OBJECTIVE： To establish a method for content determination of N-（E）-p-coumaroyltyrosine in leaves of Abrus cantoniensis， and to optimize its purification technology. METHODS： HPLC method was adopted for the content determination of N-（E）-p-coumaroyltyrosine in A. cantoniensis. The determination was performed on Hypersil BDS C18 column with mobile phase consisted of 0.1％ formic acid water （A）-methanol （B） （gradient elution） at a flow rate of 1.0 mL/min. The column temperature was set at 25 ℃. The detection wavelength was set at 300 nm， and sample size was 10 μL. Using polyamide resin as material， the yield of N-（E）-p-coumaroyltyrosine as indicators， single factor test was used to optimize the purification technology of N-（E）-p-coumaroyltyrosine， such as concentration， sample size， stationary adsorption time. RESULTS： The linear range was 2.575-51.50 μg （r＝0.999 9） for N-（E）-p-coumaroyl-tyrosine. The limit of quantification was 0.000 618 μg， and the detection limit was 0.000 129 μg. RSDs of precision， stability and reproducibility tests were all lower than 3％. The recoveries were 97.04％-102.43％ （RSD＝2.06％， n＝6）. The optimal purification technology was as follows： the concentration of the sample solution was 0.04 g /mL （by the leaves of A. cantoniensis）； sample capacity 50 mL； the sample flow rate was 1.0 mL/min； the stationary adsorption time was 20 min； the eluting solvents were ammonia containing water （containing 0.1％ acetic acid）， 20％ ethanol （containing 0.1％ acetic acid） and ammonia（pH 10）. Average yield was 98.94％，average dry paste content was 61.17 mg/g， and average dry paste purity was 19.73％ by optimal purification technology. CONCLUSIONS： Established method is simple， accurate and stable. The optimized technology is stable and feasible.