Environmental Entomology (1991) 20, 410-418
David R. Horton, Phillip L. Chapman and John L. Capinera (1991)
Detecting local adaptation in phytophagous insects using repeated measures designs
Environmental Entomology 20 (2), 410-418
Abstract: Substantial variation among experimental units may occur in studies that address effects of the host plant on insect performance or preference. Efficiency in demonstrating host effects can be enhanced by recognizing that variation among insects is often important and, more particularly, by using an experimental design appropriate for these conditions. Performance data from studies that address local adaptation in phytophagous insects are often analyzed with a two factor ANOVA, testing for host species, population, and host species x population effects. Local adaptation to different hosts would be indicated by a significant interaction term. Commonly, a completely randomized design is employed. A repeated measures design involving an allocation of siblings across host species can provide considerably more powerful tests on the host x population interaction than a completely randomized design. Allocation of relatives across host species can induce positive correlations among hosts and, depending upon magnitude of correlations, provide conditions suitable for powerful tests of the interaction term. We develop power curves for several patterns of correlation to, describe the amount by which the repeated measures design will be more powerful than the completely randomized design. A literature review indicates that conditions appropriate for repeated measures analysis are very common. Finally, we provide an example using two populations of a phytophagous beetle feeding on three plant species, and show that the repeated measures design demonstrates host x population effects missed by the more traditional design.
(The abstract is excluded from the Creative Commons licence and has been copied with permission by the publisher.)
Database assignments for author(s): John L. Capinera, David R. Horton
Research topic(s) for pests/diseases/weeds:
population dynamics/ epidemiology
Pest and/or beneficial records:
| Beneficial | Pest/Disease/Weed | Crop/Product | Country | Quarant.
|
|---|---|---|---|---|
| Leptinotarsa decemlineata | ||||
| Myzus persicae | ||||
| Spodoptera frugiperda | ||||
| Chrysomela knabi | ||||
| Alsophila pometaria | ||||
| Liriomyza sativae |
