Landscape Distributional Pattern and Craniometry of Suncus etruscus (Mammalia: Insectivora, Soricidae) in South-East Bulgaria

ACTA ZOOLOGICA BULGARICA Acta zool. bulg., 56 (3), 2004: 299-312

Landscape Distributional Pattern and Craniometry of Suncus etruscus (Mammalia: Insectivora, Soricidae) in South-East Bulgaria Vasil V. Popov*, Bojan Miltchev**, Valeri C. Georgiev***, Hristo A. Dimitrov****, Tsenka Chassovnikarova* Abstract: The distribution of S. etruscus in South-East Bulgaria was traced on the basis of Barn Owl pellet collections. The species was found at 39 out of the 41 collecting sites. The effect of eight environmental variables, describing the proportion of the main habitat types within the owl’s hunting territory and the climatic and landscape context, on the structure of the shrew prey spectra was investigated by means of redundancy analysis. It has been found out that the percentage of moist areas within the hunting territory and the temperature sums affect significantly the structure of the shrew prey spectra. The percentages of S. etruscus were positively related to the temperatures. On the basis of the obtained results and comparisons with published data, it was hypothesized that S. etruscus was a recent invader in the area. The spreading route probably follows the valleys of the Maritsa and the Tundzha rivers and curves eastward north of the Sakar and Strandzha Mountains reaching the Black Sea shore. The proportion of S. etruscus in the Barn Owl’s food spectra tends to be lower compared with the true Mediterranean areas in Europe. Comparisons with craniometric data from literature indicate that the SE Bulgarian population represents a relatively large form in agreement with a well expressed size cline from south to north that complies with Bergmann’s rule. Within the analyzed data set from S Europe the palatal length and zygomatic breadth increase significantly with decreasing of the mean annual temperature. Key words: Pygmy white-toothed shrew, Owl pellets, landscape distributional pattern, craniometry, geographical variability.

Introduction The Barn Owl is highly specialized to feed with small mammals (SCHMIDT 1973, LOVARI et al. 1976, TAYLOR 1994). The bone remains concentrated in its pellets provide important information on the regional occurrence of shrew and rodent species, which are rarely captured by the conventional trapping methods (VOHRALIK, SOFIANIDOU 2000, KRYŠTUFEK, VOHRALIK 2001). The Barn Owl can be considered a key predator of * Institute of zoology, Bulgarian Academy of Sciences, 1, Tsar Osvoboditel Blvd., 1000 Sofia, Bulgaria; e-mail: [email protected] ** Faculty of Biology, University of Sofia “St. Kliment Ohridski”, 8 Dragan Tsankov, 1164 Sofia, Bulgaria; e-mail: [email protected] *** Ministry of Environment and Water, “National Nature Protection Service” Directorate, 22, Maria Luisa Blvd., 1000 Sofia, Bulgaria; e-mail: [email protected] **** University of Plovdiv “Paisii Hilendarski”, Department of Zoology, 24, Tsar Asen St., 4000 Plovdiv, Bulgaria

299

shrews among the European owls (KORPIMÄKI, NORRDAHL 1989). In particular, the shrews have a rather high percentage in its total prey taken in Southern Europe, especially of such small-sized species as the pygmy white-toothed shrew, Suncus etruscus (SAVI, 1822), and the lesser white-toothed shrew, Crocidura suaveolens (PALLAS, 1811), (HERRERA 1974, CONTOLI 1975, LOVARI et al. 1976, CONTOLI et al. 1977, RZEBIKKOWALSKA 1985, NIETHAMMER 1989, ALIVIZATOS, GOUTNER 1999, TEMME 2000). The recent investigation on the feeding ecology of Barn Owl in SE Bulgaria based on pellet analyses has revealed a curiously wide distribution of Suncus etruscus across the region. The collected data, subject of the present report, allow to analyze the regional species’ distribution in detail. The review of the literature data indicates that the occurrence of shrews in the diet of owls is primarily affected by the composition of the prey community (KORPIMÄKI, NORRDAHL 1989), which in its turn depends on the geographical location of the owl’s hunting territory and its landscape context (LOVARI et al. 1976, TAYLOR 1994). Although the contents of owl pellets do not represent the true proportions of the small mammal species in the community the owl preys upon (ANDREWS 1990), the comparative analysis of such data offers a possibility to relate the spatial changes of the prey spectra to the environmental heterogeneity within the area of interest. The multivariate quantitative techniques can assist in revealing such patterns in a repeatable and potentially unbiased way. As far as the present study is concerned, this approach will help to put the quantitative changes of S. etruscus across pellet samples within the context of shrew distributional patterns of regional significance and to reveal its environmental affinities. Beside this, the obtained large sample of cranial fragments allows to provide a biometrical characteristics of the species in this part of its range, which is of particular importance having in mind the scarce data in this respect (LAAR, DAAM 1967, SPITZENBERGER 1970, 1978, VOHRALIK 1985, NIETHAMMER 1962, 1989, POPOV, NIJAGOLOV 1991, KRYŠTUFEK, VOHRALIK 2001).

Material and Methods Study area. The region has a relatively uniform topography with elevations ranging from 0 to 500 m above sea level. It has a continental Mediterranean climate, characterized by mild and short winters and warm or moderately hot summers. The mean annual temperature varies between 100 and 13.30 C. The mean January temperatures range between 2o and 4o C along the Black Sea coast, and between 00 and 20 in the adjacent inlands. The mean July temperatures vary between 220 and 230 for the hilly areas, and between 230 and 240 for the lowlands. According to the temperature sums for the period with temperatures above 100 (growing season) the region supports three climatic types: hot (39000 - 41000 C), associated with the lowlands in the north-western part of the region, between the rivers of Tundzha and Maritsa, as well as the Burgas lowland and the adjacent coastal strip to the south; moderately hot (37000 - 39000 C), associated mainly with the Manastirski and Sveti-Ilijski hilly areas, the left bank of the Tundzha River, and the western part of the Burgas lowland; warm (35000 - 37000 C), associated with the Sakar Mts., northern foothills of the Strandzha Mts., and Bakadzhitsite - Hisar hills. The summer (June-August) drought expressed as a difference between rainfall and evaporation (mm) varies from -400 - -300 (high) for the lowlands west of the Tundzha River through -300 - -200 (moderate) for the hilly areas and the lowlands near the sea shore, to -200 - - 100 (low) for the semi-mountain areas of the northern foothills of the Strandzha Mts. The region is dominated by agricultural fields, replacing the native oak

300

301

302

17 1 0 0 0 1 1 0 1 1 0 1 0 16 1 1 1 1 1 1 2 1 1 1 2 2 15 2 2 2 2 2 2 2 2 2 2 1. 2 14 0 0 1 0 2 1 0 0 0 1 2 1 13 2.1 0.6 0.6 0.5 2.1 0.8 1.1 5.8 1.2 0.3 1.5 1.1 12 73.6 76.6 71.5 74.3 62.0 93.6 74.1 57.7 78.4 61.1 57.6 72.1 11 4.6 2.9 2.2 8.6 13.4 2.1 0.7 9.5 3.7 2.3 18.4 0.9 10 19.8 19.9 25.8 16.4 22.5 3.5 24.2 27.0 16.7 36.3 22.5 25.9 9 481 51 227 2 31 2 249 104 293 33 159 1 129 97 20 10 129 978 186 14 174 24 322 8 1.46 1.96 17.62 100 6.45 100 7.63 4.80 2.04 9.09 6.92 100 3.87 27.83 10.00 5.43 1.33 7.14 2.87 8.33 1.86 7 45.53 66.67 59.03 41.93 59.84 78.85 67.23 42.42 40.88 62.79 52.58 55.00 60.00 51.16 39.36 27.42 35.71 54.02 79.17 42.86 6 36.80 27.45 23.35 25.80 28.11 15.38 22.52 30.30 38.36 28.62 19.58 20.00 10.00 37.98 45.91 61.29 21.43 22.41 12.50 41.61 5 10.19 3.92 25.80 2.41 4.78 15.15 13.20 1.55 5.00 10.00 2.32 11.04 10.75 21.43 9.77 8.38 3 2.30 3.42 2 MG05 MH10 MH20 MG26 MG24 MG24 MG39 MG38 MG37 MG48 MG48 MG47 MG45 MH60 MG59 MG69 MG68 MG66 MG64 MG76 MH80 MG89 MH90

4 6.03 2.01 0.96 3.41 3.03 0.63 3.10 1.03 10.00 20.00 3.10 2.35 0.54 14.28 8.62 3.73

v8 S. a.

No on map 1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

UTM

S. m.

N. a.

C. l.

C. s.

S. e.

N

v1

v2

v3

v4

v5

v6

v7

Landscape characteristics Percentage of habitat types within the barn owl’s hunting area Species (%) Location

Fig. 1. Sample localities in UTM grid. Circles, denoted with numbers (1-33), represent large collections of owl pellets; black circles - samples with S. etruscus, open circles - samples without S. etruscus. Asterisks, denoted with letters (A-H), represent remains of S. etruscus from isolated pellets. Shaded -areas above 300 m above sea level. For more details see Table 1. 120 C - mean annual temperature isotherm is also shown.

Table 1. Location of the Tyto alba pellet samples from SE Bulgaria, species percentages, number of shrew specimens (N), and landscape characteristics Abbreviations: Species: S. a. - Sorex araneus, S. M. - S. minutus, N. a. - Neomys anomalus, C. l. - Crocidura leucodon, C. s. - Crocidura suaveolens, S. e. - Suncus etruscus. v1 - v 8: environmental variables, described in the text for the most numerous samples (N>100).

forests. As a whole, the remains of the forests are preserved as restricted patches on the hilly areas, but these form a relatively large cover on the southeastern part of the region. The secondary vegetation is presented by forests and shrubs of Oriental hornbeam and Christ’s thorn occurring on the rough terrain. The above information is summarized from HERSHKOVICH (1984), YORDANOVA, DONTCHEV (1997). Prey spectra and shrew remains. We located and visited nests, roosts and loafing sites of the barn owls to collect regurgitated pellets. During the period 1999 - 2002 a total of 33 collections of pellets were obtained from 31 squares of the 10 x 10 km UTM grid (Table 1, Fig. 1: 1-33). Additionally, remains of S. etruscus were found in isolated owl pellets, collected accidentally from 8 localities (Fig. 1: A-H). The pellets were dissected by hand and the contents was examined with the aid of a binocular microscope. Mammals were identified and counted on the basis of skulls and lower jaws; the latter were sorted into left and right dentaries, the highest number of any category being taken as the minimum number of individuals (MNI), represented in a collection of pellets. The data concerning the shrews from the large samples of pellets are shown on Table 1. Environmental data. In order to assess the effect of environment on the quantitative structure of the shrew spectra in the prey of Tyto alba (SCOPOLI, 1769), and on the quantitative representation of S. etruscus in particular, two groups of variables were recorded around the localities of the 21 largest collections (with more than 100 shrew specimens), (Table 1). The first group of variables (v1 - v4) characterizes the hunting territory of the owl and represents the local conditions. The localities were indicated on

17 1 0 1 1 1 1 1 1 16 2 2 1 2 2 2 2 1 15 2 1 2 2 2 2 2 2 14 0 2 0 1 0 0 1 0 13 0.8 0.8 1.2 6.3 40.2 1.5 1.7 20.6 12 78.2 84.1 57.3 80.9 37.7 69.6 66.7 24.5 11 4.7 1.2 10.5 7.9 7.3 8.5 29.1 25.8 10 16.2 13.9 31.0 4.9 14.8 20.5 2.5 29.1 9 988 826 65 271 1516 360 823 9 857 494 8 1.42 5.93 7.69 2.58 0.86 0.55 0.24 11.11 0.47 0.20 7 52.93 45.57 35.38 67.90 66.82 55.00 52.85 44.44 69.43 65.79 6 36.94 30.79 35.38 26.57 26.71 10.00 40.82 22.22 27.54 10.53 5 4.55 4.48 9.23 1.48 5.14 34.44 6.07 22.22 2.10 23.28 4 3.95 7.27 4.61 2.21 0.26 0.47 0.20 3 0.20 6.06 7.69 1.84 0.20 2 MG99 NH01 NH00 NG09 NG18 NH30 NG39 NG38 NG49 NG67 1 24 25 26 27 28 29 30 31 32 33

Table 1. Continued.

a map (1: 25 000) and a crude description of the surroundings (a circle with a radius of 1 km) was given in a form of a percentage cover of four habitat types: v1 - buildings; v2 - forests and shrubs; v3 open terrain and cultivated fields; v4 - moist areas. The second group of variables aimed to describe the landscape context. The following four (v5 - v8) variables were evaluated on the basis of interpolation between isopleths within the respective 10 x 10 km UTM squares: v5 - altitude (0 - the whole square area is below 200 m a.s.l., 1 - part of the square area is above 200 m a.s.l.; 2 - the whole square area is above 200 m a.s.l.); v6 - summer temperature conditions, based on the temperature sums for the period with temperatures above 10 o C (growing season), see above (1 - warm ; 2 - moderately hot, and 3 - hot); v7 - summer drought, for definitions see above (1 - low; 2 - moderate; 3 - high); v8 presence (1) / absence (0) of water basins (rivers, lakes, marshes). Analytical procedures. The most numerous samples and the respective environmental variables (Table 1), after arcsine transformation (ZAR 1999) of the percentage data, were analyzed by redundancy analysis (RDA) - a canonical form of principal component analysis (BRAAK 1987a, b), in order to determine the effect of environmental variables on the quantitative structure of the studied shrew spectra. With this method, it is possible to extract variance explained by one or more environmental variables introduced a priori in the analysis. The most important environmental variables were revealed by a forward selection and their significance tested by a permutation test. On the ordination diagram the species and environmental variables are presented by arrows. The angles between arrows represent the correlations between species’ percentages and environmental variables. The environmental variables with long arrows and the sharp angles with ordination axes are the most important in the analysis (BRAAK 1987a, b, 1990). Craniometry. The cranial, mandibular and dental measurements, presented in millimetres and taken using a stereo-microscope, fitted with a micrometric lens, were as follows: 1. condylobasal length (CBL); 2. zygomatic breadth (ZB); 3. interorbital width (IW); 4. palatal length (PALL); 5. postglenoid width (PGW); 6. length of I1 - M3 (LI1-M3); 7. length of P4 – M3 (LP4 – M3); 8. length of M1-M3 (LM1-M3); 9. length of A1-A4; 10. length of P4 (LP4); 11. width of P4

303

(WP4); 12. length of M1 (LM1); 13. width of M1 (WM1); 14. length of M2 (LM2); 15. width of M2 (WM2); 16. length of M3 (LM3); 17. width of M3 (WM3); 18. length of mandible (LMd); 19. height of processus coronoideus (HPC); 20. height of horizontal branch of mandible under M2, measured lingually (HMd/M1); 21. height of condyle (HC); 22. width of condyle (WC); 23. length of I1-M3 (LI1-M3); 24. length of P3-M3 (LP3-M3); 25. length of M1-M3 (LM1-M3); 26. length of P3 (LP3); 27. width of P3 (WP3); 28. length of P4 (LP4); 29. width of P4 (WP4); 30. length of M1 (LM1); 31. width of M1 (WM1); 32. length of M2 (LM2); 33. width of M2 (WM2); 34. length of M3 (LM3); 35. width of M3 (WM3) . The measurements, except the length of mandible, were taken according LÓPEZ-FUSTER et al. (1979). The length of mandible was measured according to CONTOLI et al. (2000). Measurements depending on tooth wearing were taken from specimens with unworn teeth only. In Table 2, the following descriptive statistics are shown: N - sample size; Min - Max - minimal and maximal observed values; X - mean; SD - standard deviation. To test whether temperature conditions play an important role in the microgeographical morphometric variation of S. etruscus within the studied area, a MANOVA was performed. Only measurements showing univariate normality and homogeneity of variances were included in this analysis (LMd, HPC, HMd/M2, LP3-M3, LM1-M3, LP4-M3, PGW, ZW, and IW). The measured specimens were combined into three groups according to the summer temperature conditions (v6, see above). In order to study the effect of temperature within a larger geographical scale, Spearman rank correlations were calculated between the annual average temperature and the arithmetic averages of some measurements of several populations reported in the literature. The temperature data were those presented in the respective sources or, if missing, were taken from appropriate climatic maps. The comparisons with literature data, presenting sample size, mean and variance, were based on F- and t-tests. If the F-test indicated that the variances differed significantly, an unequal variance t-statistic was implemented.

Results Regional distributional pattern. Of the environmental variables included in the analysis, statistical testing using forward selection with Monte Carlo permutation tests (BRAAK 1990) showed that only two variables, v4 and v6, were contributing significantly (P=0.05, after 99 permutations) to the pattern shown in Fig. 2. The first ordination axis (eigenvalue 0.22) was highly correlated with v4 (r=0.83) and described the increasing of the percentage of N. anomalus CABRERA, 1907 in the shrew bulk prey taken in areas with a large proportion of moist habitats. The remaining species did not show any pronounced correlation with this axis. The second axis (eigenvalue 0.11) was correlated with the temperature conditions (v6, r = -0.55). The shrew species (except N. anomalus) formed two well pronounced groups along this axis. The percentages of S. araneus LINNAEUS, 1758, S. minutus LINNAEUS, 1766, and C. leucodon (HERMANN, 1780) in barn owl’s food spectra were negatively related to the temperature conditions during the growing season, while the proportions of S. etruscus and C. suaveolens were positively affected by this factor. Both axes were statistically significant (P=0.01, Monte Carlo permutation test, 99 permutations) and captured 32 % of the variation of shrew spectra in relation to environmental variables. Craniometry. The results of MANOVA showed that the differences among the three groups, defined on the basis of the summer temperature conditions, were not

304

5 4 3

C.leucodon

S.araneus

19

S.minutus

23

2 Axis 2

Table 2. Cranial, mandibular and dental measurements (mm) of Suncus etruscus from SE Bulgaria. For abbreviations – see text.

18

N.anomalus

1

25 24

1

3 01 1 2 1

17

28

0

27 13 7

-1 -2 -3

S.etruscus 14

3

32

29 33

9

v4

C.suaveolens 8

v6 -4 -3,5 -3,0 -2,5 -2,0 -1,5 -1,0 -0,5 0,0 0,5 1,0 1,5 2,0 2,5 3,0 3,5 4,0 4,5 Axis 1 Fig. 2. Ordination diagram, showing the results of redundancy analysis, based on thå data presented in Table 1. Points - samples, numbered according to their position on Fig. 1; Solid arrows - species vectors; dotted arrows - vectors of environmental variables (v4 - percentage of moist areas; v6 summer temperature conditions).

significant. Consequently, the data were pooled and the univariate statistics of the measurements are given in Table 2. In order to evaluate the morphometric distinctiveness of the Bulgarian population of S. etruscus, pair wise comparisons, based on t- tests, were performed. The comparisons with data from the Apennine Peninsula and Sicily (CONTOLI et al. 2000) indicated that the mean values of ZW, PALL, and HPC for the studied population were higher than, or rarely equal, to the Italian samples. In many cases, the differences were statistically significant (P< 0.05). In two measurements, PGW and LMd, the Bulgarian population occupied an intermediate location within the range of the Italian populations. The comparisons with the data from the West Mediterranean (LÓPEZ-FUSTER et al. 1979, SANS-COMA et al. 1981, 1985) showed that in all characters (PALL, PGW, LP4M3, BLP4, WP4, BLM1, WM1, BLM2, WM2, WM3, LM1-M3, LM1, LM2, and LM3) but one (PGW), the mean values in the Bulgarian population were significantly greater (P< 0.05). The mean value of PGW of the Bulgarian population was significantly lower than in the West Mediterranean populations. In order to test the environmental determination of skull size, Spearman rank correlation coefficients were calculated between mean annual temperatures and mean values of 5 skull measurements (ZW, PALL, PGW, LMd, and HPC) in 19 samples given in the literature (LÓPEZ-FUSTER et al. 1979, SANS-COMA et al. 1981, 1985, CONTOLI et al. 2000) and observed in this study. Although four out of five analyzed measurements showed a negative relation to temperature, only PALL and ZW had significant coefficients: -0.58 (P=0.009) and -0.47 (P=0.049), respectively. These inverse relationships between mean annual temperature and respective measurements are shown on Fig. 3.

305

No: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35

Measurement CBL ZW IW PALL PGW LI1-M3 LP4-M3 LM1-M3 LA1-A4 LP4 WP4 LM1 WM1 LM2 WM2 LM3 WM3 LMd HPC HMd/M2 HC WC LI1-M3 LP3-M3 LM1-M3 LP3 WP3 LP4 WP4 LM1 WM1 LM2 WM2 LM3 WM3

N 5 210 184 214 169 195 209 212 26 90 90 92 92 87 87 87 87 90 195 208 89 89 185 81 197 34 34 67 67 89 89 91 91 91 91

X 12.48 4.06 2.82 4.91 4.40 5.59 3.17 2.17 1.80 1.27 1.33 1.07 1.47 0.97 1.38 0.45 0.96 6.37 3.02 0.89 1.33 0.97 5.18 3.79 2.67 0.75 0.51 0.81 0.63 1.10 0.78 1.13 0.71 0.84 0.53

Min 11.80 3.65 2.60 3.90 4.10 5.15 2.97 2.05 1.67 1.12 1.15 1.00 1.25 0.90 1.22 0.40 0.60 6.00 2.85 0.65 1.22 0.90 4.75 3.57 2.52 0.62 0.42 0.70 0.55 1.02 0.52 1.02 0.65 0.75 0.45

Max 13.20 4.30 3.10 5.40 4.95 6.00 3.40 2.32 1.92 1.40 1.47 1.15 1.62 1.05 1.45 0.52 1.05 6.70 3.17 1.03 1.45 1.10 5.45 3.95 2.85 0.85 0.57 0.90 0.70 1.25 1.07 1.25 0.80 0.95 0.62

SD 0.576 0.110 0.082 0.139 0.130 0.121 0.068 0.048 0.078 0.044 0.065 0.030 0.061 0.036 0.049 0.025 0.054 0.145 0.066 0.045 0.046 0.040 0.132 0.082 0.059 0.050 0.033 0.033 0.036 0.045 0.051 0.052 0.030 0.038 0.033

These comparisons indicate that the population from Bulgaria represents a largesized form, in agreement with its geographical location within the context of the observed morphometric cline.

Discussion Distribution. The RDA model presented above reflects a general pattern meaningful to us in relation to the known ecology of the shrew species included. Although the studied area is relatively homogeneous, RDA reveals the occurrence of three types of environmentally determined shrew prey spectra: 1) spectra from moist areas dominated by N. anomalus, 2) spectra from low elevations with hot growing seasons, dominated by C.

306

5,06 1

A

5,00

PALL

4,94

3

2

4 4,88 5 4,82

10

8 9

7

6 14

12

15 16

13

17

4,76 18

19

4,70 11

12

13

14

15

16

17

18

19

MEAN ANNUAL TEMPERATURE 4,12

B

4,08

3

4 4,04

ZW

4,00 10 3,96

15

1

3,92

6

11

7

5

9

12 14

3,88

18

19

3,84 11

12

13

14

15

16

17

18

19

MEAN ANNUAL TEMPERATURE Fig. 3. Palatal length, PALL (A) and zygomatic width, ZW (B) of S. etruscus relative mean annual temperature with negative exponential regression lines. Samples: 1.Veneto, 2. Sardinia (subfossil) 3. SE Bulgaria, 4. Emilia, 5. Abruzzi, 6. Apulia, 7. Umbria, 8. South France (skulls), 9. Liguria, 10. Latinum, 11. Latinum - Rome, 12. Calabria, 13. Sardinia (recent), 14. Campania, 15. Sardinia, 16. South France (owl pellets), 17. Catalunya, 18. North Sicily, 19. East Sicily. Sources: 2., 13. - SANS-COMA et al. (1985); 8, 16 - SANS-COMA et al., (1981); 17. - LÓPEZ-FUSTER et al. (1979); 1, 4 - 7, 9 - 12, 14 - 15, 18 - 19. - CONTOLI et al. (2000); 3. present study.

307

suaveolens and locally with a large percentage of S. etruscus, and 3) spectra from hilly areas with warm growing seasons, showing a great share of C. leucodon and an occurrence of Sorex minutus and S. araneus. It appears that the higher temperatures, which in this area are connected with low elevations, favour S. etruscus. The proportion of S. etruscus in our largest samples (with more than 100 shrew specimens, N=20) varies between 0 and 27.83 %. The distribution is highly asymmetrical with a median of 2.22 %. This value is similar to the species’ percentage in the winter diet of Barn Owl from Northeastern Greece - 2.94 % (a sample of 68 shrew specimens), (ALIVIZATOS, GOUTNER 1999). However, the comparisons with other data on the shrew prey spectra of Barn Owl from the Mediterranean area indicate that these percentages are relatively low. The share of S. etruscus in the island of Kos and the adjacent Anatolian coast (NIETHAMMER 1989) varies between 5.42 and 18.93 % with a median value of 11.57 % (3 samples with more than 90 shrew specimens). The percentage of S. etruscus in Central Tuscany (Italy) is higher, varying between 1.51 and 30.55 % with a median value of 15.53 % (6 samples with sizes varying from 66 to 229 shrew specimens), (LOVARI et al. 1976). In South Portugal (Prov. Alentejo), the share of S. etruscus is 20.93 % (a sample of 172 shrew specimens), (TEMME 2000). It should be mentioned that the Italian area has a more northern location than the other ones, including our study area. Thus, the regional climatic conditions are probably more important than the geographical latitude. In all of the above-mentioned areas, where the share of S. etruscus is greater, the climate is warmer. For example, the sums of temperatures for the period with temperatures above 100 C are as follows: South Portugal - above 60000 C, Central Italy - 4500 - 50000 C, West Anatolian coast - 5000 - 60000 C. These data indicate again that the share of S. etruscus in the Barn Owl prey spectra from South Europe is positively related to the temperatures. For Bulgaria Suncus etruscus was first recorded in 1980 from the southern part of the Black Sea coast (Burgasko Ezero Lake) (VOHRALIK 1985) and the subsequent findings were confined to this area, too (POPOV, NIJAGOLOV 1991, POPOV 2000). A record (1987, Barn Owl pellets) from the adjacent inland (Krumovo, Yambol district) has been reported (VOHRALIK, SOFIANIDOU 2000). Interestingly enough, a large sample of small mammals (4727 specimens) obtained from Barn Owl pellets gathered earlier (1979) in the same region (Atanassovsko Ezero Lake) lacked S. etruscus (SIMEONOV et al. 1981). The species was not recorded by the previous studies on the prey spectra of other owls in this part of Bulgaria either (SIMEONOV 1968, 1981, 1983 a, b, 1985). In contrast to the hitherto data, remains of S. etruscus were found in the pellets of Bubo bubo (LINNAEUS, 1758) and Asio otus (LINNAEUS, 1758), collected in this part of the country during the last years (MILCHEV, unpubl. data). On the basis of these evidences it can be supposed either that S. etruscus is a recent invader in the area, or that it remained undetected for a long time. Although in some cases the correct determination of the shrew material in owl pellets in some earlier studies is questionable (e. g. SIMEONOV 1985), the size distinction of S. etruscus is of such magnitude, that its confusion with Crocidura spp. seems unlikely. So, the first possibility seems more plausible. This interpretation is consistent with the data from Turkmenia, where it was found that the species has probably extended its areal during the last decades (KOLODENKO 1971, 1975). It is usually claimed that S. etruscus prefers relatively humid places overgrown with ruderal vegetation and shrubs. This is especially true for the more arid Mediterranean areas. For instance, in the coastal areas of Algeria its remains constituted between 5 and 16 % of the total numbers of insectivores, while further inlands it was less numerous

308

(RZEBIK-KOWALSKA 1985). Within the studied region, however, the species tends to occur in a greater percentage in the relatively drier sites and its greatest amount in the shrew spectra is in the inland localities, not along the sea shore, where the climate is more humid. In Turkmenia, the species also tends to occur in drier sites (KOLODENKO 1971). Based on the results of the above analyses and comparisons, it can be supposed that S. etruscus has invaded this part of the country through the valleys of the Maritsa and Tundzha rivers not via the sea coast. These valleys offer a wide and continuous climatic connection to the adjacent warmer Mediterranean parts of the Thracian Plain. Moreover, large territories along these rivers and the adjacent plains and hills are deforested and dominated by xerotherm shrubby sub-Mediterranean vegetation. In agreement with this assumption, the greatest percentages of S. etruscus in the prey spectra (more than 5 %) were concentrated in the western portion of the studied area (UTM squares MH20, MG39, MG48, MG59, MH50, MG68, MG76). A secondary concentration of large percentages occurred in the nearby western part of the Burgas lowland (UTM squares MG89, NH01, NH00), while the percentages near the sea coast were generally low. It can be supposed that the large invasion along the sea coast is prevented by the relatively narrow climatic connection and the largely forested landscapes. The available data on the occurrence of Myomimus roachi (BATE, 1937) reveal similar spatial pattern in SE Bulgaria. It can be suggested that these two species represent the northerly distributional routes of the more xerophilous Mediterranean species. In contrast, the more mesophilous Euxinian species, such as Talpa levantis THOMAS, 1906, invaded the country through the more humid and forested areas of the Strandzha Mts. (POPOV, MILCHEV 2001). One unchanged conclusion with respect to our previous study (POPOV, NJAGOLOV 1991) is that the distribution of S. etruscus is related to the warmest parts of Southern Bulgaria and most probably is restricted within the 120 C isotherm (KAHMANN, ALTNER 1956). The role of other factors such as “the type of biotope, precipitation, presence of snow cover, competition” (VOHRTALIK, SOFIANIDOU 2000) although potentially important, still needs to be proved. Morphometry. The variability within our large sample corresponds to the scarce recent and subfossil data from the Near East (KOCK, NADER 1983), The East - Mediterranean islands (REUMER, PAYNE 1986, REUMER, OBERLI 1988, BESENECKER et al. 1972), the Island of Korfu (NIETHAMMER 1962), and Caucasus (SOKOLOV, TEMBOTOV 1989). In previous studies, the occurrence and size of S. etruscus were correlated to temprerature conditions, latitude, and competition (KAHMANN, ALTNER 1956, LÓPEZFUSTER et al. 1979, SANS-COMA et al. 1981, 1985, CONTOLI et al. 2000). It was found out that the size of S. etruscus is partly controlled by abiotic factors, such as mean annual temperature (CONTOLI et al. 2000) and isolation (SANS-COME et al. 1985), while some biotic factors, like competition with Crocidura spp., do not seem to be important (CONTOLI et al. 2000). The studied population from SE Bulgaria fits well in these patterns. In nearly all skull dimensions it is larger than the population from the warmer regions of S Europe. The exceptions concern two characters - the length of mandible and the postglenoid width. These measurements had relatively low mean values in the Bulgarian population in contrast to the patterns observed in the other measurements. It can be supposed that these differences reflect the regional aloofness of the studied population, not related to the environmental conditions. Received: 12.07.2004 Accepted: 10.11.2004

309

References ANDREWS P. 1990. Owls, caves and fossils. London, Nat. Hist. Mus. publ., VIII + 231 p. ALIVIZATOS H., V. GOUTNER 1999. Winter diet of the Barn owl (Tyto alba) and Long-eared owl (Asio otus) in Northeastern Greece: A comparison. - Journal of Raptor Research, 33: 160-163. BESENECKER H., F. SPITZENBERGER AND G. STORCH 1972. Eine Holozäne Kleinsäuger-Fauna der Insel Chios, Ägäis (Mammalia: Insectivora, Rodentia). - Senckenbergiana biologica, 53 (3-4): 145-177. BRAAK C. J. F., TER 1987. 5. Ordination. - In: Jongman R. H. et. al. (eds). Data Analysis in Community and Landscape Ecology. Wageningen, Pudoc., 91-170 p. BRAAK C. J. F., TER 1987b. CANOCO a FORTRAN program for canonical community ordination by [partial][detrended][canonical] correlation analysis, principal components and redundancy analysis. Wageningen, Agricultural Mathematics Group, 95 p. BRAAK C. J. F., TER 1990. Update notes: CANOCO version 3.10. Wageningen. Agricultural Mathematics Group. 35 p. CONTOLI L. 1975. New data on the role of the Barn Owl in the control of mammals in Central Italy. - In: Proc. of World Conference on birds of prey, Vienna, 280-282 p. CONTOLI L., A. DE MARCHI AND D. PENKO 1977. Sul sistema trofico “Micromamimmiferi Tyto alba” nel parco “Boschi di Carrega” (Parma). L’Ateneo Parmanese. - Acta Naturalia, 13 (4): 705-728. CONTOLI L., C. BATTISTI AND A. BUSCEMI 2000. On morphology of Suncus etruscus (Mammalia, Soricidae): a negative relation between size and temperature. - Italian Journal of Zoology, 67: 329-332. HERRERA C. M. 1974. Régimen alimenticio de Tyto alba en España Sudoccidental. - Ardeola 19: 359-394. HERSHKOVICH E. 1984. Agroclimatic resources of Bulgaria. Sofia, Publ. House of the Bulg. Acad. of Sciences, 115 p. (In Bulgarian, English summary). KAHMANN H., H. ALTNER 1956. Die Wimperspitzmaus Suncus etruscus (SAVI, 1822) auf der Insel Korsika und ihre circummediterrane Verbreitung. - Säugetierkundlische Mitteilungen, 4: 72-81. KOCK D., I. A. NADER 1983. Pygmy shrew and rodents from the Near East (Mammalia: Soricidae, Rodentia). - Senckenbergiana biologica, 64: 13-23. KOLODENKO A. I., 1971. On distribution and ecology of Suncus etruscus Savi, 1822 in Turkmenia. - Proceedings of Akademy of Turkmenian SSR (biol.), 3: 60-643. KOLODENKO A. I., 1975. New data on the distribution of Suncus etruscus Savi, 1822 in Turkmenia. - Proceedings of Akademy of Turkmenian SSR (biol.), 1: 92-93. KORPIMÄKI E., K. NORRDAHL 1989. Avian and mammalian predators of shrews in Europe: regional differences, between-year and seasonal variation, and mortality due to predation. - Annales Zoologici Fennici, 26: 389-400. KRYŠTUFEK B., V. VOHRALIK 2001. Mammals of Turkey and Cyprus. Introduction, checklist and Insectivora. Koper. 140 p. LAAR V. VAN, S. DAAN 1967. The Etruscan shrew, Suncus etruscus (Savi, 1822) found on Samos, Greece. - Zeitschrift für Säugetierkunde, 32: 174-175. LÓPEZ-FUSTER M., V. SANS-COMA, I. VESMANIS AND R. FONS 1979. Sorbe el musgaño enano, Suncus etruscus (SAVI, 1822), en Catalunya Iberica. (Mammalia, Insectivora). - Miscelanea Zoologica, 5: 109-124. LOVARI S., A. RENZONI AND R. FONDI 1976. The predatory habits of the Barn Owl (Tyto alba SCOPOLI) in relation to the vegetation cover. - Bolletino di zoologia, 43: 173-191. NIETHAMMER J. 1962. Die Säugetiere von Korfu. - Bonner zoologische Beiträge, 13: 1-49.

310

NIETHAMMER J. 1989. Gewöllinhalte der Schleiereule (Tyto alba) von Kos und aus Südwestanatolien. - Bonner zoologische Beiträge, 40: 1-9. POPOV V. V. 2000. Epigeobiont animal assemblages from two landscapes of the Bulgarian Black Sea coast: relationships to environmental gradients, assemblage structure and biodiversity. III. Small mammals (Mammalia: Insectivora, Rodentia). - Acta zoologica bulgarica, 52 (3): 79-96. POPOV V., B. MILTCHEV 2001. New data on the morphology and distribution of Talpa levantis Thomas, 1906 (Mammalia: Insectivora) in Bulgaria. - Acta zoologica bulgarica, 53 (3): 79-94. POPOV V. V., K. K. NIJAGOLOV 1991. A new record of Suncus etruscus (SAVI, 1822) (Mammalia, Soricidae) from Bulgaria. - Acta zoologica bulgarica, 41: 69-71. REUMER J. W. F., U. ORBELI. 1988. Shrews (Mammalia: Soricidae) from a Bronze age deposit in Cyprus, with the description of a new subspecies. - Bonner zoologische Beiträge, 39 (4): 305-314. REUMER J. W. F., S. PAYNE 1986. Notes on Soricidae (Insectivora, Mammalia) from Crete. II. The shrew remains from Minoan and Classical Kommos. - Bonner zoologische Beiträge, 37 (3): 173-182. RZEBIK-KOWALSKA B. 1985. Records of Suncus etruscus in Algeria. - Acta Zoologica Fennica, 173: 225-226. SANS-COMA V., R. FONS AND I. VESMANIS 1981. Eine morphometrische Untersuchungen am Schädel der Etruskerspitzmaus, Suncus etruscus (SAVI, 1822) aus Süd-Frankreich (Mammalia, Insectivora, Soricidae). - Zoologische Abhandlungen, 37 (1): 1-31. SANS-COMA V., J. A. ALCOVER AND M. J. LÓPEZ-FUSTER 1985. Morphometrischer Vergleich rezenter und subfossiler Etruskerspitzmäuse Suncus etruscus (SAVI, 1822) von der Insel Sardinien. - Säugetierkunde Mitteilungen, 32: 151-158. SIMEONOV S. D. 1968. Materialen über die Nahrung des Steinkauzes (Athene noctua SCOPOLI) in Bulgarien. - Fragmenta Balcanica, 6 (17): 157-165. SIMEONOV S. D. 1981. Studies on the nesting and the diet of the Scops Owl (Otus scops (L.)) in Bulgaria. - Ekologija, Sofia, 9: 51-58. (In Bulgarian, English summary). SIMEONOV S. D. 1983a. New data on the diet of the Little Owl (Athene noctua (SCOP.)) in Bulgaria. - Ekologija, Sofia, 11: 53-60. (In Bulgarian, English summary). SIMEONOV S. D., 1983b. Studies in the diet of the Short-earred Owl (Asio flammeus (PONT.)) in Bulgaria. - Ekologija, Sofia, 11: 61-66. (In Bulgarian, English summary). SIMEONOV S. D. 1985. A study on nest biology and food range of the Tawny Owl (Strix aluco L.) in Bulgaria. - Ekologija, Sofia, 17: 42-48. (In Bulgarian, English summary). SIMEONOV S. D., T. M. MICHEV AND P. S. SIMEONOV 1981. Materials on the nesting distribution and the diet of the Barn Owl (Tyto alba (SCOPOLI) in Bulgaria. - Ekologija, Sofia, 8: 49-54. (In Bulgarian, English summary). SCHMIDT E., 1973. Die Nahrung der Schleiereule (Tyto alba) in Europa. - Zeitschrift für angewandte Zoologie, 60: 43 - 70. SOKOLOV V. E., A. K. TEMBOTOV 1989. Mlecopitajuscie Kavkaza: Nasekomojaddnye [Mammals of Caucasus: Insectivores]. Moskva, Nauka. (In Russian). SPITZENBERGER F. 1970. Erstnachweise der Wimperspitzmaus (Suncus etruscus) für Kreta und Kleinasien und die Verbreitung der Art im südwestasiatischen Raum. - Zeitschrift für Säugetierkunde, 35: 107-113. SPITZENBERGER F. 1978. Die Säugetierfauna Zyperns. Teil I: Insectivora und Rodentia. Annalen des Naturhistorischen Museums in Wien, 81: 443-446. TAYLOR I. 1994. Barn Owls: predator-prey relationships and conservation. Cambridge, Cambridge Univ. Press. 304 p.

311

TEMME, M. 2000., Gewöllinhalte der Schleiereule Tyto alba bei Albertnoa, Provinz Alentejo, Portugal. - Ornithologische Mitteilungen, 52 (6/7): 219-228. VOHRALIK V. 1985. Notes on the distribution and the biology of small mammals in Bulgaria (Insectivora, Rodentia). I. - Acta Universitas Carolinae - Biologica, 1981: 445-461. VOHRALIK V., T. S. SOFIANIDOU 2000. New records of Suncus etruscus (Mammalia: Insectivora) in Bulgaria and Greece and distribution of the species in the Balkans. - Lynx, n.s., 31: 143-148. YORDANOVA M., D. DONTCHEV 1997. Geography of Bulgaria. Sofia, Academic Publishing House “Prof. Marin Drinov”, 730 p. (In Bulgarian, English summary). ZAR J. H. 1999. Biostatistical analysis. New Jersey, Prenice-Hall, Inc.

Ëàíäøàôòíè îñîáåíîñòè íà ðàçïðîñòðàíåíèåòî è êðàíèîìåòðèÿ íà Suncus etruscus (Mammalia: Insectivora, Soricidae) â Þãîèçòî÷íà Áúëãàðèÿ Â. Ïîïîâ, Á. Ìèë÷åâ, Â. Ãåîðãèåâ, Õ. Äèìèòðîâ, Ö. ×àñîâíèêàðîâà (Ðåçþìå) Ïðîñëåäåíî å ðàçïðîñòðàíåíèåòî íà S. etruscus â Þãîèçòî÷íà Áúëãàðèÿ íà îñíîâàòà íà ïîãàäêè îò çàáóëåíà ñîâà. S. etruscus áåøå óñòàíîâåí â 39 îò îáùî 41 èçñëåäâàíè íàõîäèùà íà ïîãàäêè. ×ðåç êàíîíè÷åí àíàëèç íà ãëàâíèòå êîìïîíåíòè å îöåíåíî âëèÿíèåòî íà 8 åêîëîãè÷íè ïàðàìåòúðà, îïèñâàùè îòíîñèòåëíàòà ïëîù íà îñíîâíèòå ìåñòîîáèòàíèÿ â ëîâíàòà òåðèòîðèÿ íà ñîâàòà êàêòî è ïàðàìåòðè, îòðàçÿâàùè êëèìàòè÷íèÿ è ëàíäøàôòåí êîíòåêñò, âúðõó ñúîòíîøåíèåòî íà âèäîâåòå çåìåðîâêè â õðàíèòåëíèÿ ñïåêúð íà õèùíèêà. Óñòàíîâåíî å, ÷å îòíîñèòåëíàòà ïëîù íà âëàæíèòå ìåñòîîáèòàíèÿ â ðàìêèòå íà ëîâíàòà òåðèòîðèÿ è ñóìàòà íà òåìïåðàòóðèòå îêàçâàò çíà÷èìî âëèÿíèå âúðõó ñòðóêòóðàòà íà õðàíèòåëíèÿ ñïåêúð îò çåìåðîâêè. Îòíîñèòåëíèÿ äÿë íà S. etruscus å ïîëîæèòåëíî êîðåëèðàí ñ òåìïåðàòóðèòå. Íà îñíîâàòà íà ïîëó÷åíèòå ðåçóëòàòè è ñðàâíåíèÿ ñ ïóáëèêóâàíè äàííè ñå èçêàçâà õèïîòåçàòà, ÷å S. etruscus îòñêîðî ñå ðàçïðîñòðàíÿâà â ðàéîíà. Ïúòÿò íà ðàçñåëâàíå âåðîÿòíî ñëåäâà äîëèíèòå íà Ìàðèöà è Òóíäæà è çàâèâà íà èçòîê ñåâåðíî îò Ñàêàð è Ñòðàíäæà, äîñòèãàéêè ×åðíîìîðñêîòî êðàéáðåæèå. Ïðîöåíòíîòî ó÷àñòèå íà S. etruscus â õðàíèòåëíèÿ ñïåêòúð íà çàáóëåíàòà ñîâà å íàé-÷åñòî ïî-íèñêî â ñðàâíåíèå ñúñ ñðåäèçåìíîìîðñêèòå ÷àñòè íà Åâðîïà. Ñðàâíåíèÿòà íà ïîëó÷åíèòå ìîðôîìåòðè÷íè äàííè ñ òàêèâà îò ëèòåðàòóðàòà ïîêàçâàò, ÷å ïîïóëàöèÿòà îò Þãîèçòî÷íà Áúëãàðèÿ ñå îòëè÷àâà ñúñ ñðàâíèòåëíî ãîëåìè ðàçìåðè íà ÷åðåïà â ðàìêèòå íà äîáðå ïðîÿâåíà êëèíàëíà èçìåí÷èâîñò îò þã íà ñåâåð â ñúîòâåòñòâèå ñ ïðàâèëîòî íà Áåðãìàí.  ðàìêèòå íà àíàëèçèðàíèÿ íàáîð îò äàííè â ãåîãðàôñêè àñïåêò äúëæèíàòà íà êîñòíîòî íåáöå è çèãîìàòè÷íàòà øèðèíà íàðàñòâàò çíà÷èìî ñ íàìàëÿâàíå íà ñðåäíàòà ãîäèøíà òåìïåðàòóðà.

312