S. Mahunka szerk.: Folia Entomologica Hungarica 56. (Budapest, 1995)

Fig. 2. Dominance curves of the grasshopper communities on the three different sites. The shape of the rank dominance curves on the grazed sites is similar to each other Other similarity indices also confirmed the stronger similarity of the communities of the PL and the AL sites (Table 3). One-way analysis of variance at the p = 0.05 probability level does not show significant differences between the groups of the three communities (Table 4). In spite of this the Chi-square test showed some significant differences between the three communities (Table 5). DISCUSSION Sheep grazing had little effect on the grasshopper communities. One of the probable explanation is that the grazing was not heavy enough to basically change the structure of the vegetation. Fielding & Brusven (1993) found changes in the fundamental community structure - e. g. density and species composition - when grazing of the livestock was intensive. Neither grazing nor vegetation structure exerted strong effect on the grasshopper communities in any site in early summer (July). Their structures were very similar to each other. In this period, when most of the individuals begin to transform into imagoes, Table 3. Renkonen, Jaccard, S0rensen and Bray-Curtis similarity coefficients of the three grasshopper communities. 1= alfalfa site, 2= pasture-land, 3= meadow Similarities 1,2 1,3 2,3 Renkonen 43.3 28.8 39.5 Jaccard 0.5396 0.3076 0.3076 S0rensen 0.7 0.4705 0.4705 Bray-Curtis 0.1406 0.1199 0.2214