Allen, W. P., L. A. Meyerson, D. Cummings, J. Anderson, G. P. Bhattarai, and J. T. Cronin. 2017. Biogeography of a plant invasion: drivers of latitudinal variation in enemy release. Global Ecology and Biogeography26: 435-446.

ABSTRACT

Aim

The relationship between herbivory and latitude may differ dramatically between sympatric native and invasive plant taxa, which can generate biogeographic heterogeneity in the strength of local enemy release for invasive plants. The aim of this study was to compare latitudinal gradients in herbivory between native and invasive plants and investigate whether gradients are driven by local adaptation and/or phenotypic plasticity.

Location

North America.

Methods

Using co-occurring native and invasive lineages of the wetland grass Phragmites australis and the specialist gall-fly Lipara rufitarsis, we conducted a field survey to examine whether the relationship between herbivory (proportion of stems galled) and latitude is parallel for each lineage. In a subsequent common garden experiment, we assessed whether latitudinal gradients in herbivory were genetically based or driven by phenotypic plasticity.

Results

In the field, L. rufitarsis herbivory on the native P. australis lineage increased from 27% in southern populations (36.5°) to 37% in northern populations (43.8°), whereas no relationship existed for the invasive lineage (i.e., non-parallel latitudinal gradients). Similar relationships were evident in the common garden, indicating a genetic basis to latitudinal variation in L. rufitarsis herbivory. The invasive lineage suffered five times less herbivory than the native lineage, supporting the local enemy release hypothesis. However, a genetic basis to this pattern was not found in the common garden, suggesting that local environmental conditions were responsible for the strong enemy release observed in the field. In particular, environmental conditions that affect stem height, diameter and density during the L. rufitarsis oviposition period appear to be important drivers of herbivory.

Main conclusions

Our study provides evidence for biogeographic heterogeneity in the strength of local enemy release, owing to non-parallel latitudinal gradients in herbivory between native and invasive taxa. We suggest that this is likely to be a common phenomenon across a range of invaded systems and types of species interactions.

View or print this publication