He therefore maintained, with Davy, the hypothesis of a non-oxidized nucleus which becomes an inexhaustible chemical source of heat by contact with the already oxidized crust. In this view of the matter, a volcano is simply a permanent fissure and channel of intercourse between the non-oxygenated nucleus and the liquids lying upon the oxygenated bed. Whenever this passage of the liquids to the nucleus takes place, elevations of the earth occur from the increase of volume due to oxidation. The heat generated by these chemical actions is propagated at once toward the exterior and interior of the globe, and, in proportion as the oxidation of the crust progresses, the seat of these chemical actions is carried deeper. This, a difficult theory to sustain, has no longer any adherents.
To the objection of the power of the tides it may be further said that, examined more carefully, they would probably be found to produce an entirely inappreciable flexion of the solid crust far from sufficient to cause any disruption. Indeed, it is a question to be considered whether earthquake phenomena do not indicate the existence of subterranean tides. This is a matter that has formed the object of over thirty years' researches by M. Alexis Perrey, Professor of the Faculty of Sciences at Dijon. Professor Perrey has compiled all the observations of earthquakes from the middle of the last century to the present time, and, in grouping the various facts collected during these one hundred and twenty-five years, he has been able to adduce evidence of a connection between the frequency of earthquakes and the phases of the moon. If the phenomena are compared with the lunar month, two maxima will be seen at the periods of the syzigia (new and full moon), and two minima during the first and last quarter. In the following table the results of observations for three periods are given, the earthquake-days being grouped in the weeks corresponding to the moon's four phases, the new and full moon groups and the quadrature groups being separated:
| OBSERVATIONS. | First Period. 1751-1800. |
Second Period. 1801-1850. |
Third Period. 1843-1872. |
| Total number of earthquakes | 3,655 | 6,595 | 17,249 |
| During the syzigia | 1,901 | 3,434 | 8,838 |
| During the quadratures | 1,754 | 3,161 | 8,411 |
| Difference in favor of syzigia | 147 | 273 | 427 |
It would thus appear that the shocks occur more frequently at those periods when the sun and moon can combine their action on the liquid particles of the interior of the globe.
M. Perrey has also compared the earthquake-periods with the times of perigee and apogee, that is, the moon's nearest point in her orbit to
