An awful lot of research papers have been published on eyespots, but scientists still differ in their opinions about exactly how they function. They agree that eyespots intimidate or startle predators, but whether they do this by fooling them into thinking they are looking at one of their own predators – mimicking the eyes of big owl, for example – or whether the eyespots do this simply by being highly conspicuous patterns, which impact upon their visual system, is a moot point.
First of all, there are many different types of spots (and other shapes) on butterfly wings, with lots of different functions, including: avoiding being eaten by deflecting attacks; marks for sexual selection; warning patterns and colours; or as markings to improve camouflage. Genuine eyespots are highly conspicuous, circular (or semi-round) shapes, consisting of a series of concentric circles. The large, colourful eyespots on the ventral (under) side of a Blue Morpho’s wing (below), are classic examples. In this case, have they evolved to startle predators, or to deflect the beaks of pecking birds away from the body of the butterfly? I’m not sure. It’s possible that the large size of these eyespots make them highly intimidating, and eyespots to deflect pecks or blows from a predator, would probably be better located further towards the wing margins ( Kodandaramaiah et al., 2013)
So, there are two main theories about large central eyespots. The first, called the ‘eye mimicry-hypothesis’ contends that large, centrally located eyespots intimidate potential predators of the butterfly – birds and lizards for example – by looking very much like the eyes of the predators’ own enemies. This explanation applies in particular to the sort of ‘eye-like’ eyespots, that are often found on the dorsal (upper) sides of the hindwings, where they can be exposed suddenly, to startle or frighten a bird. The dorsal hindwing eye-spots on this Peacock Pansy (Junonia almana) butterfly (below), perfectly fit this categorisation. Notice how the large hindwing eyespots differ markedly from the circular roundels on the forewings (below), which look more like the sort of symbols used by the RAF on their aircraft wings! There are also two types of eyespots on the Morpho wings (above).
Another butterfly with a very convincing, ‘eye-like’ eyespot – this one on the ventral hindwing – is the Tufted Jungleking butterfly (top and below), a nymphalid (Morphinae) from Thailand. It’s almost as though it were painted by an artist! There is a black patch, a white stripe, a white crescent and lots of tiny white dots. These all create the illusion of a very real eye (at least to my way of seeing). It is very hard not to think that the butterfly has evolved an ‘eye-like’ eyespot via natural selection. But although these eyespots look to us like very convincing fake eyes, there is not very much definitive scientific evidence (Stevens, 2005), at least until fairly recently, to back this eye-mimicking theory. It may be that we are just very good at ‘making sense’ of patterns such as this?
The second, or alternative theory about eyespots – the ‘conspicuousness-hypothesis’ – has it that it is the conspicuous colouration of the pattern that induces the startle or avoidance behaviour, in whatever wants to eat the butterfly. In this theory, eyespots work because they are a novel or highly unexpected sight; a shock to the system perhaps and best avoided? Experiments by Professor Martin Stevens and colleagues (2007; 2008) showed that circles were no more effective than a range of other shapes – bars and squares – in preventing predation, by bird predators. These researchers concluded therefore, that the anti-predator function of eyespots was based purely on their conspicuousness. One might ask why evolution has gone to all the trouble of selecting such complex eye-like eyespots (as above) when a simple bar or roundel would do? And, this question was addressed in more recent research (below).
For perfectly round, roundel-like eyespots – the RAF insignia as I call them – it could be that they are ‘developmentally easier to produce’ (Stevens, 2008). These round, concentric eyespots can, it is thought, be formed by simple concentration gradients during the development of the wing. For example, the eyespots on the Peacock Pansy – and the other round eyespot on the Tufted Jungleking (above) – could be formed by colour-inducing chemical signals, originating from a point source inside the eyespot and then expanding outwards during development. The chemical signal for the outer ring is emitted first, and then the signal for the inner core ring is released; the gap between these signals can be expressed as a coloured ring, for example (see Otaki, 2011). In other words, it is not very difficult to make circular eyespots during the development of a butterfly in this way.
But we are still left trying to explain all the beautiful and highly complex, ‘eye-like’ eyespots we see on butterflies and moths. A butterfly whose eyespots were undoubtedly painted😊, or designed 🤣, by an artist of great genius – a veritable Picasso of the butterfly decorating world! – is the Peacock (below). Are these eyespots just highly conspicuous shapes? Or does the butterfly want us, as stand-in predators, to think of them as real eyes? It is difficult to be sure but they do work (!) and have been shown to effectively deter avian predators (Vallin et al., 2005, 2007).
So although the large eyespots of the dorsal forewings don’t look like very convincing eyes, we must remember they did not evolve to fool or startle us, but their own predators. We don’t really know how they appear to birds and lizards, but they may look for certain cues in the eyespot, that sends them the message that it is an eye. The eyespots on the dorsal surface of the hindwings (below) of the Peacock are rather different: there is a large white concentric circle around a black oval, within which are many tiny bluish-white scales.
Are the eyespots on the hindwings of the Peacock butterfly more eye-like, more convincing? It is difficult to say for certainty, but when the butterfly is viewed upside-down, as shown here on the UK Butterflies website, it really does look like an owl! Especially as the peacock butterfly eyespots are usually hidden when resting and only exposed by the butterfly when disturbed. Olofsson et al. (2012) found that chickens were fearful and ‘typically fled’ when peacock butterflies suddenly exposed their hindwing eyespots. EEKS! These researchers concluded that predators – well chickens at least! – really did think of these large eyespots as belonging to the eyes of a potential predator. Other experiments, using pied flycatchers reacting to on the wings of Peacock butterflies, also showed that these large eyespots act as a deterrent to feeding, and these researchers also preferred the suggestion that eye mimicry – rather than mere conspicuousness – explains their intimidating effect (Merilaita et al., 2011).
If eye-spots are functioning just as highly conspicuous shapes or patterns, it might be better to call them ‘wingspots’ as Stevens et al. (2007) suggest. But I am not sure. Why would evolution have gone to the trouble of producing such convincingly eye-like structures that require much more complex developmental pathways to produce them. Take the Owl butterfly for example (below). This huge eyespot on the ventral hindwing does contain concentric circles that could have developed by simple diffusion gradients, but it also contains a white hemisphere. These sort of elements are called ‘sparkles’: i.e. circular and crescent-shaped elements in the eyespot which are thought to create the illusion of a spherical eyeball, with the sort of sparkle that real eyes have! (Blut et al., 2012).
Blut et al (2012) found that these sparkle-shapes were very common and remarkably uniform: about half of all the eyespots they examined had a central, pinpoint-like ‘sparkle’ (like the forewing eyespot on the Peacock Pansy), 12% had a marginal, crescent-shaped ‘sparkle’ (like the Tufted Jungleking I think), 13% have a semi-circular ‘sparkle’ (like the Owl butterfly), and 22% have an intermediate semi-circular to crescent-shaped ‘sparkles’. They also showed – by artificially varying the position of the ‘sparkles’ – that the more natural-looking eyespots (to us as well as range of birds they tested), confirmed that ‘eyespots with a sparkle’ in a natural position have a stronger deterrent effect than those with a ‘sparkle’ in an unnatural position. They concluded that their ﬁndings supported the eye mimicry hypothesis better than the conspicuousness theory, but they could not be sure that improving the illusion of a three-dimensional object, might also have enhanced the conspicuousness of the eyespots! The same conundrum.
So they carried out another experiment (Blut &Lunau, 2015), comparing the effects of eyespots with the same level of conspicuousness (to birds), with eyespots that differed in terms of their similarity to the lens of an eye. The different eyespots had dots, bars and crescents in the middle. In this case, the birds were more deterred by the realistic, circular and crescent-shaped ‘sparkles’, than those with rectangular-shaped ‘sparkles’. The researchers also concluded that the shape and alignment of an eyespot’s ‘sparkle’ might be more important to birds than the overall colour pattern. So we are starting to see how eyespots which don’t look like perfect eyes us, might work perfectly well for other animals.
Incidentally, the eyespots on an Owl butterfly, make a very convincing owl, when viewed upside-down! (below).
Finally, in an experiment playing around with computer generated images like the one above, and with pictures of real owl’s eyes, both with and without eyes and eyespots manipulated, researchers (Da Bona et al., 2015) showed that mimetic eyespots were as good as real owl eyes in scaring Great tits (Parus major). Furthermore, they found that mimetic eyespots – the ones on the butterfly wings – were more effective in inducing an aversive response in the birds than the less mimetic, but equally contrasting eyespots. They concluded that ‘Predator mimicry, not conspicuousness, explains the efficacy of butterfly eyespots’.
Is this the end of the story? I doubt it. There is still an awful lot we don’t understand about butterflies and their beautiful wing patterns. It is also worth remember that in real life, eyespots will be seen in a wide range of lighting conditions, as he butterflies move and change positions. There’s a lot more to find out about these fascinating features.
In my next blog on eyespots, I will look at their deflective functions.
Blut, C., & Lunau, K. (2015). Effects of lepidopteran eyespot components on the deterrence of predatory birds. Behaviour, 152(11), 1481-1505.
Blut, C., Wilbrandt, J., Fels, D., Girgel, E. I., & Lunau, K. (2012). The ‘sparkle’in fake eyes–the protective effect of mimic eyespots in Lepidoptera. Entomologia experimentalis et applicata, 143(3), 231-244.
De Bona, S., Valkonen, J. K., López-Sepulcre, A., & Mappes, J. (2015). Predator mimicry, not conspicuousness, explains the efficacy of butterfly eyespots. Proc. R. Soc. B, 282(1806), 20150202.
Labandeira, C. C., Yang, Q., Santiago-Blay, J. A., Hotton, C. L., Monteiro, A., Wang, Y. J., … & Dilcher, D. L. (2016). The evolutionary convergence of mid-Mesozoic lacewings and Cenozoic butterflies. Proc. R. Soc. B, 283(1824), 20152893.
Kodandaramaiah, U., Lindenfors, P., & Tullberg, B. S. (2013). Deflective and intimidating eyespots: a comparative study of eyespot size and position in Junonia butterflies. Ecology and evolution, 3(13), 4518-4524.
Merilaita, S., Vallin, A., Kodandaramaiah, U., Dimitrova, M., Ruuskanen, S., & Laaksonen, T. (2011). Number of eyespots and their intimidating effect on naive predators in the peacock butterfly. Behavioral Ecology, 22(6), 1326-1331.
Olofsson, M., Løvlie, H., Tibblin, J., Jakobsson, S., & Wiklund, C. (2012). Eyespot display in the peacock butterfly triggers antipredator behaviors in naïve adult fowl. Behavioral Ecology, 24(1), 305-310.
Otaki, J. M. (2011). Artificially induced changes of butterfly wing colour patterns: dynamic signal interactions in eyespot development. Scientific reports, 1, 111.
Stevens, M. (2005). The role of eyespots as anti-predator mechanisms, principally demonstrated in the Lepidoptera. Biological Reviews, 80(4), 573-588.
Stevens, M., Hopkins, E., Hinde, W., Adcock, A., Connolly, Y., Troscianko, T., & Cuthill, I. C. (2007). Field experiments on the effectiveness of ‘eyespots’ as predator deterrents. Animal Behaviour, 74(5), 1215-1227.
Stevens, M., Hardman, C. J., & Stubbins, C. L. (2008). Conspicuousness, not eye mimicry, makes “eyespots” effective antipredator signals. Behavioral Ecology, 19(3), 525-531.
Vallin, A., Jakobsson, S., Lind, J. & Wiklund, C. (2005). Prey survival by predator intimidation: an experimental study of peacock butterﬂy defence against blue tits. — Proc. Roy. Soc. Lond. B: Biol. Sci. 272: 1203-1207.
Vallin, A., Jakobsson, S. & Wiklund, C. (2007). An eye for an eye? On the generality of the intimidating quality ofeyespots in a butterﬂy and a hawkmoth. —Behav. Ecol. Sociobiol. 61: 1419-1424.
I am an entomologist with a background in quarantine pests and invasive invertebrates. I studied zoology at Imperial College (University of London) and did a PhD on the population dynamics of a cereal aphid (Metopolophium dirhodum) in the UK. I spent 5 years with the British Antarctic Survey studing cold hardiness of Antarctic invertebates and 17 years with the Food and Environment Research Agency. My main interests now are natural history, photography, painting and bird watching.