The account's most extreme figures — 50,000, 10,000, and 8,000 warriors, and villages four or five leagues long — appear in the Medina/Muñoz manuscript branch but not in Oviedo's version, which preserves the populous-settlement picture and one figure of about 5,000 warriors.

AMZ-E005 states the branch difference verbatim from Denevan's collation; no new material bears on it.

Direct comparison of the manuscript branches confirms that the extreme population figures only exist in the Medina/Muñoz lineage, while the Oviedo version retains the qualitative description without the vast numbers. This effectively uncouples the chronicle's qualitative geography from its most heavily criticized mathematics. The strongest opposing consideration is that Oviedo might have simply redacted numbers he personally found unbelievable, meaning their absence doesn't conclusively prove they were added by later copyists rather than present in the lost original.

Denevan’s manuscript comparison is explicit and verbatim: the Medina/Muñoz branch includes the extreme warrior counts and village lengths, while Oviedo’s does not, though it retains one figure of about 5,000 warriors and the general picture of crowding. This directly supports the claim that the famous largest numbers are textually unstable whereas the qualitative description is shared. The main objection is that Oviedo might have edited out numbers he considered implausible, so absence in his version does not prove absence in the lost autograph; nonetheless, the stated fact about divergent manuscript content is firmly established.

AMZ-E005 quotes Denevan verbatim: Oviedo has no 50,000, 10,000, or 8,000 warriors and no villages four or five leagues long, and one mention of 5,000 warriors. That is a concrete textual observation, not an interpretive dispute, and it is what licenses treating the qualitative geography and the extreme arithmetic as separable. Strongest opposing consideration: absence from Oviedo's redaction does not prove absence from Carvajal's lost original — Oviedo may have trimmed figures he disbelieved — so the instability cuts against relying on the numbers either way, which is already the claim's point.

Denevan's collation, verified verbatim, states that the 50,000/10,000/8,000-warrior figures and the four-to-five-league villages appear in the Medina/Muñoz branch and not in Oviedo's, which carries a single 5,000-warrior figure. That is a direct documentary observation about the witnesses, not an inference. The strongest opposing consideration is that Oviedo could have trimmed numbers he disbelieved, so absence from his redaction does not prove absence from Carvajal's lost original. That is true — but the claim asserts only cross-branch instability, and that is exactly what the collation shows, whichever direction the changes ran. This single fact reorganizes the whole Carvajal debate, licensing trust in the qualitative account while refusing the counts.

Denevan's verbatim statement (AMZ-E005) is a specific, checkable philological observation about which figures appear in which branch, and it is the kind of claim a specialist is unlikely to get wrong in a peer-reviewed venue. Confidence is moderate rather than high only because a single collation by one scholar carries the whole weight — no synoptic witness table exists (AMZ-R002 is proposed precisely for this). The strongest opposing consideration is Denevan's own implied caveat: Oviedo's silence does not prove the figures were absent from the lost original, since a redactor who disbelieved them could have cut them. Either way the figures are unstable, which is what the claim is used for.

Denevan's collation states the branch difference verbatim; verified against the text.

Denevan's collation (AMZ-E005) states the branch difference verbatim; no new material affects it.

Denevan's manuscript comparison states this directly, verified verbatim: "There is no mention of 50,000 or 10,000 and 8,000 warriors as there is in the Medina (and Muñoz) versions… nor of villages 4 or 5 leagues in length. There is one mention of 5,000 warriors in the Oviedo version."

Textual collation by William Denevan demonstrates that the most extreme figures—such as 50,000, 10,000, and 8,000 warriors, and villages four to five leagues in length—are found strictly in the Medina/Muñoz manuscript tradition and are omitted from Oviedo's version, which retains only a single warrior count of about 5,000 alongside the qualitative description of dense settlement. This discrepancy establishes that the headline demographic numbers are textually unstable and cannot be cited uncritically as Carvajal's unified testimony. The strongest opposing consideration is the possibility that Oviedo selectively abridged or censored numerical details present in an earlier common exemplar.

The verbatim Denevan passage establishes that the 50,000/10,000/8,000 figures and villages-of-4-or-5-leagues appear only in the Medina/Muñoz branch while Oviedo preserves one ~5,000 figure and the qualitative picture. Strongest opposing consideration: verification rests on a Wayback/OCR copy of Denevan rather than the manuscripts themselves, and absence from Oviedo does not by itself distinguish later embellishment from Oviedo's abridgement — the direction of the textual change remains an inference.

AMZ-E005 directly states that figures of 50,000, 10,000, and 8,000 warriors and villages four or five leagues long occur in the Medina/Muñoz versions but not Oviedo, while Oviedo includes one figure of 5,000. The strongest opposing consideration is reliance on a modern collation verified through OCR rather than a synoptic inspection of the manuscripts, but the record exactly supports the claim as stated.

Denevan’s side-by-side manuscript comparison (AMZ-E005) states verbatim that the 50,000/10,000/8,000 warrior figures and four-to-five-league villages appear in Medina/Muñoz but not in Oviedo, which retains one ~5,000-warrior figure and the populous-settlement picture. No contrary manuscript evidence appears in the ledger; the only residual uncertainty is OCR/pagination of the verified copy.

AMZ-E005 reproduces Denevan's statement verbatim: the 50,000, 10,000 and 8,000 warrior figures and the four-to-five-league villages are absent from Oviedo, which retains the populous picture and one 5,000-warrior mention. This is a bibliographic fact about extant witnesses, checkable and checked. The opposing consideration is provenance quality — verification came via a Wayback abstract plus OCR full text, so page locators are OCR pagination — but the substantive collation is unambiguous and AMZ-R002 would only sharpen it.

AMZ-E005 quotes Denevan's branch comparison verbatim: the 50,000, 10,000 and 8,000 warrior figures and the multi-league villages are in the Medina and Muñoz versions and absent from Oviedo, which carries one figure of about 5,000. Verification was via OCR full text. Which figure sits in which specific witness within the Medina/Muñoz family has not been re-read at page level; that is AMZ-R002's job and does not alter the branch-level finding.

E005 (strong) quotes Denevan's branch comparison verbatim: the 50,000/10,000/8,000 figures and the four-to-five-league villages are absent from Oviedo, which keeps one figure of about 5,000. Verified from an OCR copy; no page-level collation yet.

Denevan states the branch difference in the words the claim uses (E005), verified from OCR full text; no admitted record contradicts it.

AMZ-E005 (strong, supports) quotes Denevan's branch comparison verbatim: the 50,000, 10,000 and 8,000 figures and the four- and five-league villages are in the Medina/Muñoz versions and absent from Oviedo's, which has one figure of about 5,000. The limitation is that the comparison was verified from an OCR copy and is a geographer's, not a philologist's collation; a blocked attempt to place each figure by witness at page level remains open.

Direct textual comparison between the surviving manuscript branches confirms that the extreme figures—50,000, 10,000, and 8,000 warriors, and multi-league settlements—are confined to the Medina/Muñoz tradition and absent from Oviedo's text, which contains a single figure of 5,000 warriors. Denevan's (1996) collation establishes this textual divergence. Strongest opposing consideration: Oviedo admitted to incorporating supplementary oral testimonies from Orellana and his crew in Santo Domingo, which leaves the exact redactional genealogy of the differing numbers open to competing interpretations.

Denevan 1996 states it verbatim (AMZ-E005), checked against a full-text OCR copy. The strongest opposing consideration: the underlying manuscripts were not collated by this ledger, and no synoptic witness table exists (AMZ-R002 is pending), so the finding is one respected scholar's comparison rather than an independent collation.

AMZ-E005 directly states that the 50,000, 10,000, and 8,000 warrior figures and villages four or five leagues long occur in the Medina/Muñoz versions but not Oviedo, which retains one figure of 5,000. This matches the claim's manuscript distinction exactly. The strongest opposing consideration is that the comparison was verified through an OCR copy rather than a new synoptic collation of the manuscripts.

Denevan 1996 states verbatim that the 50,000 / 10,000 / 8,000 warrior figures and 4–5 league villages appear in Medina/Muñoz but not Oviedo, which retains one figure of about 5,000 warriors (AMZ-E005). The opposing consideration is reliance on a single modern collator’s page comparison rather than a published full synoptic table (AMZ-R002), which does not disturb the stated branch difference.

AMZ-E005 quotes Denevan verbatim: the 50,000, 10,000 and 8,000 warrior figures and the four-or-five-league villages are absent from Oviedo, which carries one mention of 5,000. This is a direct textual comparison by a specialist, checked against an OCR full text, and it is the kind of claim a single competent collation can settle. The opposing consideration is that the whole finding rests on one scholar's reading, verified from an OCR copy rather than the manuscripts, and the synoptic collation that would make it unimpeachable (AMZ-R002) has not been done.

Denevan's (1996) detailed manuscript collation directly verifies that the extreme numbers (50,000, 10,000, and 8,000 warriors, and villages 4–5 leagues long) appear in the Medina and Muñoz manuscript traditions but are absent from Oviedo's text, which reports at most 5,000 warriors. The strongest opposing consideration is that both traditions still convey an overarching picture of substantial riverine populations, meaning the difference between branches is quantitative rather than qualitative.

Denevan's branch comparison is quoted verbatim and verified; the asymmetry it documents is the strongest single fact in the Carvajal dossier. The opposing consideration is that this is one modern scholar's collation, and the ledger's own AMZ-R002 synoptic witness table — which would make the comparison independent — has not been performed.

AMZ-E005 directly states that the 50,000, 10,000, and 8,000 warrior figures and four- or five-league villages occur in the Medina/Muñoz versions but not in Oviedo, which retains a 5,000-warrior figure. The strongest opposing consideration is that the ledger relies on Denevan's comparison rather than presenting a complete synoptic collation of the manuscript witnesses. That limitation does not materially undermine the specific textual contrast reported.

Denevan’s direct manuscript comparison (E005) states the extreme figures and league-long villages are absent from Oviedo while present in Medina/Muñoz, with only one ~5,000-warrior mention retained. No contrary collation appears in the ledger.

E005 quotes Denevan's collation verbatim. The Oviedo version has no 50,000, 10,000, or 8,000 warriors, no villages four or five leagues long, and one mention of 5,000 warriors; the Medina (and Muñoz) versions carry the larger figures. The clause that Oviedo 'preserves the populous-settlement picture' rests on the claim's anchor rather than the quoted sentence, though it is implicit in the comparison. The strongest opposing consideration is transmissional: Oviedo rewrote the friar's account with Orellana's oral particulars (E089), so his text is a redaction, not a clean second manuscript. The divergence is real, but it cannot show in which direction the figures moved, and no synoptic table exists (AMZ-R002).

AMZ-E005 quotes Denevan's comparison verbatim: the 50,000, 10,000 and 8,000 warrior figures and the multi-league villages are absent from Oviedo, which has one mention of 5,000. It is one specialist's comparison, checked from an OCR copy. A page-level attempt to place each figure by witness did not complete, and no synoptic collation exists.