Resumo. ABSTRACT Currently, it is aimed to use substrates composed primarily of organic waste in order to reduce costs with chemical fertilizer. In that sense, the experiment was to evaluate different types of substrates for the emergence and early development of seedlings of Cassia grandis L. f. The experiment was carried out in a greenhouse at the Departament of Vegetable Sciences of Federal Rural do Semi- Árido (UFERSA), in Mossoró, Rio Grande do Norte state, Brazil. The substrates used were: sand + carbonized rice husk (in ratios of 1:1, 1:3 and 3:1), plant sand + wood dust (in ratios of 1:1, 1:3 and 3:1), plant sand + coconut fiber (in ratios of 1:1, 1:3 and 3:1), plant sand + chestnut bark (in ratios of 1:1, 1:3 and 3:1) and plant sand (control). The experimental design was a completely randomized design with 13 treatments (combinations of substrates), with four replications of 25 seeds and the means to test for Skot-knot (pr <0.05). We evaluated the following variables: plant emergence, emergence speed, plant height, root length and weight of the aerial part of the plant, root dry mass and total dry mass. According to the results, the substrate plant sand + sawdust (1:1) and plant sand + coconut fiber (1:3) showed higher values, providing better results in emergence, weight of the aerial part of the plant, length of the aerial part of the plant, root dry mass and root length. It was found that the substrate compounds sand plant and chestnut bark (1:3) is harmful to the emergence of seedling of Cassia grandis .