Summary and Conclusions
Seismograms recorded at epicentral distances between 4° and 26° are investigated, and revised travel-time curves for P and S are given. Both consist of two branches, the first almost a straight line ending in a shadow zone, the second beginning with relatively large amplitudes and with a delay relative to the first branch. The end of the shadow zone for P seems to be near 15° in most regions; for S this distance seems to vary somewhat and to be usually a few degrees greater. The “20-degree discontinuity” of Lehmann and Jeffreys, following Byerly (1926), as well as observations by Neumann (1933) and others, probably can be referred to the same phenomenon. Differences in distance are partly due to local effects, partly to the difficulty of tracing the first branch of the travel-time curves of P and S across the shadow zone.
Huancayo, Peru, seismograms from near-by earthquakes at intermediate depth are used for studying the change of amplitude with focal depth at a given distance in and near the shadow zone. All observations are explained on the assumption that the velocity of P as well as S waves decreases somewhat (between 1/2 and 3 per cent?) at a depth below the Mohorovičić discontinuity to a minimum value at a depth near 80 to 100 km. The data do not permit decision whether the decrease in velocity is gradual over a range of about 50 km. or (almost) sudden at a given depth. The layer with lower velocity extends downward to a depth of about 150 to 180 km.; below, the velocities increase rather rapidly, but Poisson's ratio remains near 0.30, contrasting with a value of slightly less than 1/4 in the upper layers. Correspondingly, the rigidity remains relatively small as compared with the bulk modulus, apparently down to the surface of the core.
It is of interest to note that such a low-velocity layer has now been established for the solid earth, the ocean, and the atmosphere, providing shadow zones for elastic (sound) waves in all three. It is also noteworthy that only a relatively small decrease in velocity is required to produce a rather extensive and pronounced shadow zone. The theory of free waves connected with the surface of minimum wave velocity probably will throw some light on the relatively large amplitudes and persistence of surface waves with a velocity of about 4.45km/sec. in seismograms.