Abstract: Based on the data of Pinus elliottii, R software was used to analyze the data. Least square method was used to fit the bark thickness at breast height, bark thickness at any height, relative bark thickness and bark diameter, and the model was optimized. The results showed that 4 models were selected to fit the bark thickness at breast height, bark thickness at any height, relative bark thickness and bark diameter. The models were BBT=0.372 + 0.049DBH, BT=0.964 - 0.676CR - 0.055RH + 0.034dob, RBT = 0.095 + 0.902RD^1.718 and dib = 0.906dob^0.995 respectively. The above 4 models have high fitting accuracy, no multiple collinearity problem, and the model has no heteroscedasticity or has been modified by variable transformation. Besides the crown length rate, the data needed by the fitting model are easy to obtain in forestry production practice. The addition of crown length to bark thickness model is better for slash pine. The study on the bark thickness model of P.elliottii plantation can provide reference for estimating the bark volume, yield and bark growth.