Abstract: To understand the characteristics of community structure and spatial distribution of small mammals in agricultural area of Yunnan Province, a systematic investigation was carried out in 104 quadrats of 25 regions in Yunnan Province from August 2010 to April 2018 by rat trap night method. The spatial variation of community characteristics along environmental gradients was analyzed by community ecological indicators. The results showed that a total of 3240 small mammals were captured and cold be classified into 42 species in 21 genera, 9 families, and 4 orders. The largest number of small mammal was rodents, dominated by Apodemus chevrieri and Rattus tanezumi. The 25 regions were clustered into three classes. The altitudinal distribution of small mammals was similar to the latitudinal distribution in agricultural areas. The number of species was relatively less in the low latitude and altitude range, with Rattus spp. and Mus spp. as the dominant species. In the high latitude and altitude region, the dominant species changed into Apodemus, Niviventer and Eothenomys. With the increases of altitude, the diversity index showed unimodal distribution, with the highe-st species diversity occurred in the mid-altitude area. The diversity index of small mammal showed the “V” type pattern in longitude, being the highest in the 98°-99° E gra-dient zone. At the latitude level, it showed an overall upward trend from south to north. Results from the GAM analysis showed that the degree of influence on the small animals in the agricultural area was in order of longitude, altitude and latitude. The similarity analysis in the composition of small mammals showed that the moderate similarity occurred in the adjacent gradient zone, and the highest similarity occurred in middle altitude zone, middle latitude zone, and low longitude zone. The farther the distance between different gradient zones, the lower the similarity of community structure. There was high spatial heterogeneity in different dimensions of small mammals’ community structure in Yunnan Province. The geographical distribution trend of species diversity showed different distribution patterns across environmental gradients.