The effect patterns of food ,predation and Apodemus agrarius as competition species on the aggressive behavior of reed voles, Microtus fortis, were measured using repeat 2×2×2 factorial experiment design under the field enclosure. The correlation between aggressive behavior and densities of reed vole populations in different treatments was very complicated. The result was not consistent with what was predicted by Chitty’s polymorphic behavior hypothesis ; aggressive behaviors of dyadic encounters were not correlated to body mass but to the difference of body mass negatively which means that there were more aggressive behaviors between dyadic encounters with similar body mass. Aggressive behaviors of male reed voles were not related with their reproductive status while those of females were. There were more aggressive behaviors between estrous females than between nonestrous females. The independent effects of food and the difference of body mass on aggressive behavior of male reed voles were extremely significant, the independent effects of predation and interspecific competition on male aggressive behavior were not significant while those of their interaction were. The effects of body mass difference, food, predation and interspecific competition on aggressive behaviors of estrous female reed voles were all extremely significant; effects of interaction among three external factors were also extremely significant. But, the effects of these three external factors on nonestrous females were not significant. The results in this paper supported the hypothesis that there were independent and additive intergrative effects of external factors of population such as food, predation and interspecific competition on aggressive behavior of voles.