Abstract: In view of the narrow detection range, baseline drift and spectral line interference caused by high salinity of HJ 1082-2019, a method mentioned by the title was proposed. The sample (5.0 g) was taken, and 50 mL of alkali extracting solution containing 0.5 mol·L^-1 sodium hydroxide and 0.28 mol·L^-1 sodium carbonate (pH>11.5), 0.4 g of magnesium chloride and 0.5 mL of phosphate buffer sohution (pH 7.0) were added. A 40 mm magneton was added into the above mixture, and stir was conducted at 95 ℃ for 60 min at rate of 600 r·min^-1. After cooling, the mixed solution was centrifuged at speed of 4 000 r·min^-1 for 5 min. An aliquot (25 mL) of filtrate was taken, and its acidity was adjusted to pH 7.0-7.5, then the volume was made up to 50 mL with water. An aliquot (1 mL) was diluted 10 times with water, in which hexavalent chromium was determined by inductively coupled plasma atomic emission spectrometry. As shown by the results, the linear range of standard curve for hexavalent chromium was within 2 mg·L^-1, and the detection limit (3.143s) was 0.096 mg·kg^-1. The determined values of 5 standard substances for the analysis of hexavalent chromium in soil were within the uncertainty ranges of the identified values, and RSDs (n=6) of the determined values were in the range of 1.1%-3.7%. Test for recovery was made on the actual samples according to the standard addition method, giving results in the range of 90.0%-112%.