_HACQUETIA 11/2 • 2012, 209-225_
DOI: 10.2478/v10028-012-0010-5
functional response traits and plant community strategy indicate the stage of secondary succession
Andrej PAUŠIČ & Andraž CARNI1' 2' *
Abstract
Changes of species composition, plant community strategy and functional response trait turnover were studied in a succession from dry pastures to a forest community (oak-hornbeam forests). The following question was asked: are functional response traits and plant community strategies indicators of TAA (time since agricultural land use abandonment), thus of a specific succession stage.
Indirect gradient analysis (DCA) was used in order to observe the position of the relevés along the axis and to correlate it with TAA. It was found that the position of relevés on DCA axis 1 is our proxy for TAA. Correlations (Spearman's rho) between the occurrence of plant functional traits and TAA were performed.
Low-growing herb species with scleromorphic leaves and green or red flowers are the predominant plant type on grassland areas, while plant species with digitate, hydro or mesomorphic leaves and white flowers typically prevail in forest. The proportion of chamaephytes increases immediately after land abandonment (afforestation). In a closed forest stand, there are many more herb species with vegetative propagation (bulbils). Herbal species in those stands most often reward pollinators with pollen. The ecological strategy of the entire plant community changes with spontaneous afforestation. On grassland, stress-tolerant species are dominant. After 10 years, the community is defined as CS and after 200 years as a community with a C-CS strategy. Key words: plant functional traits, land use transformations, secondary succession, Bela krajina, Slovenia.
Izvleček
Raziskava se ukvarja s spreminjanjem funkcionalnih rastlinskih znakov, ekoloških značilnosti vrst in ekološke strategije združbe skozi posamezne stadije zaraščanja pašnikov v odvisnosti od časa opustitve kmetijske rabe (TAA).
V raziskavi smo uporabili multivariatno DCA analizo in opazovali položaj florističnih popisov v DCA prostoru in jih korelirali s TAA. Izračunali smo Spearmanov korelacijski koeficient med pojavnostjo posameznega rastlinskega funkcionalnega znaka in TAA.
Nizkorastoče zeliščne vrste s sklerofilnimi listi in cvetovi rumenih in rdečih barv so prevladujoč tip rastlin na pašnikih. V gozdovih prevladujejo vrste z deljenimi, hidro ali mezomorfnimi listi in s svetlejšimi (belimi) cvetovi.
Delež hamefitov se v združbi po opustitvi kmetijske rabe močno poveča (proces zaraščanja). V sklenjenem gozdnem sestoju je opazen večji delež zeliščnih vrst, ki se razmnožujejo vegetativno (zarodni brstiči ipd.). Omenjene zeliščne vrste privabljajo opraševalce največkrat s cvetnim prahom.
Ekološka strategija celotne združbe se preko sekundarne sukcesije spreminja. Na pašnikih prevladujejo stres-toleratorji. Po desetih letih ima združba strategijo kompetitor/ stres tolerator, po dvesto letih pa kompe-titor-kompetitor/ stres tolerator.
Ključne besede: funkcionalni rastlinski znaki, sprememba krajinske zgradbe, sekundarna sukcesija, Bela krajina, Slovenija.
1	Institute of Biology, Scientific Research Centre of the Slovenian Academy of Sciences and Arts, Novi trg 2, SI-1000 Ljubljana, Slovenia. E-mail: andrej.pausic@zrc-sazu.si
2	University of Nova Gorica, Vipavska cesta 13, SI-5000 Nova Gorica, Slovenia.
* Corresponding author: carni@zrc-sazu.si
1. INTRODUCTION
Changes in biodiversity are most commonly evaluated through changes in the species composition. Many studies provide a precise assessment of the influence of land use transformations on vegetation composition and ecosystemic structure (at regional and global levels). That has led to new attempts to measure plant functional traits and establish which plant strategies reflect ecological and morphological adaptations (Par-tel & Zobel 1999, Garnier et al. 2001, Cousins & Eriksson 2002, Garnier et al. 2004, Castro et al. 2010, Saatkamp et al. 2010, Catorci et al. 2011, Prevosto et al. 2011, Vitasovic Kosic et al. 2011).
The change from grassland to forest as a consequence of the absence of human disturbance results in potential natural vegetation (Odum 1980). That is the final, stable stage resulting from the climatic and edaphic conditions in the area. Many studies have shown that during the afforestation process, the appearance of communities, ecological and morphological species turnover (Castro et al. 2010, Rehunkova & Prach 2010, Saatkamp et al. 2010, Latzel et al. 2010).
Rehunkova & Prach (2010) analysed the role of local site and landscape factors in the course of spontaneous succession in disused gravel-sand pits over a broader geographical area. They recognised plant functional traits as a powerful tool for predicting the colonization success of plants available in the local species pool. The next important study of secondary succession with functional response traits was by Castro et al. (2010). The authors assessed the response of species richness, composition and functional traits to decreasing land use intensity. They found changes in community strategy and species composition through the secondary succession.
In our study, we observed the changes in species composition, in functional response traits and in plant community strategy that occurred subsequent to the abandonment of agricultural land use. We were particularly interested in the ecological and morphological changes in the plant communities during the process of spontaneous afforestation (secondary succession).
We tried to complete a detailed study of the afforestation stages that result in the potential natural vegetation of the region. Our aims were (1) to study the species turnover process between different succession stages, (2) to understand the
process of functional trait and species habitat preference turnover between individual succession stages and (3) to study the appearance of a particular functional response trait in each succession stage and to correlate it with the time, since the agricultural land use was abandoned (TAA).
Our hypothesis: (A) specific functional response traits may be related to TAA and may therefore be strongly linked to plant species growing in a specific succession stage, (B) functional response traits and ecological plant characteristics are a good indicator of the succession stage and therefore of TAA.
2. METHODS
Study area
The research took place in the region of Bela krajina in SE Slovenia (Figure 1). This is a karst solution plain formed mainly by calcareous rocks, limestone and dolomites. On the surface, these rocks weather into chromic cambisols and luvi-sols, which sporadically even completely cover them. Annual precipitation in this part of Slovenia is 1300 mm and mean annual air temperature is 10.9 °C (ARSO 2011).
An area of ca. 1000 ha was selected (45.514535° -45.539406° N and 15.209397° - 15.246939° E) for the purpose of this research. The region lies on a Pleistocene karst corrosion plain, at an altitude between 160 and 420 metres and is fairly homogeneous in terms of geomorphology and climate.
The region experienced the gradual formation of a cultivated landscape, which began to change intensively at the beginning of the 20th century. This was a result of a period of migration of the local inhabitants out of Bela krajina. There were three major migration flows, with the first at the beginning of the 20th century, when people migrated to Western Europe and North and South America. The second wave of migration took place during WWII and the third wave resulted from delayed industrialisation in the 1960s, when the local inhabitants emigrated to larger industrial hubs (Orozen-Adamic et al. 1995). Today, the area is forested or under the process of secondary succession, as a consequence of land use abandonment.
Vrhov-Jica ^^
L	■ n . - ^ •	J^eliki vrh ^ q
Cresnjevec pri Dragatusu ^-jy	Q.
■-wr	) - v^0 25i
TAA (Years)	
O	0
O	10
•	20
•	45
•	210
0	500	1000 m
	1_i_i_i_i_I_i_i_i_i_I
Figure 1: Study area in Bela krajina (SE Slovenia). The dots represent relevé plots with attached TA A attribute.
Slika 1: Območje raziskav v Beli krajini. Prikazane so lokacije popisov s podatkom o času od opustitve kmetijske rabe oz.
paše (TAA).
211
Brought to you by | National & University Library Authenticated | 10.248.254.158 Download Date | 8/26/14 4:56 AM
Vegetation
Eighty-nine randomised plots (10 x 10 m) were selected on flat terrain away from depressions or sinkholes, in order to get samples with similar geomorphologic characteristics. The plots were located in different stages of forest succession (illyrian oak-hornbeam forests). The minimum distance between sample plots was 100 meters (less for plots of the stage A, since there were not enough appropriate sites to be included in the study). The period of time since agricultural land use abandonment of each plot (the TAA) was estimated by overlaying different digitalized old cadastral maps (1790, 1823 and 1913) and digital orthophotos (1954, 1975, 1986, 1999 and 2009) (Pausic & Carni 2012 b). If a pasture was clearly recognizable on a cadastral map of the year 1823, but abandoned in the next observed time interval (year 1913), we assumed that the time since agricultural land use was abandoned can be considered to be approximately 100 years (since we do not have information about the exact year that agricultural land use was abandoned). All the studied plot sites were abandoned once; therefore no anthropogenic influences were present after the land use was abandoned on these sites.
Communities were sampled according to the Central European method (Braun-Blanquet 1964). Data were stored in the Turboveg programme (Hennekens & Schaminée 2001).
Taxa nomenclature cited in the text is in agreement with Flora Europaea (Tutin et al. 1964-1993).
Plant functional traits
We analyzed functional response traits and habitat preferences (Table 1), divided into five groups. The traits were chosen from those proposed by Weiher et al. (1999) as indicators linked to the main plant population processes: dispersal, establishment and persistence.
Using the BIOLFLOR database (Klotz et al. 2002) and data on habitat preferences (Ellenberg et al. 1992), the species (249 recorded taxa) were attributed selected functional response traits, ecological strategy of species and data on habitat preferences of species for each selected plot in turn. As a result, a table was worked out for each plot, presenting the occurrence of each particular trait (number of taxa) in the plot (relevé) as for the whole cluster group.
Statistical analysis and data processing
Based on TWINSPAN (Hill 1979) classification analysis, the relevés were classified into 5 groups (A, B, C, D and E), using Juice 6.5 software (Tichy 2002). TWINSPAN pseudospecies cut levels for species abundances were set to 0-5-25 percentage scale units. Initially, six division levels were chosen. Later, different levels of division were accepted, resulting in 5 groups of relevés interpretable in terms of ecology. Using a fidelity index (Chytry et al. 2002, Tichy 2002) for each species, we were able to calculate the diagnostic species of each of the five cluster groups.
In the next step, we investigated the actual TAA of each relevé site with the help of old cadastral maps.
Indirect gradient analysis (DCA) of floristic data from relevé plots was performed with the Canoco program (ter Braak and Smilauer 2002), in order to observe the position of the relevés along axes and to calculate their projection values along axis 1.
Instead of the TAA attribute, we decided to correlate the values from DCA axis 1 with functional response traits, since some older cadastral maps give inaccurate attributions and using this method minimised errors in our study. The position on DCA axis 1 is therefore our proxy for TAA.
The correlations (Spearman's rho) between TAA and selected functional response traits (Table 1) were calculated in Statistica 8.0 (Statsoft Inc. 2007).
The average ecological strategy of the entire community (stage) at a specific succession stage was calculated from the ecological strategy of each sample, with the C-S-R Signature Calculator 1.2 program (Hunt et al. 2004). The program consists of conversion and comparator tools. The conversion of floristic data into a C-S-R signature is carried out automatically by the first, or 'calculator', part of the new spreadsheet tool. The user pastes-in a data matrix containing quantitative records from one vegetation sample. The tool calculates the percentage abundance of each functional type. The second, or 'comparator', part of the tool accepts a selection of C-S-R signatures transferred manually from the 'calculator' part. The positions of all of these signatures are then plotted in C-S-R space. The direction and magnitude of any differences between samples with respect to C, S and R components are reported (Hunt et al. 2004).
Table 1: Selected plant functional traits and ecological characteristics. Tabela 1: Izbrani funcionalni rastlinski znaki in ekološke značilnosti vrst.
TRAITS USED IN STUDY	DESCRIPTION	% of missing data
A) Traits describing the vegetative morphology of the species		
1. Life form (Raunkiaer 1934)	Life form refers to the vertical position of vegetative buds (as an adaptation to adverse seasons)	0 1
2. Method of vegetative propagation	We distinguished runners, propagation with bulbils, fragmentation, rhizome and bulbs.	5.2
B) Traits describing the shape and morphology of photosynthesising leaves		
1. Leaf form (Günther 1987)	Plants with grass-like, simple, full, digitate, pinnate and needle were found.	0 1
2. Leaf anatomy (Frank and Klotz 1990)	We distinguished helomorphic, hygromorphic, mesomorphic and sclero-morphic leaves.	0 1
C) Traits describing the flower shape and reproductive biology of the species:		
1. Flower shapes (Müller 1881)	Müller classified insect pollinated flowers into 9 classes. The main aim was to achieve a grouping of pollinators. Nectariferous flowers were grouped according to the depth of nectar display (flowers with open, partly hidden and hidden nectar).	0 1
2. Flower shapes according to Kugler (Kugler 1970)	Kugler distinguished 10 major flower types: disk- and bowl- shaped flowers, funnel flowers, bell-shaped flowers, stalk disc flowers, lip flowers, flag blossoms, flower heads, spike flowers, brush flowers and trap flowers.	0 1
3. Flower colour	We observed flowers with blue, brown, green, red, violet, white and yellow flowers.	0 1
4. Beginning of flowering	We selected three months in which most of the species start to bloom (March, April and May).	0 1
5. Duration of flowering	Most of the species in the study area have a flowering duration from 2 to 4 months.	2.8
6. Fruit type (Bässler et al. 1996; Strasburger et al. 1998)	The fruit is defined here only as the fruit at the time of seed ripening. The	
	remaining parts of the flower are treated as "additional structures" and are identified with germinules. Fruits are categorized according to characteristics of seed maturation, pericarp or arrangement of the pericarp.	2.8
7. Diaspore type	Generative diaspores (units of dispersal) may be seeds or can be imbedded in additional structures or an additional structure can be attached to them (fruit with appendage, infructescense, seed, spore, vegetative).	3 1
8. Diaspore weight (mg)	We set three investigated weight classes: diaspores with weight up to 1 mg, from 1 to 50 mg and above 50 mg	15.2
9. Floral rewards for pollinators (Ayasse et al. 2000; Gumbert and Kunze 2001)	The plant species were distinguished in 3 groups according to the floral reward offered to pollinators (nectar, pollen and deceit).	10.8
D) Data on distribution areas and species ecology		
1. Species continentality (Meusel & Jäger 1992)	Continentality characterizes the range of a plant species from the coasts to the centres of continents.	0 1
2. Ellenbergs indicator values for individual plant species (Ellenberg 1992)	Simple ordinal classes of organisms (initially plants) with a similar realized ecological niche along a gradient. The latest edition of Ellenberg's indicator values contains values on a 9 point scale for soil acidity, productivity/ nutrients, soil humidity, continentality, soil salt content and light.	3.2
E) Ecological strategy of the species		
1. C-S-R (Grime et al. 1997)	Description of plant ecological strategies. Grime distinguishes 3 major -extreme groups (stress tolerators, ruderals and competitors) and a combination of those groups.	18 1
2. Species life span	The life span refers not only to the actual life span of species (annuals, biennials, perennials).	0 1
3. RESULTS
Twinspan classification divided the relevés into 5 separate groups, corresponding to the TAA.
We calculated median values for the TAA of each cluster group individually. The results are: group A - 0 years, B - 10 years, C - 20 years, D - 45 years and E - 210 years abandoned plots.
Vegetation on grasslands was classified into group A. The largest proportion of heliophilous species (highest fidelity; Phi coefficient) occurs in this group (Genista germanica, Leontodon hispi-dus, Potentilla erecta, Calluna vulgaris). This stage can be described as Calluna stage.
The following stage is the Pteridium-Frangula stage (group B), in which Pteridium aquilinum and shrub species (Frangula alnus) prevail. This is the initial phase of the secondary succession process.
Stage C is the Betula stage. In this stage of secondary succession, a dense forest formation is visible, with species such as Betula pendula and Populus tremula.
Stage D is a forest stage, in which Carpinus betulus is the dominant tree species. This stage is dominated by Carpinus betulus and heliophilous herbs disappear. The herb layer is dense, consisting of typical forest species. Epimedium alpinum, Fragaria moschata and Aremonia agrimonoides are characteristic herb species of this stage. Such a
stage of secondary succession occurs 45 years after land abandonment. Stage D is considered to be the Epimedium-Carpinus stage.
Stage E is forest with a characteristic herb layer composed of sciophilous forest species (e.g., Pulmonaria officinalis, Anemone nemorosa, Galium sylvaticum). The forest is two-layered, with Quercus petraea as the characteristic tree species in the upper tree layer and Carpinus betulus in the lower tree layer. We consider stage E to be the Carpinus-Quercus stage, the end stage of secondary succession.
The relevés indicate species abundance in relation to the TAA (Figure 2). Species such as Frangula alnus and Calluna vulgaris soon disappear in the afforestation process, due to the changed ecological conditions. Pteridium aquilinum is abundant in stages A, B and C and disappears completely in the final stage E. Betula pendula in the tree layer has the highest abundance in stage C.
The D stage has characteristic herb species; Aremonia agrimonoides, Fragaria moschata and Epimedium alpinum, which show the highest abundance and fidelity values (Phi coefficient).
In older forest stands (E) dominated by Quercus petraea; Anemone nemorosa, Polygonatum mul-tiflorum, Pulmonaria officinalis and Tamus com-munis have the highest fidelity coefficient values (Figure 2) and could justifiably be considered to be indicator species for forest stands older than 200 years.
Figure 2: Synoptic table showing the turnover of species in the time gradient. Clusters (stages) are indicated by capital letters: A - Calluna stage, B - Pteridium-Frangula stage, C - Betula stage, D - Epimedium-Carpinus stage and E - Carpinus-Quercus stage. The layers are indicated as: 1 upper tree layer, 3 lower tree layer, 4 shrub layer and 6 herb layer.
Slika 2: Sinoptična tabela florističnih popisov iz Bele krajine. Jasno je vidna premena vrst skozi stadije sekundarne sukcesije. Različni stadiji so označeni kot: A- stadij Calluna, B - stadij Pteridium-Frangula, C - stadij Betula, D - stadij Epimedium-Carpinus in E - stadij Carpinus-Quercus.
Figure 3: DCA ordination diagram with 89 relevés shows the separation of relevés into five groups (A-E), which correspond to the passive projection of the TAA.
Slika 3: DCA ordinacija prikazuje razporeditev florističnih popisov v pet skupin (A-E)(kot so bile določene po Twinspan klasifikaciji) po stadijih sekundarne sukcesije.
DCA diagram analysis shows the position of 89 relevés, divided into five distinct groups (Figure 3). The distinction corresponds to groups made by Twinspan analysis. The relevés are arranged in groups according to the TAA and are along axis 1 (eigenvalue 0.636) in DCA. Axis 2 has a much lower eigenvalue of 0.261.
We subsequently performed a correlation statistical test and calculated Spearman's (rho) correlation coefficients for each measured functional plant trait to TAA; these are shown in Table 2.
Table 2: Correlation (Spearman rho) of selected plant traits and ecological characteristics with TAA. Legend: * p<0.05, ** p<0.01, n.s. non significant
Tabela 2: Spearmanov koeficient korelacije (rho) izbranih funkcijskih rastlinskih znakov in ekoloških značilnosti s TAA.
Legenda: * p<0.05, ** p<0.01, n.s. ni korelacije.
Traits describing vegetative morphology of the species
Traits describing leaf form and morphology LEAF FORM (Raunkiaer 1934)
digitate full
grass-like needle pinnate simple
0.41 (**) 0.16 n.s. -0.68 (**) -0.30 (*) -0.17 n.s. 0.14 n.s.
LEAF ANATOMY (Frank & Klotz 1990) helomorphicl -0.16 n.s. hygromorphic -0.51 (**) mesomorphic -0.02 n.s. scleromorphic -0.64 (**)
Traits describing flower shape and reproductive biology of the species
Flower classes according to MÜLLER (Müller 1881)
flowers with open nectar
flowers with totally hidden nectar
flower associations with totally hidden nectar
butterfly flowers
bee flowers
-0.03 n.s. 0.33 (*) 0.69 (**) -0.17 n.s. -0.31 (*)
VEGETATIVE PROPAGATION
bulb Bulbil
fragmentation
rhizome
runner
0.19 n.s. 0.31 (*) -0.05 n.s. -0.09 n.s. -0.59 (**)
Flower types according to KUGLER (Kugler 1970) pollen flower	-0.21 n.s.
disc flowers with nectar open	-0.01 n.s.
disc flowers with nectar ± hidden
nectaries at base of stamens	0.58 (**)
funnel, tube flowers (large)	-0.26 (*)
funnel, tube flowers (small)
bell shaped flowers with sticky pollen
true lip flowers
lip flowers, Orchidaceae type
flower heads, Asteraceae
-0.13 n.s. Q.Q8 n.s. Q.Q5 n.s. -Q.18 n.s. -Q.4Q C**)
FLOWER COLOUR
blue brown green red, purple violet
-Q.14 n.s. -Q.Q2 n.s. -Q.38 C**) -Q.5Q C**) -Q.Q8 n.s.
white yellow
-Q.3Q C*) -Q.Q8 n.s.
BEGINNING OF FLOWERING
March	0.30 (*)
April	-0.31 (*)
May	-0.01 n.s.
DURATION OF FLOWERING
2 month	0.29 (*)
3	month
4	month
-Q.Q4 n.s. -Q.63 C**)
FRUIT TYPE (Bässler et al. 1996; Strassburger 1998) berry	(*)
capsule	-0.66 (**)
nut ^^^^^^Ho.33 (*) schizocarp	0.02 n.s.
DIASPORE TYPE
infructescence
fruit with appendage
seed
spore
vegetative
-Q.12 n.s. Q.31 C*) -Q.33 C*) Q.65 C**) Q.Q7 n.s.
DIASPORE WEIGHT (mg)
Q.l - 1	-Q.68 C**)
1 - 5Q	-Q.12 n.s.
5Q+	-Q.37 C**)
FLORAL REWARDS
Kunze 2001)
CAyasse et al. 2QQQ; Gumbert &
nectar pollen deceit
-Q.Q2 n.s. -Q.33 C*) -Q.32 C*)
Habitat preferences - ELLENBERG VALUES
CEllenberg 1992)
-Q.83 C**) -Q.36 C**) -Q.15 n.s. Q.ll n.s. Q.68 C**)
light
temperature continentallity moisture soil reaction nutrients 0.71 (**
SPECIES LIFE SPAN
annual	-0.32 (*)
perennial	0.03 n.s.
Data on distribution areas, species ecology
OCEANITY - CONTINENTALLITY (Mäusel & Jäger 1992)
species of continental climate but ranging
into sea climate	-0.01 n.s.
species of sea climate	-0.19 n.s. species missing in extreme continental climate
and extreme sea climate	0.30 (*)
In the group of traits describing the vegetative morphology of the species (Table 2), the formation of the bulbils trait and runner trait show a significant correlation trend with TAA in the vegetative propagation group. The runner trait has a high negative correlation with the time gradient, while the bulbil trait has a positive one.
In the observed traits describing leaf anatomy, there was a positive correlation between TAA and the hygromorphic leaf trait, while the sclero-morphic leaf trait showed a negative correlation with TAA.
Among the traits describing the flower shape and reproductive biology of the species, the number of flowers with totally hidden nectar in the receptacle and flower associations (inflorescences) with totally hidden nectar increase along the TAA gradient. On the other hand, the bee flower trait negatively correlates with increasing TAA. There is also a strong positive correlation between the category of disc flowers with nectar ± hidden nectaries at the base of stamens and TAA and a negative correlation between the funnel flower trait and TAA.
In the category flower colour, we found that a white colour has a positive correlation with TAA, while on open grassland; plants with red and green flower colours prevail.
The beginning of flowering, in April (on grasslands) and in March (in forests) is also significant. A flowering duration of 2 months correlates positively with TAA, while a flowering duration of 4 months correlates strongly negatively with TAA.
In the fruit type group, berry and nut correlate positively with TAA, which indicates that the number of species with these traits increases with the land use abandonment stage. The capsule trait has a strong negative correlation with TAA.
Within the diaspore group, the spore trait and the fruit with appendage trait have a high posi-
tive correlation with TAA. The seed trait has a negative correlation. The diaspore weight class up to 1 mg correlates negatively with TAA and the class weight more than 50 mg positively.
In the floral rewards category, pollen is the only trait which significantly correlates positively with TAA, the deceit trait shows a negative correlation.
Among data on habitat preferences and species ecology: the trait "species missing in an extreme continental climate" correlates positively with TAA. Ellenberg values of temperature, soil reaction and nutrients correlate with TAA positively, while the Ellenberg value for light trait shows a strong negative correlation.
In relation to plant life form changes (Figure 4), we determined that grassland (Calluna stage) was dominated by hemicriptophytes (63.6%), followed by macro phanerophytes with 15.1%, geophytes contributed 7.6% and 6% of all recorded species were nano-phanerophytes. There were 4.2% chamaephytes and 3.2% thero-phytes. The annual life span trait shows a negative correlation with TAA.
In the first stage of afforestation (Frangula-Pteridium stage), 44% of all species were hemic-ryptophytes, 19.7% of the species were nano-phanerophytes, 18% were macro-phanerophytes, 13.1% chamaephytes, 3.3% geophytes and 1.6% of the recorded species were therophytes.
The Betula stage is a forest stand that indicates an already slightly altered state. Groups were represented as follows: macro-phanerophytes 46%, hemicriptophytes 25.7%, nano-phanerophytes 19%, chamaephytes 5.4%, geophytes 2.7% and thero-phytes 1.2% of all recorded species.
The species composition in young forest with a median TAA of 45 years (Epimedium-Carpinus stage) was as follows: macro-phanerophytes predominate with 40.6%, followed by hemicriptophytes with 28.3%, nano-phanerophytes with 20.76%, geophytes with 9.4% and chamaephytes with 0.94%. The species composition in old forests with a median TAA of 210 years (Carpinus-Quercus stage) was: macro-phanerophytes 42.1%, hemicriptophytes 25.5%, nano-phanerophytes 18.72%, geophytes 12.7% and chamaephytes 0.98%.
Figure 4: Life form changes in different succession stages. A -Calluna stage, B - Pteridium-Frangula stage, C - Betula stage, D - Epimedium-Carpinus stage and E - Carpinus-Quercus stage.
Slika 4: Premena življenjskih oblik rastlin v različnih stadijih sekundarne sukcesije. A - stadij Calluna, B - stadij Pteridium-Frangula, C - stadij Betula, D - stadij Epimedium-Carpinus in E - stadij Carpinus-Quercus.
Change of plant community strategy in the TAA gradient.
Grasslands and grasslands under afforestation are dominated by S-CS strategists (Figure 5). In the course of secondary succession, through abandonment of grazing and mowing, both species and the community as a whole in the C-S-R triangle show an increasing tendency toward the C-CS category (in the last observed succession stage E).
Plant community
Herb layer
Figure 5: Changes in CSR signatures of the whole plant community and herbal layer in secondary succession. A - Calluna stage, B - Pteridium-Frangula stage, C - Betula stage, D - Epimedium-Carpinus stage and E - Carpinus-Quercus stage. Slika 5: Premena strategije rastlinske združbe za celotno združbo in zeliščno plast. A - stadij Calluna, B - stadij Pteridium-Frangula, C - stadij Betula, D - stadij Epimedium-Carpinus in E - stadij Carpinus-Quercus.
C
C
R
S
R
S
Additional analysis was performed for the herb layer change through secondary succession. Grassland herbs mainly possess a stress-tolerator strategy, while in the last stage (state E), forest herbs demonstrate the traits of C-CS strategists (the herb layer is more pronounced than the whole plant community). The turnoff of the whole community life strategy in succession stages B, C and D is very similar to the life strategy turnoff of the herbal layer (Figure 5).
4. DISCUSION
Landscape abandonment has resulted in changes in floristic composition in Bela krajina. Similar results have also been found elsewhere in Europe and the world in general (Sternberg et al. 2000, Vesk & Westoby 2001, Adler et al. 2004, Carni et al. 2007, Johansson et al. 2010, Rehunkova & Prach 2010, Carni et al. 2011, Pausic & Carni 2012).
Detailed studies on the early successional patterns of forest species were already carried out in the 1980s. Halpern (1989) studied the interactions of life history traits and disturbance in clear cut and burned Pseudotsuga forests in the Cascade Range of Oregon. The detailed study by Leps & Stursa (1989) on the life history strategies change in secondary succession in the Krkonose Mountains already included the plant strategy turnover but encompassed a shorter successional interval (20 years). The results of these studies refer to short-term experiments, while the changes documented in our study cover a much longer time period.
The plant community composition co-develops in newly developed ecological conditions, as is also evident from the results of our study: during the first 10 years after land abandonment, grassland alters into shrubland of the Pteridium-Frangula stage. After the next 10 years, this stage transforms into the Betula stage, which needs an additional 25-30 years (or 50 years in total) to be dominated by Carpinus betulus (Epimedium-Carpinus stage). The last stage in secondary succession in the study area is the Carpinus-Quercus stage, which occurs after 200 years of land abandonment. The course of succession is fast at the beginning, since abandonment results in the increased growth of competitive plant taxa that were unable to grow in the previous disturbed environment.
One of the characteristics of species in a closed stand is an ability to propagate vegetatively via bulbils (Cardamine bulbifera). The occurrence of species with such a form of vegetative propagation in older forests is due to the higher proportion of competitors (Grime et al. 1997) in closed forest stands. Runners are plants found mostly on grasslands and shrubland (Ajuga reptans, Carex pilosa, Fragaria vesca). Such a form of propagation enables the species to populate a larger distribution area over a relatively short time period (Mony et al. 2010).
Changes in leaf morphology and method of vegetative propagation in relation to TAA
Scleromorphic leaves possess a special cuticle and epidermis. They often comprise additional structures to facilitate water retention or to reduce evapotranspiration (trichomes, specialized sunken leaf stomata, special leaf dyes etc.). Species with sclero-morphic leaves are common in grasslands and do usually not appear in closed forest stands (Doryc-nium germanicum, Carlina acaulis, Calluna vulgaris) (Carni et al. 2010, Rehunkova & Prach 2010).
It has been determined that the number of species with hygromorphic leaves increases with forest stand age (Rehunkova & Prach 2010). Hy-gromorphic leaves have a characteristic soft tissue typical of sciophilous species adapted to high air humidity. The leaf cuticle and epidermis are very thin. Such leaves are characteristic of species from shady forests (Cyclamen purpurascens, Anemone nemorosa).
Over the time since abandonment, leaf shape changes greatly. On grasslands, species with grass-like leaves ((Poaceae) and solid, needle-like leaves (Juniperus communis) dominate, while plant species with digitate leaf shapes (Acer, Crataegus) occur in forests.
Shape, colour, morphology of reproductive organs and fruits
The general flower shape (Müller 1881) shows a tendency to change along secondary succession phases. On open grasslands, forest edges and shrubland, the prevailing types of flowers are funnel or tube shaped (Gentiana asclepiadea, Primula vulgaris, Vinca minor, Buphthalmum salicifolium). According to the Kugler classification (Kugler
1970), through spontaneous afforestation and transformation of areas into forest, flowers assume the characteristics of disc-type flowers with hidden nectar (Crataegus laevigata). Flowers with totally hidden nectar are much more frequent in forests than in open grasslands (Fragaria moschata).
The colour of the flowers in an individual succession stage certainly has a significant role.
Most species in open grasslands have red or purple flowers, such as Calluna vulgaris, Centaurea jacea, Dianthus barbatus, Orchis morio, Polygala amara (Forrest & Thomson 2009). A red flower is black in the UV spectra, in which most insect pollinators sense (Chittka & Raine 2006). It is therefore in strong contrast with the surrounding green area (which remains the same colour in UV). In forest stands, the prevalent flower colour is white (Anemone, Convallaria, Galium, Polygona-tum). White appears in the UV spectra as a strong blue, which helps plants to be visible when they grow in the (dense) forest herbal layer (Forrest & Thomson 2009, Miller et al. 2011).
Most plants of open grasslands start to flower in April and have a flowering duration of 4 months. In forests, plants already start to flower in March, with an average duration of flowering of about 2 months (Forrest & Thomson 2009, Pel-lissier et al. 2010). The flowering period of grassland species is thou longer, lasting for almost (depending on the species) the entire vegetation period. Strong winds in spring, however, facilitate the pollination of some wind pollinated (anemo-gamic) plants in forest from the family Betulace-ae, such as Corylus avelana and Carpinus betulus. The sample plots in forest contained several early flowering insect pollinated forest species (Daphne mezereum, Pulmonaria officinalis, Vinca minor, Primula vulgaris, Corydalis cava).
Other species on grasslands, though, have much more time to develop flowers, since they are clearly visible to pollinators for most of the vegetation season. Grassland species thus tend to have a long flowering period (Debussche et al. 1996, Casado et al. 2004, Peco et al. 2005, Pellis-sier et al. 2010).
The most prevalent fruit type in pastures (myr-meco, anemochorous) is capsules (Orchis morio, Ophrys sphegodes) (Lengyel et al. 2010), while in closed forest stands, plant species most often pro-
duce nuts (Fagus sylvatica, Corylus avellana) and berries (Actaea spicata, Berberis vulgaris, Convallaria majalis, Frangula alnus, Polygonatum multiflorum).
Spore is of the diaspore type, which is mostly bound to old forest stands. Most fern species (Pteridopsida) grow in forest (Asplenium tricho-manes, Athyriumfilix-femina, Dryopterisfilix-mas), while they are not so common on (often dry) grasslands, since the prothallium (gametophyte stage) needs a lot of moisture for successful fertilization (Hua et al. 2010). Seed is predominantly of the diaspore type among heliophylus grassland species (Dianthus barbatus), while fruit with an appendage characterizes forest phanaerophytic species, such as Carpinus betulus or Betulapendula. The reason is the different dispersion vector and strategy (Tackenberg et al. 2006). With the beginning of the afforestation process, the diaspore weight increases. A diaspore weight up to 1 mg is typical of (anemochorous, zoohorous) grassland species (Agrostis capillaris, Briza media, Clinopodi-um vulgare), while forest species have much heavier diaspores, normally more than 50 mg (Cornus mas, Quercus petraea, Fagus sylvatica). Grassland plant species have relatively smaller, lighter seeds (often with pappi or narrow wings), since they are normally spread by anemochory (Vittoz et al. 2009, Rehunkova & Prach 2010), while forest ones need to be stronger, bigger and often with an edible appendage, since they are mostly spread by autochory and zoochory (insects, mammals) (Heinken et al. 2006, Rehunkova & Prach 2010).
The floral reward for pollination by an insect in forest species is usually pollen (Convallaria majalis, Cyclamen purpurascens), while a zoochorous plant on open grassland or pasture rewards prospective pollinators with deceit pollination. The latter is especially typical of mimicry species from the family Orchidaceae (Ophrys, Orchis) on dry grassland (Pellissier et al. 2010). We conclude that the land abandonment process increases zo-ochory and autochory and decreases anemochory (Rehunkova & Prach 2010).
Ellenberg values for light and temperature indicate that land abandonment and the afforestation process change the ecological conditions for the plant community. During secondary succession after agriculture land use abandonment, organic matter in the soil increases and some minerals and nutrients drawn up by plants (e.g. Betula) from lower soil horizons accelerate decomposition and mineralization by higher decomposition activity. With the decomposition of organic litter,
the concentrations of exchangeable Ca2+, P, Mg2+, as well as other nutrients, increase, as does the soil pH (Carni et al. 2007, Pausic & Carni 2012).
Changes of plant life forms, changes in functional response traits and habitat preferences in relation to land use change have previously been observed in other studies (Castro et al. 2010). It has been established that the process of afforestation of (open) grassland areas fundamentally changes the life form of the entire community (Fig. 4).
With agricultural land use change (abandonment), the ecological strategies of both plant species and the entire community change. We were also interested in this change. We observed a change in the species strategy of the entire community and in the herbal layer over the course of secondary succession periods (A, B, C, D and E). Our premise was that the community as a whole and the herb layer change according to a similar pattern.
During the course of secondary succession, from grasslands to forests, a shift from S-CS towards C-CS strategy is evident (i.e., species for which competition is the ecological strategy that dictates the existence of individual species on a specific site, a feature of phanerophytes). This is in fact characteristic of older forest.
Additional analysis was performed only for the herb layer change through secondary succession. The range of changes in the herb layer during secondary succession is slightly more extreme than in the community as a whole. It is evident that the community strategy in succession stages B, C and D is very similar, whereas in the herb layer, the transition from one succession stage to another is also reflected in a change of strategy of the herb layer.
Species of grassland and pastures are far more adapted to stress, to human disturbance. Constant mowing or grazing therefore prevent the occurrence of species other than herb species and occasional trees or shrubs. Spontaneous afforestation also reduces the accessibility of certain essential living resources (such as light, space), so species require another strategy to prosper in such environments. They become competitors.
Our methodological approach permitted the identification of a series of individual attributes that are positively or negatively associated with TAA. There is no doubt that the results from this study are linked to a series of methodological decisions. The choice of the traits used was necessarily pragmatic (Peco et al. 2005), in an attempt to minimise the number of chosen traits and the
effort required for their measurement, while at the same time maximising their functional relationship with secondary succession.
5. CONCLUSIONS
Overall, our study showed that secondary succession in Bela krajina is a result of high species turnover, as well as change in plant functional traits and structural changes of the whole community (Sternberg et al. 2000, Vesk & Westoby 2001, Adler et al. 2004, Carni et al. 2007, Johansson et al. 2010, Rehunkova & Prach 2010).
After human activity has been abandoned, species from the forest edge or from nearby forest areas start migrating to the grassland (a changed ecological regime) due to the changed ecological conditions. Over the course of several years, not only does the species composition change, but also the ecological strategy of the whole community, the manner of propagation, duration of flowering, morphology of the plant habitus, leaf forms (species are replaced by those that are best adapted to the new ecological conditions). After land has been abandoned, around 50 years are needed for the development of Carpinus forest and 200 years for the development of two-layered Quercus-Carpinus forest, which can be seen as the potential vegetation of the study area.
Our research findings indicate that:
1.	the rate and direction of development, changes in the plant community (and ecosystem), correlate with abiotic ecological factors (in our case, above all the time gradient) but this impact becomes obvious only after the abandonment of land use (after cessation of the disturbance that hindered the process of secondary succession);
2.	changes of anthropogenic activity lead to a change in the floristic inventory of the community. Secondary succession starts when anthropogenic disturbance is absent and the species and the community as a whole show an entirely different morphological-ecological turnover;
3.	land use results in an artificially stable system (quasi-equilibrium), which collapses after the land is abandoned and tends toward the natural equilibrium.
Figure 6: Steljniki near the village of Vinomer. Steljniki represent a transition phase between the Pteridium-Frangula (B) and Betula (C) stages of secondary succesion. The annual removal of litter protect steljniki from overgrowing. Slika 6: Belokranjski steljniki pri kraju Vinomer. Steljniki predstavljajo prehod med sukcesivnima stadijema Pteridium-Frangula (B) in Betula (C). Letna košnja in odstranjevanje stelje steljnike varujeta pred zaraščanjem.
Figure 7: Approximately 45 years old forest stand, Epimedium-Carpinus stage. Slika 7: Gozdni sestoj, star približno 45 let. Stadij Epimedium-Carpinus.
6. POVZETEK
Funkcionalni rastlinski znaki in ekološka strategija združbe označujejo posamezen stadij sekundarne sukcesije
Raziskava je imela tri cilje: (1) študija premene vrst med sekundarno sukcesijo, (2) razumevanje premene izbranih funkcionalnih rastlinskih znakov in ekoloških značilnosti vrst med stadiji sekundarne sukcesije in (3) študija pojavljanja rastlinskih funkcionalnih znakov v stadijih sekundarne sukcesije in korelacija le-teh s časom od opustitve kmetijske obdelave oz. paše (TAA).
Raziskava je potekala v Beli krajini, na območju med vasmi Bojanci, Butoraj, Dragatuš in Tribuče. Območje je bilo do začetka 20. stoletja še intenzivna kmetijska krajina, nato pa se je zaradi izseljevanja prebivalstva in opuščanja kmetijske dejavnosti krajinska zgradba spremenila. Za območje so značilna tri obdobja izseljevanja prebivalstva; prvo v začetku 20. stoletja (v države Zahodne Evrope in v obe Ameriki), drugo med 2. svetovno vojno in tretje v večja industrijska središča kot rezultat zakasnele industrializacije v 60-tih let. Med in po emigracijskih valovih se je pokrajina zaraščala z gozdom.
Na območju smo izdelali 89 florističnih popisov (Braun-Blanquet 1964), ki so bili med seboj oddaljeni najmanj 100 metrov. Vsi popisi so bili izdelani na območju pašnikov ali pašnikov v različnih fazah zaraščanja. Podatke o času od opustitve kmetijske rabe (TAA) smo pridobili s pomočjo prekrivanja katastrskih načrtov, vojaških kart in letalskih posnetkov iz različnih obdobij (Paušič & čarni 2012b).
Analizirali smo premeno funkcionalnih rastlinskih znakov in ekoloških značilnosti vrst (Tabela 1), ki smo jih razdelili v pet skupin. Nabor izbranih funkcionalnih rastlinskih znakov povzemamo po Weiher et al. (1999). Vsaki od 249 najdenih rastlinskih vrst smo pripisali funkcionalen rastlinski znak in ekološke značilnosti.
V	programu Twinspan (Hill 1979) smo opravili klasifikacijsko analizo popisov. S pomočjo uporabe indeksa navezanosti vrst (Chytry et al. 2002) smo izračunali diagnostične vrste za vsakega izmed pet snopov.
V	naslednjem koraku smo opravili DCA analizo popisov in analizirali položaj popisov v multivari-atnem prostoru. Ugotovili smo, da je postavitev skupin popisov po abscisni osi sorazmerna z njihovo povprečno vrednostjo TAA. Zato smo za vsak
popis izračunano projekcijo vrednost na abscisni osi DCA upoštevali kar kot proxy vrednost TAA. Izračunane vrednosti abscisne osi TAA za vsakega izmed 89 popisov smo v nadaljevanju korelirali s pojavnostjo izbranih rastlinskih funkcionalnih znakov in njihovih ekoloških značilnosti. Izračunali smo Spearmanov koeficient (rho), Tabela 2.
Na koncu nas je zanimala še premena strategije celotne rastlinske združbe in posebej zeliščne plasti za vsak stadij sekundarne sukcesije.
Klasifikacija v programu Twinspan je naše popise razdelila v 5 snopov, ki so sorazmerni TAA vsakega posameznega popisa. TAA (mediana) za vsak snop posebej znaša: A - 0 let, B - 10 let, C - 20 let, D - 45 let, E - 210 let (Slika 2).
Združbo na pašnikih sestavljajo heliofilne vrste in jo imenujemo stadij Calluna. Temu stadiju sekundarne sukcesije sledi stadij Pteridium-Frangula (TAA = 10). Stadij C poimenujemo Betula. Stadij D (TAA = 45) je značilen gozd, kjer je Carpinus be-tulus dominantna drevesna vrsta, v zeliščni plasti pa dominira Epimedium alpinum. Stadij imenujemo Epimedium - Carpinus. Stadij E zaznamujejo številne skiofilne vrste v zeliščni plasti, gozd je značilno dvoplasten, kjer v zgornji plasti prevladuje Quercuespetraea, v spodnji drevesni plasti pa Carpinus betulus (Slika 2, Slika 4, Tabela 2). Ostale korelacije opazovanih funkcionalnih rastlinskih znakov s TAA so predstavljene v Tabeli 2.
Analiza premene ekološke strategije združbe med sekundarno sukcesijo (Slika 5) je pokazala, da ima združba na pašnikih značilno strategijo stres tolerator, po desetih letih kompetitor/ stres tolerator, po dvesto letih pa kompetitor- kompetitor/ stres tolerator. Podobno se spreminja strategija zeliščne plasti skozi stadije sekundarne sukcesije.
Rezultati študije kažejo na spremembo funkcionalnih rastlinskih znakov in ekoloških značilnosti, kakor tudi celotne strategije združbe skozi sekundarno sukcesijo kot posledica opustitve paše.
7. ACKNOWLEDGEMENTS
For technical assistance, we thank Mr Iztok Sajko. We also thank Mr Iztok Vranicar (Surveying and Mapping Authority, Crnomelj) and Mr Martin Cregeen, who kindly corrected our English. The authors acknowledge financial support from the Slovenian Research Agency (project nos. L1-9737 and P1-0236).
8. REFERENCES
Adler, P.B., Milchunas, D. G., Lauenroth, W.K., Sala, O.E., Burke, I.C. 2004: Functional traits of graminoids in semi-arid steppes: a test of grazing histories. Journal of Applied Ecology 41: 653-663.
ARSO, 2011: Agencija Republike Slovenije za okolje. Slovenia, Ljubljana. http://www.arso. gov.si
Ayasse, M., Schiestl, F.P., Paulus, H.F., Lofstedt, C., Hansson, B., Ibarra, F., Francke, W. 2000: Evolution of reproductive strategies in the sexually deceptive orchid Ophrys sphegodes. How does flower-specific variation of odor signals influence reproductive success? Evolution 54: 1995-2006.
Bässler, M., Jäger, E.J., Werner, K. 1996: Exkursionsflora von Deutschland. Bd. 2 Gefässpflanzen. 16. Ed. Fischer, Jena, 640 pp.
Braun-Blanquet, J. 1964: Pflanzesoziologie. Grundzüge der Vegetationskunde. 3. Edition, Springer Verlag, Wien, 865 pp.
Casado, M.A., Castro, I., Ramirez, L., Costa, M., Miguel, J.M., Pineda, F.D. 2004: Herbaceous plant richness and vegetation cover in Mediterranean grasslands and shrublands. Plant Ecology 170: 83-91.
Castro, H., Lehsten, V., Lavorel, S., Freitas, H. 2010: Functional response traits in relation to land use change in Montado. Agriculture, Ecosystems & Environment 137: 183-191.
Catorci, A., Vitanzi, A., Tardella, F. M. 2011: Variations in CSR strategies along stress gradients in the herb layer of submediterranean forests (central Italy). Plant Ecology and Evolution 144: 299-306.
Chittka, L. & Raine, E.N. 2006: Recognition of flowers by pollinators. Current Opinion in Plant Biology 9: 428-425.
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.
Cousins, S.A.O. & Eriksson, O. 2002: The influence of management history and habitat on plant species richness in a rural hemiboreal landscape, Sweden. Landscape Ecology 17: 517-529.
Čarni, A., Košir, P., Marinšek, A., Šilc, U., Zelnik, I. 2007: Changes in structure, floristic composition and chemical soil properties in a succession of birch forests. Periodicum biologorum 109: 13-20.
Čarni, A., Matevski, V., Šilc, U. 2010: Morphological, chorological and ecological plasticity of Cistus incanus in the southern Balkans. Plant Biosystems 144: 602-617.
Čarni, A., Juvan, N., Košir, P., Marinšek, A., Paušič, A., Šilc, U. 2011: Plant communities in gradients. Plant Biosystems 145: 54-64.
Debussche, M., Escarre, J., Lepart, J., Houssard, C., Lavorel, S. 1996: Changes in Mediterranean plant succession: old-fields revisited. Journal of Vegetation Science 7: 519-526.
Ellenberg, H., Weber, H.E., Düll, R., Wirth, V., Werner, W., Paulissen, D. 1992: Zeigerwerte von Pflanzen in Mitteleuropa. Erich Goltze, Göttingen, 262 pp.
Forrest, J. & Thomson, J.D. 2009: Pollinator experience, neophobia and the evolution of flowering time. Proceedings of the Royal Society of Biological Sciences 276: 935-943.
Frank, D. & Klotz, S. 1990: Biologisch-ökologische Daten zur Flora der DDR. Wissenchaftli-che Beiträge der Martin Luther Universität, Halle, 167 pp.
Garnier, E., Shipley, B., Roumet, C., Laurent, G. 2001: A standardized protocol for the determination of specific leaf area and leaf dry matter content. Functional Ecology 15: 688-695.
Garnier, E., Laurent, G., Bellmann, A., Debain, S., Berthelier, P., Ducout, B., Roumet, C., Navas, M.L. 2004: Consistency of species ranking based on functional leaf traits. New Phytolo-gist 152: 69-83.
Grime, J.P., Thompson, K., Hunt, R., Hodgson, J.G., Cornelissen, J.H.C. 1997: Integrated screening validates primary axes of specialisation in plants. Oikos 79: 259-281.
Gumbert, A. & Kunze, J. 2001: Colour similarity to rewarding model plants affects pollination in a food deceptive orchid, Orchis boryi. Biological Journal of the Linnean Society 72: 419-433.
Günther, J. 1987: Blatmerkmale und deren ökologische Aussagekraft in ausgewählten Segetal-gesellschaften. Martin Luther Universität, Halle, 410 pp.
Halpern, C.B. 1989: Early successional patterns of forest species: interactions of life history traits and disturbance. Ecology 70: 704-720.
Heinken, T., Schmidt, M., Von Oheimb, G., Kriebitzsch, W.U., Ellenberg, H. 2006: Soil seed banks near rubbing trees indicate dispersal of plant species into forests by wild boar. Basic and Applied Ecology 7: 31-44.
Hennekens, S.M. & Schaminée, J.H.J. 2001: Tur-bo-Veg: A comprehensive database management systeme for vegetation data. Journal of Vegetation Science 12: 589-591.
Hill, M.O. 1979: TWINSPAN, a fortran program for arranging multivariate data in an ordered two-way table by classification of the individuals and attributes. Cornell University, Ithaca.
Hua, W., Xiu- Qun, L., Hua, J., Long- Qing, C. 2010: Effects of light, macronutrients, and sucrose on germination and development of the endangered fern Adiantum reniforme var. sinense (Adiantaceae). Scientia Horticulturae 125: 417-421.
Hunt, R., Hodgson, J.G., Thompson, K., Bungener, P., Dunnett, N.P., Askew, A.P. 2004: A new practical tool for deriving a functional signature for herbaceus vegetation. Applied Vegetation Science 7: 163-170.
Johansson, V.A. & Cousins, S.A.O. 2010: Remnant populations and plant functional traits in abandoned semi- natural grasslands. Folia Geobotanica 45: 46-49.
Klotz, S., Kühn, I., Durka, W. 2002: BIOLFLOR-Eine Datenbank zu biologisch-ökologischen Merkmalen der Gefäspflanzen in Deutschland. Bundesamt für Naturschutz, Bonn.
Kugler, H. 1970. Blütenökologie. 2. Ed. Gustav Fisher Verlag, Jena, 278 pp.
Latzel, V., Klimešová, J., Doležal, J., Pyšek, P., Tackenberg, O., Prach, K. 2010: The association of dispersal and persistence traits of plants with different stages of succession in Central European man-made habitats. Folia Geobotanica 46: 289-302.
Lengyel, S., Aaron, D.G., Andrew, M.L., Jonathan, D.M, Robert, R.D. 2010: Convergent evolution of seed dispersal by ants, and phy-logeny and biogeography in flowering plants: a global survey. Perspectives in Plant Ecology, Evolution and Systematics 12: 43-55.
Lepš, J. & Štursa, J. 1989. Species-area curve, life history strategies, and succession: a field test of relationships. Vegetatio 83: 249-257.
Meusel, H. & Jäger, E. 1992: Vergleichende Chorologie der zentraleuropäischen Flora. Fischer Verlag, Jena. 421 pp.
Miller, R., Owens, S.J., Rorslett, B. 2011: Plants and colour: Flowers and pollination. Optics and Laser Technology 43: 282-294.
Mony, C., Mercier, E., Bonis, A., Bouzille, J.B. 2010: Reproductive strategies may explain plant tolerance to inundation: A mesocosm
experiment using six marsh species. Aquatic Botany 92: 99-104.
Müller, H. 1881: Alpenblumen, ihre Befruchtung durch insekten und ihre Anpassungen an dieselben. Leipzig, 611 pp.
Odum, E.P. 1980: Grundlagen der Ökologie. 1. Thieme, Stuttgart, 370 pp.
Orozen-Adamic, M., Perko, D., Kladnik, D. 1995: Krajevni leksikon Slovenije. DZS, Ljubljana, 638 pp.
Pausic, A., Carni, A. 2012: Records of past land use are best stored in soil properties. Plant Biosystems. DOI: 10.1080/11263504.2012.748100
Pausic, A., Carni, A. 2012 b: Landscape transformation in the low karst plain of Bela krajina (SE Slovenia) over the last 220 years. Acta geographica Slovenica 52 (1): 35-60.
Pärtel, M. & Zobel, M. 1999: Small-scale plant species richness in calcareous grasslands determined by the species pool, community age and shoot density. Ecography 22: 153-159.
Peco, B., Pablos, D.I., Traba, J., Levassor, C. 2005: The effect of grazing abandonment on species composition and functional traits: the case of dehesa grasslands. Basic and Applied Ecology 6: 175-183.
Pellissier, L., Vittoz, P., Internicola, A.I., Gigord, L.D.B. 2010: Generalized food-deceptive orchid species flower earlier and occur at lower altitudes than rewarding ones. Journal of Plant Ecology 3: 243-250.
Prevosto, B., Kuiters, L., Bernhardt-Römermann, M., Dölle, M., Schmidt, W., Hoffmann, M., Van Uytvanck, J., Bohner, A., Kreiner, D., Stadler, J., Klotz, S., Brandl, R. 2011: Impacts of land abandonment on vegetation: succes-sional pathways in European habitats. Folia Geobotanica 46: 303-325.
Raunkiaer, C. 1934: Life forms of plants and statistical plant geography. Oxford University Press, Oxford, 410 pp.
Rehunkova, K. & Prach, K. 2010: Life-history traits and habitat preferences of colonizing plant species in long- term spontaneous succession in abandoned gravel-sand pits. Basic and Applied Ecology 11: 45-53.
Saatkamp, A., Römermann, C, Dutoit, T. 2010: Plant functional traits show non-linear response to grazing. Folia Geobotanica 45: 239-252.
Statsoft Inc. 2007: Statistica (data analysis software system), version 8.0. www.statsoft.com.
Strassburger, E., Sitte, P., Ziegler, H., Ehrendorfer, F., Bresinsky, A. 1998: Lehrbuch der Bota-
nik für Hochschulen. 34. Ed. Fischer Verlag, New York, 1024 pp.
Sternberg, M., Gutman, M., Perevolotsky, A., Ungar, E.D., Kigel, J. 2000: Vegetation response to grazing management in a Mediterranean herbaceous community: a functional group approach. Journal of Applied Ecology 37: 224-237.
Tackenberg, O., Römermann, C., Thompson, K., Poschlod, P. 2006: What does diaspore morphology tell us about external animal dispersal? Evidence from standardized experiments measuring seed retention on animal coats. Basic and Applied Ecology 7: 45-58.
ter Braak, J.F.C. & Smilauer, P. 2002: CANOCO Reference manual and CanoDraw for Windows, User's guide to Canoco for Windows: Software for canonical community ordination (version 4.5). Microcomputer Power. Ithaca.
Tichy, L. 2002: JUICE, software for vegetation classification. Journal of Vegetation Science 13: 451-453.
Tutin, T.G., Heywood, V.H., Burges, N.A., Moore, D.M., Valentine, D.H., Walters, S.M., Webb, D.A. 1964-1993: Flora Europaea. Vol 1-5. Cambridge University Press, Cambridge.
Vesk, P.A. & Westoby, M. 2001: Predicting plant species responses to grazing. Journal of Applied Ecology 38: 897-909.
Vitasovic Kosic, I., Tardella, F.M., Ruscic, M., Ca-torci, A. 2011: Assessment of floristic diversity, functional composition and management strategy of North Adriatic pastoral landscape (Croatia). Polish Journal of Ecology 59: 765-776.
Vittoz, P., Dussex, N., Wassef, J., Guisan, A. 2009: Diaspore traits discriminate good from weak colonisers on high-elevation summits. Basic and Applied Ecology 10: 508-515.
Weiher, E., Van der Werf, A., Thompson, K., Roderick, M., Garnier, E., Erikson, O. 1999: Challenging Theophrastus: a common core list of plant functional traits for functional ecology. Journal of Vegetation Science 10: 609-620.
Received 1. 8. 2012 Revision received 3. 9. 2012 Accepted 10. 9. 2012