Toward the end of cartoonist Gary Larson's There's a Hair in My Dirt (1998), the hapless protagonist, Harriet, is momentarily frightened by “a pair of large, ominous eyes staring back at her. Oh, I recognize you now, Mr. Owl!' she laughed.” Larson's comment: “Ha! Did Harriet ever get taken in by one of the oldest tricks in Nature's book –the old I'm-a-scary-creature-with-giant-eyeballs gag. You see, Mr. Owl' was really a Royal moth, an insect that uses its large wingspots to mimic a much more frightening animal.” Though widely accepted in common lore, how well do eyespots actually protect real Lepidoptera? Save a few older studies,notably one by Blest in 1957, remarkably little is known.

Enter Adrian Vallin and colleagues, with an elegant demonstration of the eyespot fear factor. They studied whether eyespots defend peacock butterflies(Inachis io) against foraging blue tits (Parus caeruleus). Peacock butterflies at rest depend on crypsis – holding their wings folded above the body, so that the ventral sides, which mimic leaves, show. But crypsis can fail, and when it does the butterfly opens its wings to reveal ventral eyespots embedded in a background of bright colors. Simultaneously, it emits a hissing noise and two clicks, manufactured by rubbing fore- and hindwing veins together.

Vallin and colleagues analyzed the anti-tit effectiveness of eyespots and hissing alone and in combination, using an old and effective set of tools:scissors and a black permanent marker. They created eyespotless butterflies by coloring over the four eyespots and created controls by blacking out other wing areas. To create non-hissing butterflies, the authors cut off the lower margin of the forewings, and they removed the lower margin of the hindwings to create controls. On a final set of butterflies, the authors removed both devices and constructed appropriate control individuals.

The authors placed each butterfly, experimental or control, in a sparsely furnished room (willow log for the butterfly, perch for the bird). They set room temperature low enough to prevent butterfly flight but not so low that they were immobilized. Blue tits invariably found their intended victims, but whether or not they killed them depended almost entirely on whether or not the butterfly had eyespots – indeed, combining groups, only one of 34 butterflies with eyespots was killed, whereas 13 of 20 with covered eyespots were killed. Butterflies waited until birds were close, or even touching them,before flashing their wings. Those with intact eyespots were able to scare birds away repeatedly. Surprisingly, sound production had no discernible additional effect.

So it's true: the flash-some-scary-eyeballs gag really seems to work. But it would be premature to announce case closed. Why do peacock butterflies produce sound? Perhaps it magnifies the effects of eyespots in other more natural contexts or against other attacking predators. Which and who? Also,why are blue tits so easily conned? Fright-and-flight makes sense the first time a blue tit is flashed, but shouldn't the bird rapidly see through the ploy? Perhaps the bird-brain circuitry engaged in procuring the next meal is completely overruled by a fright reflex stemming from one of the most important and ancient of rules: eyeballs portend gullets. Butterfly eyespots only pretend gullets, but that's enough.

Vallin, A., Jakobsson, S., Lind, J. and Wiklund, C.(
2005
). Prey survival by predator intimidation: an experimental study of peacock butterfly defence against blue tits.
Proc. Roy. Soc. B
272
,
1203
-1207.