Spatial variation in selection in a plant-pollinator system in the wadis of Sinai, Egypt.

We studied an insect-plant pollination system in adjacent steep-sided wadis and a connecting plain in the mountains of southern Sinai (Egypt): this environment creates a strongly divided habitat, which may promote the local differentiation of sub-populations. We tested for spatial differences in phenotypic reproductive characters of the only plant flowering abundantly in early spring, Alkanna orientalis (Boraginaceae), and its major pollinator at that time of year, Anthophora pauperata (Apoidea, Anthophoridae). There were significant morphological differences between sub-populations of Alkanna, mainly between plants from the narrower wadis and those on the interconnecting plain. Flowers on the plain were larger, with wider corollas and more nectar standing crop; these plants retained more flowers on the inflorescence, but received many fewer visits to flowers. There was a significant selection gradient between flower size and maternal fitness (seed set) in the plain, but not elsewhere. Natural selection may have increased resources devoted to attracting insect visitors in response to fewer pollinating visits in the plain. Consistent with this explanation, by experimentally manipulating flower number per plant, we showed that within a wadi having more flowers on a plant secured more visits.

Climate change: bees and orchids lose touch.

Spring temperature increases could differentially affect flowering times and pollinator flight periods, leading to asynchrony and reduced pollination. A specialist orchid-bee study combining herbarium, museum and field data shows that bee flight dates are advancing faster than orchid flowering, which could lead to significant future uncoupling.

Ecology: pollinator-plant synchrony tested by climate change.

Flowering plants could lose their pollination service if climate warming potentially uncouples timing of flowering from pollinator availability. Recent evidence might suggest this effect may be less than feared.

Nectar theft and floral ant-repellence: a link between nectar volume and ant-repellent traits?

As flower visitors, ants rarely benefit a plant. They are poor pollinators, and can also disrupt pollination by deterring other flower visitors, or by stealing nectar. Some plant species therefore possess floral ant-repelling traits. But why do particular species have such traits when others do not? In a dry forest in Costa Rica, of 49 plant species around a third were ant-repellent at very close proximity to a common generalist ant species, usually via repellent pollen. Repellence was positively correlated with the presence of large nectar volumes. Repellent traits affected ant species differently, some influencing the behaviour of just a few species and others producing more generalised ant-repellence. Our results suggest that ant-repellent floral traits may often not be pleiotropic, but instead could have been selected for as a defence against ant thieves in plant species that invest in large volumes of nectar. This conclusion highlights to the importance of research into the cost of nectar production in future studies into ant-flower interactions.

Bidirectional flower color and shape changes allow a second opportunity for pollination.

Flowers act as "sensory billboards" with multiple signals (color, morphology, odor) attracting and manipulating potential pollinators. Many use changing signals as indicators that visitation and/or pollination have occurred). Floral color change is commonly used to transmit this information (often correlated with reduced nectar reward) and can be specifically triggered by pollination or visitation. By retaining color-changed flowers, plants benefit from larger floral displays but also indicate at close range which flowers are still rewarding (and still unpollinated), so that visitors forage more efficiently. However, the legume Desmodium setigerum shows a unique ability, if inadequately pollinated, to reverse its flowers' color and shape changes. Single visits by bees mechanically depress the keel and expose stigma and anthers (termed "tripping"); visits also initiate a rapid color change from lilac to white and turquoise and a slower morphological change, the upper petal folding downwards over the reproductive parts. But flowers receiving insufficient pollen can partially reopen, re-exposing the stigma, with a further color change to deeper turquoise and/or lilac. Thus, most flowers achieve pollination from one bee visit, but those with inadequate pollen receipt can reverse their signals, earning a "second chance" by eliciting attention from other potential pollinators.