Plants can go to extraordinary lengths to tempt pollinators to do their bidding. From donning female insect disguises (complete with pheromones!) to lure lustful males, oozing the stench of rotting flesh to temp hungry flies, or, most commonly, offering bribes of sweet rewards for any who visits – no lengths are too gross or pricy to spread their precious pollen.

Most pollinators visit multiple plant species, which generally isn't a problem when there are plenty of bees, flies, and other pollen-dusted tourists to go around. But research now reveals that when pollinators become scarce these lengths could include choosing what passes for violence among plants.

Ecological theory predicts that competition due to declines in pollinators could drive plants to even more varied strategies of interactions and habitat use to allure pollinators, which would increase plant diversity.

But another theory counters that as the number of pollinators declines, so do the chances that the same pollinator will visit the same type of rarer plants, meaning more common species will outcompete them, and that would reduce biodiversity.

As this is a scenario the world is heading towards – in the face of worrying insect declines and disease impacting our closest pollinating ally, the European honey bee – a team of researchers led by Princeton University ecologist Christopher Johnson put these competing theories to the test in field experiments.

Using 80, 2.25 m2 plots of paired annual plant species native to Switzerland, the researchers controlled how much pollination happened in some plots by hand. The rest had to rely on the normal environmental levels of pollination. Johnson and their team compared population and fitness measurements for each.

All five plant species used – field mustard (Sinapis arvensis), corn gromwell (Buglossoides arvensis), the common poppy (Papaver rhoeas), cornflower (Centaurea cyanus), and wild fennel (Nigella arvensis) – relied on general insect pollination, but corn gromwell could also self pollinate.

Another 22 replicate plots were enclosed, with half exposed only to one pollinator species and the rest to background environmental levels to simulate pollinator decline. For these plots, the researchers also measured floral visits by pollinators.

"For nine of ten species pairs, competition for pollinators weakened stabilizing niche differences between competitors," Johnson and colleagues write in their paper, meaning the paired species did not strike a new balance of interactions with each other within the shared pockets of the environment when pollination was reduced.

"These results support the hypothesis that pollinators destabilize plant competition by favoring more common plant species at the expense of their rarer competitors," the team concludes.

It seems when our incredible assortment of insect pollinators becomes scarce, each plant is in it for itself, grabbing at all the remaining pollinators' attention at the expense of their neighbors.

The ability to breed was three times greater for the common poppy, wild fennel, and cornflower when hand-pollinated than those relying on background pollination, showing these species are naturally self-limiting in order to strike balance with their neighbors.

But these plants are only that considerate to a point: The team identified weakening of this within-species competition as a major driver for destabilizing the multi-species communities.

The result created competition imbalances and exacerbated the average fitness differences between plant species: It reduced all pairs of plant species' ability to coexist with each other.

This could lead to the most common groups of plants exiling the rarer ones from their once thriving communities.

The researchers did not expect that this would occur so uniformly across all pairs of species. Since the study was done across a small area, and a short timescale, the researchers say that over time a new coexistence equilibrium could be established with some of the species.

Johnson and colleagues did however check to see if the pairwise interaction they measured differed when the plants were grown in more specious communities (three, four, and five species) and found that overall they did not, further supporting their conclusions.

These are concerning findings if they hold true over wider scopes. If plant communities do become so easily unstable as pollinators decline, broader interactions between species and their implications will be difficult to predict. At least some understanding of these will be crucial in our attempts to maintain as much biodiversity as we can as conditions worsen.

Meanwhile, we can all help reduce further insect declines by growing native plants, reverting lawns to natural habitats, avoiding pesticides, herbicides, and other toxic chemicals, limiting outside lighting, and supporting groups and leaders who actually take these tiny but critical cogs of our living world into consideration.

This research was published in Nature.