THE VENTRAL CRANIAL SIZE AND SHAPE VARIATION BETWEEN MALES AND FEMALES OF EUROPEAN BROWN FROGS: RANA DALMATINA, R. GRAECA AND R. TEMPORARIA (ANURA, AMPHIBIA)

Sexual size and shape dimorphism is a very common phenomenon widely studied in the field of evolutionary biology. The differences between sexes are related to their life strategies and driven by the two evolutionary processes, sexual and natural selection. In amphibians, females are larger sex due to high correlation with fecundity, while dimorphism in body shape is often related to intrasexual competition for opposite sex during the breeding season. The main aim of this study is to describe patterns of ventral cranial size and shape variation between males and females of three species of European brown frogs, Rana dalmatina, R. graeca, R. temporaria, from the Balkan Peninsula. Our results showed that species R. dalmatina and R. graeca are sexually dimorphic for the ventral cranial size while species R. graeca and R. temporaria for the ventral cranial shape. Sexual dimorphism in cranial size is most probably an indirect consequence of natural selection favoring larger body for higher fecundity in explosive breeders like brown frogs. Cranial shape variation is under a strong influence of size variation but without allometric scaling between sexes. Sexual shape differences of analyzed brown frog species were most pronounced in the level of connection between cranium and jaw which indicate that differential diet of males and females can be a factor affecting observed sexual dimorphism patterns. More detailed studies of males and females microhabitats are necessary to conclude if differences in intersexual ecology affect intersexual size and shape differences in the cranium.


INTRODUCTION
Size  and shape variations are the principal sources of biological diversity (Hallgrímsson & Hall, 2005) and their origin and maintenance are in the focus of many biological disciplines. Sexual size and shape dimorphism, the morphological difference between sexually mature males and females, is a very common phenomenon widely studied in the field of evolutionary biology (Fairbairn, 2013). The differences between sexes are related to their life strategies and driven by the two evolutionary processes, sexual and natural selection (Shine, 1989;Monnet & Cherry, 2002).

MATERIAL AND METHODS
In this study, we included 112 adult skulls (R. dalmatina: 20 males, 19 females; R. graeca: 19 males, 17 females; R. temporaria: 27 males, 10 females) obtained from the Batrachological Collections of the Institute for Biological Research "Siniša Stanković," Belgrade (Dţukić et al., 2015). The sample size in this geometric morphometric study was large enough for the appropriate estimation of different parameters (Cardini & Elton, 2007).
Skulls were taken from adult specimens (determined by gonad examination) and they were cleared with trypsin and potassium hydroxide (KOH), stained with Alizarin Red S to distinguish cranial elements and their articulations better and then preserved in glycerol. Images of the ventral cranium were obtained with a Sony DSC-F828 digital camera (resolution 8.0 MP; Sony Corp., Tokyo, Japan). Eighteen specific twodimensional landmarks for the ventral cranium were digitized using TpsDig2 software (Rohlf, 2008). The chosen configuration of landmarks provides an adequate summary of specific aspects of the ventral cranium morphology. The specific positions of the chosen landmarks are shown in Figure 1. We applied Generalized Procrustes analysis (GPA) (Dryden & Mardia, 1998) to analyse interspecific, sexual, and body sizerelated variation in the ventral cranium of brown frogs'. A GPA was used to obtain a matrix of shape co-ordinates from which differences due to the position, scale, and orientation were removed (Dryden & Mardia, 1998). Variation in ventral cranial size (centroid size -CS) was evaluated by ANOVA with sex as factors. Determining the level of sexual dimorphism in size, the standard index is calculated by CS values: I SSD = size of the larger sex (females)/size of the smaller sex (males). The differences in shape between sexes are described as Procrustes distances. Procrustes distance, a linear measure of shape differences between landmark configurations, was used as an index of sexual dimorphism in shape (I SShD ).
To access the optimal estimate of the impact of allometry on shape changes, we employed two-way permutational MANCOVA with species and sex as factors and CS as covariable. Factor × CS interaction would indicate that sizedependent shape changes differ between the species or sexes. The percentage of predicted allometry with the statistical significance of the allometric regressions was tested with permutation tests against the null hypothesis of allometry independence. Residuals from the multivariate regression of shape variables on CS were used to visualize non-allometric shape changes in the ventral cranium between sexes by discriminant analysis, and to calculate size corrected index of sexual dimorphism in shape (I SShDcorr .).

RESULTS
Cranial size differs significantly between sexes for R. dalmatina and R. graeca with larger cranium in females, but not for R. temporaria (Table 1). Females had larger body size in our sample (R. dalmatina: females L = 63.9 mm, males L = 57.2 mm; R. graeca: females L = 68.1 mm, males L = 61.9 mm; R. temporaria: females L = 85.1 mm, males L = 75.7). The calculated SSD indexes (Table 1) and mean CS values for R. dalmatina and R. graeca showed that females had 12% larger cranium than males. Regression analyses showed that the ventral cranium of males and females for all three analysed brown frog species share the same allometric trajectory (F = 0.6744, P = 0.6438), with 16.91% of shape variation explained by the size variation (P BIOLOGY < 0.0001). Sexual shape differences were significant for R. graeca and R. temporaria, and not significant for R. dalmatina, before and after removing the allometric component of the ventral cranium variation ( Table 2). The calculated SShD indexes were similar for R. graeca and R. temporaria (Table 2). Patterns of the intersexual shape variation were the same before and after removing the allometric component. Females of the R. graeca had wider skull than males in the posterior part of the cranium (landmarks 1, 2, 3, 4) but shorter snout (landmarks 14, 15, 16, 17), with anteriorly displaced contact between pterygoid and maxilla (landmarks 11, 12) ( Figure 2A). In the R. temporaria females compared to males the tips of lateral pterygoid process and quadrate bones were displaced posteriorly (landmarks 1, 2, 3, 4), contacts between the palatine and maxilla and pterygoid and maxilla anteriorly (landmarks 9, 10, 11, 12), and snout was slightly longer (landmarks 16, 17) ( Figure 2B).

Figure 2.
Shape changes associated with non-allometric SShD between sexes in R. graeca (A) and R. temporaria (B) (from males to females).

DISCUSSION
Three species of the European brown frogs are sexually dimorphic for the ventral cranial size in R. dalmatina and R. graeca and for the ventral cranial shape in R. graeca and R. temporaria. Cranial shape variation is under a strong influence of size variation but without allometric scaling between sexes. The most pronounced shape dissimilarities between sexes are found in the posterior part of the ventral cranium.
Two evolutionary processes, sexual and natural selection, have the highest impact on the morphological variability of males and females (Shine, 1989;Andersson, 1994;Fairbairn, 1997;Hendry et al., 2014). Therefore, the difference between sexes is a consequence of differential interaction of each sex with environment and interaction between sexes (Herrel et al., 2001;Sacchi et al., 2009;Peiman & Robinson, 2010;Lailvaux et al., 2012). In the context of sexual selection, sexually dimorphic size and shape of the cranium could evolve through intrasexual competition where cranial traits provide advantages in breeding opportunities to some individuals through contests of dominance or attracting the opposite sex. For example, maximum bite force linked to features of cranial-jaw complex contributes to successful breeding in reptiles (Lappin & Husak, 2005;McBrayer & Anderson, 2007). In addition, sexually dimorphic size and shape of the cranium could arise through natural selection and intersexual niche divergence in order to reduce intersexual competition for resources and habitat use. Indeed, many studies showed that skull dimorphism is mainly restricted to differences in the size and shape of the parts that can affect bite force and prey size (Vincent, et al., 2004;Herrel et al., 2007;Kaliontzopoulou et al., 2007;Ljubisavljević et al., 2010).
Our results showed that females had larger ventral cranium and bodies than males, therefore sexual dimorphism in cranial size is most probably an indirect consequence of natural selection favoring larger body for higher fecundity in explosive breeders like brown frogs (Woolbright, 1983). In addition, brown frogs do not display male to male combats or territorial contests, or intersexual interactions like copulatory bites, so sexual dimorphism in cranial size is probably not a consequence of sexual selection. Sexual shape differences of analyzed brown frog species were most pronounced in the level of connection between cranium and jaw which indicate that differential diet of males and females can be a factor affecting observed sexual dimorphism patterns. However, studies about the diet of brown frogs males and females are almost not existing (Cicort-Lucaciu et al., 2011) so the direct link between observed intersexual cranial shape variation and diet variation cannot be established. Study of the dorsal cranium variation in the European brown species showed that ecologically similar species (in habitat characteristics and type of locomotion) shared the cranial morphology (Krstičić Račković et al., 2019). However, more detailed studies of males and females microhabitats are necessary BIOLOGY to conclude if differences in intersexual ecology affect intersexual size and shape differences in the cranium.
Observed patterns of sexual dimorphism in cranial size and shape of three brown frog species indicate that further studies are necessary in order to find causes and consequences of intersexual variation of the cranium.