The loss of avian understory insectivores could potentially itself instigate another trophic cascade, resulting in herbivorous arthropod release with the potential for further cascading effects on the plant community. Accelerating global declines in avian diversity and abundance threaten ecosystem consequences, due to loss of both biodiversity and functionally important species, yet this potential has been little studied (Sekercioglu et al. 2002, 2004; Sekercioglu 2006a, b). Insectivores, including insectivorous birds, impact the evolution, abundance, and behavior of their arthropod prey (Holmes 1990). Exclosure experiments show that insectivores in natural and managed tropical habitats effectively control arthropod herbivores at sub-outbreak levels (Gradwohl and Greenberg 1982, Greenberg et al. 2000, Van Bael et al. 2003, Boege and Marquis 2006, Van Bael et al. 2007), and may limit the extent of herbivore outbreaks in tropical forest (Van Bael et al. 2004).

The biggest failure of many previous exclosure experiments was to attribute all insectivory to birds, when bats could contribute greatly, particularly in tropical forests. Foliage-gleaning bats consume arthropods in similar size-ranges (~4-30mm), from similar taxa as birds (LaVal and LaVal 1980, Sherry 1984), and in large quantities: gleaning bats consume over half of their body mass in arthropods nightly (Kalka and Kalko 2006). Two recent studies have distinguished bird and bat predation: Mexican shade coffee bird predation exceeded that of bats, with non-additive bird+bat effects (Williams-Guillén and I. Perfecto 2007), while bat exceeded bird predation by 50% on Barro Colorado Island (M. Kalka, unpubl. data). The higher bat predation on Barro Colorado Island may be due to diminished avian insectivore function due to declines experienced by this guild (Robinson 1999), but may also indicate bat compensation in the absence of avian competitors (Kremen 2005).

Additionally, whereas understory insectivorous birds are highly sensitive to fragmentation and disturbance, experiencing widespread declines and extirpations (Sekercioglu et al. 2002, Sodhi et al. 2004, Lindell 2007), insectivorous bats in Trinidad were not adversely impacted by logging (Clarke et al. 2005). At La Selva, where insectivorous birds accounted for 50% of all avian declines (Sigel et al. 2006), the bat community structure has remained virtually unchanged, with 45 insectivorous bat species remaining abundant (Heller and Volloth 1995) and slight, but non-significant declines in abundance of some species since the 1970s (Sanchez 2007). Further, La Selva supports 19 species of Phyllostomid bats (Timm 1994), a family considered highly sensitive to disturbance among bats (Medellín et al. 2000). Their persistence at La Selva suggests they are less disturbance-sensitive than their complementary diurnal avian guild, or at least less impacted by humans to date. Thus, bats' role in controlling herbivorous arthropods may be of even greater importance than birds' in fragmented forests. Effective land-use and management plans aimed at preventing or mitigating community-level cascading effects of increased herbivory necessitate understanding the impacts and interactions of insectivorous birds with other insectivores such as bats, and the potential for compensation by one or both groups (Kremen 2005).

The second component involves two experimental studies designed to investigate the potential for cascading community consequences following avian insectivore declines. The first experiment, conducted between June 2007 - April 2008, excluded all aerial vertebrate insectivores to assess the potential for "top-down" (i.e., predation) and "bottom-up" (i.e., vegetative) interactions in arthropod regulation. This study found evidence of arthropod control by insectivores in the rainforest understory, as well as "bottom up" effects of vegetation and leaf-litter density on arthropod densities, particularly spiders.

The second experiment, to be conducted between May-October 2008, will distinguish bird from bat predation impacts, and investigate the potential existence and relative effects of functional compensation by bats in absence of avian insectivores at La Selva Biological Station, a large rainforest fragment with a depauperate avian insectivore community, compared to a control site (Refugio Bartola, an unfragmented site with an intact avian insectivore community. In this study I will investigate the relative functional importance of avian versus chiropteran insectivores in limiting abundances of herbivorous arthropods and their damage to plants in the rainforest understory.