There are many obvious similarities between Lyme disease and syphilis. The major ones are their spirochetal etiology, the ability of the spirochetes to stay alive in human tissue for years, occurrence of clinical manifestations in stages, early disease in the skin and later disease in the brain, and susceptibility to antibiotic treatment. Thus, one can assume that many of the same lessons learned from the centuries of experience with syphilis will apply to Lyme disease. One of these lessons that should be constantly borne in mind is that spirochetal disease of the brain can mimic many other neurological diseases. Thus, the "effective clinician" must take special care to consider Lyme disease primarily because of the excellent response to antibiotics early in its course in relationship to some of the diseases it mimics. Lyme meningitis, occurring in the "second stage" of the disease, usually is fairly easily recognized because it occurs in the summer or early fall, often is associated with ECM or a recent history of it, and has a characteristic clinical picture of lymphocytic meningoradiculoneuritis. Many patients with Lyme meningitis or ECM have very mild symptoms, and it is likely that a large percentage of patients go undiagnosed and untreated. The frequency of progression of these patients to third-stage disease is unknown but may be quite high. This can be inferred from a similar situation in the other major late manifestation of Lyme disease: Lyme arthritis. A large number of patients present with joint involvement as their only manifestation of Lyme disease. Similarly, patients may present with symptoms of third-stage Lyme disease affecting the CNS, but they may not be recognized because of the lack of earlier stages usually associated with the disease. Thus, serology has become a very important tool for identifying patients exposed to B. burgdorferi. At the present time, serologic tests are the key to diagnosis of Lyme disease in its later stages, since, as in neurosyphilis, cultures and tests for antigen have not proven useful. Lyme arthritis and acrodermatitis atrophicans (ACA) both are associated with quite high antibody titers to the organism, while the test is understandably unreliable for identification of patients with ECM. Antibody titers in Lyme meningoradiculoneuritis are generally positive but often are not as high as those in ACA or arthritis. The antibody response in serum in CNS Lyme disease seems to be related to the presence of other manifestations; patients who have had both arthritis and CNS disease have quite high titers, while those with only CNS disease sometimes do not.(ABSTRACT TRUNCATED AT 400 WORDS)