The evolution of wind pollination in angiosperms

Theresa M. Culley* Stephen G. Weller Ann K. Sakai Dept of Ecology and Evolutionary Biology, University of California–Irvine, Irvine, CA 92697, USA. *e-mail: [email protected] Although much emphasis has been placed on specialist pollination systems, attention is shifting towards systems in which plants are visited by a variety of pollinators [1,2]. Until recently, examples of generalist pollination systems have involved biotic pollinators (mainly insects). Although abiotic pollination systems are usually considered separately from their biotic counterparts, the recent discovery of both wind and insect pollination (ambophily) in several plant species has highlighted the importance of a mixture of insect (entomophily) and wind pollination (anemophily) in generalist pollination systems. Here, we review features of anemophily and analyse the extent of specialization towards wind pollination. Wind pollination has evolved repeatedly in numerous lineages and is presently found in ~18% of angiosperm families [3], including the Juglandaceae, Betulaceae and Poaceae. Wind pollination is considered a derived condition, because phylogenetic analyses suggest an insect-pollinated ancestor for many wind-pollinated species [4–8] and pollination of many basal angiosperms is facilitated by beetles and flies [9]. By contrast, wind pollination is the ancestral state in gymnosperms. A change from animal to wind pollination has occurred in many angiosperm taxa, including at least twice in Schiedea spp. (Caryophyllaceae) [10], once in the Chelidonioideae (Papaveraceae) [6], and three times in the Hamamelidoideae (Hamamelidaceae) [7] and the Fagaceae [11]. Nonetheless, there is still little understanding of mechanisms underlying the transition in these taxa, even though other shifts, such as from biotic pollination to selfing, have been studied extensively [12–15]. Although the evolution of wind pollination has previously received considerable attention [16–18], the recent availability of molecular techniques in combination with phylogenetic analyses [5–8,11,19] has provided new insights into factors favoring the evolution of this pollination system. Wind pollination is thought to evolve in response to changes in the abiotic environment that render biotic pollination less advantageous whilst enhancing the success of wind pollination [16,17,20]. Compared with animal pollination, wind pollination is a relatively passive process in which pollen release, transport and deposition depend largely on abiotic factors other than just wind, such as humidity, rainfall and temperature. Wind-pollinated species typically inhabit relatively open areas and disperse pollen in dry environments, where it has a greater probability of reaching receptive surfaces. Many of these species also possess morphological features that maximize reproductive success within such environments (Table 1; Fig. 1). Wind pollination is common in higher latitudes and elevations, especially in temperate areas, but is considered rare in the tropics, especially in lowland rainforests [18]. The evolution of wind pollination has generally been explained in terms of environmental variation. For example, wind pollination might be facilitated by migration of individuals into an area with a distinct dry season, where conditions are unfavorable for insect pollination [10,16]. Variation in the biotic environment could also promote wind pollination through fluctuations in pollinator abundance [21,22]. In areas of low density, or in newly colonized habitats, a decline in pollinator availability could result in reduced reproductive success of insect-pollinated plant species. Over time, repeated failure of insect pollination might favor the selection of morphological features that improve wind pollination, even at the cost of effective biotic pollination [19].