Effects of forest fragmentation, anthropogenic edges and fruit colour on the consumption of ornithocoric fruits

Abstract

We investigated effects of fruit colour (red, black or white), habitat (anthropogenic edges and forest interior) and fragment size on the removal of artificial fruits in semideciduous forests in south-east Brazil. Eight forest fragments ranging from 251 to 36,000 ha were used. We used artificial fruits, which were placed on shrubs between 1 and 2 m in height and checked after 48 and 96 h for peck marks in the fruits. All three variables affected the probability of consumption of our fruit models. Red and black fruits were statistically more pecked than the white fruits. The probability of fruit consumption was lower in the interior than at the edge and less in small than in large fragments. However, the decrease fruit consumption in small compared with large fragments was more accentuated for red and black fruits than for white fruits. Our results show that habitat reduction and edges affect the chances of a fruit being eaten by birds, which may ultimately affect plant fitness in forest fragments.

Introduction

Since the conversion rate of tropical forests into disturbed habitats exceeds 1.2% annually, studies on the effects of forest fragmentation on the ecology of plants and animals are especially relevant to management of isolated forest remnants (Laurance and Bierregaard, 1997). One of the major changes brought by habitat fragmentation is an increase in the proportion of the edge (Kapos et al., 1997). Edge effects include decreasing humidity and increasing light and temperature (Murcia, 1995, Kapos et al., 1997) which may ultimately affect plant recruitment and survival (Benetiz-Malvido et al., 1999). Another important issue related to fragmentation is potential disruption of key mutualistic interactions between animals and plants, including pollination and seed dispersal (Aizen and Feinsinger, 1994, Santos and Telleria, 1994).

Forest fragmentation and the resulting anthropogenic edges may alter the abundance of fruiting plants and frugivorous birds (Stiles, 1980, Restrepo and Gomez, 1998, Restrepo et al., 1999). However, very few studies have evaluated how anthropogenic edges and habitat fragmentation affect the mutualistic process of frugivory and seed dispersal (Restrepo and Gomez, 1998, Restrepo et al., 1999).

A large proportion of vascular plants in Neotropical rain forests is dispersed by birds or mammals (Foster et al., 1986, Howe and Westley, 1988, Morellato and Leitao-Filho, 1992). These include canopy tree species with large seeds dispersed by specialized large frugivores (such as toucans, guans, and large cotingas) (Howe and Smallwood, 1982) which are prone to become vulnerable due to dispersal failure in fragmented forests (Silva and Tabarelli, 2000, Galetti, 2001). Understorey small fruits, on the other hand, are consumed by a wide variety of generalist birds, including manakins, tanagers and flycatchers (Wheelwright et al., 1984, Loiselle and Blake, 1999), which are not as highly affected by fragmentation (Willis, 1979, Aleixo and Vielliard, 1995).

A key challenge to understand the relationship between fruit and frugivore is to relate fruit morphology and accessibility to removal rate by birds (Levey et al., 1984, Moermond et al., 1986, Murray et al., 1993). Most bird-dispersed understorey fruits are easily accessible and coloured black (or dark purple), red or sometimes white (Wheelwright and Janson, 1985), such as Melastomataceae and Rubiaceae's fruits. This plant guild, although important to maintain at least small frugivore birds (Gilbert, 1980, Galetti and Stotz, 1996, Silva et al., 2002), has not considered threatened by habitat modification.

In this paper we examined how forest size, anthropogenic edge, and fruit colour influence the probability of understorey fruit consumption by birds and if each one these factors can influence the effect of the other.