Skip to content
Karma Sakha
Sign in

The Texts · Further Classical Jātaka / Muhūrta / Praśna

Ṣaṭpañcāśikā

षट्पञ्चाशिका

Attributed to Pṛthuyaśas (the same traditional son-of-Varāhamihira attribution as Horāsāra) · If the attribution holds, later 6th century CE — confirmation lacking · prashna

On the dating: The attribution is traditional; many of its verses closely parallel Praśna-jñāna and Daivajña Vallabha.

What this text is

The classic fifty-six-verse horary primer.

Query judgment from the ascendant, aspects and planetary significations, long read with Utpala-tradition commentary.

Its role in this app

Praśna corroboration beside Praśna Mārga and Praśna Tantra — and the first text this project has published in full: all 56 verses appear below in Sanskrit with our own English translation, under CC BY 4.0.

Editions our research holds

  • vak-ayer-english

    V.A.K. Ayer English translation — the tier’s best horary seat.

    in copyright — read for verification only, never republished · scan quality: good

  • sastri

    V.S. Sastri edition (82 pp) — the printed Sanskrit we transcribed the verses below from. Its English is in copyright and was consulted only as a sense-check after our own translation was drafted.

    in copyright — read for verification only, never republished · scan quality: usable

  • sanskrit-text

    A handwritten manuscript scan, six folios and incomplete — held as a variant witness, not used as the source.

    public domain · scan quality: unusable

What our reading found

Findings from this project's own verification passes — recorded because a source's defects are part of citing it honestly.

The work has seven adhyāyas of 7+17+5+5+5+5+12 = 56 verses — the number its own title names, and the strongest completeness check the text affords. There is therefore no Adhyāya VIII, so the commentary’s forward reference to “VIII-14”, for a line that actually stands at VI.5, points outside the text as this edition numbers it. The edition’s running heads print the wrong chapter or śloka in at least two places and are not reliable locators; its index, by contrast, held across some seventy checked entries and independently settled the worst-printed verse in the book. The printing sets visarga as a low dot that often fails to scan, spells drekkāṇa two ways within a single chapter, and carries real metrical defects at II.14 and IV.2 that belong to the text rather than to our reading. Three akṣaras defeated all three of our readers and are marked rather than guessed.

The text itself

The Ṣaṭpañcāśikā is here in full. All 56 verses, which is the whole work — its title means "the fifty-six", and the edition's own śloka index counts exactly that many. On 2026-09-23 the mūla verification of 2026-09 ruled every one of the 56 verses against every printing this library holds: 1 correction written, 55 verses cleared, and 21 variant readings put on the ledger. A book that barely moves in a pass is the expected result here and not a thin one: this text was already at tier constituted, so its verses had been weighed between two exemplars before this pass began, and the pass was asking a narrower question. The length stands at 56 in 7 chapters on the basis already recorded — the Sastri 1941 edition’s own śloka-anukramaṇikā at printed pp. 65–66 indexes exactly 56 ślokas, pratīka by pratīka, and the work’s title means "the fifty-six". ONE THING A READER SHOULD NOT INFER FROM THE TIER. Of the 47 checked-no-change rulings this book carried from the round of 2026-09-08 on the 1941 first edition, 14 say in their own words that they claim "the openings and the numbering, not a word-by-word plate read of the bodies", and 6 more are index-only. So 20 of the 56 verses had their opening words and their number checked against that printing rather than every akṣara of their body. That is recorded here because the tier does not say it. 56 of this text's 56 verses carry an English translation; 7 of 7 chapters are begun or finished (1–7). This is the whole work: 56 verses against the 56 it is known to have. Every verse is tier 'constituted': two independent exemplars weighed, every disagreement adjudicated and recorded. ⚠ 7 of 7 chapter(s) carry a visual-validation record and 0 have been audited by the gate, so the third clause of 'constituted' — a visual gate — is NOT yet met.

56 verses in the original Sanskrit with our own English translation, which is made from that Sanskrit and not adapted from any existing translation. The Sanskrit was read from the printed plates of V. Subrahmanya Sastri's edition by two independent readers working separately; where they disagreed the page was re-read at higher magnification and the disagreement recorded. THAT EDITION IS NOT PUBLIC DOMAIN. Its title page prints "(Copy right Registered)", and this sentence said "public-domain printed plates" until 2026-09-08, when somebody opened the title page. What we take from it is the mūla, which is ancient and nobody's copyright; none of its English is used, and its wording is checked against by the copyright rail. The copy ingested is the THIRD edition (Sri Mallikarjuna Press, Bangalore, 1978, seen through the press by K. Subrahmanyam). On 2026-09-08 the FIRST edition (Sri Vijaya Lakshmi Vilas Press, Bangalore City, 1941; original preface signed 23 December 1940; IGNCA R-223) was opened for the first time and collated: it is a different setting of the same editor's text, so it is NOT an independent exemplar, but it prints many sorts the worn 1978 setting drops, and nine readings were settled on it that day — including the flagship crux at II.1. An independent encoded e-text of a different transmission line was collated in 2026-08. On 2026-08-13 the whole book — all 56 verses — was swept a second time by readers who were given only page images and had never seen our text, and every place they differed from us was decided by looking at the ink at up to 26x magnification. The length of the work is not our inference: the 1941 edition carries a śloka-anukramaṇikā (printed pp. 65–66) which indexes exactly 56 ślokas, in the division 7·17·5·5·5·5·12 that this corpus uses.

ABOUT THIS PRINTING, because most of what looks like disagreement is one fact about it. The 1978 plates are a worn letterpress that drops superscript and subscript marks throughout: the anusvāra above all, then the visarga and the repha, and sometimes a whole consonant sort. Of the 36 verses where the blind sweep differed from our text, the majority differ only in such a mark. We restore them, because the grammar and the metre require them and because they are present in some occurrences of the same word elsewhere in the same book; the verse's diplomatic reading, where one is given, shows what the page actually sets. SINCE 2026-09-08 THIS IS NO LONGER AN INFERENCE. The 1941 FIRST edition of the same text was opened and prints the very sorts the 1978 drops — वाच्यं for वाच्य at I.2, धातुं मूलं for धातुमूल and मित्यो for मिथ्यो at I.7, शुभैः for शुभै at III.2, वदन्ति for वदति at IV.2, तत्तुल्य for ततुल्य at II.14, and प्रमदाकृतं where the 1978 sets an avagraha and a gap. So the restorations were right, and the ones that could not be restored are now visible as what they are. The printing also carries outright compositor's errors — रिथर for स्थिर at II.15, विदर्ययः for विपर्ययः at IV.1 — which are recorded and emended, never adopted. TWO PLACES ARE STILL OURS AND NO WITNESS HAS THEM, and they are labelled as ours wherever they appear: द्व्यात्मकांशके at II.15, where every witness reads ह्यात्म or a variant of it and no witness reads द्व्य; and the तु of वदेत्तु at II.10, where the 1941, the 1978 and the independent e-text all print the non-word वदेत्त, so the defect is older than any copy we hold.

This is a first edition, under correction. This text has been audited three times adversarially and once mechanically, and the mechanical pass is the one that found most. Readers given only the page images, with no access to our record, re-transcribed all 56 verses; a diff then put their reading beside ours and every difference was adjudicated on the plate. Four readings changed as a result. One of them, at VII.9, we had printed as अधः where the plate and an independent exemplar both read भ्रष्टः — and our own apparatus already held the evidence, unused. One closed a gap we had published as unreadable. One removed a non-word. One, at IV.1, showed that two witnesses which were supposed to record what the page said had quietly normalised it into what they expected, and that our text had inherited the normalisation as though it were attested. Everything found, including what remains open and the occasions where we stated evidence stronger than the evidence we held, is written down in the project's corrections file.

81 reading disagreements were adjudicated and are recorded verse by verse. Every akṣara was legible.

Adhyāya 1 · horādhyāyaḥ (On General Principles)

1.1a reading here is contestedtwo exemplars weighed

प्रणिपत्य रविं मूर्ध्ना वराहमिहिरात्मजेन पृथुयशसा । प्रश्ने कृतार्थगहना परार्थमुद्दिश्य सद्यशसा ॥ १ ॥

praṇipatya raviṃ mūrdhnā varāhamihirātmajena pṛthuyaśasā | praśne kṛtārthagahanā parārthamuddiśya sadyaśasā || 1 ||

Having bowed his head to the Sun, the far-famed son of Varāhamihira, of true renown, sets forth this composition — profound in accomplishing its purpose in query — with the good of others as its aim.

5 notes on this verse
  • Where the Sanskrit is open: The verse carries no finite verb and no explicit noun for the work: kṛtārthagahanā (feminine nominative) presupposes an understood feminine noun such as kṛti, 'composition'. 'Sets forth this composition' supplies that ellipsis; the Sanskrit itself leaves it implicit.
  • What the page would not yield: In this printing the रा of वराह is ink-damaged and superficially resembles ग; both witnesses nonetheless read वराह, and the edition's own gloss below the verse names Varāhamihira.
  • About this edition: Before Adhyāya 1 the edition prints Bhaṭṭotpala's maṅgala verse (keśājārka… karoty utpalaḥ), the commentator's benediction — transcribed in the witness file, not numbered as text.
  • Corroboration: The edition's index attests both the author's name and the subject of this verse at I—1, independently of the running heads (which this printing gets wrong elsewhere).
  • About this edition: Annotation moved out of the `witnessA` field (CORRECTIONS.md C6): a witness field records the reading and nothing else, so this gloss now lives where notes live — “uncertainties flagged: rā of varāha ink-damaged; possible anusvāra dot on mūrdhnā”
1.2two exemplars weighed

च्युतिर्विलग्नाद्धिबुकाच्च वृद्धिर्मध्यात्प्रवासोऽस्तमयान्निवृत्तिः । वाच्यं ग्रहैः प्रश्नविलग्नकालाद्गृहं प्रविष्टो हिबुके प्रवासी ॥ २ ॥

cyutirvilagnāddhibukācca vṛddhirmadhyātpravāso'stamayānnivṛttiḥ | vācyaṃ grahaiḥ praśnavilagnakālādgṛhaṃ praviṣṭo hibuke pravāsī || 2 ||

A fall from position, reckoned from the ascendant; increase, from the fourth; a journey abroad, from the tenth; the return, from the seventh — all this is to be declared by the grahas, counted from the moment of the query's ascendant. A graha in the fourth shows one who has gone indoors; [elsewhere,] one who is abroad.

What the printed page actually sets, here

The Sanskrit above is our constituted text — what we judge the work to say, after weighing two independent exemplars and recording every disagreement. This edition's page differs, and the difference is shown rather than smoothed away:

च्युतिर्विलग्नाद्धिबुकाच्च वृद्धिर्मध्यांत्प्रवासोऽस्तमयान्निवृत्तिः । वाच्य ग्रहैः प्रश्नविलग्नकालाद्गृहं ⟨प्र⟩विष्टो हिबुके प्रवासी ॥ २ ॥

4 notes on this verse
  • About this edition: The printed वाच्य stands where the syntax would more naturally take वाच्या or वाच्यं. Both witnesses read वाच्य and a 600dpi crop shows no ā-mātrā, so it is transcribed as printed.
  • Corroboration: The edition's index attests this verse's content at I—2 under three separate headings — fall from one's position, transfer, and prosperity from the fourth house — matching cyutiḥ and hibukāc ca vṛddhiḥ.
  • About this edition: RE-READ AND SETTLED (2026-08-13). The plate does print a bindu here — at 20x a distinct diamond blob sits above the headline of ध्या, so मध्यांत् is what the ink says, and two independent blind readers reported it. It is nevertheless not the text: every other member of this list is an ablative (विलग्नात्, हिबुकात्, अस्तमयात्) and मध्यांत् is not a form. Read as a stray sort or a broken piece; मध्यात् stands in the mūla, मध्यांत् is recorded diplomatically. A blind reader also found that the press failed to ink one akṣara at the head of ⟨प्र⟩विष्टो, which is now marked as the lacuna it is rather than passed over in silence.
  • About this edition: SUPERSEDED 2026-09-08 — notes[0] above. It said the printed वाच्य 'is transcribed as printed' because both witnesses read it and a 600 dpi crop showed no ā-mātrā. Both facts are still true of the 1978 plate and both are beside the point: the missing sign is an ANUSVĀRA, which nobody looked for, and the 1978 setting drops anusvāras throughout the book. The 1941 first edition prints वाच्यं (PDF p.11, printed p.3, read at 900 dpi), Ayer prints वाच्यं, and witness D prints वाच्यः. The mūla now reads वाच्यं.
1.3two exemplars weighed

यो यो भावः स्वामिदृष्टो युतो वा सौम्यैर्वा स्यात्तस्य तस्यास्ति वृद्धिः । पापैरेवं तस्य भावस्य हानिर्निर्देष्टव्या पृच्छतां जन्मतो वा ॥ ३ ॥

yo yo bhāvaḥ svāmidṛṣṭo yuto vā saumyairvā syāttasya tasyāsti vṛddhiḥ | pāpairevaṃ tasya bhāvasya hānirnirdeṣṭavyā pṛcchatāṃ janmato vā || 3 ||

Whatever house its own lord sees or stands with, or benefics do, of that house there is increase. By malefics, in the same way, harm to that house is to be declared, whether for the querent or from the birth chart.

4 notes on this verse
  • Another reading: The edition itself records another reading for the fourth pāda: vijñātavyā jāyatāṃ pṛcchatāṃ vā, replacing nirdeṣṭavyā … janmato vā.
  • About this edition: The printing breaks hāniḥ + nirdeṣṭavyā across the third and fourth pādas with a hyphen. That is typographic; the word is single.
  • About this edition: Annotation moved out of the `witnessA` field (CORRECTIONS.md C6): a witness field records the reading and nothing else, so this gloss now lives where notes live — “11 syllables in each of the four pādas”
  • About this edition: The 2026-08-12 note on this verse claimed an independent plate reading disagreed with us. It was withdrawn on 2026-08-13: the diff that produced it was defective (see CORRECTIONS.md C8) — it keyed verses by number alone in a text whose chapters restart numbering, and kept only the first line of each verse. The finished sweep, which read all 56 verses, places this verse's divergence among the anusvāra/visarga/repha marks that this worn printing drops throughout the book. Those are recorded as a property of the exemplar in the text-level note, not as an open question about the reading.
1.4two exemplars weighed

सौम्ये विलग्ने यदि वास्य वर्गे शीर्षोदये सिद्धिमुपैति कार्यम् । अतो विपर्यस्तमसिद्धिहेतुः कृच्छ्रेण संसिद्धिकरं विमिश्रम् ॥ ४ ॥

saumye vilagne yadi vāsya varge śīrṣodaye siddhimupaiti kāryam | ato viparyastamasiddhihetuḥ kṛcchreṇa saṃsiddhikaraṃ vimiśram || 4 ||

If a benefic occupies the ascendant, or its own division, in a head-rising sign, the undertaking reaches its accomplishment. The reverse of this is the cause of failure; a mixed condition brings it about, but with difficulty.

What the printed page actually sets, here

The Sanskrit above is our constituted text — what we judge the work to say, after weighing two independent exemplars and recording every disagreement. This edition's page differs, and the difference is shown rather than smoothed away:

सौम्ये विलग्ने यदि वास्य वर्गे शीर्पोदये सिद्धिमुपैति कार्यम् । अतो विपर्यस्तमसिद्धिहेतुः कृच्छ्रेण संसिद्धिकरं विमिश्रम् । ४ ॥

5 notes on this verse
  • About this edition: DIPLOMATIC vs CONSTITUTED. The Sanskrit above is the CONSTITUTED text — what we judge the work to say. This printing differs: शीर्पोदये with प — a compositor's wrong sort for ष, metrically invisible. Terminal punctuation is a SINGLE daṇḍa before the numeral, not the double our text prints. Both are now recorded separately, because an edited reading and a read one are different objects and sharing one field is what let this record claim more precision than it had (CORRECTIONS.md C1.5).
  • About this edition: The plate prints śīrpodaye — a compositor's wrong sort, प for ष. The constituted text is śīrṣodaye: the edition's own commentary on printed p.7 glosses śīrṣodaya and pṛṣṭhodaya immediately after this verse with a true ष, and VI.3 sets the word correctly. The misprint is metrically invisible, since र्षो and र्पो both scan long, which is presumably how it passed proofing. Recorded rather than silently normalised — this edition's misprints are part of citing it honestly.
  • About this edition: Witness A first read śīghrodaye, which was simply wrong. Witness B supplied the right word, but by normalising the misprint rather than reading it; the record then turned that into a claim about the ink which the adversarial pass refuted. The reading is right and the reason given for it was not.
  • Corroboration: The edition's index attests this verse at I—4 under both success and failure in one's undertaking, matching siddhim upaiti kāryam and ato viparyastam asiddhihetuḥ.
  • About this edition: Annotation moved out of the `witnessA` field (CORRECTIONS.md C6): a witness field records the reading and nothing else, so this gloss now lives where notes live — “all four pādas count 11; A recorded no doubtful akṣara here — wrongly”
1.5two exemplars weighed

होरास्थितः पूर्णतनुः शशांको जीवेन दृष्टो यदि वा सितेन । क्षिप्रं प्रनष्टस्य करोति लब्धिं लाभोपयातो बलवाञ्छुभश्च ॥ ५ ॥

horāsthitaḥ pūrṇatanuḥ śaśāṃko jīvena dṛṣṭo yadi vā sitena | kṣipraṃ pranaṣṭasya karoti labdhiṃ lābhopayāto balavāñchubhaśca || 5 ||

The Moon with her disc full, standing in the ascendant, seen by Jupiter or else by Venus, swiftly brings about the recovery of what is lost — and so does a strong benefic gone to the eleventh.

4 notes on this verse
  • Corroboration: The fourth pāda recurs verbatim as the fourth pāda of VI.3, and the two verses' third pādas restate one another. The repetition is the text's own; it also means the two transcriptions, made thirty plates apart, corroborate each other.
  • About this edition: The printing spaces lābhopayāto as two words; the syllable count shows it is one (lābha + upayāta).
  • About this edition: Annotation moved out of the `witnessA` field (CORRECTIONS.md C6): a witness field records the reading and nothing else, so this gloss now lives where notes live — “śaśāṃko printed with anusvāra, not the conjunct”
  • Corroboration: PRAŚNA MĀRGA QUOTES THIS VERSE WHOLE AT 29.53, AND IT SETTLES THE य. PM prints होरास्थितः पूर्णतनुश्शशाङ्को जीवेन दृष्टो यदि वा सितेन। क्षिप्रं प्रणष्टस्य करोति लब्धिं लाभ्रोपयातो बलवाञ्छुभश्च — the same verse, with the Kerala doubling sandhi (पूर्णतनुश्शशाङ्को for पूर्णतनुः शशांको) and one obvious slip (लाभ्रो for लाभो). It reads उपयातो, and its own English construes it that way: "and so does a strong benefic that has gone to the eleventh (lābha)". Bhaṭṭotpala's 1858 lithograph prints उपघातो and glosses लाभे एकादशस्थाने उपघातः; against it now stand the 1941 first edition, the 1978 third edition and an independent Kerala line. A self-check points the same way and needs no book: the pāda लाभोपयातो बलवाञ्छुभश्च stands again at VI.3 of this same work.
1.6a reading here is contestedtwo exemplars weighed

स्वांशे विलग्ने यदि वा त्रिकोणे स्वांशे स्थितः पश्यति धातुचिंताम् । परांशकस्थश्च करोति जीवं मूलं परांशोपगतः परांशम् ॥ ६ ॥

svāṃśe vilagne yadi vā trikoṇe svāṃśe sthitaḥ paśyati dhātuciṃtām | parāṃśakasthaśca karoti jīvaṃ mūlaṃ parāṃśopagataḥ parāṃśam || 6 ||

If a graha standing in its own navāṃśa, whether in the ascendant or in a trine, looks upon its own navāṃśa, the thing enquired after is mineral. Standing in another's navāṃśa it makes it animal; and one gone to another's navāṃśa, looking upon another's, makes it vegetable.

What the printed page actually sets, here

The Sanskrit above is our constituted text — what we judge the work to say, after weighing two independent exemplars and recording every disagreement. This edition's page differs, and the difference is shown rather than smoothed away:

स्वांशे विलग्ने यदि वा त्रिकोणे स्वांशे स्थित पश्यति धातुचिंताम् परांशकस्थश्च करोति जीवं मूल परांशोपगतः परांशम् ॥ ६ ॥

4 notes on this verse
  • About this edition: DIPLOMATIC vs CONSTITUTED. The Sanskrit above is the CONSTITUTED text — what we judge the work to say. This printing differs: स्थित and मूल WITHOUT visarga/anusvāra, and NO daṇḍa closing pāda b. Witness D (the independent e-text) prints स्थितः and मूलं, so the constituted forms are attested elsewhere in the tradition — our plate simply omits them. Both are now recorded separately, because an edited reading and a read one are different objects and sharing one field is what let this record claim more precision than it had (CORRECTIONS.md C1.5).
  • Where the Sanskrit is open: Both witnesses read the plate WITHOUT the visarga of sthitaḥ and without the anusvāra of mūlaṃ. Both are required by the eleven-syllable count, and both belong to the class of marks this edition sets as a low dot that routinely fails to scan. They are therefore supplied on metre, with the printed state recorded here.
  • About this edition: dhātuciṃtām is printed with anusvāra rather than the conjunct dhātucintām; transcribed as printed.
  • About this edition: The 2026-08-12 note on this verse claimed an independent plate reading disagreed with us. It was withdrawn on 2026-08-13: the diff that produced it was defective (see CORRECTIONS.md C8) — it keyed verses by number alone in a text whose chapters restart numbering, and kept only the first line of each verse. The finished sweep, which read all 56 verses, places this verse's divergence among the anusvāra/visarga/repha marks that this worn printing drops throughout the book. Those are recorded as a property of the exemplar in the text-level note, not as an open question about the reading.
1.7two exemplars weighed

धातुं मूलं जीवमित्योजराशौ युग्मे विन्द्यादेतदेव प्रतीपम् । लग्ने योंऽशस्तत्क्रमाद्गण्य एव संक्षेपोऽयं विस्तरात्तत्प्रभेदः ॥ ७ ॥

dhātuṃ mūlaṃ jīvamityojarāśau yugme vindyādetadeva pratīpam | lagne yoṃ'śastatkramādgaṇya eva saṃkṣepo'yaṃ vistarāttatprabhedaḥ || 7 ||

Mineral, vegetable, animal — so in an odd sign; in an even sign one should find this same order reversed. Whatever navāṃśa stands in the ascendant is to be counted in that order. This is the matter in brief; its subdivisions follow from expanding it.

What the printed page actually sets, here

The Sanskrit above is our constituted text — what we judge the work to say, after weighing two independent exemplars and recording every disagreement. This edition's page differs, and the difference is shown rather than smoothed away:

धातु मूल जीवमिंथ्योजराशौ युग्मे विन्द्यादेतदेव प्रतीपम् । लग्ने योंशस्तत्क्रमाद्गण्य एव संक्षेपोऽयं विस्तरात्तत्प्रभेदः ॥ ७ ॥

6 notes on this verse
  • About this edition: DIPLOMATIC vs CONSTITUTED. The Sanskrit above is the CONSTITUTED text — what we judge the work to say. This printing differs: An anusvāra on मिं that our text omitted, and NO avagraha in योंशस् where our text printed one. Both are now recorded separately, because an edited reading and a read one are different objects and sharing one field is what let this record claim more precision than it had (CORRECTIONS.md C1.5).
  • About this edition: Two of the three witnesses read थ्य here, so the plate is transcribed as printed: dhātumūlajīvamithyojarāśau. The sense, however, requires iti + oja — 'thus, in an ODD sign' — which the verse's own contrast with yugme (even) demands. The likeliest account is that this printing sets थ्य where त्य is meant. Recorded as printed with the discrepancy named, rather than corrected into the text.
  • About this edition: The first pāda is set with word-spaces although the metre requires it as a single eleven-syllable run.
  • About this edition: Annotation moved out of the `witnessA` field (CORRECTIONS.md C6): a witness field records the reading and nothing else, so this gloss now lives where notes live — “Śālinī; all four pādas 11”
  • About this edition: The 2026-08-12 note on this verse claimed an independent plate reading disagreed with us. It was withdrawn on 2026-08-13: the diff that produced it was defective (see CORRECTIONS.md C8) — it keyed verses by number alone in a text whose chapters restart numbering, and kept only the first line of each verse. The finished sweep, which read all 56 verses, places this verse's divergence among the anusvāra/visarga/repha marks that this worn printing drops throughout the book. Those are recorded as a property of the exemplar in the text-level note, not as an open question about the reading.
  • About this edition: SUPERSEDED 2026-09-08 — notes[1] and notes[5] above. Both said the printing sets मिथ्यो, that the mūla transcribes it as printed, and that the sense requires इति + ओज which only the English and Hindi carry. That gap between the mūla and the renderings is now closed from the evidence rather than by argument: the 1941 FIRST edition prints धातुं मूलं जीवमित्योजराशौ (PDF p.20, printed p.12, at 900 dpi), its own śloka index lists the verse under ध as धातुं मूलं जीवमित्योज I—7, the 1858 Bhaṭṭotpala lithograph lemmatises धातुं मूलं जीवम् इति ओजराशौ, and the Surat manuscript reads मित्यो. The 1978's थ्य is a wrong sort in a worn setting. The two accusatives धातुं मूलं are restored with it — the 1978 drops their anusvāras and our text had run them into a compound.

Adhyāya 2 · gamāgamādhyāyaḥ (Travelling and Returning)

2.1a reading here is contestedtwo exemplars weighed

वृषसिंहवृश्चिकघटैर्विद्धि स्थानं गमागमौ न स्तः । न मृतं न चापि नष्टं न रोगशांतिर्न चाभिभवः ॥ १ ॥

vṛṣasiṃhavṛścikaghaṭairviddhi sthānaṃ gamāgamau na staḥ | na mṛtaṃ na cāpi naṣṭaṃ na rogaśāṃtirna cābhibhavaḥ || 1 ||

Know that with Vṛṣabha, Siṃha, Vṛścika or Kumbha ⟨as the ascendant⟩ there is a position; going and coming are not. Neither death, nor loss, nor the cure of illness, nor defeat.

What the printed page actually sets, here

The Sanskrit above is our constituted text — what we judge the work to say, after weighing two independent exemplars and recording every disagreement. This edition's page differs, and the difference is shown rather than smoothed away:

वृषसिंहवृश्चिकघटैर्विद्धि स्थान गमागमौ नास्तिः । न मृतं न चापि नष्ट न रोगशांतिर्न चाभिभवः ॥ १ ॥

7 notes on this verse
  • Corroboration: The contested word is read as sthānaṃ ('position'), not the syānna that Witness A saw on the plate. Five independent lines of evidence agree: the Āryā count closes at exactly 30 mātrās against the second half's 27 only on this reading; the commentary on printed p.16 restates the rule as sthānaprāptiṃ vaden na cāgamanam; the edition's index summarises this verse as the acquisition of a position; the same index attests its three denials — death, cure of disease, defeat — in the verse's own order; and Witness B, reading the plate independently, produced sthānaṃ directly.
  • What the page would not yield: This is the worst-inked page of the book, and re-rendering it at 600 DPI produced no additional detail — the limit is the scan, not the raster. The reading therefore rests on convergent evidence rather than on the ink, and that is recorded rather than glossed over.
  • Where the Sanskrit is open: Witness B read viddhi where Witness A read vilagne. vilagne is kept: it is what the mātrā count requires (viddhi leaves the half two short) and what the sense needs. Witness B also confirms ghaṭaiḥ (with the ai) and nāsti without a visarga, both of which Witness A had flagged.
  • About this edition: Reader's annotation moved out of the `witnessA` field (CORRECTIONS.md C6) — a witness field records the reading and nothing else: “original read; witness A saw स्यान्न and नास्तिः on the plate. The change to स्थानं/नास्ति is a RECONSTRUCTION recorded in the notes, deliberately kept distinct from what was seen”
  • Where the Sanskrit is open: THE MĀTRĀ ARGUMENT FOR sthānaṃ TURNS ON A CONVENTION, AND THE APPARATUS DOES NOT SAY SO. Raised 2026-09-01 by the Hindi stage, which recounted this half-line as THIRTY-ONE mātrās — one over the Āryā's thirty — and concluded that our constituted reading is metrically impossible, the surplus coming from our mixing witness A's vilagne with witness B's sthānaṃ and nāsti. Checked by hand: vṛ1 ṣa1 siṃ2 ha1 vṛś2 ci1 ka1 gha1 ṭair2 vi1 lag2 ne2 = 17, then sthā2 naṃ2 ga1 mā2 ga1 mau2 nā2 = 29, plus the final sti. EVERYTHING TURNS ON THAT LAST SYLLABLE. Counted light it is 30 and the parallel note's claim stands exactly; counted heavy — and a half-line's final syllable is anceps, so it may be — it is 31. Neither count is wrong; they use different conventions for the same syllable. WHAT THIS DOES AND DOES NOT CHANGE. It does NOT unseat sthānaṃ, which rests on four further and independent supports the count never touched: the commentary on printed p.16 restating the rule as sthānaprāptiṃ vaden na cāgamanam, the edition's index summarising the verse as the acquisition of a position, that index attesting the three denials in the verse's own order, and witness B reading sthānaṃ directly off the plate. It DOES remove one of the five supports the parallel note lists, because a count that depends on an unstated convention is not evidence until the convention is stated. The same caveat applies to that note's use of the count against viddhi: viddhi is two mātrās shorter on either convention, so that argument survives; it is only the sthānaṃ/syānna discrimination that the count cannot make, both being three mātrās. This is recorded rather than resolved. Fixing it means stating which convention this edition's verses are scanned under, once, and applying it everywhere — not adjudicating this verse alone.
  • Where the Sanskrit is open: विलग्ने IS A REJECTED CONJECTURE, AND IT WAS NEVER ANYBODY'S READING OF A PAGE. It entered as witness A's transcription of the 1978 plate on 2026-08-11 and was kept by a metre argument recorded three separate times and wrong each time — viddhi was scanned as three mātrās when it is four, because dhi stands before sthānaṃ. On 2026-09-08 both plates were rendered at 900 dpi and read: the 1941 first edition sets विद्धि … न स्तः and the 1978 third edition sets विद्धि … नास्तिः. There is no ल on either page. The constituted text takes विद्धि from both settings and न स्तः from the 1941, witness D, candidate-202001, the Surat manuscript and Bhaṭṭotpala's gloss; the 1978's नास्तिः is a singular where the subject गमागमौ is dual, and is recorded in the diplomatic field rather than adopted.
  • Where the Sanskrit is open: SUPERSEDED 2026-09-08 — notes[2] and notes[5] above, and the retraction is arithmetical. notes[2] kept विलग्ने because 'the mātrā count requires it'; notes[5] re-checked the count, found the sthānaṃ argument turned on an unstated anceps convention, and concluded that 'viddhi is two mātrās shorter on either convention, so that argument survives'. It does not survive. विद्धि is FOUR mātrās, not three: धि stands before स्थानं and a short vowel before a conjunct is heavy by position. Both readings reach the Āryā's thirty, so the metre never chose between them. What does choose is that no witness anywhere reads विलग्ने. Everything notes[2] and notes[5] say about स्थानं is untouched by this and still stands.
2.2two exemplars weighed

तद्विपरीतं तु चरैर्द्विशरीरैर्मिश्रितं फलं भवति । लग्नेन्द्वोर्वक्तव्यं शुभदृष्ट्या शोभनमतोऽन्यत् ॥ २ ॥

tadviparītaṃ tu carairdviśarīrairmiśritaṃ phalaṃ bhavati | lagnendvorvaktavyaṃ śubhadṛṣṭyā śobhanamato'nyat || 2 ||

With the movable signs, the reverse of that; with the two-bodied signs the result is mixed. Of the ascendant and the Moon it is to be pronounced: by a benefic aspect, well — otherwise, the contrary.

3 notes on this verse
  • Where we differ: Both of Witness A's flagged spots were decided against it here. The plate at 600 DPI reads miśritaṃ (not miśraṃ) and vaktavyaṃ (not the vīkṣyaṃ A conjectured), and Witness B had both. The second is a real change of sense — vaktavyaṃ is 'is to be declared', where vīkṣyaṃ would have made the clause about aspecting. With miśritaṃ the Āryā count also closes, which A's reading did not.
  • Corroboration: The commentary on printed p.16 restates this verse's structure as cararāśau viparītaṃ, miśraṃ vācyaṃ dvimūrtyudaye — matching its movable-sign and dual-sign clauses respectively.
  • About this edition: Annotation moved out of the `witnessA` field (CORRECTIONS.md C6): a witness field records the reading and nothing else, so this gloss now lives where notes live — “lagnendvoḥ is a genitive DUAL — of the lagna and the Moon”
2.3two exemplars weighed

सुतशत्रुगतैः पापैः शत्रुर्मार्गान्निवर्तते । चतुर्थगैरपि प्राप्तः शत्रुर्भग्नो निवर्तते ॥ ३ ॥

sutaśatrugataiḥ pāpaiḥ śatrurmārgānnivartate | caturthagairapi prāptaḥ śatrurbhagno nivartate || 3 ||

With malefics gone to the fifth and the sixth, the enemy turns back from the road. With them in the fourth as well, the enemy, though he has arrived, turns back broken.

3 notes on this verse
  • Corroboration: The edition's index attests an enemy suffering defeat at II—3 and II—4, matching this verse and the next.
  • About this edition: Annotation moved out of the `witnessA` field (CORRECTIONS.md C6): a witness field records the reading and nothing else, so this gloss now lives where notes live — “Anuṣṭubh, all four pādas exactly 8; no doubtful akṣara; the house-words suta=5th and śatru=6th are in the verse itself”
  • Corroboration: PRAŚNA MĀRGA QUOTES THIS VERSE AT 24.64, AND THE DEFECTIVE COPY IS ITS OWN. PM prints चतुर्थगैरपि प्राप्त: शन्रुमार्गान्निवर्तते । चतुर्थगैरपि प्राप्तः शत्रुर्भग्नो निवर्तते — four pādas in which the opening चतुर्थगैरपि प्राप्तः stands twice. Pāda a of the verse, सुतशत्रुगतैः पापैः, has dropped out of PM's line and a dittography of pāda c has taken its place; PM's शन्रु is a further slip for शत्रु. Pādas b, c and d agree with our line letter for letter. So the testimonium confirms three quarters of this verse and locates a loss in the OTHER work — a collation paying in the direction it is not usually expected to.
2.4two exemplars weighed

झषालिकुंभकर्कटा रसातले यदा स्थितः । रिपोः पराजयस्तदा चतुष्पदैः पलायनम् ॥ ४ ॥

jhaṣālikuṃbhakarkaṭā rasātale yadā sthitaḥ | ripoḥ parājayastadā catuṣpadaiḥ palāyanam || 4 ||

When Mīna, Vṛścika, Kumbha or Karkaṭa stands in the fourth, then comes the enemy's defeat, and flight with the four-footed.

5 notes on this verse
  • About this edition: The 400dpi plate shows sthitaḥ with a clear visarga, i.e. the SINGULAR, although the subject is four named signs and the plural would agree. The disagreement is between the printing and its own grammar, and it is recorded rather than corrected.
  • About this edition: karkaṭā legitimately carries NO visarga here — karkaṭāḥ + ra gives karkaṭā ra by regular sandhi — so it must not be 'restored'.
  • Corroboration: The edition's index places the retreat of quadrupeds at II—4 under one heading and at II—1 under another. The II—1 entry is a typo: this verse carries the printed numeral 4, and the index's own RETREAT entry gives II—4.
  • About this edition: Annotation moved out of the `witnessA` field (CORRECTIONS.md C6): a witness field records the reading and nothing else, so this gloss now lives where notes live — “16+16; jhaṣa=Meena, ali=Vṛścika, rasātala=the 4th house”
  • Corroboration: PRAŚNA MĀRGA 24.87 QUOTES THIS VERSE AND PRINTS THE PLURAL. PM sets झषालिकुम्भकर्कटा - रसातले यदा स्थिताः। रिपोः पराजयस्तदा चतुष्पदैः पलायनम् — identical apart from स्थिताः, the nominative plural the four named signs require. Witness A had already flagged the ending as doubtful (स्थित⟨ाः?⟩), and notes[0] recorded the singular as a disagreement between the printing and its own grammar. An independent Kerala line now carries the grammatical form, so the singular is a defect of our printing rather than of the text. It is still recorded and not corrected: the 400 dpi plate shows the visarga plainly, and no plate we hold sets स्थिताः.
2.5a reading here is contestedtwo exemplars weighed

चरोदये शुभः स्थितः शुभं करोति यायिनाम् । अशोभनैरशोभनं स्थिरोदयेऽपि वा शुभम् ॥ ५ ॥

carodaye śubhaḥ sthitaḥ śubhaṃ karoti yāyinām | aśobhanairaśobhanaṃ sthirodaye'pi vā śubham || 5 ||

A benefic placed at the rising of a movable sign brings what is auspicious to those who set out; with the inauspicious, the inauspicious. At the rising of a fixed sign too, auspicious.

3 notes on this verse
  • Another reading: The edition records another reading for the fourth pāda: sthire'ṣṭame'pi vā śubham. This changes the doctrine rather than the spelling — the received pāda speaks of a fixed sign rising, the variant of the eighth house.
  • Where the Sanskrit is open: Both witnesses read the plate without the visarga of sthitaḥ; the eight-syllable count requires it, and this edition's visarga is a low dot that routinely fails to scan. Supplied on metre, with the printed state recorded.
  • About this edition: The 2026-08-12 note on this verse claimed an independent plate reading disagreed with us. It was withdrawn on 2026-08-13: the diff that produced it was defective (see CORRECTIONS.md C8) — it keyed verses by number alone in a text whose chapters restart numbering, and kept only the first line of each verse. The finished sweep, which read all 56 verses, places this verse's divergence among the anusvāra/visarga/repha marks that this worn printing drops throughout the book. Those are recorded as a property of the exemplar in the text-level note, not as an open question about the reading.
2.6a reading here is contestedtwo exemplars weighed

स्थिरे शशीचरोदये न चागमो रिपोर्यदा । तदागमं रिपोर्वदेद्विपर्यये विपर्ययम् ॥ ६ ॥

sthire śaśīcarodaye na cāgamo riporyadā | tadāgamaṃ riporvadedviparyaye viparyayam || 6 ||

When a movable sign rises and the Moon stands in a fixed one, the enemy does not come. In the reverse case one should declare his coming: the reverse for the reverse.

2 notes on this verse
  • Where we differ: Witness B read rāśī where Witness A read śaśī. A 600 DPI crop settles it for A: the plate reads śaśī, the Moon. rāśī would not be a well-formed locative here in any case.
  • Where the Sanskrit is open: Both witnesses read tadāgama without the anusvāra; the sixteen-syllable line requires tadāgamaṃ.
2.7two exemplars weighed

स्थिरे तु लग्नमागते द्विरात्मके तु चन्द्रमाः । निवर्तते रिपुस्तदा सुदूरमागतोऽपि सन् ॥ ७ ॥

sthire tu lagnamāgate dvirātmake tu candramāḥ | nivartate ripustadā sudūramāgato'pi san || 7 ||

But when the ascendant has come into a fixed sign and the Moon is in a two-bodied one, then the enemy turns back, though he has come a very long way.

3 notes on this verse
  • Corroboration: Witness A could not choose among dvirātmake, dviśarīre and dvidehe'tha for the dual-sign word, having established only that the metre needs four syllables there and so rules out the standard dvyātmake (three). Witness B read dvirātmake directly, and the two witnesses now agree. The edition's index independently indexes this verse under dual signs.
  • About this edition: Annotation moved out of the `witnessA` field (CORRECTIONS.md C6): a witness field records the reading and nothing else, so this gloss now lives where notes live — “pādas a, c, d each exactly 8”
  • Corroboration: PRAŚNA MĀRGA 24.62 QUOTES THIS VERSE VERBATIM — the cleanest testimonium the collation found in this chapter. PM prints स्थिरे तु लग्नमागते द्विरात्मके तु चन्द्रमाः । निवर्तते रिपुस्तदा सुदूरमागतोऽपि सन् with not one character different. It settles द्विरात्मके in particular: this chapter's ledger left the dual-sign word unresolved among द्विरात्मके, द्विशरीरे and द्विदेहेऽथ until witness B read it off the plate, and an independent transmission now prints it.
2.8two exemplars weighed

चरे शशी लग्नगतो द्विदेहः पथोऽर्द्धमागत्य निवर्तते रिपुः । विपर्यये चागमनं द्विधा स्यात्पराजयः स्यादशुभेक्षिते तु ॥ ८ ॥

care śaśī lagnagato dvidehaḥ patho'rddhamāgatya nivartate ripuḥ | viparyaye cāgamanaṃ dvidhā syātparājayaḥ syādaśubhekṣite tu || 8 ||

The Moon in a movable sign and a two-bodied sign in the ascendant: the enemy turns back having covered half the road. In the reverse case the coming would be twofold; but when aspected by an inauspicious graha, there would be defeat.

2 notes on this verse
  • About this edition: The two witnesses agree on this verse letter for letter, INCLUDING the second pāda, which counts twelve syllables against eleven in the other three. Since the reading is not in doubt, the metrical irregularity belongs to the printed text rather than to the transcription — which is worth stating plainly, because a flagged count usually means a misreading and here it does not.
  • Corroboration: PRAŚNA MĀRGA 24.63 QUOTES THIS VERSE AND DIVIDES THE LAST COMPOUND DIFFERENTLY. PM prints चरे शशी लग्नगतो द्विदेहः पथोऽर्धमागत्य निवर्तते रिपुः। विपर्यये चाऽऽगमनं द्विधा स्यात् पराजयः स्यादशुभे क्षते तु and renders the tail "when there is affliction by a malefic". Our line reads अशुभेक्षिते, aśubha + īkṣite, "aspected by a malefic" — the standard technical term and this text's own idiom. The two scan identically, so the metre cannot referee them; the lexicon can. क्षत, "wounded", is not how this text speaks of a malefic's action, while ईक्षित is (III.4 पापदृष्टाः, IV.5 सौम्यनिरीक्षिताः). PM's reading is best explained as ईक्षिते misdivided. Recorded, not adopted. PM also sets पथोऽर्धम् for the geminated पथोऽर्द्धम् of our line — the same word.
2.9two exemplars weighed

अर्कार्किज्ञसितानामेकोऽपि चरोदये यदा भवति । प्रवदेत्तदाशुगमनं वक्रगते नेति वक्तव्यम् ॥ ९ ॥

arkārkijñasitānāmeko'pi carodaye yadā bhavati | pravadettadāśugamanaṃ vakragate neti vaktavyam || 9 ||

When even one among the Sun, Saturn, Mercury or Venus stands at the rising of a movable sign, one should declare a swift going; when it has gone retrograde, 'no' is what must be said.

What the printed page actually sets, here

The Sanskrit above is our constituted text — what we judge the work to say, after weighing two independent exemplars and recording every disagreement. This edition's page differs, and the difference is shown rather than smoothed away:

अर्कार्किज्ञसितानामेकोऽपि चरोदये यदा भवति । प्रवदेत्तदाशुगमन वक्रगते नेति वक्तव्यम् ॥ ९ ॥

4 notes on this verse
  • About this edition: The printing runs the words together as vakragateneti. The parsing is vakragate + na + iti — 'when retrograde, "not" is to be said'. Read as an instrumental vakragatena, the negation vanishes entirely.
  • About this edition: Annotation moved out of the `witnessA` field (CORRECTIONS.md C6): a witness field records the reading and nothing else, so this gloss now lives where notes live — “arka=Sun, arki=Saturn, jña=Mercury, sita=Venus — the four grahas are named inside the verse”
  • About this edition: RE-READ AND SETTLED (2026-08-13). This locus was reported on 2026-08-12 as a divergence at a CASE ENDING — ours वक्रगते-, a plate reading वक्रगतैर्- — which would have changed the construction. It does not stand. Cropped at 14x WITH THE LINE ABOVE INCLUDED, the marks that read as a second stroke of ai belong to the descenders of चरोदये यदा in the line above, which hang into this line's mātrā zone; the marks that actually rise from this line's headline are single strokes. The plate reads वक्रगते नेति, which is witness A's reading and ours. A third blind reader of the same plate produced a third reading again (वक्रगतें नैति), which is what a worn line does to readers and is why this was settled on the ink rather than by vote.
  • About this edition: Verified at 14x in the same pass: the plate prints आशुगमन with NO anusvāra. The accusative गमनं is required by प्रवदेत् and is what we print, so this is our emendation and is now recorded as one rather than left to look like the plate's own reading.
2.10a reading here is contestedtwo exemplars weighed

स्थिरोदये जीवशनैश्चरेक्षिते गमागमौ नैव वदेत्तु पृच्छतः । त्रिपञ्चषष्ठा रिपुसङ्गमाय पापाश्चतुर्थां विनिवर्तनाय ॥ १० ॥

sthirodaye jīvaśanaiścarekṣite gamāgamau naiva vadettu pṛcchataḥ | tripañcaṣaṣṭhā ripusaṅgamāya pāpāścaturthāṃ vinivartanāya || 10 ||

At the rising of a fixed sign aspected by Jupiter and Saturn, one should declare neither going nor coming for the querent. Malefics in the third, fifth and sixth make for an encounter with the enemy; in the fourth, for his turning back.

What the printed page actually sets, here

The Sanskrit above is our constituted text — what we judge the work to say, after weighing two independent exemplars and recording every disagreement. This edition's page differs, and the difference is shown rather than smoothed away:

स्थिरोदये जीवशनैश्चरेक्षिते गमागमौ नैव वदेत्त पृच्छतः । त्रिपञ्चषष्ठा रिपुसङ्गमाय पापाश्चतुर्थां विनिवर्तनाय ॥ १० ॥

5 notes on this verse
  • Where the Sanskrit is open: ⚠ SUPERSEDED 2026-09-08 — read the later note on this verse first. The ⟨?⟩ this note describes is gone from the mūla and the word is settled as वदेत्तु, our own supply of the तु; see the ambiguity note and the closing note below. The note as written stands below, unaltered, because a record of what was believed is worth keeping. ——— Both witnesses read a doubled ta after vadet, and at 600 DPI the doubling is visible; the pāda then counts twelve as the metre requires, where A's original vadet gave eleven. The exact word is still open — vadettu (vadet tu) would be the natural form — so the doubling is transcribed with the doubt marked. Separately, caturthāṃ is confirmed for A at 600 DPI against B's caturthā, and the fourth pāda still counts one short.
  • About this edition: The 2026-08-12 note on this verse claimed an independent plate reading disagreed with us. It was withdrawn on 2026-08-13: the diff that produced it was defective (see CORRECTIONS.md C8) — it keyed verses by number alone in a text whose chapters restart numbering, and kept only the first line of each verse. The finished sweep, which read all 56 verses, places this verse's divergence among the anusvāra/visarga/repha marks that this worn printing drops throughout the book. Those are recorded as a property of the exemplar in the text-level note, not as an open question about the reading.
  • About this edition: ⚠ SUPERSEDED 2026-09-08 — read the later note on this verse first. The mūla no longer carries any ⟨?⟩ at this word, so the asymmetry this note explains no longer exists; the Hindi and the Sanskrit now agree in carrying no mark. The note as written stands below, unaltered, because a record of what was believed is worth keeping. ——— HINDI RENDERING — The mūla carries ⟨?⟩ at वदेत्त⟨?⟩ and the Hindi does not. The doubt is over a particle after the verb (वदेत् तु?), not over the verb: वदेत्, 'one should declare', is certain on either reading, so there is nothing in the Hindi for the mark to attach to. Recorded here so the absence of the bracket reads as a decision rather than as a silent resolution.
  • Where the Sanskrit is open: तु IS OURS AND NO WITNESS HAS IT. The 1941 first edition, the 1978 third edition and the sanskritdocuments e-text — two settings and two transmission lines — all print वदेत्त. The word is not Sanskrit, and the pāda is already a complete twelve-syllable Vaṃśastha, so what has dropped is a vowel sign and not an akṣara: this is the one place in the book where the defect is older than our exemplar. वदेत्तु is the only candidate that is both a word and metrically possible. Anyone who finds a witness for a different one should take it; the printed reading is preserved in `diplomatic` so nothing has to be recovered from a note.
  • About this edition: SUPERSEDED 2026-09-08 — notes[0] and notes[2] above, which described the ⟨?⟩ as a live doubt in the mūla. The mark is gone. It claimed an akṣara could not be read from the page; the page can be read, in two settings, and what it says is वदेत्त. notes[0] had already named वदेत्तु as the natural form, and that is now the constituted reading, marked in the ledger and in notes[3] as OUR supply. The Hindi's lack of a bracket, which notes[2] explains, is still right: the supply is a particle after a verb whose meaning is certain either way.
2.11a reading here is contestedtwo exemplars weighed

नागच्छति परचक्रं यदार्कचंद्रौ चतुर्थभवनस्थौ । बुधगुरुशुक्रा हिबुके यदा तदा शीघ्रमायाति ॥ ११ ॥

nāgacchati paracakraṃ yadārkacaṃdrau caturthabhavanasthau | budhaguruśukrā hibuke yadā tadā śīghramāyāti || 11 ||

The enemy's force does not come when both the Sun and the Moon stand in the fourth house. When Mercury, Jupiter and Venus are in the fourth, then it comes swiftly.

1 note on this verse
  • Where the Sanskrit is open: The anusvāra of paracakraṃ is faint. The two dual endings in the line — arkacandrau and bhavanasthau — are legible and corroborate one another.
2.12a reading here is contestedtwo exemplars weighed

मेषधनुःसिंहवृषा यद्युदयस्था भवंति हिबुके वा । शत्रुर्निवर्तति तदा ग्रहसहिता वा वियुक्ता वा ॥ १२ ॥

meṣadhanuḥsiṃhavṛṣā yadyudayasthā bhavaṃti hibuke vā | śatrurnivartati tadā grahasahitā vā viyuktā vā || 12 ||

If Meṣa, Dhanus, Siṃha or Vṛṣabha stands at the rising, or in the fourth, then the enemy turns back — whether those signs are joined with grahas or empty of them.

2 notes on this verse
  • About this edition: Both witnesses read nivartati, not the nivartate used at II.7 and II.8. Since the two agree, the non-standard form is what this edition prints, and it is kept as printed.
  • Where the Sanskrit is open: Witness B dropped the repha of śatrur; śatruḥ + nivartati gives śatrur nivartati by regular sandhi, so A's reading is kept.
2.13two exemplars weighed

स्थिरराशौ यद्युदये शनिर्गुरुर्वा स्थितस्तदा शत्रुः । उदये रविर्गुरुर्वा चरराशौ स्यात्तदागमनम् ॥ १३ ॥

sthirarāśau yadyudaye śanirgururvā sthitastadā śatruḥ | udaye ravirgururvā cararāśau syāttadāgamanam || 13 ||

If, at the rising of a fixed sign, Saturn or Jupiter stands there, then the enemy [halts]. At the rising of a movable sign, with the Sun or Jupiter, there would be his coming.

2 notes on this verse
  • Corroboration: Witness A flagged the same doubtful cluster twice, once after śani- and once after ravi-, noting only that it must be one form and not two. Witness B read gurur vā in both places, which is the disjunctive the verse's parallel structure needs: 'Saturn or Jupiter' in the fixed-sign case, 'the Sun or Jupiter' in the movable.
  • Corroboration: PRAŚNA MĀRGA 24.68 QUOTES THIS VERSE WITH TWO REPHAS OUR PRINTING DOES NOT HAVE. PM sets स्थिरराशौ यद्युदये शनिर्गुरुर्वा स्थितस्तदा शत्रोः। उदये रविर्गुरुर्वा चरराशौ स्यात्तदाऽऽगमनम् — शनिर्गुरुर्वा and रविर्गुरुर्वा against शनिगुरुर्वा and रविगुरुर्वा, and the genitive शत्रोः against the nominative शत्रुः. This matters because a dropped repha is precisely what the 1978 setting does everywhere, as the text-level `exemplar` note records, and the 1941 collation of 2026-09-08 did not open this locus. So the divergence is left OPEN in the honest sense: it is recorded here, and the 1941 plate at printed p.23 has still to be read at this word. The metre was deliberately not used to decide it — this book has already had three cruxes settled by a mātrā count that turned out to be wrong. [2026-09-23: the mūla now reads शनिर्गुरुर्वा and रविर्गुरुर्वा; this sentence names शनिगुरुर्वा and रविगुरुर्वा, the forms our text carried before the mūla-verification pass of 2026-09-23 corrected them, and they are kept as history.]
2.14two exemplars weighed

ग्रहः सर्वोत्तमबलो लग्नाद्यस्मिन् गृहे स्थितः । मासैस्तत्तुल्यसंख्याकैर्निवृत्तिं यातुरादिशेत् ॥ १४ ॥

grahaḥ sarvottamabalo lagnādyasmin gṛhe sthitaḥ | māsaistattulyasaṃkhyākairnivṛttiṃ yāturādiśet || 14 ||

In whichever house from the ascendant the graha of greatest strength stands, by months equal in number to it one should appoint the return of him who has gone.

3 notes on this verse
  • Where we differ: CORRECTED 2026-08-11 by the adversarial verify pass. Two errors stood here. First, the mūla still read lagnāccāsmin although this file's own ledger had already resolved the spot to lagnād yasmin — the resolution was recorded and never applied, so the record contradicted itself. Second, and worse, the line reads māsais tat-tulya-, with the tat that Witness B supplied and that I had 'refuted' from a 600 dpi crop. Re-read at maximum magnification there are unmistakably two akṣara bodies between सै and ल्य. Gemini was right; I was wrong; the line counts sixteen syllables as the metre requires, and there is no defect here.
  • Corroboration: The edition's index places the time of return of a person gone out to fight at II—14, matching this verse's month-counting rule.
  • About this edition: CONFIRMED ON THE PLATE 2026-09-08. notes[0] and the ledger row had said since 2026-08-11 that तत् IS printed and that the line counts sixteen syllables, while the mūla and the row's own resolution still spelled ततुल्य — one virāma short of a word. The 1941 first edition prints मासैस्तत्तुल्यसंख्याकैः (PDF p.29, printed p.21, at 900 dpi); witness D prints तत्तुल्य. Both now agree with the mūla. ⚠ features/jyotisha/text-catalog.ts still tells readers this verse carries 'real metrical defects'; that claim was retracted on 2026-08-11 and is a code fix outside this corpus.
2.15a reading here is contestedtwo exemplars weighed

चरांशस्थे ग्रहे तस्मिन् कालमेवं विनिर्दिशेत् । द्विगुणं स्थिरभागस्थे त्रिगुणं द्व्यात्मकांशके ॥ १५ ॥

carāṃśasthe grahe tasmin kālamevaṃ vinirdiśet | dviguṇaṃ sthirabhāgasthe triguṇaṃ dvyātmakāṃśake || 15 ||

When that graha stands in a movable navāṃśa, one should state the time so; doubled, when it stands in a fixed portion; trebled, in a ⟨dual-natured⟩ navāṃśa.

What the printed page actually sets, here

The Sanskrit above is our constituted text — what we judge the work to say, after weighing two independent exemplars and recording every disagreement. This edition's page differs, and the difference is shown rather than smoothed away:

चरांशस्थे ग्रहे तस्मिन् कालमेवा विनिर्दिशेत् । द्विगुण रिथरभागस्थे त्रिगुण ह्यात्मकांशके ॥ १५ ॥

6 notes on this verse
  • About this edition: DIPLOMATIC vs CONSTITUTED. The Sanskrit above is the CONSTITUTED text — what we judge the work to say. This printing differs: कालमेवा, not कालमेवं. The final compound reads ह्यात्म — and Witness D prints ह्यात्मकांशके too, so this is what the tradition transmits, however little sense it makes; only the doctrine argues for द्व्यात्मक. Both are now recorded separately, because an edited reading and a read one are different objects and sharing one field is what let this record claim more precision than it had (CORRECTIONS.md C1.5).
  • What the page would not yield: The word rendered ⟨two-bodied⟩ is NOT read. All three witnesses produced something different (hyātma / dyātma / hvātma) and none is a word. Angle brackets mark it as supplied by us: the verse's three cases are movable, fixed and — by the sequence — dual, and dvyātmaka is what that requires, but no reader has seen it on the page. Until 2026-08-12 the translation delivered 'two-bodied' plainly, i.e. a conjecture presented as text; that is corrected here.
  • Where the Sanskrit is open: Both witnesses read dviguṇa and triguṇa without the anusvāra the accusative wants.
  • Where we differ: A blind reader who had never seen our text transcribed this plate on 2026-08-12 (docs/VISION_EXTRACTION.md); the whole 56-verse book was swept and every divergence adjudicated on the plate at magnification. A LACUNA CLOSED. We published ⟨?⟩यात्मकांशके with the first akṣara unread. The plate prints ह्यात्मकांशके and witness D agrees, so the ink is not in doubt — but ह्यात्मक is not a word here. The edition's own commentary settles it: the tripled period is the one it assigns to a dual-natured navāṃśa, and dual-natured is dvyātmaka. द्व्य and ह्य are a known confusion in this face. Emended to द्व्यात्मकांशके, with the printing recorded diplomatically.
  • About this edition: The same line prints रिथर for स्थिर — a transposed-sort misprint, not a variant. Kept in the diplomatic reading only. The plate also drops the anusvāra of द्विगुणं and त्रिगुणं, which is this printing's habit rather than a reading (see the text-level note).
  • About this edition: PARTLY CLOSED 2026-09-08. notes[1] said no reader had seen द्व्यात्मक on a plate. That is still true and is now true of one more plate: the 1941 FIRST edition also prints ह्यात्मकांशके (PDF p.30, printed p.22, at 900 dpi), so both Sastri settings, witness D, the Maharishi copy and all three vision witnesses agree on a reading that is not a word. द्व्यात्मक remains OURS. What has changed is that the English no longer hides it — notes[5]'s complaint that '⟨⟩ appears in the Hindi and not in the English' is answered, and the ledger now carries a row that records the emendation itself rather than only the disagreement. In the emendation's favour, and the record did not note it: the 1941's own English on that page reads 'The period has to be trebled when the Navamsa in question is a dual one'.
2.16two exemplars weighed

यातुर्विलग्नाज्जामित्रभवनाधिपतिर्यदा । करोति वक्रमावृत्तेः कालान्तं ब्रुवते परे ॥ १६ ॥

yāturvilagnājjāmitrabhavanādhipatiryadā | karoti vakramāvṛtteḥ kālāntaṃ bruvate pare || 16 ||

Should the ruler of the seventh house, counted from the traveller's ascendant, turn retrograde — others pronounce that to be the limit of the time of his return.

2 notes on this verse
  • About this edition: jāmitra, the term used here for the seventh house, is one of several Greek-derived technical terms sitting inside this mūla (with drekkāṇa at VI.2 and āsphujit in the commentator's opening benediction). Recorded as an observation about the text's vocabulary; the etymological claim itself belongs in research/sources with a citation.
  • Corroboration: The verse closes bruvate pare, 'others say', i.e. it reports a rival school rather than the author's own view. The edition's index names that school at II—16.
2.17two exemplars weighed

उदयर्क्षाच्चन्द्रर्क्षं भवति च यावद्दिनानि तावद्भिः । आगमनं स्याच्छत्रोर्यदि मध्ये न ग्रहः कश्चित् ॥ १७ ॥

udayarkṣāccandrarkṣaṃ bhavati ca yāvaddināni tāvadbhiḥ | āgamanaṃ syācchatroryadi madhye na grahaḥ kaścit || 17 ||

By as many days as there are signs from the rising sign to the Moon's sign, the enemy's arrival would come — provided no graha stands between.

2 notes on this verse
  • Corroboration: The edition's index places the time at which an enemy's arrival may be expected at II—17, matching this verse's day-count rule.
  • About this edition: Reader's annotation moved out of the `witnessA` field (CORRECTIONS.md C6) — a witness field records the reading and nothing else: “cchatroḥ = syāt + śatroḥ). Chapter colophon follows on the same plate: गमागमाध्यायो द्वितीयः, confirming 17 ślokas”

Adhyāya 3 · jayaparājayaḥ (Victory and Defeat)

3.1a reading here is contestedtwo exemplars weighed

दशमोदयसप्तमगाः सौम्या नगराधिपस्य विजयकराः । आरार्की ज्ञगुरुसिताः प्रभङ्गदौ विजयदा नवमे ॥ १ ॥

daśamodayasaptamagāḥ saumyā nagarādhipasya vijayakarāḥ | ārārkī jñagurusitāḥ prabhaṅgadau vijayadā navame || 1 ||

Benefics gone to the tenth, to the ascendant and to the seventh bring victory to the lord of the city. In the ninth, Mars and Saturn are the two that bring utter rout, while Mercury, Jupiter and Venus are the givers of victory.

2 notes on this verse
  • Where the Sanskrit is open: The verse interleaves two clauses rather than stating them in sequence, and only the number-endings disentangle them: ārārkī is DUAL (āra = Mars, arkī = Saturn) and pairs with prabhaṅgadau (dual); jñagurusitāḥ is PLURAL (Mercury, Jupiter, Venus) and pairs with vijayadā (plural). A translation that follows the word order rather than the endings will attach the wrong outcome to the wrong grahas. Witness B dropped the repha of ārārkī, which would collapse the dual that carries the whole distinction.
  • Corroboration: The edition's index attests victory for the governor of a town at III—1 and a complete defeat at III—1, matching the verse's two clauses.
3.2two exemplars weighed

पौरास्तृतीयभवनाद्धर्माद्वा यायिनः शुभैः शुभदाः । व्ययदशमाये पापाः पुरस्य नेष्टाः शुभा यातुः ॥ २ ॥

paurāstṛtīyabhavanāddharmādvā yāyinaḥ śubhaiḥ śubhadāḥ | vyayadaśamāye pāpāḥ purasya neṣṭāḥ śubhā yātuḥ || 2 ||

The townsmen are read from the third house, and those who march from the ninth; with benefics there, these give what is auspicious. Malefics in the twelfth, the tenth and the eleventh are unwelcome for the city — and benefics there, for him who marches.

4 notes on this verse
  • Corroboration: śubhai is confirmed at 400dpi, as Witness B read it. dharmādvā (dharmāt + vā) is confirmed independently by the facing commentary, which discusses the vā in that word explicitly — useful, because Witness B garbled this line badly (paurāsmṛtīya-, kramāddvā) and carries little weight on it.
  • Corroboration: The edition's index attests this verse at III—2 under the townsmen, the besieging party and the town's chief — matching paurāḥ, yāyinaḥ and purasya.
  • About this edition: Annotation moved out of the `witnessA` field (CORRECTIONS.md C6): a witness field records the reading and nothing else, so this gloss now lives where notes live — “house-words internal: tṛtīya 3rd, dharma 9th, vyaya 12th, daśama 10th, āya 11th”
  • About this edition: CLOSED ON THE PLATE 2026-09-08. The mūla read शुभै, which is not a Sanskrit form — the word is the instrumental plural शुभैः and its visarga had been dropped, as this printing drops visargas throughout. The 1941 first edition prints शुभैः (PDF p.32, printed p.24); witness D carries the visarga too. The other three plurals in the same verse — शुभदाः, पापाः, नेष्टाः — were already right, which is what made the missing one visible.
3.3two exemplars weighed

नृराशिसंस्था ह्युदये शुभाः स्युर्व्ययायसंस्थाश्च यदा भवन्ति । तदाशु सन्धिं प्रवदेन्नृपाणां पापैर्द्विदेहोपगतैर्विरोधम् ॥ ३ ॥

nṛrāśisaṃsthā hyudaye śubhāḥ syurvyayāyasaṃsthāśca yadā bhavanti | tadāśu sandhiṃ pravadennṛpāṇāṃ pāpairdvidehopagatairvirodham || 3 ||

When benefics stand in human signs at the rising, and when they stand also in the twelfth and the eleventh, then one should swiftly declare a treaty between the kings; with malefics gone to two-bodied signs, declare conflict.

2 notes on this verse
  • About this edition: The pāda boundary does not fall at the printed line-break: pāda a ends at śubhāḥ syur and pāda b is vyayāyasaṃsthāśca yadā bhavanti. Split at the printed break instead, the pādas come out 11 and 12 and look defective; split metrically, all four count 11.
  • Corroboration: The edition's index attests a treaty of peace between two kings at III—3, matching sandhiṃ pravaden nṛpāṇām.
3.4two exemplars weighed

केन्द्रोपगताः सौम्याः सौम्यैर्दृष्टा नृलग्नगाः प्रीतिम् । कुर्वन्ति पापदृष्टाः पापास्तेष्वेव विपरीतम् ॥ ४ ॥

kendropagatāḥ saumyāḥ saumyairdṛṣṭā nṛlagnagāḥ prītim | kurvanti pāpadṛṣṭāḥ pāpāsteṣveva viparītam || 4 ||

Benefics gone to the angles, aspected by benefics, standing in a human ascendant, bring about amity. Malefics, aspected by malefics, bring about the reverse in those very places.

1 note on this verse
  • Corroboration: Witness A could not read the fourth pāda's cluster at all, recording only that it looked like pāpābhyetyeva and did not parse. Witness B read pāpāsteṣveva — pāpāḥ + teṣu + eva, 'malefics … in those very houses' — and a 600 DPI crop confirms it letter for letter. A genuine gap in Witness A, closed by the second witness and then verified on the plate.
3.5a reading here is contestedtwo exemplars weighed

द्वितीये वा तृतीये वा गुरुशुक्रौ यदा तदा । आश्वेवागच्छते सेना प्रवासो च न संशयः ॥ ५ ॥

dvitīye vā tṛtīye vā guruśukrau yadā tadā | āśvevāgacchate senā pravāso ca na saṃśayaḥ || 5 ||

When Jupiter and Venus stand in the second or in the third, then the army comes back at once; and there is a journey, beyond doubt.

6 notes on this verse
  • What the page would not yield: ⚠ SUPERSEDED 2026-09-08 — read the later note on this verse first. The 1941 first edition prints आश्वेवागच्छते with a long आ and unbroken ink, so nothing here was unread, ⟨at once⟩ is a rendering rather than a supply, and both marks are gone from the mūla. The note as written stands below, unaltered, because a record of what was believed is worth keeping. ——— ⟨at once⟩ is supplied, not read: the initial vowel is marked ⟨?⟩ in the mūla and the two witnesses differ (āśu 'at once' vs the aśva iva 'like a horse' that the note itself warns is easy to fall into). Until 2026-08-12 the English simply said 'at once', resolving a spot the record calls unresolved.
  • Where we differ: pravāso is printed NOMINATIVE. The earlier translation ('about the journey there is no doubt') rendered it as the locative pravāse that Witness A had only floated as a possibility, and narrowed na saṃśayaḥ — a formulaic whole-clause 'without doubt' — onto that one word. Now rendered from the case that is actually printed.
  • Corroboration: The edition's index attests the return of one's army at III—5, and separately mislocates the same statement at III—9, which is impossible (Adhyāya III has five ślokas).
  • About this edition: Reader's annotation moved out of the `witnessA` field (CORRECTIONS.md C6) — a witness field records the reading and nothing else: “Chapter colophon follows: जयपराजयो नाम तृतीयोऽध्यायः, confirming 5 ślokas and giving the chapter its name”
  • Where the Sanskrit is open: AYER READS प्रवासी AND गच्छति; WE READ प्रवासो AND गच्छते, AND THE 1941 PLATE IS WITH US. V.A.K. Ayer's Taraporevala printing sets आश्वेवाऽऽगच्छति सेनाप्रवासी च न संशयः and translates 'without doubt the army or person from abroad will arrive soon' — a nominative प्रवासी coordinated with सेना by च. The 1941 first edition and the 1978 third edition both print प्रवासो and गच्छते, and witness D prints प्रवासो and गच्छते too, so two transmission lines agree against Ayer. On गच्छते the metre is also with us: pāda c is odd, position 6 must be heavy, and ति is light. Recorded, not adopted.
  • About this edition: SUPERSEDED 2026-09-08 — notes[0] above. It said ⟨at once⟩ was 'supplied, not read', because the initial vowel was marked ⟨?⟩ and the witnesses split between आशु ('at once') and अश्व इव ('like a horse'). The 1941 plate prints आश्वेवागच्छते with a long आ and unbroken ink (PDF p.35, printed p.27, at 900 dpi), and the Surat manuscript reads the same. आशु is in the text, so 'at once' is a rendering and not a supply, and the brackets come off in both languages. The अश्व इव reading belongs to witness D's short अ and is recorded, not adopted.

Adhyāya 4 · śubhāśubhalakṣaṇādhyāyaḥ (Auspicious and Inauspicious Indications)

4.1two exemplars weighed

केन्द्रत्रिकोणेषु शुभस्थितेषु पापेषु केन्द्राष्टमवर्जितेषु । सर्वार्थसिद्धिं प्रवदेन्नराणां विपर्यवस्थेषु विपर्ययः स्यात् ॥ १ ॥

kendratrikoṇeṣu śubhasthiteṣu pāpeṣu kendrāṣṭamavarjiteṣu | sarvārthasiddhiṃ pravadennarāṇāṃ viparyavastheṣu viparyayaḥ syāt || 1 ||

When benefics stand in the angles and the trines, and malefics stand clear of the angles and the eighth, one should declare for people the accomplishment of every aim. When they stand reversed, the reverse would follow.

What the printed page actually sets, here

The Sanskrit above is our constituted text — what we judge the work to say, after weighing two independent exemplars and recording every disagreement. This edition's page differs, and the difference is shown rather than smoothed away:

केन्द्रत्रिकोणेषु शुभस्थितेषु पापेषु केन्द्राष्टमवर्जितेषु । सर्वार्थसिद्धिं प्रवदेन्नराणां विपर्यवस्थेषु विदर्ययः स्यात् ॥ १ ॥

5 notes on this verse
  • Another reading: The edition records śubhānviteṣu as another reading for śubhasthiteṣu in the first pāda, and remarks that the variant states the rule more clearly.
  • About this edition: The initial of the second kendrāṣṭamavarjiteṣu is malformed in this printing and reads to the eye almost as ne rather than ke. Witness B read kendrāṣṭama- plainly, pāda a establishes kendra, and the rule itself requires it ('malefics anywhere EXCEPT the angles and the eighth'), so the doubt is graphic only and is recorded here rather than carried in the text.
  • Corroboration: The edition's index attests the fulfilment of one's cherished objects at IV—1, matching sarvārthasiddhim.
  • Where we differ: A blind reader who had never seen our text transcribed this plate on 2026-08-12 (docs/VISION_EXTRACTION.md); the whole 56-verse book was swept and every divergence adjudicated on the plate at magnification. CORRECTED HERE, and the correction is really about our witnesses. We published विपर्ययस्थेषु on the strength of witnesses A and B, both of which are readings of THIS plate. The plate does not say that: at 20x it prints विपर्यवस्थेषु, and witness D — a different exemplar altogether — prints विपर्यवस्थेषु too. So A and B silently normalised what they saw into what they expected, and our text inherited the normalisation as though it were attested. viparyavastha ("being in a reversed state") is good Sanskrit and both independent exemplars carry it, so it is the text. Both readings scan as Indravajrā, so the metre never flagged it.
  • About this edition: The plate prints विदर्ययः for विपर्ययः in the last pāda — a plain compositor's error (द set for प; the two are adjacent in the case and the word recurs correctly earlier in the same line). Recorded in the diplomatic reading, emended in the mūla.
4.2a reading here is contestedtwo exemplars weighed

त्रिपञ्चलाभास्तमयेषु सौम्या लाभप्रदा नेष्टफलाश्च पापाः । तुलाथ कन्या मिथुनं घटश्च नृराशयस्तेषु शुभं वदन्ति ॥ २ ॥

tripañcalābhāstamayeṣu saumyā lābhapradā neṣṭaphalāśca pāpāḥ | tulātha kanyā mithunaṃ ghaṭaśca nṛrāśayasteṣu śubhaṃ vadanti || 2 ||

Benefics in the third, the fifth, the eleventh and the seventh give gain, and malefics there give unwanted fruit. Tulā, then Kanyā, Mithuna and Kumbha are the human signs; in these, the auspicious is declared.

What the printed page actually sets, here

The Sanskrit above is our constituted text — what we judge the work to say, after weighing two independent exemplars and recording every disagreement. This edition's page differs, and the difference is shown rather than smoothed away:

त्रिपञ्चलाभास्तमयेषु सौम्या लाभप्रदा नेष्टफलाश्च पापाः । तुलाथ कन्या मिथुनं घटश्च नृराशयस्तेषु शुभं वदति ॥ २ ॥

6 notes on this verse
  • Where we differ: The plate reads vadati, and the record originally justified that with a syllable count which was simply wrong: I had claimed vadati gives ten syllables against vadanti's eleven. Both give eleven. There is no metrical defect in this pāda; the earlier claim that there was one is retracted.
  • Where the Sanskrit is open: Witness B also read naṣṭa- for neṣṭa-; A's reading is kept, since na + iṣṭa ('not desired') is the sense the verse needs.
  • Corroboration: The commentary on printed p.27 quotes this verse's biped-sign list with mithuno and an ending nṛrāśaya iti, where the mūla here prints mithunaṃ. The edition cites its own text with a different reading three pages earlier.
  • About this edition: Annotation moved out of the `witnessA` field (CORRECTIONS.md C6): a witness field records the reading and nothing else, so this gloss now lives where notes live — “11×4 exact; houses internal: tri 3rd, pañca 5th, lābha 11th, astamaya 7th”
  • Where we differ: RE-REVERSED 2026-09-08 — notes[0] above. That note retracted a metre argument for वदन्ति and said flatly 'There is no metrical defect in this pāda'. The retraction was right about the syllable COUNT (both readings give eleven) and wrong about the metre, because it counted syllables instead of weighing them. Upajāti pins position 10 heavy and this verse's own three sister-pādas obey it — …yeṣu SAUmyā · …lāś ca PĀpāḥ · …naṃ ghaTAŚ ca — while वदति puts a light syllable there. And the question is now settled without metre at all: the 1941 FIRST edition prints वदन्ति (PDF p.37, printed p.29). The 1978 setting dropped the न् sort, leaving a visible gap before ॥ २ ॥. The 1978 reading is kept in the diplomatic field.
  • Corroboration: THE MANUSCRIPT WRITES A NASAL HERE, SO IT IS ON THE SIDE OF वदन्ति. MS CGV-PSS-0172-1245, p.5 upper leaf, was magnified on 2026-09-08 at the word this verse turns on: it sets …नृराशयस्तेषु शुभं वदंत°, with an unmistakable nasal between द and त, closing ॥२९॥. The final vowel could not be resolved at that magnification and is not claimed. What is claimed is the nasal, and it is what the 1978 plate lacks: that plate sets वदति with a conspicuous gap before the numeral where the न् sort belongs. So a fourth line — after witness A, witness D and the Maharishi copy — carries the consonant, and the 1941 first edition prints वदन्ति whole.
4.3two exemplars weighed

स्थानप्रदा दशमसप्तमगाश्च सौम्या मानार्थदाः स्वसुतलग्नगता भवंति । पापा व्ययायसहिता न शुभप्रदाः स्युर्लग्ने शशी न शुभदो दशमे शुभश्च ॥ ३ ॥

sthānapradā daśamasaptamagāśca saumyā mānārthadāḥ svasutalagnagatā bhavaṃti | pāpā vyayāyasahitā na śubhapradāḥ syurlagne śaśī na śubhado daśame śubhaśca || 3 ||

Benefics gone to the tenth and the seventh give position; gone to the second, the fifth and the ascendant, they give honour and wealth. Malefics attended by the twelfth and the eleventh would give nothing auspicious. The Moon in the ascendant is no giver of the auspicious; in the tenth she is auspicious.

4 notes on this verse
  • Corroboration: Witness A first read sva before lagne, giving svalagne, and corrected it to syur on metre alone: this verse is Vasantatilakā at fourteen syllables per pāda, and sva leaves pāda c at thirteen and pāda d at fifteen. Witness B read syur directly, and the commentary on the same page quotes the clause as beginning at lagne — placing the word boundary exactly where the metre put the pāda break. Three independent lines, same answer.
  • About this edition: sva DOES occur in this verse, in pāda b, where it is the second house (wealth) — the commentary says so explicitly. A real word in the neighbourhood is what made the wrong reading plausible.
  • Corroboration: The edition's index attests a yoga bringing honour and wealth at IV—3, matching mānārthadā (māna = honour, artha = wealth) in the same order.
  • Corroboration: PRAŚNA MĀRGA 24.47 QUOTES THIS VERSE ALMOST WHOLE, AND IT CONFIRMS THE VISARGA RESTORED ON 2026-09-08. PM prints स्थानप्रदा दशमसप्तमगाश्च सौम्या मानार्थदास्स्वसुतलग्नगता भवन्ति। पापव्ययाऽऽयसहिता न शुभप्रदास्स्यु- लग्ने शशी न शुभदो दशमे शुभश्व. Its मानार्थदास्स्व is the Kerala doubling of मानार्थदाः स्व, so an independent line carries the visarga this verse gained on 2026-09-08 when the 1941 first edition was opened; and its शुभप्रदास्स्यु is the same doubling of शुभप्रदाः स्युर्, which independently confirms the word boundary this chapter's ledger settled at discrepancies[2]. PM's शुभश्व at the end is a slip for शुभश्च. One real divergence: PM makes पापव्ययाऽऽय° a compound where our line has पापा standing free as a nominative plural.
4.4a reading here is contestedtwo exemplars weighed

इन्दुं द्विसप्तदशमायरिपुत्रिसंस्थं पश्येद्गुरुः शुभफलं प्रमदाकृतं स्यात् । लग्नत्रिधर्मसुतनैधनगाश्च पापाः कार्यार्थनाशभयदाः शुभदाः शुभाश्च ॥ ४ ॥

induṃ dvisaptadaśamāyariputrisaṃsthaṃ paśyedguruḥ śubhaphalaṃ pramadākṛtaṃ syāt | lagnatridharmasutanaidhanagāśca pāpāḥ kāryārthanāśabhayadāḥ śubhadāḥ śubhāśca || 4 ||

Should Jupiter look upon the Moon as she stands in the second, the seventh, the tenth, the eleventh, the sixth or the third, an auspicious result would follow, brought about by a young woman. And malefics gone to the ascendant, the third, the ninth, the fifth and the eighth give ruin of one's undertaking, loss of wealth and fear; benefics there give what is auspicious.

What the printed page actually sets, here

The Sanskrit above is our constituted text — what we judge the work to say, after weighing two independent exemplars and recording every disagreement. This edition's page differs, and the difference is shown rather than smoothed away:

इन्दु द्विसप्तदशमायरिपुत्रिसंस्थ पश्येद्गुरुः शुभफलं ऽमादाकृत स्यात् । लग्नत्रिधर्मसुतनैधनगाश्च पापाः कार्यार्थनाशभयदाः शुभदाः शुभाश्च ॥ ४ ॥

4 notes on this verse
  • About this edition: DIPLOMATIC vs CONSTITUTED. The Sanskrit above is the CONSTITUTED text — what we judge the work to say. This printing differs: इन्दु, संस्थ and कृत all WITHOUT the accusative anusvāra our text supplies; the same line prints clear anusvāras elsewhere, so their absence is real. Both are now recorded separately, because an edited reading and a read one are different objects and sharing one field is what let this record claim more precision than it had (CORRECTIONS.md C1.5).
  • What the page would not yield: ⚠ SUPERSEDED 2026-09-08 — read the later note on this verse first. Nothing in this slot was damaged. The 1978 plate prints a clean avagraha and the 1941 first edition prints the word whole as प्रमदाकृतं; the mark is gone from the mūla. The note as written stands below, unaltered, because a record of what was believed is worth keeping. ——— Pāda b carries a five-syllable slot that neither witness could read. The metre pins its length exactly — paśyed guruḥ śubhaphalaṃ is eight syllables and syāt is the fourteenth — and a 600 DPI crop shows the ending clearly as -mādākṛtaṃ with only the FIRST akṣara damaged. So the gap has narrowed from five unknown syllables to one, and that one is left unread. The edition's English and index both speak of gain through a woman, which gives the sense and not the letters; supplying rāmā- on that basis would be exactly the substitution this protocol exists to prevent.
  • Corroboration: Witness A read pāda c as lagnatridharmeṣu-tanaidhana- and could not account for one syllable. Witness B read lagnatridharmasutanaidhana-, and a 600 DPI crop confirms suta. The pāda then counts exactly fourteen, and the five houses it names are the 1st, 3rd, 9th, 5th and 8th.
  • About this edition: CLOSED ON THE PLATE 2026-09-08. The mūla carried ⟨?⟩मादाकृतं and the app told readers an akṣara could not be read from the page. Nothing was unreadable. On the 1978 plate the first sign is a cleanly printed avagraha and the pāda is one syllable short, so a sort had dropped; the 1941 FIRST edition prints the word whole — प्रमदाकृतं, pramadā-kṛtaṃ, 'brought about by a young woman' (PDF p.38, printed p.30, magnified from a 900 dpi render). The म is short, which rules out both प्रमादाकृतं and witness D's प्रमादकृतं, and Vasantatilakā requires a light syllable at position 10, which only प्रमदा° gives. The edition's own English on the same page says the same thing — 'through a damsel' — and repeats it on the next page as 'gain and other things through the help of women'. Witness B's ऽमदाकृत was right about the short म and was overridden in August.
4.5a reading here is contestedtwo exemplars weighed

शुभग्रहाः सौम्यनिरीक्षिताश्च विलग्नसप्ताष्टमपञ्चमस्थाः । त्रिषड्दशाये च निशाकरः स्याच्छुभं भवेद्रोगनिपीडितानाम् ॥ ५ ॥

śubhagrahāḥ saumyanirīkṣitāśca vilagnasaptāṣṭamapañcamasthāḥ | triṣaḍdaśāye ca niśākaraḥ syācchubhaṃ bhavedroganipīḍitānām || 5 ||

Benefic grahas, aspected by benefics, standing in the ascendant, the seventh, the eighth and the fifth — and the Moon in the third, the sixth, the tenth or the eleventh — it would be auspicious for those afflicted by disease.

7 notes on this verse
  • About this edition: The sandhi spans the pāda join: pāda c ends syāt and pāda d opens śubhaṃ, printed together as syācchubhaṃ. Counting the printed line as it stands gives 10 and 12 and looks defective; splitting metrically gives 11 and 11.
  • Where the Sanskrit is open: ⚠ SUPERSEDED 2026-09-08 — read the later note on this verse first. The witnesses were right and the supply was wrong: the plate prints the nominative निशाकरः, so no locative ending is needed and none is supplied. The note as written stands below, unaltered, because a record of what was believed is worth keeping. ——— Both witnesses read niśākara without the locative ending; the eleven-syllable count requires niśākare.
  • Corroboration: The edition's index attests recovery from illness at IV—5 with the Moon in the 3rd, 6th, 10th or 11th — independently confirming both this verse's house-list (tri, ṣaṭ, daśa, āya) and the Moon.
  • About this edition: Reader's annotation moved out of the `witnessA` field (CORRECTIONS.md C6) — a witness field records the reading and nothing else: “Chapter colophon follows, giving only the number and no name — so this chapter's Sanskrit title rests on the chapter head on printed p.29”
  • About this edition: ⚠ SUPERSEDED 2026-09-08 — read the later note on this verse first. There is no ⟨?⟩ at this word any more and no supplied ending to mark: the plate prints the nominative निशाकरः. The note as written stands below, unaltered, because a record of what was believed is worth keeping. ——— HINDI RENDERING — The locative ending of निशाकरे is ours, supplied on the syllable count — both witnesses read निशाकर without it — and the mūla marks the doubt ⟨?⟩. Our English absorbed the supply silently into 'the Moon in'; the Hindi carries the same sense but marks ⟨स्थित हो⟩ so the relation still shows as our filling rather than the verse's. निरीक्षित and विलग्न are kept as the mūla's own words where our English used the plain 'aspected' and 'the ascendant'.
  • About this edition: SUPERSEDED 2026-09-08 — notes[1] and notes[4] above. Both said the locative ending of निशाकरे is ours, supplied on the eleven-syllable count because both witnesses read निशाकर without it. The witnesses were right and the supply was wrong: the plate prints the NOMINATIVE निशाकरः, with the visarga the 1978 setting dropped (1941, PDF p.39, printed p.31), and witness D prints निशाकरः too. The construction is 'and let the Moon be in the 3rd, 6th, 10th or 11th', which is how the edition's own English takes it. The ⟨?⟩ that stood after the supplied ending — so the mūla read as though a further akṣara were illegible — is gone with it. Separately, त्रिषट्दशाये was wrong sandhi: षट् before द becomes षड्, and the plate prints त्रिषड्दशाये.
  • Corroboration: THE MANUSCRIPT READS THE NOMINATIVE TOO. MS CGV-PSS-0172-1245, p.5 upper leaf, magnified on 2026-09-08, sets …त्रिषड्दशाये वा निशाकरः स्या° , closing ॥३१॥ — निशाकर followed by the visarga assimilated into the स of स्यात्, which is the nominative, not the locative निशाकरे that this verse carried until 2026-09-08 as our own supply. So the correction made on the 1941 plate that day is confirmed by a witness of a different kind. The manuscript reads वा where our line reads च, which changes nothing.

Adhyāya 5 · pravāsacintādhyāyaḥ (Enquiry Concerning One Who Is Abroad)

5.1a reading here is contestedtwo exemplars weighed

दूरगतस्यागमनं सुतधनसहजस्थितैर्ग्रहैर्लग्नात् । सौम्यैर्नष्टप्राप्तिं लघ्वागमनं गुरुसिताभ्याम् ॥ १ ॥

dūragatasyāgamanaṃ sutadhanasahajasthitairgrahairlagnāt | saumyairnaṣṭaprāptiṃ laghvāgamanaṃ gurusitābhyām || 1 ||

The return of one gone far away is shown by grahas standing in the fifth, the second and the third from the ascendant; by benefics, the recovery of what was lost; and by Jupiter and Venus together, a swift return.

3 notes on this verse
  • Corroboration: Witness A could not read the tail of the first line, offering sarvaiḥ or syāt. Witness B read lagnāt, and a 600 DPI crop confirms grahair lagnāt — 'by planets situated in the 5th, 2nd and 3rd FROM THE LAGNA'. The ablative is what makes the house-list intelligible, so this was a substantive gap and it is now closed.
  • Where the Sanskrit is open: laghvāgamanaṃ is laghu + āgamana, a QUICK return; the speed is the point of the clause, not merely the fact of return. gurusitābhyām is an instrumental DUAL — Jupiter and Venus acting together, which is the stated condition.
  • About this edition: Annotation moved out of the `witnessA` field (CORRECTIONS.md C6): a witness field records the reading and nothing else, so this gloss now lives where notes live — “houses internal: suta 5th, dhana 2nd, sahaja 3rd”
5.2two exemplars weighed

जामित्रे त्वथवा षष्ठे ग्रहः केन्द्रेऽथ वाक्पतिः । प्रोषितागमनं विद्यात् त्रिकोणे ज्ञे सितेऽपि वा ॥ २ ॥

jāmitre tvathavā ṣaṣṭhe grahaḥ kendre'tha vākpatiḥ | proṣitāgamanaṃ vidyāt trikoṇe jñe site'pi vā || 2 ||

A graha in the seventh, or else in the sixth, with Jupiter in an angle: one should know the return of him who is away. So too with Mercury or Venus in a trine.

2 notes on this verse
  • About this edition: jāmitra for the seventh house is glossed as such by the commentary on this page, which corroborates the reading. vākpati ('lord of speech') is Jupiter; jña is Mercury; sita is Venus — the whole verse names its grahas by epithet.
  • About this edition: Annotation moved out of the `witnessA` field (CORRECTIONS.md C6): a witness field records the reading and nothing else, so this gloss now lives where notes live — “8×4 exact; no doubtful akṣara”
5.3two exemplars weighed

अष्टमस्थे निशानाथे कण्टकैः पापवर्जितैः । प्रवासी सुखमायाति सौम्यैर्लाभसमन्वितः ॥ ३ ॥

aṣṭamasthe niśānāthe kaṇṭakaiḥ pāpavarjitaiḥ | pravāsī sukhamāyāti saumyairlābhasamanvitaḥ || 3 ||

With the Moon standing in the eighth and the angles free of malefics, the traveller comes home in comfort; and with benefics in them, attended by gain.

2 notes on this verse
  • About this edition: kaṇṭaka is used here for what the text elsewhere calls kendra (V.2 above uses kendra), and the commentary glosses the two as the same set of houses. Worth a glossary row: this text treats them as synonyms. niśānātha ('lord of night') is the Moon.
  • Corroboration: The edition's index attests an absentee returning home safely at V—3, matching pravāsī sukham āyāti.
5.4a reading here is contestedtwo exemplars weighed

पृष्ठोदये पापनिरीक्षिते वा पापास्तृतीये रिपुकेन्द्रगे वा । सौम्यैरदृष्टा वधबन्धदाः स्युर्नष्टा विनष्टा मुषिताश्च वाच्याः ॥ ४ ॥

pṛṣṭhodaye pāpanirīkṣite vā pāpāstṛtīye ripukendrage vā | saumyairadṛṣṭā vadhabandhadāḥ syurnaṣṭā vinaṣṭā muṣitāśca vācyāḥ || 4 ||

At a back-rising sign aspected by a malefic; or with malefics in the third, or gone to the sixth or to an angle, unaspected by benefics — they would be givers of violence and of bondage, and the man is to be declared lost, destroyed, or robbed.

3 notes on this verse
  • About this edition: Third instance in this text of the same typographic pattern: pāda c ends syur and pāda d opens naṣṭā, printed joined as syurnaṣṭā across the pāda break.
  • Where the Sanskrit is open: The verse's own outcome-words are severe and specific — vadha (killing or torture), bandha (bondage), naṣṭa, vinaṣṭa, muṣita (lost, destroyed, robbed). Under the Witness Protocol these are rendered at their weight in translation; the softening that governs the app's own prose applies to our notes, never to the text's verdict.
  • Corroboration: The edition's index attests torture and imprisonment at V—4, and a person being decoyed by thieves at V—4 — matching vadhabandhadāḥ and muṣitāḥ.
5.5two exemplars weighed

ग्रहो विलग्नाद्यतमे गृहे तु तेनाहता द्वादश राशयः स्युः । तावद्दिनान्यागमनस्य विद्यान्निवर्तनं वक्रगतैर्ग्रहैस्तु ॥ ५ ॥

graho vilagnādyatame gṛhe tu tenāhatā dvādaśa rāśayaḥ syuḥ | tāvaddinānyāgamanasya vidyānnivartanaṃ vakragatairgrahaistu || 5 ||

In whichever house from the ascendant a graha stands, let twelve be multiplied by that number: by so many days one should know his coming. But when the grahas have gone retrograde, reckon instead his turning back.

4 notes on this verse
  • Corroboration: Witness B confirms the anusvāra of nivartanaṃ that Witness A had flagged as faint.
  • About this edition: Fourth instance in this text of a cross-pāda join: pāda c ends vidyān and pāda d opens nivartana, printed together as vidyānnivartana. Note also the word-split vilagnād yatame (vilagnāt + yatame, 'from the lagna, in whichever house'), easily mis-split as vilagnā-dyatame.
  • Corroboration: The verse states an arithmetic rule whose numbers are inside it — dvādaśa (twelve) and tenāhatā ('multiplied by that') — and the edition's worked example on the next page agrees with the Sanskrit.
  • About this edition: Reader's annotation moved out of the `witnessA` field (CORRECTIONS.md C6) — a witness field records the reading and nothing else: “Chapter colophon follows: प्रवासचिन्ताध्यायः पञ्चमः, confirming 5 ślokas”

Adhyāya 6 · naṣṭaprāptyadhyāyaḥ (The Recovery of What Is Lost)

6.1a reading here is contestedtwo exemplars weighed

स्थिरोदये स्थिरांशे वा वर्गोत्तमगतेऽपि वा । स्थितं तत्रैव तद्द्रव्यं स्वकीयेनैव चोरितम् ॥ १ ॥

sthirodaye sthirāṃśe vā vargottamagate'pi vā | sthitaṃ tatraiva taddravyaṃ svakīyenaiva coritam || 1 ||

At the rising of a fixed sign, or in a fixed navāṃśa, or in a vargottama one — that property stands in that very place, and was stolen by one of the household itself.

3 notes on this verse
  • Where the Sanskrit is open: Both witnesses read sthita without the anusvāra that agrees with taddravyaṃ; supplied on grammar, with the printed state recorded.
  • About this edition: The commentary on this page gives the edition's own definition of the vargottama navāṃśa — in movable, fixed and dual signs respectively the first, middle and last — which agrees with the convention the app already uses.
  • Corroboration: The edition's index attests fixed signs at VI—1, and separately indexes here the case where the thief belongs to the enquirer's own household — matching sthirodaye and svakīyenaiva coritam.
6.2two exemplars weighed

आदिमध्यावसानेषु द्रेकाणेषु विलग्नतः । द्वारदेशे तथा मध्ये गृहान्ते च वदेद्धनम् ॥ २ ॥

ādimadhyāvasāneṣu drekāṇeṣu vilagnataḥ | dvāradeśe tathā madhye gṛhānte ca vadeddhanam || 2 ||

By the first, the middle and the last drekkāṇa reckoned from the ascendant, one should declare the wealth to be at the gateway, in the middle, and at the far end of the house.

3 notes on this verse
  • About this edition: Confirmed at 400dpi with a SINGLE ka: drekāṇeṣu. VI.5 by contrast prints dreṣkāṇaiḥ. The edition genuinely spells this term two different ways within one chapter, and both are transcribed as printed.
  • Corroboration: The edition's index attests at VI—2 both the decanate as the locating device and the ground behind a dwelling as the place indicated — matching drekāṇeṣu and gṛhānte.
  • About this edition: Annotation moved out of the `witnessA` field (CORRECTIONS.md C6): a witness field records the reading and nothing else, so this gloss now lives where notes live — “8×4 exact; vaded dhanam = vadet + dhanam”
6.3two exemplars weighed

पूर्णः शशी लग्नगतः शुभो वा शीर्षोदये सौम्यनिरीक्षितश्च । नष्टस्य लाभं कुरुते तदाशु लाभोपयातो बलवाञ्छुभश्च ॥ ३ ॥

pūrṇaḥ śaśī lagnagataḥ śubho vā śīrṣodaye saumyanirīkṣitaśca | naṣṭasya lābhaṃ kurute tadāśu lābhopayāto balavāñchubhaśca || 3 ||

The full Moon standing in the ascendant, or a benefic, in a head-rising sign and aspected by benefics, then swiftly brings about the recovery of what is lost — and so does a strong benefic gone to the eleventh.

2 notes on this verse
  • Corroboration: The fourth pāda is verbatim identical to the fourth pāda of I.5, and the third pādas of the two verses restate one another; both state the same rule. The repetition is the text's own, not a transcription slip — and since the two plates were read thirty apart without looking for a match, the agreement corroborates both transcriptions.
  • Corroboration: THE PĀDA लाभोपयातो बलवाञ्छुभश्च STANDS TWICE IN THIS WORK — here and at I.5 — AND PRAŚNA MĀRGA QUOTES THE I.5 OCCURRENCE. PM 29.53 prints क्षिप्रं प्रणष्टस्य करोति लब्धिं लाभ्रोपयातो बलवाञ्छुभश्च, reading उपयातो (its लाभ्रो is a slip for लाभो) and rendering it "a strong benefic that has gone to the eleventh (lābha)". Bhaṭṭotpala's 1858 lithograph prints उपघातो at I.5 and glosses लाभे एकादशस्थाने उपघातः; against it stand both Sastri settings, an independent Kerala line, and this work repeating the pāda unchanged here. The reading is settled, and the settling needed no new book.
6.4two exemplars weighed

दिग्वाच्या केन्द्रगतैरसंभवे वा वदेद्विलग्नर्क्षात् । मध्याच्च्युतैर्विलग्नान्नवांशकैर्योजना वाच्या ॥ ४ ॥

digvācyā kendragatairasaṃbhave vā vadedvilagnarkṣāt | madhyāccyutairvilagnānnavāṃśakairyojanā vācyā || 4 ||

The direction is to be declared by the grahas gone to the angles; or, failing those, one should declare it from the ascendant's sign. By the navāṃśas fallen away from the middle of the ascendant, the yojanas are to be declared.

2 notes on this verse
  • Corroboration: vilagnarkṣāt (vilagna + ṛkṣa, 'from the lagna-SIGN') was a reconstruction at reconciliation, since both witnesses garbled the conjunct. The 400dpi validation pass shows the repha above kṣā, so it is now a reading rather than a conjecture.
  • Where we differ: The edition glosses yojana in English as roughly four krośas or fourteen kilometres. That equivalence is the modern translator's, not the mūla's, and must not migrate into our translation as though the text stated it.
6.5two exemplars weighed

अंशकाज्ज्ञायते द्रव्यं द्रेष्काणैस्तस्कराः स्मृताः । राशिभ्यः कालदिग्देशा वयो जातिश्च लग्नपात् ॥ ५ ॥

aṃśakājjñāyate dravyaṃ dreṣkāṇaistaskarāḥ smṛtāḥ | rāśibhyaḥ kāladigdeśā vayo jātiśca lagnapāt || 5 ||

From the navāṃśa the substance is known; by the drekkāṇas the thieves are told; from the signs, the time, the direction and the place; and the age and the caste, from the lord of the ascendant.

What the printed page actually sets, here

The Sanskrit above is our constituted text — what we judge the work to say, after weighing two independent exemplars and recording every disagreement. This edition's page differs, and the difference is shown rather than smoothed away:

अंशकाज्ज्ञायते द्रव्य द्रेष्काणैस्तस्कराः स्मृताः । राशिभ्यः कालदिग्देशा वयो जातिश्च लग्नपात् ॥ ५ ॥

6 notes on this verse
  • About this edition: DIPLOMATIC vs CONSTITUTED. The Sanskrit above is the CONSTITUTED text — what we judge the work to say. This printing differs: द्रव्य without the anusvāra. Both are now recorded separately, because an edited reading and a read one are different objects and sharing one field is what let this record claim more precision than it had (CORRECTIONS.md C1.5).
  • About this edition: Confirmed at 400dpi as dreṣkāṇaiḥ, with ṣ, as Witness B read it — against drekāṇeṣu at VI.2. The two spellings stand as printed.
  • About this edition: The commentary on printed p.13 quotes this verse's first pāda and refers it to VIII—14. This edition has only seven adhyāyas — its own final colophon closes the work at VII.12 — so that reference cannot point inside the text as numbered here; the line stands at VI.5. Either the reference is an error in the apparatus, or it follows a different recension's division. The first is likelier: the book's other internal cross-reference, to IV—2, does resolve correctly.
  • Corroboration: The edition's index attests the kind of substance stolen, the description of the thieves, and the caste of the thief all at VI—5 — matching dravyaṃ, taskarāḥ and jātiḥ in this one verse.
  • About this edition: Reader's annotation moved out of the `witnessA` field (CORRECTIONS.md C6) — a witness field records the reading and nothing else: “; lagnapāt = lagnapa + ablative). Chapter colophon follows: नष्टप्राप्त्यध्यायः षष्ठः, confirming 5 ślokas”
  • Corroboration: PRAŚNA MĀRGA 29.4 QUOTES THIS VERSE WHOLE AND WORD FOR WORD, AND 29.5 RESTATES IT. PM 29.4 prints अंशकाज्ज्ञायते द्रव्यं द्रेष्काणैस्तस्कराः स्मृताः । राशिभ्यः कालदिग्देशा वयोजातिश्च लग्नपात् — every akṣara the same, with only the word space inside वयो जातिश्च closed and a stray typographic mark before it in our copy of Praśna Mārga. Two things follow. FIRST, THE SPELLING: this chapter's ledger records the edition setting द्रेकाण at VI.2 and द्रेष्काण here, and an independent line prints द्रेष्काण, so the ष belongs to the text and the variation to the printing. PM 29.5, which restates the same rule in its own words (द्रव्यमंशकतश्चोरा द्रेक्काणरादिशेत्सुधीः । दिग्देशकालराशिभ्यो वयोजातिं च लग्नपात्), spells it द्रेक्काण — so both spellings are old. SECOND, THE CASTE LOCUS: वयो जातिश्च लग्नपात् is quoted verbatim by another classical work, so "the age and the caste, from the lord of the ascendant" is this text's own statement and not our construal. It is annotated, never softened.

Adhyāya 7 · miśrakādhyāyaḥ (Miscellaneous)

7.1two exemplars weighed

विषमस्थितेऽर्कपुत्रे सुतस्य जन्मान्यथाङ्गनायाश्च । लभ्या वरस्य नारी समस्थितेऽतोऽन्यथा वामम् ॥ १ ॥

viṣamasthite'rkaputre sutasya janmānyathāṅganāyāśca | labhyā varasya nārī samasthite'to'nyathā vāmam || 1 ||

With Saturn placed in an odd house, the birth of a son; otherwise, of a girl. With him placed in an even house, a wife is to be had for the suitor; otherwise than this, the adverse.

3 notes on this verse
  • Corroboration: The edition's index attests both of this verse's clauses at VII—1: the prediction of a boy or a girl, and the acquisition of a bride.
  • About this edition: Annotation moved out of the `witnessA` field (CORRECTIONS.md C6): a witness field records the reading and nothing else, so this gloss now lives where notes live — “clean; arkaputra = Saturn; viṣamasthite/samasthite = the odd/even switch that is the whole rule; vāmam = the adverse case”
  • Corroboration: PRAŚNA MĀRGA 18.95 QUOTES THIS VERSE AND READS परस्य WHERE OUR LINE READS वरस्य. PM prints विषमस्थितेऽर्कपुत्रे सुतस्य जन्माऽन्यथाऽङ्गनायास्तु । लभ्या परस्य नारी समस्थितेऽप्यन्यथा वामम्, and renders the second half "another man's wife is to be won". Witnesses A, B and D and both Sastri settings read वरस्य, "of the suitor" — which is what an adhyāya on marriage and progeny wants, and what the edition's own index attests ("the acquisition of a bride"). The two scan identically, so the metre cannot referee them. परस्य is Praśna Mārga's own idiom in this material: it prints परजायाम् in the parallel to VII.10. Recorded, not adopted. PM also sets °ङ्गनायास्तु for °ङ्गनायाश्च and समस्थितेऽप्यन्यथा for समस्थितेऽतोऽन्यथा — particle swaps that leave the sense intact.
7.2a reading here is contestedtwo exemplars weighed

गुरुरविसौम्यैर्दृष्टस्त्रिसुतमदायारिगः शशी लग्नात् । भवति च विवाहकर्ता त्रिकोणकेन्द्रेषु वा सौम्याः ॥ २ ॥

gururavisaumyairdṛṣṭastrisutamadāyārigaḥ śaśī lagnāt | bhavati ca vivāhakartā trikoṇakendreṣu vā saumyāḥ || 2 ||

The Moon gone to the third, the fifth, the seventh, the eleventh or the sixth from the ascendant, and under the aspect of Jupiter, of the Sun and of Mercury, becomes the maker of a marriage; or the benefics do so, in trines and angles.

4 notes on this verse
  • Where the Sanskrit is open: Witness B read bhaumyaiḥ where Witness A read saumyaiḥ. A is kept: the commentary on the facing page and the edition's index both name Mercury here, and saumya is that graha; bhaumyaiḥ is not a well-formed alternative. The sa/bha confusion is a known hazard of this face.
  • About this edition: saumya carries BOTH of its senses in this one verse: in gururavisaumyaiḥ it is MERCURY, listed beside Jupiter and the Sun as a graha, while the last pāda's saumyāḥ means the BENEFICS as a class. This verse is the example to cite for the glossary rule that saumya must always be disambiguated.
  • Corroboration: The commentary on the facing page glosses each member of the five-house compound trisutamadāyāri separately — 3rd, 5th, 7th, 11th and 6th — and its gloss agrees with the compound member for member.
  • Corroboration: THE MANUSCRIPT PRINTS THIS VERSE TWICE, IN TWO STATES, AND THE SECOND READS गुरुबुध FOR गुरुरवि. MS CGV-PSS-0172-1245 was re-rendered at 220 dpi on 2026-09-08 and its two occurrences were magnified. On p.5's LOWER leaf it reads गुरुरविसौम्यैर्दृष्टस्त्रिसुतमदनायांगिरः शशी लग्नात् ॥ भवति विवाहकर्ता त्रिकोणकेंद्रेषु वा सौम्याः, closing with a numeral in the thirties — that is our reading, letter for letter. On p.6 it reads गुरुबुधसौम्यैर्दृष्टस्त्रिसुतमदनायांगिरः शशी लग्नात्, closing ॥५४॥ — Jupiter and MERCURY where our line has Jupiter and the SUN. (The 2026-09-07 review recorded that second numeral as ॥४५॥; magnified it is ॥५४॥.) The variant is worth knowing because this verse already has a ledger row about which benefic is meant: discrepancies[0] rejected भौम्यैः for सौम्यैः on the ground that the commentary and the index both name Mercury here. The manuscript is the first witness to put Mercury in the Sanskrit itself. It is recorded and not adopted: both Sastri settings, witness B, witness D and the manuscript's own first occurrence all read गुरुरवि.
7.3a reading here is contestedtwo exemplars weighed

चन्द्रार्कयोः सप्तमगौ सितार्की सुखेऽष्टमे वाऽपि तथा विलग्नात् । द्वितीयदुश्चिक्यगतौ तथा च वर्षासु वृष्टिं प्रवदेन्नराणाम् ॥ ३ ॥

candrārkayoḥ saptamagau sitārkī sukhe'ṣṭame vā'pi tathā vilagnāt | dvitīyaduścikyagatau tathā ca varṣāsu vṛṣṭiṃ pravadennarāṇām || 3 ||

Venus and Saturn gone to the seventh, reckoned from the Moon and from the Sun; or likewise in the fourth or the eighth from the ascendant; and so too gone to the second and the third — in the rains one should declare rain for men.

7 notes on this verse
  • What the page would not yield: One akṣara of the final pāda is unread. A wide 600dpi crop shows the pāda as varṣāsu vṛṣṭiṃ pravadejja(?)ṇām, which counts ELEVEN syllables — correcting Witness A's earlier report that the count failed here. The printed form yields no standard word; jagatām ('of creatures') and janānām ('of people') both fit the count and the sense, and the flash-lite third witness read something closer to the latter. It is left as one unread akṣara with two named candidates rather than emended, because the sense is obvious from context and that is exactly what makes a fabricated ending easy to accept.
  • Where the Sanskrit is open: sitārkī is a dual (Venus and Saturn) and saptamagau and gatau are dual too — three dual endings agreeing across the verse, which is what secures the distributive reading.
  • Corroboration: The edition's index attests rain at VII—3 and VII—4.
  • About this edition: Annotation moved out of the `witnessA` field (CORRECTIONS.md C6): a witness field records the reading and nothing else, so this gloss now lives where notes live — “pādas a, b, c count 11 each; duścikya = the 3rd house”
  • Where we differ: A blind reader who had never seen our text transcribed this plate on 2026-08-12 (docs/VISION_EXTRACTION.md); the whole 56-verse book was swept and every divergence adjudicated on the plate at magnification. CORRECTED HERE. We published प्रवदेज्ज⟨?⟩णाम्, which is not a word and which we had left flagged as unresolved rather than settled. Read at 26x the plate gives -दे + a stacked conjunct + an akṣara carrying the ā-mātrā + णाम् — प्रवदेन्नराणाम्, "one should declare [rain] for men". Witness D, an independent exemplar, reads प्रवदेन्नराणम्. The reading is grammatical, scans as the Upajāti requires, and removes a non-word from the published text.
  • About this edition: One place the blind reader and this edition part company, recorded because it is the kind of divergence that must NOT be adopted: the reader read दुखे where we print सुखे. Witnesses A, B and the independent e-text D all read सुखे, and the edition's own English glosses the house as "the 4th" — which is sukha. The glyph is degraded; three witnesses and the sense agree against one reading of worn ink, so सुखे stands.
  • Corroboration: PRAŚNA MĀRGA 25.85 QUOTES THIS VERSE WITH A DIFFERENT FOURTH PĀDA. PM prints चन्द्रार्कयोस्सप्तमगौ सिताकीं सुखेऽष्टमे वापि यथा विलग्नात् । द्वितीयदुश्चिक्यगतौ तथा वा वर्षासु वृष्टिं सुवदेत्विपश्चित् — "the wise man should confidently predict rain". Our pāda d ends प्रवदेन्नराणाम्, "one should declare it for men", which was read off the plate at 26× and is printed by witness D. Both endings are eleven syllables and both scan as Upajāti, so this is a genuine variant and not a corruption on either side. Ours stands: two Sastri settings and the independent e-text agree against one Kerala line. PM's सिताकीं is a slip for सितार्की, and its यथा विलग्नात् / तथा वा are particle swaps.
7.4two exemplars weighed

सौम्या जलराशिस्थास्तृतीयधनकेन्द्रगाः सिते पक्षे । चन्द्रे वाप्युदयगते जलराशिस्थे वदेद्वर्षम् ॥ ४ ॥

saumyā jalarāśisthāstṛtīyadhanakendragāḥ site pakṣe | candre vāpyudayagate jalarāśisthe vadedvarṣam || 4 ||

Benefics standing in watery signs, gone to the third, the second and the angles, in the bright fortnight; or else the Moon risen and standing in a watery sign — one should declare rain.

3 notes on this verse
  • Corroboration: tṛtīya is read by both Witness B and the flash-lite third witness, against Witness A's turya+āya — a 2-of-3 majority, and the house-list then matches the commentary's. Separately the plate reads vadedvarṣam ('should declare rain'), not A's dadedvarṣam; the declarative is this text's idiom elsewhere (vaded dhanam at VI.2).
  • About this edition: Annotation moved out of the `witnessA` field (CORRECTIONS.md C6): a witness field records the reading and nothing else, so this gloss now lives where notes live — “site pakṣe = the bright fortnight; jalarāśi = a watery sign”
  • Corroboration: PRAŚNA MĀRGA 25.83 QUOTES THIS VERSE AND CONFIRMS वदेद्वर्षम्. PM prints सौम्या जलराशिस्थास्तृतीयधनकेन्दरगास्सिते पक्षे। चन्द्रोऽप्युदयं याते जलराशिस्थे वदेदर्षम् — the first half ours apart from PM's केन्दर for केन्द्र, and the last word वदेदर्षम्, which is वदेद्वर्षम् with the conjunct broken in printing. This chapter's ledger weighed ददेद् against वदेद् and resolved to वदेद्वर्षम् at 600 dpi; an independent line now prints the same verb. PM restates the Moon's condition as a nominative (चन्द्रोऽप्युदयं याते) where our line has the locative absolute चन्द्रे वाप्युदयगते; the sense is identical and it is recorded, not adopted.
7.5a reading here is contestedtwo exemplars weighed

पुंवर्गे लग्नगते पुंग्रहदृष्टे बलान्विते पुरुषः । युग्मे स्त्रीग्रहदृष्टे स्त्रीबुधयुक्ते तु गर्भयुता ॥ ५ ॥

puṃvarge lagnagate puṃgrahadṛṣṭe balānvite puruṣaḥ | yugme strīgrahadṛṣṭe strībudhayukte tu garbhayutā || 5 ||

When a masculine division holds the ascendant, aspected by a masculine graha and endowed with strength: a man. In an even sign, aspected by a feminine graha: a woman; but when joined with Mercury, one carrying a child.

2 notes on this verse
  • Where the Sanskrit is open: Witness B read garbhiyutā; garbhayutā is the well-formed compound and is kept.
  • Corroboration: The verse is a strict masculine/feminine mirror — puṃvarga, puṃgraha, puruṣaḥ against yugma, strīgraha, garbhayutā — and the symmetry is itself a check on the akṣaras. The edition's index attests both the masculine and the female varga at VII—5, and pregnancy at VII—5.
7.6a reading here is contestedtwo exemplars weighed

कुमारिकां बालशशी बुधश्च वृद्धां शनिः सूर्यगुरू प्रसूताम् । स्त्रीं कर्कशां भौमसितौ विधत्त एवं वयः स्यात्पुरुषेषु चैवम् ॥ ६ ॥

kumārikāṃ bālaśaśī budhaśca vṛddhāṃ śaniḥ sūryagurū prasūtām | strīṃ karkaśāṃ bhaumasitau vidhatta evaṃ vayaḥ syātpuruṣeṣu caivam || 6 ||

The young Moon and Mercury appoint a maiden; Saturn, an old woman; the Sun and Jupiter, one newly delivered; Mars and Venus, a rough-bodied woman. Such would the age be — and so likewise among men.

5 notes on this verse
  • Where the Sanskrit is open: The accusative anusvāra of strīṃ would match karkaśāṃ beside it, but is not clearly inked.
  • About this edition: The verse ends by extending its own rule to men (evaṃ vayaḥ syāt puruṣeṣu caivam), so although every category it lists is female the rule is not. A translation that stops at the list would misstate its scope.
  • Another reading: The edition records another reading for the first pāda: bālā kumārī ca indur budhaś ca. It says the variant's sense differs — the Moon then denotes a girl not yet of age and Mercury an unmarried maiden, splitting between the two grahas what the received pāda gives jointly.
  • Corroboration: The edition's index attests the young Moon at VII—6 and a robust or rough-bodied woman at VII—6 (printed V—6, an impossible reference since Adhyāya V has only five ślokas), matching bālaśaśī and karkaśāṃ.
  • Corroboration: PRAŚNA MĀRGA 28.7 QUOTES THIS VERSE AND DIFFERS IN ONE SANDHI ONLY. PM prints कुमारिकां बालशशीबुधश्च वृद्धां शनिः सूर्यगुरू प्रसूताम्। स्त्रीं कर्कशां भौमसितौ विधत्ते एवं वयः स्यात् पुरुषेषु चैवम्. The one difference is विधत्ते एवं against विधत्त एवं: विधत्ते + एवम् gives विधत्त एवम् by regular sandhi, so PM prints the hiatus unresolved and our line prints it resolved. Seventy-seven characters run identical either side of it — the second-longest verbatim testimonium the collation found in this book.
7.7two exemplars weighed

आत्मसमं लग्नगतैर्भ्राता सहजस्थितैः सुतः सुतगैः । माता वा भगिनी वा चतुर्थगैः शत्रुगैः शत्रुः ॥ ७ ॥

ātmasamaṃ lagnagatairbhrātā sahajasthitaiḥ sutaḥ sutagaiḥ | mātā vā bhaginī vā caturthagaiḥ śatrugaiḥ śatruḥ || 7 ||

By grahas gone to the ascendant, one like oneself; by those standing in the third, a brother; by those gone to the fifth, a son; by those gone to the fourth, a mother or a sister; by those gone to the sixth, an enemy.

2 notes on this verse
  • Another reading: The edition records another reading for the first line: ātmasamaṃ lagnagatais tṛtīyagair bhrātaraḥ sutaḥ sutagaiḥ. It differs at two points — tṛtīyagaiḥ for sahajasthitaiḥ (the same house, a different word) and the PLURAL bhrātaraḥ for the singular bhrātā. The second is substantive.
  • Corroboration: The edition's index attests brothers at VII—7, a mother at VII—7 and a sister at VII—7, matching the verse's list.
7.8a reading here is contestedtwo exemplars weighed

भार्या सप्तमसंस्थैर्नवमे धर्माश्रितो गुरुर्दशमे । स्वांशपतिमित्रशत्रुषु तथैव वाच्यं बलयुतेषु ॥ ८ ॥

bhāryā saptamasaṃsthairnavame dharmāśrito gururdaśame | svāṃśapatimitraśatruṣu tathaiva vācyaṃ balayuteṣu || 8 ||

By those standing in the seventh, a wife; in the ninth, one given to dharma; in the tenth, a preceptor. In just the same way it is to be declared of the lord of one's own navāṃśa, of his friend and of his enemy — when these are endowed with strength.

4 notes on this verse
  • Where we differ: Witness A read svāṃga- ('own body'); the plate reads svāṃśa-, as Witness B had it, and the commentary settles it: it discusses the lord of the RISING NAVĀṂŚA, his friend and his enemy — which is exactly svāṃśapati, mitra, śatru. A technical term recovered that Witness A had turned into an ordinary word.
  • Where the Sanskrit is open: Both witnesses read guru without the repha; guruḥ + daśame gives gurur daśame, supplied on grammar with the printed state recorded.
  • About this edition: The closing balayuteṣu ('when they are possessed of strength') governs the whole house-to-relation list of VII.7 and VII.8 together. A reader meeting VII.7 alone would miss the strength condition entirely.
  • Corroboration: The edition's index attests a friend, a wife and a virtuous action at VII—8. Its preceptor entry prints VI—8, which cannot exist (Adhyāya VI has five ślokas) and belongs here — correcting that typo confirms guru in the tenth.
7.9two exemplars weighed

चरलग्ने चरभागे मध्याद्भ्रष्टे प्रवासचिन्ता स्यात् । भ्रष्टः सप्तमभवनात्पुनर्निवृत्तो यदि न वक्री ॥ ९ ॥

caralagne carabhāge madhyādbhraṣṭe pravāsacintā syāt | bhraṣṭaḥ saptamabhavanātpunarnivṛtto yadi na vakrī || 9 ||

With a movable ascendant and a movable navāṃśa fallen past the middle, the thought would be of a journey abroad. Fallen from the seventh house and turned back again — if it is not retrograde.

What the printed page actually sets, here

The Sanskrit above is our constituted text — what we judge the work to say, after weighing two independent exemplars and recording every disagreement. This edition's page differs, and the difference is shown rather than smoothed away:

चरलग्ने चरभागे मध्याद्भ्रष्टे प्रवासचिन्ता स्यात् । भ्रष्टः सप्तमभवनात्पुनर्निवृत्तो यदि न वक्री ॥ ९ ॥

4 notes on this verse
  • Where we differ: Witness A read araḥ, which is not a word, and offered āraḥ (Mars) or arkaḥ (the Sun) as corrections — i.e. it assumed the verse named a graha. Witness B read adhaḥ ('below'), which is an adverb, parses with what follows (adhaḥ saptamabhavanāt, 'below the seventh house'), and matches the rule the commentary states. adhaḥ is adopted, with the note that a 600 DPI crop is consistent with it but not decisive on its own. Worth recording plainly: Witness A's error here would have invented a planet.
  • Corroboration: The edition's index attests the return of a traveller to his place at VII—9 and movable signs at VII—9, matching caralagne / carabhāge and the return clause.
  • About this edition: Annotation moved out of the `witnessA` field (CORRECTIONS.md C6): a witness field records the reading and nothing else, so this gloss now lives where notes live — “vakrī = retrograde; the condition yadi na vakrī governs the clause”
  • Where we differ: A blind reader who had never seen our text transcribed this plate on 2026-08-12 (docs/VISION_EXTRACTION.md); the whole 56-verse book was swept and every divergence adjudicated on the plate at magnification. CORRECTED HERE. We published अधः; the plate reads भ्रष्टः, and so does witness D, which is a DIFFERENT exemplar and not this printing. Our own apparatus already contained the evidence and we did not use it: witness A could not resolve the word and offered ⟨अरः? / आरः? / अर्कः?⟩ — every one of its guesses carries an र, i.e. A saw the र-element of भ्र and guessed at a graha name. Witness B alone read अधः. Re-read at 18x against the भ्र of मध्याद्भ्रष्टे in the line above — the same conjunct, the same fount — the subscript र is there. भ्रष्टः also answers भ्रष्टे in the first line, which अधः does not.
7.10a reading here is contestedtwo exemplars weighed

अस्ते रविसितवक्रैः परजायां स्वां गुरौ बुधे वेश्याम् । चन्द्रे च वयः शशिवत्प्रवदेत्सौरेऽन्त्यजातीनाम् ॥ १० ॥

aste ravisitavakraiḥ parajāyāṃ svāṃ gurau budhe veśyām | candre ca vayaḥ śaśivatpravadetsaure'ntyajātīnām || 10 ||

In the seventh, by the Sun, Venus and Mars: another man's wife; with Jupiter: his own; with Mercury: a courtesan. And with the Moon, one should declare her age according to the Moon's own; with Saturn, a woman of the lowest castes.

3 notes on this verse
  • Where the Sanskrit is open: This verse classifies women by graha — parajāyā (another man's wife), veśyā (courtesan), antyajāti (the lowest castes). It is a caste-and-status judgement in the text itself. The faithfulness rule applies: it is rendered at its weight and not omitted, and the honest handling is a note stating plainly that this is the framework of a text of its period, not a claim the app endorses. Do not soften the translation; do annotate it.
  • About this edition: asta is a fifth synonym for the seventh house in this text, alongside jāmitra, mada and saptama. vakra is Mars, saura is Saturn — the grahas are named entirely by epithet.
  • Corroboration: PRAŚNA MĀRGA 28.6 QUOTES THIS VERSE, CONFIRMS शशिवत् INDEPENDENTLY, AND CORROBORATES THE CASTE CLAUSE. PM prints अस्ते रविसितवक्रैः परजायां स्वां गुरौ बुधे वेश्याम्। चन्द्रे च वयश्शशिवत् प्रवदेन्मान्देऽन्त्यजातिजातीयताम् — the first half identical letter for letter. Two things follow. FIRST: V.A.K. Ayer's printing reads राशिवत् here and the 2026-09-07 review proposed adopting it. PM's वयश्शशिवत् is the Kerala doubling of वयः शशिवत्, so a third line, unrelated to either Sastri setting, prints शशिवत् too — and the checked-no-change recorded on this verse now rests on two independent transmissions instead of one plate and a mātrā count. SECOND: the caste clause is neither ours nor a translator's. An independent classical work quotes it, reading अन्त्यजातिजातीयताम् where our line reads अन्त्यजातीनाम्. The one real divergence is the graha: PM names Māndi (मान्दे) where our line names Saturn (सौरे). Māndi is Saturn's own upagraha, so the two lines make the same prediction through different agents. Recorded, not adopted — both Sastri settings and witness D read सौरे.
7.11a reading here is contestedtwo exemplars weighed

मन्दः पापसमेतो लग्नान्नवमे शुभैरयुतदृष्टः । रोगार्तः परदेशे चाष्टमगो मृत्युकर एव ॥ ११ ॥

mandaḥ pāpasameto lagnānnavame śubhairayutadṛṣṭaḥ | rogārtaḥ paradeśe cāṣṭamago mṛtyukara eva || 11 ||

Saturn joined with a malefic, in the ninth from the ascendant, neither joined with nor aspected by benefics: the man is afflicted by disease in a foreign land. And gone to the eighth, that same Saturn is a very causer of death.

2 notes on this verse
  • Corroboration: lagnān navame ('in the 9th from the lagna') is confirmed at 400dpi, as Witness B read it; Witness A's lagnāccavame was not well-formed. It matches the index entry for sickness in a foreign place at VII—11.
  • Where the Sanskrit is open: śubhair ayutadṛṣṭaḥ is a compound NEGATIVE condition — neither joined nor aspected by benefics. Losing the privative a- would invert the rule. mṛtyukara eva is a death-clause; our own audits have caught received translations dropping exactly this kind of clause, so it ships rendered at weight, with framing in a note and never a silent omission.
7.12two exemplars weighed

सौम्ययुतोऽर्कः सौम्यैः सन्दृष्टश्चाष्टमर्क्षसंस्थश्च । तस्माद्देशादन्यं गतः स वाच्यः पिता तस्य ॥ १२ ॥

saumyayuto'rkaḥ saumyaiḥ sandṛṣṭaścāṣṭamarkṣasaṃsthaśca | tasmāddeśādanyaṃ gataḥ sa vācyaḥ pitā tasya || 12 ||

The Sun joined with a benefic, aspected by benefics, and standing in the eighth sign: his father is to be declared to have gone from that country to another.

3 notes on this verse
  • Corroboration: The edition's index attests the father at VII—12 and the querist's father having quitted a foreign country at VII—12, matching pitā tasya.
  • About this edition: The last verse of the work. The chapter colophon (miśrakādhyāyaḥ saptamaḥ) and the work colophon (ṣaṭpañcāśikā samāptā) both follow on the next plate, and the total then stands at 56 ślokas across seven adhyāyas — 7 + 17 + 5 + 5 + 5 + 5 + 12 — which is the number the work's own title names.
  • About this edition: Annotation moved out of the `witnessA` field (CORRECTIONS.md C6): a witness field records the reading and nothing else, so this gloss now lives where notes live — “aṣṭamarkṣa = aṣṭama + ṛkṣa, the 8th sign”

Use it freely — CC BY 4.0. This Sanskrit transcription and this English translation are released under a Creative Commons Attribution 4.0 licence. Copy them, print them, build on them; the only condition is attribution to Karma Sakha (karmasakha.com). The underlying work is ancient and in the public domain everywhere.

If you find an error, it is ours and we want it. Every reading on this page is recorded with the evidence behind it, so a disagreement can be argued about concretely rather than asserted.

Readings on this site cite texts at chapter level, with verse numbers only where our research verified the passage in a named edition. How conflicts between texts are resolved is documented on the conventions page.