Home » Flowers play a key role in flight control for bumblebees

Flowers play a key role in flight control for bumblebees

A bee can approach a flower from any direction, but markings on the flower help guide bees in the most effective way, like a natural air traffic control.

You can listen to this page as an audio file.

Some flowers can have patterns on their petals. These can help pollinators recognise favoured species when out foraging, but do they also aid pollinators in pointing the way to the reward as nectar guides? Robin Richter and colleagues have examined how patterns on flowers help direct bumblebees in their search for nectar. Surprisingly, they did not find that nectar guides helped direct the bees when they landed. However, they found the guides helped as a kind of ‘air traffic control’ improving approach, landing and even departure for the bees. The effect can reduce flower handling time by up to 30%.

The scientists got their results from a combination of real bumblebees and artificial flowers. All the artificial flowers were the same size and shape, meaning that the critical difference between the experiments was the patterns on the flowers. In the first set of experiments, the scientists released bumblebees to fly towards flowers that were either plain yellow or blue, flowers with a line marking across the diameter of the flower, and flowers with a cross marker across the flower.

In the second set of experiments, the team used crosses to see what sort of guide performed best, a cross across the whole flower, a cross just on the inner part of the flower, or a cross that reached from the rim to halfway into the flower.

A white flower viewed from above, so that it looks like it's in plan. There are six large petals, with three holding a yellow stain from around halfway along the petal to the centre of the flower. Above where the six petals meet are three smaller petals. These have purple marks from the edge of the petal to the centre. The purple marks on the petals above align with the yellow marks on the petals below, so the impression is of a large Y shape, with arms at angles of 120 degrees.
Y marks the spot for the Fortnight Lily, Dietes grandiflora. Image: Canva.

Using video, Richter and colleagues found that the bees landed faster when they could see a pattern. Also, the lines and crosses changed where the bees landed. For the plain flowers, the bees landed anywhere but with a preference for the rim. For the patterns, the bees landed on the pattern.

The experiments with the different crosses showed how the bees used the patterns to land. The bees used them as a guide if the patterns reached the rim. If there was only a small inner cross on the flower, the bees landed wherever took their fancy. The scientists checked research on the bees’ vision and found that they could still see the inner cross pattern. It wasn’t too small. They simply didn’t care about it.

Once the bees landed, they walked straight to the nectary for their reward, except in one case. If there was an outer cross pattern, the bees followed it till it disappeared. Then they were a little confused. This result shows that the bees can perceive the pattern when they’re on the flower, and the patterns could have an effect.

This result contradicts earlier results that show patterns can guide how bumblebees walk across the flower, but Richter and colleagues think this might be down to how the bees were trained in earlier experiments. Commenting on earlier work by Leonard and Papau, they write.

They showed that foraging experience drastically reduces walking time to the nectary in both flower types, with and without patterns. They furthermore showed that the influence of patterns on the approach time depended on which patterns had been rewarded during training.

Bees that were initially rewarded on un-patterned flowers did not decrease their approach time with patterned flowers. The opposite was the case for the reversed order of presentation. Without patterns during learning, bees might thus learn the nectary configuration relative to the flowers’ shape, but if patterns are present, bees might learn the nectary position relative to the patterns, and thus require longer to locate the nectary without them. In our study, the foragers were extensively trained to pattern-less grey flowers, and likely learnt the nectary configuration. This would explain that the approach walking times did not differ between the patterned and un-patterned flowers, as they might have mainly relied on their knowledge about the flower configuration.

Richter et al. 2023

The patterns also affected departure. You might expect that, once rewarded, the bees would take-off immediately. Instead, they wander for a little exploration. It appears from the heat maps Richter and colleagues made of the bees’ rambles that if there’s no pattern in the centre of the flower, they hang about there a lot longer.

Of the patterns, the full cross most effectively guides bees away from the centre but not along the arms of the pattern. The bees wander away from the centre before take-off, but in any direction. The inner cross pattern has a similar effect but, to my eyes, it looks like they might not wander so far from the centre before take-off. Oddly, for the line pattern, they do seem to tend to stay closer to the lines.

While the results are interesting, do they have any value? If the flowers are artificial, do they reflect what the bees encounter in the wild? The authors admit that this isn’t certain, as there hasn’t been a lot of work quantifying the patterns bees see in the wild. Their results indicate at least a reason for studying and classifying patterns. Another issue is whether their flowers feel like the real thing, and here they are more confident.

Importantly, our results demonstrate that even for experienced foragers who have learned the position of the nectary on the flower, flower patterns are important to guide their approach flight and landing on the flower, thus significantly reducing handling time. Since approach guidance and landing control cannot be supported by other sensory modalities (except for mechanosensation in the legs and antennae upon contact with the flower (Reber et al. 2016)), we expect this effect also to be effective in a natural setting. Indeed, we expect approach guidance and landing control to be of even higher importance than in the laboratory, as flowers in nature are encountered at different heights, tilt angles, and vary in size and shape…, making it harder to transfer learned approach-and-landing strategies between flowers, and therefore relying more on sensory inputs.

Richter et al. 2023

Richter and colleagues suggest that recently-developed imaging techniques open up new avenues of research and close with a couple of interesting new questions to examine. One is whether plants can have deceptive nectar guides. If they do, the obvious follow-up question would be Why? The other question they have is about ornamental plants. These are often bred specifically to develop new patterns. So what effect do these human-attracting patterns have on pollinators? Are they causing havoc for pollinators, or do their striking designs aid attraction? It also raises the possibility of breeding specifically more pollinator-friendly plants with more helpful nectar guides.


Richter, R., Dietz, A., Foster, J., Spaethe, J. and Stöckl, A. (2023) “Flower patterns improve foraging efficiency in bumblebees by guiding approach flight and landing,” Functional Ecology. Available at: https://doi.org/10.1111/1365-2435.14262.

Alun Salt

Alun (he/him) is the Producer for Botany One. It's his job to keep the server running. He's not a botanist, but started running into them on a regular basis while working on writing modules for an Interdisciplinary Science course and, later, helping teach mathematics to Biologists. His degrees are in archaeology and ancient history.

Read this in your language

The Week in Botany

On Monday mornings we send out a newsletter of the links that have been catching the attention of our readers on Twitter and beyond. You can sign up to receive it below.

@BotanyOne on Mastodon

Loading Mastodon feed...