This is the first time when garnet was discovered in calcalkaline volcanic rocks of Mesozoic era in western Qinling, the composition of which is almost in coincidence with that in data reported from abroad, that is, the garnet is an almandine characterized by abundant Fe, less Mg and rare Mn, but of higherproportion of Ca. Almost all of the garnet in the rocks are the corroded ones and surrounded by euhedral-subhedral plagioclases(some are with zonary structure), which makes us believe that theseplagio clases are not the reaction rims between garnet and magma, but the products of magmatic crystallization. The crystallizing mechanism is that while the uprising magma was corroding early crystallized garnet, Ca and At contents increased around it so that it helped to form the plagioclases around the residual garnet. In the volcanic rocks there are no xenolithes of metamorphic, plutonic rocks and other minerals with the same crystallizing conditions as garnet; and in the garnet crystal there are some silicate glass inclusions, etc; and considering the data which have been published and the geologic background of the Mesozoic volcanic rocks, we think that the garnet is not the xenolith crystal of the magma, but the eary-crystallized mineral of the same magma as the mother rock. And based on that we infer that the magma which formed the rock originated from the bottom of the lower crust at a depth of more than 50 km.