Pollinator-driven floral evolution is known to occur via two distinct routes: shifts between pollination systems (eg, from one pollinator to another or away from animal pollination altogether) and shifts in how a given pollinator is used (such as the placement of pollen on the insect body). Either process can eventually result in speciation due to reproductive isolation. In order to quantify the macroevolutionary importance of these two types of pollination shifts, they must be examined in a phylogenetic context that demonstrates the frequency of each type of shift within a group.
In a new article published in Annals of Botany, lead author Saroj Ruchisansakun and colleagues used a phylogeny of 282 species of Balsaminaceae, which includes the large genus Impatiens, to reconstruct the evolution of traits related to pollination syndromes. A key trait examined was the evolution of floral asymmetry, which indicates a probable shift in pollination deposition from one part of an insect body to another. Shifts in pollination syndromes were inferred from previously corroborated changes in the shape and colour of the corolla.
The researchers found that although both floral symmetry and pollination syndrome shifted multiple times, the latter changed more frequently. Floral symmetry shifts tended to be unidirectional moves toward asymmetry, which was associated with bee pollination. Asymmetry appeared in 19% of species studied, despite being relatively uncommon in angiosperms, demonstrating its importance in this group.
“Small differences in floral structures mean that a bumble bee may carry pollen from one species on its head, from another species on the central part of the thorax, and from another species only on the left hand side of the thorax. This way, the same bumble bee can visit three co-flowering Impatiens species, even in a single, quite small patch of flowers without polluting stigmas of any of these species with the “wrong” kind of pollen,” explains co-author Dr. Timo van der Niet, a senior lecturer at the University of KwaZulu-Natal in South Africa. “The literature review we did for our study revealed that pollination systems are only known for about 5% of all species,” he adds. “This lack of natural history knowledge is currently the most important limiting factor to understanding of pollinator-driven evolution in flowering plants as a whole. I therefore hope that scientists are encouraged to perform detailed studies, based on natural history observations, to characterize plant-pollinator interactions.”