchenko7,

Yonas Meheretu8, Judith S. Mbau9 and Josef Bryja1,2

1Institute of Vertebrate Biology of the Czech

Academy of Sciences, 603 65 Brno, Czech

Republic, 2Department of Botany and

Zoology, Faculty of Science, Masaryk

University, 602 00 Brno, Czech Republic,3Department of Zoology, Faculty of Science,

University of South Bohemia, 370 05 �Cesk e

Bud�ejovice, Czech Republic, 4Institute of

Animal Physiology and Genetics of the Czech

Academy of Sciences, 602 00 Brno, Czech

Republic, 5Center for Conservation Genomics,

Smithsonian Conservation Biology Institute,

National Zoo, Washington, DC 20008, USA,6Department of Vertebrate Zoology, National

Museum of Natural History, Smithsonian

Institution, Washington, DC 20560-0108,

USA, 7A.N. Severtsov Institute of Ecology and

Evolution RAS, 119081 Moscow, Russia,8Department of Biology, College of Natural

and Computational Sciences, Mekelle

University, Mekelle, Tigray, Ethiopia,9Department of Land Resource Management

and Agricultural Technology, College of

Agriculture and Veterinary Sciences,

University of Nairobi, Nairobi, Kenya

*Correspondence: Tatiana Aghov�a, Institute of

Vertebrate Biology ASCR, Research Facility

Studenec, 675 02 Studenec, Czech Republic.

Current address: Department of Zoology,

National Museum, 115 79 Prague, Czech

Republic.

E-mail: [email protected]

Dedication

We dedicate this paper to the late Bill Stanley

who devoted much of his career to the study

of small mammals in East Africa. Bill

facilitated acquiring numerous samples used in

this study and the promise of future

collaborations was cut short as Bill passed

away unexpectedly during his field trip in

Ethiopia in 6 October 2015.

ABSTRACT

Aim The rodent genus Gerbilliscus is widespread in savannas throughout sub-

Saharan Africa. The eastern clade comprises four species with distributions cen-

tred in the Somali-Masai biogeographical region of East Africa. We investigated

the genetic diversity of the group with a view to illuminating the historical

(Plio-Pleistocene) processes that formed contemporary biota of the understud-

ied Somali-Masai region.

Location Somali-Masai savanna, East Africa.

Methods We performed multilocus genetic analyses of 240 samples from 112

localities, combining genotyping of recently collected samples (N = 145), 454-

pyrosequencing of museum material (N = 34) and published sequences

(N = 61). We used Bayesian and maximum likelihood approaches for phyloge-

netic reconstructions, and coalescent-based methods to delimit species. We also

estimated divergence times and modelled recent and past distributions to

reconstruct the major evolutionary influences in the Somali-Masai region dur-

ing the Plio-Pleistocene.

Results Genetic analyses provided evidence for six lineages, possibly corre-

sponding to distinct species. The two main species groups (with two and four

putative species, respectively) have overlapping distributions, but species within

each group are distributed parapatrically. The origin of the eastern clade dates

back to the Pliocene, while individual species diverged in the Pleistocene. The

distribution of genetic diversity and ecological niche modelling point to the

importance of the Rift Valley and the presence of unsuitable xeric habitats in

the allopatric diversification of Gerbilliscus in the Somali-Masai savanna within

the last 5 Myr.

Conclusions This is the first detailed phylo(bio-)geographical study of ani-

mals with predominant distribution in the Somali-Masai region. It revealed

currently underestimated diversity of eastern clade of Gerbilliscus and proposed

a scenario of its evolution during Plio-Pleistocene. Conspicuous genetic struc-

ture of these taxa can be now used to test detailed phylogeographical hypothe-

ses related to Plio-Pleistocene history of gerbils and, to some extent, also biota

of Somali-Masai bioregion in general.

Keywords

biogeography, Gerbillinae, historical DNA, murid rodents, phylogeography,

Plio-Pleistocene climate change, pyrosequencing, Rift Valley, species delimita-

tion, tropical Africa

ª 2017 John Wiley & Sons Ltd http://wileyonlinelibrary.com/journal/jbi 2295doi:10.1111/jbi.13017

Journal of Biogeography (J. Biogeogr.) (2017) 44, 2295–2307

110


INTRODUCTION

Tropical grasslands, savannas and shrublands are the most

widespread terrestrial habitats in sub-Saharan Africa (Sayre

et al., 2013). These ecosystems, hosting some of the most

abundant and diverse mammalian communities on Earth,

are threatened by increased human activities (Craigie et al.,

2010). Within the continent they can be divided into four

main bioregions (sensu Linder et al., 2012): South African,

Zambezian, Sudanian and Somali-Masai, with the last one

corresponding to the ‘Somalia-Masai’ region, originally pro-

posed by White (1983) as a distinct dry phytochorion,

extending from Eritrea to Tanzania (Fig. 1). International

Vegetation Classification (IVC; naturserve.org; Faber-Langen-

doen et al., 2014) identifies this division as Eastern Africa

Xeric Scrub and Grassland with four ecoregions: (1) Somali,

(2) Southern, (3) Northern Acacia-Commiphora bushlands

and thickets and (4) Masai xeric grasslands and shrublands

(Dixon et al., 2014), referred to herein as Somali-Masai

savanna (Fig. 1). The major geomorphological feature here is

the East African Rift system which has had a long-term

influence on the climatic regime in Africa (Sepulchre et al.,

2006). While this region has relatively low floristic and fau-

nistic species diversity, a high level of endemism exists for

plants (Thulin, 1993), reptiles (Burgess et al., 2007) and

rodents (Varshavsky et al., 2007). This region is home to the

Horn of Africa biodiversity hotspot, which ranks among the

oldest and most stable arid regions of Africa (Kingdon,

1990). In addition, it is one of the least known African

bioregions and possibly also one of the most threatened due

to the rapid increase in human populations and climatic

changes (Geist & Lambin, 2004).

Despite the biogeographical uniqueness of the Somali-

Masai region, genetic data on organisms living here are

scarce and fragmentary – usually included as part of large-

scale phylogenetic reconstructions with under-representative

geographical sampling. Previous studies found that taxa from

this region are phylogenetically distinct, thereby providing

Figure 1 Distribution of species and intraspecific haplogroups of Gerbilliscus in East Africa. Species of the eastern clade are represented

with different colours and intraspecific haplogroups are indicated by different symbol shapes. Empty squares represent genetically

confirmed records of Gerbilliscus species from the western clade, empty circles those from the southern clade (sensu Granjon et al.,

2012), and black crosses indicate localities where Gerbilliscus was not captured during our recent field trappings. The map is based on

data from Colangelo et al. (2007, 2010), McDonough et al. (2015), this study and our unpublished data of DNA-barcoded specimens

from western and southern clades. Orange background indicates the extent of Somali-Masai savanna, i.e. the distribution of Somali,

Northern and Southern Acacia-Commiphora bushlands and thickets (based on WWF records http://www.eoearth.org/). [Colour figure

can be viewed at wileyonlinelibrary.com]

Journal of Biogeography 44, 2295–2307

ª 2017 John Wiley & Sons Ltd

2296

T. Aghov�a et al.

111


evidence that the Somali-Masai biota evolved in isolation for

an extended period of time (e.g. �Sm ıd et al., 2013; Mikula

et al., 2016). Noteworthy examples include the plant Wajira

(Thulin et al., 2004), Grevy’s zebra (Kebede et al., 2016) and

warthogs (Randi et al., 2002). However, detailed phylogeo-

graphical studies of the structure and evolutionary history of

genetic lineages living predominantly in the Somali-Masai

bioregion are lacking.

Eastern Africa experienced a turbulent climatic and geo-

morphological history during the Plio-Pleistocene. The evolu-

tion of its biota was influenced by several major clima-

tic transitions (Maslin & Christensen, 2007), including the

intensification of the Northern Hemisphere glaciation (3.2–

2.5 Ma), the development of the Walker circulation

(2.0–1.7 Ma), and the early-middle Pleistocene transition

(1.2–0.8 Ma) (Maslin & Christensen, 2007; Trauth et al.,

2009). In particular, the tropical oscillations between warm

and relatively wet pluvials and cooler and drier interpluvials

during the Pleistocene resulted in repeated expansions

and contractions of either savanna-like or forest-like habitats

(e.g. Cowling et al., 2008). These oscillations, which roughly

correspond to interglacial and glacial periods in the northern

and southern latitudes (Dupont, 2011), significantly affected

diversification and speciation processes. Beside these global

transitions, several additional local factors, such as climate-

driven vegetation changes (Sepulchre et al., 2006; Cerling

et al., 2011; Uno et al., 2016), or the presence of the amplifier

lakes in the bottom of the Rift Valley (Trauth et al., 2010),

probably influenced the evolution of taxa adapted to eastern

African savannas, woodlands and grasslands. Unfortunately,

without studies of detailed organismal genetic structure, it is

difficult to evaluate the roles of climatic and geomorphologi-

cal factors (e.g. Trauth et al., 2010) in forming current

biological diversity in the area. Carefully selected model spe-

cies can provide valuable information for understanding our

own history, as numerous crucial localities for studies of

human evolution are located in eastern Africa (Maslin et al.,

2015).

Small mammals such as rodents are often good models

for phylogeographical reconstruction because they are habi-

tat specialists, exhibit low dispersal ability, and have rela-

tively high genetic substitution rates. Gerbils of the genus

Gerbilliscus (Thomas, 1897) are widespread rodents living

in the savannas, woodlands, grasslands and semi-deserts of

sub-Saharan Africa. The phylogeny of the genus was

recently investigated using karyology, morphology and

DNA sequences (e.g. Granjon et al., 2012). The mono-

phyletic eastern Gerbilliscus clade (also called ‘robustus

group’, but referred to here as the ‘eastern clade’) repre-

sents the first cladogenetic split within the genus and com-

prises four currently recognized species: G. nigricaudus

(Peters, 1987), G. vicinus (Peters, 1987), G. robustus (Cret-

zschmar, 1826) and G. phillipsi (deWinton, 1989) (sensu

Monadjem et al., 2015). Unlike the western (Granjon et al.,

2012) and southern (McDonough et al., 2015) clades, the

phylogeny and distribution of the eastern clade have never

been studied in detail. Previously published genetic data

have been limited to a few specimens genotyped at one

mitochondrial and one nuclear marker with low variability

(Colangelo et al., 2007) and morphological analyses were

geographically biased towards Tanzanian localities (Colan-

gelo et al., 2010).

Species distributions based on morphological identifica-

tions (Monadjem et al., 2015) indicate that the eastern clade

occurs within the borders of the Somali-Masai region and its

genetic analysis may provide insights into the evolutionary

processes that formed the contemporary biota of this region.

We analysed how gerbil divergence dates reflect the geomor-

phology and climate of the area. More specifically, we tested

if (1) local geomorphological features such as mountain

ranges and the Great Rift Valley with its amplifier lakes and

(2) periods of pronounced climatic variation in the Plio-

Pleistocene influenced genetic structure and divergence

within the eastern clade of Gerbilliscus.

MATERIALS AND METHODS

DNA extraction, PCR amplification and sequencing

Our genetic dataset includes sequences from recent collec-

tions (145 individuals/72 localities), museum samples (34/14)

and georeferenced sequences from Genbank (61/26) (Fig. 1;

for details see Table S1 in Appendix S1 in Supporting Infor-

mation). DNA from 96% ethanol-preserved tissue samples

was extracted using a DNeasy Blood & Tissue kit (Qiagen,

Hilden, Germany). For phylogenetic analysis we selected four

genetic markers; all specimens were sequenced for one mito-

chondrial gene, cytochrome b (CYTB), and a subset of indi-

viduals (1–5 individuals per mitochondrial lineage), for three

nuclear markers that have previously been used successfully

to resolve intrageneric phylogenies in mammals: Intron 7 of

the gene for b-fibrinogen (FGB) and exons of breast cancer

susceptibility gene (BRCA1) and Interphotoreceptor Binding

Protein gene (IRBP). For genotyping protocols see

Appendix S2. Museum samples (taken mostly from dry

skins) were pyrosequenced on a GS Junior (Roche, Basel,

Switzerland) using the CYTB mini-barcode protocol (Galan

et al., 2012; see Appendix S2). This method allows for the

separation of focal sequences in samples contaminated by

distantly related organisms (e.g. human DNA; Bryja et al.,

2014).

Phylogenetic analysis

Sequences were aligned using Mafft 7 (Katoh & Standley,

2013) and the supermatrix (4473 bp) was created in Sea-

View 4 (Gouy et al., 2010). The final dataset for phyloge-

netic analyses consisted of 185 unique sequences of CYTB,

25 sequences of BRCA1, 27 sequences of FGB and 26

sequences of IRBP. The remaining 55 CYTB sequences (iden-

tical and/or shorter sequences from the same or neighbour-

ing localities) were unambiguously assigned to particular



2297

Phylogeny of Somali-Masai gerbils

112


lineages in RAxML 7.2.8 (Stamatakis, 2006). These data were

used to increase the precision with which we mapped the

geographical distribution of clades and assigned type material

to particular genetic groups. As outgroups we used two

sequences of G. giffardi from western clade, which is the sis-

ter lineage to eastern Gerbilliscus (see Table S1 in

Appendix S1 for the GenBank numbers).

The best partitioning scheme and substitution models were

determined using PartitionFinder 1.1.1 (Lanfear et al.,

2014). Evolutionary relationships were estimated using maxi-

mum likelihood (ML) in RAxML 7.2.8 (Stamatakis, 2006)

and Bayesian inference (BI) in MrBayes 3.2.2 (Ronquist

et al., 2012) (see Appendix S2). For visualization of

intraspecific structure we also constructed haplotype net-

works for the two most numerous species, G. vicinus and G.

nigricaudus. Haplotype files were produced in DnaSP 5

(Librado & Rozas, 2009) and median-joining networks in

Network 4.6.1.3 (Bandelt et al., 1999).

Species delimitation

Our results indicated six lineages that might represent sepa-

rate species. We tested the distinctiveness of their gene pools

using the three nuclear loci in a Bayesian framework and

BP&P 3 (Yang & Rannala, 2014), which samples shared phy-

logenetic space and possible mergers of predefined candidate

species under assumptions of a multispecific coalescent

model (Degnan & Rosenberg, 2009). The analysis was run

with three combinations of gamma priors for root age (s0)

and effective population size (h). First, we used G(10,1300)

as an informative prior for s0, based on an assumption that

deeply divergent and broadly sympatric G. nigricaudus and

G. vicinus are heterospecific and s0 is approximately equal to

half of their mean divergence in terms of Jukes-Cantor dis-

tance (see Table S6-S8, Appendix S3). We have no specific

information about h and hence our prior was either diffuse,

G(2,300), and hence uninformative, or suggestive of large

values, G(10,1300) and hence conservative (tending to lump

the species). Finally, we used the diffuse prior for h in com-

bination with diffuse prior for s0, G(0.5,25). Dirichlet distri-

bution with a = 1 is fixed in the software as a prior for

internal branch lengths and we also used it as a prior for

between-gene variability in evolutionary rates. Each analysis

was run twice to check if it converged on a similar posterior

distribution.

Species tree

The species tree was calculated under the Bayesian frame-

work implemented in *BEAST package (Heled & Drum-

mond, 2010), an extension of BEAST 1.8.2 (Drummond

et al., 2012). Alignments for each of the four genes were

imported into BEAUti 1.8.2 where they were assigned sepa-

rate and unlinked substitution, clock and tree models.

Sequences were assigned to species as delimited by the best

model in BP&P. Two independent runs were carried out for

10 9 106 generations with sampling every 1000 generations

in beast. We discarded first 25% as burn-in and the result-

ing parameter and tree files were examined for convergence

and effective sample sizes (> 200) in Tracer 1.6 (Rambaut

et al., 2014). The two runs were combined in LogCombiner

1.8.2 and the species tree was visualized in DensiTree

(Bouckaert, 2010).

Molecular dating

The times to most recent common ancestors (TMRCA) were

inferred in BEAST 1.8.2 from a three-locus dataset (CYTB,

BRCA1, IRBP); sequences of FGB were excluded because of a

lack of sequenced outgroup taxa. The final dataset consisted

of 17 species of Gerbilliscus as well as Desmodillus auricularis

and Tatera indica, their closest relatives in the phylogeny of

gerbillines (Granjon et al., 2012). Sequences analysed are

listed in Table S2 in Appendix S1. We used partition-specific

substitution models (see Table S4 in Appendix S2), gene-spe-

cific lognormal relaxed clocks (Drummond et al., 2006) and

birth–death tree priors (Gernhard, 2008), as well as three fos-

sil calibrations: (1) Abudhabia pakistanensis from the Potwar

Plateau, Pakistan (Flynn & Jacobs, 1999) for the whole Ger-

billiscus-Tatera-Desmodillus clade (Denys & Winkler, 2015);

(2) Gerbilliscus sp. from Lemudong0o, Kenya (Manthi, 2007)

for Gerbilliscus and (3) Gerbilliscus sp. from Hadar, Ethiopia

(Sabatier, 1982) for the eastern clade. The calibration density

was always lognormal (Mean in real space = 1.6, SD = 0.9)

with an offset corresponding to the particular fossil age,

which is 8.6 million years (Ma) for Abudhabia (Flynn &

Jacobs, 1999), 6.1 Ma for the Lemudong0o Gerbilliscus

(Deino & Ambrose, 2007) and 3.3 Ma for the Hadar Gerbil-

liscus (Reed & Geraads 2012). For more details see

Appendix S2.


The present distribution of the East African Gerbilliscus clade

was predicted by maximum entropy (MaxEnt) modelling

(Phillips et al., 2006) interfaced with R computing environ-

ment by packages ‘dismo’ (Hijmans et al., 2016) and ‘ENMe-

val’ (Muscarella et al., 2014). Only CYTB-barcoded

specimens were included to avoid taxonomic confusion

resulting in 66 unique presence records (in 0.5° resolution).

We focused on climatic conditions as predictors of relative

occurrence rates (RORs). More specifically, we used 19 BIO-

CLIM variables downloaded from the WorldClim website

(Hijmans et al., 2005). Following corrected Akaike informa-

tion criterion (AICc) based model selection (Warren & Sei-

fert, 2011) we used linear and quadratic features

(transformations of original variables) and a regularization

coefficient of 1.68 (see Appendix S2 for details). Under these

parameters, RORs were predicted for all background sites as

well as the corresponding sites in palaeoclimatic layers pro-

vided by WorldClim for climatic conditions during the Last

Glacial Maximum (21 ka) and during the last interglacial



2298

T. Aghov�a et al.

113


(140–120 ka). These layers were produced as estimates from

global circulation models built by Braconnot et al. (2007) for

the last glacial and by Otto-Bliesner (2006) for the last inter-

glacial. The results were visualized using the R packages

‘maptools’ (Bivand & Lewin-Koh, 2016) and ‘raster’ (Hij-

mans, 2015). The importance of each predictor was quanti-

fied by its spatial randomization, calculating the Spearman

correlation (rs) of predictions before and after randomization

and subtracting it from unity.

RESULTS

Phylogeny, distribution and species delimitation of

East African Gerbilliscus

Both BI (Fig. 2) and ML (see Fig. S2, Fig. S5 in

Appendix S3) phylogenetic analyses recovered six main lin-

eages whose relationships were only partly resolved. The first

split separated the eastern Gerbilliscus into the nigricaudus

group (consisting of two lineages G. nigricaudus and G. cf.

bayeri; see Discussion for more details on the use of names

for particular lineages) and the robustus group (consisting of

four lineages G. vicinus, G. phillipsi, G. robustus, G. sp. n.

(Babile)). The distribution of the eastern clade is largely con-

cordant with the borders of the Somali-Masai region and

parapatric with respect to western and southern clades of

Gerbilliscus. The only exceptions are records of G. robustus in

Sudan and Chad, where they may be sympatric with species

of the western clade (Fig. 1).

The BP&P analysis supported all six main lineages as phylo-

genetically distinct entities (putative species) regardless of the

combination of priors (posterior probability > 0.98 in all

cases). Posterior samples of both s0 and mean h were substan-

tially narrower than the corresponding priors, largely coinci-

dent across analyses, and their median corresponded to 0.23–

0.40 quantile of prior density (see Fig. S3 in Appendix S3).

Thus, the support for distinctiveness of our candidate species

was neither due to some specific priors nor due to the sam-

pling getting stuck at an extreme combination of large diver-

gence times and low effective population sizes. The *BEAST

species tree of these six lineages recovered a fully resolved

topology showing that G. sp. n. (Babile) is the sister species of

G. robustus and G. phillipsi is sister to G. vicinus (Fig. 3).

Phylogenetic analysis (Fig. 2) as well as haplotype net-

works (see Fig. S1 in Appendix S3) revealed significant

genetic variation within the three species. We defined two

haplogroups for G. robustus, and four for both G. vicinus

and G. nigricaudus; genetic K2P distances among intraspeci-

fic haplogroups ranged from 0.036 to 0.100 (Table 1). In G.

robustus, one haplogroup occurred in the Rift Valley in

Ethiopia (R1), while the second was found in Chad and

Sudan (R2). In the two remaining species, the distribution

of intraspecific lineages was mostly parapatric, either sepa-

rated by the Rift Valley (e.g. V2 versus V1, V3,V4) or along

the north-south axis (e.g. N4-N2-N1-N3) (Fig. 2, see

Fig. S1 in Appendix S3).

Divergence time estimates

Divergence dating in BEAST estimated the TMRCA of the

genus Gerbilliscus in the Late Miocene (7.03 Ma; 95% HPD:

6.17–10.13; Fig. 4). TMRCA of the eastern clade, i.e. the split

between the nigricaudus and robustus groups, was estimated

as 4.17 Ma (95% HPD: 3.38–6.78) and the basal split in the

robustus group as 2.42 Ma (95% HPD: 1.50–4.30). Three

pairs of sister species diverged almost simultaneously about

1.70 Ma: G. robustus and G. sp. n. (Babile) at 1.62 Ma (95%

HPD: 0.73–2.95), G. vicinus and G. phillipsi at 1.75 Ma (95%

HPD: 0.73–3.31) and G. nigricaudus and G. cf. bayeri at

1.80 Ma (95% HPD: 0.86–3.53).


The suitability of climatic conditions for the eastern Gerbillis-

cus clade appears to have deteriorated from the last inter-

glacial, through the last glacial to the present (Fig. 5).

According to our model the Horn of Africa was more suitable

in the last interglacial, and less so from the last glacial maxi-

mum to the present. In Tanzania, on the southern border of

the Somali-Masai area (see Fig. 1) the suitability of condi-

tions also fluctuated. Nevertheless, in the core of the present

day distribution (Somali-Masai savanna on both sides of the

Rift Valley) climate was apparently favourable during the last

glacial cycle. The most important variable affecting the mod-

elled distribution was precipitation during the wettest quarter

(importance = 0.76), but the effect of mean temperature dur-

ing the coldest quarter, precipitation during the wettest

month and annual precipitation were also significant (impor-

tance ≥ 0.20; see Fig. S4 in Appendix S3 for details).

DISCUSSION

Cryptic diversity and distribution of the eastern

clade of Gerbilliscus

In this study, we used multi-locus genetic data to reconstruct

the evolutionary history of the eastern African clade of ger-

bils, genus Gerbilliscus, distributed mostly in the Somali-

Masai region. Because of a previous lack of sampling and

genetic data, precise distributions and phylogenetic relation-

ships of four currently recognized species in this clade (i.e.

G. nigricaudus, G. robustus, G. vicinus, G. phillipsi; sensu

Colangelo et al., 2005) remained obscure. Using phylogenetic

and species delimitation approaches we evaluated the status

of these four species, analysed their intraspecific variability

and provided details on their distributions.

In accordance with previous work (Colangelo et al., 2005,

2007; Granjon et al., 2012), we identified two lineages that

may represent separate but unrecognized species. The first,

Gerbilliscus sp. n. (Babile), was first reported as a genetically

distinct lineage (12.5–17.6% CYTB genetic distance from

other species in the eastern clade; see Table 1) from the

Babile Elephant Sanctuary in eastern Ethiopia (Lavrenchenko



2299


114


Figure 2 Bayesian inference phylogeny for East African Gerbilliscus based on concatenated dataset, i.e. supermatrix of one mitochondrial

(CYTB) and three nuclear loci (FGB, BRCA1, IRBP). Black points indicate nodes with posterior probability support > 0.95. [Colour

figure can be viewed at wileyonlinelibrary.com]



2300

T. Aghov�a et al.

115


et al., 2010). We captured additional individuals at the same

site, and increased our sample by adding museum specimens

from additional localities in Ethiopia and Somalia. A coales-

cence-based species tree demonstrated that G. sp. n. (Babile)

is a sister lineage to G. robustus, from which it seems to be

separated by a mountain ridge in the eastern Ethiopian high-

lands (Fig. 1). The second currently unrecognized species we

referred to as G. cf. bayeri. We sequenced a syntype of G.

nigricaudus bayeri (RMCA5183-M) from Maroon River in

western Kenya, and it clustered within a clade distributed

west of the Rift Valley, as sister to G. nigricaudus. Based on

the results of a coalescent species delimitation and high

CYTB genetic distances (13.4–15.1% from G. nigricaudus),

this clade may deserve the status of a distinct species status

(see also Bates, 1988). Taxonomic revision of the group,

including morphological analyses and formal descriptions of

these taxa, is currently in preparation and will be published

elsewhere (M.M. McDonough et al., unpublished data).

Three taxa, G. sp. n. (Babile), G. phillipsi and G. cf. bayeri,

exhibited relatively small geographical distributions (Fig. 1),

although this could be a result of limited sampling in some

regions. For example, G. sp. n. (Babile) may be more wide-

spread in southeastern Ethiopia and part of Somalia, and

museum records of G. robustus or G. phillipsi from Somalia

(Monadjem et al., 2015) may in fact represent this species

(M.M. McDonough et al., unpublished data). On the other

hand, present records of G. sp. n. (Babile) suggest its occur-

rence is associated with transitional semi-evergreen bushland

(van Breugel et al., 2016), which is localized only in the nar-

row belt along mountain chains in Ethiopia. Gerbilliscus cf.

bayeri has a limited distribution in western Kenya and South

Sudan (and potentially also in northern Uganda). On the

Figure 3 Coalescent based species tree for Gerbilliscus in East

Africa generated using *BEAST, visualized by DensiTree.

Posterior probability of nodes is indicated in boxes.

Figure 4 Ultrametric Bayesian tree generated using BEAST with dated divergences for East African Gerbilliscus (i.e. TMRCAs) and 95%

highest posterior densities (HPDs). Stars indicate the position of palaeontological records used for calibration of the tree.



2301


116


other hand, the three remaining species (G. robustus, G. vici-

nus, G. nigricaudus) have large distributions. Conspicuous

intraspecific structure of these species (e.g. Fig. S1 in

Appendix S3) can be used to propose and test phylogeo-

graphical hypotheses related to the Plio-Pleistocene history

of gerbils and, to some extent, also the biota of the Somali-

Masai bioregion in general.

Evolutionary history of the eastern clade of

Gebilliscus and a scenario of diversification in the

Somali-Masai savanna

This study is the first, to our knowledge, to focus on the

detailed phylogeography of any group living predominantly

in the Somali-Masai savanna. We estimated the origin of the

genus Gerbilliscus as dating back to the late Miocene. Previ-

ous estimates of TMRCA varied, according the calibration

points and genetic markers used, between 6.3 and 8.5 Ma

(Chevret & Dobigny, 2005; Colangelo et al., 2005, 2007),

which is in agreement with our results. Based on our esti-

mated divergence dates, the Plio-Pleistocene evolution of the

eastern Gerbilliscus clade seems to have been influenced by

several major climatic events. Very little is known about the

effect of the Messinian salinity crisis on eastern African cli-

mate (Maslin & Christensen, 2007), but it is generally held

that overall aridification at the Miocene/Pliocene boundary

promoted the expansion of very dry habitats (Hodell et al.,

2001). Due to the fact that arid parts of the extant Somali-

Masai savanna (e.g. Masai-xeric shrubland) are likely to by

uninhabitable by Gerbilliscus (as is northern Kenya today;

Figs 1 & 5), we hypothesize that such aridification led to the

early-Pliocene split between the robustus group in the north

and the nigricaudus group in the south.

The intensification of Northern Hemisphere glaciation is

likely to have had similar effects, reflected in the African as

an increase in aridity c. 2.7 Ma (deMenocal, 2004). Our esti-

mates of divergence dates suggest that prolonged intervals of

wet-dry climatic oscillations (Potts, 2013) generated direc-

tional selection pressures driven by the expansion and con-

traction of grasslands. The oldest divergence within the

robustus group occurred around 2.5 Ma. Two Pleistocene

climatic transitions, the development of the Walker circula-

tion (2.0–1.7 Ma; Ravelo et al., 2004) and the early-middle

Pleistocene transition (1.2–0.8 Ma; Berger & Jansen, 1994),

also significantly affected the ecosystems in eastern Africa.

The Walker circulation increased the interannual variation in

rainfall and the early-mid Pleistocene transition prolonged

and intensified glacial-interglacial climatic cycles. Both transi-

tions also helped the spread of C4 plants (Cerling et al.,

1988). Both hominins (Homo ergaster and Homo erectus) and

a number of ungulate species associated with grassland habi-

tats first appeared between 2.0 and 1.9 Ma (Bobe, 2004). In

eastern Gerbilliscus, this period seems to be associated with

synchronous diversification of extant sister species (Fig. 4),

while the subsequent intensification of climatic cycles appar-

ently affected allopatric divergences within species with large

distribution areas (see e.g. Fig. S1 in Appendix S3).

Additionally, the local geomorphology played an impor-

tant role in evolution of the East African biota. For example,

Gerbilliscus robustus is separated from G. sp. n. (Babile) by a

mountain chain that apparently prevents the dispersal of

both taxa across high altitude areas with very different envi-

ronmental conditions (Fig. 1). One of most influential land-

scape features in the evolution of small mammals in the

Somali-Masai region is the Rift Valley (Trauth et al., 2010).

Rifting generated numerous depressions which were repeat-

edly filled with water, and which were highly sensitive to

changes in the local precipitation-evaporation regime (Maslin

et al., 2014). Rift basins, either filled with water or extremely

dry with halophytic vegetation (White, 1983) could prevent

the dispersal of small terrestrial animals (Trauth et al., 2010).

Gerbilliscus biogeography appears to reflect the influence of

the Rift Valley both from east to west (G. cf. bayeri versus G.

nigricaudus, G. nigricaudus N2 versus N4 and N1 versus N3,

Figure 5 MaxEnt results of (a) the present distribution of suitable habitats for East African Gerbilliscus clade as typical representatives

of Somali-Masai fauna (red dots indicate confirmed presence data); (b) MaxEnt prediction for last glacial maximum (21 ka). (c)

MaxEnt prediction for last interglacial period (120–140 ka). Green colour represent habitats that are suitable for the eastern

Gerbilliscus, while yellow-orange are not. [Colour figure can be viewed at wileyonlinelibrary.com]



2302

T. Aghov�a et al.

117


G. vicinus V2 versus V3) and from north to south (G. vicinus

versus G. phillipsi versus G. robustus; Fig. 1).

An interesting observation is the migration of G. robustus

R2 into Central Africa, because very few taxa originating in

the Somali-Masai region have spread westwards or south-

wards. Conversely, numerous species typically inhabiting the

Sudanian savanna have their origins in the east, including

rodents (e.g. Brouat et al., 2009; Dobigny et al., 2013) and

ungulates (Lorenzen et al., 2012). Our ecological niche mod-

elling (Fig. 5) suggests that conditions have been suitable,

especially during interglacials, for the spread of savanna spe-

cies into the belt of Sudanian region, although it is not clear

why other Somali-Masai taxa (e.g. Gr�evy’s zebra or pouched

mice) remained restricted to the easternmost part of Africa

(IUCN 2016, Mikula et al., 2016).

Gerbilliscus as a model for understanding the

evolution of the Somali-Masai savanna

According to White (1983), the Somali-Masai region com-

prises several vegetation types. The most widespread vegeta-

tion in eastern Africa including the Horn of Africa is Acacia-

Commiphora bushland and thicket (shown as orange back-

ground in Fig. 1; Dixon et al., 2014). Overlap between this

vegetation type and the distribution of eastern Gerbilliscus

spp. suggests that gerbils may be informative regarding the

Plio-Pleistocene evolution of this bioregion, although the

actual distribution of the genus may be more restricted than

can be predicted from vegetation maps.

The absence of Gerbilliscus in large areas of southeastern

Ethiopia, Somalia and northeastern Kenya may partly be

caused by a lack of sampling or of sampling at appropriate

times of the year. However, two indicators argue against this

interpretation. First, our recent sampling in southeastern

Ethiopia and northern Kenya (Fig. 1) failed to find any Ger-

billiscus, and there is also a dearth of museum specimens

from this area (Monadjem et al., 2015; ‘Free and Open

Access to Biodiversity Data|GBIF.Org’). These arid areas

belonging to the Masai-xeric shrubland ecoregion (sensu

White, 1983) have an annual rainfall of < 150 mm, and are

often almost devoid of vegetation, covered only by stones.

Small mammal communities are dominated by Acomys, Ger-

billus and Arvicanthis, rodents that can tolerate extreme arid-

ity. Second, our species distribution modelling predicts

(Fig. 5) that virtually all of the Horn of Africa represents

unsuitable habitat for Gerbilliscus, which should be explored

more extensively in the field. In the contrast, climatic condi-

tions in southern Somalia seem to support the occurrence of

Gerbilliscus (Fig. 5), which is in agreement with previous

biodiversity surveys (Bates, 1988; Varty, 1988).

Our study provides the first detailed genetic structure of

any mammal from the Somali-Masai savanna, including

identification of the main factors responsible for their evolu-

tionary history since the Miocene. The general validity of

evolutionary patterns observed in Gerbilliscus should now be

tested with comparative data from other taxa that haveTable

1K2-Pgenetic

distancesam

onglineagesofEastAfrican

Gerbilliscusclade(below

diagonal)calculatedfrom

mitochondrial

CYTBgenein

mega.Values

above

diagonal

represent

standarderrors.

G.vicinus

V1

G.vicinus

V2

G.vicinus

V3

G.vicinus

V4

G.phillipsi

G.robustus

R1

G.robustus

R2

G.sp.n.

(Babile)

G.nigricaudus

N1

G.nigricaudus

N2

G.nigricaudus

N3

G.nigricaudus

N4

G.cf.

bayeri

G.vicinusV1

0.010

0.009

0.008

0.018

0.019

0.022

0.019

0.021

0.022

0.023

0.023

0.030

G.vicinusV2

0.053

0.008

0.011

0.019

0.022

0.023

0.020

0.021

0.023

0.023

0.021

0.030

G.vicinusV3

0.043

0.044

0.010

0.018

0.021

0.023

0.021

0.023

0.024

0.024

0.023

0.029

G.vicinusV4

0.036

0.057

0.053

0.020

0.020

0.023

0.019

0.022

0.023

0.023

0.023

0.030

G.phillipsi

0.146

0.159

0.158

0.152

0.014

0.019

0.017

0.024

0.024

0.025

0.024

0.024

G.robustusR1

0.161

0.179

0.159

0.163

0.113

0.015

0.017

0.026

0.025

0.025

0.026

0.027

G.robustusR2

0.169

0.181

0.184

0.171

0.141

0.100

0.022

0.026

0.024

0.025

0.025

0.028

G.sp.n.(Babile)

0.156

0.170

0.153

0.161

0.144

0.125

0.176

0.022

0.024

0.021

0.022

0.023

G.nigricaudusN1

0.213

0.228

0.223

0.204

0.225

0.208

0.236

0.201

0.011

0.010

0.013

0.018

G.nigricaudusN2

0.201

0.226

0.216

0.207

0.238

0.230

0.231

0.224

0.075

0.014

0.010

0.020

G.nigricaudusN3

0.210

0.221

0.221

0.198

0.228

0.224

0.224

0.205

0.069

0.101

0.015

0.016

G.nigricaudusN4

0.217

0.227

0.224

0.215

0.232

0.220

0.235

0.214

0.080

0.054

0.099

0.018

G.cf.bayeri

0.242

0.263

0.254

0.239

0.219

0.221

0.220

0.225

0.134

0.151

0.139

0.136



2303


118


similar ecological requirements distributions (‘community

phylogeography; e.g. Hickerson & Meyer, 2008). Several

other small mammals associated with Somali-Masai savanna

habitats (e.g. spiny mice, genus Acomys; grass rats, Arvican-

this; meadow mice, Myomyscus; arid-adapted shrews, genus

Crocidura) are suitable models for such studies. The results

of a comparative study could provide robust conclusions

regarding biogeographical processes operating during the

Plio-Pleistocene in these dry and endangered ecosystems in

northeastern Africa.

ACKNOWLEDGEMENTS

This study was supported by two projects of the Czech Science

Foundation, nos. P506/10/0983 and 15-20229S, the Ministry

of Culture of the Czech Republic (DKRVO 2017/15, National

Museum, 00023272) and the Russian Foundation for Basic

Research (project no. 15-04-03801-a). For help during the

field work we acknowledge V. Mazoch, H. Konvi ckov!a, J.

Vrbov!a Kom!arkov!a, J. Kr!asov!a, A. H!anov!a, A. Kone cn!y, L.

Cuypers, C. Sabuni, A. Katakweba, A. Massawe, J. Skl!ıba, K.

Welegerima, A. Ribas, F. Sedl!a cek and all local collaborators.

We would like to thank also J. Vot!ypka and S. Gryseels for

providing samples. For help with genotyping we acknowledge

H. Konvi ckov!a and A. Bryjov!a. A. Dehne-Garcia, G. J. Ker-

goat, J. Sm!ıd and A. Drummond helped with the design of

data analysis. We thank P.J.J. Bates, P. van Breugel and I. Friis

for additional information. Most analyses were run on the

Czech-grid infrastructure METACENTRUM, minor part then

on CIPRES Gateway or CBGP cluster. We also thank the cura-

tors that allowed us to study the tissue collections in their

care: J. Phelps and B. Stanley (FMNH), N. Duncan (AMNH),

D. Moerike (SMNS), and W. Wendelen (RMCA). We also

would like to thank Judith Masters and four anonymous

reviewers for their useful comments on an earlier version of

the manuscript.

REFERENCES

Bandelt, H.J., Forster, P. & R€ohl, A. (1999) Median-joining

networks for inferring intraspecific phylogenies. Molecular

Biology and Evolution, 16, 37–48.

Bates, P.J.J. (1988) Systematics and zoogeography of Tatera

(Rodentia: Gerbillinae) of north-east Africa and Asia. Bon-

ner zoologische Beitr€age, 39, 265–303.

Berger, W.H. & Jansen, E. (1994) Mid-Pleistocene climate

shift: the Nansen connection. The polar oceans and their

role in shaping the global environment (ed. by O.M. Johan-

nessen, R.D. Muench and J.E. Overland). AGU Geophysical

Monograph Series, 85, 295–311.

Bivand, R. & Lewin-Koh, N. (2016) maptools: Tools for

reading and handling spatial objects. R package version

0.8-39. Available at https://CRAN.R-project.org/package=

maptools

Bobe, R. (2004) The expansion of grassland ecosystems in

Africa in relation to mammalian evolution and the origin

of the genus Homo. Palaeogeography, Palaeoclimatology,

Palaeoecology, 207, 399–420.

Bouckaert, R.R. (2010) DensiTree: making sense of sets of

phylogenetic trees. Bioinformatics, 26, 1372–1373.

Braconnot, P., Otto-Bliesner, B., Harrison, S. et al. (2007)

Results of PMIP2 coupled simulations of the Mid-Holo-

cene and Last Glacial Maximum - Part 1: experiments and

large-scale features. Climate of the Past, 3, 261–277.

van Breugel, P., Friis, I. & Demissew, S. (2016) The transi-

tional semi-evergreen bushland in Ethiopia : characteriza-

tion and mapping of its distribution using predictive

modelling. Applied Vegetation Science, 19, 355–367.

Brouat, C., Tatard, C., Ba, K., Cosson, J.-F., Dobigny, G.,

Fichet-Calvet, E., Granjon, L., Lecompte, E., Loiseau, A.,

Mouline, K., Piry, S. & Duplantier, J.-M. (2009) Phylo-

geography of the Guinea multimammate mouse (Masto-

mys erythroleucus): a case study for Sahelian species in

West Africa. Journal of Biogeography, 36, 2237–2250.

Bryja, J., Mikula, O., Sumbera, R., Meheretu, Y., Aghov!a, T.,

Lavrenchenko, L.A., Mazoch, V., Oguge, N., Mbau, J.S.,

Welegerima, K., Amundala, N., Colyn, M., Leirs, H. &

Verheyen, E. (2014) Pan-African phylogeny of Mus (sub-

genus Nannomys) reveals one of the most successful mam-

mal radiations in Africa. BMC Evolutionary Biology, 14,

256.

Burgess, N.D., Butynski, T.M., Cordeiro, N.J., Doggart, N.H.,

Fjelds, J., Howell, K.M., Kilahama, F.B., Loader, S.P.,

Lovett, J.C., Mbilinyi, B., Menegon, M., Moyer, D.C.,

Nashanda, E., Perkin, A., Rovero, F., Stanley, W.T. & Stu-

art, S.N. (2007) The biological importance of the Eastern

Arc Mountains of Tanzania and Kenya. Biological Conser-

vation, 134, 209–231.

Cerling, T.E., Bowman, J.R. & O’Neil, J.R. (1988) An iso-

topic study of a fluvial-lacustrine sequence: the Plio-Pleis-

tocene koobi fora sequence, East Africa. Palaeogeography,

Palaeoclimatology, Palaeoecology, 63, 335–356.

Cerling, T.E., Wynn, J.G., Andanje, S.A., Bird, M.I., Korir,

D.K., Levin, N.E., Mace, W., Macharia, A.N., Quade, J.

& Remien, C.H. (2011) Woody cover and hominin

environments in the past 6 million years. Nature, 476,

51–56.

Chevret, P. & Dobigny, G. (2005) Systematics and evolution

of the subfamily Gerbillinae (Mammalia, Rodentia, Muri-

dae). Molecular Phylogenetics and Evolution, 35, 674–688.

Colangelo, P., Corti, M., Verheyen, E., Annesi, F., Oguge,

N., Makundi, R.H. & Verheyen, W. (2005) Mitochon-

drial phylogeny reveals differential modes of chromoso-

mal evolution in the genus Tatera (Rodentia:

Gerbillinae) in Africa. Molecular Phylogenetics and Evolu-

tion, 35, 556–568.

Colangelo, P., Granjon, L., Taylor, P.J. & Corti, M. (2007)

Evolutionary systematics in African gerbilline rodents of

the genus Gerbilliscus: inference from mitochondrial genes.

Molecular Phylogenetics and Evolution, 42, 797–806.

Colangelo, P., Castiglia, R., Franchini, P. & Solano, E. (2010)

Pattern of shape variation in the eastern African gerbils of



2304

T. Aghov!a et al.

119


the genus Gerbilliscus (Rodentia, Muridae): environmental

correlations and implication for taxonomy and systemat-

ics. Mammalian Biology – Zeitschrift f€ur S€augetierkunde,

75, 302–310.

Cowling, S.A., Cox, P.M., Jones, C.D., Maslin, M.A., Peros,

M. & Spall, S.A. (2008) Simulated glacial and interglacial

vegetation across Africa: implications for species phyloge-

nies and trans-African migration of plants and animals.

Global Change Biology, 14, 827–840.

Craigie, I.D., Baillie, J.E.M., Balmford, A., Carbone, C., Col-

len, B., Green, R.E. & Hutton, J.M. (2010) Large mammal

population declines in Africa’s protected areas. Biological

Conservation, 143, 2221–2228.

Degnan, J. H., & Rosenberg, N. A. (2009). Gene tree discor-

dance, phylogenetic inference and the multispecies coales-

cent. Trends in Ecology & Evolution, 24, 332–340.

deMenocal, P.B. (2004) African climate change and faunal

evolution during the Pliocene-Pleistocene. Earth and Plan-

etary Science Letters, 220, 3–24.

Dixon, A.P., Faber-Langendoen, D., Josse, C., Morrison, J. &

Loucks, C.J. (2014) Distribution mapping of world grass-

land types. Journal of Biogeography, 41, 2003–2019.

Dobigny, G., Tatard, C., Gauthier, P., Ba, K., Duplantier,

J.M., Granjon, L. & Kergoat, G.J. (2013) Mitochondrial

and nuclear genes-based phylogeography of Arvicanthis

niloticus (Murinae) and sub-Saharan open habitats Pleis-

tocene history. PLoS ONE, 8, e77815.

Drummond, A.J., Ho, S.Y.W., Phillips, M.J., & Rambaut, A.

(2006). Relaxed phylogenetics and dating with confidence.

PLoS Biology, 4, e88.

Drummond, A.J. & Rambaut, A. (2007) BEAST: Bayesian

evolutionary analysis by sampling trees. BMC Evolutionary

Biology, 7, 214.

Drummond, A.J., Suchard, M.A., Xie, D. & Rambaut, A.

(2012) Bayesian phylogenetics with BEAUti and the BEAST

1.7. Molecular Biology and Evolution, 29, 1969–1973.

Dupont, L. (2011) Orbital scale vegetation change in Africa.

Quaternary Science Reviews, 30, 3589–3602.

Faber-Langendoen, D., Keeler-Wolf, T., Meidinger, D., Tart,

D., Hoagland, B., Josse, C., Navarro, G., Ponomarenko, S.,

Saucier, J.P., Weakley, A. & Comer, P. (2014) EcoVeg: a

new approach to vegetation description and classification.

Ecological Monographs, 84, 533–561.

Galan, M., Page, M. & Cosson, J.F. (2012) Next-gseneration

sequencing for rodent barcoding: species identification

from fresh. Degraded and environmental samples. PloS

ONE, 7, e48374.

Geist, H.J. & Lambin, E.F. (2004) Dynamic causal patterns of

desertification. BioScience, 54, 817.

Gernhard, T. (2008) The conditioned reconstructed process.

Journal of Theoretical Biology, 253, 769–778.

Gouy, M., Guindon, S. & Gascuel, O. (2010) SeaView Ver-

sion 4: a multiplatform graphical user interface for

sequence alignment and phylogenetic tree building. Molec-

ular Biology and Evolution, 27, 221–224.

Granjon, L., Colangelo, P., Tatard, C., Colyn, M., Dobigny,

G. & Nicolas, V. (2012) Intrageneric relationships within

Gerbilliscus (Rodentia, Muridae, Gerbillinae), with charac-

terization of an additional West African species. Zootaxa,

25, 1–25.

Heled, J. & Drummond, A.J. (2010) Bayesian inference of

species trees from multilocus data. Molecular Biology and

Evolution, 27, 570–580.

Hickerson, M.J. & Meyer, C.P. (2008) Testing comparative

phylogeographic models of marine vicariance and dispersal

using a hierarchical Bayesian approach. BMC Evolutionary

Biology, 8, 322.

Hijmans, R.J. (2015) raster: Geographic data analysis and

modeling. R package version 2.5-2. Available at: https://

CRAN.R-project.org/package=raster

Hijmans, R.J., Cameron, S.E., Parra, J.L., Jones, P.G. & Jarvis,

A. (2005) Very high resolution interpolated climate sur-

faces for global land areas. International Journal of Clima-

tology, 25, 1965–1978.

Hijmans, R.J., Phillips, S., Leathwick, J. & Elith, J. (2016) dismo:

Species distribution modeling. R package version 1.0-15.

Available at: https://CRAN.R-project.org/package=dismo

Hodell, D.A., Curtis, J.H., Sierro, F.J. & Raymo, M.E. (2001)

Correlation of Late Miocene to Early Pliocene sequences

between the Mediterranean and North Atlantic. Paleo-

ceanography, 16, 164–178.

IUCN (2016) The IUCN red list of threatened species. Version

2016-2. Available at: www.iucnredlist.org (accessed 4

November 2016).

Katoh, K. & Standley, D.M. (2013) MAFFT Multiple

Sequence Alignment Software Version 7: improvements in

performance and usability. Molecular Biology and Evolu-

tion, 30, 772–780.

Kebede, F., Rosenbom, S., Khalatbari, L., Moehlman, P.D.,

Beja-Pereira, A., & Bekele, A. (2016). Genetic diversity of

the Ethiopian Grevy’s zebra (Equus grevyi) populations

that includes a unique population of the Alledeghi Plain.

Mitochondrial DNA Part A, 27, 397–400.

Kingdon, J. (1990) Island Africa: the evolution of Africa’s rare

animals and plants. Collins, London.

Lanfear, R., Calcott, B., Kainer, D., Mayer, C. & Sta-

matakis, A. (2014) Selecting optimal partitioning

schemes for phylogenomic datasets. BMC Evolutionary

Biology, 14, 1.

Lavrenchenko, L.A., Kruskop, S.V., Bekele, A., Belay, G.,

Morozov, P.N., Ivlev, Y.F. & Warshavsky, A.A. (2010)

Mammals of the Babille elephant sanctuary (Eastern Ethio-

pia). sRussian Journal of Theriology, 9, 47–60.

Librado, P. & Rozas, J. (2009) DnaSP v5: a software for com-

prehensive analysis of DNA polymorphism data. Bioinfor-

matics, 25, 1451–1452.

Linder, H.P., de Klerk, H.M., Born, J., Burgess, N.D., Fjelds!a,

J. & Rahbek, C. (2012) The partitioning of Africa: statisti-

cally defined biogeographical regions in sub-Saharan

Africa. Journal of Biogeography, 39, 1189–1205.



2305


120


Lorenzen, E.D., Heller, R., & Siegismund, H.R. (2012). Com-

parative phylogeography of African savannah ungulates.

Molecular Ecology, 21, 3656–3670.

Maslin, M.A. & Christensen, B. (2007) Tectonics, orbital

forcing, global climate change, and human evolution in

Africa: introduction to the African paleoclimate special

volume. Journal of Human Evolution, 53, 443–464.

Maslin, M.A., Brierley, C.M., Milner, A.M., Shultz, S.,

Trauth, M.H. & Wilson, K.E. (2014) East african climate

pulses and early human evolution. Quaternary Science

Reviews, 101, 1–17.

Maslin, M.A., Shultz, S. & Trauth, M.H. (2015) A synthesis

of the theories and concepts of early human evolution.

Philosophical Transactions of the Royal Society B: Biological

Sciences, 370, 20140064.

McDonough, M.M., Sumbera, R., Mazoch, V., Ferguson,

A.W., Phillips, C.D. & Bryja, J. (2015) Multilocus phylo-

geography of a widespread savanna-woodland-adapted

rodent reveals the influence of Pleistocene geomorphology

and climate change in Africa’s Zambezi region. Molecular

Ecology, 24, 5248–5266.

Mikula, O., Sumbera, R., Aghov!a, T., Mbau, J.S., Katak-

weba, A.S., Sabuni, C.A., & Bryja, J. (2016). Evolution-

ary history and species diversity of African pouched

mice (Rodentia: Nesomyidae: Saccostomus). Zoologica

Scripta, 45, 595–617.

Monadjem, A., Taylor, P.J., Denys, C. & Cotterill, F.P.D.

(2015) Rodents of Sub-Saharan Africa. A biographic and

taxonomic synthesis. Walter de Gruyter GmbH, Berlin/

Munich/Boston.

Muscarella, R., Galante, P.J., Soley-Guardia, M., Boria, R.A.,

Kass, J.M., Uriarte, M. & Anderson, R.P. (2014) ENMeval:

an R package for conducting spatially independent evalua-

tions and estimating optimal model complexity for Max-

ent ecological niche models. Methods in Ecology and

Evolution, 5, 1198–1205.

Otto-Bliesner, B.L. (2006) Simulating Arctic climate warmth

and icefield retreat in the last interglaciation. Science, 311,

1751–1753.

Phillips, S.J., Anderson, R.P. & Schapire, R.E. (2006) Maxi-

mum entropy modeling of species geographic distribu-

tions. Ecological Modelling, 190, 231–259.

Potts, R. (2013) Hominin evolution in setings of strong envi-

ronmental variability. Quaternary Science Reviews, 73, 1–13.

Rambaut, A., Suchard, M.A., Xie, D. & Drummond, A.J.

(2014) Tracer v1.6. Available at: http://beast.bio.ed.ac.uk/

Tracer.

Randi, E., D’Huart, J.P., Lucchini, V. & Aman, R. (2002)

Evidence of two genetically deeply divergent species of

warhthog, Phacochoerus africanus and P. aethiopicus

(Artiodactyla: Suiformes) in Eastern Africa. Mammalian

Biology, 67, 91–96.

Ravelo, A.C., Andreasen, D.H., Lyle, M., Olivarez Lyle, A. &

Wara, M.W. (2004) Regional climate shifts caused by

gradual global cooling in the Pliocene epoch. Nature, 429,

263–267.

Ronquist, F., Teslenko, M., van der Mark, P., Ayres, D.L.,

Darling, A., H€ohna, S., Larget, B., Liu, L., Suchard, M.A.

& Huelsenbeck, J.P. (2012) MrBayes 3.2: efficient Bayesian

phylogenetic inference and model choice across a large

model space. Systematic Biology, 61, 539–542.

Sayre, R., Comer, P., Hak, J., et al. (2013). A new map of

standardized terrestrial ecosystems of Africa. Washington,

DC: Association of American Geographers. 24 pp.

Sepulchre, P., Ramstein, G., Fluteau, F., Schuster, M., Tierce-

lin, J.-J. & Brunet, M. (2006) Tectonic uplift and Eastern

Africa aridification. Science, 313, 1419–1423. Sm!ıd, J., Carranza, S., Kratochv!ıl, L., Gvo zd!ık, V., Nasher,

A.K. & Moravec, J. (2013) Out of Arabia: a complex bio-

geographic history of multiple vicariance and dispersal

events in the gecko genus Hemidactylus (Reptilia: Gekkoni-

dae). PLoS ONE, 8, e64018.

Stamatakis, A. (2006) RAxML-VI-HPC: maximum likeli-

hood-based phylogenetic analyses with thousands of taxa

and mixed models. Bioinformatics (Oxford, England), 22,

2688–2690.

Thulin, M. (ed.) (1993) Flora of Somalia: Vol. 1 – Pterido-

phyta; Gymnospermae; Angiospermae (Annonaceae-Faba-

ceae). Royal Botanic Gardens, Kew.

Thulin, M., Lavin, M., Pasquet, R. & Delgado-Salinas, A.

(2004) Phylogeny and biogeography of Wajira (Legumi-

nosae): a monophyletic segregate of Vigna centred in the

Horn of Africa region. Systematic Botany, 29, 903–920.

Trauth, M.H., Larrasoa~na, J.C. & Mudelsee, M. (2009)

Trends, rhythms and events in Plio-Pleistocene African cli-

mate. Quaternary Science Reviews, 28, 399–411.

Trauth, M.H., Maslin, M.A., Deino, A.L., Junginger, A.,

Lesoloyia, M., Odada, E.O., Olago, D.O., Olaka, L.A.,

Strecker, M.R. & Tiedemann, R. (2010) Human evolution

in a variable environment: the amplifier lakes of Eastern

Africa. Quaternary Science Reviews, 29, 2981–2988.

Uno, K.T., Pratigya, J.P. & Jackson, K.E. (2016) Neogene

biomarker record of vegetation change in eastern Africa.

Proceedings of the National Academy of Sciences USA, 113,

6355–6363.

Varshavsky, A.A., Karimova, T.Y. & Neronov, V.M. (2007)

Specific features of the rodent fauna formation in Somali

Peninsula. Zoologichesky Zhurnal, 86, 1498–1506.

Varty, N. (1988) Mammals and problems of their conserva-

tion in the riverine forests of the Jubba Valley, southern

Somalia. Tropical Zoology, 6975, 179–192.

Warren, D.L. & Seifert, S.N. (2011) Ecological niche model-

ing in Maxent : the importance of model complexity and

the performance of model selection criteria. Ecological

Applications, 21, 335–342.

White, F. (1983) Vegetation of Africa – A descriptive memoir

to accompany the UNESCO/AETFAT/UNSO vegetation map

of Africa. Natural Resources Research Report XX.

UNESCO, Paris.

Yang, Z. & Rannala, B. (2014) Unguided species delimitation

using DNA sequence data from multiple loci. Molecular

Biology and Evolution, 31, 3125–3135.



2306

T. Aghov!a et al.

121


SUPPORTING INFORMATION

Additional Supporting Information may be found in the

online version of this article:

Appendix S1 Details on used individuals.

Appendix S2 Additions to materials and methods.

Appendix S3 Additions to results.

DATA ACCESSIBILITY

New sequences used in this study are available in GenBank

under accession numbers KU965898–KU966133 and

KY352044–KY352064. Further details on used specimens

and museum vouchers are specified in Table S1 in

Appendix S1.

BIOSKETCH

Tatiana Aghov�a is PhD student working on phylogeogra-

phy of African savanna ecosystems using rodents as a model

organism, under supervision of Josef Bryja and Radim Sumbera.

Author contributions: J.B. and R.S. conceived the idea; T.A.,

R.S., O.M., M.M., L.L., Y.M., J.M. and J.B. participated to

sampling in the field; T.A., L.P. and M.M. genotyped the

material. T.A. and O.M. analysed the data, T.A., R.S. and

J.B. wrote the first version of the manuscript that was

approved by all authors.

Editor: Judith Masters



2307


123


Fossils know it best: using a new set of

fossil calibrations to improve the

temporal phylogenetic framework of

murid rodents (Rodentia: Myomorpha:

Muroidea: Muridae)

Aghová T., Kimura Y, Bryja J., Dobigny G., Granjon L.,

Kergoat G.J. (Molecular Phylogenetics and Evolution, Major

revision); bioRxiv 180398, https://doi.org/10.1101/180398

Study design: GJK, GD, LG, JB

Collecting genetic data: TA

Collecting paleontological

data: TA, YK

Data analysis: TA, GK

Writing: TA, JB, GJK


Paper

IV

Paper

125


1

Title 1

Fossils know it best: using a new set of fossil calibrations to improve the temporal 2

phylogenetic framework of murid rodents (Rodentia: Myomorpha: Muroidea: 3

Muridae) 4

5

Authors 6

Tatiana Aghováa,b,c*, Yuri Kimurad, Josef Bryjaa,c, Gauthier Dobignye,f, Laurent Granjone, Gael 7

J. Kergoatg 8

9

Authors’ affiliations 10

a Institute of Vertebrate Biology of the Czech Academy of Sciences, !"#$%&'(&603 65, Brno, 11

Czech Republic 12

b Department of Zoology, National Museum, )%*+,!-./&$%0"-#1& 2'(& 334&56(&Prague, Czech 13

Republic 14

c Department of Botany and Zoology, Faculty of Science, Masaryk University, 7#+%8-.%&9(&611 15

37, Brno, Czech Republic 16

d Department of Geology and Paleontology, National Museum of Nature and Science, 4-1-1 17

Amakubo, Tsukuba, 305-0005 Ibaraki, Japan 18

e CBGP, IRD, CIRAD, INRA, Montpellier SupAgro, Univ. Montpellier, Montpellier, France 19

f :*7+;& <7+=#;*>$?@A;& BCDE70;=-Calavi, Abomey-Calavi University, 01BP2009, Cotonou, 20

Benin 21

g CBGP, INRA, CIRAD, IRD, Montpellier SupAgro, Univ. Montpellier, Montpellier, France 22

*Corresponding author: [email protected] 23

24

Running title: Dated phylogeny of Muridae 25

126


2

Abstract 26

Murid rodents (Rodentia: Myomorpha: Muroidea: Muridae) represent the most diverse and 27

abundant mammalian group. In this study, we reconstruct a dated phylogeny of the family using 28

a multilocus dataset (six nuclear and nine mitochondrial gene fragments) encompassing 160 29

species representing 82 distinct murid genera from four extant subfamilies (Deomyinae, 30

Gerbillinae, Lophiomyinae, and Murinae). In comparison with previous studies on murid or 31

muroid rodents, our work stands out for the implementation of multiple fossil constraints within 32

the Muridae thanks to a thorough review of the fossil record. Before being assigned to specific 33

nodes of the phylogeny, all potential fossil constraints were carefully assessed; they were also 34

subjected to several cross-validation analyses. The resulting phylogeny is consistent with 35

previous phylogenetic studies on murids, and recovers the monophyly of all sampled murid 36

subfamilies and tribes. Based on nine controlled fossil calibrations, our inferred temporal 37

timeframe indicates that the murid family likely originated in the course of the Early Miocene, 38

23.0-16.0 million years ago (Ma), and that most major lineages (i.e. tribes) have started 39

diversifying ca. 10 Ma. Historical biogeography analyses support the Paleotropical origin for 40

the family, with an initial internal split (vicariance event) followed by subsequent migrations 41

between Afrotropical and Indomalayan lineages. During the course of their diversification, the 42

biogeographic pattern of murids is marked by several dispersal events toward the Australasian 43

and the Palearctic regions, mostly from the Indomalaya. The Afrotropical region was also 44

secondarily colonized at least three times from the Indomalaya, indicating that the latter region 45

has acted as a major centre of diversification for the family. 46

47

Keywords 48

Fossils, historical biogeography, molecular dating, Muridae, Mus, Rattus 49

50

127


3

1. Introduction 51

With about 150 genera and more than 730 recognized species, Muridae is the most diverse 52

family of mammals (Musser and Carleton, 2005). Collectively murids have colonized highly 53

distinct ecological niches, adapting to a wide array of environments ranging from warm (deserts 54

or tropical forests) to cold habitats (high altitude mountain ranges, tundra; Vaughan et al., 55

2011). Life habits in murids are also diverse, as the family encompasses amphibious, arboreal, 56

fossorial, or terrestrial taxa (Michaux et al., 2007; Musser and Carleton, 2005). 57

All murid species are native to the Old World (Musser and Carleton, 2005), but some 58

species (especially the black rat Rattus rattus Linnaeus, the Norway rat Rattus norvegicus 59

(Berkenhout) and the house mouse Mus musculus Linnaeus) now have a worldwide distribution 60

due to commensalism and dissemination by humans. Murid species diversity is especially high 61

in the Australasian and Indomalayan regions which accommodate half of the species diversity 62

of the family (Rowe et al., 2016a). Second to that is the species diversity in the Afrotropical 63

region (more than 200 species; Musser and Carleton, 2005). By contrast, there are much less 64

native murid taxa in the Palearctic region (e.g. Apodemus Kaup, Diplothrix Thomas, or 65

Tokudaia Kuroda). 66

The history of murid systematics is complex and convoluted with numerous changes 67

occurring in the past sixty years (see Table 1 for a summary). Simpson (1945) divided 68

representatives of family Muridae (as currently understood) into two separate families: 69

Cricetidae (with subfamilies Gerbillinae, Lophiomyinae and others) and Muridae (subfamilies 70

Murinae and Otomyinae). Chaline et al. (1977) considered “murid” rodents to belong to four 71

families: Cricetidae (including Lophiomyinae), Gerbillidae, Muridae (exclusively Murinae) 72

and Nesomyidae (including Otomyinae). Lavocat (1978) simplified this classification by 73

recognizing only two families: Muridae (Murinae) and Nesomyidae (in which he included 74

Gerbillinae, Lophiomyinae and Otomyinae). Another major change was later made by Carleton 75

128


4

and Musser (1984), who defined family Muridae in the broad sense with no less than 14 76

subfamilies (including Gerbillinae, Lophiomyinae, Murinae and Otomyinae). Following the 77

introduction of molecular systematics, changes in the classification of family Muridae 78

continued at a fast rate. Using molecular phylogenetics Chevret et al. (1993a) demonstrated that 79

Acomys I. Geoffroy is not a member of the subfamily Murinae but belongs to a separate 80

monophyletic clade including Deomys Thomas, Lophuromys Peters and Uranomys Dollman. 81

All four genera were assigned to the subfamily Deomyinae, which is closely related to the 82

Gerbillinae. In another study, Chevret et al. (1993b) showed that Otomyinae are closely allied 83

to the tribe Arvicanthini, thus unequivocally constituting a subset of the subfamily Murinae at 84

the tribe level (Ducroz et al., 2001; Jansa and Weksler, 2004). Jansa and Weksler (2004) also 85

strongly suggested that Lophiomyinae belonged to the Muridae. Only part of these proposals 86

was followed by Musser and Carleton (2005) who recognized the following five subfamilies in 87

the family Muridae: Deomyinae, Gerbillinae, Leimacomyinae, Murinae and Otomyinae. 88

Nowadays the most consensual classification agrees on five subfamilies: Deomyinae (four 89

genera and ca. 42 species), Gerbillinae (16 genera and ca. 103 species), Leimacomyinae (only 90

one species, possibly extinct; Kingdon, 2015), Lophiomyinae (only one species) and Murinae 91

(129 genera and ca. 584 species; see the review of Granjon and Montgelard, 2012). 92

Musser and Carleton's (2005) comprehensive catalogue listed 730 species in the family 93

Muridae. Estimates of species diversity in this family are very likely not definitive, as new 94

murid taxa are being regularly described (e.g. Carleton et al., 2015; Esselstyn et al., 2015; 95

Missoup et al., 2016; Mortelliti et al., 2016; Rowe et al., 2016a). Expected and ongoing raise in 96

species number can be accounted for by an increased focus on poorly known regions with high 97

levels of endemism, especially in tropical Asia and Africa. It is also linked with the 98

development of integrative taxonomy studies, where molecular genetic approaches are able to 99

129


5

detect taxa and geographical regions with high cryptic diversity (e.g. Bryja et al., 2014, 2017; 100

Ndiaye et al., 2016). 101

Because of the high species richness of the family, determining the precise timing of its 102

radiation is of particular paleobiogeographic interest. Several dated estimates for the age of 103

Muridae are available owing to studies either focusing on the order Rodentia (Adkins et al., 104

2001, 2003; Fabre et al., 2012; Montgelard et al., 2008), on the superfamily Muroidea 105

(Muroidea; Schenk et al., 2013 ; Steppan et al., 2004) or on various murid subsets (e.g. Bryja 106

et al., 2014; Chevret and Dobigny, 2005; Dobigny et al., 2013; Fabre et al., 2013; Pagès et al., 107

2016; Rowe et al., 2008, 2011, 2016b). However, no clear consensus could be reached for the 108

age of the family Muridae. Indeed, age estimates derived from all aforementioned studies are 109

far from being congruent, likely because their datasets have not been designed for this particular 110

purpose. In addition, all these studies used very diverse dating procedures, some of them relying 111

on substitution rate calibrations (e.g. Arbogast et al., 2001; Nicolas et al., 2008) whereas others 112

used fixed ages (e.g. the putative Mus/Rattus split at 12 Ma; Steppan et al., 2004), very distant 113

fossil constrains (Adkins et al., 2003; Fabre et al., 2012; Montgelard et al., 2008) or primary 114

calibrations using various fossil constraints within or outside the family Muridae (e.g. Bryja et 115

al., 2014; Pagès et al., 2016; Rowe et al., 2016b; Schenk et al., 2013). 116

For fossil-based calibrations of molecular clocks, it is crucial: (i) to properly assign and 117

place fossils on the tree, and (ii) to correctly estimate the age of fossil-bearing formations 118

(Parham et al., 2012; Sauquet et al., 2012). Unfortunately the fossil record of oldest murids is 119

quite fragmentary and mostly consists of isolated teeth and mandible remains, thus sometimes 120

making taxonomic identification difficult. The earliest representatives for the family Muridae 121

include the tribe Myocricetodontini with genera such as †Myocricetodon Lavocat, †Dakkamys 122

Jaeger and †Mellalomys Jaeger (Jacobs and Flynn, 2005; Lazzari et al., 2011). Extinct members 123

of the genus †Potwarmus Lindsey could be considered as a stem group of the subfamily 124

130


6

Murinae based on detailed analyses of dental morphology (Lazzari et al., 2011; López 125

Antoñanzas, 2009; Wessels, 2009). The earliest unequivocal representative of the subfamily 126

Murinae is the genus †Antemus Jacobs (Jacobs and Downs, 1994; Jacobs and Flynn, 2005; 127

Kimura et al., 2015). †Antemus possesses a new cusp (anterostyle, also known as t1), which is 128

a synapomorphy of Murinae. The earliest record of †Antemus chinjiensis is dated at 13.8 Ma 129

(Jacobs et al., 1990) based on specimens from the locality YGSP 491, Chinji Formation in the 130

Potwar Plateau, Pakistan (Jacobs, 1977). In the fossil record of the Potwar Plateau, two more 131

derived fossil genera are of particular interest: †Karnimata Jacobs and †Progonomys Schaub. 132

Based on the relative position of the anterostyle to the lingual anterocone on M1, Jacobs (1978) 133

hypothesized that †Karnimata is related to Rattus and that †Progonomys is a member of the 134

lineage including Mus. Hence, their first stratigraphic occurrence has been used to define the 135

widely used Mus/Rattus calibration (ca. 12 My; Jacobs and Downs, 1994). However, in 2015, 136

Kimura et al. revisited these fossils from a paleontological perspective and proved this 137

calibration point to be controversial. They showed that †Karnimata is a member of the 138

Arvicanthini-Millardini-Otomyini clade rather than a member of the lineage encompassing the 139

genus Rattus and its relatives (i.e. tribe Rattini). Therefore, they demonstrated that the 140

continuous fossil record of the murine rodents from the Potwar Plateau actually provides a 141

minimum age for the most recent common ancestor of the lineages leading to Arvicanthis 142

Lesson and Mus (= Mus/Arvicanthis split). 143

Recent progresses in divergence dating analyses lead us to revisit results previously 144

obtained by favouring the implementation of a new set of well-justified primary fossil 145

calibrations within a Bayesian framework. In comparison to previous studies (listed above), our 146

study can be considered as medium-sized in terms of taxonomic sampling, and essentially 147

focused on the family Muridae. But our study stands out for rigorous evaluation of the fossil 148

data for this highly diverse mammalian family. The present study has four main objectives: (i) 149

131


7

to design a comprehensive multi-marker molecular dataset for the family Muridae, (ii) to review 150

the murid fossil record in order to identify reliable and suitable primary fossil calibrations, (iii) 151

to provide a reliable estimate of the timing of diversification of the family using multiple fossil 152

calibrations, and (iv) to lean on the resulting dated phylogeny to reconstruct the biogeographic 153

history of the family using up-to-date analytical approaches. 154

155

2. Material and Methods 156

2.1. Taxon sampling 157

For this study, new DNA sequences were generated for five murid species (Acomys cf. 158

cineraceus, Acomys subspinosus (Waterhouse), Acomys wilsoni Thomas, Arvicanthis niloticus 159

(Desmarest), Arvicanthis neumanni (Matschie) see Appendix A). Though we largely relied on 160

GenBank data for this work, it is worth underlining that our research group generated thousands 161

of murid sequences (all deposited in GenBank) in the past 15 years (we used some of these 162

sequences for 44 species included in this study). In total, our dataset (Appendix A) encompasses 163

160 murid species representing 82 of the 151 known murid genera. All four extant subfamilies 164

(if considering the Togo mouse from Leimacomyinae to be extinct) of Muridae are included. 165

For the largest subfamily Murinae, we included representatives of all 10 tribes that have been 166

defined by Lecompte et al. (2008): Apodemini, Arvicanthini, Hydromyini, Malacomyini, 167

Millardini, Murini, Otomyini, Phloeomyini, Praomyini and Rattini. As outgroup taxa, we 168

selected five species of the family Cricetidae (from subfamilies Arvicolinae, Cricetinae, 169

Neotomyinae and Tylomyinae), which constitutes the sister group of Muridae (Fabre et al., 170

2012). Finally, the tree was rooted using Calomyscus baluchi Thomas, a representative of the 171

more distant family Calomyscidae (Fabre et al., 2012). All species names followed Musser and 172

Carleton (2005) and Monadjem et al. (2015). 173

174

132


8

2.2. DNA extraction, sequencing and molecular matrix 175

DNA was extracted using a Qiagen® DNeasy Blood and Tissue kit (Qiagen, Hilden, 176

Germany) following the manufacturer’s instructions. Two nuclear gene fragments were 177

targeted using the following combinations of polymerase chain reactions (PCR) primers: 178

IRBP217 and IRBP1531 (Stanhope et al., 1992) for the fragment of the ‘interphotoreceptor 179

retinoid binding’ (IRBP) gene; RAG1F1705 and RAG1R2951 (Teeling et al., 2000) for a 180

fragment of the ‘recombination activating gene 1’ (RAG1) gene. For PCR protocols, see Bryja 181

et al. (2017) and Teeling et al. (2000), respectively. PCR products were Sanger sequenced in 182

both directions using the BigDye® Terminator chemistry (Thermo Fisher Scientific) either in 183

the Institute of Vertebrate Biology on an ‘Applied Biosystems® 3130xl Genetic Analyzer’, or 184

commercially through the GATC Biotech company (Konstanz, Germany). New sequences were 185

deposited in GenBank under accession numbers KY634246 to KY634250. 186

The newly generated sequences were further combined with data from GenBank. The 187

resulting matrix (see Appendix A) encompasses the following six nuclear and nine 188

mitochondrial gene fragments: ‘acid phosphatase 5’ (AP5), BRCA1, intronic portion of 189

‘Peripheral benzodiazapine receptor variant’ (BZRP), ‘growth hormone receptor’ (GHR), IRBP 190

and RAG1, for the nuclear genes, and ‘12S ribosomal RNA’ (12S), ‘16S ribosomal RNA’ 191

(16S), ‘ATP synthase 8’ (ATPase8), ‘cytochrome c oxidase I’ (COI), ‘cytochrome oxidase II’ 192

(COII), ‘cytochrome b’ (Cytb), ‘Aspartic acid transfer RNA’ (tRNA-Asp), ‘Lysine transfer 193

RNA’ (tRNA-Lys), ‘Serine transfer RNA’ (tRNA-Ser), for the mitochondrial genes. For nine 194

taxa (Acomys cf. cineraceus, Acomys wilsoni Thomas, Aethomys chrysophilus (de Winton), 195

Aethomys hindei (Thomas), Aethomys kaiseri (Noack), Aethomys silindensis Roberts, 196

Arvicanthis nairobae J.A. Allen, Arvicanthis neumanni and Thallomys paedulcus (Sundevall), 197

gene fragments were concatenated from two individuals to minimize the amount of missing 198

data. For all protein-coding genes, we used Mesquite 3.2 (Maddison and Maddison, 2007) to 199

133


9

check the coding frame for possible errors or stop codons. The sequences of several markers 200

(i.e. 12S, 16S, intronic portion of BZRP, tRNA-Asp, tRNA-Lys and tRNA-Ser) were variable 201

in length; their alignment was accomplished using MUSCLE (Edgar, 2004) with default 202

settings. 203

204

2.3. Phylogenetic analyses 205

Phylogenetic analyses were conducted using both Bayesian inference (BI) and 206

maximum likelihood (ML). Analyses were performed on the online computer cluster CIPRES 207

Science Gateway (Miller et al., 2010; www.phylo.org) and on the high performance computing 208

(HPC) cluster hosted in the Centre de Biologie pour la Gestion des Populations (CBGP) in 209

Montferrier-sur-Lez, France. For both phylogenetic analytical approaches, we carried out 210

partitioned analyses to improve phylogenetic accuracy (Nylander et al., 2004). The molecular 211

dataset was divided a priori into 33 partitions: we used three partitions for each of the protein-212

coding genes (AP5, ATPase8, BRCA1, COI, COII, Cytb, GHR, IRBP and RAG1) and one 213

partition for each of the rRNA-tRNA genes (12S, 16S, tRNA-Asp, tRNA-Lys and tRNA-Ser) 214

as well as the BZRP intronic portion. The best partitioning scheme and substitution models 215

were determined with PartitionFinder 1.1.1 (Lanfear et al., 2014) using a greedy heuristic 216

algorithm; because of the risk of over-parameterization associated with the high number of 217

specified partitions, the ‘unlinked branch lengths’ option was chosen over the ‘linked branch 218

lengths’ option. The Bayesian information criterion (BIC) was also preferentially used to 219

compare partitioning schemes and substitution models following the recommendation of 220

Ripplinger and Sullivan (2008). 221

PartitionFinder (based on BIC) identified the same three partitions for both BI and ML 222

analyses: two partitions are associated with a Generalized-Time-Reversible ( !"#$%#$&) model 223

134


10

and one partition is associated with a Hasegawa-Kishino-Yano ( !"#$%#$&) model (see Table 224

2). 225

Bayesian inference analyses were carried out using MrBayes v3.2.6 (Ronquist et al., 226

2012b). Two independent runs with four MCMC (one cold and three incrementally heated 227

chains) were conducted: they ran for 50 million generations, with trees sampled every 1,000 228

generations. A conservative 25% burn-in was applied after checking for stability on the log-229

likelihood curves and the split-frequencies of the runs. Support of nodes for MrBayes analyses 230

was provided by clade posterior probabilities (PP) as directly estimated from the majority-rule 231

consensus topology. Following Erixon et al. (2003), nodes supported by PP 0.95 were 232

considered strongly supported. 233

Maximum likelihood analyses were performed using RAxML v8.2.8 (Stamatakis, 2014). 234

Because this software does not allow simpler substitution models, we used three partitions with 235

a General Time Reversible ('()#$%#$&) model (see Table 2). The best ML tree was obtained 236

using heuristic searches with 100 random addition replicates. Clade support was then assessed 237

using a non-parametric bootstrap procedure with 1,000 replicates. Following Hillis and Bull 238

(1993), nodes supported by bootstrap values (BV) 70 were considered strongly supported. 239

240

2.4. Evaluation of suitable fossil calibrations 241

Following the recommended criteria of Parham et al. (2012) for fossil calibrations, we 242

rigorously compiled a list of potential candidates from the paleontological literature and 243

eventually retained 18 candidate fossils (see Table 3). The candidate fossils possess the 244

information for the collection site, unique identification number, and the state of preservation 245

along with justification for the age of the fossil (i.e., age of fossil-bearing formation and 246

stratigraphic level, preferably with an absolute age by radiometric dating and/or reliable relative 247

age estimates, for example, by magnetostratigraphy). 248

135


11

In the next step, diagnostic morphological characters were reassessed to determine 249

whether they could be reliably used as minimum age constraints in our phylogeny, either as 250

crown or stem calibrations. Seven fossils were discarded following this step (see the ‘Results’ 251

section). 252

For the remaining 11 fossils, we used the cross-validation procedure developed by Near 253

and Sanderson (2004) and Near et al. (2005). The following approach was used: (i) we 254

identified potential inconsistencies within the 11 remaining fossil calibrations, and (ii) we 255

explored the impact of the inclusion of each of these fossils on our divergence time estimates. 256

Each of the 11 fossil constraints was enforced at a time in a specific Bayesian relaxed-clock 257

(BRC) analysis to estimate the ages of the remaining nodes (see also section 2.5). First, the sum 258

of the squared differences between the molecular and fossil age estimates (SS) was calculated 259

(for more details see Near and Sanderson, 2004). All calibration points were then ranked based 260

on the magnitude of its SS score; here the fossil with the greatest SS score is assumed to be the 261

most inconsistent with respect to all other fossils in the analysis (Near and Sanderson, 2004). 262

Second, we calculated the average squared deviation, s, for all fossil calibrations in the analysis. 263

Following the method of Near et al. (2005), we removed the fossil with the greatest SS score 264

and recalculated s with the remaining fossil calibration points. This process was pursued until 265

only the two fossil calibration points with the lowest and second lowest magnitudes of SS 266

remained (Near and Sanderson, 2004). The rationale behind this procedure is to assess whether 267

calibration points are approximately equally informative and accurate (Near et al., 2005): if it 268

is the case the magnitude of s should only decrease by a small fraction whenever a fossil 269

calibration is removed. 270

271

2.5. Bayesian relaxed-clock analyses 272

136


12

Although divergence time dating is now a well-established cornerstone of evolutionary 273

biology, there is still no widely accepted objective methodology for converting data from the 274

fossil record to calibration information of use in molecular phylogenies (Drummond and 275

Bouckaert, 2015). In the last few years, several methodological approaches to better implement 276

fossil calibrations have been developed, for instance allowing one to directly include fossil 277

lineages in phylogenies (‘total-evidence dating’; Pyron, 2010; Ronquist et al. 2012a) or to 278

account for information on the density of the fossil record (‘fossilized birth-death (FBD) 279

process’; Stadler, 2010; Heath et al., 2014). However, for our study, a ‘total evidence dating’ 280

approach was not applicable since it would have required the coding of a morphological matrix 281

for both fossils and extant taxa, which is problematic given the fragmentary nature of muroid 282

fossils. The use of the FBD methodology was also not envisioned because the fossil record of 283

muroid rodents is too sparse. Instead, we relied on a node-dating approach in which fossil 284

information is enforced on specific nodes through the use of parametric distributions. 285

Following our assessment of the murid fossil record and the results of cross-validation 286

analyses, nine fossil calibrations were finally retained for the dating procedure (for more 287

information, see Tables 3 and 4, and Appendix B). Five of them were defined based on fossil 288

material collected in the Siwalik Group of Pakistan (†Antemus chinjiensis, †Karnimata darwini 289

Jacobs, †cf. Karnimata sp., †Mus sp. and †Abudhabia pakistanensis Flynn and Jacobs). Three 290

additional accepted fossils originate from 6.1 Ma fossils discovered in the Lemudong'o locality 291

in Kenya (†Aethomys sp., †Arvicanthis sp. and †Gerbilliscus sp.), and the last retained fossil 292

calibration constraint is defined by the 9.6 Ma fossil of †Parapodemus lungdunensis Schaub. 293

Priors for fossil constraints were defined by using either uniform or lognormal statistical 294

distributions in two separate analyses. Statistical distributions were bounded by the minimum 295

ages provided by the fossil constraints and a conservative maximum age (ca. 25 Ma) for the 296

root derived from the study of Schenk et al. (2013; see Table 4). In a preliminary way (see 297

137


13

section 2.4 of the Material and Methods), BRC analyses were also conducted using one fossil 298

constraint at a time to carry out cross-validation analyses. 299

Bayesian relaxed-clock analyses were conducted with BEAST v1.8.4 (Drummond et al., 300

2012) using uncorrelated lognormal relaxed clocks (Drummond et al., 2006). To limit the risk 301

of over-parameterization: (i) we used three clock models (based on PartitionFinder results, 302

Table 2); and (ii) we enforced a guide tree that corresponds to the topology with the best clade 303

support (this topology corresponds to the topology obtained with MrBayes; see the ‘Results’ 304

section). For the tree speciation model, a birth death process (Gernhard, 2008) was used in order 305

to better account for extinct and missing lineages. 306

BEAST .xml files were modified to implement the path-sampling procedure for Bayes 307

factor (BF) estimation following the recommendations of Baele et al. (2013). Out of the two 308

calibrations, the calibration procedure with lognormal prior has the best harmonic mean (-309

208117.74 versus -208262.58 for the procedure with a uniform prior) and is recovered as the 310

best-fit calibration procedure with a statistically significant BF of 289.68 (BF 10, Kass and 311

Raftery, 1995). The final analysis (with nine verified fossil constraints and lognormal prior 312

distribution for calibration constraints) was carried out by two independent runs each with 50 313

million generations and trees sampled every 5,000 generations. We used a conservative burn-314

in of 12.5 million generations per run. Post burn-in trees from both analyses were further 315

combined using the LogCombiner module of BEAST. Convergence of runs was assessed 316

graphically under Tracer v.1.6 and by examining the ESS of parameters. 317

318

2.6. Historical biogeography 319

Ancestral biogeography was reconstructed using the R package ‘BioGeoBEARS’ 320

(Matzke, 2013). Data for species’ ranges were obtained from the International Union for 321

Conservation of Nature website (https://www.iucn.org/). Five major biogeographic areas were 322

138


14

defined on the basis of Olson et al. (2001): A, West Palearctic (from Western Europe to the 323

Ural Mountains, including North Africa); B, East Palearctic (from the Urals to Japan); C, 324

Indomalaya (from Afghanistan through the Indian subcontinent and Southeast Asia to lowland 325

southern China, and through Indonesia as far as Java, Bali, and Borneo, west of the Wallace 326

line); D, Australasia (Australia, New-Zealand, Papua-New-Guinea and neighbouring small 327

islands); E, Afrotropics (Africa, northern part excluded). Dispersal rate between adjacent areas 328

was fixed to 1 (A-B; B-C), whereas the dispersal of 0.7 (A-E; C-D) and 0.3 (B-D; B-E) was 329

specified for long-distance dispersal or whenever a geographical barrier had to be crossed. 330

Dispersal was disallowed between geographical areas separated by two or more areas (A-D; D-331

E). Six models of geographic range evolution were compared in a likelihood framework: (i) 332

Dispersal-Extinction Cladogenesis model (DEC) similar to Lagrange (Ree and Smith, 2008), 333

which parameterizes dispersal and extinction; (ii) DEC +J model (Matzke, 2013; 2014), which 334

adds founder-event speciation with long-distance dispersal (cladogenesis, where daughter 335

lineage is allowed to jump to a new range outside the range of the ancestor; Matzke, 2013) to 336

the DEC framework; (iii) Dispersal Vicariance Analysis (DIVA; Ronquist, 1997); (iv) DIVA 337

with long-distance dispersal (DIVA +J; Matzke, 2013); (v) Bayesian inference of historical 338

biogeography for discrete areas (BayArea; Landis et al., 2013); and (vi) BayArea with long-339

distance dispersal (BayArea +J; Matzke, 2013). Model fit was assessed using the Akaike 340

information criterion (AIC) and likelihood-ratio tests (LRT). 341

342

3. Results 343

3.1. Phylogeny of Muridae 344

Our multilocus dataset representing all major lineages of the family Muridae is 10,482 bp 345

long with 42.5% missing data. Both BI and ML analyses yield similar topologies (see Fig. 1 for 346

the topology inferred under BI, and Fig. S1 in Appendix D for the best-fit ML tree), as indicated 347

139


15

by a high proportion of shared nodes (160 out of 162). BI and ML analyses differ only in the 348

position of Pelomys fallax (Peters) and Zyzomys argurus (Thomas), but their placements are 349

not significantly supported in either analysis. Clade support is moderate to high on average; if 350

considering the number of nodes that are supported by PP ! 0.95 or BV ! 70%, BI analyses 351

yield a slightly more robust topology (135 well-supported nodes) compared to the ML tree (122 352

well-supported nodes). 353

Phylogenetic analyses confirm the monophyly of the family Muridae, of all its four 354

constituent subfamilies, as well as of the previously defined tribes of the subfamily Murinae 355

(Fig. 1). On the contrary, the phylogenetic position of some genera (e.g. Acomys I. Geoffroy, 356

Dasymys Peters, Golunda Gray, Melomys Thomas, Micaelamys Ellerman, Pelomys Peters, 357

Oenomys Thomas and Otomys F. Cuvier) within particular tribes was only partly supported. 358

359

3.2. Evaluation of suitable fossil calibrations 360

We summarized all fossils considered in this study in Table 3 and Appendix B regarding 361

taxonomic information and specification for prior settings (see also Figure 2 for their respective 362

positions within the tree). Five out of 18 preselected fossils (i.e. †Parapelomys robertsi Jacobs, 363

†Potwarmus primitivus, †Preacomys kikiae Geraads, †cf. Progonomys sp. Schaub and †aff. 364

Stenocephalomys Frick) were excluded from further analyses because the scarcity of 365

paleontological interpretation about their phylogenetic relationships impeded assigning them to 366

specific nodes of the phylogeny (Appendix B). We also excluded fossils of Acomys and 367

Lemniscomys Trouessart from the Lemudong’O locality, Kenya (Manthi, 2007), because first 368

upper molars, which possess the most diagnostic characters in the murine dentition, are not 369

described from the locality (Table 3; see more details also in Appendix B). The two-step cross-370

validation procedure resulted in a further reduction of the fossil set of possible calibration 371

points. Specifically, we excluded two fossils: one is a 2.4 Ma fossil identified as the genus 372

140


16

Gerbillus Desmarest, while the other one corresponds to a 6.1 Ma fossil identified as the genus 373

Mastomys Thomas (Appendix C). The rationale is that (i) these two fossil calibrations exhibited 374

the largest magnitude of SS, and that (ii) their removal also resulted in a very high (fivefold) 375

decrease in s (see Appendix C for more details). As a result of the latter series of selection steps, 376

nine fossils were finally retained for divergence dating (see Figure 2 for their position on the 377

tree, Table 4 for specification of priors, and Appendix B for more details on all considered 378

fossils). 379

380

3.3 Historical biogeography and divergence dating 381

Among the six models of geographic-range evolution compared in a likelihood 382

framework in BioGeoBEARS, the Dispersal-Extinction Cladogenesis model with founder-383

event speciation (DEC +J) was chosen because of its best likelihood and AICc associated scores 384

(lnL=-117.1, AICc=240.4; Table 5). 385

The dated tree resulting from the BRC analyses is shown in Figure 2 while dating 386

estimates for all internal nodes are provided in Table 6, and results of ancestral distribution 387

reconstructions are presented in Figure 3. The most recent common ancestor (MRCA) of 388

Muridae originated during the early Miocene (median age of 19.3 Ma; 95% highest posterior 389

density (HPD): 17.06-21.92 Ma) in the Afrotropical and Indomalayan bioregions. 390

Three subfamilies (Deomyinae, Gerbillinae and Lophiomyinae) belonging to the same 391

clade started their diversification in the Afrotropics. Within this clade, a first split occurred ca. 392

18.6 Ma (95% HPD: 16.35-21.11 Ma) between the Lophiomyinae and the clade encompassing 393

the Deomyinae and Gerbillinae. Deomyinae started their diversification ca. 14.6 Ma (95% 394

HPD: 12.68-16.82 Ma) while Gerbillinae started theirs ca. 12.2 Ma (95% HPD: 10.46-14.16 395

Ma). In Deomyinae and Gerbillinae, several lineages were able to colonize the Palearctic region 396

141


17

from the Afrotropics (in our dataset, this concerns Acomys russatus (Wagner) and Gerbillus 397

gerbillus Olivier). 398

The subfamily Murinae originated in the Indomalayan region during the middle Miocene 399

(median age of 14.2 Ma; 95% HPD: 12.70-16.07 Ma); the corresponding basal split separated 400

the Phloeomyini and all remaining murines (‘core murines’ sensu Steppan et al. 2005). The 401

next major split occurred in the Indomalaya between Rattini and the remaining tribes of 402

Murinae (median age of 12.3 Ma; 95% HPD: 11.28-13.62 Ma), with an origin of Rattini 403

estimated at 10.4 Ma (95% HPD: 9.23-11.93 Ma). The ancestral area of Rattini was also 404

inferred to be the Indomalaya; during the course of their diversification, a few taxa colonized 405

Australasia (e.g. Bunomys andrewsi (J.A. Allen), Melasmothrix naso Miller and Hollister, 406

Paruromys dominator Thomas and Rattus leucopus (Gray)) as well as the West and East 407

Palearctic (e.g. Micromys minutus (Pallas) and Diplothrix legata (Thomas)). The Hydromyini 408

split from the remaining Murinae at ca. 11.9 Ma (95% HPD: 10.95-13.10 Ma); although basal 409

lineages of this tribe are currently found in the Indomalaya (e.g. Archboldomys luzonenzis 410

Musser, Apomys datae (Meyer), Apomys hylocoetes Mearns, Chrotomys gonzalesi Rickart and 411

Heaney, Chiropodomys gliroides (Blyth) and Rhynchomys isarogensis Musser and Freeman), 412

a specific and diverse lineage of Hydromyini also colonized and radiated in the Australasia ca. 413

8.1 Ma (95% HPD: 7.22-9.09 Ma). The clade gathering Apodemini, Malacomyini, Murini and 414

Praomyini likely originated in the Afrotropics, with several lineages secondarily colonizing the 415

Indomalaya and the West and East Palearctic. The split between Malacomyini (which remained 416

in the Afrotropics) and Apodemini (which dispersed and differentiated mainly in the West and 417

East Palaearctic) is estimated at ca. 10.2 Ma (95% HPD: 9.33-11.39 Ma). Murini started to 418

diversify in the Indomalaya at ca. 7.2 Ma (95% HPD: 6.24-9.29 Ma). The intense radiation (51 419

extant species, Monadjem et al., 2015) of Praomyini occurred in the Afrotropics (median age 420

of 6.8 Ma for the MRCA of Praomyini; 95% HPD: 6.06-7.77 Ma). The Indomalayan Millardini 421

142


18

split from the predominantly Afrotropical Arvicanthini + Otomyini tribes at ca. 10.8 Ma (95% 422

HPD: 9.88-12.04 Ma). The respective first diversifications within Arvicanthini, Millardini and 423

Otomyini are estimated at ca. 8.8 Ma (95% HPD: 7.92-9.72 Ma), 9.2 Ma (95% HPD: 7.72-424

10.75 Ma) and 4.9 Ma (95% HPD: 4.12-5.98 Ma), respectively. The position of Asian Golunda 425

within Arvicanthini is not resolved (Fig. 1); the dispersal to Indomalaya of the lineage leading 426

to the extant Golunda species at ca. 8.5 Ma (95% HPD: 7.33-9.57 Ma; as suggested in Fig. 3) 427

should therefore be taken with caution. 428

4. Discussion 429

4.1. Selection of taxa and molecular markers 430

Our sampling of 160 species from 82 genera represents 22% of known murid species 431

diversity and more than half of the generic diversity of the family Muridae. When one compares 432

our sampling effort to previous studies (Table 7), only the study of Fabre et al. (2012) relied on 433

a better sampling for the family Muridae (302 species from 105 genera, i.e. about 41% of known 434

species diversity). In the study of Schenk et al. (2013), 18% of murid species are included. The 435

number of sampled murid species is also lower in Lecompte et al. (2008) and Rowe et al. (2008) 436

because their studies focussed on specific tribes and subfamilies. 437

438

4.2. Calibration of molecular clock and divergence dating 439

Using the classical Mus/Rattus calibration as prior for divergence dating often lead to an 440

underestimation of the age of the subfamily Murinae, inferring median ages that are generally 441

comprised between 13.3 to 12.0 Ma. Only the most recent studies (e.g. Rowe et al., 2016b) used 442

the correct Mus/Arvicanthis calibration with a prior median age of 11.1 Ma (as suggested by 443

Kimura et al., 2015). This resulted in the estimation of Murinae age of ca. 14.0 Ma (Rowe et 444

al., 2016b), which is consistent with our study (Table 6). 445

143


19

Other fossils frequently used for molecular clock calibration are from the genus 446

†Parapodemus. Recent studies used these fossils in two ways: the first occurrence of 447

†Parapodemus sp. (Martín-Suáres and Mein, 1998) in the late Miocene in Europe (Lungu, 448

1981; see Appendix B) was used to calibrate the MRCA of ‘Apodemurini’ (representing the 449

split between Apodemini+Malacomyini and Murini+Praomyini; Fabre et al., 2013; Rowe et al., 450

2011). Another calibration point is based on the discovery of †Parapodemus pasquierae 451

Aguilar and Michaux, from ‘Lo Fournas 6’ site (Roussillon, France). Authors postulated that 452

the latter species and the smaller †Parapodemus lugdunensis co-occurred during the same time 453

period ‘MN10’ (Aguilar et al., 1999; Montuire et al., 2006), dated approximately at 9.7 Ma 454

(Mein, 2003). Michaux et al. (2002) considered the differences between these two species as 455

representative of the split between the large Apodemus mystacinus Danford and Alston and all 456

smaller Apodemus species from the subgenus Sylvaemus, but they used a younger age of 7.0 457

Ma as a prior for their divergence. Numerous authors followed this calibration (e.g. Bryja et al., 458

2014; Fabre et al., 2013; Lecompte et al., 2008; Schenk et al., 2013) even if there is no clear 459

rationale for it. The estimated dates of MRCA of Apodemini range from 7.5 Ma (Rowe et al. 460

2016b; Schenk et al., 2013) to 9.6 Ma (Bryja et al., 2014; Michaux et al., 2002; Lecompte et 461

al., 2008). In our study, we conservatively used the †Parapodemus lugdunensis fossil as a stem 462

constraint for Apodemini and this placement resulted in an estimation of their MRCA at 9.0 Ma 463

(95% HPD: 7.92-10.13 Ma). 464

There are several localities in Africa (e.g. Lukeino Formation, Winkler, 2002; 465

Lemudong’o, Ambrose et al., 2007; Manthi, 2007) where fossil representatives of Arvicanthis 466

were identified. These fossils were used for molecular clock calibration in several studies with 467

a prior MRCA for the genus at 6.1 Ma (e.g. Fabre et al., 2013; Rowe et al., 2011). The fossils 468

from Lemudong’o were also used, in a less conservative way, by Bryja et al. (2014): based on 469

early records of †Otomys sp. (ca. 5.0 Ma; Denys, 1990), †Aethomys sp., †Arvicanthis sp., 470

144


20

†Lemniscomys sp. (from Lemudong’o = 6.1 Ma; Ambrose et al., 2007; Manthi et al., 2007) and 471

other relevant samples where these and related genera were absent (9.50-10.50 Ma; Mein et al. 472

2004), they set the split between Arvicanthini and Otomyini within 6.08-9.54 Ma. This 473

calibration resulted in an estimation of the MRCA for the tribe Arvicanthini at 7.8 Ma (Bryja 474

et al., 2014), which is about 1 million year younger than our own estimate (Table 6). Rowe et 475

al. (2016b) used as minimum age 8.7 Ma for the split between Arvicanthis and Otomys based 476

on the study of Kimura et al. (2015) that set minimum and maximum ages for locality Y388, 477

where †Karnimata darwini was found. This calibration resulted in an estimated MRCA of 478

Arvicanthini at 8.5 Ma. In our study, we instead used the age of an older locality (Y182; median 479

age of 9.2 Ma) where †Karnimata darwini was also found (Jacobs, 1978; Kimura et al., 2015), 480

in order to set a crown calibration for the Arvicanthini/Millardini/Otomyini clade. This 481

placement resulted in an estimation of their MRCA at 8.8 Ma (95% HPD: 7.92-9.72 Ma) (Table 482

6). 483

484

4.3. Historical biogeography with focus on faunal exchanges between the Afrotropics and 485

the Indomalaya 486

Our study could not resolve the origin of murid rodents, but it was either in the Afrotropics 487

or in the Indomalaya. Our inferred ancestral tropical range for the MRCA of murids is consistent 488

with the fact that most extant murid taxa are still distributed in warm and moist tropical areas. 489

During the Early Miocene (23.0-16.0 Ma), the rotation of Africa and Arabia, and finally the 490

collision with Eurasia formed a landbridge between Africa and Eurasia (the so-called 491

‘Gomphotherium landbridge’; Rögl, 1999). During this time period, early murids colonized 492

both geographical regions. The subsequent reopening of the Mediterranean-Indo-Pacific 493

seaway (‘Indo-Pacific recurrence’; Rögl, 1999) separated Africa from Eurasia again, thus 494

giving rise to the main clades of Afrotropics and Indomalaya rodents. Three subfamilies, 495

145


21

Deomyinae, Gerbillinae and Lophiomyinae, then likely diversified in the Afrotropics (Chevret 496

and Dobigny, 2005; Ndiaye et al., 2016; Schenk et al., 2013, this study; Figure 3). This 497

hypothesis is supported by paleontological records since the oldest fossils tied to these 498

subfamilies were found in the Afrotropics (e.g. late Miocene Acomys, Gerbilliscus, Lophiomys 499

and †Preacomys from East Africa; Winkler et al. 2010 and references therein). The subfamily 500

Murinae started to diversify in Indomalaya, most probably in Southeast Asia, where we can 501

also find the hitherto highest phylogenetic diversity, including the oldest offshoots of this clade 502

(e.g. the ancestor of Phloeomyini probably lived in the Philippines, those of Rattini and 503

Hydromyini in South-east Asia, etc.; Fabre et al., 2013). 504

During the Middle Miocene (16.0-11.6 Ma), the Mediterranean-Indo-Pacific seaway 505

closed again at the beginning of the Serravallian ca. 13.8 Ma (‘Parathethys Salinity Crisis’; 506

Rögl, 1999), co-incidentally with a global cooling that caused vegetation shifts and a general 507

aridification (Prista et al., 2015). The newly formed landbridge (Rögl, 1999) allowed repeated 508

dispersals of murine rodents from Asia to both Africa and Eurasia. Murine fossil records 509

provide clear evidence for connections between the Indomalaya, the Palearctic, and the 510

Afrotropics. Among them, there are two conspicuous examples: (i) †Progonomys was recorded 511

in many Indomalayan Middle Miocene localities (Jacobs and Flynn, 2005) as well as in the 512

Palearctic region (Algeria: Wessels, 2009; China: Qiu et al., 2004; Egypt: Heissig, 1982; 513

France: Mein et al., 1993; Spain: Weerd, 1976); and (ii) the oldest records of †Parapelomys 514

spp. were found synchronously in Africa (8.5 Ma; Chorora, Ethiopia; Geraads, 2001) and in 515

Pakistan (ca. 8.0 Ma; Jacobs and Flynn, 2005). During this period, representatives of several 516

murine tribes occurred in the Afrotropics (Arvicanthini, Malacomyini, Otomyini and 517

Praomyini) and the Indomalaya (Millardini, Murini, Rattini, and basal lineages of Hydromyini). 518

The last faunal interchange of murid taxa between Africa, Asia and Western Palearctic 519

(Benammi et al., 1996; Sabatier, 1982; Sen, 1977, 1983; Winkler, 2002) is coincident with 520

146


22

Messinian Salinity Crisis ca. 6 Ma during the Late Miocene (Hsü et al., 1973, 1978). During 521

this period of global sea level depression (Haq et al., 1987) Africa and Arabia were reconnected 522

through Neguev-Sinai landbridge (‘Levantine corridor’, Fernandes et al., 2006) and landbridge 523

in the Bab-el-Mandeb (Bosworth et al., 2005). In murids, evidence to support this faunal 524

exchange can be found in the African subgenus Nannomys (genus Mus), which colonized 525

Afrotropics and started there its radiation ca. 5.2 Ma (Bryja et al., 2014). A possible example 526

for an opposite west-to-east migration is the genus Golunda, which belongs to the Arvicanthini 527

tribe. In a predominantly Afrotropical clade, Golunda is the only genus that occurs in the 528

Indomalaya, probably since the end of Miocene (Ducroz et al., 2001; Fig. 3). However, one 529

should be cautious with this scenario since the position of Golunda within Arvicanthini is not 530

well supported (Fig. 1). Africa-to-Asia dispersals at the Miocene/Pliocene boundary have been 531

also recorded in other taxa, such as rodents (e.g. Myomyscus yemeni (Sanborn and Hoogstraal); 532

our unpubl. data), reptiles (e.g. Varanus yemenensis: Böhme et al., 2003, Portik and Papenfuss 533

2012; Hemidactylus geckos: Šmíd et al. 2013; Echis vipers: Pook et al. 2009) and hamadryas 534

baboons (Winney et al. 2004). 535

536

5. Conclusion and perspectives 537

In this study, we provided an improved multilocus dated phylogeny for the highly 538

speciose family Muridae. Both our dating and historical biogeography analyses suggest that the 539

family originated during the Early Miocene, and subsequently gave rise to four extant 540

subfamilies: three in the Afrotropical region (Deomyinae, Gerbillinae and Lophiomyinae) and 541

one in the Indomalaya (Murinae). Our study also supports a dynamic biogeographic scenario 542

in which repeated colonisation events occurred in the Australasian (Hydromyini, Rattini), 543

Afrotropical (Malacomyini, Praomyini, Arvicanthini, Otomyini) and Palearctic (Apodemini) 544

regions. One of the strong aspects of this study lies in the assessment and treatment of fossil 545

147


23

data (Appendix B); such data is likely to be useful for further studies investigating the timing 546

of diversification of rodents, or even mammals in general. For an easy access to all 547

corresponding fossil records, we have made data available on the Date-a-Clade Website 548

(http://palaeo.gly.bris.ac.uk/fossilrecord2/dateaclade/index.html), Paleobiology Database 549

(http://fossilworks.org/) and TimeTree Database (http://timetree.org). 550

551

Acknowledgments 552

This study was supported by two projects of the Czech Science Foundation, no. 15-553

20229S, the Ministry of Culture of the Czech Republic (DKRVO 2017/15, National Museum, 554

00023272) and JSPS KAKENHI JP15H06884 (Grant-in-Aid for Young Scientists Start-up). 555

Most analyses were run on CBGP HPC computational platform, while a minor part was 556

performed on CIPRES Gateway. We thank Alexandre Dehne Garcia for his help on the HPC 557

CBGP cluster and Nick J. Matzke for introduction into BioGeoBEARS analysis. We would 558

also like to thank Alisa Winkler and Christiane Denys for discussion about some 559

paleontological aspects, as well as Arame Ndiaye and Pascal Chevret for providing additional 560

sequences. We are grateful to Vincent Lazzari, Fredrick K. Manthi, Pierre Mein, Rajeev 561

Pantanik and Sevket Sen who provided permission for the reproduction of the figures in 562

Appendix B. 563

564

Authors´ contributions: GJK, GD, LG, JB conceived the ideas; TA collected genetic 565

data (including new genotyping); YK, TA collected and analysed paleontological data; TA, 566

GJK performed phylogenetic analyses; TA, JB, GJK wrote the first version of the manuscript 567

that was then implemented by all authors. 568

569

148


24

References 570

Adkins, R.M., Gelke, E.L., Rowe, D., Honeycutt, R.L., 2001. Molecular phylogeny and 571

divergence time estimates for major rodent groups: evidence from multiple genes. Mol. 572

Biol. Evol. 18, 777–91. 573

Adkins, R.M., Walton, A.H., Honeycutt, R.L., 2003. Higher-level systematics of rodents and 574

divergence time estimates based on two congruent nuclear genes. Mol. Phylogenet. Evol. 575

26, 409–420. 576

Aguilar, J.P., Escarguel, G., Michaux, J., 1999. A succession of Miocene rodent assemblages 577

from fissure fillings in southern France: Palaeoenvironmental interpretation and 578

comparison with Spain. Palaeogeogr. Palaeoclimatol. Palaeoecol. 145, 215–230. 579

Ameur, A., 1984. Découverte de nouveaux Rongeurs dans la formation miocène de Bou 580

Hanifia (Algérie occidentale). Géobios, 17, 167-175. 581

Ambrose, S.H., Nyamai, C.M., Mathu, E.M., Williams, M.A.J., 2007. Geology, Geochemistry, 582

and Stratigraphy of the Lemudong’o Formation, Kenya Rift Valley. Kirtlandia, Clevel. 583

Museum Nat. Hist. 53–64. 584

Arbogast, B. S., Browne, R. A., Weigl, P. D., 2001. Evolutionary genetics and Pleistocene 585

biogeography of North American tree squirrels (Tamiasciurus). J. Mammal. 82, 302-319. 586

Baele, G., Li, W.L.S., Drummond, A.J., Suchard, M. A., Lemey, P., 2013. Accurate model 587

selection of relaxed molecular clocks in bayesian phylogenetics. Mol. Biol. Evol. 30, 239–588

43. 589

Benammi, M., Calvo, M., Prévot, M., Jaeger, J.-J., 1996. Magnetostratigraphy and 590

paleontology of Ait Kandoula Basin (High Atlas, Morocco) and the African-European late 591

Miocene terrestrial fauna exchanges. Earth Planet. Sci. Lett. 145, 15-29. 592

Bernor, R.L., Brunet M., Ginsburg, L., Mein, P., Pickford, M., Rögl, F., Sen, S., Steininger F., 593

Thomas, H., 1987. A consideration of some major topics concerning old world Miocene 594

149


25

mammalian chronology, migrations and paleogeography. Geobios. 20, 431-439. 595

Blanga-Kanfi, S., Miranda, H., Penn, O., Pupko, T., DeBry, R.W., Huchon, D., 2009. Rodent 596

phylogeny revised: analysis of six nuclear genes from all major rodent clades. BMC Evol. 597

Biol. 9, 71. 598

Bosworth, W., Huchon, P., McClay, K., 2005. The Red Sea and Gulf of Aden Basins. J. Afr. 599

Earth Sci. 43, 334–378. 600

Böhme, W., 2003. Checklist of the living monitor lizards of the world (family Varanidae). Zool. 601

Verhandel. 341, 1–43. 602

Bryja, J., Aghova, T., Bryjova, A., Denys, C., Verheyen, E., 2017. Evolutionary history of the 603

thicket rats (genus Grammomys) mirrors the evolution of African forests since late 604

Miocene. J. Biogeogr. 44, 182–194. 605

Bryja, J., Mikula, O.E., Umbera, R., Meheretu, Y., Aghová, T., Lavrenchenko, L. A, Mazoch, 606

V., Oguge, N., Mbau, J.S., Welegerima, K., Amundala, N., Colyn, M., Leirs, H., 607

Verheyen, E., 2014. Pan-African phylogeny of Mus (subgenus Nannomys) reveals one of 608

the most successful mammal radiations in Africa. BMC Evol. Biol. 14, 256. 609

Campisano, C.J., Feibel, C.S., 2008. Depositional environments and stratigraphic summary of 610

the Pliocene Hadar formation at Hadar, Afar Depression, Ethiopia. in: Quade, J., Wynn, 611

J.G. (Eds.), The Geology of Early Humans in the Horn of Africa. Geol. Soc. Am., vol. 612

446, pp. 179-201. 613

Carleton M.D., Musser G.G., 1984. Muroid rodents. in: Anderson S., Jones J.K., Jr., (Eds.), 614

Orders and families of recent mammals of the world. Wiley, New York, pp. 289–379. 615

Chaline, J., 1977. Les rongeurs au pilori? Analyse critique des reconstitutions paléoécologiques 616

en microthériodontologie. Approche écologique de l’Homme fossile. Paris, 77-81. 617

Cheema, I. U., S.Mahmood Raza, L. J. Flynn, A. R. Rapjar, Tomida, Y., 2000. Miocene 618

smallmammals from Jalalpur, Pakistan, and their biochronologic implications. Bull. Natl. 619

150


26

Sci. Mus., Tokyo, C 26, 57–77. 620

Chevret, P., Denys, C., Jaeger, J.J., Michaux, J., Catzeflis, F.M., 1993a. Molecular evidence 621

that the spiny mouse (Acomys) is more closely related to gerbils (Gerbillinae) than to true 622

mice (Murinae). Proc. Natl. Acad. Sci. U.S.A. 90, 3433–3436. 623

Chevret, P., Denys, C., Jaeger, J.-J., Michaux, J., Catzeflis, F., 1993b. Molecular and 624

palaeontological aspects of the tempo and mode of evolution in Otomys (Otomyinae: 625

Muridae: Mammalia). Biochem. Syst. Ecol. 21, 123–131. 626

Chevret, P., Dobigny, G., 2005. Systematics and evolution of the subfamily Gerbillinae 627

(Mammalia, Rodentia, Muridae). Mol. Phylogenet. Evol. 35, 674–688. 628

Daxner-Höck, G., 2003. Cricetodon meini and other rodents from the Mühlbach and Grund, 629

Lower Austria (Middle Miocene, late MN5). Ann. Nat.hist. Mus. Wien (A) 104, 267-291. 630

DeBry, R.W., 2003. Identifying conflicting signal in a multigene analysis reveals a highly 631

resolved tree: the phylogeny of Rodentia (Mammalia). Syst. Biol. 52, 604–617. 632

Deino, A.L., Ambrose, S.H., 2007. 40Ar/39Ar dating of the Lemudong’o late Miocene fossil 633

assemblages, southern Kenya rift. Kirtlandia. 56, 65–71. 634

Denys, C., 1990. The oldest Acomys (Rodentia, Muridae) from the lower Pliocene of South 635

Africa and the problem of its murid affinities. Paleontogr. Abt. A, 210, 79–91. 636

Denys, C., Michaux, J., Catzeflis, F., Ducroq, S., Chevret, P., 1995. Morphological and 637

molecular data against the monophyly of Dendromurinae (Muridae:Rodentia). Bönner 638

Zool. Beitr. 45, 173–190. 639

Dobigny, G., Tatard, C., Gauthier, P., Ba, K., Duplantier, J., Granjon, L., Kergoat, G.J., 2013. 640

Mitochondrial and Nuclear Genes-Based Phylogeography of Arvicanthis niloticus 641

(Murinae) and Sub-Saharan Open Habitats Pleistocene History. PLoS One 8. 642

Donoghue, P. C., Benton, M. J., 2007. Rocks and clocks: calibrating the Tree of Life using 643

fossils and molecules. Trends Ecol. Evol. 22, 424-431. 644

151


27

Drummond, A.J., Bouckaert, R.R., 2015. Bayesian evolutionary analysis with BEAST 2. 645

Cambridge University press. 646

Drummond, A.J., Ho, S.Y.W., Phillips, M.J., Rambaut, A., 2006. Relaxed phylogenetics and 647

dating with confidence. PLoS Biol. 4, e88. 648

Drummond, A. J., Suchard, M. A., Xie, D., Rambaut, A. 2012. Bayesian phylogenetics with 649

BEAUti and the BEAST 1.7. Mol. Biol. Evol. 29, 1969-1973. 650

Ducroz, J.F., Volobouev, V., Granjon, L., 2001. An Assessment of the Systematics of 651

!"#$%&'(#&)* +,-)&'.* /.#&0* 1#',$(,&-!#%2* 34 * 5)67)&$).89* :",27'#,&%!y and 652

Biogeographical Implications J. Mammal. Evol. 8, 173-206. 653

Edgar, R.C., 2004. MUSCLE: multiple sequence alignment with high accuracy and high 654

throughput. Nucleic Acids Res. 32, 1792–7. 655

Erixon, P., Svennblad, B., Britton, T., Oxelman, B., 2003. Reliability of Bayesian posterior 656

probabilities and bootstrap frequencies in phylogenetics. Syst. Biol. 52, 665–673. 657

Fabre, P.H., Hautier, L., Dimitrov, D., Douzery, E.J., 2012. A glimpse on the pattern of rodent 658

diversification: a phylogenetic approach. BMC evolutionary biology, 12(1), 88. 659

Fabre, P.H., Pagès, M., Musser, G.G., Fitriana, Y.S., Fjeldså, J., Jennings, A., Jønsson, K. A., 660

Kennedy, J., Michaux, J., Semiadi, G., Supriatna, N., Helgen, K.M., 2013. A new genus 661

of rodent from wallacea (Rodentia: Muridae: Murinae: Rattini), and its implication for 662

biogeography and Indo-Pacific Rattini systematics. Zool. J. Linn. Soc. 169, 408–447. 663

Fernandes, C.A., Rohling, E.J., Siddall, M., 2006. Absence of post-Miocene Red Sea land 664

bridges: Biogeographic implications. J. Biogeogr. 33, 961–966. 665

Flynn, L.J., Jacobs, L.L., 1999. Late Miocene small mammal faunal dynamics: the crossroads 666

of the Arabian peninsula. in: Whybrow, P.J., Hill, A., (Eds.), Fossil vertebrates of Arabia. 667

Yale University Press, New Haven, pp. 412–419. 668

Gavryushkina, A., Heath, T. A., Ksepka, D. T., Stadler, T., Welch, D., Drummond, A. J., 2017. 669

152


28

Bayesian total-evidence dating reveals the recent crown radiation of penguins. Syst. 670

Biol. 66, 57-73. 671

Geraads, D., 2001. Rongeurs du Miocene superieur de Chorora (Ethiopie): Dendromuridae, 672

Muridae et conclusions. Palaeovertebrata. 30, 89–109. 673

Geraads, D., Alemseged, Z., Bellon, H., 2002. The late Miocene mammalian fauna of Chorora, 674

Awash basin, Ethiopia: systematics, biochronology and the 40K–40Ar ages of the 675

associated volcanics. Tertiary Research. 21, 113–127. 676

Gernhard, T., 2008. The conditioned reconstructed process. J. Theor. Biol. 253, 769–778. 677

Granjon, L., Colangelo, P., Tatard, C., Colyn, M., Dobigny, G., Nicolas, V., 2012. Intrageneric 678

relationships within Gerbilliscus (Rodentia, Muridae, Gerbillinae), with characterization 679

of an additional West African species. Zootaxa. 25, 1–25. 680

Granjon, L., Montgelard, C., 2012. The Input of DNA Sequences to Animal Systematics: 681

Rodents as Study Cases. in: Munshi, A. (Ed.), DNA Sequencing - Methods and 682

Applications, InTech, pp. 104-140. 683

Haq, B.U., Hardenbol, J., Vail, P.R., 1987. Chronology of fluctuating sea levels since the 684

Triassic. Science. 235, 1156-1167. 685

Heath, T. A., Huelsenbeck, J.P., Stadler, T., 2014. The fossilized birth-death process for 686

coherent calibration of divergence-time estimates. Proc. Natl. Acad. Sci. U.S.A. 111, 687

E2957-66. 688

Heissig, K., 1982. Kleinsäuger aus einer obermiozänen (Vallesium) Karstfüllung Ägyptens. 689

Mitt. Bayer. Staatssamml. Paläontol. Hist. Geol. 22, 97-101. 690

Hillis, D.M., Bull, J.J., 1993. An empirical test of bootstrapping as a method for assessing 691

confidence in phylogenetic analysis. Syst. Biol. 42, 182–192. 692

Hsü, K.J., 1978. Stratigraphy of the lacustrine sedimentation in the Black Sea. Initial Reports 693

of the DSDP. 42B, 509–524. 694

153


29

Hsü, K.J., Cita, M.B., Ryan, W.B.F., 1973. The origin of the Mediterranean evaporites. Initial 695

Reports of the DSDP. 42, 1203–1231. 696

Jacobs, L. L. 1977. A new genus of murid rodent from the miocene of Pakistan and comments 697

on the origin of the Muridae. PaleoBios. 25, 1–11 698

Jacobs, L. L., 1978. Fossil rodents (Rhizomyidae and Muridae) from Neogene Siwalk deposits, 699

Pakistan. Mus. North. Ariz. Bull. 52, 1–103. 700

Jacobs, L.L., Downs, W.R., 1994. The evolution of murine rodents in Asia. Natl. Sci. Museum 701

Monogr. 8, 149–156. 702

Jacobs, L. L., Flynn, L.J., Downs, W.R., 1989. Neogene Rodentia of Southern Asia. in: Black, 703

C.C., Dawson, M.R. (Eds.), Papers on fossil rodents in honor of Albert Elmer Wood. 704

Natural History Museum, Los Angeles County, Science Series No. 33, pp. 157–177. 705

Jacobs, L., Flynn, L., Downs, W., Barry, J., 1990. Quo vadis Antemus? The Siwalik muroid 706

record. in: Lindsay, E.H, Fahlbusch, V., Mein, P. (Eds.), European Neogene Mammal 707

Chronology. Plenum Press. New York. pp. 573–586. 708

Jacobs L.L., Flynn J.J., 2005. Of mice ... again: the Siwalik rodent record, murine distribution, 709

and molecular clocks. in: Leiberman. D.E., Smith, R.J,, Kelley, J., (Eds.), Interpreting the 710

past: essays on human, primate, and mammal evolution in honor of David Pilbeam. Brill 711

Academic Publishers Inc, Boston. pp. 63–80. 712

Jaeger, J.-J., 1977. Les rongeurs du Miocène moyen et supérieur du Maghreb. Palaeovertebrata. 713

8, 3–166. 714

Jansa, S. A, Weksler, M., 2004. Phylogeny of muroid rodents: relationships within and among 715

major lineages as determined by IRBP gene sequences. Mol. Phylogenet. Evol. 31, 256–716

76. 717

Kass, R. E., Raftery, A. E., 1995. Bayes factors. J. Am. Stat. Assoc. 90, 773-795. 718

Kimbel, W., Walter, R., Johanson, D., Reed, K., Aronson, J., Assefa, Z., Marean, C.W., Eck, 719

154


30

G., Bobe, R., Hovers, E., Rak, Y., Vondra, C., Yemane, T., York, D., Chen, Y., Evensen, 720

N., Smith, P., 1996. Late Pliocene Homo and Oldowan tools from the Hadar formation 721

(Kada Hadar member), Ethiopia. J. Hum. Evol. 31, 549-561. 722

Kimura, Y., Hawkins, M.T.R., McDonough, M.M., Jacobs, L.L., Flynn, L.J., 2015. Corrected 723

placement of Mus-Rattus fossil calibration forces precision in the molecular tree of 724

rodents. Nat. Publ. Gr. 1–9. 725

Kimura, Y., Jacobs, L.L., Cerling, T.E., Uno, K.T., Ferguson, K.M., Flynn, L.J., Patnaik, R., 726

2013. Fossil Mice and Rats Show Isotopic Evidence of Niche Partitioning and Change in 727

Dental Ecomorphology Related to Dietary Shift in Late Miocene of Pakistan. PLoS One 728

8, 1–10. 729

Kingdon, J., 2015. The Kingdon field guide to African mammals. Bloomsbury Publishing, 730

London. 731

Lavocat, R., 1978. Rodentia and Lagomorpha. in: Maglio, V.J., Cooke, H.B.S. (Eds.), Evolution 732

of African mammals. Harvard University Press, Cambridge. pp. 68-89. 733

Landis, M.J., Matzke, N.J., Moore, B.R., Huelsenbeck, J.P., 2013. Bayesian analysis of 734

biogeography when the number of areas is large. Syst. Biol. 62, 789–804. 735

Lanfear, R., Calcott, B., Kainer, D., Mayer, C., Stamatakis, A., 2014. Selecting optimal 736

partitioning schemes for phylogenomic datasets. BMC Evol. Biol. 14, 82. 737

Lazzari, V., Tafforeau, P., Michaux, J., 2011. When homologous cusps display non-738

homologous wear facets: An occlusal reorganization ensures functional continuity during 739

dental evolution of Murinae (Rodentia, Mammalia). Arch. Oral Biol. 56, 194–204. 740

Lecompte, E., Aplin, K., Denys, C., Catzeflis, F., Chades, M., Chevret, P., 2008. Phylogeny 741

and biogeography of African Murinae based on mitochondrial and nuclear gene sequences, 742

with a new tribal classification of the subfamily. BMC Evol. Biol. 8, 199. 743

Lindsay, E.H., 1988. Cricetid Rodents from Siwalik Deposits near Chinji Village. Part 1: 744

155


31

Megacricetodontinae, Myocricetodontinae and Dendromurinae. Palaeovertebrata. 18, 95-745

154. 746

López Antoñanzas, R., 2009. First potwarmus from the miocene of Saudi Arabia and the early 747

phylogeny of murines (Rodentia: Muroidea). Zool. J. Linn. Soc. 156, 664–679. 748

Lungu, A. N., 1981. Hipparion Fauna of Middle Sarmatian of Moldavia (Insectivora, 749

Lagomorpha and Rodentia). State Tiraspol Institut Pedagogiki Publ. 1557, 1-137. 750

Maddison, W. P. and Maddison, D.R., 2017. Mesquite: a modular system for evolutionary 751

analysis. Version 3.2 http://mesquiteproject.org 752

Macholán, M., 2012, (Ed.),Evolution of the house mouse (No. 3). Cambridge University Press. 753

Manthi, F.K., 2007. A preliminary review of the rodent fauna from Lemudong’o, southwestern 754

Kenya, and its implication to the late Miocene paleoenvironments. Kirtlandia. 56, 92–105. 755

Martín-Suáres, E., Mein, P., 1998. Revision of the genera Parapodemus, Apodemus Rhagamys 756

and Rhagapodemus (Rodentia, Mammalia). Geobios. 31, 87–97. 757

Matzke, N.J., 2013. Probabilistic historical biogeography: new models for founder-event 758

speciation, imperfect detection, and fossils allow improved accurancy and model-testing. 759

Berkeley Plan. J. 26, 217–220. 760

Matzke, N.J., 2014. Model selection in historical biogeography reveals that founder-event 761

speciation is a crucial process in island clades. Syst. Biol. 63, 951–970. 762

Mein, P., Martín-Suárez, E., Agustí, J. 1993. Progonomys Schaub, 1938 and Huerzelerimys 763

gen. nov.(Rodentia); their evolution in Western Europe. Scripta Geol. 103, 41-64. 764

Mein, P., 2003. On Neogene rodents of Eurasia: distributions and migrations - in: Reumer, 765

J.W.F., Wessels, W. (Eds.), Distribution and migration of Tertiary mammals in Eurasia. 766

A volume in honour of Hans de Bruijn. Natuurmuseum Rotterdam. pp. 407-418. 767

Mein, P., Pickford, M., Senut, B., 2004. Late Miocene micromammals from the Harasib karst 768

deposits, Namibia. Part 2b-Cricetomyidae, Dendromuridae and Muridae, with an 769

156


32

addendum on the Myocricetodontinae. Communs. Geol. Surv. Namibia. 13, 43–63. 770

Michaux, J. R., Chevret, P., Filippucci, M. G., Macholán, M., 2002. Phylogeny of the genus 771

Apodemus with a special emphasis on the subgenus Sylvaemus using the nuclear IRBP 772

gene and two mitochondrial markers: cytochrome b and 12S rRNA. Mol. Phylogenet. 773

Evol. 23, 123-136. 774

Michaux, J., Chevret, P., Renaud, S., 2007. Morphological diversity of Old World rats and mice 775

(Rodentia, Muridae) mandible in relation with phylogeny and adaptation. J. Zool. Syst. 776

Evol. Res. 45 (3), 263-279. 777

Miller, M.A., Pfeiffer, W., Schwartz, T., 2010. Creating the CIPRES Science Gateway for 778

inference of large phylogenetic trees, in: 2010 Gateway Computing Environments 779

Workshop, GCE 2010. pp. 1–8. 780

Monadjem, A., Taylor, P. J., Denys, C., Cotterill, F. P., 2015. Rodents of sub-Saharan Africa: 781

a biogeographic and taxonomic synthesis. Walter de Gruyter GmbH, 782

Berlin/Munich/Boston. 783

Montgelard, C., Forty, E., Arnal, V., Matthee, C. A., 2008. Suprafamilial relationships among 784

Rodentia and the phylogenetic effect of removing fast-evolving nucleotides in 785

mitochondrial, exon and intron fragments. BMC Evol. Biol. 8, 321. 786

Montuire, S., Maridet, O., Legendre, S., 2006. Late Miocene–Early Pliocene temperature 787

estimates in Europe using rodents. Palaeogeogr. Palaeoclimatol. Palaeoecol. 238, 247-788

262. 789

Munthe, J., 1987. Small-mammal fossils from the Pliocene Sahabi Formation of Libya. in. 790

Boaz, N.T., El-Arnauti,A., Gaziry, A.W, de Heinzelin, J., Boaz, D.D, (Eds.), Neogene 791

Paleontology and Geology of Sahabi, Alan R. Liss. New York. pp. 135–144. 792

Musser, G.G., Carleton M.D., 2005. Superfamily Muroidea, in: Wilson, D.E., Reeder, D.A.M. 793

(Eds.), Mammal Species of the World. A Taxonomic and Geographic Reference. The 794

157


33

Johns Hopkins University Press. Baltimore. pp. 894–1531. 795

Ndiaye, A., Chevret, P., Dobigny, G., Granjon, L., 2016. Evolutionary systematics and 796

biogeography of the arid habitat-adapted rodent genus Gerbillus (Rodentia, Muridae): a 797

mostly Plio-Pleistocene African history. J. Zool. Syst. Evol. Res. 54, 299-317. 798

Near, T.J., Meylan, P. A., Shaffer, H.B., 2005. Assessing concordance of fossil calibration 799

points in molecular clock studies: an example using turtles. Am. Nat. 165, 137–146. 800

Near, T.J., Sanderson, M.J., 2004. Assessing the quality of molecular divergence time estimates 801

by fossil calibrations and fossil-based model selection. Philos. Trans. R. Soc. Lond. B. 802

Biol. Sci. 359, 1477–1483. 803

Nicolas, V., Mboumba, J.F., Verheyen, E., Denys, C., Lecompte, E., Olayemi, A., Missoup, 804

A.D., Katuala, P., Colyn, M., 2008. Phylogeographic structure and regional history of 805

Lemniscomys striatus (Rodentia: Muridae) in tropical Africa. J. Biogeogr. 35, 2074–2089. 806

Nylander, J.A.A., Ronquist, F., Huelsenbeck, J.P., Nieves-Aldrey, J.L., 2004. Bayesian 807

Phylogenetic Analysis of Combined Data. Syst. Biol. 53, 47–67. 808

Ogg, J. G., Smith, A. G., 2004. Geologic Time Scale 2004. in: Gradstein, F. M., Ogg, J. G., 809

Smith, A. G.A (Eds.), Cambridge University Press. pp. 63–86. 810

Olson, D.M., Dinerstein, E., Wikramanayake, E.D., Burgess, N.D., Powell, G.V.N., 811

Underwood, E.C., D’amico, J. A., Itoua, I., Strand, H.E., Morrison, J.C., Loucks, C.J., 812

Allnutt, T.F., Ricketts, T.H., Kura, Y., Lamoreux, J.F., Wettengel, W.W., Hedao, P., 813

Kassem, K.R., 2001. Terrestrial Ecoregions of the World: A New Map of Life on Earth. 814

Bioscience 51, 933. 815

Pagès, M., Fabre, P.-H., Chaval, Y., Mortelliti, A., Nicolas, V., Wells, K., Michaux, J.R., 816

Lazzari, V., 2016. Molecular phylogeny of Southeast Asian arboreal murine rodents. Zool. 817

Scr. 45, 349-364. 818

Parham, J.F., Donoghue, P.C.J., Bell, C.J., Calway, T.D., Head, J.J., Holroyd, P. A., Inoue, 819

158


34

J.G., Irmis, R.B., Joyce, W.G., Ksepka, D.T., Patané, J.S.L., Smith, N.D., Tarver, J.E., 820

Van Tuinen, M., Yang, Z., Angielczyk, K.D., Greenwood, J.M., Hipsley, C. A., Jacobs, 821

L., Makovicky, P.J., Müller, J., Smith, K.T., Theodor, J.M., Warnock, R.C.M., Benton, 822

M.J., 2012. Best practices for justifying fossil calibrations. Syst. Biol. 61, 346–359. 823

Pook, C. E., Joger, U., Stümpel, N., Wüster, W., 2009. When continents collide: phylogeny, 824

historical biogeography and systematics of the medically important viper genus Echis 825

(Squamata: Serpentes: Viperidae). Mol. Phylogenet. Evol. 53, 792-807. 826

Portik, D. M., Papenfuss, T. J., 2012. Monitors cross the Red Sea: the biogeographic history 827

of Varanus yemenensis. Mol. Phylogenet. Evol. 62, 561-565. 828

Prista, G.A., Agostinho, R.J., Cachão, M.A., 2015. Observing the past to better understand the 829

future: A synthesis of the Neogene climate in Europe and its perspectives on present 830

climate change. Open Geosci. 7, 65–83. 831

Pyron, R.A., 2010. A likelihood method for assessing molecular divergence time estimates and 832

the placement of fossil calibrations. Syst. Biol. 59, 185–194. 833

Qiu, Z., Zheng, S., Zhang, Z., 2004. Murids from the late Miocene Bahe formation, Lantian, 834

Shaanxi. Vertebrata Pal. Asiatica, 42, 67-76. 835

Ree, R.H., Smith, S. A., 2008. Maximum likelihood inference of geographic range evolution 836

by dispersal, local extinction, and cladogenesis. Syst. Biol. 57, 4–14. 837

Reed, D. N., 2011. New murid (Mammalia, Rodentia) fossils from a late Pliocene (2.4 Ma) 838

locality, Hadar A. L. 894, Afar Region, Ethiopia. J. Vert. Paleontol. 31, 1326–1337. 839

Renaud, S., Michaux, J., Mein, P., Aguilar, J.P., Auffray, J.C., 1999. Patterns of size and shape 840

differentiation during the evolutionary radiation of the European Miocene murine rodents. 841

Lethaia 32, 61–71. 842

Ripplinger, J., Sullivan, J., 2008. Does choice in model selection affect maximum likelihood 843

analysis? Syst. Biol. 57, 76–85. 844

159


35

Ronquist, F., 1997. Dispersal-Vicariance Analysis: A New Approach to the Quantification of 845

Historical Biogeography. Syst. Biol. 46, 195. 846

Ronquist, F., Klopfstein, S., Vilhelmsen, L., Schulmeister, S., Murray, D.L., Rasnitsyn, A.P., 847

2012a. A total-evidence approach to dating with fossils, applied to the early radiation of 848

the hymenoptera. Syst. Biol. 61, 973–999. 849

Ronquist, F., Teslenko, M., van der Mark, P., Ayres, D.L., Darling, A., Höhna, S., Larget, B., 850

Liu, L., Suchard, M. A, Huelsenbeck, J.P., 2012b. MrBayes 3.2: efficient Bayesian 851

phylogenetic inference and model choice across a large model space. Syst. Biol. 61, 539–852

42. 853

Rowe, K.C., Achmadi, A.S., Esselstyn, J.A., 2016a. A new genus and species of omnivorous 854

rodent (Muridae: Murinae) from Sulawesi, nested within a clade of endemic carnivores. J. 855

Mammal. 97, 978-991. 856

Rowe, K.C., Achmadi, A.S., Esselstyn, J.A., 2016b. Repeated evolution of carnivory among 857

Indo-Australian rodents. Evolution. 70, 653-665. 858

Rowe, K.C., Aplin, K.P., Baverstock, P.R., Moritz, C., 2011. Recent and rapid speciation with 859

limited morphological disparity in the genus rattus. Syst. Biol. 60, 188–203. 860

Rowe, K.C., Reno, M.L., Richmond, D.M., Adkins, R.M., Steppan, S.J., 2008. Pliocene 861

colonization and adaptive radiations in Australia and New Guinea (Sahul): Multilocus 862

systematics of the old endemic rodents (Muroidea: Murinae). Mol. Phylogenet. Evol. 47, 863

84–101. 864

Rögl, F., 1999. Mediterranean and Paratethys. Facts and hypotheses of an Oligocene to 865

Miocene paleogeography (short overview). Geol. Carpathica. 50, 339–349. 866

Sabatier, M., 1982. Les rongeurs du site Pliocène à Hominidés de Hadar (Ethiopie). 867

Palaeovertebrata. 12. 1–56. 868

Sauquet, H., Ho, S.Y.W., Gandolfo, M.A., Jordan, G.J., Wilf, P., Cantrill, D.J., Bayly, M.J., 869

160


36

Bromham, L., Brown, G.K., Carpenter, R.J., Lee, D.M., Murphy, D.J., Sniderman, J.M.K., 870

Udovicic, F., 2012. Testing the impact of calibration on molecular divergence times using 871

a fossil-rich group: The case of nothofagus (Fagales). Syst. Biol. 61, 289–313. 872

Sen, S. 1977. La faune de rongeurs Pliocenes de Calta (Ankara, Turquie). Bull. Mus. Natl. Hist. 873

Nat. Sci. Terre. 465, 1–171. 874

Sen, S., 1983. Rongeurs et lagomorphes du gisement pliocène de Pul-e Charkhi, bassin de 875

Kabul, Afghanistan. Bull. Mus. Natl. Hist. Nat. 5, 33-74. 876

Schenk, J. J., Rowe, K. C., Steppan, S. J., 2013. Ecological opportunity and incumbency in the 877

diversification of repeated continental colonizations by muroid rodents. Syst. Biol. 62, 878

837-864. 879

Simpson, G. G., 1945. The principles of classification and a new classification of mammals. 880

Bull. Am. Mus. Nat. Hist. 85, 1–350. 881

Stadler, T., 2010. Sampling-through-time in birth-death trees. J. Theor. Biol. 267, 396–404. 882

Stamatakis, A., 2014. RAxML version 8 a tool for phylogenetic analysis and post-analysis of 883

large phylogenies. Bioinformatics. 30, 1312-1313. 884

Stanhope, M. J., Czelusniak, J., Si, J. S., Nickerson, J., Goodman, M., 1992. A molecular 885

perspective on mammalian evolution from the gene encoding interphotoreceptor retinoid 886

binding protein, with convincing evidence for bat monophyly. Mol. Phylogenet. Evol. 1, 887

148-160. 888

Steppan, S., Adkins, R., Anderson, J., 2004. Phylogeny and divergence-date estimates of rapid 889

radiations in muroid rodents based on multiple nuclear genes. Syst. Biol. 53, 533–553. 890

Steppan, S.J., Adkins, R.M., Spinks, P.Q., Hale, C., 2005. Multigene phylogeny of the Old 891

World mice, Murinae, reveals distinct geographic lineages and the declining utility of 892

mitochondrial genes compared to nuclear genes. Mol. Phylogenet. Evol. 37, 370–88. 893

Suwa, G., Beyene, Y., Nakaya, H., Bernor, R. L., Boisserie, J. R., Bibi, F., ... & Asfaw, B. 2015. 894

161


37

Newly discovered cercopithecid, equid and other mammalian fossils from the Chorora 895

Formation, Ethiopia. J. Anthropol. Sci. 123, 19-39. 896

!"#$%&'$%()**)+,)$%-'$%.*)/0123"4$%5'$%6307#"8$%9'$%:);2<*$%='.'$%>0*)3<1$%&'$%?@AB'%CD/%0E%897

Arabia: a complex biogeographic history of multiple vicariance and dispersal events in the 898

gecko genus Hemidactylus (Reptilia: Gekkonidae). PLoS One. 8, e64018. 899

Teeling, E.C., Scally, M., Kao, D.J., Romagnoli, M.L., Springer, M.S., Stanhope, M.J., 2000. 900

Molecular evidence regarding the origin of echolocation and flight in bats. Nature, 403, 901

188–192. 902

Vangengeim, E.A., Lungu, A.N., Tesakov, A.S., 2006. Age of the Vallesian Lower Boundary 903

(Continental Miocene of Europe). Stratigr. Geol. Correl. 14, 655-667. 904

Vaughan, T.A., Ryan, J.M., Czaplewski, N.J., 2011. Mammology. 5th ed. Jones and Bartlett 905

Publishers, Sudbury. 906

Vrba, E. S. 1995. On the connections between paleoclimate and evolution. in: Vrba, E.S., 907

Denton, G.H., Partridge,T.C., Burckle, L.H. (Eds.), Paleoclimate and Evolution with 908

Emphasis on Human Origins. Yale University Press. New Haven. pp. 24–48. 909

Weerd, A. D., 1976. Rodent faunas of the Mio-Pliocene continental sediments of the Teruel-910

Alfambra region, Spain. Doctoral dissertation, Utrecht University. 911

Wessels, W., 2009. Miocene rodent evolution and migration. Muroidea from Pakistan, Turkey 912

and Northern Africa. Geologica Ultraiectina 307. Doctoral dissertation, Departement 913

Aardwetenschappen. 914

Winkler, A.J., 2002. Neogene paleobiogeography and East African paleoenvironments: 915

contributions from the Tugen Hills rodents and lagomorphs. J. Hum. Evol. 42, 237–256. 916

Winkler, A.J., Denys, C., Avery, D.M., 2010. Rodentia, in: Werdelin, L., Sanders, W.J. (Eds.), 917

Cenozoic mammals of Africa. Univ of California Press. pp. 261–304. 918

Winney, B. J., Hammond, R. L., Macasero, W., Flores, B., Boug, A., Biquand, V., ... & Bruford, 919

162


38

M. W. 2004. Crossing the Red Sea: phylogeography of the hamadryas baboon, Papio 920

hamadryas hamadryas. Mol. Ecol. 13, 2819-2827. 921

922

923

Figure titles 924

Figure 1. Molecular phylogeny of family Muridae based on Bayesian inference (BI) in 925

MrBayes. Very similar topology was obtained also by maximum likelihood (ML) 926

analysis in RAxML (see Appendix D). Red points show nodes supported in BI 927

analysis (posterior probability PP !0.95), blue points show high bootstrap support 928

in ML analysis (bootstrap BV 0.70). Violet nodes are supported by both analyses. 929

Figure 2. Divergence dating analysis of family Muridae. Nodes show medians of times to most 930

recent common ancestor (MRCA), node bars indicate 95% HPD intervals. Latin 931

numbers in yellow squares indicate positions of fossil constrains selected by 932

multiple-step evaluation and used for final analysis (see Table 4 for more details). 933

Figure 3. Ancestral reconstruction for family Muridae with BioGeoBEARS (DEC+J; d=0.008; 934

e=0; j=0.0246; LnL=-117.11). Five biogeographical areas are represented using 935

different colours: A, West Palearctic (dark blue); B, East Palearctic (light blue); C, 936

Indomalaya (green); D, Australasia (yellow); E, Afrotropics (red). 937

938

939

940

163


39

Tables: 941

Table 1: Brief summary of taxonomic changes in the family Muridae. The taxonomic position 942

of particular lineages has changed significantly and they were either considered as 943

separate families, or included in other families outside Muridae (names of families 944

are in bold). 945

Table 2: PartitionFinder results showing optimum partitioning schemes and best fit models 946

for each analysis (RAxML, MrBayes, BEAST). Settings: BIC, unlinked branch 947

lengths, greedy algorithm. 948

Table 3: Overview of 18 candidate fossils with stratigraphic age, locality and relevant references. For 949

more details see Appendix B. 950

Table 4: Overview of fossils finally selected for divergence dating, with parameters of uniform 951

and lognormal prior distribution. 952

Table 5: Comparison of models used for BioGeoBEARS; likelihood scores (LnL), number of 953

parameters (numparams), dispersal rate (d), extinction rate (e), free parameter controlling 954

the relative probability of founder-event speciation events at cladogenesis (j), corrected 955

Akaike Information Criterion (AICc), and AICc model weights. 956

Table 6: Results of divergence dating analysis. Time to the most common ancestor (MRCA) is 957

shown as median in Ma, with 95% highest posterior density (HPD). Estimates from 958

previous studies dealing with divergence dating of murid rodents are reviewed here 959

for comparison. 960

Table 7: Comparison between previous relevant studies. 961

962 963

164


40

Supplementary material 964

Appendix A. List of taxa and genetic markers. 965

Appendix B. Description of considered fossils. 966

Appendix C. Results of cross-validation of fossil constraints. 967

Appendix D. Maximum likelihood phylogenetic tree. 968

165

Ta

tian

a A

gh

ová

/ Ph

.D. d

isserta

tion

(2018

)

Table 1: Brief summary of taxonomic changes in the family Muridae, The taxonomic position of particular lineages has changed significantly and they were either considered separate families or included in other families outside Muridae (name of families are in bold). Simpson (1945)

Chaline et al. (1977)

Lavocat (1978)

Cricetidae Lophiomyinae Cricetidae Lophiomyinae Nesomyidae Lophiomyinae Gerbillinae Gerbillidae

Gerbillinae Muridae Murinae Muridae Murinae Otomyinae

Otomyinae Nesomyidae Otomyinae Muridae Carleton and Musser (1984) Musser and Carleton (2005) Granjon and Montgelard (2012)

Muridae Lophiomyinae Cricetidae Lophiomyinae Muridae Lophiomyinae Gerbillinae Muridae Otomyinae Gerbillinae Murinae Gerbillinae Murinae Otomyinae Murinae Deomyinae

Deomyinae Leimacomyinae Leimacomyinae

166

Ta

tian

a A

gh

ová

/ Ph

.D. d

isserta

tion

(2018

)

Table 2: PartitionFinder results showing optimum partitioning schemes and best fit models for each analysis (RAxML, MrBayes, BEAST). Settings: BIC, unlinked branch lengths, greedy algorithm. Analysis N. of

part

Best sub.

model

Subsets

RAxML 3 GTR+I+G 12S, 16S, ATPase8_pos1-pos2, COII_pos1-pos2, COI_pos1-pos2, CYTB_pos1-pos2, tRNA-Asp, tRNA-Lys, tRNA-Ser GTR+I+G ATPase8_pos3, COII_pos3, COI_pos3, CYTB_pos3 GTR+I+G AP5_pos1-pos3, BRCA_pos1-pos3, BZRP, GHR_pos1-pos3, IRBP_pos1-pos3, RAG_pos1-pos3

MrBayes 3 GTR+I+G 12S, 16S, ATPase8_pos1-pos2, COII_pos1-pos2, COI_pos1-pos2, CYTB_pos1-pos2, tRNA-Asp, tRNA-Lys, tRNA-Ser

and BEAST HKY+I+G ATPase8_pos3, COII_pos3, COI_pos3, CYTB_pos3 GTR+I+G AP5_pos1-pos3, BRCA_pos1-pos3, BZRP, GHR_pos1-pos3, IRBP_pos1-pos3, RAG_pos1-pos3

167

Ta

tian

a A

gh

ová

/ Ph

.D. d

isserta

tion

(2018

)

Table 3: Overview of 18 candidate fossils with stratigraphic age, locality and relevant references. For more details see Appendix B. Fossil Final

analysis

Reason for

excluding

Age (Ma) Dating Fossils

Method References Site References

†Potwarmus primitivus

(Wessels)

No Equivocal placement

16.0 magnetostratigraphy time scale of Ogg and Smith (2004)

Pakistan, Potwar Plateau, YGSP591

Lindsay (1988), Wessels (2009)

†Antemus chinjiensis

Jacobs

Yes (1) 13.8 magnetostratigraphy Jacobs and Flynn (2005) Pakistan, Potwar Plateau, YGSP 491

Jacobs et al. (1989)

†cf. Progonomys sp.

Schaub


11.6 magnetostratigraphy time scale of Ogg and Smith (2004)

Pakistan, Potwar Plateau, YGSP 83, YGSP 504

Jacobs and Flynn (2005), Cheema et al. (2000), Kimura et al. (in prep.)

†cf. Karnimata sp.

Jacobs

Yes (2) 11.2 magnetostratigraphy Kimura et al. (2015) by time scale of Ogg and Smith (2004)

Pakistan, Siwalik Group, Nagri Formation, YGSP 791, YGSP 797,

Jacobs and Flynn (2005); Kimura et al. (2015)

†Parapodemus

lungdunensis Schau

Yes (3) 9.6 magnetostratigraphy Daxner-Höck (2003) France, Dionay Lungu (1981), Mein et al. (1993), Renaud et al. (1999)

†Karnimata darwini

Jacobs

Yes (4) 9.2 magnetostratigraphy Kimura et al. (2015) by time scale of Ogg and Smith (2004)

Pakistan, Siwalik Group, Dhok Pathan Formation, YGSP 182

Jacobs (1978); Kimura et al. (2015)

†Abudhabia

pakistanensis Flynn and

Jacobs

Yes (5) 8.7 magnetostratigraphy Flynn and Jacobs (1999) by time scale of Ogg and Smith (2004)

Pakistan, Siwalik Group, Dhok Pathan Formation, YGSP387

Flynn and Jacobs (1999)

†aff. Stenocephalomys

Frick


8.5 40K/40Ar Geraads et al. (2002), Suwa et al. (2015)

Ethiopia, Chorora Geraads (2001)

†cf. Parapelomys Jacobs No Equivocal placement



168

Ta

tian

a A

gh

ová

/ Ph

.D. d

isserta

tion

(2018

)

Table 3 (continued)

Fossil Final

analysis

Reason for

excluding

Age (Ma) Dating Fossils

Method References Site References

†Preacomys kikiae

Geraads




†Mus sp. Linneaus Yes (6) 8.0 magnetostratigraphy Kimura et al. (2015) by time scale of Ogg and Smith (2004)

Pakistan, Siwalik Group, Dhok Pathan Formation, YGSP 547

Kimura et al. (2013; 2015)

†Acomys sp. I. Geoffroy No missing M1 6.1 40Ar / 39Ar Deino and Ambrose (2007)

Kenya, Lemudong’o,

locality 1 Manthi (2007)

†Aethomys sp.Thomas Yes (7) 6.1 40Ar / 39Ar Deino and Ambrose (2007)

Kenya, Lemudong'o, locality 1

Manthi (2007)

†Arvicanthis sp. Lesson Yes (8) 6.1 40Ar / 39Ar Deino and Ambrose (2007)



†Gerbilliscus sp.

(Thomas)

Yes (9) 6,1 40Ar / 39Ar Deino and Ambrose (2007)

Kenya, Lemudong'o, locality 1

Manthi (2007)

†Lemniscomys sp.

Trouessart

No missing M1 6.1 40Ar / 39Ar Deino and Ambrose (2007)



†Mastomys sp. Thomas No cross-validation

6.1 40Ar / 39Ar Deino and Ambrose (2007)



†Gerbillus sp.

Desmarest

No cross-validation

2.4 geochronology (BKT-3 tephra), 40K/40Ar, 40Ar / 39Ar; sedimentology

Kimbel et al. (1996), Campisano and Feibel (2008)

Ethiopia, Hadar, AL894 Reed (2011)

169

Ta

tian

a A

gh

ová

/ Ph

.D. d

isserta

tion

(2018

)

Table 4: Overview of fossils finally selected for divergence dating, with parameters of uniform and lognormal prior distribution.

Fossil Position Stem/Crown

Age (Ma)

Uniform

distribution

Lognormal

distribution

Min Max Offset Log Mean

1 †Antemus chinjiensis crown Murinae crown 13.8 13.24 25.29 13.24 1.0 3.2

2 †cf. Karnimata sp. Mus/Arvicanthis split crown 11.2 10.47 25.37 10.47 1.0 4.0

3 †Parapodemus lugdunensis Apodemus/Tokudaia split stem 9.6 8.93 25.41 8.93 1.0 4.5

4 †Karnimata darwini TMRCA Millardini/Otomyini/Arvicanthini crown 9.2 8.52 25.42 8.52 1.0 4.6

5 †Abudhabia pakistanensis Gerbilliscus/Desmodillus split crown 8.7 8.01 25.43 8.01 1.0 4.7

6 †Mus sp. Murini stem 8.0 7.29 25.45 7.29 1.0 4.9

7 †Aethomys sp. Aethomys stem 6.1 5.34 25.50 5.34 1.0 5.5

8 †Arvicanthis sp. Arvicanthis stem 6.1 5.34 25.50 5.34 1.0 5.5

9 †Gerbilliscus sp. Gerbilliscus stem 6.1 5.34 25.50 5.34 1.0 5.5

170

Ta

tian

a A

gh

ová

/ Ph

.D. d

isserta

tion

(2018

)

Table 5: Comparison of models used for BioGeoBEARS; likelihood scores (LnL), number of parameters (numparams), dispersal rate (d), extinction rate (e), free parameter controlling the relative probability of founder-event speciation events at cladogenesis (j), corrected Akaike Information Criterion (AICc), and AICc model weights.

LnL numparams d e j AICc AICc_wt

DEC -133.724 2 0.005 0.000 0.000 271.525 0.05181 DEC +J -129.826 3 0.004 0.000 0.008 265.806 0.90415 DIVALIKE -146.958 2 0.007 0.000 0.000 297.993 9.2665e-08 DIVALIKE +J -132.856 3 0.004 0.000 0.012 271.866 0.04369 BAYAREALIKE -178.311 2 0.005 0.036 0.000 360.698 2.2423e-21 BAYAREALIKE +J -137.687 3 0.003 0.000 0.014 281.528 0.00035

171

Ta

tian

a A

gh

ová

/ Ph

.D. d

isserta

tion

(2018

)

Table 6: Results of divergence dating analysis. Time to the most common ancestor (TMRCA) is shown as median in Ma, with 95% highest posterior density

(HPD). Estimates from previous studies dealing with divergence dating of murid rodents are reviewed here for comparison.

TMRCA median 2.5% 97.5% Steppan et al.

(2004)

Lecompte et al.

(2008)

Rowe et al.

(2011)

Fabre et al.

(2012)

Fabre et al.

(2013)

Schenk et al.

(2013)

Bryja et al.

(2014)

Rowe et al.

(2016b)

Muridae 19.3 17.06 21.92 22.5 ~33-23 ~21.0 Lophiomyinae* 18.6 16.35 21.11

Gerbillinae 12.2 10.46 14.16 9.3 ~23-5 ~10.0 Deomyinae 14.6 12.68 16.82 13.1 ~23-5 ~13.0 Murinae 14.2 12.70 16.07 12.0 12.3 13.3 ~23-5 ~14.5 ~14.0 "core Murinae" 12.3 11.28 13.62 10.3 11.3 11.60 ~12.5 ~12.5 Phloeomyini 9.8 7.72 11.85 8.6 ~10.0 ~10.5 Rattini 10.4 9.23 11.93 9.7 8.70 ~11.0 ~11.0 Hydromyini 10.4 9.31 11.71 8.9 ~11.0 ~11.0 Malacomyini 3.4 2.35 4.60 4.4 Apodemini 9.0 7.92 10.13 9.6 ~7.5 9.5 ~7.5 Murini 7.2 6.24 8.29 6.6 5.3 ~6.0 7.4 ~6.5 Praomyini 6.8 6.05 7.77 7.6 ~5.0 6.8 ~6.0 Millardini 9.2 7.72 10.75 Otomyini 4.9 4.12 5.98 ~6.0 3.8 ~3.0 Arvicanthini 8.8 7.92 9.72 8.4 7.3 ~7.5 7.8 ~8.5 * offshoot from Gerbillinae + Deomyinae

172

Ta

tian

a A

gh

ová

/ Ph

.D. d

isserta

tion

(2018

)

Table 7: Comparison between previous relevant studies. Author Focus Subfamilies # genera # species Fossils in

Muridae # genetic markers

Fabre et al. (2012) Rodentia 4 105 302 0 11 (8)*

Schenk et al. (2013) Muroidea 4 85 136 4 4

our study Muridae 4 82 160 9 15

Rowe et al. (2008) Hydromyini (outgroup all Murinae + Gerbillinae+Deomyinae)

3 66 78 2 8

Lecompte et al. (2008) Murinae 3 46 83 2 3

* 11 genetic markers for Rodentia, only 8 for Muridae

173


Calomyscus baluchiOchrotomys nuttalliMesocricetus auratusOndatra zibethicusOtotylomys phyllotisTylomys nudicaudusLophiomys imhausiDesmodillus auricularisGerbilliscus robustusGerbillurus paebaGerbilliscus gambianusGerbilliscus giffardiGerbillus nanusGerbillus nancillusGerbillus gerbillusGerbillus nigeriaeGerbillus tarabuliUranomys ruddiLophuromys flavopunctatusLophuromys sikapusiDeomys ferrugineusAcomys subspinosusAcomys spinosissimusAcomys russatusAcomys wilsoniAcomys ignitusAcomys cineraceusAcomys chudeauiBatomys grantiPhloeomys cumingiPhloeomys sp.Micromys minutusMaxomys bartelsiiMaxomys suriferMelasmothrix nasoDacnomys millardiLeopoldamys sabanusChiromyscus chiropusNiviventer cremoriventerNiviventer confucianusNiviventer culturatusBerylmys bowersiSundamys muelleriBunomys andrewsiParuromys dominatorDiplothrix legataRattus exulansRattus leucopusRattus norvegicusChiropodomys gliroidesApomys dataeApomys hylocoetesArchboldomys luzonensisChrotomys gonzalesiRhynchomys isarogensisAnisomys imitatorLorentzimys nouhuysiChiruromys vatesMacruromys majorHyomys goliathPogonomys loriaePogonomys macrourusMammelomys lanosusAbeomelomys seviaMallomys rothschildiLeptomys elegansHydromys chrysogasterParahydromys asperPseudohydromys ellermaniXeromys myoidesLeggadina forrestiZyzomys argurusNotomys fuscusMastacomys fuscusConilurus penicillatusLeporillus conditorUromys caudimaculatusParamelomys levipesMelomys rufescensMelomys cervinipesSolomys salebrosusMalacomys edvardsiMalacomys longipesTokudaia osimensisApodemus mystacinusApodemus agrariusApodemus semotusMus cookiiMus musculusMus crociduroidesMus pahariMus platythrixMus haussaMus mattheyiMus minutoidesMus musculoidesPraomys delectorumHeimyscus fumosusHylomyscus parvusPraomys degraaffiPraomys jacksoniPraomys daltoniPraomys misonneiPraomys tullbergiPraomys morioPraomys obscurusPraomys lukolelaeMyomyscus verrauxiiComolys goslingiZelotomys hildegardeaeStenocephalemys griseicaudaStenocephalemys albipesStenocephalemys albocaudataMyomyscus brockmaniMastomys kollmannspergeriMastomys couchaMastomys hubertiMastomys natalensisMastomys awashensisMastomys erythroleucusCremnomys cutchicusMillardia kathleenaeMillardia meltadaOtomys dentiOtomys barbouriOtomys lacustrisParotomys brantsiiParotomys littledaleiOtomys irroratusOtomys angoniensisOtomys typusOtomys occidentalisOtomys orestesOtomys tropicalisOenomys hypoxanthusGolunda elliotiMicaelamys namaquensisPelomys fallaxRhabdomys dilectusRhabdomys pumilioArvicanthis nairobaeArvicanthis rufinusArvicanthis niloticusArvicanthis blickiArvicanthis neumanniLemniscomys striatusLemnsicomys zebraLemnsicomys rosaliaLemnsicomys bellieriLemniscomys macculusHybomys univittatusStochomys longicaudatusAethomys hindeiAethomys chrysophilusAethomys ineptusAethomys silindensisAethomys kaiseriAethomys nyikaeDasymys incomtusThallomys loringiThallomys nigricaudaThallomys paedulcusGrammomys dolichurusGrammomys cometesGrammomys macmillaniGrammomys surdaster

CalCalCalomyomyomyscuscuscus balbalbalbalbalbaluchuchuchuchuchiiOchOchOchrotrotrotomyomyomysCalCalomyomyomyscuscuscusCalCalomyomyomy s

nutnutnuttaltallibalbalbaluchuchiuchi

MesMesMesMesocrocrocriceicetustusOchOchrotrotrotomyomyomysOch omyomyomys nut

auraurauratuatusOndOndOndatratraMesMesocrocrocriceMesMesocrocr

zibzibzibethethethicuicusiceicetustustus aurauratuauraur

OtoOtotyltyltylomyomyssOndOndOndatratraOndatra zibzibzib

phyphyphyllollotistiszibethethethicuicuszibethicus

TylomyomyomyomysOtoOtotyltyltylomyOtotyltyltylomy

nudnudnudnudicaicaicauduudussomyomyssomyomy phyphyphyllollotistisphyphyphyllotis

LopLopLopLopLopLophiohiohiohiohiomysmysmysmysTylTylomyomyomyomyomyomysTylTylomyomyomyomyomys nudnudnudnud

imhimhimhimhimhimhimhausausausausausausiiinudicaicaicauduudusnudicaicaudus

DesDesDesDesDesDesmodmodmodmodmodillillillillillusususususLopLopLopLopLopLophiohiohiomysmysmysmysLopLopLopLopLop mysmysmys imhimh

auraurauraurauraurauricuicuicuicuicuicuicuicuicularlarlarlarlarlarlarlarimhausausausausi

GerGerbilbilbilliscuscuscusDesDesmodmodillillillusDesDesmodmod

robrobrobrobustustusususaurauraurauricuicularlaraur

GerGerbilbilbillurlurususGerGerbilbilbilliscuscuscusGerGerbilbilbillis

paepaepaepaepaebabacus robrobrobustustrobustust

GerGerbilbilbilliscuscuscusGerGerbilbilbillurlurususGerGerbilbilbillur paepae

gamgamgamgamgamgambiabiabianusnusnuspaepaepaepaepaepaebabapaepaepaepaebaba

GerGerbilbilbilliscuscuscusGerGerbilbilbilliscuscuscusGerGerbilbilbillis

gifgifgifgifgifgiffarfarfarfardididigamgamgamgamgambiabiabianusgamgamgamgambiabiabianus

GerGerbilbilbilluslusGerGerbilbilbilliscuscusGerGerbilbilbillis

nannannanusususususgifgifgifgifgiffarfarfargifgifgifgifgiffarfarfar

GerGerbilbilbilluslusGerGerbilbilluslusGerGerbilbillus

nannannancilcilcilcilcilluslusnannannanusususususus

GerGerbilbilbilluslusGerGerbilbilluslusGerGerbilbillus

gergergergerbilbilbilbilbilbilluslusnannancilcilcilcilluslusnancilcilcilcillus

GerGerbilbilbilluslusGerGerbilbilbilluslusGerGerbilbilbillus

nignignigerierierierieriaeaegergergerbilbilbilluslusgergerbillus

GerGerGerGerbilbilbilbilbilluslusluslusGerGerbilbilbilluslusGerGerbilbilbillus

tartartartarabuabuabuabuabuabuabulililinignignigerierierieriaeaenignignigerieriae

UraUraUraUranomnomnomysysysGerbillusGer tar

rudrudrudruddididitartarabuabu

LopLophurhuromyomysUraUranomnomnomysUraUranomys rud

flaflavopvopunctatusususruddiruddi

LopLophurhuromyomysLopLophurhuromyomysLopLophurhuromyomys

sikapuapusiflavopvopuncflavopvopunc

DeoDeomysmysLopLophurhuromyomyLopLop omyomy

ferrugrugineineusomysomys sikapuapusisikapuapusi

AcoAcomysmysDeoDeomysmysDeomysmys

subspispinosnosusferrugrugineineusferrugrugineus

AcoAcomysmysAcoAcomysmysAcomysmys

spinosnosississimussubspispinosnosussubspispinosnosus


russatsatusspinosnosissspinosnosiss

AcoAcomysmysAcoAcomysmysAco

wilsonsonirussatsatusrussatus


ignituitusAcoAcomysmysAcoAcomysmysAcomysmys

cineraeraceuceusignituitusign s


chudeadeauiuicineraeraceucineraceu

BatBatBatBatBatomyomyomyomyomyssssAcoAcoAcomysmysAcoAcomysmys chu

gragragrantintintintintintiPhlPhlPhloeooeooeomysmysmysBatBatBatomyomyomyomysssBatBatBatomyomyomyomy gragra

cumcumcumingingingigrantintintigra

PhlPhlPhlPhloeooeooeooeooeooeomysmysmysmysmysPhlPhlPhloeooeooeooeomysmysmysPhlPhlPhloeooeooeooeomysmysmys

spcumcum.

MicMicMicromromromysysysPhloeooeooeomysmysmysPhl mys

minminminutuutuutuutuutuutuutussmysmysmysmys spspspspspsp

MaxMaxMaxomyomyomyssMicMicMicromromromysysMic ysys

barbarteltelteltelsiiminminutuutuutusminutus

MaxMaxMaxomyomyomyssMaxMaxMaxomyomyomyssMaxomyomyomy

surifeifeifeiferbarteltelteltelsiibarteltelsii

MelMelMelasmasmasmothothothrixMaxMaxMaxomyomyomyssMaxMaxomyomyomy surifesurife

nasnasoifeifeifeife

DacDacDacnomnomnomysysysMelMelMelasmasmasmothothothMel

milmillarlarlardirixrix nasnasonasnaso

LeoLeoLeopolpolpoldamdamdamysDacDacDacnomnomnomysysysDacnomysysys mil

sabsabsabanumilmillarlarlardilardi

ChiChiChiromromromyscyscyscusLeoLeoLeopolpolpoldamdamdamysLeoLeopolpolpol ys

chichichichiropsabsabanusabanu

NivNivNiviveiveivententerrChiChiChiromromromyscyscyscusChiromyscyscyscus

crecremormormormormoriveyscusyscus chichichichiroprop

NivNivNiviveiveivententerrNivNivNiviveiveivententerrNivNivive

conconfucfucfuciancremormormormorivecremor

NivNivNiviveiveivententerrNivNivNiviveiveivententerrNivNivive

culculturturturturatuconconfucfucfucianconfucfucfucian

BerBerBerylmylmylmysNivNivNiviveiveiventerNiviveivente

bowbowersersersersirr culculturturturatu

SunSunSundamdamdamysysysBerBerBerylmylmylmysBerylmylmylmys bow

muemuemuemuellelleribowbowersersersiersi

BunBunBunomyomyomysSunSunSundamdamdamysSun ys

andandrewrewrewrewsiys muemuemuellellerimue ri

ParParParurourouromysmysmysBunBunBunomyomyomysBunomyomyomys and

domdomdomdominatorandrewrewrewrewsiand

DipDipDiplotlotlothrihrixxParParParurourouromysmysmysParParParurouromysmysmys

leglegataataataatadomdomdomdominadomdomdomdom

RatRatRattustustusDipDipDiplotlothriDipDipDiplothri

exuexuexulansshrihrixx legatalegata

RatRatRattustustusRatRatRattustustusRatRatRat

leuleuleucopususususexuexuexulanssexu

RatRatRattustustusRatRatRattustustusRat

nornornorvegvegicuicuicuicuicuicusleuleuleucopusususcopus

ChiChiChiChiChiropropropropropodoodoodomysmysmysmysRatRatRattustustusRat nornornornorvegvegicuicuicunornornorvegicu

gligligligligliroiroiroiroiroidesdesdesdesdesicuicuicusssicuicusss

ApoApoApomysmysmysChiChiropropropodoChiropropropodo

datdatdataeodoodomysmysodomysmys

ApoApoApomysmysmysApoApoApomysmysmysApoApoApomysmysmys

hylhylhylocoocoeteeteetesdatdatdataedatdatae

ArcArcArchbohbohboldoldomysmysApoApoApomysmysmysApoApomysmysmys hylhylhylocohylhylhyloco

luzluzluzoneoneonensinsisocoeteeteetesocoeteetes

ChrChrotootootomysmysArcArchbohbohboldoldomysArchbo mys

gongonzalzalzalesiesiesimysmys luzluzluzoneoneoneluzoneone

RhyRhynchnchnchomyomyomysChrChrotootootomysmysChrotomysmys gongon

isaisaisarogrogrogensensisisgongonzalzalzalesiesiesigon

AniAniAnisomsomsomysRhyRhynchnchnchomyRhyRhy omy

imiimitattattatororomyomysomyomys isaisaisarogrogisaisarogrog

LorLorentententzimzimysysAniAnisomsomsomysAnisomsomys imiimi

nounounouhuyhuyhuyhuysiimitattattatororor

ChiChirurrurruromyomyssLorLorentententzimzimysysLorLorentententzimzimysys

vatvatesesnounounouhuynounounouhuy

MacMacrurrurruromyomyomysChiChirurrurruromyomyssChi omyomy vat

majmajmajororHyoHyomysmysmysMacMacrurrurruromyMac omy

golgolgoliathhomyomyomysomyomyomys majmajmajmaj

PogPogonoonoonomysmysmysHyoHyomysmysmysHyoHyomysmysmys golgolgoliatgolgolgoliat

lorlorloriaeiaeiaeiathhiath

PogPogonoonoonomysmysmysPogPogonoonoonomysmysmysPogPog mysmysmys

macmacmacrourouMamMamMammelmelomyomyomysPogPogonoonoonomysmysmysPogPog mysmysmys mac

lanlanlanosuosumacmacrouroumac

AbeAbeAbeomeomeomelomlomlomysMamMamMammelmelomyomyomysMammelomyomyomys

sevsevsevseviaMalMallomlomlomysysAbeAbeomeomeomelomlomAbeAbeome

rotrothschschschilhillomlomysys sevsevsevseviasevsevia

LepLeptomtomtomysMalMallomlomlomysMalMallomys

eleeleganganganssrotrothschschschilrot

HydHydromromromysysLepLeptomtomtomysLepLeptomys

chrysoysoysoysogasgaseleganganganselegangangans

ParParahyahyahydrodromysmysHydHydromromromysysHydHydromromysys chrysochryso

aspaspaspysoysogasysoysogas

PsePseudoudoudohydhydhydromromromysysParParahyahyahydrodromysmysParParahyahyahydrodromysmys aspaspaspasp

ellasp

XerXerXeromyomyomysPsePseudoudoudohydPse hyd

myomyomyoideideideshydhydhydromromromysyshydhydhydromromysys

LegLeggadgadgadinainaXerXeromyomyomysXeromyomyomys myomyomyo

forresresrestitimyoideideidesmyo s

ZyzZyzomyomyomysLegLeggadgadgadinaLegLeggadgadgadina

argargarguruuruurusinainaina forresresrestiforresresresti

NotNotomyomyomysZyzZyzomyomyomysZyzZyzomyomyomys

fusfusfuscuscuscusargargarguruuruargargarg

MasMastactactacomyomyomysNotNotomyomyomysNotomyomyomys fusfusfuscusfusfusfuscus

fusfusfuscuscusfuscuscusfuscus

ConConiluiluilurusrusMasMastactactacomyomyMastactacomyomy

penpeniciiciicillallaomyomysomys fusfusfuscuscusfusfusfuscuscus

LepLeporioriorillussConConiluiluilurusrusConilurus

conditditditororpeniciiciicillallapeniciicilla

UroUromysmysmysLepLeporioriorilluLepLep llu

caucaucaudimdimdimacuaculatlatususllusslluss conditditditororcon

ParParameameamelomlomlomysUroUromysmysmysUromysmys caucaucaudimcau

levlevlevipeipeipeipesdimdimdimacuaculatlatdimdimacuacu

MelMelomyomyomysParParParameameamelomParParamelom

rufrufescescensenslomlomysys levlevlevipeipeipeipe

MelMelomyomyomysMelMelomyomyomysMelMelomyomyomys

cercervinvinvinipeipessrufrufescescensensrufrufescens

SolSolSolomyomyomyomysMelMelomyomyomysMelMelomyomyomys

salsalsalsalsalebrebrebrosuosuosuosuosuosuosusss cercervinvinvinipeipescervinipeipes

MalMalMalMalMalacoacoacoacomysmysmysmysmysmysSolomyomyomyomysSolomyomyomys sal

edvedvedvedvedvedvedvardardardardardsisalebrosuosussalebrosuosus

MalMalMalacoacoacomysmysmysmysMalMalMalacoacoacomysmysmysMalMalacomysmysmys

lonlonlongipesedvedvedvardedvard

TokTokTokudaudaudaudaudaiaiaiaMalMalMalacoacoacomysmysMalMalacomysmys

osiosiosiosiosimenmenmenmenmenmenmensissismysmysmysmys lonlonlongipgipgipeslongipgipes

ApoApoApoApodemdemdemususTokTokudaudaudaudaiaTokuda osi

mysmysmystacinuApoApoApoApodemdemdemususApoApoApodemdemdemususApoApodem

agragrariariusmysmysmystacinumysmystacinu

ApoApoApoApoApodemdemdemdemususApoApoApodemdemdemususApoApodem

semsemsemotusagragrariariusagragrariarius

MusMusMusApoApodemApodem

coocookiikiikiikiidemdemdemdemusususdemususus

MusMusMusMusMusMusMusMus

musculculuscookiikii


crocidcidurouroidesmusculculusmusculus


pahpahariaricrocidcidurocrocidciduro


plaplatyttythrixxpahpahariaripahpahariari


haussassaplatyttythriplatyttythri


matthetheyihaussassahaussa


minutoutoideideidesmatthetheyimatthetheyi


musmusculculoidoidoidesminutoutoideideidesutoideides

PraPraPraomyomyomyssMusMus musmus

deldeldeldeldelectectectectoruorummculculoidoidoidesculculoid

HeiHeiHeimysmyscusPraPraPraomyomysPraPraPraomyomys del

fumfumfumosusdeldelectectorumdel

HylHylHylomyomyscuscusHeiHeiHeimysmyscusHeiHeiHeimysmyscus fumfum

parparvusfumfumosusfumfumosus

PraPraPraomyomysHylHylHylomyomyscuHylHylHylomyomyscu

degdegdegraaraaffiscuscusscuscus parparvusparparvus

PraPraPraomyomysPraPraPraomyomysPraPra s

jacjacjacksoksonidegdegdegraaraaffidegdegdegraaffi

PraPraPraomyomysPraPraPraomyomysPraPraomyomys

daldaldaltontonijacjacjacksoksonijacjacjacksoni


mismismisonnonneiPraPraPraomyomysPraPraPraomyomysPraPraomyomys

tultultullbelbergimismismisonnonneimismisonnei


mormormorioiotultultullbelbergilbergi


obsobsobscurcurusmormormorioiomor


lukluklukoleolelaeobsobsobscurcurusobsobsobscurcurus

MyoMyoMyomysmyscuscusPraPraPraomyomysPraomyomys lukluk

ververrauxiiComComComolyolysMyoMyoMyomysmyscusMyoMyoMyomysmyscus

gosgosgoslinlingicuscus ververrauverrau

ZelZelotootootomysComComComolyolysComComComolyolys gosgos

hilhildegdegardeaeeaeeaegosgoslinlingigosgoslinlingi

SteStenocnocnocephalealealemysmysZelZelotootootomysZelZel mys hilhildegdegardhilhildegdegard

griseiseiseicaucaudaardeaeeaeeaeardeaeeae

SteStenocnocnocephalealealemysmysSteStenocnocnocephalealealemysmysSteSte ephalealemysmys

albipeipeipesgriseiseiseicaugriseiseicau

SteStenocnocnocephalealealemysmysStenocnocnocephalealealemysmysSte ephalealemysmys

albocaocaocaudaudataalbipeipeipesalbipeipes

MyoMyoMyomysmyscuscusSteStenocnocnocephalealeSteSte eph

brobrockmanianialemysmysalemysmys albocaoca

MasMasMastomtomysMyoMyoMyomysmyscusMyoMyoMyomysmyscus

kolkollmalmannsperperpergergericuscus brobrockmanianibrockm

MasMasMastomtomysMasMasMastomtomysMasMastomys

coucouchachakolkollmalmannslmanns


hubhuberterticoucouchachacoucoucha


natnatalealensishubhubertertiert

MasMasMastomtomysMasMasMastomtomysMastomys

awaawaawashensissnatnatalealensisalensi


eryeryerythrthroleucuucuucusawaawaawashensissawashensi

CreCreCreCremnomnomnomysmysmysmysMasMasMastomtomysysMasMas ys eryeryery

cutcutcutcutchichichichicuscuscuscuscuseryerythrthroleucuucuerythrole

MilMillarlarlardiadiaCreCremnomnomysmysCreCremnomnomysmys

kathlehlehlehleenaenaeemysmys cutcutcutchichicuscuscuscutcutcut

MilMilMilMillarlarlarlardiadiadiadiaMilMillarlardiadiaMilMillardia

melmeltadtadtadtadtadtadaaaakathlehlehlehleenaenakathleena

OtoOtoOtoOtomysmysmysMilMillardiadiaMil

dendendendendendentitidiadia mel

OtoOtoOtomysmysmysOtoOtomysmysmysOtomysmysmys

barbarbarbarbouboubouridendendendentidendenti

OtoOtomysmysmysOtoOtomysmysOtomysmys

laclaclacustustrisrisbarbarbarbarboubouribarbarbouri

ParParotootootootomysmysmysOtoOtomysmysmysOtomysmysmys laclaclac

brabrabrantsiilacustrisrislacust

ParParotootootootomysmysmysParParotootootomysmysmysParoto

litlittletledalbrabrabrantsiibrabrabrantsii

OtoOtoOtomysmysmysParParotootomysPar mys

irrirrirroraoratustusmysmysmysmysmysmys

OtoOtoOtomysmysmysOtoOtomysmysmysOtomysmys

angangangangonionioniensisOtoOtomysmysmysOtoOtomysmysmysOtomysmysmys

typtyptypususangangangangangang

OtoOtomysmysmysOtoOtomysmysmysOtomysmysmys

occoccoccoccideideidentalistyptyptypusustyptypus

OtoOtoOtomysmysmysOtoOtomysmysmysOtomysmysmys

oreoreorestestessOtoOtoOtoOtoOtomysmysmysmysOtoOtomysmysmysOtomysmysmys

trotrotrotropicpicpicpicalialialissoreoreorestestessoreoreoresteste

OenOenOenOenomyomyomyomysssOtoOtoOtomysmysmysmysOtoOtomysmysmys trotro

hyphyphypoxaoxaoxaoxaoxaoxaoxanthnthnthnthnthusususustropicalialistropic s

GolGolGolundundaOenOenOenomyomysOenOenOenomyomys

ellelliotiihyphypoxaoxahyphypoxaoxa

MicMicMicaelaelamyamyamyamysGolGolGolundundaGolunda ellelliot

namnamnamnamaquaquensensensensisisPelPelomyomyomysMicMicMicaelaelamyMicMic amy

falfalfalfallaxamyamyamyamysamyamyamys namnam

RhaRhaRhabdobdomysmysmysmysPelPelomyomyomysPelomyomys falfalfalfallaxfalfalfalfallax

dildilectectectuslaxlax

RhaRhaRhabdobdomysmysmysmysRhaRhaRhabdobdomysmysmysmysRhaRhabdobdomysmysmysmys

pumpumpumiliodildilectectectusectus

ArvArvicaicaicanthnthnthisisRhaRhaRhabdobdomysmysmysmysRhaRhabdomysmysmysmys

nainainairobrobaeaepumpumpumpumiliopumpumpumpumilio

ArvArvicaicaicanthnthnthisisArvArvicaicaicanthnthnthisisArvicanthnthis

rufrufrufinuinuinusnainainairobrobaenainairob

ArvArvicaicaicanthnthnthisisArvArvicaicaicanthnthnthisisArv nthnth

nilnilotiotioticuscusrufrufrufinuinuinusrufrufrufinuinus

ArvArvicaicaicanthnthnthisisArvArvicaicaicanthnthnthisisArvicanthnthis

bliblickickickinilnilotiotioticusniloticus

ArvArvicaicaicanthnthnthisisArvArvicaicaicanthnthnthisisArv nthnth

neuneuneumanmanniniLemLemLemnisniscomcomcomcomysArvArvicaicaicanthnthnthisisArvica neu

strstrstrstriatususLemLemLemnsinsicomcomcomcomysLemLemLemnisniscomcomcomcomysLemniscomys

zebzebzebzebrastrstrstrstriatusstriat

LemLemLemnsinsicomcomcomcomysLemLemLemnsinsicomcomcomcomysLemnsicomys

rosrosrosrosaliaaazebzebzebzebrara

LemLemLemnsinsicomcomcomcomysLemLemLemnsinsicomcomcomcomysLemnsicomys

belbelbelbellieririrosrosrosrosaliaaarosrosrosali

LemLemLemnisniscomcomcomcomysLemLemLemnsinsicomcomcomcomysLemnsi ys

macmacmacmacculculusususususbelbelbelbellieririlieri

HybHybHybomyomysLemLemLemnisniscomLemLemLem com

uniunivitvitvittattatususcomcomysys

StoStochochochomysmysmysHybHybHybomyomysHybHybHybomyomys uni

longicgicgicgicaudaudaudatuatuatuatusunivitvitvittattatususuni tat

AetAethomhomhomysysStoStochochochomysmysStoStochochochomysmys

hinhindeideideideimysmys longicgicgicgiclongicgicgicgic

AetAethomhomhomysysAetAethomhomhomysysAetAethomhomysys

chrchrysoysoysoysophiphiluslusluslushinhindeideideideihin


ineineptuptuptuptuschrchrysoysoysoysophichrchrysoysoysoysophi


silsilindindindindensisisineineptuptuptuptusineptuptuptus


kaikaiserserserserisilsilindindindindenssilsilindens


nyinyikaekaekaekaekaikaiserserserserikaiser

DasDasDasymyymysAetAethomhomhomysAet ys

incincomtomtomtomtusnyinyikaekaekaekaenyinyikae

ThaThaThallollomysmysmysDasDasDasymyymysDasDasymyymys inc

loringingingingiincomtomtomtomtusincomtus

ThaThaThallollomysmysmysThaThaThallollomysmysmysThaThaThallollomysmysmys

nigricricricricaudaudaaloringingingingiloringingingingi

ThaThaThallollomysmysmysmysThaThaThallollomysmysmysThaThallollomysmysmys

paepaeduldulduldulcuscuscusnigricricricricaudaudaanigricricaud

GraGraGrammommomysmysmysmysThaThaThallollomysmysmysTha mysmysmys paepaepaepae

doldoldolichuruuruuruuruuruspaeduldulduldulcuscuscuspaedulduldulcus

GraGraGrammommomysmysmysmysGraGraGrammommomysmysmysmysGra mysmysmysmys

comcomcometeetesGraGraGrammommomysmysmysmysGraGraGrammommomysmysmysmysGra mysmysmysmys

macmacmacmilmillanlanlanlanicomcomcometeeteseteetes

GraGraGrammommomysmysmysmysGraGraGrammommomysmysmysmysGra

sursursurdasdastertertertertermacmacmacmilmillanlanlanlanmacmil

imus

ingi

anuanussropropususanuanuss

iveivententerropropususroprop

ianianususiveiventente

atuatusianianusus

inatortortor

iidaleieieiiiii

ensisis

ntalislis

unctatususus

nterusnte

iaerourourusrusosuosuosusrourourusrus

iahildididiiaia

gasgasgasterteraspererergasgasgasgasgasgas

ellellellermermanianiasperereraspererer

ti

cusllatustuscuscus

orllatusllatus

aculatlatususor

ardardsisisisigipesesesardsisisi

sissississisgipgipesesgipgipes

tacinuinuinuinusususus

tacinuinuinutac

otuotussssusus

cauda

!"#$%!&

larlarlarlarisisisis

nus

larlarlarlarisisisisis

nus

BV 70, PP 0.95

PP 0.95

BV 70

! "

Fig. 1

174


Calomyscus baluchiOchrotomys nuttalliMesocricetus auratusOndatra zibethicusOtotylomys phyllotisTylomys nudicaudusLophiomys imhausiDesmodillus auricularisGerbilliscus robustusGerbillurus paebaGerbilliscus gambianusGerbilliscus giffardiGerbillus nanusGerbillus nancillusGerbillus gerbillusGerbillus nigeriaeGerbillus tarabuliUranomys ruddiLophuromys flavopunctatusLophuromys sikapusiDeomys ferrugineusAcomys subspinosusAcomys spinosissimusAcomys russatusAcomys wilsoniAcomys ignitusAcomys cineraceusAcomys chudeauiBatomys grantiPhloeomys cumingiPhloeomys sp.Micromys minutusMaxomys bartelsiiMaxomys suriferMelasmothrix nasoDacnomys millardiLeopoldamys sabanusChiromyscus chiropusNiviventer cremoriventerNiviventer confucianusNiviventer culturatusBerylmys bowersiSundamys muelleriBunomys andrewsiParuromys dominatorDiplothrix legataRattus exulansRattus leucopusRattus norvegicusChiropodomys gliroidesApomys dataeApomys hylocoetesArchboldomys luzonensisChrotomys gonzalesiRhynchomys isarogensisAnisomys imitatorLorentzimys nouhuysiChiruromys vatesMacruromys majorHyomys goliathPogonomys loriaePogonomys macrourusMammelomys lanosusAbeomelomys seviaMallomys rothschildiLeptomys elegansHydromys chrysogasterParahydromys asperPseudohydromys ellermaniXeromys myoidesLeggadina forrestiZyzomys argurusNotomys fuscusMastacomys fuscusConilurus penicillatusLeporillus conditorUromys caudimaculatusParamelomys levipesMelomys rufescensMelomys cervinipesSolomys salebrosusMalacomys edvardsiMalacomys longipesTokudaia osimensisApodemus mystacinusApodemus agrariusApodemus semotusMus cookiiMus musculusMus crociduroidesMus pahariMus platythrixMus haussaMus mattheyiMus minutoidesMus musculoidesPraomys delectorumHeimyscus fumosusHylomyscus parvusPraomys degraaffiPraomys jacksoniPraomys daltoniPraomys misonneiPraomys tullbergiPraomys morioPraomys obscurusPraomys lukolelaeMyomyscus verrauxiiComolys goslingiZelotomys hildegardeaeStenocephalemys griseicaudaStenocephalemys albipesStenocephalemys albocaudataMyomyscus brockmaniMastomys kollmannspergeriMastomys couchaMastomys hubertiMastomys natalensisMastomys awashensisMastomys erythroleucusCremnomys cutchicusMillardia kathleenaeMillardia meltadaOtomys dentiOtomys barbouriOtomys lacustrisParotomys brantsiiParotomys littledaleiOtomys irroratusOtomys angoniensisOtomys typusOtomys occidentalisOtomys orestesOtomys tropicalisOenomys hypoxanthusGolunda elliotiMicaelamys namaquensisPelomys fallaxRhabdomys dilectusRhabdomys pumilioArvicanthis nairobaeArvicanthis rufinusArvicanthis niloticusArvicanthis blickiArvicanthis neumanniLemniscomys striatusLemnsicomys zebraLemnsicomys rosaliaLemnsicomys bellieriLemniscomys macculusHybomys univittatusStochomys longicaudatusAethomys hindeiAethomys chrysophilusAethomys ineptusAethomys silindensisAethomys kaiseriAethomys nyikaeDasymys incomtusThallomys loringiThallomys nigricaudaThallomys paedulcusGrammomys dolichurusGrammomys cometesGrammomys macmillaniGrammomys surdaster

!"#$%!&

' †Antemus chinjiensis

( †cf. Karnimata sp.

†Parapodemus lugdunensis

†Mus sp.

†Abudhabia pakistanensis

†Gerbilliscus sp.

†Aethomys sp.

†Arvicanthis sp.

)

*

+

,

-

†Karnimata darwini.

'

(

)

.

*

+

,

-

/

CaloCaloCalomyscmyscmyscusus balubalubaluchiOchrOchrOchrotomotomotomotomysys

myscmyscmyscnuttnuttnuttalli

MesoMesoMesocriccriccricetusetusOchrOchrOchrotomotomotomotomysysOchrOchrotomotom

auraauraauratustusOndaOndaOndatratratra zibezibethicthicthicusOtotOtotOtotylomylomylomylomysys phylphyllotilotissTyloTyloTyloTylomysmysmysmysmysOtotOtotOtotylomylomOtotOtotOtot

nudinudinudinudicaudcaudcaudususylomysys phylphyllotilotisphylloti

LophLophLophLophLophiomyiomyiomyiomyiomysssTyloTylomysTyloTylomys

imhaimhaimhaimhausiusiusiusiusiDesmDesmDesmDesmodilodilodilodilodilodilluslusLophLophiomyiomysLophiomyiomys

auriauriauriauriculaculaculaculacularisrisrisrisrisGerbGerbGerbilliilliscusscus roburoburobustusstusstusGerbGerbGerbilluillurusrus paebpaebpaebaGerbGerbGerbilliilliscusscus

paebgambgambianuianuianuianuianuianusspaebpaebpaebaa

GerbGerbGerbilliilliscusscus giffgiffardiardiardigambgambianuianuianugambgambianuianu

GerbGerbGerbilluillus nanussGerbGerbGerbilluillus nancilluillusGerbGerbGerbilluillus gerbilluillusGerbGerbGerbilluillus nigeriaeriae

gerbilluillusgerbillu

GerbGerbGerbGerbGerbGerbGerbilluilluilluilluilluilluss tarataratarabulibulibulibulinigeriaeriaenigeriaeriae

UranUranUranomys ruddruddruddiLophLophLophuromysysUranUranomysUranUranomys

flavflavopunopunctatctatusLophLophLophuromysysLophLophuromysysLophLophurom

sikasikapusipusiflavflavopunopunflavopun

DeomDeomDeomysLophLophLoph

ferrferruginugineuseusuromysysuromys sikasikapusipusisika

AcomAcomysDeomDeomysDeom

subssubspinopinosussusferrferruginugineuseusferrugin

AcomAcomysAcomAcomAcomysAcom

spinspinosisosissimusimussubssubspinopinosussussubspinosus

AcomAcomAcomysysrussrussatusatus

AcomAcomAcomysAcomAcomAcomysAcom

wilswilsonioniAcomAcomAcomysAcomAcomysAcom

igniignitustusAcomAcomAcomys

yscinecineraceraceusigniigniigni

AcomAcomAcomysysysAcomAcomAcomysAcom

chudchudchudchudeauieauieauieauiBatoBatoBatomysmysmys

ysgrangrangrantititi

PhloPhloeomyeomyeomyssBatoBatomysmysmysBatoBato grangran

cumicumicumingigrangrantiti

PhloPhloPhloPhloeomyeomyeomyeomyeomyeomysssPhloPhloeomyeomyeomyeomyssPhloPhlo

spspsp.MicrMicrMicromysomysomys

eomyeomyeomyminuminuminuminutustusspsp

MaxoMaxoMaxomysmysmysMicrMicrMicromysomysomysMicrMicr

bartelsielsiiMaxoMaxoMaxomysmysmysMaxoMaxoMaxomysmysmysMaxoMaxo

suriferferMelaMelaMelasmotsmotsmotsmothrixhrixMaxoMaxoMaxomysmysmysMaxoMaxoMaxo surifersuri

nasonasonasoDacnDacnDacnomysomysomysomysomys millardiardiLeopLeopLeopoldaoldaoldamysmysmysDacnDacnDacnomysomysomysomysomysDacnomysomysomys

sabasabasabanusnusnusChirChirChiromysomysomysomyscuscusLeopLeopLeop mysmysmysLeopLeopLeop

chirchirchiropusopusopusNiviNiviNiviventventventerererChirChirChiromysomysomysomyscusChirChirChir

cremcremcremoriventeenteenteentercus chirchirchiropusopusopusopus

NiviNiviNiviventventventererer confconfconfucianusnusnusnusNiviNiviNiviventventventererer culturaturatus

confconfconfucianusconfucia

BeryBeryBerylmyslmyslmys bowersirsiSundSundSundamysamysamysamysBeryBeryBerylmyslmyslmysBeryBerylmys

muelmuelleriBunoBunoBunomysmysmysSundSundSundamysamysamysSund

andrandrewsiewsiamys

ParuParuParuromyromyromyromyssBunoBunoBunomysmysmysBuno

domidomidominatorrrDiplDiplDiplothrothrothrixixixParuParuParuromyromyromyromyssParuParuParuromyromyromy

legalegalegataRattRattRattusDiplDiplDiplothrDipl

exulexulexulexulansothrothrixixix legalega

RattRattRattus leucleucleucleucopusopusRattRattRattRattusus norvnorvnorvnorvnorvnorvegicegicegicegicegicegicegicusus

opusopusopus

ChirChirChirChiropodopodopodopodopodomysomysnorvnorvegicnorv

glirglirglirglirgliroideoideoideoidessegicususegicus

ApomApomApomysysopodopod

datadatadataeopodopodomysomysopodopodomys

ApomApomApomysysApomApomApomysysApom

hylohylohylocoetcoetesesArchArchArchboldboldomysomysomysApomApomApomysysApomApomysys hylohylohylocoethylo

luzoluzoluzonensnensnensisiscoetcoetesescoetcoet

ChroChroChrotomytomytomysArchArchArchboldboldomysomysArchArch

gonzgonzalesalesiomys

RhynRhynRhynchomchomchomysysChroChroChrotomytomytomysChrotomytomy gonz

isarisarogenogensissisgonzgonzalesalesigonz

AnisAnisAnisomysomysRhynRhynRhynchomchomRhynRhyn

imitimitatoratoratorchomysyschomys isarisarogenogenisarisar

LoreLoreLorentzintzimysmysmysomysomysomys

nouhnouhnouhuysiChirChirChiruromuromuromysysLoreLoreLorentzintzimysmysmysLoreLoreLore

vatevatessMacrMacrMacruromuromuromysysChirChirChiruromuromuromysysChirChir

majomajomajorHyomHyomHyomysys goligoligoliath

ysysysys majomajo

PogoPogoPogonomynomynomyssHyomHyomHyomysysHyom goligoligoli

loriloriaeaeaegoliathgoliath

PogoPogoPogonomynomynomyssPogoPogoPogonomynomynomyssPogo

macrmacrmacrouruourusMammMammMammelomelomelomysPogoPogoPogonomynomynomyPogoPogonomy

lanolanolanosusAbeoAbeoAbeomelomelomelomysmysMammMammMammelomelomelomysMammMammelom

seviseviseviaMallMallMallomysomysAbeoAbeoAbeomelomelomysAbeoAbeoAbeo

rothrothschischischildimysmysmys

LeptLeptLeptomysomysMallMallMallomysomysMall

elegelegansansansHydrHydrHydromysomysomysLeptLeptLeptomysomysLeptLeptomys

chrychrysogasogasogasterstersterelegelegansansanseleg

ParaParaParahydrhydrhydromysomysHydrHydrHydromysomysomysHydrHydromysomys chrychry

aspeaspeasperPseuPseuPseudohydohydohydromdromysysParaParaParahydrhydrhydromysomysParahydrhydromys aspeaspe

ellermanrmanrmanrmaniasperasper

XeroXeroXeromysmysPseuPseuPseudohydohyPseuPseu

myoimyoimyoidesdesdesdohydohydromdromysysdromysys

LeggLeggLeggadinadinaaXeroXeroXeromysmysXeromys myoimyoi

forrforrestiestiestiestimyoimyoidesdesmyoi

ZyzoZyzoZyzomysmysLeggLeggLeggadinadinLegg

arguarguargurusrusrusNotoNotoNotomysmysZyzoZyzoZyzomysmysZyzo

fuscfuscfuscusarguarguargurusargu

MastMastMastacomacomacomysysmysmys fuscfuscfusc

fuscfuscususfuscfusc

ConiConiConilurulurusMastMastMastacomacomMastMastMastacom

penipenicillcillcillcillatusacomysys fuscfuscususfuscfuscusus

LepoLepoLeporillrillusus condcondcondconditoritoritorpenipenicillcillcillatuscill

UromUromUromysysLepoLepoLeporillrillusLepo

caudcaudcaudimacimacimacimaculatulatusususus

ParaParaParamelomelomelomysmysUromUromUromysysUrom

levilevilevipesMeloMeloMelomysmysParaParaParamelomeloParaParaPara

ruferufescenscenscensmelomysmysmys levilevilevipes

MeloMeloMelomysmysMeloMeloMelomysmysMeloMeloMelo

cervcervinipinipinipesruferufescenscenscensscenscen

SoloSoloSoloSoloSolomysmysmysmysmysMeloMeloMelomysmysMeloMeloMelo

salesalesalesalebrosbrosbrosbrosususmys cervcervinipinipinipescervcerv

MalaMalaMalacomycomycomycomycomysSoloSoloSolomysmysmysSoloSoloSolo

edvaedvaedvardsirdsirdsiMalaMalaMalacomycomycomycomycomycomyssMalaMalaMalacomycomycomycomycomysMalaMalaMala

longlonglongipesipesipesipesipesipesTokuTokuTokuTokuTokuTokudaiadaiadaiadaiadaiaMalaMalacomyMalaMalacomy

osimosimosimensiensiensiensiensiensiscomys longipeslongipes

ApodApodApodemusemusemusemusemus mystmystacinacinususApodApodApodemusemusemusemusemusApodApodApodemusemusemusemusApodApodemus

agraagrariusriusmystmystacinacinacinacin

ApodApodApodApodApodApodemusemusemusemusemusemusemusApodApodApodemusemusemusemusApodApodemus

semosemosemosemosemosemotustustustusMusApodApod

cookcookcookiiiiApodemusemusemusemusemus

Mus muscmusculusulusMus croccrociduriduroideoidesMus pahapahapahaririMus platplatplatythrythrix

pahapahapahariripahari

Mus haushaussasaplatplatythrythrplat

Mus mattmattheyiheyiMus minuminuminutoidtoides

mattmattheyiheyimattheyi

MusMus muscmuscmuscmusculoiuloiuloiuloidesdesdesPraoPraomysmys deledelectorctorumumHeimHeimyscuyscuss

mysmysfumosussussus

HyloHylomyscmyscmyscususHeimHeimyscuyscussHeim fumo

parvususPraoPraomysmysHylomyscmyscmyscHylo

degrdegraaffaaffimyscususmyscus parvusparvus

PraoPraomysmysmysmys

jackjacksonisonidegrdegraaffaaffidegrdegraaffi

PraoPraomysmysPraomysmysPrao

daltdaltonijackjacksonijacksoni

PraomysmysPraomysmysPrao

misomisonneinneiPraomysmys

mysmystulltullbergbergi


morimoriotulltullbergberg


obscobscurusurusPraoPraomysmys

mysmyslukolukolelalelae

MyomMyomyscuyscuyscusPraomysmysPrao

verrauxiauxiauxiiComoComolyslysMyomMyomyscuyscuMyomyscu

goslgoslingiingiyscuyscus

ZeloZelotomytomytomyslyslys gosl

hildegaregaregardeaedeaegoslingiingigoslingi

StenocepocepocephalehalemysZelotomytomytomysZelotomys hildegarhild

grisgriseicaeicaudaegaregardeaedeaedeaedeae

StenocepocepocephalehalemysStenocepocepocephalehalemysStenocephalehalemys

albialbipespesgrisgriseicaeicagriseica

StenStenocepocepocephalehalemysStenocepocepocephalehalemysStenocephalehalemys

alboalboalbocaudataalbialbipespesalbi

MyomMyomyscuyscuyscusStenocepocepocephalehaleSten

brockmankmankmankmanihalemyshalemys

MastMastomysomysMyomMyomyscuyscuMyom

kollmannmannmannsperspergeriyscusyscus

MastMastomysomysMastMastomysomysMast

couchahakollmannmannkollmann


hubertirtiMastMastomysomys

omysomysnatalenslenslensis


awashenshenshensisMastMastomysomysomysomysMastMastomysomysMast

eryterythrolhrolhroleucueucusCremCremCremCremnomynomynomynomynomys

omys erytcutccutccutccutccutccutchicuhicuhicuseryteryt

MillMillMillardiardiardiaCremCremCremnomynomynomyCremCremCremnomynomynomy

kathleenleenleenaenomysnomys

MillMillMillMillardiardiardiardiardiaaa meltmeltmeltmeltadaadaadaadaadaOtomOtomOtomys dentdentdentiOtomOtomOtomysOtomOtomOtomysOtomOtom

barbbarbbarbouriouriouriOtomOtomOtomysOtomOtomOtomysOtomOtom

laculaculacustristristrisParoParoParotomytomytomytomyssOtomOtomOtomysOtomOtom

branbrantsiitsiiParoParoParotomytomytomytomyssParoParoParotomytomytomytomyssParoParo

littledaledaledaledaleileiOtomOtomOtomysParoParoParotomyParoParo

irroirroirroraturaturatustomytomytomy

OtomOtomOtomysOtomOtomOtomysOtomOtom

angoangoangoniennienniensissisOtomOtomOtomysOtomOtomOtomysOtomOtom

typutyputypusOtomOtomOtomys

ysoccioccioccidentdentdentalisalistyputyputyputypu

OtomOtomOtomysOtomOtomOtomysOtomOtom

oresoresorestestesOtomOtomOtomOtomOtomysysysOtomOtomOtomysOtomOtom

troptroptroptroptroptroptropicalicalicalicalicalisisOenoOenoOenomysmysmys

ys troptrophypohypoxantxantxantxantxanthushushustroptroptrop

GoluGoluGolundandandaOenoOenoOenomysmysmysOenoOenoOenomys

ellielliotihypoxanthypoxant

MicaMicaMicaelamelamelamelamys namanamanamaquenquenquenquensissisPeloPeloPelomysmysMicaMicaMicaelamelamMicaMicaMica

fallfallaxelamelamyselamelamys nama

RhabRhabRhabdomydomydomydomysPeloPeloPelomysmysPeloPeloPelomys fallfall

dilediledilectusctusfallaxfallax

RhabRhabRhabdomydomydomydomysRhabRhabRhabdomydomydomydomysRhabRhabRhabdomy

pumipumipumilioArviArviArvicantcantcanthishisRhabRhabRhabdomydomydomydomysRhabRhabdomydomys

nairnairnairobaeobaepumipumiliopumipumilio

ArviArviArvicantcantcanthishis rufirufinusnusArviArviArvicantcantcanthishis niloniloticuticus

rufirufirufi

ArviArviArvicantcantcanthishis blicblickikiArviArviArvicantcantcanthishis neumneumneumanniLemnLemnLemniscoiscoiscomys stristristriatusLemnLemnLemnsicosicosicomysLemnLemnLemniscoiscoiscomysLemniscoisco

zebrzebrzebrzebraaLemnLemnLemnsicosicosicomysLemnLemnLemnsicosicosicomysLemnsicosico

rosarosarosaliaLemnLemnLemnsicosicosicomysLemnLemnLemnsicosicosicomysLemnsicosico

bellbellbellbellieriLemnLemnLemniscoiscoiscomysLemnLemnLemnsicosicosicomysLemnsico

maccmaccmacculusulusulusHyboHyboHybomysmysmysLemnLemnLemniscoiscoiscoLemnLemnLemn

univittaittaittatusmysmys

StocStocStochomyhomyhomyhomysHyboHyboHybomysmysmysHyboHybomys

longlongicauicaudatudatudatudatusAethAethAethomysomysomysStocStocStochomyhomyhomysStocStoc

hindeieieis longlongicaulongicau

AethAethAethomysomysomysomysomysomys

chrysophsophsophilusilusilusAethAethAethomysomysomysAethAethAethomysomysomysAethAeth

ineptustustuschrysophsophchry

AethAethAethomysomysomysAethAethAethomysomysomysAethAeth

silindenndenndensissisineptustustusineptustus

AethAethAethomysomysomysomysomysomys

kaiserierieriAethAethAethomysomysomysAethAethAethomysomysomysAethAeth

nyikaeaeDasyDasyDasymysmysmysAethAethAethomysomysomysAeth

incomtusmtusmtusnyik

ThalThalThallomylomylomyssDasyDasyDasymysmysmysDasyDasyDasymys

loringingingiThalThalThallomylomylomyssThalThalThallomylomylomyssThalThalThallomy

nigricauicauicaudaloringingingiloringi

ThalThalThallomylomylomyssThalThalThallomylomylomyssThalThalThallomy

paedpaedpaedulcuulcusnigricauicauicaudanigricauicau

GramGramGrammomymomymomymomysThalThalThallomylomylomyssThalThal paed

dolidolichurchurusususpaedpaedpaedulcuulcuspaedpaedulcuulcus

GramGramGrammomymomymomymomysGramGramGrammomymomymomymomysGramGram

comecomecometestesGramGramGrammomymomymomymomys

momymomymomymomymacmmacmmacmillaillaillanini

GramGramGrammomymomymomymomysGramGramGrammomymomymomymomysGramGram

surdsurdsurdsurdasteasterrr

CaloOchrMesoOchrOchr

OndaOtotTyloTyloOtotOtot

LophLophTyloTylo

DesmDesmLophLoph

GerbGerbGerbGerbGerbGerbGerbGerbGerbGerbGerbUranUranLophUranUran

LophLophLophLoph

DeomDeomLophLoph

AcomAcomDeom

AcomAcomAcom

AcomAcomAcomAcomAcom

AcomAcomAcom

AcomAcomAcomAcomAcomAcom

BatoPhloBatoBato

PhloPhloPhloPhlo

MicrMicrMaxoMaxoMicrMicr

MaxoMaxoMaxoMaxo

MelaMelaMaxoMaxo

DacnDacnLeopLeopDacnDacn

ChirChirLeop

NiviNiviChirChir

NiviNiviNiviNiviBeryBerySundSundBeryBery

BunoBunoSundSund

ParuParuBunoBuno

DiplDiplParuParu

RattRattDiplDipl

RattRattRattRattChirChirApomApomApom

ArchApom

ChroArch

RhynChroChro

AnisRhyn

LoreChirLore

MacrChirChir

HyomPogoHyomHyom

PogoPogoPogo

MammPogoPogo

AbeoMamm

MallAbeoAbeo

LeptMallMall

HydrLeptLept

ParaHydrHydr

PseuParaPara

XeroPseuPseu

LeggXeroXero

ZyzoLeggLegg

NotoZyzoZyzo

MastConiMast

LepoUromLepoLepo

ParaUromUrom

MeloParaPara

MeloMeloMelo

SoloSoloMeloMelo

MalaSoloSolo

MalaMalaMala

TokuTokuMalaMala

ApodApodApod

ApodApodApodApod

MusApodApod

MusMusMusMusMusMusMusMusMusPraoHeimHyloHeimHeim

PraoHyloHylo

PraoPraoPraoPrao

PraoPraoPrao

PraoPraoPraoPrao

PraoPraoPrao

PraoMyomPraoPrao

ComoMyom

ZeloStenZeloZelo

StenStenSten

StenStenSten

MyomStenSten

MastMyomMyom

MastMastMast

MastMastMast

MastMastMastMast

MastMastMast

CremCremMillCremCrem

MillMillOtomOtomOtom

OtomOtom

ParoOtomOtom

ParoParoParo

OtomParo

OtomOtom

OtomOtom

OtomOtomOtom

OtomOtomOtom

OenoGoluOeno

MicaPeloMicaMica

RhabPeloPelo

RhabRhabRhab

ArviRhab

ArviArviArviArviLemnLemnLemnLemn

LemnLemnLemn

LemnLemnLemn

LemnLemnLemn

HyboLemnLemn

StocHyboHybo

AethStoc

AethAethAeth

AethAeth

AethAethAeth

DasyAethAeth

ThalDasyDasy

ThalThalThal

ThalThalThal

GramThal

GramGram

GramGramGram

rmani

!"#$%!&

' †Antemus chinjiensis

( †cf. cfcf Karnimata sp.

†Parapodemus lugdunensis

†Mus sp.

†Abudhabia pakistanensis

†Gerbilliscus sp.

†Aethomys sp.

†Arvicanthis sp.

)

*

+

,

-

†Karnimata darwini.

'

(

)

.

*

+

,

-

/

/

!" "!# #!

Fig. 2

175


Lophiomys_imhausiDesmodillus_auricularisGerbilliscus_robustusGerbillurus_paebaGerbilliscus_gambianusGerbilliscus_giffardiGerbillus_nanusGerbillus_nancillusGerbillus_gerbillusGerbillus_nigeriaeGerbillus_tarabuliUranomys_ruddiLophuromys_flavopunctatusLophuromys_sikapusiDeomys_ferrugineusAcomys_subspinosusAcomys_spinosissimusAcomys_russatusAcomys_wilsoniAcomys_ignitusAcomys_cf.cineraceusAcomys_chudeauiBatomys_grantiPhloeomys_cumingiPhloeomys_sp.Micromys_minutusMaxomys_bartelsiiMaxomys_suriferMelasmothrix_nasoDacnomys_millardiLeopoldamys_sabanusChiromyscus_chiropusNiviventer_cremoriventerNiviventer_confucianusNiviventer_culturatusBerylmys_bowersiSundamys_muelleriBunomys_andrewsiParuromys_dominatorDiplothrix_legataRattus_exulansRattus_leucopusRattus_norvegicusChiropodomys_gliroidesApomys_dataeApomys_hylocoetesArchboldomys_luzonensisChrotomys_gonzalesiRhynchomys_isarogensisAnisomys_imitatorLorentzimys_nouhuysiChiruromys_vatesMacruromys_majorHyomys_goliathPogonomys_loriaePogonomys_macrourusMammelomys_lanosusAbeomelomys_seviaMallomys_rothschildiLeptomys_elegansHydromys_chrysogasterParahydromys_asperPseudohydromys_ellermaniXeromys_myoidesLeggadina_forrestiZyzomys_argurusMastacomys_fuscusNotomys_fuscusConilurus_penicillatusLeporillus_conditorUromys_caudimaculatusParamelomys_levipesMelomys_rufescensMelomys_cervinipesSolomys_salebrosusMalacomys_edwardsiMalacomys_longipesTokudaia_osimensisApodemus_mystacinusApodemus_agrariusApodemus_semotusMus_cookiiMus_musculusMus_crociduroidesMus_pahariMus_platythrixMus_haussaMus_mattheyiMus_minutoidesMus_musculoidesPraomys_delectorumHeimyscus_fumosusHylomyscus_parvusPraomys_degraaffiPraomys_jacksoniPraomys_daltoniPraomys_misonneiPraomys_tullbergiPraomys_morioPraomys_obscurusPraomys_lukolelaeMyomyscus_verreauxiiColomys_goslingiZelotomys_hildegardeaeStenocephalemys_griseicaudaStenocephalemys_albipesStenocephalemys_albocaudataMyomyscus_brockmaniMastomys_kollmanspergeriMastomys_couchaMastomys_hubertiMastomys_natalensisMastomys_awashensisMastomys_erythroleucusCremnomys_cutchicusMillardia_kathleenaeMillardia_meltadaOtomys_dentiOtomys_barbouriOtomys_lacustrisParotomys_brantsiiParotomys_littledaleiOtomys_irroratusOtomys_angoniensisOtomys_typusOtomys_occidentalisOtomys_orestesOtomys_tropicalisOenomys_hypoxanthusGolunda_elliotiMicaelamys_namaquensisPelomys_fallaxRhabdomys_dilectusRhabdomys_pumilioArvicanthis_nairobaeArvicanthis_rufinusArvicanthis_niloticusArvicanthis_blickiArvicanthis_somalicusLemniscomys_striatusLemniscomys_zebraLemniscomys_rosaliaLemniscomys_bellieriLemniscomys_macculusHybomys_univittatusStochomys_longicaudatusAethomys_hindeiAethomys_chrysophilusAethomys_ineptusAethomys_silindensisAethomys_kaiseriAethomys_nyikaeDasymys_incomtusThallomys_loringiThallomys_nigricaudaThallomys_paedulcusGrammomys_dolichurusGrammomys_cometesGrammomys_macmillaniGrammomys_surdaster

20 15 10 5 0

Millions of years ago

EEEEEE

ABEEAEAEEEEEEEEEEECCC

ABCCCDCCCC

BCCCCDDBCD

ABCCDCCCCCDDDDDDDDDDDDDDDDDDDDDDDDDDEEBA

ABCC

ABCCCCEEEEEEEEEEEEEEEEEEEEEEEEEEEECCCEEEEEEEEEEEECEEEEEEEEEEEEEEEEEEEEEEEEEEEEEE !"##$#%& &'!("&)*!

!"##$#%& #"+#,--".,

!"##$#%& +$#*)*&

!"##$#%& ($-,+/'!'&

0/"--$#%& 1"*('-+'&

0/"--$#%& .,2!,+"'("

0/"--$#%& -$!,.2,

3"&%#%& ,.+$#)'&

4*)/$#%& .%,5"*

4*)/$#%& 5",&*!,

4*)/$#%& &,-,.(*.&,&

4*)/$#%& ,.*1)'&

4*)/$#%& +/!%&$1/,-'&

4*)/$#%& /,.(*,

6&)$+/$#%& -$.2,+"'('&

7%8$#%& '.9,))")'&

:*#.,&+$#%& #"++'-'&

:*#.,&+$#%& 8*--,*!,

:*#.,&+$#%& !$&"-,"

:*#.,&+$#%& ;*8!"

:*#.,&+$#%& &)!,")'&

4!9,+".)/,& .*'#"..,

4!9,+".)/,& 8-,+5,

4!9,+".)/,& .,-$),+'&

4!9,+".)/,& !'<,.'&

4!9,+".)/,& .",!$8"*

=/"8($#%& 1'#,-,$

=/"8($#%& (,-*+)'&

>*-$#%& <"--"?

@,+"*-"#%& ."#"A'*.&,&

$-'.(" *--,$),

B*.$#%& /%1$?".)/'&

B)$#%& )!$1,+"-,&

B)$#%& $!*&)*&

B)$#%& $++,(*.)"-,&

B)$#%& )%1'&

B)$#%& ".2$.,*.&,&

B)$#%& ,!!$!")'&

>"!$)$#%& -,))-*("-*,

>"!$)$#%& 8!".)&,,

B)$#%& -"+'&)!,&

B)$#%& 8"!8$'!,

B)$#%& (*.),

@,--"!(," #*)-"("

@,--"!(," 5")/-**."*

C!*#.$#%& +')+/,+'&

@"&)$#%& *!%)/!$-*'+'&

@"&)$#%& "D"&/*.&,&

@"&)$#%& .")"-*.&,&

@"&)$#%& /'8*!),

@"&)$#%& +$'+/"

@"&)$#%& 5$--#".&1*!2*!,

@%$#%&+'& 8!$+5#".,

6)*.$+*1/"-*#%& "-8$+"'(")"

6)*.$+*1/"-*#%& "-8,1*&

6)*.$+*1/"-*#%& 2!,&*,+"'("

E*-$)$#%& /,-(*2"!(*"*

C$-$#%& 2$&-,.2,

@%$#%&+'& 9*!!*"'?,,

>!"$#%& -'5$-*-"*

>!"$#%& $8&+'!'&

>!"$#%& #$!,$

>!"$#%& )'--8*!2,

>!"$#%& #,&$..*,

>!"$#%& ("-)$.,

>!"$#%& F"+5&$.,

>!"$#%& (*2!""<<,

7%-$#%&+'& 1"!9'&

7*,#%&+'& <'#$&'&

>!"$#%& (*-*+)$!'#

@'&G#'&+'-$,(*&

@'&G#,.')$,(*&

@'&G#"))/*%,

@'&G/"'&&"

@'&G1-")%)/!,?

@'&G1"/"!,

@'&G+!$+,('!$,(*&

@'&G#'&+'-'&

@'&G+$$5,,

41$(*#'& &*#$)'&

41$(*#'& "2!"!,'&

41$(*#'& #%&)"+,.'&

0$5'("," $&,#*.&,&

@"-"+$#%& -$.2,1*&

@"-"+$#%& *(D"!(&,

6$-$#%& &"-*8!$&'&

@*-$#%& +*!9,.,1*&

@*-$#%& !'<*&+*.&

>"!"#*-$#%& -*9,1*&

H!$#%& +"'(,#"+'-")'&

:*1$!,--'& +$.(,)$!

C$.,-'!'& 1*.,+,--")'&

I$)$#%& <'&+'&

@"&)"+$#%& <'&+'&

E%;$#%& "!2'!'&

:*22"(,." <$!!*&),

J*!$#%& #%$,(*&

>&*'($/%(!$#%& *--*!#".,

>"!"/%(!$#%& "&1*!

7%(!$#%& +/!%&$2"&)*!

:*1)$#%& *-*2".&

@"--$#%& !$)/&+/,-(,

48*$#*-$#%& &*9,"

@"##*-$#%& -".$&'&

>$2$.$#%& #"+!$'!'&

>$2$.$#%& -$!,"*

7%$#%& 2$-,")/

@"+!'!$#%& #"F$!

C/,!'!$#%& 9")*&

:$!*.);,#%& .$'/'%&,

4.,&$#%& ,#,)")$!

=/%.+/$#%& ,&"!$2*.&,&

C/!$)$#%& 2$.;"-*&,

4!+/8$-($#%& -';$.*.&,&

41$#%& /%-$+$*)*&

41$#%& (")"*

C/,!$1$($#%& 2-,!$,(*&

="))'& .$!9*2,+'&

="))'& -*'+$1'&

="))'& *?'-".&

3,1-$)/!,? -*2")"

>"!'!$#%& ($#,.")$!

K'.$#%& ".(!*D&,

6'.("#%& #'*--*!,

K*!%-#%& 8$D*!&,

I,9,9*.)*! +'-)'!")'&

I,9,9*.)*! +$.<'+,".'&

I,9,9*.)*! +!*#$!,9*.)*!

C/,!$#%&+'& +/,!$1'&

:*$1$-("#%& &"8".'&

3"+.$#%& #,--"!(,

@*-"&#$)/!,? ."&$

@"?$#%& &'!,<*!

@"?$#%& 8"!)*-&,,

@,+!$#%& #,.')'&

>/-$*$#%& &1L

>/-$*$#%& +'#,.2,

K")$#%& 2!".),

4+$#%& +/'(*"',

4+$#%& +<LG+,.*!"+*'&

4+$#%& ,2.,)'&

4+$#%& D,-&$.,

4+$#%& !'&&")'&

4+$#%& &1,.$&,&&,#'&

4+$#%& &'8&1,.$&'&

3*$#%& <*!!'2,.*'&

:$1/'!$#%& &,5"1'&,

:$1/'!$#%& <-"9$1'.+)")'&

H!".$#%& !'((,

*!8,--'& )"!"8'-,

*!8,--'& .,2*!,"*

*!8,--'& 2*!8,--'&

*!8,--'& .".+,--'&

*!8,--'& .".'&

*!8,--,&+'& 2,<"!(,

*!8,--,&+'& 2"#8,".'&

*!8,--'!'& 1"*8"

*!8,--,&+'& !$8'&)'&

3*&#$(,--'& "'!,+'-"!,&

:$1/,$#%& ,#/"'&,

Fig. 3

176


Appendix B: Description of fossils

Tatiana Aghová, Yuri Kimura, Josef Bryja, Gauthier Dobigny, Laurent Granjon, Gael J. Kergoat Fossils know it best: using a new set of controlled fossil calibrations to improve the temporal phylogenetic framework of murid rodents (Rodentia: Myomorpha: Muroidea: Muridae)

Figure S1: Dental terminology of murid upper and lower molars (Lazzari et al., 2010; with permission from V. Lazzari).

1. †Potwarmus primitivus (Wessels)

Fossils used in this study

Locality: YGSP 591, Potwar Plateau, Pakistan

177


Age: 16.0 Ma (median age), determined based on magnetostratigraphy, correlated to

the Geomagnetic Polarity Time Scale of Ogg and Smith (2004).

References of the fossils: Lindsay (1988)

Paleontological event: The earliest stratigraphic occurrence of the species, interpreted

to as transition between cricetine and murine dental plan.

Proposed for the fossil calibration point by: Not used.

Taxonomic Information: The genus †Potwarmus is considered to be a primitive

dendromurine (Lindsay, 1988) or more recently to be a stem murine (Wessels, 2009; Lopez-

Antonanzas, 2009). This genus represents four valid species: †Potwarmus primitivus

(Wessels) (Figure S2), originaly described as Antemus primitivus from Chinji Formation,

Pakistan; †Potwarmus thailandicus (Jaeger et al.) originaly described as Antemus thailandicus

from Li Basin, Thailand †Potwarmus flynni Lopéz Antonanzas form Al Jadidah, Saudi Arabia

and †Potwarmus minimus Lindsay. The absence of the anterostyle (t1) in the upper molar in

†Potwarmus, which is otherwise very similar to †Antemus, has been reasoned for exclusion of

the genus from the Murinae. By combining molecular data and the molar structure regarding

two or three longitudinal rows with cusps, Jansa and Weksler (2004) showed that three

longitudinal rows occur in Murinae, Acomyinae, Dendromurinae and the Cricetomyinae. This

suggests that a murine plan evolved at least four times from the cricetine plan. †Potwarmus

could be interpreted in all of these subfamilies as a stem representative although recently

Wessels (2009) and Lazzari et al., (2011) consider †Potwarmus as a stem Murinae.

Reasons to be chosen in the analysis: Because of indefinitive taxonomic assignment, we

excluded this fossil from analysis.

Figure S2: Upper and lower molar of †Potwarmus primitivus without scale (from Patnaik,

178


2014; with permission from R. Patnaik).

2. †Antemus chinjiensis Jacobs


Locality: YGSP 491, Potwar Plateau, Pakistan

Age: 13.8 Ma (median age) , determined by magnetostratigraphy, correlated to the

Geomagnetic Polarity Time Scale of Ogg and Smith (2004).

References of the fossils: Jacobs et al. (1989)

Paleontological event: The earliest stratigraphic occurrence of fossils with a murine-

like dental plan.

Proposed for the fossil calibration point by: This study.

Taxonomic Information: Although the genus †Antemus does not yet have a typical murine

dental plane, it is widely accepted to be a definite murine based on the presence of an

anterostyle (t1) in the upper molar (M1; overview of discussion in Musser & Carleton 2005).

Antemus represents two species: †Antemus chinjiensis (Figure S3) whose oldest record is 13.8

Ma (Jacobs and Flynn, 2005) from Pakistan (Jacobs et al. 1990) and †Antemus mancharensis

Wessels (Figure S4) from the Manchar Formation in Sehwan Sharif, Pakistan (Wessels,

2009). Although †Antemus mancharensis possibly appeared earlier (Wessels, 2009), the age

of the sediments, where the species was recorded, is estimated based on faunal comparisons.

According to Wessels (2009), the first appearance of †Antemus mancharensis in Sehwan was

correlated to the second oldest level of †Potwarmus primitivus in the Potwar Plateau, in

which ages of sediments are well-constrained by the magnetostratigraphic framework.

Description of the fossil: †Antemus is a murine rodent with isolated enterostyle (t4) on M1

and M2; no anterostyle (t1) on M2; paracone (t6) and metacone (t9) well separated;

enterostyle (t4) not joined to hypocone (t8); valley between t1 and t4 shallow, similar in depth

to the valley between t4 and t8; labial cingulum on M2 weak (Jacobs, 1977).

179


How was used for analysis: †Antemus chinjiensis is the oldest rodent classified as a murine,

therefore we assign this fossil as Murinae crown with lognormal prior distribution (offset

13.24, log 1.0, mean 3.2).

Figure S3: Upper and lower molar of †Antemus chinjiensis without scale (from Patnaik,

2014; with permission from R. Patnaik).

3. †cf. Progonomys sp. Schaub


Locality: Pakistan, YGSP 83, 504, Potwar Plateau

Age: 11.6 Ma, determined based on magnetostratigraphy correlated to the

Geomagnetic Polarity Time Scale of Ogg (2012).

References of the fossils: Kimura et al. (2017)

Paleontological event: The earliest stratigraphic occurrence of the genus, which has a

typical murine dental plan.

Proposed for the fossil calibration point by: This study.

Note: The fossils from YGSP 83, YGSP 504 used to be assigned to †Progonomys

hussaini (Jacobs and Flynn, 2005). Kimura et al. (2017) examined murine materials

from the same region but younger ages (10-11 Ma) and concluded that †Progonomys

fossils from the younger age are †Progonomys hussaini but that those from YGSP 83,

YGSP 504 represent species that are not yet known. They tentatively named these

fossils †cf. Progonomys sp.

Description of the genus: †Progonomys is originally named based on fossils from Europe

180


but is later recognized in southern Asia, and thereby this genus (or its hypothetical precursor)

is considered to be an important biogeographic and biochronologic role as the earliest

dispersals of murines (Jacobs and Flynn, 2005). †Progonomys includes six described species:

†Progonomys cathalai Shaub, †Progonomys debruiji, †Progonomys woelferi Bachmayer &

Wilson, †Progonomys hussaini Cheema et al., 2000 ; †Progonomys sinensis Qiu et al., 2004;

†Progonomys morganae Kimura et al. (2017).

Description of the fossil: †Progonomys differs from †Antemus in having connections of

enterostyle (t4) to protocone (t5). The M2 has and in having the anterostyle (t1) on M2.

How was used for analysis: Because of indefinitive taxonomic assignment, we excluded this

fossil from analysis.

Figure S4: Upper molar of †Progonomys cathalai without scale (from Sen, 2003; with

permission from S. Sen).

4. †cf. Karnimata Jacobs


Locality: Pakistan, YGSP 791, YGSP 797, Nagri Formation, Siwalik Group

Age: 11.2 Ma (median age), determined based on magnetostratigraphy correlated to

the Geomagnetic Polarity Time Scale of Ogg (2012).

References of the fossils: Jacobs and Flynn, 2005; Kimura et al., 2015, 2013a

181


Paleontological event: The earliest occurrence of the informal group, interpreted to

be close to the initial split between Arvicanthis and Mus.

Proposed for the fossil calibration point by: Kimura et al. (2015)

Note: The systematic assignment of †cf. Karnimata is yet informal due to the

difficulty of assessing its dental morphology based on limited isolated teeth.

Description of the genus: Stratigraphically first record of association of two morphotypes

†Karnimata and †Progonomys is found in Siwalik Group (11.1 Ma; Kimura et al., 2015). At

the appearance of the two morphotypes, they share 49% overlap area (Kimura et al., 2013b) .

This two fossils candidate for evidence of the Mus/Rattus split (Jacobs and Flynn, 2005) until

recently. Kimura et al. (2015) corrected placement of this fossils and show, that the dental

pattern and root structure of molar teeth in †Karnimata indicate a relationship closer to

Arvicanthis and †Progonomys is close related to Mus.

Description of the fossil: The first appearance of †?Karnimata, interpreted to be close to the

initiation of lineage separation between the †Progonomys and †Karnimata clades (Kimura et

al., 2015).

How was used for analysis: As suggested in Kimura et al., (2015) we placed this fossil as a

crown for Mus/Arvicanthis split with lognormal prior distribution (offset 10.47, log 1.0, mean

4.0).

5. †Parapodemus lugdunensis Schaub


Locality: France, Dionay

Age: 9.6 Ma, Chron C4Ar.2n (Aguilar et al., 2004), determined by

magnetostratigraphy

References of the fossils: Lungu (1981); Mein et al (1993); Renaud et al. (1999)

Paleontological event: The oldest reliable age of well represented Parapodemus

lugdunensis

182


Proposed for the fossil calibration point by: Rowe et al. 2011, Fabre et al. 2013,

Missoup et al. 2016, this study.

Description of the genus: Martín-Suáres and Mein (1998) proposed transfer all

†Parapodemus species (except †Parapodemus gaudryi (Dames) to Apodemus, based on

morphological continuity in development of diagnostic molars. In contrast, de Bruijn et al.

(1999) and Sen (2003) agreed, that genus †Parapodemus is valid with four species:

†Parapodemus gaudryi (Dames); †Parapodemus lugdunensis Schaub; †Parapodemus

barbarae Van de Weerd; and †Parapodemus meini Martin Suarez and Freudenthal.

Description of the fossil: The first representative of †Parapodemus is found from Buzhor 1

in Moldavia (Lungu, 1981). This specimen was originally identified as genus †Progonomys

cathalai, but later Mein et al. (1993) noticed that the specimen closely follows the taxonomic

definition of †Parapodemus (also see Martín-Suárez and Mein, 1998). The single specimen

possesses a complete stephanodonty (the cusps enterostyle (t4), protocone (t5), paracone (t6),

metacone (t9) and hypocone (t8) united in a continuous loop) and lacks posterostyle (t7)

although there is a continuous high crest between enterostyle and hypocone (Mein et al.,

1993; Figure S5).

How was used for analysis: The single specimen from Buzhor1 has been applied to the split

between the Apodemurini-Malacomyini clade and the Praomyini-Murini clade (e.g., Rowe et

al. 2011, Fabre et al. 2013, Missoup et al. 2016). However, as summarized in Sinitsa and

Delinschi (2016), the locality Buzhor1, which used to be considered in the early Vallesian age

(~11-10 Ma), is now interpreted to lie in the late Vallesian age (~10-9 Ma). The largest

diversity of Muridae recorded in the Late Vallesian is found in the locality Dionay (Mein,

1984). The material of †Parapodemus lugdunensis were studied by Renaud et al. (1999) and

the age of this locality was interpolated as 9.6 Ma (Aguilar et al., 2004; Renaud et al., 2005).

Thus, we decided to use these fossils for our analysis. We assigned this fossil as a stem

calibration of tribe Apodemini (Apodemus/Tokudaia split) with lognormal prior distribution

(offset 8.93, log 1.0, mean 4.5).

183


Figure S5: Upper molar of †Parapodemus from locality Buzhor without scale (Mein et al.,

1993; with permission from P. Mein).

6. †Karnimata darwini Jacobs


Locality: Pakistan, YGSP 182, Dhok Pathan Formation, Siwalik Group

Age: 9.2 Ma (median age), determined based on magnetostratigraphy correlated to the


References of the fossils: Jacobs (1978)

Paleontological event: The first occurrence of the derived condition (small, vertical)

of the metacone in a chronoclinal assemblage of the Karnimata clade, interpreted to be

a most common recent ancestor of Millardini/Otomyini/Arvicanthini clade

Proposed for the fossil calibration point by: Kimura et al. (2015).

Taxonomic Information: The descendant of †cf. Karnimata is considered †Karnimata

darwini (Figure S6) from Siwalik Group, Pakistan (Jacobs, 1978; Kimura et al., 2015). Mein

et al. (1993) and Wöger (2011) synonymized †Karnimata darwini from Siwalik to be

†Progonomys woelferi Bachmayer and Wilson, which is otherwise distributed only in Europe.

Storch and Ni (2002) pointed out that †Karnimata darwini differs from †Progonomys

woelferi in having enterostyle (t4) on M1 in a more anterior, symmetrical position relative to

paracone (t6), and anterostyle (t1) more closely approaching lingual antercone (t2). They

suggest that these different morphologies reflect an early divergence of two separate clades.

This view is followed by Kimura et al. (2017).

184


How was used for analysis: †Karnimata darwini is considered as a most common recent

ancestor of Millardini/Otomyini/Arvicanthini clade (Kimura et al., 2015). We placed it as a

crown with lognormal distribution (offset 8.52, log 1.0, mean 4.6).

Figure S6: Upper and lower molar of †Karnimata darwini without scale (from Patnaik, 2014;

with permission from R. Patnaik).

7. †Abudhabia pakistanensis Flynn and Jacobs


Locality: YGSP387, Dhok Pathan Formation, Siwalik Group, Pakistan (Flynn and

Jacobs, 1999)

Age: 8.7 Ma, determined based on magnetostratigraphy correlated to the Geomagnetic

Polarity Time Scale of Ogg and Smith (2004)

References of the fossils: Flynn and Jacobs (1999)

Paleontological event: The earliest occurrence of the genus with fused transverse

laminae lacking a longitudinal crest, which is a dental plan similar to extant

Gerbilliscus, Desmodillus, and Tatera (see Flynn et al., 2003; Denys and Winkler,

2015)

Proposed for the fossil calibration point by: Aghova et al. (2017)

Taxonomic Information: Among the oldest fossil Gerbillinae is †Abudhabia, which was

described initially from the United Arab Emirates †Abudhabia baynunensis, de Bruijn and

Whybrow, and †Abudhabia pakistanensis, Flynn and Jacobs is known from Pakistan,

†Abudhabia cf. A. kabulense, Patnaik from India and †Abudhabia kabulense Sen from

Afghanistan (described as Protatera). The only African records of †Abudhabia sp. are from

185


the late Miocene of Kenya (Lothagam: Winkler, 2003; Lukeino Fm.: Mein and Pickford,

2006).

How was used for analysis: We select †Abudhabia pakistanensis as a most common recent

ancestor for Gerbillinae. We used lognormal prior distribution (offset 8.01, log 1.0, mean 4.7)

and we set this calibration point as crown for Gerbilliscus/Desmodillus split.

8. †Aff. Stenocephalomys Frick


Locality: Chorora, Ethiopia

Age: 8.5 Ma, 40K/40Ar (Geraads, 2002; Suwa et al., 2015)

References of the fossils: Geraads (2001)

Paleontological event: The first occurrence of dental plan similar to extant genus

Stenocephalemys.

Proposed for the fossil calibration by: Not used.

How was used for analysis: †Aff. Stenocephalomys was not included in the final analysis,

because it was not clear assignment to the phylogeny.

9. †cf. Parapelomys Jacobs


Locality: Ethiopia, Chorora

Age: 8.5 Ma, determined by 40K/40Ar dating (Geraads, 2002; Suwa et al., 2015)


Paleontological event: The first occurrence of fossils comparable to †Parapelomys


Taxonomic Information: Among the earliest fossil of †Parapelomys is from Ethiopia as

identified to be cf. Parapelomys (Geraads, 2001). First occurrence of true †Parapelomys is

186


represented by Siwalik fossils older than 8 Ma, co-occurring with †Progonomys and

†Karnimata (Jacobs and Flynn, 2005). Till now we recognize three species: †Parapelomys

robertsi Jacobs (Figure S8), †Parapelomys orientalis Sen and †Parapelomys charkhensis

Patnaik. †Parapelomys, which is considered to be derived from a species of †Karnimata

(Jacobs and Downs, 1994) † Parapelomys displays the arvicanthine characters of well-aligned

cusps of the upper M1 prelobe, bunodont cusps and large molars (Sen, 1983; Jacobs, 1978).

There is a anterostyle (t1) and labial anterocone (t3) on the upper M2 and the upper M3 is

long (Denys and Winkler, 2015).

How was used for analysis: This fossil was not used in our analysis, because it was difficult

to assigned it to a specific node in the current topology although there is a consistent

agreement that †Parapelomys is an arvicanthine rodent (e.g., Sen, 1983; Jacobs, 1978;

Musser, 1987; Patnaik, 1997).

10. †Preacomys kikiae Geraads


Locality: Chorora, Ethiopia

Age: 8.5 Ma, 40K/40Ar (Geraads, 2002; Suwa et al., 2015)


Paleontological event: The first occurrence of dental plan similar to extant genus

Acomys.


Taxonomic Information: The earliest deomyine rodent occurred in Africa in the Late

Miocene, represented by one extinct genus †Preacomys (Geraads, 2001; Mein et al., 2004).

Three species of †Preacomys have been described: †Preacomys kikiae Geraads (Figure S9),

†Preacomys karsticus Mein and †Preacomys griffini Mein. The latter two were reported from

Namibia (Mein et al., 2004). †Preacomys kikiae, was found at Chorora (8.5 Ma; Suwa et al.,

2015), Lukeino (as †Preacomys cf. kikiae; Mein and Pickford, 2006) and Harasib 3a (Mein et

al., 2004). Based on morphological evaluation, it seems that †Preacomys karsticus is possibly

ancestral to Acomys spinosissimus, †Preacomys griffini is a possible ancestor of Acomys

subspinosus (Denys and Winkler, 2015).

187


Description of the fossil: The anterostyle (t1) is stretched out into a low oblique blade; lingal

anterocone (t2) elongated; labial anterocone (t3) very small, located at the same level as t2;

metacone (t9) almost as large as paracone (t6); posterior cingulum strong (Geraads, 2001;

Mein et al., 2004).

How was used for analysis: †Preacomys kikiae was not used for the final analysis, because

of unclear phylogenetic position of the fossil.

11. †Mus sp. Linneaus


Locality: YGSP 547, Dhok Pathan Formation, Siwalik Group, Pakistan

Age: 8.0 Ma (median age), determined based on magnetostratigraphy correlated to the


References of the fossils: Kimura et al., (2013a, 2015)

Paleontological event: The first occurrence of the genus Mus in geological time.

Proposed for the fossil calibration point by: Kimura et al. (2015).

Taxonomic Information: The fossils associated with genus Mus has been reported from a

number of localities in the Indian subcontinent. The first occurrence of the genus †Mus sp. in

geologic time is 8.0 Ma from locality YGSP 547, Dhok Pathan Formation, Siwalik Group

(Jacobs and Flynn, 2005; Kimura et al., 2013a). Other important discoveries include †Mus

auctor Jacobs (6.5 Ma; Figure S10) from the same region (Jacobs and Flynn, 2005), †Mus

flynni Patnaik (2.5 Ma) from the Indian Siwalik (Patnaik, 1991), †M. jacobsi Kotlia (2.4 Ma)

from the Kashmir intermontane basin (Kotlia, 1992), †M. linnaeusi Patnaik (2.0 Ma) from the

Upper Siwalik, India (Patnaik, 1997; 2001).

How was used for analysis: This calibration point was used as a stem calibration for tribe

Murini with lognormal prior distribution (offset 7.29, log 1.0, mean 4.9).

188


12. †Acomys sp. I. Geoffroy


Locality: Lemudong’o, locality 1, Kenya

Age: 6.1 Ma, 40Ar/39Ar (Deino and Ambrose, 2007)

References of the fossils: Manthi (2007)

Paleontological event: The first occurrence of Acomys in geological time.

Proposed for the fossil calibration point: Not used.

How was used for analysis: The Acomys fossil was exluded from our final analysis, because

upper first molars (M1), which are the most diagnostic tooth element for the taxonomy of the

Muridae, are not yet found from the locality. The presence of M1 was added to our criteria for

selection of fossils.

Figure S7: Left mandibular fragment with m1-2 Acomys KNM-NK 42315 (from Manthi,

2007; with permission from F. K. Manthi).

13. †Aethomys sp. Thomas





189


Paleontological event: The first occurrence of Aethomys similar in geological time

Proposed for the fossil calibration point: This study.

How was used for analysis: The Aethomys fossil was used as stem calibration for genus

Aethomys with lognormal prior distribution (offset 5.34, log 1.0, mean 5.5).

14. †Arvicanthis sp. Lesson





Paleontological event: The first occurrence of Arvicanthis similar in geological time.

Proposed for the fossil calibration point: Rowe et al. (2011), Fabre et al. (2013)

How was used for analysis: The Arvicanthis fossil was used as stem calibration for genus

Arvicanthis with lognormal prior distribution (offset 5.34, log 1.0, mean 5.5).

15. †Gerbilliscus sp. (Thomas)





Paleontological event: The first occurrence of Gerbilliscus similar to extant species.

Proposed for the fossil calibration point: Aghova et al. (2017)

How was used for analysis: The Gerbilliscus fossil was used as stem calibration for genus

Gerbilliscus with lognormal prior distribution (offset 5.34, log 1.0, mean 5.5).

190


16. †Lemniscomys sp. Trouessart





Paleontological event: The first occurrence of Lemniscomys similar to extant species.


How was used for analysis: The Lemniscomys fossil was excluded from final analysis,

because the fossils lack the most diagnostic tooth elements, upper first molars.

17. †Mastomys sp. Thomas





Paleontological event: The first occurrence of Mastomys similar in geological time.


How was used for analysis: Mastomys fossil was excluded after cross-validation.

18. †Gerbillus sp. Desmarest


Locality: A.L. 894, Hadar, Ethiopia

Age: 2.4 Ma, geochronology (BKT-3 tephra), 40K/40Ar, 40Ar/39Ar, sedimentology

(Campisano and Feibel, 2008; Kimbel et al., 1996)

References of the fossils: Reed, 2011; Reed and Geraads, 2012

Paleontological event: The first occurrence of Mastomys in geological time.

191



Taxonomic Information: Although the larger varieties of gerbils such as Gerbilliscus are

present in Mio-Pliocene deposits in East Africa (e.g., Lemudong’o; Laetoli), the smaller,

more arid adapated taxa, such as Gerbillus, do not appear in the East African fossil record

until Member F times in the Omo Shungura Formation (Wesselman, 1984).

Description of the fossil: The material from A.L. 894 is similar in size to the Gerbillus sp.

indet. from Omo Shungura, Member F (Wessleman, 1984), but differs from the latter in

having a longitudinal crest connecting the anterior and middle laminae of M1, similar to that

present in extant Gerbillus. At ca. 2.4 Ma the A.L. 894 specimens likely predate the Omo

material and thus represent the oldest occurrence of Gerbillus in East Africa (Reed, 2011;

Reed and Geraads, 2012). The coincident appearance Gerbillus at A.L. 894 in North-east

Ethiopia and also in Member F at Omo Shungura, Ethiopia, supports the hypothesis of a

synchronized transition in the faunas in East Africa between 3.2 and 2.4 Ma, and provides

some evidence for a transition to a more arid regional paleoenvironment (Reed and Geraads,

2012).

How was used for analysis: Gerbillus sp. was excluded after cross-validation.

References

Aghová, T., Šumbera, R., Piálek, L., Mikula, O., McDonough, M. M., Lavrenchenko, L. A.,

Meheretu, Y., Mbau, J.S., Bryja, J., 2017. Multilocus phylogeny of East African gerbils

(Rodentia, Gerbilliscus) illuminates the history of the Somali-Masai savanna. J.

Biogeogr. doi:10.1111/jbi.13017

Aguilar, J.-P., W. A. Berggren, M.-P. Aubry, D. V. Kent, G. Clauzon, M. Benammi, and

Michaux, J., 2004. Mid Neogene Mediterranean marine-continental correlations: an

alternative interpretation. Palaeogeogr. Palaeoclimatol. Palaeoecol. 204, 165–186.

Campisano, C.J., Feibel, C.S., 2008. Depositional environments and stratigraphic summary of

the Pliocene Hadar Formation at Hadar, Afar Depression, Ethiopia. Geol. Early Humans

Horn Africa 179–201.

Campisano, C. J., Feibel, C. S., 2008. Depositional environments and stratigraphic summary

of the Pliocene Hadar formation at Hadar, Afar depression, Ethiopia. Geol. Soc. Spec.

192


Pap. 446, 179-201.

Cheema, I. U., S.Mahmood Raza, L. J. Flynn, A. R. Rapjar, Tomida, Y., 2000. Miocene

smallmammals from Jalalpur, Pakistan, and their biochronologic implications. Bull. Natl.

Sci. Mus., Tokyo, 26, 57–77.

de Bruijn, H., G. Saraç, L. W. Hoek Ostende, and Roussiakis, S., 1999. The status of the

genus name Parapodemus Schaub, 1938: New data bearing on an old controversy, in:

Reumer, J.W.F, de Vos, J. (Eds.), Elephants have a snorkel! Papers in honor of Paul Y.

Sondaar, DEINSEA 7: Annual of the Natural History Museum of Rotterdam. Rotterdam,

pp. 95–112.

Deino, A.L., Ambrose, S.H., 2007. 40Ar/39Ar dating of the Lemudong’o late Miocene fossil

assemblages, southern Kenya rift. Kirtlandia 56, 65–71.

Denys, C., Winkler, A.J., 2015. Advances in integrative taxonomy and evolution of African

!"#$% "&$'()*+,% -&.% &"/-&0&1#230% )"''*% -#$'% )-'% &0'2!03"% 4&"'*)5% #(,% 6&75% 89:95%

Hautier, L. (Eds.), Evolution of the Rodents: Advances in Phylogeny, Functional

Morphology and Development. Campridge University Press, pp. 186–220.

Fabre, P.H., Pagès, M., Musser, G.G., Fitriana, Y.S., Fjeldså, J., Jennings, A., Jønsson, K. A.,

Kennedy, J., Michaux, J., Semiadi, G., Supriatna, N., Helgen, K.M., 2013. A new genus

of rodent from wallacea (Rodentia: Muridae: Murinae: Rattini), and its implication for

biogeography and Indo-Pacific Rattini systematics. Zool. J. Linn. Soc. 169, 408–447.

Flynn, L.J., Jacobs, L.L., 1999. Late Miocene small mammal faunal dynamics: the crossroads

of the Arabian peninsula. in: Whybrow, P.J., Hill, A., (Eds.), Fossil vertebrates of

Arabia. Yale University Press, New Haven. pp. 412–419.

Flynn, L. J., Winkler, A. J., Jacobs, L. L., Downs, W., 2003. Chapter 23: Tedford's Gerbils

from Afghanistan. Bulletin of the American Museum of Natural History. pp. 603-624.

Geraads, D., 2001. Rongeurs du Miocene superieur de Chorora (Ethiopie): Dendromuridae,

Muridae et conclusions. Palaeovertebrata 30, 89–109.

Geraads, D., 2002. Plio-Pleistocene mammalian biostratigraphy of Atlantic

Morocco. Quaternaire. 13, 43-53.

Jacobs, L. L. 1977. A new genus of murid rodent from the miocene of Pakistan and comments

on the origin of the Muridae. PaleoBios. 25, 1–11

Jacobs, L. L., 1978. Fossil rodents (Rhizomyidae and Muridae) from Neogene Siwalk

deposits, Pakistan. Mus. North. Ariz. Bull. 52, 1–103.

Jacobs, L.L., Downs, W.R., 1994. The evolution of murine rodents in Asia. Natl. Sci.

Museum Monogr. 8, 149–156.

193


Jacobs, L. L., L. J. Flynn, Downs, W.R., 1989. Neogene Rodentia of Southern Asia. in: Black,

C.C., Dawson, M.R. (Eds.), Papers on fossil rodents in honor of Albert Elmer Wood.

Natural History Museum, Los Angeles. pp. 157–177.

Jacobs, L., Flynn, L., Downs, W., Barry, J., 1990. Quo vadis Antemus? The Siwalik muroid

record. in: Lindsay, E.H, Fahlbusch, V., Mein, P. (Eds.), European Neogene Mammal

Chronology. Plenum Press. New York. pp. 573–586.

Jacobs L.L., Flynn J.J., 2005. Of mice ... again: the Siwalik rodent record, murine distribution,

and molecular clocks. in: Leiberman. D.E., Smith, R.J,, Kelley, J., (Eds.), Interpreting

the past: essays on human, primate, and mammal evolution in honor of David Pilbeam.

Brill Academic Publishers Inc, Boston. pp. 63–80.

Jansa, S. A., Weksler, M., 2004. Phylogeny of muroid rodents: Relationships within and

among major lineages as determined by IRBP gene sequences. Mol. Phylogenet. Evol.

31, 256–276.

Kimbel, W.H., Walter, R.C., Johanson, D.C., Reed, K.E., Aronson, J.L., Assefa, Z., Marean,

C.W., Eck, G.G., Bobe, R., Hovers, E., Rak, Y., Vondra, C., Yemane, T., York, D., Y

Chen, Evensen, N.M., Smith, P.E., 1996. Late Pliocene Homo and Oldowan Tools from

the Hadar Formation (Kada Hadar Member), Ethiopia. J. Hum. Evol. 31, 549–561.

Kimura, Y., Hawkins, M.T.R., McDonough, M.M., Jacobs, L.L., Flynn, L.J., 2015. Corrected

placement of Mus-Rattus fossil calibration forces precision in the molecular tree of

rodents. Nat. Publ. Gr. 1–9.

Kimura, Y., Jacobs, L.L., Cerling, T.E., Uno, K.T., Ferguson, K.M., Flynn, L.J., Patnaik, R.,

2013a. Fossil Mice and Rats Show Isotopic Evidence of Niche Partitioning and Change

in Dental Ecomorphology Related to Dietary Shift in Late Miocene of Pakistan. PLoS

One 8, 1–10.

Kimura, Y., Jacobs, L.L., Flynn, L.J., 2013b. Lineage-Specific Responses of Tooth Shape in

Murine Rodents (Murinae, Rodentia) to Late Miocene Dietary Change in the Siwaliks of

Pakistan. PLoS One 8.

Kimura, Y., Flynn, L. J., Jacobs, L. L. 2017. Early Late Miocene Murine Rodents from the

Upper Part of the Nagri Formation, Siwalik Group, Pakistan, with a new fossil

calibration point for the Tribe Apodemurini (Apodemus/Tokudaia). Fossil Imprint, 73.

197–212.

Kotlia, B.S., 1992. Pliocene murids (Rodentia, Mammalia) from Kashmir Basin, northwestern

India. N. JB. Geol. Palaeont. 184, 339-357.

Lazzari, V., Aguilar, J.-P., Michaux, J., 2010. Intraspecific variation and micro-

194


macroevolution connection: illustration with the late Miocene genus Progonomys

(Rodentia, Muridae). Paleobiology 36, 641–657.

Lazzari, V., Tafforeau, P., Michaux, J., 2011. When homologous cusps display non-

homologous wear facets: An occlusal reorganization ensures functional continuity during

dental evolution of Murinae (Rodentia, Mammalia). Arch. Oral Biol. 56, 194–204.

Lindsay, E.H., 1988. Cricetid Rodents from Siwalik Deposits near Chinji Village. Part 1:

Megacricetodontinae, Myocricetodontinae and Dendromurinae. Palaeovertebrata. 18, 95-

154.

López Antoñanzas, R., 2009. First potwarmus from the miocene of Saudi Arabia and the early

phylogeny of murines (Rodentia: Muroidea). Zool. J. Linn. Soc. 156, 664–679.

Lungu, A. N., 1981. Hipparion Fauna of Middle Sarmatian of Moldavia (Insectivora,

Lagomorpha and Rodentia). State Tiraspol Institut Pedagogiki Publ. 1557, 1-137.

Manthi, F.K., 2007. A preliminary review of the rodent fauna from Lemudong’o,

southwestern Kenya, and its implication to the late Miocene paleoenvironments.

Kirtlandia. 56, 92–105.

Martín-Suáres, E., Mein, P., 1998. Revision of the genera Parapodemus, Apodemus

Rhagamys and Rhagapodemus (Rodentia, Mammalia). Geobios 31, 87–97.

Mein, P., 1984. Composition quantitative des faunes de mammifères du Miocène moyen et

supérieur de la région lyonnaise. Paléobiologie continentale, 14, 339-346.

Mein, P., Pickford, M., Senut, B., 2004. Late Miocene micromammals from the Harasib karst

deposits, Namibia. Part 2b-Cricetomyidae, Dendromuridae and Muridae, with an

addendum on the Myocricetodontinae. Communs. Geol. Surv. Namibia. 13, 43–63.

Mein, P., Pickford, M., 2006. Late Miocene micromammals from the Lukeino Formation (6.1

to 5.8 Ma), Kenya. Bull. Mens. Soc. Linn. Lyon. 75, 183–223.

Mein, P., Martín-Suárez, E., Agustí, J., 1993. Progonomys Schaub, 1938 and Huerzilerimys

gen. nov. (Rodentia); their Evolution in Western Europe, Scripta Geol. 103, 41-64.

Missoup, A.D., Nicolas, V., Eiseb, S., Chung, E.K., Denys, C., 2016. Phylogenetic position of

the endemic Mount Oku rat, Lamottemys okuensis (Rodentia: Muridae), based on

molecular and morphological data. Zool. J. Linn. Soc. 177, 209–226.

Musser, G.G., 1987. The first occurence of Hadromys (Rodentia: Muridae) in early Pleistocen

Siwalik strata in northern Pakistan and its bearing on biogeographic affinities between

indian and northeastern african murine faunas. Am. Mus. Novit. 2883, 1–36.

Musser, G.G., Carleton M.D., 2005. Superfamily Muroidea, in: Wilson, D.E., Reeder, D.A.M.

(Eds.), Mammal Species of the World. A Taxonomic and Geographic Reference. The

195


Johns Hopkins University Press. Baltimore. pp. 894–1531.

Ogg, J. G., Smith, A. G., 2004. Geologic Time Scale 2004. in: Gradstein, F. M., Ogg, J. G.,

Smith, A. G.A (Eds.), Cambridge University Press. pp. 63–86.

Ogg, J. G., 2012. Geomagnetic Polarity Time Scale. In: Gradstein, F. M., Ogg, J. G., Schmitz,

M., & Ogg, G. (Eds.), The geologic time scale 2012. Elsevier.

Patnaik, R., 1991. Micropalaeontology, biostratigraphy and palaeoenvironmental analysis of

the Siwalik sequences in the Saketi-Nahan area (Himachal Pradesh). Ph.D thesis. Panjab

University, Chandigarh.

Patnaik, R., 1997. New murids and gerbillids (Rodentia, Mammalia) from Pliocene Siwalik

sediments of India. Palaeovertebrata 26, 129–165.

Patnaik, R., 2001. Late Pliocene micromammals from Tatrot Formation (Upper Siwaliks)

exposed near Village Saketi, Himachal Pradesh, India. Palaeontographica 261, 55-81.

Patnaik, R., 2014. Phylogeny of siwalik murine rodents: implications for Mus-Rattus

divergence time. J. Paleontol. Soc. Ind. 59,15–27.

Qiu, Z., Zheng, S., Zhang, Z., 2004. Murids from the late Miocene Bahe formation, Lantian,

Shaanxi. Vertebrata Pal. Asiatica, 42, 67-76.

Reed, D.N., 2011. New murid (Mammalia, Rodentia) fossils from a late Pliocene (2.4 Ma)

locality, Hadar A.L. 894 , Afar Region, Ethiopia. J. Vertebr. Paleontol. 31, 1326–1337.

Reed, D.N., Geraads, D., 2012. Evidence for a Late Pliocene faunal transition based on a new

rodent assemblage from Oldowan locality Hadar A.L. 894, Afar Region, Ethiopia. J.

Hum. Evol. 62, 328–337.

Renaud, S., Michaux, J., Mein, P., Aguilar, J.P., Auffray, J.C., 1999. Patterns of size and

shape differentiation during the evolutionary radiation of the European Miocene murine

rodents. Lethaia 32, 61–71.

Renaud S, Michaux J, Schmidt D, Aguilar J, Mein P, Auffray J. 2005. Morphological

evolution, ecological diversification and climate change in rodents. Proc. R. Soc. London

Ser. B 272, 609–17.

Rowe, K.C., Aplin, K.P., Baverstock, P.R., Moritz, C., 2011. Recent and rapid speciation with

limited morphological disparity in the genus rattus. Syst. Biol. 60, 188–203.

Sen, S., 1983. Rongeurs et lagomorphes du gisement pliocène de Pul-e Charkhi, bassin de

Kabul, Afghanistan. Bull. Mus. Natl. Hist. Nat. 5, 33-74.

Sen, S., 2003. Muridae and Gerbillidae (Rodentia). in: Fortelius, M., Kappelman, J., Sen, S.,

Bernor, R. L., (Eds.), Geology and Paleontology of the Miocene Sinap Formation,

Turkey. Columbia University Press, New York City. pp. 125–140.

196


Sinitsa, M. V., Delinschi, A., 2016. The earliest member of Neocricetodon (Rodentia:

Cricetidae): a redescription of N. moldavicus from Eastern Europe, and its bearing on the

evolution of the genus. J. Paleontol. 90, 771-784.

Storch, G., Ni, X., 2002. New Late Miocene murids from China ( Mammalia , Rodentia )

Nouveaux muridés (Mammifères, Rongeurs) du Miocène supérieur de Chine. Geobios

35, 515–521.

Suwa, G., Beyene, Y., Nakaya, H., Bernor, R.L., Boisserie, J.-R., Bibi, F., Ambrose, S.H.,

Sano, K., Katoh, S., Asfaw, B., 2015. Newly discovered cercopithecid, equid and other

mammalian fossils from the Chorora Formation, Ethiopia. Anthropol. Sci. 123, 19–39.

Wesselman, H.B., 1984. The Omo micromammals: systematics and paleoecology of early

man sites from Ethiopia. Contrib. Vertebr. Evol. 7, 1–219.

Wessels,W., 2009. Miocene rodent evolution and migration: Muroidea from Pakistan, Turkey

and North Africa. Geologia Ultraiectina, 307, 1–290.

Winkler, A.J., 2003. Rodents and lagomorphs from the Miocene and Pliocene of Lothagam,

northern Kenya. in: Leakey, M.G., Harris, J.M., (Eds.), Lothagam, the Dawn of

Humanity in Eastern Africa. Columbia University Press, New York, pp. 169–198.

Wöger, J.A., 2011. A Revision of the Murinae (Rodentia, Mammalia) of Kohfidisch (Upper-

Miocene, Burgenland, Austria). Diplomarbeit. Universität Wien.

Comparative phylogeography of the Somali-Masai region in ...

Documents

Transcript of Comparative phylogeography of the Somali-Masai region in ...