Abstract: Objective （ ） To compare the measured results of arsenic in urine by atomic fluorescence spectrometry AFS and - （ - ）， Methods inductively coupled plasma mass spectroscopy ICP MS and analyze the reasons of the difference. The samples WS/T 474-2015 Determination of Arsenic in Urine by Hydride Generation Atomic Fluorescence were pretreated according to Spectrometry， （ ∶ ∶ ∶∶ ，V/V/V） and digested with mixed acid nitric acid sulfuric acid perchloric acid=3 1 1 and then determined by - - AFS and ICP MS. The samples were diluted with 0.50% nitric acid and determined by ICP MS. The samples included urine ， ， （ arsenic quality control samples inorganic arsenic supplemented samples and organic arsenic arsenic choline and arsenic ） - betaine supplemented samples. Standard curve method was used to compare the results of AFS method and ICP MS method. Results （ ） （ ） The results of quality control samples by AFS method digestion and ICP-MS method without digestion were ， - within the range of reference values but the values obtained by AFS method were lower than those obtained by ICP MS method. - - - ， The recovery of AFS and ICP MS was 97.79% 100.82% and 99.55% 99.98% respectively. In the middle and high ， - （ P ） concentration groups the measured values of inorganic arsenic by AFS were lower than that by ICP MS all <0.01 . The （ ） - recovery of arsenic betaine and arsenic choline by AFS method digestion was only 2.17% 2.63%. The values of arsenic betaine （ ） - （ and arsenic choline measured by AFS method digestion were lower than those measured by ICP MS method without ） - （ ）（ P ）Conclusion digestion and ICP MS method digestion all <0.01 . The result of urine arsenic measured by AFS method - ， was lower than that measured by ICP MS method which may be related to the mixed acid digestion of AFS method. Keywords: ； - ； ； ； ； ；