Density has been suggested to affect variation in extra-pair paternity (EPP) in avian mating systems, because increasing density promotes encounter rates and thus mating opportunities. However, the significance of density affecting EPP variation in intra- and interspecific comparisons has remained controversial, with more support from intraspecific comparisons. Neither experimental nor empirical studies have consistently provided support for the density hypothesis. Testing the density hypothesis is challenging because density measures may not necessarily reflect extra-pair mating opportunities, mate guarding efforts may covary with density, populations studied may differ in migratory behavior and/or climatic conditions, and variation in density may be insufficient. Accounting for these potentially confounding factors, we tested whether EPP rates within and among subpopulations of the reed bunting (Emberiza schoeniclus) were related to density. Our analyses were based on data from 13 subpopulations studied over 4 years. Overall, 56.4% of totally 181 broods contained at least one extra-pair young (EPY) and 37.1% of totally 669 young were of extra-pair origin. Roughly 90% of the extra-pair fathers were from the adjacent territory or from the territory after the next one. Within subpopulations, the proportion of EPY in broods was positively related to local breeding density. Similarly, among subpopulations, proportion of EPY was positively associated with population density. EPP was absent in subpopulations consisting of single breeding pairs, that is, without extra-pair mating opportunities. Our study confirms that density is an important biological factor, which significantly influences the amount of EPP within and among subpopulations, but also suggests that other mechanisms influence EPP beyond the variation explained by density.