Negative indirect effects of an avian insectivore on the fruit set of an insect-pollinated herb T. D. Meehan, H. M. Lease and B. O. Wolf Meehan, T. D., Lease, H. M. and Wolf, B. O. 2005. Negative indirect effects of an avian insectivore on the fruit set of an insect-pollinated herb.  / Oikos 109: 297  /304. Though an abundance of research has focused on direct interactions between birds and plants, relatively few studies have reported on indirect interactions. Of those reports, all have focused on positive indirect effects of birds on plants through predation of plant natural enemies. We conducted an observational study along the Middle Rio Grande in New Mexico to determine if avian aerial insectivores had a negative, indirect impact on insect-pollinated plants through predation of insect pollinators. We found considerable taxonomic overlap, at the order and family level, between insects visiting sweet clover (Melilotus officinalis ) and those eaten by cliff swallows (Hirundo pyrrhonota ). We found a significant negative relationship between proximity of sweet clover to cliff swallow breeding colonies and sweet clover fruit set during the cliff swallow nestling period. The apparent effect of cliff swallows was strongest within 200 m of breeding colonies (approximately 50% reduction in fruit set) and decreased nonlinearly to a distance of approximately 400 m. Finally, we found that the clover fruit set gradient disappeared after the nestling period, when chicks had fledged and the colony was abandoned. T. D. Meehan, H. M. Lease and B. O. Wolf, Dept of Biology, Univ. of New Mexico, Albuquerque, NM 87131, USA. Present address for TDM, Dept of Sciences and Conservation Studies, College of Santa Fe, Santa Fe, NM, 87505 USA (tmeehan@csf.edu). Direct interactions between birds and plants have been well studied. Birds are known to play important roles as pollinators (Regal 1982, Jordano 1987), seed dispersers (Temple 1977, Hutchins and Lanner 1982, Wheelright 1988, Wenny and Levey 1998), seed predators (Castro et al. 1999, Howe and Brown 1999), and herbivores (Clout and Hay 1987, James and Burney 1997, Perrow et al. 1997). These direct interactions have been shown to affect plant population dynamics and community struc- ture, as well as the evolution and expression of plant life histories and morphologies (Brown and Kodric-Brown 1979, Paige and Whitham 1985, 1987, Murray et al. 1994, Bruneau 1997). Economic ornithologists of the late 19th and early 20th centuries suggested indirect interactions between birds and plants (Judd 1901, Beal 1904, 1907, McAtee 1911), though it is only recently that these relationships have been quantified. It has since been shown that bird predation of invertebrate herbivores can lead to de- creased herbivory and increased growth in desert grasses (Bock et al. 1992), deciduous trees (Marquis and Whelan 1994, Sipura 1999, Murakami and Nakano 2000, Strong et al. 2000, Sanz 2001, Lichtenberg and Lichtenberg 2002), and intertidal algae (Wootton 1995). These studies have illustrated how birds have a positive indirect effect on plants through predation of plant natural enemies. We have not encountered studies that have quantified the indirect effects of birds on plants through predation of plant mutualists. In particular, we are aware of no studies that have quantified the negative indirect effects of birds on plants through the predation of insect pollinators. In this study, we examine the hypothesis that avian aerial insectivores, through predation of aerial insect Accepted 21 October 2004 Copyright # OIKOS 2005 ISSN 0030-1299 OIKOS 109: 297 /304, 2005 OIKOS 109:2 (2005) 297