Flowers and pollen dispersal by individual pollinators

Individual flowers serve pollination by contributing to a plant's overall attractiveness to pollinators (including both signaling and energetics) and by controlling the transfer of pollen to and from each visitor. Because attraction typically involves all flowers open on a plant, we review this function when we discuss inflorescences (below). Here, we consider the role of individual flowers in controlling pollen exchange with pollinators. Floral characteristics mediate this exchange by determining which areas of pollinators' bodies contact a flower's pollen and stigma(s), and the intensity and duration of that contact. Most aspects of floral form, as well as the production of nectar and floral oils, contribute to these pollination functions. Before addressing the roles of specific floral features in pollen export and import, we review the general pattern of pollen dispersal by a single pollinator from a specific donor flower to recipient flowers.

For most angiosperms, individual pollinators transport donor pollen to several or many recipient flowers (reviewed by Morris et al. 1994; Harder & Wilson 1998), because each stigma receives only a fraction of the pollen carried by a pollinator from a specific donor flower. In general, successive recipient flowers receive progressively less pollen from a particular donor (see Fig. 15.1a) due to deposition on stigmas and transport losses (e.g., grooming). If the pollen carried by a pollinator behaved as a single, completely mixed population, receipt of donor pollen by successive recipient flowers would decline as expected for simple geometric decay - as in Fig. 15.1a. However, in reality the decline is more rapid among initial recipients and more gradual among later recipients (reviewed by Morris et al. 1994), suggesting that the pollen carried by a pollinator behaves as a subdivided population with heterogeneous transport conditions.

Such subdivision could arise either passively or actively (Harder & Wilson 1998). Passive segregation could arise through at least three mechanisms: differences in the ability of areas on a pollinator's body to carry pollen (e.g., hairy head versus smooth mouthparts); variation in the incidence and intensity of contact by pollinators with anthers and/or stigmas; and accumulation of pollen in layers on the pollinator. Active segregation of pollen on a pollinator arises from behaviors such as grooming or movements of the mouthparts. These behaviors affect pollen on some sites on a pollinator's body (exposed sites), but not on others (safe sites: e.g., Kimsey 1984; Thomson 1986). Such behaviors could also move pollen from exposed to safe sites, supplementing safe sites with pollen from flowers visited previously while depleting exposed sites. As a result, the proportion of pollen from a specific donor flower that is dispersed to stigmas via safe versus exposed sites increases steadily as the pollinator visits successive recipient flowers.

Variation in pollen removal from a donor flower by individual pollinators is modified by grooming and layering dynamics so that pollen export increases non-linearly as removal increases (Harder & Wilson 1997). When

Fig. 15.1. Theoretical features of pollen dispersal by a single pollinator from (a) a single flower and (b) a 50-flowered plant (see Harder & Barrett 1996). Panel (a) considers the pollination fates of pollen removed from the first of five flowers (the donor flower) visited by a pollinator on a focal plant (di = rp[i - p]i-i, where d is the proportion of donor pollen received by recipient flower i, r is the proportion of available pollen removed from each flower, and p is the proportion of pollen carried by the pollinator that is deposited on the stigma of each flower). Panel (b) illustrates how pollen export from a donor plant (E) to other plants varies with the proportion of flowers that a pollinator visits per inflorescence (E = r[i - {1 - p}vn]/p, where v is the proportion of the n open flowers visited by the pollinator). For both panels, r = 0.2, p = 0.1, and n = 50 flowers.

Fig. 15.1. Theoretical features of pollen dispersal by a single pollinator from (a) a single flower and (b) a 50-flowered plant (see Harder & Barrett 1996). Panel (a) considers the pollination fates of pollen removed from the first of five flowers (the donor flower) visited by a pollinator on a focal plant (di = rp[i - p]i-i, where d is the proportion of donor pollen received by recipient flower i, r is the proportion of available pollen removed from each flower, and p is the proportion of pollen carried by the pollinator that is deposited on the stigma of each flower). Panel (b) illustrates how pollen export from a donor plant (E) to other plants varies with the proportion of flowers that a pollinator visits per inflorescence (E = r[i - {1 - p}vn]/p, where v is the proportion of the n open flowers visited by the pollinator). For both panels, r = 0.2, p = 0.1, and n = 50 flowers.

grooming intensity (and associated pollen loss) varies positively with the amount of pollen removed from a flower (e.g., Harder 1990a), enhanced removal increases subsequent pollen export at a decelerating rate (e.g., Thomson & Thomson 1989). This increased export arises both because each recipient flower receives more donor pollen and because donor pollen reaches more recipient flowers (Harder & Wilson 1998). With layering, total export by a single pollinator initially increases with pollen removal, as each recipient receives more donor pollen. However, greater increases in removal decrease total export, because pollen becomes buried more quickly and so does not reach distant recipients (Harder & Wilson 1998). As we discuss below, the diminishing returns associated with grooming and layering influence the evolution of attractiveness and floral control of pollen removal.

The importance of pollen exchange between flower and pollinator for pollen dispersal by individual pollinators should promote selection for floral features that mediate pollinator-flower interactions to a plant's advantage. Obvious features affecting pollen removal and deposition include corolla size and shape (Murcia 1990; Campbell et ol. 1996; Kobayashi et ol. 1999; but see Wilson 1995), the amount and schedule of pollen presentation (Harder & Thomson 1989; Harder & Wilson 1994), anther position (Harder & Barrett 1993), and stigma size, structure and position (e.g., Waser & Price 1984; Murcia 1990; Campbell et ol. 1994; Conner et ol. 1995). In addition, pollen exchange often varies with the duration of pollinator visits (Harder 1990&; Murcia 1990; Conner et ol. 1995; Hurlbert et ol. 1996; but see Mitchell & Waser 1992), which depends on the amount and quality of food (nectar or pollen) present in a flower (Montgomerie 1984; Harder 1986; Thomson 1986; Martinez del Rio & Eguiarte 1987; Harder & Barclay 1994). Because food availability often increases as time elapses since the last pollinator visit (Waser & Mitchell 1990; Kadmon 1992; Williams 1997; Jones et ol. 1998), flowers exchange more pollen with individual pollinators when they receive infrequent visits than when pollinators visit often (Harder & Thomson 1989; Klinkhamer & de Jong 1993; Harder & Wilson 1994).

Was this article helpful?

0 0

Post a comment