The influence of distinct pollinators on female and male reproductive success in the Rocky Mountain columbine JOHANNE BRUNET and KARSTEN G. A. HOLMQUIST USDA-ARS, VCRU, Department of Horticulture, 1575 Linden Drive, University of Wisconsin, Madison, WI 53706, USA Abstract Although there are many reasons to expect distinct pollinator types to differentially affect a plant’s reproductive success, few studies have directly examined this question. Here, we contrast the impact of two kinds of pollinators on reproductive success via male and female functions in the Rocky Mountain columbine, Aquilegia coerulea. We set up pollinator exclusion treatments in each of three patches where Aquilegia plants were visited by either day pollinators (majority bumble bees), by evening pollinators (hawkmoths), or by both (control). Day pollinators collected pollen and groomed, whereas evening pollinators collected nectar but did not groom. Maternal parents, potential fathers and progeny arrays were genotyped at five microsatellite loci. We estimated female outcrossing rate and counted seeds to measure female reproductive success and used paternity analysis to determine male reproductive success. Our results document that bumble bees frequently moved pollen among patches of plants and that, unlike hawkmoths, pollen moved by bumble bees sired more outcrossed seeds when it remained within a patch as opposed to moving between patches. Pollinator type differentially affected the outcrossing rate but not seed set, the number of outcrossed seeds or overall male reproductive success. Multiple visits to a plant and more frequent visits by bumble bees could help to explain the lack of impact of pollinator type on overall reproductive success. The increase in selfing rate with hawkmoths likely resulted from the abundant pollen available in experimental flowers. Our findings highlighted a new type of pollinator interactions that can benefit a plant species. Keywords: Aquilegia coerulea, bumble bee, female function, hawkmoth, male function, paternity analysis, pollinator Received 17 October 2008; revision received 16 June 2009; accepted 18 June 2009 Introduction The foraging behaviour of a pollinator influences its effectiveness and its potential impact on a plant’s repro- ductive success via male and female functions (Schem- ske & Horvitz 1984; Herrera 1987; Wilson & Thomson 1991). In hermaphroditic plants, reproductive success depends on both the number of seeds set on a plant (female function) and the number of seeds sired by that same plant (male function). Large variation in repro- ductive success via both male and female functions typ- ically occurs in natural populations (Lee 1986; Campbell 1989; Meagher 1991; Smouse & Meagher 1994; Smouse & Robledo-Arnuncio 2005) and such variation has been associated with various floral and plant characteristics (Stanton et al. 1986; Young & Stanton 1990; Elle & Mea- gher 2000). The interaction between plant traits and pol- linator behaviour, however, plays an important role in influencing a plant’s reproductive success (Castellanos et al. 2003). For example, the foraging behaviour of a pollinator influences its ability to remove pollen from anthers and deposit pollen on stigmas following a sin- gle visit (effectiveness) (Young et al. 2007). Honeybees foraging for pollen typically removed large amount of pollen from anthers but deposited little on stigmas; in contrast, honeybees foraging for nectar on the same plant species removed less pollen but deposited more of it on stigmas (Young et al. 2007). The foraging behav- iour of a pollinator can influence its effectiveness and Correspondence: Johanne Brunet, Fax: (608) 262-4743; E-mail: jbrunet@wisc.edu or Johanne.Brunet@ars.usda.gov Published 2009. This article is a US Government work and is in the public domain in the USA Molecular Ecology (2009) 18, 3745–3758 doi: 10.1111/j.1365-294X.2009.04304.x