तद्धिताधिकारे आर्हीयप्रकरणम्

Covers SK 1680–1735 (56 sūtras): all present. The spine is Aṣṭādhyāyī 5.1.18–5.1.71, in order, with two kinds of departure. First, the SK splices in four rules from elsewhere where it needs them: three vṛddhi rules from the seventh adhyāya, 7.3.17, 7.3.26 and 7.3.27 (SK 1683–1685), read after asamāse niṣkādibhyaḥ because they govern the compound forms of these affixes; and 1.2.50 (id goṇyāḥ, SK 1703), read after tena krītam because its example is a "bought with" form. Second, the SK omits two Vedic rules, keeping them for the vaidikī section: 5.1.61 (saptano 'ñ chandasi, SK 3491) and 5.1.67 (chandasi ca, SK 3492). So SK 1726 is 5.1.60 and SK 1727 is 5.1.62; SK 1731 is 5.1.66 and SK 1732 is 5.1.68. SK 1708 has no kaumudī prose of its own: its examples are given under 1709.

taddhitādhikāre ārhīyaprakaraṇam — outline (p34, SK 1680–1735, all 56 sūtras)

This prakaraṇa sets up two nested headings and then teaches the senses they serve.

• prāg vateṣ ṭhañ (1680) makes ṭhañ the default affix for every sense taught before tena tulyaṃ kriyā ced vatiḥ (5.1.115, SK 1778). That heading runs far beyond this chapter.
• ārhād agopucchasaṃkhyāparimāṇāṭ ṭhak (1681) carves out a ṭhak heading inside it. It runs up to and including tad arhati (SK 1728) and applies to any base except gopuccha, a numeral (saṃkhyā) or a measure (parimāṇa). Those three keep the general ṭhañ.

The chapter falls into two halves.

• 1680–1701: affixes. The two headings, then apavādas for particular bases: ṭhan, yat, kan (with an optional iṭ), ḍvun, ṭiṭhan, añ, aṇ, kha, īkan and a second aṇ. The kaumudī counts thirteen affixes with ṭhañ (1701; BM lists them). Two rule-families are interleaved: vṛddhi of the final member in compounds ending in a measure (1683–1685), and luk of the affix after a word with adhyardha ("one and a half") as first member or after a dvigu (1693–1701).
• 1702–1735: senses. Each sense comes with its own samarthavibhakti (the case of the base).
• third case: "bought with that" (1702).
• sixth case: "the cause of that" (1704), "lord of that" (1708), "the sowing-ground of that" (1711).
• seventh case: "known in that" (1709).
• first case: "that is given in this" as interest etc. (1713); "that is its share, price or wages" (1722); "that is its measure" (1723).
• second case: "carries off, carries, brings that" (1716); "holds, takes up, cooks that" (1718); "deserves that" (1728).

In each sense the affixes already taught apply as prescribed: ṭhañ for gopuccha, numerals and measures; ṭhak elsewhere; the base-specific apavādas where they apply. The kaumudī closes at 1735: the domain of the twelve ārhīya affixes, ṭhak and the rest, has ended. ataḥ paraṃ ṭhañ eva: from here on, ṭhañ alone (SK 1736 opens p35).

Only the kaumudī blocks are translated (SK-TRANSLATE §25). The readings cited as BM (Bālamanoramā) and TB (Tattvabodhinī) are the ones recorded in the unit notes. Where the p34 audit (.work/sk-run/p34-audit-fixes.md) corrects a reading, this outline follows the corrected reading. This applies to 1680, 1683, 1685, 1687, 1689, 1690, 1693, 1694, 1699, 1701, 1703, 1705, 1707, 1711, 1714–1720, 1724 and 1728–1735. The digital source's corrupt forms are not repeated. Ordering: SK order throughout; nothing is reordered.

Affix shapes used below (standard operations, cited where the units themselves do not spell them out):

• ṭh → ik by ṭhasyekaḥ (7.3.50, SK 1170), but ṭh → k after a stem ending in u or t by isusuktāntāt kaḥ (7.3.51, SK 1221): cātvāriṃśatkaḥ, aikṣukaḥ.
• kha/khañ → īna, cha → īya, gha → iya, by āyaneyīnīyiyaḥ… (7.1.2, SK 475). vu → aka by yuvor anākau (7.1.1, SK 1247).
• ñ-, ṇ- and k-marked affixes (ṭhañ, ṭhak, añ, aṇ, ḍaṇ, khañ) bring ādivṛddhi (7.2.117–118, SK 1075–1076). ṭhan, ṭiṭhan, ṣṭhan, kan, īkan, yat, kha, cha, gha do not: śatikam, kaṃsikaḥ, pañcakaḥ.
• Marker letters for the feminine: the ṭ of ṭiṭhan gives ṅīp (kaṃsikī), the ṣ of ṣṭhan gives ṅīṣ (pātrikī, dvyāḍhakikī). ṭhan takes ṭāp (ardhikā, bhāgikā).
• Final a/ā/i/ī of the stem is deleted before the affix by yasyeti ca (6.4.148, SK 311). ḍ-marked affixes (ḍvun, ḍaṇ, ḍa, ḍati) delete the ṭi by ṭeḥ (6.4.143, SK 316). After viṃśati the syllable ti goes instead, by ti viṃśater ḍiti (6.4.142, SK 844).

A. The two headings: ṭhañ up to vati, ṭhak up to arhati — running example: gopucchena krītam → gaupucchikam

• 1680 prāg vateṣ ṭhañ 5.1.18 — adhikāra: ṭhañ governs the senses taught before the sūtra containing vati, tena tulyam (SK 1778). BM: vati names that sūtra, and ṭhañ is present in the rules before it that state only senses.
• TB, printed under sk.1678.t1: ṭhañ is made a heading, not simply stated, so that after the intervening apavāda headings (sarvabhūmipṛthivībhyām aṇañau, śīrṣacchedād yac ca), ṭhañ alone comes back into force, as in pārāyaṇa… (SK 1736).
• The source reads varti. NSP 1915 prints vatiṃ, so the corruption is confirmed.
• 1681 ārhād agopucchasaṃkhyāparimāṇāṭ ṭhak 5.1.19 — adhikāra inside ṭhañ's: ṭhak up to and including tad arhati (SK 1728), an apavāda of ṭhañ. It does not apply to gopuccha, a numeral or a measure.
• gopuccha is excluded, so it keeps the general ṭhañ: gopucchena krītam → gaupucchikam (1702). ṭhañ → ika, with ādivṛddhi o → au.
• "Up to and including" renders āṅ as abhivyāpti (BM, TB). So ṭhak also comes in the sense arhati itself: śvetacchatram arhati → śvaitacchatrikaḥ (1728; TB).
• BM: because they are named separately, a numeral is not a measure here. BM's verse-vārttika sorts unmāna (weight), parimāṇa (capacity, e.g. prastha), pramāṇa (length) and saṃkhyā (number).
• TB: arha imitates part of the finite verb arhati, not the noun arham.

B. niṣka etc., and vṛddhi in compounds ending in a measure — running example: paramaniṣkeṇa krītaḥ → paramanaiṣkikaḥ

• 1682 asamāse niṣkādibhyaḥ 5.1.20 — ṭhak after niṣka etc., when not in a compound: naiṣkikam, bought with a niṣka. In a compound the base gets ṭhañ instead.
• ārhāt is carried through seventeen sūtras, up to tena krītam (SK 1702). The count is A 5.1.20–36 or 5.1.21–37; BM's paryantam suggests the limit is included, and the kaumudī does not settle it.
• TB's niṣkādi: niṣka, paṇa, pāda, māṣa, vāha, droṇa, ṣaṣṭi. TB gives two views, both from the Padamañjarī. (1) droṇa (a measure) and ṣaṣṭi (a number) would otherwise get ṭhañ; for the other five (weights) ṭhak already follows from 1681, and the rule only excludes compounds. (2) parimāṇa in 1681 covers every means of measuring, so the rule is needed to give ṭhak to all seven.
• 1683 parimāṇāntasyāsaṃjñāśāṇayoḥ 7.3.17 — spliced in here. Before a ñ-, ṇ- or k-marked taddhita, vṛddhi falls on the first vowel of the final member (uttarapada) of a compound ending in a measure. It does not apply when the derivative is a name, or when the final member is śāṇa.
• paramaniṣka is a compound, so it takes ṭhañ (1682). The ni of the final member niṣka gets vṛddhi: paramanaiṣkikaḥ.
• Counter-example with a name: pañca kalāpāḥ parimāṇam asya → pāñcakalāpikam, with ṭhañ by 1723. Only the first syllable gets vṛddhi.
• Jñāpaka: 1682's "asamāse" would be pointless without tadantavidhi, so it shows that tadantavidhi holds before this point (the rules before 1682): sugavyam (yat after su-go, SK 1662), yavāpūpyam (SK 1664).
• From here onward, a vārttika limits it: saṃkhyāpūrvapadānāṃ tadantagrahaṇaṃ prāg vater iṣyate tac cāluki. Tadantavidhi holds before vati only for compounds whose first member is a numeral, and only when no luk has occurred: pārāyaṇikaḥ, dvaipārāyaṇikaḥ.
• Why "no luk"? dvābhyāṃ śūrpābhyāṃ krītam → dviśūrpam: the dvigu's affix undergoes luk (1693). In the sense "bought with a dviśūrpa", 1691's añ does not apply. ṭhañ comes instead, and the final member gets vṛddhi by this rule: dviśaurpikam.
• BM and TB: this ṭhañ is not deleted by 1693, because it is not the affix that caused the dvigu. TB adds that the kaumudī's own gloss at 1693 must be read that way.
• The source prints śūrpabhyāṃ; NSP 1915 reads śūrpābhyāṃ.
• 1684 ardhāt parimāṇasya pūrvasya tu vā 7.3.26 — after ardha ("half"), vṛddhi falls on the first vowel of a measure-word final member. The prior member ardha gets vṛddhi only optionally. With ṭhañ, droṇa being in a compound (BM): ardhadroṇena krītam → ārdhadrauṇikam / ardhadrauṇikam.
• 1685 nātaḥ parasya 7.3.27 — after ardha, vṛddhi does not replace a following short a of the measure-word; the prior member still takes vṛddhi optionally: ārdhaprasthikam / ardhaprasthikam.
• Why "a"? kuḍava begins with u, so the final member gets vṛddhi by 1684: ārdhakauḍavikam. The source spells it -kauḍabikam; NSP and BM have kauḍavikam.
• Why the t-marker (only short a)? Take ardhakhārī (a feminine thing "in half a khārī"). If long ā were also covered, the prohibition of puṃvadbhāva by vṛddhinimittasya… (SK 840) would fail in ardhakhārībhārya. BM and TB: the point concerns the option where the prior member has no vṛddhi. The source prints the form with ārdha-; NSP has ardha-.
• BM: 1683–1685 are from the seventh adhyāya, and the fifth adhyāya now resumes.

C. Numerals and coin-or-measure words: ṭhan/yat, kan, iṭ, ḍvun, ṭiṭhan, añ, aṇ — running example: śatena krītam → śatikam / śatyam

• 1686 śatāc ca ṭhanyatāv aśate 5.1.21 — after śata ("hundred"), ṭhan and yat, apavādas of the kan of 1687 (BM, TB). They do not come when the sense is itself a hundred (aśate).
• ṭhan brings no vṛddhi: ṭh → ik, and the a is deleted: śatikam. With yat: śatyam.
• Why aśate? śataṃ parimāṇam asya → śatakaḥ saṅghaḥ. Here the sense of the affix (the group) is in reality not different from the sense of the base (the hundred), so only kan comes.
• ca brings in asamāse: dviśatena krītam → dviśatakam, with kan. BM analyses dviśata as "a doubled hundred"; TB as a collective dvandva, "two and a hundred". The English leaves it untranslated.
• 1687 saṃkhyāyā atiśadantāyāḥ kan 5.1.22 — kan after a numeral, but not one ending in ti or śat. kan brings no vṛddhi.
• pañcabhiḥ krītaḥ → pañcakaḥ; also bahukaḥ. TB: pañca is an ordinary numeral, bahu a technical one (bahugaṇavatuḍati saṃkhyā, SK 258).
• ti-final: saptati gets ṭhañ instead, with ādivṛddhi and the i deleted: sāptatikaḥ.
• śat-final: catvāriṃśat also gets ṭhañ, and ṭh → k after the t: cātvāriṃśatkaḥ. The source has cātvāriṃtkaḥ; NSP reads cātvāriṃśatkaḥ.
• 1688 vator iḍ vā 5.1.23 — kan after a vatu-final stem optionally takes the augment iṭ, placed before it because of the ṭ. tāvat (vatup by SK 1840; a numeral by SK 258): tāvatikaḥ / tāvatkaḥ, bought for that much.
• BM, TB: the ablative vatoḥ turns the nominative kan into a genitive (SK 41), hence "to kan".
• 1689 viṃśatitriṃśadbhyāṃ ḍvun asaṃjñāyām 5.1.24 — the rule is split (yogavibhāga).
• The first part, viṃśatitriṃśadbhyām, gives kan to viṃśati and triṃśat, which 1687 had excluded. By elimination (BM), this applies when the derivative is a name: viṃśatikaḥ, triṃśatkaḥ. The source has triṃśatikaḥ; NSP reads triṃśatkaḥ.
• The second part: when the derivative is not a name, ḍvun, an apavāda of kan. vu → aka, and the ḍ deletes: in viṃśati the ti (ti viṃśater ḍiti); in triṃśat the ṭi (ṭeḥ). This gives viṃśakaḥ, triṃśakaḥ.
• 1690 kaṃsāṭ ṭiṭhan 5.1.25 — ṭiṭhan after kaṃsa. The ṭ gives ṅīp in the feminine, and the i is for pronunciation: kaṃsikaḥ, [feminine] kaṃsikī.
• vārttika ardhāc ceti vaktavyam: also after ardha: ardhikaḥ, ardhikī. TB: ardha is fixed (rūḍha) for [half] a kārṣāpaṇa, so no incapacity from its need for a correlate arises.
• vārttika kārṣāpaṇāṭ ṭiṭhan vaktavyaḥ: also after kārṣāpaṇa. The source prints ṭhiṭhan; NSP reads ṭiṭhan.
• vārttika pratir ādeśaś ca vā: prati optionally replaces kārṣāpaṇa: kārṣāpaṇikaḥ / kārṣāpaṇikī, pratikaḥ / pratikī. BM: prati is tied to ṭiṭhan (saṃniyogaśiṣṭa), which matters at 1694.
• 1691 śūrpād añ anyatarasyām 5.1.26 — añ after śūrpa, optionally: śaurpam. śūrpa is a measure, so the other option is ṭhañ (BM, TB): śaurpikam.
• 1692 śatamānaviṃśatikasahasravasanād aṇ 5.1.27 — aṇ after śatamāna, viṃśatika, sahasra and vasana. It is an apavāda of ṭhañ, ṭhak and kan, distributed by base (BM, TB).
• śatamānena krītam → śātamānam (ṭhañ blocked: a measure).
• vaiṃśatikam (ṭhañ or ṭhak blocked, by what viṃśatika names).
• sāhasram (kan blocked: a numeral).
• vāsanam (ṭhak blocked).

D. After adhyardha- and a dvigu: luk of the ārhīya affix, and the affixes that replace it — running example: dvābhyāṃ kaṃsābhyāṃ krītam → dvikaṃsam

• 1693 adhyardhapūrvadvigor lug asaṃjñāyām 5.1.28 — after a word with adhyardha ("one and a half") as first member, and after a dvigu, the ārhīya affix undergoes luk, except in a name.
• dvikaṃsa is a dvigu in a taddhita sense; its ṭiṭhan is deleted: dvikaṃsam. Likewise adhyardhakaṃsam.
• In a name the affix stays: pāñcakalāpikam. The source has pāñcā-; NSP reads pāñca-, as at 1683.
• BM, TB: dvigoḥ is ablative ("after", in a samāhāra-dvandva with adhyardhapūrva). The luk falls only on the affix that caused the dvigu, which saves dviśaurpikam (1683).
• 1694 vibhāṣā kārṣāpaṇasahasrābhyām 5.1.29 — after the same bases ending in kārṣāpaṇa or sahasra, the luk is optional.
• The luk is of the vārttika's ṭiṭhan (1690): adhyardhakārṣāpaṇam / adhyardhakārṣāpaṇikam; dvikārṣāpaṇam / [dvikārṣāpaṇikam]. The source has adhyardhakārṣāṇam and lacks the second pair's non-luk form; NSP supplies both.
• With the substitute prati there is no luk, since prati is tied to ṭiṭhan: adhyardhapratikam, dvipratikam.
• sahasra: optional luk of 1692's aṇ: adhyardhasahasram / adhyardhasāhasram; dvisahasram / dvisāhasram. Where aṇ stays, the final member gets vṛddhi by saṃkhyāyāḥ saṃvatsarasaṃkhyasya ca (7.3.15, SK 1752; BM, TB).
• 1695 dvitripūrvān niṣkāt 5.1.30 — optional luk after a dvigu ending in niṣka with dvi or tri as first member.
• The affix is ṭhañ (a compound, so not ṭhak), with final-member vṛddhi by 1683: dviniṣkam / dvinaiṣkikam, triniṣkam / trinaiṣkikam.
• vārttika bahupūrvāc ceti vaktavyam: also with bahu: bahuniṣkam / bahunaiṣkikam.
• TB: dvigoḥ is connected here, to exclude a genitive tatpuruṣa (dvayor niṣkaḥ).
• 1696 bistāc ca 5.1.31 — the same optional luk after bista preceded by dvi, tri or bahu: dvibistam / dvibaistikam, and so on. BM: bahu comes from 1695's vārttika. By ca, dvitripūrvāt is drawn in and stops here (TB).
• 1697 viṃśatikāt khaḥ 5.1.32 — kha after such a word ending in viṃśatika. The kaumudī: adhyardhapūrvād dvigoḥ holds here all the same.
• adhyardhaviṃśatikīnam, dviviṃśatikīnam (kha → īna).
• TB: kha replaces 1692's aṇ and the luk that would follow it.
• BM says dvigoḥ lapsed in 1695–1696 and resumes here. TB holds that dvigoḥ ran on through them. The English "holds here all the same" is neutral.
• 1698 khāryā īkan 5.1.33 — īkan after such a word ending in khārī, instead of ṭhañ and its luk: adhyardhakhārīkam, dvikhārīkam.
• vārttika kevalāyāś ceti vaktavyam: also after khārī on its own: khārīkam.
• TB: kan would shorten the ī (ke 'ṇaḥ, SK 834), and ikan would lose it (yasyeti ca), hence īkan.
• 1699 paṇapādamāṣaśatād yat 5.1.34 — yat after such a word ending in paṇa, pāda, māṣa or śata: adhyardhapaṇyam, dvipaṇyam; adhyardhapādyam, dvipādyam.
• In dvipādyam, pāda is not replaced by pat (pādaḥ pat, SK 414): the deletion of its a by yasyeti ca counts as the original (sthānivadbhāva).
• It is not replaced by pad either (pad yaty atadarthe, SK 991), which covers only pāda as a limb of a living being. Here, beside paṇa and māṣa, pāda is a measure (BM, TB).
• 1700 śāṇād vā 5.1.35 — yat optionally after such a word ending in śāṇa. In the other option ṭhañ, which undergoes luk (1693): adhyardhaśāṇyam / adhyardhaśāṇam.
• TB: on the principal view the other affix should be ṭhak, since śāṇa is a weight.
• 1701 dvitripūrvād aṇ ca 5.1.36 — after śāṇa with dvi or tri as first member, aṇ, and by ca yat. TB and BM call it a vārttika that the Vṛttikāra set among the sūtras.
• There are three forms.
• aṇ: dvaiśāṇam. Only the first syllable gets vṛddhi, because 1683 excludes śāṇa.
• yat: dviśāṇyam.
• ṭhañ with luk: dviśāṇam. The source has dvīśāṇam; NSP reads dviśāṇam.
• The kaumudī then sums up. Thirteen affixes, ṭhañ and the rest, have been introduced. Their samarthavibhaktis and senses are awaited, and are now stated.

E. The first sense: "bought with that" — running example: prasthena krītam → prāsthikam

• 1702 tena krītam 5.1.37 — the affixes come as prescribed after a third-case word, in the sense "bought with that".
• ṭhañ for bases excluded from ṭhak: gopucchena krītam → gaupucchikam; sāptatikam (saptati); prāsthikam (prastha).
• ṭhak: naiṣkikam (1682).
• BM, TB: the instrumental must be the price paid (karaṇa, mūladravya). "Bought by Devadatta" gets no affix, because the sense is not expressed (anabhidhāna).
• 1703 id goṇyāḥ 1.2.50 — spliced in here. When a taddhita undergoes luk, the final ī of goṇī (its feminine affix, BM) is replaced by i. This is an apavāda of the feminine affix's luk by luk taddhitaluki (SK 1408).
• pañcabhir goṇībhiḥ krītaḥ paṭaḥ: the ārhīya affix is deleted (1693), and the ī becomes i: pañcagoṇiḥ.
• BM: without the rule the ṅīṣ would go by SK 1408 and the a-final base would take ṭāp, giving *pañcagoṇā.

F. Sixth and seventh case: the cause of that, lord of that, known in that — running example: śatasya nimittaṃ dhanapatisaṃyogaḥ → śatikaḥ / śatyaḥ

• 1704 tasya nimittaṃ saṃyogotpātau 5.1.38 — the affixes come after a sixth-case word, in the sense "the cause of that", when the cause is a connection (saṃyoga = sambandha) or a portent (utpāta, a sign of good or ill).
• śatikaḥ / śatyaḥ dhanapatisaṃyogaḥ: a connection with a rich man that brings a hundred. ṭhan and yat come by 1686.
• śatikam / śatyam dakṣiṇākṣispandanam: a throbbing of the right eye that portends a hundred.
• vārttika vātapittaśleṣmabhyaḥ śamanakopanayor upasaṃkhyānam: after vāta, pitta and śleṣman, in the sense "allaying or aggravating". BM supplies ṭhak: vātikam, paittikam, ślaiṣmikam.
• vārttika saṃnipātāc ceti vaktavyam: also after saṃnipāta: sāṃnipātikam.
• 1705 godvyaco 'saṃkhyāparimāṇāśvāder yat 5.1.39 — yat, an apavāda of ṭhak, after go and after a two-vowel word, but not after a numeral, a measure or aśva etc.
• gor nimittam → gavyaḥ: o → av before y (vānto yi pratyaye, SK 63).
• Two-vowel words: dhanyaḥ, yaśasyaḥ, svargyaḥ.
• Counter-examples:
• vijaya has three vowels: vaijayikaḥ (ṭhak).
• Numerals: pañcakam, saptakam (kan). The source prints pañakam; NSP reads pañcakam.
• Measures: prāsthikam (ṭhañ), khārīkam (īkan).
• aśva etc.: āśvikam, āśmikam (ṭhak; aśman loses its ṭi by nas taddhite, SK 679). The source prints āsmikam; NSP reads āśmikam.
• vārttika brahmavarcasād upasaṃkhyānam: also after brahmavarcasa: brahmavarcasyam.
• 1706 putrāc cha ca 5.1.40 — cha after putra, and by ca yat: putrīyaḥ, putryaḥ. TB: stated because 1705 would give yat obligatorily.
• BM, TB: in putrīyā iṣṭiḥ the rite counts as a saṃyoga, "that by which one is joined" (with the fruit, a son).
• 1707 sarvabhūmipṛthivībhyām aṇañau 5.1.41 — aṇ and añ after sarvabhūmi and pṛthivī respectively, in the same sense.
• sārvabhaumaḥ: sarvabhūmi is in the anuśatikādi list, so both members get vṛddhi (anuśatikādīnāṃ ca, SK 1438). The source prints padyate; NSP reads paṭhyate ("is read").
• pārthivaḥ.
• TB (printed under sk.1678.t1): this apavāda carries its own heading, which is why 1680 has to keep ṭhañ in force after it.
• 1708 tasyeśvaraḥ 5.1.42 — the same aṇ/añ after sarvabhūmi and pṛthivī, in the sense "lord of that". It has no kaumudī of its own; its example is under 1709.
• TB, BM: tasya is restated to cut off the sense "cause" (nimitta), so that "lord" is not read as another qualifier of it.
• 1709 tatra vidita iti ca 5.1.43 — the same affixes after a seventh-case word, "known in that".
• sarvabhūmer īśvaraḥ (1708) or sarvabhūmau viditaḥ (1709) → sārvabhaumaḥ; likewise pārthivaḥ.
• BM, TB: the two senses are split into separate sūtras (yogavibhāga) so that they are not paired one-to-one (yathāsaṅkhya) with the two bases, and so that only "known in" continues. The base-to-affix pairing of 1707 still holds.
• 1710 lokasarvalokāṭ ṭhañ 5.1.44 — ṭhañ after loka and sarvaloka, "known in that": laukikaḥ. sarvaloka is anuśatikādi, so both members get vṛddhi: sārvalaukikaḥ.

G. The sowing-ground of that; "that is given in this" as interest etc. — running example: pañcāsmin vṛddhir dīyate → pañcakaḥ

• 1711 tasya vāpaḥ 5.1.45 — the affixes after a sixth-case word, "the sowing-ground (vāpa) of that". vāpa is a field, "that in which [seed] is sown".
• TB: tasya is an objective genitive, the thing sown, not the sower (so there is no affix in devadattasya vāpaḥ).
• prasthasya vāpaḥ → prāsthikam (ṭhañ: a measure).
• drauṇikam (ṭhak: droṇa is niṣkādi, BM).
• khārīkam (īkan, 1698).
• 1712 pātrāt ṣṭhan 5.1.46 — ṣṭhan after pātra, same sense: pātrikam. The ṣ gives ṅīṣ: pātrikī kṣetrabhaktiḥ, a portion of field sown with a pātra.
• 1713 tad asmin vṛddhyāyalābhaśulkopadā dīyate 5.1.47 — the affixes after a first-case word, "that is given in this" as interest (vṛddhi), revenue (āya), profit (lābha), duty (śulka) or bribe (upadā). The verb is singular because it is connected with each noun in turn.
• pañcāsmin vṛddhir … dīyate → pañcakaḥ (kan, 1687).
• śatikaḥ / śatyaḥ (ṭhan and yat, 1686); sāhasraḥ (aṇ, 1692).
• The kaumudī defines each term:
• vṛddhi is what the creditor receives over the principal;
• āya is the owner's share from villages and the like;
• lābha is the seller's excess over the price;
• śulka is the king's share, the payment for protection;
• upadā is a bribe.
• vārttika caturthyarthe upasaṃkhyānam: also in the fourth-case sense: pañcāsmai … dīyate → pañcako devadattaḥ. Or the vārttika is not needed, since the recipient can be intended as a locus, as in samam abrāhmaṇe dānam.
• 1714 pūraṇārdhāṭ ṭhan 5.1.48 — ṭhan after an ordinal (pūraṇa) and after ardha, in that sense. It is an apavāda of ṭhak and of ṭiṭhan respectively.
• dvitīyo vṛddhyādir asmin dīyate → dvitīyikaḥ; also tṛtīyikaḥ. The source has tṛtīyakaḥ; NSP reads tṛtīyikaḥ.
• ardhikaḥ. ardha is fixed (rūḍha) as half a rūpaka. BM, TB: ṭiṭhan would give ṅīp in the feminine, but ardhikā with ṭāp is wanted.
• 1715 bhāgād yac ca 5.1.49 — yat after bhāga, and by ca ṭhan. bhāga too is fixed as half a rūpaka.
• bhāgyam / bhāgikam śatam.
• bhāgyā / bhāgikā viṃśatiḥ. The source has bhagikā; NSP reads bhāgikā.

H. Second case: carries off / carries / brings; holds / takes up / cooks — running example: prasthaṃ sambhavati → prāsthikaḥ kaṭāhaḥ

• 1716 tad dharati vahaty āvahati bhārād vaṃśādibhyaḥ 5.1.50 — the affixes after a second-case word, "carries off, carries or brings that". bhāra and vaṃśa etc. are not two separate bases. The kaumudī gives two explanations of the single base:
• (1) a stem ending in bhāra that follows vaṃśa etc. vaṃśabhāraṃ harati vahaty āvahati vā → vāṃśabhārikaḥ (ṭhak); aikṣubhārikaḥ. BM: this needs the two words to be construed as not co-referential and in inverted order.
• (2) vaṃśa etc. that are loads (bhārabhūtebhyaḥ). bhārabhūtān vaṃśān harati → vāṃśikaḥ; aikṣukaḥ, where ṭh → k after the u of ikṣu. BM, TB: the singular bhārāt is used because each base is taken severally.
• 1717 vasnadravyābhyāṃ ṭhankanau 5.1.51 — ṭhan and kan respectively after vasna ("price") and dravya ("goods"), in the senses of 1716. Neither brings vṛddhi.
• vasnaṃ harati … → vasnikaḥ. The source prints vastra/vastrikaḥ; NSP reads vasna / vasnikaḥ.
• dravyakaḥ.
• 1718 sambhavaty avaharati pacati 5.1.52 — the affixes after a second-case word, "holds, takes up (avaharati), or cooks that".
• prasthaṃ sambhavati → prāsthikaḥ kaṭāhaḥ (ṭhañ: a measure), a pan that lets a prastha be contained in itself. TB: sambhavati is transitive here.
• prāsthikī brāhmaṇī (ṅīp after ṭhañ): she takes up or cooks a prastha. BM takes avaharati as measuring out so that even a slightly short quantity makes a prastha; BM's kaumudī evidently read upasaṃharati here.
• vārttika tat pacatīti droṇād aṇ ca: aṇ after droṇa in the sense "cooks", and by ca ṭhañ: droṇaṃ pacati → drauṇī / drauṇikī.
• 1719 āḍhakācitapātrāt kho 'nyatarasyām 5.1.53 — kha optionally after āḍhaka, ācita and pātra; otherwise ṭhañ (they are measures). The kaumudī gives the feminine forms:
• āḍhakīnā / āḍhakikī.
• ācitīnā / ācitikī. The source has acitīnā; NSP reads ācitīnā.
• pātrīṇā / pātrikī. Both witnesses print pātrīkī, and pātrikī is the translator's conjecture from the grammar (cf. dvipātrikī, 1720).
• 1720 dvigoḥ ṣṭhaṃś ca 5.1.54 — after a dvigu ending in āḍhaka, ācita or pātra, ṣṭhan and (by ca) kha optionally. In the remaining option ṭhañ, which undergoes luk by 1693.
• dvyāḍhakikī: ṣṭhan, and the ṣ gives ṅīṣ.
• dvyāḍhakīnā: kha.
• dvyāḍhakī: ṭhañ with luk, then ṅīp by dvigoḥ (4.1.21, SK 479).
• dvyācitikī, dvyācitīnā; with ṭhañ-luk dvyācitā, because aparimāṇabistācitakambalyebhyo na taddhitaluki (4.1.22, SK 480) bars ṅīp after ācita.
• dvipātrikī, dvipātrīṇā, dvipātrī.
• BM: the luk by 1693 takes only ṭhañ, not ṣṭhan or kha, or teaching them would be pointless.
• 1721 kulijāl lukkhau ca 5.1.55 — after a dvigu ending in kulija, luk (of ṭhañ) and kha optionally, and by ca ṣṭhan. Where none applies, ṭhañ is heard.
• Four forms:
• dvikulijī: ṭhañ with luk, then ṅīp.
• dvaikulijikī: ṭhañ kept. The final member gets no vṛddhi, because kulija is also excluded at 1683 (BM, TB).
• dvikulijīnā: kha.
• dvikulijikī: ṣṭhan.
• BM: "optionally" comes from the anuvṛtti of anyatarasyām. An NSP editorial note (printed under sk.1718.n1) disagrees: anyatarasyām does not continue, because kha is named again.

I. First case: share, price, wages; "that is its measure"; numeral names — running example: prasthaṃ parimāṇam asya → prāsthiko rāśiḥ

• 1722 so 'syāṃśavasnabhṛtayaḥ 5.1.56 — the affixes after a first-case word, "that is its share (aṃśa = bhāga), price (vasna = mūlya) or wages (bhṛti = vetana)". pañcakaḥ: one whose share, price or wages is five (kan). The source prints vastra; NSP reads vasna.
• 1723 tad asya parimāṇam 5.1.57 — the affixes after a first-case word, "that is the measure of this": prasthaṃ parimāṇam asya → prāsthiko rāśiḥ, a heap measuring a prastha (ṭhañ).
• BM, TB: here parimāṇa covers anything that delimits, number included, because 1724 qualifies number by parimāṇa: ṣāṣṭikam / ṣāṣṭikaḥ.
• TB: the fresh tad asya stops the luk of 1693 in dviṣāṣṭikaḥ. BM denies the affix after dviṣaṣṭi etc. for want of usage; BM probably means the numeral sixty-two. An NSP note takes tad asya as merely for clarity.
• 1724 saṃkhyāyāḥ saṃjñāsaṃghasūtrādhyayaneṣu 5.1.58 — 1723 continues. After a numeral, the affixes come when a name, a group, a sūtra[-text] or a recitation is meant.
• vārttika saṃjñāyāṃ svārthe pratyayo vācyaḥ: in a name the affix comes in the word's own sense: pañcaiva pañcakāḥ śakunayaḥ.
• Or else: the numeral denotes the number alone (as in dvyekayoḥ, SK 186), and the affix comes in the sense of what has that measure: "five is their measure".
• Group: pañcakaḥ.
• sūtra: aṣṭakaṃ pāṇinīyam, Pāṇini's text of eight [adhyāyas]. sūtra is named separately because saṃgha is restricted (rūḍha) to groups of living beings.
• Recitation: pañcakam adhyayanam, recited five times.
• vārttika stome ḍavidhiḥ: ḍa when a stoma (a Sāman laudation of the Chandogas) is meant.
• pañcadaśa mantrāḥ parimāṇam asya → pañcadaśaḥ (ṭi-lopa); also saptadaśaḥ.
• ekaviṃśaḥ: the ti of viṃśati is deleted by ti viṃśater ḍiti (SK 844). This is what the ḍ is for (BM).
• 1725 paṅktiviṃśatitriṃśaccatvāriṃśatpañcāśatṣaṣṭisaptatyaśītinavatiśatam 5.1.59 — nipātana in the sense "that is its measure": paṅkti, viṃśati, triṃśat, catvāriṃśat, pañcāśat, ṣaṣṭi, saptati, aśīti, navati, śata. The kaumudī calls them conventional words (rūḍhiśabda).
• BM, TB set out the Bhāṣya's derivations, e.g. paṅkti = pañcan + ti, a metre of five five-syllable pādas. BM: the Bhāṣya concludes that they are underived stems.
• 1726 pañcaddaśatau varge vā 5.1.60 — nipātana, optionally, when a group (varga) is meant.
• pañca parimāṇam asya → pañcad vargaḥ: ḍati with ṭi-lopa (BM). Likewise daśat.
• Otherwise kan: pañcakaḥ, daśakaḥ.
• 5.1.61 (Vedic) is omitted here (SK 3491).
• 1727 triṃśaccatvāriṃśator brāhmaṇe saṃjñāyāṃ ḍaṇ 5.1.62 — ḍaṇ after triṃśat and catvāriṃśat, in the sense "that is its measure", when a brāhmaṇa text is named.
• triṃśad adhyāyāḥ parimāṇam eṣāṃ brāhmaṇānām → traiṃśāni: ādivṛddhi, with the ṭi deleted by ṭeḥ.
• Likewise cātvāriṃśāni, which NSP follows with brāhmaṇāni.

J. Second case: "deserves that" (tad arhati), and the close of the ārhīya affixes — running example: śvetacchatram arhati → śvaitacchatrikaḥ

• 1728 tad arhati 5.1.63 — ṭhañ and the rest after a second-case word, "deserves that". The kaumudī glosses it as "is fit to obtain", which supplies an object for the second case (BM).
• śvetacchatram arhati → śvaitacchatrikaḥ: the ārhīya ṭhak, which reaches this sense because 1681 includes it. The source has śvetacchatrarhati; NSP reads śvetacchatram arhati.
• TB adds ṭhañ examples: prāsthikaḥ, drauṇikaḥ.
• 1729 chedādibhyo nityam 5.1.64 — after cheda etc., in the sense "constantly deserves that". The kaumudī takes nityam as repetition (ābhīkṣṇya), qualifying the sense.
• chedaṃ nityam arhati → chaidiko vetasaḥ: the cane, which grows again after cutting (ṭhak).
• gaṇasūtra virāga viraṅgaṃ ca: virāga, and viraṅga in its place: vairāgikaḥ, vairaṅgikaḥ.
• BM and TB, following the Bhāṣya, both conclude against this construal: nitya only means "not optional", and the sūtra is dispensable. The kaumudī is translated on its own terms.
• 1730 śīrṣacchedād yac ca 5.1.65 — yat, and by ca ṭhak, after śīrṣaccheda, in the same sense.
• śīrṣacchedaṃ nityam arhati → śīrṣacchedyaḥ / śairṣacchedikaḥ, one who constantly deserves beheading. The source has śīraśchedaṃ; NSP reads śīrṣacchedaṃ.
• śīrṣa for śiras is a nipātana, [only] in conjunction with yat and ṭhak. BM: otherwise śīrṣaṃś chandasi (SK 3514) allows it only in the Veda.
• 1731 daṇḍādibhyo yat 5.1.66 — yat after daṇḍa etc. The yat is carried forward from 1730 (BM, TB). The AS wording is daṇḍādibhyaḥ; the SK heading prints yat.
• daṇḍam arhati → daṇḍyaḥ.
• arghyaḥ. The source and NSP OCR read ardhyaḥ; the daṇḍādi member is argha, "price, honorific offering" (BM, Amara).
• vadhyaḥ.
• BM, TB: the reading daṇḍādibhyo yaḥ is wrong, against the Bhāṣya and the accent.
• 5.1.67 (Vedic) is omitted here (SK 3492).
• 1732 pātrād ghaṃś ca 5.1.68 — ghan after pātra, and by ca yat: gha → iya, pātriyaḥ; pātryaḥ. Both mean "deserves a vessel" (BM).
• 1733 kaḍaṅkaradakṣiṇāc cha ca 5.1.69 — cha after kaḍaṅkara and dakṣiṇā, and by ca yat.
• kaḍaṅkara is analysed kaḍaṃ karoti, with khac by nipātana from this very sūtra (and so mum). It means the stalks of māṣa, mudga and the like; the source prints mudra.
• kaḍaṅkarīyaḥ / kaḍaṅkaryaḥ (gauḥ), a cow that deserves such fodder.
• dakṣiṇām arhati → dakṣiṇīyaḥ / dakṣiṇyaḥ.
• 1734 sthālībilāt 5.1.70 — cha and yat carried forward, after sthālībila ("the inside of a cooking-pot"): sthālībilīyāḥ / sthālībilyāḥ taṇḍulāḥ, grains fit for cooking. TB: they are apavādas of ṭhak.
• 1735 yajñartvigbhyāṃ ghakhañau 5.1.71 — gha and khañ respectively after yajña and ṛtvij, in the sense "deserves that".
• yajñam arhati → yajñiyaḥ (gha → iya, the a deleted).
• ārtvijīno yajamānaḥ, a sacrificer who deserves an officiating priest (khañ → īna, ādivṛddhi). yajamāna attaches to ārtvijīnaḥ. TB, reading ca for the source's vā, refers both forms to the sacrificer.
• vārttika yajñartvigbhyāṃ tatkarmārhatīty upasaṃkhyānam: the same affixes in the sense "deserves the work of that": yajñiyo deśaḥ, ārtvijīna ṛtvik. BM: the vārttika shows that the affixes here take words only in their primary sense.
• Close: the domain of the twelve ārhīya affixes, ṭhak and the rest, has ended. ataḥ paraṃ ṭhañ eva: from here on, ṭhañ alone. The source lacks this sentence; NSP prints it, and BM cites it under sk.1736.
• BM: the thirteen of 1701, minus the heading ṭhañ, are the twelve. Only the prāgvatīya ṭhañ continues into p35 (pārāyaṇa…, SK 1736).