Māṁgallu grant of Dānārṇava

Seal

Plates

⟨4⟩ -sam(ā)sādi⟨5⟩ta- ⬦ -sam{y}āsādi⟨5⟩ta- VR; -sam¿ya?⟨ā⟩sādi⟨5⟩ta- PVPS • The vowel mark for ā is extended downward in a curve much like that for subscript r. This may be an ornamental extension of the stroke or an error made by an illiterate scribe, but definitely does not look like a subscript y.

⟨5⟩ -vaśīkr̥tārāti- ⬦ -vaś¿i?⟨ī⟩kr̥tārāti- VR PVPS.

⟨7⟩ {-viṣṇaṣ(ṇu)vi-} ⬦ {-viṣṇuṣūvi-} VR PVPS • At this locus, VR marks the text from pratā to vallabhendrasya as superfluous, while PVPS discards the string from pratā to (the first) vallabhendrasya. There is no way to tell which iteration of the (imperfectly) repeated string was inscribed in error, but PVPS’s excision of a string stretching across both iterations can not, in my opinion, be reasonably derived from scribal error; and since the second instance of vallabhendrasya contains a scribal correction (see next apparatus entry), I assume that its presence in the text was approved by a contemporary editor. Thus, by my reasoning, the whole of the first iteration was rejected, and the text re-inscribed from satyaśrāya onward, ignoring the fact that, if the rejected string were explicitly deleted, this string ought to have started with ssa. — ⟨7⟩ -va⟨⟨lla⟩⟩bhendrasya • The addition is written below the line and to the right of bha, i.e. one character to the right of where one would expect it to be. — ⟨7⟩ kubja- VR ⬦ kubba- PVPS.

⟨10⟩ paṁcaviṁśa⟦(r)⟧⟨⟨t⟩⟩i¿ma?⟨M⟩ ⬦ paṁcaviṁśa¿kr?⟨t⟩i¿ma?⟨M⟩ VR; paṁcaviṁśat¿i?⟨ī⟩¿ma?⟨M⟩ PVPS • I believe that an originally inscribed r has been corrected somewhat clumsily to t.

⟨12⟩ u◯¿py?⟨cc⟩āṭya ⬦ u◯¿py?⟨tp⟩āṭya VR PVPS • The emendation proposed by previous editors does not presuppose a smaller scribal error, and related grants use uccāṭya; utpāṭya never occurs as far as I am aware.

⟨14⟩ dhyarddha- ⬦ dhyardha- VR PVPS.

⟨15⟩ -v¿o?⟨i⟩kramāditya- ⬦ -v¿e?⟨i⟩kramāditya- VR PVPS.

⟨17⟩ ¿bhe?⟨rbha⟩ka ⬦ bheka VR PVPS • The reading is clear, but unattested elsewhere, while the form arbhaka-vijayāditya is attested in the Koḻūru grant of Bhīma II.

⟨19⟩ cā¿ṇ?⟨n⟩y¿a?⟨e⟩ ⬦ cānye VR PVPS • The principal consonant definitely has the shape of ṇ, but the subscript y is joined to its foot rather than to its wing. The character is not ne by any means, though it must have been intended for such.

⟨20⟩ saptā{śa}ntar¿o?⟨e⟩ ⬦ saptā{śa}ntare VR; saptā{sa}ntare PVPS • PVPS’s sa is probably a typo. — ⟨20⟩ -rāja⟨ṁ⟩ kr̥ta- ⬦ -rāja-kr̥ta⟨ṁ⟩ VR; -rājaṁ kr̥ta- PVPS • While there is no original anusvāra, nor space for one in its usual position, PVPS’s reading is better as an emendation and equally permitted by the metre. I have supplied hyphenation in PVPS’s reading.

⟨22⟩ bhuvaM ⬦ bhuvaṁ VR PVPS.

⟨23⟩ °ābdā¡ṁ!⟨ñ⟩ jita-ripur agama(t) VR ⬦ °ābdā(n) agama(t) PVPS • Typographic omission in PVPS, with some strangeness around the n.

⟨25⟩ ¡vacadharasya! ⬦ vac¿a?⟨o⟩(dha)¿r?⟨v⟩asya VR; {va}ca¿dha?⟨tu⟩rasya PVPS • In addition to an extra light syllable in this word (where the poet apparently availed of a licence to use three light syllables where two would be expected), the first pāda of this stanza also has a caesura between a prefix and a verb, which is also permissible only by licence. VR’s emendation, while it improves this particular word, only aggravates the metrical anomaly. PVPS’s emendation is metrically and grammatically correct, but is too heavy-handed to accept, while with his alternative proposal ca varasya is awkward and bland (not to mention that this is one of the very few instances of dha that does not look like va, so it should probably not be emended to va). If I am correct in interpreting the composer’s intent (for which see the translation), then the meaning could have been expressed in correct prosody, e.g. as vacasaḥ pateḥ kamala-saṁbhūtasya (though the inferior caesura is not remedied by this suggestion).

⟨26⟩ -¿ś?⟨g⟩rām¿ū?⟨ya⟩tvā° ⬦ -grām¿ū?⟨ya⟩tvā° PVPS; -grāmy¿ā?⟨ya⟩tvā° VR.

⟨27⟩ nindā(rha)m • The problematic character looks rather like hyā, with the subscript y unfinished.

⟨28⟩ ¿u?⟨ya⟩d-vaida◯gdhyam PVPS ⬦ udvaidagdhyam VR.

⟨30⟩ vvicāryyate ⬦ vvicāryate VR PVPS.

⟨31⟩ -mahārājā(dh)irāja- PVPS ⬦ -mahārāja- VR.

⟨32⟩ ma⟨33⟩⟦tī⟧ntri- ⬦ ma⟨33⟩ntri- VR PVPS • I cannot interpret the character tī (or possibly ti) written directly below vā in the line above and ignored by both previous editors. In the facsimile I see no indication that it has been deleted, but can only assume that it was engraved in error for the ntri required here, and the correct character was re-engraved to the right of it. It was probably not meant to be an addition, since it is equal or almost equal in size to the regular characters, whereas other interlinear additions are much smaller. There is also no apparent place where such an addition would have been required. This character was engraved before the rest of line 33 and slightly higher than the other characters of that line; the correct ntri is to the right of and below this one.

⟨33⟩ -tī{E}⟨r⟩tth¿o?⟨ā⟩dhyakṣam ⬦ -tī{E}rthādhyakṣam VR; -tī{E}rthādhyakṣ¡am!⟨ān⟩ PVPS • PVPS’s emendation of the ending is plausible, but the composer appears to have used the singular in a collective sense.

⟨34⟩ -nimitt¿e?⟨aṁ⟩ PVPS ⬦ -nimitte VR • There may in fact be an anusvāra or a final M after this word.

⟨35⟩ sāma(nta)- • The principal consonant of the problematic character in fact has the shape of g, and the subscript component is distorted. The intent must have been nta. — ⟨35⟩ -vo◯(ṭṭ/dd)i- PVPS ⬦ -voḍḍi- VR • PVPS’s preferred reading is voṭṭi, which he connects to the name viṣṭi, see the commentary. He mentions voddi as a possible reading only in his discussion. I agree that both these are possible and each is more likely than VR’s voḍḍi.Throughout the inscription, ḍ has a tail that curls conspicuously upward (cf. e.g. l32 ḍi). At face value, ddi may be the most likely reading; compare l51 ddi, which is practically identical in shape to what we have here. Conversely, ṭ should normally lack a neck, see e.g. ṭṭa in ll 12, 20 and 31. However, l53 ṭi does have a neck and looks quite identical to the upper part of the present character. Given the textual parallels cited by PVPS, I accept ṭṭi as the most likely reading.

⟨36⟩ -ka◯¿k?⟨r⟩aḥ VR ⬦ -ka¿t?⟨r⟩aḥ PVPS • Possibly a typo in PVPS. — ⟨36⟩ -ni⟨r⟩ddāri-(cāṭa)-bala- VR ⬦ -ni⟨r⟩ddāri¿cāṭa?⟨tāri⟩-bala- PVPS • Both previous editors show the text as clear. Indeed, -cāṭa- seems likely from the estampage, but if correct, it requires emendation. PVPS’s emendation is very plausible and is probably close to the intent of the composer, but I hesitate to introduce it to the text because it requires a major intervention. An equally invasive alternative emendation would be nirddārita-prabala-. — ⟨36⟩ -vī(ra?)- ⬦ -v¿i?⟨ī⟩(ra)- VR PVPS • The vowel mark of the first character has a dot, so I prefer to read it as the expected ī. For the second character, VR admits to uncertainty but PVPS does not. It can plausibly be read as ra, but several other readings may be possible, including ri. If the latter is correct, then the word may be emended to vairi, in which case ari is not needed in the previous locus. — ⟨36⟩ (bhaṭā?)ndhakāraḥ • VR admits to uncertainty on the problematic characters, while PVPS does not. The reading is plausible both in context and on the basis of the vestiges, but as far as I can tell, other readings may also be possible, such as balā°, varā° or even ghanā°.

⟨37⟩ sa{ṁ}t-pūjya- ⬦ saṁ{t}pūjya VR; satpūjya PVPS • While VR’s emendation is also plausible, I prefer my emendation as being less invasive and slightly smoother. PVPS may have been of the same opinion, though he ignores the superfluous anusvāra. — ⟨37⟩ -pravaṇa- ⬦ -pravaṇa⟨ḥ⟩ VR PVPS • Compare paropakāra-pravaṇa-prabhāvaḥ in line 40 below. — ⟨37⟩ -sva{(ḥ)}- • While I agree with VR and PVPS that a visarga is not wanted here, I am not sure that one was at all present in the original, though shown as clear by both previous editors. The dots may be random damage. — ⟨37⟩ {śrīmac-ca}⟨38⟩{lukyak(u)}¿bha?⟨ca⟩lukya- ⬦ śrīmac-cā⟨38⟩lukya{kubhalukya}- VR; {śrīmac-}ca⟨38⟩lukya{kubhalukya}- PVPS • The character at the end of line 37 seems in the estampage to be cca, though VR reads it as ccā. To reduce the dittography, both previous editors suppress the string kubhalukya. While this yields meaningful text with a small and straightforward suppression, it seems unlikely that these characters could have been erroneously engraved while the others were correct and deliberate. Much more likely in my opinion is that the engraver first started with śrīmac-calukya-ku, probably beiginning to engrave kula out of habit. Realising the mistake, he would have started again from the beginning of the stanza, but this time round, he neglected to close the bottom of ca, so the character as engraved turned out as bha. My emendation is thus more complex than that of the previous editors, but presupposes much more feasible scribal mistakes. The final problem with this locus is that of śrīmac. Suppressing this word results in correct prosody for the line, but so does suppressing the subsequent vaṁśa (and altering a retained vowel, so we are left with śrīmac-calukyodita-). I believe it is much more likely that śrīmac was, again out of habit, engraved unnecessarily (or perhaps conceived of as a bit of prose tagged on before the stanza). Moreover, with my emendation of the dittography, śrīmac is a natural part of the erroneously engraved string, whereas PVPS has to resort to a separate emendation to suppress it (while VR only notes that the stanza is metrically faulty, but does not attempt to correct it, though his editor does propose deleting śrīmac in a footnote.).

⟨38⟩ -śrīdvā(ram ic?)¿(c?)?⟨ch⟩ā¿g?⟨n⟩ugataṁ ⬦ -śrīdvāra-madhyānugataṁ VR; -śrīdvāram ¿adhyā?⟨ājñā⟩nugataṁ PVPS • My reading and emendation are tentative. See the commentary for a discussion. In the actual reading, the character dvā is mostly clear. The ostensible ra-ma is wholly indiscernible in the estampage, though read as clear by both editors. I see no explicit indication of an i here, but a smallish vowel marker (like that in bhūmipāla earlier in this line) could conceivably have fit in the damaged spot above consonant read as m. For the next character, dhyā is a plausible reading, but far from the only possibility. In fact, vyā is at least as likely, and out of context, I would be most inclined to read it as dyā. My suggested reading, ccā (intended for cchā) may be wishful thinking, but seems to be permitted by the facsimile. Next, the character read as nu by both previous editors is in fact a perfect specimen of gu, though nu is a very plausible emendation. There seems to be a short vertical punctuation mark at the end of this word, though no previous editor reads it and it is definitely superfluous. It is also possible that what looks like an anusvāra and a punctuation mark is in fact a final M, though this would likewise be out of place. — ⟨38⟩ n¿a?⟨ā⟩n¿(ū)?⟨yā⟩⟨39⟩(dh)ipa-dvāram ⬦ nanu ⟨39⟩ vipad-⟨d⟩vāram VR; ¿nanu?⟨nūnaṁ⟩ ⟨39⟩ vipad-⟨d⟩vāram PVPS • VR’s editor in a footnote proposes emending the first word of this string to nānyo. It is not clear to me how he sees this as an improvement to the text without additional alterations. Reading nanu would be unmetrical, and the plate seems rather to have nanū. I find that emending ū to yā is a minor intervention, and there are two instances in the text where u must clearly be emended to ya (lines 26 and 28), so we know the scribe is prone to this kind of mistake. I thus emend the first two characters in a way similar to that suggested by VR’s editor. But in the next line, I read dhipa-dvāram in preference to vipa-dvāram (which requires supplying an extra d), since most instances of dh in this text are indistinguishable from v. See the commentary for further discussion.

⟨39⟩ a¿ma?⟨haṁ⟩ PVPS ⬦ ama VR • VR’s editor proposes the same emendation in a footnote. It very probably reflects the intent of the composer, though given that the stanza as a whole is quite unintelligible, something else may have been intended, e.g. amuṁ, imaṁ or aho. — ⟨39⟩ viśām¿i?⟨ī⟩ty akṣ(e?)ṣṭa ⬦ viśāmi tyakṣaṣṭa- VR; viśāmīty akṣa⟨ṁ⟩¿ṣṭ?⟨st⟩a PVPS • VR’s editor in his footnote proposes viśāmīty ācaṣṭa. A dot above the right-hand side of kṣa may be seen as confirmation for PVPS’s emendation, but its position is not where anusvāra is normally placed in this inscription and the dot is not read as such by either previous editor. Moreover, I find PVPS’s interpretation unconvincing. See the commentary for further discussion. — ⟨39⟩ vā(ṭaṁ) gata- PVPS ⬦ -vāṭaṁ-gata- VR • The damaged character read as ṭaṁ might conceivably be ḍa, tha, dha, pa, ma, la or va. An additional problem is segmentation into compounds and independent words. Although PVPS’s text is spaced, he too apparently construed both of these words as in compound to the next word, which I find problematic. See the commentary for further discussion.

⟨40⟩ pa(ro)pa(kā)ra- • The character ro has an extra curly stroke on each side. These may be random, or ornamental elaborations of the vowel mark, or perhaps a remnant of correction from ko. In turn, kā lacks the expected arms. — ⟨40⟩ -¿ph?⟨p⟩rabhāvaḥ ⬦ -prabhāvaḥ VR PVPS. — ⟨40⟩ arātīn¿th?⟨dh⟩ana- ⬦ arātīndhana- VR PVPS.

⟨41⟩ -kauśal¿a?⟨e⟩na ⬦ -k¿o?⟨au⟩śalena VR PVPS.

⟨42⟩ °⟨rājaḥ| Abhūt su⟩to PVPS • PVPS’s emendation is perfectly plausible, though the intended wording may have been slightly different. VR offers no emendation, while his editor in a footnote suggests °rājaḥ suto bhavad, which is inferior because it requires inserting text both before and after the inscribed to. — ⟨42⟩ -van(d)yanāṁ(bā)(y)¿a?⟨ā⟩m • The problematic component of what we expect to be ndya may be dh. bā in fact looks like ha, with the tail of the vowel mark curved back under the body. (ya) has some extra strokes including a small loop at the bottom centre, possibly traces of correction from a different character.

⟨43⟩ guṇḍyanā⟦(ka)⟧khy⟦o⟧⟨⟨aḥ⟩⟩ ⬦ guṇḍyanā{(ta)}khyaḥ VR PVPS • Where previous editors see a superfluous ta, I believe ka was first engraved, intended for kya in error for khya. This may or may not have been explicitly deleted. There is also a stroke for o above the correct khya, so probably khyo was engraved and then corrected to khyaḥ. PVPS suggests that the deleted/superfluous character and the next one together could be alternatively read as ryyaḥ. In spite of some resemblence to ryya, I feel this is not likely, but the possibility cannot be excluded. — ⟨43⟩ pratāpāya(ti)⟨44⟩tām PVPS ⬦ pratāpāyya (ci)⟨44⟩tām VR.

⟨44⟩ °ārā¡dhdh!y(o) ⬦ °ārādhyo VR PVPS • The problematic character looks more like rdhdhya.

⟨45⟩ -nāmadheyena(|) ⬦ -nāmadheyen¿ā?⟨a⟩ VR; -nāmadheyena PVPS. — ⟨45⟩ (c/v)e(lā)⟨46⟩paṟṟu- ⬦ velā⟨46⟩paṟṟu- VR PVPS • PVPS uses the name Celāpaṟṟu in his discussion, though he prints velāpaṟṟu in his edition. It seems likely to me that the character is indeed ce, as the upper outline of its body has a a definite notch and is joined directly to the headmark rather than to the stem. Compare e.g. vi at the end of the previous line versus cā in line 30. However, see also the next entry on viddamayyaḥ, where I read the same glyph as v.

⟨46⟩ °saṁbhū⟨to⟩ (v)iddamayyaḥ ⬦ °saṁbhū⟨to⟩ ciddamayyaḥ VR PVPS • The problematic character does indeed look like ci (see also the previous entry on velāpaṟṟu). However, since the names Viddamayya, Viddamiya and Viddamaśarman are attested in several related inscriptions, while I know of no occurrence of Ciddamayya, I prefer to read this character as vi. Note that both previous editors may have vacillated on the reading, since both of them supply to for the preceding word rather than taś, as would be expected before c.

⟨47⟩ j¿a?⟨ā⟩⟨48⟩t(yā) ⬦ jā⟨48⟩tyā VR PVPS. — ⟨47⟩ j¿a?⟨ā⟩⟨48⟩t(o) ⬦ jā⟨48⟩tyā VR PVPS • In the estampage I do not see a full subscript y, much less yā where the vowel stroke ought to be attached to the end of y. I think this character is to, with vowel strokes on the right of the headmark and the bottom left of the body. The reading of the previous editors, along with their emendation (see next entry) makes the text more awkward, see also the note to the translation.

⟨48⟩ bhūdeva- ⬦ bhūdeva⟨ḥ⟩ VR PVPS • In my perception the emendation proposed by the previous editors makes the text more awkward regardless of the reading in the previous entry. See also the note to the translation. — ⟨48⟩ tasyā{ḥ}bhūn ⬦ tasyābhūn VR PVPS. — ⟨48⟩ mācemāṁbāyā⟨ṁ⟩ ⬦ mācemāṁbāyāṁ VR; macemāṁbāyāṁ PVPS • The short a in PVPS is probably a typo. I see no original anusvāra here. The composer’s intent may also have been mācemāṁbāyāḥ. Note also that the e must be short for the line to be metrical. — ⟨48⟩ sūnu⟨r⟩ ddommana- ⬦ sūnur ddommana- VR PVPS.

⟨49⟩ -¿ś?⟨g⟩u¿no?⟨ṇa⟩- ⬦ -¿ś?⟨g⟩u(ṇa)- VR PVPS. — ⟨49⟩ saṁpann¿e?⟨a⟩s VR ⬦ saṁpa¿ṇ?⟨n⟩nas PVPS.

⟨50⟩ durāpe PVPS ⬦ durāp¿e?⟨o⟩ VR. — ⟨50⟩ kākartya- PVPS ⬦ kāka{rt}tya- VR.

⟨51⟩ āha¿ppā?⟨ḫpa⟩ti- ⬦ āha⟨ḥ⟩p{p}¿ā?⟨a⟩ti- VR; āharppati- PVPS • I assume that the upper p with the vowel mark (which PVPS reads as a repha) was meant to be an upadhmānīya. This presupposes a much smaller scribal error than VR’s equivalent emendation and is preferable to PVPS’s non-standard reading. — ⟨51⟩ ka⟨r⟩ppaṭam VR ⬦ karppaṭam PVPS. — ⟨51⟩ (ābaddhaṁ) VR ⬦ āra(bdhaṁ) PVPS • Previous editors admit to uncertainty only in the last character of this string. That could indeed be bdhaṁ from what I can make out, but the preceding character is much more likely from the vestiges to be ba than ra. Compare also baddha-karppaṭakaḥ in line 21 of the Cevuru plates of Amma I.

⟨52⟩ br¿ā?⟨a⟩hma¿v?⟨c⟩aryya¿c?⟨m⟩ ⬦ br¿ā?⟨a⟩hmacaryya(m) VR PVPS. — ⟨52⟩ ¿ga?⟨I⟩ty- PVPS ⬦ gaty- VR • Given that ga is consistently inscribed instead of initial I throughout the inscription, I fully endorse PVPS’s emendation.

⟨53⟩ ka◯¿ppā?⟨rppa⟩ṭi- ⬦ ka¿ppā?⟨rppra⟩ṭī- VR; karppaṭī- PVPS • VR’s emendation to rppra seems to be a typo for rppa. — ⟨53⟩ anuṣṭhānam ⬦ anuṣ¿ṭ?⟨ṭh⟩ānam VR; anuṣ¿ta?⟨ṭhā⟩nam PVPS • PVPS’s t and a are probably typos; his actual reading would have been indentical to VR’s.

⟨54⟩ i(daṁ?) PVPS ⬦ iha VR • PVPS’s reading is far from clear (though he prints it as such). Nonetheless I prefer it because VR’s reading is unmetrical, and because there is some space after the problematic character. The noise in that space may well include an anusvāra, and what seems to be the hook of ha may in turn be random noise. — ⟨54⟩ tasm(ai) PVPS ⬦ tasme VR • VR’s editor suggests emendation to tasmai, but the second stroke for that vowel may actually be present. PVPS prints this reading as clear, but I am not sure if he does so after a scrutiny of the estampage or out of negligence. — ⟨54⟩ do¿t?⟨m⟩mana-nāmne PVPS ⬦ dotmāna-nāmne VR • The expected emendation is also suggested by VR’s editor. The ā in VR’s tmā may be a typo. PEM’s etext emends here to ātmano nāmnā. I do not know if this is her own emendation or if it comes from an unnamed source. It is in line with PVPS’s opinion (which I endorse) that Dānārṇava issued this charter in his own name rather than in lieu of Amma II, but I am confident this was not the intent of the composer.

⟨55⟩ sa¿vvā?⟨rvv⟩- ⬦ sa⟨r⟩vva- VR; sarvva- PVPS. — ⟨55⟩ -pū⟨r⟩vvam ⬦ -pūrvvam VR PVPS. — ⟨55⟩ -ni⟨56⟩mitt(e) ⬦ -ni⟨56⟩mittaṁ VR PVPS • There is definitely no anusvāra here; however, part of an e is in my opinion discernible. Even if that is noise, I would still prefer emending to nimitte in spite of the hiatus, since that is the form used in practically all related grants that mention an occasion (with a few occurrences of nimittena and none, that I know of, of nimittam).

⟨56⟩ Agrahārīkr̥ty(ā){tta(ca)}smābhir VR ⬦ Agrahārīkr̥ty(ā){ttacha}smābhir PVPS • It is possible that the text has tya rather than tyā, and that one or both of the superfluous characters was intended for A. The second superfluous character, apart from a malformed ca or cha with the headmark in the wrong place, could be a not-so-malformed rai with only the lower left vowel stroke present and the top one omitted. — ⟨56⟩ ddatta ⬦ datta VR PVPS.

⟨57⟩ pū¿m?⟨rv⟩vataḥ ⬦ pūrvvataḥ VR PVPS. — ⟨57⟩ koḍupulū¿v?⟨r⟩i VR PVPS • I adopt the emendation of the Telugu on blind faith.

⟨58⟩ laṁjiya(mā)ḍa- ⬦ laṁjayamāḍa- VR; laṁjiyavāḍa- PVPS • While the i is clear, the only problem with mā is that the vowel marker is attached to the left arm rather than the right. It may be a correction of ma to vā, or of vā to or mā.

⟨59⟩ pallikaṇṭi- PVPS ⬦ pallikaṇṭī- VR.

⟨61⟩ pulu◯gudla- ⬦ pūlagudla- VR; puluguḍla- PVPS • The word definitely begins with pulu. The second part may be guḍla, but, without understanding it and judging by the glyph alone, gudla is much more likely. — ⟨61⟩ kartta¡vv!yā ⬦ kartta(vyā) VR PVPS.

⟨62⟩ ba⟨63⟩hubhi¿m?⟨r v⟩asudhā ⬦ ba⟨63⟩hubhir vvasudhā VR PVPS.

⟨63⟩ dat⟨t⟩ā ⬦ dattā VR PVPS.

⟨65⟩ -¿(l)?⟨v⟩a(rṣ)a- ⬦ -varṣa- VR PVPS. — ⟨65⟩ k¿u?⟨r̥⟩m¿a?⟨i⟩ḥ ⬦ kr̥miḥ VR; k¿r̥?⟨ri⟩miḥ PVPS • There is definitely a typo in PVPS’s edition, but I am not sure what his actual reading may have been. — ⟨65⟩ (bh)ūy¿ā?⟨o⟩ ⬦ bhūyo VR PVPS.