1. Introduction
Pollination is crucial for sexual reproduction in flowering plants. Over 80% of Angiosperms (including 75% of crop species), rely on animal vectors for pollen transport between or within flowers (Herrera and Pellmyr 2002, Ollerton et al. 2011). When pollen deposition is inadequate to fertilize all available ovules, plants face pollen limitation (PL) (Ashman et al. 2004a). Such pollen inadequacy can be driven by various mechanisms, such as the low frequency of effective pollinator visits and/or low quantity or quality of conspecific pollen deposited in the stigmas (Ashman et al. 2004; Knight et al. 2005). Regardless of the mechanism, meta-analyses compiling the effect sizes of PL on fruit and/or seed production have shown that pollen-limited plants can produce, on average, 75% less fruits and/or seeds than pollen-supplemented flowers (not pollen-limited) (Knightet al. 2005). This negative effect of PL is observed in most Angiosperm species investigated so far (see García-Camacho & Totland 2009; Wolowski et al. 2014; Burns et al. 2019). In the near future, the negative effects of human activities on pollinators worldwide are expected to progressively increase PL (González-Varoet al. 2013; Neuschulz et al. 2016), raising concerns about the impacts of PL on the reproductive success and population dynamics of plant species in pristine and human-disturbed habitats (e.g. Knight 2004; Ward & Johnson 2005; Freitas et al. 2020; Soltaniet al. 2021).
Despite PL being a pivotal subject in plant reproductive biology, one important consequence of PL has received little to no attention: its effect on seed dispersal effectiveness, especially for plant species dispersed by animals. The seeds of many Angiosperms, including 25-80% of temperate and 50-90% of tropical species flora, are animal-dispersed (Jordano & Schupp 2000). For those plants, variation in seed recruitment and the spatial distribution of plants is dictated by the outcome of disperser-plant interactions (Jordano 1995; Galetti et al. 2013; Snell et al. 2019). Because the magnitude of PL experienced by a population directly impacts the quantity and quality of dispersal units (i.e. unit of propagation of a plant species and, in our case, the entity located and removed by the disperser animals) produced by a plant, PL could have cascading effects on the dynamics of plant-disperser interactions, patterns of post-dispersal seed success, and ultimately, on the spatial distribution of plants.
A search on the Web of Science database reveals that this link between PL and seed dispersal has been dismissed, at least in part, due to the non-integrative nature of studies evaluating the outcomes of pollination and seed dispersal to plants. In February 2023, we found 3,765 and 20,219 studies focusing on PL or plant seed dispersal, respectively (Boolean operators: “pollen limitation” OR “plant seed dispersal”). However, only 328 studies (8.7% of total PL studies and 1.1% of total dispersal studies) focused on any connection between these two ecological processes (Boolean operators: “pollen limitation” AND “plant seed dispersal”). This suggests that most studies quantifying PL and/or evaluating its effects on plant reproductive success neglect natural processes following seed formation. Similarly, studies describing the dynamic of seed dispersal by animals have focused on a spatial snapshot of a particular dispersal system, mostly neglecting the pre-dispersal processes driving the availability of dispersal units, such as PL. Considering that pollination and seed dispersal are two of the most threatened processes regulating plant demography and regeneration (Neuschulz et al. 2016), we propose an integrative approach that allow us to evaluate the interplay between these processes, improving our ability to predict and manage plant populations in different ecological scenarios.
Here, we present a new conceptual framework about the mechanisms through which PL can directly and/or indirectly affect the outcome of seed dispersal by animals (Fig. 1). We incorporate the quantitative and qualitative effects of PL on plant reproduction (pre-dispersal processes) into the Seed Dispersal Effectiveness approach (SDE) – the most common approach used to evaluate the outcome of seed dispersal for plant species. According to the SDE, the outcome (or effectiveness) of seed dispersal for a given plant is determined by the product of the number of seeds dispersed by all dispersers in the community (quantitative component) and the probability that dispersed seeds survive to adulthood (qualitative component) (Schupp et al.2010). While providing a valuable guide for SDE quantification, this approach still neglects the role of the pre-dispersal processes, such as PL. To bridge this gap, we first describe the direct effects of PL on quantitative and qualitative traits of dispersal units produced by animal-dispersed plants. Several of these effects can interfere with the value of a plant population as a foraging patch to dispersers, as well as the value of dispersal units as a food resource to dispersers (Donahue et al. 2003; Valenta & Nevo 2020). Then, we use Optimal Foraging Theory (OFT) (MacArthur & Pianka 1966) to predict how the effects of PL on dispersal units can affect disperser foraging behavior and, consequently, quantitative and qualitative components of SDE. Finally, we discuss the consequences of this indirect link between two key ecological processes for the ecological and evolutionary dynamic of animal-dispersed plants. To bolster our framework, we use two of the most common types of seed dispersal observed in nature as models: endozoochory (i.e. seed dispersal by frugivores that ingest the fruits while visiting the mother plants - Soltani et al. 2018) and myrmecochory (i.e. seed dispersal by ants that transport the diaspores to their nest – observed in more than 23,000 Angiosperm species - Lengyel et al. 2010). Using these dispersal strategies, we propose scenarios in which PL should affect specific outcomes for dispersal, hoping to spur novel research directions on the subject.