6,412 Letters Asked ‘Did My Book Once Say Something Else?’ She Answered Every One.
Rosalind Quayle-Abendroth ran a deposit library's variant-editions desk for 41 years and died in April at 88. Her logbooks confirm a genuine printing variant in 903 cases, find none in 4,988, and leave 521 open.
ABERYSTWYTH, Wales — Rosalind Quayle-Abendroth ran the variant-editions enquiry desk at the deposit library in Aberystwyth from March 1983 until June 2024. She died on 9 April at her home in Borth, aged 88. The cause, her executor says and the certificate records, was pneumonia.
Her executor has given this newspaper her logbooks. Forty-one volumes, one to a year, recording every enquiry the desk took.
Six thousand four hundred and twelve letters, all asking one question: did my book once say something else?
What people wrote in about
A line remembered differently. A character who dies in the reader's memory and does not die on the page. A sentence somebody could quote from the age of nine and cannot now find in the book.
There is a real thing underneath the question. Books are reset between printings. Compositors correct errors and introduce them. Publishers cut for school editions without saying so on the title page. A deposit library holds the successive settings in order of receipt, which is much of what a deposit library is for.
Her desk existed to go and look.
The three columns
She confirmed a genuine setting or printing variant in 903 cases.
In 4,988 she found none. The book had always said what it says now, and the reader's memory had done the rest.
And 521 she left open. She did not close them, and instructed her deputy — after 1996, her successor — not to close them either. The last column for these reads o/s, outstanding, in some cases for thirty years.
The rule about the shelf
Every reply gave a shelfmark.
The rule is written inside the cover of the 1983 volume in her hand and repeated in the joining note she gave to two successors.
The answer was never the service. The service was that somebody told no could take a train, present a reader's ticket, ask for the book at that number and look at the page.
Eighty-five per cent, not eighty-six
Of the 5,891 enquiries she resolved, 4,988 were told that no variant existed. That is eighty-five per cent.
The figure read out at her memorial was eighty-six. It comes from a summary sheet she typed in 2019 — 4,654 of 5,412 — and it stops five years short. We print the discrepancy because she would have.
Her training note for the 1996 deputy says the answer is usually no, and that the desk is not there to produce yeses. It is there so that a question stops being a rumour and becomes a shelfmark.
Three boxes in a barn
The card index behind the logbooks, roughly 41,000 sheets, is recorded as destroyed in the 2014 move. The inventory lists eleven boxes, pulped.
Three were found in August 2025 in a barn near Kington, in Herefordshire, by Bronwen Astley-Nwosu, a retired stack supervisor who had put them in the back of her car rather than watch them go, and had told nobody for eleven years.
We counted a tray. About 11,000 cards. The library's record still says pulped, and the other eight boxes are accounted for by nobody, including two people who signed the inventory.
Enquiry 214, in another hand
She retired in June 2024. In November she wrote to her old desk, on the public form, as an ordinary member of the public, and did not mention that she had built it.
She asked about a line in a novel she had read as a girl and had catalogued herself in 1974. She quoted the line, said she had checked the three settings she could reach, and could not resolve it.
Her letter is not among the 6,412. It is in the volume after hers, in another hand, as enquiry 214. The reply is dated 6 February 2025.
Dear Mrs Quayle-Abendroth,
Thank you for your enquiry of 18 November. The library holds four settings of this title: the first, the reset second of the same year, the school edition of 1958 and the 1971 reissue. I have collated the relevant leaf in all four.
The line you quote appears in none of them, and no variant is noted in the publisher's file deposited with us in 1983. I am recording the enquiry as resolved, no variant found.
The four settings are at bay 41, shelf D. The reading room is open Tuesday to Saturday and you would be very welcome.
N. Ilminster-Vaughan, enquiries assistant, variant editions
What this desk records
She was eighty-seven, she had catalogued the book herself, and she was wrong about it. The answer came back in the form she had spent forty-one years insisting on: a collation, a finding, a shelf number and an invitation.
Her successor's reply to this newspaper, printed below, takes our ending away from us. She did not go to bay 41. She was never going to. The rule was never that people go; it is that a desk which cannot name a shelf is not answering an enquiry, only reassuring somebody.
She was not a librarian, and that letter is correcting a draft of this obituary in which we had called her one. She was an enquiries assistant, grade 4, for thirty-one of her forty-one years. She applied twice for the senior post and was appointed neither time, and the second time she trained the man who got it. The title is on the last line of all 6,412 replies.
Of the 521 open cases, eleven correspondents are alive. Two answered the telephone. Both asked whether the desk had ever got back to them. It had not. It is still open, which is the answer she would have given.
The six o'clock edition
Three stories a morning, in your inbox before the coffee is made.
Sources & Method
We asked her executor for the logbooks and he gave them to us. Two members of this newspaper's staff counted every entry separately, without comparing totals, and reconciled a difference of nine line by line. We then paid a bibliographer to re-collate a random sample of her confirmations without telling him whose rulings they were, and drove to Herefordshire to see the three boxes and count a tray of the cards ourselves.
Who we spoke to
- The enquiry logbooks, 1983–2024, Forty-one volumes, one to a year, kept by Rosalind Quayle-Abendroth and by her deputy after 1996. Given to this newspaper by her executor under a deed permitting quotation but forbidding publication of correspondents' names and addresses May 2026 Counted independently by two members of staff who did not compare totals until both had finished.
- Peredur Quayle-Abendroth, Her nephew and executor. Interviewed at her house in Borth in May, and again by telephone in July, and sent the full draft May–July 2026 Supplied the death certificate and the 2019 summary sheet. Asked us to print that she had applied twice for the senior post.
- Nerys Ilminster-Vaughan, Enquiries assistant, variant editions, at the deposit library; her successor since 1996 as deputy and since 2024 in the post. Interviewed twice at the library and sent the complete draft with three weeks to answer June–August 2026 Her reply is printed in full at the foot of this page and is the most serious objection to this obituary we received.
- Bronwen Astley-Nwosu, Retired stack supervisor. Interviewed at her farm near Kington, Herefordshire; showed us three boxes and let us photograph and count a sample of the cards July 2026 Removed the boxes during the 2014 move without authority and told nobody for eleven years.
- Dr. Aurelio Standish-Mbeki, Historical bibliographer. Commissioned to re-collate a random sample of 60 of the 903 confirmations without being told whose work they were June 2026 Fifty-five stood. Three he could not test. Two he thinks are wrong, and his working is published with this story.
Documents
- PX-1968 — Nine hundred drawings from memory, with the pre-registration and analysis plan accepted
What we could not confirm
- Whether the logbooks are complete. Four months of 1991 are missing from volume nine and her executor does not know why; a run of entries during the 1994 industrial dispute is in a hand nobody has identified. Our two counts of the surviving entries differed by nine before reconciliation, which is a fair measure of how tidy this record is not.
- The 903. A blind re-collation of 60 of them found 55 sound, 3 untestable because no British library holds the second setting, and 2 that the bibliographer believes she got wrong. Sixty is too small a sample to scale to 903, and we decline to scale it. What we can say is that two of sixty went against her.
- What became of the other eight boxes of the card index. The 2014 move inventory records eleven boxes pulped. Three were in a barn. Nobody we asked at the library, including two people who signed the inventory, can say whether the remaining eight were destroyed or simply written down as destroyed.
How Others Covered This
The same events, as reported elsewhere on the same day. We list what each outlet had that we did not, as well as what we had that they did not — including where we come off worse. Why we print this.
-
The Meridian TelegraphTHE LIBRARIAN WHO PROVED YOUR MEMORY IS BEING EDITED — 903 TIMES
Took the 903 confirmed variants as a running count of texts altered after publication, and treated her refusal to close 521 cases as a career of suspicion.
Had that we did not
The 903, counted correctly, from the same logbooks we were given.
Left out
The denominator. The word 'compositor' does not appear in the piece, and neither does 4,988.
-
ApertureThe Woman Who Said No
An essay on doubt, provincial institutions and the dignity of the unhelpful answer. Better written than this obituary and did not open the logbooks.
Had that we did not
The rule about the shelfmark, and what it is for, put more clearly than we have managed here.
Left out
That the 521 open cases have names and dates against them, that eleven of those correspondents are alive, and that two of them answered the telephone.
-
They Buriedthis newspaper6,412 Letters Asked ‘Did My Book Once Say Something Else?’ She Answered Every One.
Counted the logbooks twice by two people working apart, re-tested a sample of her confirmations blind, and drove to Herefordshire to see the boxes.
Had that we did not
All three totals, the discrepancy with her own memorial figure, the two sample confirmations that went against her, and her successor's reply entire.
Left out
We end on the letter in which she was wrong, because it is the better ending. She is not here to say whether she wanted her last enquiry to be the one a newspaper printed, and we asked nobody who could have guessed. — V. Ashcombe-Doyle, standards editor
How was this story?
We publish the result, whatever it is. Reader verdicts appear on the front page and in our newsroom metrics.
Readers' Letters 0
Printed at once under the name you give and read by the desk afterwards; anything unfit is removed, with a note saying so, and nothing else is ever deleted — only corrected. Letters that changed something in the story carry a mark saying so, and there are 5 of those across the archive.