surrounded by a non-woodland matrix. For some
species groups, possession of a large expanse of
such edge habitat can increase the overall species
number found on a reserve, an apparently benefi-
cial effect (Kellman 1996). But, the argument goes, if
the additional species supported by the edge habi-
tat are those of the matrix, then they do not depend
on the reserve and should be discounted in evalu-
ating the merits of the reserve system.
In illustration, Humphreys and Kitchener (1982)
classified vertebrate species into those dependent
on reserves and those which persist in disturbed
areas in the intervening agricultural landscape of
the Western Australian wheat belt (Fig. 10.8). They
found that the small habitat islands, which were
mostly edge habitat, were disproportionately rich
in ubiquitous species. Those species restricted to
woodlots typically required a larger area (cf.
Hinsleyet al. 1994). This pattern was found sepa-
rately for lizards, passerine birds, and mammals.
By analysing all species within a taxon together,
small reserves came out in a better light than they
warranted. Similar results are reported by Watson
et al. (2004a,b) in a study of birds in littoral forests
and surrounding habitats in south-eastern Madag-
ascar. Core forest locations were found to be richer
than edge or matrix habitats, with some 68% of the
forest dependent species found to be edge-sensitive.
Frugivorous species and canopy insectivores
(including six endemic vanga species) were gener-
ally edge-sensitive, while in contrast, sallying
insectivores were edge-preferring. At least part of
the edge-sensitivity recorded was attributable to
changes in vegetation structure at the remnant edges.
Consistent with this edge sensitivity, forest
dependent species were generally lacking from
fragments of less than 10 ha, thereby demonstrat-
ing a small-island effect (Chapter 4), masked in
analyses of all bird species richness (Fig. 10.9).
Hansson (1998) used a rather different approach in
his studies of edge effects in ancient woodlands in
Sweden, but again found differences in bird com-
munity composition that appeared to relate to habi-
tat structure at the edges. He analysed the degree of
nestedness, finding that whereas census data for
birds from the centres of the patches were nested,
the samples from the edge habitat were non-nested.
In particular circumstances, ecotones may have
negative implications for core woodland species
(Wilcove et al. 1986), in that although they may sup-
port additional species or larger populations of
non-woodland core species, these populations may
have negative interactions with deep-woodland
species. Studies in North America have suggested
that the nesting success of songbirds is lower nearEDGE EFFECTS 277Passerine
birds01.0
0.8
0.6
0.4
0.20 500 1000 1500 2000Edge speciesEdge speciesEdge speciesCore speciesCore speciesEdge-core speciesEdge-core speciesEdge-core species01.0
0.8
0.6
0.4
0.201.0
0.8
0.6
0.4
0.20 500 1000 1500 20000 500 1000 1500 2000MammalsLizardsProportionProportionProportion(a)(b)(c)Area (ha)Co
respeciesFigure 10.8The proportion of the core species (those dependent
on the reserves) and the edge species (those which can also survive
outside the reserves in disturbed habitat) which were present in a
series of 21 nature reserves in the Western Australian wheatbelt in
relation to the number of each present in a reserve of 2000 ha. The
curves with negative slopes represent the difference between the
curves for edge and core species, and represent the relative excess of
edge species over core species compared with their distribution at
2000 ha. Data for (a) mammals, (b) passerine birds, and (c) lizards.
(Redrawn from Humphreys and Kitchener 1982, Fig. 1.)