This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and

Transcription

1 This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier s archiving and manuscript policies are encouraged to visit:

2 Zoology 115 (2012) Contents lists available at SciVerse ScienceDirect Zoology journa l h o me pa g e: www. els evier.com/locate/zool Variation in the cranium shape of wall lizards (Podarcis spp.): effects of phylogenetic constraints, allometric constraints and ecology Aleksandar Urošević a,, Katarina Ljubisavljević a, Dušan Jelić b, Ana Ivanović c a Department of Evolutionary Biology, Institute for Biological Research Siniša Stanković, University of Belgrade, Bulevar Despota Stefana 142, Belgrade, Serbia b State Institute for Nature Protection, Trg Mažuranića 5, Zagreb, Croatia c Institute of Zoology, Faculty of Biology, University of Belgrade, Studentski trg 16, Belgrade, Serbia a r t i c l e i n f o Article history: Received 9 September 2011 Received in revised form 26 December 2011 Accepted 17 January 2012 Keywords: Allometry Cranium shape Geometric morphometrics Lacertid lizards Morphological variation a b s t r a c t We used geometric morphometrics to explore the influence of phylogenetic and allometric constraints as well as ecology on variation in cranium shape in five species of monophyletic, morphologically similar Podarcis lizards (Podarcis erhardii, Podarcis melisellensis, Podarcis muralis, Podarcis sicula and Podarcis taurica). These species belong to different clades, they differ in their habitat preferences and can be classified into two distinct morphotypes: saxicolous and terrestrial. We found (i) no phylogenetic signal in cranium shape, (ii) diverging allometric slopes among species, and (iii) a significant effect of habitat on cranium shape. The saxicolous species (P. erhardii and P. muralis) had crania with elongated parietals, elongated cranium bases, shortened anterior parts of the dorsal cranium, reduced chambers of the jaw adductor muscles and larger subocular foramina. These cranial features are adaptations that compensate for a flattened cranium, dwelling on vertical surfaces and seeking refuge in crevices. The crania of the terrestrial species (P. melisellensis, P. sicula and P. taurica) tended to be more elongate and robust, with enlarged chambers of the jaw adductor muscle, reduced skull bases and shortened parietals. Terrestrial species exhibited more variation in cranium shape than saxicolous species. Our study suggests that shape variation in Podarcis sp. lizards is largely influenced by ecology, which likely affects species-specific patterns of static allometry Elsevier GmbH. All rights reserved. 1. Introduction Understanding the relationships among form, function, phylogeny and physical constraints is of great importance to evolutionary biology. Linking morphological variation with ecological factors has become a widespread approach in addressing this issue in recent decades (Vanhooydonck and Van Damme, 1999; Aerts et al., 2000; Irschick et al., 2005). However, the association between form and function may be difficult to detect due to physical and phylogenetic constraints, trade-offs or pronounced differences in the rates of evolutionary change in different structures and taxa. Lacertid lizards are suitable candidates for investigating the phylogenetic, ecological and mechanistic aspects of morphological variation for several reasons. First, most lacertids have a uniform cranial skeleton (Arnold, 1973, 1989; Arnold et al., 2007). Second, in general they are food and habitat generalists. Third, there is considerable overlap in the distributions of body size, prey size and home ranges among species (Arnold, 1987; Herrel Corresponding author. addresses: aurosevic@ibiss.bg.ac.rs, acaurosevic@gmail.com (A. Urošević). et al., 2001). Fourth, studies of functional morphology showed that differences in bite force are correlated with cranial morphology and, therefore, even small morphological differences in cranium size and shape can have important functional and ecological implications (Herrel et al., 1999, 2001, 2007; Verwaijen et al., 2002; Lailvaux et al., 2004). Recent studies of the lacertid cranium have found that allometric effects explain much of the interspecific variation in cranium shape and that functional constraints (specifically, differences in bite force) explain patterns of change in the dorsal and ventral cranium (Costantini et al., 2010; Ljubisavljević et al., 2010a, 2011). Kaliontzopoulou et al. (2010) suggested that ecomorphological adaptations to different habitat types could evolve very quickly in the Podarcis lacertid clade. Lacertid lizards of the genus Podarcis (wall lizards) comprise a well-supported lineage (Harris and Arnold, 1999; Fu, 1998, 2000; Arnold et al., 2007) and exhibit substantial ecomorphological variation. The genus Podarcis is the most abundant group of reptiles in southern Europe, with currently recognised species (Harris and Arnold, 1999; Sá-Sousa and Harris, 2002; Lymberakis et al., 2008). They share very similar morphologies, and phylogenetic analyses based solely on the morphological characters are often conflicting (Arnold, 1973, 1989). Recent phylogenetic studies based on molecular data have revealed cases of hidden diversity /$ see front matter 2012 Elsevier GmbH. All rights reserved.

3 208 A. Urošević et al. / Zoology 115 (2012) and paraphyly within the genus but the phylogenetic relationships between the clades still remain unresolved, with conflicting results in different studies (Castilla et al., 1998; Harris and Sá-Sousa, 2001; Poulakakis et al., 2003, 2005a,b; Harris et al., 2005; Podnar et al., 2005, 2007). We studied populations of five Podarcis species from the central and western Balkan Peninsula Podarcis erhardii, Podarcis melisellensis, Podarcis muralis, Podarcis sicula and Podarcis taurica. Three of these species (P. erhardii, P. melisellensis and P. taurica) belong to the Balkan Peninsula group (Harris and Arnold, 1999) although the relationship of P. erhardii (or the closely related Podarcis peloponnesiaca) to the remaining two species remains somewhat unclear. The association of P. erhardii or P. peloponnesiaca to the other Podarcis species varies in different studies (Harris and Arnold, 1999; Poulakakis et al., 2003, 2005a,b; Harris et al., 2005; Arnold et al., 2007). P. muralis and P. sicula clades were considered sister groups (Harris and Arnold, 1999; Poulakakis et al., 2005b). However, new analyses revealed the numt-sic pseudogene in the nuclear DNA of P. sicula which is very similar to the P. muralis mitochondrial DNA cyt b sequences. That pseudogene could have created the likely wrong pattern of association between P. sicula and P. muralis (Harris et al., 2005; Podnar et al., 2005, 2007). These five species exhibit distinct habitat preferences and can be classified into two main ecomorphotypes: the saxicolous type (P. erhardii and P. muralis), which prefers vertical, rocky habitats; and the terrestrial type (P. melisellensis, P. sicula and P. taurica), a ground-dweller in vegetated habitat (Arnold, 1987; Vanhooydonck and Van Damme, 1999). Morphologically, saxicolous species tend to be dorsoventrally compressed, with flattened bodies and heads, while the ground-dwelling species are laterally compressed, with more cylindrical bodies and more robust heads (Arnold, 1987; Vanhooydonck and Van Damme, 1999). In the present study, our aim was to analyse cranium shape variation among wall lizard species via geometric morphometrics in order to (i) identify possible constraints of shared evolutionary history (phylogenetic signal) on cranium shape, (ii) identify possible developmental constraints (allometry) on cranium shape among Podarcis species, and (iii) test for an influence of preferred habitat type on the evolution of cranium shape in this group. 2. Materials and methods 2.1. Samples All Podarcis individuals sampled were adults. Specimens were initially collected for use in other studies and were either deposited in the herpetological collections of the Institute for Biological Research, Siniša Stanković, Belgrade (P. erhardii, P. melisellensis, P. muralis and P. taurica) or the State Institute for Nature Protection of Croatia, Zagreb (P. sicula). Sample locations and sizes for each species are given in Table 1. The sexual maturity of specimens was determined through analysis of the reproductive characters of Podarcis species as described previously (Bejaković et al., 1996; Ljubisavljević et al., 2010b) Cranium preparation and landmark configuration Crania were cleared with trypsin and potassium hydroxide (KOH), stained with alizarin red S (to better distinguish cranial elements and their articulations; see Dingerkus and Uhler, 1977) and preserved in glycerol. Images of the dorsal and ventral cranium were obtained with a Sony DSC-F828 digital camera (resolution 8.0 MP; Sony Corp., Tokyo, Japan). Each cranium was submerged in glycerol with the parietal (dorsal cranium) or palate (ventral cranium) views positioned parallel to the photographic plane in the Fig. 1. Landmarks digitised on the (A) dorsal and (B) ventral cranium of a Podarcis taurica male specimen. centre of the optical field to reduce and equalise distortion. The camera was set 3 cm from the cranium. The camera set-up and the placement of each specimen relative to the camera lens was kept constant to minimise image aberration due to parallax (image distortion resulting from placing the camera too close to the specimen) and to reduce error in the subsequent geometric morphometric analysis (Mullin and Taylor, 2002). We used TpsDig software (Rohlf, 2005) to digitise 14 twodimensional landmarks on the dorsal cranium and 18 landmarks on the ventral cranium. Landmarks were digitised (by A.U.) on the right side of each specimen to avoid redundant information from symmetric structures. The landmarks were chosen based on their presence in all specimens and their reliability in providing an adequate summary of the specific aspects of cranial morphology (Fig. 1 and Table 2). The chosen landmark configurations have also been employed in previous studies of cranial size and shape variation in lacertid lizards (Ljubisavljević et al., 2010a, 2011). The landmarks represent contact points between bones, tips of processes, or the point of maximum curvature of structures (Bookstein, 1991) Statistical analyses A generalised Procrustes analysis (GPA) was applied to obtain a matrix of Procrustes coordinates (shape variables) from which differences due to position, scale and orientation were removed (Rohlf and Slice, 1990; Bookstein, 1996; Dryden and Mardia, 1998). Procrustes coordinates can be used as input variables in any conventional statistical analysis (Zelditch et al., 2004). As a measure of size, we used centroid size (CS) (calculated as the square root of the summed squared distances of each landmark from the centroid of the form; Bookstein, 1991), which is uncorrelated with shape in the absence of allometry (Bookstein, 1991). Significant sexual dimorphism in cranium size and shape characterises lacertid lizards and lizards of the genus Podarcis in particular (Herrel et al., 1999, 2001; Verwaijen et al., 2002;

4 A. Urošević et al. / Zoology 115 (2012) Table 1 Species studied, sampling localities and sample sizes. Species State Sampling locality No. No. P. erhardii Serbia Pčinja P. melisellensis Montenegro Skadar Lake: Mali Moračnik island P. muralis Serbia Belgrade P. sicula Croatia Krk island and Zagreb P. taurica Serbia Deliblato sands Table 2 Number and description of dorsal and ventral cranial landmarks. Number Description Dorsal cranium Ventral cranium 1 Tip of premaxilla (tip of the snout) Tip of premaxilla (tip of the snout) 2 Suture between premaxilla and nasals Suture between premaxilla and maxilla 3 Lateralmost point of nasal Suture between vomer and palatine 4 Suture between nasal, frontal and maxilla Anteriormost point of subocular foramen 5 Suture between maxilla, prefrontal and frontal Anteriormost point of ectopterygoid 6 Anteriormost point of postorbital Posterior tip of maxilla 7 Anteriormost point of postfrontal Lateralmost point of cranium 8 Posteriormost point of prefrontal Posteriormost point of subocular foramen 9 Suture between both frontals and parietal Suture between pterygoid and palatine 10 Suture between frontal, postfrontal and parietal Posterior tip of jugal 11 Posteriormost point of squamosal Anterior tip of basipterygoid process 12 Posterior tip of supratemporal process of parietal Posterior tip of basipterygoid process 13 Posterior tip of exoccipital Anteriormost point of quadrate 14 Posteriormost point on the curve of the occipital condyle Lateralmost point of quadrate 15 Posterior tip of pterygoid process 16 Posterior point of quadrate 17 Posterior tip of otooccipital 18 Posteriormost point on the curve of the occipital condyle Ljubisavljević et al., 2010a). Therefore, to analyse cranium shape variation we chose a conservative approach and all analyses were performed on females and males separately. To calculate variation in the size of the dorsal and ventral cranium we employed ANOVA procedures, with CS as the dependent variable and with habitat and species nested within habitat as factors. To explore variation in the shape of the dorsal and ventral crania, we employed MANOVA with Procrustes coordinates as dependent variables and habitat and the species nested within those habitats as factors. Tukey s HSD (honestly significant difference) post hoc test was performed to test for differences in CS between the species. We used the TwoGroup6 program, IMP series (Sheets, 2000) to calculate Procrustes distances between different species mean shapes and performed Goodall s F-test to estimate the statistical significance of differences between the mean shapes. To explore the variation in cranium shape among species, we conducted a principal component analysis (PCA) of the covariance matrix of shape variables, computed for each sex separately, for all five species. The PCA was performed with the MorphoJ software package (Klingenberg, 2011). To test for phylogenetic signal, we used two alternative phylogenies. One was the phylogeny of the Balkan peninsula Podarcis spp. clades published by Poulakakis et al. (2005b), which includes all five species investigated in the present study. Poulakakis et al. (2005b) inferred phylogenetic relationships from mtdna cyt b and 16S genes using methods of maximum likelihood and Bayesian inference. Since the association between P. sicula and P. muralis proposed by Poulakakis et al. (2005b) is not concordant with the more recent studies (Harris et al., 2005; Podnar et al., 2007), we also used the phylogeny proposed by Harris et al. (2005). However, two of the species from our study were not included in Harris et al. (2005). Therefore, we produced an alternative tree based on the following assumptions: (i) P. melisellensis belongs to the P. taurica P. milensis P. gaigeae group, which is strongly supported by the studies of Poulakakis et al. (2005a,b); (ii) P. erhardii is closely related to P. peloponnesiaca, which is indicated by morphological data (Arnold, 1973), proteine electrophoresis (Lutz and Mayer, 1985) and the data inferred from mitochondrial DNA, which confirms that P. erhardii is paraphyletic with regard to P. peloponnesiaca (Poulakakis et al., 2003, 2005b; Lymberakis et al., 2008). The tree topologies are given in Fig. 2. Fig. 2. Two alternative phylogenetic trees of five Podarcis species, inferred from molecular data. (A) Taken and adapted from Poulakakis et al. (2005b); (B) taken and adapted from Harris et al. (2005), modified according to Arnold (1973), Lutz and Mayer (1985), and Poulakakis et al. (2003, 2005a,b).

5 210 A. Urošević et al. / Zoology 115 (2012) Table 3 The analysis of variance in skull size between habitats and species obtained by one-way ANOVA. Sex Cranium side Factor ANOVA SS F df P Dorsal Habitat Species (habitat) < Ventral Habitat Species (habitat) < Dorsal Habitat < Species (habitat) < Ventral Habitat < Species (habitat) < We applied a recently developed procedure to test for the presence of phylogenetic signal in the morphometric data (Klingenberg and Gidaszewski, 2010). The criteria of squared-change parsimony were used to reconstruct the values of the internal nodes of the phylogeny from the shape averages of the terminal taxa (Maddison, 1991; McArdle and Rodrigo, 1994; Rohlf, 2001, 2002). We then tested for phylogenetic signal in the shape data following Klingenberg and Gidaszewski (2010). This test uses a permutation approach to simulate the null hypothesis of the complete absence of phylogenetic structure by randomly reassigning the phenotypic data (shape configurations) to the terminal nodes of the phylogeny (Klingenberg and Gidaszewski, 2010). We performed the permutation tests for both tree topologies using unweighted squared-change parsimony. The phylogenetic signal in cranium shape was quantified by using MorphoJ (Klingenberg, 2011). To assess the effect of size on shape changes of the dorsal and ventral cranium among species, we evaluated the similarity of static allometric trajectories. Static allometry denotes size-related shape changes measured in different individuals at the same developmental stage within a population or species (Klingenberg, 1998; Shingleton, 2010). The homogeneity of static allometric slopes on the interspecific level was evaluated by employing a multivariate analysis of covariance (MANCOVA) approach with shape variables as dependent variables, species as factor and CS as covariate for each sex separately. Significant species CS interaction would indicate that size-dependent shape changes diverge between species. To test if the allometric slopes differ with regard to habitat preference, we employed a multivariate analysis of covariance (MANCOVA) with shape variables as dependent variables, habitat code as factor and CS as covariate, for each sex separately. Significant habitat CS interaction would indicate that size-dependent shape changes differ between species with different habitat preference. Multivariate linear regression was used to assess the effects of ecological structuring of variables, to estimate the amount of cranium size and shape variation attributable to habitat preference and to visualise shape changes related to habitat preference. This method is commonly used in biogeographical studies to partition the effects of geography and environment (Legendre and Legendre, 1998; Ruggiero and Kitzberger, 2004; Botes et al., 2006; Cardini et al., 2007; Cardini and Elton, 2009). Habitat preference (saxicolous or terrestrial) was numerically coded and used as an independent variable. To obtain the percentage of variance attributable to habitat preference and to visualise shape changes we used TpsRegr (Rohlf, 2009) and MorphoJ (Klingenberg, 2011). Standard statistical procedures were performed using the SAS statistical package (SAS Institute Inc., Cary, NC, USA). 3. Results 3.1. Variation in cranium size and shape The differences in cranium size related to habitat were statistically significant in males, but not in females (Table 3). The significant variations in ventral cranium shape between habitats and between the species within habitats indicate that patterns of shape change vary between species (Table 4). Pairwise comparisons of cranium size and shape differences between species are given in supplementary tables (see Table S1 in Appendix A for the dorsal cranium and Table S2 for the ventral cranium). We performed a PCA to further explore variation in cranium shape between species, for each sex separately. The patterns of shape changes for females and males were fairly consistent for each cranium side. For the dorsal cranium of females, the first two principal components (PC) accounted for 41.9% of the total variation in shape. PC1 described a gradient from P. melisellensis and P. sicula, which had crania with shortened, posteriorly widened parietals, a slightly elongated midface and rostrum and reduced occipital bones, to P. erhardii and P. muralis females, which had elongated parietals, enlarged occipitals and a slightly shortened rostrum. PC2 described a gradient from narrow and elongate crania to short, wide crania with enlarged orbits. The two ecomorphs, saxicolous (P. erhardii and P. muralis) and terrestrial (P. melisellensis, P. sicula and P. taurica) were separated along PC1 with a slight overlap. PC2 separates a P. muralis + P. sicula group with more elongated crania from P. erhardii, P. melisellensis and P. taurica which are more robust (Fig. 3A). For the dorsal cranium of males, the first two PC accounted for 39.1% of the total variation in shape among species. The shape changes along PC1 spanned a gradient from P. sicula and P. taurica, which had anteriorly shortened and posteriorly widened parietals, Table 4 Multivariate analysis of variance (MANOVA) for dorsal and ventral crania shape variables. Sex Cranium side Factor Wilks F df P Dorsal Habitat < Species (habitat) < Ventral Habitat < Species (habitat) < Dorsal Habitat < Species (habitat) < Ventral Habitat < Species (habitat) <0.0001

6 A. Urošević et al. / Zoology 115 (2012) Fig. 3. Ordination of (A) the female and (B) the male specimens of five Podarcis species in the space of the first two principal axes. Thin-plate spline deformation grids illustrate cranial shape changes correlated with PC1 and PC2. an elongated, narrow midface and rostrum and reduced occipitals, to P. muralis and P. erhardii, which had anteriorly elongated and posteriorly narrowed parietals, enlarged occipitals and orbits and a shortened midface. PC2 described a gradient from P. sicula, which had a narrow and elongated cranium, to P. taurica with a short, wide cranium with enlarged orbits. Along the second axis, there was a separation between P. sicula on the one hand and the cluster P. erhardii + P. melisellensi + P. taurica on the other hand, with P. muralis being intermediate (Fig. 3B). For the ventral cranium of females, the first two axes accounted for 41.2% of the total variation in shape among species. PC1 described a gradient from P. melisellensis, which had enlarged chambers of the jaw adductor muscle (described by landmarks 6, 7, 8, 10, 11, 12, 14, 15), shortened and widened jugals, laterally positioned quadrates, a reduced cranium base and small subocular foramina, to P. muralis females, which have narrow jaw adductor muscle chambers, elongated and narrow jugals, quadrates positioned medially, an elongated cranium base and enlarged subocular foramina. PC2 described shape changes from P. melisellensis to P. sicula, involving a narrowing and elongation of the cranium, a shortening of the quadrates and a reduction of the cranium base. P. melisellensis was very well separated along PC1, and the remaining species clustered together, with P. sicula separating along PC2 (Fig. 4A). For the ventral cranium of males, both axes described 48.3% of the total variation in ventral cranium shape. PC1 described shape changes from P. melisellensis and P. sicula to P. muralis and P. erhardii. Changes involved a narrowing of the jaw adductor muscle chambers, elongation and narrowing of the jugal bones, medially positioned quadrates, enlarged subocular foramina and an enlarged cranium base. PC2 described shape changes between P. melisellensis and P. sicula involving a narrowing of the cranium, shortening of the quadrates, narrowing and elongation of the jaw adductor muscle chambers, medially positioned jugals and reduction of the

7 212 A. Urošević et al. / Zoology 115 (2012) Fig. 4. Ordination of (A) the female and (B) the male ventral crania of five Podarcis species in the space of the first two principal axes. Thin-plate spline deformation grids illustrate cranial shape changes correlated with PC1 and PC2. cranium base. P. melisellensis and P. sicula were separated from the other species along PC1 and from each other along PC2 (Fig. 4B) Effect of phylogeny We found no evidence of phylogenetic signal in the dorsal and ventral cranium shape, for neither of the phylogenies used. Results of the permutation tests are given in Table Static allometry In females, the CS species interaction was insignificant for both the dorsal and the ventral cranium, which indicates that allometric slopes in females are homogenous among species. However, the CS species interaction was significant for both cranium sides of males, which indicates that for males, allometric trajectories significantly diverge (Table 6). Table 5 Test for phylogenetic signal in cranium shape. Sex Dorsal cranium Ventral cranium Phylogeny Phylogeny

8 A. Urošević et al. / Zoology 115 (2012) Table 6 Allometric shape changes among species, tested by multivariate analysis of covariance (MANCOVA). df, degrees of freedom; CS, centroid size. The significant CS species interaction indicates that allometric slopes diverge among species. Sex Cranium side Effect Wilks F Effect df Error df P Dorsal Species CS CS species Ventral Species CS CS species Dorsal Species CS CS species Ventral Species CS CS species Table 7 Effect of habitat preference on size-dependent shape changes, tested by multivariate analysis of covariance (MANCOVA). df, degrees of freedom; CS, centroid size. The significant CS habitat interaction indicates that allometric slopes differ with habitat preference. Sex Cranium side Effect Wilks F Effect df Error df P Dorsal Habitat CS CS habitat Ventral Habitat CS CS habitat Dorsal Habitat CS CS habitat Ventral Habitat CS CS habitat Table 8 Percentage and significance of shape variance explained by habitat type, obtained by nonparametric multivariate regression of shape variables on habitat code. Cranium side Sex Wilks F df P % explained Goodall s F Dorsal Ventral Cranium shape and habitat preference The CS habitat interaction was significant for both sexes, for the dorsal and ventral cranium alike, which indicates that allometric slopes of both females and males differ in relation to habitat (Table 7). The results of the multivariate regression used to quantify the ecological component of size and shape variation are provided in Table 8. There was a significant effect of habitat preference on cranial shape. The habitat preference had the largest effect on ventral cranium shape in males. In females, changes in shape of the dorsal cranium from saxicolous to terrestrial species involved reduction of the occipitals, shortening of the parietals by moving the fronto-parietal suture posteriorly, posterior widening of the parietals, movement of the postorbital posteriorly and a slight elongation of the midface (Fig. 5). In males, shape changes were similar to those observed in females, yet were more pronounced in the occipital, midface and posterior parietal regions, and less pronounced along the frontal parietal suture. These shape changes for both females and males were similar to the among-species changes in dorsal cranium shape described by PC1. In females, the changes from saxicolous to terrestrial species in the shape of the ventral cranium involved enlargement of the jaw adductor muscle chambers, shortening and widening of the jugals, reduction of the occipitals and narrowing of the subocular foramina. The analogous changes in males were similar, only more pronounced (Fig. 6). For both males and females, changes related to habitat preference were consistent with the between-species variation in ventral cranium shape described by PC1. 4. Discussion 4.1. General remarks Phylogenetic signal in phenotypic traits within a monophyletic group is usually expected, as their shared evolutionary history and presumably shared developmental constraints lead to phenotypic similarity. On the other hand, adaptive radiation and morphological divergence are usually attributed to ecological divergence (Losos, 2010). When interspecific variation in ecological demands correlates with variation in morphological traits, it is very likely that the environment, through constraints and/or selection, has affected the variation of phenotypic traits (Hansen and Martins, 1996; Revell et al., 2008). The evolution of the lizard cranium shows different patterns of divergence and disparity between clades and the phylogeny imposes a primary signal upon which an ecological signal is imprinted (Stayton, 2005). Our analyses of variation in cranial shape in five Podarcis species and the position of their cranium shapes in morphospace indicate that cranium shape is most

9 214 A. Urošević et al. / Zoology 115 (2012) Fig. 5. Dorsal cranium shape changes associated with habitat preference, illustrated as a thin-plate spline deformation grid. For each sex, the shape changes are exaggerated three times. Landmarks 9, 10 and 12 on the dorsal skull deformation grid define the parietal bone. likely correlated with species ecological preference. We detected no phylogenetic signal in cranial shape, which suggests either that phylogenetic constraint does not affect the pattern of cranial shape variation in wall lizards or that adaptation to current environmental factors overwhelms the phylogenetic effects (Gidaszewski et al., 2009; Klingenberg and Gidaszewski, 2010). However, we cannot draw definite conclusions because the phylogenies we used are not well supported. Further, our sample consisted of only five species, which potentially reduced the power of comparative analyses and prevented us from making more general conclusions. The absence of phylogenetic signal in cranium and body shape is not unusual in lizards (Goodman and Isaac, 2008; Kohlsdorf et al., 2008; but see Vanhooydonck and Van Damme, 1999), and tracing the connection between morphology and phylogeny has always been problematic for this group (Arnold, 1973, 1989; Arnold et al., 2007). The static allometric slopes of females appeared to be homogenous among species, while the static allometric slopes of males diverged. Allometric shape changes explained the significant percentage of shape changes between sexes (Kaliontzopoulou et al., 2010; Ljubisavljević et al., 2010a, 2011). The magnitude of sexual dimorphism and allometric shape changes varied between species (Ljubisavljević et al., 2010a) and even between populations within species (Kaliontzopoulou et al., 2010). Therefore, the divergence in allometric slopes between males could be expected as a result of the divergence in the patterns of sexual dimorphism among species. Finally, we found a significant effect of habitat preference on cranial shape variation including allometry. These findings are consistent with previous studies on lacertid lizards (Vanhooydonck and Van Damme, 1999; Herrel et al., 2001; Verwaijen et al., 2002; Kaliontzopoulou et al., 2010) Cranial shape variation and habitat preference Compared to terrestrial species, saxicolous ones (P. erhardii and P. muralis) have elongated parietals and occipitals, and a shortened midface and rostrum. The point of articulation between parietals and frontals, which is important in cranial kinesis and stress reduction, is moved forward (Moazen et al., 2008, 2009). The elongation of the parietal and occipital bones could compensate for the flattened cranium which is suited for vertical habitats and for seeking shelter in crevices (Arnold, 1998; Herrel et al., 2001). The large subocular foramina are also characteristic of saxicolous lacertids and are associated with a crevice-dwelling way of life (Arnold, 1973; Kaliontzopoulou et al., 2010). Relative to terrestrial species, saxicolous Podarcis species have smaller jaw adductor chambers. The shift of the frontoparietal suture forward could also be a compensation related to bite force, due to a shortening of those parts of the dorsal cranium bones involved in biting, resulting in an increased out-lever (Verwaijen et al., 2002). Fig. 6. Ventral cranium shape changes associated with habitat preference, illustrated as a thin-plate spline deformation grid. For each sex, the shape changes are exaggerated three times. Landmarks 6, 7, 8, 10, 11, 12, 14 and 15 on the ventral skull deformation grid define the jaw adductor muscle chamber.

10 A. Urošević et al. / Zoology 115 (2012) The terrestrial species (P. melisellensis, P. sicula and P. taurica) have a longer rostrum and midface. The frontoparietal suture is moved posteriorly, which implies that the part of the dorsal cranium bones involved in biting is longer. This elongation should produce a wider gape at the expense of bite force. The shift of the frontoparietal suture posteriorly is especially pronounced in females, a pattern consistent with previous studies that found dietary divergences between the sexes in lacertid lizards (Herrel et al., 2001; Verwaijen et al., 2002). In the ventral cranium of terrestrial species, an enlargement of the jaw adductor muscle chambers is evident, together with a reduction of the occipitals and a narrowing of the subocular foramina. Terrestrial wall lizards have larger jaw adductor muscles and more robust bones, particularly in males. Such pronounced patterns in males are consistent with the results of recent sexual dimorphism studies (Herrel et al., 2007; Ljubisavljević et al., 2010a, 2011). However, there are differences in cranium shape among the terrestrial species. Whereas P. sicula has a robust and elongated cranium, P. melisellensis has a robust but wide cranium and P. taurica has a cranium shape approaching the mean of all five species. Orbits of terrestrial species appear enlarged, which could reflect selection for good eyesight in their open, vegetated habitats (Fleishman, 1992). We detected differences in the patterns of shape variation between the dorsal and ventral crania. Our results indicate that the structures of the ventral cranium, which are involved in the mechanics of feeding and jaw movement, are much more susceptible to various proximal effects (ecological, dietary), as suggested previously (Herrel et al., 2007; Ivanović et al., 2009; Ljubisavljević et al., 2011). The fact that habitat preference has the greatest effect on the ventral crania of males can be interpreted in the context of sexual dimorphism. Terrestrial Podarcis spp. males have enlarged jaw adductor muscle chambers relative to saxicolous males. Our data confirm that the structures of the ventral cranium, especially those involved in jaw movement and biting, can be an important component of sexual dimorphism in lizards. Usually, males have larger jaw adductor muscles than females. These larger muscles play a role in territoriality and ritual fighting, female choice and trophic niche partitioning between females and males (Herrel et al., 1999, 2007; Lappin et al., 2006; Ljubisavljević et al., 2010a, 2011). All the Podarcis species we studied are sexually dimorphic and tend to be territorial (Edsman, 1989). Intrasexual selection (male male competition), trophic niche partitioning and possibly intersexual selection (female preference) may have led to increased jaw muscle mass in males; however, dwelling in vertical habitats and seeking shelter in crevices can impose strong constraints against such muscular development (Vanhooydonck and Van Damme, 1999; Lappin et al., 2006; Ljubisavljević et al., 2010a, 2011). Thus, terrestrial lizards, which are free from such constraints, may have evolved larger jaw adductor muscles in response to sexual and natural selection Conclusions and further perspectives Differences in the pattern of phenotypic variation and habitat preference within closely related species of Podarcis lizards provide an excellent setting to explore the process of adaptive radiation, as well as the effects of shared evolutionary history, developmental constraints and functional requirements on the patterns of phenotypic variation. The observed patterns of variation indicate that phenotypic diversification in Podarcis species is driven largely by ecological demands, which appear to affect species-specific allometries. Studies of a wider range of species and ecotypes of lacertid lizards, detailed trophic studies, a more well-resolved molecular phylogeny and consideration of ontogenetic shape changes are needed for a better understanding of the relationships among morphology, ecology and phylogeny in lacertid lizards. Acknowledgements We would like to thank Miloš Kalezić, Georg Džukić and Tanja Vukov for helpful suggestions. This study was financially supported by the Serbian Ministry of Education and Science, grant no Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at References Aerts, P., Van Damme, R., Vanhooydonck, B., Zaaf, A., Herrel, A., Lizard locomotion: how morphology meets ecology. Neth. J. Zool. 50, Arnold, E.N., Relationships of the palearctic lizards assigned to the genera Lacerta, Algyroides and Psammodromus (Reptilia: Lacertidae). Bull. Br. Mus. Nat. Hist. (Zool.) 25, Arnold, E.N., Resource partitioning among lacertid lizards in southern Europe. J. Zool. 1, Arnold, E.N., Towards the phylogeny and biogeography of the Lacertidae: relationships within an Old-World family of lizards derived from morphology. Bull. Br. Mus. Nat. Hist. (Zool.) 44, Arnold, E.N., Structural niche, limb morphology and locomotion in lacertid lizards (Squamata, Lacertidae); a preliminary survey. Bull. Br. Mus. Nat. Hist. (Zool.) 64, Arnold, E.N., Arribas, O., Carranza, S., Systematics of the Palaearctic and Oriental lizard tribe Lacertini (Squamata: Lacertidae: Lacertinae), with descriptions of eight new genera. Zootaxa 1430, Bejaković, D., Aleksić, I., Tarasjev, A., Crnobrnja-Isailović, J., Džukić, G., Kalezić, M.L., Life history variation in a community of lacertid lizards from the Skadar Lake region (Montenegro). Herpetol. J. 6, Bookstein, F.L., Morphometric Tools for Landmark Data. Cambridge University Press, Cambridge. Bookstein, F.L., Combining the tools of geometric morphometrics. In: Marcus, L.F., Corti, M., Loy, A., Nayulor, G.J.P., Slice, D.E. (Eds.), Advances in Morphometrics. Plenum Press, New York, pp Botes, A.M.A., McGeoch, H., Robertson, G., van Niekerk, A., Davids, H.P., Chown, S.L., Ants, altitude and change in the northern Cape Floristic Region. J. Biogeogr. 33, Cardini, A., Elton, S., Geographical and taxonomic influences on cranial variation in red colobus monkeys (Primates, Colobinae): introducing a new approach to morph monkeys. Glob. Ecol. Biogeogr. 18, Cardini, A., Jansson, A.-U., Elton, S., Ecomorphology of vervet monkeys: a geometric morphometric approach to the study of clinal variation. J. Biogeogr. 34, Castilla, A.M., Fernadez-Pedrosa, V., Backljau, T., Gonzalez, A., Latorre, A., Moya, A., Conservation genetics of insular Podarcis lizards using partial cytochrome b sequences. Mol. Ecol. 7, Costantini, D., Alonso, M.L., Moazen, M., Bruner, E., The relationship between cephalic scales and bones in lizards: a preliminary microtomographic survey on three lacertid species. Anat. Rec. 293, Dingerkus, G., Uhler, L.D., Enzyme clearing of alcian blue stained whole small vertebrates for demonstration of cartilage. Stain Technol. 52, Dryden, I.L., Mardia, K.V., Statistical Shape Analysis. Wiley, New York. Edsman, L., Territoriality and competition in wall lizards. Ph.D. Thesis. University of Stockholm, Sweden. Fleishman, L.J., The influence of the sensory system and the environment on motion patterns in the visual displays of anoline lizards and other vertebrates. Am. Nat. 139, Fu, J., Toward the phylogeny of the family Lacertidae: implications from mitochondrial DNA 12S and 16S gene sequences (Reptilia: Squamata). Mol. Phylogenet. Evol. 9, Fu, J., Toward the phylogeny of the family Lacertidae why 4708 base pairs of mtdna sequences cannot draw the picture. Biol. J. Linn. Soc. 71, Gidaszewski, N.A., Baylac, M., Klingenberg, C.P., Evolution of sexual dimorphism of wing shape in the Drosophila melanogaster subgroup. BMC Evol. Biol. 9, 110. Goodman, B.A., Isaac, J.L., Convergent body flattening in a clade of tropical rock using skinks (Scicidae: Lygosominae). Biol. J. Linn. Soc. 94, Hansen, T.F., Martins, E.P., Translating between microevolutionary process and macroevolutionary patterns: the correlation structure of interspecific data. Evolution 50, Harris, D.J., Arnold, E.N., Relationships of wall lizards, Podarcis (Reptilia: Lacertidae) based on mitochondrial DNA sequences. Copeia 1999, Harris, D.J., Sá-Sousa, P., Species distinction and relationships of the Western Iberian Podarcis lizards (Reptilia, Lacertidae) based on morphology and mitochondrial DNA sequences. Herpetol. J. 11, Harris, D.J., Pinho, C., Carretero, M.A., Corti, C., Böhme, W., Determination of genetic diversity within the insular lizard Podarcis tiliguerta using mtdna sequence data, with a reassessment of the phylogeny of Podarcis. Amphibia-Reptilia 26,

11 216 A. Urošević et al. / Zoology 115 (2012) Herrel, A., Spithoven, L., Van Damme, R., De Vree, F., Sexual dimorphism of head size in Gallotia galloti; testing the niche divergence hypothesis by functional analyses. Funct. Ecol. 13, Herrel, A., Van Damme, R., Vanhooydonck, B., De Vree, F., The implications of bite performance to diet in two species of lacertid lizards. Can. J. Zool. 79, Herrel, A., McBrayer, L.D., Larson, P.M., Functional basis for sexual differences in bite force in the lizard Anolis carolinensis. Biol. J. Linn. Soc. 91, Irschick, D.J., Carlisle, E., Elstrott, J., Ramos, M., Buckley, C., Vanhooydonck, B., Meyers, J., Herrel, A., A comparison of habitat use, morphology, clinging performance and escape behaviour among two divergent green anole lizard (Anolis carolinensis) populations. Biol. J. Linn. Soc. 85, Ivanović, A., Sotiropoulos, K., Džukić, G., Kalezić, M.L., Skull size and shape variation vs. molecular phylogeny: study case of the alpine newts (Mesotriton alpestris, Salamandridae) from the Balkan Peninsula. Zoomorphology 128, Kaliontzopoulou, A., Carretero, M.A., Llorente, G.A., Intraspecific ecomorphological variation: linear and geometric morphometrics reveal habitat-related patterns within Podarcis bocagei wall lizards. J. Evol. Biol. 23, Klingenberg, C.P., Heterochrony and allometry: the analysis of evolutionary change in ontogeny. Biol. Rev. 73, Klingenberg, C.P., MorphoJ: an integrated software package for geometric morphometrics. Mol. Ecol. Resour. 11, Klingenberg, C.P., Gidaszewski, N.A., Testing and quantifying phylogenetic signals and homoplasy in morphometric data. Syst. Biol. 59, Kohlsdorf, T., Grizante, M.B., Navas, C.A., Herrel, A., Head shape evolution in Tropidurinae lizards: does locomotion constrain diet? J. Evol. Biol. 21, Lailvaux, S.P., Herrel, A., Vanhooydonck, B., Meyers, J.J., Irschick, D.J., Fighting tactics differ in two distinct male phenotypes in a lizard: heavyweight and lightweight bouts. Proc. R. Soc. Lond. B 271, Lappin, A.K., Hamilton, P.S., Sullivan, B.K., Bite-force performance and head shape in a sexually dimorphic crevice-dwelling lizard, the common chuckwalla [Sauromalus ater (=obesus)]. Biol. J. Linn. Soc. 88, Legendre, P., Legendre, L., Numerical Ecology, 2nd ed. Elsevier, Amsterdam. Ljubisavljević, K., Urošević, A., Aleksić, I., Ivanović, A., 2010a. Sexual dimorphism of skull shape in lacertid lizard species (Podarcis spp., Dalmatolacerta sp., Dinarolacerta sp.) revealed by geometric morphometrics. Zoology 113, Ljubisavljević, K., Džukić, G., Kalezić, M.L., 2010b. Lacertid reproductive strategy at the periphery of the species range within sandlands: a case study of the Balkan wall lizard (Podarcis taurica). Cent. Eur. J. Biol. 5, Ljubisavljević, K., Polović, L., Urošević, A., Ivanović, A., Patterns of morphological variation in the skull and cephalic scales of the lacertid lizard Algyroides nigropunctatus. Herpetol. J. 21, Losos, J.B., Adaptive radiation, ecological opportunity, and evolutionary determinism. Am. Nat. 175, Lutz, D., Mayer, W., Albumin evolution and its phylogenetic and taxonomic implications in several lacertid lizards. Amphibia-Reptilia 7, Lymberakis, P., Poulakakis, N., Kaliontzopoulou, A., Valakos, E., Mylonas, M., Two new species of Podarcis (Squamata; Lacertidae) from Greece. Syst. Biodivers. 6, Maddison, W.P., Squared-change parsimony reconstructions of ancestral states for continuous-valued characters on a phylogenetic tree. Syst. Zool. 40, McArdle, B.H., Rodrigo, A.G., Estimating the ancestral states of a continuous-valued character using squared-change parsimony: an analytical solution. Syst. Biol. 43, Moazen, M., Curtis, N., Evans, S.E., O Higgins, P., Fagan, M.J., Combined finite element and multibody dynamics analysis of biting in a Uromastyx hardwickii lizard skull. J. Anat. 213, Moazen, M., Curtis, N., O Higgins, P., Jones, M.E.H., Evans, S.E., Fagan, M.J., Assessment of the role of sutures in a lizard skull a computer modelling study. Proc. R. Soc. Lond. B 276, Mullin, S.K., Taylor, P.J., The effects of parallax on geometric morphometric data. Comput. Biol. Med. 32, Podnar, M., Mayer, W., Tvrtkovic, N., Phylogeography of the Italian wall lizard, Podarcis sicula, as revealed by mitochondrial DNA sequences. Mol. Ecol. 14, Podnar, M., Haring, E., Pinsker, W., Mayer, W., Unusual origin of a nuclear pseudogene in the Italian wall lizard: intergenomic and interspecific transfer of a large section of the mitochondrial genome in the genus Podarcis (Lacertidae). J. Mol. Evol. 64, Poulakakis, N., Lymberakis, P., Antoniou, A., Chalkia, D., Zouros, E., Mylonas, M., Valakos, E., Molecular phylogeny and biogeography of the wall-lizard Podarcis erhardii (Squamata: Lacertidae). Mol. Phylogenet. Evol. 28, Poulakakis, N., Lymberakis, P., Valakos, E., Pawlis, P., Mylonas, M., 2005a. Phylogeography of Balkan wall lizard (Podarcis taurica) and its relatives inferred from mitochondrial DNA sequences. Mol. Ecol. 14, Poulakakis, N., Lymberakis, P., Valakos, E., Zouros, E., Mylonas, M., 2005b. Phylogenetic relationships and biogeography of Podarcis species from the Balkan Peninsula, by Bayesian and maximum likelihood analyses of mitochondrial DNA sequences. Mol. Phylogenet. Evol. 37, Revell, L.J., Harmon, L.J., Collar, D.C., Phylogenetic signal, evolutionary process, and rate. Syst. Biol. 57, Rohlf, F.J., Comparative methods for the analysis of continuous variables: geometric interpretations. Evolution 55, Rohlf, F.J., Geometric morphometrics and phylogeny. In: MacLeod, N., Forey, P.L. (Eds.), Morphology, Shape, and Phylogeny. Taylor & Francis, London, pp Rohlf, F.J., TpsDig Department of Ecology and Evolution, State University of New York at Stony Brook, Available at: Rohlf, F.J., TpsRegr Department of Ecology and Evolution, State University of New York at Stony Brook, Available at: Rohlf, F.J., Slice, D.E., Extensions of the Procrustes method for the optimal superimposition of landmarks. Syst. Zool. 39, Ruggiero, A., Kitzberger, T., Environmental correlates of mammal species richness in South America: effects of spatial structure, taxonomy and geographic range. Ecography 27, Sá-Sousa, P., Harris, D.J., Podarcis carbonelli Pérez-Mellado, 1981 is a distinct species. Amphibia-Reptilia 23, Sheets, H.D., Integrated Morphometrics Package (IMP), Available at: sheets/morphsoft.html. Shingleton, A., Allometry: the study of biological scaling. Nat. Educ. Knowl. 1, 2. Stayton, C.T., Morphological evolution of the lizard skull: a geometric morphometrics survey. J. Morphol. 263, Vanhooydonck, B., Van Damme, R., Evolutionary relationships between body shape and habitat use in lacertid lizards. Evol. Ecol. Res. 1, Verwaijen, D., Van Damme, R., Herrel, A., Relationships between head size, bite force, prey handling efficiency and diet in two sympatric lacertid lizards. Funct. Ecol. 16, Zelditch, M.L., Swiderski, D.L., Sheets, D.H., Fink, W.L., Geometric Morphometrics for Biologists: A Primer. Elsevier Academic Press, San Diego.