अजन्तपुल्लिङ्गप्रकरणम्

Covers SK 178–286 (109 sūtras): all present.

ajanta-puṃliṅga-prakaraṇa — outline (p08, SK 178–286, all 109 sūtras)

Deriving the masculine declension of every vowel-final stem by running it through the twenty-one sup endings. Ten outcome-groups: first the shared machinery (A), then one full paradigm per stem-type (B–J), each group running a concrete stem from su (nom. sg.) to sup (loc. pl.). Saṃjñā and paribhāṣā sūtras are placed where they first bite; the ṇatva/ṣatva/na-lopa operations sit inside the paradigm whose form they alter. Ordering note: the base ṇatva rule raṣābhyāṃ no ṇaḥ (235) is printed by the SK only in group D, but its extension (197) appears in B where ṇatva first fires — the two are cross-referenced.

A. Preliminaries every declension needs — running example: what licenses attaching su … sup to rāma

• 178 arthavad adhātur apratyayaḥ prātipadikam 1.2.45 — a meaningful word-form (arthavat) that is neither a dhātu nor an affix nor affix-ending is a prātipadika. So bare rāma qualifies; the ṭi-type underived words (ḍittha) are the model.
• 179 kṛttaddhitasamāsāś ca 1.2.46 — kṛt-ending, taddhita-ending, and compound forms are also prātipadika (the samāsa mention is niyama: prātipadika-hood of a pada-containing aggregate belongs only to a compound, not a sentence).
• 180 pratyayaḥ 3.1.1 — adhikāra: whatever is henceforth taught is a pratyaya (affix), governing to the end of adhyāya 5. The su etc. attached to rāma are affixes by this heading.
• 181 paraś ca 3.1.2 — adhikāra: that affix comes after its base — hence rāma+su, not su+rāma.
• 182 ṅyāpprātipadikāt 4.1.1 — adhikāra (through adhyāya 5): the following affixes come after a ṅī-ending, āp-ending, or prātipadika base — the slot that admits sup after rāma.
• 183 svaujasamauṭ… ṅyossup 4.1.2 — the twenty-one sup affixes: su au jas / am auṭ śas / ṭā bhyām bhis / ṅe bhyām bhyas / ṅasi bhyām bhyas / ṅas os ām / ṅi os sup — attached to rāma. The u of su, e of ṅasi, and ja/śa/ṭa/ṅa/pa are it-markers.
• 184 vibhaktiś ca 1.4.104 — sup (and tiṅ) are also termed vibhakti; the seven triads carry the old names prathamā … saptamī.
• 185 supaḥ 1.4.103 — of sup, each successive triad is termed ekavacana / dvivacana / bahuvacana (sg./du./pl.).
• 186 dvyekayor dvivacanaikavacane 1.4.22 — the dual and singular are used in the senses of two-ness and one-ness respectively (rāma sg. vs. du.).
• Note: the it-reading rules 189, 190, 195 and the number-saṃjñā 187 are placed in B where they first act on rāma's affixes.

B. The a-stem paradigm — running example: rāma, nom. sg. rāmaḥ → loc. pl. rāmeṣu

• 187 bahuṣu bahuvacanam 1.4.21 — the plural is used for many; and the nom. sg. is realized here: rāma+su → rutva (su→sur) → visarga → rāmaḥ.
• 188 sarūpāṇām ekaśeṣa ekavibhaktau 1.2.64 — of like-formed words in one case, one alone remains (ekaśeṣa): "two Rāmas" rāma+rāma+au → one rāma, then vṛddhi (vṛddhir eci) → rāmau.
• 189 cuṭū 1.3.7 — a cu/ṭu at an affix's start is it: the initial j of jas (nom. pl.) is it.
• 190 na vibhaktau tusmāḥ 1.3.4 — tu-varga, s, m are not it inside a vibhakti: the final s of jas stays.
• 191 ato guṇe 6.1.97 — after non-final a, pararūpa before guṇa; but by paratva the pūrvasavarṇa-dīrgha wins here (191 is apavāda only to akaḥ savarṇe, not to prathamayoḥ): rāma+as → a+a → ā → rāmāḥ (nom. pl.).
• 192 ekavacanaṃ sambuddhiḥ 2.3.49 — the nom. sg. used in address is termed sambuddhi (voc. sg.).
• 193 eṅhrasvāt sambuddheḥ 6.1.69 — after an aṅga ending in eṅ or a short vowel, the sambuddhi's hal (its su) is elided: he rāma, he rāmau, he rāmāḥ. (The eṅ-clause secures voc. of hari/viṣṇu once sambuddhi-guṇa makes them eṅ-final: he hare, he viṣṇo.)
• 194 ami pūrvaḥ 6.1.107 — before the am of acc. sg., the pūrvarūpa (prior a) is the single substitute: rāma+am → rāmam; du. rāmau.
• 195 laśaku ataddhite 1.3.8 — initial l, ś, ku of a non-taddhita affix is it: the ś of śas (acc. pl.) is it.
• 196 tasmāc chaso naḥ puṃsi 6.1.103 — in the masculine, the s of śas after the pūrvasavarṇa-dīrgha → n: rāma+as → rāmā+s → rāmān (acc. pl.).
• 197 aṭkupvāṅnumvyavāye 'pi 8.4.2 — ṇatva (base rule raṣābhyāṃ no ṇaḥ, SK 235) fires even when aṭ / ku / pu / āṅ / num intervene between r-ṣ and n: this licenses the ṇ of rāmeṇa (r … ṇ separated by the labial m) and rāmāṇām.
• 198 padāntasya 8.4.37 — ṇatva does not touch a pada-final n: the n of rāmān (acc. pl.) is safe.
• 199 yasmāt pratyayavidhis tadādi pratyaye 'ṅgam 1.4.13 — the form beginning with the base from which an affix is taught is termed aṅga when that affix follows (rāma before sup).
• 200 aṅgasya 6.4.1 — adhikāra: the following operations apply "of an aṅga."
• 201 ṭāṅasiṅasām ināt syāḥ 7.1.12 — after an a-stem, ṭā→ina, ṅasi→āt, ṅas→sya: rāma+ina → rāmeṇa (ṇatva, inst. sg.); rāmāt (abl. sg.), rāmasya (gen. sg.).
• 202 supi ca 7.3.102 — the final a of the aṅga → long before a yañ-initial sup: rāma+bhyām → rāmābhyām.
• 203 ato bhisa ais 7.1.9 — after an a-stem, bhis→ais (anekāl, a whole-substitute): rāma+ais → rāmaiḥ (inst. pl.).
• 204 ṅer yaḥ 7.1.13 — after an a-stem, dative ṅe→ya: rāma+ya → (dīrgha by 202 via sthānivadbhāva) → rāmāya.
• 205 bahuvacane jhalyet 7.3.103 — the aṅga's final a → e before a jhal-initial plural sup: rāma+bhyas → rāmebhyaḥ (dat./abl. pl.); it also feeds the e of rāmeṣu.
• 206 vā 'vasāne 8.4.56 — at pause, a jhal optionally → car: abl. sg. rāmā+āt → rāmāt / rāmād (with doubling, four forms).
• 207 osi ca 7.3.104 — the aṅga's final a → e before os: rāma+os → rāme+os → rāmayoḥ (gen./loc. du.).
• 208 hrasvanadyāpo nuṭ 7.1.54 — ām (gen. pl.) takes the augment nuṭ after a short-vowel / nadī / āp stem: rāma+nām.
• 209 nāmi 6.4.3 — a vowel-final aṅga → long before nām: rāmā+nām → rāmāṇām (ṇatva); the sannipāta-paribhāṣā that would block it is itself set aside by ārambha-sāmarthya.
• 210 apadāntasya mūrdhanyaḥ 8.3.55 — adhikāra (through the pāda): a non-pada-final dental → mūrdhanya (ṣatva).
• 211 iṇkoḥ 8.3.57 — adhikāra: that retroflexion operates on a sound standing after iṇ or ku.
• 212 ādeśapratyayayoḥ 8.3.59 — an s that is a substitute or part of an affix, after iṇ/ku and non-final, → ṣ: loc. pl. rāma+su → (e by 205) rāme+su → rāmeṣu. (Kṛṣṇa, Mukunda decline alike.)

C. The sarvanāman (pronoun) sub-system — running example: sarva, nom. pl. sarve, gen. pl. sarveṣām

• 213 sarvādīni sarvanāmāni 1.1.27 — sarva and the sarvādi-gaṇa are sarvanāman; the saṃjñā reaches tadanta forms too (jñāpaka: dvandve ca).
• 214 jasaḥ śī 7.1.17 — after an a-pronoun, jas→śī (anekāl, whole-substitute): sarva+śī → sarve (nom. pl.).
• 215 sarvanāmnaḥ smai 7.1.14 — after an a-pronoun, dative ṅe→smai: sarvasmai.
• 216 ṅasiṅyoḥ smātsminau 7.1.15 — after an a-pronoun, ṅasi→smāt, ṅi→smin: sarvasmāt (abl. sg.), sarvasmin (loc. sg.).
• 217 āmi sarvanāmnaḥ suṭ 7.1.52 — ām after an a-varṇa pronoun takes augment suṭ (s): sarva+s+ām → (et) sarve-s-ām → (ṣatva) sarveṣām. (Lists the 35 sarvādi words.)
• 218 pūrvaparāvaradakṣiṇottarāparādharāṇi vyavasthāyām asaṃjñāyām 1.1.34 — pūrva, para, avara, dakṣiṇa, uttara, apara, adhara are sarvanāman in fixed relative sense (not as proper name); the otherwise-invariable saṃjñā turns optional before jas: pūrve / pūrvāḥ.
• 219 svam ajñātidhanākhyāyām 1.1.35 — sva is sarvanāman (optionally, before jas) when not meaning kinsman or wealth: sve / svāḥ.
• 220 antaraṃ bahiryogopasaṃvyānayoḥ 1.1.36 — antara is sarvanāman (optionally, before jas) in the senses "outer" and "undergarment": antare / antarāḥ.
• 221 pūrvādibhyo navabhyo vā 7.1.16 — after the nine (pūrva etc.), smāt/smin optionally replace ṅasi/ṅi: pūrvasmāt / pūrvāt, pūrvasmin / pūrve.
• 222 na bahuvrīhau 1.1.29 — no sarvanāman-saṃjñā when forming a bahuvrīhi: tvatka-pitṛka, matka-pitṛka (akac blocked already in the prakriyā).
• 223 tṛtīyāsamāse 1.1.30 — no sarvanāman in a tṛtīyā-tatpuruṣa: māsapūrvāya (dat.), likewise its analytic sentence māsena pūrvāya.
• 224 dvandve ca 1.1.31 — no sarvanāman (of the aggregate) in a dvandva: varṇāśrametarāṇām (gen. pl., not -eṣām).
• 225 vibhāṣā jasi 1.1.32 — in a dvandva the saṃjñā returns optionally before jas (only for śī-bhāva): varṇāśrametare / varṇāśrametarāḥ.
• 226 prathamacaramatayālpārdhakatipayanemāś ca 1.1.33 — prathama, carama, taya-ending, alpa, ardha, katipaya, nema are optionally sarvanāman before jas: prathame / prathamāḥ, dvitaye / dvitayāḥ, neme / nemāḥ (vārttika adds tīya-ending before ṅit: dvitīyasmai / dvitīyāya).

D. The pada / bha / sarvanāmasthāna apparatus and the anomalous a/ā/an-stems it governs — running examples: jarā/nirjarā, pāda, yūṣan, ahan, viśvapā

• 227 jarāyā jaras anyatarasyām 7.2.101 — jarā→jaras optionally before a vowel-initial vibhakti (substitute of the jara-part only, by nirdiśyamānasya): nirjaras-au, nirjaras-aḥ / (halādi) like rāma.
• 228 paddannomās… prabhṛtiṣu 6.1.63 — for pāda, danta, nāsikā, māsa, hṛdaya, etc., the substitutes pad, dat, nas, mās, hṛt… appear optionally from śas onward: padaḥ / pādān (acc. pl.), padā / pādena (inst. sg.).
• 229 suḍ anapuṃsakasya 1.1.43 — the suṭ (first five sup: su au jas am auṭ) are sarvanāmasthāna for a non-neuter stem.
• 230 svādiṣv asarvanāmasthāne 1.4.17 — before a non-sarvanāmasthāna sup (up to kap), the prior element is termed pada.
• 231 yaci bham 1.4.18 — before such a sup beginning with y or a vowel, the prior element is termed bha.
• 232 ā kaḍārād ekā saṃjñā 1.4.1 — adhikāra: one item takes only one saṃjñā (the para / anavakāśa one). So before ac-initial śas etc. a form is bha, not pada → no jaśtva: dataḥ, datā; jaśtva where it is pada: dadbhyām. (māsaḥ; mābhyām by rutva→yatva→ya-lopa.)
• 233 bhasya 6.4.129 — adhikāra: the following operations are "of a bha."
• 234 allopo 'naḥ 6.4.134 — the a of a stem-final an is elided in the bha context (before ya/ac-initial non-sarvanāmasthāna sup): yūṣan+as → yūṣṇ-aḥ → yūṣṇaḥ (acc. pl.), yūṣṇā (inst. sg.).
• 235 raṣābhyāṃ no ṇaḥ samānapade 8.4.1 — the base ṇatva rule: n→ṇ after r/ṣ within one pada → the ṇ of yūṣṇaḥ, yūṣṇā (its across-intervener extension was 197).
• 236 nalopaḥ prātipadikāntasya 8.2.7 — a prātipadika-pada's final n is elided; being asiddha (tripādī), it blocks dīrgha/etva/aistva: yūṣabhyām, yūṣabhiḥ, yūṣabhyaḥ (not yūṣā-/yūṣai-/yūṣe-).
• 237 vibhāṣā ṅiśyoḥ 6.4.136 — the a of stem-final an is optionally elided before ṅi or śī: yūṣṇi / yūṣaṇi (loc. sg.).
• 238 saṃkhyāvisāyapūrvasyāhnasyāhann anyatarasyāṃ ṅau 6.3.110 — for ahan preceded by a number / vi / sāya, the substitute ahan optionally appears before ṅi: dvyahni / dvyahani / dvyahne; vyahni…; sāyāhni….
• 239 dīrghāj jasi ca 6.1.105 — after a long vowel, no pūrvasavarṇa before jas (or ic): viśvapā+au → vṛddhi viśvapau (du.); viśvapā+jas → savarṇa-dīrgha viśvapāḥ (nom. pl.); decisive for gauryau, gauryaḥ.
• 240 āto dhātoḥ 6.4.140 — the final ā of an ā-ending dhātu (a bha-aṅga) is elided: viśvapā+śas → viśvapaḥ (acc. pl.). Contrast non-root hāhā: hāhā (inst.), hāhai (dat.), hāhe (loc.); by yoga-vibhāga at "ātaḥ" even non-roots drop ā (ktvaḥ, śnaḥ). [Ends the ā-stems; hariḥ, harī next.]

E. Short i- and u-stems and the ghi machinery — running example: hari (u-stems śambhu, guru decline alike)

• 241 jasi ca 7.3.109 — guṇa for a short-vowel aṅga before jas: hari+jas → hare+ayaḥ (guṇa i→e, then ay) → harayaḥ (nom. pl.).
• 242 hrasvasya guṇaḥ 7.3.108 — guṇa of a final short vowel in the sambuddhi: hari→hare, su elided (by 193) → he hare. Also harim (acc. sg.), harī (du.), harīn (acc. pl.).
• 243 śeṣo ghyasakhi 1.4.7 — a short i/u stem that is not nadī and not sakhi is termed ghi (basis for 244–247).
• 244 āṅo nā 'striyām 7.3.120 — ṭā (āṅ)→nā after a ghi, non-feminine: hari+nā → hariṇā (inst. sg., ṇatva). (Fem. matyā excepted.)
• 245 gher ṅiti 7.3.111 — guṇa for a ghi before a ṅit sup: dat. sg. hari+ṅe → hare+ya → haraye. (Not sakhye — sakhi is non-ghi.)
• 246 ṅasiṅasoś ca 6.1.110 — for eṅ, the pūrvarūpa is the single substitute before the a of ṅasi/ṅas: hare+aḥ → hareḥ (abl./gen. sg.); haryoḥ (gen./loc. du.), harīṇām (gen. pl.).
• 247 acca gheḥ 7.3.119 — the ghi's final becomes a (with ṅi→au per aut): hari+ṅi → harau (loc. sg.); haryoḥ, hariṣu. [śrī-pati, agni, ravi, kavi alike.]

F. sakhi and pati — the ghi exceptions, with the vṛddhi/anaṅ/su-lopa apparatus — running examples: sakhi, pati

• 248 anaṅ sau 7.1.93 — for aṅga sakhi, its final → anaṅ before non-voc su (ṅit → antādeśa): sakhi→sakhan- (→ … → sakhā).
• 249 alo 'ntyāt pūrva upadhā 1.1.65 — the sound before the final phoneme is termed upadhā (penult).
• 250 sarvanāmasthāne cāsambuddhau 6.4.8 — the upadhā of an n-final aṅga is lengthened before a non-voc sarvanāmasthāna: sakhan→sakhān- (→ nom. sg. sakhā).
• 251 apṛkta ekāl pratyayaḥ 1.2.41 — a single-phoneme affix is termed apṛkta.
• 252 halṅyābbhyo dīrghāt sutisyapṛktaṃ hal 6.1.68 — su/ti/si, when apṛkta and hal, are elided after a hal-final or long-ṅī/āp stem: sakhā, rājā (su-lopa). (First "hal" secures rājā's na-lopa, since saṃyogānta-lopa is asiddha.) he sakhe.
• 253 sakhyur asambuddhau 7.1.92 — after sakhi, a non-voc sarvanāmasthāna is treated as ṇidvat — feeding vṛddhi by 254.
• 254 aco ñṇiti 7.2.115 — vṛddhi of a vowel-final aṅga before a ñit/ṇit affix: sakhi (ṇidvat) → sakhāy- → sakhāyau (nom. du.), sakhāyaḥ (nom. pl.), sakhāyam (acc. sg.). No ghi → sakhyā (inst.), sakhye (dat.).
• 255 khyatyāt parasya 6.1.112 — the a of ṅasi/ṅas after khy/ty (yaṇ'd khi/ti) → u: sakhy+uḥ → sakhyuḥ (abl./gen. sg.).
• 256 aut 7.3.118 — ṅi→au after an i/u ghi: sakhy+au → sakhyau (loc. sg.). (Its u carried forward is uttarārtha.) [su-sakhi, ati-sakhi discussed under the ṇidvad/ghi rules.]
• 257 patiḥ samāsa eva 1.4.8 — pati is ghi only in a compound: uncompounded patyā, patye, patyuḥ, patyau; but compound bhūpatinā, bhūpataye. [kati always plural.]

G. Numerals and the luk / pratyaya-lakṣaṇa apparatus — running examples: kati, tri, dvi

• 258 bahugaṇavatuḍati saṃkhyā 1.1.23 — bahu, gaṇa, vatu-ending, ḍati-ending stems are saṃkhyā (number).
• 259 ḍati ca 1.1.25 — a saṃkhyā ending in ḍati is termed ṣaṭ (e.g. kati).
• 260 pratyayasya lukślulupaḥ 1.1.61 — an affix's disappearance by luk / ślu / lup is named luk / ślu / lup respectively.
• 261 ṣaḍbhyo luk 7.1.22 — after a ṣaṭ-stem, luk of the following jas and śas: kati+jas/śas → endings vanish.
• 262 pratyayalope pratyayalakṣaṇam 1.1.62 — a lost affix still conditions operations (so guṇa by jasi ca would otherwise reach kati).
• 263 na lumatāṅgasya 1.1.63 — but a lu-elision blocks pratyaya-lakṣaṇa for an aṅga-operation: hence no guṇa → kati (nom./acc. pl.), katibhiḥ, katīnām, katiṣu. [tri: trayaḥ, trīn, tribhiḥ.]
• 264 tres trayaḥ 7.1.53 — tri→traya before ām: trayāṇām (gen. pl.); triṣu (loc. pl.).
• 265 tyadādīnām aḥ 7.2.102 — the tyadādi pronominals' final → a before a vibhakti (iṣṭi: only through dvi): dvi+au → dvau, dvābhyām, dvayoḥ. [Ends i-stems; vātapramī, yayī, papī (uṇādi-ī, no su-lopa) follow.]

H. Long ī/ū and nadī stems — running examples: bahuśreyasī (nadī), kumārī (dhātu-ī), sudhī, pradhī

• 266 yū stryākhyau nadī 1.4.3 — long-ī / long-ū words inherently denoting a woman are termed nadī (vārttika: prathama-liṅga counts even under upasarjana).
• 267 ambārthanadyor hrasvaḥ 7.3.107 — an ambā-sense or nadī-ending stem is shortened in the sambuddhi: he bahuśreyasi; acc. pl. bahuśreyasīn.
• 268 āṇ nadyāḥ 7.3.112 — ṅit affixes after a nadī-stem take the augment āṭ (→ ā, prefixed).
• 269 āṭaś ca 6.1.90 — vṛddhi is the single substitute for āṭ + a following vowel: bahuśreyasyai (dat. sg.), bahuśreyasyāḥ (abl./gen. sg.); gen. pl. (nuṭ) bahuśreyasīnām.
• 270 ṅer ām nadyāmnībhyaḥ 7.3.116 — ṅi (loc. sg.)→ām after a nadī / āp / nī stem (nuṭ blocked by āṭ, paratva): bahuśreyasyām. [atilakṣmīḥ: uṇādi-ī → no su-lopa.]
• 271 aci śnudhātubhruvāṃ yvor iyaṅuvaṅau 6.4.77 — before a vowel, the y/v of a śnu-ending / i-u-ending-dhātu / bhrū aṅga → iyaṅ / uvaṅ (antādeśa): initially applicable in kumārī+au, and to sudhī.
• 272 er anekāco 'saṃyogapūrvasya 6.4.82 — but for the ī of a polysyllabic dhātu-aṅga not preceded by a dhātu-internal conjunct, yaṇ (before a vowel) supersedes: kumāryau, kumāryaḥ; pradhīḥ→pradhyau. (Monosyllabic nī→niyau by iyaṅ; vārttika blocks yaṇ when the prior member is neither gati nor kāraka: śuddhadhiyau.)
• 273 na bhūsudhiyoḥ 6.4.85 — for bhū and sudhī, no yaṇ — iyaṅ/uvaṅ instead: sudhiyau, sudhiyaḥ. (sakhīyati-kvip sakhīḥ etc. discussed here; vārttika: kvau luptaṃ na sthānivat.) [Ends i-stems; śambhu like hari.]

I. ṛ-stems and the polysyllabic ū-dhātu stems — running examples: kroṣṭṛ, dhātṛ, pitṛ, nṛ; khalapū, varṣābhū

• 274 tṛjvat kroṣṭuḥ 7.1.95 — kroṣṭu behaves like a tṛj-ending word (i.e. is read as kroṣṭṛ) before a non-voc sarvanāmasthāna.
• 275 ṛto ṅisarvanāmasthānayoḥ 7.3.110 — guṇa for an ṛ-final aṅga before ṅi or a sarvanāmasthāna (default, overridden for su by 276).
• 276 ṛd uśanaspurudaṃso 'nehasāṃ ca 7.1.94 — for ṛ-stems (and uśanas, purudaṃsas, anehas) the substitute anaṅ before non-voc su: kroṣṭṛ→kroṣṭan- → kroṣṭā.
• 277 aptṛn… praśāstṝṇām 6.4.11 — for ap, tṛn/tṛc-derived naptṛ etc., the penult is lengthened before a non-voc sarvanāmasthāna: kroṣṭā, kroṣṭārau, kroṣṭāraḥ; kroṣṭāram, kroṣṭūn. (Niyama: excludes pitṛ, bhrātṛ; includes udgātṛ.)
• 278 vibhāṣā tṛtīyādiṣv aci 7.1.97 — kroṣṭṛ is optionally tṛjvat before the vowel-initial endings from tṛtīyā on: kroṣṭrā / kroṣṭunā (inst.), kroṣṭre / kroṣṭave (dat.).
• 279 ṛta ut 6.1.111 — ṛ + the a of ṅasi/ṅas → single substitute u (r-para): pituḥ, dhātuḥ (abl./gen. sg.).
• 280 rāt sasya 8.2.24 — after r, only the s of a final conjunct is dropped (niyama), r→visarga: kroṣṭuḥ (nom. sg.). Gen. pl. kroṣṭūnām (nuṭ over tṛjvad, pūrva-vipratiṣedha); kroṣṭari (loc. sg.), kroṣṭroḥ (du.). [Then hūhū, nadī aticamū, khalapū.]
• 281 oḥ supi 6.4.83 — the u/ū of a polysyllabic dhātu-aṅga (not saṃyoga-pūrva) → yaṇ (v) before a vowel-initial sup: khalapvau, khalapvaḥ. (Monosyllabic lū→luvau by uvaṅ; svabhū→svabhuvau by 273.)
• 282 varṣābhvaś ca 6.4.84 — the ū of varṣābhū → v before a vowel-initial sup: varṣābhvau, varṣābhvaḥ (vārttikas add dṛnbhū, kārabhū, punarbhū; ṛvarṇāt ṇatva → dhātṝṇām). [Then dhātā, pitā (pitarau, pitaraḥ), nā/narau (nṛ).]
• 283 nṛ ca 6.4.6 — for nṛ, the long substitute is optional before nām: nṝṇām / nṛṇām. (Ends ṛ-stems; then imitated ṝ-stems kṝ/tṝ, ḷ-stems gaml̥/śakl̥ with laparatva gamalau, and e-stems se/smṛte.)

J. o- and au-stems — running examples: go, rai, glau

• 284 goto ṇit 7.1.90 — a sarvanāmasthāna after go behaves as ṇit → vṛddhi: gauḥ (nom. sg., o→au), gāvau, gāvaḥ.
• 285 auto 'mśasoḥ 6.1.93 — (split "ā, otaḥ") a final o before am/śas → single substitute ā: go+am → gām (acc. sg.), go+śas → gāḥ (acc. pl.); gavā, gave, goḥ. (Vārttika oto ṇit: sudyauḥ; smṛtau bahuvrīhi smṛtauḥ.)
• 286 rāyo hali 7.2.85 — for rai, the final → ā before a hal-initial vibhakti (before a vowel, āy-ādeśa): rāḥ (nom. sg.), rāyau, rāyaḥ; rāyā (inst.), rābhyām. Au-stem glauḥ, glāvau, glāvaḥ — here auto 'mśasoḥ does not apply (au/o not savarṇa per the śivasūtra ai auC).