White clover is one of the most important forages in the world, with high nutritive value and crude protein content. Crude protein traits of white clover germplasms was selected using SPAD and near infrared reflectance spectroscopy. The SPAD value was measured by Chlorophyll Meter SPAD-502, and was used to evaluate the crude protein of white clover. In the vegetative period, there was a positive relationship between SPAD value and foliar protein content (y=0.422x+4.984， R2=0.737), but in the flowering period, there was a negative relationship between the two indexes (y=-0.345x+37.50， R2=0.711). Crude protein content of white clover germplasms was predicted using near infrared reflectance spectroscopy with PLS regression and the model was validated by cross validation and external validation. The results showed that the correlation coefficient of cross validation, the RMSECV, and the correlation coefficient of external validation are 0.904, 0.988%, and 0.987, respectively. NIRS model of white clover crude protein content has good accuracy and precision. FT-NIRS was more accurate than SPAD. NIRS is feasible as a rapid analysis method, and can be used in the selection and breeding of white clover germplasms to improve the breeding efficiency.