The Loom

The Semiotics of a Leaf

A new autumn has brought another burst of red and yellow leaves. And it has also brought an interesting new idea about why trees put on this show every year.

In recent years, scientists have been roughly divided into two camps when it comes to autumn leaves. One camp holds that autumn colors are just part of preparations for winter. The other holds that the colors are a warning to insects to stay away.

The warning hypothesis came from the late biologist William Hamilton. He pointed out that trees fight off insect larvae with toxins. A more vigorous tree could produce more toxins than a weaker one. It could also produce more vibrant colors in the fall by producing more pigment molecules. (The red in leaves is created by molecules called anthocyanins, for example). Perhaps a tree could send a message to insects looking for a place to lay their eggs: stay away from me or I’ll kill your kids in the spring.

Insects that could recognize this signal benefited by staying away, laying their eggs on weaker trees. And brighter leaves were in turn favored, since trees that produced them didn’t have to struggle so much with insects devouring their equipment for gathering sunlight. Hamilton and other scientists put his warning hypothesis to the tests, and they found some hints that he was right. The color of leaves, for example, did seem to influence where insects chose to lay their eggs.

But scientists who study the physiology of trees weren’t sold. They came at the question of autumn colors from another direction: they studied the molecules that create the colors. During autumn, leaves dismantle their green chlorophyll and pump the nutrients back into their branches. It’s a vulnerable time in a tree’s life, because it becomes vulnerable to damage from ultraviolet rays. Anthocyanins and other pigment molecules protect leaves by absorbing the damage caused by UV rays, so that the dwindling chlorphyll in a leaf can continue to generate the energy required for winter preparations. (For more details, see my article last fall in the Times.)

Everyone in the debate that I spoke to agreed that there was a lot left to learn about autumn leaves, and as I mentined here, it was even possible that they might be the result of several different forces working together. And in a paper in press at the journal Bioessays, University of Frieburg biologist H. Martin Schaefer and his student Gregor Rolshausen have come up with a way to account for a lot of the evidence offered up by both camps. The core of their argument is that autumn leaves are cues, not signals.

This may all sound more like semiotics than botany, but there’s some real science here. Schaefer and Rolhausen point out that the molecules that give leaves their fall colors and the molecules that defend leaves against insects have a lot in common. In order to make some toxins, a tree actually starts out using the same recipe for making anthocyanins. Only relatively late in the process does it alter the recipe to produce toxins (which are colorless, unlike anthocyaninins). Other toxins are made with a completely different recipe, but they have another link to anthocyanins. These molecules are very delicate, and are easily damaged by sunlight, and they need anthocyanins to protect them.

Schaefer and Rolshausen argue that the primary function of anthocyanins is not to ward off insects. Its main job is to shield leaves from radiation. It just so happens that a tree with a lot of colorful anthocyanin is probably also a tree with a lot of colorless toxins. One reason is that a tree that is physically capable of producing lots of anthocyanins can also produce lots of toxins that use an almost identical recipe. Another reason is that a lot of anthocyanins can protect other toxins, so that they can attack insects.

Insects can’t see toxins, but they can see reds and yellows produced in autumn leaves. And anthocyanins may be reliable cues about how dangerous a tree will be to an insect’s eggs. While it’s a cue, Schaefer and Rolshausen argue, it’s not a signal that’s directed to the bugs.

By linking autumn colors to insect defense, Schaefer and Rolshausen can account for a lot of the evidence amassed so far. But their new hypothesis won’t win the day until more experiments can find more support for it than for the previous ones. It may even turn out that other forces are also at play. I suspect that many more autumn leaves will fall to the ground before scientists have worked them all out.

Update, 12/2/05: The paper is now out.


  1. #1 Andrew Brown
    October 31, 2005

    Carl, we’re getting into very deep water here. Does a signal require an attempt to communicate, or is it something constructed in the behaviour of the beholder? Dennett, I think, would plump unequivocaly for the latter. In this, as in most things, he is merely making explicit assumptions that a lot of scientists hold. Signals, or informaiton more generally, are not properties of things in themselves, but of their interactions with the world.

    I suppose the test, in evolutionary terms, would be to see whether the colouring of leaves was maintained only if it functioned as a signal: in other words, if it didn’t protect against UV light as well. But since protection against UV light also increases the amount of some toxins, I don’t see how the experiment could, even in theory, be done. There’s a double effect that can’t be disentangled.

  2. #2 Peter Ellis
    October 31, 2005

    The obvious approach, it seems to me, would be to study the sequence of anthocyanin-producing genes and the toxin-producing genes in a range of different species.

    That might tell you whether the colours came first or the toxins came first, which would in turn be a clue to their function. The scenario described above looks very like one where a biological pathway producing pigment molecules for defence against UV was later extended to generate toxins

    Heck, you could even look at the evolution of insect eye pigment genes. Maybe insects see red because tree leaves are red, not vice versa.

  3. #3 Bruce Amiata
    October 31, 2005

    Is it not also the case that most insects (the polinating ones, anyway) detect light shifted towards the blue, rather than the red? (Bird polinators, generally, are attracted to the red end of the spectrum). I wonder how much discrimination insects have from green through the yellow and red portion of the spectrum…

  4. #4 Perry
    October 31, 2005

    My understanding of colour change was given to me in 1960 by my Botany teachers. Put simply, abscission cells at the base of each leaf slowly grow if a leaf uses more sugar than it creates. As the green chlorophyll deteriotes the underlying colour become more evident until the leaves dry and drop. The rationale then, was that the leaves would be a liability as they would increase the “top hamper” effect during the winter gales.

    “Perhaps a tree could send a message to insects looking for a place to lay their eggs: stay away from me or I’ll kill your kids in the spring.”

    How would that work? The leaves would have died and be well on their way to become leaf mould.

    This is an example of over complication. Dare I suggest that the simple explanation is probably the more correct, however, you chaps are very fortunate to have such beautiful colours at this time of the year. In NW London, our trees just seem to go yellow and drop. No maples!

  5. #5 David Winter
    October 31, 2005


    I’ve enjoyed all your stories about the autumn colours, thanks for bringing such a cool topic to the attention of your readers.

    If one of the co-evolutionary theories is actually going on it would be cool to find trees that cheat. That is, trees that don’t produce toxins or use colours for ‘honest signalling’ like Hamilton’s idea but freeload on the effort put in by other tress doing those things. Once aphids avoid red leaves you might get away with just producing red leaves to reduce your parasite load.

  6. #6 Patrick Foss
    November 2, 2005

    Does anyone know if the colorless toxins are also odorless? Saying that the color alone deterred the insects may lead to an erroneous assuption that the color was responsible when in fact it may indeed be incidental and that the odor or taste of the toxins are apparent to an insect when it first lands on the tree.

  7. #7 Gene Norman
    November 2, 2005

    I’m don’t study these thing scientifically, but the aborists/foresters here in Wisconsin tell us that a tree which colors early, or much brighter than similar surrounding trees is under some kind of stress, including drought and insect attacks. Can you: a. dispute/corroborate this and/or
    b. include it in your theory?

  8. #8 Rob Bate
    November 6, 2005

    The summer color of trees is a combination of chlorophyll, which is a deep blueish green, carotene, a yellow orange as in carrots and anthocyanin. the red described above. The rich blue/green coupled with the yellow orange and red mix into the relatively neutral yellow green common to most midsummer leaves.

    I assumed that when the chlorophyll was reabsorbed or broke down the yellows, oranges and reds were left as a beautiful, but essentially functionless, byproduct. Has any purposeful use of fall color been established?

  9. #9 mathew
    January 8, 2006

    am a final year botany student. i wish to get some information regarding poilnators in birds. this is for my thesis work. kindly help me if possible.

New comments have been disabled.