direct proportion existing between the specific gravity and the strength, the densest wood having borne the greatest strain before breaking. Instead, however, of this point of density lying at the centre of the tree, as in the specimens of Oak that were tested in a similar manner, we find it, as is generally the case with other woods, about midway between the pith and the outer layers of duramen. The results, if compared with the mean of the firstmentioned set of experiments on Yellow Pine (Table CXLVIII.), which were upon pieces taken from several trees, show that the tree from which the seven specimens were obtained possessed a little less strength, and rather less elasticity, than the former; but then it must be borne in mind that they were selected pieces, and probably did not include the weaker wood of either the oldest or the newest layers. Further experiments were tried on six out of seven of the specimens, to ascertain their relative tensile strength (Table CLXIX., column 9). The following are the average results:—
| The pieces | 1 and 1 | s.g. | 562 | broke with | 2,607 | lbs. on the | square inch. |
| The pieces„ | 2 and„ 2′ | s.g„ | 584 | broke with„ | 2,800 | lbs. on the„ | square inch.„ |
| The pieces„ | 3 and„ 3′ | s.g„ | 567 | broke with„ | 2,870 | lbs. on the„ | square inch.„ |
The centre piece ʘ was not tried for tensile strength, as it was too much crippled under the transverse strain to be of any further value for experimental purposes.
The denser layers 2 and 2', were not in this case quite so strong as 3 and 3', which were of a less specific gravity.
