A model showing the topological distribution, functions, and serological specifities of eight distinct, monoclonal antibody-defined epitopes on the tick-borne encephalitis (TBE) virus glycoprotein has been presented in a previous publication (F. X. Heinz, R. Berger, W. Tuma, and Ch. Kunz (1983). Virology 126, 525–537.) In the present report the influence of conformational change, chemical modification, and fragmentation on the antigenic reactivity of each epitope has been analyzed by the use of blocking enzyme immunoassays and “Western blotting.” One of the two major antigenic domains (A), composed of three different epitopes, completely lost its antigenicity upon incubation at pH 5.0 or by treatment with guanidine-HCl/urea, SDS, reduction and carboxymethylation, as well as by proteolytic (trypsin, α-chymotrypsin, thermolysin) and chemical (CNBr) fragmentation. The second major antigenic domain (B), however, defined by four distinct monoclonal antibodies, three of which are hemagglutination (HA)-inhibiting, neutralizing, and protective, was shown to be resistant to low pH, guanidine-HCl/urea treatment, and proteolytic cleavage of the native protein. Also, polyclonal immune sera from mice and rabbits contained antibody populations reactive with antigenic determinants which are resistant and others which are sensitive to conformational change and fragmentation. Glycoprotein fragments with molecular weights of about 9000, generated by proteolysis of the native protein, were immunoreactive with neutralizing and protective monoclonal antibodies (defining domain B) as well as with a polyclonal mouse immune serum. Thus, these fragments appear to contain antigenic determinants which are immunodominant on the native protein and play an important role in the induction of a protective immune response against TBE virus. In addition, these results show that antibody binding to antigenic domains which are topologically and structurally completely unrelated may result in neutralization and/or HA inhibition. As the presence of two receptor-binding sites is unlikely, different effector mechanisms may account for the effects of these antibodies. The antigenic reactivity of domain A is sensitive to the same treatments which also inactivate HA activity of TBE virus, whereas domain B is resistant. These treatments include a change of domain A induced by incubation at slightly acidic pH which also results in inactivation of virus infectivity. Antibodies to domain A therefore presumably block viral activities by direct binding at or near the putative receptor-binding site whereas antibodies to domain B may cause loss of biological activities by inducing a conformational change of the receptor-binding site.