A Butterfly's Flash Coloration Distracts Predators—Read Future Textbook Knowledge in Ethology

Abstract

A highlight of visiting a tropical rainforest is watching iridescent butterflies passing by, flashing their colours in the sunlight that makes it to the forest floor. It can be quite tricky to keep track of such a butterfly because typically only the upper side of the wings is iridescent, so that the colour only flashes when this side of a wing can be seen. People think that this kind of moving flash coloration makes it difficult for visual predators to follow the flight path of the butterfly. Hence, iridescent colours may help butterflies to distract predators. Computer simulations and experiments with humans as ‘predators’ suggest that this may be true, but as of now the flash colouration hypothesis had not been tested for real.

In this issue of Ethology, Vieira-Silva et al. (2024) conducted a series of clever experiments to finally test the flash colouration hypothesis in Morpho helena butterflies. This species displays an iridescent blue colour on the upper side of its wings (as can be seen on this issues' cover image) that flashes when they move through the forest.

In a first experiment, the authors painted the cryptic underside of the wings with a colour mimicking the iridescent blue of the upper side of the wing. Thereby, the butterflies become more constantly visible during flight, because now the blue colour is exposed all of the time. The underside of the wings of a control group got painted with a brown colour similar to the original cryptic colour, thereby controlling for the effect of catching and painting the butterflies. In a capture–recapture analysis, Vieira-Silva et al. (2024) found that blue-coloured butterflies were less likely to be seen again compared to the brown-coloured control individuals. Because Morpho helena butterflies do not move around far, the most parsimonious explanation for the lower resighting rates of the blue-coloured butterflies is that a higher proportion of them got eaten by predators.

In a second experiment, Vieira-Silva et al. (2024) tested if an overall cryptic colour as such would have reduced predation. To test this, they coloured the upper side of the wing—which normally has the blue flashing colouration—with brown colour, so that the butterflies become completely cryptic during flight. This treatment, however, did not affect recapture rates, suggesting that completely cryptic butterflies did not have an advantage over individuals that flashed their blue colour during flight. Hence, the distraction effect of a flashing blue colour likely has a similar effect than complete crypsis.

To show that the higher predation of butterflies with an underside coloured in blue was really due to moving butterflies and not because such butterflies became generally more visible to predators, the authors conducted a third experiment. To test if blue colouration affected predation in non-moving butterflies, they compared how likely dead mounted butterflies got attacked from predators when the underside of their wings was either coloured with a cryptic brown colour—the controls mimicking the natural situation—or with a blue colour, similar to Experiment 1. There was no difference in attacks between control and blue-coloured butterfly mounts, suggesting that non-moving blue-coloured butterflies did not attract more predators than non-moving cryptic butterflies. With this third experiment, Vieira-Silva et al. (2024) could demonstrate that it was really the movement aspect of experiment 1 that caused higher predation rather than just the more conspicuous colour as such.

At Ethology, we attempt to publish high-quality behavioural research. Some of the research published in our journal has become textbook knowledge or citation classics. I believe this paper by Vieira-Silva et al. (2024) has the potential to become a classic citation for the flash colouration hypothesis on predator distraction. The experimental design of this study is as simple as it is clever. In times of ever more complex statistical analyses, this study still excels with the quite simple statistical approach of a chi-square test. Congratulations and chapeau to the authors!