Crustal Structure of the Earth Toshiro Tanimoto 1. INTRODUCTION The boundary between the crust and the mantle was discoveredby Mohorovicic in 1909 under the European continent. Subsquent research in this century established the major differencesbetween the continentaland oceanic crust; a typical thickness for the continentalcrust is 30-50 km while a typical thickness for me oceanic crusts is 6 km. In terms of history the continental crust contains a much longer history of 4 billion years, whereas the oceanic crust contains at most 200 million yearsof history because of recycling of oceanic plates. Becauseof its long history, the continental crust has been subjected to various tectonic processes, such as repeated episodes of partial melting, metamorphism, intrusion, faulting and folding. It is thus easier to find systematic relationships between age and structure of oceanic crusts. However, the existence of hotspotsas well as changing patterns of plate motion complicate oceanic crustal structure. In this section, we assemble crustal thickness data from various tectonic provinces and discuss their implications. T. Tanimoto, Department of Geological Sciences, University of California. Santa Barbara, Santa Barbara, CA 93106 Present Address: T. Tanimoto, Tokyo Institute of Technology, Earth and Planetary Sciences, Ookayama 2-12-l Meguro-ku, Tokyo 152, Japan Global EarthPhysics A Handbook of Physical Constants AGU Reference Shelf 1 Copyright 1995 by the American Geophysical Union. 2. OCEANIC CRUSTS 2.1. Classic Subdivision and Mean Crustal Thickness The oceaniccrust is classically divided into three layers [521; Layer 1 is the sedimentary layer, whose thickness varies widely according to sedimentsources, and Layer 2 has a thickness of 1 S-2.0 km and P-wave velocity of 4.5 5.6 km/s and Layer 3 has a thickness of 4.5-5.0 km and P- wave velocity of 6.5-7.0 km/s. Combined thickness of layer 2 and 3 is often referred to as the oceanic crustal thickness and we adopt this convention. For the continental crust, we define the thickness from the surface to the Mohorovicic discontinuity (Moho). The interpretation of oceanicvelocity structure is based on two independent sources of information; one is by comparison of seismic velocities in laboratory measurements of rocks from ocean drilling cores with the velocities measured in seismic refraction experiments. The other is based on analogy with structuresin ophiolite complexes. A commonly held view (e.g.,[65] ) is that Layer 2 starts with extrusive volcanic rocks at shallow depths which grade downward from pillow basalts into sheeted dikes. There is a transition zone at the top of Layer 2 which shows inter-fingeringof extrusive basaltic rocks and sheeted dikes. Layer 3 has properties appropriate to the massive to cumulate gabbro layer seen in ophiolite complexes. The top of Layer 3 has a transitional layer which shows interfingering of sheeted dikes (at the bottom of Layer 2) and isotropic gabbro (at the top of Layer 3). The isotropic gabbro layer is underlain by layeredgabbroand harzburgitesuccessively. The traditional seismic modelling used a few homogeneous layers, which has been replacedby layers which contain velocity gradients in recent studies (e.g., [66]). If the assumptionof a few stack of homogeneous 214