LAUSR

The relationship between 27 mechanical parameters and density, shrinkage coefficients, moisture content, anisotropy ratio, and the fiber saturation point (FSP) in small specimens of Quercus crassifolia Humb. & Bonpl. and Q. laurina Humb. & Bonpl. wood, from Oaxaca, Mexico, is described. Wood from both oaks displayed low dimensional stability, while mechanical testing results revealed a strong material in most forms of stress, as well as a high impact-resistance wood. Wood mechanical properties in dry and green state were positively correlated with density, shrinkage coefficients, and FSP. Density was generally the single best predictor for mechanical strength, although the modulus of elasticity was a better predictor of the modulus of rupture in the bending tests. Univariate models to predict mechanical properties using the best regressor of the physical properties gave R2 values from 0.104 (p = 0.015) to 0.494 (p < 0.001), whereas multivariate regression significantly increased the predictive power of the models, with R2 values from 0.190 (p = 0.004) to 0.646 (p < 0.001).