Author
 |
Swope, Sarah |
|
PARKER, INGRID - University Of California |
|
Submitted to: Ecology
Publication Type: Peer Reviewed Journal Publication Acceptance Date: 12/4/2009 Publication Date: 8/1/2010 Citation: Swope, S.M., Parker, I.M. 2010. Trait-mediated interactions and lifetime fitness of the invasive plant Centaurea solstitialis. Ecology. 91(8):2284-2293. Interpretive Summary: Plants interact with both enemies (e.g., herbivores and pathogens) and mutualists (e.g., pollinators) simultaneously and sequentially. Such multispecies interactions can give rise to indirect effects that produce nonadditive outcomes. A common example of indirect effects is that of pathogen infection on pollinator visits. In such cases, pathogen infection has the direct effect of reducing plant size which in turn has the indirect effect of making the plant less conspicuous to its pollinators, so it receives fewer visits and sets fewer seeds. Such indirect effects are likely to be common in nature but are also difficult to study and their nonadditive nature makes the outcome inherently difficult to predict. But understanding multispecies interactions is important in the use of biological control agents to control invasive plants because modern approaches to biocontrol rely on releasing multiple agents for each target weed. Centaurea solstitialis is one of the most problematic invasive weeds in California and the weevil Eustenopus villosus is its dominant biological control agent. The weevil attacks that plant in two ways. As an adult, weevils kill the youngest buds by feeding on them. Adult weevils lay their eggs in the inflorescences of the plant and the larvae consume seeds as they mature. We conducted a field experiment to quantitatively assess the direct effect of the recently-approved biocontrol pathogen Puccinia jaceae f.s. solstitialis on plant performance and any indirect effects that might arise by altering the plant’s interactions with Eustenopus or its pollinators. We documented numerous nonadditive outcomes between the pathogen and the weevil. Pathogen infection enhanced the impact of adult weevils on the plant. Infected plants produced fewer buds than uninfected plants and because adult weevils did not avoid feeding on infected plants, they destroyed proportionally more buds per infected plant than per uninfected plant. In other words, the indirect interaction between the pathogen and the weevil produced a super-additive outcome and this is the ideal result between two biocontrol agents. However, we found a very different outcome when the weevil was in the larval stage. Larval weevils consumed proportionally fewer seeds per inflorescence when they were developing on infected plants as compared to uninfected plants. Infection had no effect on pollinator visitation. The net effect was that plants attacked by both the pathogen and the weevil did not produce fewer seeds than those attacked only by the weevils. Because the consequence to the plant of interacting with one species may depend on the presence or absence of other interacting species, a more sophisticated understanding of multispecies interactions may be necessary for the selection of biocontrol agents that act in a complementary manner to reduce plant fitness. Likewise, relatively tractable weed-biocontrol systems allow us to examine multispecies interactions that can be difficult to study experimentally in native systems. Technical Abstract: Plants interact with numerous enemies and mutualists simultaneously and sequentially. Such multispecies interactions can give rise to trait-mediated indirect effects that are likely to be common in nature but which are also inherently difficult to study and predict. Understanding multispecies interactions is also important in the use of biological control agents to control invasive plants because modern approaches to biocontrol rely on releasing multiple agents for each target weed. Centaurea solstitialis is one of the most problematic invasive weeds in California and the weevil Eustenopus villosus is its dominant biological control agent. We conducted a field experiment to quantitatively assess the direct effect of the recently-approved biocontrol pathogen Puccinia jaceae f.s. solstitialis on plant performance and any indirect effects that might arise by altering the plant’s interactions with Eustenopus or its pollinators. We documented both synergy and interference between the two biocontrol agents depending on the lifestage of one. Puccinia infection increased the impact of bud-feeding by the adult weevils but reduced the impact of seed-feeding by larval weevils. Infection had no effect on pollinator visitation. The net effect was that attack by both the pathogen and the weevil did not reduce plant fitness relative to plants attacked only by the weevil. Because the consequence to the plant of interacting with one species may depend on the presence or absence of other interacting species, a more sophisticated understanding of multispecies interactions may be necessary for the selection of biocontrol agents that act in a complementary manner to reduce plant fitness. Likewise, relatively tractable weed-biocontrol systems allow us to examine multispecies interactions that can be difficult to study experimentally in native systems. |
