The element contents of wheat from four major wheat-producing regions of China were analyzed and used in multivariate statistical analysis to classify wheat according to geographical origin. The concentrations of 15 elements (Be, Na, Mg, Al, K, Ca, V, Mn, Fe, Cu, Zn, Mo, Cd, Ba, and Th) in 240 samples from the 2007/2008 and 2008/2009 harvests were determined by inductively coupled plasma mass spectrometry. The analysis of variance and linear discriminant analysis were applied to classify wheat origin, and the effects of region, variety, and harvest year on the element contents were analyzed in this study. It was concluded that the multielement analysis is a promising method to provide reliable origin information for wheat, although the element profiles and discriminant models were affected by wheat varieties, harvest years, and agricultural practices.