Skull modularity of the European ground squirrel Spermophilus citellus (Linnaeus, 1766)

Abstract

The skull is a complex structure that has frequently been studied for the
patterns of morphological integration and modularity. The ventral side of the skull can
be divided into two functional modules, the neurocranium composed of the braincase,
eyes and ears, and the viscerocranium composed of the jaw apparatus. The aim of this
study was to test the ventral cranium of the European ground squirrel Spermophilus
citellus (Linnaeus, 1766) for this partitioning as sciurid skull is believed to be highly
integrated without clear divisions into subunits. Additionally, I compared the degree
of modularity between juveniles and adults. Hypothesized modularity was tested on
159 (43 juveniles and 116 adults) skulls by applying geometric morphometric method
based on Escoufier RV coefficient. In adults, the results yielded strong support to the
hypothesis that the viscerocranium and neurocranium are separate modules. In juveniles,
two-module organization of the skull was also confirmed, but the hypothesized
modules were much more integrated with each other. Although allometry can be a
strong integrating factor, it had very little influence on the hypothesized modularity of
the S. citellus skull. A permutation test for the difference in the degree of modularity
between juveniles and adults was marginally significant. The change in the strength
of integration between the viscerocranium and neurocranium in the S. citellus skull
during ontogeny, with the higher level of modularity in adults than in juveniles, is
probably a consequence of the transition from suckling to gnawing of food and greater
specialisation of the two functional modules.

Full text

Skull modularity of the European ground squirrel Spermophilus citellus
(Linnaeus, 1766)
Modularnost lobanje evropske tekunice Spermophilus citellus (Linnaeus, 1766)
Tina Klenovšek
Department of Biology, Faculty of Natural Sciences and Mathematics, University of Maribor,
Koroška cesta 160, 2000 Maribor, Slovenia
Correspondence: tina.klenovsek@um.si
Abstract: The skull is a complex structure that has frequently been studied for the
patterns of morphological integration and modularity. The ventral side of the skull can
be divided into two functional modules, the neurocranium composed of the braincase,
eyes and ears, and the viscerocranium composed of the jaw apparatus. The aim of this
study was to test the ventral cranium of the European ground squirrel Spermophilus
citellus (Linnaeus, 1766) for this partitioning as sciurid skull is believed to be highly
integrated without clear divisions into subunits. Additionally, I compared the degree
of modularity between juveniles and adults. Hypothesized modularity was tested on
159 (43 juveniles and 116 adults) skulls by applying geometric morphometric method
based on Escouer RV coefcient. In adults, the results yielded strong support to the
hypothesis that the viscerocranium and neurocranium are separate modules. In juve-
niles, two-module organization of the skull was also conrmed, but the hypothesized
modules were much more integrated with each other. Although allometry can be a
strong integrating factor, it had very little inuence on the hypothesized modularity of
the S. citellus skull. A permutation test for the difference in the degree of modularity
between juveniles and adults was marginally signicant. The change in the strength
of integration between the viscerocranium and neurocranium in the S. citellus skull
during ontogeny, with the higher level of modularity in adults than in juveniles, is
probably a consequence of the transition from suckling to gnawing of food and greater
specialisation of the two functional modules.
Keywords: neurocranium, viscerocranium, ontogeny, allometry, Escouer RV,
morphological integration, geometric morphometrics
Izvleček: Lobanja je kompleksna struktura, pri kateri se pogosto preučujejo vzorci
morfološke modularnosti in integracije. Ventralno stran lobanje lahko razdelimo v dve
funkcionalni podenoti; nevrokranij, ki ga sestavljajo kosti možganskega dela lobanje,
oči in ušes, in viscerokranij, ki je iz kosti čeljustnega aparata. Cilj raziskave je bil
preveriti prisotnost te delitve na ventralni strani lobanje evropske tekunice Spermo-
philus citellus (Linnaeus, 1766), saj za lobanje veveric velja, da imajo visoko stopnjo
integracije brez jasnih delitev na podenote. Primerjala sem tudi stopnjo modularnosti
med mladimi in odraslimi osebki. Hipotetično modularnost sem testirala na 159 (43
mladih in 116 odraslih) lobanjah z metodami geometrijske morfometrije na osnovi
ACTA BIOLOGICA SLOVENICA
LJUBLJANA 2014 Vol. 57, [t. 1: 59–67

60 Acta Biologica Slovenica, 57 (1), 2014
Escouerjevega RV koecienta. Pri odraslih osebkih so rezultati podprli hipotezo o
delitvi lobanje na viscerokranij in nevrokranij. Hipotezo o modularnosti sem potrdila
tudi pri mladih osebkih, le da sta bila modula med seboj veliko bolj povezana. Čeprav
je alometrija lahko močan integracijski faktor, je imela na hipotetično modularnost
lobanje S. citellus zelo majhen vpliv. Permutacijski test, s katerim sem testirala razliko
v stopnji modularnosti med mladimi in odraslimi osebki, je bil na meji signikantnosti.
Sprememba v stopnji povezanosti med viscerokranijem in nevrokranijem pri lobanji
S. citellus med ontogenijo, z višjim nivojem modularnosti pri odraslih kot pri mladih,
je verjetno posledica prehoda s sesanja na glodanje hrane ter večje specializacije obeh
funkcionalnih modulov.
Ključne besede: nevrokranij, viscerokranij, ontogenija, alometrija, Escouer RV,
morfološka integracija, geometrijska morfometrija
Introduction
The skull is a complex structure composed of
many parts that have different embryonic origins
and functions (Klingenberg et al. 2004). To func-
tion as a whole, the parts of the skull are integrated
(Olson and Miller 1958). The integration is not
evenly distributed, but rather structured into mod-
ules (Klingenberg et al. 2004) that are internally
tightly correlated and relatively independent from
other modules (Klingenberg 2008). Integration
and modularity of a structure can be studied by
analyzing the covariation among its traits (Drake
and Klingenberg 2010). In morphometrics, traits are
usually measured by the use of lenghts or landmarks.
It has been shown that in studies of integration and
modularity the two methodologies, i. e. traditional,
using linear measurements, and geometric, using
landmarks, give similar results (Goswami and
Polly 2010, Jojić et al. 2012). Goswami and Polly
(2010) also compared different exploratory and
conrmatory approaches for studying integration
and modularity and did not nd statistically dis-
tinguishable differences among them.
Integration and modularity of the skull have
been the most frequently studied in primates,
whereas in rodents, analyses of the mandible modu-
larity have been more popular (e. g. Klingenberg
and Leamy 2001, Jojić et al. 2007, Klingenberg
2009, Swiderski and Zelditch 2010, Jojić et al.
2012). The ventral side of the skull can be divided
into two functional components, the neurocranium
composed of the braincase, eyes and ears, and the
viscerocranium composed of the jaw apparatus
(Emerson and Bramble 1993). However, different
analyses of the rodent skull showed that patterns
of modularity can be inconsistent and sometimes
unclear (Klingenberg 2013). Among rodents,
modularity and integration in sciurid skulls are,
especially compared to murid skulls, poorly inves-
tigated. Olson and Miller (1958) studied the fox
squirrel Sciurus niger and discovered that its skull
is well integrated without clear subdivisions into
subunits. Moreover, Roth (1996) suggested that
high integration of the sciurid skull is a general
feature of the family; probabaly because of its
conservative evolution.
European ground squirrel, Spermophilus citel-
lus (Linnaeus, 1766), inhabits dry grasslands and
open woodland throughout central and southeastern
Europe (Ramos-Lara et al. in press). It is a rela-
tively well-studied species with clear phylogenetic
stucturing and reasonably well-known ecology and
life history (ibid.). Because of its longevity and
well dened age stages, S. citellus is also a good
organism for studying changes in the strength of
integration over postnatal ontogeny, a process that
has been previously investigated in some rodent
skulls (Willmore et al. 2006, Zelditch et al. 2006,
Gonzalez et al. 2011, Klingenberg 2013).
In this study, I applied geometric morphometric
methods to test the S. citellus ventral cranium for
hypothetical partitioning to two functional modules,
the viscerocranium and neurocranium, despite the
previous ndings that the sciurid skull is highly
integrated (Olson and Miller 1958, Roth 1996).
Additionally, based on prediction that the shift in
diet during ontogeny could inuence the strength
of integration between modules, I compared the
level of modularity between juveniles and adults.

61
Klenovšek: Skull modularity of S. citellus
Material and Methods
I studied 159 skulls of S. citellus from Bur-
genland (Austria) and Banat (Serbia). Specimens
are deposited in the Slovenian Museum of Natural
History (Ljubljana, Slovenia), the Museum of
Natural History (Vienna, Austria), and the Zoologi-
cal Research Museum Alexander Koenig (Bonn,
Germany). Individuals were categorized either
as juveniles (1 - 5 months old, caught after natal
emergence to the end of September) or adults (>
5 months old, caught just before or after the rst
hibernation). Age was estimated on the basis of
molar tooth wear (Ružić 1966) and the date of
collection. Very old individuals (after the fourth
hibernation) were excluded. The sample comprised
24 juveniles and 70 adults from Burgenland, and
19 juveniles and 46 adults from Banat. Adults from
different seasons were pooled because a previous
study of the S. citellus skull ontogeny (Klenovšek
and Kryštufek 2013) showed that the skull shape
does not change after the age of ve months,
which means that the majority of shape changes
during growth are correlated with the shift from
a liquid to a solid diet that takes place before the
rst hibernation.
The ventral side of the skulls was photographed
under constant conditions, following Cardini and
Tongiorgi (2003). Twenty-two two-dimensional
landmarks were digitized on the left side of the
skull (Fig. 1), using the tpsDig program (Rohlf
2010). Landmark precision was tested for digitiz-
ing error as described in Klenovšek and Kryštufek
(2013). The digitizing error was low.
Landmark coordinates were superimposed
using the generalized Procrustes analysis (GPA)
(Rohlf and Slice 1990) to standardize size and
remove the differences in landmark congura-
tions due to position and orientation. With GPA,
I obtained centroid sizes (CS) and Procrustes
coordinates for all skulls. The CS is a geometric
measure of size calculated as the square root of the
sum of squared distances between each landmark
and the centroid of the landmark conguration
Figure 1: Ventral side of the cranium of S. citellus with 22 landmarks divided into two hypothesized modules
(white dots ‒ viscerocranium, black dots ‒ neurocranium). For denitions of landmarks see Klenovšek
and Kryštufek 2013
Slika 1: Ventralna stran lobanje S. citellus z 22 oslonilnimi točkami razdeljenimi na dva hipotetična modula (bele
pike ‒ viscerokranij, črne pike ‒ nevrokranij). Za opis točk glej Klenovšek and Kryštufek 2013

62 Acta Biologica Slovenica, 57 (1), 2014
(Bookstein 1991). Procrustes coordinates are shape
variables containing the complete information on
shape variation after superimposition.
A previous study of the ventral side of the
skull of S. citellus (Klenovšek and Kryštufek
2013), performed on the same material, detected
sexual dimorphism (SD) in the size of the skull
in adults. Because there was no SD in the skull
shape, the sexes were pooled in the current study.
Population differences between Burgenland and
Banat were also signicant in skull size as well
as shape (for results see Klenovšek and Kryštufek
2013). Nevertheless, morphometric distances
between populations at different ages showed
that differences between juveniles and adults
exceeded the differences between populations
(ibid.). I therefore pooled juveniles and adults
from Burgenland and Banat.
To evaluate the hypothesis that in S. citellus the
anterior part of the ventral cranium (the upper jaw
bones with the palate or the viscerocranium) and the
posterior part (the skull base or the neurocranium)
are modules, the conguration of 22 landmarks
was divided into subsets of 10 (viscerocranium)
and 12 (neurocranium) landmarks (Fig. 1).
I compared the degree of covariation between
the hypothesized modules to alternative spatially
contiguous partitions with the same number of
landmarks as in the hypothesized modules (Klin-
genberg 2009).
The strength of association between the sets of
landmarks was estimated with the RV coefcient, a
multivariate generalization of the Pearson correla-
tion coefcient (Escouer 1973), which represented
the amount of covariation scaled by the amount
of variation within the two sets of variables. If the
two sets of variables are completely uncorrelated,
the RV coefcient takes the value of zero, and
the value of one, if the two sets of variables are
completely interdependent (Klingenberg 2009).
If the hypothesis of modularity holds, the RV
coefcient for the selected partition should be the
lowest value, or at least near the lower extreme
of the distribution of RV coefcients of all parti-
tions (Klingenberg 2009). I separately computed
RV coefcients for the hypothesized modules for
juveniles and adults. Because allometry can have
a major effect on detection of modularity (e. g.
Hallgrímsson et al. 2006, Klingenberg 2009), I
afterwards corrected the data for allometry and
repeated the analyses of modularity with the residu-
als from the multivariate regression of shape on
the centroid size for each age group. Because the
value of RV coefcient can depend on the sample
size as well as on the difference in sample size
between groups (Fruciano et al. 2013), I computed
the Escouer RV coefcients (Escouer, 1973)
for both age classes, and performed a permuta-
tion test of the null hypothesis of no difference in
the RV coefcient between juveniles and adults
(Fruciano et al. 2013).
Statistical analyses were performed using
the IBM SPSS Statistics (2008), and analyses
of morphological modularity with the MorphoJ
software (Klingenberg 2011) and RVComparison
1.0 (Fruciano et al. 2013).
Results
In juveniles, 294 (or 3.9 %) of the 7460 parti-
tions had a lower RV coefcient than the partition
into the hypothesized modules, and in adults, none
of the 7460 alternative partitions had a lower RV
coefcient (Fig. 2). Both age groups had similar
minimal RV values (juveniles: RV = 0.282, adults:
RV = 0.212). In adults, the RV coefcient for the
hypothesized subdivision was clearly in the lower
extreme of the distribution of RV coefcients,
which was consistent with the hypothesis that
viscerocranium and neurocranium of the S. citellus
ventral side of the skull are distinct modules. A
higher RV coefcient in juveniles (RV = 0.330)
for the hypothesized partition, and the percent of
partitions with lower RV coefcients, indicated that
juveniles, compared to adults, are characterized
by a lower level of modularity.
Regressions of shape variables onto CS showed
statistically signicant effect of size on shape
in both age groups. In juveniles, the allometry
accounted for 10.77 % of shape variation, and
in adults, for 4.82 %. After the correction for
allometry, the values of RV coefcients between
the viscero- and neurocranium were in both age
groups higher than before the correction (Fig. 3).
In both age groups, the range of the distribution
of RV coefcients for all alternative partitions
broadened mainly to the right side, towards the
higher values of RV coefcients. In juveniles, the
P-value slightly lowered.

63
Klenovšek: Skull modularity of S. citellus
Figure 2: Histograms of the RV coefcients of all spatially contiguous partitions of the ventral skull landmark
congurations for juveniles and adults. The arrows indicate the values of RV coefcients beween the
hypothesized modules (viscerocranium vs. neurocranium). The P value is the proportion of partitions
with lower RV than observed for the hypothesized modules
Slika 2: Histograma RV koecientov vseh prostorsko sosednjih delitev konguracije oslonilnih točk ventralne
strani lobanje za mlade in odrasle osebke. Puščice označujejo vrednosti RV koecientov med hipotetičnima
moduloma (viscerokranijem in nevrokranijem). P-vrednost je delež delitev z nižjim RV koecientom
kot tistim za hipotetična modula
Finally, the Escouer RV coefcients of juve-
niles and adults were 0.331 and 0.158, respectively.
Permutation test for the difference in modularity
between a priori dened groups disclosed that
the difference in RV coefcient between juveniles
and adults (0.173) was marginally signicant (P
= 0.050).
Discussion
The ventral skull is a complex structure that
can be divided into two functional components, the
neurocranium and the viscerocranium (Emerson
and Bramble 1993). Olson and Miller (1958) and
Roth (1996), on the other hand, discovered that
squirrels have a highly integrated skull without
clear subdivisions into subunits. Nevertheless, in
Figure 3: Histograms of the RV coefcients of all spatially contiguous partitions of the ventral skull landmark
congurations for juveniles and adults after the correction for allometry. The arrows indicate the values
of RV coefcients beween the hypothesized modules (viscerocranium vs. neurocranium). The P value
is the proportion of partitions with lower RV than observed for the hypothesized modules
Slika 3: Histograma RV koecientov vseh prostorsko sosednjih delitev konguracije oslonilnih točk ventralne
strani lobanje za mlade in odrasle osebke po izločitvi vpliva alometrije. Puščice označujejo vrednosti
RV koecientov med hipotetičnima moduloma (viscerokranijem in nevrokranijem). P-vrednost je delež
delitev z nižjim RV koecientom kot tistim za hipotetična modula

64 Acta Biologica Slovenica, 57 (1), 2014
the current study, the analysis of the covariation
among landmarks in the ventral cranium yielded
strong support to the hypothesis that in S. citellus
the viscerocranium and neurocranium are separate
modules. I also analyzed the strength of integration
between the hypothesized modules of the S. citel-
lus ventral cranium during ontogeny and observed
that the level of modularity was higher in adults
than in juveniles, i.e. the hypothesized modules in
the juvenile skull were more integrated with each
other. In adults, the partition to viscerocranium and
neurocranium had the lowest RV value from all
alterative partitions and therefore the lowest degree
of covariation. In juveniles, the RV of the hypoth-
esized modules was near the lower extreme of the
distribution of RV coefcients. Nevertheless, both
age groups had similar minimal RV values. Minimal
RV value applies minimal covariation between sets
of landmarks and was in adults congruent with the
partition to viscerocranium and neurocranium. In
juveniles, a pattern of modularity emerged that
did not match the subdivision into functional
modules. A further study might discover a model
of association between cranial traits different from
conventional modules. Allometry can be a strong
integrating factor (Klingenberg 2009), but it had
little inuence on the hypothesized modularity
of the S. citellus skull in juveniles as well as in
adults, probably because the amount of shape vari-
ation explained by allometry was low. Although
marginally signicant, observed difference in RV
coefcients between the two age classes also sug-
gests that in S. citellus the strength of association
between viscerocranium and neurocranium was
higher in juvenile than in adult skull.
The relationship between the developmen-
tal determinants of integration and phenotypic
covariance is very complicated, therefore the
integration cannot be studied through phenotypic
covariance patterns alone (Hallgrímsson et al.
2009). Nevertheless, it is commonly known that
functionally and developmentally related traits
are more integrated than traits that do not share
functional and developmental inuences (Leamy
et al. 1999, Willmore et al. 2006). Most studies of
morphological integration showed high covariation
patterns between functionally related traits (e. g.
Cheverud 1995, Marroig and Cheverud 2001,
Klingenberg et al. 2004, Ivanović and Kalezić
2010, Jojić et al. 2012). In skulls, modularity is
well studied and dened in primates (e. g. Marriog
and Cheverud 2001, González-José et al. 2008),
whereas in rodents the correlations between func-
tionally and developmentally related structures
are less consistent and obvious (Willmore et al.
2006, Klingenberg 2013). For instance, Monteiro
et al. (1999) discovered that the orofacial region
of the Thrichomys apereoides (Lund, 1839) skull
is less integrated than the basicranium. Willmore
et al. (2006), on the other hand, found out that the
facial regions of mice are weakly but signicantly
integrated, while no integration was found for the
cranial vault. Jojić et al. (2011), who used the
same methods as the present study, conrmed the
hypothesis of a face-basicranium organization of
the Apodemus avicollis (Melchior, 1834) skull.
Tests of integration and modularity can yield
mixed results also because of the differences in
methodology (Jojić et al. 2012, Klingenberg 2013).
Olson and Miller (1958) studied morphological
integration of the cranium and mandible of the fox
squirrel using linear measurements and methods
based on statistical correlation. Roth (1996) studied
the lateral view of the cranium of several squirrel
species using landmark based geometric morpho-
metric methods and preseneted only descriptive
preliminary results with no statistical analyses of
morphological integration. Therefore, it is possible
that support for the two-module organization of
the ventral cranium observed herein or for a highly
integrated sciurid skull reported in previous studies
(Olson and Miller 1958, Roth 1996) could depend
upon the methodology.
Because the shape of bones is inuenced by
the mechanical forces during ontogeny (Sun et al.
2004, Young and Badyaev 2007) and the bones
of the viscerocranium are directly involved in the
mechanics of feeding, they undergo prominent
shape transformations after weaning when juve-
niles change their diet form suckling to gnawing.
Based on the results, I suppose that modularity in
the S. citellus ventral cranium is largely driven
by masticatory forces that apply to the bones of
the viscerocranium and form a functional module
that is in weak covariation with the neurocranium,
which grows relatively slow during postnatal on-
togeny (Herring 1993, Monteiro et al. 1999). For
a better understanding of modularity, variability
and development of the skull of S. citellus more
research is needed, ideally on bigger samples and

65
Klenovšek: Skull modularity of S. citellus
postnatal longitudinal data including the study of
uctuating asymmetry, which can be a useful tool
for determination of the boundaries of develop-
mental modules (Klingenberg and Zaklan 2000,
Klingenberg et al. 2001).
Conclusions
1. Unlike previous studies of sciurid skulls, the
analysis of the covariation among landmarks in
the ventral cranium yielded strong support to the
hypothesis that in S. citellus the viscerocranium
and neurocranium are separate modules.
2. The level of modularity was higher in adults
than in juveniles, i.e. the hypothesized modules
in the juvenile skull were more integrated with
each other.
3. Although allometry can be a strong inte-
grating factor, it had very little inuence on the
hypothesized modularity of the S. citellus skull.
4. The change in the strength of integration
between the viscerocranium and neurocranium in
the S. citellus skull during ontogeny is probably
a consequence of the change in diet, from liquid
to solid food.
Povzetek
Članek obravnava morfološko integracijo in
modularnost lobanje evropske tekunice Spermo-
philus citellus in primerja stopnjo modularnosti
med mladimi in odraslimi osebki vrste. Moduli
so notranje tesno povezani deli strukture, ki so
med seboj relativno neodvisni. Integracijo in
modularnost morfoloških struktur analiziramo
z ugotavljanjem stopnje povezanosti med posa-
meznimi deli strukture. Ventralno stran lobanje
lahko razdelimo v dve funkcionalni podenoti;
nevrokranij, ki ga sestavljajo kosti možganskega
dela lobanje, oči in ušes, in viscerokranij, ki je
iz kosti čeljustnega aparata. Cilj raziskave je bil
preveriti prisotnost te delitve na ventralni strani
lobanje evropske tekunice, saj je za lobanje veveric
znano, da imajo visoko stopnjo integracije brez
jasnih delitev na podenote (Olson and Miller 1958,
Roth 1996). Testirala sem 159 lobanj, od tega 43
mladih in 116 odraslih osebkov. Obliko ventralne
strani lobanje sem opisala z 22 oslonilnimi točkami.
Konfiguracije koordinat oslonilnih točk vseh
lobanj sem poravnala s posplošeno Procrustovo
analizo (GPA), ki konguracije koordinat pre-
makne, zavrti in skalira, tako da je vsota kvadra-
tov razlik na koncu v vzorcu minimalna. Nato
sem z delitvijo oslonilnih točk na viscerokranij
in nevrokranij postavila hipotezo o modularni
zgradbi lobanje. Hipotezo testiramo s izračunom
stopnje kovariabilnosti med hipotetičnimi moduli,
ki jo nato primerjamo z vsemi alternativnimi
delitvami točk na podskupine z enakim številom
točk, kot jih imajo hipotetični moduli. Hipoteza
o modularnosti drži, če je kovariabilnost med
hipotetičnimi moduli izrazito nižja kot med vsemi
ostalimi alternativnimi delitvami točk na module.
Stopnjo integracije med moduli sem ovrednotila z
metodami, ki temeljijo na RV koecientu. Primer-
jala sem tudi stopnjo modularnosti in integracije
med mladimi in odraslimi osebki. Tako pri mladih
kot odraslih osebkih so rezultati podprli hipotezo
o delitvi lobanje S. citellus na viscerokranij in
nevrokranij, le da je bila lobanja mladih osebkov
izraziteje integrirana. Razliko v stopnji integracije
in modularnosti lobanje med mladimi in odraslimi
osebki je potrdil tudi permutacijski test. Čeprav
je alometrija lahko močan integracijski faktor, je
imela majhen vpliv na vzorec modularnosti in
integracije. Sprememba v stopnji modularnosti
pri lobanji evropske tekunice med ontogenetskim
razvojem je najverjetneje posledica spremembe v
prehrani, t. j. prehoda s tekoče na trdo hrano, in
večje specializacije obeh funkcionalnih modulov.
Acknowledgements
I thank Vida Jojić and Franc Janžekovič for
most helpful comments that signicantly improved
the manuscript.
References
Bookstein, F.L., 1991. Morphometric Tools for Landmark Data: Geometry and Biology. Cambridge
University Press, Cambridge, 435 pp.

66 Acta Biologica Slovenica, 57 (1), 2014
Cardini, A., Tongiorgi, P., 2003. Yellow-bellied marmots (Marmota aviventris) ‘in the shape space’
(Rodentia, Sciuridae): sexual dimorphism, growth and allometry of the mandible. Zoomorphology,
122, 11-23.
Cheverud, J.M., 1995. Morphological integration in the saddleback tamarin (Saguinus fuscicollis)
cranium. American Naturalist, 145, 63-89.
Drake, A.G., Klingenberg, C.P., 2010. Large-scale diversication of skull shape in domestic dogs:
disparity and modularity. American Naturalist, 175, 289-301.
Emerson, S.B., Bramble, D.M., 1993. Scaling, allometry and skull design. In: Hanken, J., Hall, B.K.
(eds.): The Skull. The University of Chicago Press, Chicago, pp. 384-416.
Escouer, Y., 1973. Le traitement des variables vectorielles. Biometrics, 29, 751-760.
Fruciano, C., Franchini, P., Meyer, A., 2013. Resampling-based approaches to study variation in
morphological modularity. PLoS ONE 8(7), e69376.
Gonzalez, P.N., Hallgrímsson, B., Oyhenart, E.E., 2011. Developmental plasticity in covariance
structure of the skull: effects of prenatal stress. Journal of Anatomy, 218, 243-257.
González-José, R., Escapa, I., Neves, W.A., Cúneo, R., Pucciarelli, H.M., 2008. Cladistic analysis
of continuous modularized traits provides phylogenetic signals in Homo evolution. Nature, 453,
775-778.
Goswami, A., Polly, P.D., 2010. Methods for studying morphological integration and modularity. In:
Alroy, J., Hunt, G. (eds.): Quantitative Methods in Paleobiology, Paleontological Society Special
Publications, 16. Yale University Printing and Publishing Services, New Haven, CT, pp. 213-243.
Hallgrímsson, B., Brown, J.J.Y., Ford-Hutchinson, A.F., Sheets, H.D., Zelditch, M.L., Jirik, F.R., 2006.
The brachymorph mouse and the developmental-genetic basis for canalization and morphological
integration. Evolution and Development, 8, 61-73.
Hallgrímsson, B., Jamniczky, H., Young, N.M., Rolian, C., Parsons, T.E., Boughner, J.C., Marcucio,
R.S., 2009. Deciphering the palimpsest: studying the relationship between morphological inte-
gration and phenotypic covariation. Evolutionary Biology, 36, 355-376.
Herring, S.W., 1993. Epigenetic and functional inuences on skull growth. In: Hanken, J., Hall, B.K.
(eds.): The Skull. The University of Chicago Press, Chicago, pp. 237-271.
Ivanović, A., Kalezić, M.L., 2010. Testing the hypothesis of morphological integration on a skull of
a vertebrate with a biphasic life cycle: a case study of the alpine newt. Journal of Experimental
Zoology Part B Molecular and Developmental Evolution, 314B, 527-538.
Jojić, V., Blagojević. J., Ivanović, A., Bugarski-Stanojević, V., Vujošević, M., 2007. Morphological
integration of the mandible in yellow-necked eld mice: the effects of B chromosmes. Journal of
Mammalogy, 88, 689-695.
Jojić, V., Blagojević, J., Vujošević, M., 2011. B chromosomes and cranial variability in yellow-necked
eld mice (Apodemus avicollis). Journal of Mammalogy, 92, 396-406.
Jojić V., Blagojević J., Vujošević M., 2012. Two-module organization of the mandible in the yellow-
necked mouse: a comparison between two different morphometric approaches. Journal of Evo-
lutionary Biology, 25, 2489-2500.
Klenovšek, T., Kryštufek, B., 2013. An ontogenetic perspective on the study of sexual dimorphism,
phylogenetic variability, and allometry of the skull of European ground squirrel, Spermophilus
citellus (Linnaeus, 1766). Zoomorphology, 132(4), 433-445.
Klingenberg, C.P., 2008. Morphological integration and developmental modularity. Annual Review
of Ecology, Evolution and Systematics, 39, 115-132.
Klingenberg, C.P., 2009. Morphometric integration and modularity in congurations of landmarks:
Tools for evaluating a-priori hypotheses. Evolution and Development, 11, 405-421.
Klingenberg, C.P., 2011. MorphoJ: an integrated software package for geometric morphometrics.
Molecular Ecology Resources, 11, 353-357.
Klingenberg, C.P., 2013. Cranial integration and modularity: insights into evolution and development
from morphometric data. Hystrix, 24, 43-58.

67
Klenovšek: Skull modularity of S. citellus
Klingenberg, C.P., Badyaev, A.V., Sowry, S.M., Beckwith, N.J. 2001. Inferring developmental modular-
ity from morphological integration: analysis of individual variation and asymmetry in bumblebee
wings. American Naturalist, 157, 11-23.
Klingenberg, C.P., Leamy, L.J., 2001. Quantitative genetics of geometric shape in the mouse mandible.
Evolution, 55, 2342-2352.
Klingenberg, C.P., Leamy, L.J., Cheverud, J.M., 2004. Integration and modularity of quantitative trait
locus effects on geometric shape in the mouse mandible. Genetics, 166, 1909-1921.
Klingenberg, C.P., Zaklan, S.D., 2000. Morphological integration between developmental com-
partments in the Drosophila wing. Evolution 54: 1273-1285.
Leamy, L.J., Routman, E.J., Cheverud, J.M., 1999. Quantitative trait loci for early- and late developing
skull characters in mice: a test of the genetic independence model of morphological integration.
American Naturalist, 153, 201-214.
Marroig, G., Cheverud, J.M., 2001. A comparison of phenotypic variation and covariation patterns and
the role of phylogeny, ecology, and ontogeny during cranial evolution of New World monkeys.
Evolution, 55, 2576-2600.
Monteiro, L.R., Lessa, L.G., Abe, A.S., 1999. Ontogenetic Variation in Skull Shape of Thrichomys
apereoides (Rodentia: Echimyidae). Journal of Mammalogy, 80, 102-111.
Olson, E.C., Miller, R.L. 1958: Morphological Integration. University of Chicago Press, Chicago,
376 pp.
Ramos-Lara, N., Koprowski, J.L., Kryštufek, B., Hoffmann, I.E., in press. Spermophilus citellus
(Rodentia: Sciuridae). Mammalian Species.
Rohlf, F.J., 2010. TpsDig2, Version 2.16. Ecology & Evolution. SUNY at Stony Brook.
Rohlf, F.J., Slice, D.E., 1990. Extensions of the Procrustes method for the optimal superimposition
of landmarks. Systematic Zoology, 39, 40-59.
Roth, V.L., 1996. Cranial Integration in the Sciuridae. American Zoologist, 36, 14-23.
Ružić, A., 1966. Određivanje uzrasnih kategorija u populaciji tekunice Citellus citellus. Arhiv bioloških
nauka, XVIII, 1/1966, Beograd.
SPSS Statistics, 2008. Version 17.0. IBM Corporation.
Sun, Z., Lee, E., Herring, S.W., 2004. Cranial sutures and bones: growth and fusion in relation to
masticatory strain. Anatomical Record Part A, 276A, 150-161.
Swiderski, D.L., Zelditch, M.L., 2010. Morphological diversity despite isometric scaling of lever
arms. Evolutionary Biology, 37, 1-18.
Willmore, K.E., Leamy, L., Hallgrímsson, B., 2006. Effects of developmental and functional interactions
on mouse cranial variability through late ontogeny. Evolution and Development, 8(6), 550-567.
Young, R.L., Badyaev, A.V., 2007. Evolution of ontogeny: linking epigenetic remodeling and genetic
adaptation in skeletal structures. Integrative and Comparative Biology, 47, 234-244.
Zelditch, M.L., Mezey, J.G., Sheets, H.D., Lundrigan, B.L., Garland, T.Jr., 2006. Developmental
regulation of skull morphology II: ontogenetic dynamics of covariance. Evolution and Deve-
lopment, 8, 46-60.