Interacting natural enemies: Impact on the host plant

Background/Question/Methods

Every plant is host to a variety of other organisms, many of which are natural enemies of the plant. Natural enemies that share a host plant may interact, even when they are separated in space and/or time. The presence of one natural enemy may have positive or negative effects on other species that occupy the host plant. In biological weed control two or more species of natural enemies are often introduced, and it is important to know the impact of multiple natural enemies on the host plant. Of particular interest is whether these effects are independent and could be predicted from the impact of these natural enemies alone. As competition between insects sharing a host plant has been found to be more prevalent than either facilitative or neutral interactions, we hypothesised that, on average, two insects species (or other natural enemies) would have less impact on host plant performance than would be predicted based on the effects of each natural enemy alone.

We conducted a literature review and meta-analysis to determine whether natural enemies operated independently or if synergistic or antagonistic effects on plant performance were common. We considered studies where two insects, plant pathogens or a combination (potential weed biocontrol agents) were tested in a factorial design, and where plant performance parameters were recorded. We counted the proportion of plant performance parameters in each study where evidence for non-independence among the natural enemies was found. We used the log-response ratios (a meta-analysis statistic) to determine how prevalent non-independent plant performance responses were when two natural enemies occurred on the plant.

Results/Conclusions

On average, the damage caused by one natural enemy was independent of that caused by the second natural enemy which was not as we predicted. The presence of one natural enemy did not promote or inhibit the decrease in plant performance caused by the second enemy. However, cases in which plant performance was less influenced than predicted under the assumption of independence (antagonism) were more common than cases where the decline in plant performance was accelerated (synergism). We did not find any combinations of potential weed biocontrol agents (or traits of those agents) that were more likely to cause non-independent effects, either antagonistic or synergistic.