The anatomical properties of loblolly pine (Pinus taeda L.) and cottonwood (Populus deltoides Barto ex. Marsh.), both from intensively managed woodlands, were quantitatively characterized using the principles of stereology. Physical and mechanical properties were also determined for each growth increment of six sample trees of both species. Anatomical parameters measured were correlated to certain mechanical properties. The numerical values obtained for the anatomical properties were derived from simple counting measurements. They included size distribution parameters of individual anatomical elements as viewed on transverse sections. The parameters were determined both in terms of within growth ring variability and in terms of changes from pith to bark within the species studied. Average mechanical properties of the two species were shown to increase significantly from pith to bark. In the regression models constructed for predicting physical and mechanical properties, up to three anatomical variables were found to significantly account for the variation in the strength properties. Close to 87% of the variation in MOR and 86% in MOE could be accounted for by the three anatomical parameters in loblolly pine. About 82% of the variation in the crushing strength parallel to grain could be explained by the anatomical variables. The three anatomical parameters selected for predicting strength properties in cottonwood were all related to fibers, suggesting that the most important anatomical elements determining strength in cottonwood are the fibers. Overall predictability in cottonwood was not as good as in loblolly pine. Less than 40% of the variation in MOR could be explained by fiber properties alone, while almost 70% in MOE was accounted for. Over 55% of the variation in maximum compression strength parallel to grain could be explained by fiber properties alone. About 84 to 94% of the variation in specific strength properties was accounted for by three anatomical variables unrelated to specific gravity in loblolly pine. Addition of specific gravity improved the model with R² values between 92-97%. In cottonwood, 65-85% of variation in specific strength properties was accounted for by three anatomical variables unrelated to specific gravity. The R² values also improved by addition of specific gravity (82-92%).

Ph. D.