Rejecting a quotation from publication and erasing it from the record are two different decisions. This page is the record.
What this page is
Nothing on this page is in the article. Each entry is a quotation or a number that was considered, could not be confirmed against the source it names by two routes, and was set aside. The page is written for the next researcher, so it carries routes.
Three status words are used.
- REFUTED — the attribution is demonstrably wrong. The named source was opened and the text is not in it.
- UNVERIFIABLE — no route to the named source has been found yet. That is a statement about this project's reach, not about the quotation.
- PARTIAL — the source was reached and most of the wording confirmed, but a load-bearing number or word could not be read.
Entries
1. Marks on Britten, the paperback preface
Wording as it circulates: "The biochemist Roy Britten argues that the genetic similarity between humans and chimpanzees is not over 98%, but closer to 95%."
Claimed source: Jonathan Marks, What It Means to Be 98% Chimpanzee, paperback edition, preface.
Status: PARTIAL.
What was tried: archive.org item whatitmeanstobe90000mark_f0z8 through bin/iapdsearch.py, which returns whole OCR paragraphs. The sentence was found on the unnumbered preface leaf 17 up to "but closer to", and the closing figure runs onto the next line, which the endpoint did not return.
What was found: the sentence exists and the wording up to the number is confirmed. The number is not.
Verdict: do not quote the figure from Marks. The article cites Britten 2002 directly for the 95 %, which is what Marks is reporting.
What would change it: a borrowable copy or a second OCR of the preface leaf.
2. Carter's "3.6 % between one human's own autosomes"
Wording as it circulates: Robert Carter, in a Facebook group post on the 2025 telomere-to-telomere data, observed that the same gap-plus-SNV method registers about 3.6 % difference between the two autosomal chromosome sets of one human.
Claimed source: a post in a Facebook group, linked in Eric's notes.
Status: UNVERIFIABLE as a quotation.
What was tried: Facebook group posts are not fetchable without a login and are not durable. Not fetched.
What was found: the number itself is reproducible from Yoo 2025 Supplementary Figures III.11 and III.12: 3.48 % gap divergence plus 0.16 % SNV divergence within hg002 gives 3.64 %.
Verdict: the article attributes the observation to Carter in prose, without quotation marks, and shows the arithmetic from the primary. It does not quote him.
What would change it: a durable copy of the post, or the same point in one of Carter's articles.
3. Marques-Bonet on "the true overall difference"
Wording as it circulates: "while earlier studies suggested a 98% to 99% similarity, comparisons that include harder-to-align regions push that difference closer to 5% to 10%, Marques-Bonet said. 'And if we account for the regions still too complex to align properly with current technology, the true overall difference is likely …'"
Claimed source: a press interview with Tomàs Marques-Bonet about the 2025 ape genome paper, recorded in Eric's notes without a URL.
Status: UNVERIFIABLE.
What was tried: nothing beyond the notes, since the article takes quotations only from the dossier and the dossier did not work this up.
What was found: the notes file carries the wording but no locator.
Verdict: not used. The 5–10 % range it describes is consistent with the control-corrected figures in the article, but a press paraphrase is not a source.
What would change it: the interview URL and the bytes.
4. Tomkins 2011, "86–89%"
Wording as it circulates: "Genome-wide DNA alignment similarity (identity) for 40,000 chimpanzee DNA sequences queried against the human genome is 86–89%."
Claimed source: Tomkins, Answers Research Journal 4:233–241, 2011.
Status: UNVERIFIABLE as a primary; the title and range are confirmed through the 2012 Tomkins and Bergman review, which is on disk.
What was tried: the ARJ URL patterns that resolve the 2013, 2015 and 2018 papers all return 404 for this one.
What was found: the 2012 review quotes the range. The paper predates the 2015 BLASTN bug disclosure, so its status is unclear regardless.
Verdict: not cited in the article. The 2013 and 2015 papers carry the same argument with the bug history attached.
What would change it: the ARJ volume-4 index or the assets.answersresearchjournal.org/doc/v4/ asset path.