HACQUETIA 11/2 • 2012, 249-269
DOI: 10.2478/v10028-012-0012-3
acidophilous dry grasslands on the quartzite bedrock in western slovakia
Daniela DUBRAVKOVA1' 2 & Jaroslav KOŠtAL3
Abstract
The paper presents results of actual phytocoenological research of acidophilous dry grasslands of the Koelerio-Phleion phleoidis Korneck 1974 alliance on the quartzite bedrock in western Slovakia. The modified TWIN-SPAN analysis distinguished two associations in the study area. Avenulo pratensis-Festucetum valesiacae Vicherek et al. in Chytry et al. 1997 is dominated by Festuca valesiaca s.l. and it is distributed on the quartzite hills in the Tribeč Mts. Potentillo heptaphyllae-Festucetum rupicolae (Klika 1951) Toman 1970 occurs at sites in the Biele Karpaty Mts and Považsky Inovec Mts and it is dominated by Festuca rupicola. The paper brings very first data on distribution of Avenulo pratensis-Festucetum valesiacae in Slovakia and extends knowledge on occurrence of Potentillo heptaphyllae-Festucetum rupicolae. Both associations represent very rare vegetation types of Central Europe. In Slovakia, they occur locally at small sized stands and are threatened by successional changes caused by cessation of traditional management and island occurrence within other vegetation types. Main environmental gradients responsible for variation in species composition of these grasslands were revealed by the principal component analysis (PCA) and interpreted using indicator values and measured characteristics of the study sites. The major pattern of variation reflects altitude, which is negatively correlated with temperature and soil depth. The relationship between species composition and environmental factors (pedology, topography, geographical position) was analysed by redundancy analysis (RDA). The most important factor affecting the data variation was longitude followed by soil pH, slope and latitude. Key words: Biele Karpaty Mts, Koelerio-Phleion phleoidis,
phytocoenology, Považsky Inovec Mts, Tribeč Mts, xerophilous grassland vegetation.
Izvleček
V članku so predstavljeni rezultati fitocenološke raziskave kisloljubnih suhih travišč zveze Koelerio-Phleion phleoidis Korneck 1974 na kvarcitni matični podlagi na zahodnem Slovaškem. Z modificirano TWINSPAN analizo smo na raziskovanem območju ločili dve asociaciji. V sestojih asociacije Avenulo pratensis-Festucetum valesiacae Vicherek et al. in Chytry et al. 1997 prevladuje vrsta Festuca valesiaca s.l., ki je razširjena na kvarcitnem hribovju v gorovju Tribeč. Sestoje asociacije Potentillo heptaphyllae-Festucetum rupicolae (Klika 1951) Toman 1970 najdemo v gorovjih Biele Karpaty in Považsky Inovec in v njih prevladuje vrsta Festuca rupicola. V članku se prvič omenja pojavljanje asociacije Avenulo pratensis-Festucetum valesiacae na Slovaškem, ki razširja poznavanje razširjenosti asociacije Potentillo heptaphyllae-Festucetum rupicolae. Obe asociaciji predstavljata redke vegetacijske tipe v srednji Evropi. Na Slovaškem se pojavljata lokalno na majhnih površinah in sta ogroženi s sukcesijskimi spremembami zaradi opuščanja tradicionalnega gospodarjenja in zaradi izoliranega pojavljanja znotraj drugih vegetacijskih tipov. Glavne ekološke gradiente, ki so odgovorni za variabilnost vrstne sestave travišč, smo dobili z uporabo metode glavnih komponent (PCA) in jih ovrednotili z indikatorskimi in merjenimi vrednostmi. Glavni vzorec variabilnosti se odraža zaradi nadmorske višine, ki je v negativni korelaciji s temperaturo in globino tal. Povezavo med vrstno sestavo in okoljskimi dejavniki (pedologijo, topografijo, geografskim položajem) smo analizirali z analizo presežkov (RDA). Najpomembnejši dejavnik je bila geografska dolžina, ki ji sledijo pH tal, naklon in geografska širina.
Ključne besede: gorovje Biele Karpaty, Koelerio-Phleion phleoidis, spodnji in zgornji triasni kvarcit
fitocenologija, gorovje Považsky Inovec Mts, Tribeč, kserofilna travniška vegetacija.
1	Institute of Botany, Slovak Academy of Sciences, Dubravska cesta 9, SK-845 23 Bratislava, Slovakia
2	Homeland Museum in Považska Bystrica, Ul. odborov 244/8, SK-017 01 Považska Bystrica, Slovakia; daniela.dubravkova@savba.sk
8 State Nature Conservancy of the Slovak Republik, Administration of Protected Landscape Area Ponitrie, Samova 3, SK-949 01 Nitra, Slovakia; kostaljar@gmail.com
1. INTRODUCTION
The Central European acidophilous dry grasslands are conventionally classified within the Koelerio-Phleion phleoidis Korneck 1974 alliance (Festuco-Brometea Br.-Bl. et Tuxen ex Soo 1947) in the syntaxonomical classification system (Korneck 1974, Dubravkova et al. 2010). These communities are distributed mostly in France, southern Germany, Austria and the Czech Republic (Chytry et al. 1997). Some localities of acidophil-ous dry grasslands also occur in western Slovakia (located in eastern part of Central Europe) where they represent a rare vegetation type. These grasslands developed on the shallow rocky soils over acidic silicate rocks such as granite, gneiss, crystalline shale, quartzite, etc. These bedrock types are rather scarce in the territory of Slovakia. Besides the western Slovakia the shallow acidic rocky soils are locally distributed in other parts of the country as well, however various types of vegetation other than dry grasslands mostly cover the stands.
The literature sources deal with the acidophil-ous dry grasslands in Slovakia only sporadically. Chytry et al. (1997) and Zlinska (2000) documented the vegetation of Potentillo heptaphyllae-Festuce-tum rupicolae (Klika 1951) Toman 1970 (syn. Astero linosyris-Festucetum rupicolae Maglocky in Chytry et al. 1997 and Peucedano oreoselini-Festucetum rupicolae Vicherek et al. in Chytry et al. 1997) at granite bedrock at the southeastern slopes of the Male Karpaty Mts between Bratislava and Vinosady and at the acidic Pleistocene sands in the Borska nizina Lowland near Zavod. These localities were considered as the only sites of the Koelerio-Phleion phleoidis in Slovakia in the past (Michalkova 2007). However, our field research proved that the acidophilous dry grasslands in Slovakia occur in larger extends than formerly expected. Since the dry grassland vegetation on acidic bedrocks types other than granite and sands has so far been neglected in the phytoso-ciological literature we aim to fill this gap focussing particularly on the dry grassland stands on the quartzite bedrock. We (1) present new vegetation data of the acidophilous dry grasslands on quartzite in the Tribec Mts, Povazsky Inovec Mts and Biele Karpaty Mts in western Slovakia; (2) perform a numerical classification of the data and assign the clusters to associations described in the phytosociological literature; and (3) evaluate the environmental characteristics of the distinguished associations.
2. MATERIAL AND METHODS 2.1 Quartzite bedrock and study area
The Lower Triassic quartzite is one of the most acidic silicate rock types. It is very poor in minerals; the crystalline quartzite contains only a trace amount of CaO, the amount of MgO varies in insignificant values, while K2O and P2O5 are missing completely (Husenica 1964). The quartz-ite bedrock in Slovakia occurs most frequently at the edges of the Tribec Mts as the remnants of the Mesozoic (Lower Triassic) shell cover of the crystalline mountains. Here the quartzite bedrock creates typical small quartzite hills with rocky hilltops (Mederly & Hresko 1988). In these stands, various types of acidophilous vegetation occur, such as pioneer, scree and grassland communities or shrubby and forest vegetation (Elias 1986). In a small extent the quartzite is also distributed in the Povazsky Inovec Mts, Biele Karpaty Mts, Male Karpaty Mts, Revucka vrchovina Upland, Turcianska kotlina Basin, Lucenecka kotlina Basin and in a few other small localities in the Western Carpathian Mts.
The dry grasslands presented in current paper are located in the Tribec Mts, Povazsky Inovec Mts and Biele Karpaty Mts. The three mountain ranges are located at the western edges of the Carpathian Mountain Arch. They represent the southwestern foothills of the Western Carpathians (the phytogeographical unit of Prea-Carpathicum) and are in direct contact with the thermophilous flora of the Pannonian region (Pannonicum, Futak 1980). The study area is located in the warm climatic region. The everage annual air temperature reaches 8-9 °C and everage annual precipitation 600-700 mm (Lapin et al. 2002, St'astny et al. 2002, Fasko & St'astny 2002a).
The Tribec Mts and Povazsky Inovec Mts are the crystalline Central Western Carpathian mountain ranges that are built of the granite central core and the outer Mesozoic sedimentary rocks (e.g. limestone, dolomites and quartzite) that usually occur at the margins of the mountain ranges. The localities of acidophilous dry grasslands in the Tribec Mts and Povazsky Inovec Mts occur at the sites of Lower Triassic quartzite. The Biele Karpaty Mts are a part of the Outer Western Carpathians and formed mostly of flysch rocks (e.g. sandstone, claystone, shale and conglomerate). The quartzite at the dry grassland localities Skalicky vrch and Ostra horka is of the
Figure 1: The soil of acidophilous dry grasslands is shallow with high amount of rocks (locality Stary vrch, Považsky Inovec Mts). Photo: J. Koštal, June 2008.
Slika 1: Tla kisloljubnih suhih travišč so plitva in kamnita (Stary vrch, gorovje Považsky Inovec). Foto: J. Koštal, junij 2008.
Upper Triassic origin. It is a part of the Meso-zoic cliff zone called the klippen belt located between the Outer and the Central Western Carpathians (Biely 1996). In the quartzite areas, the dry grasslands developed locally at the sites with shallow rocky 'ranker' soil on south facing slopes with modest inclinations and at the hilltops (Figure 1). The soil pH (KCl) at the sites varies from 3.3 to 4.7 (Table 1).
2.2 Vegetation and environmental data
Based on the distribution of quartzite bedrock in Slovakia according to the geological maps (Geological map of the Slovak Republic, Biely 1975, Ivanicka 1998, Ivanicka 2007) and visual inspection of the satellite images of Slovakia in software Google Earth 6.0.2 we delimited localities of a potential occurrence of acidophilous grassland communities. We visited the selected localities
in the field and in the case of acidophilous dry grassland presence we made phytosociological relevés following the standard principles of the Zürich-Montpellier school (Braun-Blanquet 1964, Westhoff & van der Maarel 1973, Dengler et al. 2008) with the nine-degree scale (Barkman et al. 1964). All relevés included data on species composition, abundance and dominance as well as the standard site characteristics (altitude, latitude, longitude, slope, aspect, cover of herb layer and layer of bryophytes and lichens, height of herb layer, etc.), (Table 1, Appendix 1). All relevés were made between 2003 and 2009. The size of relevé plots was 16-25 m2; in one relevé (Table 1, rel. 4) the plot size was smaller and represented the total size of the stand covered by the study vegetation. At the relevé plots we measured the average soil depth by repeated pricking of a long needle (3 mm wide in diameter) into the soil. We also gathered soil samples mixing the soil collected in root depth from three different spots within a plot. The soil samples were analysed to
• locality
10	0	10 km of acidophilus
grassland
Figure 2: Map of study sites of the acidophilous dry grasslands located in western Slovakia (the Biele Karpaty Mts, Považsky Inovec Mts and Tribeč Mts).
Slika 2: Zemljevid proučevanih rastišč acidofilnih suhih tra-višč zahodne Slovaške (gorovja Biele Karpaty, Považsky Ino-vec in Tribeč).
determine the soil pH (KCl) in accordance with Fiala (1999).
The acidophilous dry grasslands on quartz-ite were confirmed and sampled in the following geomorphologic units of western and central Slovakia: Tribeč Mts, Považsky Inovec Mts, Biele Karpaty Mts, Revucka vrchovina Upland, Turčianska kotlina Basin and Lučenecka kotlina Basin. Seven relevés from Revucka vrchovina Upland (localities Breznička, Selce, Uderinâ), Turčianska kotlina Basin (Rudno) and L^eneckâ kotlina Basin (Hrnčiarska Ves) in central Slovakia were rather heterogeneous and their species composition was shifted in succession towards the sub-xerophilous vegetation and mesic hay meadows. These relevés were selected as dissimilar to the rest of the data set in the outlier analysis performed by DCA (CANOCO 4.5 package; ter
Braak & Smilauer 2002). For this reason we decided to present only a relatively homogeneous data set including 21 relevés from 11 geographically contiguous localities in the Tribec Mts, Povazsky Inovec Mts and Biele Karpaty Mts (western Slovakia) in the current paper (Figure 2). However, the omitted relevés are stored in the Slovak National Phytosociological Database (Hegedusovâ 2007) and available for further studies.
2.3	Classification analysis and assignment of relevés to associations
Relevés were stored in a TURBOVEG database (Hennekens & Schaminée 2001) and the data set edited using the JUICE 6.5 program (Tichy 2002). For numerical classification we used modified TWINSPAN algorithm (Rolecek et al. 2009) with three pseudospecies cut levels (0%, 5%, 25%) and total inertia as a measure of cluster heterogeneity. The constancy of differential species of the associations in Table 1 is over 50%; in few individual cases (when the species was set as a diagnostic species of the association in a literature source and was absent from all relevés of the other association) we set as differential also species with constancy > 30%. The species constancies > 50 % are given in bold.
The two clusters created by the TWINSPAN analysis were assigned to the phytosociological associations by comparing the diagnostic species of both units to the published characteristics of the associations in Chytry et al. (2007), Michâlk-ovâ (2007) and Dubravkovâ et al. (2010). Applying the Czech and Slovak expert systems for identification of syntaxa (Chytry 2007, Janisovâ et al. 2007) and comparing of the relevés from Table 1 to clusters originated in the large-scale study Dubravkovâ et al. (2010) using the function 'match to clusters' with the frequency-positive fidelity index (FPFI) in the JUICE program (Tichy 2002) verified the assignment.
2.4	Gradient analyses
We defined the main environmental gradients responsible for variation in species composition on the basis of indirect ordination techniques, using the principal component analysis (PCA) based on square-root transformed data on species cover from the CANOCO 4.5 package (ter Braak
& Smilauer 2002). For ecological interpretation of the ordination axes the average non-weighted indicator values of Borhidi (1993) for the relevés were plotted onto the PCA ordination diagram as supplementary environmental data.
The relationship between species composition and selected environmental factors was analysed by redundancy analysis (RDA) using the CANOCO 4.5 package. As the resulting gradient lengths in the DCA were short (1.98 for the first axis), the constrained linear ordination method of RDA was chosen for the evaluation of the independent (marginal), conditional, and pure effects of individual environmental variables. Species were not weighted by abundance, and no exclusion of rare species was applied. The following environmental variables were used in the RDA: average soil depth, soil pH (KCl), altitude, slope, latitude and longitude of the site. Forward selection was used for ranking environmental variables in order of importance (Palmer 1993). All studied environmental factors were tested by the Monte Carlo permutation test with unrestricted permutations (9999 permutations, P<0.05). Finally, the pure effect (where the percentage variance is explained by the variable, while the remaining significant variables were used as co-variables) was calculated (ter Braak & Prentice 1988). Pure variance is expressed as percentage of total inertia. To evaluate the independent (marginal) effects of individual variables, the variance explained by the variable when used as the only constraining variable was calculated. The conditional effect of a variable is given by the additional variance explained by the variable at the time it was included in the forward selection. Measured environmental variables that passed the forward selection in the RDA were correlated with ordination axes at P<0.001 (Herben & Munzbergovâ 2003).
2.5 Nomenclature
Nomenclature of taxa is in accordance with Mar-hold & Hindâk (1998). The syntaxa names and assignment of species to the diagnostic species of high-level syntaxa in Table 1 follow Janisovâ et al. (2007), Jarolimek & Sibik (2008) and Dubravk-ovâ et al. (2010). The taxon Festuca valesiaca s.l. includes both a diploid species Festuca valesiaca and a tetraploid F. pseudodalmatica (Janisovâ et al. 2007, Smarda 2008).
3. RESULTS
3.1	Classification of relevés
Based on the precise selection of the suitable sites on quartzite bedrock in western Slovakia and using the modified TWINSPAN analysis we defined two associations of acidophilous dry grasslands.
Syntaxonomical scheme:
Festuco-Brometea Br.-Bl. et Tuxen ex Soo 1947 Festucetalia valesiacae Br.-Bl. et Tuxen ex Br.-Bl. 1949
Koelerio-Phleion phleoidis Korneck 1974
1.	Avenulo pratensis-Festucetum valesiacae Vicherek et al. in Chytry et al. 1997
2.	Potentillo heptaphyllae-Festucetum rupicolae (Klika 1951) Toman 1970
3.2	Description of the associations
3.2.1 Avenulo pratensis-Festucetum valesiacae Vicherek et al. in Chytry et al. 1997
Table 1, relevés 1-13
Species composition and vegetation structure
The association comprises semi-closed acido-philous dry grasslands dominated by Festuca valesiaca s.l. All specimens revised by the expert were determined as a tetraploid taxon Festuca pseudodalmatica. Plants of the two ploidy levels (Festuca pseudodalmatica and F. valesiaca) are very difficult to distinguish based on morphological characters in the field so they might have been confused by botanists in the past. Recent taxo-nomical study Smarda (2008) showed that the two ploidity levels are not rigorously restricted to specific geological subsoils as well as that the ecological amplitude and extent of occurrence of Festuca pseudodalmatica is much wider then previously expected. The both species can even occur at the same sites. Regarding this knowledge, dominance of the tetraploid Festuca pseudodalmat-ica at the stands of Avenulo pratensis-Festucetum valesiacae is not surprising.
Besides the fescues, other grass species co-dominant in some stands are Anthoxanthum odo-ratum, Agrostis capillaris, and Arrhenatherum ela-tius. The generalist species of Central European
dry grasslands with wide ecological amplitude (e.g. Acosta rhenana, Eryngium campestre, Galium verum agg., and Tithymalus cyparissias) that are diagnostic of the class Festuco-Brometea (Janisová et al. 2007) are abundantly present at the sites. The differential species distinguishing this vegetation unit from the other association are Aira caryoph-yllea, Bromus hordeaceus, Cerastium pumilum, Cru-ciatapedemontana, Festuca valesiaca s.l. (F. pseudo-dalmatica), Luzula campestris, Pilosella bauhini, Poa bulbosa, Scleranthus annuus agg., Vicia tetrasperma, and Viola kitaibeliana. An important component of the species composition are the acidophytes and species of soils poor in minerals (e.g. Aceto-sella vulgaris, Avenella flexuosa, Chondilla juncea, Hypochaeris radicata, Jasione montana, Pilosella officinarum, Potentilla argentea agg., Steris vis-caria, and Trifolium arvense). Many of them are diagnostic species of the alliance Koelerio-Phleion phleoidis (Janisová et al. 2007, Dúbravková et al. 2010). Short annuals (e.g. Cerastium pumilum, Scleranthus annuus agg., Veronica verna, and Viola kitaibeliana) and bryophytes (e.g. Ceratodon pur-pureus, Hypnum cupressiforme, and Polytrichum juniperinum) are also abundant. Some species (e.g. Steris viscaria, Jasione montana, and Pilosella officinarum) bloom since the middle of May and create an attractive flower aspect (Figure 3). The relevés poor in species (e.g. rel. 2, 5) are typical of high abundance of bryophytes and their species composition resembles a pioneer community. The average cover of herbs is 78 % and that of bryophytes and lichens 34%. The height of plants at the stands is 27 cm on average.
Synecology and distribution
The current paper brings the very first data on distribution of the association in Slovakia. The stands of Avenulo pratensis-Festucetum valesiacae in western Slovakia occur exclusively at the NW and SE edges of the Tribec Mts, in the areas of warm and dry climate at lower altitudes (200310 m a.s.l.). The stands occur at the hilltops and ridges of small quartzite hills (localities Bádice, Jelenec, Krnca, Ladice, Mankovce, and Podhora-ny - Maly Bahorec) and sun-facing slopes of gentle inclinations (1°-10°, max. 35°). The secondary substitute stands can be found at the bottoms and edges of abandoned quartzite quarries (localities Bádice, Ladice, Jelenec, and Krnca). The sites are rare; they occur locally and in a small ex-
Figure 3: A stand of Avenulo pratensis-Festucetum valesiacae at locality Maly Bahorec, Tribec Mts. Photo: J. Kostal, May 2008.
Slika 3: Sestoj asociacije Avenulo pratensis-Festucetum valesiacae na lokaciji Maly Bahorec, gorovje Tribec. Foto: J. Kostal, maj 2008.
tent, mostly near the settlements where they were influenced by compressing of the soil as a result of grazing since they were used as pastureland in the past (Anonymus 1957).
The association represents a rare Central European type of dry grasslands. From other Central European countries, the association was recorded in the Czech Republic and Austria (SE edge of the Bohemian Massif, surroundings of the Neusiedler See); (Chytry et al. 1997, Dubravk-ova et al. 2010). The average soil pH (KCl) at the sites in the Tribec Mts is 4.0. Comparing to the typical soil pH of 5.3-6.7 at the sites of the association at the SE edge of the Bohemian Massif (Chytry et al. 1997), the soil pH at the quartzite sites in western Slovakia is even more acidic.
Variability
On the basis of the modified TWINSPAN analysis we distinguished two variants of the association. The first variant (Table 1, rel. 1-8) represented stands rather resembling pioneer vegetation typical of occurrence of Jasione montana, Sedum sexangulare and Steris viscaria. Comparing to the second variant the herb layer (Et) was more open (average cover 70%) and cover of layer of bryophytes and lichens (E0) was higher (42% on
average). The stands occurred at slopes of steeper inclinations (5-35°). The second variant (Table 1, rel. 9-13) represented closed stands on flat surfaces with inclination of 0-1°. Average cover of Ej was higher (94%) and E0 lower (14%). Since the two variants did not show any remarkable flo-ristical differentiation and were not supported by PCA ordination analysis (Figure 6) as well as due to the relatively small number of relevés analysed we considered these variants as adaptation of vegetation to the local environmental conditions and stage of succession.
Syndynamics and conservation status
Avenulo pratensis-Festucetum valesiacae represents a substitute vegetation type on previously clear-cut stands of acidophilous oak forests of the alliances Genisto germanicae-Quercion Neuhäusl et Neuhäuslova-Novotna 1967 and Quercion petrae-ae Zolyomi et Jakucs ex Jakucs 1960, which are considered to be the edaphic climax vegetation on shallow soils over quartzite in warm areas (Husova 1967). There occur grasslands dominated by Avenellaflexuosa, heathlands of the alliance Genistion pilosae Duvigneaud 1942 and hay meadows of Arrhenatherion elatioris Koch 1926 in the surroundings of the study stands.
The association is a very rare vegetation type of Slovakia that occurs locally at small sized stands. Due to the island occurrence within other types of vegetation and actual abandonment of the sites, the vegetation is evidently threatened by successional changes. The rare and threatened plant species occur infrequently at the sites (e.g. the EN - endangered species Aira caryophyllea and Ventenata dubia, VU - vulnerable species Orchis morio and LRnt - species of lower risk Jasione montana; Ferakova et al. 2001).
3.2.2 Potentillo heptaphyllae-Festucetum rupicolae (Klika 1951) Toman 1970
Table 1, relevés 14-21
Species composition and vegetation structure
The association includes closed acidophilous dry grasslands dominated by Festuca rupicola. Other grass species (e.g. Agrostis capillaris, Anthoxanthum odoratum, and Arrhenatherum elatius) accompany
the dominant. The species composition, similarly to the previous association, comprises the gener-alists of Central European dry grasslands (e.g. Acosta rhenana, Eryngium campestre, Galium verum agg., and Tithymalus cyparissias). The differential species of the association are Bromus erectus, Di-anthus carthusianorum, Festuca rupicola, Fragaria viridis, Poa pratensis agg., Potentilla heptaphylla, Prunus spinosa, and Teucrium chamaedrys. Higher constancies also show Crataegus monogyna and Rosa canina agg. These are mostly sub-xerophil-ous species, herbs of hay meadows and succes-sionally important shrubs that indicate a less xero- and thermophilous character of the vegetation comparing to the association Avenulo praten-sis-Festucetum valesiacae and abandonment of the sites. Occurrence of the acidophytes and species of soils poor in minerals representing mostly the diagnostic species of the alliance Koelerio-Phleion phleioidis is similar to the previous association. The bryophytes and short herbs and annuals (e.g. Cerastium brachypetalum and Veronica verna) are also present. A colourful aspect in the middle of May bring flowers of Steris viscaria and Pilosella officinarum. The average cover of herbs is 85% and that of bryophytes and lichens 32.5%. The height of plants at the stands is 28 cm on average.
Synecology and distribution
The localities of Potentillo heptaphyllae-Festucetum rupicolae occur in the Povazsky Inovec Mts (lokalities Stary vrch, Skalice and Lasid) and in the Biele Karpaty Mts (localities Skalicky vrch and Ostra horka). These sites are situated at the hills near the settlements and used for grazing in the past (Figure 4). The dry grasslands developed locally on flat surfaces or modest slopes of mostly sun facing expositions. The sites are situated at the altitudes of about 300-400 m a.s.l. that are roughly about 50-100 m higher comparing to the stands of Avenulo pratensis-Festucetum valesiacae. The stands of the association are rare, small in size and of local distribution.
In other parts of Slovakia, this vegetation occurs at the SE slopes of the Male Karpaty Mts and in the Borska nizina Lowland (Chytry et al. 1997, Zlinska 2000; Michalkova 2007). Some stands with vegetation resembling the Potentillo heptaphyllae-Festucetum rupicolae were also documented from the quartzite areas of Slovakia located further to the east (Revucka vrchovina
SOIl|t!A UliOI^l
e
<s
<N
m k
k k
SOIl|t!A UliOI^l
-
v ■C
s
3 =
• 4)
s
"3 Ci
iN
iN
00 PC rt œ
rt ^
rt ^
cS
~<U >
<u
15 p¿
00 o
lin H
iN O
^r
iN O
in
oo m <s
iN
iN
in O iN
00 O iN
00 iN
M	
<U	
<u	
M	iS
<U	
	o
s—(	
<U	
	—
m (S 00 fl
M <N
00 O ^ rn cK ^
».o o
^r t-^ rt
00 o
C^ (N
un	O
a\
oo	un
a\	rt
c^ oo m
ON (S
l>	m	vc
^r	vo	<N
C
^	£	^
co	vc
^r	oo	on
rt	<N	O
VID	on
rt	<N	O
^r	vo
®	rt	<s
^	2	«N
^	3	^
CK	,_
o^	^

in in in
o, oo m
CK vo l>
o, oo m


0s
O in
<N
■ I>
no
^ M 't
on CK on [> on
o, in vc m i> iN
^ m
<N
vo
on ^
S d ■H
'o


Ä3UBJSU03
-iS S3
i
y
-S
i
s
&
si !
Ä3UBJSU03 3§BJU33.ia¿
S3 «
i!
š
y
-S
I
! .¡s
s
j
6d
i
-S s
i <f
c
*<J
v &
=
is
iH P

in in iN <N
o o i-H			(S	(S so	(S SO	■t	TT	m	\Q ^f	00 m	
m	rt	s-		1-	rt	1-	1-	s-	s-	s-	
* m	cd	s-		+	+	1-			s-	s-	+
m	s-	s-	+	ca	+		+	s-	+	s-	
* m	rt	+	s-	ca	+		+	s-	s-	s-	
*		s-		s-	+	+		s-	s-		
-	i-		rt	1-	1-		rt				
*	-	+	s-		+				s-		
* m	cd	s-	1-	1-			s-			s-	cd
s-			+			1-					
rt	m	s-				ca		+			
m					s-	+	+				
	s-		1-					s-			
		s-	+	+		i-	+	s-			
s s s s 3
S rg S
a
s
"a
si ^
^ -S S Ö

-a s a -c
s N
Ô >3
0
■S3 =S	-5
is ¡3	„C
U^	£

ŠSf	£
•« ¡^	a
§ ^ -K	-S
¡-S -S	'S
g .o S	g
cq S S	C
100		so	CO \o	0	0 m	0	0 m	00 m	0 0 1	0 0 1	so so		in t-	so	so	so	00 m	so so	0 m	0 m	in <N	in <N	in <N	in <N			in <N	00 00	
		-		+	s-	+	s-		-	ca		s-			s-	1-	-	s-	s-		s-	+		s-			s-	-	-
		+					s-		ca	i-	ca	+	i-		1-						i-								
		-	s-	s-	s-	s-		-	+	+	ca	+	i-	+		s-	ca	+		i-								-	
			s-						ca	ca	ca	1-	+	1-				i-	ca				+				s-	+	S-
	m	-	cd	-	s-	+	s-		i-	-	s-	1-		+		+	s-	+	+					s-				-	S-
		s-	s-	-	s-	s-			ca	-	X	s-	i-	1-	1-	1-		i-		i-								-	S-
	-								-	+	-				1-	1-		-	+	i-								ca	S-
	s-		s-				+		-	ca	ca		-	s-	+			-		+			s-						S-
m <N		on	in	00					0 0 1	so	so	<s as	TT m	so	a\ so	■t m	a\ so	00 m	vo	00 m	00 m	on	<N	00	m (N	m in		so	
+ 1-
^ +
s^ + s^ s-^ +
s^ s^ s-
m +
+ s^ -— s^ s-
C3 + 1-
+ 1-
s^ s^ s^ s-
+ 1-
i—i i—i c^ ;—

Xi	ca s^ s-
+ >-
+
i
t t
+ 1-
^ +
s^ s^ + ca s-
£
=
it 5
■ $
M 3
■a s
3
? U ^ v. 3 s
S IS
3
G
3 -a
.C
IS
^ 'K &
8 » I
-I f
ta
-a „3
!= g ^ cu K &
^sgg
t 1
<s
1 £
3 .g
<3
3
I § ^ I
s §
IS
Co ^
33 O
£
M M ca
^ S 3s
. Is S 1
a, o
M M ca
cvi5

& 3
£
S
-S3 s tx:
I


Co i^.
3 -g
s
3

3	3	3
3	^	'¡3
I	I	2
^	1	H

3
! it S
J (t
«
1 s •S
I
'S
3
I
3	a
|	&
t	!
R	tC
	en					O					M			O		00	00		m		m		in	in					
	' 1	' 1	'				<N	<N			so	'				m	m	<N	' 1		' 1		<N	<N	<N				
<S				+	+	+															s-		+						
o <s				s-										i-	+														
a\				-	-	+		+		+				+	i-			s-											
SO						+				+			-					+						+					
	s-				+	-		+		+	(tí	s-	+		i-		s-		s-					l-					
SO		s-	s-		-					+	+		i-		+	i-													
					tí		s-			+	1-		s-	s-	i-	+	s-								l-				
TT					-		s-			tí	-		1-	+	+	-	s-								+				
	_!	m		in		<s			m	a\	(S	00	TT	a\	in	00	00	m	m	m	m					in			
g	on	' 1	'	so		so			<N	so	\o	m		so	' 1			' 1	' 1	<N	' 1	' 1				' 1	' 1	' 1	' 1
ro	s-			-	s-	s-				s-	s-	+		+						s-						+			
<N wH				ca	+				s-	s-	+	+	+	+				s-		s-									
i-H wH	s-	s-	s-	-	+	s-				s-	s-		+	i-		+		+	s-		s-	-					l-	l-	l-
O wH	s-			-		+				+	s-	s-	+		i-				+			-							
				S-	s-	s-					s-			+															
00				cd	m									i-															
t>	s-	s-		m	s-	s-				+	s-	-	1-	-			s-									l-			
					+				s-		-		1-	S-	s-														l-
m				s-	i-							i-																l-	
T						s-				1-			S-	-						+	s-								
ro				-		s-				1-	s-																l-		
(S				-	-					(tí				+															
			s-	s-	+	s-				+																			
-
v
■c =
3 =
• 4)
Su
"ČS «
-a
Ü o
3 |
-S
3 Ü ce
S š
co <
,g !5
o
s -g
i •s
M M (tí
t 8
g „
(X) S
S
s
ft
i § U G
s o -s
•S k
S c
Í3 1 iS t
s „3
O
tu ô
■a ^ O £
1
3 -a Ü o
hI
JS
£ 3
M M (tí 3
1	S
S	-s
¡3	S
o	^ U
3
■13 3
u s
3	3
	¡s
ÍS	;s
	"<3
Ü	ta
te	ft
^ s
-fc .s ^ £
.s
A
3 -C O
3 !? ft
3
ja
o &
¿o ft ft
s
ft tC
O U
							00	O	o	m	in		00			
		.—i	.—i	.—i	.—i	<N	00	O	m	<—i	<N	.—i		,—i		
																
							m	-					+			
								-								
							cd	-					+			
							m	-	i-		s-					
							cd	-	i-					s-		
						s-	X	S-		ca		+	+			
			s-	s-		s-	^	X	+							
		s-			s-		+	m	+							
in		00	00	00	00			(S	_!	vo	_i	00		in	in	in
' 1	' 1						00		on		m	m				' 1
							s-	s-			s-					
							cd	-		cS				+		
s-	s-						X									
s-	s-			S-	S-		^	+	+	^						
			S-				+	ca	+		-					
								-			-	ca				
							cd							+		
		S-						ca		-		-			+	
								+								-
												-				
								+				cd				
£
-O -0
-a 3


-O -0
^ t
U • •
CO
% vc
^	*
a	£P
5	3
^	-c
6	«
3	^
3 -S
s
3
s
so • • : ^ ^C
1 S S
Q ^ &
S £ -3 ^
is s
c -is
a
•3 3 3
-c 3 -a
a s
is a
- o • -
■S !U In
£

••c
s-
a S -is
O O §1
m C
<N
£ ^ a >
r - -S ^ 2
£ a
h 'J
3 J?
_J	^
^	a
5	S
a	a
a
O	<N
. ;	G
+	"t
c
si
C ¡5
^ s
> L"
3 s
I 5 - -a 1 a
3 -c 3
3
"3
^ 3
a 3 p ^ ^ £
a f
-3 x ^

I
3
"3
IS
a BS
Si
u c
■$ p3 5 £ E
> oj <□ >
id
^^ il T3 S? fl rt
JSsI
BS
hi
O
s

c
o
TS
V
V
<N
s
3
^
s g
s
3
K
S tis K E-
S


H s ^s
X
+
3 -a K 3
3 -C
Figure 4: The stands of Potentillo heptaphyllae-Festucetum rupicolae occur frequently at the hilltops of small quartzite hills in a mosaik with stands of Avenela flexuosa (locality Stary vrch, Považsky Inovec Mts). Photo: J. Koštal, June 2008. Slika 4: Sestoji asociacije Potentillo heptaphyllae-Festucetum rupicolae se pogosto pojavljajo na vrhovih manjših kvarcitnih v mozaiku s sestoji vrste Avenela flexuosa (Stary vrch, gorovje Považsky Inovec). Foto: J. Koštal, junij 2008.
Figure 5: The acidophilous dry grassland often colonise substitute stands at abandoned quartzite quarries (a quarry at Stary vrch, Považsky Inovec Mts). Photo: J. Koštal, June 2008.
Slika 5: Acidofilna suha travišča pogosto naseljujejo nadomestna rastišča v kamnolomih kvarcita (kamnolom pri lokaciji Stary vrch, gorovje Považsky Inovec). Foto: J. Koštal, junij 2008.
Upland, Turcianska kotlina Basin, Lucenecka kotlina Basin; Dubravkova ined., Kost'al ined.). These stands, however, were rather heterogeneous and more mesic. Besides Slovakia, the association is distributed in the Czech Republic and in NE Austria (Chytry et al. 1997; Chytry et al. 2007; Dubravkova et al. 2010).
Syndynamics and conservation status
Potentillo heptaphyllae-Festucetum rupicolae represents a substitute vegetation type on previously clear-cut stands of acidophilous oak forests of the alliances Genisto germanicae-Quercion and Quercion petraeae. Vegetation types neighbouring to the stands of the association are mostly hay meadows (Arrhenatherion elatioris), sub-xerophil-ous grasslands (Bromion erecti Koch 1926) and grasslands dominated by Avenella flexuosa.
The association is one of the rare vegetation types of Slovakia and it is threatened by succes-sional changes caused by cessation of traditional management. Secondary substitute sites of the association can be recently found in the abandoned quartzite quarries (Figure 5). The rare and threatened plant species occur infrequently at some sites (e.g. VU - vulnerable species Orchis morio and LRnt - species of lower risk Jasione montana; Ferâkovâ et al. 2001).
3.3 The gradient analyses
3.3.1 Indirect gradient analysis (PCA)
Results of the principal component analysis (eigenvalues: 1st axis 0.244, 2nd axis 0.158, sum of all canonical eigenvalues 0.749; Table 2) support division of the data into two groups which represent the associations Avenulo pratensis-Festucetum valesiacae and Potentillo heptaphyllae-Festucetum rupicolae. The first and second axes explained 40.2% of the variance of species data and 47.8% of variance of species-environment relation. The first axis was positively correlated with altitude (correlation coefficient 0.5289) and negatively correlated with temperature indicator values (-0.5200) and soil depth (-0.4455). Separation of the relevés to the two main groups located in the right and left parts of the ordination chart is mainly based on the gradient of temperature and altitude of the sites (Figure 6). The environmen-
tal variables with the highest correlation coefi-cients to the second ordination axis are soil pH (-0.3195) and altitude of the sites (-0.302).
SAMPLES
•	Avenulo pratensis-Festucetum valesiacae
♦	Potentillo heptaphyllae-Festucetum rupicolae
SUPPL. VARIABLES -*
Figure 6: Diagram of the principal component analysis (PCA) of the two acidophilous dry grassland associations based on square-root transformed data of species cover (first and second axes). Supplementary environmental variables represent unweighted indicator values of temperature, light, conti-nentality, nutrients, soil reaction and moisture according to Borhidi (1993), Shannon-Wiener index, number of species, measured average value of soil depth and soil pH (KCl), and altitude of the plot.
Slika 6: Diagram analize glavnih component (PCA) dveh asociacij acidofilnih suhih travišč narejen s korenjenjem pok-rovnih vrednosti (prva in druga os). Dodatne okoljske spre-meljivke predstavljajo netehtane indikatorske vrednosti za temperaturo, svetlobo, kontinentalnost, hranila, reakcijo tal, in vlažnost po Borhidiju (1993), Shannon-Wienerjev indeks, število vrst, izmerjeno povprečno globino tal in pH (KCl), ter nadmorsko višino popisne ploskve.
Relevés of the Avenulo pratensis-Festucetum valesiacae are located in the left and upper parts of the diagrams (Figures 6, 7). It is a warmth-demanding association occuring at the sites at lower altitudes (the average altitude is 50 m smaller than at sites of the other community; Table 1). Generally, the soils are deeper and sites are dryer. The higher covers of weighty species displayed in Figure 7 reached mostly therrophytes with early spring phenophase (e.g. Aira caryo-phyllea, Cruciata pedemontana, Vicia tetrasperma, Viola kitaibeliana). Comparing to the Potentillo heptaphyllae-Festucetum rupicolae, the association is more open with lower average cover of herb layer. The relevés 2, 5, and 8 were separated off the diagram centre which was caused by higher
covers of mosses, mostly Brachythecium albicans and Hypnum cupressiforme.
Group of relevés separated in the right part of the PCA diagrams (Figures 6, 7) represent the association Potentillo heptaphyllae-Festucetum rupico-lae. According to the results of PCA analysis, less thermophitic species occur in these grasslands and they are distributed at more humid sites with soil of somewhat less acidic pH (average pH 4.3; Table 1) located at higher altitudes (Figure 6). Some sub-xerophilous species (e.g. Bromus erectus, Fragaria viridis, Lotus corniculatus, Potentilla hep-taphylla) and succesionally important shrubs (Rosa canina) reach higher covers which indicates that the association is less thermo- and xerophilous, and relevés are influenced by successional shifts.
Table 2: The outputs of the PCA analysis of the phytocoenological relevés from Table 1. Tabela 2: Rezultati PCA analize fitocenoloških popisov iz Tabele 1.
Axes	1	2	3	4 Total variance
Eigenvalues	0.244	0.158	0.123	0.074	1.000
Cumulative percentage variance of species data	24.4	40.2	52.5	59.9	
Species-environment correlations	0.948	0.937	0.912	0.964	
Cumulative percentage variance of species-environment relation	29.3	47.8	61.6	70.8	
Sum of all eigenvalues					1.000
Sum of all canonical eigenvalues					0.749
1.0 1.0
SPECIES -*■
SAMPLES
•	Avenulo pratensis-Festucetum valesiacae
♦	Potentillo heptaphyllae-Festucetum rupicolae
Figure 7: Diagram of the principal component analysis (PCA) of the acidophilous dry grasslands based on square-root transformed data of species cover (first and second axes). 20 species with weight over 25% are shown. The three-character abbreviations are as follows: Aira caryophyllea, Brachythecium albicans, Bromus erectus, B. hordeaceus, Cladonia rangiformis, Cruciata pedemontana, Festuca rupicola, F. valesiaca, Fragaria viridis, Hypnum cupressiforme, Lotus corniculatus, Pilosella bauhini, Potentilla heptaphylla, Rosa canina, Sarothamnus scoparius, Scleranthus annuus, Tithymalus cyparisias, Vicia tetrasperma, Viola kitaibeliana, Vulpia myuros. Explanations of symbols are in accordance with Figure 6. Slika 7: Diagram analize glavnih component (PCA) acido-filnih suhih travišč narejen s korenjenjem pokrovnih vrednosti (prva in druga os). Prikazanih je 20 vrst s težo večjo kot 25 %. Okrajšava s tremi črkami predstavlja: Aira caryophyllea, Brachythecium albicans, Bromus erectus, B. hordeaceus, Cladonia rangiformis, Cruciata pedemontana, Festuca rupicola, F. valesiaca, Fragaria viridis, Hypnum cupressiforme, Lotus corniculatus, Pilosella bauhini, Potentilla heptaphylla, Rosa canina, Sarothamnus scoparius, Scleranthus annuus, Tithymalus cyparisias, Vicia tetrasperma, Viola kitaibeliana, Vulpia myuros. Razlaga simbolov je enaka kot pri Sliki 6.
3.3.2 Direct gradient analysis (RDA)
The relationship between species composition and selected environmental factors was analysed using redundancy analysis (RDA). Sum of all canonical eigenvalues was 0.416 (eigenvalues: 1st axis 0.219, 2nd axis 0.09; Table 3). The first RDA axis explained 21.9% of variance of the species data and 52.6% of the species-environment relationship, which means that 52.6% of the variability of our data set caused by the selected environmental factors was reflected by the first canonical axis. The 4 factors that were significant in the Monte Carlo permutation test (longitude, soil pH, slope and latitude) together explained 99.99% of compositional variance (Table 4). In the forward selection, longitude was included in the first step explaining 51.92% of data variation. Soil pH (KCl) was indicated as the second most important factor explaining 17.79% of data variation folloved by slope (15.14%) and latitude (15.14%). Of the significant environmental factors, longitude had the strongest pure effect on the variability of our data set (28.16%). All variables passing the forward selection had a significant pure effect on variability of the data set. They explained
78.64% of data variance when other significant variables were put into the RDA analysis as co-variables.
Figure 8 shows positions of relevés of the two associations in relation to four environmental varibles passing the forward selection. The first ordination axis was positively correlated with longitude (correlation coefficient 0.9461) and negatively with latitude (-0.7511). Latitude also negatively correlates with the third axis (-0.5090). The environmental variable with the highest correlation to the second axis is soil pH (-0.6824) and to the fourth axis is slope (0.7455). Relevés of the two studied associations were separated in the right and left parts of the ordination charts mainly based on longitude and latitude of the relevé plots (Figure 8). The geographical coordinates reflect the geographical position of relevés of the two studied associations which is strictly pronounced in our data set. All sites of Avenulo pratensis-Festucetum valesiacae without any exception were recorded in the Tribec Mts that is located in the southeastern part of the study area (Figure 1). Potentillo heptaphyllae-Festucetum rupicolae exclusively occurs in the northwest in the Biele Karpaty Mts and Povazsky Inovec Mts.
.0		• 5
-	Slope 21 7 21 «	8 .4 •
	17 Latitude ^	6 Longitude
-	19 15 14 18+ 20	3 12 9 2 13 2 «1 J11 10
.0 1-	PH	
-1.0	1.5
16 Slope PH \ Latitude 19	•3 • 1 110 Longitude
18 15 20 14 21 17 7	• HL- 4 12 89
-1.0	1.5
Figure 8: RDA diagrams of relevé positions of two acidophilous dry grassland associations (explanation of symbols are in accordance with Figure 6) in relation to four environmental factors passing the forward selection (9999 permutations, P<0.05). RDA diagram of the 1st and 2nd axes (left) and 1st and 4th axes (right) are depicted. The two main axes display of variance of the species data and of the species-environment relationship. The significant environmental factors represent 99.99% of the data variation. Table 4 shows the order of inclusion of individual variables in forward selection.
Slika 8: Diagram analize RDA in položaj popisov dveh asociacij acidofilnih suhih travišč (razlaga simbolov je enaka kot pri Sliki 6) glede na štiri okoljske dejavnike v metodi izbiranja (9999 permutacij, P<0.05). Prikazana sta RDA diagrama prve in druge osi (levo) in prve in četrte osi (desno). Glavni osi prikazujeta 31.4% variabilnosti in 75.4% korelacije med vrstami in okoljem. Statistično značilni dejavniki predstavljajo 99.99% variabilnosti podatkov. Tabela 4 predstavlja vrstni red vključevanja posameznih dejavnikov v metodi izbiranja.
Table 3: The outputs of the RDA analysis of the phytocoenological relevés from Table 1. Tabela 3: Rezultati analize RDA fitocenoloških popisov iz Tabele 1.
Axes	1	2	3	4 Total variance
Eigenvalues	0.219	0.095	0.062	0.041	1.000
Species-environment correlations	0.955	0.887	0.83	0.883	
Cumulative percentage variance of species data	21.9	31.4	37.5	41.6	
Cumulative percentage variance of species-environment relation	52.6	75.4	90.2	100.0	
Sum of all eigenvalues					1.000
Sum of all canonical eigenvalues					0.416
Table 4: Redundancy analysis, results of forward selection. Marginal effect - percentage variance explained by individual variable while used as the only constraining variable. Conditional effect - additional variance explained by the variable at the time it was included in the stepwise selection. Pure effect - percentage variance explained by the variable after all variables significant when alone were used as co-variables (ter Braak & Prentice 1988). Variance explained is shown as % of total inertia. Legend: ns - not significant, * P<0.05, ** P<0.01, *** P<0.001. Tabela 4: Analiza presežkov (RDA), rezultati vnaprejšnje izbire. Robni učinek - odstotek pojasnjene variance posamezne spremenljivke kot edine omejujoče spremenljivke. Pogojni učinek - dodatna pojasnjena variancaspremeljiv-ke, ko je bila ta vključena v postopno izbiro. Čisti učinek - odstotek pojasnjene variance posamezne spremenljivke po tem, ko so vse ostale spremenljivke uporabljene kot ko-spremenljivke (ter Braak & Prentice 1988). Pojasnjena variance je prikazana kot odstotek skupne variabilnosti. Legenda: ns - ni značilno, * P<0.05, ** P<0.01, *** P<0.001.
Environmental variable	Marginal effect	%	Conditional effect (in selection order)	%	Pure effect	%
Longitude	0.216""	51.92	0.216"""	51.92	0.117""	28.16
Soil pH (KCl)	0.087*	20.91	0.074"	17.79	0.080""	19.23
Slope	0.058ns	13.94	0.063"	15.14	0.067"	16.11
Latitude	0.159"""	38.22	0.063"	15.14	0.063"	15.14
Soil depth	0.080ns	19.23			0.060 ns	14.42
Altitude	0.112**	26.92			0.048 ns	11.54
Variance explained by significant variables				99.99		
4. DISCUSSION AND CONCLUSIONS
4.1 Species composition
The diagnostic species of Avenulo pratensis-Fes-tucetum valesiacae published in relevant syntaxo-nomical papers (Chytry et al. 2007, Dubravkova et al. 2010) and the differential species of the association assign in the current study (Table 1) differ partially. The species in common are the dominant grass Festuca valesiaca s.l. and also Genista pillosa and Poa bulbosa. Other species preferably occurring in Avenulo pratensis-Festucetum valesiacae that were assigned in the literature sources as diagnostic although they do not reach the threshold criteria set in this paper to be considered as differential are Asperula cynynchica, Avenella flexu-osa, Calluna vulgaris, Cladonia foliacea, Euphrasia
stricta, and Thymus pannonicus. Other diagnostic species of Avenulo pratensis-Festucetum valesiacae that equally occur in both associations in Table 1 are Acetosella vulgaris, Achillea millefolium agg., Acosta rhenana, Anthoxanthum odoratum, Eryngium campestre, Hypericum perforatum, Jasione montana, Koeleria macrantha, Sedum sexangulare, Trifolium arvense, Pilosella ojficinarum, Cladonia rangiformis, and Polytrichum piliferum. The data set from western Slovakia is totally lacking e.g. Agrostis vinealis, Armeria vulgaris, Gagea bohemica, Helichrysum are-narium, and Thymuspraecox that are the diagnostic species of higher fidelities in the above-mentioned publications. On the other hand, Aira caryophyllea occurs with higher frequencies specifically in the relevés from the Tribec Mts. This is a local peculiarity not typical for the sites of the association Ave-nulo pratensis-Festucetum valesiacae in other areas.
Comparing the diagnostic species of Potentillo heptaphyllae-Festucetum rupicolae listed in Chytry et al. (2007) and Michâlkovâ (2007) and the differential species from Table 1, we found 3 species in common (Festuca rupicola, Dianthus carthu-sianorum, and Teucrium chamaedrys). A species preferably occurring in this association that was assigned in the literature as diagnostic although in this paper it does not reach the threshold criteria for the differential species is Petrorhagia prolifera. Other diagnostic species of Potentillo hep-taphyllae-Festucetum rupicolae that equally occur in both associations in this paper are Acetosella vulgaris, Koeleria macrantha, Trifolium arvense, T. campestre, Pilosella officinarum, Potentilla argentea, Sedum sexangulare, Steris viscaria, Tithymalus cypa-risias, and Cladonia rangiformis.
In spite of the above-mentioned differences, the study sites show a close relation to the associations' descriptions in terms of species composition and the preferred environmental conditions. The reason why the data of acidophilous dry grasslands from quartzite of western Slovakia slightly differ from the published characteristics of the associations may be ascribed to the fact, that these data were not included in the datasets used for calculation of diagnostic species in Michâlkovâ (2007) and Dubravkovâ et al. (2010) with an exception of 2 relèves included in the second publication mentioned. What is more, there were not described any other acidophilous dry grassland associations dominated by Festuca valesiaca and F. rupicola in the territory of Central Europe. Therefore we considered these stands as data that shall enlarge not only the total distribution of these acidophilous dry grassland associations but also their internal variability. However, the differential species assigned in the current paper shall not to be used as diagnostic species of the two study associations because they are based on limited number of relevés originated from a geographically small area.
4.2 Environmental conditions and
ORDINATION ANALYSES
According to Chytry et al. (2007), the association Avenulo pratensis-Festucetum valesiacae occurs in dry regions with average yearly rainfall of 550 mm or lower. The more mesophilous association Potentillo heptaphyllae-Festucetum rupicolae occupies sites of similar soils but located in more
humid regions. This moisture gradient is also valid to the sites of the associations in western Slovakia. In southern Moravia in the Czech Republic, however, both associations occur practically at nearby sites on southeastern foothills of the Bohemian Massif (Chytry et al. 2007). In western Slovakia, the occurrence of the associations follows the humidity ranges more precisely and the occurrence of each association is limited in separate mountain areas. The sites of the studied associations do not occur close to each other but in climatically (considering the average yearly temperatures and rainfall) and topographically (altitude) pronounced areas. As an important factor representing the moisture gradient of the sites of the two associations is the precipitation range during the growing season and the average rainfall in July (in the Tribec Mts - up to 60 mm; Povazsky Inovec and Biele Karpaty Mts - up to 80 mm; Fasko & St'astny 2002b).
The strict spatial differentiation of sites of the two associations is most probably the reason of the great importance of longitude and latitude proved by the RDA analysis with forward selection. Longitude representing the west-east gradient also reflects partially the continental and oceanic influences expressed in the species composition of the acidophilous dry grasslands. However the portions of variance explained by significant variables passed forward selection remain surprisingly high. Out of the six studied environmental variables four passed the forward selection at P<0.05. An interesting fact is however, that the altitude did not pass the forward selection although it was the factor with the highest correlation with the first PCA axis.
4.3 The Koelerio-Phleion phleoidis
ALLIANCE
We assigned the two associations of interest to the Koelerio-Phleion phleoidis alliance following the traditional classification of Central European acidophilous dry grasslands, although Dubravkovâ et al. (2010) showed a close relation of this vegetation to the alliance Festucion valesiacae Klika 1931. We acknowledge that classification of the acidophilous dry grasslands in the Carpatho-Pannonian region of the eastern Central Europe to higher syntaxa is rather tangled, however we did not aim to answer the question of the Koeler-io-Phleion phleoidis delimitation in the current
paper. To solve this problem efficiently, following questions need to be addressed in some subsequent supraregional studies: (1) is the alliance Koelerio-Phleion phleoidis supported at all, (2) do the acidophilous dry grassland associations in eastern Central Europe belong to the same alliance as those originally described by Korneck (1974) from western Gemany, and (3) what is the relation of Koelerio-Phleion phleoidis to other higher units that include acidophilous vegetation, e.g. Koelerio-Corynephoretea Klika in Klika et Novak 1941? Although this paper does not deal with the topics outlined, the new original data published here might be an important component necessary for the later studies.
Acknowledgements
We are grateful to Milan Chytry for consulting the classification issues; Anna Kubinska and Anna Petrasova for determination of the bryo-phytes; Anna Lackovicova for identification of the lichens; Petr Smarda for determination of specimens of the genus Festuca and Jan Jahna for consultation on the geology of the sites. D. D. was supported by the Science Grant Agency of the Ministry of Education of the Slovak Republic and Slovak Academy of Sciences (VEGA 2/0181/09) and a grant through the EEA Financial Mechanism and the Norwegian Financial Mechanism and the state budget of the Slovak Republic (SK0115).
APPENDIX 1:
Origin of the relevés in Table 1.
PRILOGA 1:
Lokacije popisov iz Tabele 1.
The entries are structured as follows: relevé number; description of the locality; longitude; latitude; geology; soil; height of the herb layer; date (year/month/day); field number; a remark. Author of all relevés is J. Kost'al.
01. Tribec Mts, Ladice, a quartzite quarry NW of the village, 1 m of the SE edge of the quarry; 18° 15' 06.72" E; 48° 24' 04.35" N; quartzite; shallow soil; 25 cm; 2003/05/25; 11/2003; Festuca valesiaca s.l. = F. pseudodalmatica.
02.	Tribec Mts, Badice, a quartzite quarry located 500 m E of the village; 18° 08' 15.54" E; 48° 23' 50.99" N; quartzite; shallow rocky soil; 25 cm; 2003/06/06; 15/2003.
03.	Tribec Mts, Jelenec, an edge of a quarry located near the cemetery about 1.25 km SE of the water reservoir Jelenec; 18° 13' 37.76" E; 48° 23' 12.82" N; quartzite; shallow soil; 30 cm; 2003/05/17; 02/2003.
04.	Tribec Mts, Podhorany, Maly Bahorec Mt., 5 m to the SW of the hilltop; 18° 06' 05.58" E; 48° 23' 13.57" N; quartzite; shallow rocky soil; 15 cm; 2008/05/20; 25/2008.
05.	Tribec Mts, Jelenec, an edge of a quarry located near the cemetery about 1.25 km SE of the water reservoir Jelenec; 18° 13' 38" E; 48° 23' 12.75" N; quartzite; shallow rocky soil; 15 cm; 2003/05/17; 03/2003.
06.	Tribec Mts, Mankovce, Horka Hill (310 m a.s.l.), 15 m to the SEE of the hilltop; 18° 19' 19.67" E; 48° 25' 51.52" N; quartzite; shallow rocky soil; 50 cm; 2008/06/03; 55/2008; Festuca valesiaca s.l. = F. pseudodalmatica.
07.	Tribec Mts, Podhorany, Maly Bahorec Hill, 40 m to the SSW of the hilltop; 18° 06' 04.89" E; 48° 23' 13.05" N; quartzite; shallow rocky soil; 30 cm; 2008/05/20; 24/2008; Festuca valesiaca s.l. = F. pseudodalmatica.
08.	Tribec Mts, Krnca, a quartzite quarry at the W edge of the village; 18° 14' 36.03" E; 48° 32' 36.29" N; quartzite; shallow soil; 30 cm; 2003/05/23; 07/2003.
09.	Tribec Mts, Ladice, a quarry located 20 m to the W of the elevation point 304 m a.s.l.; 18° 14' 58.29" E; 48° 24' 02.56" N; quartzite; shallow rocky soil; 20 cm; 2008/05/23; 40/2008; Festuca valesiaca s.l. = F. pseudodalmatica.
10.	Tribec Mts, Ladice, a small quarry near a cherry orchard; 18° 15' 08.98" E; 48° 23' 58.07" N; quartzite; shallow rocky soil; 10-40 cm; 2008/06/03; 54/2008; Festuca valesiaca s.l. = F. pseudodalmatica.
11.	Tribec Mts, Ladice, a small quarry near the vineyards; 18° 15' 09.05" E; 48° 23' 58.27" N; quartzite; shallow rocky soil; 35 cm; 2008/ 06/03; 53/2008.
12.	Tribec Mts, Jelenec, an edge of a quarry located near the cemetery; 18° 13' 38.55" E; 48° 23' 15.84" N; quartzite; shallow rocky soil; 30-100 cm; 2008/05/23; 36/2008; Festuca valesiaca s.l. = F. pseudodalmatica.
13.	Tribec Mts, Jelenec, an edge of a quarry located near the cemetery; 18° 13' 38.08" E;
48° 23' 14.88" N; quartzite; shallow rocky soil; 35 cm; 2008/05/23; 37/2008.
14.	Biele Karpaty Mts, Chocholna-Velcice, Ska-licky vrch Mt., an elevation point Pod hajom (402 m a.s.l.); 17° 55' 26.82" E; 48° 53' 06.08" N; quartzite; shallow rocky soil; 15-35 cm; 2009/ 05/13; 01/2009.
15.	Biele Karpaty Mts, Chocholna-Velcice, Ska-licky vrch Mt., an elevation point Pod hajom (402 m a.s.l.); 17° 55' 27.01" E; 48° 53' 07.43" N; quartzite; shallow rocky soil; 10-35 cm; 2009/ 05/13; 02/2009.
16.	Biele Karpaty Mts, Drietoma, Ostra hor-ka Mt., SW slope; 17° 57' 20.64" E; 48° 54' 03.55" N; quartzite; shallow rocky soil; 15 cm; 2009/05/13; 03/2009.
17.	Povazsky Inovec Mts, Beckov, Skalice Mt.; 17° 54' 03.47" E; 48° 46' 38.08" N; quartzite; shallow rocky soil; 35 cm; 2008/06/20; 58/ 2008.
18.	Povazsky Inovec Mts, Salgovce, Stary vrch Mt., a NW edge of a quarry; 17° 53' 14.87" E; 48° 31' 51.52" N; quartzite; shallow rocky soil; 35 cm; 2008/06/17; 57/2008.
19.	Povazsky Inovec Mts, Beckov, Lasid Mt., SW slope, at the forest edge; 17° 55' 19.91" E; 48° 47' 08.58" N; quartzite; shallow rocky soil; 40 cm; 2008/06/20; 60/2008.
20.	Povazsky Inovec Mts, Salgovce, Stary vrch Mt., above an abandoned quarry; 17° 53' 19.47" E; 48° 31' 49.65" N; quartzite; shallow rocky soil; 50 cm; 2008/06/17; 56/2008.
21.	Povazsky Inovec Mts, Beckov, Skalice Mt., 50 m SW of the hilltop; 17° 54' 01.12" E; 48° 46' 36.88" N; quartzite; shallow rocky soil; 25-80 cm; 2008/06/20; 59/2008.
REFERENCES
Anonymus 1957: Vojenska topograficka mapa 1:25 000, mapovacie stvorce M-34-121-cd, M-34-121-dc, M-34-121-cc. (The army topographical map 1 : 25 000, map sections M-34-121-cd, M-34-121-dc, M-34-121-cc) [in Slovak]. Generalni stab ceskoslovenske armady. Depon. in Üstav krajinnej ekologie SAV, Bratislava.
Barkman, J. J., Doing, H. & Segal, S. 1964: Kritische Bemerkungen und Vorschläge zur quantitativen Vegetationsanalyse. Acta Bot. Neerl. 13:394-419.
Biely, A. 1975: Geologicka mapa Tribeca 1:50 000 (Geology map of the Tribec Mts) [in Slovak]. SGÜDS, Bratislava.
Biely, A. 1996: Vysvetlivky ku geologickej mape Slovenska 1:500 000. (Annotation to the geology map of Slovakia 1:500 000) [in Slovak]. Vydavatel'stvo D. Stura, Bratislava, 77 pp.
Borhidi, A. 1993: A Magyar flora szocialis maga-tartas tipusai, termeszetessegi es relativ okolo-giai ertekszamai. (Social behaviour types of the Hungarian flora, its naturalness and relative ecological indicator values) [in Hungarian]. Janus Pannonius Tudomanyegyetem No-venytani Tanszek, Pecs, 93 pp.
Braun-Blanquet, J. 1964: Pflanzensoziologie. Grundzuge der Vegetationskunde. 3 Aufl. Springer Verlag, Wien, 865 pp.
Dengler, J., Chytry, M. & Ewald, J. 2008. Phyto-sociology. In: J0rgensen, S. E.& Fath, B. D. (eds): Encyclopedia of ecology. Elsevier, Oxford, pp. 2767-2779.
Dubravkova, D., Chytry, M., Willner, W., Illyes, E., Janisova, M. & Kallayne Szerenyi, J. 2010: Dry grasslands in the Western Carpathians and the northern Pannonian Basin: a numerical classification. Preslia 82: 165-221.
Elias, P. 1986: Vegetacia statnych prirodnych rezervacii Hrdovicka a Solciansky Haj a pro-jektovanej SPR Kovarecka dubina (pohorie Tribec). (Vegetation of nature reserves Hrd-lovicka, Solciansky Haj and Kovarecka dubina in the Tribec Mts.) [in Slovak]. Rosalia 3: 33-79.
Fasko, P. & St'astny, P. 2002a: Mapa priemernych rocnych uhrnov zrazok, 1:2 000 000. (Map of the average annual precipitation, 1:2 000 000.) [in Slovak]. In: Hrnciarova, T. (ed.): Atlas kra-jiny Slovenskej republiky. (The landscape atlas of the Slovak Republic). MZP SR & SAZP, Bratislava, Banska Bystrica, p. 344.
Fasko, P. & St'astny, P. 2002b: Mapa priemernych uhrnov zrazok v juli, 1:2 000 000. (Map of the average precipitation in July, 1:2 000 000.) [in Slovak]. In: Hrnciarova, T. (ed.): Atlas krajiny Slovenskej republiky. (The landscape atlas of the Slovak Republic). MZP SR & SAZP, Bratislava, Banska Bystrica, p. 344.
Ferakova, V., Maglocky, S. & Marhold, K. 2001: Cerveny zoznam paprad'orastov a semennych rastlin Slovenska. (The red list of ferns and flowering plants of Slovakia.) [in Slovak]. In: Balaz, D., Marhold, K. & Urban, P. (eds.): Cerveny zoznam rastlin a zivocichov Slovenska. (The red list of plants and animals of Slovakia.) Ochr. Prir. 20 (Suppl.): 48-81.
Fiala, K. 1999: Zavazne metody rozborov pod.
Ciastkovy monitorovací systém - Poda. (Obligatory methods of the soil analyses. The partial monitoring system - the soil.) [in Slovak]. VÚPOP, Bratislava, 142 pp.
Futák, J. 1980: Fytogeografické clenenie Slovenska. Mapa 1 : 1 000 000. (The phytogeograph-ical division of Slovakia. Map 1 : 1 000 000) [in Slovak]. In: Bertová, L. (ed.): Flóra Slovenska IV/1. (Flora of Slovakia IV/1). Veda, Bratislava.
Geologická mapa Slovenskej republiky, mierka 1 : 50 000 (Geological map of the Slovak Republic, 1 : 50 000) [in Slovak]. SGÚDS, Bratislava. URL: http://www.geology.sk/in-dex.php?pg=geois.mapovy_server [accessed 2012-01-03].
Hegedüsová, K. 2007: Centrálna databáza fytoce-nologickych zápisov (CDF) na Slovensku [Central database of phytosociological relevés (CDF) in Slovakia]. Bull. Slov. Bot. Spolocn. 29: 124-129.
Hennekens, S. M. & Schaminée, J. H. J. 2001: TURBOVEG, a comprehensive data base management system for vegetation data. Journal of Vegetation Science 12: 589-591.
Herben, T. & Münzbergová, Z. 2003: Zpracování geobotanickych dat v pííkladech. I. Data o druhovém slození. (Analysis of geobotany data in examples. I. Species composition data). Katedra botaniky PrF UK, Praha, 118 pp.
Husová, M. 1967: Azidophile Eichenwälder auf Quarziten im Tribec-Gebirge, Slowakei. Fol. Geobot. et Phytotax. 2: 121-136.
Húsenica, J. 1964: Minerálna sila materskych hornín lesnych pod na Slovensku. (Minerals in the bedrocks of the forest soils in Slovakia) [in Slovak]. Vydavatel'stvo SAV, Bratislava, 246 pp.
Chytry, M. (ed.) 2007: Vegetace Ceské republiky. 1. Travinná a kerícková vegetace (Vegetation of the Czech Republic. 1. Grassland and heathland vegetation.) [in Czech with English summaries]. Academia, Praha, 526 pp.
Chytry, M., Hoffmann, A. & Novák, J. 2007: Su-ché trávníky (Festuco-Brometea). (Dry grasslands (Festuco-Brometea)). In: Chytry, M. (ed.): Vegetace Ceské republiky. 1. Travinná a keííc-ková vegetace (Vegetation of the Czech Republic. 1. Grassland and heathland vegetation.) [in Czech with English summaries]. Academia, Praha, pp. 371-497.
Chytry, M., Mucina, L., Vicherek, J., Pokorny-Strudl, M., Strudl, M., Koó, A. J. & Maglocky,
Š. 1997: Die Pflanzengesellschaften der west-pannonischen Zwergstrauchheiden und azidophilen Trockenrasen. Diss. Bot. 277: 1-108.
Ivanička, J., Polak, M., Hok, J., Hatar, J., Greguš, J., Vozar, J., Nagy, A., Fordinal, K., Pristaš, J., Konečny, V. & Šimon, L. 1998: Geologicka mapa Tribeča 1 : 50 000. (Geology map of the Tribeč Mts 1 : 50 000) [in Slovak]. MŽP SR, GÜDŠ, Bratislava.
Ivanička, J., Havrila, M. & Kohut, M. 2007: Ge-ologiska mapa Považskeho Inovca a JV časti Trenčianskej kotliny 1 : 50 000. (Geology map of the Považsky Inovec Mts and the SE part of the Trenčianska kotlina Basin 1 : 50 000) [in Slovak]. MŽP SR, ŠGÜDŠ, Bratislava.
Janišova, M., Hajkova, P., Hegedüšovä, K., Hriv-nak, R., Kliment, J., Michalkova, D., Ružičko-va, H., Rezničkova, M., Škodova, I., Tichy, L., Uhliarova, E., Ujhazy, K. & Zaliberova, M. 2007: Travinnobylinna vegetacia Slovenska -elektronicky expertny system na identifikaciu syntaxonov (Grassland vegetation of Slovakia - electronic expert system for identification of syntaxa) [in Slovak with English summaries]. Botanicky ustav SAV, Bratislava, 264 pp.
Jarolimek, I. & Šibik, J. (eds.) 2008: Diagnostic, constant and dominant species of the higher vegetation units of Slovakia. Veda, Bratislava, 332 pp.
Korneck, D. 1974: Xerothermvegetation in Rheinland-Pfalz und Nachbargebieten. Schrift-enr. Vegetationsk. 7: 1-196.
Lapin M., Faško P., Melo M. & Št'astny P. 2002: Mapa klimatickych oblasti, 1 : 1 000 000: (Map of the climatic regions, 1 : 1 000 000.) [in Slovak]. In: Hrnčiarova, T. (ed.) Atlas krajiny Slovenskej republiky. (The landscape atlas of the Slovak Republic). MŽP SR & SAŽP, Bratislava, Banska Bystrica, p. 344.
Marhold, K. & Hindak, F. (eds.) 1998: Checklist of non-vascular and vascular plants of Slovakia. Veda, Bratislava, 688 pp.
Mederly, P. & Hreško, J. 1988: Geologicky a geo-morfologicky vyskum ŠPR Solčiansky Haj, Hrdovicka a Kovarecka horka, II. čast'. (Geological and geomorphological study of the Solčiansky Haj National Nature Reserve, Hrdovicka and Kovarecka horka, 2nd part.) [in Slovak]. Rosalia: 23-46.
Michalkova, D. 2007: Koelerio-Phleion phleoidis Korneck 1974. In: Janišova, M., Hajkova, P., Hegedüšova, K., Hrivnak, R., Kliment, J., Michalkova, D., Ružičkova, H., Rezničkova, M.,
Skodova, I., Tichy, L., Uhliarova, E., Ujhazy, K. & Zaliberova, M.: Travinnobylinna vegeta-cia Slovenska - elektronicky expertny system na identifikaciu syntaxonov. (Grassland vegetation of Slovakia - electronic expert system for identification of syntaxa) [in Slovak with English summary]. Botanicky ustav SAV, Bratislava, pp. 86-89.
Palmer, M. W. 1993: Putting things in even better order - the advantages of canonical correspondence analysis. Ecology 74: 2215-2230.
Rolecek, J., Tichy, L., Zeleny, D. & Chytry, M. 2009: Modified TWINSPAN classification in which the hierarchy respects cluster heterogeneity. J. Veg. Sci. 20: 596-602.
Smarda, P. 2008: DNA ploidy level variability of some fescues (Festuca subg. Festuca, Poaceae) from Central and Southern Europe measured in fresh plants and herbarium specimens. Biologia 63: 349-367.
St'astny P., Nieplova E. & Melo M. 2002: Mapa priemernej rocnej teploty vzduchu, 1 : 2 000 000: (Map of the average annual air temperature, 1:1 000 000.) [in Slovak]. In: Hrnciarova, T. (ed.) Atlas krajiny Slovenskej
republiky. (The landscape atlas of the Slovak Republic). MZP SR & SAZP, Bratislava, Banska Bystrica, p. 344.
ter Braak, C. J. F. & Prentice, I. C. 1988: A theory of gradient analysis. Adv. Ecol. Res. 18: 271317.
ter Braak, C. J. F. & Smilauer, P. 2002: CANOCO reference manual and CanoDraw for Windows user's guide. Software for canonical community ordination (version 4.5). Biometris, Wageningen & Ceske Budejovice, 500 pp.
Tichy, L., 2002: JUICE, software for vegetation classification. Journal of Vegetation Science 13: 451-453.
Westhoff, V. & van der Maarel, E. 1973: The Braun-Blanquet approach. In: Whittaker, R. H. (ed.): Ordination and classification of communities. Dr. W. Junk Publishers, The Hague, pp. 617-727.
Zlinska, J. 2000: Vegetacia Holubyho lesostepi pri Vinosadoch v Malych Karpatoch. (Vegetation of the Holubiho lesostep near Vinosady in the Male Karpaty Mts) [in Slovak with English summary]. Acta Envir. Univ. Comenianae 10: 139-152.
Received 23. 1. 2012 Revision received 5. 7. 2012 Accepted 9. 7. 2012