L. Forró szerk.: Miscellanea Zoologica Hungarica 13. 2000 (Budapest, 2000)

Horváth, Gy.; Pintér, V.: Small mammal fauna of two abandoned field habitats, and a spatio-temporal analysis of four rodent populations

be attributed to the generally very small amount of capture data. Among the four species it is M. arvalis in which the value of patchiness sinks the most, yet remaining in the range above 1.0, implying aggregation. Correlation between number of individuals and patchiness was tested for only in the five-month period in 1998 (Fig. 17). There was no signiDensity/ha Density/ha Fig. 17. Patchiness as a function of density of the two populations in habitat B in 1998 ficant correlation between these variables in the case of M. arvalis and A. flavicollis. Presumably this was strongly associated with the low capture rates. In A. agrárius a negative non-linear correlation was significant between the two parameters for this year. In A. sylvaticus a negative linear correlation, though not significant, characterised the relationship between the aggregation index and population size. The lack of significance is probably due to the shorter sampling period than in the other plot. Nevertheless, the regression analysis in this species revealed identical correlations in the two habitats. Discussion The small mammal faunas of the two studied open habitats appeared to be similar, which is primarily caused by the commonness of the dominant species. There was no considerable difference between the two habitats in respect of their diversities. However, it must be noted that diversity calculations and comparisons were performed only for a selected period, as in plot B there were less months of trapping. In the two areas differing in their successional stages the appearance and disappearance of certain individuals could be recorded already during our sampling period. In the five-year old area turning into a marsh, Neomys species preferring wet habitats appeared beside the Crocidura species, whereas in the three-year old abandoned land even the Crocidura species became more frequent only in the second study year. As the area was gradually loosing its ploughland character, burrows of M. spicilegus kept disappearing to an extent that in the second year very few specimens were captured. The closing and thickening of the vegetation, a phenomenon typical of both habitats, favoured A. agrárius. There are several hints in the literature that this species prefers areas close to water bodies or periodically inundated lands (Zejda 1967, Babinska-Werka et al. 1979, Chelkowska et al. 1985), which we can confirm with our findings. Palotás (1986), too, evaluated his sample areas by calculating Shannon-Wiener diversity, and similarly to us found higher diversity in areas periodically under water (e.g. backwater forests, flood embankments). Pelikán & Nesvadbova (1979) compared the communities of cultivated lands with those of sur-