The ratio of the amplitude of the longitudinal wave reflected at the surface of the earth to that of the direct wave as recorded by Galitzin seismographs was examined to find whether it might be used as a means of distinguishing between Pacific and continental reflections.
Theoretically, assuming that equal energy is radiated in all directions from the source, the amplitude ratio, PP/P, for Pacific reflections should be smaller than for continental reflections at all epicentral distances, if the speed of P waves near the surface is higher under the Pacific than under the continents.
The records of 194 earthquakes with epicenters at distances from 19° to 103°, and having reflections under the continents or the Pacific Ocean, were examined for the purpose. No such pattern as expected by theory was observed; on the contrary there resulted a general scattering when these ratios were plotted, for both types of reflection, against epicentral distances.
The observed values of the apparent angle of incidence agreed better with those calculated for V = 8.00 km/sec. than with those for V = 6.00 km/sec., where V is the velocity of P waves near the surface of the earth. Since 8.00 km/sec. is the velocity of longitudinal waves below the surface layers of the continents, while 6.00 km/sec. is an intermediate velocity within the layers, it was concluded that the waves recorded by the Galitzin seismographs were not refracted into the surface layers of the earth and consequently that the amplitudes of waves of the periods recorded on Galitzin seismographs (4 sec. to 12 sec.) do not afford a means of differentiating between reflections at a layered surface such as the continent and at an unlayered surface such as Gutenberg considers the Pacific to be.
However, comparison of some twenty records of the vertical Benioff seismograph of approximately 0.7 sec. free period (recorded waves of periods 1 sec. to 2 sec.) with those of the same earthquakes recorded by the Galitzins of 12 sec. free period showed no essentially different behavior.
It was observed that the Berkeley, California, and Florissant, Missouri, stations, both using Galitzin instruments of nearly the same constants, were situated at the same distances from eight Mexican epicenters, and both received impulses from these earthquakes over the same kind of continental paths. An examination of their respective values of PP/P showed this to be larger at one station or another according as the first recorded motion at Berkeley was a condensation or a rarefaction. This variation in the values of PP/P shows that energy is not sent out equally in all directions from the source, as was assumed, and therefore that the mechanism at the focus plays an important part in the value of this ratio.