Shifts between pollinators are frequent in the evolution of flowering plants and have contributed to their incredible success. For a shift to occur, there must be intraspecific variation in floral traits that diverge to suit the new pollinator. The reproductive isolation introduced by the shift over time provides the basis for speciation. These cases can be subtle, and difficult to recognize and study in nature.
In a new article published in Current Biology, lead author Miguel Castañeda-Zárate and colleagues investigated the southern African grassland orchid Satyrium longicauda and the basis of its shift from moth pollination to bee pollination in a certain floral morph. Within a single square kilometre site, the authors identified six distinct forms of the plant that differed in floral traits including spur length, in number and position of leaves, and in habitat.
Observation showed that five of the forms were pollinated by nocturnal moths, while the sixth appeared to be pollinated by the bee species Rediviva neliana. While the moths gathered nectar by dipping their proboscis into the nectar spur and receiving pollen on the proboscis, the bees used their forelegs to probe the flower, receiving the pollen on their forelegs. The authors confirmed that bee visits had resulted in pollination, finding that all six morphs ranged from 79-90% fruit set.
Upon further investigation, the researchers found that the bees were gathering not only nectar, but floral oil, which they collect with the hairs on their forelegs. All six floral morphs tested positive for the presence of floral oil, and all contained diacetin, a cue for oil-collecting bees, though the diacetin concentration was much higher in the bee-pollinated morph.
“The presence of traces of diacetin in all moth morphs indicates a possible pre-adaptation that might explain why bees initially visited flowers adapted for moth pollination,” explains Castañeda-Zárate. “In order for a pollinator switch to occur, ancestral flowers adapted to the original moth pollinator must somehow gotten visited by the novel pollinator. The presence of traces of oil may provide a piece of the puzzle how that occurred in our study system.”
A further difference in the bee-pollinated morph is in the length and contents of its spur. The bee morph had almost no nectar in the spur, coupled with a small but significant decrease in the spur’s length, suggesting that it may be vestigial under the newer pollination system. Castañeda-Zárate points out that spur-loss is rare in floral evolution. “Gain of spurs is quite common but loss is unusual. The classical Darwinian explanation for long nectar spurs featured a process of coevolution with a pollinator in which spurs and pollinator mouthparts coevolved to ever-increasing lengths,” he says. “Nectar spurs are also considered a ‘key innovation’ associated with plant diversification.”
The advantage to the single morph that has moved to bee pollination could be lower competition for pollinators or a reproductive barrier against unfit hybrids, though this has yet to be shown. This work highlights a possible evolutionary pathway toward bee pollination and demonstrates the importance of pre-adaptations and reward chemistry as sources of mismatch between morphology and pollinators.