^{page 747 note 2} Cf. Zend-Avesta, Tome i, ii, p. 60 seq.Google Scholar

^{page 1 note 3} Cf. Zeitschrift für vergleichende Sprachforschung, xlii, 1, n. 2.Google Scholar

^{page 749 note 1} There is, however, a strange coincidence between the words of Herodotus (i, 131), τόν κύκλον пάντα Δία καλέοντες KaAeovres, and the expression in Yasna xxx, 5, that Ahura Mazda “dresses himself in the firmest heavens”. But this is a common Iranian idea.

^{page 749 note 2} Religionsgeschichtlicke Versuche und Vorarbeiten, xvii, 1 (Giessen, 1920).Google Scholar

^{page 749 note 3} Cf. Clemen, , loc. cit., p. 23.Google Scholar

^{page 749 note 4} Athen, . Deipnosophistæ, xii, 515 D.Google Scholar

^{page 750 note 1} Professor Hertel himself points to the coincidence between Herodotus, i, 140: οί δέ δή Μάγοι пάντα пλήν κυνός καί κτείνουσι καί μέγα κτείνοντες μοίως μρμηκάς τε καί Φις καί τλλα ρПετά καί пετεινά and certain regulations in the Vendīdād (cf. Die Zeit Zoroasters, p. 51); but not even this has made him hesitate. Those same regulations apparently prevailed amongst the Iranian Kambojas according to Jātaka vi, 208Google Scholar, 27 seq. (cf. E. Kühn, , Avesla Studies in Honour of Peshotanji, p. 214Google Scholar; Nariman, , JRAS., 1912, p. 255 seq.Google Scholar; Charpentier, , Zeitschr. für Indologie, ii, p. 145).Google Scholar

^{page 750 note 2} Cf. Zeitschr. für Indologie, ii, p. 140 seq.Google Scholar

^{page 751 note 1} Ackæmeniclen und Kayaniden, p. 80 seq.Google Scholar

^{page 752 note 1} Cf. Die Zeit Zoroasters, p. 44 seq.

^{page 752 note 2} Cf. Proc. Soc. Bibl. Arch., 1899, pp. 127, 132, 137 seq.Google Scholar

^{page 752 note 3} In this connexion a word of protest should be laid against the conclusions which Dr. H. Sköld has tried, in JRAS., 1924, p. 265 seq. (recently repeated in Finnisch-Ugrische Forsch., xvi, 277 seq.), to found upon this Assara Mazāš. Of course, we know nothing about the time when the name became familiar to the Assyrians. What we really know is this: (1) In the famous text from Boghaz-köi, dating from the early fourteenth century b.c., we have a form Na-ša-at-ti-ia, i.e. Nāsalya with preserved -s-; this form can neither be “Indian” (I leave it to Dr. Hüsing and others to believe that the rulers of the Mitani were Hindus) nor “Indo-iranian”. It belongs, of course, to the Iranian invaders from Media, and shows that at this time -s- was still preserved in Iranian dialects. (2) The Gāthās, the oldest literary document in an Iranian language, have throughout altered -s->-h-; they may well date from about 1000 b.c., though they were, of course, written down much later. This is all so far.

^{page 752 note 4} Cf. Die Zeit Zoroasters, p. 36 seq.

^{page 753 note 1} Cf. Zeitschr.f. vergl. Sprachforschung, 42 (1909), p. 1 seq.Google Scholar

^{page 754 note 1} I learn from Achæmeniden und Kayaniden, p. 10, that the Professor thinks the greater part of the Rigveda to have been composed outside India proper in post-Zoroastrian times—i.e. according to the same authority, after 500 b.c. These ideas are so bold that no one can take them into account without some sort of evidence.

^{page 754 note 2} We know quite well that an Aryan emigration went westwards to Media and Mesopotamia from Bactria and Sogdiana soon after 2000 b.c., and the invasion of India may well belong to the same period.