Shear wave splitting measurements along with other geophysical results provide critical constraints to construct numerous tectonic models in the Himalayas and southern Tibet, which are the locations of the prototype of active continental collision. Over the past decade, various models regarding the geometry and fate of the subducting Indian plate have been proposed based on many versions of shear wave splitting analysis from difference studies. In an effort to establish a uniform shear wave splitting database, we reassessed all the available data from over 100 seismic stations beneath the Himalayas and southern Tibetan Plateau. Our preliminary results of shear wave splitting measurements in the study area reveal clear evidence of significant anisotropy, with a splitting time of up to 1.5 s. When the PKS and SKKS in addition to SKS phases are used, remarkable azimuthal variations of the splitting parameters have been identified. Some of the splitting parameters can be interpreted as a combined effect of two layers of anisotropy. The top layer has a NE-SW fast direction which can be considered as the result of lower-crustal plastic flow, and the lower layer has a nearly E-W fast direction which can be interpreted as reflecting the asthenospheric flow associated with the motion of the Eurasian plate.