Seismic refraction measurements of the compressional wave velocity of the upper mantle, obtained in the northeast Pacific Ocean between California and Hawaii, have shown a dependence on direction indicating anisotropy of velocity with a maximum of 8.6 km/sec in a roughly easterly direction and a minimum of 8.0 km/sec. In order to test whether or not these observations represent true anisotropy, rather than geographical variation of velocity, special anisotropy experiments were conducted at two locations centered roughly at 31°N, 121°W and 35°N, 126°W, respectively. At each of these locations significant anisotropy was observed, agreeing in direction with the apparent anisotropy indicated for the entire region. The magnitude of the velocity difference, 0.3 km/sec, on any single station is half the over‐all variation of 0.6 km/sec. These observations do not prove the hypothesis of anisotropy throughout the entire region from California to Hawaii. However, they are consistent with this hypothesis and give encouragement to similar observations elsewhere.

No abstract

Two major families of faults dominate the tectonics of slow spreading ridges: the transform faults paralleling the spreading direction and the 'GLORIA' faults, observed by large-scale side-scan sonar, normal to it. The observation of microearthquakes to 8 km depth beneath the Mid-Atlantic Ridge axis implies that not only the transform faults but the GLORIA faults as well penetrate into the mantle.Since large horizontal temperature gradients are often associated with them, both types of fault provide the means of circulating large volumes of sea water through the oceanic mantle. Under these conditions, serpentinization of upper mantle rocks will occur in a predictable way. Its influence on the tectonic development of slow spreading ridges is discussed.

No abstract

Twenty‐seven seismic refraction profiles in the southern Indian Ocean are described. They straddle four major physiographic features: Two branches of the mid‐ocean ridge system, Broken ridge, and the Ninetyeast ridge. The Réunion and Crozet basins are structurally very similar to the flanks and adjacent basins of the northern mid‐Atlantic ridge. North of Amsterdam Island the mantle beneath the crest of mid‐ocean ridge has a velocity of 7.61 km/sec; away from the crest the mantle has normal velocity but is exceptionally shallow. Broken ridge has a quasi‐continental structure, and Ninetyeast ridge appears to be upfaulted oceanic crust rather than volcanic accumulation. A station in the West Australian basin revealed a shallow, low‐velocity mantle.

Fifteen seismic refraction profiles in the northwest Indian Ocean are described. The results suggest a connection between the linear Maldive ridge and the Deccan traps of the Indian mainland. No anomalous upper mantle velocities have been found near the crest of the Carlsberg ridge. A great thickness of material with velocity 6.03 km/sec west of the Saya de Malha bank has been interpreted as an extension of the Seychelles granitic block to the south. At two stations between the Chagos archipelago and the Seychelles bank the Mohorovicic discontinuity appears to be less than 8 km below sea level.

The earthquake swarm which accompanied the Fernandina Caldera collapse is shown to consist of two different types of earthquake: lowstress events occurring on the main caldera fault and higher stress events on minor faults. The difference in scale of these faults allows the two types of earthquake to be distinguished by magnitude. Events occurring on the main caldera fault have a high ' b ' value, those on minor faults a normal ' b ' value. The fall in magma chamber pressure which led to the collapse is shown to have been quite small and hence a model of the volcano is developed which can account for the magnitudes of the earthquakes observed, the stress on the main caldera fault and the seismic efficiency.The seismicity of the rift zone of the Mid-Atlantic Ridge has many similarities with that of the Fernandina collapse and it is proposed that the median valley which is characteristic of slow-spreading ridges is a zone of caldera collapse. No such collapse occurs along the axes of fastspreading ridges, so these are relatively aseismic.

No abstract