DISCUSSION
In community ecology, the relationship between plants and pollinators is critical to understanding how pollinators contribute to plant community functioning. Though observations of pollinator visitation can be used to infer how plants interact with each other through shared pollinators, visitation does not necessarily correspond to pollen transfer (Mayfield 2001; Popic et al. 2013; Ballantyne et al. 2015, 2017; Barrios et al. 2016). Metabarcoding has been critical for understanding if and when bees use certain pollen resources in plant communities (Galliot et al. 2017; Lucas et al. 2018), but to date has only reliably shown if bees use pollen resources from specific species in flowering plant communities, not to what extent (Bell et al. 2017, Bell et al. 2019). This is because relative read abundance from PCR can be unreliable for accurate use in amplicon quantitation in a sample.
Using our new approach, we have shown that shared pollinators amongClarkia species have preferences for different species ofClarkia , and are also inconstant foragers, often carrying more than one species of pollen in their pollen balls at a time. Though the trends of pollen use were similar to what we had expected from pollinator visitation observations alone, our molecular analysis added nuance to how pollinators used Clarkia resources. We have additionally shown that our new method, qAMPseq, estimates the relative amounts of pollen species in a sample.