It is not easy to distinguish pine hybrids and their parents using their morphological characteristics. The traditional method (chemical analysis or molecular marker technology) of identification is complicated, time-consuming and costly. The truth of pine hybrids must be confirmed during the program for forest tree breeding and popularization of good pine varieties. Therefore, partial least square-discriminant analysis(PLS-DA) regression modeling was constructed using the data of near-infrared spectroscopy of the pine foliage for 2 pine hybrids and their 3 parent species in southern China. The results showed that PLS-DA has a high accuracy rate , allowing a secure identification of 2 hybrids and their 3 parent species. It is clear, from the results, that near-infrared technology would be used to identify hybrids and purebreds in pine and it possesses the advantage of being simple，inexpensive，and easy to operate.