Ercole Rondinelli to Duke of Nevers, 11 November 1590

01 The letter

02 The key

32 symbols: 23 letter/vowel forms, 5 in nomenclature, 2 nulls, 2 controls

Key elsewhere

Original key in a different volume: Français 3995, f89r.

03 The decipherment

f62r, full cipher lines 15–17

… dalla necessita io non posso credere che l signor duca di Mena sia bastante per difenderla, uedendosi che egli si consiglia et si gouerna per l'auiso di quelli che desiderano la sua ruina.

Translation. …because of the necessity, I cannot believe that the Duke of Mena is capable of defending it, seeing that he takes counsel and governs himself according to the advice of those who desire his ruin.

04 Our check

This result is George Lasry’s. This project did not re-solve the cipher: it checked one sample against the Gallica image and recorded what it found.

• Batch A. Image checked: https://gallica.bnf.fr/ark:/12148/btv1b90582263/f69.item
• Folio matches the report: yes; cipher present: yes
• What we checked: f62r "Ser.mo Principe Sig.re et Pron mio Col.mo", full cipher in letter-substitution with superscript vowels; his key image is the original "Cifra col S. Cav. Giac. Gratij" (Fr. 3995 f89: a=8, b=7, c=6, d=5, f=m, n=h, r=y, s=theta, t=g...), alphabet consistent with the cipher letters; superscript (babuino) syllables made a quick run-decode inconclusive
• Verdict: partly confirmed (method read after matching with key from external sources, extent complete). The check is a sample, not a re-solution.
• Notes: Key labelled for Cav. Giacomo Gratij(?), so the matching-key attribution to Rondinelli rests on matching, not on the key title. Clear Italian postscript at foot of f62r.

05 Sources

• George Lasry, with ChatGPT Astra 6 and Anthropic Opus 5.5, BnF cipher research, 10 October 2026 (sent to Daniel Bourdeau on 13 October 2026 for publication here), p. 41; his other items are listed on his contributor card.
• The working folder: targets/rondinelli1590/ (his entry as extracted from the report, our spot-check, profile.json).