Author(s): DelWayne R. Bohnenstiehl 
Nearly 40 years ago, a study of seismicity along oceanic transform faults provided telling evidence in support of the theory of plate tectonics . Today, as exemplified by the work of McGuire and colleagues  (page 457 of this issue), seismologists are unravelling the mysteries of earthquake generation and testing the limits of prediction along these remote fault zones.
Earth's mid-ocean ridge system is where tectonic plates are moving apart and new sea floor is being created. Where one segment of this volcanic rift is offset from its neighbour, the differential motion of the plates is accommodated by a zone of lateral motion known as a transform fault (Fig. 1a). These fault zones are among the most conspicuous features on the sea floor, leaving scars that can often be traced across an ocean basin.
Like their continental cousins, such as the North Anatolian fault in Turkey and the San Andreas fault in the United States, most oceanic transforms can generate large earthquakes. Increasingly long-term seismic monitoring, however, has revealed dramatic differences between these two groups of faults. Most startlingly, an inventory of earthquakes along oceanic transforms documents a dramatic 'slip deficit', with about 85% of all transform motion occurring aseismically . In contrast, for continental transforms, plate motion occurs almost exclusively during fault slippage events that generate earthquakes (Fig. 1b).
To examine this difference and explore its implications for understanding earthquake initiation, McGuire et al .  turned to an array of moored hydrophones used by the US National Oceanic and Atmospheric Administration to monitor seismicity along parts of the equatorial East Pacific Rise . These instruments take advantage of the more efficient propagation of sound in the oceans, relative to the solid Earth, to detect and locate earthquakes that would otherwise go unnoticed by distant land-based seismometers. Within the remote region...