Hung out to dry: diminished flowers offer less to pollinators and us

Abstract

The effects of drought conditions on flowering plants have been studied for decades by numerous researchers (Waser & Price, 2016; Phillips et al., 2018; Höfer et al., 2021, 2023; Kuppler et al., 2021; Cordeiro & Dötterl, 2023; Barman et al.). For example, under dry conditions, wildflowers are often stunted, not attaining their normal height and failing to produce normal levels of fruit or seed production. Drought stress can alter flowering time, causing either early flowering or delayed flowering depending on plant species and timing of drought conditions. Drought can also negatively impact pollen germination rates and reduce the quality and quantity of pollen that flowers produce. This in turn can affect pollination, fertilization and seed or fruit set. Impacts on flowering time, floral morphology and pollen viability have also been observed. There is evidence of drought or xeric environments producing flowers with reduced amounts of nectar and pollen and changes in floral fragrances. Additionally, studies have shown the potential effects of drought on pollination behavior. In Table 1, we present ideas for the possible effects of drought conditions on plants and their bee pollinators, on both short- and long-term timescales.

The results of Barman et al. are generally consistent with most previous studies of the effects of drought stress on plant biology, save for the failure to find effects of water stress on floral fragrance. However, most previous studies were done using species with hermaphroditic (perfect) flowers. Some analysis of drought effects has been made on corn, a wind-pollinated monoecious plant (Campbell et al., 2014). Barman et al.'s study is the first to document detailed drought effects on pollination and pollinator behavior in a monoecious species. This is interesting because monoecy is associated with patterns of adaptive phenotypic plasticity in response to environmental stress, including drought. How, for example, might drought affect reproductive allocation to male vs female flowers in the context of pollination? Life history theory predicts that environmental stress will result in greater investment in male flowers, because female flowers are costlier than male flowers. Barman et al.'s results are consistent with this prediction. Female flowers showed proportionately greater reduction in weight and petal length under drought stress than males and were more likely to be aborted. Drought reduced nectar volume in both sexes. However, the reduction was proportionately greater in females which, while producing more nectar than males under normal conditions, produced no nectar at all under drought conditions. While these effects are consistent with adaptive changes in reproductive allocation, it remains possible that the effects of drought are passive rather than adaptive, that is, indicative of constraints imposed by water stress. Studies of underlying mechanisms including functional genomics are needed to address this issue (Van Kleunen & Fischer, 2005; Golenberg & West, 2013).

Adaptive or not, how did changes in floral traits under water limitation affect pollination? Barman and co-authors allowed buff-tailed bumblebees (Bombus terrestris L.) from a captive colony (in Salzburg, Austria) to visit artificially water-stressed or control plants in a flight cage. Female flowers, which are larger, but fewer in number, were readily visited by B. terrestris worker bees, even though drought-stressed female flowers offered no nectar at all. Under drought conditions, bees might do better to only visit male flowers, yet authors detected no effect of drought stress on pollinator visitation to female flowers. This intriguing result suggests that under drought stress, female flowers are acting essentially as rewardless mimics of the rewarding male flowers. Obligate intersexual mimicry involving a rewardless gender is known in plants including begonias (Russell et al., 2020) and some cucurbits (Dukas, 1987), but here the mimicry is induced by drought.

Of relevance to the notion of drought-induced intersexual mimicry, Barman et al. failed to find differences in the quantity or composition of floral volatiles between male and female flowers, whether the plants were stressed or not. While a study by Campbell et al. (2019) found that certain plants exhibited phenotypic plasticity in the production of floral volatiles under drought conditions, male and female Styrian pumpkin flowers apparently smell alike, even when water-stressed.

Despite the similarity in fragrance, visits to female flowers were consistently greater in number than visits to male flowers, suggesting that bees could in fact discriminate between males and females, perhaps using visual cues. Why aren't the female flowers avoided altogether? Possibly, the bees experience some constraints in discriminating between male and female flowers. In earlier studies of bumblebee foraging on begonias and cucurbits (Dukas, 1987; Russell et al., 2020), it was shown that although bees do learn to avoid obligately rewardless females to some extent, they never do so entirely. Bees seemed simply unable to distinguish male and female flowers with 100% accuracy. In the case of Styrian oil pumpkin, the fact that female flowers not only look and smell like male flowers but are also bigger than males, perhaps makes the female flower a ‘super-stimulus’ that is difficult for bees to ignore despite the lack of nectar reward.

As noted by Barman et al., European bumblebees such as B. terrestris are not the native and co-adapted pollinators of these cucurbits, which evolved in the Americas, specifically Mexico (Hurd et al., 1971). The bee genera Peponapis (now placed in the genus Eucera) and Xenoglossa are specialized oligolectic bees that depend entirely on the nutritious Cucurbita pollen grains and sweet nectar. These female squash and gourd bees readily seek out the oily pollen grains of cucurbit species in deserts of the Americas. It is known that male squash bees can be legitimate pollinators (Cane et al., 2011). It would be interesting to investigate the response of these native bees to Styrian oil pumpkins under drought stress. We might predict that squash bees would be equally or more deceived. Dukas (1987) found that native bees seemed to learn to avoid rewardless female flowers in the studied cucurbit over the course of the day but seemed to forget this avoidance the next morning. He suggested that solitary bees do not learn as well as social bees, and there is some (limited) independent evidence to support this.

Last year (2023) was the hottest year in the past 2000 yr based on a recent tree ring study (Esper et al., 2024). Hotter weather, along with overall climactic regimes, brings associated droughts around the world. These recurring droughts are certain to negatively affect the growth and production of agricultural crops and human welfare. Plants will suffer directly in terms of their reproduction and survival. The world's more than 350 000 flowering plant species will also produce less edible vegetation, flowers, nutritious fruits and seeds that a myriad of pollinators (c. 200 000 species), and wildlife species depend on for their own well-being and ultimate survival.

In summary, the key effects of drought highlighted are reduced flowering and floral displays, altered flowering phenology, changes in floral rewards such as nectar and pollen (and sometimes volatiles) that attract pollinators and ultimately impacts plant reproductive success through effects on pollen viability and seed or fruit production. The potential effects on pollen, seeds and fruits could be an important future direction for researchers investigating the effects of drought. Plant breeders, melittologists, farmers and conservation biologists should be concerned with Barman et al.'s findings. Perhaps, it will be possible to breed crop plants that are more drought-resistant, less susceptible to bud abortion and that produce adequate amounts of floral nectar, and bee-attracting headspace floral volatiles. In a warming and drying world, all humanity, plants and wildlife depend on one another.