Results of the published studies regarding the relationship between the area of habitat island and the number of observed animal species were rather inconsistent. Several authors, in accordance with the prediction of the classical theory of island biogeography, reported a significant positive correlation (Faeth and Kane, 1978; Mader, 1980; Nilsson et al., 1988; McCoy and Mushinsky, 1994; Bolger et al., 1997; Abensperg-Traun and Smith, 1999; Peintinger et al., 2003; Hovestadt et al., 2005; Watson et al., 2005). Others, contrary to the prediction, described a significant negative relationship (Bauer, 1989; Estades and Temple, 1999; Magura et al., 2001; Lovei et al., 2006). Furthermore, some studies found that overall animal species richness was unrelated to the habitat area (As, 1993; Hopkins and Webb, 1984; Davies and Margules, 1998; Brose, 2003; Gandhi et al., 2004; Juliao et al., 2004).
The above inconsistency could arise from the fact that the original theory of island biogeography considered real islands. There is a significant difference between real and habitat islands. In real islands, the surrounding habitat (ocean, sea, lake, river etc.) is usually inhospitable to organisms occurring on islands. In the case of habitat islands, the bordering habitat (the matrix) is usually less hostile. Consequently, species richness of real islands is not notably influenced by the surrounding habitat, while habitat islands could be inhabited by colonists from the matrix: "species can colonize the islands from the sea" (Cook et al., 2002). The above difference is increasingly emphasized when studying the predictions of island biogeography theory on habitat islands. Clear distinction should be drawn between those species that occur in both the habitat patch and the matrix (e.g. generalist species) and the specialist species that truly perceive the habitat patches as islands (Bauer, 1989; Magura et al., 2001, 2008; Cook et al., 2002; Lovei et al., 2006). The specialist species are unable to survive in the surrounding matrix.
In this case study, we demonstrated that depending on the ratio of habitat specialist and generalist species in an assemblage, the species-area relationship may be positive or negative. Based on the significant negative correlation between the total number of ground beetle species and the grassland area one can easily draw the (seriously false) conclusion that it is sufficient to conserve the small patches because they support most species. In fact, however, this was due to the increasing ratio of generalist species with decreasing patch size. Previous studies on ground beetles also emphasized that generalist species from the neighboring matrix and from the edge may cause increased overall species richness in habitat patches with limited size and/or high degree of isolation (Bauer, 1989; Halme and Niemela, 1993; Desender et al., 1999; Magura et al., 2001; Lovei et al., 2006; Magura and Kodobocz, 2007). Removing the non-habitat specialist species from the assemblages and analyzing the importance of only habitat specialist species (open-habitat species associated with sandy soils), the significant negative relationship turned over and became significant positive as predicted by the theory of island biogeography. This duality in the species-area relationship concerning ground beetles is not a special case, as several studies reported similar results in different habitats. Bauer (1989) also found that the relationship between the size of limestone outcrop and the overall ground beetle species richness was significantly negative, while the number of limestone specialist species and the area of outcrop showed a significant positive relation. Species richness of farm woodland ground beetles also correlated positively with the area of woodland, while the total number of species did not (Usher et al., 1993). Similar patterns exist for conifer forest patches, where only the forest specialist species showed a positive species-area relationship (Halme and Niemela, 1993). De Vries (1994) also reported that the relationship between the area of heath fragments and the species richness of heath specialist ground beetles was significantly positive, while the area-total species correlation was statistically not significant. The species-area relationship in deciduous forest patches was also significantly negative when all species were considered, while it was significantly positive for the forest specialist species (Magura et al., 2001; Lovei et al., 2006).
Was this article helpful?