Assyrian Personal Names, XXV 
Iranian y between two vowels appears sometimes as zz or z (= v); XSayarsa — Ahfi- 
maru, AhSumarsi, AhSuarsu, etc.; cf. Bi vtr. 
The Iranian diphtongs are in Semitic cuneiform writing weakened: aë, ai to & or 7, 
e. g. Arta-raeva* = Arta-ri(-e)-mu BE IX; ai$a, ae$a = 28u in Dagi-, Misda-e-fu BE IX, 
XSapra-a&Sa* — Ass. Satar-e-$u; pairi = pri in Ass. Pi-ri-Sati; au, ao to z, cf. Gaubruva = 
Gübarü, raocana-däta* = Rusundatu, and so forth. 
In order to facilitate the pronunciation of two consonants, especially in the beginning 
of a word, but not infrequently also in the interior of it, an additional vowel is used: a) before 
the first consonant, e. g. Ahsizarsu — X$ayar$a, /s-pi-i-ta-am-mu UMBS U 1 = Spitama, /p-ra- 
a-du-pirna BE IX = Dparapepvng; cf. Bi. bYBYItnN from xSaprapävan; b) between the two 
consonants; e. g. Z/aXi| HiXi-zartu, Haxatrittu — XSaprita; Ba. $atar, Ass. satar = xSapra; Sitir 
(beside Kira) = éipra; par(a) = fra in Pa-ra-da- = Fräda, Partama = fratama, Parumartis 
= Fravartis, etc.; katir (beside £atri) = x'apra; firrina = farna, in Pi-ir-ri-na-sätu, Baga- 
bukisu = Baga-buy$a, and so forth; c) before and after the first consonant; e. g. Akkasiiar su 
= XSayarSa. 
The Iranian sound group pr is in Babylonian and Assyrian represented by 7r, e. s. Mipra— 
7 Mi-it-ri, XSaprita = Hasatrittu, and so forth, while the corresponding Old Persian single character 
p' appears in Babylonian as Zr, is, ss and $, e. g. &ipra — dr, Xira in Sit(i)rantalma — 
Cip antayma, Artaksatsu, Artahsassu etc.— Artax$apr'a, Aina = Ap'ina. The change of this 
double sound into a sibilant took place in a comparatively early time, as may be concluded from 
the examples just mentioned and such Greek forms of names as Apraéecons and Tiocoapepvic. 
Hence I have presumed that this change likewise occurs in the name Umissu < Va^umisa < 
Valumip'a; but I quite admit that this explanation is doubtful, since the etymology of the name 
in question, according to Iranian scholars, is not clear. As for the later change into Zr, which is 
frequent in Middle and New Persian, it is difficult to believe that it could have led to the names 
Baga-miri and Artahsar from the time of Artaxerxes I (B. C. 462—424) as HÜSING supposes !. 
Baga-miri is, in all probability, identical with Bagavira?. "On the other hand, Ar-ta-ah-Sa-ar 
BE IX or Ar-tah-Sa-ri TNB z. e. Apra£apnc, to my mind, corresponds to Ass. Ay-fa-sa-ri, which 
name is known from 830 B. C. Now if Artahsar — Artasari were identical with ArtaySahr we 
should be obliged to assume that the sound change pr > hr occurred as early as the 9'^ century, 
which is improbable. Hence we may accept the explanation given by JUSTI, SCHULZE? and 
others, according to which (Artahsar — Artasari —) Aprafapns is a "Kosename" formed with 
-ara from the abbreviated form Artahs*. 
Finally it will be noted that Iranian vowel stems in a in the Assyrian-Babylonian ren- 
dering frequently appear with the ending a, which perhaps corresponds to the Avestan nomi- 
native ending à and Old Persian at; cf. Ardar& (aredra) Arsaka, Arzaramna, Atarsitra, Data, 
Dalta(darsta), Parada (Frada), Mitirriadada (UMBS, II, 1), Umimana (Vivana), Umidarna (Vidarna), 
Umittana (TNB; Utana), Uppamma (Av. upama), Sura, etc.; cf. also Mispara (= Vayaspara). 
The investigation of Elamite names, which especially the excavations at Susa provided 
with a rich material, is still in its infancy. In treating of the not very numerous Elamite names 
that occur in this work, I have availed myself of JENSEN's! and HÜSING's? important researches 
and suggestions. 
I) Cf. KZ 36, p. 562. 2) Cf. NOLDEKE, BE IX sub voce; SCHEFTELOWITZ, ZDMG, 57, p. 166. 
3) Justı, Zranisches Namenbuch, p. 36; SCHULZE, KZ, 33, p. 220 ff. 
4) P. JENSEN, Zlamitische Eigennamen, in NZKM, VI (1891), pp. 47—70, 209—226. 
5) G. HÜsING, Reduplication and Iteration in Elamischen Eigennamen, in OLZ, 1900, col. 83 f.; Die Zlamische 
Sprachforschung, in Memnon, 1910, pp. 5—40. 
No; 1. D 
