Extra-pair mating could drive sexual selection in socially monogamous species, but support for this hypothesis remains equivocal. We used lifetime fitness data and a unique model species, the dimorphic white-throated sparrow (Zonotrichia albicollis), to examine how extra-pair mating affects the potential for sexual selection. The morphs of this species employ distinct reproductive strategies, with white males pursuing extra-pair mating at higher rates than tan counterparts. Social and extra-pair mating is disassortative by morph, with paternity exchange occurring primarily between pairs composed of white males and tan females. We predicted stronger sexual selection as quantified by Bateman gradients and standardized variance in reproductive success in white compared to tan males. Furthermore, because males drive extra-pair mating, we predicted costs of multiple mating and a negative Bateman gradient in tan females. The Bateman gradient for lifetime reproductive success was larger in white than tan males, and extra-pair mating contributed more to the Bateman gradient for white males. However, the Bateman gradient was positive in tan females. White males had higher variance in annual reproductive success than tan males or females, but variance in lifetime reproductive success did not differ between the morphs or sexes. Moreover, extra-pair mating did not increase variance in male reproductive success relative to apparent patterns, and within-pair success accounted for much more variance than extra-pair success. Although extra-pair mating in white males increases Bateman gradients, and potential for sexual selection via mate numbers, these latter results call the overall importance of extra-pair mating in driving selection into question.