Predator-prey interactions and life history of Orius laevigatus and O. majusculus feeding on flower and leaf-inhabiting thrips
Description
Thrips (Thysanoptera: Thripidae) are major pests in horticulture worldwide. Longstanding biological control strategies that have been developed for flower thrips such as Frankliniella occidentalis (Pergande) are being disrupted by the recent introduction of leaf-inhabiting thrips such as Echinothrips americanus Morgan and Thrips setosus Moulton in Northern Europe. In this study, we evaluated the predator–prey interactions, predation capacity, juvenile development and adult reproduction of the two commercial anthocorid predators Orius laevigatus (Fieber) and Orius majusculus (Reuter) (Hemiptera: Anthocoridae) on these thrips. In behavioral assays, predators were more successful in subduing and consuming sedentary leaf-inhabiting thrips adults compared to the highly mobile F. occidentalis. Furthermore, O. laevigatus was more successful in subduing prey compared to the bigger predator O. majusculus. Female adults of O. laevigatus and O. majusculus killed 18 and 20 F. occidentalis adults, respectively, in 24 h, while the kill rate was around two times higher when predators were offered E. americanus or T. setosus as prey. Developmental and reproductive parameters of both Orius predators were more favorable when feeding on the leaf-inhabiting thrips compared to F. occidentalis. This was further evident in the higher intrinsic rates of increase (rm) we recorded on a diet of E. americanus compared to F. occidentalis (0.162 and 0.148 females/female/day for O. laevigatus, respectively; 0.148 and 0.127 for O. majusculus, respectively). Our findings show that E. americanus and T. setosus constitute high quality prey for anthocorid predators, highlighting the potential of these predators for effective pest control.
Files
Steps to reproduce
Continuous and count data that fulfilled the normality and homoscedasticity assumptions were analyzed with an ANOVA. When these requirements were not met, count data were analyzed with a generalized linear model (GLM) with Poisson error distribution. Binary data were analyzed through GLM with binomial error distribution and probit link. To account for under- or overdispersion in the GLMs when necessary, we changed the error distributions to quasipoisson for count data and quasibinomial for binary data (McCullagh and Nelder, 1989). To account for pseudo-replication in the observation experiment where a single predator had multiple encounters within the studied timeframe, we included the experimental individual as random effect in a Generalized linear mixed model (GLMM) analysis. For the GLMM, negative binomial error distribution with linear parameterization was chosen as the best fitting, based on AICc criteria (Hardin and Hilbe, 2018). For all studied parameters, two-way factorial analyses were initially applied. When interaction between factors was not significant, we performed post hoc analysis using Tukey’s HSD to separate means on significant main effects. When a significant interaction between factors was found, means were compared pairwise. Sex ratios were compared to an equal distribution (1:1) using Chi-square tests. All statistical analyses were performed using the statistical software R 4.0.2 (R Core Team, 2021). We used the ‘multcomp’ package to perform post hoc analyses (Hothorn et al., 2008), the ‘glmmTMB’ package to fit GLMM (Brooks et al., 2017), and the ‘DHARMa’ package to perform residual diagnostics for all models (Hartig, 2022).