Brintou.] lUu [April 19, 



schek and others claim that they cannot be analyzed as of Ar^-an 

 extraction.* 



Such an opinion seems to me without foundation. We find 

 such place-names wherever the Celtic stock of central Europe 

 left its traces. For the termination in -ess, I need but instance 

 Yindouissa, Yogessus, Sigonessus, Bodiocassus, etc. Its sig- 

 nification is well known. It means " the seat " (sedes, sessio^ 

 positio) of the person or tribe, and in this sense was especiall}' 

 employed as a suffix in the Celtic dialects. f 



The termination -anda in the form -anta or -ante is a familiar 

 Celtic suffix of tribal names, as Bi'igantes, Trinobantes, etc. 

 From these were derived place-names, as Carantia, Brodentia,etc. 

 The later terminations in -anza or -enza, as in Braganza, Pia- 

 cenza, etc., were corruptions of this, as was also the German 

 termination in words like " Pegnitz," etc.| 



Many other proper names of places and persons from south- 

 ern Asia Minor have lately been analyzed b}' M. Georges Radet, 

 and his researches appear to place beyond doubt these two 

 theses — 1. That the original Anatolians constituted an ethnic 

 unit; 2. That they spoke a tongue of Ar^'an affiliation. § 



Many of these names have a Celtic physiognomy. Thus the 

 Hittite roj'al names, Thargathazas, Tarthisebu, etc., simulate 

 the Gallic Thartontis, Turones, etc., in which the prefix tar {thai% 

 tar or dor) means " above, across," and by metaphor, superior, 

 leading, etc.|| 



A more striking coincidence is offered hy some religious 

 terms. 



It is generally conceded that the Ephesian Diana was origi- 

 nally a " Hittite " deity, and that her name Artemis is an Ana- 

 tolian word. It is also known that the original form under 

 which she was worshiped was that of a black conical stone, 

 thought to have been a meteorite. Now in Celtic arlan means 

 " a stone," and it often forms a part of proper names, as Artgal, 



* Paiili has been industrious in collecting such place-names. A long list will be found 

 in his IiiHnhrift von Lenivos, above quoted. 



fTliis is the explanation of Zeuss, Grammatka Ccltica, pp. 61 and 747-749. I am sur- 

 prised that it has been overlooked. 



I See Zeuss, Grammatica C'dtica, pp. 759, 7G0. It has been suggested that this termina- 

 tion is the Old Indian inda, sindim, river, whence Indus, etc., applied to tribes, towns, 

 etc., on a river. 



I See the Rtvac ArcMologiquc, Tome -xxii (1893), p. 209, sqq. 



II The Celtic tar esai (see above; is translated " super locum, in loco." Zeuss, «. s., p. 613. 



