Abstract: In order to find out a fast measure method of the content of soybean fatty acid based on near infrared spectroscopy. There were 25 soybean samples collected in 2005 from the major soybean areas in Heilongjiang province were used to measure fatty acid content. Using Perten8620 type NIR to scan the collected samples and get the spectrum data. Then all data were analyzed by multi linear regression(MLR) and partial least square(PLS). Results show that NIR between 1700 nm and 2300 nm is applicable to the determination of soybean fatty acid, and PLS can produce a better prediction than MLR. Filer group selection was discussed in this study, the results show that the measurement accuracy will be influenced by filer group.