During the quadrennium results of tomographic experiments on the world's major Cenozoic continental rifts (Green et al., 1991; Davis et al., 1993; Glahn et al., 1993; Gao et al., 1994a) suggest that significant asthenospheric upwarp occurs beneath the east African, Rio Grande and Baikal rifts, but is absent beneath the Rhine Graben. The Kenya Rift International Seismic Project (KRISP, Keller et al., 1992) carried out a large scale refraction and teleseismic tomography project in Kenya. A new crustal model across the east African rift shows systematic crustal thinning across the rift axis, as well as variable crustal thickness along axis. Some of the largest P-wave velocity contrasts in the upper mantle beneath an extending region (up to 12 percent) were observed teleseismically beneath the Kenya rift, suggesting the mantle beneath the rift is partially molten.
A new book ``Seismic Tomography: Theory and Practice'' (Iyer and Hirahara, 1993B) describes other developments in the field. One of the best examples of tomography of a slab was presented by Zhao and Hasegawa (1993) for the Japan Islands.
Zhao et al. (1993) present a reflection seismic section from the Tethyan Himalaya showing a spectacular mid-crustal reflection which they suggest marks the fault along which the Indian plate is underthrusting Tibet. A deeper dipping reflector is thought to represent the Moho of the underthrusting Indian plate (Figure 2). Levander et al., 1994a performed a 700 channel seismic reflection/refraction experiment which imaged the Brooks range, Alaska, fold and thrust belt. The combined refraction-reflection acquisition used seismic velocities from the refraction analysis to perform stacking and migration of the reflected signals.
Observations of layered reflectors in the lower crust are often associated with actively extending continental crust; though an exception is the recent observation of laminated crust of the Fennoscandian craton (Babel working group, 1991). The source of such reflections in extending crust has been variously attributed to fluid or solid intrusions. Parsons et al. (1992) use waveform modelling to identify and interpret reflections from the lower crust at the transition zone from the Colorado Plateau to the Basin and Range province. They propose the laminated reflectivity is due to layers of positive impedance contrast, approximately 200 m thick or less, probably due to sill-like basaltic intrusions into the base of the crust, rather than low impedance, fluid filled, intrusions. Levander et al., (1994b) use maps of exposed crustal rocks to compute synthetic reflection seismograms, and suggest that deep crustal layering in regions of multiple scattering could be an artefact of processing. Seismic reflection studies from around the world are reviewed by Mooney and Meissner (1992) who emphasize multiple origins for crustal reflectivity.