1. Sequential damage by attackers is hypothesized to result in reciprocal antagonism (cross-talk) between jasmonic acid (JA) and salicylic acid (SA) defence signalling pathways in plants. However, evidence for this cross-talk is not universal and several studies have found positive (synergistic) or no interaction whatsoever between JA and SA pathways. 2. Here we conducted a meta-analysis of studies on plant-mediated effects of initial attackers on performance of subsequent attackers to test the hypothesis of cross-talk between plant hormonal signalling pathways.3. We found a significant negative mean effect size of JA-inducing initial attackers on both JA- and SA-inducing subsequent attackers, but a non-significant effect of SA-inducing initial attackers on both JA- and SA-inducing subsequent attackers. Effects on subsequent herbivores were contingent on the biology of the initial attacker, with negative effects of JA-inducing initial herbivores and no effect of SA-inducing initial herbivores, whereas pathogens on average did not influence subsequent herbivores. Further, negative effects of JA-inducing initial herbivores on JA- and SA-inducing subsequent herbivores held when both attacks occurred on the same plant part, but when attacks were on different plant parts the effect of JA-inducing herbivores on SA-inducing herbivores was non-significant. 4. Synthesis. These results indicate that reciprocal antagonism between JA and SA pathways is not universal, and suggest asymmetry and specificity in plant-mediated interactions.