As has previously been reported by many investigators the anticentromere antibody is considered to be a useful serologic marker for the CREST (calcinosis, Raynaud’s phenomenon, esophageal involvement, sclerodactyly and telangiectasia) variant of systemic sclerosis. However, this antibody also appears in other conditions. By screening antinuclear antibody tests using HEp-2 cells as substrate, 39 patients were shown to have the anticentromere antibody. These patients were divided into 4 groups: group 1 = 17 patients with systemic sclerosis; group 2 = 9 patients with Raynaud’s phenomenon alone; group 3 = 7 patients with other connective-tissue diseases, and group 4 = 6 patients with conditions other than those present in groups 1–3. Follow-up over years revealed that some patients suffering solely from Raynaud’s phenomenon (group 2) developed the symptoms of systemic sclerosis. In contrast, of the patients who did not have Raynaud’s phenomenon (group 4), none developed any symptom suggesting systemic sclerosis. We suggest that the simultaneous presence of Raynaud’s phenomenon and the anticentromere antibody predicts the occurrence of systemic sclerosis. In contrast, the presence of the anticentromere antibody alone cannot necessarily be used to predict the development of systemic sclerosis.