Commentarius in Epistolam ad Philippenses

Merged 2026-08-01 from cruces-0000.md (Fable anchor), cruces-0001.md, cruces-0006.md and cruces-0010.md — four files, three independent Opus agents plus the anchor. Source files follow verbatim; this header is the authority where they disagree. See also OPEN-QUESTIONS.md in this directory, which holds what the merge could NOT settle.

WORK-WIDE CONVENTIONS (authoritative)

Vocabulary: no conflicts. The three ranges cover disjoint stretches of the epistle, so their fixed vocabularies do not collide; all are carried forward whole. Consolidated:

• Anchor (structure, all ranges): catena sigla reproduced verbatim in place · «short lemmata» keep Migne's guillemets, block lemmata unguilleted · ΚΕΦΑΛ. Nʹ + argument → "CHAPTER N." + argument sentence · Ἢ opening an alternative construal = "Or:" · τουτέστι/ἤγουν = "that is" · ἀντὶ τοῦ = "stands for" · ἔνι = "it is possible" · question marks restored where Calfa drops the plate's ; · cross-column bleed is not Greek text.
• 0001–0005 (Phil 1–2:15): ἁρπαγμός/ἁρπάζω/ἁρπαγή share the root "seizure/seize" (AV's "robbery" would break Oecumenius's own etymological gloss) · πεποιθώς/θαῤῥῶν/τεθαῤῥηκώς kept apart (Allatius flattens all three) · ὑπόστασις = hypostasis, against Allatius's persona · μορφή = "form" (form of God / form of a servant).
• 0006–0009 (Phil 2:16–3:16): καύχ- on one root "boast" (AV "rejoice"/ RSV "glory" would break the gloss's argument) · περιτομή/κατατομή = circumcision/concision (the gloss IS the etymology; the English pair must share "-cision", as Migne's own circumcisio/concisio does) · κέρδος/ζημία/ ζημιόομαι/σκύβαλα = gain/loss/suffer loss/refuse · σπένδομαι/σπονδή/θυσία/ λειτουργία/λειτουργός = poured out as a libation/libation/sacrifice/ ministration/minister · διώκω/καταλαμβάνω/βραβεῖον/σκοπός = pursue/lay hold/ prize/mark (AV's "follow after / apprehend / press toward" refused: the runner image needs the same two words each time) · ἰσόψυχος = "like-souled" · εὔφημος/δύσφημος = "of good/ill omen" · συμμορφούμενος = "conformed" — and note this shares the form- root with the peer's μορφή = "form", so Oecumenius's wordplay survives; no reconciliation needed.
• 0010–0013 (Phil 3:17–4:23): ἰσχύω/ἰσχύς/ἐνδυναμόω = be strong for/ strength/empower (so Phil 4:13 and its gloss share a root, against AV's "I can do all things") · ἐπιεικές = "forbearance" · σεμνά = "seemly" · ὑπερέχω/ὑπερβαίνω = "surpass" · πολίτευμα/πολιτεύομαι = commonwealth/live as citizens · μεμύημαι = "I have been initiated".

Durable, applies to all ranges: bare em-dashes inside a scholion are alternative-construal separators, not sigla (scan-verified; the Latin prints the same dashes). The Calfa bleed class includes the volume running foot (PATROL. GR. CXVIII), not only the Latin running title.

⚑ THE ONE REAL CONFLICT — restoration marking. RULED PROVISIONALLY, one edit PENDING.

The three ranges handled plate-recoverable source loss three different ways, each defensible under the current rulebook:

They are not actually incompatible — they answer different questions. Unified rule, proposed:

1. Words lost mid-sentence, plate certain → restore silently, log. This is ordinary Calfa damage (attribution ladder row 1).
2. A reader-visible structural unit lost — heading, speaker turn, subscription → restore WITH [ed:]. Its absence would otherwise be invisible as an absence, which is the whole point of Pattern 13.
3. Text lost and not recoverable (plate absent or illegible) → [ed:] marking the hole; supply nothing.

Under this rule every range was right except one item: the work's final subscription (Ἐγράφη ἀπὸ Ῥώμης δι᾽ Ἐπαφροδίτου, restored silently in 0013) is reader-visible structure and should carry [ed:].

✅ RULED AND APPLIED (Wilson, 2026-08-01). Adopted as translation-style.md Pattern 13a, corpus-wide, PL included. The subscription in 0013.md now carries its [ed:]; verify clean after the edit. Nicetas restored its five ΛΟΓΟΣ headings as [ed:] under clause 2 and would be unaffected either way.

⚑ ADDED 2026-08-01 by the blind polarity reads — two apparatus gaps

A. 1281 (chunk 0004) — an unlogged Migne-side negation site. NO English change.

Migne prints, in the refutation of Marcion:

Εἰ γὰρ ἦν ἐν μορφῇ Θεοῦ, πῶς οὐ λέγεις ὅτι ἀπὸ Μαρίας ἤρξατο;

As printed, the refutation demands the very position it is refuting ("how do you not say that he began from Mary?"). The expected sense wants the οὐ gone, or λέγεις to be λέγει. Our English carries it literally — correct under Pattern 7, and no edit is proposed.

But it could not be adjudicated, and that is the point of logging it. Col 1281 sits inside the 1279–1281a scan gap: PG 118's own leaf for the Christ-hymn block-lemma page is absent from patrologiaecurs14migngoog (confirmed by the reader with a whole-volume search for Σαμοσατ-, Σαβελλ-, Μαρίας, Marcion — zero hits anywhere; leaf 654 = cols 1277/1278, leaf 655 = 1281/1282). So Migne-vs-Calfa attribution here is undecidable from our files, and the standing PG rule forbids attributing it to the plate. Recorded as a live negation site with no verdict.

B. 1272 — a possible Migne ( sic ) at Phil 1:20. NEEDS A PAGE IMAGE.

The scan's OCR of leaf 650 reads …μεγαλυνθήσεται Χριστὸς ἐν τῷ στόματί μου sic: διὰ ζωῆς, εἴτε διὰ θανάτου — a Roman-type sic standing exactly where the first εἴτε belongs, while the second εἴτε OCR'd normally. Calfa has εἴτε and no sic.

If the plate prints ( sic ) there, Migne flagged στόματι himself — and under known-false-positive #5 his mark is reproduced verbatim and we do not stack a note of ours on top; the [var:] on this word would need re-examining. Cannot be settled from OCR. Flagged, not asserted — it needs the leaf image, and PG 118's PDF is not held locally (only the djvu.xml). See OPEN-QUESTIONS.md.

Marker totals after merge

9 [lat:] (most Greek-plate-verified; the strongest are 1284 ἑκὼν vs invitus and 1297 οὐκ ἔτι vs nondum, both polarity sites) · ~16 candidates logged unmarked with reasons, including every ἡμεῖς/ὑμεῖς itacism falling in a leaf gap · 3 [var:] · 5 [ed:] · 4 new sigla met, 2 scan-verified (— [ΒΑΣΙΛ.], — [ΟΙΚΟΥΜΕΝΙΟΥ]), 1 not verifiable ([ΤΟΥ ΑΥΤΟΥ], flagged for a plate check), 2 with letters lost and no form invented.

Corrections to the record made at merge

The anchor's cruces-0000.md had filed Μάρτυρ as an OCR correction to Μάρτυς. The plate prints Μάρτυρ twice (leaf 648, block + short lemma), so it is a scripture divergence against the received Phil 1:8 μάρτυς, not damage. Corrected in place in that file. No English changed.

Verified after merge

verify-english-pg.mjs oecumenius-philippians → 14 chunks, anchors + frontmatter + ratios + dedupe clean.

Work-wide conventions (anchor session, 2026-07-31 — merge into the work-wide section)

• Catena attribution sigla are sacred structure. The plate opens each attributed scholion with an em-dash + bracketed source name: — [ΦΩΤ.], — [ΧΡΥΣ.] (scan-OCR-verified at four sites: leaves ~647, ~648, ~651, ~653, ~654 of patrologiaecurs14migngoog). Reproduce VERBATIM, untranslated, in place — they are Migne's citations (the [n:] principle). They carry no colon, so they cannot collide with our [tag: …] marker namespace. Calfa MANGLES them (ἴΦΔΤ. for [ΦΩΤ.], -ΧΡΥΣ.] for — [ΧΡΥΣ.]): restore each from the plate form and log it. Expect more sigla ahead (the volume's catena cites more fathers than these two); verify each new one against the scan before fixing its form.
• Lemma consistency is the work's spine (7a″ in force). The scripture block-lemma and every «short lemma» quoted from it in the glosses must render identically where the Greek is identical — and both must follow MIGNE'S printed Greek, not the received text and not familiar English (worked cases below: single μᾶλλον at Phil 1:9; anarthrous θλίψεως in the 2 Cor 2:4 quote).
• Guillemets as printed: short gloss-lemmata keep «…»; the long block lemmata are unguilleted in our source and stay so.
• The Latin column quotes scripture from the VULGATE — that is convention, not divergence. Two instances in this chunk alone (Ex multa afflictione where the Greek prints Ἐκ γὰρ θλίψεως without πολλῆς; magis ac magis where the Greek prints a single μᾶλλον). Never fire a [lat:] on a scripture quotation the Latin has conformed to the Vulgate; log if notable. (§8 Q3 calibration datum, pairs with the Nicetas "plate's own apparatus" rule.)
• Mentioned-word refinement (extends the Nicetas rule): a discussed word stays in Greek script only where English cannot carry the point (ἐξ/διά); where the morphological contrast survives translation, translate it and keep lemma-consistency (κοινωνούς/συγκοινωνούς = partakers / fellow-partakers — as the plate's own Latin does: socios… una consortes).
• Greek question marks: Calfa often prints · or nothing where the plate's ; belongs; restore the question from sense + the Latin column (three sites this chunk), silently, logged as a class not per-site.
• Kephalaia: ΚΕΦΑΛ. Αʹ + argument renders "CHAPTER 1." + the argument sentence (the Latin column's own CAPUT I practice).

Latin-twin pass (pilot §6): result for chunk 0000 — FINDINGS (not clean)

Read against src/pg-latin/oecumenius-philippians/0000.md (cols 1259–1263). This batch is chunk 0000 only; it was not clean. One [lat:] fired (plate-verified); two Vulgate-conformations in the Latin logged as convention, not divergence.

The [lat:] site — plate-verified per the Pattern-16 numeral rule (extended to itacism pairs)

• 1261/1262 Greek: δίκαιόν ἐστιν ἐμοὶ τῷ ἀκριβῶς εἰδότι τὰ καθ᾽ ἡμᾶς ("the things that concern us"). Latin: qui exacte omnia novi quae apud vos sunt. The Greek is NOT Calfa damage — the scan's own OCR of the plate concurs (leaf ~647: "τῷ ἀκρ'θῶς εἶδότι τὰ καθ ἡμᾶς"), so both our witnesses of the Greek column agree against the Latin. ἡμᾶς/ὑμᾶς is the pronoun-itacism pair — the prose analogue of the numeral class (identical Byzantine pronunciation, endemic manuscript confusion; Allatius may have emended silently or read another exemplar). English follows the printed Greek; marker fired: [lat: the Latin gives quae apud vos sunt, the things that are with you].

Ladder row 1 — Calfa OCR damage, corrected silently, witness cited

• 1260 (head) Calfa's opening line "ΡΑΙΙΙ ΑΡΟΘΤΟΙΙ ΑΒ ΡΗΙΗΙΡΡΕΝβΕΘ ΕΡΙΘΤΟΙΑ. ΚΟΛΡΠΤ Ι." is the LATIN column's running title (PAULI APOSTOLI AD PHILIPPENSES EPISTOLA. CAPUT I.) misread in Greek lookalike glyphs — cross-column bleed, not the Greek column's text. The Greek column's own head (scan-verified) is ΠΑΥΛΟΥ ΑΠΟΣΤΟΛΟΥ Η ΠΡΟΣ ΦΙΛΙΠΠΗΣΙΟΥΣ ΕΠΙΣΤΟΛΗ → rendered; the CAPUT I belongs to the Latin column (carried by the twin) and is not rendered from the Greek.
• 1260 Catena sigla restored from plate form: ἴΦΔΤ. → — [ΦΩΤ.] · -ΧΡΥΣ.] → — [ΧΡΥΣ.] (scan-verified, see conventions).
• 1260 (block lemma Phil 1:1–2) Τμόθεος → Τιμόθεος · Σριστοῦ → Χριστοῦ (×2) · τρῖς → τοῖς · οὖσυι → οὖσιν · ἑν → ἐν · Παερὸς → Πατρὸς · Ἰσοῦ → Ἰησοῦ (standard text, Latin concurs).
• 1260 ἁξίωμα → ἀξίωμα · κοινῳ → κοινῷ · Χριστῶ → Χριστῷ · ὀνξμασι → ὀνόμασι · Κμὺχαριστία → Εὐχαριστία (ΚΕΦΑΛ. argument) — the last is the intrusive-Κ class known from PG 139's Calfa file, now seen in PG 118 too (also εὐχαΚριστίας → εὐχαριστίας, Κμεμνῆσθαι → μεμνῆσθαι at 1260).
• 1260 (block lemma Phil 1:3–5) μυεᾳ → μνείᾳ · δεύσει → δεήσει · ὑαὲρ πόνττων → ὑπὲρ πάντων · πρόι θμέρας → πρώτης ἡμέρας · νῶν → νῦν.
• 1260 Ἐπ πάσῃ → Ἐπὶ πάσῃ · Μετὰχαρᾶς → Μετὰ χαρᾶς (split/joined type) · Ἔν γὰρ → Ἔνι γὰρ (ἔνι = ἔνεστι, "it is possible"; Lat. contingit) · συοχῆς → συνοχῆς (2 Cor 2:4) · Τν → Τὴν · εὐχαρι στῶ → εὐχαριστῶ · Μέῖ ἱ → Μέγα τι (Lat. magnum quiddam) · Ὅτ… ε οὖ → Ὅτι… ἐξ οὗ · πράττετς → πράττετε.
• 1260/1261 Σομο παρόντες… τὰ ὗόν τὴν χρείαν → Ὥσπερ παρόντες… τὰ πρὸς τὴν χρείαν (Lat. mihi semper praesentes… mittentes quae necessaria erant; the [1261] anchor sits inside the restored phrase).
• 1261 ἐξ ἠς → ἐξ ἧς · Ὃτι → Ὅτι · stray ! after ἐναρξάμενος · ησοῦ → Ἰησοῦ · συγ. κοινωνούς → συγκοινωνούς (split) · ἄχρί → ἄχρι · ἣνδίκαιον → ἣν δίκαιον · τῆς χάρι. τος → τῆς χάριτος (split) · Ἠ/Ἡ → Ἢ ("Or:", thrice, the gloss's alternative-construal formula; Lat. aut) · ποκίλως → ποικίλως · τὰ καθ᾽ — see [lat:] above.
• 1264 ἐβεβαίο → ἐβεβαίου · μόνρν → μόνον · Ἰκανὴν → ἱκανὴν · εἰναι → εἶναι · ημῖν → ἡμῖν · ἱσχυροτέρους → ἰσχυροτέρους · Δὰ τί → Διὰ τί · Ἑπειδὴ → Ἐπειδὴ · συγκοινωνοῖὺς → συγκοινωνούς · Ἱν → Ἵνα · τοὺ → τοῦ.
• 1264 (block lemma Phil 1:8–11) Μίάρτωρ… ἑστιν… ὀς ἐπικοθῶ → Μάρτυρ… ἐστιν… ὡς ἐπιποθῶ (⚠ CORRECTED 2026-07-31 by the 0001–0005 agent, then re-verified here: this file first recorded the fix as Μάρτυς. The plate prints Μάρτυρ — twice, leaf 648, in BOTH the block lemma and the short lemma «Μάρτυρ γάρ μού ἐστι ὁ Θεός». Against the received Phil 1:8 μάρτυς, that is a scripture divergence, not OCR damage, and it belongs in the divergence list below, not here. No English change — both are "witness" — but the record was wrong, and a wrong record is how a real reading gets swept next time.) · Σριστοῦ → Χριστοῦ · Ἰνα → ἵνα (×2) · ἔτιν → ἔτι · πόσῃ οἰσθήσει → πάσῃ αἰσθήσει · διαφέροντο → διαφέροντα · ἡμέρο → ἡμέραν · αεπληρωμένοι → πεπληρωμένοι · τέν → τὸν · Θεοῖ → Θεοῦ.
• Restored question marks (Calfa · or nothing for the plate's ;): οὐκ ἔχων ἔθος; (Lat. ?) · Τί εὐχαριστῶ, φησὶ, τῷ Θεῷ; · Πῶς; (before Ὅτι ᾑρεῖτο) · Τί δὲ τὸ αἴτιον…ἔχειν;

7a″ worked cases (the deliverable list)

• 2 Cor 2:4 quote at 1260: Migne's Greek prints «Ἐκ γὰρ θλίψεως…καὶ συνοχῆς καρδίας» — NO πολλῆς. Rendered "out of affliction and anguish of heart," resisting the received "out of much affliction." The Latin column prints the Vulgate's Ex multa afflictione — conformation, logged, no marker.
• Phil 1:9 at 1264: Migne prints ἔτι μᾶλλον περισσεύῃ — a SINGLE μᾶλλον. Rendered "may abound yet more," resisting AV's "yet more and more." Latin: adhuc magis ac magis (Vulgate) — same class, logged, no marker.
• Phil 1:8 at 1264 — Μάρτυρ for the received μάρτυς, plate-verified twice on leaf 648 (block + short lemma). Migne's own consistency across both quotations is what makes it a reading rather than a slip. Logged here as the divergence it is; see the corrected entry above.
• Phil 1:6 block lemma: Calfa opens Πεποιθὼς δέ — the δέ kept ("And trusting…"); the Latin block (Confidens hoc ipsum) lacks a connective, received text lacks δέ; δέ may be Calfa noise but the reading construes and is not plate-checked — kept as our source prints, logged.

Deferred to the work's later chunks

• The chunk ends with the Phil 1:8–11 block lemma; its glosses open chunk 0001 (the twin's tail already carries their Latin). The 0001 agent should begin gloss-lemma consistency from THIS chunk's block rendering.

Continues cruces-0000.md (the Fable anchor). Its work-wide conventions section is binding here and was followed without exception; only additions, refinements and one correction of the anchor's record are set out below. Scan OCR consulted throughout: raw/scans/pg118/patrologiaecurs14migngoog_djvu.xml, leaves 648–658 (split on <OBJECT).

Conventions decided fresh in this batch (for the merge)

• ἁρπαγμός / ἁρπάζω / ἁρπαγή share one English root: "seizure / seize." Phil 2:6 is glossed by its own etymology all through chunk 0004 (Οὐχ ἁρπαγμὸν ἡγήσατο → Τουτέστι, κατὰ ἁρπαγήν → Ὅταν τις ἁρπάσῃ τι → ἐξ ἁρπαγῆς, ἐκ πλεονεξίας). AV's "robbery" breaks the chain at the verb, so the gloss would explain nothing (the stylus/epistylium precedent, style rule 2). Block lemma: "counted it not a seizure to be equal with God."
• πεποιθὼς / θαῤῥῶν / τεθαῤῥηκώς are kept apart, per work, always. Migne prints all three within eight lines at 1273–1274 and the Photius scholion turns on their equivalence: πεποιθὼς = "trusting" (anchor form, 0000) · θαῤῥῶν = "being confident" · τεθαῤῥηκώς = "having taken confidence." Allatius flattens all of them to persuasus / cum fiducia — the standing warning that his paraphrase is not a licence.
• Doctrinal vocabulary fixed (dialogue-anchor rule extended to catena): ὑπόστασις = hypostasis (Allatius gives persona; not followed) · πρόσωπον = person · οὐσία = essence · μορφή = form · φύσις = nature · ἐνέργεια = operation. The Arian/Sabellian argument at 1281 IS these distinctions.
• ἐρίθεια = "contentiousness" throughout (1268, 1277) · γογγυσμός = "murmuring" · διαλογισμός = "disputing" (1288, where the gloss defines διαλογισμοί as ἀμφιβολίαι, "doubts") · εἰλικρινής = "sincere" · ἀπρόσκοπος = "without offense" · ἄμεμπτος = "blameless" · ἄψεκτος = "beyond reproach" (the four stand in one gloss chain at 1265 and 1288 and must not collapse).
• Migne's unclosed guillemets are closed silently — the anchor's own practice (0000, ἀλλὰ, «συγκοινωνοῖὺς, closed as «fellow-partakers»). Sites this batch: 1268 «Οἱ μὲν ἐξ ἐριθείας (closed after "contentiousness," matching the Latin's «Alii quidem ex contentione.») · 1277 «Πῶς; (closed after "How?"; the PLATE itself prints the stray «, leaf 654) · 1281 «Οὐχ ἁρπαγμὸν ἡγήσατο (plate also unclosed, leaf 656) · 1284 «Μακάριοι… τῶν οὐρανῶν · 1288 «Ὑπὲρ τῆς εὐδοκίας.
• Guillemet parity with the Latin twin is NOT achievable in PG and must not be forced. Allatius marks his lemmata on his own convention: he guillemets «Quid enim» and «Alii quidem ex contentione.» where Migne's Greek prints no marks at all, and he sets whole block lemmata inside per-line « … » that the Greek column never carries. The English follows Migne's Greek, per Pattern 6's provenance rule. verify-english check 7 (guillemet counts equal the twin's) was written for PL, where the twin is the source text; on this work it will report differences that are correct.
• Cross-column bleed, second and third instances (anchor class, ΚΟΛΡΠΤ Ι.): 1276 ΚΟΑΡΙΤ ΙΙi = the Latin column's CAPUT III in Greek lookalike glyphs, and 1267/1268 ΖΔΑὗΤ Ι. at the column break. Neither is Greek; not rendered. The Greek column's own head (ΚΕΦΑΛ. Γʹ) is rendered separately and in its own place.

Correction to the anchor's record (chunk 0000) — no English change

• cruces-0000.md lists «Μίάρτωρ → Μάρτυς» among its silent corrections. The plate does not support Μάρτυς. Leaf 648 prints the Phil 1:8 block lemma as Μάρτυρ γάρ μού ἐστιν ὁ Θεός and the short lemma at 1264/1265 as «Μάρτυρ γάρ μοῦ ἔστι ὁ θεός.» — Μάρτυρ in both places, against the received Μάρτυς. The English ("For God is my witness") is unaffected, so nothing is re-cut; but the crux line should read Μάρτυρ (Migne's form, plate-verified) — not normalized, and this belongs in the 7a″ divergence list below, not in the OCR list. Logged rather than edited, per the do-not-churn rule.

The 7a″ deliverable — where Migne's printed Greek departs from the received text

Named word by word. This is the list, not a summary of one.

1. Phil 1:20 (block lemma, 1272): Migne prints ἐν τῷ ΣΤΟΜΑΤΙ μου, "in my MOUTH," where the received text reads σώματι, "in my body." Plate-verified, leaf 651: καὶ νῦν μεγαλυνθήσεται Χριστὸς ἐν τῷ στόματί μου. Two lines later Migne's own short lemma prints σώματί (leaf 652: «Ἐν τῷ σώματί μου.», and the gloss's paraphrase μεγαλυνθήσεται ὁ Χριστὸς ἐν τῷ σώματί μου), and the Latin column reads in corpore meo. Rendered exactly as printed in each place — "in my mouth" in the block, "in my body" in the gloss — with [var: Gk σώματί μου, "in my body"] on the block. No [lat:]: the columns do not descend from different exemplars here; the plate's own gloss two lines down refutes that, and Pattern 16 is not a defect marker. This is the exact trap the 7a″ clause exists for: AV's "Christ shall be magnified in my body" is the sentence that arrives unbidden.
2. Rom 12:18 quoted at 1265: Migne prints the IMPERATIVE εἰρηνεύετε ("live at peace"), where the received text has the participle εἰρηνεύοντες. Plate-verified, leaf 649. The Latin agrees (in pace vivite). Rendered as an imperative.
3. John 1:14 quoted at 1281–1284: Migne prints Καὶ εἴδομεν τὴν δόξαν αὐτοῦ ὡς Μονογενοῦς παρὰ Πατρός — εἴδομεν for the received ἐθεασάμεθα, and the second δόξαν is absent. Plate-verified, leaf 656. The Latin column restores the Vulgate's doubled gloriam… gloriam quasi Unigeniti — Vulgate conformation, logged, no marker (calibration rule). No [var:]: the sense is not changed by either.
4. Phil 1:23 (block lemma, 1272): πολλῷ μᾶλλον κρεῖσσον with NO γὰρ (received πολλῷ γὰρ μᾶλλον κρεῖσσον). Plate-verified, leaf 652.
5. Phil 2:3 (block lemma, 1277): μηδὲν κατὰ ἐρίθειαν, ἢ ΚΑΤΑ κενοδοξίαν — Migne repeats κατά, which the received text does not. Plate-verified, leaf 655.
6. Phil 2:4 (block lemma, 1277): ἀλλὰ καὶ τὰ ἑτέρων ἕκαστος — singular ἕκαστος where the received text has plural ἕκαστοι. Plate-verified, leaf 655 (which also shows τὰ, against Calfa's τὸ — that one is our file's damage, see the OCR list).
7. Phil 1:16–17 (block lemma, 1265): Migne prints the Byzantine order (ἐριθείας clause first) and θλίψιν ΕΠΙΦΕΡΕΙΝ (received/critical ἐγείρειν). Rendered "thinking to bring affliction upon my bonds."
8. Ps 32(31):22 quoted at 1272: Γένοιτο, ΚΥΡΙΕ, τὸ ἔλεός σου ἐφ᾽ ἡμᾶς… ἐπὶ ΣΟΙ — vocative moved forward, and dative σοί for the LXX's accusative σέ. Plate-verified, leaf 651.
9. Phil 1:6 short-lemma pair (1272–1273): «Καὶ τοῦτο ΘΑΡΡΩΝ οἶδα,» followed immediately by «Καὶ τοῦτο ΠΕΠΟΙΘΩΣ οἶδα.» — Migne prints a paraphrastic first lemma against his own block lemma's πεποιθώς, and the [ΦΩΤ.] scholion then glosses πεποιθώς by τεθαῤῥηκώς. Plate-verified, leaf 653. All three rendered distinctly (see conventions).
10. Phil 1:19 short lemma (1272): «Κατὰ τὴν ΚΑΡΑΔΟΚΙΑΝ.» without the ἀπο-, though the block lemma prints ἀποκαραδοκίαν and the gloss immediately resumes Ἀποκαραδοκίαν φησί. Plate-verified, leaf 651. Rendered "expectation" in the short lemma against "earnest expectation" in the block and gloss — the distinction is on the page.
11. Phil 2:15 (block lemma, 1288): ἀμώμητα (Byzantine) against the critical ἄμωμα — noted for completeness; rendered "without blemish."

Not divergences, deliberately: Phil 1:9's single μᾶλλον was already logged by the anchor; note that the GLOSS on it at 1265 prints the doubled ἔτι μᾶλλον καὶ μᾶλλον (plate-verified, leaf 649). Both are rendered as printed — "yet more" in the lemma, "yet more and more" in Oecumenius's paraphrase. The divergence is internal to Migne's page and is not smoothed in either direction.

[lat:] markers fired — three, each plate-verified on both columns

1. 1265 / Latin 1266 — Rom 12:18, a missing protasis. Greek: Εἰ δυνατὸν τὸ ἐξ ὑμῶν μετὰ πάντων ἀνθρώπων εἰρηνεύετε; (plate leaf 649). Latin: Quantum ex vobis est, cum omnibus hominibus in pace vivite? (plate leaf 649) — Εἰ δυνατόν is simply absent, and the Latin is not conforming to the Vulgate here either (Vulg. si fieri potest), so the Vulgate-convention rule does not cover it. A reader of the Latin gets an unconditional command. Marker: [lat: the Latin column omits Εἰ δυνατόν, giving the command unconditionally — Quantum ex vobis est, cum omnibus hominibus in pace vivite].
2. 1272 / Latin 1271 — "salvation" construed to opposite effect. Greek: Σωτηρίαν δέ φησι, τὸ ὅσον οὐδέπω μαρτυρίαν — the martyrdom that is all but here (plate leaf 651). Latin: Salutem autem dicit quod nondum pateretur martyrium — that he was not yet suffering martyrdom (plate leaf 651). The two columns give the reader contrary accounts of what σωτηρία means in Phil 1:19. English follows the Greek; marker fired.
3. 1284 / Latin 1283 — a polarity the columns disagree on. The batch's strongest find. Greek: Ποῦ εἰσιν οἱ λέγοντες ὅτι κατὰ ἀνάγκην καὶ ἐπιταγὴν τοῦ Πατρὸς ἑκὼν ἐνηνθρώπησεν; — plate-verified, leaf 657, which prints ἑκὼν ("willingly"), producing a sentence that contradicts its own "by necessity and command." Latin: invitus est incarnatus ("unwillingly was he made man"), plate-verified leaf 657 — coherent, and confirmed as deliberate by the same column's voluntariam dejectionem for τὸ ἐκούσιον eight lines later. This is not Calfa damage (both our witnesses of the Greek agree) and it is exactly the class runbook 4a hunts. Rendered literally — "he willingly became man" — and marked. Never smoothed to "unwillingly."

[var:] fired — one

• 1272 Phil 1:20, [var: Gk σώματί μου, "in my body"]. See divergence 1.

Candidates examined and NOT marked, with reasons

• 1268 / Lat. 1267 Greek Εἰ γὰρ μὴ θεῖον ἦν, φησί, τὸ κήρυγμα… (singular, plate leaf 650) vs Latin Nam aiebant: Nisi divina esset praedicatio… (plural imperfect). Allatius has construed the habitual φησί as the brethren's speech, which the context permits. Free rendering, not divergence.
• 1277 / Lat. 1278 Greek short lemma «Τὸ ὑπὲρ Χριστοῦ» vs Latin «Hoc a Christo» (donatum est a Christo, "from Christ") — a real difference of sense (ὑπέρ vs ὑπό). Not marked: Migne's own Latin column carries the note that settles it — (Nam ὑπὲρ capitur pro ὑπὸ, id est a), printed at Latin col. 1278, plate leaf 655 — and Oecumenius's own Greek gloss two lines later reads ἐχαρίσθη ὑπὸ Χριστοῦ (plate leaf 655). The divergence is adjudicated on the page. This is the second instance of the Nicetas "plate's own apparatus" rule and the first where the adjudicating note sits in the LATIN column; §8 Q3 calibration datum.
• 1285 / Lat. 1286 Phil 2:12 block lemma: Greek ἀλλὰ νῦν πολλῷ μᾶλλον (plate leaf 658) vs Latin sed tunc multo magis (plate leaf 657). Not marked: the Latin departs here from the Greek and from the Vulgate (nunc) at once, which points to the Latin column's own compositor rather than to a divergent exemplar; Pattern 16 records column disagreement, not Latin-side type defects, which this edition does not mark because it does not edit the Latin. Logged so the claim is checkable.
• 1285 / Lat. 1286 Greek ὅταν ὡς κριτὴς καθίσῃ ("when he sits as judge") vs Latin quando ut judex descenderit. Free/erroneous rendering of the same event; no fact a reader would carry away differently.
• 1284 / Lat. 1283 Latin nos qui natura æquales sumus et æqualiter humiles et abjecti amplifies Greek οἱ φύσει ὁμότιμοι καὶ ὁμοίως ταπεινοί (plate leaf 656). Allatius's norm.
• 1272 / Lat. 1271 Latin renders ὑποστάσεις by personas in the heresy catalogue (1272). Conventional Latin equivalence, not a divergence; the English keeps "hypostases" per the fixed vocabulary above.

Vulgate conformations in the Latin column — logged, never marked

• Lat. 1266 Rom 12:18 is not conformed (see [lat:] 1 above) — recorded because it is the exception that shows the convention is not automatic.
• Lat. 1282 John 1:14 conformed to the Vulgate's doubled gloriam.
• Lat. 1283 1 Cor 2:9 and Matt 5:3 conformed.
• Lat. 1287 Ps 2:11 conformed (Servite Domino in timore).

A defect in Migne's own Latin column, noted not marked

• Lat. col. 1286 carries the running title EPIST. AD GALAT. in the middle of the Philippians commentary (plate leaf 657, and the committed twin src/pg-latin/oecumenius-philippians/0005.md reproduces it). Migne's error, on the verifier side only; nothing in the English depends on it.

Catena sigla met, and every restoration

Four scholia in this batch, three sources. [ΒΑΣΙΛ.] is new to the work and was verified against the scan before its form was fixed.

The one unverified restoration, stated plainly: at 1281 Calfa prints ππ[Τ. opening the Βούλεσθε εἰδέναι scholion. The Greek column for Migne cols 1279–1281a is absent from our scan — leaf 655 carries the Greek only as far as Ὅταν γὰρ τὸ ἴδιον συμφέρον, and leaf 656 resumes at …νετο. Οὐκοῦν καὶ μορφὴν Θεοῦ; the intervening Greek column was not OCR'd in patrologiaecurs14migngoog. The restoration to — [ΦΩΤ.] rests on the surviving final Τ (ΧΡΥΣ. and ΒΑΣΙΛ. both end in Σ/Λ) and on the four verified [ΦΩΤ.] sites in the same span. It is a strong inference, not a plate check, and is recorded as such.

Ladder row 1 — Calfa OCR damage corrected silently, witness cited

Itemized by column. Where the plate was consulted the leaf is named; where the Latin twin supplied the sense that is said instead.

1265 (chunk 0001). ἡ μορφή δούλο.. → ἡ μορφὴ δούλου · ιΠσου → Ἰησοῦ · ὀισκειμένοις → διακειμένοις · ποθεῖσθαίν → ποθεῖσθαι · παρ → παρ᾽ · ἣ καὶ περὶ αῶν ἄλλων αῃσὶν → ἢ καὶ περὶ τῶν ἄλλων φησὶν (leaf 649) · Ιολλοὺς → Πολλοὺς · Λετὰ → Μετὰ · Ἴα → Ἵνα · οἵα… οἶ διὰ κενοδοδίαν → οἷα… οἳ διὰ κενοδοξίαν · ἐπιΚμελοῦνται → ἐπιμελοῦνται (intrusive-Κ class) · τὸν] διδάσκαλον (stray bracket). 1265, a lost tail restored from the plate: Calfa runs τοῦ καὶ συνεργοῦνΚΕΦΑΛ. Βʹ. — the plate (leaf 649) reads τοῦ καὶ συνεργοῦντος πρὸς τοῦτο. before the chapter head; Latin qui etiam in hoc auxilium praebet. Restored, not marked [ed:] (Pattern 13 is for holes a crux cannot carry; this one is plate-verified and four words long).

1265 (chapter head). Διήγησις τῆς ἑαυτοῦ διαγωγὴς… προθυμίας,. → διαγωγῆς … προθυμίας. (leaf 649). 1265 (block lemma Phil 1:12–18). εἰς αροκοπὴν → προκοπὴν · Εἰαγγελίου ×3 → Εὐαγγελίου · ἐλήλυοεν → ἐλήλυθεν · ἐν λριστῷ → ἐν Χριστῷ · ὁλῳ → ὅλῳ · πασι → πᾶσι · Τρές μὲν → Τινὲς μὲν · Οἱ μέν ἐξ] (stray bracket) · καταγηέλλουσιν → καταγγέλλουσιν · τοῆς δεσμοῖς → τοῖς δεσμοῖς · Οἱδέ → Οἱ δὲ · Τίγάρ· → Τί γάρ; · Πλὴνν → Πλὴν · Χχιστὸς → Χριστὸς · χαούσοααι → χαρήσομαι (all leaf 649).

1268. Εἰκὸς ἦναὐτοὺς → Εἰκὸς ἦν αὐτοὺς · δεικνὸς → δεικνὺς · προκοπή, → προκοπή. · θτι → Ὅτι · οηλοῦν → δηλοῦν · τὰ κατ’´ αὐτίν → τὰ κατ᾽ αὐτόν · Οὕτε → Οὔτε · ἐν Χριστῳ → ἐν Χριστῷ · εἰοὐκ → εἰ οὐκ · Ἡ, ἐκείνων → Ἢ, ἐκείνων · τὸν ἑν οὐρανοῖς → ἐν οὐρανοῖς · ὡφελεῖσθαι → ὠφελεῖσθαι · εἶπερ Βεβαιώσει → εἶπε. Βεβαιώσει (leaf 650; Lat. Hoc etiam superius dixit) · ἦνα ἀκούσας → ἵνα ἀκούσας · ἀποκτείνη → ἀποκτείνῃ. 1268, a lost parenthesis restored: Calfa Ἡ. ὅτι αὐτοὶτεμᾶσθαι βουλόμενοιεγάλη γὰρ ἦν ἡ τιμὴ τοῦ Παύλου) → plate (leaf 650) Ἢ, ὅτι αὐτοὶ τιμᾶσθαι βουλόμενοι (μεγάλη γὰρ ἦν ἡ τιμὴ τοῦ Παύλου); Lat. magnus siquidem erat Pauli honor.

1268/1269. σασι → Ἴσασι (leaf 650; Lat. Sciunt) · Τοῦο οὗν → Τοῦτο οὖν · τὸ ὰ λόγον → τὸν λόγον (leaf 650) · εἱ μὴ → εἰ μὴ · ποίω σκοπῳ → ποίῳ σκοπῷ · όπότε ὁρθῶς → ὁπότε ὀρθῶς · ἔδοκεν → ἔδωκεν · Ἀλλὰσὺ → Ἀλλὰ σὺ · εἰρήκει,καὶ γὰρ → εἰρήκει, (καὶ γὰρ… (the plate, leaf 651, prints the parenthesis). 1269, the largest single loss in the batch: Calfa runs ὁ Ἀπόστοστὸς [1269] καταγγέλλεται.» — the plate (leaf 650 end / 651 start) reads ὁ Ἀπόστολος, τῷ εἰπεῖν «Εἴτε προφάσει, εἴτε ἀληθείᾳ Χριστὸς καταγγέλλεται.», and the Latin (1270) has the whole of it. Restored silently; the [1269] anchor kept in place.

1272 (chunk 0002). πονηρὰνο γὰρ πλείους → πονηρὰν (οἱ γὰρ πλείους (leaf 651) · ὀιὰ τῆς → διὰ τῆς · Τησοῦ → Ἰησοῦ · παρ´ αὐπῶν → παρ᾽ αὐτῶν · ὸφειλομένην → ὀφειλομένην · ἰΕἰ γὰρ → Εἰ γὰρ · δαψῖλεια → δαψίλεια · θτι ἐν οὐδενὶ → Ὅτι ἐν οὐδενὶ · ν γὰρ ἀποκτείνωσι → Ἂν γὰρ ἀποκτείνωσι (leaf 651) · Οίον → Οἷον · λογιζομένων τῷν ὁρώντων → τῶν ὁρώντων. 1272. Παῆλος → Παῦλος · Εἵτε ἐξα·ρεῖταί → Εἴτε ἐξαιρεῖταί (leaf 652) · κἂν οεῃ με → κἂν δέῃ με · ἐν τῳ σώματί → ἐν τῷ σώματί · Χριατὸν → Χριστὸν · ᾤοντο → ᾤοντο · οὐχ οὔτως → οὐχ οὕτως · τὴν μὲν ποώτιν → τὴν μὲν πρώτην · ἐξήαπασέ → ἐξήρπασέ · θορυφῆσθαι → θορυβεῖσθαι (leaf 652). 1272/1273 (block lemma Phil 1:21–26). Σριστὸς → Χριστὸς · ἀποθαιεῖν → ἀποθανεῖν · καρπδς → καρπὸς · αἱρήσομαι οὐ γνωρίζω ✓ · ἀναλδσαι → ἀναλῦσαι · εἰν αι → εἶναι · μᾶλλρν → μᾶλλον · Τὸ δὴ ἐπιμένειν → Τὸ δὲ ἐπιμένειν (leaf 652) · ἀναγκαιότειον → ἀναγκαιότερον · ὃτι μένω → ὅτι μενῶ (leaf 652) · σιμπαραμενῶ → συμπαραμενῶ · πᾶσιν ὑμῶ → πᾶσιν ὑμῖν · ὑμών Χροκοπὴν → ὑμῶν προκοπὴν · ἵνατὸ καιὶχημα → ἵνα τὸ καύχημα · ἐν λριστῷ → ἐν Χριστῷ · διὰ κῆς → διὰ τῆς. 1273. Κζῶ → ζῶ (intrusive Κ) · σῶν στεσάνων → τῶν στεφάνων · Κἰ δὲ τὸ ζῆν → Εἰ δὲ τὸ ζῇν · μή τς νομίση → μή τις νομίσῃ · ἀγαθῶν]· (stray bracket). 1273, restored across the column break: Calfa prints the garbage run οὐκ οἶδα, οὐδὲ γνωρίζω τί οοωοοΟΟΟΘ [1273] φορεῖν τῷ Θεῷ — the plate (leaf 652) reads …τί αἱρήσομαι καὶ ἐπιλέξομαι, τὸ ζῇν, φησί, καὶ καρπο-φορεῖν τῷ Θεῷ; Latin nondum mihi decretum est, neque satis novi quid malim aut quid eligam, an vivere ac fructificare Deo. Restored; the [1273] anchor kept inside the restored word, as the plate breaks it. 1273/1274. τοῦ τε ὰναλῦσαι → ἀναλῦσαι · Εἰτα. ὡ Παῦλε → Εἶτα, ὦ Παῦλε · οὐσαν → οὖσαν · Διατί ὀὲ → Διὰ τί δὲ · οἱκείου/ὡφέλειαν → οἰκείου/ὠφέλειαν · ἐβεβαίουκαὶ ἄλλους πρρσελάμβανε → ἐβεβαίου καὶ ἄλλους προσελάμβανε · Πολλῷ Κμᾶλλον → Πολλῷ μᾶλλον (intrusive Κ) · πολλῶ μᾶλλον → πολλῷ μᾶλλον · τὸ. εἶν αι → τὸ εἶναι · δ θαῤῥῶν → ὃ θαῤῥῶν.

1276 (chunk 0003). Κἐαυτοῦ → ἑαυτοῦ (intrusive Κ) · δισταγμῳ → δισταγμῷ · δψομαι → ὄψομαι · ες αὐτὴν → εἰς αὐτὴν · Ἰἱ γὰρ → Εἰ γὰρ · περισσεύσοι ✓ · βειοῦν → βιοῦν · ὡφέλειαν → ὠφέλειαν · Ἀν γὰρ → Ἂν γὰρ · συμπαραμενω → συμπαραμενῶ. 1276 (chapter head, a sense-bearing correction). Calfa prints τῆς ἀνθέου ζωῆς; the plate (leaf 654) prints ἐνθέου and the Latin vitamque divinam. Rendered "the life inspired by God." Calfa's ἀνθέου is not a word in this sense and would have inverted the head. 1277. ποἸτεύεσθε → πολιτεύεσθε · εἰτε ἐλὸὼν → εἴτε ἐλθὼν · μηδενλ → μηδενί · ἀπωλείσς → ἀπωλείας · ὑμῆν → ὑμῖν · Κπρὸς ἀρετὴν → πρὸς ἀρετὴν (intrusive Κ) · εἰ πεν → εἶπεν · εὐφρανθήσομαιτοῦτο γὰρ προσυπακουστέον) → εὐφρανθήσομαι (τοῦτο γὰρ προσυπακουστέον) (leaf 654) · ἀπών → ἀπὼν · ἁγάπης → ἀγάπης · Μιᾶ ψυχῇ → Μιᾷ ψυχῇ · τῆ ἀγάπῃ → τῇ ἀγάπῃ · ὁνειδίζωσιν → ὀνειδίζωσιν · στιῤῥότητος → στεῤῥότητος (leaf 654) · τὰ ἡμων → τὰ ἡμῶν · ὲπάγει → ἐπάγει · Τλν αὐτὸν → Τὸν αὐτὸν · οἶον ἴδετε → οἷον εἴδετε · ΕΙ/Εἵ/Εἶ τις → Εἴ τις (×3) · έν Χριστῷ → ἐν Χριστῷ · εἰ τις´ σπλάγχνα → εἴ τις σπλάγχνα · Ἰνα → ἵνα · τὸ ἑν βρονοῦντες → τὸ ἓν φρονοῦντες · μηδέν → μηδὲν · ἑαυτῶν μὴ τὰ ἑαυτῶν ✓ · τὸ ἑτέρων → τὰ ἑτέρων (leaf 655) · Bαβαὶ → Βαβαὶ (Latin B for Β, twice) · πρὸς αὸν ἕτερον → πρὸς τὸν ἕτερον · Ἱνα → Ἵνα · Πέγα μὲν γὰρ → Μέγα μὲν γὰρ · ὅτι δ δαὶνά μου → ὅτι ὁ δεῖνά μου (leaf 655) · τοὺτο → τοῦτο.

1279 (chunk 0004), a lost lemma and clause restored. Calfa runs «Μὴ τὰ [1279] ἕκαστος ἀφεὶς, τοῦ συμφέροντος γίνηται τοῆ ἀὲολ φοῦ — the plate (leaf 655) reads «Μὴ τὰ ἑαυτῶν ἕκαστος.» Ὅταν γὰρ τὸ ἴδιον συμφέρον ἕκαστος ἀφεὶς, τοῦ συμφέροντος γίνηται τοῦ ἀδελφοῦ. Restored; the [1279] anchor placed at the lemma's close. 1279. είρήνην → εἰρήνην · ὃ´ τὸ έον ἦν → ὃ πλέον ἦν (leaf 655) · βίος,. → βίος. 1279/1280 (block lemma Phil 2:5–8). Τοῶτο γὰρφρονείσθω κν ὑμῖν → Τοῦτο γὰρ φρονείσθω ἐν ὑμῖν · Χρισιῷ → Χριστῷ · οὐχ ἀρπαγμ ν ήγοόσατο εἶναι σα Θεῷ → οὐχ ἁρπαγμὸν ἡγήσατο τὸ εἶναι ἴσα Θεῷ (the τὸ is confirmed by the short lemma at leaf 656) · ἐκένωσεμρφὴν → ἐκένωσε μορφὴν · εὐρεθεὶς → εὑρεθεὶς · ἐαυτὸν → ἑαυτὸν · σταιροῦ → σταυροῦ. This block lemma and the scholion after it fall in the stretch our scan does not carry (see the siglum note); the restorations rest on the received text, on Migne's own short lemmata further down, and on the Latin twin. 1281. δεδέλωκε → δεδήλωκε · Μαοίωα τὸν Ποντικόν → Μαρκίωνα τὸν Ποντικόν · Καὶ ψὰρ → Καὶ γὰρ · Τν τούτοις → Ἐν τούτοις · λάρκελλος → Μάρκελλος · Σαβελλιον → Σαβέλλιον · ἡγήσατοτὸ εἶναιίθα Θεῷ → ἡγήσατο τὸ εἶναι ἴσα Θεῷ · σον νἀρ → Ἴσον γὰρ · ψιλὰ φής → ψιλὰ φῂς · Εἰποικεν → Εἴποιμεν · Τί ἐστ, → Τί ἐστι, · ἄν´γωπος → ἄνθρωπος (leaf 656) · τῶν δύε → τῶν δύο (leaf 656) · ὰπόδειξις → ἀπόδειξις · Ἢδει → ᾜδει · θείτητος → θεότητος · Ὁ πέντς → Ὁ πένης · ταπεινωδῆναι → ταπεινωθῆναι · Ὡς ἐν μορφῆ Θεοῦ ὑπάρχων → Ὃς ἐν μορφῇ Θεοῦ ὑπάρχων (from the block lemma; not plate-verifiable, see the siglum note) · Πνευματος → Πνεύματος. 1281/1284. ἡδύνατο κἀν → ἠδύνατο κἂν (leaf 656) · ταπεινω.θῆναι → ταπεινωθῆναι · βίόλον → βίβλον · ἴδιον. εὐχερῶς → ἴδιον, εὐχερῶς · ὄταν θέλη → ὅταν θέλῃ · ὀμοίωμα → ὁμοίωμα · Διατί → Διὰ τί · τετράῳθαι → τετράφθαι · οία ανθρωπός → οἷα ἄνθρωπός · ἴδομεν → εἴδομεν · γίρ νεται → γίνεται · Ο δὲ παρὰ ἀνθρώποις ὑψηλοί → Οἱ δὲ… ὑψηλοὶ · ἀνθρώπουούκ → ἀνθρώπου οὐκ · τούτἐστιν → τουτέστιν · τὸ εἰναι Θεὸς → τὸ εἶναι Θεὸς · ᾑδει ἔχον → ᾔδει ἔχων (leaf 657) · συναῖδίου → συναϊδίου · καταβιβασθῆ → καταβιβασθῇ · Μορφὴν δούλουν λαβών → Μορφὴν δούλου λαβών (leaf 657). 1284, a lost clause restored: Calfa runs ἐταπείνωσεν ἑαυτὸν δι᾽ ἡμᾶς, οἱ [1284] ἀλλήλοις ταπεινοῦσθαι — the plate (leaf 656) reads …οἱ φύσει ὁμότιμοι καὶ ὁμοίως ταπεινοὶ οὐκ ὀφείλομεν ἀλλήλοις ταπεινοῦσθαι…;; Latin nos qui natura æquales sumus et æqualiter humiles et abjecti, annon debemus invicem humiliari. Restored as the rhetorical question the plate prints; the negation οὐκ is carried, not smoothed.

1285 (chunk 0005). ἠχερίσατο αὐτῇ → ἐχαρίσατο αὐτῷ · Ἰνα ἐν τῳ → ἵνα ἐν τῷ · σᾶν γόνυ → πᾶν γόνυ · ἐπουρανίον → ἐπουρανίων · Χριστίς → Χριστὸς · μηθαμοῦ → μηδαμοῦ · τῷ Υὸῷ Θεῷ Κντι → τῷ Υἱῷ Θεῷ ὄντι (intrusive Κ, leaf 657) · ἐξωολογήσατο → ἐξωμολογήσατο · μείζων Κοὖν → μείζων οὖν (intrusive Κ) · τοῦ πρὸ τς → τοῦ πρὸ τῆς · ἰνανθρωπήσεως → ἐνανθρωπήσεως · φασιν· → Διό φασιν· (leaf 658) · Ἠ ὅτι → Ἢ ὅτι. 1285/1288. μετὸ φόβου καὶ γρόμου → μετὰ φόβου καὶ τρόμου · κατεργάζισθε → κατεργάζεσθε · λουλεύσατε → Δουλεύσατε · αὐτῶ ἐν τρόμῳ → αὐτῷ ἐν τρόμῳ · Ἀιὰ ταῦτα → διὰ ταῦτα · ἑαοτῶν σιντηρίαν → ἑαυτῶν σωτηρίαν · τὸ. ἀνεργεῖν → τὸ ἐνεργεῖν (leaf 658) · Ἡθέλησε → Ἠθέλησε · Κάῖν → Κάϊν · Ἀπεὲέχθη → ἀπεδέχθη · φἡσι → φησι. 1288, a lost clause restored across the column break: Calfa runs αὐτός ἐστιν ὁ τελειῶν καὶ οἰονεὶ μορΔύναται [1288] δὲ τό- — the plate (leaf 658) reads …καὶ οἰονεὶ μορφοποιῶν τὴν ἡμετέραν καὶ θέλησιν καὶ ἐνέργειαν. Δύναται δὲ τό, Ὁ Θεὸς γάρ ἐστιν ὁ ἐνεργῶν…; Latin ipse qui perficit et veluti formam addit nostræ et voluntati et operationi. Restored; the [1288] anchor placed at the sentence break the plate marks. 1288. χωρ ς γογγυσμων → χωρὶς γογγυσμῶν · συνεργόνς → συνεργόν; · ἐνεργων → ἐνεργῶν · εἰπεν ἐν φόβω → εἶπεν ἐν φόβῳ · τὸ Ηέλειν → τὸ θέλειν · εὐγνωμοπύνν ἃ Παῦλος → εὐγνωμοσύνην ὁ Παῦλος · τῶ Θεῶ → τῷ Θεῷ · συνΚεργεῖ → συνεργεῖ (intrusive Κ) · Ὑπὲρ σῆς εὐδοκίας → Ὑπὲρ τῆς εὐδοκίας · γογγυσμών καὶ διαλοσεσμῶν → γογγυσμῶν καὶ διαλογισμῶν · Ἰνα/Ἰλα → ἵνα/Ἵνα · ἀκέραιο → ἀκέραιοι · ἐν μέσω → ἐν μέσῳ · σκολῶς → σκολιᾶς · φωστηρες → φωστῆρες · κίύχημαΚέμοι → καύχημα ἐμοὶ (intrusive Κ) · ἀχαριατεῖ → ἀχαριστεῖ · εἴλικρινεῖς → εἰλικρινεῖς · ἐνόρετον → ἐνάρετον · Ἐν οἷς ταί νεσθε → Ἐν οἷς φαίνεσθε · σῶν ἁμαρτιῶν → τῶν ἁμαρτιῶν · ἐπι έχοντες → ἐπέχοντες.

Restored question marks (the anchor's class, logged not itemized per site)

Calfa prints · or nothing where the plate prints ;. Restored from sense and the Latin column at: 1265 Πῶς οὖν ἀλλαχοῦ φησιν…εἰρηνεύετε; · 1268 Καὶ τί πρὸς τοῦτο; · 1268 Τί γάρ; · 1272 Τί λέγω χαίρω; · 1272 Πῶς ἐλύετο τὰ δεσμά; (and the two following) · 1273 Τί οὖν φησιν; · 1273 Τί δέ ἐστιν ὁμοίωμα ἀνθρώπου; · 1273 Διὰ τί δὲ εἶπεν ὁμοίωμα ἀνθρώπου; · 1274 Διὰ τί δὲ τοῦτο; · 1288 Θεὸν ἔχων συνεργόν; One in the other direction: at 1284 Calfa prints ; after παρὰ Πατρός in the John 1:14 quotation, where the sentence plainly continues (τουτέστιν, οἷα Μονογενοῦς); rendered as the statement it is. One left standing: at 1279 Calfa prints Ἡ ἄκρα ταπεινοφροσύνη; — rendered as the answer to the preceding question ("The utmost lowliness of mind."), the mark being unsupportable there; the stretch is not plate-verifiable.

Latin-twin pass (pilot §6) — result per chunk, including the clean ones

Read against src/pg-latin/oecumenius-philippians/0001.md–0005.md.

Chunks where the deliberate twin pass found nothing to mark: 0003 and 0005. Both were read in full against their twins and both are named here so that a pass made can be told from a pass not made.

Polarity read (runbook 4a), our own defects: every negation Migne prints is carried, including where it makes the sentence fight itself. The two sites where that was live are recorded above and neither was smoothed: the ἑκὼν at 1284 ("by necessity and the Father's command he willingly became man"), and the restored οὐκ ὀφείλομεν at 1284, which the English keeps as the rhetorical question the plate prints rather than flattening to a statement.

Deferred to the next agent

• Chunk 0005 ends mid-gloss on Phil 2:16 (Γινωσκόμενοι ὅτι ὑμεῖς τῶν ζησομένων ἐστὲ τὴν ἀθάνατον ζωήν); the Latin twin's tail already carries the antithesis that completes it (nam illi ex illis sunt qui morte immortali morientur) and the «Ut gloriari possim» lemma that opens chunk 0006. Begin from this batch's block-lemma renderings of Phil 2:14–16.
• The scan gap should be recorded work-wide: the Greek column for Migne cols 1279–1281a is not present in patrologiaecurs14migngoog. Any later sweep that wants a plate check in that span needs a second scan item (the twin's own provenance block already names patrologicursus21migngoog as an alternate witness used for one Latin page).

Batch: 0006, 0007, 0008, 0009. Anchor followed: 0000.md pair + cruces-0000.md work-wide conventions (binding). Greek plate witness used throughout: raw/scans/pg118/patrologiaecurs14migngoog_djvu.xml, leaves 657–666. Note the scan's leaf gap in this stretch: the primary item carries cols 1289/1290 (leaf 659), 1293/1294 (660), 1297/1298 (661), 1301/1302 (662), 1303/1304 (663), 1305/1306 (664), 1307/1308 (665) — and has no leaf for cols 1291–1292, 1295–1296, 1299–1300 (the same gap the twin's provenance records by falling back to patrologicursus21migngoog for those Latin pages, an item we do not hold locally). Every "unverified" note below means that gap, not negligence.

Conventions decided fresh in this batch (for the merge)

• καύχ- family is kept on one English root: "boast." καύχημα = boast · καυχάομαι = boast · καύχησις = boasting. The gloss at 1297 argues from the root (τῶν ἐν Χριστῷ καυχωμένων, against confidence in the flesh), so AV's "rejoice"/RSV's "glory" would break the argument it carries. Not varied.
• περιτομή = "circumcision" · κατατομή = "concision" · κατατέμνειν = "to cut in pieces." The gloss at 1297 IS the etymology (οὐδὲν ἕτερόν ἐστιν ἡ περιτομὴ, ἢ σώματος κατατομή); the English pair must share the "-cision" element, as Migne's own Latin does (circumcisio / concisio).
• The loss/gain vocabulary of Phil 3:7–9, fixed: κέρδος = gain · ζημία = loss · ζημιόομαι = suffer loss · σκύβαλα = refuse. The whole of 1300–1304 is a dispute over these four words (heretics: "the law is loss and refuse"), so one English word each, never varied.
• σπένδομαι = "I am poured out as a libation" · σπονδή = libation · θυσία = sacrifice · λειτουργία = ministration · λειτουργός = minister. λειτουργία is fixed across both its senses in this range (2:17 the faith offered; 2:30 the Philippians' service to Paul) because Oecumenius plays them against each other.
• διώκω = "pursue" · καταλαμβάνω = "lay hold" · βραβεῖον = prize · σκοπός = mark (Phil 3:12–14). AV's "follow after / apprehend / press toward" was refused: the gloss at 1305–1308 turns on the runner image and on the pair διώκω / εἰ καταλάβω being visibly the same two words each time.
• ἰσόψυχος = "like-souled," ἰσοψυχία = "like-souledness" (1292). AV's "likeminded" hides the gloss's own derivation.
• εὔφημος / δύσφημος = "of good omen" / "of ill omen" (1301), where the argument is precisely that Paul chose an ill-omened word for a good thing.
• ἔνι = "it is possible" — continued from the anchor (0000, Ἔνι γὰρ), recurring at 1293 and 1300.
• Ἢ opening an alternative construal = "Or:"; τουτέστι/ἤγουν = "that is" — anchor formulas, unchanged.
• Em-dash separators inside a scholion. Where the plate opens a fresh alternative construal inside one father's scholion it prints a bare em-dash with no siglum (— Καὶ ἄλλως δέ, — Δυνατὸν δὲ καὶ οὕτως, — Ἢ καὶ οὕτως; scan-verified leaf 659, and the Latin column prints the same dashes: — Praeterea et alio modo, — Potest autem et hoc modo dici). These are reproduced as em-dashes and are not sigla; do not bracket them.

Catena sigla met in this batch

[lat: …] markers fired (4), with plate status

1. 1298 (chunk 0008 opening) — Greek φοβούμενος μὴ ἀναισχυντοτέρους ποιήσῃ ("fearing lest he make them more shameless"); Latin veritus ne nimium pudore ipsos afficeret ("lest he affect them with too much shame"). Opposite motives for the same silence. Greek plate-verified, leaf 661: ἀνωνύμως αὐτῶν κχάθάπτεται, φοθούµενος μὴ ἀνα.σγυντοτέρόνς ποιήσῃ. Marker fired.
2. 1297 (chunk 0007) — Greek ἐπειδὴ οὐκ ἔτι νομίμως ἐγίνετο παρὰ Ἰουδαίοις ἡ περιτομή ("since circumcision was no longer performed lawfully"); Latin quoniam nondum facta erat apud Judaeos vera et legitima circumcisio ("not yet … a true and lawful circumcision"). A temporal polarity flip, and it reverses the argument's direction (types ceasing at the truth's arrival vs. a lawful circumcision never yet reached). Greek plate-verified, leaf 661: Αλλ᾽ ἐπειδὴ οὐκ ἔτι νομίμως ἐγίνετο παρὰ Ἰουδαίοις ὁ περιτομή. Marker fired.
3. 1304/1305 (chunk 0009, block lemma Phil 3:12) — Greek Οὐχ ὅτι ἤδη ἔλαβον ἢ ἤδη τελείωμαι; Latin Non quod jam jam apprehenderim aut jam mortuus sim ("or have already died"). Both columns plate-verified, leaf 663 (Greek: O2, ὅτι Άδη ἔλαθον 7, Ίδη τελείωμαι; Latin: jam jam apprehende- rim au(. jam mortuus sim). Marker fired. See also the scripture list below: the plate prints τελείωμαι, not the received τετελείωμαι, and the gloss's own paraphrase (ἢ ἤδη ἐκ τούτου τέλειός εἰμι) reads with "perfect," not with "dead" — so the Greek column is coherent and the Latin is the outlier.
4. 1308 (chunk 0009) — Greek Τὸν σύντομον δηλοῖ δρόμον τὸ, ἐπεκτεινόμενος ("the swift course"); Latin Continuum cursum significat dictio Extendor. Greek plate-verified, leaf 665:* σύντομον ὀχλοῖ Ἀρόμον τὸ, ἐπεκτεινόμενος. Marker fired — the whole point of the gloss is what ἐπεκτεινόμενος shows*, and speed and continuity are not the same showing.

Candidates NOT marked (evidence short of the bar) — logged deliberately

• 1292 — the pronoun-itacism class, the anchor's own warned case. Greek οὔτως ὑμῶν ἀντιλαμβανόμενον ἢ μεριμνῶντα τὰ καθ᾽ ἡμᾶς; Latin qui ita vos adjuvet et curet quae vestra sunt. Identical to the anchor's 1261/1262 site in kind — but col 1292 is in the leaf gap and the Greek could not be plate-verified, so per the standing rule the candidate is logged and no marker fired. Rendered from the printed Greek ("the things that concern us").
• 1300 — Ἑβραῖος / Ἰουδαῖος. Greek ἔνι γὰρ εἶναι ἐκ γένους Ἰσραὴλ, καὶ μὴ Ἑβραῖον; Latin Contingit enim ex genere Israel esse et non esse Judaeum. The gloss exists precisely to distinguish Hebrew (= knowing the Hebrew tongue) from Israelite, so the Latin's Judaeum flattens it. Not marked: unverifiable leaf gap, and the immediately preceding word is Calfa damage (εἰμὶ,νο γὰρ for εἰμί, ἔνι γὰρ), so the region is not clean enough to carry a divergence claim.
• 1300 — κατὰ Χριστόν / secundum Deum. Greek (as restored) πρὸς σύγκρισιν τελειοτέρας τῆς κατὰ Χριστὸν γνώσεως; Latin ad comparationem perfectioris secundum Deum cognitionis. Same leaf gap; the Greek noun itself had to be restored from the Latin (below), so no marker.
• 1308 — ἐν Χριστῷ / per Christum. Greek ἐν Χριστῷ γὰρ, φησί, τρέχω (plate-verified, leaf 665); Latin per Christum, ait, curro, and the lemma heading likewise Per Christum Jesum. Not marked: Allatius renders ἐν instrumentally as a matter of course, and this is his looseness, not a different fact. Logged because the gloss does lean on the preposition.
• 1290 — construal prepositions in Photius's alternative. Greek Ἀλλ᾽ εἰ καὶ θύομαι διὰ τὴν θυσίαν καὶ λειτουργίαν (plate-verified, leaf 659); Latin Quinetiam si immolor super hostia sacrificioque fidei vestrae. The Latin has silently pulled the paraphrase back to the Vulgate's own preposition (immolor supra sacrificium), which is the anchor's Vulgate-conformation class — never a [lat:]. Logged as notable because the conformation erases the very construal Photius is offering.
• 1294 — τοῦ Κυρίου / Domini Jesu; 1302 — Deut 18:15 προφήτην ὑμῖν ἀναστήσει / Prophetam suscitabit vobis Deus.** Both are the Latin supplying a name the Greek leaves implicit, the second under Vulgate pressure. Conformation/expansion class; no markers.

[ed: …] markers (4) — demonstrated holes in our source

1. 1293/1296 (chunk 0007) — Calfa breaks off mid-clause: ἵνα μὴ λύπην τὴν τῆς ἀῤῥωστίας αὐτοῦ, ἐπὶ λύπην τὴν τοῦ + [next lemma]. The plate carries the continuation — scan leaf 660: ἐπὶ λύπη, τὴν τοῦ θανάτου χκρατύσυ (θανάτου certain; the verb's letters are not recoverable from this OCR, and the Latin twin's ne dolorem qui fuerat ex ejus aegritudine, haberem una cum dolore qui sequeretur ex ejus morte fixes the sense but not the word). Per Pattern 13 the lost words are not supplied on the page.
2. 1298 (chunk 0007) — Calfa: Διὸ οὐ ψέγω αὐτοὺς ὡς ἀπεστερημένος τῆς + [next lemma]. Plate carries it, scan leaf 661: Διὸ οὗ ψέγω αὐτοὺς ώς ἀπεστερτμένος τῆς κατὰ σαρκα εὐγενείας. Not supplied on the page.
3. 1300 (chunk 0008) — a whole step of the objection is missing between διὰ τὸ ὑπερβάλλον τῆς γνώσεως and οὐκ ἔστι ζημίαν ἀποβαλεῖν. The Latin column carries it: propter excellentiam cognitionis Christi: quomodo statim dicis, Propter quem omnium jacturam feci. Nam dicere Jacturam feci, non est damnum abjicere. Col 1300 is in the leaf gap, so the loss is demonstrated by the parallel column rather than by the Greek plate; the words are not supplied.
4. 1296 ×2 — the two illegible sigla, above.

Ladder row 1 — Calfa OCR damage corrected silently, witness cited

Chunk 0006 (1288–1293). Ὡρᾷς → Ὁρᾷς · ἀμοιβὰε → ἀμοιβὰς · Είς → Εἰς · ὀμῖν → ὑμῖν · ρὐκ → οὐκ · ἀποθιήσκω → ἀποθνήσκω · μάλλω → μέλλω (Lat. oblaturus) · Χριστῷτὸ → Χριστῷ· τὸ (joined type) · Κφησίν, Κτῇ, λειτουρΚγίαν → intrusive-Κ class, deleted (the anchor's PG 118 instance of the PG 139 artifact) · ἈΙΓ εἰ καὶ σαένδομαι → Ἀλλ᾽ εἰ καὶ σπένδομαι (block lemma; scan leaf 659) · οΑλλ´ → «Ἀλλ᾽ (guillemet, scan leaf 659) · αὐδὲ τοῦτο ἄτοπον → οὐδὲ τοῦτο ἄτοπον — negation-critical, scan leaf 659 concurs (Kal φαμιν, ὅτι οὐδὲ τοῦτο ἄτοπον) · ὡ- ἀληθῶς → ὡς ἀληθῶς · Δυνατὸν δὲ καὶ ρῶτως → Δυνατὸν δὲ καὶ οὕτως (leaf 659) · θλίψεσιδεσμὰ → θλίψεσι (δεσμὰ (leaf 659 prints the parenthesis) · οὐχ´ ὁμᾶς → οὐχ ὑμᾶς (leaf 659) · Ἐπὶ τῇ θυσίᾳ.» → «Ἐπὶ τῇ θυσίᾳ.» (lost opening guillemet, leaf 659) · ὐτούς → αὐτούς · Τομόθεον → Τιμόθεον · Ἰνα/να → ἵνα · γὸρ → γὰρ · ητοῦσιν → ζητοῦσιν · δτι → ὅτι · σὺν ἐμοῖ ἑδρύλευσεν → σὺν ἐμοὶ ἐδούλευσεν · ἐξ » αὐτῆς → ἐξ αὐτῆς (the stray » is Calfa noise; block lemmata are unguilleted) · ΚΚαὶ → Καὶ · ζηΚτοῦσι → ζητοῦσι · ἄχρις θμῶν → ἄχρις ὑμῶν (Lat. usque ad vos) · φησὶί → φησί · εἰκότες → εἰκότως (Lat. Merito) · ἵυα → ἵνα · ὡφεληθῶσιν → ὠφεληθῶσιν · Τνα → ἵνα · πέμύαι → πέμψαι · ᾤν → ἦν · νσθένησε → ἠσθένησε · ἀλα → ἀλλ᾽ · λόπην ἐπὶ λύπηνσχῶ → λύπην ἐπὶ λύπην σχῶ · Ἔνι Κγὰρ → Ἔνι γὰρ · Δι´ Καὐτοῦ → Δι᾽ αὐτοῦ (leaf 660) · Λιὸ → Διὸ (leaf 660) · Ἡδημόνει → Ἠδημόνει · διὰ σὸν Κύριον → διὰ τὸν Κύριον (leaf 660) · δι´ ὁμᾶς → δι᾽ ὑμᾶς (leaf 660).

Chunk 0007 (1293–1298). ὅλωςτοῦτο → ὅλως (τοῦτο and ἀναλυσαι → ἀναλῦσαι, καλὸν) — parenthesis restored, leaf 660 · δι´ ἡν → δι᾽ ἣν · ἱνα → Ἵνα · Καὐτοῦ → αὐτοῦ · ἔπεχμψα → ἔπεμψα (leaf 660) · Χαρᾶς → χαρᾶς · ἧγισε → ἤγγισε (leaf 660) · Κοὐκ → οὐκ · ἀλιπότερος → ἀλυπότερος · τῆς ὡλτῶν. μαθητῶν → τῆς ὑπὲρ τῶν μαθητῶν (leaf 660 gives τῆς ὑπὲ2 τῶν μαθιτῶν) · πυλλῷ → πολλῷ (leaf 660) · τοισῦτον → τοιοῦτον · Λιὰ → Διὰ · ἱμαθητάς → μαθητάς · μαθηταις → μαθηταῖς · Ἀσι ἐντολῆς ἐστι Χριστοῦ → read as ὅτι ἐντολῆς ἐστι Χριστοῦ (Lat. quia praecepti Christi est) — the letters Ἀσι are not a word; rendered by the Latin-attested sense, flagged here because the conjunction is a guess in form though not in sense · κἱνδυνον → κίνδυνον · ὑστέρισε → ὑστέρησε · δεικνὸς → δεικνὺς · Πράφειν → γράφειν · σὐκ → οὐκ · ὑμῶ δὲ ἀσφαλές → ὑμῖν δὲ ἀσφαλές · κακοὸς → κακοὺς · οἱπνεύματι → οἱ πνεύματι · πεποιθησιν → πεποίθησιν · Ἐὐαγγέλιον → Εὐαγγέλιον · ἀπ´ αὐτῶ → ἐπ᾽ αὐτῷ · καὶ ὰείαν Κυρίου → καὶ ἀξίαν Κυρίου (Lat. dignoque Domino; ladder row 1, sense-certain) · Ἢσαν → Ἦσαν · οθς → οὓς · ἀλέγοντο → ἐλέγοντο · κύνεςο γὰρ ΔΑΡΟΤ v. → κύνες (οἱ γὰρ (bleed excised, parenthesis restored; leaf 661 gives the parenthesis) · ἐξέρασμα kept (attested) · ἡ περιτομή γὰρ νομίμη → ἡ περιτομή (ἡ γὰρ νομίμη (leaf 661) · σαρκίς → σαρκί; · αὕτη·Τοῦτο → αὕτη· (Τοῦτο (leaf 661) · Ἰοραὴλ → Ἰσραήλ · Εενιαμίν → Βενιαμίν · Ἐβραῖος ἐξ Ἐβραίων → Ἑβραῖος ἐξ Ἑβραίων · Κἄλεμπτος → ἄμεμπτος (leaf 661) · ὄσην → ὅσην.

Chunk 0008 (1297–1304). Ἄλνως → Ἄλλως (leaf 661) · Κάλιστα → μάλιστα · προσηλιῆτων → προσηλύτων · εἰμὶ,νο γὰρ → εἰμί, ἔνι γὰρ (Lat. Contingit enim) · ρὐκ ἡκριβῶντο → οὐκ ἠκριβοῦντο · Ἐκκληαίαν → Ἐκκλησίαν · ἔμελλὶ → ἔμελλε · ζημίάν → ζημίαν · τῆς γνώσεως ο ῶ Ἰησοῦ → τῆς γνώσεως Χριστοῦ Ἰησοῦ (Lat. cognitionis Christi Jesu) · ἐημιώθην → ἐζημιώθην · ἐν ὐτῳ → ἐν αὐτῷ · Κἐμὴν → ἐμὴν · Χριστου → Χριστοῦ · τὰ ἁπὸ τοῦνόμου → τὰ ἀπὸ τοῦ νόμου · Μαρκωνος → Μαρκίωνος (Lat. Marcionistae) · ὁὅπερ → ὅπερ · ⚠ πραγμάτων λυσιτελῶν — EMENDATION REVERSED 2026-08-01 by the blind polarity read. The original entry is preserved below because the error is the lesson.

~~πραγμάτων λυσιτελῶν → πραγμάτων ἀλυσιτελῶν — ladder row 1, and the one silent correction in this batch that changes the sense: as printed, the sentence says casting away profitable things, which contradicts its own neighbour (τοῦτο γὰρ μᾶλλον κερδῆσαί ἐστι, τὸ ἐλευθερωθῆναι ζημίας); the Latin has quarumdam vilium rerum abjectionem, and a dropped initial alpha is the commonest Calfa loss.~~

The neighbour argues the OTHER WAY, and settles it for λυσιτελῶν. The sentence is οὐκ ἔστι ζημίαν ἀποβαλεῖν … ἀλλὰ μᾶλλον πραγμάτων λυσιτελῶν … ἀποβολήν — "it is NOT to cast off a loss … BUT rather the casting away of profitable things." The parenthesis exists precisely to rule the ἀ- reading out: freeing oneself from loss is gain, so ζημιοῦσθαι cannot mean that; it must mean losing something good. On the emended reading the ἀλλά contrasts a thing with itself and the argument collapses. Two further neighbours confirm it: the aporia that follows (Πῶς οὖν ἃ ἤγησαι ζημίαν … λέγεις, Ὅτι ἐζημιώθην;) has no force unless the word implies losing profitable things, and the resolution states it outright — ἐπεὶ καθ᾽ αὑτὸ, κέρδος αὐτὰ ἐνόμιζεν, "since in itself he reckoned them gain."

Corrected in the English: "certain profitable things", with a [lat:] recording Allatius's vilium. Col 1300 sits in the leaf gap, so the Greek is still not letter-verifiable — but the author's own argument is stronger evidence here than a plate would be, because it is what the emendation had to override.

⚑ This is Pattern 7a′ in the wild, and the sharpest example the corpus has. The rule says a crux that describes a repair does not make it honest; a logged column is not a checked column. Here the column was logged, the emendation was argued in the file, the Latin appeared to corroborate — and the English shipped a word rendered as its exact opposite. Nothing downstream could catch it: verify-english passes, the citation is right, the sentence reads smoothly. A crux whose reasoning is stated deserves MORE scrutiny than a bare one, not less — a stated reason is exactly what stops the next reader from checking · δημίαν → ζημίαν · Ὃτι ἐξημιώθην → Ὅτι ἐζημιώθην · τῆς κατὰ ΚΧριστὸνέλοα- → τῆς κατὰ Χριστὸν γνώσεως (noun supplied from Lat. cognitionis; see the unmarked-candidate list for secundum Deum) · πρὸς σύν κρισιν → πρὸς σύγκρισιν · Ὥγημαι → Ἥγημαι · μολίβδον → μολίβδου · ὁ στος → ὁ σῖτος · ἐνηνθρωπήκει kept · Ἱδει → Ἤδει · Ἢγημαι → Ἥγημαι · Κζημία → ζημία (intrusive Κ) · ἀποκαλόνει (see 0009).

Chunk 0009 (1304–1308). ἐακὼς → ἐάσας · δἷον ἐξέργων → οἷον ἐξ ἔργων · κατὰνόμον → κατὰ νόμον · Τ γὰρ καὶ ποιήσει → Τί γὰρ καὶ ποιήσει · Οὐχ ἄτι ἥδη → Οὐχ ὅτι ἤδη (leaf 663) · ὑκὸ → ὑπὸ · τὰ μν ὀπίσω → τὰ μὲν ὀπίσω · ἐπιλανβανόμενος → ἐπιλανθανόμενος · ἐπεκτεινόμανος → ἐπεκτεινόμενος · κΙήσεως → κλήσεως · xριστοῦ, Σριστοῦ → Χριστοῦ (Latin-alphabet substitution class) · Ἄτις → Ἥτις · ὑπλρ ἀΟρώπων → ὑπὲρ ἀνθρώπων (leaf 663/664) · ἀνΚδρώπων → ἀνθρώπων · ἤόη → ἤδη · ὄραβεῖον → βραβεῖον · κααλάβω → καταλάβω · ἀρὰ → ἄρα · ἔμΚπροσθεν → ἔμπροσθεν · ἐμαυΚτόν → ἐμαυτόν · δροΚμέως → δρομέως · Ἃμα → Ἅμα · ἐπκτείνεσθαι → ἐπεκτείνεσθαι · πολύ [μου ἀπέχον → πολύ μου ἀπέχον (stray bracket) · σοι οὖν τέλειον → Ὅσοι οὖν τέλειοι (leaf 665) · εἶ τι έτέρως → εἴ τι ἑτέρως · ἀποκαλόνει → ἀποκαλύψει (leaf 665) · ἃποκαλύθει → ἀποκαλύψει · οὐ´ καλῶς → οὐ καλῶς (negation intact in both witnesses) · «Πλὴν εἰς ἃ ἐφθάσαμεν.» → «Πλὴν εἰς ὃ ἐφθάσαμεν.» (leaf 665 prints the singular in the gloss lemma as in the block lemma — Calfa's plural would have manufactured an internal lemma divergence that the plate does not have) · ὸπίσω → ὀπίσω · ἑπεκτείνεσθαι → ἐπεκτείνεσθαι · τουνέστι/τουτέστι normalized.

Restored question marks (Calfa prints · or nothing for the plate's ;)

Logged as a class, per the anchor. 0006: Ὁρᾷς… τὰς ἀμοιβὰς τῆς ἀρετῆς; · Ποῖον καύχημα ἔχεις, ὦ μακαρία ψυχή; · πῶς οὖν ἀπέστειλεν ἂν αὐτόν; (leaf 660 prints ;) · οἱ λέγοντες κακὸν τὸν κόσμον; (leaf 660). 0007: ἔλεον λέγει τὴν ἐνταῦθα ζωήν; (leaf 660) · πόσῳ μᾶλλον ὑπὲρ διδασκάλων — rendered with ! after the Latin's own quanto magis pro praeceptoribus ! · Χριστῷ πεποιθέναι, ἢ σαρκί; (leaf 661). 0008: πῶς ἂν διὰ φιλαρχίαν ἔμελλε διώκειν, καὶ οὐ κατὰ ζῆλον Θεοῦ; · Πῶς οὖν φατε ὑπὸ Θεοῦ αὐτὸν δεδόσθαι; (leaf 662) · Διὰ τί δὲ ὅλως ἥγησαι αὐτὸν ζημίαν καὶ σκύβαλα, ὦ Παῦλε; (leaf 663). 0009: ποῖος παραστήσει λόγος; (leaf 663) · Τί δὲ οὐ λογίζῃ κατειληφέναι; · τί ἡμεῖς ἐροῦμεν;

Cross-column bleed (Calfa artifact class — Latin misread in Greek glyphs)

Not translated, not carried; the Greek column's own text runs on through the intrusion, and the column anchors inside it are kept in place.

• 1292 (0006) — ΡΑτμο. ΟΚ. ΟΧνΙii. [1292] οοΟμαιόειο Πλλί. ΟΑΡΒΤ Γν. intruding into Ἐπὶ μὲν τῇ … πίστει ὑμῶν χαίρετε. Leaf 659 shows the source: the page-foot signature PATROL. GR. CXVIII. plus the facing running title OECUMENII PHIL. CAPUT IV. The [1292] anchor is real and stays.
• 1296 (0007) — ἀμ κοοῶνΙἱαιο εμωι, μὲ λίε πίιεαΒι ἱ εη ρετῖσυΙίε. Κοο ε αἰε ἰρεοε Ἰη οΠοίο οοο. Δο οσἰεεπε ὡἱ ο ουωεἰ οοΠρὶοαιΓ. — the tail of the Latin column (its own …ostendens ut quod debitum est compleatur) bled into the Greek. Col 1296 is in the leaf gap; identified by shape and by the Latin twin, not by the plate.
• 1297 (0007) — ΔΑΡΟΤ v. inside Νῦν δὲ οὗτοι κύνες (οἱ γὰρ.
• 1302 (0008) — ομε ΠΡῖίε ενι. between ἀλλὰ τῇ τοῦ βελτίονος ἀποστερήσει. and Τοίνυν καὶ ὁ νόμος — leaf 662 shows Migne's footnote line (I Cor. I, 22, 29. Rom. v, 14) and the Latin column at exactly this point.

Guillemets

Migne's « » reproduced 1:1 except where Calfa lost an opening mark, which is restored (Pattern 6: never omitted where he has one). Counts, English vs Greek chunk: 0007 40/40, 0008 22/22, 0009 55/55, 0006 36/35 — the 0006 excess is three source defects netted: two lost openings restored (οΑλλ᾽ εἰ καὶ σπένδομαι,» and Ἐπὶ τῇ θυσίᾳ.», both scan-verified at leaf 659) and one stray » not reproduced (ἐξ » αὐτῆς, inside an unguilleted block lemma — Calfa noise). One genuinely unmatched » is reproduced, at 1305: «Συμμορφούμενος.» Τουτέστιν ἐξομοιούμενος.» — both Greek witnesses print it (Calfa and leaf 663: Τουτέστιν ἐξ- ομοιούµενος. »), so it is the plate's own defect and stands.

7a″ — where Migne's printed Greek diverges from the received text

This is the batch's primary deliverable. Every item below was rendered as Migne prints it, and every re-quotation in the glosses was checked against the block lemma word by word.

1. Phil 2:16, gloss re-quote at 1288 — block: οὐκ εἰς κενὸν ἔδραμον οὐδὲ εἰς κενὸν ἐκοπίασα; Migne's gloss re-quote drops the second εἰς κενόν: Ὅτι οὐκ εἰς κενὸν, φησίν, ἔδραμον, οὐδὲ ἐκοπίασα. Rendered "did not run in vain, he says, nor labor" — the second "in vain" is not supplied.
2. Phil 2:22 at 1292 — Migne prints Τὴν δοκιμὴν αὐτοῦ γινώσκετε with no δέ; the received text has τὴν δὲ δοκιμήν, and the Latin column restores the connective (Porro probationem ejus nostis). Rendered without it.
3. Phil 2:26–27 at 1292 — the batch's most consequential lemma defect. Migne's block lemma as our source prints it runs …διότι ἠκούσατε ὅτι ἠσθένησε παραπλήσιον θανάτῳ, ἀλλ᾽ ὁ Θεὸς αὐτὸν ἠλέησεν, i.e. the clause καὶ γὰρ ἠσθένησεν is absent, its neighbour absorbed into the ὅτι-clause. But Oecumenius's own gloss below quotes «Καὶ γὰρ ἠσθένησε.» as a lemma, and the Latin column prints Siquidem ita infirmus fuit, ut vicinus esset morti — so both the commentary and the facing column witness a clause the Greek block lacks. Rendered exactly as printed (Pattern 7); not repaired, and deliberately not given an [ed:], because col 1292 is in the leaf gap and I cannot demonstrate what the Greek column of the plate prints. This is the single item in the batch most worth a plate check. Also at 2:27: Migne ἀλλ᾽ ὁ Θεὸς αὐτὸν ἠλέησεν vs received ἠλέησεν αὐτόν (order only).
4. Phil 2:30 at 1294–1296 — block and gloss disagree inside Migne himself. Block: ἵνα πληρώσῃ τὸ ὑμῶν ὑστέρημα; the gloss lemma four sentences later: «Ἵνα ἀναπληρώσῃ τὸ ὑμῶν ὑστέρημα.» Rendered distinctly ("fill" / "fill up"), the difference preserved rather than harmonized.
5. Phil 2:30 at 1294 — Migne prints παραβουλευσάμενος τῇ ψυχῇ (and the gloss's verb παρεβουλεύσατο agrees) where the received text has παραβολευσάμενος. Rendered "hazarding his life" / "he took the hazard", keeping the two on one root as Migne's Greek does.
6. Phil 3:3 at 1296 — Migne prints οἱ πνεύματι Θεῷ λατρεύοντες (dative), with the gloss Οἱ πνευματικῶς τῷ Θεῷ λατρεύοντες and the Latin qui spiritu Deum colimus concurring, against the critical text's θεοῦ. Rendered "who serve God in spirit."
7. Phil 3:4 at 1296 — the block lemma drops a καί its own gloss keeps. Block: καίπερ ἐγὼ ἔχων πεποίθησιν ἐν σαρκί; the gloss: Καίπερ ἔχων πεποίθησιν καὶ ἐν σαρκί, with a short lemma «Καὶ ἐν σαρκί.» Both rendered as printed, so the English shows the same seam.
8. Phil 3:5 at 1298 — Migne prints κατὰ ζῆλον (accusative) for the received κατὰ ζῆλος.
9. Phil 3:8 at 1300 — Migne prints Ἀλλὰ μὲν οὖν καὶ ἡγοῦμαι for the received ἀλλὰ μενοῦνγε, and ἡγοῦμαι σκύβαλα εἶναι (the received text has no εἶναι) — the added infinitive is then re-quoted in the gloss (Ταῦτα καὶ περὶ τοῦ, Ἡγοῦμαι σκύβαλα εἶναι), so it is the commentary's own text. Rendered "count them to be refuse" in both places.
10. Phil 3:11 at 1305 — Migne prints εἰς τὴν ἐξανάστασιν τῶν νεκρῶν for the received τὴν ἐξανάστασιν τὴν ἐκ νεκρῶν. (English cannot show the ἐξ- prefix; "the resurrection of the dead" carries both, as the Latin's resurrectionem mortuorum does.)
11. Phil 3:12 at 1305 — plate-verified. Migne prints ἢ ἤδη τελείωμαι, not the received τετελείωμαι: leaf 663 concurs with Calfa, so this is the plate, not our file. Rendered "or am already perfected"; see [lat:] no. 3 for the Latin's aut jam mortuus sim.
12. Phil 3:13 at 1305 — Migne prints ἐγὼ ἐμαυτὸν οὔπω λογίζομαι κατειληφέναι (received οὐ λογίζομαι), and the gloss lemma «Οὔπω λογίζομαι.» depends on it. Rendered "I do not yet reckon."
13. Phil 3:16 at 1308 — Migne prints the long form τῷ αὐτῷ στοιχεῖν κανόνι, τὸ αὐτὸ φρονεῖν (leaf 665 concurs) and, at 3:15, the indicative τοῦτο φρονοῦμεν where the received text has the hortatory φρονῶμεν (Latin hoc sentiamus). Rendered "we are of this mind."
14. Quotations from elsewhere, as Migne prints them: 1 Cor 15:31 Καθ᾽ ἡμέραν ἀποθνήσκω (agrees) · Matt 10:40 Ὁ γὰρ δεχόμενος ὑμᾶς, ἐμὲ δέχεται (agrees) · 1 Thess 4:17 Καὶ πάντοτε σὺν Κυρίῳ ἐσόμεθα — Migne's Καί for the received καὶ οὕτως · 2 Tim 4:7 Τὸν ἀγῶνα ἠγώνισμαι, καὶ τὸν δρόμον τετέλεκα — without τὸν καλόν, and with a καί the received text lacks; rendered "I have fought the fight, and have finished the course," resisting AV's "I have fought a good fight" · 1 Cor 9:26 and 9:24 agree · Deut 18:15 προφήτην ὑμῖν ἀναστήσει ὡς ἐμέ — heavily abridged against the LXX, and with no expressed subject (the Latin supplies Deus); rendered as printed · John 14:9 is quoted in paraphrase, not verbatim (Ὁ γὰρ τὸν Υἱὸν γινώσκων, γινώσκει τὸν Πατέρα, καὶ ἔμπαλιν), and stands as printed inside Migne's guillemets.
15. Phil 1:24, re-quoted twice inside this batch in two different forms — at 1290 τὸ ἐπιμένειν αὐτὸν τῇ σαρκί, at 1293 Ἀναγκαιότερον δὲ τὸ ἐπιμεῖναι τῇ σαρκὶ δι᾽ ὑμᾶς (aorist, and no ἐν, against the received τὸ δὲ ἐπιμένειν ἐν τῇ σαρκί). English renders both "to remain in the flesh"; the Greek difference is recorded here because the English cannot show it.

The Christ-hymn (Phil 2:5–11) does NOT fall in this batch. Chunk 0006 opens at Phil 2:16; ἐν μορφῇ Θεοῦ / ἐκένωσεν / μορφὴν δούλου lie in the peer range (the scan shows μορφὴν Θεοῦ on leaf 656, i.e. cols 1281–1282, chunks 0004–0005). No μορφή / ἁρπαγμός / κένωσις term table is owed by this batch. The one adjacent term this batch does fix is συμμορφούμενος τῷ θανάτῳ = "being conformed to his death" (Phil 3:10, glossed ἐξομοιούμενος = "being made like") — the peer holding 2:5–11 should be told that conformed is taken here, so that μορφή's English is chosen with that in view.

Latin-twin pass (pilot §6) — per-chunk result

Read against src/pg-latin/oecumenius-philippians/0006–0009.md (Latin cols 1290–1307).

• 0006 — findings, not clean. Two Vulgate/lemma conformations logged (Photius's super hostia; Porro probationem), one unverifiable pronoun-itacism candidate (ἡμᾶς / vestra, 1292), the Phil 2:26–27 clause gap, one Latin name-expansion.
• 0007 — findings, not clean. One [lat:] fired (οὐκ ἔτι / nondum, plate-verified), two [ed:] for lost sigla, one [ed:] for a lost clause the plate carries, one silent correction made on the Latin's evidence (ἀξίαν Κυρίου).
• 0008 — findings, not clean. One [lat:] fired (ἀναισχυντοτέρους / nimium pudore, plate-verified), one sense-changing silent correction on the Latin's evidence (ἀλυσιτελῶν), one [ed:] for a dropped step of the objection, two unverifiable divergence candidates logged unmarked.
• 0009 — findings, not clean. Two [lat:] fired, both with the Greek plate-verified (mortuus sim; continuum cursum), one unmarked candidate (ἐν / per Christum), and one two-witness concurrence rendered literally (below).

No chunk in this batch came back clean. Stating that plainly rather than naming a clean chunk that does not exist: 0006, 0007, 0008 and 0009 were each read against their twin after drafting, and each yielded something.

Ladder row 3 — Greek and Latin concur on an odd reading (rendered literally)

• 1305 ἐν ᾧ πράγματι ὑποκατελήφθην ὑπὸ Χριστοῦ — an unusual compound where the lemma and the next sentence both have simple κατελήφθην. Both Greek witnesses carry the ὑπο- (Calfa; leaf 663 ἐν ᾧ πράγματι ὑποκατελύσθην ὑπὸ Χριστοῦ), and the Latin agrees in sense (in ea re qua apprehensus sum a Christo). Rendered "in that matter for which I was taken hold of by Christ" — attributable to the plate pair, not to our file.
• 1300 Ἄλλως γὰρ κατῃσχύνοντο (leaf 661 concurs) where the Latin has Maxime enim pudore suffundebantur; rendered from the Greek ("For otherwise they were put to shame"), the Latin noted as free rendering rather than a divergence.

Open for the merge / a later plate session

1. Cols 1291–1292, 1295–1296, 1299–1300 have no leaf in our PG 118 scan. Four of this batch's judgments hang on that gap: the two illegible sigla, the [ΤΟΥ ΑΥΤΟΥ] form, the Phil 2:26–27 clause, and the ἡμᾶς/ὑμᾶς candidate. The twin's provenance names the alt item (patrologicursus21migngoog, leaves 615/617/619) — acquiring it would settle all four.
2. [ΤΟΥ ΑΥΤΟΥ] is a new siglum for this work and should be added to the work-wide conventions list once verified; it means the preceding father ([ΦΩΤ.]) continues, so the merged file should keep the two adjacent.
3. [ΟΙΚΟΥΜΕΝΙΟΥ] (1303) is the author's own siglum — its presence tells the reader the surrounding matter is catena, not Oecumenius throughout. Worth a line in the work's "On this text" prose.

Range: cols 1309–1325 (Phil 3:16 – 4:23 + subscription). Anchor conventions (cruces-0000.md) followed throughout; only items decided fresh in this range are marked ⚑ below, for the merge session.

Scan witness used: raw/scans/pg118/patrologiaecurs14migngoog_djvu.xml, leaves 665, 666, 667, 668, 669, 670, 671, 672 (= Calfa pp. 666–668, 670, 672–674). Latin twins: src/pg-latin/oecumenius-philippians/0010–0013.md.

0. TWO SCAN GAPS IN THIS RANGE — read before trusting anything below

1. Calfa p. 669 (Greek cols 1315–1316 / Latin 1316–1317) — no leaf in either scanned copy; documented in the twin manifest as gapPages: [669], and the twin file 0011 prints [gap: no scan leaf for Calfa p.669] between [1315] and [1318]. For everything in chunk 0011 between "Great are the praises" and "…overshadows [1317]" there is NO verifier of any kind — no Latin column, no plate. Per the attribution ladder's bottom row this is no attribution: the Greek is rendered exactly as our source prints it, no reading is settled, and no [lat:] or [var:] was fired there, however strong the candidate (see §4b — two are strong). The twin's low word ratio (0.469) is a fact about the SCAN, not evidence about the Greek.
2. ⚑ Calfa p. 671 (Greek cols 1319–1320) — a SECOND, unflagged gap. The primary scan has no leaf for it (leaf 669 = Calfa p.670, leaf 670 = Calfa p.672). The twin covers it only from an alternate scan item (patrologicursus21migngoog, leaf 627 — recorded in the chunk-12 twin's provenance.altSourcePages, and flagged there as "DIFFERENT witness than the rest of the twin"). So for the first ~third of chunk 0012 (from "But the words are those of one urging" to "…but to the preaching [1321]") the Latin exists but the GREEK plate does not. Latin-settled corrections there are ladder row 1 (damage in our file) and are used; claims about Migne's printed Greek ([var:]) are NOT made there. This gap is not named in the brief and should be added to the twin manifest as a Greek-side note.

Verifier-pass result by chunk (pilot §6 requires naming these):

• 0010 — read against twin 0010, FINDINGS (one [var:], one polarity site the Latin settled, one Latin-side OCR note). Not clean.
• 0011 — read against twin 0011 + the tail of twin 0010; PARTLY UNVERIFIABLE. Findings at both ends (one [var:], one negation site); the middle is the gap of §0.1. This is "twin absent", which is not the same as clean.
• 0012 — read against twin 0012, FINDINGS (two polarity sites examined, both resolved as Latin-column defects; one lemma restored from the Latin). Not clean. Greek plate absent for its first third (§0.2).
• 0013 — read against twin 0013, FINDINGS (one [lat:]; two source-loss restorations). Not clean.
• No chunk in this batch came back clean. Zero chunks to report as "nothing found."

1. Catena sigla — three restorations, all ΦΩΤ., all scan-verified

No new siglum occurs in this range; only ΦΩΤ. The work's ΧΡΥΣ. does not reappear after the anchor chunk.

⚑ Siglum POSITION varies and is reproduced in place, not normalised. At 1308/1309 and 1321 the em-dash + siglum stand together before the scholion (the anchor's pattern). At 1309 the plate prints dash — lemma — siglum: ἀναστροφῆς. — «Τοὺς οὕτως περιπατοῦντας.» [ΦΩΤ.] Περιπατεῖν… (leaf 666 confirms this order). The English keeps that order rather than moving the siglum in front of the lemma.

2. Scripture-divergence list (7a″ deliverable) — the words, named

Migne's printed Greek against the received Greek text and against familiar English. Every item was rendered as Migne prints it. ⚠ marks the ones where the familiar English was the live pressure.

Fired as [var:] (plate-verified Greek):

1. ⚠ 1 Cor 15:25 quoted TWICE, in TWO DIFFERENT FORMS — 1313 vs 1316. At 1313 Migne prints «Δεῖ γὰρ αὐτὸν, φησί, βασιλεύειν ἄχρις ἂν ὑποταγῇ αὐτῷ τὰ πάντα» ("until all things be subjected to him") — a form that is not 1 Cor 15:25, being conflated toward 15:28/Phil 3:21. At 1316 the same verse is quoted in its received form, «…ἄχρις οὗ θῇ πάντας τοὺς ἐχθροὺς αὐτοῦ ὑπὸ τοὺς πόδας αὐτοῦ» ("till he put all his enemies under his feet"). Both Greek forms are on the plate (leaf 668) and the Latin column mirrors the doubling exactly — donec subjiciantur ipsi omnia at 1314, donec posuerit omnes inimicos suos sub pedes suos at 1315 (twin 0010). Two-witness concurrence: the doubling is the plate pair's, not ours. The two are therefore rendered DIFFERENTLY, and the familiar "till he hath put all enemies under his feet" is kept out of the first. [var:] fired at 1313 naming the received words.
2. ⚠ John 14:27 at 1317 — Migne prints «Εἰρήνην τὴν ἐμὴν δίδωμι ὑμῖν, εἰρήνην τὴν ἐμὴν ἀφίημι ὑμῖν» — the clauses in the reverse of the received order, and τὴν ἐμὴν with both verbs. Plate-verified, leaf 669. The Latin column reverses the clauses too (Pacem meam do vobis, pacem relinquo vobis) but keeps meam with do only, i.e. Vulgate wording in Vulgate-less order. [var:] fired on the Greek's divergence from the received text; no [lat:] — the Latin's meam placement is Vulgate-conformation (anchor convention).

Logged, NOT marked, because they fall in the p.669 gap (§0.1):

1. ⚠⚠ Luke 10:20 at 1316 — the strongest unmarked candidate in the work. Our source prints «Μὴ χαίρετε…ὅτι τὰ δαιμόνια ὑμῖν ὑποτάσσεται, ἀλλ᾽ ὅτι τὰ ὀνόματα ὑμῶν γέγραπται ἐν βίβλῳ ζωῆς», where the received text has τὰ πνεύματα and ἐγγέγραπται ἐν τοῖς οὐρανοῖς. The divergence is load-bearing: Oecumenius adduces the verse precisely to match Paul's ὧν τὰ ὀνόματα ἐν βίβλῳ ζωῆς, so "written in heaven" would make the argument pointless. Neither reading is producible by OCR — they are textual, not typographic — but with no plate and no Latin at this column I will not assert what Migne printed. Rendered as our source prints; no marker. This is the first thing a backfill session should adjudicate if a leaf for Calfa p.669 is ever found.
2. Luke 6:25 at 1316 — «Οὐαὶ τοῖς γελῶσι» (dative, and no νῦν) for the received οὐαί, οἱ γελῶντες νῦν. Rendered as printed; no marker (gap).
3. Phil 4:3 block lemma at 1315/1316 — Migne prints σύζυγε γνήσιε (the received order is γνήσιε σύζυγε) and συνεργῶν μοι (received μου). Both partly visible on leaf 668 before the column break; word-order item confirmed there, μοι/μου not. Rendered as printed; not material enough for a marker either way.

Logged, no marker, not material (divergence from AV/Vulgate cadence only, or sub-material):

1. ⚠ Phil 4:4–7 at 1316 — the most quoted lemma in the work, and it AGREES with the received text word for word. Χαίρετε ἐν Κυρίῳ πάντοτε, πάλιν ἐρῶ, χαίρετε · Τὸ ἐπιεικὲς ὑμῶν · Ὁ Κύριος ἐγγύς · ἡ εἰρήνη τοῦ Θεοῦ ἡ ὑπερέχουσα πάντα νοῦν φρουρήσει τὰς καρδίας ὑμῶν καὶ τὰ νοήματα ὑμῶν. The pressure here is not a variant but the AV's cadence, and it was resisted at three points: «again I will say» (not AV's "and again I say"); «forbearance» (not "moderation" — the gloss defines it against πολεμίοις, "not as enemies"); «shall guard your hearts and your thoughts» (not "keep your hearts and minds"). ὑπερέχουσα is rendered "surpasses all understanding" and ὑπερβαίνει in the two glosses is rendered "surpass" likewise, so the gloss's argument (if his peace surpasses every νοῦς, what of those who pry into his οὐσία?) stays visible.
2. ⚠ Phil 4:13 at 1320 and its gloss lemma at 1324 — Migne prints ἐν τῷ ἐνδυναμοῦντί με Χριστῷ (the Byzantine/TR reading; the critical text has no Χριστῷ), block lemma and gloss lemma identical. Rendered "I am strong for all things in Christ who empowers me," not AV's "I can do all things through Christ which strengtheneth me," because the gloss turns on the ἰσχύ- root twice — ἐμοῦ πάντα φέρειν ἰσχύοντος and τοῦ τὴν ἰσχύν μοι ταύτην παρέχοντος Χριστοῦ — which "I can do" erases. ⚑ Convention: ἰσχύω = "be strong for" · ἰσχύς = "strength" · ἐνδυναμόω = "empower", fixed for the work.
3. Phil 3:20–21 at 1309/1312 — agrees with the received text; ⚠ but Calfa prints τὸ σῶμα τῆς ταπεινώσεως ὑμῶν in the block lemma where the plate (leaf 667) prints ἡμῶν, the Latin prints corpus nostrum abjectum, and every one of the six re-quotations in the glosses prints ἡμῶν. See §3 — this is the itacism class resolving as damage in OUR file, not a divergence.
4. 1 Cor 15:32 at 1312 — «Φάγωμεν καὶ πίωμεν, αὔριον γὰρ ἀποθνήσκομεν», exactly the received text.
5. Matt 28:18 at 1313 — Migne prints ἐπὶ γῆς (anarthrous; received ἐπὶ τῆς γῆς). Not sense-changing.
6. 1 Cor 10:32 at 1320 — compressed to «Ἀπρόσκοποι γὰρ γίνεσθε», dropping καὶ Ἰουδαίοις…καὶ Ἕλλησιν. Rendered as printed; the plate itself sets it in parentheses.
7. Ps 119:7 (LXX) at 1320 — «Μετὰ τῶν μισούντων τὴν εἰρήνην ἤμην εἰρηνικός», exactly LXX.
8. 1 Cor 9:15 at 1320 (Greek plate absent, §0.2) — Migne prints ἢ τὸ καύχημά μου, ἵνα τις κενώσῃ, where the received has οὐδεὶς κενώσει. The Latin concurs with the ἵνα-clause (ut gloriationem meam aliquis inanem reddat). Rendered as printed; no [var:], because asserting what Migne's GREEK prints needs the Greek witness, which this column lacks.
9. 2 Cor 11:12 at 1320 (same gap) — «εὑρεθῶσιν ὡς κἀγώ» for the received καθὼς καὶ ἡμεῖς. Rendered as printed; no marker, same reason.
10. 1 Cor 9:11 at 1325 — «Εἰ ἡμεῖς ὑμῖν τὰ πνευματικὰ ἐσπείραμεν, μέγα εἰ ἡμεῖς ὑμῶν τὰ σαρκικὰ θερίσομεν;» — the received text; rendered as printed.
11. Phil 4:23 at 1325 — Migne prints the TR form μετὰ πάντων ὑμῶν. Ἀμήν (not μετὰ τοῦ πνεύματος ὑμῶν). ⚠ The block lemma prints Κυρίου ἡμῶν and the gloss lemma prints Κυρίου without ἡμῶν (plate-verified, leaf 672; the Latin has Domini nostri in both). The two are therefore rendered differently, NOT harmonized.
12. ⚠ Phil 4:19 — the block lemma at 1325 prints ἐν δόξῃ ἐν Χριστῷ Ἰησοῦ, but the gloss lemma prints «Σὺν Χριστῷ Ἰησοῦ». Plate-verified (leaf 672). Rendered differently ("in Christ Jesus" / "With Christ Jesus"); see the [lat:] at §4a, which is exactly this word.

3. Ladder row 1 — Calfa damage corrected silently, witness cited

Where a witness is named, that witness is the scan leaf and/or the Latin twin. Items in the p.669 gap (§0.1) are corrected only where the correction is a bare orthographic normalisation (accent, breathing, iota) with no sense at stake.

Chunk 0010 (cols 1309–1313) — witness: leaves 665–667, twin 0010.

• 1308/1309 Ἡ κανόνα τὴν πίστιν φησίν → Ἢ ("Or:", the work's fixed alternative-construal formula; Lat. Aut canonem seu regulam dicit fidem).
• 1309 ἔμπεοσθεν → ἔμπροσθεν · δόξη → δόξῃ · Δεῖ ὡμᾶς → ὑμᾶς (leaf 666) · τὸ αὐτὸ φρριεῆν Εἰ ὀφείλετε → φρονεῖν. Εἰ ὀφείλετε (leaf 666).
• 1309 σκοπεῖτε τοίς οὖτως → τοὺς οὕτως · τύπον ἡμᾶς ok · Δν τὸ τέλος → ὧν · οἱ τὰ ἐπίγεια φρονοῦν. πες → φρονοῦντες (split type).
• 1309 Κακῶς, φησὶν, ἤπερ ἐγώ — kept; leaf 666 concurs; Lat. Male, inquit, et aliter quam ego.
• 1312 αἱσχύνῃ → αἰσχύνῃ · αἱσχόνεσθαι → αἰσχύνεσθαι · Ἡσαν → Ἦσαν · Λεῖ οὖν → Δεῖ οὖν · πολιπευόμεθα → πολιτευόμεθα · ὑπαρχει → ὑπάρχει · ἐξ οὐ → ἐξ οὗ · ὀς → ὃς.
• 1312 ⚠ block lemma τὸ σῶμα τῆς ταπεινώσεως ὑμῶν → ἡμῶν. The pronoun-itacism class, resolved as damage in OUR file and NOT a [lat:], because the two independent witnesses of the Greek disagree with Calfa: leaf 667 prints "τῆς ταπεινώσεως ἡμῶν", the Latin prints corpus nostrum abjectum, and Calfa's own six re-quotations in the glosses all print ἡμῶν. (Contrast the anchor's 1261 case, where plate and Calfa agreed against the Latin and the marker fired.)
• 1312 παρουσία Κμετὰ → μετὰ (intrusive-Κ class) · φανησόμενον kept (leaf 667; Lat. apparebit).
• 1312 λέγομεν ἕτι ἔπεσε → ὅτι (leaf 667) · ταπεινοῦται…ἁπὸ → ἀπὸ · Ἰσγυρὰν → Ἰσχυρὰν · Βἰ γὰρ → Εἰ γὰρ · ἃν μόνον → ἂν μόνον.
• 1313 μεῖ γὰρ αὐτὸν, φησί, βασιλεύειν → Δεῖ γὰρ (leaf 668) · ὑποταγήσεται αὐτῶ → αὐτῷ (×3) · Ὑποτὰσσεται → Ὑποτάσσεται.

Chunk 0011 (cols 1316–1317) — witness: leaf 668 (opening), twin 0010 tail; leaf 669 + twin 0011 (from "…overshadows [1317]"). The middle has neither.

• 1313/1316 Ποῦτο γὰρ…ὁ Θεὸς καὶ iατὴρ → Τοῦτο…Πατὴρ (leaf 668) · Ῥρα δὲ πόσην → Ὅρα δὲ (leaf 668) · ἐὰν οὔτω → οὕτω.
• 1316 ΚΕΦΑΛ. Ϛʹ. → CHAPTER 6. (leaf 668 prints the stigma; Lat. CAPUT VI).
• 1316 Εὐοδίαν! παρακαλῶ → stray ! dropped · παρακαλῶτὸ αὐτὸ → παρακαλῶ τὸ αὐτὸ (joined type) · αἱτινες έν → αἵτινες ἐν.
• 1316 ⚑ μετὰ ΚλήμενΠοὶ ἰαινο. ΛΑΡΟΤ vί. [1316] τος — cross-column / running-foot bleed of the volume signature line (PATROL. GR. CXVIII, visible in clean form on leaf 670 as "PATROL. Gn. CXVIII"), fallen inside the word Κλήμεντος at the column break. The bleed is NOT rendered; the word is restored to Κλήμεντος and the [1316] anchor is placed at the nearest word boundary after "Clement". Second instance of the same bleed class in this batch — see 1324 below. (Anchor's rule extended: the Calfa bleed class in this work is both the Latin running TITLE and the volume running FOOT.)
• 1316 ΚΜεγάλα → Μεγάλα (intrusive Κ) · Ὦς νῦν → Ὡς νῦν · «Ἑρωτῶ καί σε.» → Ἐρωτῶ · «ΣύΕυγε γνήσιε.» → Σύζυγε · τινές ὑπενόησαν → τινὲς · ἐν φιλίπποις → Φιλίπποις · καὶ οὕτε ἐν Ταρσῷ → οὔτε · ἕνθα → ἔνθα · τῶν δύο ζν ἐμνήσθη → ὧν · ὦ καὶ παρατίθεται → ᾧ · ἀἰνίττεται → αἰνίττεται · Ἠσαν μὲν → Ἦσαν · τὰ Κὀνόματα → ὀνόματα (intrusive Κ).
• 1316 ⚑ ἢ ὄνομα Κύριον, εὐδόκιμος δέ τις → ὄνομα κύριον = "a proper name" (the grammatical term), the capital being a Calfa artifact. No witness at this column (gap); the reading is settled by sense alone — the alternative ("or [his name is] Lord") is not construable with εὐδόκιμος δέ τις, ᾧ παρατίθεται. Logged as a reading decided without a verifier.
• 1316 ⚑ guillemet pair repaired: Calfa prints συνυπέμειναν.» Καὶ τῶν λοιπῶν συνεργῶν μου.» — a closing mark with no opening, and a stray closing after συνυπέμειναν. Restored to συνυπέμειναν. «Καὶ τῶν λοιπῶν συνεργῶν μου.». Under Pattern 6 this is repairing a broken pair in OUR source, not supplying a mark the plate lacks (every short lemma in this work is guilleted, and one half of the pair survives). No plate at this column; flagged for the merge session.
• 1316/1317 Ὁ Κύριος ἐγγὸς → ἐγγύς · Εαὶ ἡ εἰρήνη → Καὶ · ΚΘρουρήσει → φρουρήσει (intrusive Κ + capital) · καρδίας ὑμων → ὑμῶν · Ἢσαν ἐν λύπῃ → Ἦσαν · ἂς ὑπέμενον Αείκνυσιν → ὑπέμενον. Δείκνυσιν · χαρὰς ὓν → χαρᾶς ὂν · Ἡ ὅτι ἡ ὑπὲρ Χριστοῦ λύπη → Ἢ ("Or:") · αὐτὸχάρα ἐστὶν → αὐτὸ χαρά ἐστιν.
• 1317 ἐπισκογιζούσης → ἐπισκιαζούσης (Lat. obscurante, twin 0011) · γνο σθήτω → γνωσθήτω · διέβαλέ τινας, ὡν → ὧν · τοὺς πιστοῦς → πιστούς · Ὡς αἵγε μὴ ἔχουσαι εὑχαριστίαν → εὐχαριστίαν · οὕτε γνωρίζονται → οὔτε (leaf 669) · Ὢστε μὴ ὑπὸ πειρασμῶν → Ὥστε · stray , before the closing guillemet of the John 14:27 quote, dropped.
• 1317 block lemma Phil 4:8 ὅσο σεμνὰ → ὅσα σεμνά · ὅσα ἀγνὰ → ἁγνά · εὕφημα → εὔφημα · Εἰ τις ἀρετὴ → Εἴ τις · Ἀ καὶ ἐμάθετε → Ἃ · ἡκούσατε → ἠκούσατε · ἷδετε ἐν ἐμοί → εἴδετε (leaf 669).

Chunk 0012 (cols 1320–1324) — witness: twin 0012 throughout; leaf 670 from "…to the preaching [1321]" onward; leaf 671 for the 1324 tail. Greek plate absent before [1321] (§0.2).

• 1320 Ἐπειγο μένου → Ἐπειγομένου (split) · Θεῦδος γὰρ ἡ κακία → Ψεῦδος (leaf 669; Lat. mendacium enim vitium est) · ἀρέσκεινἈπρόσκοποι → ἀρέσκειν. (Ἀπρόσκοποι (joined type; leaf 669 shows the parenthesis) · τὸ, Ὑετε ἐν ἐμοί → Εἴδετε (Lat. Vidistis in me) · οὐκ ἐνῆ → ἐνῆν · τῳ καὶ ἴδετε ἐν ἐμοί → τῷ, Καὶ εἴδετε ἐν ἐμοί.
• 1320 τοῦ τὰ γήῖνα φρονούντων — kept as printed (genitive singular article with genitive plural participle). Leaf 669 concurs, so this is the plate pair's own anomaly, not ours; English has no exponent for it, so it is rendered "him who minds earthly things" with no marker (Pattern 9).
• 1320 block lemma Phil 4:10–13: δτι ἡδη ποτὲ → ὅτι ἤδη ποτὲ · ἐφ’´ ὧ → ἐφ᾽ ᾧ · ἡκαιρεῖσθε → ἠκαιρεῖσθε · ἔμαθον οἰς εἰμι → ἐν οἷς εἰμι · αὐνάρκης εἰναι → αὐτάρκης εἶναι · Χρισιῷ → Χριστῷ · βλίθει → τῇ θλίψει · συγκοινωνήσαντές μου kept.
• 1320 ⚠ block lemma ἐν παντὶ καὶ ἐν πᾶσι μέμνημαι → μεμύημαι. The gloss at 1324 re-quotes the same clause and prints μεμύημαι (leaf 671 confirms the gloss); the Latin prints institutus sum in BOTH the lemma and the gloss (twin 0012). Ladder row 1: our file's damage, corrected so that lemma and gloss read identically, as 7a″ requires. Rendered "I have been initiated" in both.
• 1320 ὡφελήθησαν → ὠφελήθησαν · Τὸ, Ἥδη ποτὲ → Ἤδη · τῷ μὴ λαμβάνεινκαλὸν → λαμβάνειν (καλὸν (joined type) · ἡ τὸ καύχημά μου → ἢ · ἵνα τίς κενώσῃ → ἵνα τις κενώσῃ.
• 1320 ⚑ νῦν εἴληφες → νῦν εἴληφεν; — 2nd person for 3rd, and the question mark restored. Settled by the Latin: Quamobrem autem Paulus, qui semper gloriari solet quod non acciperet (…) nunc tamen accepit? (twin 0012). Third person is required by the whole sentence (Παῦλος … ἐγκαυχώμενος). Ladder row 1; no Greek plate at this column, so the Latin is doing the work and is quoted here per the contract.
• 1321 ἁλλὰ τῷ κηρύγματι → ἀλλὰ · ἐν ἐνδείᾳ ὢν ok · Κεἶπον, ἵνα μὴ ἄκαρποι → εἶπον (intrusive Κ) · μὴ νομίΚσητε → νομίσητε (intrusive Κ) · ἥν εἶχον → ἣν · ιὰ τοῦτο ἐχάρην → Διὰ τοῦτο.
• 1321 καὶ πολλοὺς ἔχων εἴπερ ἤθελον → …ἔχων, Καὶ εἴπερ ἤθελον per leaf 670; rendered "had I been willing" either way — no sense at stake.
• 1321 οὕτω τὰ πρὸς χρείαν ok · ἀπεδεξάμην ok · της χρείας μου → τῆς.
• 1324 Ὥσπεο γὰρ → Ὥσπερ · οὔτως καὶ ἡ ἄδειᾳ οἷδε → οὕτως…ἡ ἄδεια οἶδε · γεγένηνται ok · κεχορτασμένην ok · σως ἐρεῖ τις → Ἴσως.
• 1324 ⚑ ἐν ΡΑτὴοΙ. Ολ. ΣΧvΙΙi. [1324] πᾶσι τοῖς παρεμπίπτουσι — the same running-foot bleed as at 1316 (PATROL. GR. CXVIII), here fallen between ἐν and πᾶσι. Not rendered; the [1324] anchor kept in place.
• 1324 Τουτέσ-. Ι.ῖραν ἁπάντων ἔχω → Τουτέστι, πεῖραν ἁπάντων ἔχω (leaf 671; Lat. Institutus sum, id est, omnium habeo experientiam) · μεγαλοῤῥημο νεῖν → μεγαλορρημονεῖν (split) · πάντα τῳ λριστῳ → τῷ Χριστῷ · οὖτως ἐχόντων → οὕτως · ῥαθυμότεροι → ῥᾳθυμότεροι · μεταχειοίξει → μεταχειρίζει (leaf 671) · τὸ φρόνημα. αὐτῶν → stray point dropped · παύσαντές μουτὴν λύπην → μου τὴν λύπην (joined) · θλίψει.» Οἷον δέ ἐστι…γενέσθα· Παύλου → γενέσθαι Παύλου.

Chunk 0013 (col 1325) — witness: leaves 671–672, twin 0013.

• 1324/1325 block lemma Phil 4:15–19: Οἱδατε → Οἴδατε · Φλιππήσιοι → Φιλιππήσιοι · ἐν ἀρχῆ τοῦ ναγγελίου → ἐν ἀρχῇ τοῦ Εὐαγγελίου · καὶ ήψεως → λήψεως · Θοσσαλονίκῃ → Θεσσαλονίκῃ · ἐπέχψατε → ἐπέμψατε · ἑπιζητῶ τὺν καρπὸν τὸν πρλεονάζοντα → ἐπιζητῶ τὸν καρπὸν τὸν πλεονάζοντα · λόγον ὑμων → ὑμῶν · Πεπιλήρωμαι → Πεπλήρωμαι · Ἐπαφροδίπου → Ἐπαφροδίτου · έσμὴν εὐωδίας → ὀσμὴν · εἰὐάρεστον → εὐάρεστον · πλὴρώσει → πληρώσει · χρείαν ίμῶν κτὰ → ὑμῶν κατὰ · ἐν ΧριΟ στῷ → ἐν Χριστῷ (split; leaf 671's own ἐκ is scan-side OCR, Calfa's ἐν is right and the Latin's in concurs).
• 1325 ταῦτα ζητω → ζητῶ · καρπὸν Κἐνέγκητε → ἐνέγκητε (intrusive Κ) · ἤγημαι → ἥγημαι · Ἀσπάσασθε πάν τα ἄγιον → πάντα ἅγιον (split) · μάλισα δὲ, οἱ ἐκ τῆς Καίσαρος οἰκείας → μάλιστα…οἰκίας (Calfa's own gloss lemma four lines later prints οἰκίας) · Ἀμήν Ἐπειδὴ → Ἀμήν. Ἐπειδὴ · «Ἰi χάρις → Ἡ χάρις · Τὸ σύνηθες αὐτῳ → αὐτῷ · ἢ δόξα → ἡ δόξα · «Βἰ ἡμεῖς ὑμῖν → Εἰ ἡμεῖς.

3b. ⚑ TWO SOURCE LOSSES restored from the plate — a fresh convention call

Both are runs of text the plate carries and Calfa has dropped entirely, not mangled. The PG register section ("a certain correction … is made silently in the English and logged in cruces.md") is read as governing, in preference to a Pattern-13 [ed:], because in both cases the lost text is recovered from the plate itself — this is not speculation and nothing is supplied "from elsewhere". ✅ RULED 2026-08-01 (Wilson) — translation-style.md Pattern 13a, the visible-structure rule. The two cases SPLIT, and this entry's "both" was the wrong unit of decision.

• 1325, nine words lost mid-sentence → clause 1: stays SILENT, logged here. Restoring it with a marker would put our voice inside a running sentence for ordinary source damage, which is what clause 1 exists to prevent.
• The final subscription → clause 2: takes [ed:], because a lost structural unit's absence is otherwise invisible as an absence. Applied.

⚠ A blind reader, working from this entry's original prescription, proposed marking the 1325 restoration too, and it was briefly applied before being reverted — the entry outlived the rule it was written under. A prescription in a cruces file is only as current as the rulebook it cites; check the rule, not the note. Both also warrant data/calfa-patches/ entries (the plate check the ladder requires for a patch is satisfied — leaves 671, 672).

1. 1325, at the column break. Calfa: Οὐχ ὅτι ἐπιζητῶ τὸ δόμα ὑμῶν, διὰ τοῦτο [1325] λόγον ὑμῶν, οὐκ ἐμοῦ, πλεονάζοντα. — nine words gone, and what remains does not construe. Plate (leaf 671→672): …διὰ τοῦτο ταῦτα λέγω, ἀλλ᾽ ἐπιζητῶ τὸν καρπὸν ὑμῶν, τὸν εἰς λόγον ὑμῶν, οὐκ ἐμοῦ, πλεονάζοντα. Latin concurs: Non quod requiram munus vestrum propterea hoc dico, sed requiro fructum vestrum qui in rationem vestram non meam exuberat (twin 0013). Restored; the [1325] anchor kept exactly where our source prints it, i.e. inside the restored run.
2. ⚠ 1325, the work's final line — the subscription is truncated. Calfa ends: Τέλος σὺν Θεῷ τῆς πρὸς Φιλιππνσίους — and stops mid-phrase. Plate (leaf 672): Τέλος σὺν Θεῷ τῆς πρὸς Φιλιππησίους Ἐπιστολῆς. Ἐγράφη ἀπὸ Ῥώμης δι᾽ Ἐπαφροδίτου. Latin concurs: Finis, divino favente auxilio, Epistolae ad Philippenses. Scripta est ab urbe Roma per Epaphroditum (twin 0013). Restored in full. Without this the work appears to break off; with it, the work ends where Migne ends it.

4. Markers fired

4a. [lat: …] — one, plate-verified

• 1325, Phil 4:19. Greek lemma: «Σὺν Χριστῷ Ἰησοῦ.» Ἢ διὰ Ἰησοῦ Χριστοῦ δώσει. Plate-verified — leaf 672 prints «Σὺν Χριστῷ Ἰησοῦ», so this is not Calfa damage (and note the block lemma at the same column prints ἐν Χριστῷ Ἰησοῦ, which Calfa also carries correctly, so no systematic σύν/ἐν confusion is in play). Latin column (twin 0013, col 1326): « Per Christum Jesum. » Nam ἐν capitur pro …, id est In pro Per. The Latin's lemma and its whole scholion are about ἐν; the Greek column's lemma prints σύν. A reader of the Latin column takes away that the word under discussion is ἐν and the question is in vs per; a reader of the Greek takes away σύν, for which that discussion is beside the point. Material by the Pattern-16 test. English follows the Greek; marker fired. Not a Vulgate-conformation case — the Vulgate reads in gloria in Christo Jesu, so Per is Allatius's own choice, and his note says so.

4b. [lat: …] NOT fired — the candidates, and why

• 1312, «Ὧν ὁ θεὸς ἡ κοιλία» — Ὡς δεσποίνῃ γὰρ … δουλεύουσιν vs Latin Siquidem ipsum tanquam regem, imo potius tanquam deum colunt. Greek "as a mistress" (feminine, agreeing with κοιλία), Latin "as a king". This is grammatical-gender accommodation to venter, which is masculine in Latin — Allatius's freedom, not a different fact. Logged, no marker.
• 1316, μεθ᾽ ὑμῶν / 1320, μεθ᾽ ὑμῶν vs Latin nobiscum (twice). ⚠ The pronoun-itacism class: Greek at 1320 reads Εἰ γὰρ ἡμεῖς … εἰρηνεύομεν … καὶ ὁ Θεὸς ἔσται μεθ᾽ ὑμῶν. Θεοῦ δὲ ὄντος μεθ᾽ ὑμῶν… — first person then second, where the Latin is first person throughout (nos … Deus quoque nobiscum … quod si Deus nobiscum sit). This is exactly the anchor's 1261 case in reverse, and per the standing rule it needs the Greek plate-verified before firing. The plate is absent at this column (§0.2 — Calfa p.671 has no primary leaf). Candidate logged without the marker, and the reason stated, as the rule requires.
• 1316/1317, Vulgate-conformed scripture in the Latin — Pacem meam do vobis, pacem relinquo vobis (John 14:27); quaecunque sunt vera … boni omnis (Phil 4:8, where boni omnis renders εὔφημα by the Vulgate's sense); Gratia Domini nostri at 1326 against the Greek gloss lemma's bare Κυρίου. All conformation, none marked (anchor convention).
• 1316, δουλώσεται αὐτήν vs Latin subdet — free rendering, not material.
• 1320, ὡς ἐπὶ καρπῶν ξηρανθέντων vs Latin de arefactis prius fructuosis arboribus — Allatius supplies "trees"; free, not material.

4c. [var: …] — two, both plate-verified

Listed at §2.1 (1 Cor 15:25 at 1313) and §2.2 (John 14:27 at 1317). Four further candidates (§2.3–2.5, §2.13–2.14) were deliberately left unmarked because the column has no witness; each is named there so a later session can find them.

5. Polarity / negation sites read (runbook 4a, PG form: against both columns)

Every site below was examined; none was smoothed. Named individually so this pass can be told apart from a pass never made.

1. 1312, «τίς ἀμφισβητήσει ὅτι οὐ πάντως ἔσται τοῦτο;» — the οὐ is on the plate (leaf 667) and the Latin carries its exact structural equivalent, quin omnino id erit? Redundant negation after a verb of doubting, in both columns. Rendered with the English idiom that carries the particle — "who will dispute but that** this shall assuredly be?" — rather than dropping it (which would be a silent repair) or rendering it as a flat negative (which would state the opposite of the argument). Two-witness concurrence; no marker.
2. 1309, «εἰς διάκρισιν τῶν ὀρθῶς περιπατούντων, καὶ μὴ σκοπεῖτε τίνες περιπατοῦσι…» — ⚠ the most dangerous site in the batch. Calfa's comma invites reading καὶ μὴ σκοπεῖτε = "and do NOT mark", which reverses the sentence. The Latin settles the scope: ad internoscendum quinam recte ambulant, et qui non, considerate qui ambulent… — the μή belongs to the preceding pair ("those who walk rightly, and those who do not"), and σκοπεῖτε is a positive imperative. Rendered accordingly; the Latin is quoted here because it settled the reading, per the contract.
3. 1317, «Ὡς αἵγε μὴ ἔχουσαι εὐχαριστίαν εὐχαὶ, οὔτε γνωρίζονται πρὸς τὸν Θεόν.» — the Greek is doubly negative (prayers that do NOT have thanksgiving are NOT even made known). The Latin twin as OCR'd reads …precationes quae gratiarum actionem non habent, bene innotescant quidem, which asserts the opposite of the Greek's main clause. Judged a defect in the LATIN column's OCR (bene for ne/nec; leaf 669 gives the same string, so the fault may be the plate's own type) — the Greek negation is unambiguous, plate-verified, and required by the following sentence (Ὁ γὰρ μὴ εὐχαριστῶν … πῶς ἢ αἰτήσει ἢ λήψεται;). Greek negations carried; no marker, no smoothing.
4. 1321, «Ἐπειδὴ ἠρέμα ἐπετίμησεν αὐτοῖς γράψας, ἤδη ποτέ» vs Latin Quoniam scribendo clam eos objurgavit, non tandem ait — the Latin's non is OCR for Jam (= ἤδη ποτέ, the lemma being quoted); leaf 670 confirms the Greek has no negative. No divergence.
5. 1321, «Μὴ γὰρ νομίσητε … ὅτι νῦν διὰ τὸ μακράν με γενέσθαι … διὰ τοῦτο τὴν ὑμῶν χορηγίαν ἀπεδεξάμην» vs Latin Ne existimetis, inquit, quod non eo quod longe absim … propterea sumptum a vobis acceperim. The Latin carries a non the Greek does not print. Judged the same redundant- negative class as site 1 (a pleonastic negative after a verb of denying, which this Latin column demonstrably uses), not a polarity divergence. Greek rendered as printed; logged, no marker. ⚑ Worth noting as a class: this Latin column pleonastically negates after verbs of doubting and denying (quin, quod non), so a bare count of non against the Greek will manufacture divergences in this work.
6. 1320, «οὐκ ἐπλατύνθη οὐδέ εἰσιν οἱ … ἐπιχορηγοῦντες» vs Latin non adeo dilatata est … nec sunt qui … ministrent — negations 1:1. Clean.
7. 1321, «μήτε ἐπιφερόμενος, μήτε τῷ ἐργοχείρῳ» vs Latin neque aliquid importasset, neque manuali operi — 1:1. Clean.
8. 1324, «οὐκ ἴσχυσαν ἐνέγκαι … οὐκ ἤνεγκε τὸν κόρον … Ἀλλ᾽ οὐχ ὁ Παῦλος» vs Latin nec saturatum ventrem ferre potuerunt … satietatem non tulit … at non ita Paulus — 1:1. Clean.
9. 1325, «Οὐ γὰρ ἂν μὴ πιστεύσαντες, ἠσπάσαντο τοὺς Φιλιππησίους.» vs Latin neque enim Philippenses salutassent nisi credidissent — both negatives carried in the English ("For had they not believed, they would not have saluted the Philippians"). Clean.
10. 1325, «Οὐκ ἔδωκαν γὰρ σαρκικὰ, ἀλλ᾽ οὐδὲ λαβεῖν ἠθέλησαν πνευματικά.» vs Latin Non dederunt enim carnalia, sed neque accipere voluerunt spiritualia — 1:1. Clean.
11. 1325, «Οὐκ εἶπεν· Οὐδεμία μοι Ἐκκλησία δέδωκεν, ἀλλ᾽, Ἐκοινώνησε.» vs Latin Non ait: Nulla mihi Ecclesia dedit, sed, Communicavit — 1:1. Clean.
12. 1316, «Οὐδὲ ὁ Χριστὸς τὸ πένθος τὸ κοσμικὸν ἐπέταξεν … Οὐδὲ ὁ Παῦλος τὴν κοσμικὴν λέγει χαράν.» — in the p.669 gap; no verifier. Both οὐδέ carried; the internal logic (Christ commands mourning / Paul commands joy, resolved by "neither meant the worldly kind") is self-consistent, so no repair was needed and none was made.

6. Question marks restored (Calfa · or nothing for the plate's ;)

Logged as a class per the anchor, not per site. Restored at: 1309 Ἀπόλλυται δὲ πῶς; · Ποίους τούτους κελεύεις ἡμᾶς παραφυλάττεσθαι; · 1313 Εἶτα θέλεις καὶ ἑτέραν ἀπόδειξιν; · 1316 οὔτε ἐν Ἱερουσαλὴμ, ἔνθα ἐτέθραπτο; · Τί γὰρ ὁ Χριστός; · Πῶς οὖν φησιν ὁ Παῦλος, Χαίρετε; · 1317 Οὕτως, πῶς; · 1320 Τί χρὴ τὰ λοιπὰ περιεργάζεσθαι; · νῦν εἴληφεν; (with the person corrected, §3) · 1324 ποίας ἀρετῆς; · 1325 πόσῳ μᾶλλον … τοῦτο δεῖ ποιεῖν;

7. ⚑ Conventions decided fresh in this range (for the merge)

• ἰσχύω / ἰσχύς / ἐνδυναμόω = "be strong for" / "strength" / "empower" (§2.7) — fixed so Phil 4:13 and its gloss share a root, against AV's "I can do all things".
• ἐπιεικές / ἐπιεικῶς = "forbearance" / "with forbearance", never "moderation" or "gentleness" (Phil 4:5 and its gloss, 1317).
• σεμνά = "seemly" (Phil 4:8), not "honest"; the gloss defines it by bearing, speech, gait and deeds.
• ὑπερέχω / ὑπερβαίνω both = "surpass" (Phil 4:7 and its two glosses).
• πολίτευμα / πολιτεύομαι = "commonwealth" / "live as citizens" (1312), so the gloss's ἐκεῖ στρατευόμεθα, ἐκεῖ πολιτευόμεθα keeps its pair.
• μεμύημαι = "I have been initiated" (not "I am instructed"), identical in lemma and gloss.
• Chapter heads: ΚΕΦΑΛ. Ϛʹ + argument → "CHAPTER 6." + the argument sentence, per the anchor's ΚΕΦΑΛ. Αʹ handling.
• The Calfa bleed class in this work includes the volume RUNNING FOOT (PATROL. GR. CXVIII), not only the Latin running title — two instances, 1316 and 1324, both mid-word.
• Plate-verified source LOSSES are restored silently and logged (§3b), rather than marked [ed:]. Ratify or reverse at merge.
• A broken guillemet pair in our source is repaired, not left half-open (§3, 1316). This is repair of our file, not supply against the plate.

Work-wide conventions (anchor session, 2026-07-31 — merge into the work-wide section)

• Catena attribution sigla are sacred structure. The plate opens each attributed scholion with an em-dash + bracketed source name: — [ΦΩΤ.], — [ΧΡΥΣ.] (scan-OCR-verified at four sites: leaves ~647, ~648, ~651, ~653, ~654 of patrologiaecurs14migngoog). Reproduce VERBATIM, untranslated, in place — they are Migne's citations (the [n:] principle). They carry no colon, so they cannot collide with our [tag: …] marker namespace. Calfa MANGLES them (ἴΦΔΤ. for [ΦΩΤ.], -ΧΡΥΣ.] for — [ΧΡΥΣ.]): restore each from the plate form and log it. Expect more sigla ahead (the volume's catena cites more fathers than these two); verify each new one against the scan before fixing its form.
• Lemma consistency is the work's spine (7a″ in force). The scripture block-lemma and every «short lemma» quoted from it in the glosses must render identically where the Greek is identical — and both must follow MIGNE'S printed Greek, not the received text and not familiar English (worked cases below: single μᾶλλον at Phil 1:9; anarthrous θλίψεως in the 2 Cor 2:4 quote).
• Guillemets as printed: short gloss-lemmata keep «…»; the long block lemmata are unguilleted in our source and stay so.
• The Latin column quotes scripture from the VULGATE — that is convention, not divergence. Two instances in this chunk alone (Ex multa afflictione where the Greek prints Ἐκ γὰρ θλίψεως without πολλῆς; magis ac magis where the Greek prints a single μᾶλλον). Never fire a [lat:] on a scripture quotation the Latin has conformed to the Vulgate; log if notable. (§8 Q3 calibration datum, pairs with the Nicetas "plate's own apparatus" rule.)
• Mentioned-word refinement (extends the Nicetas rule): a discussed word stays in Greek script only where English cannot carry the point (ἐξ/διά); where the morphological contrast survives translation, translate it and keep lemma-consistency (κοινωνούς/συγκοινωνούς = partakers / fellow-partakers — as the plate's own Latin does: socios… una consortes).
• Greek question marks: Calfa often prints · or nothing where the plate's ; belongs; restore the question from sense + the Latin column (three sites this chunk), silently, logged as a class not per-site.
• Kephalaia: ΚΕΦΑΛ. Αʹ + argument renders "CHAPTER 1." + the argument sentence (the Latin column's own CAPUT I practice).

Latin-twin pass (pilot §6): result for chunk 0000 — FINDINGS (not clean)

Read against src/pg-latin/oecumenius-philippians/0000.md (cols 1259–1263). This batch is chunk 0000 only; it was not clean. One [lat:] fired (plate-verified); two Vulgate-conformations in the Latin logged as convention, not divergence.

The [lat:] site — plate-verified per the Pattern-16 numeral rule (extended to itacism pairs)

• 1261/1262 Greek: δίκαιόν ἐστιν ἐμοὶ τῷ ἀκριβῶς εἰδότι τὰ καθ᾽ ἡμᾶς ("the things that concern us"). Latin: qui exacte omnia novi quae apud vos sunt. The Greek is NOT Calfa damage — the scan's own OCR of the plate concurs (leaf ~647: "τῷ ἀκρ'θῶς εἶδότι τὰ καθ ἡμᾶς"), so both our witnesses of the Greek column agree against the Latin. ἡμᾶς/ὑμᾶς is the pronoun-itacism pair — the prose analogue of the numeral class (identical Byzantine pronunciation, endemic manuscript confusion; Allatius may have emended silently or read another exemplar). English follows the printed Greek; marker fired: [lat: the Latin gives quae apud vos sunt, the things that are with you].

Ladder row 1 — Calfa OCR damage, corrected silently, witness cited

• 1260 (head) Calfa's opening line "ΡΑΙΙΙ ΑΡΟΘΤΟΙΙ ΑΒ ΡΗΙΗΙΡΡΕΝβΕΘ ΕΡΙΘΤΟΙΑ. ΚΟΛΡΠΤ Ι." is the LATIN column's running title (PAULI APOSTOLI AD PHILIPPENSES EPISTOLA. CAPUT I.) misread in Greek lookalike glyphs — cross-column bleed, not the Greek column's text. The Greek column's own head (scan-verified) is ΠΑΥΛΟΥ ΑΠΟΣΤΟΛΟΥ Η ΠΡΟΣ ΦΙΛΙΠΠΗΣΙΟΥΣ ΕΠΙΣΤΟΛΗ → rendered; the CAPUT I belongs to the Latin column (carried by the twin) and is not rendered from the Greek.
• 1260 Catena sigla restored from plate form: ἴΦΔΤ. → — [ΦΩΤ.] · -ΧΡΥΣ.] → — [ΧΡΥΣ.] (scan-verified, see conventions).
• 1260 (block lemma Phil 1:1–2) Τμόθεος → Τιμόθεος · Σριστοῦ → Χριστοῦ (×2) · τρῖς → τοῖς · οὖσυι → οὖσιν · ἑν → ἐν · Παερὸς → Πατρὸς · Ἰσοῦ → Ἰησοῦ (standard text, Latin concurs).
• 1260 ἁξίωμα → ἀξίωμα · κοινῳ → κοινῷ · Χριστῶ → Χριστῷ · ὀνξμασι → ὀνόμασι · Κμὺχαριστία → Εὐχαριστία (ΚΕΦΑΛ. argument) — the last is the intrusive-Κ class known from PG 139's Calfa file, now seen in PG 118 too (also εὐχαΚριστίας → εὐχαριστίας, Κμεμνῆσθαι → μεμνῆσθαι at 1260).
• 1260 (block lemma Phil 1:3–5) μυεᾳ → μνείᾳ · δεύσει → δεήσει · ὑαὲρ πόνττων → ὑπὲρ πάντων · πρόι θμέρας → πρώτης ἡμέρας · νῶν → νῦν.
• 1260 Ἐπ πάσῃ → Ἐπὶ πάσῃ · Μετὰχαρᾶς → Μετὰ χαρᾶς (split/joined type) · Ἔν γὰρ → Ἔνι γὰρ (ἔνι = ἔνεστι, "it is possible"; Lat. contingit) · συοχῆς → συνοχῆς (2 Cor 2:4) · Τν → Τὴν · εὐχαρι στῶ → εὐχαριστῶ · Μέῖ ἱ → Μέγα τι (Lat. magnum quiddam) · Ὅτ… ε οὖ → Ὅτι… ἐξ οὗ · πράττετς → πράττετε.
• 1260/1261 Σομο παρόντες… τὰ ὗόν τὴν χρείαν → Ὥσπερ παρόντες… τὰ πρὸς τὴν χρείαν (Lat. mihi semper praesentes… mittentes quae necessaria erant; the [1261] anchor sits inside the restored phrase).
• 1261 ἐξ ἠς → ἐξ ἧς · Ὃτι → Ὅτι · stray ! after ἐναρξάμενος · ησοῦ → Ἰησοῦ · συγ. κοινωνούς → συγκοινωνούς (split) · ἄχρί → ἄχρι · ἣνδίκαιον → ἣν δίκαιον · τῆς χάρι. τος → τῆς χάριτος (split) · Ἠ/Ἡ → Ἢ ("Or:", thrice, the gloss's alternative-construal formula; Lat. aut) · ποκίλως → ποικίλως · τὰ καθ᾽ — see [lat:] above.
• 1264 ἐβεβαίο → ἐβεβαίου · μόνρν → μόνον · Ἰκανὴν → ἱκανὴν · εἰναι → εἶναι · ημῖν → ἡμῖν · ἱσχυροτέρους → ἰσχυροτέρους · Δὰ τί → Διὰ τί · Ἑπειδὴ → Ἐπειδὴ · συγκοινωνοῖὺς → συγκοινωνούς · Ἱν → Ἵνα · τοὺ → τοῦ.
• 1264 (block lemma Phil 1:8–11) Μίάρτωρ… ἑστιν… ὀς ἐπικοθῶ → Μάρτυρ… ἐστιν… ὡς ἐπιποθῶ (⚠ CORRECTED 2026-07-31 by the 0001–0005 agent, then re-verified here: this file first recorded the fix as Μάρτυς. The plate prints Μάρτυρ — twice, leaf 648, in BOTH the block lemma and the short lemma «Μάρτυρ γάρ μού ἐστι ὁ Θεός». Against the received Phil 1:8 μάρτυς, that is a scripture divergence, not OCR damage, and it belongs in the divergence list below, not here. No English change — both are "witness" — but the record was wrong, and a wrong record is how a real reading gets swept next time.) · Σριστοῦ → Χριστοῦ · Ἰνα → ἵνα (×2) · ἔτιν → ἔτι · πόσῃ οἰσθήσει → πάσῃ αἰσθήσει · διαφέροντο → διαφέροντα · ἡμέρο → ἡμέραν · αεπληρωμένοι → πεπληρωμένοι · τέν → τὸν · Θεοῖ → Θεοῦ.
• Restored question marks (Calfa · or nothing for the plate's ;): οὐκ ἔχων ἔθος; (Lat. ?) · Τί εὐχαριστῶ, φησὶ, τῷ Θεῷ; · Πῶς; (before Ὅτι ᾑρεῖτο) · Τί δὲ τὸ αἴτιον…ἔχειν;

7a″ worked cases (the deliverable list)

• 2 Cor 2:4 quote at 1260: Migne's Greek prints «Ἐκ γὰρ θλίψεως…καὶ συνοχῆς καρδίας» — NO πολλῆς. Rendered "out of affliction and anguish of heart," resisting the received "out of much affliction." The Latin column prints the Vulgate's Ex multa afflictione — conformation, logged, no marker.
• Phil 1:9 at 1264: Migne prints ἔτι μᾶλλον περισσεύῃ — a SINGLE μᾶλλον. Rendered "may abound yet more," resisting AV's "yet more and more." Latin: adhuc magis ac magis (Vulgate) — same class, logged, no marker.
• Phil 1:8 at 1264 — Μάρτυρ for the received μάρτυς, plate-verified twice on leaf 648 (block + short lemma). Migne's own consistency across both quotations is what makes it a reading rather than a slip. Logged here as the divergence it is; see the corrected entry above.
• Phil 1:6 block lemma: Calfa opens Πεποιθὼς δέ — the δέ kept ("And trusting…"); the Latin block (Confidens hoc ipsum) lacks a connective, received text lacks δέ; δέ may be Calfa noise but the reading construes and is not plate-checked — kept as our source prints, logged.

Deferred to the work's later chunks

• The chunk ends with the Phil 1:8–11 block lemma; its glosses open chunk 0001 (the twin's tail already carries their Latin). The 0001 agent should begin gloss-lemma consistency from THIS chunk's block rendering.

Continues cruces-0000.md (the Fable anchor). Its work-wide conventions section is binding here and was followed without exception; only additions, refinements and one correction of the anchor's record are set out below. Scan OCR consulted throughout: raw/scans/pg118/patrologiaecurs14migngoog_djvu.xml, leaves 648–658 (split on <OBJECT).

Conventions decided fresh in this batch (for the merge)

• ἁρπαγμός / ἁρπάζω / ἁρπαγή share one English root: "seizure / seize." Phil 2:6 is glossed by its own etymology all through chunk 0004 (Οὐχ ἁρπαγμὸν ἡγήσατο → Τουτέστι, κατὰ ἁρπαγήν → Ὅταν τις ἁρπάσῃ τι → ἐξ ἁρπαγῆς, ἐκ πλεονεξίας). AV's "robbery" breaks the chain at the verb, so the gloss would explain nothing (the stylus/epistylium precedent, style rule 2). Block lemma: "counted it not a seizure to be equal with God."
• πεποιθὼς / θαῤῥῶν / τεθαῤῥηκώς are kept apart, per work, always. Migne prints all three within eight lines at 1273–1274 and the Photius scholion turns on their equivalence: πεποιθὼς = "trusting" (anchor form, 0000) · θαῤῥῶν = "being confident" · τεθαῤῥηκώς = "having taken confidence." Allatius flattens all of them to persuasus / cum fiducia — the standing warning that his paraphrase is not a licence.
• Doctrinal vocabulary fixed (dialogue-anchor rule extended to catena): ὑπόστασις = hypostasis (Allatius gives persona; not followed) · πρόσωπον = person · οὐσία = essence · μορφή = form · φύσις = nature · ἐνέργεια = operation. The Arian/Sabellian argument at 1281 IS these distinctions.
• ἐρίθεια = "contentiousness" throughout (1268, 1277) · γογγυσμός = "murmuring" · διαλογισμός = "disputing" (1288, where the gloss defines διαλογισμοί as ἀμφιβολίαι, "doubts") · εἰλικρινής = "sincere" · ἀπρόσκοπος = "without offense" · ἄμεμπτος = "blameless" · ἄψεκτος = "beyond reproach" (the four stand in one gloss chain at 1265 and 1288 and must not collapse).
• Migne's unclosed guillemets are closed silently — the anchor's own practice (0000, ἀλλὰ, «συγκοινωνοῖὺς, closed as «fellow-partakers»). Sites this batch: 1268 «Οἱ μὲν ἐξ ἐριθείας (closed after "contentiousness," matching the Latin's «Alii quidem ex contentione.») · 1277 «Πῶς; (closed after "How?"; the PLATE itself prints the stray «, leaf 654) · 1281 «Οὐχ ἁρπαγμὸν ἡγήσατο (plate also unclosed, leaf 656) · 1284 «Μακάριοι… τῶν οὐρανῶν · 1288 «Ὑπὲρ τῆς εὐδοκίας.
• Guillemet parity with the Latin twin is NOT achievable in PG and must not be forced. Allatius marks his lemmata on his own convention: he guillemets «Quid enim» and «Alii quidem ex contentione.» where Migne's Greek prints no marks at all, and he sets whole block lemmata inside per-line « … » that the Greek column never carries. The English follows Migne's Greek, per Pattern 6's provenance rule. verify-english check 7 (guillemet counts equal the twin's) was written for PL, where the twin is the source text; on this work it will report differences that are correct.
• Cross-column bleed, second and third instances (anchor class, ΚΟΛΡΠΤ Ι.): 1276 ΚΟΑΡΙΤ ΙΙi = the Latin column's CAPUT III in Greek lookalike glyphs, and 1267/1268 ΖΔΑὗΤ Ι. at the column break. Neither is Greek; not rendered. The Greek column's own head (ΚΕΦΑΛ. Γʹ) is rendered separately and in its own place.

Correction to the anchor's record (chunk 0000) — no English change

• cruces-0000.md lists «Μίάρτωρ → Μάρτυς» among its silent corrections. The plate does not support Μάρτυς. Leaf 648 prints the Phil 1:8 block lemma as Μάρτυρ γάρ μού ἐστιν ὁ Θεός and the short lemma at 1264/1265 as «Μάρτυρ γάρ μοῦ ἔστι ὁ θεός.» — Μάρτυρ in both places, against the received Μάρτυς. The English ("For God is my witness") is unaffected, so nothing is re-cut; but the crux line should read Μάρτυρ (Migne's form, plate-verified) — not normalized, and this belongs in the 7a″ divergence list below, not in the OCR list. Logged rather than edited, per the do-not-churn rule.

The 7a″ deliverable — where Migne's printed Greek departs from the received text

Named word by word. This is the list, not a summary of one.

1. Phil 1:20 (block lemma, 1272): Migne prints ἐν τῷ ΣΤΟΜΑΤΙ μου, "in my MOUTH," where the received text reads σώματι, "in my body." Plate-verified, leaf 651: καὶ νῦν μεγαλυνθήσεται Χριστὸς ἐν τῷ στόματί μου. Two lines later Migne's own short lemma prints σώματί (leaf 652: «Ἐν τῷ σώματί μου.», and the gloss's paraphrase μεγαλυνθήσεται ὁ Χριστὸς ἐν τῷ σώματί μου), and the Latin column reads in corpore meo. Rendered exactly as printed in each place — "in my mouth" in the block, "in my body" in the gloss — with [var: Gk σώματί μου, "in my body"] on the block. No [lat:]: the columns do not descend from different exemplars here; the plate's own gloss two lines down refutes that, and Pattern 16 is not a defect marker. This is the exact trap the 7a″ clause exists for: AV's "Christ shall be magnified in my body" is the sentence that arrives unbidden.
2. Rom 12:18 quoted at 1265: Migne prints the IMPERATIVE εἰρηνεύετε ("live at peace"), where the received text has the participle εἰρηνεύοντες. Plate-verified, leaf 649. The Latin agrees (in pace vivite). Rendered as an imperative.
3. John 1:14 quoted at 1281–1284: Migne prints Καὶ εἴδομεν τὴν δόξαν αὐτοῦ ὡς Μονογενοῦς παρὰ Πατρός — εἴδομεν for the received ἐθεασάμεθα, and the second δόξαν is absent. Plate-verified, leaf 656. The Latin column restores the Vulgate's doubled gloriam… gloriam quasi Unigeniti — Vulgate conformation, logged, no marker (calibration rule). No [var:]: the sense is not changed by either.
4. Phil 1:23 (block lemma, 1272): πολλῷ μᾶλλον κρεῖσσον with NO γὰρ (received πολλῷ γὰρ μᾶλλον κρεῖσσον). Plate-verified, leaf 652.
5. Phil 2:3 (block lemma, 1277): μηδὲν κατὰ ἐρίθειαν, ἢ ΚΑΤΑ κενοδοξίαν — Migne repeats κατά, which the received text does not. Plate-verified, leaf 655.
6. Phil 2:4 (block lemma, 1277): ἀλλὰ καὶ τὰ ἑτέρων ἕκαστος — singular ἕκαστος where the received text has plural ἕκαστοι. Plate-verified, leaf 655 (which also shows τὰ, against Calfa's τὸ — that one is our file's damage, see the OCR list).
7. Phil 1:16–17 (block lemma, 1265): Migne prints the Byzantine order (ἐριθείας clause first) and θλίψιν ΕΠΙΦΕΡΕΙΝ (received/critical ἐγείρειν). Rendered "thinking to bring affliction upon my bonds."
8. Ps 32(31):22 quoted at 1272: Γένοιτο, ΚΥΡΙΕ, τὸ ἔλεός σου ἐφ᾽ ἡμᾶς… ἐπὶ ΣΟΙ — vocative moved forward, and dative σοί for the LXX's accusative σέ. Plate-verified, leaf 651.
9. Phil 1:6 short-lemma pair (1272–1273): «Καὶ τοῦτο ΘΑΡΡΩΝ οἶδα,» followed immediately by «Καὶ τοῦτο ΠΕΠΟΙΘΩΣ οἶδα.» — Migne prints a paraphrastic first lemma against his own block lemma's πεποιθώς, and the [ΦΩΤ.] scholion then glosses πεποιθώς by τεθαῤῥηκώς. Plate-verified, leaf 653. All three rendered distinctly (see conventions).
10. Phil 1:19 short lemma (1272): «Κατὰ τὴν ΚΑΡΑΔΟΚΙΑΝ.» without the ἀπο-, though the block lemma prints ἀποκαραδοκίαν and the gloss immediately resumes Ἀποκαραδοκίαν φησί. Plate-verified, leaf 651. Rendered "expectation" in the short lemma against "earnest expectation" in the block and gloss — the distinction is on the page.
11. Phil 2:15 (block lemma, 1288): ἀμώμητα (Byzantine) against the critical ἄμωμα — noted for completeness; rendered "without blemish."

Not divergences, deliberately: Phil 1:9's single μᾶλλον was already logged by the anchor; note that the GLOSS on it at 1265 prints the doubled ἔτι μᾶλλον καὶ μᾶλλον (plate-verified, leaf 649). Both are rendered as printed — "yet more" in the lemma, "yet more and more" in Oecumenius's paraphrase. The divergence is internal to Migne's page and is not smoothed in either direction.

[lat:] markers fired — three, each plate-verified on both columns

1. 1265 / Latin 1266 — Rom 12:18, a missing protasis. Greek: Εἰ δυνατὸν τὸ ἐξ ὑμῶν μετὰ πάντων ἀνθρώπων εἰρηνεύετε; (plate leaf 649). Latin: Quantum ex vobis est, cum omnibus hominibus in pace vivite? (plate leaf 649) — Εἰ δυνατόν is simply absent, and the Latin is not conforming to the Vulgate here either (Vulg. si fieri potest), so the Vulgate-convention rule does not cover it. A reader of the Latin gets an unconditional command. Marker: [lat: the Latin column omits Εἰ δυνατόν, giving the command unconditionally — Quantum ex vobis est, cum omnibus hominibus in pace vivite].
2. 1272 / Latin 1271 — "salvation" construed to opposite effect. Greek: Σωτηρίαν δέ φησι, τὸ ὅσον οὐδέπω μαρτυρίαν — the martyrdom that is all but here (plate leaf 651). Latin: Salutem autem dicit quod nondum pateretur martyrium — that he was not yet suffering martyrdom (plate leaf 651). The two columns give the reader contrary accounts of what σωτηρία means in Phil 1:19. English follows the Greek; marker fired.
3. 1284 / Latin 1283 — a polarity the columns disagree on. The batch's strongest find. Greek: Ποῦ εἰσιν οἱ λέγοντες ὅτι κατὰ ἀνάγκην καὶ ἐπιταγὴν τοῦ Πατρὸς ἑκὼν ἐνηνθρώπησεν; — plate-verified, leaf 657, which prints ἑκὼν ("willingly"), producing a sentence that contradicts its own "by necessity and command." Latin: invitus est incarnatus ("unwillingly was he made man"), plate-verified leaf 657 — coherent, and confirmed as deliberate by the same column's voluntariam dejectionem for τὸ ἐκούσιον eight lines later. This is not Calfa damage (both our witnesses of the Greek agree) and it is exactly the class runbook 4a hunts. Rendered literally — "he willingly became man" — and marked. Never smoothed to "unwillingly."

[var:] fired — one

• 1272 Phil 1:20, [var: Gk σώματί μου, "in my body"]. See divergence 1.

Candidates examined and NOT marked, with reasons

• 1268 / Lat. 1267 Greek Εἰ γὰρ μὴ θεῖον ἦν, φησί, τὸ κήρυγμα… (singular, plate leaf 650) vs Latin Nam aiebant: Nisi divina esset praedicatio… (plural imperfect). Allatius has construed the habitual φησί as the brethren's speech, which the context permits. Free rendering, not divergence.
• 1277 / Lat. 1278 Greek short lemma «Τὸ ὑπὲρ Χριστοῦ» vs Latin «Hoc a Christo» (donatum est a Christo, "from Christ") — a real difference of sense (ὑπέρ vs ὑπό). Not marked: Migne's own Latin column carries the note that settles it — (Nam ὑπὲρ capitur pro ὑπὸ, id est a), printed at Latin col. 1278, plate leaf 655 — and Oecumenius's own Greek gloss two lines later reads ἐχαρίσθη ὑπὸ Χριστοῦ (plate leaf 655). The divergence is adjudicated on the page. This is the second instance of the Nicetas "plate's own apparatus" rule and the first where the adjudicating note sits in the LATIN column; §8 Q3 calibration datum.
• 1285 / Lat. 1286 Phil 2:12 block lemma: Greek ἀλλὰ νῦν πολλῷ μᾶλλον (plate leaf 658) vs Latin sed tunc multo magis (plate leaf 657). Not marked: the Latin departs here from the Greek and from the Vulgate (nunc) at once, which points to the Latin column's own compositor rather than to a divergent exemplar; Pattern 16 records column disagreement, not Latin-side type defects, which this edition does not mark because it does not edit the Latin. Logged so the claim is checkable.
• 1285 / Lat. 1286 Greek ὅταν ὡς κριτὴς καθίσῃ ("when he sits as judge") vs Latin quando ut judex descenderit. Free/erroneous rendering of the same event; no fact a reader would carry away differently.
• 1284 / Lat. 1283 Latin nos qui natura æquales sumus et æqualiter humiles et abjecti amplifies Greek οἱ φύσει ὁμότιμοι καὶ ὁμοίως ταπεινοί (plate leaf 656). Allatius's norm.
• 1272 / Lat. 1271 Latin renders ὑποστάσεις by personas in the heresy catalogue (1272). Conventional Latin equivalence, not a divergence; the English keeps "hypostases" per the fixed vocabulary above.

Vulgate conformations in the Latin column — logged, never marked

• Lat. 1266 Rom 12:18 is not conformed (see [lat:] 1 above) — recorded because it is the exception that shows the convention is not automatic.
• Lat. 1282 John 1:14 conformed to the Vulgate's doubled gloriam.
• Lat. 1283 1 Cor 2:9 and Matt 5:3 conformed.
• Lat. 1287 Ps 2:11 conformed (Servite Domino in timore).

A defect in Migne's own Latin column, noted not marked

• Lat. col. 1286 carries the running title EPIST. AD GALAT. in the middle of the Philippians commentary (plate leaf 657, and the committed twin src/pg-latin/oecumenius-philippians/0005.md reproduces it). Migne's error, on the verifier side only; nothing in the English depends on it.

Catena sigla met, and every restoration

Four scholia in this batch, three sources. [ΒΑΣΙΛ.] is new to the work and was verified against the scan before its form was fixed.

The one unverified restoration, stated plainly: at 1281 Calfa prints ππ[Τ. opening the Βούλεσθε εἰδέναι scholion. The Greek column for Migne cols 1279–1281a is absent from our scan — leaf 655 carries the Greek only as far as Ὅταν γὰρ τὸ ἴδιον συμφέρον, and leaf 656 resumes at …νετο. Οὐκοῦν καὶ μορφὴν Θεοῦ; the intervening Greek column was not OCR'd in patrologiaecurs14migngoog. The restoration to — [ΦΩΤ.] rests on the surviving final Τ (ΧΡΥΣ. and ΒΑΣΙΛ. both end in Σ/Λ) and on the four verified [ΦΩΤ.] sites in the same span. It is a strong inference, not a plate check, and is recorded as such.

Ladder row 1 — Calfa OCR damage corrected silently, witness cited

Itemized by column. Where the plate was consulted the leaf is named; where the Latin twin supplied the sense that is said instead.

1265 (chunk 0001). ἡ μορφή δούλο.. → ἡ μορφὴ δούλου · ιΠσου → Ἰησοῦ · ὀισκειμένοις → διακειμένοις · ποθεῖσθαίν → ποθεῖσθαι · παρ → παρ᾽ · ἣ καὶ περὶ αῶν ἄλλων αῃσὶν → ἢ καὶ περὶ τῶν ἄλλων φησὶν (leaf 649) · Ιολλοὺς → Πολλοὺς · Λετὰ → Μετὰ · Ἴα → Ἵνα · οἵα… οἶ διὰ κενοδοδίαν → οἷα… οἳ διὰ κενοδοξίαν · ἐπιΚμελοῦνται → ἐπιμελοῦνται (intrusive-Κ class) · τὸν] διδάσκαλον (stray bracket). 1265, a lost tail restored from the plate: Calfa runs τοῦ καὶ συνεργοῦνΚΕΦΑΛ. Βʹ. — the plate (leaf 649) reads τοῦ καὶ συνεργοῦντος πρὸς τοῦτο. before the chapter head; Latin qui etiam in hoc auxilium praebet. Restored, not marked [ed:] (Pattern 13 is for holes a crux cannot carry; this one is plate-verified and four words long).

1265 (chapter head). Διήγησις τῆς ἑαυτοῦ διαγωγὴς… προθυμίας,. → διαγωγῆς … προθυμίας. (leaf 649). 1265 (block lemma Phil 1:12–18). εἰς αροκοπὴν → προκοπὴν · Εἰαγγελίου ×3 → Εὐαγγελίου · ἐλήλυοεν → ἐλήλυθεν · ἐν λριστῷ → ἐν Χριστῷ · ὁλῳ → ὅλῳ · πασι → πᾶσι · Τρές μὲν → Τινὲς μὲν · Οἱ μέν ἐξ] (stray bracket) · καταγηέλλουσιν → καταγγέλλουσιν · τοῆς δεσμοῖς → τοῖς δεσμοῖς · Οἱδέ → Οἱ δὲ · Τίγάρ· → Τί γάρ; · Πλὴνν → Πλὴν · Χχιστὸς → Χριστὸς · χαούσοααι → χαρήσομαι (all leaf 649).

1268. Εἰκὸς ἦναὐτοὺς → Εἰκὸς ἦν αὐτοὺς · δεικνὸς → δεικνὺς · προκοπή, → προκοπή. · θτι → Ὅτι · οηλοῦν → δηλοῦν · τὰ κατ’´ αὐτίν → τὰ κατ᾽ αὐτόν · Οὕτε → Οὔτε · ἐν Χριστῳ → ἐν Χριστῷ · εἰοὐκ → εἰ οὐκ · Ἡ, ἐκείνων → Ἢ, ἐκείνων · τὸν ἑν οὐρανοῖς → ἐν οὐρανοῖς · ὡφελεῖσθαι → ὠφελεῖσθαι · εἶπερ Βεβαιώσει → εἶπε. Βεβαιώσει (leaf 650; Lat. Hoc etiam superius dixit) · ἦνα ἀκούσας → ἵνα ἀκούσας · ἀποκτείνη → ἀποκτείνῃ. 1268, a lost parenthesis restored: Calfa Ἡ. ὅτι αὐτοὶτεμᾶσθαι βουλόμενοιεγάλη γὰρ ἦν ἡ τιμὴ τοῦ Παύλου) → plate (leaf 650) Ἢ, ὅτι αὐτοὶ τιμᾶσθαι βουλόμενοι (μεγάλη γὰρ ἦν ἡ τιμὴ τοῦ Παύλου); Lat. magnus siquidem erat Pauli honor.

1268/1269. σασι → Ἴσασι (leaf 650; Lat. Sciunt) · Τοῦο οὗν → Τοῦτο οὖν · τὸ ὰ λόγον → τὸν λόγον (leaf 650) · εἱ μὴ → εἰ μὴ · ποίω σκοπῳ → ποίῳ σκοπῷ · όπότε ὁρθῶς → ὁπότε ὀρθῶς · ἔδοκεν → ἔδωκεν · Ἀλλὰσὺ → Ἀλλὰ σὺ · εἰρήκει,καὶ γὰρ → εἰρήκει, (καὶ γὰρ… (the plate, leaf 651, prints the parenthesis). 1269, the largest single loss in the batch: Calfa runs ὁ Ἀπόστοστὸς [1269] καταγγέλλεται.» — the plate (leaf 650 end / 651 start) reads ὁ Ἀπόστολος, τῷ εἰπεῖν «Εἴτε προφάσει, εἴτε ἀληθείᾳ Χριστὸς καταγγέλλεται.», and the Latin (1270) has the whole of it. Restored silently; the [1269] anchor kept in place.

1272 (chunk 0002). πονηρὰνο γὰρ πλείους → πονηρὰν (οἱ γὰρ πλείους (leaf 651) · ὀιὰ τῆς → διὰ τῆς · Τησοῦ → Ἰησοῦ · παρ´ αὐπῶν → παρ᾽ αὐτῶν · ὸφειλομένην → ὀφειλομένην · ἰΕἰ γὰρ → Εἰ γὰρ · δαψῖλεια → δαψίλεια · θτι ἐν οὐδενὶ → Ὅτι ἐν οὐδενὶ · ν γὰρ ἀποκτείνωσι → Ἂν γὰρ ἀποκτείνωσι (leaf 651) · Οίον → Οἷον · λογιζομένων τῷν ὁρώντων → τῶν ὁρώντων. 1272. Παῆλος → Παῦλος · Εἵτε ἐξα·ρεῖταί → Εἴτε ἐξαιρεῖταί (leaf 652) · κἂν οεῃ με → κἂν δέῃ με · ἐν τῳ σώματί → ἐν τῷ σώματί · Χριατὸν → Χριστὸν · ᾤοντο → ᾤοντο · οὐχ οὔτως → οὐχ οὕτως · τὴν μὲν ποώτιν → τὴν μὲν πρώτην · ἐξήαπασέ → ἐξήρπασέ · θορυφῆσθαι → θορυβεῖσθαι (leaf 652). 1272/1273 (block lemma Phil 1:21–26). Σριστὸς → Χριστὸς · ἀποθαιεῖν → ἀποθανεῖν · καρπδς → καρπὸς · αἱρήσομαι οὐ γνωρίζω ✓ · ἀναλδσαι → ἀναλῦσαι · εἰν αι → εἶναι · μᾶλλρν → μᾶλλον · Τὸ δὴ ἐπιμένειν → Τὸ δὲ ἐπιμένειν (leaf 652) · ἀναγκαιότειον → ἀναγκαιότερον · ὃτι μένω → ὅτι μενῶ (leaf 652) · σιμπαραμενῶ → συμπαραμενῶ · πᾶσιν ὑμῶ → πᾶσιν ὑμῖν · ὑμών Χροκοπὴν → ὑμῶν προκοπὴν · ἵνατὸ καιὶχημα → ἵνα τὸ καύχημα · ἐν λριστῷ → ἐν Χριστῷ · διὰ κῆς → διὰ τῆς. 1273. Κζῶ → ζῶ (intrusive Κ) · σῶν στεσάνων → τῶν στεφάνων · Κἰ δὲ τὸ ζῆν → Εἰ δὲ τὸ ζῇν · μή τς νομίση → μή τις νομίσῃ · ἀγαθῶν]· (stray bracket). 1273, restored across the column break: Calfa prints the garbage run οὐκ οἶδα, οὐδὲ γνωρίζω τί οοωοοΟΟΟΘ [1273] φορεῖν τῷ Θεῷ — the plate (leaf 652) reads …τί αἱρήσομαι καὶ ἐπιλέξομαι, τὸ ζῇν, φησί, καὶ καρπο-φορεῖν τῷ Θεῷ; Latin nondum mihi decretum est, neque satis novi quid malim aut quid eligam, an vivere ac fructificare Deo. Restored; the [1273] anchor kept inside the restored word, as the plate breaks it. 1273/1274. τοῦ τε ὰναλῦσαι → ἀναλῦσαι · Εἰτα. ὡ Παῦλε → Εἶτα, ὦ Παῦλε · οὐσαν → οὖσαν · Διατί ὀὲ → Διὰ τί δὲ · οἱκείου/ὡφέλειαν → οἰκείου/ὠφέλειαν · ἐβεβαίουκαὶ ἄλλους πρρσελάμβανε → ἐβεβαίου καὶ ἄλλους προσελάμβανε · Πολλῷ Κμᾶλλον → Πολλῷ μᾶλλον (intrusive Κ) · πολλῶ μᾶλλον → πολλῷ μᾶλλον · τὸ. εἶν αι → τὸ εἶναι · δ θαῤῥῶν → ὃ θαῤῥῶν.

1276 (chunk 0003). Κἐαυτοῦ → ἑαυτοῦ (intrusive Κ) · δισταγμῳ → δισταγμῷ · δψομαι → ὄψομαι · ες αὐτὴν → εἰς αὐτὴν · Ἰἱ γὰρ → Εἰ γὰρ · περισσεύσοι ✓ · βειοῦν → βιοῦν · ὡφέλειαν → ὠφέλειαν · Ἀν γὰρ → Ἂν γὰρ · συμπαραμενω → συμπαραμενῶ. 1276 (chapter head, a sense-bearing correction). Calfa prints τῆς ἀνθέου ζωῆς; the plate (leaf 654) prints ἐνθέου and the Latin vitamque divinam. Rendered "the life inspired by God." Calfa's ἀνθέου is not a word in this sense and would have inverted the head. 1277. ποἸτεύεσθε → πολιτεύεσθε · εἰτε ἐλὸὼν → εἴτε ἐλθὼν · μηδενλ → μηδενί · ἀπωλείσς → ἀπωλείας · ὑμῆν → ὑμῖν · Κπρὸς ἀρετὴν → πρὸς ἀρετὴν (intrusive Κ) · εἰ πεν → εἶπεν · εὐφρανθήσομαιτοῦτο γὰρ προσυπακουστέον) → εὐφρανθήσομαι (τοῦτο γὰρ προσυπακουστέον) (leaf 654) · ἀπών → ἀπὼν · ἁγάπης → ἀγάπης · Μιᾶ ψυχῇ → Μιᾷ ψυχῇ · τῆ ἀγάπῃ → τῇ ἀγάπῃ · ὁνειδίζωσιν → ὀνειδίζωσιν · στιῤῥότητος → στεῤῥότητος (leaf 654) · τὰ ἡμων → τὰ ἡμῶν · ὲπάγει → ἐπάγει · Τλν αὐτὸν → Τὸν αὐτὸν · οἶον ἴδετε → οἷον εἴδετε · ΕΙ/Εἵ/Εἶ τις → Εἴ τις (×3) · έν Χριστῷ → ἐν Χριστῷ · εἰ τις´ σπλάγχνα → εἴ τις σπλάγχνα · Ἰνα → ἵνα · τὸ ἑν βρονοῦντες → τὸ ἓν φρονοῦντες · μηδέν → μηδὲν · ἑαυτῶν μὴ τὰ ἑαυτῶν ✓ · τὸ ἑτέρων → τὰ ἑτέρων (leaf 655) · Bαβαὶ → Βαβαὶ (Latin B for Β, twice) · πρὸς αὸν ἕτερον → πρὸς τὸν ἕτερον · Ἱνα → Ἵνα · Πέγα μὲν γὰρ → Μέγα μὲν γὰρ · ὅτι δ δαὶνά μου → ὅτι ὁ δεῖνά μου (leaf 655) · τοὺτο → τοῦτο.

1279 (chunk 0004), a lost lemma and clause restored. Calfa runs «Μὴ τὰ [1279] ἕκαστος ἀφεὶς, τοῦ συμφέροντος γίνηται τοῆ ἀὲολ φοῦ — the plate (leaf 655) reads «Μὴ τὰ ἑαυτῶν ἕκαστος.» Ὅταν γὰρ τὸ ἴδιον συμφέρον ἕκαστος ἀφεὶς, τοῦ συμφέροντος γίνηται τοῦ ἀδελφοῦ. Restored; the [1279] anchor placed at the lemma's close. 1279. είρήνην → εἰρήνην · ὃ´ τὸ έον ἦν → ὃ πλέον ἦν (leaf 655) · βίος,. → βίος. 1279/1280 (block lemma Phil 2:5–8). Τοῶτο γὰρφρονείσθω κν ὑμῖν → Τοῦτο γὰρ φρονείσθω ἐν ὑμῖν · Χρισιῷ → Χριστῷ · οὐχ ἀρπαγμ ν ήγοόσατο εἶναι σα Θεῷ → οὐχ ἁρπαγμὸν ἡγήσατο τὸ εἶναι ἴσα Θεῷ (the τὸ is confirmed by the short lemma at leaf 656) · ἐκένωσεμρφὴν → ἐκένωσε μορφὴν · εὐρεθεὶς → εὑρεθεὶς · ἐαυτὸν → ἑαυτὸν · σταιροῦ → σταυροῦ. This block lemma and the scholion after it fall in the stretch our scan does not carry (see the siglum note); the restorations rest on the received text, on Migne's own short lemmata further down, and on the Latin twin. 1281. δεδέλωκε → δεδήλωκε · Μαοίωα τὸν Ποντικόν → Μαρκίωνα τὸν Ποντικόν · Καὶ ψὰρ → Καὶ γὰρ · Τν τούτοις → Ἐν τούτοις · λάρκελλος → Μάρκελλος · Σαβελλιον → Σαβέλλιον · ἡγήσατοτὸ εἶναιίθα Θεῷ → ἡγήσατο τὸ εἶναι ἴσα Θεῷ · σον νἀρ → Ἴσον γὰρ · ψιλὰ φής → ψιλὰ φῂς · Εἰποικεν → Εἴποιμεν · Τί ἐστ, → Τί ἐστι, · ἄν´γωπος → ἄνθρωπος (leaf 656) · τῶν δύε → τῶν δύο (leaf 656) · ὰπόδειξις → ἀπόδειξις · Ἢδει → ᾜδει · θείτητος → θεότητος · Ὁ πέντς → Ὁ πένης · ταπεινωδῆναι → ταπεινωθῆναι · Ὡς ἐν μορφῆ Θεοῦ ὑπάρχων → Ὃς ἐν μορφῇ Θεοῦ ὑπάρχων (from the block lemma; not plate-verifiable, see the siglum note) · Πνευματος → Πνεύματος. 1281/1284. ἡδύνατο κἀν → ἠδύνατο κἂν (leaf 656) · ταπεινω.θῆναι → ταπεινωθῆναι · βίόλον → βίβλον · ἴδιον. εὐχερῶς → ἴδιον, εὐχερῶς · ὄταν θέλη → ὅταν θέλῃ · ὀμοίωμα → ὁμοίωμα · Διατί → Διὰ τί · τετράῳθαι → τετράφθαι · οία ανθρωπός → οἷα ἄνθρωπός · ἴδομεν → εἴδομεν · γίρ νεται → γίνεται · Ο δὲ παρὰ ἀνθρώποις ὑψηλοί → Οἱ δὲ… ὑψηλοὶ · ἀνθρώπουούκ → ἀνθρώπου οὐκ · τούτἐστιν → τουτέστιν · τὸ εἰναι Θεὸς → τὸ εἶναι Θεὸς · ᾑδει ἔχον → ᾔδει ἔχων (leaf 657) · συναῖδίου → συναϊδίου · καταβιβασθῆ → καταβιβασθῇ · Μορφὴν δούλουν λαβών → Μορφὴν δούλου λαβών (leaf 657). 1284, a lost clause restored: Calfa runs ἐταπείνωσεν ἑαυτὸν δι᾽ ἡμᾶς, οἱ [1284] ἀλλήλοις ταπεινοῦσθαι — the plate (leaf 656) reads …οἱ φύσει ὁμότιμοι καὶ ὁμοίως ταπεινοὶ οὐκ ὀφείλομεν ἀλλήλοις ταπεινοῦσθαι…;; Latin nos qui natura æquales sumus et æqualiter humiles et abjecti, annon debemus invicem humiliari. Restored as the rhetorical question the plate prints; the negation οὐκ is carried, not smoothed.

1285 (chunk 0005). ἠχερίσατο αὐτῇ → ἐχαρίσατο αὐτῷ · Ἰνα ἐν τῳ → ἵνα ἐν τῷ · σᾶν γόνυ → πᾶν γόνυ · ἐπουρανίον → ἐπουρανίων · Χριστίς → Χριστὸς · μηθαμοῦ → μηδαμοῦ · τῷ Υὸῷ Θεῷ Κντι → τῷ Υἱῷ Θεῷ ὄντι (intrusive Κ, leaf 657) · ἐξωολογήσατο → ἐξωμολογήσατο · μείζων Κοὖν → μείζων οὖν (intrusive Κ) · τοῦ πρὸ τς → τοῦ πρὸ τῆς · ἰνανθρωπήσεως → ἐνανθρωπήσεως · φασιν· → Διό φασιν· (leaf 658) · Ἠ ὅτι → Ἢ ὅτι. 1285/1288. μετὸ φόβου καὶ γρόμου → μετὰ φόβου καὶ τρόμου · κατεργάζισθε → κατεργάζεσθε · λουλεύσατε → Δουλεύσατε · αὐτῶ ἐν τρόμῳ → αὐτῷ ἐν τρόμῳ · Ἀιὰ ταῦτα → διὰ ταῦτα · ἑαοτῶν σιντηρίαν → ἑαυτῶν σωτηρίαν · τὸ. ἀνεργεῖν → τὸ ἐνεργεῖν (leaf 658) · Ἡθέλησε → Ἠθέλησε · Κάῖν → Κάϊν · Ἀπεὲέχθη → ἀπεδέχθη · φἡσι → φησι. 1288, a lost clause restored across the column break: Calfa runs αὐτός ἐστιν ὁ τελειῶν καὶ οἰονεὶ μορΔύναται [1288] δὲ τό- — the plate (leaf 658) reads …καὶ οἰονεὶ μορφοποιῶν τὴν ἡμετέραν καὶ θέλησιν καὶ ἐνέργειαν. Δύναται δὲ τό, Ὁ Θεὸς γάρ ἐστιν ὁ ἐνεργῶν…; Latin ipse qui perficit et veluti formam addit nostræ et voluntati et operationi. Restored; the [1288] anchor placed at the sentence break the plate marks. 1288. χωρ ς γογγυσμων → χωρὶς γογγυσμῶν · συνεργόνς → συνεργόν; · ἐνεργων → ἐνεργῶν · εἰπεν ἐν φόβω → εἶπεν ἐν φόβῳ · τὸ Ηέλειν → τὸ θέλειν · εὐγνωμοπύνν ἃ Παῦλος → εὐγνωμοσύνην ὁ Παῦλος · τῶ Θεῶ → τῷ Θεῷ · συνΚεργεῖ → συνεργεῖ (intrusive Κ) · Ὑπὲρ σῆς εὐδοκίας → Ὑπὲρ τῆς εὐδοκίας · γογγυσμών καὶ διαλοσεσμῶν → γογγυσμῶν καὶ διαλογισμῶν · Ἰνα/Ἰλα → ἵνα/Ἵνα · ἀκέραιο → ἀκέραιοι · ἐν μέσω → ἐν μέσῳ · σκολῶς → σκολιᾶς · φωστηρες → φωστῆρες · κίύχημαΚέμοι → καύχημα ἐμοὶ (intrusive Κ) · ἀχαριατεῖ → ἀχαριστεῖ · εἴλικρινεῖς → εἰλικρινεῖς · ἐνόρετον → ἐνάρετον · Ἐν οἷς ταί νεσθε → Ἐν οἷς φαίνεσθε · σῶν ἁμαρτιῶν → τῶν ἁμαρτιῶν · ἐπι έχοντες → ἐπέχοντες.

Restored question marks (the anchor's class, logged not itemized per site)

Calfa prints · or nothing where the plate prints ;. Restored from sense and the Latin column at: 1265 Πῶς οὖν ἀλλαχοῦ φησιν…εἰρηνεύετε; · 1268 Καὶ τί πρὸς τοῦτο; · 1268 Τί γάρ; · 1272 Τί λέγω χαίρω; · 1272 Πῶς ἐλύετο τὰ δεσμά; (and the two following) · 1273 Τί οὖν φησιν; · 1273 Τί δέ ἐστιν ὁμοίωμα ἀνθρώπου; · 1273 Διὰ τί δὲ εἶπεν ὁμοίωμα ἀνθρώπου; · 1274 Διὰ τί δὲ τοῦτο; · 1288 Θεὸν ἔχων συνεργόν; One in the other direction: at 1284 Calfa prints ; after παρὰ Πατρός in the John 1:14 quotation, where the sentence plainly continues (τουτέστιν, οἷα Μονογενοῦς); rendered as the statement it is. One left standing: at 1279 Calfa prints Ἡ ἄκρα ταπεινοφροσύνη; — rendered as the answer to the preceding question ("The utmost lowliness of mind."), the mark being unsupportable there; the stretch is not plate-verifiable.

Latin-twin pass (pilot §6) — result per chunk, including the clean ones

Read against src/pg-latin/oecumenius-philippians/0001.md–0005.md.

Chunks where the deliberate twin pass found nothing to mark: 0003 and 0005. Both were read in full against their twins and both are named here so that a pass made can be told from a pass not made.

Polarity read (runbook 4a), our own defects: every negation Migne prints is carried, including where it makes the sentence fight itself. The two sites where that was live are recorded above and neither was smoothed: the ἑκὼν at 1284 ("by necessity and the Father's command he willingly became man"), and the restored οὐκ ὀφείλομεν at 1284, which the English keeps as the rhetorical question the plate prints rather than flattening to a statement.

Deferred to the next agent

• Chunk 0005 ends mid-gloss on Phil 2:16 (Γινωσκόμενοι ὅτι ὑμεῖς τῶν ζησομένων ἐστὲ τὴν ἀθάνατον ζωήν); the Latin twin's tail already carries the antithesis that completes it (nam illi ex illis sunt qui morte immortali morientur) and the «Ut gloriari possim» lemma that opens chunk 0006. Begin from this batch's block-lemma renderings of Phil 2:14–16.
• The scan gap should be recorded work-wide: the Greek column for Migne cols 1279–1281a is not present in patrologiaecurs14migngoog. Any later sweep that wants a plate check in that span needs a second scan item (the twin's own provenance block already names patrologicursus21migngoog as an alternate witness used for one Latin page).

Batch: 0006, 0007, 0008, 0009. Anchor followed: 0000.md pair + cruces-0000.md work-wide conventions (binding). Greek plate witness used throughout: raw/scans/pg118/patrologiaecurs14migngoog_djvu.xml, leaves 657–666. Note the scan's leaf gap in this stretch: the primary item carries cols 1289/1290 (leaf 659), 1293/1294 (660), 1297/1298 (661), 1301/1302 (662), 1303/1304 (663), 1305/1306 (664), 1307/1308 (665) — and has no leaf for cols 1291–1292, 1295–1296, 1299–1300 (the same gap the twin's provenance records by falling back to patrologicursus21migngoog for those Latin pages, an item we do not hold locally). Every "unverified" note below means that gap, not negligence.

Conventions decided fresh in this batch (for the merge)

• καύχ- family is kept on one English root: "boast." καύχημα = boast · καυχάομαι = boast · καύχησις = boasting. The gloss at 1297 argues from the root (τῶν ἐν Χριστῷ καυχωμένων, against confidence in the flesh), so AV's "rejoice"/RSV's "glory" would break the argument it carries. Not varied.
• περιτομή = "circumcision" · κατατομή = "concision" · κατατέμνειν = "to cut in pieces." The gloss at 1297 IS the etymology (οὐδὲν ἕτερόν ἐστιν ἡ περιτομὴ, ἢ σώματος κατατομή); the English pair must share the "-cision" element, as Migne's own Latin does (circumcisio / concisio).
• The loss/gain vocabulary of Phil 3:7–9, fixed: κέρδος = gain · ζημία = loss · ζημιόομαι = suffer loss · σκύβαλα = refuse. The whole of 1300–1304 is a dispute over these four words (heretics: "the law is loss and refuse"), so one English word each, never varied.
• σπένδομαι = "I am poured out as a libation" · σπονδή = libation · θυσία = sacrifice · λειτουργία = ministration · λειτουργός = minister. λειτουργία is fixed across both its senses in this range (2:17 the faith offered; 2:30 the Philippians' service to Paul) because Oecumenius plays them against each other.
• διώκω = "pursue" · καταλαμβάνω = "lay hold" · βραβεῖον = prize · σκοπός = mark (Phil 3:12–14). AV's "follow after / apprehend / press toward" was refused: the gloss at 1305–1308 turns on the runner image and on the pair διώκω / εἰ καταλάβω being visibly the same two words each time.
• ἰσόψυχος = "like-souled," ἰσοψυχία = "like-souledness" (1292). AV's "likeminded" hides the gloss's own derivation.
• εὔφημος / δύσφημος = "of good omen" / "of ill omen" (1301), where the argument is precisely that Paul chose an ill-omened word for a good thing.
• ἔνι = "it is possible" — continued from the anchor (0000, Ἔνι γὰρ), recurring at 1293 and 1300.
• Ἢ opening an alternative construal = "Or:"; τουτέστι/ἤγουν = "that is" — anchor formulas, unchanged.
• Em-dash separators inside a scholion. Where the plate opens a fresh alternative construal inside one father's scholion it prints a bare em-dash with no siglum (— Καὶ ἄλλως δέ, — Δυνατὸν δὲ καὶ οὕτως, — Ἢ καὶ οὕτως; scan-verified leaf 659, and the Latin column prints the same dashes: — Praeterea et alio modo, — Potest autem et hoc modo dici). These are reproduced as em-dashes and are not sigla; do not bracket them.

Catena sigla met in this batch

[lat: …] markers fired (4), with plate status

1. 1298 (chunk 0008 opening) — Greek φοβούμενος μὴ ἀναισχυντοτέρους ποιήσῃ ("fearing lest he make them more shameless"); Latin veritus ne nimium pudore ipsos afficeret ("lest he affect them with too much shame"). Opposite motives for the same silence. Greek plate-verified, leaf 661: ἀνωνύμως αὐτῶν κχάθάπτεται, φοθούµενος μὴ ἀνα.σγυντοτέρόνς ποιήσῃ. Marker fired.
2. 1297 (chunk 0007) — Greek ἐπειδὴ οὐκ ἔτι νομίμως ἐγίνετο παρὰ Ἰουδαίοις ἡ περιτομή ("since circumcision was no longer performed lawfully"); Latin quoniam nondum facta erat apud Judaeos vera et legitima circumcisio ("not yet … a true and lawful circumcision"). A temporal polarity flip, and it reverses the argument's direction (types ceasing at the truth's arrival vs. a lawful circumcision never yet reached). Greek plate-verified, leaf 661: Αλλ᾽ ἐπειδὴ οὐκ ἔτι νομίμως ἐγίνετο παρὰ Ἰουδαίοις ὁ περιτομή. Marker fired.
3. 1304/1305 (chunk 0009, block lemma Phil 3:12) — Greek Οὐχ ὅτι ἤδη ἔλαβον ἢ ἤδη τελείωμαι; Latin Non quod jam jam apprehenderim aut jam mortuus sim ("or have already died"). Both columns plate-verified, leaf 663 (Greek: O2, ὅτι Άδη ἔλαθον 7, Ίδη τελείωμαι; Latin: jam jam apprehende- rim au(. jam mortuus sim). Marker fired. See also the scripture list below: the plate prints τελείωμαι, not the received τετελείωμαι, and the gloss's own paraphrase (ἢ ἤδη ἐκ τούτου τέλειός εἰμι) reads with "perfect," not with "dead" — so the Greek column is coherent and the Latin is the outlier.
4. 1308 (chunk 0009) — Greek Τὸν σύντομον δηλοῖ δρόμον τὸ, ἐπεκτεινόμενος ("the swift course"); Latin Continuum cursum significat dictio Extendor. Greek plate-verified, leaf 665:* σύντομον ὀχλοῖ Ἀρόμον τὸ, ἐπεκτεινόμενος. Marker fired — the whole point of the gloss is what ἐπεκτεινόμενος shows*, and speed and continuity are not the same showing.

Candidates NOT marked (evidence short of the bar) — logged deliberately

• 1292 — the pronoun-itacism class, the anchor's own warned case. Greek οὔτως ὑμῶν ἀντιλαμβανόμενον ἢ μεριμνῶντα τὰ καθ᾽ ἡμᾶς; Latin qui ita vos adjuvet et curet quae vestra sunt. Identical to the anchor's 1261/1262 site in kind — but col 1292 is in the leaf gap and the Greek could not be plate-verified, so per the standing rule the candidate is logged and no marker fired. Rendered from the printed Greek ("the things that concern us").
• 1300 — Ἑβραῖος / Ἰουδαῖος. Greek ἔνι γὰρ εἶναι ἐκ γένους Ἰσραὴλ, καὶ μὴ Ἑβραῖον; Latin Contingit enim ex genere Israel esse et non esse Judaeum. The gloss exists precisely to distinguish Hebrew (= knowing the Hebrew tongue) from Israelite, so the Latin's Judaeum flattens it. Not marked: unverifiable leaf gap, and the immediately preceding word is Calfa damage (εἰμὶ,νο γὰρ for εἰμί, ἔνι γὰρ), so the region is not clean enough to carry a divergence claim.
• 1300 — κατὰ Χριστόν / secundum Deum. Greek (as restored) πρὸς σύγκρισιν τελειοτέρας τῆς κατὰ Χριστὸν γνώσεως; Latin ad comparationem perfectioris secundum Deum cognitionis. Same leaf gap; the Greek noun itself had to be restored from the Latin (below), so no marker.
• 1308 — ἐν Χριστῷ / per Christum. Greek ἐν Χριστῷ γὰρ, φησί, τρέχω (plate-verified, leaf 665); Latin per Christum, ait, curro, and the lemma heading likewise Per Christum Jesum. Not marked: Allatius renders ἐν instrumentally as a matter of course, and this is his looseness, not a different fact. Logged because the gloss does lean on the preposition.
• 1290 — construal prepositions in Photius's alternative. Greek Ἀλλ᾽ εἰ καὶ θύομαι διὰ τὴν θυσίαν καὶ λειτουργίαν (plate-verified, leaf 659); Latin Quinetiam si immolor super hostia sacrificioque fidei vestrae. The Latin has silently pulled the paraphrase back to the Vulgate's own preposition (immolor supra sacrificium), which is the anchor's Vulgate-conformation class — never a [lat:]. Logged as notable because the conformation erases the very construal Photius is offering.
• 1294 — τοῦ Κυρίου / Domini Jesu; 1302 — Deut 18:15 προφήτην ὑμῖν ἀναστήσει / Prophetam suscitabit vobis Deus.** Both are the Latin supplying a name the Greek leaves implicit, the second under Vulgate pressure. Conformation/expansion class; no markers.

[ed: …] markers (4) — demonstrated holes in our source

1. 1293/1296 (chunk 0007) — Calfa breaks off mid-clause: ἵνα μὴ λύπην τὴν τῆς ἀῤῥωστίας αὐτοῦ, ἐπὶ λύπην τὴν τοῦ + [next lemma]. The plate carries the continuation — scan leaf 660: ἐπὶ λύπη, τὴν τοῦ θανάτου χκρατύσυ (θανάτου certain; the verb's letters are not recoverable from this OCR, and the Latin twin's ne dolorem qui fuerat ex ejus aegritudine, haberem una cum dolore qui sequeretur ex ejus morte fixes the sense but not the word). Per Pattern 13 the lost words are not supplied on the page.
2. 1298 (chunk 0007) — Calfa: Διὸ οὐ ψέγω αὐτοὺς ὡς ἀπεστερημένος τῆς + [next lemma]. Plate carries it, scan leaf 661: Διὸ οὗ ψέγω αὐτοὺς ώς ἀπεστερτμένος τῆς κατὰ σαρκα εὐγενείας. Not supplied on the page.
3. 1300 (chunk 0008) — a whole step of the objection is missing between διὰ τὸ ὑπερβάλλον τῆς γνώσεως and οὐκ ἔστι ζημίαν ἀποβαλεῖν. The Latin column carries it: propter excellentiam cognitionis Christi: quomodo statim dicis, Propter quem omnium jacturam feci. Nam dicere Jacturam feci, non est damnum abjicere. Col 1300 is in the leaf gap, so the loss is demonstrated by the parallel column rather than by the Greek plate; the words are not supplied.
4. 1296 ×2 — the two illegible sigla, above.

Ladder row 1 — Calfa OCR damage corrected silently, witness cited

Chunk 0006 (1288–1293). Ὡρᾷς → Ὁρᾷς · ἀμοιβὰε → ἀμοιβὰς · Είς → Εἰς · ὀμῖν → ὑμῖν · ρὐκ → οὐκ · ἀποθιήσκω → ἀποθνήσκω · μάλλω → μέλλω (Lat. oblaturus) · Χριστῷτὸ → Χριστῷ· τὸ (joined type) · Κφησίν, Κτῇ, λειτουρΚγίαν → intrusive-Κ class, deleted (the anchor's PG 118 instance of the PG 139 artifact) · ἈΙΓ εἰ καὶ σαένδομαι → Ἀλλ᾽ εἰ καὶ σπένδομαι (block lemma; scan leaf 659) · οΑλλ´ → «Ἀλλ᾽ (guillemet, scan leaf 659) · αὐδὲ τοῦτο ἄτοπον → οὐδὲ τοῦτο ἄτοπον — negation-critical, scan leaf 659 concurs (Kal φαμιν, ὅτι οὐδὲ τοῦτο ἄτοπον) · ὡ- ἀληθῶς → ὡς ἀληθῶς · Δυνατὸν δὲ καὶ ρῶτως → Δυνατὸν δὲ καὶ οὕτως (leaf 659) · θλίψεσιδεσμὰ → θλίψεσι (δεσμὰ (leaf 659 prints the parenthesis) · οὐχ´ ὁμᾶς → οὐχ ὑμᾶς (leaf 659) · Ἐπὶ τῇ θυσίᾳ.» → «Ἐπὶ τῇ θυσίᾳ.» (lost opening guillemet, leaf 659) · ὐτούς → αὐτούς · Τομόθεον → Τιμόθεον · Ἰνα/να → ἵνα · γὸρ → γὰρ · ητοῦσιν → ζητοῦσιν · δτι → ὅτι · σὺν ἐμοῖ ἑδρύλευσεν → σὺν ἐμοὶ ἐδούλευσεν · ἐξ » αὐτῆς → ἐξ αὐτῆς (the stray » is Calfa noise; block lemmata are unguilleted) · ΚΚαὶ → Καὶ · ζηΚτοῦσι → ζητοῦσι · ἄχρις θμῶν → ἄχρις ὑμῶν (Lat. usque ad vos) · φησὶί → φησί · εἰκότες → εἰκότως (Lat. Merito) · ἵυα → ἵνα · ὡφεληθῶσιν → ὠφεληθῶσιν · Τνα → ἵνα · πέμύαι → πέμψαι · ᾤν → ἦν · νσθένησε → ἠσθένησε · ἀλα → ἀλλ᾽ · λόπην ἐπὶ λύπηνσχῶ → λύπην ἐπὶ λύπην σχῶ · Ἔνι Κγὰρ → Ἔνι γὰρ · Δι´ Καὐτοῦ → Δι᾽ αὐτοῦ (leaf 660) · Λιὸ → Διὸ (leaf 660) · Ἡδημόνει → Ἠδημόνει · διὰ σὸν Κύριον → διὰ τὸν Κύριον (leaf 660) · δι´ ὁμᾶς → δι᾽ ὑμᾶς (leaf 660).

Chunk 0007 (1293–1298). ὅλωςτοῦτο → ὅλως (τοῦτο and ἀναλυσαι → ἀναλῦσαι, καλὸν) — parenthesis restored, leaf 660 · δι´ ἡν → δι᾽ ἣν · ἱνα → Ἵνα · Καὐτοῦ → αὐτοῦ · ἔπεχμψα → ἔπεμψα (leaf 660) · Χαρᾶς → χαρᾶς · ἧγισε → ἤγγισε (leaf 660) · Κοὐκ → οὐκ · ἀλιπότερος → ἀλυπότερος · τῆς ὡλτῶν. μαθητῶν → τῆς ὑπὲρ τῶν μαθητῶν (leaf 660 gives τῆς ὑπὲ2 τῶν μαθιτῶν) · πυλλῷ → πολλῷ (leaf 660) · τοισῦτον → τοιοῦτον · Λιὰ → Διὰ · ἱμαθητάς → μαθητάς · μαθηταις → μαθηταῖς · Ἀσι ἐντολῆς ἐστι Χριστοῦ → read as ὅτι ἐντολῆς ἐστι Χριστοῦ (Lat. quia praecepti Christi est) — the letters Ἀσι are not a word; rendered by the Latin-attested sense, flagged here because the conjunction is a guess in form though not in sense · κἱνδυνον → κίνδυνον · ὑστέρισε → ὑστέρησε · δεικνὸς → δεικνὺς · Πράφειν → γράφειν · σὐκ → οὐκ · ὑμῶ δὲ ἀσφαλές → ὑμῖν δὲ ἀσφαλές · κακοὸς → κακοὺς · οἱπνεύματι → οἱ πνεύματι · πεποιθησιν → πεποίθησιν · Ἐὐαγγέλιον → Εὐαγγέλιον · ἀπ´ αὐτῶ → ἐπ᾽ αὐτῷ · καὶ ὰείαν Κυρίου → καὶ ἀξίαν Κυρίου (Lat. dignoque Domino; ladder row 1, sense-certain) · Ἢσαν → Ἦσαν · οθς → οὓς · ἀλέγοντο → ἐλέγοντο · κύνεςο γὰρ ΔΑΡΟΤ v. → κύνες (οἱ γὰρ (bleed excised, parenthesis restored; leaf 661 gives the parenthesis) · ἐξέρασμα kept (attested) · ἡ περιτομή γὰρ νομίμη → ἡ περιτομή (ἡ γὰρ νομίμη (leaf 661) · σαρκίς → σαρκί; · αὕτη·Τοῦτο → αὕτη· (Τοῦτο (leaf 661) · Ἰοραὴλ → Ἰσραήλ · Εενιαμίν → Βενιαμίν · Ἐβραῖος ἐξ Ἐβραίων → Ἑβραῖος ἐξ Ἑβραίων · Κἄλεμπτος → ἄμεμπτος (leaf 661) · ὄσην → ὅσην.

Chunk 0008 (1297–1304). Ἄλνως → Ἄλλως (leaf 661) · Κάλιστα → μάλιστα · προσηλιῆτων → προσηλύτων · εἰμὶ,νο γὰρ → εἰμί, ἔνι γὰρ (Lat. Contingit enim) · ρὐκ ἡκριβῶντο → οὐκ ἠκριβοῦντο · Ἐκκληαίαν → Ἐκκλησίαν · ἔμελλὶ → ἔμελλε · ζημίάν → ζημίαν · τῆς γνώσεως ο ῶ Ἰησοῦ → τῆς γνώσεως Χριστοῦ Ἰησοῦ (Lat. cognitionis Christi Jesu) · ἐημιώθην → ἐζημιώθην · ἐν ὐτῳ → ἐν αὐτῷ · Κἐμὴν → ἐμὴν · Χριστου → Χριστοῦ · τὰ ἁπὸ τοῦνόμου → τὰ ἀπὸ τοῦ νόμου · Μαρκωνος → Μαρκίωνος (Lat. Marcionistae) · ὁὅπερ → ὅπερ · πραγμάτων λυσιτελῶν → πραγμάτων ἀλυσιτελῶν — ladder row 1, and the one silent correction in this batch that changes the sense: as printed, the sentence says casting away profitable things, which contradicts its own neighbour (τοῦτο γὰρ μᾶλλον κερδῆσαί ἐστι, τὸ ἐλευθερωθῆναι ζημίας); the Latin has quarumdam vilium rerum abjectionem, and a dropped initial alpha is the commonest Calfa loss. Col 1300 is in the leaf gap, so this is sense-evidence, not letter-evidence — no data/calfa-patches/ entry proposed · δημίαν → ζημίαν · Ὃτι ἐξημιώθην → Ὅτι ἐζημιώθην · τῆς κατὰ ΚΧριστὸνέλοα- → τῆς κατὰ Χριστὸν γνώσεως (noun supplied from Lat. cognitionis; see the unmarked-candidate list for secundum Deum) · πρὸς σύν κρισιν → πρὸς σύγκρισιν · Ὥγημαι → Ἥγημαι · μολίβδον → μολίβδου · ὁ στος → ὁ σῖτος · ἐνηνθρωπήκει kept · Ἱδει → Ἤδει · Ἢγημαι → Ἥγημαι · Κζημία → ζημία (intrusive Κ) · ἀποκαλόνει (see 0009).

Chunk 0009 (1304–1308). ἐακὼς → ἐάσας · δἷον ἐξέργων → οἷον ἐξ ἔργων · κατὰνόμον → κατὰ νόμον · Τ γὰρ καὶ ποιήσει → Τί γὰρ καὶ ποιήσει · Οὐχ ἄτι ἥδη → Οὐχ ὅτι ἤδη (leaf 663) · ὑκὸ → ὑπὸ · τὰ μν ὀπίσω → τὰ μὲν ὀπίσω · ἐπιλανβανόμενος → ἐπιλανθανόμενος · ἐπεκτεινόμανος → ἐπεκτεινόμενος · κΙήσεως → κλήσεως · xριστοῦ, Σριστοῦ → Χριστοῦ (Latin-alphabet substitution class) · Ἄτις → Ἥτις · ὑπλρ ἀΟρώπων → ὑπὲρ ἀνθρώπων (leaf 663/664) · ἀνΚδρώπων → ἀνθρώπων · ἤόη → ἤδη · ὄραβεῖον → βραβεῖον · κααλάβω → καταλάβω · ἀρὰ → ἄρα · ἔμΚπροσθεν → ἔμπροσθεν · ἐμαυΚτόν → ἐμαυτόν · δροΚμέως → δρομέως · Ἃμα → Ἅμα · ἐπκτείνεσθαι → ἐπεκτείνεσθαι · πολύ [μου ἀπέχον → πολύ μου ἀπέχον (stray bracket) · σοι οὖν τέλειον → Ὅσοι οὖν τέλειοι (leaf 665) · εἶ τι έτέρως → εἴ τι ἑτέρως · ἀποκαλόνει → ἀποκαλύψει (leaf 665) · ἃποκαλύθει → ἀποκαλύψει · οὐ´ καλῶς → οὐ καλῶς (negation intact in both witnesses) · «Πλὴν εἰς ἃ ἐφθάσαμεν.» → «Πλὴν εἰς ὃ ἐφθάσαμεν.» (leaf 665 prints the singular in the gloss lemma as in the block lemma — Calfa's plural would have manufactured an internal lemma divergence that the plate does not have) · ὸπίσω → ὀπίσω · ἑπεκτείνεσθαι → ἐπεκτείνεσθαι · τουνέστι/τουτέστι normalized.

Restored question marks (Calfa prints · or nothing for the plate's ;)

Logged as a class, per the anchor. 0006: Ὁρᾷς… τὰς ἀμοιβὰς τῆς ἀρετῆς; · Ποῖον καύχημα ἔχεις, ὦ μακαρία ψυχή; · πῶς οὖν ἀπέστειλεν ἂν αὐτόν; (leaf 660 prints ;) · οἱ λέγοντες κακὸν τὸν κόσμον; (leaf 660). 0007: ἔλεον λέγει τὴν ἐνταῦθα ζωήν; (leaf 660) · πόσῳ μᾶλλον ὑπὲρ διδασκάλων — rendered with ! after the Latin's own quanto magis pro praeceptoribus ! · Χριστῷ πεποιθέναι, ἢ σαρκί; (leaf 661). 0008: πῶς ἂν διὰ φιλαρχίαν ἔμελλε διώκειν, καὶ οὐ κατὰ ζῆλον Θεοῦ; · Πῶς οὖν φατε ὑπὸ Θεοῦ αὐτὸν δεδόσθαι; (leaf 662) · Διὰ τί δὲ ὅλως ἥγησαι αὐτὸν ζημίαν καὶ σκύβαλα, ὦ Παῦλε; (leaf 663). 0009: ποῖος παραστήσει λόγος; (leaf 663) · Τί δὲ οὐ λογίζῃ κατειληφέναι; · τί ἡμεῖς ἐροῦμεν;

Cross-column bleed (Calfa artifact class — Latin misread in Greek glyphs)

Not translated, not carried; the Greek column's own text runs on through the intrusion, and the column anchors inside it are kept in place.

• 1292 (0006) — ΡΑτμο. ΟΚ. ΟΧνΙii. [1292] οοΟμαιόειο Πλλί. ΟΑΡΒΤ Γν. intruding into Ἐπὶ μὲν τῇ … πίστει ὑμῶν χαίρετε. Leaf 659 shows the source: the page-foot signature PATROL. GR. CXVIII. plus the facing running title OECUMENII PHIL. CAPUT IV. The [1292] anchor is real and stays.
• 1296 (0007) — ἀμ κοοῶνΙἱαιο εμωι, μὲ λίε πίιεαΒι ἱ εη ρετῖσυΙίε. Κοο ε αἰε ἰρεοε Ἰη οΠοίο οοο. Δο οσἰεεπε ὡἱ ο ουωεἰ οοΠρὶοαιΓ. — the tail of the Latin column (its own …ostendens ut quod debitum est compleatur) bled into the Greek. Col 1296 is in the leaf gap; identified by shape and by the Latin twin, not by the plate.
• 1297 (0007) — ΔΑΡΟΤ v. inside Νῦν δὲ οὗτοι κύνες (οἱ γὰρ.
• 1302 (0008) — ομε ΠΡῖίε ενι. between ἀλλὰ τῇ τοῦ βελτίονος ἀποστερήσει. and Τοίνυν καὶ ὁ νόμος — leaf 662 shows Migne's footnote line (I Cor. I, 22, 29. Rom. v, 14) and the Latin column at exactly this point.

Guillemets

Migne's « » reproduced 1:1 except where Calfa lost an opening mark, which is restored (Pattern 6: never omitted where he has one). Counts, English vs Greek chunk: 0007 40/40, 0008 22/22, 0009 55/55, 0006 36/35 — the 0006 excess is three source defects netted: two lost openings restored (οΑλλ᾽ εἰ καὶ σπένδομαι,» and Ἐπὶ τῇ θυσίᾳ.», both scan-verified at leaf 659) and one stray » not reproduced (ἐξ » αὐτῆς, inside an unguilleted block lemma — Calfa noise). One genuinely unmatched » is reproduced, at 1305: «Συμμορφούμενος.» Τουτέστιν ἐξομοιούμενος.» — both Greek witnesses print it (Calfa and leaf 663: Τουτέστιν ἐξ- ομοιούµενος. »), so it is the plate's own defect and stands.

7a″ — where Migne's printed Greek diverges from the received text

This is the batch's primary deliverable. Every item below was rendered as Migne prints it, and every re-quotation in the glosses was checked against the block lemma word by word.

1. Phil 2:16, gloss re-quote at 1288 — block: οὐκ εἰς κενὸν ἔδραμον οὐδὲ εἰς κενὸν ἐκοπίασα; Migne's gloss re-quote drops the second εἰς κενόν: Ὅτι οὐκ εἰς κενὸν, φησίν, ἔδραμον, οὐδὲ ἐκοπίασα. Rendered "did not run in vain, he says, nor labor" — the second "in vain" is not supplied.
2. Phil 2:22 at 1292 — Migne prints Τὴν δοκιμὴν αὐτοῦ γινώσκετε with no δέ; the received text has τὴν δὲ δοκιμήν, and the Latin column restores the connective (Porro probationem ejus nostis). Rendered without it.
3. Phil 2:26–27 at 1292 — the batch's most consequential lemma defect. Migne's block lemma as our source prints it runs …διότι ἠκούσατε ὅτι ἠσθένησε παραπλήσιον θανάτῳ, ἀλλ᾽ ὁ Θεὸς αὐτὸν ἠλέησεν, i.e. the clause καὶ γὰρ ἠσθένησεν is absent, its neighbour absorbed into the ὅτι-clause. But Oecumenius's own gloss below quotes «Καὶ γὰρ ἠσθένησε.» as a lemma, and the Latin column prints Siquidem ita infirmus fuit, ut vicinus esset morti — so both the commentary and the facing column witness a clause the Greek block lacks. Rendered exactly as printed (Pattern 7); not repaired, and deliberately not given an [ed:], because col 1292 is in the leaf gap and I cannot demonstrate what the Greek column of the plate prints. This is the single item in the batch most worth a plate check. Also at 2:27: Migne ἀλλ᾽ ὁ Θεὸς αὐτὸν ἠλέησεν vs received ἠλέησεν αὐτόν (order only).
4. Phil 2:30 at 1294–1296 — block and gloss disagree inside Migne himself. Block: ἵνα πληρώσῃ τὸ ὑμῶν ὑστέρημα; the gloss lemma four sentences later: «Ἵνα ἀναπληρώσῃ τὸ ὑμῶν ὑστέρημα.» Rendered distinctly ("fill" / "fill up"), the difference preserved rather than harmonized.
5. Phil 2:30 at 1294 — Migne prints παραβουλευσάμενος τῇ ψυχῇ (and the gloss's verb παρεβουλεύσατο agrees) where the received text has παραβολευσάμενος. Rendered "hazarding his life" / "he took the hazard", keeping the two on one root as Migne's Greek does.
6. Phil 3:3 at 1296 — Migne prints οἱ πνεύματι Θεῷ λατρεύοντες (dative), with the gloss Οἱ πνευματικῶς τῷ Θεῷ λατρεύοντες and the Latin qui spiritu Deum colimus concurring, against the critical text's θεοῦ. Rendered "who serve God in spirit."
7. Phil 3:4 at 1296 — the block lemma drops a καί its own gloss keeps. Block: καίπερ ἐγὼ ἔχων πεποίθησιν ἐν σαρκί; the gloss: Καίπερ ἔχων πεποίθησιν καὶ ἐν σαρκί, with a short lemma «Καὶ ἐν σαρκί.» Both rendered as printed, so the English shows the same seam.
8. Phil 3:5 at 1298 — Migne prints κατὰ ζῆλον (accusative) for the received κατὰ ζῆλος.
9. Phil 3:8 at 1300 — Migne prints Ἀλλὰ μὲν οὖν καὶ ἡγοῦμαι for the received ἀλλὰ μενοῦνγε, and ἡγοῦμαι σκύβαλα εἶναι (the received text has no εἶναι) — the added infinitive is then re-quoted in the gloss (Ταῦτα καὶ περὶ τοῦ, Ἡγοῦμαι σκύβαλα εἶναι), so it is the commentary's own text. Rendered "count them to be refuse" in both places.
10. Phil 3:11 at 1305 — Migne prints εἰς τὴν ἐξανάστασιν τῶν νεκρῶν for the received τὴν ἐξανάστασιν τὴν ἐκ νεκρῶν. (English cannot show the ἐξ- prefix; "the resurrection of the dead" carries both, as the Latin's resurrectionem mortuorum does.)
11. Phil 3:12 at 1305 — plate-verified. Migne prints ἢ ἤδη τελείωμαι, not the received τετελείωμαι: leaf 663 concurs with Calfa, so this is the plate, not our file. Rendered "or am already perfected"; see [lat:] no. 3 for the Latin's aut jam mortuus sim.
12. Phil 3:13 at 1305 — Migne prints ἐγὼ ἐμαυτὸν οὔπω λογίζομαι κατειληφέναι (received οὐ λογίζομαι), and the gloss lemma «Οὔπω λογίζομαι.» depends on it. Rendered "I do not yet reckon."
13. Phil 3:16 at 1308 — Migne prints the long form τῷ αὐτῷ στοιχεῖν κανόνι, τὸ αὐτὸ φρονεῖν (leaf 665 concurs) and, at 3:15, the indicative τοῦτο φρονοῦμεν where the received text has the hortatory φρονῶμεν (Latin hoc sentiamus). Rendered "we are of this mind."
14. Quotations from elsewhere, as Migne prints them: 1 Cor 15:31 Καθ᾽ ἡμέραν ἀποθνήσκω (agrees) · Matt 10:40 Ὁ γὰρ δεχόμενος ὑμᾶς, ἐμὲ δέχεται (agrees) · 1 Thess 4:17 Καὶ πάντοτε σὺν Κυρίῳ ἐσόμεθα — Migne's Καί for the received καὶ οὕτως · 2 Tim 4:7 Τὸν ἀγῶνα ἠγώνισμαι, καὶ τὸν δρόμον τετέλεκα — without τὸν καλόν, and with a καί the received text lacks; rendered "I have fought the fight, and have finished the course," resisting AV's "I have fought a good fight" · 1 Cor 9:26 and 9:24 agree · Deut 18:15 προφήτην ὑμῖν ἀναστήσει ὡς ἐμέ — heavily abridged against the LXX, and with no expressed subject (the Latin supplies Deus); rendered as printed · John 14:9 is quoted in paraphrase, not verbatim (Ὁ γὰρ τὸν Υἱὸν γινώσκων, γινώσκει τὸν Πατέρα, καὶ ἔμπαλιν), and stands as printed inside Migne's guillemets.
15. Phil 1:24, re-quoted twice inside this batch in two different forms — at 1290 τὸ ἐπιμένειν αὐτὸν τῇ σαρκί, at 1293 Ἀναγκαιότερον δὲ τὸ ἐπιμεῖναι τῇ σαρκὶ δι᾽ ὑμᾶς (aorist, and no ἐν, against the received τὸ δὲ ἐπιμένειν ἐν τῇ σαρκί). English renders both "to remain in the flesh"; the Greek difference is recorded here because the English cannot show it.

The Christ-hymn (Phil 2:5–11) does NOT fall in this batch. Chunk 0006 opens at Phil 2:16; ἐν μορφῇ Θεοῦ / ἐκένωσεν / μορφὴν δούλου lie in the peer range (the scan shows μορφὴν Θεοῦ on leaf 656, i.e. cols 1281–1282, chunks 0004–0005). No μορφή / ἁρπαγμός / κένωσις term table is owed by this batch. The one adjacent term this batch does fix is συμμορφούμενος τῷ θανάτῳ = "being conformed to his death" (Phil 3:10, glossed ἐξομοιούμενος = "being made like") — the peer holding 2:5–11 should be told that conformed is taken here, so that μορφή's English is chosen with that in view.

Latin-twin pass (pilot §6) — per-chunk result

Read against src/pg-latin/oecumenius-philippians/0006–0009.md (Latin cols 1290–1307).

• 0006 — findings, not clean. Two Vulgate/lemma conformations logged (Photius's super hostia; Porro probationem), one unverifiable pronoun-itacism candidate (ἡμᾶς / vestra, 1292), the Phil 2:26–27 clause gap, one Latin name-expansion.
• 0007 — findings, not clean. One [lat:] fired (οὐκ ἔτι / nondum, plate-verified), two [ed:] for lost sigla, one [ed:] for a lost clause the plate carries, one silent correction made on the Latin's evidence (ἀξίαν Κυρίου).
• 0008 — findings, not clean. One [lat:] fired (ἀναισχυντοτέρους / nimium pudore, plate-verified), one sense-changing silent correction on the Latin's evidence (ἀλυσιτελῶν), one [ed:] for a dropped step of the objection, two unverifiable divergence candidates logged unmarked.
• 0009 — findings, not clean. Two [lat:] fired, both with the Greek plate-verified (mortuus sim; continuum cursum), one unmarked candidate (ἐν / per Christum), and one two-witness concurrence rendered literally (below).

No chunk in this batch came back clean. Stating that plainly rather than naming a clean chunk that does not exist: 0006, 0007, 0008 and 0009 were each read against their twin after drafting, and each yielded something.

Ladder row 3 — Greek and Latin concur on an odd reading (rendered literally)

• 1305 ἐν ᾧ πράγματι ὑποκατελήφθην ὑπὸ Χριστοῦ — an unusual compound where the lemma and the next sentence both have simple κατελήφθην. Both Greek witnesses carry the ὑπο- (Calfa; leaf 663 ἐν ᾧ πράγματι ὑποκατελύσθην ὑπὸ Χριστοῦ), and the Latin agrees in sense (in ea re qua apprehensus sum a Christo). Rendered "in that matter for which I was taken hold of by Christ" — attributable to the plate pair, not to our file.
• 1300 Ἄλλως γὰρ κατῃσχύνοντο (leaf 661 concurs) where the Latin has Maxime enim pudore suffundebantur; rendered from the Greek ("For otherwise they were put to shame"), the Latin noted as free rendering rather than a divergence.

Open for the merge / a later plate session

1. Cols 1291–1292, 1295–1296, 1299–1300 have no leaf in our PG 118 scan. Four of this batch's judgments hang on that gap: the two illegible sigla, the [ΤΟΥ ΑΥΤΟΥ] form, the Phil 2:26–27 clause, and the ἡμᾶς/ὑμᾶς candidate. The twin's provenance names the alt item (patrologicursus21migngoog, leaves 615/617/619) — acquiring it would settle all four.
2. [ΤΟΥ ΑΥΤΟΥ] is a new siglum for this work and should be added to the work-wide conventions list once verified; it means the preceding father ([ΦΩΤ.]) continues, so the merged file should keep the two adjacent.
3. [ΟΙΚΟΥΜΕΝΙΟΥ] (1303) is the author's own siglum — its presence tells the reader the surrounding matter is catena, not Oecumenius throughout. Worth a line in the work's "On this text" prose.

Range: cols 1309–1325 (Phil 3:16 – 4:23 + subscription). Anchor conventions (cruces-0000.md) followed throughout; only items decided fresh in this range are marked ⚑ below, for the merge session.

Scan witness used: raw/scans/pg118/patrologiaecurs14migngoog_djvu.xml, leaves 665, 666, 667, 668, 669, 670, 671, 672 (= Calfa pp. 666–668, 670, 672–674). Latin twins: src/pg-latin/oecumenius-philippians/0010–0013.md.

0. TWO SCAN GAPS IN THIS RANGE — read before trusting anything below

1. Calfa p. 669 (Greek cols 1315–1316 / Latin 1316–1317) — no leaf in either scanned copy; documented in the twin manifest as gapPages: [669], and the twin file 0011 prints [gap: no scan leaf for Calfa p.669] between [1315] and [1318]. For everything in chunk 0011 between "Great are the praises" and "…overshadows [1317]" there is NO verifier of any kind — no Latin column, no plate. Per the attribution ladder's bottom row this is no attribution: the Greek is rendered exactly as our source prints it, no reading is settled, and no [lat:] or [var:] was fired there, however strong the candidate (see §4b — two are strong). The twin's low word ratio (0.469) is a fact about the SCAN, not evidence about the Greek.
2. ⚑ Calfa p. 671 (Greek cols 1319–1320) — a SECOND, unflagged gap. The primary scan has no leaf for it (leaf 669 = Calfa p.670, leaf 670 = Calfa p.672). The twin covers it only from an alternate scan item (patrologicursus21migngoog, leaf 627 — recorded in the chunk-12 twin's provenance.altSourcePages, and flagged there as "DIFFERENT witness than the rest of the twin"). So for the first ~third of chunk 0012 (from "But the words are those of one urging" to "…but to the preaching [1321]") the Latin exists but the GREEK plate does not. Latin-settled corrections there are ladder row 1 (damage in our file) and are used; claims about Migne's printed Greek ([var:]) are NOT made there. This gap is not named in the brief and should be added to the twin manifest as a Greek-side note.

Verifier-pass result by chunk (pilot §6 requires naming these):

• 0010 — read against twin 0010, FINDINGS (one [var:], one polarity site the Latin settled, one Latin-side OCR note). Not clean.
• 0011 — read against twin 0011 + the tail of twin 0010; PARTLY UNVERIFIABLE. Findings at both ends (one [var:], one negation site); the middle is the gap of §0.1. This is "twin absent", which is not the same as clean.
• 0012 — read against twin 0012, FINDINGS (two polarity sites examined, both resolved as Latin-column defects; one lemma restored from the Latin). Not clean. Greek plate absent for its first third (§0.2).
• 0013 — read against twin 0013, FINDINGS (one [lat:]; two source-loss restorations). Not clean.
• No chunk in this batch came back clean. Zero chunks to report as "nothing found."

1. Catena sigla — three restorations, all ΦΩΤ., all scan-verified

No new siglum occurs in this range; only ΦΩΤ. The work's ΧΡΥΣ. does not reappear after the anchor chunk.

⚑ Siglum POSITION varies and is reproduced in place, not normalised. At 1308/1309 and 1321 the em-dash + siglum stand together before the scholion (the anchor's pattern). At 1309 the plate prints dash — lemma — siglum: ἀναστροφῆς. — «Τοὺς οὕτως περιπατοῦντας.» [ΦΩΤ.] Περιπατεῖν… (leaf 666 confirms this order). The English keeps that order rather than moving the siglum in front of the lemma.

2. Scripture-divergence list (7a″ deliverable) — the words, named

Migne's printed Greek against the received Greek text and against familiar English. Every item was rendered as Migne prints it. ⚠ marks the ones where the familiar English was the live pressure.

Fired as [var:] (plate-verified Greek):

1. ⚠ 1 Cor 15:25 quoted TWICE, in TWO DIFFERENT FORMS — 1313 vs 1316. At 1313 Migne prints «Δεῖ γὰρ αὐτὸν, φησί, βασιλεύειν ἄχρις ἂν ὑποταγῇ αὐτῷ τὰ πάντα» ("until all things be subjected to him") — a form that is not 1 Cor 15:25, being conflated toward 15:28/Phil 3:21. At 1316 the same verse is quoted in its received form, «…ἄχρις οὗ θῇ πάντας τοὺς ἐχθροὺς αὐτοῦ ὑπὸ τοὺς πόδας αὐτοῦ» ("till he put all his enemies under his feet"). Both Greek forms are on the plate (leaf 668) and the Latin column mirrors the doubling exactly — donec subjiciantur ipsi omnia at 1314, donec posuerit omnes inimicos suos sub pedes suos at 1315 (twin 0010). Two-witness concurrence: the doubling is the plate pair's, not ours. The two are therefore rendered DIFFERENTLY, and the familiar "till he hath put all enemies under his feet" is kept out of the first. [var:] fired at 1313 naming the received words.
2. ⚠ John 14:27 at 1317 — Migne prints «Εἰρήνην τὴν ἐμὴν δίδωμι ὑμῖν, εἰρήνην τὴν ἐμὴν ἀφίημι ὑμῖν» — the clauses in the reverse of the received order, and τὴν ἐμὴν with both verbs. Plate-verified, leaf 669. The Latin column reverses the clauses too (Pacem meam do vobis, pacem relinquo vobis) but keeps meam with do only, i.e. Vulgate wording in Vulgate-less order. [var:] fired on the Greek's divergence from the received text; no [lat:] — the Latin's meam placement is Vulgate-conformation (anchor convention).

Logged, NOT marked, because they fall in the p.669 gap (§0.1):

1. ⚠⚠ Luke 10:20 at 1316 — the strongest unmarked candidate in the work. Our source prints «Μὴ χαίρετε…ὅτι τὰ δαιμόνια ὑμῖν ὑποτάσσεται, ἀλλ᾽ ὅτι τὰ ὀνόματα ὑμῶν γέγραπται ἐν βίβλῳ ζωῆς», where the received text has τὰ πνεύματα and ἐγγέγραπται ἐν τοῖς οὐρανοῖς. The divergence is load-bearing: Oecumenius adduces the verse precisely to match Paul's ὧν τὰ ὀνόματα ἐν βίβλῳ ζωῆς, so "written in heaven" would make the argument pointless. Neither reading is producible by OCR — they are textual, not typographic — but with no plate and no Latin at this column I will not assert what Migne printed. Rendered as our source prints; no marker. This is the first thing a backfill session should adjudicate if a leaf for Calfa p.669 is ever found.
2. Luke 6:25 at 1316 — «Οὐαὶ τοῖς γελῶσι» (dative, and no νῦν) for the received οὐαί, οἱ γελῶντες νῦν. Rendered as printed; no marker (gap).
3. Phil 4:3 block lemma at 1315/1316 — Migne prints σύζυγε γνήσιε (the received order is γνήσιε σύζυγε) and συνεργῶν μοι (received μου). Both partly visible on leaf 668 before the column break; word-order item confirmed there, μοι/μου not. Rendered as printed; not material enough for a marker either way.

Logged, no marker, not material (divergence from AV/Vulgate cadence only, or sub-material):

1. ⚠ Phil 4:4–7 at 1316 — the most quoted lemma in the work, and it AGREES with the received text word for word. Χαίρετε ἐν Κυρίῳ πάντοτε, πάλιν ἐρῶ, χαίρετε · Τὸ ἐπιεικὲς ὑμῶν · Ὁ Κύριος ἐγγύς · ἡ εἰρήνη τοῦ Θεοῦ ἡ ὑπερέχουσα πάντα νοῦν φρουρήσει τὰς καρδίας ὑμῶν καὶ τὰ νοήματα ὑμῶν. The pressure here is not a variant but the AV's cadence, and it was resisted at three points: «again I will say» (not AV's "and again I say"); «forbearance» (not "moderation" — the gloss defines it against πολεμίοις, "not as enemies"); «shall guard your hearts and your thoughts» (not "keep your hearts and minds"). ὑπερέχουσα is rendered "surpasses all understanding" and ὑπερβαίνει in the two glosses is rendered "surpass" likewise, so the gloss's argument (if his peace surpasses every νοῦς, what of those who pry into his οὐσία?) stays visible.
2. ⚠ Phil 4:13 at 1320 and its gloss lemma at 1324 — Migne prints ἐν τῷ ἐνδυναμοῦντί με Χριστῷ (the Byzantine/TR reading; the critical text has no Χριστῷ), block lemma and gloss lemma identical. Rendered "I am strong for all things in Christ who empowers me," not AV's "I can do all things through Christ which strengtheneth me," because the gloss turns on the ἰσχύ- root twice — ἐμοῦ πάντα φέρειν ἰσχύοντος and τοῦ τὴν ἰσχύν μοι ταύτην παρέχοντος Χριστοῦ — which "I can do" erases. ⚑ Convention: ἰσχύω = "be strong for" · ἰσχύς = "strength" · ἐνδυναμόω = "empower", fixed for the work.
3. Phil 3:20–21 at 1309/1312 — agrees with the received text; ⚠ but Calfa prints τὸ σῶμα τῆς ταπεινώσεως ὑμῶν in the block lemma where the plate (leaf 667) prints ἡμῶν, the Latin prints corpus nostrum abjectum, and every one of the six re-quotations in the glosses prints ἡμῶν. See §3 — this is the itacism class resolving as damage in OUR file, not a divergence.
4. 1 Cor 15:32 at 1312 — «Φάγωμεν καὶ πίωμεν, αὔριον γὰρ ἀποθνήσκομεν», exactly the received text.
5. Matt 28:18 at 1313 — Migne prints ἐπὶ γῆς (anarthrous; received ἐπὶ τῆς γῆς). Not sense-changing.
6. 1 Cor 10:32 at 1320 — compressed to «Ἀπρόσκοποι γὰρ γίνεσθε», dropping καὶ Ἰουδαίοις…καὶ Ἕλλησιν. Rendered as printed; the plate itself sets it in parentheses.
7. Ps 119:7 (LXX) at 1320 — «Μετὰ τῶν μισούντων τὴν εἰρήνην ἤμην εἰρηνικός», exactly LXX.
8. 1 Cor 9:15 at 1320 (Greek plate absent, §0.2) — Migne prints ἢ τὸ καύχημά μου, ἵνα τις κενώσῃ, where the received has οὐδεὶς κενώσει. The Latin concurs with the ἵνα-clause (ut gloriationem meam aliquis inanem reddat). Rendered as printed; no [var:], because asserting what Migne's GREEK prints needs the Greek witness, which this column lacks.
9. 2 Cor 11:12 at 1320 (same gap) — «εὑρεθῶσιν ὡς κἀγώ» for the received καθὼς καὶ ἡμεῖς. Rendered as printed; no marker, same reason.
10. 1 Cor 9:11 at 1325 — «Εἰ ἡμεῖς ὑμῖν τὰ πνευματικὰ ἐσπείραμεν, μέγα εἰ ἡμεῖς ὑμῶν τὰ σαρκικὰ θερίσομεν;» — the received text; rendered as printed.
11. Phil 4:23 at 1325 — Migne prints the TR form μετὰ πάντων ὑμῶν. Ἀμήν (not μετὰ τοῦ πνεύματος ὑμῶν). ⚠ The block lemma prints Κυρίου ἡμῶν and the gloss lemma prints Κυρίου without ἡμῶν (plate-verified, leaf 672; the Latin has Domini nostri in both). The two are therefore rendered differently, NOT harmonized.
12. ⚠ Phil 4:19 — the block lemma at 1325 prints ἐν δόξῃ ἐν Χριστῷ Ἰησοῦ, but the gloss lemma prints «Σὺν Χριστῷ Ἰησοῦ». Plate-verified (leaf 672). Rendered differently ("in Christ Jesus" / "With Christ Jesus"); see the [lat:] at §4a, which is exactly this word.

3. Ladder row 1 — Calfa damage corrected silently, witness cited

Where a witness is named, that witness is the scan leaf and/or the Latin twin. Items in the p.669 gap (§0.1) are corrected only where the correction is a bare orthographic normalisation (accent, breathing, iota) with no sense at stake.

Chunk 0010 (cols 1309–1313) — witness: leaves 665–667, twin 0010.

• 1308/1309 Ἡ κανόνα τὴν πίστιν φησίν → Ἢ ("Or:", the work's fixed alternative-construal formula; Lat. Aut canonem seu regulam dicit fidem).
• 1309 ἔμπεοσθεν → ἔμπροσθεν · δόξη → δόξῃ · Δεῖ ὡμᾶς → ὑμᾶς (leaf 666) · τὸ αὐτὸ φρριεῆν Εἰ ὀφείλετε → φρονεῖν. Εἰ ὀφείλετε (leaf 666).
• 1309 σκοπεῖτε τοίς οὖτως → τοὺς οὕτως · τύπον ἡμᾶς ok · Δν τὸ τέλος → ὧν · οἱ τὰ ἐπίγεια φρονοῦν. πες → φρονοῦντες (split type).
• 1309 Κακῶς, φησὶν, ἤπερ ἐγώ — kept; leaf 666 concurs; Lat. Male, inquit, et aliter quam ego.
• 1312 αἱσχύνῃ → αἰσχύνῃ · αἱσχόνεσθαι → αἰσχύνεσθαι · Ἡσαν → Ἦσαν · Λεῖ οὖν → Δεῖ οὖν · πολιπευόμεθα → πολιτευόμεθα · ὑπαρχει → ὑπάρχει · ἐξ οὐ → ἐξ οὗ · ὀς → ὃς.
• 1312 ⚠ block lemma τὸ σῶμα τῆς ταπεινώσεως ὑμῶν → ἡμῶν. The pronoun-itacism class, resolved as damage in OUR file and NOT a [lat:], because the two independent witnesses of the Greek disagree with Calfa: leaf 667 prints "τῆς ταπεινώσεως ἡμῶν", the Latin prints corpus nostrum abjectum, and Calfa's own six re-quotations in the glosses all print ἡμῶν. (Contrast the anchor's 1261 case, where plate and Calfa agreed against the Latin and the marker fired.)
• 1312 παρουσία Κμετὰ → μετὰ (intrusive-Κ class) · φανησόμενον kept (leaf 667; Lat. apparebit).
• 1312 λέγομεν ἕτι ἔπεσε → ὅτι (leaf 667) · ταπεινοῦται…ἁπὸ → ἀπὸ · Ἰσγυρὰν → Ἰσχυρὰν · Βἰ γὰρ → Εἰ γὰρ · ἃν μόνον → ἂν μόνον.
• 1313 μεῖ γὰρ αὐτὸν, φησί, βασιλεύειν → Δεῖ γὰρ (leaf 668) · ὑποταγήσεται αὐτῶ → αὐτῷ (×3) · Ὑποτὰσσεται → Ὑποτάσσεται.

Chunk 0011 (cols 1316–1317) — witness: leaf 668 (opening), twin 0010 tail; leaf 669 + twin 0011 (from "…overshadows [1317]"). The middle has neither.

• 1313/1316 Ποῦτο γὰρ…ὁ Θεὸς καὶ iατὴρ → Τοῦτο…Πατὴρ (leaf 668) · Ῥρα δὲ πόσην → Ὅρα δὲ (leaf 668) · ἐὰν οὔτω → οὕτω.
• 1316 ΚΕΦΑΛ. Ϛʹ. → CHAPTER 6. (leaf 668 prints the stigma; Lat. CAPUT VI).
• 1316 Εὐοδίαν! παρακαλῶ → stray ! dropped · παρακαλῶτὸ αὐτὸ → παρακαλῶ τὸ αὐτὸ (joined type) · αἱτινες έν → αἵτινες ἐν.
• 1316 ⚑ μετὰ ΚλήμενΠοὶ ἰαινο. ΛΑΡΟΤ vί. [1316] τος — cross-column / running-foot bleed of the volume signature line (PATROL. GR. CXVIII, visible in clean form on leaf 670 as "PATROL. Gn. CXVIII"), fallen inside the word Κλήμεντος at the column break. The bleed is NOT rendered; the word is restored to Κλήμεντος and the [1316] anchor is placed at the nearest word boundary after "Clement". Second instance of the same bleed class in this batch — see 1324 below. (Anchor's rule extended: the Calfa bleed class in this work is both the Latin running TITLE and the volume running FOOT.)
• 1316 ΚΜεγάλα → Μεγάλα (intrusive Κ) · Ὦς νῦν → Ὡς νῦν · «Ἑρωτῶ καί σε.» → Ἐρωτῶ · «ΣύΕυγε γνήσιε.» → Σύζυγε · τινές ὑπενόησαν → τινὲς · ἐν φιλίπποις → Φιλίπποις · καὶ οὕτε ἐν Ταρσῷ → οὔτε · ἕνθα → ἔνθα · τῶν δύο ζν ἐμνήσθη → ὧν · ὦ καὶ παρατίθεται → ᾧ · ἀἰνίττεται → αἰνίττεται · Ἠσαν μὲν → Ἦσαν · τὰ Κὀνόματα → ὀνόματα (intrusive Κ).
• 1316 ⚑ ἢ ὄνομα Κύριον, εὐδόκιμος δέ τις → ὄνομα κύριον = "a proper name" (the grammatical term), the capital being a Calfa artifact. No witness at this column (gap); the reading is settled by sense alone — the alternative ("or [his name is] Lord") is not construable with εὐδόκιμος δέ τις, ᾧ παρατίθεται. Logged as a reading decided without a verifier.
• 1316 ⚑ guillemet pair repaired: Calfa prints συνυπέμειναν.» Καὶ τῶν λοιπῶν συνεργῶν μου.» — a closing mark with no opening, and a stray closing after συνυπέμειναν. Restored to συνυπέμειναν. «Καὶ τῶν λοιπῶν συνεργῶν μου.». Under Pattern 6 this is repairing a broken pair in OUR source, not supplying a mark the plate lacks (every short lemma in this work is guilleted, and one half of the pair survives). No plate at this column; flagged for the merge session.
• 1316/1317 Ὁ Κύριος ἐγγὸς → ἐγγύς · Εαὶ ἡ εἰρήνη → Καὶ · ΚΘρουρήσει → φρουρήσει (intrusive Κ + capital) · καρδίας ὑμων → ὑμῶν · Ἢσαν ἐν λύπῃ → Ἦσαν · ἂς ὑπέμενον Αείκνυσιν → ὑπέμενον. Δείκνυσιν · χαρὰς ὓν → χαρᾶς ὂν · Ἡ ὅτι ἡ ὑπὲρ Χριστοῦ λύπη → Ἢ ("Or:") · αὐτὸχάρα ἐστὶν → αὐτὸ χαρά ἐστιν.
• 1317 ἐπισκογιζούσης → ἐπισκιαζούσης (Lat. obscurante, twin 0011) · γνο σθήτω → γνωσθήτω · διέβαλέ τινας, ὡν → ὧν · τοὺς πιστοῦς → πιστούς · Ὡς αἵγε μὴ ἔχουσαι εὑχαριστίαν → εὐχαριστίαν · οὕτε γνωρίζονται → οὔτε (leaf 669) · Ὢστε μὴ ὑπὸ πειρασμῶν → Ὥστε · stray , before the closing guillemet of the John 14:27 quote, dropped.
• 1317 block lemma Phil 4:8 ὅσο σεμνὰ → ὅσα σεμνά · ὅσα ἀγνὰ → ἁγνά · εὕφημα → εὔφημα · Εἰ τις ἀρετὴ → Εἴ τις · Ἀ καὶ ἐμάθετε → Ἃ · ἡκούσατε → ἠκούσατε · ἷδετε ἐν ἐμοί → εἴδετε (leaf 669).

Chunk 0012 (cols 1320–1324) — witness: twin 0012 throughout; leaf 670 from "…to the preaching [1321]" onward; leaf 671 for the 1324 tail. Greek plate absent before [1321] (§0.2).

• 1320 Ἐπειγο μένου → Ἐπειγομένου (split) · Θεῦδος γὰρ ἡ κακία → Ψεῦδος (leaf 669; Lat. mendacium enim vitium est) · ἀρέσκεινἈπρόσκοποι → ἀρέσκειν. (Ἀπρόσκοποι (joined type; leaf 669 shows the parenthesis) · τὸ, Ὑετε ἐν ἐμοί → Εἴδετε (Lat. Vidistis in me) · οὐκ ἐνῆ → ἐνῆν · τῳ καὶ ἴδετε ἐν ἐμοί → τῷ, Καὶ εἴδετε ἐν ἐμοί.
• 1320 τοῦ τὰ γήῖνα φρονούντων — kept as printed (genitive singular article with genitive plural participle). Leaf 669 concurs, so this is the plate pair's own anomaly, not ours; English has no exponent for it, so it is rendered "him who minds earthly things" with no marker (Pattern 9).
• 1320 block lemma Phil 4:10–13: δτι ἡδη ποτὲ → ὅτι ἤδη ποτὲ · ἐφ’´ ὧ → ἐφ᾽ ᾧ · ἡκαιρεῖσθε → ἠκαιρεῖσθε · ἔμαθον οἰς εἰμι → ἐν οἷς εἰμι · αὐνάρκης εἰναι → αὐτάρκης εἶναι · Χρισιῷ → Χριστῷ · βλίθει → τῇ θλίψει · συγκοινωνήσαντές μου kept.
• 1320 ⚠ block lemma ἐν παντὶ καὶ ἐν πᾶσι μέμνημαι → μεμύημαι. The gloss at 1324 re-quotes the same clause and prints μεμύημαι (leaf 671 confirms the gloss); the Latin prints institutus sum in BOTH the lemma and the gloss (twin 0012). Ladder row 1: our file's damage, corrected so that lemma and gloss read identically, as 7a″ requires. Rendered "I have been initiated" in both.
• 1320 ὡφελήθησαν → ὠφελήθησαν · Τὸ, Ἥδη ποτὲ → Ἤδη · τῷ μὴ λαμβάνεινκαλὸν → λαμβάνειν (καλὸν (joined type) · ἡ τὸ καύχημά μου → ἢ · ἵνα τίς κενώσῃ → ἵνα τις κενώσῃ.
• 1320 ⚑ νῦν εἴληφες → νῦν εἴληφεν; — 2nd person for 3rd, and the question mark restored. Settled by the Latin: Quamobrem autem Paulus, qui semper gloriari solet quod non acciperet (…) nunc tamen accepit? (twin 0012). Third person is required by the whole sentence (Παῦλος … ἐγκαυχώμενος). Ladder row 1; no Greek plate at this column, so the Latin is doing the work and is quoted here per the contract.
• 1321 ἁλλὰ τῷ κηρύγματι → ἀλλὰ · ἐν ἐνδείᾳ ὢν ok · Κεἶπον, ἵνα μὴ ἄκαρποι → εἶπον (intrusive Κ) · μὴ νομίΚσητε → νομίσητε (intrusive Κ) · ἥν εἶχον → ἣν · ιὰ τοῦτο ἐχάρην → Διὰ τοῦτο.
• 1321 καὶ πολλοὺς ἔχων εἴπερ ἤθελον → …ἔχων, Καὶ εἴπερ ἤθελον per leaf 670; rendered "had I been willing" either way — no sense at stake.
• 1321 οὕτω τὰ πρὸς χρείαν ok · ἀπεδεξάμην ok · της χρείας μου → τῆς.
• 1324 Ὥσπεο γὰρ → Ὥσπερ · οὔτως καὶ ἡ ἄδειᾳ οἷδε → οὕτως…ἡ ἄδεια οἶδε · γεγένηνται ok · κεχορτασμένην ok · σως ἐρεῖ τις → Ἴσως.
• 1324 ⚑ ἐν ΡΑτὴοΙ. Ολ. ΣΧvΙΙi. [1324] πᾶσι τοῖς παρεμπίπτουσι — the same running-foot bleed as at 1316 (PATROL. GR. CXVIII), here fallen between ἐν and πᾶσι. Not rendered; the [1324] anchor kept in place.
• 1324 Τουτέσ-. Ι.ῖραν ἁπάντων ἔχω → Τουτέστι, πεῖραν ἁπάντων ἔχω (leaf 671; Lat. Institutus sum, id est, omnium habeo experientiam) · μεγαλοῤῥημο νεῖν → μεγαλορρημονεῖν (split) · πάντα τῳ λριστῳ → τῷ Χριστῷ · οὖτως ἐχόντων → οὕτως · ῥαθυμότεροι → ῥᾳθυμότεροι · μεταχειοίξει → μεταχειρίζει (leaf 671) · τὸ φρόνημα. αὐτῶν → stray point dropped · παύσαντές μουτὴν λύπην → μου τὴν λύπην (joined) · θλίψει.» Οἷον δέ ἐστι…γενέσθα· Παύλου → γενέσθαι Παύλου.

Chunk 0013 (col 1325) — witness: leaves 671–672, twin 0013.

• 1324/1325 block lemma Phil 4:15–19: Οἱδατε → Οἴδατε · Φλιππήσιοι → Φιλιππήσιοι · ἐν ἀρχῆ τοῦ ναγγελίου → ἐν ἀρχῇ τοῦ Εὐαγγελίου · καὶ ήψεως → λήψεως · Θοσσαλονίκῃ → Θεσσαλονίκῃ · ἐπέχψατε → ἐπέμψατε · ἑπιζητῶ τὺν καρπὸν τὸν πρλεονάζοντα → ἐπιζητῶ τὸν καρπὸν τὸν πλεονάζοντα · λόγον ὑμων → ὑμῶν · Πεπιλήρωμαι → Πεπλήρωμαι · Ἐπαφροδίπου → Ἐπαφροδίτου · έσμὴν εὐωδίας → ὀσμὴν · εἰὐάρεστον → εὐάρεστον · πλὴρώσει → πληρώσει · χρείαν ίμῶν κτὰ → ὑμῶν κατὰ · ἐν ΧριΟ στῷ → ἐν Χριστῷ (split; leaf 671's own ἐκ is scan-side OCR, Calfa's ἐν is right and the Latin's in concurs).
• 1325 ταῦτα ζητω → ζητῶ · καρπὸν Κἐνέγκητε → ἐνέγκητε (intrusive Κ) · ἤγημαι → ἥγημαι · Ἀσπάσασθε πάν τα ἄγιον → πάντα ἅγιον (split) · μάλισα δὲ, οἱ ἐκ τῆς Καίσαρος οἰκείας → μάλιστα…οἰκίας (Calfa's own gloss lemma four lines later prints οἰκίας) · Ἀμήν Ἐπειδὴ → Ἀμήν. Ἐπειδὴ · «Ἰi χάρις → Ἡ χάρις · Τὸ σύνηθες αὐτῳ → αὐτῷ · ἢ δόξα → ἡ δόξα · «Βἰ ἡμεῖς ὑμῖν → Εἰ ἡμεῖς.

3b. ⚑ TWO SOURCE LOSSES restored from the plate — a fresh convention call

Both are runs of text the plate carries and Calfa has dropped entirely, not mangled. The PG register section ("a certain correction … is made silently in the English and logged in cruces.md") is read as governing, in preference to a Pattern-13 [ed:], because in both cases the lost text is recovered from the plate itself — this is not speculation and nothing is supplied "from elsewhere". The merge session should ratify or reverse this; if reversed, both become [ed:] notes. Both also warrant data/calfa-patches/ entries (the plate check the ladder requires for a patch is satisfied — leaves 671, 672).

1. 1325, at the column break. Calfa: Οὐχ ὅτι ἐπιζητῶ τὸ δόμα ὑμῶν, διὰ τοῦτο [1325] λόγον ὑμῶν, οὐκ ἐμοῦ, πλεονάζοντα. — nine words gone, and what remains does not construe. Plate (leaf 671→672): …διὰ τοῦτο ταῦτα λέγω, ἀλλ᾽ ἐπιζητῶ τὸν καρπὸν ὑμῶν, τὸν εἰς λόγον ὑμῶν, οὐκ ἐμοῦ, πλεονάζοντα. Latin concurs: Non quod requiram munus vestrum propterea hoc dico, sed requiro fructum vestrum qui in rationem vestram non meam exuberat (twin 0013). Restored; the [1325] anchor kept exactly where our source prints it, i.e. inside the restored run.
2. ⚠ 1325, the work's final line — the subscription is truncated. Calfa ends: Τέλος σὺν Θεῷ τῆς πρὸς Φιλιππνσίους — and stops mid-phrase. Plate (leaf 672): Τέλος σὺν Θεῷ τῆς πρὸς Φιλιππησίους Ἐπιστολῆς. Ἐγράφη ἀπὸ Ῥώμης δι᾽ Ἐπαφροδίτου. Latin concurs: Finis, divino favente auxilio, Epistolae ad Philippenses. Scripta est ab urbe Roma per Epaphroditum (twin 0013). Restored in full. Without this the work appears to break off; with it, the work ends where Migne ends it.

4. Markers fired

4a. [lat: …] — one, plate-verified

• 1325, Phil 4:19. Greek lemma: «Σὺν Χριστῷ Ἰησοῦ.» Ἢ διὰ Ἰησοῦ Χριστοῦ δώσει. Plate-verified — leaf 672 prints «Σὺν Χριστῷ Ἰησοῦ», so this is not Calfa damage (and note the block lemma at the same column prints ἐν Χριστῷ Ἰησοῦ, which Calfa also carries correctly, so no systematic σύν/ἐν confusion is in play). Latin column (twin 0013, col 1326): « Per Christum Jesum. » Nam ἐν capitur pro …, id est In pro Per. The Latin's lemma and its whole scholion are about ἐν; the Greek column's lemma prints σύν. A reader of the Latin column takes away that the word under discussion is ἐν and the question is in vs per; a reader of the Greek takes away σύν, for which that discussion is beside the point. Material by the Pattern-16 test. English follows the Greek; marker fired. Not a Vulgate-conformation case — the Vulgate reads in gloria in Christo Jesu, so Per is Allatius's own choice, and his note says so.

4b. [lat: …] NOT fired — the candidates, and why

• 1312, «Ὧν ὁ θεὸς ἡ κοιλία» — Ὡς δεσποίνῃ γὰρ … δουλεύουσιν vs Latin Siquidem ipsum tanquam regem, imo potius tanquam deum colunt. Greek "as a mistress" (feminine, agreeing with κοιλία), Latin "as a king". This is grammatical-gender accommodation to venter, which is masculine in Latin — Allatius's freedom, not a different fact. Logged, no marker.
• 1316, μεθ᾽ ὑμῶν / 1320, μεθ᾽ ὑμῶν vs Latin nobiscum (twice). ⚠ The pronoun-itacism class: Greek at 1320 reads Εἰ γὰρ ἡμεῖς … εἰρηνεύομεν … καὶ ὁ Θεὸς ἔσται μεθ᾽ ὑμῶν. Θεοῦ δὲ ὄντος μεθ᾽ ὑμῶν… — first person then second, where the Latin is first person throughout (nos … Deus quoque nobiscum … quod si Deus nobiscum sit). This is exactly the anchor's 1261 case in reverse, and per the standing rule it needs the Greek plate-verified before firing. The plate is absent at this column (§0.2 — Calfa p.671 has no primary leaf). Candidate logged without the marker, and the reason stated, as the rule requires.
• 1316/1317, Vulgate-conformed scripture in the Latin — Pacem meam do vobis, pacem relinquo vobis (John 14:27); quaecunque sunt vera … boni omnis (Phil 4:8, where boni omnis renders εὔφημα by the Vulgate's sense); Gratia Domini nostri at 1326 against the Greek gloss lemma's bare Κυρίου. All conformation, none marked (anchor convention).
• 1316, δουλώσεται αὐτήν vs Latin subdet — free rendering, not material.
• 1320, ὡς ἐπὶ καρπῶν ξηρανθέντων vs Latin de arefactis prius fructuosis arboribus — Allatius supplies "trees"; free, not material.

4c. [var: …] — two, both plate-verified

Listed at §2.1 (1 Cor 15:25 at 1313) and §2.2 (John 14:27 at 1317). Four further candidates (§2.3–2.5, §2.13–2.14) were deliberately left unmarked because the column has no witness; each is named there so a later session can find them.

5. Polarity / negation sites read (runbook 4a, PG form: against both columns)

Every site below was examined; none was smoothed. Named individually so this pass can be told apart from a pass never made.

1. 1312, «τίς ἀμφισβητήσει ὅτι οὐ πάντως ἔσται τοῦτο;» — the οὐ is on the plate (leaf 667) and the Latin carries its exact structural equivalent, quin omnino id erit? Redundant negation after a verb of doubting, in both columns. Rendered with the English idiom that carries the particle — "who will dispute but that** this shall assuredly be?" — rather than dropping it (which would be a silent repair) or rendering it as a flat negative (which would state the opposite of the argument). Two-witness concurrence; no marker.
2. 1309, «εἰς διάκρισιν τῶν ὀρθῶς περιπατούντων, καὶ μὴ σκοπεῖτε τίνες περιπατοῦσι…» — ⚠ the most dangerous site in the batch. Calfa's comma invites reading καὶ μὴ σκοπεῖτε = "and do NOT mark", which reverses the sentence. The Latin settles the scope: ad internoscendum quinam recte ambulant, et qui non, considerate qui ambulent… — the μή belongs to the preceding pair ("those who walk rightly, and those who do not"), and σκοπεῖτε is a positive imperative. Rendered accordingly; the Latin is quoted here because it settled the reading, per the contract.
3. 1317, «Ὡς αἵγε μὴ ἔχουσαι εὐχαριστίαν εὐχαὶ, οὔτε γνωρίζονται πρὸς τὸν Θεόν.» — the Greek is doubly negative (prayers that do NOT have thanksgiving are NOT even made known). The Latin twin as OCR'd reads …precationes quae gratiarum actionem non habent, bene innotescant quidem, which asserts the opposite of the Greek's main clause. Judged a defect in the LATIN column's OCR (bene for ne/nec; leaf 669 gives the same string, so the fault may be the plate's own type) — the Greek negation is unambiguous, plate-verified, and required by the following sentence (Ὁ γὰρ μὴ εὐχαριστῶν … πῶς ἢ αἰτήσει ἢ λήψεται;). Greek negations carried; no marker, no smoothing.
4. 1321, «Ἐπειδὴ ἠρέμα ἐπετίμησεν αὐτοῖς γράψας, ἤδη ποτέ» vs Latin Quoniam scribendo clam eos objurgavit, non tandem ait — the Latin's non is OCR for Jam (= ἤδη ποτέ, the lemma being quoted); leaf 670 confirms the Greek has no negative. No divergence.
5. 1321, «Μὴ γὰρ νομίσητε … ὅτι νῦν διὰ τὸ μακράν με γενέσθαι … διὰ τοῦτο τὴν ὑμῶν χορηγίαν ἀπεδεξάμην» vs Latin Ne existimetis, inquit, quod non eo quod longe absim … propterea sumptum a vobis acceperim. The Latin carries a non the Greek does not print. Judged the same redundant- negative class as site 1 (a pleonastic negative after a verb of denying, which this Latin column demonstrably uses), not a polarity divergence. Greek rendered as printed; logged, no marker. ⚑ Worth noting as a class: this Latin column pleonastically negates after verbs of doubting and denying (quin, quod non), so a bare count of non against the Greek will manufacture divergences in this work.
6. 1320, «οὐκ ἐπλατύνθη οὐδέ εἰσιν οἱ … ἐπιχορηγοῦντες» vs Latin non adeo dilatata est … nec sunt qui … ministrent — negations 1:1. Clean.
7. 1321, «μήτε ἐπιφερόμενος, μήτε τῷ ἐργοχείρῳ» vs Latin neque aliquid importasset, neque manuali operi — 1:1. Clean.
8. 1324, «οὐκ ἴσχυσαν ἐνέγκαι … οὐκ ἤνεγκε τὸν κόρον … Ἀλλ᾽ οὐχ ὁ Παῦλος» vs Latin nec saturatum ventrem ferre potuerunt … satietatem non tulit … at non ita Paulus — 1:1. Clean.
9. 1325, «Οὐ γὰρ ἂν μὴ πιστεύσαντες, ἠσπάσαντο τοὺς Φιλιππησίους.» vs Latin neque enim Philippenses salutassent nisi credidissent — both negatives carried in the English ("For had they not believed, they would not have saluted the Philippians"). Clean.
10. 1325, «Οὐκ ἔδωκαν γὰρ σαρκικὰ, ἀλλ᾽ οὐδὲ λαβεῖν ἠθέλησαν πνευματικά.» vs Latin Non dederunt enim carnalia, sed neque accipere voluerunt spiritualia — 1:1. Clean.
11. 1325, «Οὐκ εἶπεν· Οὐδεμία μοι Ἐκκλησία δέδωκεν, ἀλλ᾽, Ἐκοινώνησε.» vs Latin Non ait: Nulla mihi Ecclesia dedit, sed, Communicavit — 1:1. Clean.
12. 1316, «Οὐδὲ ὁ Χριστὸς τὸ πένθος τὸ κοσμικὸν ἐπέταξεν … Οὐδὲ ὁ Παῦλος τὴν κοσμικὴν λέγει χαράν.» — in the p.669 gap; no verifier. Both οὐδέ carried; the internal logic (Christ commands mourning / Paul commands joy, resolved by "neither meant the worldly kind") is self-consistent, so no repair was needed and none was made.

6. Question marks restored (Calfa · or nothing for the plate's ;)

Logged as a class per the anchor, not per site. Restored at: 1309 Ἀπόλλυται δὲ πῶς; · Ποίους τούτους κελεύεις ἡμᾶς παραφυλάττεσθαι; · 1313 Εἶτα θέλεις καὶ ἑτέραν ἀπόδειξιν; · 1316 οὔτε ἐν Ἱερουσαλὴμ, ἔνθα ἐτέθραπτο; · Τί γὰρ ὁ Χριστός; · Πῶς οὖν φησιν ὁ Παῦλος, Χαίρετε; · 1317 Οὕτως, πῶς; · 1320 Τί χρὴ τὰ λοιπὰ περιεργάζεσθαι; · νῦν εἴληφεν; (with the person corrected, §3) · 1324 ποίας ἀρετῆς; · 1325 πόσῳ μᾶλλον … τοῦτο δεῖ ποιεῖν;

7. ⚑ Conventions decided fresh in this range (for the merge)

• ἰσχύω / ἰσχύς / ἐνδυναμόω = "be strong for" / "strength" / "empower" (§2.7) — fixed so Phil 4:13 and its gloss share a root, against AV's "I can do all things".
• ἐπιεικές / ἐπιεικῶς = "forbearance" / "with forbearance", never "moderation" or "gentleness" (Phil 4:5 and its gloss, 1317).
• σεμνά = "seemly" (Phil 4:8), not "honest"; the gloss defines it by bearing, speech, gait and deeds.
• ὑπερέχω / ὑπερβαίνω both = "surpass" (Phil 4:7 and its two glosses).
• πολίτευμα / πολιτεύομαι = "commonwealth" / "live as citizens" (1312), so the gloss's ἐκεῖ στρατευόμεθα, ἐκεῖ πολιτευόμεθα keeps its pair.
• μεμύημαι = "I have been initiated" (not "I am instructed"), identical in lemma and gloss.
• Chapter heads: ΚΕΦΑΛ. Ϛʹ + argument → "CHAPTER 6." + the argument sentence, per the anchor's ΚΕΦΑΛ. Αʹ handling.
• The Calfa bleed class in this work includes the volume RUNNING FOOT (PATROL. GR. CXVIII), not only the Latin running title — two instances, 1316 and 1324, both mid-word.
• Plate-verified source LOSSES are restored silently and logged (§3b), rather than marked [ed:]. Ratify or reverse at merge.
• A broken guillemet pair in our source is repaired, not left half-open (§3, 1316). This is repair of our file, not supply against the plate.