B. Papp szerk.: Studia Botanica Hungarica 39. 2008 (Budapest, 2008)

Dobolyi, K., Erős-Honti, J.; Botta-Dukát, Z.: Habitat preference of Linum dolomiticum (Linaceae)

methods when used on noisy data sets (LAMBERT and WILLIAMS 1966, GAUCH 1982). Ward's algorithm of minimum increment of sum of squares was used for dendrogram construction (WARD 1963, LEGENDRE and LEGENDRE 1998). Both ordination and classification were performed by SYN-TAX 2000 package (PODANI 2001). The number of interpreted groups was selected based on proportion of clusters with more than 4 character species (i.e. species withp value lower than 10~

9 in the Fisher exact test; TiCHY et al. 2008). Thus five clusters were separated and interpreted. The relevé groups were characterised by faithful (u

hyp > 3) (CHYTRY et al. 2002), constant (frequency higher than 60%) and dominant (cover higher than 15% at least 25% of the relevés belonging the group) species and analysed on the basis of the following species characteristics: phytogeographical area (floral element) (HORVÁTH et al. 1995), phytosociological character, social behaviour type, water and continentality preference categories (BORHIDI 1993). The values of features characterising the groups of relevés were calculated according to percentage pro rata of plant mass, that is, the sum of the cover values concerning one feature has been expressed in percentage of sum of all cover values of all species. Testing the association between Linum dolomiticum and other species the u

hyp statistics was calculated (CHYTRY et al. 2002). Regarding the high number of pair-wise comparisons, the critical value was set up to 3, instead of the generally used 2 (that correspond the 5% probability of Type I error). The lists of faithful, constant and dominant species were drawn up and the associations between species were calculated by JUICE programme (TiCHY 2002). RESULTS The study of species groups has resulted in five groups of separated relevés. Group I Number of relevés: 50. Characteristics of the group: - typical exposure: NW, W; - total cover of vegetation: 90-100%; - vegetation type: slightly degraded closed dolomite grassland; Bromus pannonicus is absent; cover of Quercuspubescens and Fraxinus ornus is 5-15%; - diagnostic species: Anthyllis vulneraria, Fraxinus ornus, Genista pilosa, Linum dolomiticum, Phyteuma orbiculare and Quercus pubescens; - constant species: Anthericum ramosum, Anthyllis vulneraria, Carex humilis, Festuca pallens, Fraxinus ornus, Genista pilosa, Globularia punctata, Helianthemum canum, Jurinea mollis, Linum dolomiticum, Phyteuma orbiculare, Quercus pubescens, Teucrium montanum;