It has been widely reported that disorder of lipid metabolism has a close relationship with type Ⅱ diabetes mellitus(T2DM)[l].In order to explore the main lipids that distinguish T2DM from the normal,an optimized protocol was established and validated to seperate fatty acids in rat serum by GC/MS.Serum were obtained from GK rats (n=23)and wistar rats(n=l8),which were considered to be the animal models of type Ⅱ diabetes mellitus(T2DM).Serum samples were derivatived by NaOH-CH3OH and H2SO4-CH3OH to extract the esterified fatty acids(EFA) and free fatty acids(FFA)[2],respectively.Subsequently,extracted by hexane,and analyzed by GC/MS after mixed with 50uL hexane.18 fatty acids were well separated,as shown in Figure.l and Figure.2.And they were identified and validated by authentic standards.The result showed that C16∶0,C18∶1,C18∶0 was the main components of EFA,and C16∶0,C18∶1,C18∶2 was the main components of FFA.Since no obvious differences were discovered between health rats and T2DM rats,partial least squares-linear discriminant analysis(PLS-LDA)[3] was used to identify the differential metabolites and pertinent potential biomarkers in response to the type Ⅱ diabetes mellitus.The results suggested that small molecules of serum lipid metabolism have an obvious reflection on type 11 diabetes mellitus(T2DM),with the application of GC/MS.