Purpose: This study was to investigate the effects of different whole-body vibration (WBV) on the surface electromyography (EMG) activity of the quadriceps and performance during the 30-s Wingate test on a rowing ergometer. Methods: Fourteen elite male rowers (height, 175.6±4.7 cm; body mass, 76.4±5.2 kg) were assigned to four treatments by randomized and counter-balanced design, which including the high-frequency high-amplitude (HFHA; 30 Hz, 0.6 mm), low-frequency low-amplitude (LFLA; 15 Hz, 0.375 mm), high-frequency low-amplitude (HFLA; 30 Hz, 0.375 mm) and non-vibration exercise (CON; 0 Hz and 0 mm). Maximum voluntary contraction (MVC) of knee extensors was measured before the treatments, and then the athletes completed the 30-s Wingate test on the rowing ergometer immediately after the 10×1 min (1 min rest interval) vibration or no vibration stimulation. The EMG activities of the quadriceps were measured via the surface EMG and the power output of the screen on the rowing ergometer was recorded using a video camera. Results: No significant differences were identified in both peak power and average power output among the treatments (p＞.05). There were no significant differences on the root-mean-square EMG (EMGrms) of the dominant leg among the treatments (p＞.05). However, the HFLA had significantly higher EMGrms in the vastus lateralis of the non-dominant leg at the peak power output than the HFHA and CON treatments (p＜.05). Moreover, the quotient of EMGrms/peak mechanical power had no significant difference among the treatments (p＞.05). Conclusion: This study suggested that a single bout of WBV could not improve the anaerobic exercise performance, and the HFLA treatment might reduce the efficiency of recruited motor units on the non-dominant leg during the anaerobic power test in elite male rowers.