¹ Benveniste, E., Titres et noms prupres en iranien ancien, Paris, 1966Google Scholar, referred to in what follows as ‘apud B.’, or simply as ‘B.’.

² ‘Amber at Persepolis’, in Studia classica et orientalia Antonino Pagliaro oblata, which volume had not yet appeared at the time of writing (May 1969). This article is here referred to as ‘Amber’. Names discussed in it, if mentioned in the present article, are followed by the signature ‘(A)’.

³ In the list which now follows, the alphabetic order is, as in ‘Amber’, (h)a, (h)i, k, l, m, n, p/b, r, s, š, t/d, (h)u, y, z; non-initially the order of consonants (double ones being counted as single, and h and w being disregarded) takes precedence over that of vowels; the OP emphatic sibilant (〈 θr) is represented by . I reproduce from Hallock question marks relating to reading, underlining where he did so the letter(s) to which each query refers. As explained in ‘Amber’, experience has taught me to consider that Hallock's queried readings are as deserving of attempts at identification as his certain ones. From him also comes the occasional ‘(1)’, which in his long typescript of 1963, but not in the later lists, stands after names which occur only once. Where Hallock gave variants, these too are reproduced. The reader is referred to the introduction of ‘Amber’ for a discussion of the semantic and circumstantial connotations of the names (often nicknames, or retrenched from compounds), the formation of patronymics, El. spelling conventions, and dialect features other than Persian. All names are of men (or boys) unless otherwise stated.

⁴ That viθa is a misreading was kindly imparted to me by M. Benveniste already in 1964. Unwilling, until he had published his discovery, to deny that reading, and unable to uphold it, I had no option but to refrain from mentioning it altogether.

⁵ Mayrhofer's complaint of ‘Unschärfe’ (n.69) in my distinction between dialect variation and optional realizations is not borne out by the precision with which he quotes the latter from the only available source, my article. Nor indeed would it have made sense to introduce a new notion, if I were not going to distinguish it clearly from the old one. Naturally I ‘sprachmoist von Dialektvarianten’, since I was saying that certain features so regarded were not that, though some were. But is it the number of times a word is used which matters, or the use made of the word?

⁶ I cannot account for my failure to cite even in TPS, 1964, P. 2, n. 2, what I should have prominently displayed in the earlier article there quoted, namely the genuine NBš, form of the word for ‘before, in the presence of’. It is *pῑh, as NP čῑ is from *čih (〈 OP *čiθi 〈 OP *čisi 〈*čisčit [as one would expect in A.V.], 〈 čitčit), see art. cit., II, at the end of n. 2. We thus have all three of the conceivable OP forms: *pasyā (Bal. pēs-, SBš, pes-), its alternative realizaton *paθyā (NBš. pī), and their inner-Persian dialect variant *pašyā (NP pēš). The bounty of attestation is as in the case of *syāva- ‘black’ (NP siyāh), *θyāva- (EI. Tiyama, see ‘Amber’), and *syāva- (Bal. ša Arm. šau).

⁷ Mayrhofer admittedly does not think it common-sense to regard the OP word for ‘king’ as Persian, and finds it natural that the Akkadian translator of Darius's ipsissima verba should have rendered the Great King' Persian pronunciation of another Persian king's name, by its Median equivalent (p. 18). No disproof of the second view can be offered. On the first all I can do is to refer to p. 28, n. 1, of my article, and to hope that the interpretation offered below of Hisatiya may prove one day not to have been far off the mark.

⁸ Were the Iranian name *vi-ramp/fa-, one would expect an El. spelling -ranp-, cf. Tanbara below.

⁹ BSOAS, xxx, 1, 1967, p. 19, n. 8Google Scholar. From MacKenzie's quoting in the same breath Pahl. 'slswmwk, whose ancestor never had a theta, an innocent reader of his note might suppose me to have said that no Pahl. s is pseudo-historical. MacKenzie also doubts, again with Mayrhofer's approval, the one case of MP initial h 〈 θw which I had been able to think of, on the ground that ‘there is no other example of initial θw 〉 h’. There is no example of initial θw becoming anything else either, while θw 〉 h is the rule internally; what other outcome does MacKenzie expect initially ? Is not one example better guidance than none, or than inference on sound-changes drawn from irrigation ?

To the available instances of internal MP h 〈 θw a new one can now be added: the verb nhrys-, discussed by Henning, in NGWG, Fachgruppe III, 1932, 2, p. 218, n. 5Google Scholar, suits its context best if one assigns to it the meaning of, as part of etymologically identifying it with, Av. niθwǝrǝsa- ‘to carve out’. °hris- was probably a metathesis of *°hirs-, as MP šr's- was of š'rs-. We note in passing that not only MP šr' syn-, but also MP sr'xšyn- meant ‘to put to shame’, cf. šrmjd kyrd in the Parth. passage parallel to Mir. Man., II, p. 302, 1. 6 which Henning adduced on the same page, n. 3. srāxš- (with ā analogical to šr's-) was from *srafš- (cf. Pers. daraxšīdan: Parth. drfš-) 〈 *šrafs- 〈 *šfars- (as in Parth. and Sogd.) 〈*fšarsa-. sr' xšyn- also occurs in 1. 20 of the hymn quoted by Henning in TPS, 1942, 56.

¹⁰ Semantically one may prefer a haplology of *apa-pauša- ‘diverter of thriving’.

¹¹ NP saxt does not belong here if the etymology proposed for it by Morgenstierne is correct, see IIFL, II, 249, and cf. Henning, , ‘Mitteliranisch’, p. 109, n. 2Google Scholar.

¹² Šarkuniya alone would not confirm the postulate of an u-stem, as this would also be a correct El. rendering of *šarg -vanya-, cf. ‘Amber’, s.v. Bakunda.

¹³ Note that El. zarnupasaš in PTT, 178, should not represent *zarani-piθa- (as Cameron has it on p. 42), but *zarnu-paθa- (to Av. pas-). The men so described would be responsible inter alia for the zaranyapaxšta.pad- appearance of royal seats.