Based on molecular technology, researchers find that extra-pair paternity (EPP) prevails among socially monogamous bird species. This phenomenon challenges traditional views of sexual selection and mating system, and has become one of the hot-spots in the avian behavior ecology. This review focuses on the evolutionary causes leading to EPP and the potential factors affecting it. Early studies of EPP in birds used a wide variety of tools, including plumage color polymorphism, polymorphic enzymes, and sex-differences in estimation of the heritability of morphological traits. Although each of these methods can be used to estimate the likelihood that EPP are present or absent in a population, none of them provide enough accurate estimation to allow meaningful cross-species comparison. Currently, studies of EPP mainly use “DNA-methods”, namely multi-locus mini-satellite fingerprints, single-locus mini-satellite fingerprints, and micro-satellite genotyping, because they can provide accurate outcome of paternity identification and their results are reproducible. From the point of the female, the driving forces for EPP are that females may gain direct benefits or indirect genetic benefits from EPP. Hypotheses explaining the benefits include fertility assurance, good genes, genetic compatibility, and genetic diversity. Despite large numbers of theoretically plausible explanations for EPP, there have been few direct empirical tests that can provide unambiguous support for only one type of explanation. The most straightforward test of the genetic benefit hypothesis of extra-pair copulation is a comparison of the performance of maternal half-siblings from multiply sired broods. Some highly-cited landmark studies spectacularly support the genetic benefit hypothesis, while other studies failed to reveal any systematic differences in maternal half-sibling performance, even in the same species or in taxonomically closely-related species. Briefly, the point that purpose of extra-pair copulation is to gain genetic benefits is still facing great challenges. Therefore, some researchers suggest that more attention should be focused on the interactions between parties involved in the extra-pair paternity phenomenon. Others, however, believe that such interactions can also be of a cooperative nature, and involve exchange of direct benefits. In brief, an approach focusing on interactions (involving either conflict or cooperation) seems to be the most promising direction to improve our understanding of the phenomenon of extra-pair paternity. As for the factors leading to intra- and inter-specific variation in the level of EPP, current researches mainly focus on the breeding density, breeding synchrony, the complexity of the habitat, paternal care, adult mortality, food availability, and genetic diversity. Explaining intra- and inter-specific variation in the extent of EPP has been difficult, but an appreciation of the problems of small sample sizes, and an ever-increasing comparative database have led to several recent advances. It now seems probable that differences between species in the rate of EPP are due to a combination of differences in life history, pattern of parental care, and local opportunities for promiscuity. In a word, although there have been a lot of theoretical and empirical researches about EPP of birds, no consensus on the basic questions has been received in this area. Thus, more scientific statistical analysis method and more empirical experiments are still badly needed to improve the theoretical system.