Endocrine-disrupting chemicals (EDCs) pose a serious risk to the health of wildlife. Nonylphenol (NP), one of the most conerned EDCs, is a degradation product of nonylphenol ethoxylates (NPEs), which has been extensively used in industrial and household production and is commonly released from industrial and municipal sewage treatment plants into the aquatic environment. In recent years, studies demonstrated that NP is capable of mimicking the biochemical behavior of natural estrogen and interacting with the estrogen receptor. Behavior repertoire is the cumulative manifestation of biochemical, physiological and environmental cues, reflecting the outcome of an animal′s integrated physiological response and biochemical alterations resulted from contamination. Inappropriate behavioral performance caused by EDCs is of high ecological significance, as toxicants can often impair or even eliminate the behavioral patterns that are necessary for the fitness and survival of the whole population. Courtship behavior in fish is one of the key endpoints for assessing the effect of EDCs. If courtship behavior is suppressed by EDCs, spawning potential would be depressed and, subsequently, population structure and dynamics also might be disturbed. Thus, it is important to evaluate the effects of EDCs on courtship behavior of fishes. However, a standard protocol for assessing the effect of EDCs on courtship behavior in zebrafish (Danio rerio) has not yet been established. The present work examined the effects of NP on courtship behavior and reproductive success of zebrafish. Fish were randomly divided into six groups. Five groups were receiving 0.1, 1, 10, 50 and 100 μg/L nominal concentrations of NP for 40 d, respectively. A sixth control group was given the same treatment as the other five groups, but received acetone only. The nominal concentration of acetone in waterborne didn’t exceed 10 μL/L. The contaminants were administered via the water, as 50% of the water was daily exchanged and replaced with water contaminated with NP. Courtship behavior was observed during a 15 min period. Time of courting, average time of each courting, time of courting that exceeds 5 sec., frequency of courting, frequency of courting that exceeds 5 sec. and frequency of courting that less than 5 sec. were used to quantify the courtship behavior performance of zebrafish. The fecundity and fertility were studied to assess reproductive success. Significant differences in courtship behavior and reproductive success were observed in different NP exposure groups. The time of courting, fecundity and fertilization rate were significantly inhibited in 100 μg/L NP exposure group (P﹤0.05). A positive correlation (P﹤0.05) was found between the fecundity and the time of courting, including the total time of courting (R=0.637), the average time of each courting (R = 0.414) and total time of courting that exceeds 5 sec (R=0.619), whereas no significant correlation (R = 0.009, P = 0.951) was found between the fecundity and the frequency of courting. It was indicated that the effects of NP on fecundity of zebrafish may partly be attributed to the shortened time of courtship. Time of courtship might be promising biomarker or indicator of NP contamination in aquatic ecosystem.