Surveying patterns in the cost of resistance in plants

Despite the general belief that traits conferring resistance in plants to pests also confer costs, disagreement persists about the frequency of costs and the conditions under which they are most likely to be evident. In this article, we analyze 88 published comparisons to explore when trade-offs between resistance and fitness traits can be detected. Among the patterns revealed are that costs were most often associated with resistance to herbicides, followed by resistance to pathogens, and least often associated with resistance to herbivores; costs were more often found in crops versus wild species; greater control of the genetic background increased the probability of detecting costs of resistance; there was large variation in the cost associated with the same resistance trait in different genetic backgrounds; and many examples of costs of resistance appeared to be due to linkage rather than pleiotropic effects. We discuss these and other results and emphasize that dependencies among the data invalidate statistical verification of these patterns. We hope that our results stimulate experiments that test the observed patterns critically.