However, there are many additional indirect pathways by which damage to a neighbor might affect herbivory and plant fitness. For example, plant cues are information available to any organism that can access them and may either directly attract or repel herbivores. In this case, the signal from a damaged “emitter” may attract herbivores under field conditions, which may in turn increase herbivory to a neighbor without any direct information exchange between the plants. This hypothesis is consistent with our observation that damage to a neighbor sometimes increased herbivory experienced by receivers in the field, but it cannot account for cases in which neighbor damage increased feeding by an herbivore in no-choice laboratory CUDC-907 1339928-25-4 palatability trials. As another possibility, the herbivore community in natural settings is often diverse, and many plant defenses are specific to particular groups of herbivores. In this context, specific defenses elicited by damage to a neighbor may have no effect on some herbivores, and may even attract others. Consistent with this possibility is the fact that, in feeding assays using Sinapis arvensis, exposure to a damaged neighbor increased palatability to a generalist herbivore, but decreased palatability to a specialist herbivore. In this situation, the net effect of damage to a neighbor will depend on the types of defensive responses and the relative abundances of alternative herbivore types in the field. Alternatively, many plants have unique responses to different herbivores based on herbivorespecific cues, and the use of mechanical damage in this study to elicit damage-induced cues may prompt a different cue than real herbivore damage. In summary, the cues released from damaged plants have highly context-specific effects on the palatability, actual herbivore damage, phenology, and fitness of their neighbors. In a realistic multi-species field setting, we found that the consequences of receiving a signal from a damaged neighbor may be either positive or negative. Moreover, regardless of the fitness impacts, we have shown that the consequences of receiving a signal from a damaged neighbor are typically greater when the neighbor is a close relative. Overall, our study paints a more complex picture of plant information exchange than has been revealed in previous studies that find a consistent benefit of interplant cues. Past studies have most convincingly demonstrated effects of plant signals on herbivory in woody plants and vines that may coordinate their own defensive response via volatile cues, but see studies that show responses of tobacco to wounded sagebrush and physiological responses of Arabidopsis to volatile cues. Field studies that assess the efficacy and consequences of plant signals in a variety of habitats and plant-life history types will be needed to understand the contexts in which plant signaling is a major component of plantherbivore interactions.