MORPHOLOGICAL CHARACTERISTICS OF THE HERMANN ’ S TORTOISE ( TESTUDO HERMANNI GMELIN , 1789 ) IN THE SOUTH-EASTERN PART OF KOSOVO AND METOHIJA

In this paper we present the results of the analysis of morphological characteristics of the Hermann’s tortoise from three localities in southeastern part of Kosovo and Metohija. 75 adult Hermann's tortoises (33 males and 42 females) have been analyzed. Morphometric analisys has been done on 7 traits. We considered the plastron black pigmentation patterns and frequencies for each pigmentation states were calculated. Females were larger than males for all morphometric traits. Based on the most important dimensions of carapace and plastron length and width, the Hermann's tortoises from the south-eastern part of Kosovo and Metohija are the most similar to the individuals of the population from the southern part of the Balkan Peninsula (the southern part of Serbia and the northern part of Greece). In relation to the populations from central and eastern Serbia, individuals from our studed area have higher values.


INTRODUCTION
Hermann's tortoise (Testudo hermanni Gmelin, 1789) is distributed over the northern Mediterranean region.It is distributed over most of the Balkan Peninsula while in the western Mediterranean (continental Spain, France and Italy, Balearic Islands, Corsica, Sardinia and Sicily) has an isolated distribution.Although it is most prevalent in areas with a Mediterranean climate, it can be found in inland regions of the Balkan Peninsula under the influence of continental climate (Bour, 1997), (Cheylan, 2001), (Fritz et al., 2006), (Ljubisavljević et al., 2014).
The Hermann's tortoise is widely distributed in Serbia, where it inhabits hilly areas and lower mountains (below 1200 m) south of the Sava and Danube rivers (Ljubisavljević et al., 2014), (Tomović et al., 2014).
It is generally accepted that two subspecies have been recognized within species Testudo hermanni (Fritz et al., 2006), (Bertolero et al., 2011), nominotypic Testudo hermanni hermanni and Testudo hermanni boettgeri.T. h.hermanni occupies the patchy western Mediterranean range, from Spain to Italy, while T. h.boettgeri inhabits the coastal areas of the Balkan Peninsula, from Croatia to Greece, including many islands in the Adriadic and Ionian seas, the European part of Turkey, and further inland in the Bosnia and Herzegovina, Serbia, Montenegro, Macedonia, Romania and Bulgaria.
The main aim of this paper is to provide basic data about morphological characteristics of the Hermann's tortoise from south-eastern part of Kosovo and Metohija.Our aim was also to test the potential presence of gender differences.
Morphometric analysis has been done on 7 traits: SCLmidline straight carapace length, from the front of the nuchal scute, to the rear of the carapace; MCW mid-body carapace width, straight width of the carapace, at the level of the 6th marginals; MaxCWmaximal carapace width, width of the carapace at the widest point, usually at the level of the 8th marginals; CHcarapace height; MPLmidline plastron length; MaxPLmaximal plastron length, length of the plastron from the tips of gulars to the tips of anals; MaxPWmaximal plastron width.All linear measurements were taken with a mechanical calliper (precision 0,01 mm) while, midline straight carapace length was taken by a flexible measuring tape (precision 1 mm), by the same person (A.S.).2) black spots on humeral scutes are isolated; 3) black spots on anal scutes are isolated; 4) black spots on anal and humeral scutes are isolated; 5) black spots on anal and femoral scutes are isolated; 6) black spots on anal, humeral and femoral scutes are isolated; 7) black spots are isolated on each scute and reduced in size; 8) black spots are lacking on anal scutes, while on humeral and femoral scutes are isolated; 9) black spots are lacking on anal scutes, while on other scutes are isolated and reduced in size; 10) black spots are lacking on anal and humeral scutes, while on other scutes are isolated and reduced in size; 11) black pigmentation is lacking.
We considered the extended plastron black pigmentation (Fig. 1) and frequencies for each pigmentation states were calculated.This was done according to modified scheme of (Guyot & Devaux, 1997), used in morphological study of the Hermann's tortoise by (Ljubisavljević et al., 2012).Since, preliminary analysis of (Ljubisavljević et al., 2012), revealed no significant sex related variations in the frequencies of black pigmentation, we performed qualitative analysis of these trait on the whole sample.
Programme package STATISTICA (version 7.0) has been used for statistical data analysis.For morphometric characters basic parameters of descriptive statistics have been separately calculated by genders: mean ( x ), standard error (SE), range and standard deviation (SD).Analysis of variance (ANOVA) has been used to establish the significance of differences in respect of morphometric characters between the genders.

Morphometric characters
Descriptive statistics of body size measurements of adult males and females of Hermann's tortoises from south-eastern part of Kosovo and Metohija are presented in Table 1.

Biology
Based on the most important dimensions of carapace and plastron length and width, the Hermann's tortoises from the south-eastern part of Kosovo and Metohija are the most similar to the individuals of the population from the southern part of the Balkan Peninsula (the southern part of Serbia and the northern part of Greece).In relation to the populations from central and eastern Serbia, individuals from our studed area have higher values that is also in accordance with literature data (Djordjević et al., 2011), (Ljubisavljević et al., 2012).
Table 1.Descriptive statistics of morphometric characters of male and female Hermann's tortoise from southeastern part of Kosovo and Metohija.Abbreviations of characters are given in "Materials and Methods".

Character
Male (n = 33) Female (n = 42) x The analysis of variance (ANOVA) showed that sexes significantly differed in all 7 characters (Table 2).Females were larger than males for all traits.These results are consistent with published dates for sexual dimorphism of the Testudo genus (Willemsen & Hailey, 1999), (Willemsen & Hailey, 2003), (Djordjević et al., 2011), (Ljubisavljević et al., 2012).The observed direction of sexual size dimorphism, with females larger than males is most common in chelonians and is usually explained by fecundity selection (Ceballos et al., 2013).

Qualitative traits
Analysis of plastron black pigmentation level has shown that in individuals of the Hermann's tortoise from the southeastern part of Kosovo and Metohija seven out of 11 states are present.The most common are the conditions that determine the fragmented and reduced black pigmentation (states 7 and 10).They were present in 37.7% and 29.5% of individuals.This type of plastron pigmentation is characteristic for the Hermann's tortoises from the southern part of Serbia (Ljubisavljević et al., 2012).On the other hand, conditions that show high level of black pigmentation (states 2 and 4), are present in only 11.4% of individuals.Pigmentation states 1, 3, 8 and 9 were not identified in the analyzed sample.

CONCLUSION
In this paper we present the results of the analysis of morphological characteristics of the Hermann's tortoise from south-eastern part of Kosovo and Metohija.Results of descriptive statistics showed that the Hermann's tortoises from the south-eastern part of Biology Kosovo and Metohija are the most similar to the individuals of the populations from the southern part of the Balkan Peninsula (the southern part of Serbia and the northern part of Greece).Variance analysis found the statistically significant difference in body size between the genders.Females had higher values for all characters.Fragmented and reduced plastron black pigmentation prevailed in 2/3 of individuals in the population.
Our results of analysis of external morphology of the Hermann's tortoise from south-eastern part of Kosovo and Metohija, are in accordance with previous studies of this species from Serbia.

Fig. 1 .
Fig.1.Range of plastral pigmentation (modified scheme ofGuyot & Devaux, 1997 according to  Ljubisavljević et al., 2012): 1) continuous black pigmentation; 2) black spots on humeral scutes are isolated; 3) black spots on anal scutes are isolated; 4) black spots on anal and humeral scutes are isolated; 5) black spots on anal and femoral scutes are isolated; 6) black spots on anal, humeral and femoral scutes are isolated; 7) black spots are isolated on each scute and reduced in size; 8) black spots are lacking on anal scutes, while on humeral and femoral scutes are isolated; 9) black spots are lacking on anal scutes, while on other scutes are isolated and reduced in size; 10) black spots are lacking on anal and humeral scutes, while on other scutes are isolated and reduced in size; 11) black pigmentation is lacking.

Table 2 .
ANOVA with sex as the factor.Abbreviations of characters are given in "Materials and Methods".Bold statistically significant P values (P < 0.05).