Recently, many exceptions have been reported that undermine the concept that the epitopes for thyroid-stimulating autoantibodies (TSAb) and thyrotropin-blocking autoantibodies (TBAb) lie within the N-terminal and C-terminal portions of the thyrotropin receptor (TSHR) ectodomain, respectively. Here we have used a new approach to examine the issue by attempting to neutralize autoantibodies with the purified, N-terminal 289 amino acids of the TSHR ectodomain (TSHR-289), essentially the A subunit. Immunoglobulin G (IgG) with TSAb activity from 10 patients with Graves' disease was preincubated with or without purified active or inactive TSHR-289. Active, but not inactive, TSHR-289 completely neutralized TSAb activity in all sera. We then tested the ability of active TSHR-289 to neutralize TBAb activity in two rare hypothyroid patients with pure TBAb activity lacking agonist activity. IgG from both patients was extremely potent in the TBAb assay. Unlike with TSAb, preincubation of the IgG with a large excess of active TSHR-289 (20 μg/mL) revealed a remarkable divergence. TBAb activity from one patient was totally neutralized whereas in the other patient TBAb activity was totally unaffected. In conclusion, using a new approach, we confirm that the major portion of the TSAb epitope in the 418 amino acid ectodomain lies upstream of residue 289 (within the A subunit). In contrast, TBAb that cause hyperthyroidism can be more heterogeneous, with epitopes that lie largely upstream (A subunit) or downstream (B subunit) of residue 289.