Matskási István (szerk.): A Magyar Természettudományi Múzeum évkönyve 89. (Budapest 1997)

Szollát, Gy., Stollmayer-Boncz, E. , Standovár, T. ; Kecskés, F.: Vegetation of the Réti-dűlő and Csömöri-tó (Hungary)

Table 5) at the margins. The western shore is bordered by willow scrub, whereas former gallery wood (Salicetum albae-fragilis) remnants adjoin other parts of the lake shore. PHYTOSOCIOLOGICAL SURVEY MATERIALS AND METHODS Phytosociological survey was carried out by using 2 x 2 m quadrats and the traditional BRAUN-BLANQUET-SOÓ method (ZÓLYOMI 1951). Fig. 4 shows the locations of the quadrats. In this paper only relevés from the Csömöri-tó and two communities (Succiso-Molinietum, Caricetum appropinquatae) of the Réti-dűlő are discussed. Unlike those of other communities, these samples are appropriate to characterize the respective vegetation. Relevés of wet and mesic meadows and of reed swamp are either too few in number or too heterogeneous for proper interpretation. For analysing Succiso-Molinietum (16 relevés) numerical classification was also used. Several hierarchical agglomerative clustering methods were tried. Here - as an example - only the results obtained by weighted averaging using Euclidean distance (JONGMAN et al. 1987). For calculating the importance of nature conservation ranks, they were weighted by species abundance. RESULTS AND DISCUSSION Basic information of all the 25 relevés are summarized in Table 6. The 16 relevés describing Succiso-Molinietum are shown in Table 7. According to the results of several numerical classifications (an example is shown in Fig. 1), two consistent groups can be distinguished. There were 4 relevés (No. 9,10 13, 22) that behaved inconsistently (either separated from the otherwise consistent two clusters, or grouped with one of them) in different analyses. The separation of the two quadrat groups can be explained by two main differences. The first group (No. 4, 8, 16, 21, 26 ) is characterized by less species (13-16) than the other group (No. 7, 14, 15, 18, 19, 23, 27) (21-28 species) ). A significantly higher dominance of Molinia hungarica is also characteristic of the first group. According to the comparision with classical phytosociological tables (KOVÁCS 1962) the first group resembles Succiso-Molinietum molinietosum, whereas the second group is similar to Succiso-Molinietum caricetosumpaniceae. It is worthy of noting, that these two types are not separated spatially (cf. Fig. 4). It is also interesting that these numerical analyses were insensitive to the otherwise sensible impression of the field botanist. More damp parts with many protected species (cf. Fig. 2) are not necessarily grouped together. On the other hand this conspicuous separation of Molinia-nch and Molinia-poor relevés indicates, that the applied quadrat size coincide with the spatial scale of a characteristic pattern of this community. Table 4 shows the phytosociological data of sedge tuft (Caricetum appropinquatae) association. The comparison with classical data (KOVÁCS 1957) indicates that some valuable fen species (e.g. Menyanthes trifoliata, Eriophorum latifolium) are missing from these fragments.