The Brazilian Atlantic Forest is one of the world's ‘hot spots’ for conservation because of its high level of endemism and number of endangered species (Myers et al. 2000). After centuries of deforestation, most of the remaining Atlantic Forest is scattered as small fragments on private land (Fonseca 1985, Jorge & Garcia 1997). Pollination could be impaired in these fragments because isolation often negatively influences the dispersal of insects, the most important pollinating agents in the tropics.

Euglossine bees (Hymenoptera: Apidae: Euglossini) are considered keystone species in the neotropics because of their role as pollinators of several plant species, particularly orchids (Dodson et al.1969, Roubik 1992). Pollination by male euglossine bees occurs when they visit flowers to collect fragrances, which may be used for courtship (Eltz et al. 1999) or attraction of other males and females (Peruquetti 2000). Synthetic products that mimic those fragrances have been used frequently in studies of euglossine bee ecology and population structure (Armbruster & McCormick 1990, Powell & Powell 1987, Roubik & Ackerman 1987). The ability of euglossine bees to disperse and find isolated flowers and distant baits (Dressler 1968, Janzen 1971) has led Janzen (1981) and Janzen et al. (1982) to suggest that bees attracted to fragrances come from a wide area that may include different habitats. According to this hypothesis, individuals collected at baiting stations are part of the same pool of bees. However, Armbruster (1993) found significant variation in the number of bees collected at nearby baiting stations, and he considered these differences as demonstration of within-habitat heterogeneity of the euglossine bee community. In Armbruster's model, results of bait collections are strongly affected by the concentration of resources in ‘hot spots’, therefore a sampling station would not necessarily represent the habitat, but only particular microhabitats.