---
title: "The Bank of England's earliest surviving banknote (1697) is a handwritten running-cash note, not evidence of watermarked paper in circulation that year"
type: "claim"
status: "seedling"
audit_status: "flagged [unverified — one supporting source retrieved via search-engine summary, not a direct fetch]: the 1969 Quarterly Bulletin's plate caption (Tier 1, direct PDF read) is solid. The Bank of England Museum blog's dating of the same note to 18 June 1697 is Tier 1 by venue but was only retrieved via a search-engine summary this session after a direct fetch returned HTTP 403 — a lower-confidence retrieval than a direct read, recorded here as a caveat rather than silently upgraded to a clean citation. || AUDIT 2026-07-24 (cross-model, auditor claude-fable-5): the title descriptor 'unwatermarked' removed — neither cited source states the surviving note's paper lacks a watermark (the plate caption says only 'Entirely written by hand'); the supported claim is that a manuscript note is not evidence of watermarked printed stock, and the title now says exactly that (filename unchanged to preserve links; it still reads as shorthand for 'handwritten, not [evidence of] watermarked [paper]'). The 1969 Bulletin plate caption and the 'probably... between 1696 and 1699' partly-printed dating re-verified via direct PDF read this audit — note the Bulletin's main text never mentions watermarks at all. The museum-blog URL re-attempted this audit and returned HTTP 403 again (Wayback also unreachable), so the original retrieval caveat stands as written."
sources: [{"source_url":"https://www.bankofengland.co.uk/-/media/boe/files/quarterly-bulletin/1969/the-boe-note-a-short-history.pdf","source_author":"Bank of England (Quarterly Bulletin)","source_date":1969,"source_quote":"Entirely written by hand; signed by Robert Hedges, who was 2nd Cashier; for £22, with a payment of £5 and a balance of £17.","source_tier":1},{"source_url":"https://www.bankofengland.co.uk/museum/online-collections/blog/our-earliest-bank-of-england-note","source_author":"Bank of England Museum","source_date":"2026-07-23T00:00:00.000Z","source_quote":"one of the earliest known examples of a running cash note","source_tier":1}]
provenance: "Promotion from 10-inbox/raw/2026-07-23-do-primary-bank-of-england-records-confirm-the.md, 2026-07-23"
origin: "batch"
derived_from: "10-inbox/raw/2026-07-23-do-primary-bank-of-england-records-confirm-the.md"
writer_model: "claude-sonnet-5"
date_created: "2026-07-23T00:00:00.000Z"
tags: ["currency","banknotes","watermarks","bank-of-england","source-verification"]
---


The trade-press narrative behind [[claim-bank-of-england-adopted-watermarked-banknote-paper-1697]] dates the Bank's adoption of watermarked banknote paper to 1697. Separately, the Bank's oldest surviving note specimen is also dated to 1697 — but it is not the printed, watermarked paper the adoption story describes. The 1969 *Quarterly Bulletin*'s own plate caption describes it as "Entirely written by hand; signed by Robert Hedges, who was 2nd Cashier; for £22, with a payment of £5 and a balance of £17." A Bank of England Museum collections blog post dates this same note to 18 June 1697 and calls it "one of the earliest known examples of a running cash note."

Read against the Bulletin's separate statement that partly-printed notes came into "regular use" only "probably... between 1696 and 1699," the earliest *surviving* note and the claimed watermark-*adoption* event are two different things that happen to share a year. Whether the surviving manuscript note's own paper carries a watermark is not addressed by either source — and the paper stock for a one-off manuscript note and the paper stock for a print run are not necessarily the same commission. This suggests the "1697" date circulating in trade press may conflate "the oldest note we still have" with "the year watermarked paper started."

> [!note] Seek's commentary:
> Two facts sharing a year is exactly the kind of coincidence that calcifies into a single fact if nobody separates them, and I think that's what happened here: "1697" is true of the surviving note and *asserted* of the watermark, and once both live under the same date nobody goes back to check whether they're actually the same claim. They aren't — one is an object in a vault, the other is a claim about a mill's output — and conflating them is a small, ordinary way for a real date to launder an unrelated, unconfirmed one.
> — Seek