Flowers stay open for pollinators, not daylight

A honeybee explores the depths of a dandelion, one of the species used in Fründ et al.‘s experiments. Photo by je-sa.

ResearchBlogging.orgIf you’ve ever stopped to admire morning glory flowers opening first thing in the morning, then noticed they’ve closed by evening, you’re at least dimly aware of one of the longest-established ideas in plant biology: that flowers open and close on a reliable daily schedule. Different species are open at different times of day, of course, but each flowering plant has its preferred open period, and it sticks to that schedule during its flowering season.

This idea led Carolus Linneaus, the father of modern biological taxonomy, to propose an Horologium florae, or “floral clock” using plantings of species with known flowering times to mark the hours. You can find his table of proposed species in the online version of Linneaus’ 1783 treatise Philosophia Botanica, if you’re not averse to Latin. Studies of flowers’ daily schedules go back to well before English was the language of international science, and continue to the present day [$a].

Yet no one seems to have spent much time considering how flowers’ schedules might respond to the activity of their very reason for being: pollinators. Flowers don’t open just to be open in a particular kind of sunlight—they’re open to attract animals that can carry pollen to another plant, and maybe leave some, too. If a flower receives enough pollen to make seeds by noon, why would it stay open until two o’clock?

According to some new experimental results, the answer to that question is that they don’t [$a].

Jochen Fründ, Carsten F. Dormann, and Teja Tscharntke set out to see whether a selection of European wildflowers adjusted their opening schedules in response to pollination, with two major experiments and a broader-scale observation project. The experiments address whether pollinator activity could change flowers’ schedules; the observations help determine how important those changes might be in studies of plant-pollinator interaction.

A floral clock in Geneva—not quite what Linneaus had in mind. Photo by aranmanoth.

In the first experiment, the team planted wildflowers—Crespis capillaris, a close relative of common dandelions—in experimental plots spaced across a field. Plots were either caged or left open to insect visitors, and Fründ et al introduced bees into some of the caged plots. So some plots had a controlled set of pollinators, some had none at all, and some had whatever pollinators were already active in the field.

The team then watched the flowers’ daily opening and closing in the experimental plots. (They had a lot of help—a long list of names in the paper’s Acknowledgements section ends with “and many others.”) Over the same period of time, flowers in the un-caged plots received more insect visitors than flowers in either other treatment, and had mostly closed by midafternoon; flowers in the caged plots with bees introduced received fewer visitors and closed hours later; and flowers in the plots with no pollinators at all stayed open till evening.

So flowers experiencing the same daylight pattern closed earlier if they received more pollinator visits. The team followed up this result by hand-pollinating flowers of C. capillaris and a handful of closely related species growing in the same field, including dandelions—and flowers of three out of four species closed more rapidly when hand pollinated. Dandelions didn’t respond to hand pollination, a result the authors explain by noting that dandelions often self-pollinate, and so don’t need to wait for animal pollinators.

Finally, the team compiled observations of plant-pollinator interactions from sites similar to their study field located across Germany, and divided them into observations taken before solar noon, when the focal flower species from the experiments above tend to be open, and after solar noon. Which pollinator species visited which flowering plants depended significantly on when the observations were made—to the extent that the apparent importance of C. capillaris and its relatives is entirely different before and after noon.

Of course, these results apply directly to only a handful of species representing a particular group of flowering plants—but it’s a group with a lot of widespread and abundant members, and the result is straightforward and striking. Animal-pollinated plants may not behave much like clocks at all. Instead, they’re more like the patrons of a singles bar: they show up at about the same time and hang around until they find someone to buy them a drink. That’s a dynamic worth keeping in mind for studies of plant-pollinator interaction, since it suggests that the partners a pollinator chooses will depend, at least in part, on whether or not it’s out after closing time. ◼


Ewusie, J., & Quaye, E. (1977). Diurnal periodicity in some common flowers. New Phytologist, 78 (2), 479-485 DOI: 10.1111/j.1469-8137.1977.tb04854.x

Fründ, J., Dormann, C., & Tscharntke, T. (2011). Linné’s floral clock is slow without pollinators – flower closure and plant-pollinator interaction webs. Ecology Letters DOI: 10.1111/j.1461-0248.2011.01654.x

von Hase, A., Cowling, R., & Ellis, A. (2005). Petal movement in cape wildflowers protects pollen from exposure to moisture Plant Ecology, 184 (1), 75-87 DOI: 10.1007/s11258-005-9053-8