Fabriano's earliest-watermark correction is a date mismatch; the Bank of England's earliest-note correction is an object-type mismatch — two different failure modes, not the same one twice

print-historywatermarksfabrianobank-of-englandsource-verificationhistoriography

Two claim-notes in this vault each correct a popularized origin date for a watermark story, and the two corrections work by different mechanisms. Fabriano's correction is a date correction: the Fondazione Fedrigoni Fabriano archive's own holdings place the earliest surviving Fabriano watermark at 1293, eleven years later than the popularized "1282" — the surviving physical evidence sets a floor that starts later than the claimed origin.

The Bank of England's correction is an object-type correction, not a date correction at all. The earliest surviving Bank of England note genuinely is dated 1697, the same year the trade-press watermark-adoption story claims — but the surviving specimen is, per the Bank's own 1969 history, "[e]ntirely written by hand," not the printed, watermarked stock the adoption narrative describes. The Bank's history separately states that partly printed notes probably did not come into regular use until "some time between 1696 and 1699," and that its oldest surviving partly printed note payable to bearer is dated 19 December 1699 — two years after the handwritten specimen.

So the Fabriano case narrows a date range using surviving evidence; the Bank of England case shows a matching date does not guarantee a matching object. Both are instances of the general risk Stevenson's "runs and remnants" rule names — a single surviving specimen carries less evidentiary weight than a consistently marked run — but they fail along different axes of that risk: one about when, the other about what kind.