The Caatinga is one of the most degraded biomes in the country. The degraded area in this biome is estimated at more than 200,000 square kilometers. Thus, the objective of the present work was to evaluate the recovery of artificially degraded areas as a function of doses, sources, presence of litter and occurrence of degradation in a Red-Yellow Argisol after five years of degradation. For this, two experiments were carried out in neighbouring areas at the Rafael Fernandes experimental farm with a randomized block design with four replications, with plots of 4,0 m x 10,0 m. The degradation was done with a mat tractor, removing the 25-cm surface layer from the soil. The litter used was obtained from five preserved areas of the Caatinga, located in Mossoró-RN. The litter was obtained by removing the approximately 0 to 5 cm layer of the existing soil surface material from the sampled areas. The treatments evaluated in the first experiment were: soil without degradation and without application of litter, soil degraded and without application of litter and degraded soil + the application of 2, 4, 6, 8 and 10 t ha-1. In the second experiment, the effects of the application of 4 t ha-1 from the litter of five sites (sources) were evaluated, as well as the presence of litter and the occurrence of soil degradation. The results were compared by the Dunnett and Tukey tests, in addition to the regression analysis, in the case of experiment 1. Increasing litter dose determined increases in organic matter, nitrogen, calcium and magnesium contents, as well as the enrichment of values of base sum, CTC's and in base Saturation and decrease of aluminium saturation in the soil. There was variation of soil chemical attributes between litter sources only, for copper content. With the presence of litter, the chemical attributes improved, in relation to the dry matter and with degradation, the chemical attributes were inferior in comparison to the treatment without degradation