HACQUETIA 9/2 • 2010, 207-220 DOI: 10.2478/v10028-010-0010-2 THE ALLIANCES ARRHENATHERION, CTNOSURION AND TRIFOLION MEDII IN WESTERN BULGARIA - ENVIRONMENTAL GRADIENTS AND ECOLOGICAL LIMITATIONS Nikolay VELEV^, Iva APOSTOLOVAi, Zuzana ROZBROJOVÄ2'3 & Petra HÄJKOVÄ2'3 Abstract The vegetation of mesic grasslands distributed in the central part of Western Bulgaria is studied. A significant number of 533 releves were made following the Braun-Blanquet approach. Classification and ordination of the vegetation was performed. The vegetation is classified up to alliance level. The result of the classification are four groups which correspond to the alliances Arrhenatherion elatioris, Cynosurion cristati and Trifolion medii. Within Cynosurion alliance two well distinct subgroups are observed, differing in moisture of the substrate - provisionally called "wet" and "dry" Cynosurion. Ellenberg Indicator Values are used for assessment of ecology of the habitats. The pasture/mowing management of the studied vegetation types is commented. Key words: Arrhenatherion, Bulgaria, classification, Cynosurion, formal definitions, mesic grasslands, ordination, Trifolion medii. Izvleček Proučevali smo vegetacijo mezofilnih travnikov razširjeno v srednjem delu zahodne Bolgarije. Naredili smo pomembno veliko število (553) popisov po Braun-Blanquetovem pristopu in jih analizirali s klasifikacijo in ordinacijo. Vegetacijo smo nato uvrstili do nivoja zveze. Rezultat klasifikacije so štiri skupine, ki odgovarjajo zvezam Arrhenatherion elatioris, Cynosurion cristati in Trifolion medii. Znotraj zveze Cynosurion smo izločili dve posebni skupini popisov, ki se ločita po vlažnosti tal in ju provizorno imenujemo "vlažni" in "suhi" Cynosurion. Za opis ekologije rastišč smo uporabili Ellenbergove Indikatorske Vrednosti. Komentirali smo tudi gospodarjenje obravnavanih vegetacijskih tipov s košnjo oziroma pašo. Ključne besede: Arrhenatherion, Bolgarija, klasifikacija, Cynosurion, formalne definicije, mezofilni travniki, ordinacija, Trifolion medii. 1. INTRODUCTION berg (1988) has suboceanic submeridional distribution. In South-Eastern Europe this vegetation The mesic grasslands encompass mostly mead- type is less frequent as compared to Western Eu- ows of high productivity in the lowlands up to rope due to limited moisture. Horvat et al. (1974) the sub-mountain level, that are of significant explain also this limitation with distribution value for practical and nature conservation pur- boundaries for many diagnostic species in the reposes. We considered as target vegetation during gion. The diversity of these community types in the field work the content of Arrhenatheretalia ela- Central Europe has been well studied (Ellmauer tioris Tüxen 1931 order which according to Ellen- & Mucina 1993, Dierschke 1995, 1997, Chytry 1 Institute of Botany, Bulgarian Academy of Sciences, Acad. Georgi Bonchev Street, Bl. 23, 1113 Sofia, Bulgaria, e-mail: nvelev@bio.bas.bg; iva@bio.bas.bg 2 Institute of Botany, Czech Academy of Sciences, Pof^č^ 3b, CZ-60300 Brno, Czech Republic 3 Department of Botany and Zoology, Faculty of Science, Masaryk University, Kotlafska 2, CZ-61137 Brno, Czech Republic, e-mail: zuzkaroz@sci.muni.cz; buriana@sci.muni.cz hacq,uetia 9/2 . 2010, 207-220 2007a, Janišova 2007). For the Balkan Peninsula, a significant amount of information about the vegetation concerned is provided by Horvat et al. (1974), Kojič et al. (1998) and Zelnik (2007). Development and implementation of the recent nature conservation initiatives, namely designation of High Nature Value grasslands (see Veen et al. 2009) requires better knowledge on the species composition, ecological conditions and syntaxo-nomical diversity of this vegetation. For the territory of Bulgaria there was no comprehensive information about the diversity of mesic grasslands based on the floristic-ecological studies (see Apostolova & Slavova 1997). Tzonev et al. (2009) report three associations and one community, assigned to Arrhenatheretalia order, published in the last ten years for the country. The deficiency in the knowledge of Bulgarian meadows has started to be rectified by recent studies on Cy-nosurion cristati (Dimitrov 2001, 2004; Apostolova & Meshinev 2006; Pavlov et al. 2006), Arrhenath-erion elatioris (Pavlov et al. 2006; Apostolova et al. 2007) and Trifolion medii (Meshinev et al. 2005). Despite the remarkable progress in implementation of the Braun-Blanquet approach in Bulgaria in the past decade it could be argued that the classification of mesic grasslands is still at the beginning. The aim of the current study is to establish the high rank syntaxa of the mesic meadows and pastures from the central part of Western Bulgaria, and to provide coherent environmental characteristics for their habitats. 2. MATERIAL AND METHODS 2.1 Study area The studied area is situated in the central part of Western Bulgaria (Figure 1). The total observed area is 10 796 sq. km., which is approximately a 1/10 part of the territory of the country. According to the climatic division of Bulgaria (Velev 2002) the studied area falls into temperate-conti-nental and transitional-continental regions. The climate in these regions is characterized by warm summers and cold winters and by a high annual amplitude of air temperature. An ombrothermic climatic diagram from Sofia observatory (situated at the centre of the investigated area) is given in Figure 2 according to Lieth et al. (1999). The base rock is mainly silicate. ''-J ROMANIA ............... shribia'^: VTJ. BULGARIA Iv Black Sea \ ^^ s. ^^ ..... \ ""■-''•v.. 100km OREKOfc 1 Figure 1: Map of Bulgaria. The studied area is marked in dark gray. Slika 1: Zemljevid Bolgarije. Obravnavano območje je sivo označeno. BULGARIA 42.65°N / 23.38°E / 586m SOFIA (OBSERV.) [43-43] +10.0°C 571mm -10- Figure 2: Ombrothermic climatic diagram from the studied area. Slika 2: Ombrotermični klimatski diagram obravnavanega območja. 2.2 Field sampling The sample plots were set in a manner to cover uniformly the whole area. Altogether 533 releves within the period 2007-2009 were obtained. During the field sampling we focused on the vegeta- tion that ecologically and floristically seemed to belong to Arrhenatheretalia order. The releves are sampled following the classical approach (Braun-Blanquet 1965, Westhoff & van der Maarel 1973). This releve type is suitable both for classification and ordination (Whittaker 1973). The form of the sample plots is square with an area of 16 sq. metres which is recommended for mesic meadows (Mueller-Dombois & Ellenberg 1974, Knapp 1984, Kent & Coker 1992, Chytry & Otypkova 2003, van der Maarel 2005, Otypkova & Chytry 2006). The sample plots are chosen in the most homogeneous part of the community. Species are estimated by the expanded combined scale for cover/abundance of Braun-Blanquet (Barkman et al. 1964, van der Maarel 1979, Parolly 2003). This scale is a good alternative to the classical (7-point) scale of Braun-Blanquet, which is relatively insensitive in the range covered by category "2". The expanded 9-point scale gives greater importance to the abundance (van der Maarel 1979, Parolly 2003). soil samples were taken from the depth range of 0-10 cm. Determination of pH was provided by pH-meter "Jenway 3310" in H20 solution with a soil:water ratio of 1:5, which macerates for at least 18 hours (ISO 10390: Soil quality - Determination of pH). After every 10 soil samples the pH-meter was calibrated by HANNA Instruments buffer solutions (pH 4.01; pH 7.01; pH 10.01). Pasture management is estimated categorically: 0 - not grazed, 1 - low intensity of grazing, 2 -moderate intensity of grazing, 3 - high intensity of grazing. Mowing management is estimated by two categories: 0 - not mown, 1 - mown. 2.3 Data processing All the releves are stored in TURBOVEG database (Hennekens & Schaminee 2001). Before conducting any analysis the data were normalized by "square root transformation" (McDonald 2008). The classification was made using the program TWINSPAN (Hill 1979) and is performed within the software package JUICE 7.0 (Tichy 2002, Tichy & Holt 2006). Formal definitions were assessed for all alliances using combinations of sociological species groups and species dominance. The sociological groups were generated by the Cocktail method (Bruelheide 1995, 2000) within all releves and contain species with the highest positive fidelity: Sociological group Trifolium montanum: Briza media, Chamaespartium sagittale, Danthonia alpina, Helianthemum nummularium, Trifolium mon-tanum; Sociological group Leontodon autumnalis: Agros-tis capillaris, Anthoxanthum odoratum, Cynosurus cristatus, Holcus lanatus, Leontodon autumnalis; Sociological group Trisetum flavescens: Medi-cago lupulina, Trifolium campestre, Trisetum flaves-cens, Vicia cracca, V grandiflora. Delimitation of syntaxa is obtained using the logical operators: AND, oR and NoT. These operators determine which sociological group is to be presented or absent in a releve, so that the releve can be referred to the relevant syntaxon (Bruelheide 1997, Chytry 2007b). Indirect Gradient Analysis (by Detrended Correspondence Analysis - DCA) using the software product CANOCO 4.5 (ter Braak & Šmilauer 2002, Lepš & Šmilauer 2003) is applied. Environmental conditions in the habitats were assessed using the Ellenberg Indicator Values (EIV) according to Ellenberg et al. (1992). All variables (EIV, total cover, altitude, soil pH) are summarized and presented by seven box and whisker plots. The observed differences in variables among the studied vegetation units were tested by one-way ANOVA with Bonferroni (in case of equal variances) and Tamhane (in case of unequal variances) post-hoc tests. The homogeneity (equality) of variance assumption was tested by Levene's test (Levene 1960). A multiple comparison of the summarized variables is revealed in order to show which of them differ statistically significantly among the studied vegetation units. The univariate statistic analyses were performed by SPSS 16.0 software. The species in the synoptic table (Table 1) are represented by two indicators: Fidelity measure, expressed by the Phi-coefficient (Chytry et al. 2002) and Constancy, expressed in percentages. The Phi-coefficient values depend on the size of the syntaxa. This problem is avoided by virtual standardizing of the volume of the obtained clusters using the software package JUICE. Phi-coefficient calculated under the standardized database is independent of the actual differences in the volume of any group (Tichy & Holt 2006). The values of Phi-coefficient are corrected using Fisher's exact test. This is done in order to eliminate the values of Phi-coefficients which are statistically insignificant. The quality of the classified syntaxa is ex- pressed by Sharpness and Uniqueness indexes. A vegetation unit has a high sharpness index if most of its species are frequent in it, while they are rare or absent in the other clusters in the database. The Uniqueness index searches syntaxa which are unique in a database. It shows whether there are similar vegetation units of the same rank in the dataset. These two indices consider the diagnostic species. The analyses of these indicators express well-defined and poorly defined vegetation units (Chytry & Tichy 2003). Dominant and constant species are listed for every alliance. Species with coverage above 25 % in 5 % of the releves in any cluster are considered as dominant. Constant species are those having at least 50 % presence in a cluster. Before determination of formal definitions, dominant and constant species of the two subgroups of Cynosurion alliance were merged into one cluster. 2.4 Nomenclature Taxonomy of the species follows Kozhuharov (1992) and Delipavlov & Cheshmedjiev (2003). Diagnostic species are selected according to Resmeritä (1970), Bo§caiu (1971), Dihoru (1975), Lopez (1978), Zuidhoff et al. (1995), Dierschke (1997), Micina (1997), Sanda et al. (1997), Čarni (1997, 2005), Pop et al. (2002), Šeffer et al. (2002), Chytry & Tichy (2003), Dutoit (2003), Valachovič (2004), Čarni et al. (2000, 2005), Gandullo & Faggi (2005), Meshinev et al. (2005), Snowarski (2005), Bärbos (2006), Hajkova et al. (2007) and Stančic (2008). Biogeographical affiliation of the species follow Oberdorfer (1994) and for the Balkan endemics Assyov & Petrova (2006). versity per sample plot is lowest (36) as compared to the other studied groups, the total species sampled within this alliance numbers 411. Arrhenath-erion alliance is defined by a very low Sharpness index (9.14), which is a result of many generalist species presented. Similar could be stated for the "dry" Cynosurion which has a Sharpness index of 3.78. On the other hand the "wet" Cynosurion includes species which are not presented in the rest of the data, and the Uniqueness index has the highest value of 42 within the studied samples. -1 SAMPLES oTrifolionmedii □ dryCynosurion O wetCynosurion □ Arrhenatherion ENV. VARIABLES Figure 3: Indirect gradient analysis. Biplot diagram of DCA. The main axes correlate with the gradients of moisture and soil reaction. ED - Empirical Data, EIV - Ellenberg Indicator Values. Slika 3: Indirektna gradientna analiza. DCA dvorazsežnostni diagram. Glavni osi z gradientoma vlažnosti in reakcije tal. ED - empirični podatki, EIV - Ellenbergove Indikatorske Vrednosti. 3. RESULTS 3.1 Classification and ordination The TWINSPAN classification and the DCA-ordination analysis of the releves distinguished four groups (Table 1, Figure 3) which correspond to the alliances Trifolion medii, Cynosurion cristati with two well delimited subgroups with respect to the soil moisture provisionally called "dry" and "wet" type and Arrhenatherion elatioris. The largest number of releves is confined to the Arrhenatherion elatioris. Although species di- 3.2 Ecological differences among alliances Trifolion medii occupies habitats with most dry soils, the lowest values of pH and with a minimum of nutrients (Figure 5). This alliance shows a resemblance to "dry" Cynosurion. The latter, in turn, shows the least expressed specifics for environmental factors in comparison with the other three vegetation units. The wettest sites are occupied by the "wet" Cynosurion. Arrhenatherion alliance is well distinguished by many variables compared to the other vegetation units (Table 3). Its habitats are located at lowest altitudes and are characterized by their highest values of temperature and soil pH (Figure 5). 3.3 Description of particular alliances In order to present the characteristics of the obtained alliances in the studied area a short description is provided for each of them: Trifolio-Geranietea Müller 1962 Origanetalia vulgaris Müller 1962 Trifolion medii Müller 1962 Diagnostic species: Centaurea jacea, Trifolium medium, T. montanum. Constant species: Achillea millefolium, Agrostis capillaris, Anthoxanthum odoratum, Briza media, Chamaespartium sagittale, Danthonia alpina, Festuca nigrescens, F. valesiaca, Filipendula vulgaris, Galium verum, Leontodon hispidus, Lotus corniculatus, Plantago lanceolata, Rumex acetosa, Stellaria graminea. Dominant species: Danthonia alpina. Formal definition: Group Trifolium montanum NOT Group Leontodon autumnalis. Soil pH optimum: medium acidic (5.4-6.0). Altitude range: 798-1242 m. Base rock: silicate. This is the best separated alliance in our database. The Sharpness index (24.08) and the 70606040302010- n LL 3 CD 3 ^^ ^ g srt» is1 Arrhinathirion Cynosurion Cynosurion lry typi wit typi Trifolion milii Figure -l:. The mowing/grazing; iraarpagtument of the stimdied vegeti^tion tyjses. Slika 4: Gospodarjenje (paša/košnja) s proučevanimi vegetacijskimi tipi. Uniqueness index (38.80) values prove its distinctness within the data set. It contains a relatively small number of releves but the species diversity is high - total species number is 220 and average species per releve is 39 (Table 2). The vegetation is used for hay-making and pasturing (Figure 4). Molinio-Arrhenatheretea Tüxen 1937 Arrhenatheretalia elatioris Tüxen 1931 Cynosurion cristati Tüxen 1947 Diagnostic species: Agrostis capillaris, Antho-xanthum odoratum, Cichorium intybus, Cynosurus cristatus, Leontodon autumnalis. Constant species: Achillea millefolium, Agrostis capillaris, Anthoxanthum odoratum, Alopecurus pra-tensis, Convolvulus arvensis, Cynosurus cristatus, Dactylis glomerata, Festuca pratensis, Galium verum, Holcus lanatus, Leontodon hispidus, Lotus corniculatus, Poa pratensis, Plantago lanceolata, Ranunculus acris, Rhinanthus rumelicus, Rumex acetosa, Stellaria graminea, Taraxacum officinale, Trifolium pratense, T. repens. Dominant species: Agrostis capillaris. Formal definition: Group Leontodon autumna-lis NOT Group Trisetumflavescens NOT Chamaespartium sagittale cover > 5 % NOT Trifolium medium cover > 5 %. Soil pH optimum: from medium acidic to slightly acidic (5.5-6.3). Altitude range: 359-1629 m. Base rock: silicate. The TWINSPAN divides these releves into two groups which share the diagnostic species of the alliance. Irrespective of this division, the two groups relate to Cynosurion. The floristic similarity can be seen in Figure 3 where the two groups are not clearly separated but overlap to a large extent, unlike the other two alliances. The vegetation of alliance Cynosurion is used as pastures and hay-making meadows. In the dry subgroup grazing and mowing are dealt evenly, but in the wet subgroup there is observed a significant prevalence of mowing. This vegetation has the highest total cover (Figure 5), which emphasizes its high agricultural value. The "dry" Cynosurion is the only vegetation type managed at some places by an intensive grazing regime (Figure 4). Arrhenatherion elatioris Luquet 1926 Diagnostic species: Cirsium arvense, Medicago lupulina, Poa pratensis, Silene vulgaris, Trifolium campestre, Trisetum flavescens, Vicia cracca. 1500- 1000- 100- Trifolionmedii dryCynosurion wetCynosurion Arrhenatherion 7- 6- 5- 4- Soil reaction (EM) _ O O O < c 3 3 Trifolion medii dry Cynosurion wet Cynosurion Arrhenatherion Trifolionmedii dryCynosurion wetCynosurion Arrhenatherion Trifolion medii dry Cynosurion wet Cynosurion Arrhenatherion Trifolionmedii dryCynosurion wetCynosurion Arrhenatherion Trifolionmedii dryCynosurion wetCynosurion Arrhenatherion Figure 5: Box and whiskers plots of the most important variables. Slika 5: Graf (škatla in brki) najpomembnejših spremenljivk. Constant species: Achillea millefolium, Agrimo-nia eupatoria, Arrhenatherum elatius, Cirsium ligu-lare, Convolvulus arvensis, Dactylis glomerata, Ely-mus repens, Festuca pratensis, F. valesiaca, Galium verum, Lotus corniculatus, Plantago lanceolata, Poa pratensis, Taraxacum officinale, Trifolium campestre, T. pratense, T. repens, V. grandiflora. Dominant species: Arrhenatherum elatius. Formal definition: Group Trisetum flavescens not Group Leontodon autumnalis NoT Group Trifolium montanum. Soil pH optimum: from medium acidic to slightly acidic (5.9-6.7). Altitude range: 472-1143 m. Base rock: in 90 % of the localities is silicate and in 10 % is limestone. Communities confined to Arrhenatherion alliance are widespread in the studied area. They occupy mainly the lowest altitudes, close to settlements and agricultural fields. Some sample plots are arable lands, abandoned since 1990s, which have been recovered by the species pool of the adjacent not utilized areas. This is a reason for high fidelity coefficients for Lathyrus aphaca, Dipsacus laciniatus, Vicia pannonica, V. grandiflora, Convolvulus arvensis (Table 1). Currently these grasslands are utilized as pastures and haymaking meadows with a noticeable prevalence of mowing (Figure 4). 4. DISCUSSION The mesic grasslands in the lowlands and hilly regions of western Bulgaria are classified into two distinct classes (Trifolio-Geranietea Müller 1962 and Molinio-Arrhenatheretea Tüxen 1937). It was not surprising that within the mesic grasslands the Trifolion medii alliance was also recognized, because it is pointed out by Čarni (2005) as "the most humid and rich in nutrients" representative of the class Trifolio-Geranietea. The presence of Trifolion medii within the vegetation targeted in this study reveals its ecological similarity with Arrhenatherion and Cynosurion. The fringe vegetation (like Trifolion medii) is a good example of ecotone habitats that contain a significant number of species, coming from surrounding areas (Valachovič 2004). The floristic composition of fringe communities as semi-natural grasslands largely depends on human activity, and despite the very low intensity of mowing and grazing, the management preserves the grasslands from further development into shrublands. In Bulgaria, similarly to other parts of Europe, Trifolion medii communities contain species diagnostic for other syntaxa (see Table 1). This is a reason why its position into the higher syntaxa varies significantly among different authors. Čarni (1997, 2005), Rodwell et al. (2002), Dengler (2004), Valachovič (2004) and Čarni et al. (2005) refer this alliance to Trifolio-Geranietea. According to Chytry et al. (2007) the alliance belongs to Festuco-Brometea Br.-Bl. & Tüxen ex Soo 1947. Some other authors classify Trifolion medii communities into Carpin-ion betuli Issler 1931 or Fagion Luquet 1926 and also into class Molinio-Arrhenatheretea (see Čarni et al. 2000, Valachovič 2004). The close vicinity of other semi-natural vegetation types managed as pastures or meadows for a long time supports transference of many species diagnostic for Fes-tuco-Brometea and Molinio-Arrhenatheretea into the fringe communities where these species fall into similar ecological conditions. In the present study we obtained data showing that, floristically and ecologically, Trifolion medii is close to Cynosu-rion and especially to its "dry type" (Figure 3, Table 3). According to Dierschke (1999) Cynosurion cristati is characterized by a pronounced absence of good diagnostic species. This alliance has been defined negatively in comparison with the haymaking meadows of the alliances Arrhenatherion and Polygono-Trisetion and with the pastures of the alliance Mesobromion (Zuidhoff et al. 1995). Horvat et al. (1974) mention that in the Balkans the Cynosurion vegetation is managed in a different way as compared to Central Europe, where it is managed as pastures and in many cases is maintained by manuring. In Bulgaria neither of the Cy-nosurion types is enhanced by man. Nevertheless the vegetation is composed of the same species which have emphasized European or Euro-Asiatic distribution (see table 1). Only a few specific Balkan plants are included in the floral diversity. The same is obvious also for Arrhenatherion. This brings us to the conclusion that lowland mesic grasslands in the studied area do not exhibit a particular South-East European specificity. Further, this fact also allows us to use Ellenberg Indicator Values for Bulgarian mesic meadows, in spite of their origin in the Central Europe. 5. ACKNOWLEDGEMENT A part of this investigation was implemented within an exchange project of the Czech and Bulgarian Academies of Sciences (2008-2010) and the long-term research plans of Masaryk University no. MSM0021622416 and GACR 526/09/H025. We also thank the two anonymous reviewers for the helpful comments and critical remarks on a previous version of the manuscript. 6. REFERENCES Apostolova, I. & Slavova, L. 1997: Compendium of Bulgarian plant communities published during 1891-1995. "Kiril Belivanov" Inc., Sofia, 340 pp. Apostolova, I., Meshinev, T. & Petrova, A. 2007: Hay meadows with Trisetum flavescens in Bulgaria: syntaxonomy and implications for nature conservation. Phytologia Balcanica 13(3): 401-414. Assyov, B. & Petrova, A. (eds). 2006: Conspectus of the Bulgarian Vascular Flora. Distribution Maps and Floristic Elements. Ed. 3. BBF, Sofia, 453 pp. Bärbos, M. I. 2006: Montane grasslands dominated by Agrostis capillaris and Festuca rubra in Maramure§ County. I. Phytosociological analysis. Contributii Botanice 41(2): 41-52. Barkman, J., Doing, H. & Segal, S. 1964: Kritische Bemerkungen und Vorschläge zur quantitativen Vegetationsanalyse. Acta Bota-nica Neederlandica 13: 394-419. Bo§caiu, N. 1971: Flora §i vegetatia muntilor Tar-cu, Godeanu §i Cernei. Acad. R. P. R. Bucharest, 494 pp. (in Romanian). Braun-Blanquet, J. 1965. Plant sociology: The study of plant communities. (Transl. rev. and ed. By C.D. Fuller and H.S. Conard.) Hafner, London, 439 pp. Bruelheide, H. 1995: Die Grünlandgesellschaften des Harzes und ihre Standortsbedingungen. Mit einem Beitrag zum Gliederungssystem auf der Basis von statistisch ermittelten Artengruppen. Dissertationes Botanicae 244: 1-338. Bruelheide, H. 1997: Using formal logic to classify vegetation. Folia Geobotanica Phytotaxo-nomica 32: 41-46. Bruelheide, H. 2000: A new measure of fidelity and its application to defining species groups. Journal of Vegetation Science 11: 167-178. Čarni, A. 1997: Syntaxonomy of the Trifolio-Gera-nietea (saum vegetation) in Slovenia. Folia Geobotanica Phytotaxonomica 32(2): 207-219. Čarni, A. 2005: Trifolio-Geranietea vegetations in south and southeast Europe. Acta Botanica Gallica 152(4): 483-496. Čarni, A., Kostadinovski, M. & Matevski, V. 2000: "Saum" (fringe) vegetation (Trifolio-Geranie-tea) in the Republic of Macedonia. Acta Bo-tanica Croatica 59(1): 279-329. Čarni, A., Franjic, J., Šilc, U. & Škvorc, Ž. 2005: Floristical, ecological and structural diversity of vegetation of forest fringes of the Northern Croatia along a climatic gradient. Phyton 45: 287-303. Chytry, M. (ed.) 2007a: Vegetation of the Czech Republic. Grassland and Heathland Vegetation. Vol. 1. Academia, Prague, 526 pp. Chytry, M. 2007b: Project Vegetation of the Czech Republic: Preface and summary of methods. In: Chytry, M. (ed.): Vegetation of the Czech Republic. Grassland and Heathland Vegetation. Vol. 1, pp. 35-52. Academia, Prague. Chytry, M., Tichy, L., Holt, J. & Botta-Dukat, Z. 2002: Determination of diagnostic species with statistical fidelity measures. Journal of Vegetation Science 13: 79-90. Chytry, M. & Otypkova, Z. 2003: Plot sizes used for phytosociological sampling of European vegetation. Journal of Vegetation Science 14: 563-570. Chytry M. & Tichy L. 2003: Diagnostic, constant and dominant species of vegetation classes and alliances of the Czech Republic: a statistical revision. Folia Fac. Sci. Nat. Univ. Masa-ryk. Brun., Biol., 108:1-231. Chytry, M., Hoffmann, A. & Novak, J. 2007: Dry grasslands. In: Chytry, M. (ed.): Vegetation of the Czech Republic. Grassland and Heathland Vegetation. Vol. 1. Academia, Prague, pp. 371-470. Delipavlov, D. & Cheshmedzhiev, I. (eds) 2003: Key to the plants of Bulgaria. Acad. Press Agrarian Univ., Plovdiv, pp. 65-88 (in Bulgarian). Dengler, J. 2004: Klasse Trifolio-Geranietea sangu-inei T. Müller 1962 - Licht-und wärmebedürftige Saumgesellschaften und Staudenfluren magerer Standorte. In: Berg, C., Dengler, J., Abdank, A. & Isermann, M. (eds.): Die Pflanzengesellschaften Mecklenburg-Vorpommerns und ihre Gefährdung - Textband. WeisdornVerlag Jena, pp. 362-379. Dierschke, H. 1995: Syntaxonomical survey of Molinio-Arrhenatheretea in Central Europe. Colloques Phytosociologiques 23: 387-399. Dierschke, H. 1997: Molinio-Arrhenatheretea (E 1). Kulturgrasland und verwandte Vegetationstypen. Teil 1: Arrhenatheretalia. Wiesen und Weiden frischer Standorte. Göttingen, 74 pp. Dierschke, H. 1999: Short survey of Arrhenatheretalia grasslands in Germany. Annali di Bota-nica 57: 7-15. Dihoru, G. 1975: The vegetal cover of the Siriu Mountain. Edit. Acad. Press, Bucharest, 216 pp. (in Romanian). Dimitrov, M. 2001: Syntaxonomic analysis of herbaceous vegetation in the Yundola Forestry Experimental Station. In: Temniskova, D. (ed.): Sixth Natl. Conf. Bot. Univ. Sofia, Sofia pp. 263-276 (in Bulgarian). Dimitrov, M. 2004: Floristic classification of the vegetation in the Yundola Forestry Experimental Station. PhD Thesis, Univ. of Forestry Press, Sofia (in Bulgarian, unpubl.). Dutoit, T., Buisson, E., Roche, P. et alard, D. 2003: Land use history and botanical changes in the calcareous hillsides of Upper-Normandy (north-western France): new implications for their conservation management. Biological Conservation 115: 1-19. Ellenberg, H. 1988: Vegetation ecology of Central Europe. Cambridge University Press, Cambridge, 731 pp. Ellenberg, H., Weber, H., Düll, R., Wirth, W., Werner, W. & Paulissen, D. 1992: Zeigerwerte von Pflanzen in Mitteleuropa. Ed. 2. Scripta Geobotanica 18: 1-258. Ellmauer, T. & Mucina, L. 1993: Mollinio-Arrhe-natheretea. In: Mucina, L., Grabherr, G. & Ellmauer, T. (eds.): Die Pflanzengesellschaften Österreichs. Teil I. Gustav Fischer Verlag, Jena, pp. 297-401. Gandullo, R. & Faggi, A. 2005: Interpretacion sintaxonomica de los humedales del noroeste de la provincia de Neuquen, Argentina. Dar-winiana 43(1-4): 10-29. Hajkova, P., Hajek, M., Blažkova, D., Kučera, T., Chytry, M., Rezničkova, M., Šumberova, K., Černy, T., Novak, J. & Simonova, D. 2007: Meadows and mesic pastures (Molinio-Arrhe-natheretea). In: Chytry, M. (ed.): Vegetation of the Czech Republic. Grassland and Heath-land. Vol. 1. Academia, Prague, pp. 165-280. Hennekens, S. & Schaminee, J. H. J. 2001: TUR-BOVEG, a comprehensive data base management system for vegetation data. Journal of Vegetation Science 12: 589-591. Hill, M. 1979: TWINSPAN a FORTRAN for arranging multivariate data in an ordered two-way table by classification of individuals and attributes. Cornell University, Ithaca, 45 pp. Horvat, I., Glavač, V. & Ellenberg, H. 1974: Vegetation Südosteuropas. G. Fisher Verl. Stuttgart, 766 pp. ISO 10390: 1994: Soil quality - Determination of pH. Janišova, M. (ed.) 2007: Grassland vegetation of Slovak Republic - electronic expert system for identification of syntaxa. SAV, Bratislava, 263 pp. Kent, M. & Coker, P. 1992: Vegetation description and analysis. A practical approach. John Wiley & Sons, London, 336 pp. Kojic, M., Popovic, R. & Karadžic, B. 1998: A syntaxonomic Review of Vegetation in Serbia. Inst. za biol. istrazivanja "Siniša Stankovic", Belgrade, 218 pp. (in Serbian). Knapp, R. 1984: Sample (releve) areas (distribution, homogeneity, size, shape) and plot-less sampling. In: Knapp, R. (ed.): Handbook of Vegetation Science 4. Junk, The Hague, pp. 101-119. Kozhuharov, S., (ed.). 1992: Field Guide to the Vascular Plants in Bulgaria. Naouka & Izkous-tvo, Sofia, 788 pp. (in Bulgarian). Lepš, J. & Šmilauer, P. 2003: Multivariate Analysis of Ecological Data using CANOCO. Cambridge University Press, Cambridge, 269 pp. Levene, H. 1960: Robust tests for equality of variances. In: Olkin, I. et al. (eds): Contributions to Probability and Statistics: Essays in Honor of Harold Hotelling. Stanford Univ. Press, Stanford, pp. 278-292. Lieth, H., Berlekamp, J., Fuest, S. & Riediger, S. 1999: Climate Diagram World Atlas. CD-Series. Climate and Biosphere, Backuys Publishers, Leiden. Lopez, G. 1978: Contribucion al conocimiento fitosociologico de la Serrania de Cuenca, II. Comunidades herbaceas: vegetacion de rocas y pedreras; comunidades acuaticas; prados humedos y juncales; praderas y pastizales; malezas ruderales y arvenses. Anal. Inst. Bot. Cavanilles, 34(2): 597-702. McDonald, J. 2008: Handbook of biological statistics. Sparky House Publishing, Baltimore, Maryland, 319 pp. Meshinev, T., Apostolova, I., Georgiev, V., Dimit-rov, v., Petrova, A. & Veen, P. 2005: Grasslands of Bulgaria. Final report on the National Grasslands Inventory Project - Bulgaria, 2001-2004. Dragon 2003 Ltd. Publishers, Sofia, 104 pp. Mucina, L. 1997: Conspectus of classes of European vegetation. Folia Geobotanica Phytotax-onomica 32: 117-172. Mueller-Dombois, D. & Ellenberg, H. 1974: Aims and Methods of Vegetation Ecology. John Wiley & Sons, Inc., New York, 547 pp. Oberdorfer, E. 1994: Exkursionsflora. Verlag Eugen Ulmer., Stuttgart, 1049 pp. Otypkova Z. & Chytry M. 2006: Effects of plot size on the ordination of vegetation samples. Journal of Vegetation Science 17: 465-472. Parolly, G. 2003: Towards common standards in phytosociological papers submitted to the Turkish Journal of Botany: A Letter from the Editor. Turkish Journal of Botany 27: 163-165. Pavlov, D., Dimitrov, M. & Malinova, D. 2006: Syntaxonomic analysis of the vegetation in landscape types in Bulgarka Nature Park. Nauka Gorata, 1: 3-29 (in Bulgarian). Pop, I., Cristea, V. & Hodi§an, I. 2002: The Vegetation of Cluj District (a phytocoenological, ecological, bioeconomic and ecoprotective study). Contributii Botanice, 35(2): 5-254. Resmeritä, I. 1970: Flora, vegetation and productivity potential of the Vlädeasa massif. Acad. Press, Bucharest, 318 pp. (in Romanian). Rodwell, J., Shaminee, J., Mucina, L., Pignatti, S., Dring, J. & Moss, D. 2002. The diversity of European vegetation. An overview of phytoso-ciological alliances and their relationships to EUNIS habitats. Wageningen, 168 pp. Sanda, V., Popescu, A. & Baraba§, N. 1997: Syn-taxonomy and characteristics of Romanian vegetation. Bacau, 366 pp. (in Romanian). Šeffer, J., Lasak, R., Galvanek, D. & Stanova, V. 2002: Grasslands of Slovakia. Final report on National Grassland Inventory 1998 - 2002. DAPHNE - Institute of applied Ecology, Bratislava, 56 pp. Snowarski, M. 2002-2005: Vascular Plants of Poland photoflora - http://www.atlas-roslin.pl/. (accessed 17.02.2010). Stančic, Z. 2008: Classification of mesic and wet grasslands in northwest Croatia. Biologia 63(6): 1089-1103. ter Braak, C. 1994: Canonical community ordination. Part 1: Basic theory and linear methods. Ecoscience 1(2): 127-140. ter Braak, C. & Šmilauer, P. 2002: CANOCO Reference Manual and CanoDraw for Windows User's Guide: Software for Canonical Community Ordination (version 4.5). Microcomputer Power, Ithaca, New York, 351 pp. Tichy, L. 2002: JUICE, software for vegetation classification. Journal of Vegetation Science 13: 451-453. Tichy, L. & Holt, J. 2006: JUICE program for management, analysis and classification of ecological data. Program manual. Brno, 98 pp. - http://www.sci.muni.cz/botany/juice/jc06_ man.htm (accessed 11.02.2009). Tzonev, R., Dimitrov, M., & Roussakova, V. 2009: Syntaxa according to the Braun-Blanquet approach in Bulgaria. Phytologia Balcanica 15(2): 209-233. Valachovič, M. 2004: Syntaxonomy of the fringe vegetation in Slovakia in relation to surrounding areas - preliminary classification. Hac-quetia 3(1): 9-25. van der Maarel, E. 1979: Transformation of cover-abundance values in phytosociology and its effect on community similarity. Vegetatio 39: 97-114. van der Maarel, E. 2005: Vegetation ecology - an overview. In: van der Maarel, E. (ed.): Vegetation ecology. Blackwell Publ., Padstow, Cornwall, pp. 1-51. Veen, P., Jefferson, R., de Smidt & van der Straa-ten, J. (eds) 2009: Grasslands in Europe of high nature value. KNNV Publishing, Zeist, The Netherlands, 319 pp. Velev, St. 2002: Climatic regioning. In: Kopralev, I. (ed.): Geography of Bulgaria. ForCom, pp. 155-156. Westhoff, V. & van der Maarel, E. 1973: The Braun-Blanquet approach. 2nd Ed. In: Whit-taker, R. (ed.): Classification of plant communities. Junk, The Hague, pp. 287-399. Whittaker, R. 1973: Direct gradient analysis: Techniques. In: Whittaker, R. (ed.): Handbook of Vegetation Science 5. Junk, The Hague, pp. 7-32. Zelnik, I. 2007: New grassland association Triseto-Centaureetum macroptili ass. nova in Slovenia. - Hacquetia 6(1): 77-89. Zuidhoff, C., Rodwell, S. & Schaminee, H.J. 1995: The Cynosurion cristati Tx. 1947 of central, southern and western Europe: a tentative overview, based on the analysis of individual rele-ves. Ann. Bot. 53: 25-47. Received 1. 4. 2010 Revision received 6. 8. 2010 Accepted 12. 8. 2010 Table 1: Synoptic table of the studied vegetation units. The species are represented by two indicators: Fidelity measure, expressed by the Phi-coefficient (Chytry & al. 2002) and Constancy, expressed in percentages. Tabela 1: Sinoptična tabela proučevanih vegetacijskih tipov. Vrste so predstavljene z dvema indikatorjema: mero naveznosti, kot Fi koeficient (Chytry & al. 2002) in prisotnostjo v odstotkih. Syntaxa Cynosurion cristati Arrhenatherion Biogeographical affiliation 1 viT/~tii/~tvi 1 f ilUHUil fflt-dll dry type wet type elatioris of the species Number of releves 30 139 66 298 Fidelity / Constnacy Phi % Phi % Phi % Phi % Trifolion medii Eur-subMed Trifolium montanum 47,5 50 - 14 - 5 -4 Euras Trifolium medium 39,6 27 - 4 - 2 -1 Euras Centaureajacea Trifolio-Geranietea 36,8 47 - 12 - 18 -6 Epralp Thesium bavarum 43,9 37 -6 - 2 -4 Eur-subMed Geranium sanguineum 29,4 13 - . - 2 -. Med-subMed Dorycnium herbaceum 27,8 23 - 11 - 2 -2 EEur Ferulago sylvatica 27,7 10 -. - -. subMed-Eur Lathyrus niger Festuco-Brometea 27,2 13 -3 - -. subMed Helianthemum nummularium 54,6 43 -2 - 3 -1 Ealp Hieracium hoppeanum 50,1 43 -9 - -2 Med-Eur Danthonia alpina 46,1 63 - 15 3,5 30 -2 Euras Hypochaeris maculata 41 23 -. - 2 -. Euras Inula salicina 38,5 27 -6 - -1 Euras Briza media 37,6 73 - 39 3,1 44 -9 subMed Asperula cynanchica 35,7 23 -3 - 2 -1 Bale Dianthus cruentus 35,5 20 -3 - -. Eur Veronica austriaca 35,5 17 -. - -1 Euras Thymus pulegioides 33,3 30 - 12 - 3 -2 Eur Thymus glabrescens 32,6 33 - 12 - 2 -9 subMed Euphorbia cyparissias 32,5 40 5 22 - 3 -8 Euras Koeleria macrantha 28,2 20 -2 - 3 -4 Pont-Med Chrysopogon gryllus 27,1 13 -1 - 2 -1 Biogeogr. affiliation Number of releves 30 139 66 298 Eur Prunella grandiflora Cynosurion dry type 24,6 10 — — 2 — Euras Agrostis capillaris 26,5 83 42,8 97 — 59 — 4 Euras Cichorium intybus - 13 32,8 57 — 12 — 40 Eur-subatl Leontodon autumnalis - 10 32,7 50 — 33 — 9 Euras Phleum pratense Cynosurion wet type — 13 23,5 38 — 26 — 8 satl-subMed Cynosurus cristatus - 47 18,6 75 41 94 — 20 Euras Anthoxanthum odoratum — 60 14,2 68 31,7 83 — 13 Euras Trifolium repens - 33 — 64 24,2 79 — 56 satl-subMed Holcus lanatus Arrhenatherion — 37 16,7 53 23,7 59 — 7 Euras Medicago lupulina — — 10 — 17 43,4 48 Euras Vicia cracca — 17 — 22 — 17 38,2 58 Euras Silene vulgaris — — 1 — 2 32 17 Euras Poa pratensis — 40 — 59 — 68 29,6 89 preAlp-subMed Trisetum flavescens — 17 — 16 — 14 27,8 42 subMed-satl Trifolium campestre — 30 — 23 — 29 25 55 Euras Cirsium arvense — 3 — 14 — 9 25 29 Euras Galium album — 3 — 11 — 5 24,3 23 satl-subMed Arrhenatherum elatius — 20 — 47 — 38 23,9 62 Euras Dactylis glomerata Molinio-Arrhenatheretea — 23 — 65 — 58 21,7 73 Euras Rumex acetosella 38,9 37 2,6 15 — 2 — 1 Eur Potentilla alba 35,5 23 — 1 — 5 — Euras Avenula pubescens 33,4 23 — 3 — 3 — 2 Euras Festuca rubra 31 43 0,4 22 — 17 — 4 Euras Luzula campestris 29,7 27 — 6 — 9 — 1 Euras Deschampsia caespitosa — 3 29,4 39 — 29 — 5 Euras Lotus corniculatus — 53 20,7 80 — 61 — 56 Eur Festuca nigrescens — 57 10,2 45 — 20 — 23 Eur Cirsium canum — — 9 56,5 59 — 12 subMed-Med Trifolium patens — 3 — 6 51,3 47 — 4 Eur Centaurea phrygia — 10 — 11 48,2 50 — 1 Euras Alopecurus pratensis — 7 — 42 45,3 80 — 38 Euras Carex hirta — 7 — 10 39,8 42 — 8 Euras Prunella vulgaris — 13 6,9 34 39,3 59 — 7 Euras Lathyrus pratensis — 3 — 30 38,9 62 — 28 Euras Lysimachia nummularia — — 3 34,7 21 — 2 Euras Rumex acetosa 24,7 73 — 40 32,8 80 — 14 Euras Festuca pratensis — 37 — 58 32,1 91 — 72 preAlp Carum carvi — 20 — 14 29,6 44 — 12 Euras Ranunculus acris — 33 — 58 28,7 76 — 36 EsubMed Leucanthemum vulgare — 47 — 34 25,8 65 — 27 Euras Cerastium holosteoides — 43 — 42 24,7 67 — 30 Euras Stellaria graminea — 70 — 55 23,5 82 — 41 subMed-satl Colchicum autumnale — 7 — 4 23,2 20 — 4 EsubMed Rhinanthus rumelicus — 50 — 53 23,1 74 — 41 Euras Carex pallescens — — 1 19,5 6 — Euras Mentha longifolia — — 4 — 5 23,4 15 Euras Vicia tetrasperma — 10 — 26 — 15 21,8 38 Euras Poa trivialis — — 1 — 6 19,7 12 Biogeogr. affiliation Number of releves 30 139 66 298 subatl-subMed Chamaespartium sagittale 79,5 73 - 1 - 2 - 1 Orobanche sp. 42,1 23 - 1 - - Alp Plantago subulata 32,2 13 - - - subatl-subMed Hypochaeris radicata 28,6 23 - 9 - 3 - 1 Balc Armeria rumelica 27,7 10 - - - Eur Silene otites 27,7 10 - - - Euras Cuscuta europaea 26,2 10 - 1 - - Euras Poa compressa - 7 31,4 29 - 3 - 8 Euras Rosa canina - 30 25 46 - 14 - 18 subMed-subatl Centaurium erythraea - 3 21,9 13 - 3 - subatl-subMed Dianthus armeria - 21,3 6 - - 1 Euras Hypericum perforatum - 47 21 47 - 8 - 21 EsubMed Hieracium praealtum - 13 19,1 27 - 9 - 12 Euras Bromus arvensis - - 13 43,1 58 - 30 Euras Vicia angustifolia - 3 - 9 27,3 30 - 13 Euras Poa sylvicola - - 26,9 12 - 2 Euras Filipendula vulgaris 25,7 57 - 22 26,8 58 - 5 Euras Festuca rupicola - - 23,5 11 - 3 Eur-Med Silene italica - - 21,5 6 - subatl-subMed Ornithogalum umbellatum - 7 - 1 21,3 14 - 1 Euras Linum catharicum - 7 - 20,6 14 - 2 subMed-pralp Allium carinatum - - 6 20,1 12 - 1 Eur Gagea pratensis - - 1 19,5 6 - EsubMed Alliumflavum - 3 - 1 18,8 9 - subMed-subatl Geranium dissectum - 3 - 1 - 3 42,5 31 Eur Salvia nemorosa - 3 - 1 - 5 40,3 31 EsubMed Vicia grandiflora - 20 - 25 - 20 35,6 60 subMed Tragopogon dubius - - 5 - 2 31,9 21 Med-subMed Lathyrus aphaca - - - 2 30,8 14 EsubMed-Med Vicia pannonica - - - 2 30,4 14 Euras Convolvulus arvensis - 10 - 58 - 52 29,9 74 EMed Dipsacus laciniatus - - 8 - 3 29,5 22 Eur Onobrychis arenaria - - 1 - 6 28,9 19 Med-subMed Xeranthemum annuum - - - 2 28,1 12 Euras Elymus repens - 10 - 31 - 33 27,4 54 Euras Myosotis arvensis - 3 - 5 - 11 26,8 26 Euras Cirsium vulgare - - 4 - 2 26,5 15 Euras Echium vulgare - - 1 - 26,4 11 subMed-Med Lathyrus nissolia - - 7 - 5 25,5 20 EsubMed Vicia villosa - 3 - 2 - 5 23,7 17 subMed Dasypyrum villosum - - - 23,2 7 subMed Bromus sterilis - - 4 - 3 22,9 14 Eur Coronilla varia - 7 - 4 - 5 22,7 20 Euras Astragalus glycyphyllos - 3 - 3 - 21,3 12 Euras Chamomilla recutita - - 2 - 21,3 9 Eur Cirsium ligulare - 13 - 45 - 29 21,1 52 Euras Chondrilla juncea - - 4 - 20 9 subMed-Med Galium aparine - - - 3 19,8 9 Euras Alopecurus myosuroides - - - 19,6 5 Euras Arctium lappa - - - 19,6 5 Euras Medicago falcata - - 1 - 5 19,3 10 Euras Artemisia vulgaris - 3 - - 2 19,3 10 Eur Cardaria draba - - - 18,9 5 Biogeogr. affiliation Number of releves 30 139 66 298 subMed-subatl Clematis vitalba - . -. -. 18,9 5 EsubMed Melissa ^officinalis - . -. -. 18,9 5 Euras Taeniatherum caput-medusae -. - 2 -. 18,9 7 Euras Euphorbia esula -. - 2 -. 18,3 7 Euras Erodium cicutarium -. -. -. 18,2 4 Med-subMed Medicago minima -. -. -. 18,2 4 Euras Tanacetum vulgare -. -1 -. 18,1 5 subatl-subMed Trifolium striatum - 7 - 5 - 27 13,3 24 Table 2: Summarized statistical components of the obtained vegetation units. Tabela 2: Statističen povzetek lastnosti vegetacijskih enot. Statistical components Vegetation unit