The mating consequences of rewarding vs. deceptive pollination systems: Is there a quantity–quality trade‐off?

Abstract

Plant mating commonly involves quality–quantity trade‐offs. Such a trade‐off features explicitly in the cross‐promotion hypothesis for the evolution of pollination by deceit. According to this hypothesis, rewardlessness enhances mating quality by altering pollinator behavior in ways that reduce self‐pollination and increase outcrossing, thus compensating for the reduced pollen dispersal and seed production resulting from lack of reinforcement of pollinator visitation behavior. The high prevalence of rewardlessness in orchids suggests that deceit pollination conveys benefits that outweigh the fecundity advantages of reward production. We test the cross‐promotion hypothesis in Disa, an African orchid genus in which nectar rewards have evolved repeatedly from rewardless ancestors. To address this hypothesis directly, we quantified pollinator behavior and the associated dispersal of stained pollen for four nectar‐producing (N+) and 10 nectarless (N−) species. We also assessed more extensive, if less direct, evidence from a survey of lifetime pollination outcomes for 15 N+ and 32 N− species. Pollinators visited N+ species more frequently and visited 1.5 times more flowers per inflorescence than in N− species. In contrast, pollinators skipped more conspecific plants between visits to donor and recipient plants when visiting N− species than when visiting N+ species. Over floral lifetimes, both N− and N+ species exhibited similar overall pollen‐transfer efficiency (8.2%) and did not differ in the overall fractions of pollen involved in self‐pollination (26% of all pollen deposited), geitonogamy (82% of all self‐pollination), or export to other conspecific plants (74%). N+ species set more fruit per flower (68.8% vs. 48.8%), but equivalent fractions of ovules per fruit were fertilized and subsequently developed into seeds in N− and N+ species. These findings provide only limited evidence that deceit pollination enhances mating quality (nectar production causes more local mating in N+ species than in N− species). Contrary to the hypothesis, deceit does not generally affect pollinator‐mediated self‐pollination. The repeated evolution of nectar production from rewardless ancestors in Disa likely occurred under severe visit limitation, favoring nectar‐producing mutants with higher reproductive output. Given that we did not find any mating quality advantage for rewardlessness in Disa, its persistence in some lineages remains an evolutionary enigma.

Mating consequences of contrasting hermaphroditic plant sexual systems

“…; Hobbhahn et al. ); the tendency of pollen to remain on pollinators’ bodies, rather than being deposited on stigmas (pollen carryover: Waser and Price ; Thomson and Thomson ); and postpollination processes that bias ovule fertilization and seed development against neighbors being parents of mature seeds (e.g., Harder et al. ; Souto et al.…”

“…1C). Three processes determine intermate distance as inferred from genotyped seeds: pollinator movement between successive plant visits (e.g., Thomson and Thomson 1989;Jersáková et al 2006;Hobbhahn et al 2017); the tendency of pollen to remain on pollinators' bodies, rather than being deposited on stigmas (pollen carryover: Waser and Price 1983; Thomson and Thomson 1989); and postpollination processes that bias ovule fertilization and seed development against neighbors being parents of mature seeds (e.g., Harder et al 1985;Souto et al 2002). Of these processes, emasculation of male flowers seems most likely to affect pollinator movement.…”

Pollination Mechanisms are Driving Orchid Distribution in Space

“…In general, plants of nectariferous species are visited more frequently than nectarless plants [29][30][31] . Pollinators also visit more flowers per inflorescence of nectariferous than in nectarless species 16,31,32 .…”

“…In general, plants of nectariferous species are visited more frequently than nectarless plants [29][30][31] . Pollinators also visit more flowers per inflorescence of nectariferous than in nectarless species 16,31,32 . Nectariferous species are less pollinator-specific than deceptive species, among which the most pollinator-specific are sexually deceptive species 32,33 .…”

Do local conspecific density and floral display size influence fruit set via pollinator visitation in Orchis militaris?

“…The impact of conspecific density on fruit set has been investigated in several deceptive orchid species (Fritz and Nilsson 1996, Meléndez-Ackerman and Ackerman 2001, Juillet et al 2007). Its negative effect on fruit set (Gumbert andKunze 2001, Internicola et al 2006) is hypothesised to be due to learning avoidance (Anderson andJohnson 2006, but see Juillet et al 2011), which lasts at least in the short term (Hobbhahn et al 2017). In addition, deceptive small floral displays can experience reduced pollination success or fruit production compared to larger displays (Aragón andAckerman 2001, Sletvold andÅgren 2011).…”