L. Hably szerk.: Studia Botanica Hungarica 20. 1988 (Budapest, 1988)
Szerdahelyi, Tibor: Vegetation studies on rocky grasslands in the Pilis Mountains Hungary
RESULTS The phytogeographical spectrum shows that the number of cosmopolitan and adventive species is almost the same in both sites. There is no essential difference between the floral element groups. From the point of view of cenological character site I is considerably the poorer of the two sites in Festuco-Brometea and Quercetea elements. Instead of these species indifferent elements and the members of weed associations appeared (e. g. Eryngium campestre , Sideritis montana , Chondrilla juncea , Fumaria schleicheri ). The life-form distribution unambiguously shows the dominance of therophyton and hemitherophyton species in site I. It can be seen in the disturbed grassland communities the proportion of geophyton, chamaephyton and hemikryptophyton species decreased and these spots are occupied by therophytons and hemitherophytons of great reproductive capacities. These species abundances are probably proportionate to the degree of degradation. I also made attempts to examine other parameters as well. It can be seen in Fig. 2 that the R Q values are high, indicating that several species have no special soil requirements. They occur in several grassland communities. There is no notable difference between the sites in this respect. The high Rg and R4 values indicate the great number of basophilous species. I could distinguish between the two sites from the point of view of temperature requirement. The number of species in common is low. This is shown by the small absolute value of TQJJ. The large number of species of site II belongs to the groups of the normal and thermophilous species. The value T Q . indicates that many more different species occur in site I. (T 0 j^. TQJJ ). The most characteristic figure of all is the distribution of water requirements. Water is the most important parameter in xerophilous communities like the ones studied, water limits the life processes, the species' range, etc. So the indifferent species are very few in both sites. Comparing the two groups from the point of view of nitrogen requirement, they are obviously different from each other (N Q ^^N 0 j ). There are less indifferent species in the "Öreg-szirt" site. Species of nitrophilous character occurred exclusively in site I, this character does not appear in site II. Summarizing the differences of species composition of the two areas, therophytons and hemitherophytons partly take the place of the kryptophytons and hemikryptophytons because of erosion, disturbed grass surface (treading of the grass), and the overcrowded stock of game. Generally the species composition becomes poor and patches consisting of few species appear in the communities ( Bromus squarrosus , Agropyron repens ). Some are observed in the Pilis Mountains where the degree of degradation is so high that the plant cover consisted of only therophytons of short life-cycle. In site I I observed the expansion of Agropyron re- pens , A. intermedium , Arenaria serpyllifolia , Bromus squarrosus and Centaurea micran- thos. According to microclimate measurements (HORÁNSZKY 1964), the appearance of the above-mentioned grass species changes the temperature conditions above the soil surface at 30 cm height. So in the grass patches the originally observed extreme temperature fluctuations decrease. The changed microclimate provides an opportunity for the establishment and expansion of other therophyton species which avoid temperature fluctuations. So in some degraded patches (near site I) the succulent species completely disappeared (e.g. Sempervi- vum hirtum , S. marmoreum , Sedum spp . ). In Fig. 3 we can see the dendrogram of a cluster analysis of a matrix of CZEKANOWSKI's similarity index. Eight groups are obtained at 0. 3 similarity value. The first four clustered groups form the sample plots (24 quadrates) of the degraded area (Klastrom-szirtek). The other four groups comprise the sample plots (2 6 quadrates) of the control area (Öreg-szirt). Only one quadrate from site I was allocated to site II. The dendrogram clearly shows the separation between two areas. Acknowledgements are due to Dr. M. RAJCZY for his advice on suitable methods for the computer analysis.
