A major earthquake occurrence zone means a place where M ≥6 events have occurred since the Holocene and similar shocks may happen again in the future. The dynamic context of the major earthquake occurrence zones in China is primarily associated with the NNE-directed push of the India plate, next with the westward subduction of the Pacific plate. The Chinese mainland is a grand mosaic structure of many crust blocks bounded by faults and sutures. When it is suffered from boundary stresses, deformation takes place along these faults or sutures while the block interiors remain relatively stable or intact. Since the Quaternary, for example, left slip on the Xianshuihe-Xiaojiang fault zone in southwestern China has produced a number of fault-depression basins in extensional areas during periods Q1 and Q2. In the Q3, the change of stress orientation and enhancement of tectonic movement made faults of varied trends link each other, and continued to be active till present day, producing active fanlt zones in this region. Usually major earthquakes occur at some special locations on these active fault zones. During these events, in the epicenter areas experience intensive deformation character- ized by large-amplitude rise and fall of neighboring sections, generation of horst-graben systems and dammed rivers. The studies on palaeoearthquakes suggest that major shocks of close magnitudes often repeated for several times at a same place. By comparison of the Chi-Chi, Taiwan event in 1999 and Yuza, Yunnan event in 1955, including contours of accelerations and intensities, destruction of buildings, and in contrast to the Xigeda formation in southwestern China, a sandwich model is established to account for the mechanism of deformation caused by major earthquakes. This model consists of three layers, i.e. the two walls of a fault and the ruptured zone intercalated between them. This ruptured zone is just the loci where stress is built up and released, and serves as a channel for seismic waves.