As they feed, caterpillars can trigger a biochemical response in plants that, within as little as 30 minutes, deters further feeding by any caterpillar. The effect lasts for days, weeks or even years. This so called induced response takes the form of an increase in chemicals such as digestion-inhibiting phenolics, alkaloids or proteinase inhibitors. Such chemical changes affect how caterpillars forage, forcing them to move farther or more frequently. They also slow the growth of the caterpillars so that they are likely to be eaten by predators before they can affect the plant too severely. Or so the story goes.

For a decade or more, groups from the universities of York, Southampton and Tuku in Finland have tried to find out whether phenolics are induced defenses against caterpillars feeding on birch. They discovered that if trees are damaged by caterpillars, there is a large increase in phenolics, in both damaged and adjacent undamaged leaves. Intriguingly, natural damage from caterpillars has a much greater effect than artificial damage caused by punching holes in leaves.

Some groups question the idea of induced defenses. Measuring the amounts of phenolics precisely is difficult, so some considered changes in the activity of an enzyme called Phenylalanine Ammonia Lyase (PAL). This enzyme catalyzes the first irreversible step in the synthesis of phenolics. She discovered that PAL responded more to damage by caterpillars than to artificial damage. Even more interesting, artificially damaged leaves painted with the saliva of caterpillars showed a greater response than leaves left untainted.

However these chemical changes are small compared with natural variations in phenolics, both among different trees and among different leaves on the same tree. Sometimes, undamaged trees still have more phenolics than damaged trees. Another unanswered question is: do the caterpillars take any notice of the phenolics? When some tests were done on caterpillar’s species, it was found that the most caterpillars were indifferent to damage. In fact, some even preferred the damaged leaves. And there was no simple relationship between feeding decisions and phenolic chemistry.

Some thought that the caterpillars might be responding to something else entirely. They tested this by spraying leaves with an inhibitor of PAL, to block the synthesis of phenolics. Caterpillars of two species responded identically to damage, whether or not phenolic synthesis was blocked. To prove the induced defense theory, researchers will have to go further and link it to insect population dynamics. They may find that the chemical response has evolved to counter something else,. Indeed the relative effects on enzyme activity of damage by caterpillars and by artificial means are identical to those of wounding with or without fungal attack.