Mimicry: polymorphism for camouflage in a caterpillar

This finding was published in 1989 by Erick Greene (reference and link below), who sent me these photos, but it’s such an astonishing case of mimicry that I showed it to my students last week.

It involves the caterpillar Nemoria arizonaria, the juvenile stage of a moth that lives in Arizona, New Mexico, northern Mexico, Texas and California.  It has two generations per year, one in the late winter/early spring and the other in the summer. In both cases the caterpillars, after hatching, live on oak trees and eat parts of them.

If the caterpillars hatch in the winter or early spring, they feed on oak “catkins” (flowers), and, sure enough, their bodies take on the appearance of a catkin, almost certainly to hide them from visual predators like birds. Here’s a “catkin morph” (to the right) next to some real catkins.

When the summer brood hatches, however, the catkins are long gone, and the caterpillars feed on the only food available: oak leaves.  This generation looks not like flowers, but like oak twigs:

It’s camouflage again, but a different type.  And it’s obviously adaptive, if you have several broods per year, to evolve an appearance that matches the environment in which you hatch. There are also differences not just in appearance, but in their heads and jaws: catkin morphs have smaller jaws suitable for eating the pollen grains, while twig morphs have larger mouthparts and jaws to nom the tougher leaves.  Finally, they differ in their behavior:  if you put the catkin morph on a twig, it moves back to the flowers, but the twig morph does the opposite.

The interesting thing about these two morphs is that they are genetically identical: a caterpillar of this species has genes that can make it look either like an oak flower, or like an oak twig.  Within its genome are two distinct developmental programs coding for its appearance, and which program is activated depends on the season (this temporally varying appearance of a single species is called a developmental polymorphism or a polyphenism).  How does the caterpillar know which set of genes to turn on, and when?

The two obvious environmental cues are photoperiod (which differs between winter/spring and summer) and diet.  Greene captured moths in the field and reared them on different diet and photoperiod regimes. It turned out that the only factor affecting appearance was diet: caterpillars reared on catkin diets assumed the catkin appearance; those raised on leaves turned into twig morphs.  Greene hypothesized that the critical chemical difference involved tannins (polyphenols), which are high in leaves and low in catkins.  Sure enough, caterpillars raised on artificial diets supplemented with polyphenols developed into twig morphs, even when they were also fed catkins.

The evolutionary advantage of producing two broods per year is obvious.  Although catkins seem to be a superior diet, they’re available only once a year during the short flowering period.  Any catkin morph that developed into a moth who was also able to produce twig morphs would leave many more copies of its genes than would a moth constrained to reproduce only once per year.

The precise evolutionary sequence of change, however, is unknown, since all we have is the endproducts.  Developmental polymorphisms are not unique to this species—they’re also found in aphids, rotifers, water striders and, of course, the social insects, where every female has the genes for becoming either a queen or a worker.

Despite our ignorance of the evolutionary path, the precision of the mimicry (to use a Stangroomism, look at them) tells us that when the proper mutations are available, natural selection can make an animal look almost identical to its background.  In this case we know the “targets” of selection: the appearance of a flower and a twig. And in both cases natural selection gets it spot on.  This precision also tells us that the predators—certainly birds—are sharp sighted. If they couldn’t see all that well, there would be no selective advantage to such a precise resemblance.  But we all know that birds have keen sight!

By the way, both forms of the caterpillar turn into this lovely geometrid moth, which itself seems to be a leaf mimic:


Greene, E. 1989. A diet-induced developmental polymorphism in a caterpillar.  Science 243:643-646.

18 thoughts on “Mimicry: polymorphism for camouflage in a caterpillar

  1. “How does the caterpillar know which set of genes to turn on, and when?”

    Isn’t God amazing? (ducks)

    Posts like this make me wish I had taken the population-biology track rather than the biochemistry path. Really, really cool stuff. Thanks.

    1. But the biochemistry – the tannins eaten by the caterpillar – is the cue… surely not too late for you to polymorphise into population biology?!

      Fascinating insect. On of the things that amazes me is those creatures – often parasites – that have a life cycle going through several other animals before ending up as sexually mature forms. Looks-wise though, this moth takes the cake!

      1. It’s true; I don’t run screaming at the mention of polyphenols. And if I had taken the populational track, I still would have graduated 3 years before this particular finding was published. Damn.

  2. Astounding! Life is awesome!

    How can one not be fascinated with coming to know how this happens and marvel at our ability to figure it out through dedication and exacting methodology?

    Why don’t people who short-circuit this ability with pseudo-answers of oogity boogity intervention and intentional poof-making design appreciate just deeply they are shortchanging themselves and blurring their understanding of the world they inhabit?

  3. So, the moths have some amazing adaptations to avoid predation from birds.

    Are there any indications of the trees adapting to avoid predation from the moths? Are the catkins varying in appearance? Are the tannins varying in ways that might mess up the caterpillar’s polymorphism mechanisms?

    Or are the trees using non-visual anti-caterpillar defense mechanisms, or are the caterpillars not that much of a threat, or do they provide some offsetting benefit, or…?



    1. Or how about the birds developing better prey detection strategies. Heat seeking vision? Better hearing to catch the munching sounds?

      Answers always lead to more questions!

      1. But nature can only work with what is available. It is a war – the struggle for survival. Tannins seem to have a function in protecting plants from some micro-organisms but oaks produce plenty yet get eaten by plenty of creatures. Some friendly botanist can maybe tell us?

        By the way, well done prof Coyne getting the word ‘nom’ in!

      2. Yes – more questions! I wonder if the moths need to lay as many eggs as other types of moth to get the same number of successful adult offspring?

  4. The precise evolutionary sequence of change, however, is unknown, since all we have is the endproducts.

    Aha! Aha! You admit it. It’s unknown; that means goddidit.

    Alert the school boards!

  5. Fascinating. Raises so many more questions! I wonder how long the genetic code for catkin mimicry would last if something threw off the cycle–for instance a catkin fungus that killed off the caterpillars that ate them. It seems that in any given genome there could be codes for all kinds of things whose trigger has disappeared.

    I’m also curious about the moth. In a lot of insects the leaf mimicry of their wings seems to be a combination of two things: the coloration and the veining. I wonder if the veining is primarily determined by mimicry or the fact that distributing resource to the leaf or the wing is efficiently accomplished by the same branching structure.

  6. I love this phenomenon, and so does my advisor. He’s raised lots of Nemoria from Arizona, not sure if they were this exact species, but he’s talked about this example plenty of times. Pretty neat!

  7. I like the summary of the sequence of investigation: What triggers the difference? Could be photoperiod or diet. Check both: it’s the diet. What about the diet? Look for differences: could be the polyphenols. Check by controlling polyphenols: it works! And now we know how to make the caterpillar grow fluffy or not.

  8. When I read stories like these I wonder why everyone doesn’t want to be an evolutionary biologist.

Comments are closed.