Multilayer Cipher Architecture in Fontana's Secretum and Bellicorum: A Twenty-Lens Template Return with Negative-Control Differential

Giovanni Fontana (Venice, c.1395 - c.1455; Padua medical doctorate 1421) produced two surviving cipher-bearing manuscripts: the Bellicorum instrumentorum liber (Munich, Bayerische Staatsbibliothek, Cod.icon. 242) and the Secretum de thesauro experimentorum imaginationis hominum (Paris, Bibliotheque nationale de France, NAL 635). A third Fontana witness, the autograph-candidate De trigono balistario in Bodleian Canon Misc 47, is plain Latin with embedded astronomical and trigonometric apparatus. Battisti and Battisti (1984) decoded the substitution layer of the cipher works (single-character substitution with a stable key inventory). The present paper reports a 2026-05-11 template-return outcome applying a twenty-lens analytical instrument across all three Fontana witnesses, and the 2026-06-11 negative-control run that applies the same template, under the same per-lens rubric (Appendix A, pre-registered relative to the 2026-06-11 negative-control run and applied retrospectively to the Fontana 2026-05-11 audit notes; see Section 2.1 temporal-order note), to three contemporaneous Padua-corridor non-Fontana witnesses: BSB Cgm 7958 (Hartlieb, Bavaria 1456), the Hartlieb Kraeuterbuch nine-witness corpus, and the Cermisone Consilia (1441 print closure). The Fontana 3-witness result is seven SIGNAL lenses for NAL 635, seven SIGNAL lenses for the Bellicorum, and eight SIGNAL lenses for the Bodleian witness against an internal return threshold of five SIGNAL lenses across two or more works; the Fontana counts are reported as Candidate-tier rescores pending blinded inter-rater re-scoring. The negative-control 3-witness result is three SIGNAL lenses for Cgm 7958, two SIGNAL lenses for the Hartlieb Kraeuterbuch corpus, and two SIGNAL lenses for the Cermisone Consilia. The differential, Fontana mean 7.3 SIGNAL lenses against negative-control mean 2.3 SIGNAL lenses, ratio 3.17x, is a tentative discriminating signal pending category-matched controls (cipher-bearing and engineering-codex Quattrocento witnesses); on the present three controls the template is not behaving as a generic Quattrocento-codex detector. The single-substitution decode of Battisti 1984 corresponds to lens five (cipher engineering); on the per-lens bibliographic-coverage audit (Appendix B), the surveyed Fontana literature offers no engagement on lenses one (iconographic), three (user instruction), six (layout geometry), seven (cross-folio motif mapping), eleven (balneological flow), twelve (paleographic hands), thirteen (scribe workflow), seventeen (recipe indexing), and nineteen (myth-to-math) as a structural set. A structural-parallel-development line (fourteenth-century Italian computus through the Padua-orbit medical-and-instrument cohort c.1410-1470 to Fontana's c.1430s cipher specialisation, structurally parallel to Trithemius 1518 and Vigenere 1586) is described in Section 3 as a contemporaneous-network reading anchored on Padua institutional density, not as a sequential lineage with documented hand-to-hand transmission. The pre-Tepenecz frontier of Voynich Manuscript provenance is the documented March 1599 Augsburg Hofkammer transaction; the competing pre-Rudolfine Hartlieb-Anna-Nannhofen-Elsenheimer chain attributed in a 2024 paper is substantially undermined on documented archival grounds. Comparator discipline (Section 5): the Voynich Manuscript is an independent point of investigation that sits inside the same chronological-and-architectural envelope as Fontana, not a numbered stage in the parallel sequence.