Abstract: The feed sample (0.200 0 g to 0.500 0 g) was digested with 5 mL of HNO₃ under the mode of programmed temperature elevation. The digested solution was then concentrated to ca. 1 mL by heating at 100℃. The solution was then cooled and diluted to 50.0 mL with water, and 9 major and beneficial elements, i.e., K, Na, Ca, P, Mg, Fe, Cu, Zn and Mn, and 5 harmful trace elements, i.e., Cr, Pb, As, Cd and Hg, were determined in this sample solution by ICP-MS. MS interferences by isobaric elements were avoided by selecting the isotope with higher relative abundance as the analytical species for the analyte. For instance, ⁴⁴Ca was selected instead of ⁴⁰Ca as the determining species to avoid the interference of ⁴⁰Ar; ⁵⁶Fe was used instead of ⁵⁴Fe to avoid the interference of ⁵⁴Cr, and etc. Another kind of MS interference was the formation of poly-atomic ions, e.g. ArCl and poly-atomic ions formed between N and O. But concentrations of the 9 major elements were significantly higher than that of the polyatomic ions possibly formed, hence their interferences were neglected. While in the determination of the 5 trace elements, such interferences could be avoided by applying mathematical equation of interference correction. Besides, in the determination of Hg, Au was added to form Hg-Au amalgam in order to avoid the adsorption effect of mercury. Furthermore, for control and correction of signal drifts and matrix effects in the determination of the 14 elements, 8 elements, including Sc, Ni and etc., were added as internal standards separately in the relative determination. As shown by the testing results, linear relationships were found between values of the singal strength and mass concentration of 14 elements in definite ranges, and their lower limits of determination (10s/k) ranged from 1.65×10^-4 to 6.60 mg·kg^-1. Tests for recovery were made by standard addition method, giving values of recovery in the range of 83.0%-115%, and values of RSDs (n=7) ranged from 0.4% to 12%.