Seismic reflection images of the upper crust of a large part of the Archaean Yilgarn Block in Western Australia contain very few reflections, implying that the upper crust is either compositionally homogeneous and largely unstructured or so completely folded and faulted that no laterally continuous surfaces remain to reflect seismic energy back to the surface. Seismic refraction data have, in the past, also been interpreted to show a homogeneous upper crust, at least at the broad reconnaissance scale sampled by previous experiments, which could not resolve structure or variations in composition over several kilometres or even tens of kilometres. Refraction data from the Pilbara Craton were similarly interpreted. However, in the light of more recent data with better spatial and temporal sampling from other shield areas, including the Kimberley Block and the Baltic Shield in Russia, the refraction data from the Yilgarn Block and Pilbara Cratons can be seen as consistent with a crust with more detail than was previously thought. The long offset data from the Kimberley Block and Baltic Shield, although different in detail because of geological differences, have some important similarities. Both data sets contain bands of wide-angle reflections that are laterally continuous, but made up of several individual wavelets that have less lateral continuity and appear to constructively and destructively interfere with each other. This effect has been modelled for the Baltic Shield, using the variation of velocity with depth measured in basement penetrated in the Kola Superdeep Bore Hole, and is attributed to the effects of seismic energy refracted through and reflected off thin (200-500 m) layers of low-velocity (5 .50 km.s-1) to high-velocity (6.60 km.s-1) rock interbedded in an upper crust with a median velocity less than 6.2 km.s-1. Thus, although the upper crust of the Kimberle y and Yilgarn appears transparent to seismic reflection imaging, it is likely to contain heterogeneities comparable to the region of the Kola Superdeep Bore Hole, at least in terms of seismic velocity and density. The Pilbara Craton probably has similar heterogeneities. No estimate has been made of the relative distribution of vertical and horizontal heterogeneities. Despite the heterogeneities, original estimates of upper crustal seismic velocity are probably still a good indication of the average velocity in the Yilgarn Block and Pilbara Craton.
