Microfibril angle (MFA) was determined at each growth ring from disks at breast height (1.3 m) from four scots pine (Pinus sylvestris) trees grown in northeastern China. Significant variation in microfibril angle was observed among growth rings. MFA at breast height showed a decreasing trend from pith to bark for each tree. The modified logistic model with nonlinear mixed-effects was used for modeling earlywood MFA. The NLME procedure in S-Plus is used to fit the mixed-effects models for the MFA data. The results showed that logistic model with two random parameters and could significantly improve the model performance. The CS, AR(1), MA(1), and ARMA(1,1) correlation structures were incorporated into mixed-effects models. The mixed model with the AR(1), MA(1), and ARMA(1,1) correlation structures improved model performance (P<0.0001).