तिङन्तक्र्यादिप्रकरणम्

Covers SK 2554–2562 (9 sūtras): all present. Every unit has its own kaumudī. The spine is not an Aṣṭādhyāyī run. The SK follows the Dhātupāṭha order of the kryādi roots (ours 1473 krī … 1533 grah; the print numbers one higher) and teaches each rule at the root where it first bites, drawing on AS books 3, 6, 7 and 8 (3.1.81 → 3.1.82 → 8.3.77 → 3.1.83 → 7.3.80 → 6.4.2 → 7.2.46 → 6.4.19 → 7.2.37). The AS-contiguous vikaraṇa run 3.1.81–83 is split by 8.3.77 (SK 2556), which the SK puts beside skanbhu, the root it names. Every AS-ref below is checked against references/ashtadhyayi/sutraani-data.json.

• Anuvṛtti from elsewhere in the SK. Most rules here carry words from sūtras taught earlier:
• kartari from kartari śap (3.1.68, SK 2167) at 2554;
• śiti from ṣṭhivuklamucamāṃ śiti (7.3.75, SK 2320) at 2558;
• saṃprasāraṇasya (6.3.139, SK 1004) and dīrghaḥ from ḍhralope… (6.3.111, SK 174) at 2559;
• iṭ from ārdhadhātukasyeḍ valādeḥ (7.2.35, SK 2184) at 2560 and 2562, and vā from svaratisūti… (7.2.44, SK 2279) at 2560;
• ekācaḥ from ekāca upadeśe 'nudāttāt (7.2.10, SK 2246) at 2562.
• One comes from later: cchvoḥ śūḍ anunāsike ca (2561) draws in kvijhaloḥ kṅiti by ca from anunāsikasya kvijhaloḥ kṅiti (6.4.15, SK 2666), which the SK teaches only later.
• 2556's nityam is there to stop the vā of sivādīnāṃ vā… (8.3.71, SK 2359) running on (TB).

tiṅantakryādiprakaraṇam — outline (p52, SK 2554–2562, all 9 sūtras)

p51 (tiṅantatanādiprakaraṇam) closed at 2553 (suṭ kāt pūrvaḥ, the saṃskṛ forms).

• Where the prakaraṇa begins. The print heads it "tiṅante kryādayaḥ 49". Our prakaraṇa number 52 differs systemically. sk2554 opens with the Dhātupāṭha entry 1473 ḍukrīñ dravyavinimaye (print 1474), then kryādibhyaḥ śnā (3.1.81).
• Where it ends. At 2562 with 1533 grah (print 1534): grahītā … agrahīṣata. The print closes "iti kryādayaḥ" ("Here end the kryādi roots"). Ours lacks it, so the English supplies it in brackets (scan page-0427). p53 opens with the print heading "tiṅante curādayaḥ 50" and 1534 cura steye (print 1535; SK 2563).
• Contents in outline.
• 2554: krī and the roots to skuñ. śnā for śap, ī before a hal-initial kṅit, ā-loss before a vowel: krīṇāti, krīṇītaḥ, krīṇanti; pramīṇāti, mamau; sināti.
• 2555–2557: skuñ and the sautra roots stambhu etc.; yuñ to pūñ. śnu beside śnā, ṣ after vi, and śānac for śnā before hi: skunoti / skunāti, viṣṭabhnoti / viṣṭabhnāti, viṣkabhnāti, stabhāna; yunāti, knūnāti, drūṇāti.
• 2558: pū and the pvādi roots, the ṝ-final run, jyā. Short vowel before a śit: punāti, lunāti, stṛṇāti, vṛṇāti, dhunāti, śṛṇāti, ṛṇāti.
• 2559: jyā, then rī … kuṣ. Long vowel for a samprasāraṇa after a hal, which the short vowel of 2558 then shortens: jināti; jānāti, badhnāti, abhānsīt, śrathnāti, mṛdnāti.
• 2560: kuṣ with nir, then kṣubh … khac. Optional iṭ: niṣkoṣitā / niṣkoṣṭā; kṣubhnāti, kliśnāti, iṣṇāti, khacñāti.
• 2561: khav, heṭh, grah. ūṭh for v: khaunāti, khaunīhi; hiṭhṇāti, gṛhṇāti.
• 2562: grah. Long iṭ outside liṭ: grahītā, agrahīt, but jagrahitha.

Threads that run through the prakaraṇa:

• śnā is the class-mark. kryādibhyaḥ śnā (2554) puts śnā where śap would come. As an apit sārvadhātuka it is ṅit (sārvadhātukam apit, 1.2.4, SK 2234; BM). That has three effects:
• no guṇa of the root: krīṇāti, punāti, yunāti;
• loss of a penultimate n (aniditāṃ hala upadhāyāḥ kṅiti, 6.4.24, SK 415): viṣṭabhnāti, badhnāti, śrathnāti, kuthnāti;
• samprasāraṇa by grahijyā… (6.1.16, SK 2412): jināti, gṛhṇāti.
• What happens to the ā of śnā. Before a hal-initial kṅit it becomes ī (ī haly aghoḥ, 6.4.113, SK 2497): krīṇītaḥ, krīṇīte, punīte. Before a vowel-initial kṅit it is lost (śnābhyastayor ātaḥ, 6.4.112, SK 2483): krīṇanti, krīṇate. At 2554 the ant for jh and the at for jh come first, by paratva, nityatva and antaraṅgatva.
• What happens to the n of śnā. It becomes ṇ after pra (hinumīnā, 8.4.15, SK 2530): pramīṇāti. ṇ is barred after kṣubh (kṣubhnādiṣu ca, 8.4.39, SK 792): kṣubhnāti. ścutva gives khacñāti, but it is blocked after ś by śāt (8.4.44, SK 112): kliśnāti. ṣṭutva gives mṛḍṇāti and hiṭhṇāti.
• Before hi. After a hal-final root, śnā → śānac and hi is lost (2557): stabhāna, badhāna, mṛdāna, kṣubhāna. After a vowel it stays: krīṇīhi, śṛṇīhi. ūṭh, being later than śānac, leaves khav no longer hal-final, so there is no śānac: khaunīhi (2561).
• The ṝ-final roots (stṝ, kṝ, vṝ, śṝ, pṝ, dṝ, ṝ …). They share the same set of operations:
• the vikaraṇa takes the short ṛ of 2558: stṛṇāti;
• guṇa even in the kit liṭ by ṛcchaty ṝtām (7.4.11, SK 2383): tastaratuḥ, cakare. For śṝ, dṝ and pṝ that guṇa alternates with a short vowel (śṝdṝprāṃ hrasvo vā, 7.4.12, SK 2495): śaśratuḥ / śaśaratuḥ;
• ir by ṝta id dhātoḥ (7.1.100, SK 2390): stīryāt, īryāt. After a labial (the dento-labial v included) it is ur by udoṣṭhyapūrvasya (7.1.102, SK 2494): vūryāt, pūryāt, avūrṣṭa;
• an optionally long iṭ (vṝto vā, 7.2.38, SK 2391): staritā / starītā. It is not long before liṅ (na liṅi, 7.2.39, SK 2529) or before the parasmaipada sic (sici ca parasmaipadeṣu, 7.2.40, SK 2392): stariṣīṣṭa, astārīt.
• Gaṇa-internal boundaries. "badhnātyantāḥ parasmaipadinaḥ" (2558): the roots from śṝ up to bandh are parasmaipadin. At plī the print has "vṛt", which closes the lvādi roots, and some say the pvādi too (2559). On that view vrī, bhrī and kṣīṣ keep the long vowel. "itaḥ parasmaipadinaḥ" follows after śranth (2559).
• The kaumudī lists forms bare. Lakāra and person labels are the translator's, given in Pāṇinian terms. Derivation stages follow BM and TB as recorded in the unit notes.
• Where the commentary lies. Only the kaumudī blocks are translated (SK-TRANSLATE §25).
• TB printed under a neighbouring unit: 2554 and 2555 at the tail of sk.2546.t1, after TB's "iti tanādayaḥ"; 2562 inside sk.2561.t1. sk.2562.t1 carries, after TB's colophon, the TB on SK 2563, which belongs to p53.
• BM lead-in printed under the preceding unit: 2559, the jyā derivation, is at the tail of sk.2558.b1, which ends "… jinātīti sthite —".
• NSP footnotes: p. 416 fn 1, on 2559's dīrghanirdeśasāmarthyāt (the Śekhara on the aṅgavṛtta-paribhāṣā), is OCR'd into sk.2560.t1 after "niraku-". It is not TB. p. 417 fn 1 is on 2562's ekācaḥ and fn 2 on SK 2563's svārthe. Both are filed as sk.2560.n1.
• Wikisource (uttara-2/0075–0076) covers 2554–2558 only; 2559 onward falls in its gap.
• Vārttika marks: in 2559 k2, the ✻ sentence is Haradatta's dictum. BM says no such vārttika is seen, so it is not labelled a vārttika. The vārttika in 2560 k2 on iṣ with śyan is a genuine one.

Where the p47–53 audit (.work/sk-run/p47-53-audit/sk2554-2562.md) corrects a reading, this outline follows the corrected reading. The fixer is applying these to the en.yml notes in parallel. The audit found no must-fix items.

• 2558: on avūrṣṭa, TB (scan page-0426) says that because liṅ and sic are kit by uś ca (1.2.12, SK 2368), guṇa does not occur, and so u for ṝ results (by 2494). The en.yml note's "u after guṇa is barred" gets TB backwards.
• 2559: BM's "śaśrāya śaśraya" is OCR (थ read as य) for śaśrātha śaśratha, which is also the print kaumudī's reading. It is not a BM variant.
• 2559: the cross-references are SK 2558 (ours and the print have 2258) and SK 2509 (ours 1509, the print 2509).
• 2560: ścutva in kliśnāti / khacñāti is the palatal substitution for the n of śnā by stoḥ ścunā ścuḥ (8.4.40, SK 111). It is not a "ś-class" substitution.
• 2562: BM's counter-example for vihitasya is grāhitam, the ṇic form with vṛddhi (TB, scan page-0427). BM's "grahitam" is OCR. The tail of that BM sentence is garbled, and only its certain part is used.
• 2554: the cikriyitha note should add that Wikisource agrees with BM's cikrayitha.

Print readings (NSP, scan-confirmed) that the outline follows:

• krīṇate (2554; ours repeats krīṇāte);
• cuskāva given once (2555);
• udoṣṭhyapūrvasya (2558; ours udeṣṭhya-);
• ārṇāt and ārṇītām (2558; ours arṇāt; BM agrees with the print);
• bharaṇa ity eke, śrathnāti and śranthe[ḥ] (2559);
• the colophon "iti kryādayaḥ" (2562).

Still open for Ajai (mūla):

• 2554 krītāt in the parasmaipada āśīrliṅ slot. It is in ours and the print (scan page-0425), but Wikisource has krīyāt, the form grammar expects. The English keeps krītāt as transmitted. The note should become a textual note.
• 2554 cikriyitha (ours = print) vs cikrayitha (BM and Wikisource). thal is pit, so there is no iyaṅ. The English follows BM. Already flagged.
• 2555 cuskave (ours = print) vs cuskuve (Wikisource), which is expected for the kit ātmanepada liṭ. Already flagged.
• 2555 sūtra spelling: stanbhu… (ours = Wikisource) vs stambhu… (print and BM's pratīka). Our kaumudī itself says stambhvādayaḥ. This is noted but not escalated.
• 2559 abhānsīt (ours = print = TB's pratīka) vs abhāntsīt (BM). abhāntsīt is the required form: vṛddhi, bhaṣ, then dh → t before the kept s, parallel to bhantsyati and abhāntsuḥ. It is not yet flagged in the notes.
• 2561 heṭha ca … hiṭhṇāti (ours) vs hiṭha ca … hiṭhṇāti (print, scan page-0427). BM has heṭha. A root heṭh cannot yield hiṭhṇāti. This is not yet escalated.

Ordering: SK order throughout.

A. krī: śnā for śap; ī for, or loss of, śnā's ā — running example: krī + śap + ti → krī + śnā + ti → krīṇāti; krī + nā + tas → krīṇītaḥ; krī + nā + jhi → krī + nā + anti → krīṇanti

• 2554 kryādibhyaḥ śnā 3.1.81 — after krī and the kryādi roots, śnā comes before a sārvadhātuka denoting the agent, in the root's own sense. It is an apavāda of śap.
• kartari comes from kartari śap (SK 2167; BM).
• Root entry 1473 ḍukrīñ dravyavinimaye (print 1474).
• krīṇāti: śnā is apit, so ṅit (SK 2234), so the ī of krī takes no guṇa (BM).
• krīṇītaḥ: the ā of śnā → ī before the hal-initial kṅit tas (SK 2497).
• krī + nā + jhi: ant for the jh by jho 'ntaḥ (7.1.3, SK 2169) comes before the ī, as para, nitya (it applies whether or not ī has been made) and antaraṅga. Then the ā is lost before the vowel (SK 2483): krīṇanti.
• Ātmanepada: at for the jh by ātmanepadeṣv anataḥ (7.1.5, SK 2258; TB) likewise comes first. The forms are krīṇīte, krīṇāte, and krīṇate (print).
• liṭ: cikrāya, cikriyatuḥ.
• Madhyama singular: cikrayitha (BM, Wikisource; ours and the print cikriyitha) / cikretha. iṭ is optional before thal by the Bhāradvāja restriction (ṛto bhāradvājasya, 7.2.63, SK 2296).
• cikriyiva: iṭ by the kṛādi restriction (7.2.13, SK 2293). cikriyiṣe.
• kretā, kreṣyati.
• āśīrliṅ slot: krītāt (ours = print; Wikisource krīyāt; for Ajai). Ātmanepada kreṣīṣṭa.
• luṅ akraiṣīt, akreṣṭa. BM adds krīṇātu, krīṇīhi, akrīṇāt, krīṇīyāt.
• 1474 prīñ "satisfying and desiring": prīṇāti, prīṇīte. 1475 śrīñ "cooking".
• 1476 mīñ "injuring":
• pramīṇāti: ṇ after pra by hinumīnā (SK 2530).
• ā where an ec would arise (mīnātiminoti…, 6.1.50, SK 2508): mamau, mimyatuḥ, mamitha / mamātha; mimye; mātā, māsyati; mīyāt, māsīṣṭa; amāsīt, amāsiṣṭām, amāsta.
• 1477 ṣiñ "binding": sināti, sinīte, siṣāya, setā.
• 1478 skuñ āpravaṇe. The meaning is unexplained by either commentator. It leads into 2555.

B. skuñ and the sautra roots stambhu etc.: śnu beside śnā, ṣ after vi, śānac before hi — running example: vi + stanbh + śnu / śnā + ti → vi-stabh-nu-ti / -nā-ti → viṣṭabhnoti / viṣṭabhnāti; stabh + nā + hi → stabh + āna + hi → stabhāna

• 2555 stanbhustunbhuskanbhuskunbhuskuñbhyaḥ śnuś ca 3.1.82 — after stanbhu, stunbhu, skanbhu, skunbhu and skuñ, śnu comes, and by ca śnā as well (carried from 2554).
• skuñ with śnu: skunoti, skunute. With śnā: skunāti, skunīte.
• liṭ cuskāva, and ātmanepada cuskave (ours = print; Wikisource cuskuve; for Ajai). skotā; askauṣīt, askoṣṭa.
• The four stambhu etc. are sautra roots and n-penultimate (BM, TB). Some say all four mean obstructing. Mādhava says the first and third mean stiffening, the second extracting and the fourth holding. All four are parasmaipadin.
• vi + stanbh + nu + ti: the n is lost before the ṅit vikaraṇa (SK 415), and the s → ṣ after vi (stambheḥ, 8.3.67, SK 2272): viṣṭabhnoti, viṣṭabhnāti.
• avaṣṭabhnoti: ṣ after ava (8.3.68, SK 2273).
• liṭ avataṣṭambha: ṣ in the abhyāsa too (8.3.64, SK 2277).
• luṅ: aṅ is optional (jṝstambhu…, 3.1.58, SK 2291). vyaṣṭabhat has aṅ, with ṣ across the aṭ by prāk sitād aḍvyavāye 'pi (8.3.63, SK 2276). vyaṣṭambhīt has none.
• stunbhu: stubhnoti, stubhnāti.
• 2556 veḥ skabhnāter nityam 8.3.77 — after vi, ṣ replaces the s of skabhnāti, invariably.
• "skabhnāti" means the root skanbhu, not only its śnā form, so viṣkambhitā is covered too (TB).
• nityam blocks the vā of SK 2359 (TB).
• vi + skabh + nu + ti → viṣkabhnoti, viṣkabhnāti.
• skunbhu, without vi: skubhnoti, skubhnāti.
• 2557 halaḥ śnaḥ śānaj jhau 3.1.83 — after a hal, śānac replaces śnā when hi follows.
• stabh + nā + hi → stabh + āna + hi: the ś and c of śānac are it. hi is lost after a (ato heḥ, 6.4.105, SK 2202; BM), giving stabhāna.
• TB: the sannipāta-paribhāṣā does not save hi, because ato heḥ was taught at all.
• stubhāna, skabhāna, skubhāna. TB: the rule reaches the sautra roots too, against those who give stabhnīhi.
• With śnu, hi stays after the u: stabhnuhi.
• 1479 yuñ: yunāti, yunīte, yotā. 1480 knūñ: knūnāti, knavitā. 1481 drūñ: drūṇāti. 1482 pūñ pavane leads into 2558.

C. pū, the ṝ-final roots, jyā: the short vowel before a śit, and the long samprasāraṇa after a hal — running example: pū + nā + ti → pu + nā + ti → punāti; jyā + nā + ti → ji + nā + ti → jī + nā + ti → ji + nā + ti → jināti

• 2558 pvādīnāṃ hrasvaḥ 7.3.80 — for pū and the rest, a short vowel replaces the root vowel before a śit. śiti comes from ṣṭhivuklamucamāṃ śiti (SK 2320; BM, TB).
• punāti, punīte. pavitā: seṭ, being ū-final.
• 1483 lūñ: lunāti, lunīte.
• 1484 stṝñ: stṛṇāti, stṛṇīte.
• tastāra. tastaratuḥ: guṇa despite the kit ending, by ṛcchaty ṝtām (SK 2383).
• staritā / starītā: the iṭ is optionally long (vṝto vā, SK 2391). stṛṇīyāt.
• āśīrliṅ stīryāt: ir by SK 2390, lengthened by hali ca (8.2.77, SK 354).
• Ātmanepada āśīrliṅ: iṭ is optional (liṅsicor ātmanepadeṣu, 7.2.42, SK 2528), and the iṭ is not long before liṅ (na liṅi, SK 2529). This gives stariṣīṣṭa. Without iṭ, liṅ is kit by uś ca (SK 2368), so there is no guṇa: stīrṣīṣṭa.
• luṅ astārīt, astāriṣṭām: no long iṭ, by SK 2392. Ātmanepada astariṣṭa / astarīṣṭa / astīrṣṭa.
• 1485 kṝñ: kṛṇāti; cakāra, cakare (SK 2383).
• 1486 vṝñ: vṛṇāti; vavāra, vavare; varitā / varītā.
• vūryāt: u for ṝ after a labial, by udoṣṭhyapūrvasya (SK 2494, print reading). The dento-labial v counts (TB). Then r and lengthening.
• variṣīṣṭa / vūrṣīṣṭa; avārīt.
• Ātmanepada luṅ avariṣṭa, avarīṣṭa, and avūrṣṭa: liṅ and sic are kit by uś ca, so there is no guṇa, and u comes (TB, per audit).
• 1487 dhūñ: dhunāti.
• dudhavitha / dudhotha: iṭ is optional (SK 2279). dudhuviva.
• dhotā / dhavitā.
• adhāvīt: iṭ by stusudhūñbhyaḥ parasmaipadeṣu (7.2.72, SK 2385). Ātmanepada adhaviṣṭa / adhoṣṭa.
• "badhnātyantāḥ parasmaipadinaḥ": from here to bandh the roots are parasmaipadin.
• 1488 śṝ:
• śaśratuḥ: the short vowel of SK 2495, then yaṇ. Otherwise śaśaratuḥ, with guṇa by SK 2383.
• śaśariva / śaśriva: the kṛādi restriction overrides śryukaḥ kiti (7.2.11, SK 2381).
• śaritā / śarītā; śṛṇīhi; śīryāt; aśāriṣṭām.
• 1489 pṝ: papratuḥ / paparatuḥ; pūryāt (SK 2494).
• 1490 vṝ; 1491 bhṝ (bharaṇa ity eke); 1492 mṝ: mṛṇāti, mamāra.
• 1493 dṝ: dadaratuḥ / dadratuḥ. 1494 jṝ (some read jhṝ or dhṝ); 1495 nṝ; 1496 kṝ.
• 1497 ṝ: ṛṇāti; arāṃcakāra (ām, the root counting as having a heavy vowel; BM); aritā / arītā.
• laṅ ārṇāt, ārṇītām (print). īryāt; ārīt.
• 1498 gṝ. 1499 jyā: samprasāraṇa by grahijyā… (SK 2412) and pūrvarūpa give ji + nā + ti. That leads into 2559 (BM, tail of sk.2558.b1).
• 2559 halaḥ 6.4.2 — a long vowel replaces an aṅga ending in a samprasāraṇa vowel that comes after a hal belonging to the aṅga.
• saṃprasāraṇasya comes from SK 1004 and dīrghaḥ from SK 174. aṅgasya is read twice: once as a genitive of the part, "a hal that is part of the aṅga", and once as a genitive of the place, with tadantavidhi (BM).
• BM's examples: vītaḥ, śūnaḥ, jīnaḥ. The counter-examples: utaḥ (no hal); nirutam, durutam (the hal is not part of the aṅga); vidhyati (the aṅga does not end in the samprasāraṇa) (BM, TB).
• ji + nā + ti → jī + nā + ti by this rule, then → ji + nā + ti by pvādīnāṃ hrasvaḥ (SK 2558; ours and the print misprint 2258): jināti. jijyau, jijyatuḥ.
• 1500 rī. reṣaṇa is the howl of wolves.
• 1501 lī: optional ā where an ec would arise (vibhāṣā līyateḥ, SK 2509; our cross-reference reads 1509). lalau / lilāya; lātā / letā.
• 1502 vlī: vlināti. 1503 plī: "vṛt". The lvādi roots close here, and some say the pvādi too.
• 1504 vrī, 1505 bhrī (bharaṇa ity eke, print) and 1506 kṣīṣ take the short vowel if the pvādi do not close at vṛt. Some say they do not.
• 1507 jñā: jā before a śit (jñājanor jā, 7.3.79, SK 2511): jānāti. Because jā is stated with a long vowel, no short vowel comes. TB defends this against the objection that the long vowel's force is spent as a jñāpaka of the aṅgavṛtta-paribhāṣā.
• 1508 bandh:
• badhnāti: n lost by SK 415.
• babandhitha. BM also gives babanddha.
• bandhā: the t of tās → dh (jhaṣas tathor dho 'dhaḥ, 8.2.40, SK 2280).
• bhantsyati: b → bh by ekāco baśo bhaṣ… (8.2.37, SK 326), dh → t by car.
• badhāna: śānac by 2557.
• luṅ abhānsīt (ours = print = TB; BM abhāntsīt; for Ajai): vṛddhi (vadavrajahalantasyācaḥ, 7.2.3, SK 2267) and bhaṣ.
• abāndhām: by pūrvatrāsiddham (8.2.1, SK 12), sic is first lost by jhalo jhali (8.2.26, SK 2281). That loss is asiddha for pratyayalakṣaṇa (1.1.62, SK 262), so no bhaṣ comes from the s-initial affix. t → dh by SK 2280. The dh may be lost by jharo jhari savarṇe (8.4.65, SK 71; TB), and BM writes abānddhām.
• abhāntsuḥ.
• 1509 vṛṅ saṃbhaktau: vṛṇīte; vavre, vavṛṣe, vavṛḍhve; avṛta.
• 1510 śranth. From here on the roots are parasmaipadin. śrathnāti (print), with the n lost.
• k2: Haradatta and others say that where liṭ is kit (by "śranthi-granthi…", a rule of another grammar; BM), e and the loss of the abhyāsa must also be stated.
• k3: śrethatuḥ, śrethuḥ. Sudhākara extends the kit status to the pit endings: śrethitha, and with ṇal śaśrātha.
• In the uttama, Mādhava gives śaśrātha / śaśratha. "The basis must be sought": BM says there is no basis and no such vārttika is seen.
• 1511 manth.
• 1512 śranth again and 1513 granth. śranth is read twice for its different meaning.
• 1514 kunth: kuthnāti, cukuntha. Durga reads the root as kuth: cukotha.
• 1515 mṛd: mṛdnāti, mṛdāna (śānac). 1516 mṛḍ: mṛḍṇāti (ṣṭutva of the n).
• 1517 gudh: gudhnāti. 1518 kuṣ: kuṣṇāti, koṣitā. This leads into 2560.

D. kuṣ after nir: optional iṭ; then kṣubh … khac — running example: nis + kuṣ + tās → niṣkoṣitā / niṣkoṣṭā

• 2560 niraḥ kuṣaḥ 7.2.46 — after kuṣ preceded by nir, a val-initial ārdhadhātuka takes iṭ optionally.
• iṭ comes from SK 2184 and vā from SK 2279 (BM, TB). "Why nir?" koṣitā, koṣitum (TB).
• niṣkoṣitā / niṣkoṣṭā.
• luṅ nirakoṣīt. Without iṭ, nirakukṣat, with ksa by śala igupadhād aniṭaḥ ksaḥ (3.1.45, SK 2336).
• 1519 kṣubh: kṣubhnāti, kṣubhnītaḥ, with no ṇ by kṣubhnādiṣu ca (SK 792). kṣobhitā; kṣubhāna (śānac).
• 1520 ṇabh and 1521 tubh: nabhnāti, tubhnāti. With śap: nabhate, tobhate. With śyan: nabhyati, tubhyati.
• 1522 kliśū:
• kliśnāti: the palatal substitution (ścutva, SK 111) of the n of śnā is barred after ś by śāt (SK 112), per audit.
• kleśitā / kleṣṭā; akleśīt / aklikṣat.
• 1523 aś: aśnāti; āśa (no nuṭ, the root not having two hals; BM).
• 1524 udhras: dhrasnāti, taking the u as it. Others make it part of the root: udhrasāṃcakāra.
• 1525 iṣ: iṣṇāti.
• "They say" tīṣasaha… (7.2.48, SK 2340) takes only the tudādi iṣ, by its association with sah (bhvādi, being named with śap). On that view iṣṇāti gets only eṣitā.
• Bhaṭṭoji says the option is right for iṣṇāti too. k2 gives the vārttika: the prohibition before t is for the iṣ that takes śyan.
• 1526 viṣ: viṣṇāti, veṣṭā. 1527 pruṣ and 1528 pluṣ: pruṣṇāti, pluṣṇāti (BM reads prūṣ).
• 1529 puṣ: poṣitā, seṭ, since only the śyan puṣ is aniṭ (BM, TB). 1530 muṣ: moṣitā.
• 1531 khac bhūtaprādurbhāve, "being born after the due time": khacñāti (ścutva). Some say the root ends in v (khav), which leads into 2561.

E. khav: ūṭh for v before the nasal of śnā — running example: khav + nā + ti → kha + ū(ṭh) + nā + ti → khaunāti; khav + nā + hi → khaunīhi

• 2561 cchvoḥ śūḍ anunāsike ca 6.4.19 — ś replaces ch (with its tuk) and ūṭh replaces v before a nasal. By ca the same happens before kvi and a jhal-initial kṅit, drawn in from SK 2666 (BM, TB).
• The c in the sūtra's "cch" stands for the tuk. BM gives the ś examples: praśnaḥ, viśnaḥ (naṅ, from pracch and vicch).
• khav + nā + ti: the n of nā is the nasal, and the v → ūṭh. vṛddhi of a + ū by etyedhatyūṭhsu (6.1.89, SK 73; BM) gives khaunāti.
• cakhāva, khavitā.
• khav + nā + hi: ūṭh is later than śānac. Once ūṭh is made the root no longer ends in a hal, so 2557 cannot apply: khaunīhi (BM).
• 1532 heṭh (print hiṭh; for Ajai), also bhūtaprādurbhāve: hiṭhṇāti (ṣṭutva).
• 1533 grah upādāne: svaritet, so ubhayapadin.
• gṛhṇāti, gṛhṇīte: samprasāraṇa before the ṅit śnā (SK 2412), then pūrvarūpa.
• BM adds the liṭ paradigm jagrāha, jagṛhatuḥ … jagṛhe. It leads into 2562.

F. grah: the long iṭ, except in liṭ — running example: grah + iṭ + tās → grah + ī + tā → grahītā; but liṭ jagrahitha

• 2562 graho 'liṭi dīrghaḥ 7.2.37 — a long vowel replaces the iṭ enjoined after monosyllabic grah, but not in liṭ.
• iṭ is carried from SK 2184 and recast as a sixth-case word; ekācaḥ comes from SK 2246 (BM).
• grahītā. In liṭ, jagrahitha.
• "Why ekācaḥ?" So that it does not apply in yaṅluk: jāgrahitā (BM, TB).
• "Why vihitasya?" The ṇic form grāhitam (per audit; BM's OCR has grahitam): after ṇi is lost the iṭ follows grah but was not enjoined after it (TB).
• NSP footnote (filed as sk.2560.n1): ekācaḥ is said in deference to Haradatta. It is inconsistent with agrahīṣṭām if aṭ comes at the lakāra stage.
• gṛhyāt.
• grahīṣīṣṭa: na liṅi (SK 2529) does not bar the length, because it carries vṝtaḥ (BM, TB).
• luṅ: no vṛddhi, by hmyantakṣaṇa… (7.2.5, SK 2299): agrahīt, agrahīṣṭām.
• Ātmanepada: agrahīṣṭa, agrahīṣātām, agrahīṣata.
• Close: "iti kryādayaḥ" (Here end the kryādi roots). It is supplied from the print; BM and TB close alike. p53 opens with curādi, 1534 cura (SK 2563).