European Journal of Taxonomy 88: 1-28

http://dx.doi.org/lQ.5852/ejt.2014.88

This work is licensed under a Creative Commons Attribution 3.0 License.

ISSN 2118-9773

WWW. europeanj oumaloftaxonomy. eu

2014 • Tabatabaei Yazdi F. et al.

Research article

um:lsid:zoobank.org:pub:3605D81E-A754-4526-ABCC-6D14B51F5886

Cranial phenotypic variation in Meriones crassus and M. libycus

(Rodentia, Gerbillinae), and a morphological divergence in M. crassus

from the Iranian Plateau and Mesopotamia (Western Zagros Mountains)

Fatemeh TABATABAEI YAZDT, Dominique ADRIAENS^ & Jamshid DARVISff

^Faeulty of Natural resourees and Environment, Ferdowsi University of Mashhad, Azadi Square,

91735 Mashhad, Iran

Email: [email protected] : [email protected] (eorresponding author)

2 Ghent University, Evolutionary Morphology of Vertebrates, K.E. Eedeganekstraat 35,

9000 Gent, Belgium

^Rodentology Researeh Department and Institute of Applied Zoology, Ferdowsi University of Mashhad,

Azadi Square, 91735 Mashhad, Iran

^ um:lsid:zoobank.org:author:0E8C5I99-I642-448A-80BB-C342C2DB65E9

" um:lsid:zoobank.org:author:38C489B9-2Q59-4633-8E3D-C53IFE3EDD8B

^ um:lsid:zoobank.org:author:9F7A7QC9-C460-495A-9258-57CB5E7E9DFA

Abstract. Jirds (genus Meriones) are a diverse group of rodents, with a wide distribution range in Iran.

SundevalFs jird {Meriones crassus Sundevall, 1842) is one sueh speeies that shows a disjunet distribution,

found on the Iranian Plateau and Western Zagros Mountains. Morphologieal differenees observed

between these two populations, however, laek quantitative support. Morphologieal differenees between

geographieal populations of Meriones crassus were analysed and eompared with those of the sympatrie

M. libycus. Similarities in the eranial morphology of these speeies were found, e.g. in a relatively large

and inflated bulla. A two-dimensional geometrie morphometrie analysis was done on the skull of 275

M. crassus and 220 M. libycus from more than 70 different loealities in their distribution range. Results

eonflrm eranial differenees between speeimens of M. crassus from the Western Zagros and those from

Afriea and Arabia, mainly at the level of the relative size of the tympanie bulla, that were signifleantly

eorrelated with the annual rainfall and elevation. Moreover, the study supports the hypothesis that the

Western Zagros speeimens are both a geographieally and phenotypieally distinet group eompared to the

other Iranian M. crassus speeimens, suggesting that the former might be a distinet speeies.

Key words, geometrie morphometries, morphology. Middle East, Muridae, skull shape.

Tabatabaei Yazdi F., Adriaens D. & Darvish J. 2014. Cranial phenotypic variation in Meriones crassus and M. libycus

(Rodentia, Gerbillinae), and a morphological divergence in M. crassus from the Iranian Plateau and Mesopotamia

(Western Zagros Mountains). European Journal of Taxonomy 88: 1-28. http://dx.doi.org/10.5852/ejt.2014.88

Introduction

Jirds, belonging to the genus Meriones Illiger, 1811 (Rodentia, Gerbillinae), are a diverse group of murid

rodents that are distributed from North Afriea to China, with 17 reeognized speeies. The SundevalFs

European Journal of Taxonomy 88 : 1-28 ( 2014 )

jird (Meriones crassus Sundevall, 1842) and Libyan jird (M libycus Lichtenstein, 1823) are known

to show both similarities in their morphologieal features and their distribution, with the Middle East

and the Iranian Plateau being a biogeographieally important part of their distribution range (Misonne

1959, 1975; Firouz 2005). In addition, Iran and the Iranian Plateau are of eonsiderable zoogeographieal

interest within the Palearetie region beeause of their great biodiversity (Anderson 1989).

Aeross their distribution range, SundevalLs jirds live in diverse habitats and geoelimatie eonditions.

Intraspeeifie variation among geographie populations of M. crassus (having formed the basis for the

reeognition of numerous subspeeies) eould aetually refleet morphologieal plastieity in relation to

environmental eonditions. As sueh, this study deals with geoelimatie eorrelations of skull variation

within this speeies, in order to test whether this relationship ean be quantifled. Also, sinee the level of

interspeeifle morphologieal differenees in jirds is poorly understood, M. crassus is eompared with its

elosest eongener, M. libycus.

Identifleation keys to M. crassus and M. libycus are mostly based on hind-elaw eolour, the auditory bulla

and the suprameatal triangle eondition (Senegas 2001; Pavlinov 2008), with their respeetive subspeeies

generally not sharply deflned and with similar appearanees within the two speeies. Several keys and

reviews provide some diagnostie features for these speeies, but most of them, e.g. for Libya (Ranek

1968), Egypt (Osborn & Helmy 1980), Arabia (Harrison 1972) and Iran (Eay 1967), are regional in

seope.

Meriones crassus is eranially distinguishable from M. libycus by a less robust eranium and a more

inflated mastoid ehamber and tympanie bulla. Nevertheless, some of the M. libycus subspeeies, sueh as

M. 1. arimalius Cheesman and Hinton, 1924, have been distinguished by their larger tympanie bullae, so

being similar to that ofM crassus (Ellerman 1948). On the other hand, subspeeies ofM. crassus, sueh as

the poorly known M. c. longifrons Eataste, 1884 mdM. c. charon Thomas, 1919, were deflned by their

less swollen bulla (Thomas 1919; Chaworth-Musters & Ellerman 1947; Ellerman 1948; Wittmer and

Btittiker 1982). The natural boundaries between these and other subspeeies ofM crassus are very poorly

known, partly due to diagnostie features often being based on a few speeimens. For M. c. longifrons,

diagnostie features were even based on speeimens kept in eaptivity and on speeimens eolleeted in the

vieinity of just a single eity (Harrison 1972).

Moreover, the impaet of geoelimatie eonditions on inter- and intraspeeifle morphologieal differenees

has been suggested in the past, e.g. in the studies by Petter (1961), Pavlinov & Rogovin (2000), Senegas

(2001), Chevret & Dobigny (2005), Darvish (2009) and Tabatabaei Yazdi & Adriaens (2011). The

relationship between environmental variation, espeeially aridity, and morphologieal variation in jirds

- mainly on the hypertrophy of the bulla - has long been reeognized (Misonne 1959; Harrison 1972;

Musser & Carleton 2005; Darvish 2009; Tabatabaei Yazdi & Adriaens 2011; Tabatabaei Yazdi et al.

2012). Thus, the existenee of environmental eorrelations with the eranial morphology in the studied jirds

ean be expeeted. A limited number of studies that ineluded traditionally measured body and skull size

in some jird speeies have already mentioned the existenee of a eorrelation between the shape variation

(i.e. hypertrophy of bulla) and aridity in jirds (Pavlinov & Rogovin 2000; Chevret & Dobigny 2005;

Momenzadeh et al. 2008; Darvish 2009). However, the degree to whieh elimate and morphologieal

variations in jird speeies ean be eorrelated has never been quantifled. Henee, this study deals with the

eorrelation between eranial shape variation, proven to be essential for systematie and taxonomie studies

in jirds (e.g. by Chaworth-Musters & Ellerman 1947; Pavlinov 2008), and some elimatie variables that

are important limiting faetors for the jirds’ habitats (aridity and altitude).

Meriones crassus oeeurs aeross north Afriea from Moroeeo to Egypt south to Sudan; also in Asia through

the Arabian Peninsula north to Turkey and east to Afghanistan and western Pakistan (Musser & Carleton

TABATABAEIYAZDIF., ADRIAENS D. & DARVISH J., Morphological variation in jirds

2005). In Iran, M. crassus comprises two geographically disjunct populations (Misonne 1959; Eay 1967;

Koffler 1972), one oeeurring in the eentral and eastern part of the Iranian Plateau and the other presently

known from the lowlands of the Khuzestan Plain in the southwest of Iran. The Western Iran M. erassus

speeimens have been eonsidered as a distinet subspeeies (M. c. charon) in the literature (Chaworth-

Musters & Ellerman 1947; Petter 1961), and henee some phenotypie differenees ean be expeeted

eonsidering the different environmental eonditions assoeiated with the habitat. Thus, the existenee of an

assoeiation between geoelimatologieal eonditions and morphologieal variation in M. crassus is tested in

this study, relying on a large sample representing the known distribution range of this speeies and using

detailed deseriptors of skull size and shape. To this end, we performed a landmark-based geometrie

morphometrie analysis on the skulls of Meriones speeimens diagnosed as M. crassus and originating

from the loealities known to span the distribution of the subspeeies reeognised within it, ineluding type

speeimens and speeimens originating from type loealities.

More eoneretely, this study aimed to evaluate whether (1) speeimens of M. crassus from the Western

Zagros eonsidered as M. crassus charon (Ellerman 1948; Harrison 1956; Hatt 1959) ean be distinguished

from those speeimens distributed aeross the Iranian Plateau, as well as from M crassus speeimens that

originated from other parts of the speeies’ distribution range; (2) morphologieal differenees between the

Western Zagros M. crassus and a eluster eontaining the other M. crassus populations are at a similar level

as that observed at the inter-speeifie level; (3) eranial variation in Sundevall and Eibyan jirds reveals

different patterns in relation to geoelimatie variation, and (4) the eranial differenees are eorrelated with

the geoelimatie variables.

Material and methods

Abbreviations

2B-PES

CVA

MANOVA

MANCOVA =

NPMANOVA =

OTU

PAST

PCA

SPSS

ET^GMA

two-bloek partial least square

eanonieal variate analysis (CV plots)

seeond molar

multivariate analysis of varianee

multivariate analysis of eovarianee

neighbour j oining

non-parametrie MANOVA

operational taxonomie unit

PAlaeontologiea STatisties

prineipal eomponent analysis

relative warps

Statistieal Paekage for the Soeial Seienees

unweighted pair-group average eluster analysis

BMNH

FAO

FMNH

MNHN

USNM

British Museum of Natural History (= Natural History Museum, Eondon)

United Nations Food And Agrieulture Organization, Rome

Field Museum of Natural History, Chieago

Museum national d’Histoire naturelle, Paris

Smithsonian National Museum of Natural History, Washington D.C.

Specimens analysed

For this study, a total of495 skulls (275 for Meriones crassus and 220 for M. libycus) was used. Juvenile

speeimens, identified on the basis of the eruption and amount of wear on the molars (M2) (Petter 1959;

Tong 1989; Pavlinov 2008), were exeluded from the analyses. The studied speeimens ofM. crassus range

from the Western Sahara to the Baloehistan provinee of Pakistan, loeated at the southeastern edge of the

European Journal of Taxonomy 88 : 1-28 ( 2014 )

Iranian Plateau, whereas the speeimens of M. libycus have been eolleeted from Algeria to Afghanistan

(provinee of Kandahar). The speeimens were obtained from the eolleetions of the Smithsonian Museum

of Natural History (Washington D.C., USA), the Field Museum of Natural History (Chieago, USA),

the British Museum of Natural History (London, UK), the Museum national d’Histoire naturelle (Paris,

Franee), the Zoologieal Museum of Ferdowsi University of Mashhad (Mashhad, Iran) and the Royal

Belgian Institute of Natural Seienees (Brussels, Belgium). A list of speeimens, with eatalogue numbers,

is available in Appendix 1. The speeies names on the museum labels were ineluded as metadata.

However, speeimens were re-identified based on all external and eranial data available, following the

keys of Chaworth-Muster & Ellerman (1947) and Osborn & Helmy (1980). External eharaeters eould

be verified on most speeimens (as eolleetion speeimens ineluded both skull and skin). In order to be able

to re-assign the speeimens to nominal taxa, the available type speeimens (Table 1) in these eolleetions

were examined and ineluded in the analyses. The samples of M crassus for the analyses of the variation

patterns were pooled into seven groups. In order to test shape differenees, speeimens were pooled into

four groups (aeeording to their geoelimatie proximity in habitat and geographieal range), namely:

Iranian Plateau, Western Zagros, Arabia and Afriea, with 137, 48, 32 and 54 speeimens, respeetively.

The speeimens from Kuwait, Jordan and northwestern Saudi Arabia (NW Arabia) were ineluded in

the Arabie group. The Jeddah loeality was exeluded from CVA beeause of the low sample size (four

speeimens) and the inelusion of some eaptive-bred speeimens.

The speeimen sampling loealities are listed in Appendix 2, their geographie eoordinates were plotted

with AreGlS ( http://www.esri.eom L AreMap 9.2 (Fig. 1).

Morphological and geoelimatie data acquisition

The use of landmark data to deseribe variation in both skull size and shape in mammals is well

established for taxonomie and phenotypie evolutionary studies (Rohlf & Mareus 1993; Fadda & Corti

2001; Bareiova & Maeholan 2006; Cardini et al. 2007; Maeholan et al. 2008). Beeause studies amply

indieated geometrie morphometries as a powerful tool for studying shape variation (e.g. Bookstein 1991;

Rohlf & Mareus 1993; Adams & Rohlf 2000), this approaeh was followed in this study. As the skull

phenotype is genetieally and funetionally more eomplex than teeth (Caumul and Polly 2005; Cordeiro-

Estrela et al 2008), it probably represents relevant biologieal variation of underlying natural groups.

40'0'0'N

30'0'0'N

20'0'0'N

20"0'0"W 10'0’0'W 0°0'0'

10WE 20'0'0'E 30°0'0’E 40WE 50WE 60°0'0'E 70WE

Fig. 1. Map showing the sampling loealities of Meriones crassus Sundevall, 1842 (eireles) andM. libycus

Eiehtenstein, 1823 (squares) and groups of sampling loealities indieated by ellipses (see more detail

about the grouping in Material and Methods). The dark elosed symbols are the sampling loealities of the

type speeimens (synonyms of Meriones crassus and M. libycus, see Table 1). The ellipses (from left to

right) show the following groups: Afriean, Jeddah, Arabian, Western Zagros and Iranian Plateau.

TABATABAEIYAZDIF., ADRIAENS D. & DARVISH J., Morphological variation in jirds

Table 1. Eist of type specimens included in this study. Synonyms are according to Musser & Carleton

(2005).

Taxon

Synonym

Type locality

Collection

M. caudatus amplus

M. libycus

Libya, Fezzan, Gatrun

USNM

M. caudatus luridus

M. libycus

Libya, Cyrenaica, Bahr El Tubat, 21 km E of Bahr El Tubat

USNM

M. crassus asyutensis

M. crassus

Egypt, Asyut, 13 mil SE of Asyut

FMNH

M. crassus charon

M. crassus

Iran, Khuzestan, Ahwaz

BMNH

M. crassus ismahelis

M. crassus

Arabia, Hufuf

BMNH

M. crassus longifrons

M. crassus

Arabia, Jedda

BMNH

M. crassus palUdus

M. crassus

Sudan, Atbara

BMNH

M. crassus pelerinus

M. crassus

Arabia, Tebuk

BMNH

M. crassus perpallidus

M. crassus

Egypt, Wadi Gedeed, Cairo, 4 km W of Cairo-Alexandria rd

FMNH

M. crassus tripolidus

M. crassus

Eibya, Gebel Eimhersuk

BMNH

M. crassus tuareg

M. crassus

Niger, Teguidda, near Tisem, west of Asben

BMNH

M. erythrourusfarsi

M. libycus

Iran, Ears, Bariz, 3 km N of Bariz, 50 km N of Ear

USNM

M. libycus aquilo

M. libycus

China, Zungaria, Gutschen

BMNH

M. libycus caudatus

M. libycus

Eibya, Tripolitania, Bir Ferdjan

BMNH

M. libycus gaetulus

M. libycus

Algeria

BMNH

M. libycus schouesboeii

M. libycus

Algeria, Tieret, Tibremt

BMNH

M. syrius

M. libycus

Syria, Al Qariatayn, Syrian Desert

BMNH

M. syrius edithae

M. libycus

Arabia, Hufuf, Khudud Spring

BMNH

M. syrius evelynae

M. libycus

Arabia, Hufuf, Khorassan Spring

BMNH

Eandmark data in two-dimensional planes were collected from photographs taken with a Nikon

D70 digital reflex camera using a Sigma 105 mm macro lens at flve megapixels in a standardized

manner. The camera was placed on a tripod parallel to the ground plane. The intact and cleaned skulls

were mounted in a box with glass pearls. Eeft-right symmetry on the ventral and dorsal sides, and

a perfect overlap at the level of bullae, teeth rows and the optic canals on the lateral side, were the

most important criteria to position the skulls in a standardized way. The ventral, dorsal and lateral

sides of the skulls were photographed. In each photograph, a scale was included by adding a piece

of millimeter paper. This allowed the acquisition of a scaling factor for calculating centroid sizes

(an overall size measure based on landmark coordinates). Eandmark conflgurations representing

the three sides of the skulls were digitized using the software TpsDig 2.12 (Rohlf 2004a). To avoid

digitisation of both sides of asymmetrical skulls (hence generating a substantial increase in the number

of shape variables), we tested for left-right differences on the dorsal and ventral faces of the skulls in a

subsample of 20 specimens of each species of the dataset. A Monte Carlo randomization (based on

partial warp scores, see below) showed that skulls were symmetrical (p > 0.726). As such, only one side

(left) of the skulls was further digitised and included in the analyses. This first test revealed, that digitised

coordinates of one side could be reliably mirrored to the other side along the midline in case of broken

skulls (Elewa 2004), which could thus still be included in this study. On the ventral, dorsal and lateral

sides of the cranium, respectively 20,19 and 22 landmarks were chosen on the condition that they include

all relevant structures which may be expected to reflect the morphological variation (Fig. 2). In order to

efficiently capture the shape and size of the tympanic bulla, three additional landmarks were digitized

(Fig. 2, open circles). The landmark positions are defined in Appendix 3, based on the terminology used

by Popesko et al. (2002) and Tong (1989). Doing the analyses without the semi-landmarks (that could be

correlated to other landmarks and hence could influence the results) revealed no significant differences.

Hence, all the landmarks were included to represent a reference of the mediosagittal plane.

European Journal of Taxonomy 88 : 1-28 ( 2014 )

For the analyses, data from different skull views were eonsidered to eneompass all variation patterns

exhibited by stmetural eomponents that are not visible in all the views (Monteiro et al. 2003).

Geoelimatie data were taken from published reeords of the Iranian Meteorologieal Organization and the

FAO elimate database (FAO 2007): latitude, longitude, elevation (m), temperature (°C) (mean annual,

monthly maximum and monthly minimum), and annual rainfall (mm). For sampling loealities for whieh

no speeifie elimatie data was available (e.g. small villages), the geoelimatie data of the elosest eity were

used. The average annual rainfall data for eaeh sampling loeality was ealeulated based on data over a

period of ten years (1995-2005).

Data analysis

Shape analysis

In order to standardise the data, a Generalised Proerustes Analysis was performed (non-shape differenees

were removed), using TpsRelw 1.46 (Rohlf 1999,2008) and PAST (PAlaeontologiea STatisties) ver. 1.74

(Hammer et al. 2001). As sueh, landmark eonfigurations were aligned by Proerustes superimposition and

were sealed to unit eentroid size (to remove morphologieal variation due to size, rotation and translation)

(Gower 1975; Bookstein et al. 1985; Rohlf & Sliee 1990; Rohlf 1995, 1999; Rohlf & Mareus 1993).

Relying on the thin plate spline approaeh (see Rohlf 1995; Zelditeh et al. 2004), variation in shape

eoordinates was deeomposed in partial warps (shape variables explaining shape variation at different

seales within the landmark eonfigurations), thus generating a matrix of partial warp seores for eaeh of

the speeimens. For this analysis, both uniform and non-uniform partial warps were ineluded to span the

full speetrum of shape variation, thus yielding the so-ealled weight matrix (Rohlf & Bookstein 2003).

This matrix was then used for further multivariate statistieal analyses. A elassifier analysis was done

in PAST on the eombination of the partial warp seores of the dorsal, ventral and lateral sides of the

speeimens. We tested how well speeimens were assigned to the a priori taxa. Sinee the shape variation

at different levels of loeal seale were given equal weight (alpha =1), the PC A (see below) eorresponds

to a prineipal eomponent analysis on Proerustes eoordinates. But, just as in a PC A, the prineipal axes

explain different pereentages of the total amount of variation, where these prineipal eomponents are then

referred to as ‘relative warps’ (RW) (Rohlf 1993).

Prineipal eomponent analysis (PCA), using PAST, and Canonieal variate analysis (CVA), using

STATISTICA (StatSoft, ver. 7.0), were performed on partial warp seores of eaeh data set separately for

the dorsal, ventral and lateral view to investigate the intraspeeifie variation and morphologieal differenees

among the OTUs. The seatter plots, illustrating the results of PCA and CVA analyses, were generated

in STATISTICA to visualize how speeimen groups are distributed in morphospaee. Visualization of

shape ehanges by deformation grids was generated using TpsSplin 1.20 (Rohlf2004b). For the PC plots

(Figs 3, 6), deformation grids represent shape differenee between eonfigurations eorresponding to lowest

and highest RW-values. For the CV plots (Figs 4, 7), these grids visualize the shape differenees when

following the trajeetory within the morphospaee along the arrows and between the groups’ eonsensuses

(ealeulated using TpsSmall 1.20) (Rohlf2003). To further test reliability of the outeome of the eanonieal

variate analyses, a elassifier analysis was done on the pooled partial warp seores of the dorsal, ventral

and lateral views. To test for shape differenees among the groups, the basie assumptions for doing a

parametrie test were not met. Henee, partial warp seores of the defined groups were subjeeted to a non-

parametrie MANOVA (NPMANO VA in PAST with 10 000 permutations, bonferroni-eorreeted p-values).

The pairwise testing for shape differenees between all eompared groups was done with a Monte Carlo

randomization (10000 permutations, bonferroni-eorreeted p-values) on Euelidean distanees between

group means using the Poptools 3.2.3 plug-in (Hood 2010) for Mierosoft Offiee Exeel 2007. This test

was done both for eaeh side of the skull (ventral, dorsal and lateral) separately, as well as for the pooled

dataset of the M. crassus groups. However, as this yielded a very high number of shape variables (110 in

TABATABAEIYAZDIF., ADRIAENS D. & DARVISH J., Morphological variation in jirds

total), the number of informative variables was redueed by using PCA-seores of this dataset of those PCs

that explained 95 % of the shape variation. In that way, the number of variables eould be redueed to 50.

In order to eompare to what degree geographieal affinities refleeted shape similarities between the

M. crassus groups and to evaluate the overall morphologieal similarities between them, a neighbour

joining (NJ elustering) and unweighted pair-group average eluster analysis (UPGMA), were performed

on the matrix of shape distanees (Euelidean Distanees) between the group means (using PAST). For

the latter analyses, the partial warp seores for the three views (ventral, dorsal and lateral) were pooled

to ealeulate the group average. Sinee the Euelidean distanee is a robust and widely applieable measure

for testing morphologieal similarities (Raup & Criek 1979) and the proerustes distanees were highly

eorrelated to the tangent distanees, the eorresponding matrix of Euelidean distanees (ealeulated using

Poptools 3.2.3) was used for the pairwise testing. The robustness of the resulting elustering trees (braneh

supports) was estimated by performing a bootstrapping of 10 000 randomizations using PAST.

To test to what degree the observed shape dififerenees between the M. crassus groups were explained

by size-independent differenees, a MANCOVA was performed on the RW-seores (explaining more than

Fig. 2. Eandmarks positioned on the eranium of Meriones crassus Sundevall, 1842 shown in (A) ventral

(B) dorsal and (C) lateral views. The straight lines on the ventral and lateral views were used for deflning

semi-landmarks based on two other landmarks. Open eireles on the ventral side: the most rostral and

on the most eaudal point of the tympanie bulla, and on the lateral side: the most rostral margin of the

tympanie bulla. Short lines are drawn to highlight sutures whieh are unelear here and on whieh the

landmarks have been deflned.

European Journal of Taxonomy 88 : 1-28 ( 2014 )

95 % of shape variations) of ventral shape variables, with eentroid size as the eo-variate and M. crassus

groups as the fixed faetor, using SPSS (Statistieal Paekage for the Soeial Seienees ver. 15.0).

A faetorial MANCOVA was performed on the same shape variables (explaining more than 95 % of

shape variations) using SPSS, with eentroid size as the eo-variate and M. crassus groups and sex (sex

per OTU interaetion in MANCOVA) as the fixed faetors to test for sexual dimorphism as well.

A two-bloek partial least square (2B-PLS) analysis (Rohlf & Corti 2000) was used to analyse the

eovariation between the overall skull shape and geoelimatie variables. The 2B-PLS eonstruets pairs of

veetors, representing linear eombinations of the variables within eaeh data set, in sueh a way that the

veetors aeeount for as mueh of the eovariation between the two original sets of variables as possible.

The new variables deseribe what patterns, if any, of eovariation exist between the two sets of original

variables (see Rohlf & Corti 2000 for further details). The eorrelation eoeffieient for eaeh geoelimatie

variable and the pereentage of eovariation for the first two PLS veetors, explaining more than 90 % of

variation, were ealeulated using MorphoJ 1.02d (Klingenberg 2008).

Size analysis

Centroid sizes of the ventral, dorsal and lateral skull views and the eentroid sizes of the ventral and lateral

views of the bullae were ealeulated using PAST (Bookstein 1991). Sinee the ealeulated sizes based on

the different landmark eonfigurations eould be different, the size analyses were done on different size

variables, namely: eentroid size of the eomplete skull (were ealeulated for the three views separately),

eentroid size of the bulla in ventral view (deseribed by landmarks 12, 13 and both open eireles), eentroid

size of the bulla in lateral view (deseribed by landmarks 14, 18, 19, 22 and the open eirele), and the

relative bulla size with respeet to the skull size (as ealeulated for both the ventral and lateral views).

Inter-group skull and bulla size differenees between all the M. crassus groups, as defined for the shape

analysis at an intra-speeifie level, and M. libycus were tested by a Kruskal-Wallis test (with Bonferroni

eorreetion) using PAST. As the ventral eranium eontains a larger number of evolutionarily independent

eomponents of variation (Caumul & Polly 2005) as well as the main part of the tympanie bulla, only the

size differenees obtained from the ventral view are plotted in Fig. 5.

The eorrelations between the bulla size (whieh explained the main observed eranial differenee at the

intraspeeifie level) and the geoelimatie variables were explored using STATISTICA.

Results

Interspecific shape differences

The highly signifieant Hotelling’s p-values (p < 0.001, Bonferroni eorreeted) from the elassifier analysis

of two speeies shows that only one speeimen (type of Meriones libycus gaetulus Lataste, 1882) was

miselassified and all other speeimens were eorreetly elassified to their a priori defined groups.

The plot of the prineipal eomponent seores (RWl vs RW2, Fig. 3) demonstrates that type speeimens

elustered within the speeimens identified as the eorresponding speeies (based on Musser & Carleton

2005). It also shows that skulls of Meriones libycus are slightly different from those of M. crassus,

exeept for those of the Western Zagros. With respeet to the shape variation explained by RWl (mainly

representing bulla infiation), these speeimens show an intermediate shape in all the skull views. For the

RW2 seores, the Western Zagros speeimens are within the upper range of the M. crassus seores for the

dorsal view, but not for the ventral and the lateral views.

In ventral view, the tympanie bulla protrudes laterally and rostrally with inereasing RWl-value,

(landmarks 12 and 13). Also, the zygomatie areh beeomes slightly less eonvex (landmarks 16 and 19).

Shape variation explained by the seeond prineipal eomponent involves the zygomatie areh beeoming

TABATABAEIYAZDIF., ADRIAENS D. & DARVISH J., Morphological variation in jirds

more convex (landmarks 16 and 19) and the braincase becoming narrower, with RW2-scores going in a

positive direction.

In dorsal view, when moving from negative to positive RWl-values, the tympanic bulla becomes

substantially inflated, thereby markedly protruding rostrally and laterally (landmarks 15 and 16). Also

the interparietal and occipital regions change, becoming distinctly narrower (landmarks 18 and 19).

The lateral part of the zygomatic plate protrudes slightly rostrally (landmark 9). With increasing RW2-

scores, the premaxilla (landmark 7) shifts slightly rostrally, the zygomatic plate shifts slightly laterally

(landmark 9), and the interorbital portion of the skull roof becomes narrower (landmarks 10, 11 and 12).

In lateral view (graph not included), also along the first principal component, the tympanic bulla shows

a substantial inflation. The distinction, however, of the OTU’s is not as clear as the dorsal view data

show. Although the ventral view does not reveal a clear distinction either, the results are incorporated in

this study, as the ventral cranium contains a larger number of evolutionarily independent components of

variation and is known to be the most informative view (Dobigny et al. 2002; Caumul & Polly 2005).

Eooking at the general pattern of shape variation, patterns of skull shape variation (in dorsal and ventral

view) are clearly different at inter- versus intraspeciflc level. Interspecific variation explained by RW1

Fig. 3. Scatter plot of RWl versus RW2 of the (A) ventral and (B) dorsal cranium of Meriones crassus

Sundevall, 1842 and M. libycus Eichtenstein, 1823. Eegends: o = M. crassus (other than Western

Zagros), • =M. crassus of Western Zagros, □ = M. libycus. Below: thin-plate spline deformation grids

visualize shape variation as expressed by the first two RWs axes (grids represent shape difference

between configurations corresponding to lowest and highest RW-values). For the numbering of

landmarks, see Fig. 2.

European Journal of Taxonomy 88 : 1-28 ( 2014 )

Table 2. Results of the Monte Carlo simulation test for pairwise eomparisons based on Euelidean

distanees between the group means based on 10 000 randomizations (*p < 0.0001). WZ: Western Zagros.

‘M crassus" refers to speeimens not from Western Zagros.

Pairwise grouping

Ventral

Dorsal

Lateral

M crassus - M. crassus WZ

0.0269*

0.0335*

0.0384*

M crassus - M. libycus

0.0288*

0.0447*

0.0436*

M. crassus WZ - M libycus

0.0190*

0.0318*

0.0278*

refleets mainly variation in bulla width. At the intraspeeifle level (RW2), variation is more eonstrained

(espeeially for the dorsal view) for the M. crassus from the Western Zagros than for those from other

loeations and for M. libycus (although eaution has to be taken due to different sample sizes). As sueh,

the latter M. crassus and M. libycus show variation in the antero-posterior position of the meatus more

than M. crassus Ifom the Western Zagros.

The NPMANOVA revealed a highly signifleant differenee between the OTU’s (p < 0.001), with

F 72 = 40.7, F 7 Q = 88.0 and gg^ = 58.2 for the ventral, dorsal and lateral shape datasets, respeetively.

The post-hoe testing showed that all groups are signifleantly different from eaeh other, thus ineluding

both M. crassus groups (Table 2). The groups are well separated in CV1-CV2 morphospaee, exeept

for four speeimens elustering within the Western Zagros group (the M. c. longifrons speeimens from

Jeddah, Fig. 4).

In ventral view. Western Zagros M. crassus show elear dififerenees with regard to other M. crassus

speeimens (Fig. 4A1). They have a markedly less inflated tympanie bulla, a more eonvex zygomatie areh

(landmarks 16 and 19), a broader zygomatie plate and a faeial tuber being positioned laterally (landmark

17). Compared to M. libycus (Fig. 4A2), Western Zagros M. crassus show a bulla that is slightly inflated

eaudally (landmarks 12 and 13); their zygomatie areh is more eonvex, the ineisive foramen is longer and

the palatine Assure beeomes slightly longer (landmarks 3 and 4) and relatively eloser to the tooth row.

In dorsal view, the Western Zagros M. crassus, as eompared to other M. crassus speeimens, show a

elearly less inflated tympanie bulla (landmarks 15 and 16). In the Western Zagros M. crassus speeimens,

the zygomatie plate is positioned more anteriorly (landmarks 9 and 10) (Fig. 4B1). In the latter group,

as eompared to M. libycus (Fig. 4B2), the bulla is more inflated and protruded rostrally and toward the

zygomatie proeess of the squamous part, the lateral edges of the zygomatie plate and premaxilla lie more

rostrally (landmarks 7 and 9) and the inter-orbital part of the eranial roof is narrower (Fig. 4).

Based on the lateral shape, the tympanie bulla is elearly less inflated in the Western Zagros M. crassus

as eompared to the other M. crassus speeimens, where the former group shows a tendeney for the

suprameatal triangle to elose posteriorly (landmarks 15 and 16) and the zygomatie plate to be positioned

eaudally (landmark 4). Compared to the eondition in the M. libycus group, the posterior proeess of the

suprameatal triangle beeomes more open and the skull in the bulla part beeomes less high in the Western

Zagros speeimens (not illustrated).

The Monte Carlo simulation on the pooled data from all views (ventral, lateral and dorsal) indieates that

all three groups are signifleantly different from one another (p < 0 . 0001 ).

Size differences between M. libycus and the M. crassus groups

The results of a Kruskal-Wallis test on skull size (with eentroid size obtained from all three skull views)

revealed signifleant (p < 0.0001) inter-group differenees among M. crassus and M. libycus. Bulla size

TABATABAEIYAZDIF., ADRIAENS D. & DARVISH J., Morphological variation in jirds

was also significantly different (p < 0.0001) between the groups in ventral and lateral views, both in

absolute and relative terms (relative to the skull size).

The Kruskal-Wallis test on the ventral skull size data shows a signifieant differenee between the M. crassus

from the Western Zagros and those from the Iranian Plateau, with the latter having signifieantly smaller

skulls (Fig. 5A). The M. crassus from the Western Zagros and M. libycus have a signifieantly larger

eranium (p < 0.05), with the latter having the largest. The Western Zagros group is not different from

the Arabian and Afriean groups (p > 0.05). For the absolute bulla size (both ventral and lateral data, not

illustrated here), the M. crassus from the Western Zagros and those from the Iranian Plateau do not show

a signifieant differenee (p >0.05). However, they show a signifieantly smaller bulla size eompared to the

other groups (p < 0.05), whieh eould be the eonsequenee of the overall small skull size in the M. crassus

speeimens from the Iranian Plateau. The Western Zagros M. crassus speeimens have a smaller, though

not signifieantly, relative bulla size than the M. libycus speeimens (p > 0.05) (Fig. 5B). The Iranian

Plateau speeimens have the largest relative bulla size, being signifieantly different from all the other

groups (p <0.05) exeept the Arabian group (p = 0.062) (Fig. 5B).

Fig. 4. Seatter plot of the CVA results of the (A) ventral and (B) dorsal shape data of Meriones crassus

Sundevall, 1842 (two groups) andM. libycus Fiehtenstein, 1823. Fegends: o = M. crassus (other than

Western Zagros), • = M crassus of Western Zagros, □ = M. libycus. The grids below show deformation

along the arrows, when moving from the M. crassus group mean shape to the Western Zagros group

mean shape (A1 and Bl), and from the M. libycus mean shape to the mean shape of the Western Zagros

(A2 and B2) (shape differenees magnified three times for better visualization). For the numbering of

landmarks, see Fig. 2.

European Journal of Taxonomy 88 : 1-28 ( 2014 )

Patterns of intraspecific shape variation in M. crassus

The plot of the first two prineipal eomponents shows three groups that ean more or less be separated

from eaeh other (Fig. 6): (A) the Iranian Plateau group, (B) the Western Zagros group and (C) the

Afiiean group with the Arabian speeimens and those from Kuwait, Jordan and NW Arabia seattered

around them.

The first two prineipal eomponents aeeount for almost 40 % of the total shape variation and show that the

Western Zagros speeimens have low RWl-seores (mainly eorresponding to having a less infiated bulla)

eompared to the other groups and oeeupy a somewhat distinet part of morphospaee (Fig. 6). Within the

M. crassus morphospaee, the following shape ehanges are the most obvious (as explained by RW1 and

RW2, Fig. 6): in ventral view, the speeimens from the Iranian Plateau (with highest RWl-seores) are

eharaeterized by a tympanie bulla being more infiated and protruding laterally and rostrally (landmarks

13, 14 and 15) and a zygomatie areh being less eonvex (landmarks 16 and 19). The speeimens with

higher RW2-seores, mostly from the Iranian Plateau and the Western Zagros Mountain and Jeddah,

are eharaeterized by a less eonvex zygomatie areh, but with the most lateral point of the zygomatie

areh being positioned more rostrally (landmarks 16 and 19) and the zygomatie plate being less wide

(landmarks 17 and 18).

For the dorsal view (Fig. 6B), speeimens from the Iranian Plateau have the highest RWl-seores and

are eharaeterized by a substantially infiated tympanie bulla that protrudes rostrally and more laterally

(landmarks 15 and 16) and a narrower eranium at the level of the zygomatie plates and oeeiput. The high

RW2-seores for speeimens from the Western Zagros and the Iranian Plateau refieet a lateral eompression

of the zygomatie proeess of the squamous and tympanie bulla (landmarks 14, 15, 16 and 12), and a less

eonvex zygomatie plate (landmarks 9 and 8).

For the lateral view (Fig. 6C), the Western Zagros and Jeddah groups eluster together and lie separate

from the other groups. The latter have high RWl-seores, refieeting a more infiated bulla protruding

rostrally (landmarks 10 to 14 and 22) and a substantially wider suprameatal triangle with eompletely

open posterior proeesses (landmarks 15 and 16 shifted away from eaeh other). With inereasing RW2-

seores, the skull roof beeomes less eonvex (the posterior part of the skull shifts eaudodorsally and the

rostral part shifts dorsally as well).

Fig. 5. Box-and-whisker plots of (A) skull size and (B) relative bulla size of the ventral eranium. The

boxes indieate the 25-75 % quartiles; the whiskers represent the minimal and maximal values.

TABATABAEIYAZDIF., ADRIAENS D. & DARVISH J., Morphological variation in jirds

Fig. 6. Scatter plot of PCA results on shape variables of the (A) ventral, (B) dorsal and (C) lateral sides

of Meriones crassus Sundevall, 1842 speeimens. Eegends: o = Iranian Plateau, • = Western Zagros,

* = Kuwait, A = Arabian, A = Jeddah, □ = Jordan/NW Arabia, ■ = Afriean. Deformation grids (two times

magnified) along the first prineipal eomponents, representing shape differenees between eonfigurations

eorresponding to minimal and maximal seores, are shown to the right of eaeh plot. For the numbering

of landmarks, see Fig. 2.

European Journal of Taxonomy 88 : 1-28 ( 2014 )

The highly significant Hotelling’s p-values (Bonferroni corrected) from the classifier analysis (p<0.001)

shows that most specimens were correctly assigned to their a priori defined M. crassus groups (Iranian

Plateau specimens: 98 %, Western Zagros specimens: 100 %, Arabian specimens: 69 % and African

specimens: 86 %).

The CVA ordination shows that the Western Zagros group is clearly separated from the other groups by

CVl (Fig. 7). The Iranian Plateau group and the African groups are also fairly well separated from each

other along CV2.

For the ventral shape, when comparing the consensus of the Western Zagros group with that of the

African group (Fig. 7A1), the former has a less infiated tympanic bulla (landmarks 12 and 13) and the

zygomatic plate is slightly narrower (landmarks 17, 18 and 20). When the Western Zagros consensus is

compared to that of the Iranian Plateau group (Fig. 7A2), the tympanic bulla is less infiated (landmarks 12

and 13), the zygomatic arch is more convex (landmarks 16 and 19), and the temporal bone is positioned

more caudally (landmarks 14 and 15 are positioned more toward the median line and caudally).

When comparing the Western Zagros and the African groups, with regard to the dorsal shape consensus,

it is found that the former has a tympanic bulla that is clearly less infiated (landmarks 15 and 16),

the zygomatic process of the squamosal is narrower (landmarks 12 and 14), the interparietal is wider

(landmarks 18 and 19), and as the bulla is less infiated, a smaller part of the bulla is visible from a dorsal

view. Comparing it to the mean shape of the Iranian Plateau group, the consensus of the Western Zagros

group has a less infiated tympanic bulla (landmarks 15 and 16), the zygomatic process of the squamosal

CVl (58.58%) CV1 (71.51%) CVl (69.81%)

Fig. 7. CVA scatter plot (axes 1 and 2) on shape variables of the (A) ventral, (B) dorsal and (C) lateral side

of the Meriones crassus groups (Jeddah group not included). Legends: o = Iranian plateau, • = Western

Zagros, A = Arabian and ■ = African. Grids show deformation (3 x magnified) when following the

trajectory within the morphospace along the arrows and between the groups’ consensus (from African

to Western Zagros -Al, B1 and Cl; and from Iranian plateau to Western Zagros -A2, B2 and C2). For

the numbering of landmarks, see Fig. 2.

TABATABAEIYAZDIF., ADRIAENS D. & DARVISH J., Morphological variation in jirds

Table 3. Euclidean distances between the Meriones crassus group mean shapes obtained from the Monte

Carlo pairwise eomparisons (*p <0.01, **p <0.001).

Pairwise grouping

Ventral

Dorsal

Lateral

Iranian plateau-Western Zagros

0.0192**

0.0222**

0.0415**

Iranian plateau-Arabian

0.0095*

0.0212**

0.0193**

Iranian plateau-African

0.0127**

0.0309**

0.0190**

Western Zagros-Arabian

0.0152**

0.0287**

0.0355**

Western Zagros-African

0.0171**

0.0123**

0.0351**

Arabian-African

0.0080

0.0222*

0.0174**

extends more laterally (landmarks 13 and 14), the nasal tip is projeeted less rostrally, and the premaxilla

is positioned more rostrally (landmarks 1 and 7).

Compared to the eonsensus of the Afriean group, the lateral faee of the skull of the Western Zagros

group (Fig. 7C1) has a less inflated bulla, the suprameatal triangle shows a tendeney to elose dorsally

and the zygomatie plate tends to move eaudally (landmark 4). Compared to the Iranian Plateau group

(Fig. 7C2), ehanges in the bulla follow the same pattern as deseribed in Fig. 7C1, but the posterior

proeesses of the suprameatal triangle are more elosed in the Western Zagros eonsensus; the skull is

higher at the level of the maxillary part, and the nasal is shorter.

The MANOVA reveals a high level of morphologieal differentiations among the M. crassus groups

(p < 0 . 001 ) with = 13.83, Fjq 296 o^ ^^^^ 120675 ^ 13.75 for the ventral, dorsal and lateral shape

datasets, respeetively. The pair-wise eomparisons show that all the M. crassus groups are signifleantly

different from eaeh other, exeept for the Arabian and Afriean ones (ventral view. Table 3). The Monte

Carlo test on the pooled data supports an overall eranial shape differenee between these groups as well

(p< 0.001).

In both the UPGMA and NJ analyses, three major elusters emerge, though eomprising different groups.

In the NJ dendogram, a Western Arabian + Iranian eluster is observed, whieh is more similar to the

Afriean group rather than to the Eastern Arabian group (ineluding Kuwait). The UPGMA, however,

suggests a eloser morphologieal affinity between the Eastern and NW Arabian groups, in whieh the

Jeddah

Western Zagros

100

Iranian Plateau

NW Arabia / Jordan

African

Eastern Arabia

Kuwait

100

Fig. 8. Dendograms obtained from the (A) NJ elustering and (B) UPGMA, using Euelidean distanees

between group means by eombining all data (shape information from dorsal, ventral and lateral views).

Braneh bootstrap support shown at the nodes, 10 000 replieates.

European Journal of Taxonomy 88 : 1-28 ( 2014 )

Table 4. The 2B-PLS results derived from speeies-speeifie analyses in Meriones crassus Sundevall, 1842

and M. libycus Liehtenstem, 1823. The relationship between eaeh of the first two veetors and the geoelimatie

variables are represented by the eorrelation eoeffieient and the pereentage of eovariation (*p < 0.05).

M. crassus

M. libycus

Variable

Elevation (m)

-0.554

-0.141

-0.065

-0.586

Rainfall (mm)

-0.115

0.684

-0.275

0.482

Mean T (°C)

0.541

-0.024

0.561

0.290

Max T (°C)

0.463

-0.299

0.099

0.436

Min T (°C)

0.415

0.340

0.637

0.083

% cov explained

80.544*

12.234*

57.987*

26.966*

Afriean group lies nested. As in the NJ, the Western Zagros and Jeddah groups are very similar, but are

most distinet from all other groups based on the UPGMA. Based on these rather eonflieting results, it is

diffieult to draw any eonelusion, although the UPGMA results are supported by higher bootstrap values.

Still, both analyses show that the Western Zagros and Jeddah group do form a distinet group.

The MANCOVA results show that even when taking size variation between the M. crassus groups

into aeeount, they are still signifieantly different in shape (Wilks’ lambda = 0.067, F-value = 6.723;

p = 0.000). The results show that there is no apparent sexual dimorphism in shape when eonsidering

the size-related variation (Wilks’ lambda = 0.846, F-value = 0.982; p = 0.51), but there is an interaetion

effeet of group on sex (Wilks’ lambda = 0.522, F-value = 1.3; p = 0.018). This means sex and group

(together) have been affeeted the observed inter-group differenees in M. crassus.

Geoelimatie eorrelations with eranial variation

Testing the assoeiation between the morphologieal variation and the geoelimatie variation shows that

only in M. crassus is there a tendeney for eorrelation of bulla size with elevation, but not with rainfall.

No eorrelation for either of these geoelimatie variables is observed in M. libycus (Table 4). Overall skull

shape variation in M. crassus proved to be mainly eorrelated with elevation and temperature, versus with

temperature and rainfall in M. libycus (Table 4).

Discussion

Is the Meriones crassus from Western Zagros phenotypically distinct?

The analyses of the separate datasets (ventral, dorsal and lateral views) show that the distinetion between

M. crassus and M. libycus is not always elear. However, the pooled data show that they are signifieantly

different in their overall skull shape, espeeially at the level of the dorsal view. Moreover, the investigation

of the phenotypie variation within the entire geographie range of M. crassus shows that the speeimens

from the Iranian Plateau reveal a eonsiderably distinet eranial form. It also shows that the Afriean and

Arabie populations seem more similar to those from the Iranian Plateau, although being geographieally

more distant. The results thus elearly support the hypothesis that M. crassus from Western Zagros has

a distinet eranial phenotype, whieh eould indieate that it is a separate natural group (see below). Sinee

M. crassus of the Western Zagros is morphologieally (eranial size and external eharaeters) distinet from

M. libycus, and sinee M. libycus eo-oeeurs with M. crassus in the Khouzestan Plain and Mesopotamia,

it ean be exeluded from being synonymous. This differenee is most pronouneed in the dorsal skull

view. It eould also be eonfirmed that the observed shape differenees were not just eorrelated with the

eranial differenees in size, where M. libycus has a substantially larger skull than the M. crassus groups.

TABATABAEIYAZDIF., ADRIAENS D. & DARVISH J., Morphological variation in jirds

Eooking at skull size, specimens from the Western Zagros represent a unique configuration in having a

significantly smaller skull compared to that of M. libycus, although they are not different in the relative

size of the bulla (Fig. 5). This is in contrast to the other M. crassus specimens, that clearly have larger

relative bulla sizes (Fig. 5B). The observed small bulla size in M. crassus from the Western Zagros

corresponds with the observations by Fetter (1961) and Thomas (1919). Our observations also seem to

confirm that, in agreement with Thomas (1919) and Harrison (1972), Jeddah specimens have a skull shape

most similar to those of the Western Zagros specimens. Still, this does not provide sufficient support to

make any further claims about the taxonomic status of M. c. longifrons, a subspecies described, based

on specimens from Jeddah, as this requires further research on a larger sample.

Do Meriones crassus from Western Zagros form a distinct geographical and natural group?

Considering the apparent geographical isolation of the Western Zagros group from the other Iranian

M. crassus, it must be noted that the Khuzestan Plain, where M. crassus specimens from the Western

Zagros are distributed, is geographically separated from the Iranian Plateau by the Zagros Mountains

(they form the Iranian Plateau’s western boundary with eastern Mesopotamia).

The analyses in this study (e.g. the cluster analysis) clearly support the hypothesis that the Western

Zagros group represents a distinct group (not considering the Jeddah group, see previous paragraph) and

also confirmed the dimorphic nature of the skull shape within Iranian populations of M. crassus. It can

thus be concluded that the Western Zagros M. crassus is both morphologically (mainly in having a less

swollen bulla and darker, pinkish buff fur) and geographically distinct from other Iranian populations

of Meriones crassus. Whether or not this now reflects reproductive isolation, and hence the result of

a recent or ongoing speciation event, cannot be ascertained based on our data. Population genetic and

phylogenetical studies should be carried out to verily this.

Geoclimatic correlation of the morphological variation

The impact of environmental factors, especially aridity, on the morphological variation in jirds has long

been recognized, thereby to some degree explaining patterns of inter- and intraspeciflc morphological

differences (e.g. Fetter 1961; Senegas 2001; Chevret & Dobigny 2005; Darvish 2009; Pavlinov et al.

2010; Tabatabaei Yazdi & Adriaens 2011; Tabatabaei Yazdi et al. 2012). This is confirmed in our study,

where the covariation between the bulla and environmental factors is prominent. The increase in tympanic

bulla size can be considered to be adaptive (Fay 1972; Webster & Webster 1984; Van der Straeten &

Dieterlen 1992; Webster & Plassmann 1992; Huang et al. 2002), as it causes an increased amplification

of sound that allows rodents to increase low-frequency hearing and detect an approaching predator more

rapidly (Burda et al. 1990; Mason 2001, 2003; Huang et al. 2002; Schleich & Vasallo 2003). As such, a

bulla hypertrophy may reflect a morphological adaptation for efficient vocal communication in a desert

environment, where rodent population densities are low, and in an underground environment as well

(Fetter 1961; Harrison 1972; Fay 1972; Darvish 2009; Colangelo et al. 2010). This is corroborated by

Vaughan et al. (2000) and our study, as smaller tympanic bullae were found in populations occurring

in regions with higher levels of rainfall (and vice versa). Smaller bullae were found in specimens

occurring in the lowlands of the Khuzestan Plain and Mesopotamia, which have a high humidity during

the summer and higher annual rainfall (ranges from 250 to 500 mm; Britannica Online Encyclopedia,

FAO 2007). Considering that climatic conditions in the Khuzestan Plain and Jeddah habitats are similar,

the observed morphological similarities between specimens originating from these localities could

reflect convergent phenotypes. Additionally, specimens originating from the Iranian Plateau, where

drier climatic conditions exist, have larger bullae (in most of the country, yearly precipitation averages

250 mm or less; Firouz 2005).

European Journal of Taxonomy 88 : 1-28 ( 2014 )

The need for a revision of identification keys

The manifest intraspeeifie and elimate-dependent morphologieal plastieity in the skulls of the studied

jird speeies (espeeially involving the bulla hypertrophy and suprameatal triangle eondition) is often used

as an important diagnostie eharaeter in regional identifieation keys (e.g., Chaworth-Muster & Ellerman

1947). It is elear from this study, that these keys mostly do not take the natural range of intraspeeifie

variability into aeeount. Henee, they should be used with eaution and are in need of revision.

Acknowledgments

We are very grateful to J. Darvish (Ferdowsi University of Mashhad), M.D. Carleton (USNM), L.R.

Heaney (FMNH), J. Cuisin (MNHN) and P. Jenkin and R. Portela Miguez (BMNH), for providing us

aeeess to their eolleetions. Thanks to Tom Greeninekx and the anonymous referees of earlier drafts of

this manuseript for their valuable eomments and assistanee in improving this artiele.

References

Adams D.C. & Rohlf F.J. 2000. Eeologieal eharaeter displaeement in Plethodon: Biomeehanieal

differenees found from a geometrie morphometrie study. Proceeding of the Natural Academic of

Sciences of the U.S.A. 97 (8): 4106^111. http://dx.doi.Org/10.1073/pnas.97.8.4106

Anderson S.C. 1989. Fauna of Persia. Encyclopedia Iranica 9 (4): 437^46.

Bareiova F. & Maeholan M. 2006. Morphometrie study of two speeies of wood miee Apodemus

sylvaticus and A. flavicollis (Rodentia: Muridae): traditional and geometrie morphometrie approaeh.

Acta Theriologica 51 (1): 15-27. http://dx.doi.org/10.1007/BF03192651

Bookstein F.F. 1991. Morphometric Tools for Landmark Data: Geometry and Biology. Cambridge

University Press, Cambridge, UK.

Bookstein F.F., ChemoffB., Elder R.F., Humphries J.M., Smith G.R & Strauss RE. 1985. Morphometries

in Evolutionary Biology, the Geometry of Size and Shape Change with Examples from Fishes. Speeial

publieation 15, Aeademy of Natural Seienees, Philadelphia.

Burda H., Bruns V. & Muller M. 1990. Sensory adaptations in subterranean mammals. Progress in

Clinical and Biological Research 335: 269-293.

Cardini A., Jansson A.-U. & Elton S. 2007. A geometrie morphometrie approaeh to the study of elinal

variation in vervet monkeys. Journal of Biogeography 34 (10): 1663-1678. http://dx.doi.org/10.llll/

i.l365-2699.2007.01731.x

Caumul R. & Polly P.D. 2005. Phylogenetie and Environmental Components ofMorphologieal Variation:

Skull, Mandible, and Molar Shape in Marmots (Marmota, Rodentia). Evolution 59 (11): 2460-2472.

http://dx.doi.Org/10.llll/i.0014-3820.2005.tb00955.x

Chaworth-Musters J.F. & Ellerman J.R. 1947. A Revision of the Genus Meriones. Proceedings of the

Zoological Society of London 117 (2-3): 478-504. http://dx.doi.Org/10.llll/j.1096-3642.1947.tb00533.x

Chevret P. & Dobigny G. 2005. Systematies and evolution of the subfamily Gerbillinae (Mammalia,

Rodentia, Muridae). Molecular Phylogenetics and Evolution 35 (3): 674-688. http: //dx. doi. org/10.1016/j.

vmpev.2005.01.001

Colangelo P., Castiglia R., Franehini P. & Solano E. 2010. Pattern of shape variation in the eastern Afriean

gerbils of the genus Gerbilliscus (Rodentia, Muridae): Environmental eorrelations and implieation

for taxonomy and systematies. Mammalian Biology 75 (4): 302-310. http://dx.doi.org/10.1016/j.

mambio.2009.05.001

TABATABAEIYAZDIF., ADRIAENS D. & DARVISH J., Morphological variation in jirds

Cordeiro-Estrela R, Baylac M., Denys C. & Polop J. 2008. Combining geometric morphometries and

pattern recognition to identify interspecific patterns of skull variation: case study in sympatric Argentinian

species of the genus Calomys (Rodentia: Cricetidae: Sigmodontinae). Biological Journal of the Linnean

Society 94 (2): 365-378. http://dx.doi.Org/10.llll/i.1095-8312.2008.00982.x

Darvish J. 2009. Morphometric comparison of fourteen species of the genus Meriones Illiger, 1811

(Gerbillinae, Rodentia) from Asia and North Africa. Iranian Journal of Animal Biosystematics 5 (1): 59-77.

Dobigny G., Baylac M. & Denys C. 2002. Geometric morphometries, neural networks and diagnosis

of sibling Taterillus species (Rodentia, Gerbillinae). Biological Journal of the Linnean Society 77 (3):

319-327. http://dx.doi.Org/10.1046/i.1095-8312.2002.00074.x

Elewa A.M.T. (ed.) 2004. Morphometries: Applications in Biology and Paleontology. Springer, Berlin.

Ellerman J.R. 1948. Key to the Rodents of South-West Asia in the British Museum. Proceeding of the

Zoological Society of London 118 (3): 765-816. http://dx.doi.Org/10.llll/i.1096-3642.1948.tb00406.x

Fadda C. & Corti M. 2001. Three-dimensional geometric morphometries of Arvicanthis: implications

for systematics and taxonomy. Journal of Zoololgical Systematics and Evolutionary Research 39 (4):

235-245. http://dx.doi.Org/10.1046/i.1439-0469.2001.00169.x

FAO 2007. FAO World Climate Data [online]. Available from http://freegeographytools.eom/2007/fao-

world-climate-data [accessed 20 July 2012]

Firouz S. 2005. The Complete Fauna of Iran. LB. Tauris Publishers, Fondon.

Gower J.C. 1975. Generalized procrustes analysis. Psychometrika 40 (1): 33-51. http://dx.doi.

org/10.1007/BF02291478

Hammer 0., Harper D.A.T. & Ryan P.D. 2001. PAST: Paleontological Statistics Software Package for

Education and Data Analysis. Palaeontologia Electronica 4(1). Available from http://palaeo-electronica.

org/2001_l/past/issuel_01.htm [accessed 20 Jan. 2013]

Harrison D.F. 1956. Gerbils from Iraq, with description of a new gerbil. Journal of Mammalia 37 (3):

417^22. http://dx.doi.org/10.2307/1376743

Harrison D.F. 1972. The Mammals of Arabia. Vol. 3. Lagomorpha, Rodentia. Ernest Been Ftd, Fondon.

Hatt R.T. 1959. The Mammals of Iraq. Miscellaneous Publications 106, Museum of Zoology, University

of Michigan, Michigan.

Hood G.M. 2010. PopTools version 3.2.3. Available from http://www.cse.csiro.au/poptools [accessed

20 Aug. 2013]

Huang G.T., Rosowski J.J., Ravicz M.E. & Peake WT 2002. Mammalian ear specializations in arid

habitats: structural and functional evidence from sand cat {Felis margarita). Journal of Comparative

Physiology A 188 (9): 663-681.

Klingenberg C.P 2008. MorphoJ. Faculty of Fife Sciences, University of Manchester, UK. Available

from http ://www. fry wings. org.uk/Morpho J_page .htm [accessed 2 Sep. 2012]

Koffler B.R. 1972. Meriones crassus. Mammalian Species 9: 1^. http://dx.doi.org/10.2307/3504037

Fay D.M. 1967. A Study of the Mammals of Iran, Resulting from the Street Expedition of 1962-63.

Fieldiana Zoology 54, Field Museum of Natural History, Chicago, http://dx.doi.org/10.5962/bhl.

title.2961

European Journal of Taxonomy 88 : 1-28 ( 2014 )

Lay D.M. 1972. The Anatomy, Physiology, Functional Significance and Evolution of Specialized Hearing

Organs of Gerbilline Rodents. Journal of Morphology 138 (1): 41-120. http://dx.doi.org/10.1002/

jmor.1051380103

Macholan M., Mikulaa O. & Vohralikc V. 2008. Geographic phenetic variation of two eastem-

Mediterranean non-commensal mouse species, Mus macedonicus and M. cypriacus (Rodentia: Muridae)

based on traditional and geometric approaches to morphometries. Zoologischer Anzeiger 247 (1): 67-

80. http://dx.doi.Org/10.1016/j.icz.2007.07.003

Mason M.J. 2001. Middle ear structures in fossorial mammals: a comparison with non-fossorial

species. Proceedings of the Zoological Society of London 255 (4): 467-86. http://dx.doi.org/10.1017/

S0952836901001558

Mason M.J. 2003. Morphology of the middle ear of golden moles (Chrysochloridae). Proceedings of the

Zoological Society of London 260 (4): 391^03. http://dx.doi.org/10.1017/S095283690300387X

Misonne X. 1959. Analyse Zoogeographique des Mammiferes de I Tran. Memoires de ITnstitut royal des

Sciences naturelles de Belgique, deuxieme serie 59, Institut Royal des Sciences Naturelles de Belgique,

Brussels.

Misonne X. 1975. The Rodents of the Iranian Deserts. In: Prakash 1. & Ghosh PK. (eds) Rodents

in Desert Environments: 47-58. Monographiae Biologicae 28, Springer Netherlands, http://dx.doi.

org/10.1007/978-94-010-1944-6 3

Momenzadeh M., Darvish J., Toutoniyan F., Sarmad M. & Sadeghi Shakib F. 2008. Investigation of

climatic effects on the shape and volume of tympanic bulla of Meriones libycus and Meriones persicus

(Muridae: Rodentia) from northeastern Iran: an evolutionary approach. Iranian Journal of Animal

Biosystematics A: 63-73.

Monteiro L.R., Duarte L.C. & Reis S.F. dos 2003. Environmental correlates of geographical variation

in skull and mandible shape of Thrichomys apereoides (Rodentia: Echimyidae). Journal of Zoology 261

(1): 47-57. http://dx.doi.Org/10.1017/S0952836903003893

Musser G.G. & Carleton M.D. 2005. Superfamily Muroidea. In: Wilson D.E. & Reeder D.M. (eds)

Mammal Species of the World: A Taxonomic and Geographic Reference (3rd ed.): 894-1531. Johns

Hopkins University Press, Baltimore.

Osborn D.J. & Helmy 1. 1980. The Contemporary Land Mammals of Egypt (Including Sinai). Fieldiana

Zoology, new series 5, Field Museum of Natural History, Chicago, http://dx.doi.org/10.5962/bhl.

title.2801

Pavlinov 1. Y. & Rogovin K.A. 2000. Relation between size of pinna and of auditory bulla in specialized

desert rodents. Journal of General Biology 61: 87-101. [In Russian with English summary]

Pavlinov l.Y. 2008. A Review of Phylogeny and Classification of Gerbillinae (Mammalia: Rodentia).

Moscow University Publishing, Moscow.

Pavlinov l.Y, Eissovsky A.A. & Obolenskaya E.V. 2010. Geographic variation of skull traits in the

Eibyan jird, Meriones libycus (Rodentia: Gerbillinae), over its entire distribution area. Russian Journal

ofTheriology9: 19-26.

Petter F. 1959. Evolution du dessin de la surface d’usure des molaires des Gerbillides. Mammalia 23 (3):

304-315. http://dx.doi.Org/10.1515/mamm.1959.23.3.304

Petter F. 1961. Repartition geographique et ecologique des rongeurs desertiques (du Sahara occidental a

ITran oriental). Mammalia 25 (si): 1-222.

TABATABAEIYAZDIF., ADRIAENS D. & DARVISH J., Morphological variation in jirds

Popesko R, Rajtova V. & Horak J. 2002. Colour Atlas of the Anatomy of Small Laboratory Animals. Vol.

2. Rat, Mouse and Golden Hamster. Wolfe Publishing, Bratislava.

Raup D. & Crick R.E. 1979. Measurement of Faunal Similarity in Paleontology. Journal of Paleontology

53 (5): 1213-1227.

Ranck G. E. 1968. The Rodents of Libya: Taxonomy, Ecology and Zoogeographical Relationships.

Bulletin of the United States National Museum 275, Smithsonian Institution Press Washington.

Rohlf F.J. & Slice D.E. 1990. Extensions of the Procrustes methods for the optimal superimposition of

landmarks. Systematic Zoology 39 (1): 40-59. http://dx.doi.org/10.2307/2992207

Rohlf F.J. 1993. Relative Warp Analysis and an Example of its Application to Mosquito Wings. In\

Marcus E.F., Bello E. & Garcia-Valdecasas A. (eds) Contributions to Morphometries'. 131-159.

Monografias, Museo Nacional de Ciencias Naturales, Madrid, http://dx.doi.org/10.5962/bhl.title.15368

Rohlf F.J. & Marcus E.F. 1993. A revolution in Morphometries. Trends in Ecology & Evolution 8 (4):

129-132. http://dx.doi.org/10.1016/0169-5347/93)90024-J

Rohlf F.J. 1995. Multivariate Analysis of Shape Using Partial-warp Scores. In'. Mardia K.V. & Feeds

C.A. (eds) Proceedings in Current Issues in Statistical Shape Analysis'. 154-158. Feeds University

Press, Feeds.

Rohlf F.J. 1999. Shape Statistics: Procrustes Superimpositions and Tangent Spaces. Journal of Classification

16 (2): 197-223. http://dx.doi.Org/10.1007/s003579900054

Rohlf F.J. & Corti M. 2000. Use of Two-Block Partial Feast Squares to Study Covariation in Shape.

Systematic Biology 49 (4): 740-753. http://dx.doi.Org/10.1080/106351500750049806

Rohlf F.J. 2003. TpsSmall: Thin Plate Spline Small Variation Analysis, version 1.20. Department of

Ecology and Evolution, State University of New York at Stony Brook. Available from http://life.bio.

sunysb.edu/morph/ [accessed 20 Jun. 2012]

Rohlf F.J. & Bookstein RF. 2003. Computing the Uniform Component of Shape Variation. Systematic

Biology 52 (1): 66-69. http://dx.doi.org/10.1080/10635150390132759

Rohlf F.J. 2004a. TpsDig: T hin Plate Spline Digitise, version 2.12. New York: Department of Ecology

and Evolution, State University of New York at Stony Brook.. Available from http://life.bio.sunvsb.edu/

morph/ [accessed 20 Jun. 2012]

Rohlf F.J. 2004b. TpsSplin: Thin Plate Spline, version 1.20. Department of Ecology and Evolution, State

University of New York at Stony Brook. Available from http://life.bio.sunysb.edu/morph/ [accessed 20

Jun. 2012]

Rohlf F.J. 2008. TpsRelw: Thin Plate Spline Relative Warps, version 1.46. Departement of Ecology

and Evolution, State University of New York at Stony Brook. Available from http://life.bio.sunysb.edu/

morph/ [accessed 20 Jun. 2012

Schleich C.E. & Vasallo A.I. 2003. Bullar Volume in Subterranean and Surface-Dwelling

Caviomorph Rodents. Journal of Mammalogy 84 (1): 185-189. http://dx.doi.org/10.1644/1545-

1542t2003)084<0185:BVISAS>2.Q.CO:2

Senegas F. 2001. Interpretation of the dental pattern of the South African fossil Euryotomys (Rodentia,

Murinae), and origin of otomyine dental morphology). In'. Denys C., Granjon F. & Poulet A. (eds)

African Small Mammals. Proceedings of the 8th International Symposium on African Small Mammals'.

151-160. Colloques et Seminaires, Institut de recherche pour le developpement, Paris.

European Journal of Taxonomy 88 : 1-28 ( 2014 )

StatSoft, Inc. 2004. STATISTICA (data analysis software system), version 7. Available from www.

StatSoft, eom [aeeessed 20 Jun. 2012]

Tabatabaei Yazdi F. & Adriaens D. 2011. Patterns of phenotypie skull variation in M. persicus (Rodentia:

Muridae) in relation to geoelimatie eonditions. Iranian Journal of Animal Biosystematics 1 (2): 129-142.

Tabatabaei Yazdi F., Adriaens D. & Darvish J. 2012. Geographie pattern of eranial differentiation

in the Asian Midday Jird Meriones meridianus (Rodentia: Muridae: Gerbillinae) and its taxonomie

implieations. Journal of Zoological Systematics and Evolutionary Research 50 (2): 157-164. http://

dx.doi.org/10.1111/i. 1439-0469.2011.00642.x

Thomas O. 1919. Notes on Gerbils referred to the Genus Meriones, with Deseriptions of new

Speeies and Subspeeies. The Annals and Magazine of Natural History 3: 263-273. http://dx.doi.

org/10.1080/00222931908673822

Tong H. 1989. Origine et evolution des Gerbillidae (Mammalia, Rodentia) en Afrique Du Nord.

Memoires de la Soeiete geologique de Franee, nouvelle serie 155, Soeiete geologique de Franee, Paris.

Vaughan T.A., Ryan J.M. & Czaplewski N.J. 2000. Mammalogy (4th ed.). Saunders College Publishers,

Orlando, Florida.

Van der Straeten E. & Dieterlen F. 1992. Craniometrieal eomparison of four populations of Praomys

jacksoni eaptured at different heights in Eastern Zaire (Kivu). Mammalia 56 (1): 125-31.

Webster D.B. & Webster M. 1984. The Speeialized Auditory System of Kangaroo Rats. Contributions to

Sensory Physiology 161-196. http://dx.doi.org/10.1016/B978-0-12-151808-0.50012-5

Webster D.B. & Plassmann W. 1992. Parallel Evolution of Eow-Frequeney Sensitivity in Old World and

New World Desert Rodents. In: Webster D.B., Popper A.N. & Fay R.R. (eds) The Evolutionary Biology

of Hearing: 625-631. Springer-Verlag, New York, http://dx.doi.org/10.1007/978-l-4612-2784-7_39

Wittmer W. & Btittiker W. 1982. Fauna of Saudi Arabia. Vol. 4. Pro Entomologia, e/o Natural History

Museum, California.

ZelditehM.E., SwiderskiD.E., Sheets H.D. &FinkW.E. 2004. Geometric Morphometries for Biologists:

a Primer. Elsevier Aeademie Press, Eondon.

Manuscript received: 11 August 2013

Manuscript accepted: 1 April 2014

Published on: 25 June 2014

Topic editor: Rudy Joeque

Desk editor: Charlotte Thionois

Printed versions of all papers are also deposited in the libraries of the institutes that are members of

the EJT eonsortium: Museum National d’Histoire Naturelle, Paris, Franee; Botanie Garden Meise,

Belgium; Royal Museum for Central Afriea, Tervuren, Belgium; Natural History Museum, Eondon,

United Kingdom; Royal Belgian Institute of Natural Seienees, Brussels, Belgium; Natural History

Museum of Denmark, Copenhagen, Denmark.

TABATABAEIYAZDIF., ADRIAENS D. & DARVISH J., Morphological variation in jirds

Appendix 1. Catalogue numbers of the analysed speeimens.

Meriones crassus

Smithsonian National Museum of Natural History (Washington D.C., USA)

328235, 328236, 350507, 350508, 350509, 350479, 350481, 329156, 329157, 350468, 350470, 350471,

350472, 350475, 350476, 350477, 350478, 354739, 354741, 354743, 369465, 369466, 350500, 350501,

350503, 350505, 326904, 326905, 326906, 326910, 326911, 326914, 326915, 326916, 326918, 326919,

326920, 326921, 326922, 326923, 326924, 326926, 326928, 326931, 326932, 326935, 326937, 326940,

326947, 328228, 328230, 328231, 328232, 328233, 328245, 328246, 328248, 354732, 354733, 354734,

354735, 354736, 350484, 350487, 350488, 350489, 350490, 350491, 350492, 350493, 350494, 350196,

350497, 350498, 350512, 354710, 354711, 354712, 354713, 354714, 354715, 3564716, 354717,

354718, 354719, 354721, 354722, 354723, 354724, 354727, 354729, 354699, 354701, 354702, 354703,

354704, 354705, 354706, 354707, 354708, 354709, 328237, 328238, 328239, 328240, 328243, 329151,

329153, 329154, 329155, 354738, 369467, 369468, 369470, 369471, 369472, 369473, 369474, 369476,

369486, 369489, 369490, 369491, 369492, 369493, 369505, 369506, 369507, 322693, 322659, 322660,

322661, 322662, 322663, 322664, 322665, 322666, 322667, 322668, 322694, 322695,282504,282505,

282506, 282507,282508, 282509,282512,282513, 283230,283231, 321872, 321873, 321874, 401157,

401158, 401159, 401160, 401161, 401162, 401163, 401165, 401166, 401167, 401168, 401169, 401171,

401172, 401173, 401174, 401175, 401176, 401177, 401178, 401179, 401180, 401181.

Field Museum of Natural History (Chicago, USA)

103308, 103311, 103312,103313, 103314,103315,103317, 103318,103319, 103321,103322,103323,

97234, 97238, 97227, 97231, 97232, 97249, 97214, 97216, 97220, 97221, 97223, 97248, 111954,

111655, 153162, 153161, 179005, 179008, 87612, 87618.

British Museum of Natural History (London, UK)

19.7.7.2356, 19.7.7.2808, 47.508, 47.514, 47.510, 47.518, 47.509, 47.513, 95.350, 47.523, 47.529,

66.2584,471537,471535,471538,1979.2052,1968.949,1968.947,1968948,206814,1957.10,206813,

846146, 1023243, 51311, 196124, 196125, 54247, 23241.

Museum national d’Histoire naturelle (Paris, France)

1997-403, 1957-1340, 1955-21, 1957-519, 1953-4230, 1957-1036, 1957-1039, 1957-1043, 1953-425,

1958-104, 1957-1042, 1957-1037, 1957-1038, 1991-1220, 1952-532, 1957-329, 1957-325, 1957-314,

1958-63, 1957-321, 1957-528, 1957-316, 1958-61,1958-62, 1957-825,1957-318, 1957-328, 1957-330,

1966-1054, 1957-823, 1957-311, 1957-315, 1957-309, 1957-307, 1957-312, 1957-320, 1958-6423,

1957-306, 1957-326, 1957-323, 1958-65, 1957-322, 1958-107, 1958-106, 1957-1354, 1995-1700.

Meriones libycus

Smithsonian National Museum of Natural History (Washington D.C., USA)

326941, 326942, 326943, 326944, 326969, 326970, 326971, 326973, 328266, 328267, 328269, 369798,

369799, 369801, 369802, 369806, 369808, 369810, 369812, 369814, 369815, 369817, 369818, 369819,

369822, 369824, 369826, 369827, 369828, 369829, 369831, 369832, 369833, 369834, 354682, 326968,

328257, 328258, 328260, 328262, 328264, 328265, 354690, 354691, 354692, 354695, 341256, 341258,

341261, 341262, 341263, 341264, 341265, 341266, 341267, 328224, 328225, 328271, 328272, 328273,

328274, 328275, 329188, 329189, 329190, 329191, 354696, 354661, 350558, 350559, 350572, 354647,

354649, 354650, 354652, 354653, 354664, 354665.

British Museum of Natural History (London, UK)

40.259, 1983.314, 40.257, 25.4.3.43, 25.4.3.23, 10.3.12.4, 25.4.3.25, 5.123, 25.4.3.20, 47.383,

56.2.29.5, 19.12.10.3, 47.1412, 47.1414.

European Journal of Taxonomy 88 : 1-28 ( 2014 )

Zoological Museum of Ferdowsi University of Mashhad (Mashhad, Iran)

M564, M799, M810, M785, 322, 245, M-35, 1021, 30, 1499, 909, 1473, 1504, 303, 208, 474, 1470,

1468, M844, M571, 1506, 1471, 217, M-17, M-35, 157, M-3, 39, M-28, M-36, 205, 212, M-21, 174,

M-29, 1469, 24, M-34, M-7, 518, M-37, 176, 563, M-1, 1482, 574, 461, 88, 9, M-572, M-579, M-12,

M-808, 4, M-432, M343, 2, M-331, M-188, M-646, M-196, M-11, 850, 11, M-111, M-842, M-573, 7,

M-175, M-548, 14, M-694, M-568, M-236, M-251, 840, 577, 106, M-3, M-12.

Museum national d’Histoire naturelle (Paris, France)

1950-468, 1950-508, 1957-1330, 1957-1338, 1957-1339, 1958-89, 1958-91, 1958-330, 1958-331,

1958-336, 1958-341, 1958342, 1958343, 1958-344, 1958-345, 1958-346, 1958-347, 1958-351, 1958-

352, 1958-750, 1957-334, 1957-335, 1957-332, 1957-336.

Royal Belgian Institute of Natural Sciences (Brussels, Belgium)

10186,10190, 10189,10187, 10348,10322, 10192,10078, 10137,10283, 10184,10497, 10185,10188,

10318, 10319, 10082, 10282, 10194.

TABATABAEIYAZDIF., ADRIAENS D. & DARVISH J., Morphological variation in jirds

Appendix 2. Overview of sampling loealities of Meriones crassus Sundevall, 1842 (M c.) andM libycus

Eiehtenstein, 1823 (M /.) used in this study. F = female; M = male. Total = both sexes ineluding

speeimens of unknown sex.

Taxon

Country

Locality

Latitude

Longitude

Sample size

Total

M c.

Iran

Abas abad

36.38° N

53.05° E

M c.

Iran

Esfahan

33.22° N

51.68° E

M c.

Iran

Kashan

33.98° N

51.45° E

M c.

Iran

Kerman

30.28° N

57.10° E

M c.

Iran

Khash, Iranshahr

28.22° N

61.20° E

M c.

Iran

Kashmar

35.17°N

57.42° E

M c.

Iran

Dasht-e- Lut

32.58° N

59.03° E

M c.

Iran

Mashad

36.05° N

58.85° E

M c.

Iran

Qazvin

36.81°N

51.05° E

M c.

Iran

Sabzevar

36.25° N

57.75° E

M c.

Iran

Shahrud

36.42° N

54.96° E

M c.

Iran

Torbat-e-heidariyeh

35.36° N

59.22° E

M c.

Iran

Zahedan

29.21° N

60.87° E

M c.

Iran

Mahallat

33.75° N

50.50° E

M c.

Iran

Tazuki

30.40° N

61.14°E

M c.

Iran

Andimeshk

33.20° N

48.25° E

M c.

Iran

Ahvaz

31.18°N

49.60° E

M c.

Iran

Qhasr-e-shirin

34.51° N

45.58° E

M c.

Iran

Taj maleki

29.76° N

50.56° E

M c.

Afghanistan

Qala Bist

31.33°N

64.20° E

M c.

Pakistan

Baluchistan

28.58° N

65.41° E

M c.

Pakistan

Nok kundi

28.81°N

62.77° E

M c.

Pakistan

Panjgur

26.75° N

64.00° E

M c.

Iraq

Kirkuk

35.47° N

44.13° E

M c.

Iraq

Ali al Gharbi

32.46° N

46.68° E

M c.

Kuwait

A1 ahmadi

29.07° N

48.08° E

M c.

Saudi Arabia

Qariya, Hafar-al-Batin

27.20° N

44.00° E

M c.

Saudi Arabia

Jeddah

21.51°N

39.22° E

M c.

Israel

Ein Hussub

31.78°N

35.22° E

M c.

Jordan

Zarqa,Ma'am

31.59°N

37.24° E

M c.

Egypt

Al Qahriah

30.10° N

31.33°E

M c.

Libya

Awbari, El Gatrun

29.06° N

15.78° E

M c.

Mauritania

Fort Gouraud

22.61° N

-12.73° W

M c.

Western Sahara

Aguera

20.83° N

-17.09° W

M /.

Iran

Geno

27.41° N

56.18° E

M /.

Iran

Jajarm

36.93° N

56.36° E

M /.

Iran

Kashmar

35.18°N

57.41° E

M /.

Iran

Sabzevar

29.60° N

52.51° E

M /.

Iran

Bajestan

34.51° N

58.18° E

European Journal of Taxonomy 88 : 1-28 ( 2014 )

Taxon

Country

Locality

Latitude

Longitude

Sample size

Total

M /.

Iran

Kouhak, Zabol

31.06° N

61.75° E

M /.

Iran

Garmsar

35.13°N

52.18° E

M /.

Iran

Kerman

29.90° N

56.53° E

M /.

Iran

Gonbad-e-Kavus

37.23° N

55.08° E

M /.

Iran

Shirvan

37.45° N

57.91° E

M /.

Iran

Torbat-e-Heidariyeh

35.28° N

59.21° E

M /.

Iran

Zahedan

29.53° N

60.83° E

M /.

Iran

Torbat-e-Jam

35.21° N

60.61° E

M /.

Iran

Nehbandan

31.53°N

60.03° E

M /.

Iran

Ghoochan

37.10° N

58.50° E

M /.

Iran

Sabzevar

36.20° N

57.71° E

M /.

Iran

Khash

35.18°N

58.45° E

M /.

Iran

Mashhad

36.26° N

59.61° E

M /.

Iran

Sarakhs

36.50° N

61.05° E

M /.

Iran

Bojnurd

37.46° N

57.31° E

M /.

Iran

Dashli Borun

37.63° N

54.81° E

M /.

Iran

Birjand

32.86° N

59.20° E

M /.

Iran

Mansourabad

28.25° N

54.03° E

M /.

Iran

Robat e-Gharabil

37.35° N

56.31° E

M /.

Iran

Qazvin

36.26° N

50.01° E

M /.

Iran

Aghbulagh

35.61°N

48.43° E

M /.

Iran

Tehran

35.85° N

50.86° E

M /.

Iran

Hamedan

35.70° N

48.20° E

M /.

Iran

Ahvaz

31.18°N

49.60° E

M /.

Afghanistan

Kandahar

31.60° N

65.70° E

M /.

Iraq

Ali al Gharbi

32.46° N

46.67° E

M /.

Saudi Arabia

Hufuf

25.37° N

49.58° E

M /.

Saudi Arabia

Shari wells, NW Arabia

24.49° N

44.38° E

M /.

Syria

Al Qaryatayn

34.23° N

37.23° E

TABATABAEIYAZDIF., ADRIAENS D. & DARVISH J., Morphological variation in jirds

Appendix 3. Definition and numbering of the landmarks used for the skull morph analyses.

Landmark

Definition

Ventral view

rostral tip of internasal suture

most lateral junction point of incisive alveolus and body of premaxillary bone

most rostral point of incisive foramen

most caudal point of incisive foramen

most rostral point of palatine foramen

most caudal point of palatine foramen

most rostral point on the alveolus of the first molar

most caudal point on the alveolus of the third molar

most caudal point of median suture of palatine bone

most rostral point of foramen magnum

most lateral point of occipital condyle

most caudal point of acoustic tympanic bulla

rostral curvature point at level of the meatus

most caudal point of zygomatic process concavity formed by temporal bone

intersection between frontal squama, wing of presphenoid bone and wing of basisphenoid bone

most lateral point of zygomatic arch at maximum width of skull

rostral point of zygomatic plate

maximum curvature of zygomatic plate in infraorbital foramen

intersection of zygomatic arch and vertical line passing through most caudal point of third molar

intersection of the zygomatic plate and line connecting landmarks 13 and 18

Dorsal view

rostral tip of internasal suture

intersection of naso-frontal suture with the internasal suture

intersection of frontal-parietal suture and the interparietal suture

intersection of suture between left and right parietals, and parietal-interparietal suture

midline point of suture between interparietal and occipital

midline point of caudal margin of the occipital

most rostral point of suture between nasal and premaxilla

rostral end of zygomatic plate

most lateral point of zygomatic plate

lateral end of the maxillary-frontal suture

rostral point of upper orbital crest at level of interorbital depression

intersection of temporal line and suture between parietal and squamosal bones

tip of concavity of squamosal root of zygomatic arch

caudal tip of squamosal root of zygomatic arch

rostrolateral end of tympanic bulla convexity

caudal end of tympanic bulla on lateral edge of suprameatal process (supramastoid part of squamosal bone)

distal tip of lateral process of supraoccipital

caudal end of suture between the mastoid part of tympanic bulla and supraoccipital

intersection of parietal-interparietal and interparietal-occipital sutures

European Journal of Taxonomy 88: 1-28 (2014)

Lateral view

most rostral point of nasal

inner extreme point of incisor at body of premaxillary bone

point at intersection between premaxillary and posterior end of incisive alveolus

most rostral end of infraorbital foramen edge on zygomatic plate

most ventral point at the margin of zygomatic plate

most caudal point of infraorbital foramen on zygomatic plate

most rostral point of suture between lacrimal and zygomatic plate

most rostral point of molar on alveolar process of maxilla

most caudal point of molar on alveolar process of maxilla

most caudal point of optic canal

middle of alisphenoid canal

most caudal point of suture between jugal and squamosal root of zygomatic arch

intersection between rostral edge of tympanic bulla and most caudal point of gap between tympanic bulla

and occipital process of temporal bone

rostral point of suprameatal triangle

lateral tip of supraoccipital process

tip of hamular process of temporal on suprameatal triangle

rostral end of suture between stylomastoid suture and stylomastoid foramen

most rostral point of paraoccipital process

intersection of suture between parietal and supraoccipital with suprameatal process of squamosa

intersection of temporal line and suture between parietal and squamosal

junction of suture between parietal and squamosal bone and suture between frontal and squamosal part of

temporal bone

intersection of tympanic part of bulla and line connecting landmarks 16 and 17