Recently, xenograft is being increasingly used in spinal fusion to reduce complications such as pain or bleeding following autograft bone grafting. The purpose of this study was to compare the fusion rate and clinical results between autograft and xenograft in 69 patients who had posterolateral lumbar fusion with instrumentation from March 1989 to April 1997. The xenograft was mixed with autogenous bone chips obtained from decompresion. Fusion and clinical results were evaluated with Furguson's method and Modified Smiley-Webster' scale, respectively. The mean follow-up time was 26 months with a minimum of 12 months and a maximum of 41 months. The results are as follows: fusion rates of autograft were 64.6% in Furguson grade A, 25.8% in grade B, 6.4% in grade C and 3.2% in grade D. Xenograft was 15.7% in grade A, 47.6% in grade B, 23.6% in grade C and 13.1% in grade D. Clinical results of autograft were excellent in 29.0%, good in 61.3%, fair in 6.5% and poor in 3,2%. Xenograft was excellent in 23.8%, good in 57,8%, fair in 15.8% and poor in 2.6%. Autograft showed a higher fusion rate of 90.4% than xenograft of 63.3% in grade A and B (P=0.012). Clinical results were not different between autograft and xenograft groups (P=0.494). However, the xenograft group showed lower fusion rate than the autograft group. The results indicated that xenograft is less useful for posteolateral fusion of the lumbar spine.