Since Vening-Meinesz's realization that the East African Rift represented an extensional, not compressional, feature and since the widespread acceptance of plate tectonics two decades later, research on the nature and causes of extensional tectonism within continental lithosphere has intensified. Among the manifestations of extensional processes affecting continental lithosphere are passive margins (Atlantic margin), discrete intracontinental rift zones (East African Rift), diffuse rifts (Basin and Range Province), strike-slip dominated rifts (Dead Sea Rift), and rifts in zones of regional compression (Tibetan grabens). Although no two rift zones are alike, continental rifts can generally be characterized by normal faulting with subsidiary strike-slip faulting, lithospheric thinning which outpaces crustal stretching, varying amounts of alkaline magmatism, heat flow that is locally elevated near faults and magmatic centers, and crust that has experienced magmatic underplating and some amount of magmatic intrusion. Most aspects of rift related deformation can be explained in terms of three parameters: (1) lithospheric and (sometimes) asthenospheric thermal structure, (2) lithospheric (particularly crustal) rheology, and (3) temporal factors such as the absolute age, timing, and rate of extension. The interaction of these physical parameters determines the eventual outcome of rifting (failure or progression to complete continental breakup), the patterns of subsidence and uplift, and the mode of extensional deformation. Modes of rifting (the lithosphere's mechanical response to extensional stress) can be broadly divided into pure shear, simple shear, and lower crustal flow mechanisms.

In a general sense, these categories of rift mechanisms can account for observations at rifted margins, in the Basin and Range Province, and at metamorphic core complexes respectively. The mechanisms of continental lithosphere rifting (the effects) are here distinguished from the processes which actually drive extension (the causes). Following the terminology of previous authors, the causal processes are categorized as either passive or active: Passive processes originate at plate boundaries or in response to convective drag on the base of the lithosphere, while active processes are seated in the sublithospheric mantle and typically involve the interaction of mantle plumes and the lithospheric plates. A peculiar set of factors, including the local stress state, the retreat of a subducting plate, and convective instability of thickened crustal roots, may drive extension in compressional settings like Tibet. In the future, rifting studies are likely to focus on further clarification of the active and passive rifting terminology, better constraints on deformation rates and lithospheric rheology and thermal structure, along-axis segmentation of rifts, and the analogies and differences between rifting on Earth and nearby planets. ¿ American Geophysical Union 1995