Abstract: Forest fires are mainly caused by human factors or natural factors. Once they occur, they will be out of control and spread everywhere. China adopts the forest fire prevention strategy of "prevention first and active elimination", divides the fire resistance intensity of the forest through scientific methods, rationally distributes the fire belt, and realizes the effective prevention and control of forest fires. In this paper, the fire resistance research of 7 kinds of dominant forest materials (cypress forest, birch forest, broad-leaved mixed forest, poplar forest, spruce forest, mixed forest, mixed forest, mixed forest) was carried out, to determine the four fire resistance indexes (dry-fresh ratio, balanced moisture content, calorific value, ignition point) of shrub layer I, herb layer, humus layer and litter layer (d<0.6cm), V(0.6cm ≤ d <2.5 cm) and litter layer (2.5cm ≤ d<7.62cm), analyzing the differences among the various fire resistance indicators and their relationship, preliminary evaluation of forest fire resistance. The results show as follows. (1) the dry-fresh ratio is located in 35.61% ~ 86.71%, the largest dry-fresh ratio in conimixed forest, and the difference between different combustible types is very significant (P <0.01). The variation of equilibrium water content of humus layer and litter-layer was 38.67%~43.23%, 45.32%~55.32%, 56.83%~63.85% and 63.81%~75.32%, respectively. The equilibrium water content was significantly different among the four forest fuel types (P <0.01). The ignition points of humus layer and dead litter layer-mixed sample were 265.12~287.68 ℃ and 242.63~270.82 ℃ respectively, and the calorific values are 842.43~1 283.97 J/g and 1 165.21~1 521.14 J/g respectively, and significantly between different forest combustible types (P <0.01).(2) Pearson's correlation analysis shows that there is a positive correlation between the dry and fresh ratio, between balanced water content, and between balanced water content and ignition point. The dry-fresh ratio is negatively correlated with the equilibrium moisture content, and between the dry-fresh ratio and the ignition point.(3) The entropy weight method is used to rank the fire resistance of 7 dominant forests and obtain the fire resistance from strong to weak: poplar forest> broad-leaved mixed forest> needle and broad-leaved mixed forest> spruce forest> coniferous mixed forest> birch forest> cypress forest.(4) The system clustering divides the fire resistance strength into 3 categories: category 1 is poplar forest with strong fire resistance; category 2 is broadleaved mixed forest, coniferous and broad-leaved mixed forest and spruce forest, with moderate fire resistance; category 3 is coniferous mixed forest, birch forest and cypress forest, with weak fire resistance. By studying the correlation between the physical and chemical properties of forest combustible materials and their fire resistance, we obtain the influence of the physical and chemical properties on the fire resistance of the two, which provides an important basis for the division of different fire risk levels in the study area; strengthen the early warning and monitoring of the relatively weak forest and study the fire and potential fire behavior of forest, forecast and monitor the occurrence and extinction of forest fire