This work is part of a wide study carried out at the EEPF «Indio Hatuey» in which the nutritive composition of mulberry forage is studied under different cultivation conditions and utilization moments. The main objective is to use the NIRS for carrying out predictions of the chemical composition and the energetic value of mulberry forages, remarkably reducing the costs of the analyses. Thirty-one samples of forages from different seasons (rainy and dry), different fertilization rates, and different fractions (leaf and fresh stem) were used, attempting to get a wide range of variability for performing the calibrations. The parameters determined were: dry matter (DM), crude protein (CP), crude fiber (CF), ethereal extract (EE), organic matter (OM), neutral detergent fiber (NDF), acid detergent fiber (ADF), acid detergent lignin (ADL) and crude energy (CE). An InfraAlyzer 450 (Bran+Luebbe), with 19 filters was used. The calibrations were carried out with 25 samples and six samples were used for validating the calibrations obtained. The software used in the mathematical processing and statistical analysis was the one supplied with the equipment (IACAL P01) and the calibrations were carried out by means of Multiple Linear Regression (MLR). The results obtained in the chemical analysis indicate that the samples used for the calibration and the validation covered similar ranges. The best calibrations were obtained using three filters. All the calibrations obtained showed satisfactory determination coefficients (R2) and which varied from 0,907 to 0,995 for crude protein. The calibrations that show CVCd»10% are considered to be good. In our calibrations the values obtained have been in all cases lower than 6,7%. Another criterion that determines the validity of the calibrations is SEC/SDd»0,3, in the nine calibrations developed for the mulberry forage, the values obtained in the performance of this index oscillated between 0,03 in the calibration of the CE to 0,32 in that of DM. The results obtained in this study show the possibility of NIRS calibrations for the prediction of the chemical composition and the nutritive value of the mulberry forage, although it would be convenient to widen the number of samples of the different calibrations with the objective of obtaining higher hardiness.