In view of the recognised influence of smoking on the disease course of psoriasis and ulcerative colitis, and the association of these diseases with seronegative spondyloarthritis, we investigated a possible effect on outcome in ankylosing spondylitis. Thirtyone non-smokers, 12 exsmokers, and 10 smokers, with definite ankylosing spondylitis of a median duration of 20 years, were studied using clinical values (modified Schober, finger floor distance, total spinal movement), radiological (lumbar spine x-ray score), functional index, and laboratory assessment (ESR, CRP, Igs). There were statistically significant differences in the outcome between smokers and non-smokers for finger floor distance (p < 0.01), Schober test (p < 0.01), total spinal movement (p < 0.001), occiput-wall distance (p < 0.01), functional index (p < 0.01), stiffness (p < 0.01), and spine x-ray scores (p < 0.02). There was no statistically significant difference between the groups with regard to disease duration or age at onset. We conclude that smoking is associated with poor long term outcome in patients with ankylosing spondylitis.