In my original Paper (Dynamic, &c.), I had suggested, as an important object, to ascertain by actual experiment what might be the wave's transit rate in various rocky and incoherent formations; and having proposed this in my first "Report upon the Facts of Earthquake" to the British Association, I was enabled by its liberality to commence those experiments, in which I was ably assisted by my eldest son, then quite a lad—Dr. Jno. William Mallet, now Professor of Chemistry at the University of Virginia, U.S.; and to give account of the results, in my second Report ("Report, British Association for 1851") to that body.
Those experiments were made by producing an impulse at one end of an accurately measured base line, by the explosion of gunpowder in the formation experimented upon, and noting the time the elastic wave generated required to pass over that distance, upon a nearly level surface. Special instruments were devised and employed, by which the powder was fired and the time registered, by touching a lever which completed certain galvanic contacts. The media or formations in which these experiments were conducted were, damp sand—as likely to give the minimum rate—and crystalline rock (granite), as likely to give the maximum. The results were received, not with doubt, but with much surprise, for it at once appeared that the actual velocity of transit was vastly below what theory had indicated as derivable from the density and modulus of elasticity of the material, taken as homogeneous, etc. The actual velocities in feet per second found were:
| In sand | 824·915 | feet | per second. |
| In discontinuous and much shattered granite | 1,306·425 | " | " |
| In more solid granite | 1,664·574 | " | " |
This I at once attributed, and as it has since been proved correctly, to the loss of vis viva, and consequently of speed, by the discontinuity of the materials.
And some indication of the general truth of the fact was derivable from comparing the rude previous approximations to the transit rate of some great Earthquakes. In the case of that of Lisbon, estimated by Mitchell at 1,760 feet per second. It was still desirable to extend similar experiments to the harder classes of stratified and of contorted rocks. This I was enabled to carry into effect, at the great Quarries at Holyhead (whence the slate and quartz rocks have been obtained for the construction of the Asylum Harbour there), taking advantage of the impulses generated at that period by the great mines of powder exploded in these rocks.
The results have been published in the "Philosophical Transactions for 1861 and 1862 (Appendix)." They show that the mean lowest rate of wave transit in those rocks, through measured ranges of from 5,038 to 6,582 feet, was 1,089 feet per second; and the mean highest, 1,352 feet per second; and the general mean 1,320 feet per second.
By a separate train of experiments on the compressibility of solid cubes of these rocks, I obtained the mean modulus of elasticity of the material when perfectly continuous and unshattered, with this remarkable result—that in these rocks, as they exist at Holyhead, nearly seven-eighths of the full velocity of wave transmission due to the material, if solid and continuous, is lost by reason of the heterogeneity and discontinuity of the rocky masses as they are found piled together in Nature.
I also proved that the wave-transit period of the unshattered material of these rocks was greatest in a direction transverse to the bedding, and least in line parallel with that; but the effect of this in the rocky mass itself may be more than counterbalanced by the discontinuity and imperfect contact of the adjacent beds.
These results indicate, therefore, that the superficial rate of translation of the solitary sea-wave of earthquakes may, when over very deep water, equal or even exceed the transit rate (in some cases) of the elastic wave of shock itself.