The purpose of this study is to determine the effects of vibration on primary (e.g. plasma cortisol), secondary (e.g. plasma glucose, aspartate aminotransferase (AST), alanine aminotransferase (ALT), Na+, K+ and Cl- and tertiary (e.g. mortality) stress responses in cultured eel, Anguilla japonica. For this purpose, three groups (including one control group and two stress groups) were set up. The control group was made exposed to vibration corresponding to 48 decibel (dB, V) (produced using electric vibrators) for 15 minutes per hour every day, and the two stress groups was made exposed to vibration corresponding to 58 and 68 dB (V) (produced using the same electric vibrators), equally, for 15 minutes per hour every day. Blood was sampled at day 0 (before starting vibration stress tests, BS), and days 1, 3, 5, 7, 9 and 11 (after starting vibration stress tests). As a result, plasma cortisol showed trend to continuously rise by consecutive stress from 4.1±0.1 ng/ml in BS. In 48 dB group (control), cortisol showed the highest level with 7.6±0.9 ng/ml after 7 days (p <0.05), but at 9 and 11 days was not significantly compared with BS level. In 58 dB group, the cortisol showed the highest level with 43.1±4.8 ng/ml after 1st day. Cortisol of 68 dB group increased significantly during the experimental period (14.4±2.3~32.0± 5.7 ng/ml) (p<0.05). In 58, and 68 dB groups during the experimental period differed significantly compared to 48 group (p <0.05). Glucose in 48 dB were increased from 42.0 ±5.7 (BS) to 52.5±2.1 (1 day), the level was not significantly from 1 to 11 days. Glucose in 58 and 68 dB groups was increased significantly than BS during experimental period (p <0.05). K+ in 68 dB increased significantly (p <0.05) from 2.3±0.2 mE/ql (BS) to 3.3± 0.5 mE/ql at 5 days. In 48 and 58 dB groups during the experimental period differed significantly (p <0.05). Na+ and Cl- levels were not differed significantly during the experimental period. AST and ALT in 58 and 68 dB groups showed trend to continuously rise by consecutive stress. At 7 and 9 days in AST, between 48, 58 and 68 dB groups differed significantly (p <0.05). In 48, 58 and 68 dB groups at 1 day, blood hematocrit increased significantly higher than BS. The 11 days after vibration stress, the mortality in 48, 58 and 68 dB groups was 1.1, 5.1 and 5.8%, respectively. The present results have shown that A. japonica exhibited ''typical'' physiological responses when exposed to chronic vibration stress. These data suggested that chronic vibration stress caused substantial stress in the fish; especially the persisting elevated plasma AST and ALT levels observed would be expected to adverse effect. In conclusion, chronic vibration stress could greatly affects the hematological characteristics in A. japonica.