Final oocyte maturation is a key step for successful spawning and fertilization. Like most other vertebrates, teleosts have full-grown postvitellogenic oocytes in the ovary that are physiologically arrested at the G2/M border in the first meiotic prophase and cannot be fertilized. Shortly before ovulation, the oocytes must become fully mature, which involves breakdown of the germinal vesicle(GVBD), chromosome condensation, assembly of the meiotic spindle, and exclusion of the first polar body. Meiosis is again arrested at the second metaphase. The time period from the resumption of meiosis to the second meiotic metaphase has been referred to as final oocyte maturation. Knowledge of the molecular events and the role of various factors involved in final oocyte maturation is a focus of research in the field of reproductive biology. Quantitative real-time PCR (qPCR) is often used to quantify transcript abundance in such studies; however the technique is subject to considerable experimental error and variation. A stably expressed housekeeping gene is typically used as a reference gene to normalize this variation. However an ideal universal reference gene that is stable under all experimental circumstances has not been described. Researchers should validate their reference genes before performing qPCR analysis. We documented the expression of six candidate reference genes: 18S rRNA(18S), 28S rRNA(28S), cathepsin Z (CTSZ), elongation factor 1-α(EF-1α), glyceraldehyde-3-phosphate dehydrogenase (gapdh), and β-actin during 17α, 20β-dihydroxy-4-pregnen-3-one (DHP) induced final oocyte maturation in common carp (Cyprinus carpio). We used a range of software to evaluate the expression stability of these reference genes. Analysis with Bestkeeper revealed that EF-1α had the lowest CV% (2.65) and STDEV (0.71) values, but had the highest Bestkeeper index value (0.956) among the six candidate reference genes, suggesting that EF-1α was the best reference gene in the present study. Analysis with geNorm revealed that EF-1α and CTSZ had the lowest M (gene expression stability) values which indicated that they were the most stable genes among the six. Additionally, the value of pairwise variation of these two genes (V2/3) was much lower than the default (0.15), suggesting that EF-1α and CTSZ should be used concurrently as a reference gene pair to normalize the expression of the target genes. Consistently, analysis with NormFinder and RefFinder demonstrated that EF-1α was the most stable gene among the six candidates. Hence, EF-1α can serve as a reference gene to normalize the expression of target genes during final oocyte maturation induced by DHP in common carp. The mRNA expression of luteinizing hormone receptor gene(LHR) during final oocyte maturation induced by DHP was normalized by 6 candidate reference genes respectively, and the relative expression was varied significantly depending on the reference gene that was used. Selection of a stably expressed reference gene is critical for all qPCR analysis to ensure accurate target gene mRNA expression information。