The aetiology of morphoea and lichen sclerosus et atrophicus is still unknown. Since the detection of Borrelia burgdorferi (B. burgdorferi) as the causative agent of Lyme disease, there has been debate about a possible association between B. burgdorferi and morphoea. Initial serological and cultural studies showed controversial results. The introduction of polymerase chain reaction (PCR) initially suggested an association between B. burgdorferi and morphoea. We reviewed the literature on B. burgdorferi (specific serology, immunohistology, culture, lymphocyte stimulation and DNA detection by PCR) since 1983, using Medline and Current Contents. Histological and immunohistological detection of B. burgdorferi was reported in 0-40% (20 of 82) of the cases with morphoea and in 46-50% (17 of 36) of the cases with lichen sclerosus et atrophicus. Cultivation of spirochetes from lesional skin succeeded in five patients (five of 68) with morphoea, but failed in patients with lichen sclerosus et atrophicus. In Europe and Asia, serological detection of antibodies against B. burgdorferi was described in 0-60% (138 of 609) of patients with morphoea and in 19% (six of 32) in the U.S.A. For lichen sclerosus et atrophicus 0-25% of the published cases (three of 23) in Europe and Asia were seropositive. DNA from B. burgdorferi was detected by PCR in 0-100% (17 of 82) of the tissues of patients with morphoea in Europe and Asia, but not a single case among 98 patients was reported to be positive from the U. S.A. In Europe and Asia, borrelial DNA was detected in 0-100% (nine of 28) of the cases with lichen sclerosus et atrophicus, whereas in the U.S.A. none of 48 patients was positive. There are two possible explanations for these contradictory findings: the most likely is that B. burgdorferi is not a causative agent for morphoea. Another possible explanation could be that a subset of morphoea is caused by a special subspecies of B. burgdorferi that is present in Europe and Asia but does not occur in the U.S.A.