Western blotting (WB; immunoblotting) is a widely used tool for the serodiagnosis of Lyme borreliosis (LB), but so far, no generally accepted criteria for performance and interpretation have been established in Europe. The current study was preceeded by a detailed analysis of WB with whole-cell lysates of three species ofBorrelia burgdorferi sensu lato (U. Hauser, G. Lehnert, R. Lobentanzer, and B. Wilske, J. Clin. Microbiol. 35:1433–1444, 1997). In that study, interpretation criteria for a positive WB result were developed with the data for 330 serum samples (from patients with LB in different stages [n = 189] and from a control group [n = 141]) originating mostly from southern Germany. In the present work, the interpretation criteria for strains PKo (Borrelia afzelii) and PBi (Borrelia garinii) developed in the previous study were reevaluated with 224 serum samples (from patients with LB in different stages [n = 97] and from a control group [n = 127]) originating from throughout Europe that were provided by the European Union Concerted Action on Lyme Borreliosis (EUCALB). De novo criteria were developed on the basis of the reactivities of the EUCALB sera and were evaluated with the data for the samples from southern Germany. Comparison of all results led to the following recommendations: For WB for immunoglobulin G (IgG), at least two bands among p83/100, p58, p43, p39, p30, OspC, p21, p17, and p14 for PKo and at least one band among p83/100, p39, p30, OspC, p21, and p17b for PBi; for WB for IgM, at least one band among p39, OspC, and p17 or a strong p41 band for PKo and at least one band among p39 and OspC or a strong p41 band for PBi. WB with PKo was the most sensitive, and this strain is recommended for use in WB for the serodiagnosis of LB throughout Europe.