PURPOSE: To determine the optimal parameters of abdominal HASTE imaging by means of a comparison of intermediate and long TE (echo time). MATERIALS AND METHODS: We evaluated 30 consecutive patients who had undergone liver MR during a three-month period. Twelve patients were diagnosed as normal, four as having liver cirrhosis, and 14 were found to be suffering form hepatic hemangioma. On the basis of measured signal intensity of the liver, spleen, pancreas and gallbladder, and of fat, muscle, hemangioma, and background, we calculated the ratios of signal to noise (S/N), signal difference to noise (SD/N), and signal intensity (SI). Image quality was compared using these three ratios, and using two HASTE sequences with TEs of 90 msec and 134 msec, images were qualitatively evaluated. RESULTS: S/N ratio of the liver was higher when TE was 90 msec(p<.05), though S/N, SD/N and SI rations of the spleen, gallbladder, and pancreas -and of hemangiom- were higher when TE was 134 msec (p<.05). However, in muscle, all these three ratios were higher at a TE of 90 msec. SD/N ratio and SI of fat were higher at a TE of 134 msec. Overall image quality was better at a TE of 134mesc than at one of 90msec. CONCLUSION: A HASTE sequence with a TE of 134msec showed greater tissue contrast and stronger T2-weighted images than one with a TE of 90msec.