Comparison of Rayleigh-wave dispersion computations on 11 models of shear-velocity structure of the continental and oceanic crust-mantle systems permits a detailed explanation of observed mantle Rayleigh-wave dispersion for periods less than 250 seconds. Velocity distributions for the continental crust-mantle obtained by Gutenberg and by Lehmann from body-wave data, both of which include a region of low velocity in the upper mantle, are consistent with the dispersion data. Furthermore, it is shown that a structure which agrees with results from travel times of body waves from explosions and from distant earthquakes and with Rayleigh-wave dispersion must necessarily contain a low-velocity region of this type.
Shear-wave data for oceanic areas are meager, but from Rayleigh-wave dispersion there is firm evidence that the shear velocity immediately below the M discontinuity of deep ocean basins is about 4.6 to 4.7 km/sec., about the same as under the continental M discontinuity. It is also clear that the velocity distribution below the depth of the continental M discontinuity cannot be the same under continents and oceans. Instead, an oceanic model obtained by successive approximation to oceanic Rayleigh-wave dispersion data shows that the region of low shear velocity extends to much shallower depths under the oceans, thus being a much more prominent feature under oceans than under continents. This is similar to results obtained by Landisman et al. using oceanic Love-wave dispersion, and provides a more detailed knowledge of the structure.
Relatively rapid downward increase of shear velocity between depths of about 400 to 500 km. is the chief cause of the observed minimum group velocity at about 225 seconds period. Curvature of the earth is probably the cause of systematic differences beyond periods of 200 to 250 seconds between observed group velocities and group velocities computed according to the flat-earth hypothesis.