Cistanche deserticola, an endemic species in China, has been one of the grade II national key conservation rare and endangered plants. The spectra of cultivated and wild C. deserticola samples were determined by Fourier transform infrared (FTIR) spectrometry. Based on the fingerprint infrared spectrum from 450 to 2 000 cm(-1), C. deserticola samples were rapidly classfied and closely studied by using the method of clustering analysis. Results showed that although there were tiny differences between the spectra of different origin, including the wild and cultivated C. deserticola samples, these samples could be successfully classified by soft independent modeling of class analogy (SIMCA). Recognition rate and rejection rate of different C. deserticola samples were up to 90%. When testing with the blind sample which the authors picked out from the chosen samples, the accuracy of clustering reaches up to 95%. On the whole, combined with clustering analysis, FTIR provides a effective way to evaluate the origin of the Chinese medicines rapidly and undamagedly.