Heredity (2005), 1–9 & 2005 Nature Publishing Group All rights reserved 0018-067X/05 $30.00 www.nature.com/hdy SHORT REVIEW To self, or not to selfy A review of outcrossing and pollen-mediated gene flow in neotropical trees M Ward1, CW Dick2,4, R Gribel3 and AJ Lowe1 1School of Integrative Biology, University of Queensland, Brisbane, Queensland 4072, Australia; 2Smithsonian Tropical Research Institute, Apartado 0843-03092, Balboa, Republic of Panama; 3Instituto Nacional de Pesquisas da Amazonia (INPA), Avenida Andre´ Arau´jo 2936, Manaus, Amazonas, Brazil Despite the typically low population densities and animal- ranging from a mean of 200 m to over 19 km for species mediated pollination of tropical forest trees, outcrossing and pollinated by small insects or bats. Future research needs to long-distance pollen dispersal are the norm. We reviewed the examine (1) the effect of inbreeding depression on observed genetic literature on mating systems and pollen dispersal for outcrossing rates, (2) pollen dispersal in a wide range of neotropical trees to identify the ecological and phylogenetic pollination syndromes and ecological classes, (3) and the correlates. The 36 studies surveyed found 490% outcrossed range of variation of mating system expression at different mating for 45 hermaphroditic or monoecious species. Self- hierarchical levels, including individual, seasonal, population, fertilization rates varied inversely with population density and ecological, landscape and range wide. showed phylogenetic and geographic trends. The few direct Heredity advance online publication, 10 August 2005; measures of pollen flow (N ¼ 11 studies) suggest that pollen doi:10.1038/sj.hdy.6800712 dispersal is widespread among low-density tropical trees, Keywords: mating system; gene flow; neotropical trees; pollination; pollen dispersal Introduction pollination syndromes (eg Boshier et al, 1995a, b; Stacy et al, 1996; Loveless et al, 1998; Nason et al, 1998). Mating systems, the genetic relatedness and patterns of Genetic marker-based studies have also revealed that pairings between gametes (as distinct from breeding outcrossing rate and pollen dispersal distances are systems, the morphological and physiological character- sensitive to ecological factors, and show variation over istics of pairing; Neal and Anderson, 2005), play a central both spatial and temporal scales (Nason and Hamrick, role in evolutionary theories of the origin of tree species 1997). For example, factors such as population density richness in the tropics (Bawa, 1992). In species-rich and pollinator abundance and composition change over tropical rainforests, the population density of adult trees the range of a species, have a concomitant impact on of a given species is generally low and thus distances outcrossing rate and pollen-mediated dispersal at a between the crowns of flowering conspecifics will be landscape scale (Franceschinelli and Bawa, 2000; Dick large (Hubbell and Foster, 1983). This observation et al, 2003; Degen et al, 2004). Mating system analyses also originally led botanists to predict that most tropical tree indicate the potential for variation in the relative rates of species should be highly self-fertilizing or inbred (eg selfing and outcrossing at a micro-scale, where variation Corner, 1954; Baker, 1959; Fedorov, 1966), under the occurs among individuals within populations, among premise that animal pollinators are unable to move populations over years and from one flowering event among widely spaced conspecifics. Over the last 30 to another (Murawski and Hamrick, 1991; Nason and years, this early view of tropical pollinators and the Hamrick, 1997). Finally, phylogenetic constraints on mating systems of tropical tree species has been floral morphology and self-compatibility system are also completely revised. Community-level studies of breed- expected to influence the observed outcrossing rates and ing systems in tropical trees revealed high levels of patterns of pollen dispersal (Gribel et al, 1999). dioecey (420%), and cross-pollination studies provided This review examines studies that use molecular evidence of self-incompatibility in hermaphroditic or genetic techniques to quantify mating systems and monoecious species (Bawa et al, 1985). Genetic marker- pollen-mediated gene dispersal in neotropical tree based analyses of mating system have tended to confirm species. The review is limited to the neotropics, as the these field studies, revealing high rates of outcrossing authors are familiar with the study species in this region. and long-distance pollen dispersal for a range of However, given the pantropical distribution of most tropical tree families and many genera (Pennington and Correspondence: M Ward, School of Integrative Biology, University of Dick, 2004), our conclusions should have broad geo- Queensland, St Lucia, Brisbane, Queensland 4072, Australia. graphic application. Our discussion of pollen dispersal is E-mail: [email protected] focussed on undisturbed populations, as Lowe et al 4Current address: Department of Ecology and Evolution and Herbarium, University of Michigan, Ann Arbor, MI 48109-1048, USA (2005) reviews studies in fragmented habitats. Our Received 1 March 2005; accepted 13 June 2005 objectives are to (1) examine the relationship between Outcrossing and pollen-mediated gene flow M Ward et al 2 the multilocus outcrossing rate (tm) and phylogenetic documented in the Dipterocarpaceae (Kaur et al, 1978). and ecological factors, such as population density, Apomixis may have an important role in the breeding pollination syndrome and habitat change, and (2) structure of some populations. In a fascinating Malaysian determine the distances over which pollen is dispersed study of a species in the pantropical genus Garcinia (G. among low-density, animal-pollinated trees in species- scortechinii), Thomas (1997) found an entirely female rich lowland neotropical forests. population that persisted solely through apomixis. High outcrossing rates result from self-incompatibility, Review of the literature and may be phylogenetically constrained, but inbreeding depression at the embryo stage can also produce largely A total of 36 case studies encompassing 45 neotropical outcrossed offspring (Seavey and Bawa, 1986). In order tree species were surveyed. For 29 of these case studies, a to distinguish between the effects of self-incompatibility multilocus mixed-mating model (eg Ritland and Jain, and inbreeding depression, Hufford and Hamrick (2003) 1981; Ritland, 1986) was used to determine the mating documented the change in outcrossing rate in P. elegans system. Henceforth, ‘mating system’ is used synony- (Fabaceae) at three early life stages: aborted embryos, mously with the outcrossing rate. Some estimate of mature seeds and seedlings. The outcrossing rate pollen dispersal was made in 10 studies. Allozymes were increased across each life stage (tm ¼ 0.79, 0.82 and employed in 22 case studies, whereas microsatellites 0.91), suggesting that inbreeding depression may explain were utilized in 14 case studies. The results of the high outcrossing observed in seedlings, the stage at literature survey are summarized in Table 1. which many genetic marker studies of mating system are carried out. Several neotropical trees also have late- Correlates of mating system acting self-incompatibility mechanisms (eg Tabebuia; Phylogeny: As the anatomical and biochemical aspects Bittencourt and Semir, 2005), and so a genetic estimation of self-incompatibility systems may be phylogenetically of outcrossing rate would be strongly influenced by the constrained (Gribel et al, 1999), we looked for developmental stage at which seeds are assessed. phylogenetic association with the mating system. The taxonomic breadth encompassed by the relevant Individual, population, range and landscape variation: literature presently includes 15 plant families (Table 1), Individual outcrossing rates can vary widely. Murawski with an average of 2.9 species examined per family. and Hamrick (1992a) reported outcrossing rates in C. It is noteworthy that species in the Malvaceae pentandra (Malvaceae) ranging from complete self- (subfamily Bombacoideae; former Bombacaceae) fre- fertilization to complete outcrossing. This variation was quently exhibit outcrossing rates lower than 80% explained by asynchrony in flowering times and (mixed-mating systems). Mixed-mating systems were nonrandom foraging behaviour of the bat pollinators. found in Pachira quinata (Fuchs et al, 2003), Cavanillesia Similar results for C. pentandra reported by Gribel (in platanifolia (Murawski and Hamrick, 1992b), Ceiba pen- Wilson et al, 2001) showed the percentage of selfed seeds tandra (Murawski et al, 1990; Murawski and Hamrick, to range from zero to 97.8% for individual trees. 1991, 1992a; Gribel et al, 1999; Lobo et al, 2005) and Multilocus outcrossing rates for 25 individual trees Pseudobombax munguba (Gribel and Gibbs, 2002). This varied from 0.38 to 1.00 in an Amazonian population of pattern may apply more generally to trees in the order Swietenia macrophylla (Meliaceae), although 23 individuals Malvales, which includes the Sterculiaceae, Ebenaceae were predominantly or completely outcrossing (Lemes, and Tiliaceae. Self-fertilization rates higher than 50% 2000). Similarly, Latouche-Halle´ et al (2004) observed have been documented in Shorea and Stemonoporus of the marked variation in outcrossing rates among individuals Dipterocarpaceae (Murawski and Bawa, 1994; Murawski of D. guianensis (Fabaceae), likely due