Abstract: 5.0 g of the sample of infant formula milk powder, were hydrolyzed in an ammoniacal medium containing 2 mL of ethanol, 4 mL of H₂O and 20 mL of aq. ammonia at 75℃ in a constant temperature water bath for 40 min with shaking at each 5 min intervals. The mixture, after the hydrolysis was ended, was cooled to room temperature and mixed with 10 mL of ethanol. The hydrolyzed fat in the reaction mixture was then extracted with 30 mL of a mixed solvent of ether and petroleum ether with a volum ratio of 1:1 by shaking for 10 min. After centrifuging for 5 min, the upper etheric layer was collected and kept in a flask. The lower layer was extracted with the ether mixture again for 2 times in the same way described above. The upper etheric layers were collected and combined in the same flask. The combined etheric layer was then evaporated to dryness under swirling. The fatty compounds in the milk powder sample were then remained in the residue. An aliquot of 60.0 mg of the residue was taken and dissolved in 4 mL of iso-octane. Then the fatty acids in the solution were methyl-esterifred by shaking for 30 s with 200 μL of 2 mol·L^-1 potassium hydroxide in methanol. The reaction mixture was kept under steady condition until it was turned to clear and transparent state, thus to have the esterification ended. 1 g of sodium bisulfate was added to the solution for neutralization of the potassium hydroxide, and after centrifuging for 3 min, the solution of upper layer was taken for GC-MS analysis under the instrumental working condition. 1.0 μL of the solution was introduced and the GC-separation was carried out by using the highly polar HP-88 capillary chromatographic column under the mode of programmed temperature elevation from 130℃ to 240℃. In the MS analysis, the EI ionization source and whole scanning mode in the range of m/z 45 to 450 were adopted for qualitative and quantitative determinations. Twelve trans-fatty acids were used as analyte in the related tests. Linear relationships between values of peak areas measured and mass concentrations of the respective trans-fatty acid methyl ester were found in definite ranges, and the detection limits (3S/N) found were between 0.42-1.45 mg·L^-1. Recovery was tested by standard addition method, giving recoveries ranged from 75.5%-107%. Values of RSDs (n=5) found were in the range of 1.1%-8.5%. Fatty acids were determined by area normalization method. The proposed method was used to analyze 3 infant formula milk powder samples, giving contents of saturated fatty acids over 50% and of trans-fatty acids all less than 0.5%, and showing that these milk powder were in conformity to the GB standards of our country.