ABSTRACT.To obtain accurate and reliable results for the expression of genes of interest using quantitative real-time polymerase chain reaction (RT-qPCR) techniques, it is necessary to normalize the data by comparing them to constitutive genes that exhibit uniform expression levels under experimental conditions. In this study, the stability of expression was evaluated for the following ten candidate reference genes in rice leaves (Oryza sativa L.) from the BRS Bojuru and BRS Ligeirinho genotypes that were subjected to salt stress (150 mM): actin 11 (ACT11), beta-tubulin (β-TUB), eukaryote elongation factor 1-α (Eef-1), eukaryotic initiation factor 4-α (eIF-4-α), E2 ubiquitinconjugating enzyme (UBC-E2), ubiquitin 5 (UBQ5), ubiquitin 10 (UBQ10), glyceraldehyde-3-phosphate dehydrogenase (GAPDH), TIP41-like, and cyclophilin. The stability of expression for the aforementioned genes was then compared to that of three LTP genes using UBQ10, Eef-1, and eIF-4-α as references. After analyzing the Reference genes for RT-qPCR in rice under salt stress expression levels using analysis of variance tests, the results indicated that UBQ10 was the most stable in all treatments (M = 0.404 and SV = 0.327). Furthermore, the eIF-4-α, TIP41-like, and cyclophilin genes exhibited the highest total coefficient of variation (CV = 269, 169.2, 179.2, respectively), which signifies that they exhibited the least stable expression. The expression levels of each candidate gene (LTP7, LTP10, and LTP13) were in contrast to the reference genes. Therefore, we concluded that UBQ10 is the best reference gene for RT-qPCR reactions under the experimental conditions. The expression analysis of LTP7, LTP10, and LTP13 confirmed the importance of validating reference genes to achieve accurate RT-qPCR results.