The slow growth-high mortality hypothesis (SG-HG) predicts that slower growing herbivores suffer greater mortality due to a prolonged window of vulnerability.Given diverse plant-herbivore-natural enemy systems resulting from different feeding ecologies of herbivores and natural enemies,this hypothesis might not always be applicable to all systems.This is evidenced by mixed support from empirical data.In this study,a metaanalysis of the SG-HM hypothesis for insects was conducted,aiming to find conditions that favor or reject SG-HM.The analysis revealed significant within-and between-group heterogeneity for almost all explanatory variables and overall did not support SG-HM.In this analysis,SG-HM was supported when any of the following 5 conditions was met:(1) host food consisted of artificial diet;(2) herbivore growth was measured as larval mass;(3) herbivores were generalists;(4) no or multiple species of natural enemies were involved in the study;and (5) parasitoids (i.e.,parasitic insects) involved in the study were gregarious.SG-HM was rejected when any of the following 5 conditions was met:(l) herbivores were from the order Hymentoptera;(2) parasitoids from more than 1 order caused herbivore mortality;(2) parasitoids were specialists;(3) parasitoids were solitary;(4) parasitoids were idiobionts or koinobionts;and (5) single species of natural enemy caused mortality of specialist herbivores.All known studies investigated herbivore mortality for a short period of their life cycle.Researchers are encouraged to monitor herbivore mortality during the entire window of susceptibility or life cycle using life tables.Studies involving multiple mortality factors (i.e.,both biotic and abiotic) or multiple natural enemy species are also encouraged since herbivores in nature face a multitude of risks during the entire life cycle.More comprehensive studies may increase our understanding of factors influencing the relationships between herbivore growth and mortality.