Butterfly wings play a crucial role during flight, but also in thermoregulation, intraspecific signalling and interaction with predators, all of which vary across different habitat types and may be reflected in wing morphology or colour pattern. We focused on the morphological variability of Erebia medusa in order to examine patterns and variations in the colouration and morphology of wings from areas representing different habitat types with different environmental characteristics. The barrier (larger fragments of forest) between populations of Erebia medusa along the elevation gradient of Kojšovská hoľa might be the aspect that hinders the movement of the population. The wing characteristics (shape, size, spotting) of males representing populations of Carpathian mountain habitats (Volovské vrchy, Ondavská vrchovina) located at different elevations were measured. The forewing shape analysis, using geometric morphometry based on 16 landmarks, showed significant differences between populations from different elevation levels. The pattern of the forewings also varied between populations. Morphological changes among individuals of Erebia medusa populations along the elevation gradient in the Carpathian Mountains showed that in the cold, highland habitats we observed smaller, narrower and elongated forewings with a reduced number of spots, while males from warmer habitats at low elevations had rounder, larger and more spotted forewings.<br> <br> Introduction<br> <br> The ecological role of individual butterfly species is largely reflected in the wings, whose shape, size and colour pattern often have adaptive value and provide information about important differences, even at the population level (Altizer and Davis 2010; Mega 2014). The variability of butterfly wing shape or size, which reflects flight performance (Cespedes et al. 2015; Le Roy et al. 2019a, b), can even provide insight into the suitability of the habitat (Pellegroms et al. 2009; Chazot et al. 2016) and the dispersal rate (Wells et al. 2018; Taylor-Cox et al. 2020). The final wing shape and size of adults depends on conditions of larval development, which can be affected by aggregation behaviour (Allen 2010; Montejo‐Kovacevich et al. 2019; Palmer et al. 2019) but also by environmental conditions (Karl and Fischer 2008; Gibbs et al. 2011; Van Dyck et al. 2016; Palmer et al. 2019). Phenotypic clines along environmental gradients can sometimes be explained by ecological rules, whose use on insects can be debatable (Blanckenhorn and Demont 2004). Bergmann’s rule is the classic ecogeographic principle that relates the body size of endotherms with environmental temperature (or latitude) (Shelomi 2012). The converse of Bergmann’s rule (Park 1949; Mousseau 1997), based on the season length effect, predicts a decrease of body size with elevation. Various clines in body size can also be explained by a combination of several other theories or hypotheses, such as the north-south cline theory (Nylin and Svärd 1991) or the “temperature – size rule” (Angilletta and Dunham 2003).<br> The wing eyespot pattern, which may serve different functions, can also play an irreplaceable role. While the pattern on the dorsal side is usually used for intraspecific communication (Oliver et al. 2009; Westerman et al. 2012; Tokita et al. 2013), the eyespots on the ventral side are rather used to deceive predators by intimidation or deflection by distracting predators from the vital, vulnerable body parts (Lyytinen et al. 2003; Stevens 2005; Stevens et al. 2007; Kodandaramaiah 2011; Prudic et al. 2015; Ho et al. 2016). Moreover, in several butterfly species, wing colour modifications are related to thermoregulation (Dennis and Shreeve 1989; Taylor-Cox et al. 2020).<br> Previous studies (Nice et al. 2005; Jugovic et al. 2018) have demonstrated that populations separated by time, space or geographical barrier may undergo changes in the shape, size and colouration of external traits (Tatarinov and Kulakova 2013). Restrictions of the movement and migration of butterfly species have an impact on the intensity and direction of gene flow between populations (Andrews 2010; Slatkin and Excoffier 2012). Characterisation of the morphological traits of E. medusa, a species inhabiting a wide range of environments, can provide insight into the selection pressures that affect adaptive responses (Cespedes et al. 2015; Taylor-Cox et al. 2020). <br> For the sedentary butterfly Erebia medusa, high intraspecific variability (numerous subspecies) and mosaic distribution throughout most of its Euro-Siberian region is characteristic (Warren 1936; Schmitt et al. 2000; Polic et al. 2014). Our study focused especially on the influence of elevation differences in the Carpathian region on intraspecific variation. For this species, large fragments of forests (Schmitt et al. 2000) may be a serious obstacle for movement. According to the study by Kleckova and Klecka (2016), E. medusa prefers a warm environment, so the adaptations to high elevation habitats needed for the activity of this species can be expected. Lower activity due to low temperature can cause a decrease of chances of escape; therefore, selection will act against some individuals (large sized or with large eyespots) (Dennis et al. 1986). A higher number of eyespots, which are important especially in escape mechanisms, may reflect increased rates of predation with rising temperature (Hillebrand et al. 2009; Vucic-Pestic et al. 2011) but also by sexual selection (Tokita et al. 2013). Based on morphological features (wing size, shape, colour pattern) examined by traditional and geometric morphometry, we focused on the morphological differences between populations from habitats differing in elevation and separated by forest areas. <br> We predicted that the morphological diversity between E. medusa populations would show changes that correlate with the average annual temperature, which varies within the elevation gradient. Our study is based on the hypothesis that i) morphological traits of males (size, shape and pattern of forewings) vary in response to various environmental conditions within an elevation gradient. We also focused on examining whether ii) the forewing size of individuals from higher elevations is smaller than the forewing size of individuals from lower and warmer regions, which induce longer feeding periods during larval development (Juhász et al. 2016). Further, iii) males from higher elevation habitats with lower temperatures were expected to have aerodynamically (narrower, angular) shaped wings that reduce energy costs (Dudley 2002; Lentink et al. 2007; Kovac et al. 2012). Finally, iv) a reduction in the eyespot number with elevation, involving various selection pressures, was expected (Slabý 1950; Tatarinov and Kulakova 2013).
