Do flower-colonizing microbes influence floral evolution? A test with fast-cycling Brassica

Pollinators are thought to be the main drivers of floral evolution.
Flowers are also colonized by abundant communities of microbes that can
affect the interaction between plants and their pollinators. Very little
is known, however, about how flower-colonizing microbes influence floral
evolution. Here we performed a six-generation experimental evolution study
using fast-cycling Brassica rapa, in which we factorially manipulated the
presence of pollinators and flower microbes to determine how pollinators
and microbes interact in driving floral evolution. We measured the
evolution of six morphological traits, as well as plant mating system and
flower attractiveness. Only one of the six traits (flower number) evolved
in response to pollinators, while microbes did not drive the evolution of
any trait, nor did they interact with pollinators in driving evolution of
morphological traits. Moreover, we did not find evidence that pollinators
or microbes affected the evolution of flower attractiveness to
pollinators. However, we found an interactive effect of pollinators and
microbes on the evolution of autonomous selfing, a trait that is expected
to evolve in response to pollinator limitation. Overall, we found only
weak evidence that microbes mediate floral evolution. However, our ability
to detect an interactive effect of pollinators and microbes might have
been limited by weak pollinator-mediated selection in our experimental
setting. Our results contrast with previous (similar) experimental
evolution studies, highlighting the susceptibility of such experiments to
drift and to experimental artefacts.