^{page 583 note 1} So doubtless, as the Heavenly Source of Waters. Readers-may be surprised at the purity and fervour of these pieces as addressed to the Holy and Heavenly Stream; but it should not be forgotten that She especially represents Ahura in His attribute of ‘purity.’ She is, like Âramaiti, His daughter; she expelsdisease, and with it all unholiness.

^{page 583 note 2} Ašaya not as equalling ‘for a reward.’

^{page 583 note 3} Yasna LXVII (Sp. LXVI) here follows, but it is included with in other sections elsewhere treated.

^{page 584 note 1} Acc. in the original.

^{page 584 note 2} So elsewhere.

^{page 584 note 3} Reading zaoθra n.s.f.

^{page 584 note 4} The original is genitive.

^{page 584 note 5} The datives are not recognised.

^{page 584 note 6} The 2nd singular is not seen; was taken as a 3rd sing, ? compare the other supposed cases with their Vedic analogues.

^{page 584 note 7} Māvayačca is here first taught us to be a dative of azēm. We should not regard the throng of such initiatives on the part of the Pahl. trlr. as a matter of no importance.

^{page 584 note 8} Zaoθrē, as dat. sg. m. I varied the usual translation with a vocative in S.B.E. xxxi, see Zaoθre above, which is surely not dative; but I now recede to the Avesta dative here. There might be two words, zaoθra (see zaoθrãm, etc.) aui zaoθrya; this latter as the word usually deciphered zaoθré (paiti jamyāt), here zaoθrya; but on the whole I would now change the reading above to zaoθra with some MSS.

^{page 584 note 9} Hu + ahu (anhu). I can no longer accede to the advanced idea of ‘preparation’ here adopted, on account of a mi-ssing letter which may well have fallen out from a text. I now agree with the Pahlavi translation.

This suggestion of the Pahlavi is valid in spite of the form of the original Avesta text, which lacks a needed letter to make it correspond; havanhāi as = dat. of hu + anhu (ahu) is defective; we formerly read an havaiha, Indian ‘*savasā,’ to ‘sávas’ after the pattern of mánasa to mánas; and affording the suspiciously intelligent meaning of ‘preparation.’ Nēr. has only uttamāya at Y. XI, 26.

But the Pers. MSS. with nēk-sāhibī here and nēk-axūī at Y. XI, 26, puts us still further upon tfte track of discovering that ‘v’ has fallen out of havanhāi, which cannot mean ‘felicity,’ etc. We may read hu + anhva = havanhva after this hint, which should have been known years ago. It maybe that the ‘v’ fell out on account of the foregoing ‘v’ in.. ya.. before it in the word havanhāi; recall the curious intin. terminal anuhe for -anhve = -ase. This question was, as seen above, first pushed forward by the Persian translator. Nēr. has uttamohāya at Y. XV, 2, to -ub-; otherwise the word does not occur with him, Nēr. At Y. LXI, 17, the Pers. treats diffeiently; ste J.K.A.S. at the place. As alternative we may have ‘good mastership,’ ‘discipline.’

^{page 585 note 1} The va would make frēhdātar ī gēhānīh (so) independent of pērōzgarīh; but see the original.

^{page 585 note 2} Pavan.

^{page 585 note 3} Begarding the original as ablative, where is the Pahlavi plural form for zōhar = zaoθra ?

^{page 585 note 4} The MS. DJ, A, C 1 has nafšā; but there is no such idea in the original. ‘This nafšā might allude by anticipation to havanhai in 10 when understood as if the ‘hava’ equalled ‘hva’ = ‘own.’ DJ in Gáthas.

^{page 585 note 5} So I would again improve upon my old translation of 1886.

Is havanhāi an adjective, agreeing with urune: ‘for the beatified soul, for life-giving progress to the settlements’ ? C., the Pers. MSS., again translates nēk sāhibī the ‘good mastership’; here ‘the good conscience,’ hū + ahu (anhū), would be too advanced an idea. Too advanced meanings are the fatal errors of beginners. Perhaps, as said, the naišā of the MSS. A at section 7 came from the hava of havanhai, mistaken for ‘hava’ = ‘hva’ = ‘svá’ = ‘own’;, its original should be hu = ‘well,’ not hava = svá.

^{page 585 note 6} Free, or erroneous, for ‘the most holy ones’; see the original.

^{page 586 note 1} Or ‘all happy,’ ‘delectable.’

^{page 586 note 2} In this sense, better than ‘manly.’ I formerly rendered ‘legitimate,’ but I now think a ‘not-adopted offspring’ more probable.

^{page 586 note 3} Although the Ahurãni was the heavenly source of all waters, rain, dews, snow, rivers, gulfs, etc., yet it seems notable that an Orontes (Alvand)* or the Oxus was often had in view, with the Caspian, into which the latter flowed, referred to in the immediate connection. Is the ‘p’ in ‘Caspian’ really that of ‘ap,’ ‘kaš ap-,’ the vouru having fallen away? Or was Crumiya this mythic sea, or the Aral ? Views as to the identity of the particular body of water held in riew in these mythical allusions would change with expositors from generation to generation accordingly as their geographical information would become developed or restricted.

^{page 586 note 4} So the Pers. translates; and so A and B mark gīrān. Waters which flow from obvious gravitation, whereas wraters whose sources wrere springs weresupposed to ascend without obvious physical cause. Persian dērān-, ‘long’ (flowing), would not be so probable.

^{page 586 note 5} Katasīg; Pers. kārēz (so).

^{page 586 note 6} So the Persian ‘barf.’

^{page 586 note 7} So the Persian ‘bārān.’

^{page 586 note *} Mt. Alvand must have been named from some river.

^{page 587 note 1} B. om. dātīhātūm; and has for it ‘bayen tan’ i martūmān,‘ adding over ‘man’ ‘yehabūndāk.’ Therefore B. would reader ‘those waters within the body of man which are of a contributive nature’ (‘sources of strength’). Is not datihatum, however, a kind of apology for the fervour of this Yašt addressed to the Heavenly Source of waters ? The assurance is given that it is in accordance with Asa Vahista, who especially rules all ritual as well as all higher law, aša, rita, rite.

^{page 587 note 2} The passage in the original is evidently an interpolation.

^{page 587 note 3} So for the various objects in sacrifice.

^{page 587 note 4} Notice the idea that the sacrifice is the moving force in nature.

^{page 587 note 5} Āzi, demon of ‘evil desire.’ One is much tempted to regard this word as a degeneration from aži, the Aži Dahāka, ‘the scorching anguisher’; the áhī of the Veda who coiled his folds about the cloud cows, keeping off the rain. The holy sacrifice withstands and removes him or her; and the rainfall comes on again. But we had better for the present hold to the more clumsy derivation.

^{page 587 note 6} Whether this ‘Mūš’ refers to a plague of mice or rats, or to a kleptomaniac epidemic (muš, ‘to steal’), or not, this entire 24 should be referred only inclusively to the Mūš by this gloss. I am still pleased with my old suggestion, ‘that cheat, the P.’

^{page 587 note 7} Avāz astešnīh rāī (raδ(?)), for the dative paitištātayaēča.

^{page 587 note 8} nasēnešnih rāī(rāδ(?)), for paitisČaptayaēČa.

^{page 587 note 9} So, tarpēšnīh (so B.) rāī (or read ‘tarvēšnīh rāī,’ or ‘tarvēnešnīh rāī’), for paititaretayaēČa.

^{page 588 note 1} See note 9 on p. 587.

^{page 588 note 2} So B. and C.

^{page 588 note 3} Patīrak yehemtūnešnīh-, lor paityaoget; no rāī here; notice rāī for other datives.

^{page 588 note 4} As said above, this 24 should not refer exclusively to the Mūš; the Ašemaoγa is more directly involved, but practically both are present.

^{page 588 note 5} The point of everything here is the zaoθra; the offering effects all this.

^{page 588 note 6} So most critically and for the first time in exegesis, giving us the rendering of āhīša.

^{page 588 note 7} In error, of course, from the original 2nd person; unless, indeed, the form could be 3rd sg., with some. Was not, however, an offerer intended?— ‘Let one, the faithful, come with an abundant offering’; a hint to the contributors.

^{page 588 note 8} Notice that was first of all explained to us here by the Pahl. trlr., as = kabed. It is, as I advanced in A. J.P. not long since, = ašā, the š having an inherent ‘a.’ It equals rtéca in a sense of emphasis; recall where the cows bellowed ‘ṛténa’ in the Ṝik, i.e. ‘right lustily.’ Aša is properly arša.

^{page 589 note 1} The MSS. A, B, C insert tōrāān, ‘of oxen,’ having some reference to the sacred herds; hut K° seems to omit it. I think it may have originated from the syllable ‘-tā-’ in vahištābyō, which -tā in an earlier Avesta-Pahlavi character might possibly have looked like the signs for tōrā, one stroke having been accidentally lengthened, nēvak’, generally = ‘good,’ seems to render sraēšta, which we should rather reproduce with ‘beautiful,’ ‘fitting.’

^{page 589 note 2} The purified zaoOra of the pious. The idea of close inspection involved an attempted test of purification, pairianharštābyō = nigērit’.

^{page 589 note 3} ‘Do ye give’ is from 37.

^{page 589 note 4} ‘Riches’ ? for rāī (so), in spite of rayē (-āō-)mand as applied to the Deity.

^{page 589 note 5} A career (a continuance) (ravešnīh (sic)) of health of body, as often, a form as from i, ae = ‘to go,’ was seen ia -ātem as read āītem, so in Pahlavi = āī.

^{page 589 note 6} The Pers, has fatah, as seeing zi = ji = ‘to conquer,’ in zīvarīh; should vazdvare be vizdvare ?

^{page 589 note 7} Not adoptive; now so preferring to (‘legitimate’ ); it must have been soon adopted as legitimate.

Was this expressive iteration an accidental growth ? Hardly; see the metre.

The last word ‘long’ is part of a compositum in the original.

^{page 589 note 8} Zaoθre, here first recognised as a dative of zaotar.

^{page 589 note 9} Notice the gen. pl. force in ahniakem, here first recognised.

^{page 590 note 1} So for aēθryanãmea.

^{page 590 note 2} Kār is determined to this sense by vāstryā.

^{page 590 note 3} ‘For the thing is not his’; that is, ‘it is not the possession of the impeller (of the raid).’ Hardly the more subtle ‘for the removal of him whose interests are alien, whose the thing is not.’ It looks as if this loīt mindavam aš had been occasioned by the ‘a’ privative of the following word. It is also quite possible that something in the form of aurvaθaēibyyacĉa suggested mindavam; arθa may have been recalled; lā mindavam may have been thus original Pahlavi text and not gloss; adōštār (Pers. bī-dūst) would then be additional Pahlavi text. It is better to take rānēnītār in an evil sense, and as added explanation for vōiγ(n), and to take ‘not is the matter His’ as further explanation possibly occasioned by the anticipation of the ‘a’ privative in aurvaθaēibyasĉa; perhaps the ‘instigator (of evil)’ would be more suitable. Has vôiγna anything to do with vi + γan, or ‘vij, vaēj’ ? Vōīγ(n), so, to reproduce, but ‘ōī’ may have AT. value as ‘ōī.’

^{page 590 note 4} We might possibly regard Spiegel's form as meant for āvīndešn. The Pers., however, reads vāzešn and does not translate, feeling the difficulty. Certain signs for ‘vāzešn’ might also partly mean ‘vaeδešn,’ and the sign for ‘z’ was suggested by the form in the Av. original, which often represents a Pahlavi z.

^{page 590 note 5} Notice the frequent occurrence of this Gathic expression, which also appears on the Behistān inscriptions.

^{page 591 note 1} One would suppose that āfrīnāmi meant originally a praj'er used at an āfrīn (āprī) ceremony, but vice versa might have been the case. The ceremony must have he^n here named from the word; see its idea.

^{page 591 note 2} B. ins. harvisp; but it has no equivalent in the original. May it not, however, suggest an emendation of the original, which has vīspāyāi vise below. The form of the ite, (so some MSS.), may have been affected by that of the termination of the previous word. The point of ‘each’ is to define ‘those.’ It was most proper that the particular villages whose representatives should be there present with their offerings should be first officially noticed.

^{page 591 note 3} Orig. ušta.

^{page 591 note 4} Orig. vañta.

^{page 591 note 5} So ‘av’ with error; it should be vocative; but there is no preceding ‘tava.’

^{page 591 note 6} Joyful tasting, good appetite.

^{page 591 note 7} Plural for singular.

^{page 591 note 8} Not necessarily voc.; see the original.

^{page 591 note 9} This seems especially addressed to the Congregation.

^{page 592 note 1} B. here gives us an interesting rubric: after the word ‘thousand’ water is to be put, or added, into the appropriate vessel; after the word ‘healing’ water is to be put into the Zaoθra, or added to it (saying again), ‘a thousand healings, ten thousand healings.’

^{page 592 note 2} Aside from the original, we might fancy a response of the Ahurāni here; ‘so may it happen as the Zaotar (has now) prayed us.’

^{page 592 note 3} See the place in this Journal.

^{page 592 note 4} = afzūnīg I rendered ‘religious zeal.’

^{page 592 note 5} Meant for the verbal form of the nominal yaoštayō, which I rendered ‘capabilities’; there may be a closely similar, but not identical, form, meaning ‘joint,’ to ‘yuj,’ so of the Dragon.

^{page 592 note 6} This looks as if he divided = ‘zaidyam.’ In this case he did not mean ‘even to us ’ as gloss.

^{page 592 note 7} Is (? rayasĉa) = ‘riches’ here, while undoubtedly ‘splendour’ elsewhere ?

^{page 593 note 1} ‘Vid,’ seen in it.

^{page 593 note 2} Allusions to the order of some intervening texts in the ancient Yasna; but this order need not have remained unchanged. Yasna LXII (Sp. LXI) might be the Fire Yašt, and LXV (Sp. LXIV), etc., would be a Water Yašt.

^{page 593 note 3} Not so in the original. (Or ‘to that one, me, who am ….’ ? )

^{page 593 note 4} One might be tempted to refer ‘his own soul,’ as often elsewhere, to some particular offerer then present, possibly to one more prominent in station. Then, again, we might regard it as an expression of individualism, meaning ‘not superficially,’ nor ‘perfunctorily,’ though we must always be on our guard against that ever threatening fault of beginners, viz., seeing too much, or too refined, a meaning in an ancient text. (‘Who of his own soul … ?; who, M., as his own soul …’ ?, etc.)

^{page 594 note 1} The dawning of the especial Miθra cult is here obvious, or else, a clear trace of it, as long since established.

^{page 594 note 2} So the Parsi-Pers. MSS.; and this alone makes full sense, ranj = ‘trouble,’ not rōz = ‘day’ (same signs).

We cannot, however, refuse to see that rōĉ = rōz = rūz = ‘day’ is an idea well comporting with that of the Sun? So deciphering, we should have to make out some such advanced and improbable rendering as this: ‘for him it is done who meets the day (sunrise) by doing duty and good works, and does not hold (back) in the day, for I also do not hold back in the day in this path, where I come and go.’ (‘He rushes on with his swift horses; so his worshipper should meet the rising day, and course on steadily in duty and good works.’) Notice the great difference naturally resulting from differing decipherments.

^{page 594 note 3} Singular for dual; notice the plural iu the original rather than the dual (in some MSS.).

^{page 595 note 1} Here reference seems decidedly to be made to the worshipper rather than to the object of worship.

The ‘eye looking favourably towards all creatures of Aūharmazd’ is not so naturally the ‘eye of Aūharmazd’; and this reference to the deserving beneficence of the offerer affects the other passages.

^{page 595 note 2} The instrumental case of the original is missed throughout; a ‘pavan’ is needed. Notice the abstract ideas everywhere.

^{page 595 note 3} We seem forced to supply ‘increase’ here from above, and in analogy with ‘make me more straightforward.’ Otherwise we should avoid all these interior suggestions as much as possible, rendering in the most superficial sense admissible: ‘and what also are the duty and good works done by me (through) Aša, for these do thou give me a benefit, (to me) in person.’

^{page 595 note 4} 65 is Y. XXXIII, 10, line c, which see, Gāθas, pp. 124, 494.

^{page 596 note 1} Reading -ēnī, 1st conj.; -ēnd and -ēnī have the same signs.

^{page 596 note 2} The wording here is founded upon a mere fragment taken out of its connection from Y. XXXVI, 15; which see in this Journal.

The 2nd sg. improp. conj. jasō apparently mistaken for a n.s.m. of the present participle jasō = jasas rendered by the 3rd sg. conj.

^{page 596 note 3} 67 is a fragment from Y. XLIII, 6, taken out of its connection, see Gāfθas, pp. 162, 512.