Main content area

Tree species diversity alters plant defense investment in an experimental forest plantation in southern Mexico

Rosado‐Sánchez, Silvia, Parra‐Tabla, Víctor, Betancur‐Ancona, David, Moreira, Xoaquín, Abdala‐Roberts, Luis
Biotropica 2018 v.50 no.2 pp. 246-253
chemical defenses, ecological function, forest plantations, herbivores, leaves, phenolic compounds, prediction, species diversity, tree and stand measurements, tree growth, trees, tropical plants, Mexico
How plant species diversity affects traits conferring herbivore resistance (e.g., chemical defenses), as well as the mechanisms underlying such effects, has received little attention. One potential mechanism for the effect of diversity on plant defenses is that increased plant growth at high diversity could lead to reduced investment in defenses via growth–defense trade‐offs. We measured tree growth (diameter at breast height) and collected leaves to quantify total phenolics in 2.5‐year‐old plants of six tropical tree species (N = 597 plants) in a young experimental plantation in southern Mexico. Selected plants were classified as monocultures or as polycultures represented by mixtures of four of the six species examined. Tree species diversity had a significant negative effect on total phenolics, where polycultures exhibited a 13 percent lower mean concentration than monocultures. However, there was marked variation in the effects of diversity on defenses among tree species, with some species exhibiting strong reductions in phenolic levels in mixtures, whereas others were unresponsive. In addition, tree species diversity had no effect on growth, nor was the negative effect of diversity on chemical defenses mediated by a growth–defense trade‐off. These results demonstrate that tree diversity can alter investment in chemical defenses in long‐lived tree species but that such effect may not always be under strong control by plant endogenous resource allocation trade‐offs. Regardless of the underlying mechanism, these findings have important implications for predicting effects on consumers and ecosystem function.