Since writing the above, we have looked over some experiments by Messrs Barlow, Beaufoy, Couch, and others, on the strength of timber. These show so much discrepancy of result, as leads us to conclude, {222} that experimenters have not attended sufficiently to the structure and nature of the timber, the position and quality of the different layers, &c. Take, for example, the stem of a tall tree, 100 years old: At the cross section, it is found to consist of a certain number of layers of matured timber, and of sap timber. These layers having been gradually formed, the external, after those more internal have partly dried, and the internal and matured wood being also filled up to more solid consistency than the external, the stem, on being barked, contracts in drying much more externally than internally. As soon as the surface has dried, the outer layers contracting laterally are not sufficient to surround the undried internal layers, thence split in longitudinal rifts; and as the drying proceeds inwards, the cracks deepen till they reach nearly to the heart—these rifts, when the timber is thoroughly dry, being generally wider in the sap timber than in the matured, more than in the proportion of the size of the respective circles. This effect of drying is what every body is acquainted with. Besides lateral contraction, there is also a disposition to contract longitudinally by drying, much greater in the external than internal layers. While the tree is undivided, this greater contraction of the {223} exterior layers is prevented, by the adhesion to the drier more filled up central column (which probably had contracted a little during the formation of the exterior sap-wood layers), the contractile force of the exterior balancing equally around this central column. Should this balance be destroyed by the stem being cleft up the middle, the longitudinal contraction will immediately take place, and the two halves will bend outward, from the outside layers contracting more than the inside layers. We have seen an ash tree rend up the middle from the cross section above the bulb, nearly to the top, on being cut across in felling, owing to the longitudinal contractile force of the exterior existing even before drying.