PDF VERSION

It has survived long enough for the copyright to expire and the book to enter the public domain. A public domain book is one that was never subject  to copyright or whose legal copyright term has expired. Whether a book is in the public domain may vary country to country. Public domain books  are our gateways to the past, representing a wealth of history, culture and knowledge that’s often difficult to discover. Marks, notations and other marginalia present in the original volume will appear in this file - a reminder of this book’s long journey from the  publisher to a library and finally to you. Usage guidelines Google is proud to partner with libraries to digitize public domain materials and make them widely accessible. Public domain books belong to the  public and we are merely their custodians. Nevertheless, this work is expensive, so in order to keep providing this resource, we have taken steps to  prevent abuse by commercial parties, including placing technical restrictions on automated querying.

We also ask that you: + Make non-commercial use of the files We designed Google Book Search for use by individual  personal, non-commercial purposes.  and we request that you use these files for + Refrain from automated querying Do not send automated queries of any sort to Google’s system: If you are conducting research on machine  translation, optical character recognition or other areas where access to a large amount of text is helpful, please contact us. We encourage the  use of public domain materials for these purposes and may be able to help. + Maintain attribution The Google “watermark” you see on each file is essential for informing people about this project and helping them find  additional materials through Google Book Search. Please do not remove it. + Keep it legal Whatever your use, remember that you are responsible for ensuring that what you are doing is legal. Do not assume that just  because we believe a book is in the public domain for users in the United States, that the work is also in the public domain for users in other  countries. Whether a book is still in copyright varies from country to country, and we can’t offer guidance on whether any specific use of  any specific book is allowed. Please do not assume that a book’s appearance in Google Book Search means it can be used in any manner  anywhere in the world. Copyright infringement liability can be quite severe. About Google Book Search Google’s mission is to organize the world’s information and to make it universally accessible and useful. Google Book Search helps readers  discover the world’s books while helping authors and publishers reach new audiences. You can search through the full text of this book on the web  ai[http: //books . google. com/| Ms i i Bed ® ( 2") ' wr  de ‘i ri. @ ; a  oe al * ‘ A '  r  aad ae ¢ 7 ae oa 7 “| UNIVERSITY OF MICHIGAN HENRY VIGNAUD LIBRARY THE ARMENIAN ORIGIN OF THE ETRUSCANS. “BY ROBERT ELLIS, B.D., | PELLOW OF 87. JOHN'S COLLEGE, CAMBRIDGE; AND AUTHOR OF ‘A TREATISE ON HANNIBAL’S PASSAGE OF THE ALPS." ~ “‘LAW@UAGES ARB THE PEDIGRER OF NATIONS,’”’—JOHNSON. | LONDON: PARKER, SON, AND BOURN, WEST STRAND. M DOCCLXI. >  x ’ . ¢ ~ _7 < *  eel + le Ms 4  e «© * * we  e LONDON : 28, PRINTERS, CHANDOS STREET, TENT GARDEN.

Vtgye ok  a CONTENTS. THE ARYAN RAOE, AND ITS DIVISIONS . . . «© «6 « « « THE ETRUSCANS A BRANCH OF THE THRACIAN RACE—<HIS- TORICAL EVIDENCE OF THE EXTENSION OF THE THRACIAN RACE FROM ARMENIA TO ETRURIA . . . -«© «© «6 « « LINGUISTIC EVIDENCE OF THE EXTENSION OF THE THRAOIAN RACE FROM ARMENIA TO ETRURIA. ©. © 1. «© 6 « CAPPADOOIAN WORDS... . . «6 « «© «© © © @ se PHRYGIAN WORDS... . .- 1. s+ 6 «© © © © eo PHRYGIAN INSCRIPTIONS . . 1. - 6 e+ + ee ew we LYDIAN WORDS. . . . «+ © 6 © © «© © © © ww CARIAN WORDS. . . . . 1 «© © © «© © © © @ oo THE LYCIANS AND THE CAUCASIAN NATIONS . . . . .'. LYCIAN INSORIPTIONS. . . + + © © © «© © «© © «@ LYCIAN WORDS. . . . 6 ee ee ew ee THRACIAN WORDS .. . . 2. - ee + 6 @ ee es SCYTHIAN WORDS . . . . . + 6 2 © es te ee ee DACIAN NAMES OF PLANTS . . . . . =. ee ee ee ALBANIAN WORDS ALLIED TO THE ARMENIAN. .... . RHAZTO-ROMANSCH WORDS ALLIED TO THE ARMENIAN .. . ETRUSOAN WORDS . . . . «1 ee ew ew ew ee ETRUSCAN INSORIPTIONS . . . . 1 1 ee we ee SEPULCHRAL INSCRIPTIONS . .-. . 1 1 ee ee VOTIVE INSCRIPTIONS . . . «1. « © © «© «© © © © VOCABULARY OF VOTIVE WORDS .......... 4271382 101 103 118 130 1V CONTENTS. PAGE THE INSCRIPTION OF OERVETRI ... .. «6 « « « « 139 THE PERUGIAN INSCRIPTION. . . . 2. © «© «© « « « « LBL GEOGRAPHICAL NAMES IN ETRURIA AND OTHER COUNTRIES . 161 THE PELASGIANS .... + «ee © © e« « «© ee « « LIS POSSIBLE EXTENSION OF THE THRACIANS TO THE WEST OF ETRURIA e e e e e e e e ® e e e e e e e 179 NAMES OF GLACIERS . . . . . «© «© « © © «© « « « 180 THE BEBRYCES OF THE EASTERN PYRENEES .... . » 182 CONCLUSION . . . « 6 © «© © © © «© «© © «© « « « 183 APPENDIX . 2. «© «© © © © © © © «© «© « «© « « « 189 INDEX OF ETRUSCAN WORDS ......-. «ee « .« 198 Tae Armenian letters are represented by the following  equivalents :—  l. a. 21. y (partakes of the sounds 2. 6, off and y: asa final 3. 9. commonly mute). 4. d. 22. 2. 5. ¢ (Eng. e or ye). | 28. § (Eng. sf). 6. 2. 24. 0. 7. € 25. é (Eng. ch). 8. é (neutral vowel: Germ. 6, | 26. p. Sansk. a). 27. § (between Eng. and sf). 9. th. : 28. r (strong 7). 10. 2 (Fr. y, Eng. ¢ in mea- | 29. ».  sure). 30. w (as v when beginning a 1], 2. syllable). 12. 2. 81. ¢. 18. &h. 82. 7. 14. z (Eng. ds). 33. # (Eng. Zz). 15. &. 84. v (wu, v: never a vowel 16. &. when alone). 17. 2 (Eng. ¢s). 35. ph. 18. 2? (Welsh 22, Polish thick?). | 36. cd. 19. g§ (Eng. ,y). 37. d (broad o, like Fr. az). 20. m. 38. 7. | The effect of the (.) is to strengthen, of the (’) to soften,  the consonant which it qualifies. The letters, 37 and 38, are  of late introduction : 37 was formerly written av: 88 is only  used in some foreign words. The following are regularly  diphthongs, when followed by a consonant:—  ev, Eng. w or you. ov, Eng. 00, Germ. w.  iv, Germ. ti, Fr. #, occa- | ow, long 0, Germ. oo, Gr. w.

sionally Eng. w. Some make 2, 8, 4, tenues, and 26, 15, 31, medials.

vi The Albanian is written in Latin characters, and the  following equivalents are adopted for the alphabet of Dr. Hahn :—  l. 18. ly. 2. v 19. m. 3. 20. n. 4. sy Ql. my. 5. § Qla. &. 6. g. 22. @. 1. JY: 23. 0. 8. dh. 24. 9. 9. ad. 25. 47. 10. e. 26. 8. Ill. é. 27. 4. 12. 2. 28. §. 13. th. 29. ¢. 14. 4. 80. 4. 15. &. 81. ph. 16. hy. 82. ch. 17. 2. 33. kh. Y is always a consonant, even at the end of words: thus  kaly, ‘ a horse,’ is pronounced nearly like Fr. casdle, as ly = Ital.  gl. Ny=Ital. or Fr. gn. The sound of @ is as in the Fr. on.

The following abbreviations have been employed— Abas. = Abasian: Alb. = Albanian: Ang.-Sax. = Anglo- Saxon: Arab.= Arabic: Arm.= Armenian: Bret. = Breton : Cappad. = Cappadocian : Circas. = Circassian : Eng. = English : Esth. = Esthonian: Etrusc. = Etruscan: Fr.= French: Gael. = (Scotch) Gaelic: Georg. = Georgian: Germ.= German: Goth. = (Meso-) Gothic: Gr. = Greek: Heb. = Hebrew: Hung. = Hungarian : Ital. = Italian: Kurd. = Kurdish: Lapp. = Lapponic: Lat.= Latin: Lith. = Lithuanian: Lyd. = Ly-  dian: O. Pers. = Old Persian: Osset. = Ossetic: Pers. = Persian: Phryg.= Phrygian: Pied.= Piedmontese: Pol. = Polish: Rhawet-Rom. = Rhxto-Romansch: Sansk. = Sanskrit: Scyth. = Scythian: Serv. = Servian: Span. = Spanish: Swed. = Swedish: Thrac. = (Proper) Thracian: Turk. = Turkish. ERRATA ET ADDENDA.

p. 15, 1. 4 from bottom, for okro read ochro.  p. 17, 1. 18, for Atkka read kikkas.  p. 80, 1. 28, for etak read etag.  p- 36, |. 5, for notcay read povcav.  p. 74, 1. 7, for ovsparpoup read oveparpoup.  p- 75, 1. 4, for poéBouu read pobBoup. Kindred terms to the Arm. etag, ‘ fossa,’ and perhaps to the Phryg. lachit (p. 80) and the Etrusc. z-lach-nke (p. 115),  would be the Ital. /acca, ‘descensus, cavum, fossa,’ and the Gael. dag, ‘cavum, specus.’ In p. 187, 1. 9 from bottom, after gizh, ‘imber,’ add cé, gen.  céithe, * flos lactis.’ THE ARMENTAN ORIGIN OF THE ETRUSCANS. [TH subject of the following pages is an extension of the  argument in the latter part of my Contributions to the Ethnography of Italy and Greece. Upon the earlier part I  need not enter here: it will merely be necessary to recapitulate  the views I entertain upon the peopling of Europe.

The Indo-Germanic, or, to adopt the shorter and now well-  established term, the Aryan race, may be divided into two  great divisions: the Northern or European Aryans, and the Southern or Asiatic Aryans. The European Aryans are sub-  divided, reckoning by language, and overlooking smaller  members, into three great branches: the Latin, the German,  and the Slavonian branch. But the Latin is an intrusive  language, derived from a single city or a very small district,  and having spread over countries mainly occupied by Celts,  who are also Aryans. The three great branches of European Aryans would therefore become, when considered ethnically,  the Celts, the Germans, and the Slavonians.

The Southern or Asiatic Aryans may be similarly divided  into three principal branches: the Armenians, the Persians,  and the Indians. The Kurds and Afghans are of less ethnical  importance. The Armenians, like the Celts, are now few in  number. It will be my endeavour to prove that the race to  which they belong once occupied a much greater extent of  country, and were spread westward from Armenia to Italy  under the names of Phrygians, Thracians, Pelasgians, Etrus-~ , B 2 THE ARMENIAN ORIGIN OF THE ETRUSCANS.

cans, and other designations. As the expansion of the Latin  language from its original seat at Rome obliterated in its ad-  vance the greater number of the Celtic dialects, so the expan-  sion of the same language in part, but yet more the expansion  of the Greek, obliterated in Europe and Asia Minor the dialects  akin to the Armenian, until it was only in the original seat of  the race, in Armenia itself, that a representative of those  dialects survived.

The only members of the Aryan stock with which I shall  have to deal, will be the Thracians, i.¢., the race of which the Armenians are a part, and the Celts: or with the western  branch of the Northern Aryans, and the western branch of the Southern Aryans. These two branches would, by their posi-  tion, be the first settlers in Europe of the divisions to which  they belong. The Celts, I conceive, entered Europe chiefly  through Scythia or Russia; and the Thracians, at least mainly,  through Asia Minor and Turkey. Having entered Europe,  one portion of the Celts would turn to the S. W., and  eventually settle in Italy. To these would belong the Um-  brians, said by some to be a branch of the o/d Celts, z.¢., the Celts who were in Italy before the Celtic invasion which ulti-  mately terminated in the capture of Rome. The Thracians,  on the other hand, advancing W. and N. W. from Asia Minor,  would form the original Aryan population in Turkey and Greece, would come in contact with the Celts along the line  of Dacia, Pannonia, Noricum, and Rhetia, and would pass into Italy, mostly under the name of Tyrrhenians, at a period  subsequent to the date of the Celtic settlements in that  peninsula. | But there would have been two races in Europe before the  arrival of any Aryan settlers. One of these races would have  been the Basque or Iberian race, with which I shall have little  to do. ‘The other is more important. If there were a race in Europe, not, Basque, and earlier than the Celtic, it is almost a  direct ethnological consequence that it must have been Fin, THE ARMENIAN ORIGIN OF THE ETRUSCANS. 3 Of this race I consider the Ligurians to have been the remnant  in Italy, and that the Fins, or some kindred Turanians, formed  the substratum over all Italy, Greece, Turkey, the Austrian States, and Asia Minor. Thus the representatives of the three Italian races and languages, the Ligurian, the Umbrian, and  the Tyrrhenian or Etruscan, would now, I believe, be found  respectively—in Lapland, Finland, and Esthonia; in Brittany, Wales, Ireland, and Scotland; andin Armenia. To substan-  tiate the last of these points is my present object. The repre-  sentatives of the fourth of the earliest European races, the Iberian, would in like manner be found in Biscay. > — Historical evidence of the extension of the Thracian race from Armenia to Etruria. 4 In endeavouring to determine the family of nations to which  the Etruscans belonged, it is necessary first to inquire from  what nation they are traditionally derived,.or with whom they  have been identified. With regard to the former of these  points there is great unanimity. The general voice of anti-  quity derived them from the Lydians, an opinion which the Etruscans themselves were also willing to accept. The Etruscans were also generally identified with the widely  spread race called Pelasgian, although the Tyrrhenians and Pelasgians in Etruria, and also in Campania, are usually  spoken of as distinct peoples, but continually associated toge-  ther. The Etruscans would thus appear to be akin to the Lydians and the Pelasgians.! Of this last race, which was 1 These affinities are disputed by one ancient writer, Dionysius of Hali-  carnassus (lib. i, cc. 29, 30). He says: ‘In my opinion, however, all are  in error, who believe the Etruscan (Tufpyxéy) and Pelasgian nations to  be the same.’ Then follows an argument in proof of this, which is  grounded on the well-known mistake of Crotoniates, i.e. Cortoniates, for Orestoniates in Herodotus. Dionysius then proceeds :{ For this reason,  therefore, I believe the Etruscans to be different from the Pelasgians. Nor do I think that the Etruscans were Lydian colonists. For they do  not use the same language as the Lydians: nor can-it be said that, B2 4 THE ARMENIAN ORIGIN OF THE ETRUSCANS,  almost entirely extinct in name in the time of Herodotus,  scarcely anything is known, and nothing need yet be said. We have now to enquire to what family the Lydians belonged :  for the story of the descent of the Etruscans from the Lydians  can hardly be regarded as rigidly historical, but merely as im-  plying that both nations belonged to the same family; an  ethnical fact which appears in the historical form of a migra-  tion from Lydia to Etruria. .

The Lydians are in the first instance connected with the Mysians and Carians. It is mentioned by Herodotus that  the Mysians and Lydians were xaofyvnror to the Carians, and  that the mythic Mysus, Lydus, and Car, were brothers.! He  also says that the Mysians were Lydian colonists.? In  reference to the more general affinities of the Lydians and Carians we have no particular statements, but are left to de-  duce them from those of the Mysians, who are described as Thracian colonists. We may therefore infer that the Lydians  and Carians belonged also to the Thracian family.

We have now probably arrived at the name of the great  family to which the Etruscans belonged, 7.¢., the Thracian. For the Thracians were not a single tribe or people. The  name, like that of Celt or German, describes one of the Aryan  families, which was, according to Herodotus, the most nume-  although they are not indeed like (the Lydians) in speech, yet they still  retain some indications of their mother-country. For they do not wor-  ship the same gods as the Lydians, nor do they resemble them in their  laws and customs; but in these things they differ yet more from the Lydians than from the Pelasgians. Their account, therefore, seems to  be more probable, who declare this (Etruscan) nation not to be foreign,  but indigenous (in Etruria) ; since it is very ancient, and is not found to  be like any other, either in speech or manners.’ There were, therefore,  three opinions concerning the Etruscans. They were considered as— 1. Pelasgians, arace which was a mere name in the time of Dionysius ;— 2. Lydians ;—3. An isolated race allied to no other (a proposition affirmed  of the Armenians half a century ago). The solution, I believe, i is, that  the Pelasgians, Lydians, and Etruscans derived their origin at a remote  period from Armenia.

11.171. 3 vii. 74. 3 Strabo, pp. 542, 566. ‘THE ARMENIAN ORIGIN OF THE ETRUSCANS. 5  rous of all races next to the Indian.! The Thracians may in-  deed be traced from the frontiers of Media to Italy and the Alps, and would have included the Armenians on the east, and  the Rhetians and Etruscans on the west. The following are  the chief indications of a chain of kindred nations within  these limits, which the ancients have transmitted to us.

The most easterly of all the tribes expressly said to be Thracian were the Sarapare, who are mentioned by Strabo as a Thracian tribe dwelling beyond Armenia, near the Medes and  the Guranians.? In the Armenian province of Persarmenia,  a district bordering on or containing the Lake of Ourmia, the Armenian Geography attributed to Moses of Chorene mentions  a part called Thraki.$ The Armenians themselves belonged to the same family as  the Phrygians, from whom they were said to be descended,  nd to whose language their own bore much resemblance. * *Appévioe 76 "Apuévoe «ww ee) Bpuywv arroxo :  yévog tx Ppvylag xal rg gwvy wodAa gPpvylZovaw.® ) But the Phrygians were well known as a Thracian tribe:* the Arme-  niang, therefore, probably belonged to the same family. Con-  tinuing westward from Armenia, we come to Cappadocia, a  country possessed by a Syrian race, who probably advanced  from the south at some remote period, and separated the Armenians from the kindred race of the Phrygians. The  eastern part of the province called Cappadocia was, however, Armenian, and formed the district of Armenia Parva. Cap-  padocia forms the only break in the chain of Thracian countries  between Media and Helvetia.

The Phrygians, as just noticed, were a Thracian race: of Ppbyec Opgxwy azomod cio. The Mysians, Lydians, and Carians belonged likewise to the same family. So, also,  according to Strabo, did the Mygdones, Bebryces, Medobithy- 1 iy. 3. 3 p. 531. 3 Mos. Chor. p. 359. Ed. Whiston. . 4 Herod. vii. 73. ° Eustath.on Dion. v. 694. © Strabo, pp. 295, 471. 6 THE ARMENIAN OBIGIN OF THE ETRUSCANS.

nians, Bithynians, Thynians, and Mariandynians, as well as  the Mosians on the Danube and the Getew.! He says, more-  over, that the Gete were éudyAwrroe with the Thracians,” who  thus extended to the frontiers of Macedonia, Illyria, and Pan-  nonia. But the Pierians in Macedonia were a tribe of Thra-  cians,® and the Macedonians also stated that Phrygians formerly  dwelt in their country under the name of Briges.* Strabo,  again, speaks of Brygi in the south of Llyria;° so that we  should find Phrygians in Macedonia and Illyria, as well as Asia, and thus trace the Phrygian name from Armenia to the Adriatic.® There were then, it appears, Thracians in Illyria. The Istrians, also, are described as Thracians.’?’ The Veneti, again,  are classed by Herodotus among the Ilyrians,® and were re-  ported by one writer to have come from Cappadocia, and to 1 p. 295. 3 p. 303. 3 Strabo, p. 410. 4 Herod. vii. 73. 5 yn. 326.

© For a full and excellent account of the Phrygians, see the art. Phrygia in Dr. Smith’s Geography. The writer has, however, fallen _ into one error, which was also made by Dr. Hahn in his Albanesische Studien, and through him by myself. The words, Aadw@rac ’Appévor elval  pot Soxodcr xal Ppiyes (Cramer. Anecd. Gv. Oxon. v. iv. p. 257), do not  refer to men, but to horses. It appeared that the Dalmatian breed of  horses was the same as that found in Phrygia and Armenia, and also in Cappadocia (ib.), the horses of the two latter countries being in high  repute among the ancients. : Three Llyrian chiefs bore the name of Bato. One was a Dardanian,  one a Dalmatian, and the third a Pannonian. This gives rise to a con-  jecture that Bato, like Brennus, signified ‘ chief.’ The Arm. vet, = Sansk.  pati, ‘dominus,’ = Zend paiti, = Pers. bad, has this signification. He-  rodotus mentions that the Scythians called themselves Scoloti, and we  finda Scythian king called Scolo-pitus. This looks as if pt¢ may have sig-  nified ‘ chief’ in Scythian. Another Scythian king, mentioned by Hero-  dotus, was Ariapithes, which might be explained from the Arm. ayr, ‘man,’ and pet, ‘chief,’ ‘ chief of men.’ Herodotus also mentions a king  of the Agathyrsi, called Spargapithes, and a leader (orpariyéorra) of the Massagetee, the son of Tomyris, called Spargapises. These names re-  semble the Arm. sparapet, ‘general, chief.’ Compare also, Arm. 26>, ‘army, zérapet, ‘ general,’ azg, ‘nation,’ azgapet, ‘prince,’ karapet, ‘leader,’ wardapet, ‘teacher.’ The remains of the Scythian language will  be subsequently noticed.

7 Scym. Ch. v. 390. 83. 196.

THE ARMENIAN ORIGIN OF. THE: ETRUSCANS. 7  have settled in company with Thracians on the Adriatic, ze, _ in Venetia.! Thracians were likewise to be found in Pannonia and Nori-  cum. The three Celtic, or mainly Celtic nations, the Bou, Taurisei, and Scordisci, eover the whole area of these two  countries. But ali these three tribes are said to have been  mixed with Thracians,? and the Seordisci, also, with Ilyrians.® When the Etruscan people possessed the plains of the Po,  before the Gallic invasion, their frontier would have extended  to Venetia and Rhetia. We have already traced the Thracians  into Venetia and Noricum, the countries bordering upon Rheetia and the older Etruria. The addition of the kindred  nations of the Rhetians and Etruscans, the one directly, and  the other indirectly, derived by tradition from the Lydians,  completes the list of Thracian nations stretching from the Caspian to the Tyrrhenian Sea.‘ Two other important ethnical names, Tuscans and Dardans, 2 Strabo, p. 552. 4 Ib. p. 296. 3 Tb. p. 313.

4 The Vindelicians are even by one writer, of indifferent authority,  ranked among the Thracians. ‘Kaum Erwahnens werth ist, was Scho-  liasten von der Abstamrhung der Vindelici traumen, Servius, der im Commentar zu Virgil (Aen. i. 244) aus seinem Dichter findet, sie seien Liburnen, und in ihnen wegen des Horaz (iv. 4) Abkémmlinge der Ama-  zonen sieht, und “Porphyrio zum Horaz, in anderer der vorigen gleich-  werthen Ansicht, sie seien von der Amazonen vertriebenen Thraker  gewesen. —Zeuss, Die Deutschen, p.231. These opinions are doubtless  of little value: and yet it is remarkable that they should be so nearly  confirmed by the evidence of language in their intimation of the original  seats of the nations in these parts. For they merely substitute Cappa-  docia, from which Thracians fad in all probability been expelled, when  the Armenians were divided from the Phrygians, in the place of Armenia. In the different accounts, there is also much consistency. One writer  reports the Veneti to have come from Cappadocia, and to have settled  in company with Thracians on the Adriatic; another derives the Vinde-  licians (considered by a third as Liburnians, while the Veneti are ranked  by a fourth among the Illyrians) from the Amazons, t.e., from Cappadocia;  while a fifth considers these same Vindelicians as Thracians expelled by Amazons, in all probability from some part of Asia Minor, if not from Cappadocia. The Lydian extraction of the Rhetians and Etruscans is  an-extension of the same belief to the origin of two other contiguous  nations. 8 THE ARMENIAN ORIGIN OF THE ETRUSCANS.

have nearly the same extent. Both are connected with Etruria,  the Dardans, however, only through the mythic Dardanus. Yet the name is Italian, as Dardanus was the name of an Apulian town in the marshes of Salpi,! and Dardi of an old Apulian people.* Both these names seem Illyrian; for the Dardanii or Dardaniate were an Illyrian nation, and there was  a country extending into Mesia, called Dardania. A large  part of the modern.Albanians are called Toscans. In Asia  these names occur again. There was a Dardanus in Mysia. Herodotus speaks of Dardans in what is now Kurdistan.’ Ptolemy mentions Tusci in Asiatic Sarmatia, and there is still  a tribe called Tuschi or Tuschethi at the head of the river Alazan, a tributary of the Cyrus, These are the chief historical or traditional indications of  the extension of one great family of nations, to which the  name Thracian may be applied, from Armenia to Etruria.‘ The languages of the ‘different branches of this family would  have been distinguished from one another by dialectic varia-  tions ; and the discrepancy arising from such a cause would  also have been increased by admixture with- other races in  the countries which the Thracians occupied. In Asia. their  language might have been partially affected by Semitic  elements from the south. The European Thracians, on ‘the La Martiniére, Geog. Dict. 2 Plin. H.W. iii. 16.. * 1.189. The root might be the Arm. dar, ‘height,’ or the Osset.  dard, ‘ distant,’ ‘ The passages in the Greek, which would prove or imply an affinity  between the Etruscans and Armenians, are these, the links of connexion  being—Etruria, Lydia, Mysia, Thrace, Phrygia, Armenia :— ™ Pacl 5é avroi Avdol, . . - « Gua be ratras re éfevpePfvar wapa  aglot Néyouet, cal Tuponvlyny dsrouxloat, k. T. \.—Herod. i. 94. "Amodecxvior 5¢ év Muddoao: Ards Kaplov lpov dpxatov, roo Musoion pev rat Aviotos pérectt, ws kaovyrjrow éotciToict Kapol, Tov yap Avddv xal roy Mucdy Aéyouas elvar Kapds ddedpeo’s. Tovrowsr pev dh péreori* Sool 5, ebvres ANAov | £Oveos, épbywooot Toiot Kapol éyévovro, rovrotor 5¢ ob wéra.— Herod. i. 171. Mugol 8é elot Avidy &roxot.—Herod. vii. 74, IIpds 5¢ rovras, drt rods Mucods, ol uev Opdxas, ol 5¢ Avoods elpjxacs, . . Baprupeiy be Thy Siddexrov > putodddioy ydp rus elvgt, xal mitoppt-yrov. THE ARMENIAN ORIGIN OF THE ETRUSCANS. 9  other hand, as we find from Strabo, were mixed with Celts  and Scythians,! while the Etruscans would have been associated  with Umbrians, a nation of reputed Celtic origin. Finally,  all the Thracian nations would have been more or less affected  by the presénce of Finnish or Turanian tribes, the predecessors  of the Aryan Celts and Thracians in probably all the countries  over which the Thracians at any time extended themselves. Tnnguistie evidence of the extension of the Thracian race from Armenia to Etruria. The inference; that there was one family of nations extend-  ing from Armenia to Etruria, must now be put to the great  test, that of language. We must endeavour to ascertain  whether it can be shown that there was one language spoken  in the countries which lie between those limits—one language,  that is to say, in a wide sense, as English, German, Swedish, Dutch, and other ‘Teutonic dialects, all form one language. Now a complete chain of nations, for the extent required,  would be formed by Armenia, Cappadocia, Phrygia, Mysia, Thrace, Illyria, Rhestia, and Etruria; as Illyria and Thrace  completely fill the space between Rhetia and Italy on the one —Strabo, p. 572. Xanthus the Lydian, and Menecrates of Ela, Strabo’s  authorities for the character of the Mysian language, were, as Mr. Grote - observes, competent judges of the fact. Efpyra: 8’8re xal aéroi ol Mucot Opgxady srockor ell riSv viv Neyoudvww Mucdv (the Moesians).—Strabo, p. 542.

Adrél 8’ol bpiyes Bptyes clot, Opdxidv re E6v0s.—Strabo, p. 295.

Adrot ol Bpiyes Opaxay drotxot elolv.— Strabo, p. 471. ~ "Apuévioe 5¢ xard wep Ppiryes éceadxaro, dbvres Ppvywy droxor..— Herod.  vi. 73.

Kat Evsofos 5¢ év yijs wepiddy gyolv, ‘’ Appévor 7d yévos ex Spvylas, xat TH  gurg woddrd dpvylfover.’—Eustath. on Dion. y. 694.

These drocla:, of the Etruscans and Mysians from the Lydians, of the “Mysians and Phrygians from the Thracians, and of the Armenians from  the Phrygians, may all be interpreted in the same manner, as imaginary  or uncertain migrations founded on the existence of ethnical affinities. Such affinities are likewise expressed, in another manner, by the frater-  hity of Mysus, Tadua, and Car.

1 p. 313. 10 THE ARMENIAN ORIGIN OF THE ETRUSCANS.

side, and Asia Minor on the other. For Strabo, in the begin-  ning of his seventh book, speaks of Illyrians and Thracians, ©  partly mixed with Celts, as occupying the whole country on  the south of the Danube as far as the frontiers of Greece. A  little further on (p. 812) he adds the Macedonians and Epirots, who would have been the connecting links between  the Illyrians and the Greeks, Of the eight countries which form the chain between — Armenia and Etruria, the language of only one has entirely  perished. This is Mysia: but as the Mysian language was  credibly affirmed to be half Lydian and half Phrygian, the  chain of countries and languages will still be unbroken, if the Lydian language be added to the list. It would be thus com-  posed : 3 Country. Language.

Armenia .... Oldor literary Armenian ;a complete language,  still written, but unspoken, and dating  from about 400 a.p.

Cappadocia .. Three Cappadocian words. .

Phrygia) ... Fifty Phrygian words, and a few inscriptions. My aia . Thirty-eight Lydian words.

Lydia Thrace..... Twenty-eight Thracian words.

Hlyria ..... The residuary element in Albanian, after the  elimination of all Turkish, Greek, Latin,  and other foreign words. Rhetia..... The residuary element in Rheto-Romansch,  after the elimination of all German, Latin, | and other foreign words. | Etruria..... Several words, and a great number of inscrip-  tions. ; From this table it appears that we should have only one Thracian dialect left, the Armenian; and that of all the other  kindred dialects only some slight relics would have survived. The Armenian would thus be the representative of the Thra- THE ARMENIAN ORIGIN OF THE ETRUSCANS. ll  cian tongue; and if it be true that all the other dialects in  the countries above named belonged, mainly at least, with the  exception of Cappadocia, and that also perhaps in part, to the  same Thracian family, then the relics of all those dialects  ought to exhibit Armenian affinities. This I shall now at-  tempt to prove for each in its order, beginning with the Cappadocian. And here one circumstanée may be adduced at the  outset as tending to show that the existence of such affinities  was a thing which might have been surmised. The Armenian  language stands alone, a distinct branch of the Aryan stock,  and one marked by such decided individuality as to have led  some to consider it as an isolated form of speech, and to ex-  clude it from the Aryan group. Now, when a language holds  this peculiar position, when it is not merely, like the English,  one member of a family, the Teutonic; but when it represents,  alone, and in a small and mountainous area, almost like the Basque, a complete and peculiar family ; the question may not  unreasonably arise—what has become of the other members of  this family of languages? It is this question which these  pages, in their widest scope, will endeavour to resolve. Cappadocian Words 1. “ Apulejus med. herb. 89: ‘ruta hortensis’ apud Cappa-  docas appellata est moly, a quibusdam armada, a Syris besasa, (Golius Arab. armal prebet).” The Arab. word signifies ‘ruta sylvestris.’ Arm. molakhot, ‘ weed;’ mo/akhind, ‘hemlock.’ 2. “ Hesychius: wne&ic (Is. Voss. ob ordinem makle) év Karradoxla yevduevog pic, dv axloupdy tiveg ALyouaw.” This word may be compared with the Sansk. xakula, Pers.  nighéah, Gael. neas, ‘a weasel,’ and with the Esth. nuggis, Hung. syest, ‘a marten.’ To show that animals of the 1 All the Cappadocian, Phrygian, Lydian, Carian, Thracian, and Scy-  thian words in the following pages are derived from Botticher’s Arica. In their examination, whatever is inclosed within marks of quotation is  also borrowed from that work, to which I am under great obligations. 12 THE ARMENIAN ORIGIN OF THE ETRUSCANS.

weasel kind are sometimes ranked with mice, cf. Lat. mus-tela,  and Germ. ratz, ‘ferret, dormouse, marmot.’ ‘ Hesychius:  oluwp mapa TWapBo¢g xadkiral re pvdg ayplov eldoc, ov raic¢ Sopaic ypwvrat mpd¢g xirwvac. Pers. samir, “Arm. samoyr, ‘mustela, martes Scythica’” (Arica, p. 27). Nne&ic would  likewise be Arm., and also Semitic. For we have in Arm.  kovz, ‘a pole-cat, a marten,’ and kznachis (= kovz-nachis), ‘amarten.’ Nachis differs little from vneEle or vraéle. In th Arm. achis, ‘a weasel’ (cf. Sansk. dkiu, ‘a mouse’), the » of  nachis has been lost. This may be readily explained from the Heb. ach-bar, ‘a field-mouse’ (ar, ‘ afield’) : for the word SY  may represent either ach or fiach.

3. © Curtius vii, 4: Sivos vocabant barbari (Bactrii),  quos ita sollerter abscondunt, ut nisi qui defoderunt invenire  non possunt ; in iis condite fruges erant. Plinius, xiii. 73:  utilissime servantur tritica in scrobibus, quos siros vocant ut  in Cappadocifi et Thracié. Hiymol.mgn.: . . . . 6 Eliot 82 épbypard riva aipol.” Arm. sirim, ‘a tomb’ (cf. ragoc,  radpoc, and Oarrw) ; sor, ‘a hollow.’ Georg. soro, ‘a hole,  a burrow. The word s:rus, it appears, was common to Cap-  padocia and Thrace.! These are all the remains of the Cappadocian language, with 1 The Georgians still keep their corn in subterranean magazines of this  kind. ‘This pit is about eight feet deep. When it is nearly full, fern is  laid over the corn, and the mouth, which has a diameter of two or three  feet,’—the floor is about six feet broad,—‘ is covered with strong boards,  and then earth laid over all till it is on a level with the adjacent ground. This is so carefully done, that the place may be passed over without no-  tice by a stranger, and even waggons may be driven across it; so that  these magazines, which are for the most part in the open court-yard, are  well secured from thieves.’—Parrot’s Journey to Ararat, Cooley’s trans-  lation, p. 67. The form of these magazines must be very like that of the  pit-dwellings of the Armenians described by Xenophon, with an entrance  like the mouth of a well, but increasing in width towards the bottom. Their descendants dwell in the same manner to this day. ‘It is not  uncommon for a traveller to receive the first intimation of his approach  to a village by finding his horse’s fore feet down a chimney, and himself  taking his place unexpectedly in the family circle through the roof.’— Layard, Nineveh and Babylon, p. 14.

THE ARMENIAN ORIGIN OF THE ETRUSCANS. 13  the exception of the names of the months, which closely re-  semble those of the ancient Persian months, as preserved in the Zend and Pehlvi languages,! but which may have been borrowed  from the Persians. It would obviously be unsafe to draw any  certain conclusions from three words only. Two of them, how-  ever, are Armenian, and one of the two, also, Thracian. The  third is Semitic, as well as one of the other two. These facts  would be in accordance with the supposition, that a Thracian  race, to which the kindred tribes of Armenia and Phrygia  belonged, once occupied the intervening country of Cappa-  docia, and that these Phrygians and Armenians were after-  wards separated by an advance of the Syrians from the south,  which gave origin to a mixed Thraco-Syrian dialect in Cappa-  docia, | | Phrygian Words. 1. ’Adapveiv, ‘ pirciv’: adduva, ‘dpidov.’ “Pers. hamdam, ‘socius, amicus, maxime familiaris ;’? proprie, ‘simul vel una  spirans. ”’? Arm. hamadam, ‘< delicious.’ Gael. daimh,. ‘affection ;’ daimheach, ‘a friend.” Heb. dam, ‘blood. Arab.  damm, ‘blood ;’ damg, ‘a friend.’ The form of adauva may  be compared with the Arm. atamn, ‘ dens.’ 2. "AZevov, ‘auywva.” “Arm. azelel, ‘barbam secare:’”  azeh, ‘arazor. Georg. zveri, ‘a beard.’ 3. ’Axeorhy, ‘iarpdv.’ Gr. axearne. Gael. ic, ‘to cure.’ Lapp. wekke, ‘auxilium.’? Arm. égn, ‘aid, support :’ akastan, ‘a strong place,’ th. ak-. Osset. agaz, ‘whole, sound ;’ aghaz, ‘aid.’ The Arm. 6gn may perhaps be found in the Bithynian  dkvoc, ‘ dlpoou rivde yuvackalov eldog.’ 4. “Axptotiv, ‘kXerrplav adgerplda.” “Sopingius adérpiav, Is. Vossius avAnrpiéa conjecit.” Esth. wargus, ‘theft.’ Lith. wagis, ‘a thief ;’ wagysté, ‘robbery.’  song, an alr; erazist, ‘a musician.’ 5. ’Apyvirag, ‘Aamav.’ ' “ Quum ordo literarum hac gloss& Arm. erg, ‘a 1 Cf. Benfey und Stern, Monatsnamen einiger alter Volker. 14 THE ARMENIAN ORIGIN OF THE ETRUSCANS,  interruptus sit, varie restituere conati sunt. Fortasse Arm.  arlez, ‘animale chimerico,’ doctiores ad indagandam veram  lectionem adjuvat. Pers. drddv, ‘spectrum in desertis ap-  parens.’” Lapp. wardali, ‘spectrum.’ Gael. arrachd, ‘a  spectre.’ Arm. aradéch, ‘spectral ;’ araééch tesil, ‘a spectre’ (lit. ‘spectral appearance’). Araééch is properly the instru-  mental case of araéch, the plural form of arad, ‘vision,’ and  tesil is derived from ¢es, ‘sight.’ Perhaps apyvirac, or what-  ever may be the correct reading, = aradates (araé-tes).

6. ‘“Apuay, ‘adéAguov.’ Gael. arm, ‘a weapon.’ Arm.  harovmn, gen. harman, ‘a blow ;’ war, ‘ arms.’ 1. "Arrnyoo, attagus, ‘hircus.’ “Pers. ftakka, ‘caper,  heedus :’? Arm. atakel, ‘valere, posse :’ Zend. takhma, < fortis.’ "Arrnyoc erit aititakhma, ‘perquam validus.’” Arm. th, ‘a goatskin (Fr. douc).’ Georg. thkavt, ‘a goat, = Germ.  ziege, = Osset. sagh. Heb. attud, ‘ hircus.’ 8. ’ArraAn, ‘ papvécc.” “ Quum precedat arradaywaerat, © podvvOhcera, bene videtur Is. Vossius correxisse pdpukic.” Arm, atatz, ‘ timber, carpentry.’ 9. “Bayaiocg 6 paratog. 4 Zede dpbyioc. péyac Toddc  raxtc. In inscriptionibus Persicis, Jaga, ‘deus,’ Arm.  bagin, ara.’”’? Pol. bog, ‘ God.’ Arm. pakas, ‘ faulty,  deficient.’ —-—-Arm. Jazovm; Sansk. dahu; ‘much, great,’ Lyd. tBu, ‘76 woAdb. =“ Bod 76 péya cal woAd Adkwvec.” Arm. wazel, ‘to run;’ wagr, ‘a tiger:’ “a celeritate Tigris  incipit vocari” (Plin. H. NW. vi. 81). Osset. bach, ‘a horse ?’  ef. Pegasus. Here the Arm., by supplying the four words,  pak(as), bag(in), baz(ovm), and waz(el), enables us to explain  all the discordant senses attached by Hesychiug to a single Phrygian word, Bay(aioc).

10. Badhy, ‘ Bactrebc.” Sansk. pdla, ‘king, lord.’ Heb.  baal, ‘lord. Pers. palwdn, ‘a hero” Lyd. wédApve, ‘Bacwrsbc.”’ Lith. wala, ‘power’ Germ. walten.  Dece-  balus was a Dacian king.

11. BayPBaroy, ‘iudrioy, aigotov. Arm. baméak, ‘ cotton,’ THE ARMENIAN ORIGIN OF THE ETRUSCANS, 15 Lat. bombyz, &e. Arm. bambasel, ‘to defame, to speak ill  of” The termination seems to be the Arm. gsel, ‘to say,’  which would leave dam, ‘ reproach.’ 12. Bédu, ‘ bdwp.” Arm. Aivth, vivth, ‘water, element ;’  wét, ‘a wave; wtak, ‘astream.’ Eng. wet, water. Pol. woda, ‘water.’ Lat. wdus. Macedon. é0u, ‘air.’ Pers. wid, bdd, ‘air” Arm. 6d, ‘air. Gipsy wodt, ‘soul.’ 13. Bexde, ‘aproc.’ Alb. bouké, ‘ bread;’ bak, ‘to make  warm.’ Eng. dake. Arm. bowch, ‘a furnace. Rhot-Rom.  butschalla, ‘a kind of bread.’ Dr. Parrot, in his Journey to Ararat, mentions a kind of Armenian bread called bockon (Eng. Ed. p. 169).

14. Bepextvdar, ‘Satuovec.? Alb. perndi, perendi, ‘God? Lith. perkénas, ‘the Thunder-God.’ Arm. weragoyn, ‘superior,  higher, above,’ the comparative of wer, ‘on high,’ = Pers. dar ;  wernakan, ‘ celestial,’ compounded of werim, ‘high,’ and the  adjectival termination -akan. Berecynthus was a mountain. Compare also Arm. erkin, ‘heaven,’ a word remarkably like  the Welsh erchynu, ‘elevated,’ from which Zeuss explains Hercynius.

15. Boextopara, ‘dpynore ppvytach.’ See inf. in Thrac. s. v.  xoAafspropde.

16. avoe rapadscog xappa ga avyn Aevdrng AauTndwv Hoorn kat y Yava bd Popvywy xat Bvywyv. Heb. gan, ‘a  garden.’ Gael. cain, ‘white.’ Esth. canna, ‘fair Lat.  candidus: Gr. yavog. Arm. kanaé, ‘ green.’ Gael. caoin, ‘ pleasing. ——Arm. ganéel, ‘to cry.’ Lat. gannire.

17. TéAapog, ‘adeAgov yuh.’ Esth. kali, ‘man’s brother's  wife.’ Gr. ydAwe, ‘husband’s sister” Arm. & etbér, © OhAcca adeA Goi.’ 18. TAoupde, ‘ xpuod¢.’ Arm. getavor, ‘beautiful, fair’ Gr. yeléw. Germ. gelb. Eng. gold. Georg. okro, ‘ gold,’ 19. Adog, ‘Adxog.” Lapp. dur, ‘ lupus 5’ : tar, ‘canis.’ Pers.  tdzi, ‘a greyhound ;’ ¢éz, ‘a ranning.’ The Arm. for ‘ wolf?  is gayl, 16 THE ARMENIAN ORIGIN OF THE ETRUSCANS- 20. Aapsiog, ‘ird Ilspcdiv 6 ppdvmoc, iwd O82 Dovyoy Exrwp. Pers. ddrahk,‘ascythe.’ Osset. zérch, ‘sharp.’ Esth.  terraw, ‘sharp, cunning;’ Jterras, ‘steel;’ ¢ark, ‘prudent.’ Lapp. #arrok, ‘rigidus, asper ;’ Yarfo, ‘instrumentum haste  instar, quo glaciem pertundunt.’ Arm. sayr, ‘edge, point ;’  zayr, ‘point, head, top, height ;’ dar, ‘ height.’ 21, "EXupog, ‘avAdc.’ Arm. etégn, ‘ a reed, a pipe’ (kdAapog, Mark xv. 19). The ancients mention two places in Armenia  called Hlegia: cf. Germ. Ried. The Arm. etégn appears allied,  as Botticher intimates, to tAcyoc, a word probably not Greek,  but Asiatic. Cf. Miiller and Donaldson, Hist. Gr. Int. v. i.  p. 142. If it be likely, as Miiller thinks, that the Ionians  received the word EAeyo¢ from their neighbours, then it might  be inferred that it was a Lydian, as well as an Armenian word. The Arm. words for ‘ flute’ are, sring (= Gr. obpryé—cf. also Arm. Sriné, ‘sound’) and etégnaphot (lit. ‘ reed-trumpet’). The Arm. chnar, ‘lyre,’ evidently = Gr. xevépa and Heb.  kinnor. Hlegium was a place in Noricum.

22. "EE, ‘éyivov.’? Arm. ozni, ‘echinus.’” So Arm. 6z  and i¢ = Exe, and ayz = al€.

28. “Evot caBot pvorna ply tare eripdiypara ag  ov kat 6 aaaZiog Ardvucog. Sabazius = cdvatih yazata, ...  t.é., ‘generator, creator.” Arm. zevel, ‘to form.’ Evol  seems a mere interjection, like the Arm. ehé! oh oh! éh! &e.,  and similar words in many other languages. Compare, how-  ever, Lat. ave, have, and Arm. chaved, ‘ expia!’ ° 24. ZéAxta, ‘Aayxava.” “Russ. zelen’; Serv. zlak vel zelie ; Lat. olus”’ Arm. setkh, sekh, ‘a melon;’ sokh, ‘an onion;’  sotgam, ‘a radish, a turnip,’ = Arab. salfam, = Georg. thalgami. Alb. Selkyif, §alkyi, ‘a water-melon.’ Kurd. sdk, ‘beet.’ Gr. ofxvg = Arm, sekh, the less perfect form of setkh, ZéAx(cov),  or Selk(yiv%). Compare, also, Arm. zath, ‘a stalk.’ 25. ZéusAev, ‘BdpBapov avdparodoy.’ “ Anne ex Limd et  anya (Arm, ayl), qui aliam terram patriam habet?’’ Arm. Zamach, ‘\and,’= Pol. ztemia; ayl, ‘other.’ But the. Arab. 3 ‘THE ARMENIAN ORIGIN OF THE ETRUSCANS. 17  zamm, ‘binding,’ zimmi, ‘a client, a subject,’ seem to give a  better derivation. If we take into Arm. the root zam or zem, ‘lig(are),’ we may form the participle zemeal, ‘ ligatus, a bond-  man, avdpdmodov,’ in the acc. zemealn. The root of Zéuerev,  like the person it described, may have been barbarous or  foreign. | 26. Zérva,‘abAn. Arm. zat, yayt, yaytni, ‘ open.’ 27. Zevudv, ‘anyhv. “Arm. zow, ‘sea, lake, reservoir.’ ” Pers. zay, ‘ reservoir ;’ awdn, ‘water. Arab. zamzam, ‘a  well.’ Georg. zghwa, ‘ sea.’ 28. KixAny, ‘rév Goxroy rd dorpov.” “ Quum nos gallinam  czlestem habeamus, quid ni xikAnv ad Sansk. Aukkut?a revoce-  mus.” LEsth. kikka, ‘a cock.’ Cf. Gr. xfyAn. The Latins  called this constellation ¢emo, and we call it the wain. Now  temo is in Arm. cheti, which would be pronounced XEN and  might = kfkAn.

29. Kiuspoc, ‘vouc. Arm. shimebs ‘the palate, caprice,  fancy ;’ kamch, ‘will, design, judgment.? Chimch and kamch  are both plural forms: their singular would be chim and kam. Pers. kém, ‘ desire, design, the palate.’ 80. KuBfixn, ‘ darodjpara.’ Arm. k6sik or havik; Pers.  haf; ‘a shoe? Kun also signified ‘the mother of the Gods,’ and.‘ Aphrodite.’ See Arica, p. 35.

31. Arruépoac, ‘Ppvyey dopua.’ Perhaps an ipoag for the  lituus. ‘ Arm. erg, ‘a song, an air;’ erazist, ‘a musician. The  name of the lituus is supposed to be derived from its bent  form. Eng. Uithe. Arm. Mor, “pliable. Welsh lleddy, * pliable.’ 82. Ma, ‘apdéBara.’ Arm. mayel, ‘to bleat ;’ mack ‘an  ewe. Lesgi mazza, ‘a sheep.’ 88. “ Plutarchus de Iside et Osiride, p. 860 B: Pobyec 2 péype viv ra Aaurpa xal Oavpacra trwv Epywv  pavixa xaAdovor. Sia rd Mavev tia rev wara Pacréwv  ayabdy dvdpa xat Suvardy yevica wap’ avroic, Sv Enior Mdodnv xadovow. Gravissimus hic locus veram Ahuramazda Cc ‘18 THE ARMENIAN ORIGIN OF THE ETRUSCANS.

etymologiam tandem indigitat. Zend mazda nihil aliud quam  quod mainyu significat ac sicut Sansk. mastaka, ‘caput,’ dictum  a man, ita Arm. imast, ‘ intellectus,’ ab imanal, ‘ intelligere,’ et Phrygice Manis idem qui Masdes.” Compare also Arm.  mazdezn, ‘as hero’ ——moyn, ‘beauty, grace.’ Lat. amenus. Gr. auetywy. Etrusc. (prob.) manus, ‘ good.’ 84. Mitra, ‘pileum Phrygiom.’ Arm. mithr, ‘ mitre,’ pro-  bably a borrowed term. Arab. midray, ‘a horn;’ midrah, ‘a  prince,’ 85. Nal phy, ‘val.’ Arm. aa, ‘but, however, rather, in  fact ;’ 1mm, ‘really, in fact.’ Jmn appears to identify ufv, pév,  with amo, imo, 86. Nnvlarov, viwhatoc, ‘Ppbyiov proc”  “ Nanta Romanorum in mentem venit, et radix au, ‘laudare.’” Arm.  novag, ‘a song ;’ noval, ‘to mew:’ th. nov (mw). Pers. ndnd, ‘a lullaby song;’ nay, ‘a reed, a flute ;? naw, ‘crying; nuswd, ‘voice, modulation ;? nuwdg, ‘a singer.’ Nini, nexig, and  vnvia-rov, appear to be the nay-nu or nay-nuwd, ‘ the flute-  song;’ and thus zexia and z\eyog would have nearly the  same meaning etymologically, just as they have in fact. We  know that the flute was the proper instrument, both for the  nenia and tAsyoc.! | | 37. Népxov, vwpxov, ‘aoxdy. Pers. nahkrak, ‘a jug, a jar,  a basin.’ | 88. “Opov, ‘dyw.’? Arm, i wer, 74 weray, ’t weroy, ‘ above.’ 89. Ovavody, ‘ vulpem.’ Arab. dways, ‘a wolf,’ Arm. yowaz, ‘a panther,’ Esth. Aunt, ‘a wolf’? Germ, huad. Eng. hound, 40. Tlutptov, ‘Bobrupov’ “Radix pydi, ‘ pinguescere.’ ”? Pers: pth ; Osset. jiu; ‘ fat, grease.’ Arm. panir, ‘ cheese.’ Esth.  pirm, ‘milk.’ Lat. pinguis. Gr. riap, raybe, whxa, wuxvde. [IxpoAéa seems to have signified in Lydian, rAnotoy, 4.¢.,  wuxvioc. See inf, in Lyd. 8. v. Baoxe.

*"Layns 6 Spf atdods xpiros yiper.--Marm. Arund. (Bentley's Pha-  laris, p. xcv,) ‘Edeyeia xal Opfvor wpoogdéueva rots ad\ots (Pausan. x. 7). THE ARMENIAN ORIGIN OF THE ETRUSCANS. 19 41. “ Apulejus de medicaminibus herbarum 5: hyoscyamum Phryges remenia appellant.” Hyoscyamus = Germ. bilsenkraut = Esth. marro rohhi or marro hain ; marro signifying ‘ violent,’  rohhi, ‘ kraut, gras, gewiirz, arnzei (medicamen),? and ain, ‘ hay, grass (herba).’ The same language gives emmis, ‘a sow,’  a word allied to emma, ‘ mother.’ Remenia might be made up  of ro(hht)—em(ma)—hain (medicamen—sus—herba), with the  termination -a ; but the word is difficult to explain. Compare  also Lapp. ruciwa, ‘hemp;’ ruomse, remse, ‘moss ;’ and Arm.  rehan, ‘ basil (ocymum),’ = Kurd. riahn, 42. “In risco, cista pelle contecta, nomen Phrygium.” Gael.  rusg ; Welsh rhisg ; ‘any external covering, rind, skin, husk,  bark, fleece.’ Ital. riccio. Riscus looks like a Celtic, and  would, perhaps, be a Galatian, or even a Cimmerian word, if  the Cimmerians should prove to have been Celts. The Cim-  merians often invaded and partly held possession of Asia Minor  during the ninth, eighth, and seventh centuries, 3.c, 43. SaBor Evoe Ppvylag. A€yovrat cal avril rov Baxyor rapa PpvElv. Ante 28.

44; Sminthos, ‘mures.’ “Sminthus terrigenam significat : Sansk. ASam,. ‘terra,’ gen. kimas, locativus si flexionem pro-  nominum hfc in re antiquiorem sequamur simi (Zend idem  zemi, Neopers. zamin thema), Sansk. did apud Grecos riBéva,  ut thus pro dhita (= hita) Phrygicum esse potuerit. Sminthus, ‘in terré creatus.’”? Arm. zamach, ‘land;’ dnel, perf. edt,  th. d-, ‘to place.’ 45. Lovea, ‘Aclpra.’ Arm. sovsan ; Pers. sdsan ; Heb. susan; ‘hiium,’ 48. Séxyou, ‘irodqpara.’ Arm. sek, ‘ dressed leather, cor-  dovan.’ (Cf. Eng. cordwainer). Lat, soceus.

47. Tiara, “Juvenalis vi. 516: ‘ Phrygia vestitur bucca  tiard.’—Isidorus Orig. xix. 80: Perse tiaras gerunt, sed reges  rectas, satrape incurvas. Reperta autem tiara a Semiramide Assyriorum reginf, quod genus ornamenti exinde usque hodie  gens ipsa retinet.” Arm, dar, ‘height.’ Pers. tar, ‘top.’ — C2 20 THE ARMENIAN ORIGIN OF THE ETRUSCANS, Arm. ¢é, gen. tearn, ‘lord.’ Compare Arm. thag, ‘crown ;’  thagavor, ‘king ;’ thagovhi, ‘queen,’ “Plato, Cratylus, p.410 A: ‘8pa rovro rovvopa ro TUP  ph re BapBapixdy y. Tovro yap ore pgdiov rpocaypa toriy ‘EAAnvecg pwvy, pavepol r’ cioiv obrwe avr kaAdovrrec Pobyec,  ouxpdv re wapaxAlvovrec. Kat rd ye tOwp kal rag xbvac  kat GAAa roAAd.” Arm. hovr, ‘fire;’ hivth, ‘water ;’ Sovn, ‘dog.’ Germ. feuer, wasser, hund. Eng. fire, water, hound.

Phrygian inscriptions.

What few Phrygian inscriptions remain are sepulchral. They will be found in Texier’s Asie Mineure, and have been  given and conjecturally interpreted by Mr. Rawlinson in his Herodotus, v. i. p. 666. They require especial notice here as  genuine relics of the Phrygian language, and also because it  appears to be in a great measure on the strength of these in-  scriptions that the learned translator of Herodotus has dis-  regarded, I cannot but think erroneously, the opinion enter-  tained by the ancients of the affinity between the Armenian  and Phrygian nations and languages.! Instead of classing the 1 p. 652. ‘The statement of Herodotus that the Armenians were  colonists of the Phrygians, though echoed by Stephen,’ (Mr. Rawlinson  should have said, by Eudoxus, who, if of Cnidus, lived 800 or 900 years  before Stephen, and was a man of the highest repute,) ‘ who adds that “they had many Phrygian forms of expression,” is not perhaps entitled  to great weight, as Herodotus reports such colonisations far too readily (as ’——it is said in a note——‘ when he accepts the Lydian colonisation  of Etruria (1.94) and the derivation of the Venetians’ (not the Venetians,  but the Sigynne) ‘ from the Medes, v. 9), and his acquaintance with the Armenians must have been scanty.’ (This is probably true; yet both Herodotus and Eudoxus lived in Asia Minor, both were travellers, and Herodotus may have passed through Armenia on his way to the East. Cf. v.52. Thousands of Greeks, again, would have heard Armenian and Phrygian, and also Persian, spoken, in the time of Xenophon, Agesilaus,  and Eudoxus.) ‘Still, as far as it goes, it would imply that the ethnic  change by which a(n) Indo-European had succeeded a Tatar prepon-  derance in Armenia was prior to his own time, and on the whole there are  perhaps sufficient grounds for assigning the movement to about the close of  the seventh century before our era.’ Ge the evidence of Herodotus and ‘ Eudoxus in favour of the Phrygian  ity of the Armenians be rejected, & THE ARMENIAN ORIGIN OF THE ETRUSCANS. 21 Phrygians with the Armenians, as Herodotus and Eudoxus have  done, and as I should wish to do, Mr. Rawlinson ranks the Phrygians, as well as the Mysians, Lydians, Carians, and Pe-  lasgians, with the Greeks (p. 676), and not with the Ar.  menians. Yet I believe that his own explanation of the Phrygian epitaphs will go far to show that his theory is only  partially true with respect to the Greeks, and not tenable with  respect to the Armenians; for several of his interpretations  are unconsciously Armenian, and not Greek. Nor is this  affinity between the Armenian and Phrygian merely in sense,  but also frequently in form, though, in this last respect, there  are some points of difference. These points chiefly are, that Armenian nouns want the nom. sing. in 8, the gen. in s, and  the acc. in » or m; while the Phrygian, like the Latin and ‘should it be used to prove an Indo-European, afterwards limited to a Medo-Persian, affinity P For Mr. Rawlinson (p. 676) ranks those Aryans,  whom he considers to have gained the prevalence over the Tatars in Armenia towards the year 600 B.c., with the Medes, Persians, Bactrians, Sogdians, and Cappadocians, and not with the Thracians, or with the Phrygians and Lydians. If this be true, the Etruscans can hardly have  been allied to the Armenians, as the supposition of any westward extension  of the Armenian race subsequent to 600 B.c. is inadmissible. Whatever Armenian affinities are found in Europe must be of far earlier introduc-  tion: for no Armenian conquests or migrations can have been made so  far to the west except in pre-historical, or even pre-traditional times. Nor is it only with respect to the Phrygian affinities of the Armenians  that I should put faith in Herodotus. I think that the two other prin-  ciples derived from him, that the Lydians and Etruscans were of one  family, and that the Hellenes and Pelasgians were not of one family, will  be eventually found, in conjunction with the affinity between the Phry-  gians and Armenians, to. form the basis on which the true system of  ethnology in Italy and Greece is to be founded. Both these principles  are generally rejected. Mr. Rawlinson holds (p. 664) that the Greek or Hellenic race was Pelasgian, and that the Pelasgic was ‘an early  stage of the very tongue which ripened ultimately into the Hellenic;’  while (p. 359) ‘ all analysis of the Etruscan language leads to the conclu-  sion that it is in its non-Pelasgic element altogether sus generis, and quite  unconnected, as far as it appears, with any of the dialects of Asia Minor. The Lydians, on the other hand, who were of the same family as the Carians, who are called Leleges, must have spoken a language closely  akin to the Pelasgic ; and the connexion of Lydia with Italy, if any, must  have been through the Pelasgic, not through the Italic element in the  population.’ With this last decision I should be inclined to concur, be- 22 THE ARMENIAN ORIGIN OF THE ETRUSCANS. Greek, is in possession of them. But this will not make the Phrygian a Classic tongue, as such forms are Sanskrit, or  generally Aryan. They are, indeed, Etruscan, which is cer-  tainly a foreign language to Greek and Latin, and, in Mr. Rawlinson’s opinion, not even Aryan. The Armenian genitive  usually terminates in i, ov (ie., uw), or oy. As in the ist, 2nd, and 5th declensions in Latin, the Sanskrit genitive in ae is  replaced by the locative or dative. The Armenian nominative  and accusative are without inflexions. | The simplest of the Phrygian epitaphs is the inscription on  the tomb of Midas :— Ates arkiaefas akenanogafos Midai gafagtae fanaktes edaes ;  which Mr, Rawlinson renders Ates-Arciaéfas, the ‘Acenanogafus, built (this) to Midas, the  warrior-king. y  lieving the Italic, or Aboriginal element in Italy, that element to  which the Oscans, Umbrians, and Sabines belonged, to be Hellenic ; and  the Pelasgian, on the other hand, not to be Hellenic, but enian, and  allied to the Phrygian, Lydian, and other Thracian dialects.§ As we are  not acquainted with the meaning, or certain of the existencé, of a single Pelasgian word, the cause of ethnology would probably be advanced by  avoiding ‘ Pelasgian,’ as far as possible, as a defining term. Greek or Hellenic, Latin, Greco-Latin or Classic, Armenian, German, Gaelic, Celtic, &c., are terms which have a linguistic, and therefore a determinable  ethnic sense. Pelasgian has none, unless some previous hypothesis be  made. It is generally used in the sense of ‘Greek’ or ‘ Classic’; but  this involves an assumption, and one of these latter terms might be more  definite in investigations of ethnic affinity. Sometimes, however, the  obscurity arising from the term ‘ Pelasgian’ may be partly remedied, as  in the two propositions above noticed relating to that unknown race. If ‘the Lydians must have spoken a language closely akin to the Pelasgic,’  and if ‘the Pelasgic ripened ultimately into the Hellenic,’ it would then  follow that the Lydians must have spoken a language nearly allied to the Greek ; a conclusion which admits of being tested, as the Lydian lan-  guage, unlike the Pelasgian, has left some relics. Yet as these relics, as  far as I can judge, are not Greek, I should infer that one of the two pro-  positions relative to the Pelasgians, and in all probability the second pro-  position, was erroneous; and that Herodotus was right in his opinion, Foay ol Hedao-yol BdpBapov yNdooay lévres.

1 I do not know any analogy for gafagtaei, ‘warrior’: but there can  be little doubt that the two other words which are explained, fanaktes  edaes, are correctly rendered, ‘ &vaxri £6nxe.’ THE ARMENIAN ORIGIN OF THE ETRUSCANS. 23 The Armenian would explain the inscription in the following  manner :— .

Ates ..... A proper name, Atys. 3  arkiaefas . . . Arm. archay,‘a king; archayazn, ‘a prince.’ Gr. apyh, dpxwy. Archayazn is compounded  of archay, ‘king,’ and azn, ‘race’ <A  synonym of azn is azg,so that the root is.  az; and archayaz(n) signifies ‘a man of  royal race.’ But perhaps the termination of  arkvaefas may be merely formative. Lassen  suggests here the Sansk. arka, ‘reverence,’  which = Arm. yarg, whence yargs, ‘ respect-  able.’ The name, fartknafas, was found on  an earthen pot near Este, and éarchnas at Cervetri. Compare tariknafas and tarchuas  in form with the Phryg. arkiaefas and the Arm. archayaz(n).  akenanogafos . Arm. akanavor, ‘illustrious,’ primarily, ‘having  eyes ;’ a word compounded of akn, gen. akan, ‘an eye,’ and of the Arm. termination, -avor. Compare Arm. thag, ‘a crown,’  thagavor, ‘a king.’ The last member of  aken-anogafos might be explained from the Arm. ang, ‘due, fit;? angov, ‘ worthy of ;’  yangavor, ‘ proper;’ which would give for  aken-anogafos the sense, ‘worthy of note (merkwiirdig), illustrious.’ Akn, ‘an eye,’  enters into the composition of several Arm.  adjectives ; as aknazov, ‘ respectful ;’ aknerev, ‘ evident ;? aknkaroyz, ‘regarding atten-  tively.” In another Phryg. inscription we  meet with Bonok akenanogafos and Inanon  akenanogafos, so that akexanogafos would be  in all probability a title of honour. 24 THE ARMENIAN ORIGIN OF THE ETRUSCANS. ‘ Midai .... ‘To Midas.’ Miday would be the Arm. form of  the gen. and dat. of Midas.  gafagtaci . . Arm, gah (= Pers. gdh), gahak, ‘throne, dignity,  eminence ;’ gahakiz, gen. and dat. gahakze, ‘sharer of a throne, colleague.’ The de-  clension of gafagt-aei seems to resemble that ° of the Arm. archay, ‘king,’ gen. and dat.  arch-ayi; or of Hermés, ‘ Hermes,’ gen. and  dat. Herm-eay.  fanaktei ... Arm. nakh, ‘first ;’ nakhki, ‘chief.’ Gr. avak. Pers. Inscript. naga, ‘king.’ In form, Janaktei may be compared with the Arm.  anakhti, gen, and dat. of anakht, ‘pure;’ or  with Anahtay, gen. and dat. of <Anahit, ‘ Anaitis,’ the great goddess of the Armeni-  ans.| In the Arm. nakhagah, ‘president,’  we find nakh and the previous word, gah, in  combination. 1 J should be inclined to connect the name Anahit etymologically with  avat. The Milesian traditions reported by Pausanias may throw some  light on this subject. He says (lib. i. c. 25) :— "Kori 82 Midnolots wpd rijs worews Addn vijcos, dreppuryace 52 dx’ abrijs vnoldes. "Acreplou rh» érépay dvoudfovor, xal row ’Aorépioy ev airy rapivat Néyoucu" .

elvas 5¢ ’Aorépiov yey “Avaxros, ”“Avaxra d¢ T'fjs raida.

And again (lib. vii. c. 2) :— Mcdrjocos 6é avrol rodde 7d dpxatérard ogiow elvar Aéyovew emi yeveds per 5h *Avaxroplayv xadeiobat ri viv,” Avanrdés te ad7éxGovos kal ’Aoreplov Bacued-  ovrosvod “Avakros. - The name of Anaz,.the autochthon and the son of Ge, cannot be better  derived than from the Arm. zak, ‘primus.’ Such a derivation would  also be suitable to Anahit: the Greeks called their gods dvaxes; voc.  sing. dva. The name Asterius, again, is readily connected with the Gr.  dorip. Now, in mythology, Asteria is the daughter of Phobe, who was  the daughter of Terra: and in the Milesian (or Carian) traditions, Aste-  rius is the son of Anax, who was the son of Ge. It is obvious that Aznar  here corresponds to Phebe, Diana, or Anahit; and when we compare  together the three datives, the Arm. anahtay, the Phryg. fanaktei, and  the Gr. dvaxri, the resemblance is very close. Indeed, Anahit may be  considered as the same deity as ‘ Astarte, gueen of heaven with crescent  horns.’ The worship of Anaitis was not confined to Armenia. She twas the THE ARMENIAN ORIGIN OF THE ETRUSCANS, 25  edacs ..... Arm. ed, ‘placed? Gr. 20nxe. The Arm.  dnel, ‘ to place,’ perf. edt, is an irregular verb. Similar verbs, regularly conjugated, like gua, ‘to go,’ and gnel, ‘to buy,’ make gnai, ‘he  went,’ and gneaz, ‘he bought.’ In general,  in the case of the third pers. sing. perf. ind. in Arm., regular verbs have the root and the  termination, but not the augment, and  irregular verbs the augment and the root,  but not the termination. Thus a regular  verb, zital, ‘ to contrive,’ makes #2¢-azZ, ‘he  contrived ;’ while an irregular verb, ¢a/, ‘ to  give,’ makes e-t, ‘he gave,’ the complete  form being evidently ¢-¢-az, ‘he gave. A  similar complete form, in the -e/ conjuga-  tion, would be edeaz, instead of ed, ‘he  placed.’ Now edeaz, as ea is a diphthong, is  very like the Phryg. edaes, which apparently  consists of the augment e-, the root -d-, and  the termination -aes. Another illustration Lydian Diana, the ‘great goddess’ of the Ephesians, aa well as the  great goddess of the Armenians (Pausan. iii. 16, Plin. H. NW. xxxiii. 4). She was also worshipped in Cappadocia (Strabo, p. 733). We find, again,  that Anu or Ana was one of the three great Assyrian or Babylonian  deities. The name signified ‘the God,’ xar’ éfox4» (Rawlinson, Herod.  v.i. p. 591). ‘ One class of his epithets refer undoubtedly to “ priority ”  or “ antiquity ”,’ and thus point to a word like the Arm. naka, ‘first,’ as  a root. The wife of Anu or Ana was Anuta or Anata (p. 593), a name  which is not far from Anahit. ‘She had precisely the same epithets as  himself.’ It is also noticed (p. 603) that the ‘ great goddess’ of the Baby-  lonians was Mulita or Enutad Now Anahit was the ‘great goddess’ of «+-  the Armenians and Ephesians, and the same as Mylitta or Enuta, whose  dissolute rites her own resembled. She was identified with Venus as  well as Diana.

Thus, when we come to consider the word ana or anaz, we trace it in Greece, Caria, Lydia, Phrygia, Cappadocia, Armenia, Assyria, and Babylonia: and its root is Armenian. The word is not found in Latin. The natural inference would be that 4vaé was a Pelasgian, not a Hellenic  term ; or, in other words, that it was of Thracian origin, and did not  belong to the Classic or Greeco-Latin atock. ) , 26 THE ARMENIAN ORIGIN OF THE ETRUSCANS.

may be supplied by the Arm. for ‘to lick,’  which is conjugated in three forms, /2ze/,  lizovl, and lizanel, of which the first makes  the perfect regularly, and the two last irre-  gularly. We thus have for ‘dazié,’ the  forms, lizea?, and liz or eliz, the complete  form being e-liz-ceaz. The interpretation of the Phrygian inscription, as derived  from the Armenian, would thus be: Atys, the illustrious prince, made (this tomb) for Midas, the  enthroned king. By comparing the Phrygian inscription with the Armenian,  we may see what the latter language has apparently lost— Phryg. Ates arkiaefas akenanogafos Midai gafagtaei Arm, Ates archayaz(n) akn-yangavor Miday gahakit Phryg. fanaktei edaes. Arm, | rth pe ed. Neither language seems to have possessed the article, but  the Phrygian has the digamma. The Phryg. /anaktei, so  closely resembling the Gr. avaxri, we are obliged to render in Arm. by 73 nakh or 74 nakhki, both nakh and nakhki being  indeclinable, just as we should be obliged to render the Lat.  regi by the Ital. a ré. In like manner, the Phryg. edaes is  reduced to the Arm. ed, as the Lat. posuit is reduced to the Ital. pose. The Arm. has, however, in general, preserved the  ancient inflexions much better than the Ital. Thus the termi-  nation of ed-aes is, as I have shown, still preserved in Arm., as  wellas of fanakt-ei, though not in the equivalent Arm. words :  so that, upon the whole, there is no part of the inscription on  the tomb of Midas, whether the root or form of the words be  considered, but what appears to exist in the Armenian language.

There are two other Phrygian sepulchral inscriptions. The  first of these, which is on the side of the tomb of Midas, is as  follows, accompanied by Mr. Rawlinson’s interpretation :— THE ARMENIAN ORIGIN OF THE ETRUSCANS. 27 Baba Meméfais protafos kphi ganafepos Lord Memefais, son of Pratas,  larvae Sikeman . edaes.

a native of Sica, —_ built (this) Here we meet again with edaes, the meaning of which can  thus hardly be doubtful. To explain the rest of the inscrip-  tion, it is necessary to resort to conjecture. Proitafos and  ganafepos seem to be the titles of Baba Memefais ; and conse-  quently phi, judging from its position, may signify ‘and,’ as  afe will be found probably to do in Etruscan. Cf. also Arm.  kap, ‘janction.” In proit-afos we find again the termination -afos, asin akenanog-afos. There remains as a root, prot¢, which,  as the mark of dignity in rulers, governors, consuls, prestors,  was a staff, we might perhaps explain from the Arm. Ji,  wirg, ‘astick,’ prtov, ‘a rush;’ Gael. dior, ‘a stick,’ druid, ‘a stab;’ Lat. virga. This would make proit-afos= Arm.  br-avor, i.e. bir-avor, ‘stick-bearing, oxnwrovyoc.’ Strabo  says of the Heniochi near Colchis (p. 496): Suvacredovrac 82 Kat ovrot Urd Tov KaAdoupvwy oxnrrobxwy’ Kal avrot d  ovroe brd Tupavvog 7h Baotrsiow eiclv. There was also a  oxnrrovxoc at the court of Persia. For the second title,  ganafepos, from which the nominative termination -os, deficient  in Armenian, is to be subtracted, we may form from the Arm.  words, gan, ‘punishment,’ and wép, ‘ zroc,’ the word ganawép,  judge” Cf. ju-dex, vin-dex, and Arm. bazmawép, ‘one who  relates many (bazovm) things.’ Baba, if allied to wamrac,  would be so also to the Arm. synonym pap, Pers. b¢0: and  the whole inscription might be thus interpreted, but with  great doubt— | Baba Memefais proitafos kphi ganafepos Sikeman Ilarmac Mepegaic oxnTrovxo¢g Kat Saorne  LKaioe  edaes, 2Onxe.

The remaining inscription is of greater length, and the last 28 THE ARMENIAN ORIGIN OF THE ETRUSCANS.

few words are not quite clear. I give it as far as Mr. Rawlin-  son has interpreted it :— Kelokes fenaftun aftas materes sosesait, materes Celoces sepulcram sue matris extruxit, matris  avutig pnrépoc LEpheteksetis ofefinonoman lachit ga  materan aresastin, Ephetexetis ex Ofefinone. Sortita est tellus matrem amatam. "EXaxe yi pnrtpa étpaorhy.

. aplorny. Bonok akenanogafos erekun _—_ telatos sostut- Bonok qui Acenanogafus erat hordeum sacrificii obtulit.  snanonakenanogafosaer, . . . « . Inanon Acenanogafus. . . . . .

The Armenian will explain, as interpreted above, several of  these words, which I shall take in their order :— Fenaftun, ‘sepulchrum.’ Arm, andth or anavth, ‘ vessel,  pot, box, piece of furniture. Another Arm. word, ¢apan,  signifying ‘ box, urn, ark,’ and therefore nearly a synonym of  anavth, means also ‘sepulchre,’ and is commonly employed in  that sense.! Arca and ayyetoy are similarly used. 1 See the Journal Asiatique (Février-Mars, 1855), Voyage a Sis, the  former capital of the medisval Armenian kingdom in Cilicia. I subjoin  one of the epitaphs (p. 277), correcting some few inaccuracies in the re-  duction to literary Armenian, and rendering the Armenian letters by  their equivalents, with the omission of the diacritical marks of the con-  sonants. Yays tapanés kay — edeal In hoc tumulo manet positus , Tér Michayél mezn — éntreal. Dominus Michael magnus electus. Sa wsemagoyn werabereal Tile superior  elevatus Yovsakan 6 werakoceal. Desiderabilis est cognominatus. Thovoys hazar ev erkov harivr Anno mille et duo centum (a.p. 1751.) E hangéstiv i = Tér hangeal. Est pace in Domino quietus. The Armenian of the epitaph differs slightly from correct Armenian.

THE ARMENIAN ORIGIN OF THE ETRUSCANS. 29 Aftas, ‘sup.’ Gr. avrov. Arm. ivr, ‘of him;’ troy, ‘of  his ;? aysr, or aydr, or aynr, ‘ of this, of that.’ Sansk. efasya, ‘of this.’ | Materes, ‘matris. Gr.pnrnp. Lat. mater. Gael. mathair. Sansk. mdétri. Pers. mddar. Arm. mayr. Osset.mdd. Arm.  matak, ‘female (of animals),’? = Pers. médak. The Arm. has,  in mayr, suffered the same loss as the Lat. in puer. Cf. Sansk.  putra = Zend puthra = Pers. pusar = Osset. ftirth = Arm. ordi = Lat. puer. - os Sosesait, ‘ exstruxit.’ Pers. séé, ‘furniture’. . . Arm. sar, ‘ furniture.’  sdtidan, ‘ to prepare.’  sézid, ‘he prepared.’  sézis, ‘contrivance’. . . . saras, ‘form, shape.’  sarasel, ‘ to form, to shape.’  saraseaz, ‘he formed, shaped.’ The Pers. verb, it will be seen, is formed from ééZ, not from  sé4is, which corresponds to the Arm. saras. Had it been  formed from sé@Z7:§, we should have had Pers. séz:3/d = Arm.  saraseaz = (prob.) Phryg. sosesait. The termination of soses-  ait seems =that of ed-aes. Compare lives, liveth, and Germ.  lebt; has, hath, and Germ. hat. The Arm. =z. In the Gael. sas, ‘an instrument,’ another kindred root to the Phryg. Thus ¢ér is written tr, and edeal, etheal. The terminations in J, so com-  mon in Etruscan, are here exemplified in Armenian. deal, ‘ positus,’  is the participle of ed, ‘ posuit,’=Phryg. edaes. The terminations of  tapanés, mezn, and thovoys, are not inflexions, but superfluous additions. These final letters in Armenian, consisting of s, d, and , originally indi-  cated a reference to the first, second, or third person, being really parts  of the pronouns, ‘J,’ ‘ thou,’ ‘he.’ The sin saand yays is in like manner  borrowed from es, ‘I ;’ the actual demonstrative pronoun being a or ay,  though it is never used without one of the letters, s, d,. The initial  letter in yays is the preposition t, before a vowel y. In hangstiv, pro-  nounced, and written in poetry, hangéstiv, the instrumental case of hang-  ist, of which the th.is found in hang-eal, we may perceive a form like the Phrygian aresast(in). The Arm. hazar, ‘thousand, mille,’ is found in Sanskrit, Zend, Persian, Gipsy, Crimsan Gothic, Hungarian, and the Slavonian of Carniola, Styria, and Carinthia. See Diefenbach, Lez. Comp. 8. V. hazer. 30 THE ARMENIAN ORIGIN OF THE ETRUSCANS,  sos- might be found, as well as in the Arm. &s§aphel, ‘to  handle,’ a verb in -aphel from a root 568. Dealing with 568 as  the Arm. has done above with sar, we should obtain sésaseaz, ‘he handled,’ 2.¢., ‘managed, made,’ a word very like the Phryg. sosesait. The verb sdSaphel forms ssapheaz. Ofefinonoman, ‘ex Ofefinone.’ Sikeman, ‘ a native of Sica,’! . Armenian.  ayd, ‘this, that? .... . da, ‘ this, that.’  aydr, ‘of this’ ..... . dora, ‘of this.’ (y)aydmané, ‘from this’ . dmané, ‘from this.’  ayg, ‘morning.’ (y)aygman, ‘in the morning.’ Osset. ay or a, ‘this;’ aman, ‘to this;’ amiy, ‘in this;’  ama, ‘at this ;’ amey, ‘from this,’ Lachit, ‘tdaxe.’ This sense eannot be obtained from the Armenian. The Arm. word which most nearly resembles  lachit is lakeaz, ‘ (it) consumed, swallowed, absorbed.’ But  lachit, as the Phryg. perfect seems to terminate in -aes or -ait,  should, perhaps, rather be taken as a present tense, The Arm.  present of dukel, ‘to consume,’ is /aké, ‘ it (z.¢., the earth) con-  sumes, swallows,’ If the Pers. had a “corresponding verb from  a root /ék-, the pres. would be /ékad, and the perf. /dékid. Lachit might also be compared with the Arm. efak, ‘ fossa,’  and be interpreted ‘ sepelit.’ Ga, ‘earth. Gr. yn = Gael. ce = Sansk. go. Arm. hav, ‘clay.’ Germ. gaw= Arm. gavar. Lachit ga, ‘ devorat tellus,’  or ‘ sepelit (Celoces) in tellure.’ These seem to be the chief points requiring notice. In the  nouns, as I said before, the Phrygian, like the Latin and Greek, has preserved the Sanskrit or Aryan terminations where  they are deficient in Armenian : thus the Phryg. fexaftun aftas  materes sosesatt appears = Arm. (z)anavth mavr wroy saraseaz. It is only in the past and future participles, and then not in- 1TI do not know any place called Sica in Asia Minor. We have Sicum  in Illyria, and Siculs in Italy. ‘THE ARMENIAN ORIGIN OF THE ETRUSCANS. 31  variably, that the Arm. acc. sing. presents the termination -2:  e.g. sireain, ‘amatum;’ srelinm, ‘amandum.’ The Etruscan  possesses both the gen. in -s and the acc.in -m. Many Arm.  pronouns, however, form the dat., though not the acc., in - or -oum(um): as ayd, ‘this;’ gen. aydr ; dat. aydm; acc. (z)ayd —— im,‘ my;’ gen. imoy; dat. imovm; ace. (z)im; abl. (y)tmoy; instr. tmow (ow = Gr. w). Here wm seems = tude or Meus; Woy = nov OL met; UnovM = indy Or menin; and imow =  gu or meo. The locative of nouns may also be formed in -oum, as 4 mard or *4 mardovm, ‘in man.’ Mr. Rawlinson has noticed the resemblance in form between  the Phryg. ares-astin, and such Gr. words as ap-lorny and 2p-aarhy, and has even translated aresastin, ‘amatam,’ which  would require that the Phryg. ares- should = Gr. ép-. Similar  forms appear also in the Phrygian vocabulary, in ax-earhy and  axp-cortv; though the first, being actually Greek, may have  been a borrowed word, if not modified by the Greek reporter. The same forms are found in Arm., with the exception of the  acc. termination in 2: e. g. Phrygian. Armenian.  tmaast, ‘intelligence ;’ th. tm(anal) * ‘to under-  stand.’  ares-ast(in) . . nao aatit ‘a sailor:’ th. nav, ‘a ship.’  ovr-ast, ‘a denier :’ th. ovr(anat), ‘ to deny.’  tap-ast, ‘ fallen, laid down :’ th. tap(al) ‘ to fall.’  gow-est, ‘ praise :’ th. gow(e/), ‘to praise.’  ovt-est, ‘food :’ th. ovt(el), ‘ to eat.’ ‘ax-cor(hy) - "| pah-est, ‘reservation:’ th. pah(el), ‘to re-  serve.’  dixp-tor(wv) . .{ hang-ist, ‘repose: th. hang(eal), ‘ quiet.’  naz-tst, ‘a female servant.’  ber-ovet, ‘tendency :’ th. der(el), ‘to bear,’  gal-ovst, arrival :’ th. gal, ‘ to come.’  thag-ovst, concealment :’ th. thag(ovn) ‘ hid.’ | eraZ-13¢, a musician.’ 32 THE ARMENIAN ORIGIN OF THE ETRUSCANBS: In order to explain the word aresastin, we may take the Arm. yaraganal, ‘ preire,’ th. ara or yarag, ‘ pre;’ and form  yaragast in precisely the same manner as ¢mast and ovrast are  formed from imanal and ovranal. This would give— Phrygian, Armenian,  ares-ast(in) . yarag-ast, ‘ precellens, preestans :’ th. yarag(anal), ‘ preecellere,’  and the meaning of lachit ga materan aresastin would be, ‘devorat tellus matrem prestantem.’ Imast is declined ——  nom. imast; gen. and dat. tmasti; acc. (z)imast; abl. (y)tmasté ;  instr. emastiv. The other forms ending in -s¢ or -3 are mostly  declined in the same manner. Aresastin might also be ex-  plained, ‘ departed,’ from the Arm. Arazeét, ‘ abdication, resig-  nation, farewell.’ The meaning of the remainder of the inscription is extremely  doubtful. There are two proper names in the nominative,  each bearing the title of akenanogafos. The rest is uncertain,  except perhaps in construction. All that can be done in such  a case, when comparing the Phrygian with any other language  with a view to prove affinity, is to select such words in that;  language as resemble the Phrygian, and see if they will give  a satisfactory meaning. The first clause to consider will be, Bonok akenanogafos erekun telatos sostut. The first two words  being known, the Arm. suggests for the remaining three, con-  sisting apparently of an acc. in -wm, a gen. in -08, and a verb— | Hrekun. Arm. erk, erkn, erkch, ‘tpyov;’ herk, ‘ cultivation.’ Telatos. Arm. thatovmn, ‘burial;’ that(el), ‘to bury;  thatar,‘an earthen vessel, a basin;’ ¢hat, ‘a territory ;’ ¢el2, ‘aplace. Sansk. ¢al, ‘condere;’ ¢ala, ‘solum, fundus.’ Gr. Oadrapoc. Lat. tellus. Gael. talamh, ‘earth.’ Etrusce. (as will  afterwards appear) tudar, ‘atomb.’ Arab. ¢alhid, ‘burying.’ For the form of te/at(os) from a root tel, compare Arm. armat, ‘root,’ th. arm ; Sinaz, ‘a building,’ th. sn(el), ‘to build ;’ and  for its declension, Gr. répac, réparog, or Arm. kin, knog, ‘ yuvi,  yuvakdc.’ - THE ARMENIAN ORIGIN OF THE ETRUSCANS. 33 Sostut. Arm. sasté, ‘he reprehends, érirmiq’ (Matt. viii. 26),  th. sast = Pers. zust, ‘strong, severe.’ Sansk. ¢ds, ‘jubere,  regere, docere, punire.’ Sostut and Jachit would respectively  belong to conjugations like the Arm. hetov, ‘he pours,’ and .  kami, ‘he wishes.” The whole clause becomes, ‘ Bonok the Illustrious forbids  work of sepulture;’ 7. ¢., no other interments were to be  allowed in the same place, a common prohibition with respect  to ancient tombs.

The remaining clause begins with another proper name,  which Texier reads Jnanon. Steuart reads it very differently,  which is of no importance in the case of a proper name.

- The two readings of the remainder are— Tex. akenanogafos aer atanisen kursaneson tanegirtog2 Ste. akenanogafos  atanisen kursaneson tanegertos. .

I shall take the reading to be—  akenanogafos aer atanisen kursaneson tanegertos.

The last two words seem to be an acc. and gen. As noverb  appears, sostut is probably understood from the preceding  clause. Akenanogafos is a known word: for the rest the Arm.  would give— Aer, Arm, ayr, ‘man.’ | Atanisen. Arm. atean, ‘a tribunal, a magistrate, a senate ;’  atenakan, ‘belonging to a tribunal, a magistrate, a judge.’ I  am not clear as to the termination sen. It might be compared  with the Arm. suffix -sé” (see inf.), but that ought rather to  be represented in Phrygian by -senos.

Kursaneson. Arm. korzanovthivn, ‘ruin, destruction,’ th.  korzan. The termination of kursan-es(on) might perhaps be  compared with the Alb. and Arm. suffixes, -e§ and -7¢. See  inf. in Alb. s. v. lezoueé.

Tanegertos. Arm. tovn, gen. tan, ‘a house ;’ kert, ‘a build- 1 We have in Arm. both lizé and lizov, ‘lingit, Aelxet.’ * In his text, tageirtog. The reading of the plate is to be preferred. D 34 THE ARMENIAN ORIGIN OF THE ETRUSCANS.

ing.’ Tanegert(os) would be a compound like Tigranoceri(a), ‘the building of Tigranes.’ Compare also Arm. gomakert, ‘who builds (kerte) a stable (gom) ;’ tnasén, ‘ who builds (sé)  a house (éovn) ;’ and mezaésén, ‘ great (mez),’ archayasén, ‘ built  by a king (archay), royal, magnificent,’ where sé = ker¢ has a  passive sense equivalent to factus. Thus the Arm. might form  tnakert, ‘ oixoddpog, oixodouh. Compare mezasén and atantisen.

By this last clause, ‘Inanon the Illustrious, a man with  judicial power, forbids any injury to the sepulchre;’ which  coincides with the second prohibition continually found in  sepulchral inscriptions. | For the whole epitaph, the actual Armenian, unmodified,  would correspond with and explain the Phrygian in the fol-  lowing manner :— Phryg. Kelokes fenaftun aftas materes sosesait, (2) (1). Arm.  Kelokés (z)anavth wwroy mavr  saraseaz, Lat. Celoces sepulcrum sue matris exstruaxit, Phryg. wateres Epheteksetis Ofefinonoman. Lachit Arm. mavr Lphetéchsetay yOwewinoneay. Laké Lat. matris Ephetexetis eax Ofefinone. Devorat Phryg. ga  materan aresastin. Bonok  akenanogafos Arm. kav (2)mayr aragin. Bonok akanavor Lat. tellus matrem gprastantem. Bonocus Itlustris Phryg. erekun ~ telatos sostut ; Inanon akenanogafos, ' - ¢ thatman Arm,  (z)erkn thatari sasté; Inanon akanavor, : tetvoy  sepulchri Lat. wsum UTNE _ vetat; Inanon Illustris, Phryg. aer atanisen, kursaneson __ tanegertos. (2) (1) Arm.  ayr atenakan, (z)korzanovthwn tan-kertt. (2) (1) Lat. vir judicialis, destructionem domiis-structure. (adifici) THE ARMENIAN ORIGIN OF THE ETRUSCANS., 385 ; ( The advantage of the Armenian over the Greek in the  interpretation of the Phrygian epitaphs seems sufficiently  clear.: It 1s true that one or two words are more nearly Greek than Armenian. 4/¢as is nearer to avrov than it is to  wroy, and materes is nearer to pnrépog than it isto mavr. But  such partial resemblances would be frequently delusive, if relied  on, and lead to very erroneous ethnological results. Thus, to  take exactly parallel cases, the Germ. sezm and seinmige are nearer  to the Fr. sen than to the Eng. fis; and the Lat. mater,  matris, are nearer to the Eng. mother, mother’s, and the Germ.  mutter, mutters, than they are to the Fr. mére, de la mere. It  must also be remembered that, while the Phrygian and Greek are ancient languages, the Armenian, in the oldest form  that we possess it, is comparatively modern.) ( The conjugation of Phrygian verbs is Armenian rather  than Greek, as appears in edaes and sosesatt ; but the declen-  sion of Phrygian nouns, on the other hand, is more Greek  than Armenian. Yet what the Armenian is here deficient in,  are merely such Aryan characteristics as are easily lost, as will  appear by the following table of declensions.:—Here I have  illustrated the Phrygian declensions, by comparing them with Armenian, Sanskrit, Etruscan, Greek, and Latin declensions. From the Armenian declensions, which are very numerous, I  have selected the most common, as exemplified in Movzay, ‘ Musa,’ which is regularly declined, though a borrowed word ;  san, ‘a nurseling, a godson;’ Levonidés, ‘ Leonidas ;’ Anahit, ‘ Anaitis ;? anakht, ‘ purus ;’ armat, ‘ radix ;’ kin, ‘femina;’ ayd, ‘is, ea, id;’ szredi, ‘amandus, -a, -um;’ and zav, ‘navis.’ In  the Sanskrit, I have taken the two regular types, au, ‘ navis,’  and harit, ‘viridis.’ As we do not seem to meet with any Phrygian plurals, I have not given any in Armenian. The Ar-  menian plural is usually formed thus: in the nom. and instr.,  ch (x) is added to those cases in the sing. : in the acc., s is added  to the nom. or acc. sing.: and in the gen., dat., and abl., 2, pre-  ceded by some vowel or diphthong, is added to the nom. sing.

- D2 THE ARMENIAN ORIGIN OF THE ETRUSCANS, 36  avapuoaay aoune arhvzaopy  guve ghozaopy * °  ups «= hvzaopg °°  kvopuoaay aoune rhvzaopyy : °  kwapuonay ~aouve = rhvzaopy * *  avavu 00919118 aophp avuy nojvuso Mgyynun avjyyoupy  paou hojanes gunupkho ghouy hojvmso gayynun hojyouy hvopruoasy  anu wyase pho ury quusD = =6fyyoUD you eppyuoaaT  cavu = hoqa.ss upho  fouy hogouin wyyyoun hoqyouy  canu horas spho fouy hoyouso Yyyyouv hoyyouy  auu = Yas pho wey joudo 8 8=6fyyouD yDUp — eppluoaayT “UBIMOULIY  uvuouowa fo  unyasa uviazour unyfouaf UIqEneaLD  toyyounf — wp tambo fob 8a.LoqnUt 8090994  sode founb 07079  uns Aryy  pen = bane Skonon Sosxnan Soka penu aavu  uyyUnry woUuryz7— f avonon = vixnan dana wosnu, § waanu 1018) Gono =isxpan ka Dene = = tan - wy,uniyy = sDUry7zj24 Sonor Sosman Soka pen sav Yuu =«vuryzq—a4 ponon gvap Sapa DEN = 81D “UBOSTUIP TT “WoeIH “UlFerT  ue = hvzaopy  supfo °  eT v6 ° PHADY =— pay * *  suliny = svayu * UDiNIDY = Uvagu * *  aguiny = aap °  soptoy = Swapu * *  quiny 8 =8§6snvu ° “POSTS THE ARMENIAN ORIGIN OF THE ETRUSCANS. 37  am Although the resemblance between the Phrygian and Armenian languages may not be always perfect, yet it is, I  think, sufficiently close to confirm the opinion of the ancients,  that the Phrygians and Armenians were of the same race, and  that the Armenian language was like the Phrygian in many  points. The Latin would hardly, under similar circumstances,  approach the Greek more nearly than the Armenian does the Phrygian, as may be readily tested by the aid of the Phrygian  inscriptions and vocabulary / Take, for instance, the meaning  of the first six words in the vocabulary, ¢fAov, rwywva, iarpdy,  kAerrplav aXerp{da, Admrayv, and wéAeuov. The affinity to the Latin is here by no means complete or remarkable. So, on  the other hand, if we take what seems the correct inter-  pretation of the Phrygian Kelokes fenaftun aftas materes sose-  sait, 1.€., Celoces sepulchrum sua matris exstruxit, we shall per-  ceive little obvious affinity in these Latin words to the Greek,  either in root or form, except in matria. « ( There appears to be no other language but the Armenian so  near the Phrygian as to claim to belong to the same Aryan  family. The next in order of affinity would probably be the Persian.) Yet both the classic languages exhibit some signs of  affinity to the Phrygian, although of a different family ; such ‘instances of affinity being probably either Aryan generally, or Thracian words borrowed from the Pelasgians and Etruscans  by the nations of the Classic or Old Italian stock, the Latins, Oscans, and Hellenes. “Avat I have already noticed. Nenia and  dituus are Phrygian, but not really Greek. Soccus is Armenian , Phrygian, Greek, and Latin. “EXeyo¢e is Greek, Armenian, Phrygian, and probably also Lydian. The roots of soccus and — EAXeyoc are Armenian. The Phrygian words, mater and ga,  are Aryan generally. From these affinities we may begin to  perceive the position which the Thracian race once held in Europe, and its influence upon the other inhabitants of Greece  and Italy. - ~ _-iThe Phrygian language is succeeded by the Lydian) which 38 THE ARMENIAN ORIGIN OF THE ETRUSCANS.

represents, in addition to itself, and in conjunction with the Phrygian, the Mysian language also, which is described as  puEoAbsiov cai wEopptyiov. Some of the words transmitted  as Lydian are of a doubtful character, and their interpretation  is not always obvious. Others, on the eontrary, are of peculiar  weight in an ethuological disquisition ( Two may he especially  mentioned, xavdadAn¢ and ravdobptov ; both of which, besides  being Armenian, and one of them Assyrian, appear to have  travelled far westward into Europe, where the first may be  recognised in Rhetian Switzerland, as well as in Albania, and  the second in Spain, a country still more remote from Assyria  and Armenia. Xdpde is another Lydian word which may  deserve more particular attention, as it intimates that there  was an ancient form of the Armenian language still closer to  the Lydian than the literary Armenian is, which, although it  is the oldest form we possess, only extends back abont 1500  years.) : Lydian Words. 1. ’Ayxwv, ‘refyoug ywvla.” Arm. ankiva, ‘corer, angle. Gr. ayxwv. Lat. angulus. Germ. winkel.

2. “AxvAor, ‘ Badravov rpwivnv.” Arm. hatin, ‘an acorn? Germ. erchel. Liat. galla. Pers. guliék, ‘ pease, a ball.’ 3. “Apgtravov, ‘ dicxoc.” Rhet.-Rom. arfidar; Pied.  arfude ; ‘to reject.’ Lat. repudio. Macedon. dpote, ‘inde.’ Arm. phovthal, ‘to hasten’ (inu); ar, ‘to’; ar-achel, ‘to  despatch’ (ach, ‘aleg’) ; arkanel,‘ to throw ;’ ar-arkanel, ‘ to  oppose.’ <Ar-phovthal, therefore, = ‘ accelerare.’ 4. “ Hesychius: ’AorpaXlay rov Opaxa Aveo’. Sequitur:  aorpaddc 6 Wapoe bd Oerraddv. Thraces a loquacitate et  garrulitate vel quod lingua uterentur Lydis non intellecta  sturnaceos appellatos crediderim. Pers. béstarak, ‘a starling.’” Arm. sareak, tarm, ‘ astarling ;’ sarel, ‘to ery;’ sarot, © crying.’ Alb. tserlé, ‘a blackbird.’ 5. “Arradog, ‘mata. Arm. hateal, ‘cut’ (cf. xdppa); THE ARMENIAN ORIGIN OF THE ETRUSCANS. 39 énd-hatel, ‘to interrupt, to cease,’ pres. and past part. énd-  hatot and énd-hateal. Lapp. ajetet, ‘ detinere ;? ayeteje, ‘ mora-  tor.’ | 6. “ Attis, ‘scitulus.’? Vide ii. (¢. e. Phryg.) 7. Quum ex Hesychio glossa ceteroquin sat obscura id eluceat nomen etiam Hattes pronunciatum fuisse, Arm. at, ‘sectio, pars, granum,’  hatanel, ‘secare, perrumpere, decidere,’ Aatanil, ‘ secari, cessare,  micare, desperare,’ in auxilium voco.” 7. “© Hesychius: BaSuppnydAn lerivec td Avowy. Quum  nos avem aliquem rapacem nominemus enéenstésser, inesse  putarim Pers. bath vel bat, Arm. bad, ‘anas.’ De parte altera  nihil definio, licet BonyaAn emendare suadeam.” The termi-  nation of this Lydian word may be explained from the Arm.  arag, ‘swift ;? aragil, ‘a stork :’ Etrusc, araeus, ‘itoat.’ See  onf. in Etruse. s. v. aracua.

8. “ Hesychius : Baxxapic pbpov roby ard Boravne éuwvd-  prov, Error 8 awd pupatvne, GAXor 82 pbpov Abdiov. “Ears 82 cat Enpov Sidracpa ro awd rie piénc. Pollux vn. 104: pbpa Heecav cal Baxxapev cal audpaxov Kal ipiwov.” Pers. bukhir, ‘perfume, odour.’ Arab. bukhdr, ‘ vapour.’ Lat. acca. Gael. dachar,‘an acorn.’ Arm. dbaklay ; Pers. bakhlah; Arab.  bdkidy ;‘a bean.’ Pers. dakhkalah, ‘a walnut.’ 9. “Hesychius: Bacavrarine AlSo0¢ otrw Afyerat AvoKOd NiSou yévog cat ALSo¢g Bacavog, w waparplBovrec rb xpvalov 2oxfuaZov. Sansk. pdsdiia (‘lapis ’).’ Arm. yesan, ‘a grind:  stone,’ = Pers. asydnah, = Sansk. ¢diia. | 10. Bacdpa. See inf. in Thrac. s. v. bassara.

11. “ Hesychius: Bdoxe wixpoArA£éa rAnolov teddaZe dv- Storl et BaariZa KpdAea Yacaov Epxov Avorori. Utrumque  corruptum. Sansk. vdga, ‘festinatio.”’ Arm. wazel, ‘to  hasten.’ Arab. daskay, ‘ going fast.’ For mxpodéa, see ante  in Phryg. 8, v. wiéptov.

12. Bpévov, ‘ubpov. Arm. dovrel, ‘to exhale, to smell ;’  bovrovmn, ‘odour. Rhet.-Rom. brainéa, ‘mist.’ Bng. breath.

13. Bolyee, €2debSepor.’ Arm. phrkel, ‘ to deliver.’ Eng. free. 40 THE ARMENIAN ORIGIN OF THE ETRUSCANS. 14. “Festus viii. p. 99: ‘helvacea genus ornamenti lydu  dictum a colore boum, qui est inter rufum et album appellatur-  que helvus.’? Non est lydicum quia si esset per Z inciperet, cf. Sansk. harit, Lith. zalas, Pers. zird (‘viridis’).” Esth. had, ‘grey ;’ haljas, ‘ green, bright, shining.’ Arm. atov, atovakan, ‘soft, tender, delicate ;’ atdét, ‘feeble, indistinct, obscure, less  bright ;’ atovés, ‘a fox.’ 15. “ Hesychius: iBpe (read iu) revie rd Boav, of Sd? rd  mov. “Eort 68 Avéwy.” See if. in Scyth. s. v. ae, and  ante in Phryg. 8. v. Bayaioc.

16. "IuBove. See inf. in Thrac. s. v. féuBpov.

17. “Iwmt, ‘Sevpo. Arm. hovp,‘near;’ hovp linel, ‘to ap-  proach.’ 18. KavéabAne, ‘ oxvAAorvierne, kuvayxn. “Arm. khendel, ‘avlyev,’ et khendot (otherwise khendavt and khendét), ‘ xvi-  ywv.’”  Aucher gives shetdel, instead of khendel, as does also  the great Arm. and Lat. Dict.; but 7 represents in the kin-  dred Asiatic languages (Arica, pp. 90,91. See also Rawlinson, Asiat, Journ. Behist. Inscript. p. 34). Alb. kyendis, ‘I choke.’ Rhet.-Rom. candarials, ‘a disease of the glands, which severely  oppresses the breathing’ (eine Art Driiseniibel, das das Ath-  men sehr erschwert). By this word, the languages of Arme-  nia, Lydia, Illyria, and Rhetia are connected together. The Arm. participial termination, avi, dt, or of, may also be recog-  nised in xavd-abA(ne). Similar forms in /, so characteristic of  the Armenian and Etruscan languages, may not improbably  be discerned in the Lydian words, dorp-aA(la), arr-aX(oe), BaSuppny-aX(n), and wicp-od (éa). There is another derivation  of xavdabAnc from the Sansk. ¢gvan, ‘ canis,’ and adhd, ‘ agitare,  vexare.’ | 19. “ Athensus xu. p.516 CD: xavdavAdv ria 2Aeyov of Avéol ovx Eva adAd tpEic, ottwe tEhaknvro wpd¢ Tac HouTa- Selac. TlvecSac Sairév gpnow 6 rapavrivog “Hyhotrmog 2 Epov kptwe Kal kynorov dprov kal dpuylou rupov avyYov re kat Cwpov miovoc. Pollux vi. 69: ein Say wpoojxov roicg 7d0- THE ARMENIAN ORIGIN OF THE ETRUSCANS. 4!  cpact cat 6 KavdvAog 2& apubdov kal rupov cal yaAaxrog Kat  pédtrog.” Perhaps from the same root as xavdatAn¢: or the Arm. sandal, ‘to desire greatly,’ might be suggested.

20. “ Athenzus xil. p. 516 C: wpwrot Avdol rv capixny &Ecvpov.. Suidas: capv«n ieoua ee rwoAAwy ovyxeluevwy. Bpwpa Abstov 2% aiparog cal adAAwy eeouarwy . . . Kal  kapuKo7rolsiy TO Koopeiv mola rive pnuarwy Tov Adyov. — Sansk. fri, ‘xepav,’ unde Sansk. farbura et Arm. kharn, ‘ varius,’ ”” 21. “Stephanus Byzantinus: KaorwAove Awpueic of Avdol  gactv. Augurer nomen litoris accolas significare.” Arm.  kovsht, koys, ‘side.’ Lat. costa, Eng. coast. The termination -wrdé¢ might be explained as in xavéd-abAne.

22. Koadadktv, ‘ BacAéa. “ Quum habeamus apud eundem (Hesychium) xoaXseiv, ‘BdpBapov ESvoc,’ et xdadrol, ‘Bap- Bapo.,’ si notissimi lius sebvog, ‘BactAedbc,’ memineremus, ‘populi regem’ interpretari possnmus.” Gael. cuadlas; Arm. Zotow ; ‘an assembly ;’ Arm. Zotowovrd, ‘people, multitude.’ Osset. koar, ‘company, multitude.’ Pers. galah, ‘a crowd.’ Gael. dion. ‘presidium;’ dein, ‘fortis.’ Pers. tanu, ‘power.’ Arab. din, ‘faith, religion, decree, empire, king.’ Arm. den, ‘faith, religion.” Arm. ateane‘a tribunal, a ma-  gistrate.’} 23. AdBpuv, ‘wéAexuv.” Lapp. Jadtet, ‘findere;’ labmet, ‘verberare.’ Pers. Jad, ‘a blow.’ Arab. /adi, ‘ striking.’ 24. Aafdac, ‘6 répavvog, 6 py & yévoug répavvoe.’ Hin-  dustani /é/d¢,‘a master.’ Pers. /dlak,‘acrown.’ Arm. /av, ‘ good, fine, better.’ Esth. /dila, ‘ bad, severe.’ Gr. aiAay. Arm. Jikel, ‘to vex, torment.’ Sansk. Jul, ‘agitare, per-  turbare.’ 25. MavAornptov, ‘rAbdtov A€utopa Aerrdyv ti.’ “ Varie  emendaverunt yduopua, Aérioua, pédAcopa.” Arm. matzmay, ‘a plate.’ 1 Compare Gael. basal, ‘a judge,’ with Baoeds. 42 THE ARMENIAN ORIGIN OF THE ETRUSCANS. 28. Mnoeic, ‘6 Zedc.’ “ Sansk. medhas, in Vedis dei excelsi  titulus, medhira, ‘ sapiens.’”? Arm. mtazi, mtavor, ‘ sapiens.’ 27. Muade, ‘d&in.? Georg. mukhi, ‘an oak.’ Arm. mogay, ‘a tamarisk,’ = Gr. uupfen. Kurd. mitéh, ‘a box-tree.’? The  name of J/ysia was supposed to be derived from pwadc. If we  may trust Homer, the tamarisk was common in the Troad. " See JZ. vi. 89; x. 466; xxi, 18, 350.

28. Madaé, ‘el8oc olvov.’ Gipsy mol, ‘wine.’ Pers. mui, “wine.” Arm. molt, ‘ intoxicated.’ 29. Mwic, ‘4 yin.’ Esth. ma, ‘earth, land.’ Lapp. mdiwe, ‘pulvis.’ Georg. miza, ‘earth.’ Lespi, misa, musea, ‘ earth.’ 80. “ Photius s. v. Néugat: cal ai Movoac 82 tre Avdoy  viuga. Stephanus Byz. 8. v. TéppnBoc: g0oyyiicg Nupgaov  axobaac, ac cat Moboacg Avool xaAover. Utram verum alii  dicant.” Cf. Arm. zovag, ‘a song;’ noval, ‘to mew;’ Pers.  nutwd, ‘voice, modulation ;’ nuedg, ‘singer, musician.’ 31. Néypua, ‘dvedoc.”’ Arm. nakhat, ‘reproach, disgrace.’ Pers. ndkas, ‘ worthless, base.’ Arab. nakdh, ‘ contempt.’ $2. TdApue, ‘ Baoirsdbc.’ See ante in Phryg. s. v. Barhy.

88. “ Tlavdobprov, ‘Abdioy Spyavov ywplic -wAf«rpov Waddo-  pévov. Pollux iv. 60: rpixopdov "Acobpio: ravdotpayv wrdpa- Zov. Arm. phandirn. Osset. fandur. Ital. mandora.” Phan-  dirn signifies ‘a trumpet.’ The Arm. has also bandirn or  bambirn or phambirn, ‘ castanets,’ and bamb, ‘thorough-base.’ Span. pandero, ‘a tabor,’ pandorga, ‘a concert.’ 34. THapauhyn, ‘9 trav Seov potpa.”’ “ Sansk. parimd, ‘destinare, decernere;’ subst. parimdfia. Etiam apud Ar-  menos para quod Sansk. part; sic (Sansk.) partdhé = (Arm.)  paraditel et repipépea = (Arm.) paraberovthiwn.” Arm. hramayel, ‘to ordain ;’ hraman, ‘a decree,’ = Pers. farmdn, = Sansk.  pramdfa.

85. “Servius 4m. x. 179. Alii incolas ejus oppidi Teutas  fuisse et ipsum oppidum Teutam nominatum, quod postea Pisas Lydi lingua sua lunarem (al. singularem) portum sig-  nificare dixerunt, quare huic urbi a portu dun@ nomen im- | wad THE ARMENIAN ORIGIN OF THE ETRUSCANS. 43  positum.” If we read pis-@, the Turkish would give ay, ‘luna;’  and if we read pi-sa, the Mantschu would give pia, ‘luna,’ and  also sata, ‘lebes,’ = Lapp. saya,= Arm. san. Cf. Germ. ha/fen, ‘pot, haven, port.’ The Arm. has—apé, ‘the hollow of the  hand ;’ apn, ‘ashore ;’ and apheé or aphseay, ‘a bowl ;’ per-  haps = pzs-. The Finnish or Turanian element in Asia Minor,  which seems to be discerned beneath the Thracian in Phrygia  and Lydia, will appear more plainly beyond the Taurus in Lycia. The Lydians of Servius are evidently the Etruscans.

36. “Johannes Laurentius Lydus mens. 3, 14: véov capdev  ro véov Erog Ere cal vuv AbyeaIat rH WAHIE avvoporoyeirac’  siot S62, of dace ry Avdwy apyxalg gwvg roy tvavrov KaAsio3at  cdpo.” Botticher notices (p. 26) that sard would once have  signified ‘year’ in Arm., as the ancient name of what was  formerly the first month in the Armenian year was Navasard, It nearly corresponded to our August, and in all probability  signified ‘ New-year,’ Nav-sard. In Arm., dar and éart signify ‘age’ and ‘ year,’ the sibilant in the more ancient form sard,  having apparently been hardened. Dar and sar have still the  same meaning in Arm., #.¢, ‘height.’ The Chaldwan sar (sdpoc), in Arm. gar (Euseb. Chron.), contaiing a fabulous  period of 3600 years (ten kings reign 120 sarz), may be the  same word. Compare also Arm. ar, ‘series, rank, chain,’ = Germ. schar, Ital. schiera. “With regard to the final d in sard,  it is one of the three letters, s, d, and 2, which are frequently  added in Arm. Thus we have spand, ‘slaughter,’ instead of  span. In Ossetic, sdérd or sdérde means ‘summer.’ The following are the names of the ancient Armenian  months!:— 1. Navasard (Aug.) Lyd. véog capdie, ‘ véov Frog.’ ; Georg. ori, ‘two,’ Cf. Mantschu ori, 2. Hors (Sept.) { é twenty,’ and Arm. erkov, ‘ two.’ 3. Sahmi (Oct.) . Georg. samt, ‘three.’ 1 I am partly indebted here to an article in the Journal Asiatique  for 1832, vol. x. p. 527. - 44, THE ARMENIAN ORIGIN OF THE ETRUSCANS. Cappad. TipE; Pers. Tir; Georg. Tiris- 4. Tré (Nov.) . | tint; Zend Tistrja; ‘the name of  the fourth month.’ Arm. chat-el, ‘to gather, to amass, to  squeeze, to weed,’———chatZov, ‘ must (of wine).’? The root of Chatoz# ought  to be chat. Chatoz,‘ Vendémiaire.’ (?) Arab. ariz, ‘ frost, hoar-frost ;? whence Arad = Frimaire.

Zend Mithra (Pers. Mihr) ; Cappad. M:Opf ; ‘the name of the 7th month.’ Arm. Mihir, ‘Mithras;’ mehean, ‘a  temple ;’ th. meh. The Arm. HMehe-  kan would not be borrowed from the Zend, but derived from a common L origin with it.

8. Areg (March) Arm. areg, ‘ the sun.’ 9. Ahekan (April) Arm. ahekan,‘on the left hand, sinister.’ Georg. mareli, ‘the name of the 10th  month.’ Arm. marg, ‘a meadow ;’ whence Mar-  gaz = Prairial, 5. Chatoz (Dec.) 6. Araz (Jan.) . 7. Mehekan or Meheki (Feb.) 10. Mareri (May) 11. Margaz (Sune) | Arm. hrovt, ‘burning ;’ hrat, ‘ excessive  heat :’ whence Hrotiz = Thermidor. Of the twelve names of the Armenian months, six may be  known to be genuine Armenian, either by root or form, or by  both—Chatoz, Mehekan, Areg, Ahekan, Marga?, and Hrotiz. Three others, Hori, Sahmi, and Mareri, are Georgian. <Araz  seems Arabic, though its form is Armenian. TZré may be con-  sidered as Persian: and Navasard would be allied to the Lydian, though probably at the same time ancient Armenian.) 87. “Tapyavov, dfo¢* Avéol. 1d raparrov. 7 rd ard 12. Hrotiz (July)  aTeugdbAwy woua. Kat wéa, qj Kat oxopmiovpoc. Heinsius et Salmasius d£o¢ scripserunt.” Pers. tarkhwdnah, ‘thick pot- THE ARMENIAN ORIGIN OF THE ETRUSCANS. 45  tage ;’ tarkhan, ‘milk soured ;’ ¢arkh,‘an orange.’ Arab.  darrak, ‘treacle, wine’? sth. farri, ‘ grapes, berries.’ Arm.  taraz, ‘agaric.’ Pers. tarsdn, ‘timid, fearful.’ Arm. fartam, ‘ timid, perplexed.’ i 38.- Teyouv, ‘Ayorhv.’ Arm. é&, ‘a spear.’ Esth. Zagi, ‘a  pike, a sword.’ Teyovy, like datro, might signify ‘ swordsman.’  or ‘soldier,’ as well as ‘ robber.’  a The affinities of the Lydians appear from their language to  be the same as those of the Phrygians. Both nations may be  classed in the same family as the Armenians, and the Aryan  family next in order of relationship would be the Persian. I  cannot perceive that near affinity between the Lydian and Greek languages which Mr. Rawlinson’s ethnological system  would require.) There is, indeed, the word ayxwy given as Lydian: but the root is common to many languages, the ter-  mination goes for little in a vocabulary, the reporter is a Greek,  and the Arm. ankivn, ‘ayxwv,’ would be written ayxéy in Greek. If ‘the Lydians must have spoken a language closely akin to  the Pelasgic,’ which ‘ripened ‘ultimately into the Hellenic,’  we could hardly fail to observe more decided Greek affinities  in thirty-eight Lydian words than we are able to discover.

( As the Carians are so nearly connected with the Mysians  and Lydians, the few words which have been preserved in the Carian language may as well be added here: Carian Words. 1. "AXa, ‘tog.’ Gael. al, ‘a horse.’ Alb. ala, ‘ quick.’ Arm. wat, ‘quick.’ Arab. walus, ‘ going quick’ (a camel).

2. Bavda, ‘ vfcn.’? Arm. wan-el, ‘to conquer.’ For the -8.,  compare span-anel, ‘to kill,’ and spand, ‘slaughter.’ Pers,  wand, * praise.’ 8. Téda, ‘Bacrebe.’ Arm. get, ‘beauty ;’ getant, ‘ fair,  good.’ Cf. Gr. yeAéw and yeAfovrec Arm, chett, ‘ temo,  gubernaculum.’ 46 THE ARMENIAN ORIGIN OF THE ETRUSCANS, 4. Ileoa, ‘XlSoc.’ Arm. hig, ‘ marble.’ Ticoa appeared in Movd-yiooa, the name of a Carian town. Cf. Arm. moyn, ‘beauty.’ Gissa was an Illyrian town, Cissa-a Thracian town,  and Cissus a Macedonian mountain.

5. KodraBpropude, ‘ Spqxiov Spynua wai xapixdy.’ See inf. in Thrac. s. v.

6. “ Machon apud Atbenzum xii. p. 580 D: Kb3S Eby rprw3drov.  tle 8 ovmirphbwv orl cor, onoiv, radav éuta y ty AShvac xapixoic xpiovat oradpoic ; Forsan cognatum xamwf{3n: v. Persica.” “ Xenophon Aad. i. 5, 6: rpiacSat ov« Fy ci py ev Ty Avdla  ayopa tv ry Kbpou BapBapmy rv Karis nv adsipwr 7 GAdirwv  rerrapwv afyAwv. ‘O o8 ofyAoe Séivarat irrd dBoAo0d¢ Kat  jyuodALoy arreode, 4 O& camlSn dbo yolviKcac arrikag Exwpet. Polysenus iv. 8, 82: 4 xawéric tort yotmE arruh. Pollux iv. 168 pérpwv dvduara recensens: cam {Sn we Zevogwr.” Arm.  kapig, gen. kapgi,‘a kind of measure (kadBoc, xoivt, youve,  congius), the socket of the eye ;’ th. sap, = Lat. cap(ere) ; kapel, ‘to fasten ;’? Aapovt, ‘plunder;’ kaptel, ‘to plunder.’ Heb.  kab. _ , 7. Sova, ‘ragoc.” Arm. sovzanel, ‘ condere, tegere,’ th.  sovz; zZov, ‘a ceiling, a roof (fectum).’ 8. TovoabAo, ‘ Wuypator. Osset. tyiisiil, ‘little, = Arm.  doyzn. Arm. thzovk, ‘a pygmy,’ th. thiz, ‘a span.’ KdrrovZa, ‘adrt¢ Opaxne, gv yg Karwxouv of IIvypaio.? Arm. kay, ‘dwelling.’ Pers. kad, ‘house.’ Osset. ghau, gau, ‘ village.’ 9. “Stephanus: Tupvnocdbc wédtc Kaptag ard rupvnscod  paBdov. avd yap thy paBdov ruuvlav Abyovow.” Arm.  tophel, ‘riwrav.’ Eng. thump. Arm. Zovp, ‘stick, rod.’ The Lycians and the Caucasian Nations. The singular resemblance of the Carian rovesbA(o1) to the Ossetic Zyisiil (where ¢y is a single letter) raises a question of  considerable interest, which calls here for a digression: The THE ARMENIAN ORIGIN OF THE ETRUSCANS. 47  language of the Lycians, the neighbours of the Carians, cannot  be explained from the Armenian, especially in its structure,  where in the affluence of vowels the Lycian resembles the Zend, and presents a remarkable contrast to the Armenian. No langnage, however, of any kind, has hitherto been found  to explain the Lycian. In this failure of all ordinary languages,  and especially of the Armenian, we may be led, from our find-  ing the Osset. ¢yi#s#l, rather than the Arm. doyzn, thiz, or  thzovk, in the rovacbdo¢ of Caria, a country bordering on Lycia, to enquire whether the Lycian language was allied to  the Ossetic, or to any other language used in the Caucasus, The Ossetic vocabulary, as we possess it, is not very extensive ;  but the grammar is well known, as we have an excellent treatise  on the language in Sjogren’s Ossetische Spracklehre.

The Ossetes are commonly considered, but not incontestably  admitted, to be a branch of the Aryan stock. The rest of the Caucasus is occupied, with the exception of some Turkish  settlers round Mount Elbruz, by the Caucasian race. This is  divided into three branches, differing greatly from each other,  and separated by dialects into many subdivisions. The West- Caucasians possess about one-half of the chain: they are  divided into Circassians and Abasians. The Zychi and Achas  of the ancients, two tribes who dwelt in this district on the Euxine, have been supposed to derive their names from the Circassian zug. or dsyg, and the Abasian agu, both signifying ‘man,’ Next in order come the Ossetes, who occupy a small  tract on both sides of the chain, in its centre. The sources of  the rivers Terek and Aragua, and the intervening great Cau-  casian pass of Dariel, lie in their country. The two divisions  of the Osset¢es, in language, are the Digort and Tagauri. These  names seem to be found in those of the Jssi and Tagori, two Scythian tribes to the east of the Tanais, mentioned by Pliny. The root of the name Jssi may be the Osset. osse, ‘ woman,’ or  rather the Lesgi oss, ‘people.’ In Osset., ossethad means ‘women.’ In the same parts Ptolemy mentions the Tusci:  bt {. AY 48 THE ARMENIAN ORIGIN OF THE ETRUSCANS:  and on the central ridge of the Caucasus, within twenty miles  of the Ossetes, and bordering on the Lesgi, who will be men-  tioned presently, we find a little tribe called Zuschi, belonging  to the Middle-Caucasian race, the Mizdschegi or Kisti. The East-Caucasian race, the Lesgi, occupy a large portion of the  ancient Albania, where, according to Strabo, twenty-six  languages were spoken: the dialects of the Lesgi are nearly  half as many. These Lesgi, called by the Georgians Lekhethi,  and by the Armenians Lékch, are the Leges or Lege of an-  tiquity ; and the root of their name would probably be found  in the Osset. /ag or dag, or the Lesgi /es,‘man.’! Glossaries  of the Caucasian dialects, necessarily scanty, have been formed  by Klaproth. Specimens are also given in an Account of the Caucasian Nations (London, 1788) .? / Between the Caucasians and Ossetes on the north, and the Armenians on the south, lies the Iberian or Georgian race,  divided into Georgians, Mingrelians, Lazi, and Suani. The  last two names are ancient, and the Lazi are probably the ‘men.’ The Suani connect the Iberians with the West-Cau-  easians, and the Tuschi and two other little tribes are con-  sidered to be Iberian as well as Middle-Caucasian. A tolerably  copious Georgian vocabulary has been published by Klap-  roth. — | The country occupied by the Caucasian nations has a length  of about 600, and a breadth of about 100 miles. The area  would be equal to that of England. The Armenian area  would be about the same, and the Georgian nearly that of Ireland. ‘ 1 So the meaning of the name Mardi: is given by the Arm. and Pers.  mard, ‘man,’ 4.e. Bporés: Osset. mard, ‘ death.’ 2 Can the name of the Ossetes, if=ossethd, ‘women,’ in any way ex-  plain the story of the Amazons, who are placed nearly in the country of  the Ossetes (Strabo, p. 503)? The fabulous Amazons in Asia Minor,  who are said to have founded Ephesus, Smyrna, and other towns, might  then be the mythic representatives of the Caucasians in that country. Some of the Amazonian traditions are noticed above, p. 7, note 4. THE.ARMENIAN ORIGIN OF THE ETRUSCANS, 49 Lycian inscriptions. The Lycian inscriptions have been elucidated by Mr. Sharpe  in Sir C. Fellows’ Zycia and in Spratt and Forbes’ Lycia, by Grotefend in the Zeitschrift fiir die Kunde des Morgeniandes, v  iv., and by Lassen in the Zeitschrift der Deutschen morgenlan-  dischen Gesellschaft, v. x. A resemblance has been traced  between the Lycian and the Zend, but Lassen considers that  the two languages have no intimate affinity (innigere Ver-  wandschaft). This limits the field of enquiry towards the  east, and the obviously un-Semitic character of the Lycian  will exclude the Syriac and other cognate languages. There  remain, in the immediate neighbourhood of Lycia, on the west  the Greek language, and on the north of Greece and Lycia  various Thracian dialects. As neither Thracian, 7.¢., Ar-  menian, nor Greek will explain the Lycian, and as no Aryan ‘European, as well as Asiatic, language seems likely to do so,  and as on the south of the Aryan country we fall into Semitic |  dialects, it is therefore to the north of the Aryans that we are  induced to look for the ancestors of the Lycians.

The data for the determination of the Lycian language are  singularly good. We have three bilingual, as well as a number  of other epitaphs. These contain words of the greatest im-  portance, such as terms of relationship, besides a variety of  expressions for ‘tomb.’ In the three bilingual ‘epitaphs,  which here follow, the Lycian epitaph is in the original  completely given, and then followed in like manner by the Greek. I have altered this arrangement for the sake of com-  parison. I. Gwéeya  Grafazeya méte prinafatu . Sederéya . TO puna TOOE eTroltnaaro LSaprog  this tomb here made - Sidarius Pé...neuw  tedéeme urppe Eétle Euwe sé lade Tlappevrog. vlog . EauTWwt Kae THE yUvatke  of Parmens - theson for’ self his and wife 50 THE ARMENIAN ORIGIN OF THE ETRUSCANS. éuwe sé tedéeme P.é. liyé. Kat ULtwe TluBcadne.  his and son Pybiales. Il. Gouinu tatu méné  prinafilu Polénida TOUTO = To fev nya epyacavro §=—. AroAAw dnc  this tomb here made Apollonides Mollewéséu = 8S Lapara Polénidau Porewemétéu MoAXtstog = kat «Ss Aarrapag = AroAAwidou Tlvpiuarioc  of Mollises, and lLaparas of Apollonides, of Purimates  prinézey€we urppe lada Epttéwié = 86 —stedéemé OLKELOL emt Taicyuvakiy ratc eavTwy Kat TOLC EyyovoLC  the domestics, for wives their and children.  sé eyé—ites&~—sretideteé = tatu ss Ewéwe 8 méeyé Kat av TIC aducnont To pynua Touro  and if anyone injures tomb this here (oéte ponamakke adadawa(l)e ada 4) (cEwAea Kat ravwrea ELN aUTWt TavTwv) The Lycian and Greek in brackets do not correspond: We  find in other epitaphs—the references are to Fellows’ Lycia—  sé eyé itadu tése meite adadawéle ada 1 (p. 482).

(and if buries any one here let him pay adas two).  and—  sé eyé itatadu meite adadéwale ada O—(p. 483).  and—  sé eyé itatitu tése meite adadawale ada 111. (p. 486).

There is a fac-simile of the third bilingual epitaph in the Lycian room in the British Museum. In the only part which  is of additional importance, the Greek does not enable us to  arrive at a literal translation of the Lycian. What I have given  as such is therefore partly conjectural. III. Gouinu prinufo méte prinafatu -utta ulan TOUTO TO pLynjia npyasaro . Iraohda .  this tomb here made -utta of Ula  tedéeme — urppe lade Guwe AvripedXrng . QuTWwt Kal = yuvate  the son, an Antiphellitan, for self and wife his THE ARMENIAN ORIGIN OF THE ETRUSCANS. 51  sé tedéemé éuweé 386 eyé teéde tehké mutu Kat TEKVOIC tay de TiC  and children his. And _ if -any one makes a purchase  méné wastto une ulawe  éweyéue séeéarefeu.téze  adiknone 7 ayopacnt To phynua n Anrw aUTO. ETITY.  here, injures or tomb this, The words in the preceding inscriptions which are most im-  portant ethnologically are obviously the words of relationship,  tedéeme, ‘son,’ and Jade, ‘ wife,’ which are neither Aryan nor Semitic. They may, I think, be shown to be Caucasian; and Caucasian terms for ‘ wife’ and ‘son’ seem also to be found in Armenian by the side of the Aryan words, kim, ‘ yuvi,’ and  ordi, ‘putra, puer” The chief difficulty in the comparison lies  in constructing ‘the perfect form of the. Caucasian word from  so many different dialects. I have therefore not merely cited  the Caucasian terms, but also analysed them, in the more im-  portant comparisons of the following list— Ewéeya, Ewéwe, Eweyéwe, Ewuinu, ‘this.’ The first term is  supposed to be a fem. acc. case, and the fourth a neut. acc. Osset. ay, ‘ this’ (nom.), ay, ayyt, ayyey, (gen. and acc. : there are  no genders). Lesgi hat, hoi, ua, ‘ this.’ Kisti woye, ‘he.’ Georg.  ese, ‘this, he.’ Kurd. au, ava, ‘this ;’ au, avi, ‘that;’ Alb. ai,  ayu, ‘this, he.’ Pers. Behist. Auwa, ‘ille, hic;’ awa, ‘id.’ érafazeya or arafazeya, ‘tomb.’ The root here appears to be  some word like altus, signifying ‘deep’ and also ‘high.’ The  word will require analysis and illustration— Arab. zamdn (pl. azmdn), ‘ time,  age, world.’  dsmdn, ‘ heaven.’  zamin, ‘country, ground.’ Pers gar = azman, ‘the highest the  heaven.’ jae  ger ———_—_———_—-—_——ezman, ‘a tomb.’ word Arm. 1% e ry ° e ° ‘ up ’  wer E2 52 THE ARMENIAN ORIGIN OF THE ETRUSCANS.  or —$ $$ ___ ——_ ——-mi, ‘a tomb.’  or —_——___—-—_—_—_——mo, ‘a hole, a ditch, a well.’ Georg. ghr ma, * deep.’  ghr tat, . ‘a hole.’ Basque gar ————aiza, ‘height.’ Leagi gwoar ida, . ‘depth.’  retsh —adaa, ‘height.’  urth c ’ Kisti | yurk a } a hole.  trakh-——_——-uua, ‘a hill.’  urk, ‘a hole or deep place Esth. | in a river.’  urg, ©a fox-hole.’ Arm. arpkh-———1, . ‘heaven, sun.’  arv, ‘ heaven.’ Osset. | «0 oe 6 « ee) f deep.’  arf ade, . . . .. ‘depth.’ Arm. wirap, . 2 ee ‘a very deep ditch.’ Lycian Géraf ——azeya, “a tomb.’ The Lycian érafazeya is the acc. sing. The acc., and also  the gen. sing. of the Osset. arfade is arfadiy. Mr. Sharpe  observes that the Lycian acc., both in the sing. and plur.,  invariably ends in a vowel. It is the same in Ossetic, where  the terminations of the gen. and acc., in both numbers are— 1, y, y, ty. The Lycian gen. sing. terminates in uw. But  there are very few Lycian words zot ending in a vowel.  mén’, méeyé, méte, meite, ‘here.’ Osset. mand, ‘here;’ am, ‘here ;’ amzy, ‘in- this, here ;’? ama or amd, ‘at this;’ amey, ‘from this,’ or ‘through this.’ The -¢e in méfe or meite may  be the Osset. 44d, an enclitic like ré or 8 Méné would then -: mind, méeyé = amiy, méte = amd-tha, and meite = amiy-tha.  prinafatu, ‘he made;’ prinafiitu, ‘they made.’ For the  form of the Osset. perfect, we must go to the auxiliary verb,  fa-un, ‘to continue to be, to complete.’ It gives—/adan, Sidi, faziy, festam, festuth, festiiy, ‘fa-i, -isti, -it, -imus, -istis, -erunt.’ ‘rat,’ from um, ‘ esse,’ is udiy. A verb like /tssiin, ‘scribere,’ makes /itista, ‘ scribebat ;’ fiistoy, ‘scribebant;’ and THE ARMENIAN ORIGIN OF THE ETRUSCANS. 53 Just faziy, ‘scripsit.? The form of /aziy is not unlike that of  prinafatu. Aswe have prina-fatu,‘ made ;’ prinu-fo and prina- Ju, ‘tomb,’ and prin-ézeyéwe, ‘ domestics,’ with the singular  priméze in another. place, the root prin would be some word  like déu(w) or dSdu(oc). It may, perhaps, be found in the Arm. wran, ‘a tabernacle.’ For the change of p- into w- (v),  compare Arm. wrip-ak, ‘prav-us,’ and wat=Eng. and Pers:  bad. We may here compare Lycian and Ossetic forms :— - (prin-éze Lycian {rine (nom. plur.) ; éraf-azeya (acc. sing.)  arf-ade (nom. sing.) | Osset. arf-adiy (acc. sing.)  eset arf-adethé (nom. plur.)  arf-adethiy (acc. plur.) | |  tedéeme, ‘ son, child.’ The following words all signify ‘son, ‘child,’ or ‘ boy,’ in their respective languages. The Arm. ?  is replaced by £4/, the orthography being bh here phonetic :— Arm. ¢t ——_——_khl - ay  dse  tzu — {————4 Georg. shyi—-——1 —____1  tsh ———-k—- Y Osset. sve ———l/—____—_—_0-_-———_- Abas, {#k-——# =u ——— $1 ——_—_tskh ———u ————-n ( 881-————— 88 ——_—-u Circas. je —adl a  tshva-—— ay , kh —_ } mir . {——_—_—_ i—. mir Lesgt t -——_——_——_- a mat é ———____f 4-1 —— mat Lycian te——_—_—_dé ——_—__e —-—__me Basque 9 8————_—————_-e — ——__-me¢(a) Pers. 2&——dah For the variation between the Lesgi ¢imr and the Lycian 54 THE ARMENIAN ORIGIN OF THE ETRUSCANS.

tedéeme, compare the Lesgi forms, Aélim, hili, chlim, ‘ water.” In chlim, the ¢ is lost and the final m preserved, nearly as ¢vmr  has lost the d of ¢edéeme, but retained a final 7. The reverse  takes place in ti. So again, Osset. fiid = Arm. hayr, for both = Lat. pater, Fr. pére.  urppe, universally considered to mean ‘for.’ Arm. ar, ar 72, ‘for lLesgizar, ‘of’? Georg. era, ‘because.’ ‘Lassen com-  pares the termination of: ur-ppe with xem-pe. Cf. Arm. péss ‘as, like:’ orpés, ‘as according to, on purpose’ (lit. ‘ which-  like’) ; hipés, ‘as’ (lit. ‘ this-like ’).

étle, atle, ‘self. Sharpe and‘ Lassen instance the Sansk.  dtman, ‘self,’ but the Lapp. eés, ‘ self,’ and the Alb. vedé, ‘self,’  may be nearer. The Lapp. has also another form, deficient in  the nom.; for the gen. of e¢s, in the third person, is eéjes or  alde, and the dat. etje-sas or alla-sas. tle and atle would be Intermediate to etye and alla.

Euwe, ‘his,’ Osset. uy, oy, ye, ‘he, his.’ Kisti woye, ‘he.’ Lesgi wa, ‘this.’ Alb. ai, ‘he.’ Fue seems to be the Lycian  for ‘he’. or ‘him’ (Fellows, p. 486). Pers. Behist. awahyd, ‘ istius.’ | épttéwe, ‘their.’ Osset. udatthé, yethd, ‘they; udatthiy, ‘of them.’ Yethd, ‘they,’ bears to ye, ‘he,’ nearly the same  relation that épité(we) does to éu(we), if we. suppose the « in éuwe to be hardened into p. Alb. aéa, ‘ they.’  sé,‘and.’ Georg. da, ‘and;’ ke, ‘also.’  Lesgi gi, ‘and.’ Lapp. ja, ‘and.’ Alb.e, ‘and? Arm, ev, ‘ and, Arm. e ——— khla r; ‘ wife,’ Abas. lkha dza Kisti xzye Uk,  t lay Circas. 1: —_ leh oes eee ew ‘ husband.’ Go Wt oo. eee,  le 8s L Csgt {, _ Pya ——— di , ° THE ARMENIAN ORIGIN OF THE ETRUSCANS. 55 For the initial ¢, which distinguishes the Lesgi ¢/’yadi from  the Lyeian Jade, compare Circas. lay, Play, ‘flesh; day,  vhlay, ‘blood.’ Terms possibly akin to those above may be  the Osset. dug, ‘man,’ and /appu and /atu, ‘ lad,’ apparently = Esth. dats, laps, ‘ child.’  ttatu, ‘a tomb.’ Lesgi ¢ataul, ‘a ditch.’ Circas. tyésha,  tshitoga, ‘a ditch’ (graben, 1.¢., grave). The Lycian has two  more words for ‘tomb,’ gopu and goru. These would be the  same as the Arm. gov and shor, both signifying ‘ ditch,  hollow.’ Gopu would also be the Georg. éubo, ‘tomb,’ and  the Sansk. /épa, ‘a ditch.’  eyé, ‘if? Arm. ethé, ‘if Gr. ci. Osset. te, aviy, ‘or.’ Lesgi iva, ¢ or?’  tesé, tése, teéde, ‘ric.’ Osset. ty2, tyizt, ‘who?’ In the three extracts between the second and third of the  bilingual epitaphs, there are some instructive points. The  verb i¢adu is rendered by Mr. Sharpe ‘ burieg ;’ t¢atadu,‘ allows  to bury ;’ and ttatdtw, ‘let’ or ‘lets bury,? The grammar of  the Caucasian languages is too little known, and the languages  themselves probably too barbarous, to allow of any comparison  with them. The Lycian forms, however, have considerable  resemblance to those of Lapponic verbs. Compare— . Lycians Lapponie.  ttadu, ‘buries’ ...... Yodyo, ‘ stat.’  itatadu, ‘allows to bury” . todtjato, ‘ substitit, stare facit.’ . ‘lets bury.’ tjodtjota, ) ‘stare permittit.’  ttatiu, { ‘let bury.’ i { tjodtjoteh, ) ‘ stare permittunt.’ The word ada is explained by Mr. Sharpe from the Arab.  ada, ‘ payment,’ which seems a very good derivation. It might  also be connected with the Lapp. wadja, ‘pretium,’ hadde, ‘pretium,’ addo, ‘ultio.’ <Adadawé fle is regarded by the same  writer with apparent justice as a compound of ada and dawéle,  which last he explains from the Arab. ¢awan, ‘a fine.’ We  might also refer it to the Lesgi ¢’e, ‘ give’ (da), tallel, ‘ to  give, without any injury to the sense. The forms, Moll-ewés- 56 THE ARMENIAN ORIGIN OF THE ETRUSCANS. és for MoAA-to-toc, and Por-ewem-étéu for Tvp-iu-artog, seem  to show that dawél might=dal. The -numerous Greek in-  scriptions found in Lycia leave no doubt as to the tenor of the  words adadatéle ada. The termination of II. is now reduced to— .  sé eyé tesé rétideteé itatu éwéwe méeyé oéte ponamakke  and if anyone injures tomb this here .  adadawale ada 4.  he pays adas four. - The th. of rétideteé may perhaps be found in the Arm. arat-  el, ‘to sully, to spoil? The meaning of oéfe ponamakke must  be left to conjecture. It would be consistent with the context  if the words were rendered ‘ fora fine.’ Cf, Lat. ué; or Esth.  ette, Lapp. auta, ‘ pre’ Lat. pena, Esth. pin-ama, ‘ punire’ Esth. maks,= Lapp. makso, = Arm. machs, = Heb. mekhes, ‘ payment, tribute, toll :’ ponamakke, ‘ strafgeld.’ . The termination of ITI., like that of II., is obscure. The  analogies which led me to interpret ¢eké... . éweyéue as I  have done, are the ee  teké, ‘makes’ . . . Lapp. ¢takh-et ; Fin. tek-a@ ; Esth. begg-ema ; ' ‘facere.’ Lapp. takka, ‘ facit.’  mutu, ‘a purchase’. Osset. miizd; Pers. muzd; Sansk. milya ; ‘pretium, merces.’ Bétticher will not  admit pucfd¢ as akin.to miizd.  méné, ‘here’ . . . So interpreted. before. The Osset. amdn, the  dative of ay, ‘this,’ might here give a . better sense, ‘ for this.’  wastto, ‘injures’. . Lapp. waste, ‘ turpis, deformis.’ Lat. vas-  tare. Wastio might also be a noun  governed by ¢eké, and the sense be, ‘ or (does) injury to this tomb.’  une, or? 2.2... Georg. anu, ‘or.’ Osset. inne, ‘ other ;’ . aniu, ‘whether.’ |  ulawe, ‘tomb’. . . Lapp. joud-et, ‘to bury.’ Mantschu oudan, ‘a.ditch.’? This seems to be the sixth THE ARMENIAN ORIGIN OF THE ETRUSCANS. 57 Lycian word for ‘tomb.’ ‘The others  are — érafazeya (Osset. Arm.), ttatu (Lesgi), goru (Arm.), gopu (Georg. Arm. Sansk.), and prinafu. (doubtful). Eweyéue, ‘this’ . . As before. Both ula-we and éweyé-ue may  be datives : ulawe éweyéue, ‘to this tomb.’ This completes the analysis of the three bilingual Lycian  epitaphs, for conjecture seems useless upon the few confused  words which remain in III. There are, however, i in the Lycian  epitaphs which are not bilingual, three words which, as they  would be terms of relationship, ought here to be noticed. We  find in Fellows’ Lyciza the following passages— |  urppe atle éuwe sé wae éuwe (p. 487). ‘for ‘self his ‘and... his  urppe lade Guwe sé tedéené sé undlatededwa (p. 479). ‘for wife his and children and.......  urppe lade Euwe Ofeté Gométeyéu ezemaze sé tedéemé  for wife his Ofeite of Gometeye . . . and children éwoeyé (p. 477). a :  his.  une. Mr. Sharpe refers to the Arab. um, ‘mother.’ We  have Abas. oan, Circas. ana, Lesgi enniu, Hung. anya, and Lapp. edue, all signifying ‘mother.’ In Arm. there is hani, ‘ grandmother,’ and in Mantschu, ounga, ‘ parents.’  uwéla-tedeéwa. This appears justly considered by Mr. Sharpe as a compound, of which the last member is either allied  to tedéeme, or is that word badly copied. _Uwéla he explains  from the Arab. weled, ‘son,’ welad, ‘being born.’ In Lapp.  welja, and in Alb. wela, signify ‘ brother.’ The Lesgi has evel,  evelyad, and ilyud, ‘mother,’ and the compound evel-thmal “would mean literally, in Lesgi, ‘ mother’s child.’ The Lycians  traced their descent through the mother. Perhaps, we may.  render the Lycian word, ‘xaolyvnrou,’ or ‘ nepotes.’ - geemaze. This is rendered by Mr. Sharpe, ‘ daughter,  which is doubtless the most obvious sense, and might bring 58 THE ARMENIAN ORIGIN OF THE ETRUSCANS.

us back to the Basque semea, ‘son,’ the Georg. dse, ‘ son,’ and  the Lesgi tema/, ‘ child.’ But the word might also be explained ‘sister,’ from the Caucasian and Georgian—— Ab thé. . . . . ‘brother.’  khsha . . . » ~ ‘sister.’ Circas. stsht . . . . . ‘brother,’  a. (vasha . . . . . ‘brother,’ Kiots 1 tok » ee « +) Sister.’  vaag «ww Se SC CS brother, boy.’ | yas. . . » . + ‘sister, daughter.’  yasss «ww CS girl.’ Georg { ma) ‘brother’ (dse, ‘son ;’ mama; * Udye ————- a ‘ father.’) ——————m aze One other expression may be noticed. In Fellows, p. 476,  we meet with this epitaph :— Ewuinu gork mute prinafath ésédéplume wrppe lade éuwe This tomb here made Esedeplume for wife his  sé tedésaeme Euweyé woméleyé.  and children his . : The most probable meaning of woméleyé is, I think, ‘dead.’ It may be compared with the Lapp. james, jabmes, sabma, ‘ mortuus ;’ jabm-et, ‘mori ;’ jabmel-et, ‘ cito vel mature mori.’ That Esedeplume did not make the tomb ‘ for himself’ as well,  is rather in favour of‘ his wife and children’ being dead. There  is also in Lapp., dme or dbme, ‘ old;’ dbme peiwe (pewe, ‘ day,  dies’) de defunctis dicitur, quando illoram mentio fit: eg. Nila dime peiwe, ‘Nicolaus defanctus? . Lycian Words. Five words are given by Botticher in his Arica as Lycian. | They are all names of places, and their affinities are sometimes  doubtful :— 1. Kadpeua, ‘ cirov dpuypde.” Gr. xalw. Arm. hizel, ‘to  burn;’ shah, ‘ cooked (meat),’ Suani der; Arm. Zorean ; *alrog.’ Hme would appear from the Lycian coins to be a com- THE ARMENIAN ORIGIN OF THE ETRUSCANS. 59  mon termination in the nomenclature of towns: thus we find Fégssérd-eme, Pegasa or Pedasa, and Jroown-eme, Tros or Tlos. Compare Perg-amus and Berg-omum. The roots of Kadp-eua  would therefore probably be, xa and 8-p, which are very  nearly the Arm. and Suani shah-dier, ‘ roasted corn.” 2. Ilarapa,‘xlorn.” Arm. pat-el, ‘to envelop ;’ pateanch, ‘an envelope, a case, a sheath, a shell ;’ patat, ‘an envelope, a  bale, a bundle ;’ patan, ‘an envelope, a band.’ 8. Ilivapa, ‘ orpoyybda.’ = Arm. padel, ‘ orpayyev.’ (?)—— Arm. puak, ‘a plate (cf. Ital. ¢ondo), a bowl, a pot,’ = Gr.  alvat, = Osset. jfivie,=Georg. pina. Sir C. Fellows supposes  the city of Pinara to have been so called from a remarkable  round crag there, not unlike the Table Mountain in character,  and containing ‘some thousands’ of tombs. He gives a pic-  ture of it in his Lycia, p. 189, which seems to confirm his  opinion. The battle of Issus was fought on a river Psnarua.

4. Téunva, ‘the name of a certain he-goat.? Lesgi ¢eng, ‘a he-goat;’ deon, ‘a ram;’ haiman,‘asheep.’ Abas. shima, ‘a goat.’ Gr. xfuatpa.

5. "YAapog, ‘xaproc.” Esth. will, ‘ fruit” Mantschu oui,  oulana, oulouri, three different kinds of fruit. Georg. khili, ‘an apple.’ In a Lycian inscription (Fellows, p. 479) we find 2+ « « ené oulame fofeto oulame mée toféte tehé. . . We have here again a termination -ame, similar to Kadép-eua,  tedé-eme, &c. May the Lapp. dme, ‘ thing, res,’ be compared ? It is, on the whole, rather difficult to form a judgment upon  the affinities of the Lycians: but my impression would be that  their kindred lay to the north of Armenia and in the Cau-  casus, and that they were, in consequence, rather to be classed  among the Turanians than the Aryans. Yet there is much  difference of opinion as to the classification of the Caucasian  and Iberian races, even among the most profound ethnologists. Rask considers them as Ugrians or Turanians, agreeing in  substance with Klaproth, who is inclined to rank them, or at 60 THE ARMENIAN ORIGIN OF THE ETRUSCANS.

least the Lesgi, with the Fins and Samoyedes. Bopp, on the  other hand, discerns an Aryan affinity in the Iberians and Aba-  sians, while Pott regards the Georgian language as entirely  foreign to the Aryan, although it may have borrowed some Persian words. The Lycians are usually placed without doubt,  as by Mr. Rawlinson, among the Aryans, though the linguistic  proofs seem hardly so decisive as he considers them in favour  of such a conclusion.! There are many names resembling DLyci, and mostly lymg  on the edge of the Thracian area: 1. The Leges of the Cau-  casus, whom Strabo calls Scythians; 2. the Ligyes placed by Zonaras near the Caucasus; 3. the Colchian Zigyes, mentioned  by Eustathius; 4. the Lzgyes of Herodotus, probably the same  as the preceding two, in or near Pontus; 5. the Ligyrit of Thrace; 6. the Jigyrisci of Noricum, the same as the Taurisci ; 7. the Lygit of Tacitus, on the Upper Oder or Vistula ;’ 8. the Ligyes of Italy, the Ligurians. These last I have previously  inferred to be Fins: and many of the other names, at least,  might be Turanian, indications of an early race which may have  been, as it were, swept into corners by the advance of. the Thracians from Armenia to Italy. Lycia in Asia Minor, and. Liguria in Italy, might correspond to Wales, Brittany, or Bis-  cay, rugged districts where the primitive inhabitants of a 1 Herod. v.i. p. 668.

2 Not far from the Lygii lay the Osi, who spoke Pannonian and not German (Tacit. Germ. c. 47). The name Ost resembles Pliny’s Jssi, the Lesgi oss, ‘ people,’ and the name Ossetes. One of the five tribes of the Lygii mentioned by Tacitus were the Aris; a name which seems like the  word Aryan, the Ossetic ir, ‘an Ossete,’ the Arm. ayr, ‘man,’ ari, ‘valiant,’ and several kindred terms. Another tribe was the Elysii,  perhaps derived from the same root as Lygit (Osset. lag, Lesgi les, ‘man’),  and reminding us of the Elisyces,a Ligurian nation between the Rhone  and the Pyrenees. A third Lygian tribe was the Manimi, a name which  might be derived from the Teutonic man, and which resembles in form  the Etrusc. arimus, ‘a monkey.’ The other two Lygian tribes were the Naharvali and the Helvecones, of which the last might perhaps be partly  compared with the Celtic Helv-it and Helv-etsi. The neighbouring Gothins are.said by Tacitus to have spoken Gallic. THE ARMENIAN ORIGIN OF THE ETRUSCANS. 61  country have kept their ground against invaders. The descend-  ants of the first possessors of the countries between the Augean  and the Caspian may have been sheltered by the Caucasus and Taurus from the Thracian race, which had deprived their ances-  tors of the intervening countries. To the same. Caucasian  stock may have belonged other nations of doubtful origin, and  similarly circumstanced to the Lycians, such as the Peonians. This people, whose country was once nearly conterminous with  the later Macedonia, represented themselves as a colony of the Teucrians from Troy (Herod. v.15). The remains of these Teucrians were called Gergithes (ib. v. 22), and probably dwelt  at Gergis, Gergithus, or Gergithion in the territory of Lamp-  sacus (Strabo, p. 589). There was a second place of the same  name in the territory of Kyme (ib.) The Teucrian Gergithes  were afterwards removed by Attalus, and placed m another  town called Gergetha, near the sources of the Caicus (ib. ‘p. 616), perhaps the same as the second Gergithion mentioned  above. In the same country Strabo also notices Pionia and Gargaris as towns of the Leleges (ib. p. 610). Now the Zroes  were a Lycian people, and indeed the Lycian Zvos is Tros, as  the Lycian inscriptions show. ergeti, again, is an Ossetic  town, and Strabo (p. 504) speaks of Gargarenses on the north-  ern edge of the Caucasus.

uv (™ return from the digression on the Lycians—the addition  of the Lydians, with their xaofyvnro., the Carians and Mysians,  to' the same family as the Armenians, will unite this last  nation to that branch of the Thracian race from which the Etruscans were directly derived by tradition. '. It also brings  the Armenian language fo the shores of the. Aigean and the Hellespont, and half the distance from Armenia to Etruria. We now cross over into Europe, to examine in the first place  the language of the nation to which the name of Thracians  more especially belonged, and which extended from the Aigean  to the Danube.: . 62 THE ARMENIAN ORIGIN OF THE ETRUSCANS. Thracian Words. 1. “ApytAog, ‘uc.’ Esth. kiir, tir, ‘a mouse.’ Gr. dpak. Lat. sorez. Arm. arnét, ‘a dormouse.’ When we compare  ar-nét, gen. ar-niti, with the Sansk. ati-nidrdlu, ‘a dormouse,’  and the Sansk. nidrd, Arm. airh, ‘sleep,’ it is sufficiently plain  that the ar of ar-nét signifies ‘mouse.’ This gives us the first  syllable of adp-y:Ao¢; and the termination might be explained  from the Arm. givt, ‘a field,’ so that apy:Ao¢ would signify ‘ field-mouse, mus araneus.” Cf. Heb. ach-bar, ‘ field-mouse’ (ante, p. 12, 8. v. vne&le).

2. Bassara,‘a kind of garment, probably made of a fox’s  hide.” Also Lydian. Arm. dag, ‘hair, mane;’ maz, ‘hair;’  mask, ‘skin, pelisse, tunic.’ Bétticher notices the rather  curious fact that Jagor signifies ‘a fox’ in Coptic. Cf. Heb.  basar, and the kindred Semitic terms signifying ‘ flesh, skin ;’  and also Arm. daéavor, ‘hairy.’ 8. Bola, ‘wdrktc.’ “ Osset. dru, ‘arx, castellum.’” Arm.  berd, ‘a castle ;’ dourgn, ‘a tower.’ Gael. dri, ‘a hill.’ Germ.  berg, burg.

4. BolZa, ‘a plant, and the seed of a plant, resembling  rign.’ Therefore a marsh-plant. “ Radix Sansk. orth, ‘ crescere ;’  vrtht, ‘ oryza.’”” Arm. brinz, ‘rice ;’ prisk, ‘the plant Zhapsia,’ Rhet-Rom. ritscha, ‘ grass growing in water.’ 5. Bouyxdy, ‘«Bdpav.’ Arm. phrnéel, phenkal, ‘ to cry.’ 6. Tévra, ‘xpéa.”’ Arm. gJamb; Sansk. jambha; Irish dia-  mann; ‘food.’ Sansk. jam; Basque yan ; Span. yantar ; Arm. Gasel ; ‘to eat.’ Lat. sentaculum.

7. (Zaporkic). Dadpdy, ‘Sopav.’? Arm. satovmn, ‘ what  envelopes ;’ gati?, ‘ a carcase;’ getmn, ‘a fleece.’ Osset. zarm, ‘a hide.’ 8. ZeitAa, SlAa, ‘olvocg.’ Arm. zetovi, hetovl, ‘to pour;’  zetkh, ‘drunken.’ Gael. si/,‘todrop.’ The Scythians, Pliny  says, called the Tanais by the name of Sidis. They had, indeed, THE ARMENIAN ORIGIN OF THE EIRUSCANS. 63  several rivers of this latter name. We meet with the S:darus,  now the Sede, near Pstum in Italy.

9. Zepa, ‘ rep{BAnua, Zoua.’ Arm. Air, ‘around, a circle ;’  zrah, ‘a cuirass,’ = Pers. zirah ; ztr, ‘a circle.’ 10. Zerpata, ‘ yérpa.’ “Vocabulum grecum et thracicum,  ni fallor, ejusdem stirpis.”? Arm. kové, ‘a pot, a jug;’ kathsay, ‘a kettle ;’ Ake#i, ‘an earthen vessel, pottery.” Rhat-Rom.  checla, ‘a pot.’ Gael. soitheach, ‘a pot.’ The Mariandynians,  a Thracian tribe in Bithynia, called a black fig yérpa (Arica,  p. 8). Here we have the Arm. thovz, ‘a fig,’ and a form like  er-pala.  ll. ZeBvOldec, ‘ yvhoror.” Arm. zavek, ‘ child, blood.’ Kurd.  zava, ‘kind, species.” Ang.-Sax. s:6,=Germ. sipp. Lat. s0-  boles, = Gael. szolack.

12. ZéuBpor, ‘ sparythagoy. > “Slav. zob7’, ‘ urus.’ ‘Arm.  zovarak, ‘ vitulus, juvencus’ ”—dovar, ‘an ox’—gomé$, ‘a buf-  falo’— gamphr, ‘ a bulldog, a mastiff’— zambik, ‘ a mare’—  smbak, ‘a hoof. —Lyd. tuBoue, ‘ Bove.’ 18. Oparrnc, ‘ AlBoc.’? Arm. char, ‘a stone;’ kharak, ‘a  rock.’? Gael. creag, carraig, ‘a rock.’ 14. KarvoBarac. “Strabo vii. 8, 38. Aéyee rode Musode & Moceddviocg xat tbixwy aréxecOa war’ evatBeaav, ...  eee + Ota S& rovro KadcicBa DeoceBeic nal cawvoParac. Arm. khaphan, ‘an obstacle ;’ khaphanel, ‘to hinder, restrain,  take away’ (améyxev) ; ‘haphanovaz, ‘ hindrance, interdiction,  prohibition.’ 15. Kijog, ‘ Sonpiby mv.  “Sansk. kdmin, ‘a climbing  plant.’” Arm. kbnil, ‘to cling, to creep along the ground.’ Gr. xbapoc.

16. KodraBpropude, ‘ Opaxcoy Spxnua cat xapixoy, hy 8...  tvdérAcov. The word divides itself into xoAa-Bpicpdc. Bétti-  cher compares the first part with the Pol. fodo, ‘ circulus, rota,’  which would be allied to the Arm. holow, ‘ rotation, circula-  tion.’ The Arm. has also khatal, ‘to leap, to play.’—Bprepude  seems allied to Speclopara, ‘ dpxnote povytaxh;’ which, as the 6+ THE ARMENIAN ORIGIN OF THE ETRUSCANS.

xoAaBpropude was évdrAvoc, may be compared with the Arm.  bir, wirg, ‘a great stick, a club’ (cf. Lat. virga, Gael. di07, ‘a  stick’), wér, ‘a wound.’ 17. Kriorac. The quotation from Strabo in (14) is thus con-  tinued: Efvae d€ rivac rwv Opaxwy, of ywpi¢c yuvamde Coon,  ove xtlorac KaAdcioOa. “Cf. Hesychium: xamrpovrec éxa- Aovvro obrwe of Oogxec.”” Arm. khiz, ‘ an obstacle,’ a synonym  of khaphan (14) ; chezel, ‘to separate, to remove, to take away ;’  khéth, ‘an obstacle ;? whence éthal, ‘ to afflict one’s self,’ gthe/, ‘tostumble.’ Thus xarvoPara, xawpovrec, and xrlarae may  all be explained in the sense of ‘amweydpevor’ from the Arm. It is observable that the Greek reporter has made out of the Thracian two genuine Greek words, carvoara and «riorat.

18. AéBa, ‘aod.’ Germ. Jeben. Arm. linel, ‘to be, to  live’ Irish Uéhearn, ‘a habitation, a ship’ (cf. Leburnz and  liburna). Arm. léray or lavray, ‘a dwelling.’ LEsth. lin, ‘a  town;’ daiw,‘aship.’- In the Prussian province of Saxony  there are a great number of places with the suffix -/eben, such  as FKisleben' and Ermisleben. AéBa does not appear among Thracian names of places, though Lebea was a. Macedonian  town: but dava must have had. nearly. the same sense in Moesia and Dacia. Could AéGa be an error for AfBa? Dava  would be allied to Georg. dada, ‘village, place;’? Arm..deh, ‘district, part ;’? Pers. dik, ‘village;’ Gael. daimh, ‘house,  domus? | | oe 19. Mayae, ‘ cavic rerpadywvoc dmdxougpog Sexonévn rie  kOdpac rac vevpa¢g xal amorcAovca Tov pOdyyov.’ »Arm.  makan, ‘a drumstick ;? mahak, ‘a great. stick” Heb. makal, ‘a twig;’ makkel, ‘a rod.’ ; — . 20. "OAEte, ‘a bear.” Arm. arg, ‘a bear.’ Lat. ursus. Lith.  lokis. | : | 21. [léArne, ‘Opaxov SxAov Kai el8o¢ raplxov.’ Gael. peall, ‘to eover ;’ peall, pill, ‘a hide” Lapp. pils, ‘ pellis rangiferina  aut alia non satis villosa.——Pol. Jett, ‘a javelin” Arm.  batkhel, ‘to strike ;’ petel, ‘to dig.’ THE ARMENIAN ORIGIN OF THE ETRUSCANS. 65 22. Teriyw or merénv, ‘Snoavpdv.? Arm. pitani, ‘ profit-  able, useful, necessary ;’ pitoych, ‘what is necessary.’ Cf. Xpiia, xphpara. | 23. ‘“Poupata, ‘Spqxtov apuvrhguov, payaipa, Elpoc, fj  axdvriov paxody.’ Arm. rovmb, ‘sarissa’ (Rivola), ‘ bombe,  grenade’ (Aucher—a sense necessarily modern); rmbackar, ‘pierre (char) de baliste.’ Kurd. rkm, ‘aspear.’? Heb. ro-  makh, ‘a spear.’ Lat. ramus. Gr. paBdoc. Gael. rong, rongas, ‘a staff, a bludgeon.’ Esth. rond, ‘a piece of wood.’ Lapp.  rémpo, ‘ truncus sive tigillum in quo in frusta carnes aut alim  res conciduntur.’ 24, Zavarat, ‘ uéSvoor, compared by Bétticher with the O. Pers. cavvaxpa, ‘krwya.’ Lapp. saja, ‘situla.? Esth. sang, ‘a  milkpail.? Arm. saz, ‘a cauldron;’ sahil, ‘to flow,’ with  which may be connected, Lat. sanguis, sanies, and Mantschu  sengut, ‘blood.’ 25. Yapamapat, ‘xeparordéuo, aroxepaXioral.’ Arm. sar, ‘summit, top ;’ sayr, ‘point, edge;’ zayr, ‘summit, head.’ Kurd. ser, ‘head, top.’ Arm. pharatel, ‘to remove, to  take away;’ pharat, ‘distant, removed, far;’ pherekel, ‘to  divide ;’ pherth, ‘a part.’ Odsset. sar, ‘a head;’ farath, ‘an Gr. capa. Arm. haraphn, ‘a head;’ karaphel, ‘to  axe.’  behead.’ _ 28. Sidalcas.or Sitaleas. ‘“ Xenopho Anad. vi. 1, 6, Thracas Sitaleam vel Sidalcam carmen popularium suorum cantantes  introducit. Pars vocis altera Sansk. arka, Arm. erg, ‘ hymnus,  carmen.’ ” If. Sid- or Sit- imply ‘ popularis,’ 1t may be com-  pared with the Lapp. sifa, ‘pagus.’ Cf. xwywdla.

27. Zeadun, ‘Spaxla paxapa.’ Lapp. skalmet, ‘ pugnare  cornubus.’ Arm. gefé, ‘ the blade of a knife” 28. Exapxn, ‘apybpia. Lapp. skarktek, ‘ fiscus sive recepta-  culum pecuniarum, qu# in templo offeruntur sive dantur.’ Arm. skavarak, ‘a plate, a disk.’ Goth. skatis, ‘ geldstiick,  geld, apytpiov, Snvapiov.’ Germ. schatz. Lapp. skatte, ‘ tri-  butum.’ Eng. scoé. ; F 66 THE ARMENIAN ORIGIN OF THE ETRUSCANS. _ The result of the examination of these Thracian words tends  greatly to confirm the hypothesis, that the (Proper) Thracians  belonged to the same family as the Armenians. Traces also  may be discerned, as in Asia Minor, of the existence of a pri-  meval Finnish or Turanian nation in Thrace. It is satisfac-  tory to find that all the Proper Thracian words known seem  capable of explanation either from the Armenian or the Fin-  nish, . .

The course of my argument would now lead me to examine  the Albanian; but it may be well, in order that the early eth-  nography of Europe may be more clearly understood, to make @ previous analysis of the remains of the Scythian language. There are also a number of Dacian words, which may be after-  wards noticed. Scythian Words. 1. "Afee, ‘EBadXrov.’ Gael. up, ‘to push ;’ 0, ‘to refuse,  deny, reject.’ Welsh ¢d, ‘to send from, to say,’ = Gr. trw. Arm. wép, ‘toc.’ Lyd. iBu, ‘rd Body.’ Hesych. “ Mox se-  quitur iBte, ‘rémra, Bog.” ‘They pushed,’ would be in Gael. dh’ up iad, or, omitting the initial dA’, as may be done,  up iad. This bears considerable resemblance to afsie. The  final consonant in zad, the constructive form of szad, ‘ they,’ is  sometimes omitted in Gael., as in dia or Jeo, ‘ with them’ (/e, ‘with ’). | 2. “AyAu, ‘xbxvog.’ Gael. ala, eala, ealag, ealadh, ‘a swan,  a wild swan.’ LEsth. Juik,‘aswan.’ Lat. olor.

3. "Adrydp, ‘rpwkadXle.’ Gael. ithche, itheadh, ‘ eating, the  act of eating ;’ -or, -oir, or -wir, Gael. termination. Sansk.  adaka, ‘edens, rpwywv.’ | 4. “AXfvda, ‘an oleaginous kind of cabbage-plant growing  in the Tanais.’ Gael. ola; Arm. ef, wt; ‘oil.’ Gael.  aileanta, ‘fragrant ;’ atllean,‘elecampane.’ Lat. allium. Gr. éXéviov. Arm. etég, ‘endive;’ etégn, eting, ‘a reed, a rush, a  nettle.’ THE ARMENIAN ORIGIN OF THE ETRUSCANS. 67 5. “ABaxnc, aBak, ‘adpwvoc, siwrnpdc, kal Svoua apxirex-  rovixov, 6 LcvdIac A€yovay avdpa xadciv.’ Sansk. a, negative  particle, vad, ‘loqui.’ Gael. 0, ‘from,’ = Lat. a, ab; bagh, ‘a  word,’ = Lat. voz.

6. “Avop, ‘vovg.” Gael. anam, ‘soul.’ Lat. animus: &e.

7. “Apia. Herodotus iv. 27. apa tv xadfovor TxbSac. Errat. Non unum sed primum significat et prestantissimum. Sansk. aryaman sensum primarium amisit, retinuit Zend. aiy-  ama.’ (Gael. air, atreamh, ‘number ;’ atreanach, ‘a beginning.’ Lapp. aremus, ‘primus.’ Gael. arm, ‘origin,’=Arm. arm,  armn,= Arab. arum. Ayab. arém,‘one,any one? __ 8. Bol&af3a, ‘xptov pérwrov.’ Georg. verzi,‘aram.’ Pol.  baran,‘aram.” Kurd. berk; Pers. barah; Lesgi bura; ‘a  lamb.’ Lat. vervex. Lapp. brekot, ‘balare.——Lapp. diwe, ‘caput.’ Pied. abd, ‘ caput.’ 9. Groucasus, ‘nive candidus. Lapp. grawes, ‘canus;’  kaisse, ‘mons altior, plerumque nive tectus.’ Esth. hahho, ‘ frost ;’ kasse jaa ( jaa, ‘ ice’), ‘ice formed by frost upon snow.’ Georg. giswa, ‘ frost.’ ~ 10. Advovare or davou[sic, ‘ airtog Suvotuyxiac.’ Gael. don,  donas, ‘mischief,’ and perhaps aobhar, ‘ a cause.’ 11. ’E&apraiog, ‘ipai ddor’.’ “Id Zend. esset asavand pdthd.” —— “ Sansk. yag, ‘ venerari ;’ Zend. yaz; Arm. yazel, ‘ sacrifi-  care; yast, sacrificium”! (hine Gr. Gyio¢).”’ Arm. ganapar, ‘away.’ 12. ‘Inrdkne, ‘ Bpwpa oxvSixdy dca yaAaxrog immelov oxev-  aZdéuevoyv.’ Lapp. Adpos, ‘equus.’ Osset. ye/s, afse, ‘ equa.’ 18. Kavaie, ‘a kind of odoriferous plant like flax.’ Arm.  kaneph; Gael. caind; Bret. cazab; ‘hemp.’ Gael. canach, ‘ cotton-grass, the herb cat’s-tail, cotton, a cotton-tree;’ caci-  neach, ‘stubble.’ 14. Kapaptec, xapayn. “ Kapaptec of sxvSucot olxot. Evioe  d2 rac xarhpec audEac. Kapapn % txt ric audtac oxnvh. Arm.  karch, ‘plaustram.’” Gael. carbad, ‘a chariot.’ 1 Cf. Lat. hostia: also ara (= asa) with yaz-el. F2 68 THE ARMENIAN ORIGIN OF THE ETRUSCANS. 15. KéAog, ‘a kind of white animal, in size between a stag  and a ram, but swifter in running.’ Arm. khatal, ‘to move,  to leap.’ Gael. clis, ‘ quick.’ See ante, p. 68, s. v. coAaBpropdc.

16. Képaxor, § pidcor Salpovec.’ Gael. car, ‘ friendly.’ Arm. ogi, ‘spirit.’ Alb. ago, ‘ God.’ 17. Maréac, ‘aotoc. Pol. mada; Esth. matia; ‘mud.’ Arm. miz, ‘dirt, filth.’ Lapp. mdiwe, ‘ pulvis;’ mdiwe atse (atse, aqua’), ‘aqua turbida.’ Matoas was a Scythian name  for the Danube, ‘turbidus et torquens flaventes Ister arenas. It might perhaps = mdiwe tatse.

18. MéowAn, ‘4 oeAqvn,’ “proprie luna plena, ni fallor.” Sansk. mds,‘moon, month.’ Gael. mos, ‘month.’ Arm.  amis, ‘month;’ mahik, ‘the horns of the crescent moon.’ Arm. 1; Gael. dan; Cornish Zen ; ‘full’ Lat. plenus. Gr.  wAéoc.

19. Oiéprara, Spparat, ‘avdpoxréva.’ Gael. fear, ‘man ;’  bat, ‘to beat ;’ das, ‘death.’ Arm. ayr, ‘man ;’ mah, ‘ death.’ 20. Tlayatn, ‘xbwy.’ Pol. pies, ‘adog.’ Lapp. piddnak, ‘a  dog.’ Germ. petze. Eng. ditch. ‘ Herod. i. 110: riv xéva  kadfquot omaxca Mido. Zend. gpd ; Afghan spay ; Pers. sak; ‘canis.’”? The sibilant in owaxa may have passed into an  aspirate, and then have been lost, in zayafy, just as Arm.  spitak = Sansk, ¢veta = Ang.-Sax. Avit = Eng. white = Germ.  weiss. axa is, however, identical in meaning with the Pol.  sobaka and suka, in which second form the labial has disap-  peared, as in the Pers. sah. : 21. IldpSove, ‘puyadac. Arm. pharatel, ‘to drive away ;’  partil, ‘to be conquered.’ _ 22, Zayapeuc. ‘ Macoayfra: ro&dra re xat alxpopdpor,  gayapec voulZovrec Exe.’ Arm. sakr, sakovr, ‘asabre.’ Lat:  securis. 23. Lavarric, ‘olmewrne. See ante, p. 65, s. v. cavarat. 24. Xleupva, ‘ xirwv oxbrivoc, Evrptyog, xetptowrde.’ Lapp.  sasne, ‘pellis depilata ;’ sassnek, ‘femoralia ex aluta facta;’  gasse, © manica,’ ~ THE ARMENIAN ORIGIN OF THE ETRUSCANS. 69 25, Sarov, ‘an eye.” Arm. spasel, ‘to observe.’ Germ.  spahen, Eng. spy. Lat. spes, specio.

26. Tapavdoc, ‘a kind of animal, 7d péye8o¢ Bode, rov Wpocwrov Tov rérov zorxog 2Aagy.’ Gael. tarbh, ‘abull;’ tar-  bhan, ‘a little bull;’ ¢arbhaii, ‘ bull-like;’ tarbhanta, ‘ grim,  bull-faced.’ The Gael. 54, =v or w, is scarcely audible in tarbh  and its derivatives.

27. Temerinda, ‘mater maris. Divide into ¢emer-inda. “ Sansk. ¢dmara, ‘aqua’ (cf. Cornish Tamar); timi, ‘ oceanus ;’ ‘tim, humidum esse.’? Arm. ¢amovk, ‘humidus.’” Gael. tadd,  tamh, ‘ oceanus ;’ taom, ‘a torrent, to pour.’ Arm. thavn or  thé, ‘moisture.’ Arab. tamm, ‘the sea.’ Inda is compared  by Zeuss (Die Deutschen, p. 296) with the Hung. anya, Turk.  anya, inya, ‘mother.’ Kindred terms would be—Lapp. edne, Circas. ana, Lesgi enniu,‘ mother :’ and also perhaps the Arm.  hant, ‘ grandmother,’ and the Lycian une, probably ‘ mother.’ 28. ®pl&a, ‘ ucomrdunpoc.’ The analogy of apaéa, ‘ uicomdp- Ocevoc, the name of a plant in the Araxes, shews that it is the  first part of ¢pt%a which signifies ‘ wicked,’ and the latter part ‘hating.’ The word is not easy to explain, but Dr. Donald-  son seems to have analysed it successfully. He compares the  first part with the Germ. /rev-el, and the Lat. prav-us, words  which would be allied to the Arm. wrép, ‘a fault ;’ wripak, ‘wicked ;’ wripil, ‘to sin ;? and also to the Irish jar, ‘ crooked,  wicked ;’ freac, ‘ wicked.’ The same philologist also compares -Ea with the Germ. scheu, the primitive idea in which is < aver-  sion’ or ‘ turning from,’ as appears by the Eng. shy, shun, and  the Germ. schief, schel, or schiel. The Arm. has Set, ‘ oblique,’ = Germ. schief; and st, ‘squinting,’ = Germ. schel or schiel. In Irish, we find seach, ‘a turn,’ and also seach, ‘to pass by, to  shun.’ On the whole, the Irish seems, in the words /reac-  seach, to approach nearest to ppfEa.

1 For the names of the Scythian deities see Varronianus, p.48. I have  merely to add their apparent Armenian or Celtic affinities. The names  are six in number :— @? 70 THE ARMENIAN ORIGIN OF THE ETRUSCANS. “When we come to compare the Proper Thracian with the Scythian, or the most ancient European language of Turkey  with the most ancient European language of Russia, we may  discern in each both Finnish and Armenian affinities. The Scythian language is, however, distinguished from the Proper Thracian by the presence of an important, it may even be said  a predominant, Celtic element. Now it is very possible that  there may have been at one time two distinct races, one Thra-  cian and the other Celtic, in Scythia; the Scythians and the Cimmerians. \; The Scythians are indeed called voc Opgxov  by Stephanus of Byzantium; and an affinity between the Cimmerts and the Kymry has frequently been conjectured. However this may be, the affinity between the Scythian  and the Celtic, and the absence of any such affinity between  the Proper Thracian and the Celtic, may be used as an -argu-  ment to prove thatthe Celtic elements of the population of Europe entered mainly through Russia, and the Armenian (or Thracian) elements mainly through Turkey,) Dacian names of Plants. ( The Dacians and Gets were always considered as Thracians  by the ancients, and were both said to speak the same Thracian  language. The relics of this language are of a peculiar nature,  consisting of the names of plants, and thus only likely to ex- Tart, ‘ Iorly.” Arm. tap, ‘fire.’ Gael. teas, ‘ ardour.’ Tlawatos, ‘Zevs.’ Arm. pap, ‘grandfather.’ "Anta, ‘TH.’ Irish bh, ‘a country.’ Gapipardbas, ‘Mocedéuv.’ Gael. tamh, ‘oceanus.’ Arm. tamovk, ‘moist.’ Arm. mazdezn, ‘a, hero.’ Olréovpos, ‘Axé\dwr.” We find in Arm. with, ‘a gazelle,’ and with-  khart, ° vast, colossal ;’ from which might be deduced with, ‘swift’ or ‘mighty.’ Zér is ‘power, force,’ in Arm. and Pers. Olrécvpos might pos-  sibly be withazér, ‘ greatly powerful,’ and thus be equivalent in sense to  the Arm. mezazér ; or else, which is perhaps more appropriate, ‘ swiftly  powerful.’ With, if interpreted ‘swift,’ might be allied to the Arm.  show | ‘swift,’ and phoyth, ‘swiftness,’ which may likewise be compared  with ol7., "Apripraca, ‘ Odpavin ’Adpodlrn.’  a THE ARMENIAN ORIGIN OF THE ETRUSCANS. 71  hibit a partial affinity to another language of the same family. Thirty-two such names have been collected by Grimm, in his History of the German Language, from Dioseorides rep? vAne  larpicnc.) . 1. BAfrov. ‘Pwpator BAlrouu, Adwor BAHe. Arm. diith, ‘bread, cake.’ Gael. dleath, bleth, bleith, ‘to grind.’ Welsh  blawd, ‘ meal, flour.” The leaves of orach are “as if they were  overstrown with meal or flour ;” and those of dite are “ almost  like the leaves of orach, but not so soft, white, nor mealy.’”* —  -@. "AvayadAte Gppny, xeAddmov, Tarot carava, Adxor  xepxepagpwyv. Grimm is inclined to consider this as a  corrupted Greek name, and is led by the Gr. agpwy to compare  the Germ. name of the anagallis, gauchheil (gowk-heal), ‘salus  stultorum.’ By a rather curious coincidence, the Arm. aprovmn,  from aprel, ‘salvare,’ signifies ‘salus;’ while kerkeri/, again,  signifies ‘to be hoarse.’ Kerkeraprovmn, therefore, which is  almost identical with xepxepa¢pwyv, would be Arm. for ‘salus  raucorum.’ Disocorides says of this herb: 6 8 yvAdc¢ abroy  avayapyapiZopevocg arogAcyparifea Keparjv.

3. XeAddioy péya. “Pwyator p4Bcouu, TédrAAot Sova, Adxor  cpovorayn. “Lith. krégéde, ‘xedidor;’ hregédyne, ‘ xer-  ddveov péya.’” (Grimm). The Arm. words for celandine have  no affinity to the Dacian. They are: gezernakhot and zizer-  nachar, ‘swallow-herb’ and ‘swallow-stone.’ Yet we may  derive the Dacian name, xpovor-, of the ‘ twittering’ swallow,  from the Arm. For in Arm., fArié is ‘chirping;’ and the  change of the suffix -:¢ into the suffix -ovst would give krovst.

4. Kevratprov. ‘Pwpatoe gpevpipobyiap, of O2 avpa povaA-  ripadil, Adcot rovABnAd. Centaury is only defined in Arm.  as ‘a certain herb (A4ot).’? The nearest approach to rovABnra  in Arm. is éovit, = Kurd. #6/h, ‘ a\Sata, bismalva, marsh-mallow,’  and ‘ podddevdpor, oleander.’ There are also thatth, ‘ sowbread,  cyclamen,’ and thovith, ‘paper.’ But rovA-B (nda), ‘ febrifuge,’ 1 For v: r. see Appendix. 3 Dodoens’ History of Plants.

—- 72 THE ARMENIAN ORIGIN OF THE ETRUSCANS,  might be explained from the Esth. tué/i-ab0:, ‘fire-help.’ The Gael. for rovA-BnAa is ceud-bhzleach, ‘ hundred-leaved.’ 5. Afibaxoc. ‘Pupatoe AGBpoun Pévepic, of Se xapdoug Bévepic, Adxor oxcaph. Welsh ysgall, ‘a thistle.’ Gipsy  karro, ‘a thistle. Arm. cher(el), Alb. skyir, ‘to tear.? The Arm. for ‘ teazle’ is hangar ; for ‘ thistle,’ ochoz, ekchan. These  last seem allied to the Esth. okkas, ‘sting, thorn,’ and ohhakas, ‘ thistle,’ as well as to the Gael. oighionnach, ‘ thistle.’ 6. "Hptyytov. ‘“Pwpaioe xarlroXoun xapdove, of o2 Kap-  repat, Adxor atxoumvot& Arm. sokh, ‘an onion,’ sekh, ‘a  melon’: Gr. aikve: Alb. segé, ‘a pomegranate.’—Arm. phovng, “a plume, a tuft.’ “Upon the tops of the branches (of the  great Eryngium) come forth round knobby and sharp prickly  heads, about the quantity of a nut, set round about full of  small flowers :” and at the top of the branches of the small Eryngium “ grow round, rough, and prickly bullets or knobs.” 7. Obuog. ‘Pwpatoe Sobduoup, Adxor péfovAa. Grimm  refers to muscus, Germ. moos. In Alb., modé is ‘ darnel ;’  modhoulé, ‘a pea;’ and misir, ‘a wallflower’ or ‘ gilliflower,’  both fragrant like thyme. The Arm. for ‘thyme’ is zothrin ;  for ‘moss,’ mamour or lér. In commenting on the Lyd.  puade, 6 d&bn,’ which I have considered as akin to the Arm.  mosay, ‘ wuptkn, Bodtticher expresses the opinion, that 1t must  be compared with pwic, ‘yn,’ so that it would have the sense  terrigena, “ut ad quamvis arborem designandum aptum sit.’ If pvade and mosay signified ‘ terrigena,’ such a meaning would  be equally applicable to pdZovAa, ‘ thyme.’ We have, too, the Kurd. miték, ‘box, a box-tree.” The termination -ovA(a)  would be Arm., as in pt-ovt, ‘ fruit,’ ovl, ‘a kid.” There would  remain yoG-, ‘terra,’ which may be explained from the Georg.  miza, ‘terra, Eng. mud, Esth. métta, ‘mire,’ Arm. miz, ‘filth. There is also Arm. mazar (cf. Span. madera and Lat.

materia), “heath, bush, furze, broom, clump of wood ;’ which  appears allied to the Esth. méts, ‘ bush, wood.’ 8. “AvnSov ro éoSidpevov, of S& worAyldoc ... . Adxor \ THE ARMENIAN ORIGIN OF THE ETRUSCANS. 73  woAmvoup. The Arm. for ‘dill, fennel,’ is samith.1 Both is ‘horse-radish ;? and Jdotbog, ‘a sprout, a bud;’ and pipe, ‘pepper.’ In Georg. balakhi 1s ‘ herb,’ daikha, ‘ mallow,’ and  boloki, * root, turnip.’ 9. "Apremiola. ‘Pwyaior ovadévria, of dt osprbAXoup, of St {oa péyta, of 8 pamrloup, of S& repravayera, TadXor  movi, Adxot Zovdarn. I find no name for mugwort in Arm. ;  but Zoudorn readily resolves itself into Arm. zow-ost: zow, “mare ;’ zowak, ‘stagnum’—os¢, ‘ramus, germen, palmites,  frondes,’ = Germ. ast. ‘“ Mugwort groweth in the borders of  fields, and about highways, and the banks of brooks or quiet  standing waters.” (Dodoens). .’Apreuicla pberat tv wapada- Aaaciowg rérotg, We To wWOAd. (Diosc.) Another of the above  names, repravayera, resembles the Arm. thrthngovk, ‘sorrel.’ . Therth is‘ a sheet (feuille) of paper.’ 10. "Oppevov fuspov. ‘“Pwpator yepevddcc, Adxoe Spypca. ‘Evidently formed from the Greek.” (Grimm). Or both  derived from a common origin. Clary is not named in Arm.,  but there is warm, ‘a net, a noose,’ which may be compared  with Sppuoc, dpa, and Spuevov. In Esth., ormid is ‘ goats-  beard or meadwort.’ 1]. ASéoreppov. ‘Pwpatoe xodobpbap, Adxor yovoAjra. Apparently Greek. The Arm. is charasermn, which is exactly ‘AiIdoreppov. Tovodjra, if genuine Dacian, might be akin  to the Arm. gonget, ‘turnip, wild turnip or rape, rape-seed.’ 12. "Ovofpuxic. “Pwuator éraxa, of 8? BpryAdara, of 82 Adwra, of 82 lovyxivaAcu, Aaxot aviacoc&é. No name in Arm., but aviac-oc&é is not unlike the Arm. anyag séz, ‘un-  lucky grass,’ or ‘ mischievous grass.’ This is very nearly the French periphrasis for ‘ weed,’ méchante herbe. Cf. also ‘ infelix  lolium,’ 13. Xapariruc. ‘Pwpator Kurpurovp, Adxot SoxeXa. In 1 Semeth is given in Dioscorides as the Egyptian name of cress. In Arab. shamér is ‘fennel ;’ shdmidh, ‘fruitfal (palm) ;’ and skdmm,  shammaé, ‘smelling.’ In Pers., shamdkh is a kind of millet. 74 THE ARMENIAN ORIGIN OF THE ETRUSCANS. -eXa we may have a word equivalent to the Arm. efevin, ‘a  pine, a cedar,’ Gr. tXdrn, Arm. etat, ‘a cedar.’ The meaning  of dox- is less easily conjectured. Thovkh signifies ‘brown’ in Arm., which presents no name for grownd-pine. But Sox-  would be better referred to the Alb. foké, ‘ earth, ground,’ which  is exactly the meaning required.

14, Aeuwvov. ‘Pwyaioe oveparpovu vlypoup, of d& riv-  tivvaBovAoup reppat, TadAot touuBapodp, Aaxoe ddxcva. “A field-flower, which Pliny (xx. 8) calls deta silvestris.’”” (Grimm) who compares our daisy and the Germ. fag. But in Arm., ‘beta rubra’ is gakndet. Subtracting det, which  signifies ‘ medicine, herb, colour,’ there remains gakn to com-  pare with Saéxeva. The word gakhin, ‘a marsh,’ is very pro-  bably not allied to gakn. In Rhet.-Rom., giaccun is ‘ plantago,  waybread,’ which “hath great large leaves, almost like to a  beet-leaf.”” (Dodoens). Now giaccun, phonetically, = gakn.} 16. Buple. “Pwpator yAadfodoup, of O8 ips aypéoreu, Aaxor &mpove. In Arm., prisk is ‘Thapsia, a kind of dill;’ but Gmpove is more nearly the Arm. aprovst, ‘salus,’ the literal  meaning of aA@ala or salvia. In Pers., ipdr is ‘thyme.’ 16. "Aypworte. “Pwpyatoe ypapev, of S aawpdrLoup, of 8  cayyourvédcp, of St obviorap, "loravol awapla, Adxot xorlara. This seems decidedly Arm., as ‘fot signifies in Arm., ‘ herb,  grass, hay, forage,’ = aypworic, = gramen. Many of the  plants previously noticed are unnamed in Arm., but merely  described as certain ‘herbs’ (thotch). The last member of  xor-fara may have defined the particular ‘herb’ (44o/),. or it  may be regarded as a termination like the Arm. -avéé. Cf. Arm. hot, ‘perfume ;’ hotavét, ‘ odoriferous.’ The Spanish name for gramen, it appears, was azrapla. May it be connected with ampovc and xepxep-appwy, and re-  ferred to the Arm. apr-zl, ‘ to live, to escape ;’ apr-ovst, ‘ safety, 1 The plant is also called in Rhet.-Rom. plantagien and luoza. The Esth. is ¢e leht or te lehhed, ‘ way-leaf.’—Cf. 34. THE ARMENIAN ORIGIN OF THE ETRUSCANS. 75  nourishment, provision’? See ane, in Lyd., 8. v. ravdobpiov ;  and znf., ‘on the possible extension of the Thracians to the  west of Etruria.’ | 17. Bdrog. ‘Pwpator oévric, of 82 pobBoup, of 52 pépa Bart-  kava, Adxoe pavreta. “I hold this for the Gr. pavreta.” (Grimm).

18. TlevrdgvAAov. “Pwpatore xeyxepdérAcoup, TadAoe wepuré-  dovAa, Adxoe mpomedovAd. The Celtic word is evidently  genuine, as Grimm notices. Bret. pemp, ‘five;’ Gael. duille, ‘a leaf.’ The Dacian is more doubtful. If we divide into  qpo-mwedovAd, we may conmipare the last part with the Dacian  p2eIera (inf. 27), the Gr. wéradov, and the Arm. thithein, ‘lamina,’ or ptovt, ‘fructus.’ Yet mpo- cannot mean ‘five :’  the Arm. is fing, which would require iyyoredovAd. But ‘cinquefoil’ was also called EvAowfraAoy (Diosc.) Now ‘wood’ is in Arm. pr-ak, ‘a stick’ is d¢7, and ‘a rush’ is préov. In Alb., phré is ‘a pole. These may explain wpo-, as well as  wpo-ctopva (28), and wpi-adnAa (82).

19. Tpaytov, tpaydxepwo. ‘“Pwpator xopvovAdxa, of 32 Berovévea, Ado: carla. Sélk is Kurd. for ‘ beet. The Arm,  has nothing nearer than se?kh, ‘a melon;’ zatik, ‘a flower;’  git, 2tot, ‘astalk, a stem;’ zatk, ‘a stalk, a rod;’ satarth, ‘a  leaf, a hair” In Esth., sa/k or sallo is ‘asmall bush.’ Gr. GadrAdéc. See ante, in Phryg., s. v. ZéAnea.

20. ‘Yoexbapyog. ‘Pwpaioe tvodva, Sevrapia, Tadd BuAt-  vouvria, Adxo: OuéXeta. Hendane has no name given in Arm. ;  but déAeca is readily derived from the Arm. de?, ‘ medicamen,  venenum, yxdproc, Bordvn, unguentum, color.’ Cf. Arm.  mkndet (m(ov)kn, ‘ mouse, rat’), ‘ arsenic,’ 2. ¢€., ‘ratebane” In  det dalar (dalar, ‘ green, Oadzpdc’), ‘herbs,’ det appears to  mean simply ‘herb.’ In gakndet, ‘beta rubra,’ we have  perhaps the Hellenised Dacian, Saxiva diéAcca. See ante, 14.

21. Srpbxvov aXrmaxaBov. ‘Pwpyator Biooddic, of 8  aroAAwapre plvwp, of 88 dpdyweu, Adxot cuewAlda. Grimm  suggests cuculus, in Arm., kkov. In Persian 46k is ‘lettuce.’ 76 THE ARMENIAN ORIGIN OF THE ETRUSCANS. Cuckoo’s meat is wood-sorrel. KuxwAlc is winter-cherry. The  nearest Arm. word is sakov?, ‘soft.’ But see Appendiz.

22. "AxaArbgn, of S2 xvldn, ‘Pwpaior ovprixa, Aaxoe dév. Welsh danad, ‘nettles.’ Gael. deanntag, ‘a nettle.” Arm.  danak, ‘a knife,’ which compare with xv{dn.

28. Tlorapoyelrwyv. ‘Pwuator Bivat pdAroup, of d& 2oBayw,  of 88 yAactardépiap, Aaxot xoaddua, TadAo ravpot.. Koaddpa nearly resembles the Arm. sotem, ‘ cress, nastur-  tium,’ one kind of which is worapoyeirwyv. In Georg., kuda  is ‘ privet.’ 24, "Aoryp arruodc, of d2 bdp0adrpov, “Pwyaiot tyyuraAte, Aaxor pa0(B.da. Grimm refers to the Old Norse rddhibidh, ‘tempus consultandi,’ and instances the ‘forget-me-not.’ From the Arm. we should get skhrat-avet, ‘ monition-  announcer.’ Avete/, ‘to announce,’ has a favourable sense :  avetaran is ‘ Gospel.’ 25. BodyAwooov. ‘“Pwpaio: AoyyalBouu, of d2 Alyyoua BéBoup, Adxos BovddAda. It does not appear that sada  signifies ‘tongue’ in any language. Grimm cites the Swiss Jél (in Stalder, alle), ‘tongue,’ comparing dingua and lingua. In Arm., Ja/ is ‘lamentation.’ Cf. Lat. dallare. The Arm.  for ‘tongue’ is /ezov; for ‘ox,’ ezn; and for ‘ bugloss,’  eznalezov. The nearest approach to SovddAAa is in the Gael.  buaghallan or buadh-ghallan, ‘ groundsel’ (lit. ‘ good-bough’).

26. Karavayxn. ‘Pwpaior Epa pircxAa, of 82 Sarfloxa, of 32 "1dr padcouc, Aaxor kaporl[SAa. For -riSia see 27. Kapo-  might be explained in many ways from the Arm. kar, ‘strength ;’ kar, ‘thread, string ;’ her, ‘hair ;’ char, ‘stone ;’  khar, ‘forage.’ No name appears in Arm. for fern, though for  actavrov we find zarkhot, i.e. ‘ mane-wort.’ 27. "Adlavrov. ‘Pwpator xeyxiwadse, of d? réppat carfAXove,  of 82 covmepxiAtoun réppat, Aaxor pitogIeera. Divide  with Grimm into ¢:30-gSeSeAd. Dido, miSda, and g&e9eda  might all be derived from one root. Grimm instances, for the  last. two, wéraXov, supposed = a (r)éraAov, and ¢édXov, perhaps ¢ THE ARMENIAN ORIGIN OF THE ETRUSCANS, 77 = 9(3)bAXov. In Arm., we have phthith, ‘the blowing of a  flower,’ phththil, ‘to blow, to bud, to sprout, BrAaordven ;’  phththot, ‘blowing ;’ phththeal, ‘blown ;’ and, which may be a  kindred term, ¢hithetn, ‘a blade, a leaf’ (but only of metal). The final'” in ¢hithetn is a common superfluity in Arm. Com-  pare |  phak, ‘closing’. . phakel, ‘to close’ . . phaket, ‘a bandage.’ (4at), th. of .. . batel, ‘to join’. . . batetn, ‘ivy.’  prededa, qu. ‘a blade.’  thithetn, ‘a blade.’ Compare also Arm. ¢hev, ‘a wing,’ thethev, ‘light, active,’ thel, ‘ thread, fibre, stalk of grass,’ and ¢theph, ‘bran,’ with filum,  fitz, wridov, and the Dacian g:3(0), midAa, and @Ie3eXa: also  arépic with mrepdv, = Arm. phetovr, = Germ. feder, = Eng. fea-  ther. In Alb., poupélyé is ‘down.’ One of the three Dacian  expressions, and perhaps the second, mi:SAa, may have signified ‘fern ;’ and xapomiSAa, our tree-fern or wall-fern, may have  meant ‘stone-fern’ (Arm. char, ‘stone’). i30-g3e3eAa was  perhaps ¢hethevathithetn, ‘ light-bladed.’? So we have in Arm.,  thethevathev, ‘ active, light’ (lit. ‘ light-winged ’); and ¢hethe-  vasarz, ‘light’ (lit. ‘lightly-moved’). The Arm. continually  forms adjectives by the combination of two substantives with-  out any inflexion. Grimm proposes to read for rpomedovAXd, ‘ cinquefoil,’ in 18, something like mirérvAa or ringSéSeXa,  thus giving @edeAd very nearly the sense of the Arm. thithetn. . The Arm. for ‘ five-bladed’ would be Aingathithetn, iyyaSt3eXa.’ TleSovAd is also like the Arm. péovt, ‘fruit,’ which, as well as  ptetn, ‘an elm,’ may be allied to rfSAa, phthith, &c.

28. "EAAEBopoce pédac. ‘“Pwpaior Beparpoup, of 82 aa-  paxa, Adxot tpodlopva. ‘ Hellebore,’ is in Arm., hog Wraii, ‘Georgian stalk;’ tog being apparently = soft, ‘ stalk, stem.’ Cf. cucwAlda (21). From the Arm. préov, ‘juncus,’ and the  suffix -drén, we might form for rpodtopva the word priovérén, ‘junceus.’ See also 18.

29. ’Axrh, of St Sévdpov apxrov, of S2 hyepov, Pwpator cap-  phthith, ‘blowing’ . phththil, ‘to blow’ . 78 THE ARMENIAN ORIGIN OF THE ETRUSCANS. Botxoup, T'aAAot cxoPijv, Adxor of DB a, xadapoedsic Exovea  kAddove. “Old Norse séf, Swed. sdf, ‘juncus’—Serv. zova, ‘sambucus nigra.’”? (Grimm). Arm. seav, gen. sevoy, ‘ black’ —cf. Georg. Savi, ‘black ;’ gavz, ‘ populus nigra’— sevni, ‘a  dress woven with palm-leaves.’ The Arm. for ‘elder’ is  thanthrment. The Servian word may be originally Dacian.

30. Xayuataxrn, of 62 EAeoc axri, of 8% aypla axri, ‘Pwpaior €BovrAXAous, T'aAXoe Sovewv?, Adxor Apa. Phryg. ZAvpoe, ‘avAdc, = Arm. etégn. Arm. etevin, ‘a pine ;’ eting, a nettle.’ 31. KoAoxuvSic. “Pwuatot xovxobpBira oABdrixa, Adxor  rovraorpa. Arm, thovz, ‘a fig;’ thovth,= Kurd. thu, ‘a  mulberry ;’ ddovm, t.e., dédovm, ‘a gourd,’ = Gipsy dudum.

82. "Aumedoc péAaiva, of 62 Ppuwvla péAava, of d Bov-  kpaviov, ‘Pwyator df3rAaphvia, of 0? Baravodra, of 62 BeriaaAxa, Adxoc mpiadhAa, of d& weyplva. No name is found for  briony in Arm. [Ipiad-hAa might possibly be compared with  mpoo-iopva (28), the Arm. suffix -e? replacing the Arm. suffix -Orén or -y6rén, See also 18. Teyp-{va, if by metathesis for repy-  fva, might be derived from the Arm. prk-e/, ‘ to bind tightly.’ Cf. vitis.

To these Dacian names Grimm adds one used by the Bessi  in Thrace— 88. Bhxiov. ‘Pwyator rovatdayw, of d2 papddpiap, of Si  moverovAdyw, Beaoot aca. Arm. haz, ‘a cough, (xe,  tussis.’ One name seems to have been overlooked by Grmm— 34. “Avdpaxvn aypla. ‘Pwyator 2drAEceBpapu, of 82 roprov- Adxa. Adxoe XAE. A kind of sengreen or houseleek. Germ.  tauch. Eng. leek. Arm. eté, ‘endive.’ Gael. dus, ‘herb.’ | When we examine such of the above names as admit of  clear explanation, we shall find that one (8) is Lithuanian,  and one (22) Celtic. Five (16, 20, 23, 31, 33), of which one  is Bessian, may be claimed, with scarcely any hesitation, as ‘Armenian, and probably even more. Thus 1 would be either THE ARMENIAN ORIGIN OF THE ETRUSCANS. #79 _ Celtic or Armenian, but rather the former, and 27 appears to  present a peculiar instance of Armenian affinity: 14 also,  when compared with 20, coincides very accurately with the Armenian. A number of other names are readily derived and  formed from the same language; as 2, 3, 6, 9, 12, 15, 24,  and 26: and several more might possibly be deduced from “Armenian roots. ( One thing is remarkable in the Dacian  names, the terminations formed with r.) Thus there are— TovAP-nrAa and mpiad-fhAa, Sox-sAa and gi30pSe3-eXd, xapo-  wf%-a, and pdZ-ovda and mpomed-ovAd. i Such terminations  are characteristic of Armenian, as will be exemplified in the  examination of the Etruscan, where they are likewise con-  tinually found ; and also in Albanian eB, modh-oulé, ‘a pea;’  tsing-ouly, ‘a kind of fig ;’ ¢rang-oul, ‘a cucumber. At  present it will be enough to refer to such Armenian words as  have been already cited in the analysis of the Dacian names—  phthth-eal, phak-et and pt-etn and bat-etn and thith-etn, kog-t,  and pt-ovt and kak-ovt. There are also d:éAeca and cada on one  side, and on the other de? and 22%, or else zatik, probably a  diminutive of zat. Nor is it without importance that, in ex-  plaining the names-of simples, s0 many Armenian names for  vegetables and the parts of vegetables should suggest them-  sebves. , The value of these coincidences cannot be better estimated  than by Grimm himself, whose object is to prove the Dacians _and Getz to be Teutonic. ‘It is,’ says the great philologist, ‘amply sufficient, if sez or ezght of my explanations be true,  and the rest more or less probable: no further proof is requi-  site, that the Dacians and Getew were Germans (Dewéssche),  nations of Teutonic race (Deutschverwandte Volker).’ Would  not this argument rather prove the Dacians to be Armenians ? I cannot think that the German resemblances, which his com-  plete mastery of al? the Teutonic dialects at every period of  their existence has enabled the author of the Deutsche Gram- 1 Cf. Arm. waroung, ‘a cucumber.’ 80 THE ARMENIAN ORIGIN OF THE ETRUSCANS.

matik to adduce, ate so many or so plain as the Armenian  language singly affords. If the English reader should not be  acquainted with the Geschichte der Deutschen Sprache, yet he  may apply a tolerably fair test to the question. . Dacia is  claimed on one side as Armenian, and on the other as Teutonic. The country lies midway between Armenia and England, and in England a Teutonic dialect is used. Can the  relics of the Dacian language be explained as well from the English as from the Armenian? If the difference of age  between the English and the Armenian should be urged, the Anglo-Saxon might be used in the place of the English. At  all events, if the Dacian were Teutonic, some one Teutonic  dialect ought to explain it as well as the Armenian does. If  all the Teutonic dialects together cannot do this, there is a  still greater reason for giving the preference to the Armenian.

If the Dacians and Thracians be ranked with the Armenians,  there will be gained in Europe, for the Armenian language, a  territory extending from the Augean to the Carpathians, and  from the Euxine to the frontiers of Pannonia and Illyria. In Scythia, to the east of Dacia, the Armenian seems to have  died away; and to the north, beyond the Carpathians, it is  not likely to have penetrated: indeed, one apparently Lithu-  anian word is found in Dacia itself. In Greece, as in Scythia,  it may have been subordinate to another element, the Hel-  lenic ; and in Asia Minor it does not seem, as a language, to  have passed over Taurus. We have. now to examine whether  it penetrated beyond the Thraco-Dacian area to the west, i. é.,  into Illyricum. Here the language to be compared with the Armenian is the residuary element in Albanian, or that which  remains after the elimination of all Turkish, Greek, Latin, and  other intrusive words. These elements render the task rather ' difficult, as many Albanian words allied to the Armenian are  borrowed from the Turkish and Greek, but especially from the Turkish, which has itself borrowed them elsewhere in its  turn. For instancé—  a THE ARMENIAN ORIGIN OF THE ETRUSCANS. 81. Alb. zaman, ‘ time,’ = Arm. Zam, 4amanak. Alb. zimécle, ‘ basket,’ = Arm. zaméit. Alb. zindan, ‘ prison,’ = Arm. zéndan.

All these Albanian words are borrowed from the Turkish,  which has itself borrowed them from the Persian or Arabic. Whether I have entirely succeeded in keeping such words out  of the following list, I cannot say: a Turkish scholar would  probably detect some additional cases. My Albanian words  are derived from Dr. Hahn’s Albanesische Studien. } Citi  tt “7 Albanian words allied to the Armenian. 1. Ago, ‘God.’ Arm. ogi, ‘spirit.’ 2. At, ayu, ‘he, this (man).’ Arm. ays, ayd, ayn, ‘this,  that.’ The terminations, -s, -d,-n, are borrowed from the  personal pronouns, ¢s, dov, inchn, ‘1, thou, he:’ therefore the  demonstrative pronoun is properly ay. Osset. ay, ‘ this.’ 3. Ala, ala, ‘quickly.’ Arm. watwat, ‘ quickly.’ 4, Amélyé, ‘sweet.’- Arm. hamet, ‘sweet.’ Esth. hamelehh, ‘ agreeable,’ 5. Ané, ‘ashore.’ Arm. aphn, ‘a shore.’ 6. Aré, ‘afield.’ Arm. art, wayr, ‘field, land,’ = Gael. ar. Gr. apoupa. : | 7. Ari, ‘a bear;’ arouske, ‘a female bear. Arm. arg, ‘a  bear.’ Gr. apxroc. : 8. Arnoig, ‘I mend.’ Arm. arnel, ‘to make. Gr. dpw.

9. Apher, ‘near.’ Arm. hovp, kpavor, ‘near.’ Gr. agap.

10. Akhour, ‘a stable” Arm. akhor, ‘a stable,’ = Pers.  dhhir.

11. Vay, ‘woe;’ vai, ‘alas!’ Arm. way, ‘ woe ;’ wdy, ‘alas | Gr. &. Lat.ve. Germ.weh. — 12. Vath, ‘acourt.? Arm. d¢h, ‘a lodging.’ 13. Valyé,‘awave.” Arm. alich,‘awave. Swed. doya, ‘a wave.’ | 14, Vapé, ‘hot.’ Arm. eph, ‘cooking.’ Gr. fiw.

G 82 THE ARMENIAN ORIGIN OF THE ETRUSCANS. 15. Vatré, ‘a fire-place.”” Arm. afr, ‘ fire.’ 16. Véla, ‘a brother.’ Lapp. wea, ‘a brother.’ Arm.  etbayr, ‘a brother.’ 17. Vend, ‘\and” Arm. and, ‘land.’ Welsh guwent, ‘a  plain.’ 18. Vets, ‘except, but.’ Arm. day¥, ‘except, but.’ Pol.  bez, ‘except, but.’ 19. Viyésge, ‘husk, shell. Arm. d/sakn, ‘a covering.’ Rhet-Rom. bleuscha, ‘husk, shell,’ = Pied. plewja, = Insh  blaosg, = Welsh dhag.

20. Viyorés, ‘the name of Aviona and its district’ (Acro-  ceraunia), Arm. d/ovr, ‘a hill.’ 21. Barth, ‘ white.’ Arm. dovr, ‘chalk.’ 22. Bats, ‘a companion.’ Arm. matdéi/, ‘to join one’s  self to.’ 23. Béigj, bafiy, ‘1 make,.I do.” Arm. day, dan, ‘word,  thing.’ 24, Bely,‘aspade.’ Arm. petel,‘todig.’ Gr. réXexve.

25. Bouké, ‘bread.’ See ante in Phryg. (p. 15) s. v. Békoc.

26. Boukhar,‘a chimney. Arm. dowch, ‘a furnace.’ 27. Yam,“Iam.’ Arm. em, ‘I am.’ 28. Gaphorre,‘a crab” Arm. khephor, khetemorth, ‘shell, -  shell.fish ;’ Aheéaphar, ‘a crab ;’ th. khez, khegek, ‘ shell, husk;’  morth, phar, ‘skin ;’ phor, ‘a sea-scorpion.’ 29. Gézoph, ‘a hide” Arm. kashi, ‘a hide;’ she, ‘a shell.” Compare the previous word.

80. Gélér, ‘gain.” Arm. yalthel, ‘to gain.’ 81. Genyeig, ‘I deceive. Arm. nengel, ‘to deceive.’ Gael. gangaid, ‘ deceit.’ 82. Goyé,: “mouth ;’ gogesif, ‘I gape;’ gyouké, ‘tongue,  language.’ Arm. gogel, khésil, ‘to speak ;’ goéel, ‘ to cry out ;’  khésaran, ‘mouth. Germ. kosen.

88. Gorgé, ‘a cave”. Arm. khorod, khor§, ‘a cave, 34. Gyt, ‘breast ;’ gyiri, ‘ relationship.’ Arm. girk, ‘ breast.’ 35. Gyizé, ‘cheese? Arm. kathn, kith, ‘milk, milk-food’ (laitage).

THE ARMENIAN ORIGIN OF ‘THE ETRUSCANS. 83 36. Gyoumé, ‘sleep.’ Arm. chovn, ‘sleep? Gr. trvoc. Lat. somnus, Gael. suain,. Rheot.-Rom. chtina, * bed’ 87. Gyus, gyusmé, gyumés, ‘half? Arm. ké&, kisamasn, ‘half’ (maen, ‘ a portion.’) 88. Dhalé, dhalté, ‘butter-milk.’ Arab. dahl, ‘milk? Arm.  dayl, dal, ‘a cow’s first milk ;’ dayeak, ‘a nurse ;’ dayeakordi, ‘ foster-brother, /rére-de-lait’ (lit. ‘ nurse-son’). Gael. dalta, ‘a  foster-child.’ Irish dairé, ‘a heifer.’ Eng. dairy.) 89. Dham, dhémb, ‘tooth.’ Arm. atamn, ‘ tooth.’ Gr. ddebe. Lat. dens.

40. Dhémpés, ‘rushing. Arm. dimel, ‘to rush,’ th. dém; -pés, Arm. adverbial termination, signifying ‘ like,’ and to be  compared with Germ. -lich, Eng. -/y: e.g. Arm. ays, ‘this,’  ayspés, ‘thus ;’ phovth-al, ‘to haste,’ phovthapés, ‘ hastily.’ 41, Dheré, deré, ‘bitter;’ tharété, ‘sour. Arm. darn, * bitter,’ 42. Dhi, ‘a goat.’ Arm, tk, ‘a goatskin. Georg. thkavi = Pers. takka = Phryg. arrnyoc = Germ. zege, ‘ a-goat.’ 43. Deky, ‘deceit.’ Arm. dav, ‘ deceit.’ 44, Deré,duré, ‘a door.’ Arm. dovrn, dovrch, ‘a door.’ Gr. Stpa, &c.

45. Des, ‘1 die ;’ dekouri, ‘dead.’ Arm, di, diakn, digoyn, ‘dead, a corpse.’ Sansk. dé, ‘perire, evanescere.’ 46. Dzah, zaf, ‘I learn.’ Arm. ganadéel, ‘to know;’ zaneay, ‘IT have known.’ 47. Dimén, dimér, ‘ winter.” Arm. zmern, ‘winter.’ Pol.  zima, ‘winter.’ Turk. zymistan, ‘ winter.’ 48. Dité, ‘day. Arm. tiv, ‘day.’ Sansk. dma, ‘ day.’ Tina, ‘the Etruscan Jupiter or Diespiter.’ 49. Doré, ‘ahand.’ Arm. zern, ‘a hand.’ Gr. 8dpoy, ‘a  hand or palm.’ Gael. dorn, ‘a fist.’ 50. Drouse, ‘perhaps.’ Arm. ¢herevs, ‘ perhaps.’ 1 Grimm remarks (Gesch. der Deut. Spr. p. 1015): ‘die Englander mit  einem wort, das ich nirgends erklart finde, nennen kuhweide und milch-  erei dairy. May it not be a word of Celtic origin P G2 84 THE ARMENIAN ORIGIN OF THE ETRUSCANS. Bl. Déind, ‘a spirit.” Arm. sound, ‘a spirit;’ Indel ‘to  breathe.’ Lat. genius.

52. #, ‘and.’ Arm. ev, ‘and.’ 53. Erré, ‘darkness ;’ errété, ‘in the evening.’ Arm. erek, ‘evening.’ Gr. éoeuvdc.

54. Ere, ‘honour;’ erséi, ‘honourable. Arm. yarg, ‘dig-  nity, esteem ;’ yargel, ‘to honour;’ yargel (participle), ‘to  be honoured.’ The Arm. termin. -el:= Lat. -andus. Germ.  ehre. . | 55. Zyarr, zyarem, ‘fever, heat in fever.’ Arm. germn, ‘heat.’ Turk. gherm, ‘heat.’ Gr. Sepude.

56, Ze, zai, ‘voice.’ Arm. zayn, ‘ voice.’ 57. Zi, ‘black.’ Arm. sev, ‘black.’ Turk. and Pers. stydh, ‘black.’ Osset. sau, ‘black.’ Sansk. ¢ydva, ‘black.’ 68. Zi, ‘grief.’ Arm. Zav, ‘grief.’ Turk. sézé, ‘ grief.’ 59. Z, ‘famine.’ Arm. sow, ‘ famine.’ 60. Zpht, ‘pitch. Arm. zivth, ‘ pitch,’ = Pers. 27/7.

61. Zok, zogou, ‘a bird, a young bird.’ Arm. zag, ‘a young  or small bird” ‘Sparrow’ (Luke x. 6) is rendered in Arm.  by zag, in Alb. by zok.

62. Thagémé, ‘a wonder.’ Arm. thagovn, ‘mysterious ;’  thagovianel, ‘to conceal, to bury,’ = Lapp. wgaket.— Gr. Java.

68. Thekéré, ‘rye” Arm. hagar, ‘rye.’ Turk. dakhl, ‘rye’ Lat. secale.

64. Im, yem, yim, ‘my.’ Arm.im, ‘my? Gr. éude.

65. Ka, ‘an ox.’ Arm. sow = Swed. ko = Germ. suk = Eng.  cow. Sansk. go = Zend gdo, ‘ bos.’ .

66. Kad, ‘an ear of corn.’ Arm. satin, ‘an acorn.’ 67. Kam, ‘I have, I am, I shall,’ (aux. verb). Arm. gom, ‘Tam;’ kam, ‘I am, I continue, I subsist.’ | 68. Kanép, ‘hemp.’ Arm. kaneph, ‘ hemp.’ 69. Karré, ‘a carriage.’ Arm. karch, ‘a carriage.’ Swed.  karra = Eng, cart.

70. Kars, ‘coarse.’ Arm. gar, ‘ coarse.’ THE ARMENIAN ORIGIN OF THE ETRUSCANS. 85 71. Katount, ‘territory, town, village. Arm. getin, ‘land,  ground.’ 72. Kéléph, ‘case, cover”? Arm. hetev, ‘skin, bark, shell.’ Gr. xaXbrrw.

73. Képoutsé, ‘ashoe.’ Arm. kéx#k or haviik, ‘a shoe,’ = Pers. £a/S.

74. Kikél, ‘a summit.’ Arm. gagathn, ‘a summit,’ Lat. _ cacumen, 75, Kodré, ‘a hill” Arm. £atar, ‘an eminence.’ 76. Kol, ‘a party, aband” Arm. sot, ‘ part, side.’ 77. Koran, ‘a trout. Arm. harmrakhayt, ‘a trout:’ th.  karmir, ‘red ;’ khayt, ‘spotted.’ 718. Korité,‘aboat.? Arm. hovr, ‘a boat.’ 79. Koské, ‘a bone.” Arm. oskr, ‘a bone.’ Pol. hosé, ‘a  bone.’ Lat. 0s. Gr. dcréov.

80. Kotse, ‘a girl’? Arm. foys, ‘a girl? Turk. iz, ‘a ‘girl, .

81. Kouth,‘a pot.’ Arm. kové,‘a pot.’ Gr. yérpa.

82. Krot8, ‘an ass’ colt’ (Jowrriquét). Arm. grast, ‘an ass’ (Lourrique) . : 83. Kyendis, ‘I choke.’ See ante in Lyd. (p. 40) 8. v.  xavSabAnc, ‘oxvAArorvlerne? L 84. Kyilis, ‘I roll’ Arm. glel, glorel, ‘to roll;’ gla,  glanak, ‘a cylinder.” Gr. xvAiw, xvAlvdw.

85. Lap, ‘I lick.’ Arm. lapf, ‘licking; Japhel, ‘to lick.’ Gr. Aclyw.

86. Lezoues, ‘a reader.” Arm. lezov, ‘tongue, language.’ Gr. Afyw. Lat. lego. Lezoues is declined as a substantive,  not as a participle (Bopp om the Alb. p. 27). It may be  compared with Arm. substantives which add -7¢ to the root;  as phrk-el, ‘to deliver,’ phrk-ié, ‘a deliverer ;’ arar-ch, ‘ crea-  tion,’ arar-ié, ‘a creator ;’ tarr, ‘substance, nature,’ ¢arr-zé, ‘a maker, a creator ;’ tovié, ‘a giver. So there might be  formed from J/ezov, ‘language,’ an Arm. subst. lezovid, ‘a  speaker,’ which is nearly identical with the Alb, /ezoues. The 86 THE ARMENIAN ORIGIN OF THE ETRUSCANS.

declension of J/ezoue§ is—nom. and acc. Jezoue§; gen., dat.,  and abl. Jezoue#. The deelension of tovid, is—nom. tovid ;  ace. (z)fovié ; gen. and dat. fovd: ; instr. tovéav.

87. Iidsé, ‘a warm-spring.’ Arm. dg, ‘a pond.’ Lat.  liquor, liz.

88. Los,‘aclub.’? Arm. lakht, ‘aclub;’ tek, ‘anoar.’ Eng.  log. lat. lignum.

89. Lyargé, ‘far’ (weit), Arm. larel, ‘to stretch, to extend.’ Lat. largus.

90. Lyaré,‘a bath.” Arm. logaran, ‘a bath;’ lovanal, ‘to  wash, to bathe.’ Lat. lavacrum, lavare. Lapp. laugo, ‘a  bath.’ 91. Lyarté, ‘high;’ also, ‘an eminence.’ Arm. darn, ‘a  mountain ;’ lernaym, ‘very high.’ 92. Lyé, lyah, ‘I leave.” Arm. lich, ‘left;’ chanel, ‘ to  leave.’ 93. Lyes, ‘a corpse.” Arm. leskamash, latkamash, ‘ skin’ (mask, ‘ skin, tegument, membrane’); /agkar, ‘a body of men.’ Kurd. /esc, ‘corpse.’ Germ. /icke.1 In Pers. la&k signifies ‘a piece,’ and /aékar, as in Arm., ‘a body of men, an army.’ 94, Lyith, ‘I fasten.” Arm. dzel, ‘to fasten.’ 95. Lyind, ‘I give birth to.” Arm. dinel, “to be, to. become,  to live;’ etanil, ‘to be made.’ 96. Mayé, ‘top, Arm. mak, ‘above.’ Esth. mdggi, ‘mountain.’ 97. Math, ‘great.’ Arm. mez, ameth, ‘great.’ Gr. yéyac.

98. Mattouné, mayouné, ‘rich.’ Arm, mezatova, ‘rich.’ 99. Marr, ‘ich fasse.” Arm, marzel, ‘to form.’ 100. Méar,‘Icarry.’ Arm. Jerel, ‘to carry.’ Gr. déow, &e.

101. Méarg, mba, ‘TL hold.’ Arm. pakel, ‘to retain.’ ~ 102, Moaré, ‘good.’ Arm. bari, ‘good.’ Lapp. puore, é good,’ 108. Myekré, ‘a beard.’ Arm. méroych, ‘a beard.’ 1 Or we might compare the Teutonic fleisc, flask, flesh, flesh. THE ARMENIAN ORIGIN OF THE ETRUSCANS. 87 104. Myergoulé, ‘a mist.2 Arm. marakhovt, mrayl,=‘a  mist.’ Lapp. murko, ‘a mist.’ 105. Melingoné, mermink, ‘an ant. Arm. mrgivn, ‘an ant,’ = Osset. mdlziig, Gr. pbppné, Swed. and Ang.-Sax. myra, Irish moirb, Pers. mur, mirdk.

106. Mérgoig, ‘I put away.’ Arm. meréel, ‘to put away.’ 107. Mis, ‘flesh, palp.? Arm. mis, ‘flesh, pulp.’ Eng.  meat.

108. Mouré,‘ north wind.’ Arm. mrrik, ‘ high wind, storm.’ Sansk. marut, ‘ ventus.’ 109. Mourk, ‘dark, black” Arm. mrel, ‘to blacken.’ Eng. murky. | 110. Mout, ‘dung’ Arm. miz, ‘filth’ Germ. mist. Eng. muck, mud.

11]. Mukouré, mukouné, ‘ grey.” Arm. mokhragoyn, ‘ grey.’ 112. Napht, ‘property, goods.’ Arm. nivih, ‘matter, sub-  stance.’ , 118. Nge, ‘leisure. Arm. ninf, ‘repose.’ Lapp. nao, ‘laxamentum.’ Rhet.-Rom. zanna, ‘cradle, bed.’ 114. Noyir, ‘depth. Arm. nerchin, ‘low. Gr. vépSe.

115. Ndé, ‘into, to.’ Arm. énd, ‘into, to.’ Gr. tvdov.

116. Oborr, ‘a court. Arm. aparanch, ‘a court.’ .

117. Ouyé, ouyété, ‘water. Arm. hwth, vivth, ‘ water ;’  wét, ‘a wave;’ wiak, ‘a brook.’ Lat. ydus. Gr. ddwp. Phryg. (édv, ‘ water.’ 118, Oulé, oudé, ‘a way, a road.’ Arm. ovlt, ‘a way, a  road,’ = Turk. yol, = Lith. ulyczia. Gr. bdd¢. Cf. "Odvaceic  and Ulysses.

119. Oufsith, ‘a bowl, a pot.? Arm. andth, ‘a vessel, a pot.’ 120. Ourth, ‘ivy.’ Arm. orth, ‘a vine, a basket.’ Gr.  dpxoc¢: th. eipyw. See m/f. in Etruse. 8. v. atesum, and come  pare also Arm. ovr, ‘a vine-branch,’ ovri, ‘ willow, osier,’  ovrkan, ‘a net;’? and Lat. vimen, vitex, vitis, vinea, vieo, and  vincio.

121. Ousta, ‘skilful? Arm. oviim, oviatov, ‘ wise, careful.’ 88 THE ARMENIAN ORIGIN OF THE ETRUSCANS. 122. Ougt, ‘invocation against the evil eye.’ Arm. ovkit, ‘vow, prayer.’ Gr. evyf.

123. Pa, ‘un’- (privative prefix.) Arm. ap-, apa-, api-, ‘un-.” Gr. az-, amo-.

124, Paké, ‘little’ Arm. pakas, ‘ deficient;’ phockr, ‘little.’ Lat. paucus. Gr. waivpoc. Ital. poco.

125. Paté, ‘a goose.’ Arm. dad, ‘aduck.’ Arab. wazz, ‘a goose, a duck.’ Compare anas and anser.

126. Pyesé, ‘a part, ashare.’ Arm, baé, bazin, ‘ part, con-  tribution.’ Lapp. pekke, ‘a piece.’ 127. Pégéré, ‘impurity, dirt.’ Arm. pagéot, ‘foul.’ Lat.  lex. Lapp. pddo, ‘ sordes.’ 128. Pelyky, ‘I defile.” Arm. pizel, ‘todefile. Eng. filth.

129. Pengoif, ‘I bind.’ Arm. padel, ‘to bind.’ 130. Poungt, ‘a fist ;? poupé, ‘a bunch, a tuft? Arm.  phovng§, ‘a bunch, a tuft.’ Lat. pugnue.

181. Prassé, ‘a leek.’ Arm. pras, ‘a leek.’ Gr. xpacov. 182. P&kyé, ‘a bladder” Arm. phéel, ‘to blow;’ phchal, © ‘to be inflated ;’ phovch, ‘wind, gust, swelling;’ pho, ‘powder.’ Gr. gucaw, dionua. Lapp. puseet, ¢ flare.’ 188. Paul, ‘a forest.’ Arm. povrak, ‘a forest.’ 134. Rouphé, ‘drinkable, soft, schliizfbar (applied to an  egg). Arm. rovph, ‘syrup of grapes.’ 185. Samar, ‘a pack-saddle.’ Arm. thambel, ‘to saddle.’ 136. Siri, ‘sex, kind.’ Arm. ser, ‘ sex, kind.’ 137. Skalis, ‘I lop, I carve.’ Arm. setd, ‘the blade of a  knife.’ Thrac. axéApn, ‘a sword.’ 138. Skyuré, ‘a plate.” Arm. skoviet, ‘a plate.’ Lat.  scutra, scutella.

139. Soudyts, ‘a bolt, a bar.’ Arm. zatk, ‘a rod, a stick ;’  zot, ‘a bar, a pole;’ sotnak, ‘a bolt.’ 140. gus, ‘I dip” Arm. sovzanel, ‘to dip; sovzi, ‘I  dipped.’ 111. sami, ‘a kerchief’? Arm. warsamak, warsakal, ‘ ker-  chief, fillet’ (warsch, ‘ hair.’) THE ARMENIAN ORIGIN OF THE ETRUSCANS. —-89 142. selkyii, ‘a melon.’ Arm. setkh, sekh, ‘amelon, Gr.  oixue.

143. Sowalé, gout, ‘sad.’ Arm. sovg, ‘mourning, grief.’ 144, Sour, ‘sand, gravel.’ Arm. éor, ‘ dry.’ 145. Sourré, ‘water.’ Arm. govr, ‘ water.’ 146. Tayé, ‘a wet-nurse.’ Arm. ¢azel, ‘to nourish.’ 147, Toké, ‘land’ Arm. yatak, ‘land.’ 148. Tserlé, ‘a blackbird. Arm. sarik, ‘blackbird, star-  ling ;? th. sarel, ‘to cry.’ Thessal. aorpadrde, ‘bapde.’ | 149. Tsingouly, ‘a kind of small fig’ Arm. ¢hovz, ‘a fig;  thzent, ‘a fig-tree.’ 150. Tsinkéré, ‘ice Arm. Zin, ‘snow.’ Lapp. ésewe, ‘nix. durior, que subtus crustam habet glacialem.’ Abas.  zeh, ‘ snow.’ 151. Zsttsé, sisé, ‘breast.? Arm. oz, ‘breast.’ Germ.  zitze. Pol. cye.

152. Tsoulyé, ‘flute, pipe.’ Arm. sovlel, ‘to pipe.’ Lat.  siiilare. 7 158. Tsokhé, ‘cloth? Arm. éovkhay, ‘cloth,’ = Turk. choha, Pol. chustka, Germ. tuch.

154. Tsap,‘astep.? Arm. daph, ‘a measurement, an ell’ ——Arm. savit, ‘a step.’ 155. Téark, ‘a circuit.? Arm. sovr§, ‘around; srgan, ‘a  circuit.’ Lat. circus.

156. THké, ‘a little, few. Arm. sakav, ‘a little, few.’ Basque chiguia, = Span. chico, ‘ little.’ 157. Ul, ‘astar” Arm. Sol, ‘a ray of light Lat. so/,= Welsh s#/, = Swed. sol.

158. Uliber, ‘a rainbow ;’ ap ultin, ‘I bend’ (ap, ‘I give,’  or ‘I give up’). Arm. ateén, ‘a bow.’ 159. Phouphoupheiké, ‘an owl.’ Arm. dov, ‘anowl.’ Lat.  bubo.

160. Phtoua, ‘a quince’ (ujAov Kudwuov). Arm. péovt, ‘fruit’ (uiAoy).

161. Cha, ‘Ieat. Arm. gaégel, ‘to eat ;’ shah, ‘cooked meat.’ 90 THE ARMENIAN ORIGIN OF THE ETRUSCANS. 162. Chapété, ‘ bright in colour’ (yAauxdéc). Arm. hapoyt ; Pers. kabéd ; Sansk. kapota ; ‘blue, azure’ (yAauxdéc).

168. Chi, chifi, ‘ashes.’ Arm. agivn, ‘ashes.’ Lat. cinzs. Lapp. una.

164, Choundé, ‘the nose.’ “ Root, chouny or owny (?), ‘ to  put down, to bend.’” Arm. ovnéch, ‘the nose ;’ ovng, ‘the  lowest part, the foot, the ground.’ Odsset. ftinz, ‘the nose.’ Abas. pintsa, ‘the nose,’ The Rhsto-Romansch language is the next in order. It is  spoken, in two principal dialects, in the Swiss Canton of the Grisons, and is almost entirely derived from the Latin and German. My authority for these dialects is Carisch’s Taschen- Worterbuch der Rhatoromanischen Spracke. Rhato-Romansch words allied to the Armenian. 1. Accla, ‘a farm (gué) with stabling (s¢al/ung) outside a  village.’ Arm. akhor; Pers. dhhdr ; ‘a stable.’ 2. Adimna, ‘always.’ Arm. andén, ‘always, immediately.’ 8. Adom, ‘a fastening’ (b¢festiguagering an den schlitten-  latten). Arm. yédel, ‘ to fasten.’ 4, Aegla. ‘Boschg (‘shrub’) d’dgla,’ ‘holm-oak. Arm.  katin, ‘an acorn,’ = Lyd. adxvAov, Germ. ezchel. Cf. also Lat.  galia, Span. agalla, 5. Assel, ‘active.’ Arm. ays, ‘ spirit,’ 6. Ancanuras, ‘sometimes.’ Arm. angam, ‘ time, fois, mal, 7. Ancarna, ‘corner, angle.’ Arm. ankivn, ‘acorner.’ Lyd.  aykov. Germ. winkel.

8. Aneg, ‘suddenly.’ Arm. anaknkal, ‘sudden :’ an-, priva-  tive; akn, ‘an eye;’ kal, ‘tobe;’ aknkalel, ‘ to expect.’ Goth.  anake, ‘suddenly.’ 9. Anfits, ‘disagreeable.’ Arm. anpét, ‘ useless, vile, detest-  able :’ an- privative ;. prii, ‘il faut” Eng. unfit.

10. Araig, ‘a king.’ Arm, archay, ‘a king.’  righ = Lat. rea, Gael. THE ARMENIAN ORIGIN. OF THE ETRUSCANS. 91  ll. Arftidar, ‘to reject? Arm. phoyth, ‘haste; phovthal, ‘to hasten.’ Gael. put, ‘to push, tothrow.’ Lyd. ap¢éravov, ‘aquoit”  Piedm. az/ude, ‘to reject.’ Lat. repudio. Eng.  refuse, refute.

12. Arguavenna, ‘bear’s-foot’ (acanthus). Arm. ar§, ‘a  bear ;’ aph, ‘ the palm of the hand’ (?).

18. Asitenf, ‘usury.’ Arm. wasch, ‘ usury.’ 14, Asdl, asoula, ‘a kid’ (ztegenlamm). Arm. ayz, ‘a goat;’  ovl, ‘a kid.’ 15. Avdar, abitar, ‘to dwell.’ Arm. oth or avth, ‘a dwell-  ing.’ Lat. habséare.

16. Aztever, ‘to fetch.’ Arm. azel, ‘to bring.’ 17. Baguord, ‘ugly. Arm. pagéot, ‘ foul,’ 18. Bambesch, ‘cotton.’ Arm. bambak, ‘ cotton” Ital.  bambagia.

19. Bap, ‘father, grandfather.’ Arm. pag, ‘ grandfather.’ 20. Bargalir, ‘to rise.” Arm. barzr, ‘high.’ 21. Bassiar, bassegiar, ‘to make haste.’ Arm. wazel, ‘to  hasten.’ Lyd. Bdoxe, ‘ 2EeOdade.’ 22. Basta, ‘enough!’ Ital. basta! Arm. bdo & ‘enough!’ The first element in da-sta would= Arm. éav, sta being = Lat. stat. So also the Lat. saf= Arm. sat, as well as Gael.  sath, .

28. Blada, ‘the consecrated wafer.’ Arm, dlth, ‘bread,  cake.” Gael. dleath, bleith, bleth, ‘to grind.’ 24. Bletsch, ‘wet.’ Arm. btkhil, ¢ to flow.’ 25. Boda, ‘pestilence.’ Arm. wat, ‘bad.’ Lith. béda, ‘calamity ;’ odus, ‘ troublesome.’ 26. Brainta, ‘mist.’ Arm. dovrel, ‘to exhale, to smell ;’  bovrovmn, ‘odour.’ Lyd. BpévOtov, ‘ pbpov.’ 27. Brél, ‘an orchard.’ _Arm. der, ‘fruit ;’ dovrastan, ‘ an  orchard.’ 28. Bural, ‘an orifice. Arm. Jderam, ‘orifice, mouth.’ Lith. durna, ‘ mouth,’ 29. Biisen, ‘a reed;’ biischen, ‘water-pipes;’ duschun, 92 THE ARMENIAN ORIGIN OF THE ETRUSCANS. ‘strong.’ Arm. doveck, ‘a plant;’ doveanil, ‘to grow, to rise.’ Lapp. and Swed. Jasun, ‘ tuba.’ 80. Candarials. See anfe, in Lyd. (p. 40) s. v. cavdadAne.

81. Cattar, ‘to find.” Arm. gtanel, ‘to find;’ givi, ‘inven-  tion.” Eng. get.

$2. Chuz, ‘ destitute.’ Arm. échott, ‘ poor, abject.’ 88. Clech, ‘amiable, tender.’ Arm. &tkath, ‘loving, affec-  tionate.’ : 84. Crap, ‘astone. Arm. char, kharak, ‘rock, stone.’ 85. Creppa, ‘a skull.’ Arm. karaphn, ‘skull, head.’ Ital.  greppa, ‘a mountain-top,’ = Welsh crib. Gr. xdpa. Cf. Car-  pates Montes.

86. Cuosp, ‘a wooden shoe.’ Arm. kégk, ‘a shoe.’ 37. Custer, ‘near? Arm. hoys, kovit, ‘side.’ Lat. costa.

88. Cuzzantar, ‘to preserve.’ Arm. zgovianal, ‘se garder.’ 89. Daya, tegen, degien, ‘a dagger.’ Arm. dasoyn, dainak, ‘a dagger ;’ dakovr, dakr, ‘an axe;’ tég, ‘aspear.’ Esth. tazi, ‘pike, sword.’ Germ. degen. Lyd. reyobv, ‘Ayorhv.’ 40. Disch, ‘length of time.’ Arm. tich, ‘age;’ tiveh, € days.’ | 41. Dischél, ‘ indigestion.” Arm. dé-=Gr. duc-; al, ¢ digestion.” 42. Dondagiar, ‘to waver, to reel.’ Arm. dandaéel, ‘to  waver, to reel.’ 48. Euna, ‘a week.’ Arm. evihn, ‘seven.’ Alb. yavé, ‘a  week.’ 44, Fadigna,‘a bough.’ Arm. phayt, ‘wood, tree.’ Gael.  frodh, ‘timber, wood.’ 45, Fale, ‘hoary.’ Arm. dal, ‘paleness, obscurity.’ Lat.  pallor. , 46, Fassui, ‘a pickaxe.’ Arm. phosel, phaparel, ‘to dig.’ Lat. fossa.

47. Fecht, ‘desire.’ Arm. phaphach, ‘ desire.’ 48. Fig, fick, ‘very;’ fig fig, ‘extremely. Arm, yoyé, ‘very ;’ yoyé yoyé, ‘ extremely.’ THE ARMENIAN ORIGIN OF THE ETRUSCANS. 93: 49, Foppa, ‘a hollow, a ravine. Arm. phap, phapar, ‘a  hollow, a cave, a hole.’ Lapp. /woppe, ‘ angustia.’ 50. Friank, ‘an asylum’ (freistdtte). Arm. phrkel, ‘to  free ;? phrkanch, ‘deliverance.’ Briges, or Phryges, signified ‘free’ in Lydian.

51. Ga, gada, giada, ‘time, fois, mal” Pers. gah, ‘ time.’ Arm. gam, ‘time, fois, mal ;’ gizak, ‘ time, temps, zeit,’ = Lith.  gadyne.

52. Gtantar, jentar, ‘to dine.’ Arm. gaégel, ‘to dine;’ Jamé, ‘food.’ Thrac. yévra, ‘meat.’ Span. yanéar, = Basque jan, “to eat.’ Lat. yentaculum.

58. Giever, ‘at least.’ Arm. géth, ‘at least.’ .

54. Gig, ‘long.’ Arm. gagathn, ‘height.’ Gr. ylyae. - 55. Gitti, ‘avaricious.’ Arm. 4222, ‘avaricious.’ Germ. geizig.

56. Ghwrer, ‘to finish.” Arm. glkhel, glkhavorel, ‘to finish.’ 57. .Gniff, muzzle.’ Arm. kagith, ¢ muzzle.’ 58. Gniocc, ‘a ball of paste.’ Arm. gndak, ‘ ball, pellet.’ 59. Guis, ‘a marten.’ Arm. kovz, kzenachis, ‘a marten, a  pole-cat.’? Pol., Serv., and Bohem., kuna, Russ. kuniza, Lith.  kiaune, ‘a marten.’ The Rhet-Rom. guis, though partly  allied to the Slavonian, can be derived, as far as I am aware,  from no European language. The Lat. for ‘marten’ is mar-  tes ; the Basque, martea ; the Germ., marder; the Fr., martre  or fouine ; the Gr., yadén; the Gael., caghan or neas ; the Lapp.,  mart or neete; and the Esth., xuggis, in which we find the  termin. of the Arm. kz-nachis. See also ante, p. 11, in Cappad.  s. v. pnette. The Slavonic ku-na and ku-niza may not im-  probably be equivalent to the Arm. kznachis, = kovz-nachis ; in  which case, one of the two Arm. words would appear to have  travelled westward with the Thracians into the Grisons, and  the other to have been taken up by the Sarmatians. The root  of the Arm. fovz and Rhet-Rom. guts may perhaps be found  in the Arm. Lhovzel, ‘ to seek, to ferret out (fureter).” Thus  the origin of a word used at the source of the Rhine is found  at the source of the Euphrates. 96 THE ARMENIAN ORIGIN OF THE ETRUSCANS.

last member would signify * pine,’ and the, ‘a tree,’ would be  found in ¢fi, ‘ an oar.’ 94. Tutta, ‘a kind of reed, of which children make squirts. ? Arm. thathar, ‘a siphon, a spout.’ 95. Uettar, ‘to anoint.’ Arm. dzanel, ‘to anoint ;’ th. éz. 96. Umblauna, ‘ptarmigan, (schneehuhn). Arm. amayi, ‘desert ;? ameyi, ‘ wild’ (?): atavnz, ‘a pigeon,’ = Osset. baldn. Lat. palumbes. Another word for ‘ ptarmigan’ is arblauna (weiss-hukn). It does not appear whether we should divide  into ar-blauna and um-dlauna, or into arb-launa and umb-  launa. 97. Verr, ‘a wild boar.’ Arm. waraz, ‘a wild boar,’ = Russ.  wekr, = Sansk. vardha. Lat. verres. 98. Verscka, uerscha, ‘the crown of the head. Arm. wer, Sabove;’ wer§, ‘end.’ Lat. vertex. 99. Zaima, ‘a drinking-glass.’ Arm. sam, ‘urn, kettle.’ Pied. sana, ‘a drinking-cup.’? Lapp. saja, saun, ‘situla.’ O. Pers. cavvaxpa, ‘acup.’ Thrac. cavara, ‘drunk,’ 100. Zapp, ‘a step;’ zaplida, ‘a sledge-road little used.’ Arm, savit ; Arab. sabi] ; ‘path, footstép’ (rpfBocg, Mark i. 8.) 101. Zezna, ‘dung.’ ~Arm. zazir, ‘foul;’ zazrovthiva, ‘ordure;’ zazaz, ‘ vile.’ 102. Zottla, ‘a wagtail.’ Arm. zit, ‘a sparrow;’ zizarn, ‘a swallow ;’ zizarnovk, ‘a nightingale’ or perhaps better  from zet, tovin, ‘a tail,’ and dotal, ‘to wag.’ Pers. sisdlak, ‘a  wagtail.’ 103. Zuncla, juncla, ‘leather thong, latchet ;’ sua, suga, ‘cord.’ Arm. éovan, ‘ cord, string.’ (‘The last, and most important, of the languages to be con-  sidered, is the Etruscan. Its relics consist of: (1) words of  which the meaning has been given; and (2) inscriptions. The interpreted words will claim the first notice: Iam in-  debted for them to the collection of Etruscan words in Dr Donaldson’s Varronianus. : THE ARMENIAN ORIGIN OF THE ETRUSCANS. 97 ° Etruscan Words. 1. Aisar, Deus?’ aicof, ‘ Seof.’? Gael. aos, ‘fire, the sun, God ;? Aosar (= Aos-fear), ‘God.’ Insh Aosar, Aesar, ‘ God.’ O. Norse ds, ‘deus.’ Arm. ays, ‘spirit, demon;’ xesar, * demon.’ - 2. Agalletor, ‘waic’ Gael. og, ogail, ¢ youthful ;? oglach, ‘lad ;? ogalachd, ‘youth.’ ‘Welsh og, og/, ‘fall of motion or  life, young’ (cf. dxbe, agilis) ; ogledd, ‘fulness of life or motion.’ Arm. ogi, ‘spirit ;’ ogeliz, ‘full of spirit.’ Lith. waikas, wai-  kélis, ‘a child.” Georg. akhah, ‘young. Turk. oghul, ‘son;  oghlan, ‘boy.’ Alb. dyallyé, ‘ child ;’ dyallythi, ‘lad.’ Agal-  letor seems = Gael. ogalachd with the Gael, termination -or,  and to the Welsh ogledd with the Welsh termination -wr, 8. Antar, ‘aeréc.’? Arab, and Pers. zasr, ‘ vulture, eagle.’ Arm. angt, ‘ vulture ;’ anzet, ‘jay.’ 4. Anta, ‘ dvenor?’ Andas, ‘ Boreas.’ Arm. anzn (in compos.  anz), =O. Norse ande, = Gael. anam, = Lat. anima. Lat. ventus. Germ. wind. Gr. dveuog. Antar, ‘an eagle,’ may be allied  to andas, as aquila is to aquilo. | 5. Aracus, ‘iépak.? Arm. arag, erag, ‘swift,’! = Gr. apyhe, = Sansk. ara, = Lapp. arwok ; aragil, ‘a stork or heron (ciconia  seu erodius) ;? arsin, ‘stork, falcon;’ argv, ‘an eagle,’ = O. Pers. dpE:poc, = Goth, ara, = Lapp. arts, = Lith. arélis ; ori, ‘a sparrow-hawk;’ ovrovr, ‘a kite.’ Aracus is the name of an Armenian in the Behistun inscription.

It appears rather probable that aragil, which signifies ‘stork’ in Armenian, or some word very like aragil, signified ‘falcon’ in Lydian. For, in that language, 3a0-vppnyaAn sig-  nified ‘ txrivoe’ (ante, p. 89). Botticher compares Ba§- with  the Arm. dad, ‘anas,” and adduces the German name of 4  bird of prey, entenstisser, We have a kind of hawk called a 1 Yonxe éocxws  dnd, paccopévy, Sor’ Gxuoros werenvav,—Il. xv. 237, H 98 THE ARMENIAN ORIGIN OF THE ETRUSCANS.

hen-driver, as well as a sparrow-hawk. BabuppnyaXn may be  equivalent to duck-hawk.

6. Arimus, ‘ rlOnxoc.’ Arm. ayr, ‘vir, homo;’ art, ‘ virile,  bold,’ é. e. ‘resembling man.'——-Heb. charum, ‘ simus.’ 7. 8. Arse verse, ‘averteignem.’ Lat. arceo. Gr. apxéw. Arm. argel, ‘ hindrance, obstacle, opposition.’ Arm. herchel, ‘to repel.’ Alb. err, ‘to keep off.’ Irish fursan, ‘ flame.’ Gael. dusrseach, ‘flame.’ Arm. hovr, war, borboch, ‘ fire.’ Germ, Jeuer. Gr. wp.

9. Atasum, ‘avadtvepac.’ Pied. autin, ‘a vine.’ Arm.  yddel, ‘to bind.’ Gael. iadh, ‘to bind.’ This would give the  first syllable of at-eswm ; while -esum may perhaps be ex-  plained from the Arm, aygi, ‘a vine ;’ Pers. ééakh, ‘a cutting  of a vine;’ Pehlvi as, asia, ‘wine;’ asiav, ‘a vine! Compare Fr. Her and ierre; Arm. batel, ‘to bind,’ batetn, ‘ivy ;’ and Gael. iadh, ‘to bind,’ eidheann, ‘ivy.’ These instances may  illustrate the derivation of vitis from veo, and of autin and  atesum from yédel or yavdel. So also the Germ. reée is con-  nected with the Ang.-Sax. repen, ‘to bind’ (Diefenbach, Les. Comp. v. ii. p. 163).

10. Balteus, ‘the military girdle.’ Eng. belt, baldric, = Iceland. de/ti, = Gael. dalt. Gael. bolt, ‘lacinia, ora, margo.’ Arm, dolor, ‘garland, circle ;? dolorel, ‘ to gird ;’ batel, ‘ to join.’ 11, Burrus, ‘xav@apog.’ Arm. bor, boret, ‘a gad-fly ;’ bzé,  bziz, ‘a beetle ;’ dzzel or bézzel, ‘to buzz’ (= Fr. bourdonner), Thus the primitive meaning of durrus would be ‘the buzzer,’ ‘the beetle’ that ‘ wheels his droning flight,’ ‘the shard-borne  beetle with his drowsy hums.’ Alb. vousé, ‘a dung-beetle ;’  bourkth, ‘a cricket.’ Pers. duzk, ‘a kind of cantharides, a  cricket, a rose-worm.’ Turk. dujik, ‘a beetle, any creeping  thing.’ Gael. durruzs, ‘a caterpillar.” Rhet-Rom. dau, ‘a  small beetle or insect.’ Georg. duzi, ‘a fly.’ 1 Hits is the name of the vine in Imeretia and Mingrelia, the ancient Colchis. Parrot, p. 274. The vine climbs there to the summits of high  trees, and then hangs down to the ground. THE ARMENIAN ORIGIN OF THE ETRUSCANS. 99 12. Capra, ‘alé.” Welsh gafyr; Bret. gavr ; Gael. gabhar ; O. Norse ha/r ; ‘a goat.’ Lapp. debra ; Arm. chésh or chavsh ; ‘ hircus.’ 18. Capys, ‘falco. Arm. gavaz, ‘a small sparrow-hawk.’ Gael. cabhar; Lapp. hapak, hapke, hauka ; Germ, habicht ; Welsh hebog ; ‘a hawk.’ 14, Cassis, ‘helmet or casque,’ properly capsis. Gael. cap,  ceap; Bret. kab; Welsh cop; ‘head.’ Lat. caput. Gr.  xeparh. Germ. hop, haupt. Lapp. kaw; Esth. kapo; ‘a  hat.’ Eng. cap. Germ. haube. Fr. casquette.

15. Celer, ‘swift.’ celeres, ‘equites.’ Pers. gelak, ‘ valde  agilis.’ Arm. shatal, ‘to go, to leap;’ chaylel, ‘to walk,’ Gr. xéAne, eéAAw. Alb. haly, ‘a horse ;’ kalyéri, kalyori, ‘riding ;’ Aalyores, ‘a rider.’ Carian ada, ‘ equus,’ = Gael. al.

16. Damnus, ‘trroc.’ Lapp. témp, ‘ equus.’ Fin. tamma ; Arm. zambik ; ‘equa’ (yument). Basque zamaria, ‘jumentum  vectorium, caballeria.? Arm. thambel, ‘to saddle;’ haméerel, ‘to bear? . Cf. onus and voc. Alb. samaros, ‘jumentum.’ Pehlvi djemna, ‘a camel.’ Mantschu femen, ‘a camel.’ Old Mexican ¢amanes, ‘ carriers.’ Gael. damh, ‘an ox, a stag.’ 17. Druna,‘apxh. Irish dron, ‘right ;’ dronadh, ‘ direc-  tion ;’ dronain, ‘throne.’ Gr. Spdvoc. Arm. té, ‘ sove-  reign ;? ¢érovni, ‘ belonging to the Lord.’ 18. Falandum, ‘celum.’ Lat. palatum, palam., Gr. gada-  péc, = Arm. phaylovn. Pers. buland, ‘high ;’ béland o pusht, ‘ high and low, heaven andearth.’ Pehlvi deland, high.” Arab. falak, ‘the sky.”—— Lapp. palwa, ‘nubes.’ Gael. JSailbhe, ‘the aerial void.’ 19. Februwm, ‘inferum.’? (Also Sabine). O. H. Germ.  furljan, ‘purgare.’ Eng. furbish, Fr. fourbir. Ital. fordire. Ang.-Sax. feormian, ‘purgare.’ Gael. foirfe, ‘ without fault ;’  fior, ‘pure ;’ feabh, ‘good; feabhas, ‘beauty, goodness, im-  provement.’ 20. Gapus, ‘dxnpua.’ Gael. cap, ‘acart.’ Lat. capsus. Gr.

carla, kaTrayn.

H2 100 THE ARMENIAN ORIGIN OF THE ETRUSCANS.: 21. Gunis, ‘yfpavoc.’ Botticher considers that Lat. auvca = Sansk. veka, ‘ grus,’ a clangore. So ginis may be considered = xhv and gans, and be compared with O. Norse gima, = Germ.  géthnen, =Gr. yalvw, and Lat. cano, In Arm. we have ganéel, ‘to cry;’ handel, ‘to call;’? and khanéel, ‘to yelp,’ = Lat.  gannire. .

The names of the crane seem generally to be derived a clan-  gore. Thus yépavog and grus may be referred to ynpéw, Lat.  garrio, = Sansk. érus ; the Germ. kranich to krahen ; the Gael.  garan to gair, ‘a shrill cry ;? and the Arm. shord and krnkan,  to khordal, ‘r&ler,’ and Arnéel, ‘ crier.’ 22. Hister, ‘ludio.” Arm. hest, ‘easy, light, pleasant ;’  hestalt, ‘agreeable, diverting ;’ Aegtalovr, ‘supple;’ heganal, ‘to amuse one’s self.’ 23. Itus, ‘idus.’? Irish zr or tdir, ‘between.’ Alb. th, ‘through.’ : 94. Lena, ‘a double cloak’ Gael. deine, ‘a shirt, a  shroud, a kind of dress.’ Lat. /ana. Pers. dayn, ‘indusium  breve angustis manicis.? Gr. yAaiva, yAavue. Arm. khiay,  lidik, ‘a cloak” lat. lodiz Swed. klade. Germ. fled. ‘Eng. cloth, clothe. | 25. Lanista, ‘gladiator. Irish Jann, ‘a sword.’ Welsh  llafyn, ‘a blade.” Lat. lanto, lamina.

26. Nanus, ‘mrAavarng.’ Arm. neng, ‘ craft, deceit ;’ nengel, “to deceive’ (xAavaw) ; nanir, ‘vain, futile,’ 27. Nepos, ‘luxurious.’ Pers. nefsant, ‘ luxuriosus, carnalis.’ -Arab. and Turk, efis, ‘delicatus.’ The th. is the Arab. xe/s, = Heb. nefeé, ‘ breath, soul, animal, body, flesh.’ Pol. napazaé, ‘to inspire, to give to drink ;’ napagé, ‘to glut.’ Alb. nepés, ‘aglutton.? Arm. nivth, ‘body, matter.’ .

28. Subulo, ‘tibicen.” Goth. svilja, ‘pfeifer, atAnrie.’ ‘Upper Germ. schwibeln, ‘ pfeifen.” Welsh gwd, ‘a pipe.’ Lat.  sibilare, = Pers. siflidan, = Arm. sovlel. Alb. tsoulyé, ‘a flute.’ .-Bhet-Rom. schiblot, ‘a flate.’ 29, Vorsus, ‘a measure of land one hundred feet square.’ THE ARMENIAN ORIGIN OF THE ETRUSCANS. 101 (Also Umbrian). Pers. warg, ‘a digging, an area round 4  house ;’ warz, ‘agriculture, a field having a raised border  round if.’ Arm. wayr, ‘ space, extent, piece of ground ;’  warel, ‘ to cultivate, to prepare ;” wargel, ‘to prepare.’ Com-  pare acre, acker, and ackern. : L (For the explanation of the words contained in this Etruscan  vocabulary, two languages are indispensable, the Armenian  and the Celtic. Now the Etruscan nation in Italy was in all  probability made up of two elements, the Tyrrhenian and the Umbrian. To the latter of these we may attribute the Celtic  element in the Etruscan language, and to the former the Armenian element. The structure of the language we should  expect to be Armenian, as the Umbrian would be the intrusive  element in Etruscan. Accordingly, when we come to examine  the Etruscan inscriptions, we shall find this to be the case.) Etruscan Inscriptions, Etruscan inscriptions are mainly of two kinds, sepulchral  and votive. By the recurrence of the same words and forms  in these inscriptions, and by comparing them, in addition,  with similar inscriptions in Latin, we are enabled, not merely  to deduce with great probability the meaning of single Etrus-  can words, but also to proceed step by step to the interpreta-  tion of some complete Etruscan sentences, and thus to become  sufficiently acquainted with the signification of Etruscan ex-  pressions, and the grammatical forms and mechanism of the  language, to pronounce with increased confidence upon its  affinities. The result of such a process appears to be in ac- |  cordance with the inferences already drawn, that the Etruscan  is a Thracian dialect, which has taken up some Celtic words,  but is still represented in substance by the Armenian. And  here it may be well to notice the degree of affinity between  the Armenian and Etruscan which it is required to prove. It 102 THE ARMENIAN ORIGIN OF THE ETRUSCANS.

is such a degree of affinity as existe between the English and German, or between the Greek and Latin; not such a degree  as exists between the English and Anglo-Saxon, or between  the Italian and Latin. The Etruscan is to be regarded as the  sister, not the daughter, of the Armenian ; and as the sister,  not of the oldest Armenian now existing, but of an Armenian  language still more ancient by several centuries. For the Ar-  menian, dating from about 400 a.p., may be a thousand years  younger than some Etruscan inscriptions. } If, therefore, we  can make such progress towards the interpretation of the Etruscan by the Armenian, as an Englishman ignorant of German, or a Latin ignorant of Greek, could make in similar  circumstances towards the interpretation of German or Greek,  we shall then have advanced as far as we ought to advance by  the aid of philology towards the establishment of the Armenian  origin of the Etruscans. Such a point, I think, we shall  succeed in gaining, even if, in the inscription of Cervetri, we  do not go beyond it.

The first Etruscan inscriptions to be noticed are sepul-  chral. | : ; 1 The commonest forms of Etruscan sepulchres are the chambers in  the rock where there are cliffs or hills, and the éwmuli or conical barrows,  with internal chambers, on more level ground. Some of the tombs in  the rocks, as in the case of the Casuccini tomb at Chiusi, consist of a  hall, with the entrance from without on one side, and openings into  smaller chambers on the three other sides. A very similar arrangement  will be observed in the rock-tomb of Van in Armenia, of which a plan  and description is given by Layard (Nineveh and Babylon, p. 396). ‘With respect to Etruscan twmuli, the writer best acquainted with them  says: ‘If the tumular form of sepulture were not one of natural sug-  gestion, and which has therefore been employed by almost every nation  from China to Peru, it might be supposed that the Lydians, who em-  ployed it extensively (see vol. i. p. 353) had copied the subterranean  huts of their neighbours the Phrygians, and introduced the fashion into Etruria. The conical pit-houses of the ancient Armenians might in the  same way be regarded as the types of the tombs of that form which  abound in southern Etruria, and which are also found south of the Tiber,  as well as in Sicily; for the description given of them (Xenophon, Anab. iv. 5, 25; cf. Diodor. xiv. pp. 258-9) closely corresponds. The  interiors of these subterranean huts of Armenia presented scenes very THE ARMENIAN ORIGIN OF THE ETRUSCANS. 103 Etruscan Epitaphs. Several words are repeatedly found in Etruscan epitaphs, in  conjunction with proper names, so that their meaning may be  like those in an Italian capanna (Dennis, Cities and Cemeteries of Etruria, p. 61, note).

The dwellings of the Armenian peasantry are scarcely altered at the  present day. A modern traveller sgys of them: ‘The houses, however;  are not properly subterraneous, in the common sense of the term. They  are generally made by excavating the earth and raising a wall of loose  stones to the required height. Trunks of trees are then laid across for  rafters and covered with branches. Then the earth is piled on until the  whole is covered, and the fabric attains a semi-globular shape. Some-  times the whole is built upon the surface, but, in both cases, the external  appearance is that of a bare mound, of earth. Ag the traveller ap-  proaches one of these villages, he discerns nothing at first but an appa-  rent unevenness in the ground. Soon the rounded tops become distin-  guished.’—(Southgate, Travels in Armenia, &c. vol.i. p. 203). Again, (p. 305), the author observes: ‘In the neighbourhood of the city (Ourmish) there are several mounds, which the missionaries conjecture  to be the hills of the ancient fire-worshippers. There are strong ap-  pearances of their being, at least, artificial. Some have been excavated,  and large walls or masses of stone, regularly laid, have been found within. Human bones have also been discovered..—Compare Micali, Monuments Inediti, Tav. lv, lvi.; also Fergusson’s Handbook of Architecture, Asia Minor and Etruria. In Sir C. Fellows’ Lycia, c. vi., it is shown how  the forms of the Lycian tombs are derived from those of domestic build-  ings. This illustrates the present question, and also explains the differ-  ence between the Etruscan and Lycian forms of sepulchres.

The form of Etruscan tombs thus favours, as far as such evidence can,  the theory of the Armenian affinities of the Etruscans. But there are  also tombs in Etruria of a third construction, and probably the work of  an earlier race. Mr. Dennis gives a representation of one at Saturnia (v. il. p. 308). There are a great many. at the same place. The author  writes (p. 316): ‘ Here the eye is startled by the striking resemblance  to the cromlechs of our own country. Not that one such monument is  actually standing above ground in an entire state; but remove the earth  from any one of those with a single cover-stone, and in the three up-  right slabs, with their shelving, overlapping lid, you have the exact  counterpart of Kit’s Cotty House, and other like familiar antiquities of Britain ; and the resemblance is not only in the form, and in the unhewn  masses, but even in the dimensions of the structures. We know, also,  that many of the cromlechs or kistvaens of the British Isles have been  found inclosed in barrows, sometimes with a circle of small upright slabs  around them; and from analogy we may infer that all (P) were origin-  ally so buried. Here is a further point of resemblance to these tombs of 104 THE ARMENIAN ORIGIN OF THE ETRUSCANS.

very probably deduced. They may be divided into three  classes: (1) words implying age; (2) words apparently im-  plying relationship ; (3) words probably referring to the tomb  or urn itself, There are also (4) a few bilingual inscriptions,  where proper names are given in Etruscan and Latin. The  epitaphs which I shall quote are all taken from Lanzi, and in-  dicated by his numbers. What are proper names I have indi-  cated by capital initials. They are sometimes contracted in  form. Words implying Age.

The first seven of the following epitaphs are consecutive in Lanzi, and are numbered 450—456 : Ray. Velan Ar. rid xlii. Jeine. L... Ste. La... vid xiili. leine. Nevile Papa avid xxu.

Tha, Leivai Ma. Krake avid xxxiii. S. Svetin L. avil rid lxv.

A. Pekni rid hii. Jeine.

Thana Kainei ri leine lv.

Av. Leku vid ixx. (10).

Comparing these with the expressions, vizit annos, annos,  anno etatis, and atatis, and observing that Jeine is jomed with  vil, and not with avi/, the following interpretations are de-  duced—  avil, ‘ stas,’ rid, ‘annus,’ leine, ¢ vixit.’  avil, ‘tas’... Arm. aveli, yavét, ar-avel, ‘more, excessive ;’  yavit-ean, ‘an age;’ hav= Lat. avus; avag, ‘elder ;’ 6 or avt, ‘a ring’ (annulus). Lapp.  gape, ‘annus, Goth. aivs, ‘time.’ Germ. Saturnia. In some of the cromlechs, moreover, which are inclosed in  tumuli, long passages, lined with upright slabs, and roofed in with  others laid horizontally, have been found; whether the similar p  in these tombs of Saturnia were also covered in, cannot be determined.’ Though cromlechs are not peculiar to the Celts, yet this close resemblance  is valuable as subsidiary evidence in support of the position, that the  predecessors of the Etruscans in Etruria were of Celtic origin. _ THE ARMENIAN ORIGIN OF THE ETRUSCANS: 105  ewig. Eng. ever, Lat. avum=Gr. aldv = O. Norse aft = = Arab. abad. Osset. afon, ‘ time ;’  afey, ‘a year.’ Ay-il seems to consist of the root av, which  is Arm., and: of the termination -2/, which is  also Arm. Thus we have #es-2/, ‘ appearance,’  from fes, ‘sight;’ arag-i/, ‘a stork, from  arag, ‘swift. - So also kath and kath-ii both  signify ‘a drop,’ and kath-tl, likewise, ‘to  drop.’ All Arm. infinitives, which partake of  the nature of nouns and are declined as such,  terminate in e/, a/,-ovl, or tl.) The Arm.  present participle, again, terminates in avt, 6¢ or ot, and the past participle in eal: so  that the terminations / and ?, which are com-  mop in Arm., imply the state of being or  having been what the root indicates, As  forms ending in / are so frequent in Etruscan, .  it may be well to illustrate the character of  the Arm. in this respect by noticing some of  the derivatives from a single root, ¢es, and  explaining what parts of speech they properly  are, |  tes, ‘ sight.’  tesanel, ‘to see,’ a form in -anel analogous  to Aap3-avw.  tesanel, ‘ aspect,’ the preceding verb used  as a noun,  tesanot, “one who sees, a seer, a prophet,’  pres, part. of tesanel.  tesaneht, ‘ visible,’ fut. part. of tesanel,  tesanelich, ‘ sight, the eyes,’ the plural form  of tesaneli. 2 Gol, ‘to be,’ is anomalous. 106 THE ABMENIAN ORIGIN OF THE ETRUSCANS,  tesot, ‘one who sees, a prophet,’ pres. part.  of a non-existent verb, fesel.  testl, ‘aspect,’ a non-existent verb, Zesi/,  the pass. of tesel, ‘to see.’  ri, ‘annus’... Arab. rig/; Heb. rege, ‘time, an age.’ The  primitive sense is ‘foot,’ from the root rag, ‘to move’ (Gesen.) = Arm. rah, as appears  from the Arm. rahe/, ‘ to go forward’ (= Gael.  rach), rah; ‘a way.’ The termination of  r-tt may be explained like that of av-27. Com-  pare also rah-il and r-2/ with nih-i and 2-2.

Arm. aral-ét, ‘time.’ Georg. roloi, ‘a clock ;’ ri, ‘ to run.’ Welsh rit, ‘ what intervenes ;’ rhi/, ‘ what  divides, a number;’ rh, ‘ an interstice,’ .e., ‘an interval (of space) :’ ‘a year’ is ‘an in-  terval (of time).’ Leine, ‘vixit?. Arm. Jine/, ‘to be, to become, to exist, to live.’ Hung. /ennt,‘ to be ;’ uy, ‘existence.’ Lapp.  lei, erat,’ = Arm. linér. Germ. leben. Gael.  dinn, ‘an age, a. period.’ Alb. dyeigj, ‘I am  born ;’ dyind, ‘I give birth. to.’ Leine, from  its form, would rather be compared with the Arm. present, limi, -‘he, is,’ than with the  imperf. linér, ‘he was.’, Linel has no per-  fect. Words apparently implying ati. Klan (passim).  klan ... Gael. clann, ‘offspring, descendants.’ Welsh plan, ‘a scion;’ plant, ‘ason. Gr. kAde, crAdSoe, KAwv,  kAdw, gAaw, rAaw. Lat. planta. Manx cleih, ‘people ;? clei, ‘a clan;’ cleuim, ‘a son-in-law ;’  cleuinys, ‘affinity ;? cloan, ‘children;’ cluight, ‘ offspring.’ THE ARMENIAN ORIGIN OF THE ETRUSCANS. 107 A more recondite and perhaps a better interpretation of kan  is given by Miiller (Ztrusker, v. i. p. 446). He compares the  two inscriptions on the same monument— La. Venete La. Lethial! etera Se. Venete La. Lethial klan  and observes: ‘If e¢tera be taken to mean ‘ other, second,’ tian  must be ‘ first, firstborn.’ tera is thus compared with the Gr. frepoc, Umbr. etre, Arm. Star. Dr. Donaldson argues in .  the same manner (Varrontanus, p. 171) : ‘If then e¢era means,  as is most probable, the second of a family’ (just as Arm. mivs, ‘other,’ does signify the second son of five—St. Martin, Mémoires, v. i. p. 174), ‘klan must mean the first or head of  the family. This might bring us to the Arm.—  kian .. Arm. glovkh, ‘head, summit, the first ‘rank ;’ glkhan  zovkn (lit. ‘ head-fish’), ‘a chub ;’ glthant, ‘ the chief  persons in a city, the nobility,’ Glkhan would become ‘khan in Etruscan ; or, omitting the  aspirate, flan.  mi Kalairu fuivs (191).

Lth. Marikane via (315).

Larthi Vetus Klaukes puza (810).

Arnth Vipis Serturis puiak Sutat...i (811). Anes Kaes put hui...(128). Suis Gael. fuil, ‘ blood, family, tribe, kindred.’ Lat. filiue,  pia Gr. vide, gvAf. Hung. fid, ‘son.’ Syriinie gi,  a ‘son’? Esth. por, ‘son,’ pots, ‘boy.’ Larthia Kaia Huzetnas Arnthalisa Kafatl eck (68). Titi Velimnias Akril seé (37). Ramthn Matulnei seh. ..... (471).  sak (Arm. zavak, ‘blood, son, child ;’ zag-tl, ‘to be born  sek or derived ;’ zag-el, ‘to produce young;’ zag, ‘a  sech young bird (veooadc) 2” Gr. réeoc. 1 Lethus is given as a Pelasgian name. (Z1. ii. 843.) * The Syrianians are a Finnish tribe in the Russian provinces of Arkhangel, Vologda, and Perm. 108 THE ARMENIAN ORIGIN OF THE ETRUSCANS. The exact meanings of t/an, fuius, and sak, are not per-  fectly clear. Klan and sak might possibly define the familia  or the gens of the deceased; and /uzus imply youth, as pa-  rentage is indicated in Etruscan without any word signifying ‘son’ or ‘child.’ The following epitaphs, 87 and 124, shew  the distinction between 4/an and such words as fuius or  pula i— Tlatisal putia Larthias Rutenei .. aural 4a line.  klan puiak Arnth Kaes Anes Ka.....

With respect to the termination of puzak it may be observed  that the termination -aé forms Arm, diminutives. Words probably referring to the tomb or urn, or to the deceased. Suthi. This term is frequently found. There is the follow-  ing inscription at the entrance of the tomb of the Volumnii,  described by Vermiglioli, near Perugia :— | Arnth Larth Velimnas Arvneal Thusiur  sutht akil theke.

The Italian antiquaries, as cited by Vermiglioli, seem to  agree in regarding suthi as a sepulchral term. Orioli says  that ‘the nature of the localities on which it is found in-  scribed does not permit a doubt on the subject.’ Migliarini  interprets eka suthi, ‘hic situs est,’ or ‘questa & la tomba.’ Vermiglioli himself, also regarding suthi as a sepulchral term,  seems inclined to follow Lanzi in Hellenising the Etruscan,  and in referring suthi to owrnpla. But the most obvious _ meaning is certainly ‘tomb,’ or ‘is buried.’ } 1 There may be some slight objections to these interpretations. We  find suthina on a statue (Micali, Mon. Tav. xxxv.), and on a patera or  mirror (Tav. xlviii.). Suthil is also found ona kind of bronze disk of  uncertain use (Lanzi, v. ii. p. 442). Suthina might possibly, though it  does not appear probable, be etymologically unconnected with suths. Its termination -uthin(a) resembles the Arm. termination -ovthiva, and  the root might be the Arm, zoh, ‘hostia;’ zoh-el, ‘0¢-ew;’ whence we  might form zohovthiun, ‘ Ovna.’ THE ARMENIAN ORIGIN OF THE ETRUSCANS.: 109 Welsh swith, ‘a.heap” Gael. suidh, ‘a seat Carian  suthi<  gova,‘atomb.’ Arm. sovzanel, perf. sovzi, th. sovz, ‘to thrust in, to cover, to hide.’ In the previous epitaph, these, found elsewhere under the  form, ¢eke, resembles the Gr. @3nxe, as Lanzi notices. It will  be found that. ¢heke or teke can hardly signify anything but ‘makes’ or ‘brings.’ The Arm. words which most nearly  resemble it are—dizé, ‘erigit;’ th. dé, ‘acervus,? = Gr. Sc, Sty: 296 ‘ducit, fert:’ zagé, ‘rhere, réve.’ Cf. rlerwy: also Lapp. éakket, Fin. dekd, Esth. teggema, ‘facere.’ -Akil may be  a proper name, Aci/ius, the nom. to theke.

Instead of eka suthi (‘ecce sepulcrum,’ or ‘hic jacet?—suthi  might be either a noun, or a verb in the 8 pers. sing., in Arm.), we sometimes meet with eka suthi nesl, followed by a  proper name.! Now the word zes/, being found on tombs,  readily leads us to the Arm. nai, ‘a bier, a coffin;’ nekh, ‘ putror ;’ nekhel, ‘ putrefacere;’ nekhot, ‘ putrefaciens ;’ nekheal, ‘ putrefactus,’ z.¢e. ‘ mortuus, cadaver, vexpdc, vécve.’ The Arm.  sovzanel, ‘condere, mergere,’ and the passive, or in this case  the reflective form, sovzani/, ‘sese condere, mergere,’ are verbs  like auapravw, which insert -an- between the root and the final  inflexion. It would be more common to derive from a root  like sovz the active and passive forms, sovzel, ‘condere,’ and  sovzil, ‘condi.? However, the Arm., as it is, will give us  ahé  sovzani mekheal ....4.  ecce sese condit putrefactus .....  from which we may explain the Etruscan—  cha suthi mest .....  mortuus \  hic conditur  here lies {  cadaver  the deceased  the body of J °° °° The Etruscan root, nes, would be intermediate in ortho- 1 See the plate of the Campanari Museum in Dennis, y. i. p. 442. - The proper name is lost. 110 THE ARMENIAN ORIGIN OF THE ETRUSCANS.

graphy, and also in meaning, to the Arm. nekh and nai. We  shall subsequently meet with another Etruscan word, sans,  resembling es? in form, and which may be also interpreted  as a participle, and identified with the Arm. Znzavt or zZnzot. We have already discerned this last Arm. participial form in  the Lyd. xavdabA-ne, = Arm. kheldavt, ‘rvlywv.’ Ante, p. 40.

Thatnei the: (76).

Laris Vete thus (80).

Larth Vete Arnthalisa ‘hui Larth Vete line (86).  thus Larth Petrni Larthalisa (318)..

Lanzi ranks thus with fuius and puta, and interprets it ‘ filius,’ or ‘filia.? This does not appear to me very probable. We find the root thw (¢hov) in the following Arm. words :—  thiv (gen. thovoy), ‘ numerus.’  thov-el, ‘ numerare.’  thov-é, ‘ numerat.’  thov-s, ‘ videtur, numeratur.’  thov-ich, ‘ sententia,’ the plural form of ¢hov:.

By extending a little the meaning of ¢hovel, we might make  it signify ‘numerare, nominare, memorare,’ and obtain for  thovt the meaning, ‘ memoratur,’ corresponding in an epitaph  to ‘in memoriam.’ We might then interpret (86)— Larth Fete Arnthalisa thu. Larth Vete lime. “4 Lartia Vettia Arnthalisa memoratur. Lartia Vettia { ft i ° 1-Cf. Fr. feu, Ital. fu, ‘ deceased,’ = Lat. fuit. In the beginning of the  inscription of the Zorre di San Manno (Lanzi, ii. p. 438) there occurs, in  conjunction with suths, ‘is buried,’ the word thwes, which, if thus signi-  fies ‘memoratur,’ ought, as thues has the form of a genitive, to signify ‘memories.’ The inscription begins thus :— Kehen suthi hinthiu thues sains «Etve Thaure Hie conditur memoris Etvus Thauru DLautne ..... coe ‘Lautnus .... These meanings of suths and thues seem not unlikely to be correct. If  we should attempt to complete the translation of these few words, the .Arm. would supply, for Ainthiu, the prep. enth, which, with the sense of ‘for,’ governs a gen. Sains appears to agree with thues: we have in THE ARMENIAN ORIGIN OF THE ETRUSCANS. lil Thovi, being implied in thovich, is both a noun and verb in Arm, -  tular Rasnal (457).  tular Hilar .. . 8 Serv (458).  tular Svariu Au. Papsinasl A. Kursnis I (460). Tetrntertular (461).  tular . Arm. thatar, ‘an earthen vessel;’ thatel, ‘to bury’ (enterrer) ; th. that, ‘tellus.’ Gael. tuda, ‘a hillock.’ Gr.rbdn, réAapoc. Phryg. felat(as). . Ante, p. 32.. Lth. Velkialu Vipinal dupw (465). L. I(n)eni Ramthai /upwu avil xxiii (463). Arni Thanie /upu avils xvii (464).  lupu . Trish lubha, ‘a corpse ;’ lu, ‘a heap.’ Gael. lodh, ‘ to  putrefy.’ | Bilingual Eqytaphe. C. Licini C. f. Migri. \ (5) V. Lekne V. Hapirnal.

The root in Hapirnal is Hapir: it appears= Niger. Cf. Arm. khavar, ‘darkness ;’ khavarin, ‘dark ;’ khavaranal, ‘ to  become dark;’ shaphiik, ‘a negro.’ Aelie Fulni Aelies Ktarthialisa. ° (7) Q. Folnius A. f. Pom. Fuscus.

The root in Kiarthialisa is kiarth or kiar. Gael. ciar means ‘fuscus.’ In the first of these epitaphs, as in many others, Etrusc. Vele= Lat. Caius. Compare Arm. e/, ‘height,’ Gael. azd/, ‘noble, a cliff,’ with Gael. caid, ‘summit’ and caidh, ‘ noble.’ Arm. hin, * old,’= Welsh hén,=Gael. sean, = Lat. sen(ez). These would  give— Kehen suth hinthiu  thues sains Etve Thaure Hic conditur caus& memoris veteris Etvus Thaurus DTautne.... so Lautnus ...

The interpretations of Ainthiu and sains are not to be relied on; but  those of suthi and thui, given in the text, may acquire some additional  probability from this inscription. 112 THE ARMENIAN ORIGIN OF THE ETRUSCANS, From the short epitaphs examined above, a tolerably com-  plete vocabulary of Etruscan sepulchral expressions may be  derived. We may now proceed to notice an epitaph of greater  length, the longest, indeed, which is given by Lanzi (Ep. 471). It is copied by him, but not accurately, from Maffei (Oss. Lt.  tom. v. p. 310), and is also found in the Mus. Htrusc. tom. 11.  tav. vil. p. 108. This epitaph was discovered, written in  black, more than a century ago, in a grotto at Corneto, the  ancient Tarquinii. It seems to run thus, in four lines, with  some /acuna, when the proper names are indicated by capital  initials :-— Ramthn Matulnei sech Markes Matulm—  purian Amke Sethres Keis—tes kisum tame—u Laf— nask Matulnask klalum he-s~ kiklenar-m—  a-avenke lupum avils—achs mealchisk Eitvapia me— Here we meet with the words, sech, puta(m), lupu(m), and  avil(s), already noticed. It remains to be seen what assistance  can be derived from the Armenian towards an interpretation  of the whole. Such an interpretation must necessarily be in  some degree empirical, but still it will show what can be done  by a particular language, even though the accuracy of the  interpretation must frequently remain doubtful, I shall take  the words in their order.

Ramthn, ..A proper name, acc. of Ramtha. The nom. Ramtha and the gen. Ramthai are found in _ Lanzi’s epitaphs, 232 and 463, Matulna . . In the original, Matudavi. The Etrusc, 7  is a digamma or F, so that F and V are easily  confounded. Matulnei, ‘of Matulna,’  sech. .... ‘belonging to the gens or familia, already ex-  plained.

Markes. . . In the original, Markvs. Markes, ‘of Marcus’  or ‘ Marca,’ .

Matulm— . Read Matulnei, supposing M=N JI, and the J  to form the beginning of an Z, the rest being THE ARMENIAN ORIGIN OF THE ETRUSCANS. 113  obliterated. In the original there is no break  between Markvs and Matulm—.  puiam ... ‘daughter,’ already explained.

Amke.... A proper name, perhaps the same.as Amycus, | a reputed Bebrycian, and therefore Thracian  name, : Sethres . . . A proper name, gen. of Sethra or Sethre, the  former of which is found three times among Lanzi’s epitaphs, as well as Sethres and Sethresa. | Keis—ies, . Read Keisiniesn. We have Keisinis in an ) epitaph in the same grotto; and Cicero, as Lanzi observes, mentions Cesennia as possess-  ing a fundus at Tarquinii.  kisum ... Arm. gé, gen. gist, ‘a corpse.” We find hize . on another tomb in.the same grotto. . Heb.

. gesem, a body.’  tame—u .. *buries,’ or ‘buries here.” Arm. damban,  dambaran, ‘tomb, sepulchre, vault, catacomb,’ = Lat. twmulus, Gr. r6uBoc, Gael. tuam.

Laf—nask \ ‘The gentes or familia of Laf—na and Ma- Matulnask §  tulna’ Arm. azg, ‘sort, race, family,’ forming  also a suffix, as in aylazg, ‘different’ (lit. ‘other-  sort’) ; davazgi, ‘noble’ (lit. ‘ good-race’).  klalum ... ‘fanera” Gr. xraiw. Arm. Jal, ‘to mourn,’  lal, lalivn, lalovmn, lalich, lalovnch, lalovthivn, ‘mourning.’  ke—s ... . Perhaps kechas or kechase. We find in Etrusc.,  kecha, kechase, and kechazt. Arm. chakea?, ‘golvit? (perf.) ; chaheaz, ‘expiavit ; kaheai, ‘ paravit.’  kiklenar-m- Read kiklena Ramtha. Arm. kaktanal, ‘to  become tender.’ Kiklena, ‘dulcissima,’ an  epithet occurring passim on children’s graves. Or kiklena might be explained ‘ mortua,’ from I 114 THE ARMENIAN ORIGIN OF THE ETRUSCANS.

the Arm. root 3, ‘extinction, death,’ and the _ suffix -etée, ‘compositus:’ ¢. g. marmnetén, ‘corporeal’ (marmtn ‘ corpus’).

a-avenke . . Read apavenke, ‘se confugit ad,’ or .‘ deponit.’  lupum...

avile ...

ache ....

mealchlek Arm. avandel, ‘to consign, to give up ;’ avandé (z)hogin, ‘he gives up the ghost’ (hogi), ‘he  dies ;? apavén, ‘security, refuge, retreat ;’  apavinil, ‘to commit one’s self to.’ Ap-, in Arm.,=Gr. ar-, Lat. ad-. The termination -t/, in apavinil, marks the passive or reflective  voice. The active would be apavinel., . Already explained. Either ‘tomb,’ or ‘body.’ . ‘statis,’ already explained. In the /acuna  after avis would have been the number of  years lived. .

Doubtful. Cornish ach, ‘soboles:? Welsh ach, ‘stem, lineage.’ Sansk. vaké ‘ growth,’ = Arm.  ag. Achs, ‘adolescens, infans’ (?).. . Also extremely doubtful. Guided by mere  assonance, we might derive from the Arm.,  meth, ‘ uadaxée,’ and ask or lesk, ‘ ocya.’ See  ante, in Alb. s.v. dye’ . Mealchlsk, ‘ pada-  xécapxog.’ Similarly, from phaphovk, ‘tender,’  marmin, ‘flesh, body, corpse,’ and morth, ‘skin,’ the Armenian forms the adjectives,  phaphkamarmin, ‘ tender-bodied,’ and phaphka-  morth, ‘ tender-skinned ;’ so that it might also  form the adj. metkalesk, ‘ naXaxdcapxog.’ Apparently a proper name. Other readings  are Hivapia and Litvapla. If apia or vapia  could be connected with arla, ‘ij,’ eit vapia,  or etté apia, would suggest ‘ sit terra.’ Arm. 126, ‘sit ;’ Irish 204, ‘ terra.’  me——~ .. Uncertain. Perhaps ‘me (facit) ;’ or ‘mitis,’ = Arm, metm, Gael. maida. ‘Sit terra levis’ THE ‘ARMENIAN ORIGIN OF THE ETRUSCANS. 115. ‘1s a common valediction at the conclusion of Latin epitaphs, The Armenian would thus enable us to arrive at the follow-  ing interpretation for the epitaph :— Ramthn, Matulnes sech,  Markes Matulnes Ramtham, Matulne prolem, Marca Matulne  puam, Amke Sethres Keis(in)ies kisum tame-u.  filiam, Amycus a Sethre Casenvia cadaver sepelit. Laf—nask, Matulnask, klalum _ike(cha)s. Laf—nia gens, Matulnia gens, funera {so F | paravit. Kiklena R(a)m(tha) a(p)avenke  lupum, Dulcissima) Ramtha = (se confugit ad tumulum, \ _ Mortua deponit corpus,  avile —, achs mealchisk. Hitvapia me— . infans tenera. 1  statis —,  adolescens I alluded above to an epitaph in the same grotto, containing  the words Kezsinis and #121. It is this— . Larth Keisinis Velus klan kia zilachnke - means munikleth methim nuphz? kanthke kalus—lupu. Here we meet with #izt, as we do -with A¢sum in the previous  epitaph ; also with meani, which may be compared with the 1 Compare the following épitdph in Muratori (p. mctxx1):— Herennis Nice V. A, III. M. VII. D. XVI. Anicetus Pater Feéc. ‘ Condita Sum Nice Que Jam Dutcissima Patri Ducens Aetatis Tenera Quat  tuor Annos Abrepta Su  peris Flentes Jam Liqui Parentes.

The resemblance of the first and third sentences of the Etruscan  epitaph to the first two sentences of the Phrygian epitaph (ante, p. 34)  is rather remarkable. The Phrygian, unlike the Etruscan, requires no Celtic for its explanation.

_ * Or aupthet. 12 116 THE ARMENIAN ORIGIN OF THE ETRUSCANS. Arm. makanal, ‘to die.’ Munikleth might be explained as a ‘sepulchral monument,’ from the Gael. muin, = Lat. mon (eo),  and the Gael. claidhe, ‘ buriul;’ cladk, ‘a sepulchre.’ Klan  and /upu are familiar expressions. As Methina and Methinal  are found in Lanzi (v. ii. p. 295) as proper names, methim  ought to= Metelium. A nominative, Methina, Methls, or Methlis, would rather be expected. The verbs appear to be  zilachnke and kanthke, of which the last resembles the Arm.  chandaké, ‘he engraves or cuts.’ Zilach-nke might be an Arm.  verb in -anakel. Compare phok-é& and phokh-anaké, ‘he  changes.’ Zilach- might contain the Arm. etag ‘ fossa,’ z  being prefixed, as in ¢eti, ‘ place,’ ze¢tetel, ‘ to place ;’ akn, ‘eye,’  zakanel, ‘to eye.’ Cf. Gael. adhidaic, ‘sepeli,’ and Phryg.  lachst (ante, p. 80). Kizi might be the dat. plur., = Arm. 72  gés, the dat. sing. being gié or 74 gé.. The change of the 8 of kisum into the z of 4izs might be thus explained ; for the Etrusc. z sometimes = ss, as in the case of Utuze, = Odvaceic. Kizi, ‘ with the dead.’ The word zilachnke appears twice on an urn found at Bo-  marzo (Giorn. Arcad. v. cxix. p. 825) in the connexion—  ztlachnke avil SI—which Orioli renders conjecturally (Gora. Arcad, V. cxx. p. 282), ‘obiit, depositus est, sepultus est (0  simile) ztatis—.’ Thus the Arm. would give for hizt zilachnke  a sense which is probable, ‘ mortuis infoditur.’ | The interpretation of the second line cannot be surmised  with any confidence, but its tenour may not improbably be,  that the grave and entombment were due to the care ofa  person named Metellus, perhaps ‘ with’ or ‘for’ nupthzs, ‘ ne-  potibus’ or ‘ nepoti.’ We meet with 41s again in another epitaph, one of those  contributed by Campanari to the Giornale Arcadico, v. cxix. It is found, p. 822 Vipinans Sethre Velthur . . . Meklasial Thanchvilu avis  kis keal X8, The last word is a number: X= Z (50) in Etruscan nume- THE ARMENIAN ORIGIN OF THE ETRUSCANS. 117  ration. The Arm. for 50 is yisovn,a modification of hing, ‘five,’ and the termination -sovnz,in which we recognize the Gr. -xovra, the Lat. -gimta, and the Goth. -hund. Campanari  interprets XS, LV, though S does not usually signify 7 in Etruscan. Let, however, X9= LV, as it will not affect the  question of affinity. We have now to explain avis kis keal.  avils . . ‘ wtatis.’  kis. . . ‘corpus,’ if its meaning has been rightly conjectured.  keal . . Arm. keal, ‘ vivere.’ The pres. part., being formed ' in Arm. by adding -o# to the root, would be seot,  instead of which the adj. hendan or kendani, ‘ vivus;’  is employed. The perf. part., being formed by  adding -eal to the root, or else to the perf. ke2(3), ‘vixi, would be head or kezeal, of which forms the  latter is in use. But seal would be a genuine Arm. form for ‘having lived,’ which may probably  be the meaning of the Etrusc. kea/. (Compare the  perf. parts, in the Armenian epitaph, azte, p. 28,  note). So also the two forms of the perf. part. of  the Arm. kal, ‘to be, to abide, to live,’ would be  keal and kazeal. The four words, avils kis keal LV, might then be interpreted, ‘ having lived as  to the body 55 (years) of age,’ or, ‘ having become  a corpse at 55 (years) of age.’ The following epitaph is also given by Campanari (Giorn.

Arcad, v. cxix. p. 884). It is on a sarcophagus— Atnas Vel. Larthal klan stalke avil DXIITI. . . .th Maruiva Tarils kenaphe lupu.

We have here to explain stal/ke and kenaphe, both probably  verbs. The Arm. suggests no root for stalke, but from the Gael. we get the verb stale, ‘ become stiff.’ Taking this root  into the Arm., we form stalke/, ‘ rigescere, torpescere, 7.¢., mori,’  and stalké, ‘ moritur.’ _ For kenaphe the Gael. gives cean, ‘a debt;’ ceannack, ‘a  reward ;? ceannaich, ‘to buy’(= Heb. anak): and the Arm.

118 THE ARMENIAN ORIGIN OF THE ETRUSCANS.

gin, ‘a price; gnel, i.e, génel, ‘to buy;’ and gné, t.¢., génd, ‘he buys.” (éné becomes in Etruscan orthography bene. For  the difference between kene and kenaphe compare the two.Arm.  onyms—  an bhoveé, \ the flies.

| khovsaphé, | This point will be more fully illustrated in dealing with the  inscription of Cervetri. Stalke avid LXIII. would thus appear  to mean, ‘dies at the age of 63,’ and henaphe lupu, ‘pays for  the sepulchre’ or ‘ entombment,’ or something similar.

I now proceed from sepulchral to votive Etruscan inscrip-  tions— : i Words on votive offerings, statues, Sc.

Kana. Lanzi interprets this word, ‘dyaAyua.’ It is found  on sculptures, perhaps (dubbiamente) on an altar (v. 1. p. 407). It appears in the two following inscriptions, both on  statues—  mi ‘ana Larthial Numthral Laukin.... (p. 465).  mi kana Larthias Vanl Velchinei Sai. . ke (p. 466).  kana . . Gael. caon, ‘resemblance’ (cixwv, simulacrum, both  votive expressions). Arm. shan, ‘a table.’ Turke and fleres.

On a candelabrum (Lanzi, v. ii. p. 421) : Au. Velskus thuplthas alpan (¢urke. On a round plate of bronze, apparently a cover (p. 422) :  mi suthil Velthura thura ¢urke Au Velthuri Fniskial. On a statue (p. 446) : Thuker Hermenas fturuke.

On a statue of Apollo (p. 448) :  mi jleres Epul .fe Aritimi © Fasti R.fr.a turke klen kecha.! 1 In the text, keka; on the plate of the atatue itself, Kecha. In  another inscription we find kecha. The missing letter in -fe appears  from Maffei, Oss. Lit., to be.a. THE ARMENIAN ORIGIN OF THE ETRUSCANS. 119 On a statue (p. 449) : Larke’ Lekne furke fleres. On a statue (p. 455):  jleres :zek sansl kver. On a picture upon a vase (Dennis, v. ii. frontispiece) :  eka erske nak achrum flerthrke.

From these examples it will be seen that ¢urke and fleres  imply ‘ giving’ or ‘ dedicating.’ Lanzi interprets turke, ‘ do-  navit,’ or ‘donum dedit.’ I believe it signifies ‘dat.’ eres  would be a corresponding. substantive. Dr. Donaldson says (Yarron. p. 178): ‘fleres clearly means donarium or something  of the kind,’ and compares it with fico and ploro.. The Arm.  will supply the following explanations— 7  turke . Arm. tovrch, ‘ gift,’ the plural form of ¢ovr. Gael.  thoir, ‘da.’ Gr. ddpov.

Jleres . Arm. ovterz, 6terz, ‘ gift, homage ;’ ovterzel, éterzel,  otorzel, ‘ to offer.’ Arm. aters, ‘ a prayer, a supplication.’. Arm. efer, ‘lamentation.’ Gael. dlaor, ‘a ery.’ The initial vowels in these Arm. words are probably  due to the circumstance, that scarcely any Arm. word  is allowed to begin with ¢=yA. In this case, ‘a  gift? would be Zerz (yAepZ) ; ‘a prayer,’ ters, (xAEpc) ;  and ‘lamentation,’ ter (xAgp). The / in fleres probably  represents this ,: indeed we may almost perceive it  does by comparing (e)ter with fieo. So Fluellen= Llewellyn, the Welsh // being = Arm. 2, or yA.

In the last of the above inscriptions, fler(es) and ¢turke seem  combined into flerthrke, which may be rendered ‘donum da-  tum,’ 7.¢., ‘a votive offering.’ For the omission of the « in -thrke, compare Arm. ¢roz and trowch, the gen. and instr. cases  of tovrch (turch). The whole of this inscription may be in-  terpreted, as we know the subject of the picture on which it is  written. It represents the parting of Admetus and Alcestis. In the centre of the picture, Admetus (A/miie) and Alcestis 120 THE ARMENIAN ORIGIN OF THE ETRUSCANS. (Alksti) are taking the farewell embrace. Behind Alcestis is Charon, the minister of Hades, with his mallet uplifted against  her. Running down between him and Alcestis, and com-  mencing above the name A/ks?s, is the inscription—  eka erske nak achrum flerthrke.

As the last word signifies ‘a votive offering,’ and the in-  scription would relate to Alcestis, it can hardly describe any-  thing but her self-devotion to death for her husband. Mr. Birch, in his Ancient Pottery, regards it as the speech of Charon relative to the parting. Achrum, therefore, probably  means ‘ Acheron,’ and nak, ‘to.’ These interpretations will  give the following analogies—  nak... Arm. nakh, ‘ first, before,’ adj. and adv.; in compos.

‘towards’ or ‘against,’ as in nakh-anz, ‘zeal,  envy,’ 1.¢., ‘animus towards or against.’ Germ.  nach, Welsh nag, ‘opposition.’ Heb. neged, ‘ be-  fore’ (coram); nekhah, ‘opposite.’ Kurd. aek, ‘near, by.’ Alb. aga, ngakha, ‘to.’  achrum . Arm. okh, ‘hatred ;’ okherim, ‘malevolent.’ Cf. Srbé and orvyiw. We might also suppose achrum = okherim, ‘ the malevolent,’ to apply to Charon,  the messenger of death, who seems about to strike Alcestis.

The first word, eka, often begins Etruscan inscriptions. It  would either signify ‘here’ or ‘behold.’ I take it = Lat. ecce, = Arm. ahd, = Gael. aca; Span. acd, ‘here.’ The inscription  then becomes— Eha  erske nak <Achrum  flerthrke. Behold ! to Hades a votive offering.

For the remaining word, erske, the Arm. again comes to  our aid, and supplies the exact word that is required—  erske . . Arm. eresé, ‘ she offers’ or ‘ presents herself’ th. eres,  eresch, * front, face.’? 1 Mr. Dennis (v. i. p. xc) is inclined to give erske the meaning, ‘she  saves,’ connecting it with the Etrurian arse, ‘averte.’ He interprets the — =~ YY ws * | 8 eee THE ARMENIAN ORIGIN OF THE ETRUSCANS. 121 If we write the Armenian, according to Etruscan ortho-  graphy, under the original inscription, we should have— Etrusc. Eka erske nak achrum  fler-thrke.

ulerz-turch Arm. Aha _ erese nach ucherim ~ alers-turch  eler-turch. Another of the previous inscriptions, on the statue of a boy,  runs thus—  fleres zek sansl ver This Dr. Donaldson compares (Varron. p. 176) with—  fleres tlen-asies sver  and observes that éver and sver are probably different forms of the  same word. They would meet in the Welsh gwaer, ‘ sister,’ which  is almost identical with the Pers. Ahwdéhar, or khwdher (khwdhr). The Irish for ‘sister’ is siur. The Arm. is choyr, gen. cher. The interpretation of the first inscription now becomes— Jleres zek sansl kver.  votum soror,} This leads us to consider another formal expression, sane/, It is found a second time in the inscription on the base of the  statue of ‘ the Orator,’ shortly to be noticed. The termination -/ would induce us to suppose, from Armenian analogy, that - sansl is a participle. Now from observing Latin votive in-  scriptions it may be inferred that there is one participle, and  only one, which must occur in such dedications. This is the  word dsbens, which would identify sans/ with the Arm. znzot, ‘rejoicing’ (gaudens, libens), the participle of znza/, ‘ gaudere.’ The meaning of the remaining word, zeé, can only be doubt-  fally conjectured. The most obvious Armenian analogies are  farnished by zge/, ‘to bring,’ and zevak, ‘ form, figure.’ Adopt-  ing this last, we should obtain the interpretation—  whole inscription: ‘Lo, she saves him from Acheron, and makes a votive  offering of herself.’ To interpret nak, ‘from,’ seems objectionable. 1 Compare the two following inscriptions in Gruter (p. meccxlviii) :— D. M. C. Egnatio Epicteto et C. Egnatio Floro modesta soror. Fortunato fratri pientissimo fecerunt sorores.

122 THE ARMENIAN ORIGIN OF THE ETRUSCANS.

fleres zek sansl  kver.  votam statuam libens soror (dat). Klen kecha. This form occurs in the inscription, already  given, on the figure of Apollo:  mi fleres pul afe Aritum  me votum Apollini et! Artemidi Fasti Rufrua turke klen kecha Fausta Rufria dat . Klen kecha also occurs on another inscription afterwards to  be analysed. The most probable meaning for kecha, judging  from Latin votive inscriptions, would be either ‘ consecrat’ or  solvit.2 The Arm. has both chaké, ‘he expiates,’ and chaké, ‘he dissolves :’ also chahanay, ‘a priest. The meaning of  kien is less clear ; but it may be connected with the Gael. glan, ‘clean, pure, sincere, righteous,’ and be nearly identical in  meaning with pivs or pia, pie, rite. Cf. Arm. sovrb, ‘clean,  pure, sacred,’ from which is derived srée/, ‘to purify, sanctify,  consecrate, dedicate.’ This exemplifies the appropriateness of  the combination, Alen kecha, ‘sacer sacrat,’ ‘sovrb erbé? He  who consecrates, kecha, ought to be holy, tien. We havea  similar reduplication in sacrosanctus and in donum dedit. The Hebrew, again, prefixes the participle to the verb to make the  sense more emphatic. The complete interpretation of the in-  scription would thus be— Me votum Apollini et Artemidi Fausta Rufria dat pia consecrat. ? (rite J ( solvit. Thuplthas alpan, This form occurs in the first of the group  of inscriptions given above (p. 118) for the determination of  turke and fleree— Au. Velskus thuplthas alpan  turke. Aulus Veliscus dat.. | We also find thuf lthas alpan in another inscription presently  to be noticed. Zhup or thuf suggests the analogies of riéwoc 1 Arm. ev, ‘and.’ THE ARMENIAN ORIGIN OF THE ETRUSOCANS. 123  and rémrrw, = Arm. tip and .fophem, which may give for the Etruscan word the meaning, ‘signum.’ thas seems well  connected by Lanzi with Arh, = Arm. atétich, the plural form  of atéth. The Arm. has also #82, ‘desire,’ gen: étzi ; étzal, ‘to  desire, to wish for ;? in which the root of dirh, Alecoua, and Alwrw, again appears. There remains a/pan, which may be  explained from the Arm. otba/, ‘ to lament, to groan.’ For the  termination; compare 2sk4el, ‘ to rule,’ with z&khan, ‘a ruler.’ The explanation of the inscription would therefore be— Au. Velskus  thuplthas alpan turke.. _ Aulus Veliscus signum-precis supplex dat.

Signum precis, ‘the sign’ or ‘memorial of a prayer,’ would ©  correspond to votum and ex voto, or to. evxog, evyh, and sdyiic Evexa.

Tinskvii. ‘A solemn form of consecration or presentation ;  already found on three other monumenta discovered in this  neighbourhood (Cortona), and which may reasonably be con-  sidered sacred offerings: i.¢., the Chimmra of the Royal Gallery of Florence found at Arezzo in the year 1554; the Griffin found at Cortona in 1720; and a small pedestal of  bronze in the Museum of Cortona, on which a statue would  have stood.”—-Micali, Mon. Ined. p. 80. .

_In the beginning of ¢inskvi/, which is to be compared with Thana and Than-kvil (Tanaquil), the name of Zina, the Etrus-  can Jupiter, has long been recognised. Kvi/, therefore, remains  to be explained.  kvil . . Arm. khilay, ‘a gift.’ Arm. chavel, chahel, ‘to expiate.”’ Cf. Arm. rake/, and | Etruse. red.

On the celebrated candelabrum of Cortona, the masterpiece  of Etruscan toreutic art, is the following inscription :  thapna lueni.  inskvil atht.  salthn Owing to a fracture, part of a letter, apparently an ¢, is lost 124 THE ARMENIAN ORIGIN OF THE ETRUSCANS.

after ath, and probably a whole letter after Juent. This last  letter Micali considers with great probability to be the ¢ want-  ing to complete the word (¢)taskvi/ ; an opinion in which Mr.

Dennis is also inclined to concur.! The inscription would thus  appear to be— | .  thapna luent tinskotl ath saltha.  thapna .. Arm. tap, ‘fire;’ thaph, ‘ardour;’ tapanal, ‘to  burn.’ The Arm. has also ¢apan, ‘ an urn, a box,  a tomb.’  lusnt . . . Arm. loys, ‘light ;’ lovsin, gen. loveni, £ the moon ;’  loven-thag (‘light-crown’), ‘the planet Jupiter.’  lovianel, ‘to light.’  tinskvil . . © offered to Tina.’  ath... A proper name. We have both Ak and AfA/ in  sepulchral inscriptions: eg., Atk Sekune Athi (Lanzi, v.11. p. $68). | 1 Cities and Cemeteries of Etruria, note, p. 443. In a note, p. 44, Mr. Dennis observes of this candelabrum: ‘It is a lychnus, such as were  hung from the ceilings of palaces or temples, and as have been found  also suspended in sepulchres—even in Etruscan ones, as in the tomb of ‘the Volumnii, at Perugia. Micali thinks it a sepulchral monument—a  funeral offering to the great god of the infernal regions, consecrated by  some lady of illustrious race, as the inscription seems to show.’ (Micali  considers thapna a proper name, and compares it with thapia, which he  conjectures to be= Apia or Appia.) ‘He suggests that it may have  hung in the chamber, where the funeral feast was wont to be celebrated,  as well as the annual inferie or parentalia. The use of sepulchral lamps  by the ancients is well known, and gave rise, in the middle ages, to  strange notions of perpetual fire; for it was asserted that some were  found still burning in thé tombs, though fifteen or twenty centuries had  elapsed since they were lighted. It seems, however, that lamps were  sometimes kept burning in sepulchres long after the interment. Micali  cites an extract from Modestinus, which shows that a certain Roman  gave freedom to his slaves at his death, on condition of their keeping a  light burning in his sepulchre: ‘Saccus servus meus et Eutychia et Hiene ancille mes omnes sub hac conditione liberi sunto, ut monumento  meo alternis mensibus lucernam accendant, et solemnia mortis peragant.’’ Cf. Grevius, Ant. Rom. p. 1451, and pp. 901-1020. It will be seen how  this bears on my interpretation of thapna lusni.

THE ARMENIAN ORIGIN OF THE ETRUSCANS, 125  salthn .. (dat-el, ‘to mix, to mould (pétrir), to envelop.’ Sat-it, ‘a carcase.’ . §at-ovmn, ‘an enve- ) Compare these terminations  lope.’ with the Etruscan names,  sat-mn, ‘an em- Volt-umna, Tol-umniue  bryo.’ (Arm. dotovmn, ‘ horror’).

archay, ‘a king.’ | archay-ovthivn, ‘a kingdom.’  charoz, ‘a herald’ («hpvé). | ciorrt ‘to proclaim’ (xcnpéacety).  charoz-ovthion, ‘a proclamation” (khpvya). So we may form from the root sa?— ’ gatovthivn, or, omitting the last two vowels, ov and  iv , Satthn, ‘a moulded work’ (répevya), or simply, ‘a work’ (opus). Salthn might also be compared in form with the Arm. Sourthn, = Pers. sirdkh, ‘an orifice.’ And the inscription would mean— | ‘A burner of light, offered to Tina, the work of Atilius.”! 1 Compare the following inscription (Gruter, p. mexlviii) :— Have Septima sit tibi  terra levis quisq.  huio tumulo posuit  ardente(m) lucernam  illius cineres aurea  terra tegat.  and the formal expression— O. F.N.D., ‘opus fecit nwmini devotus.’ Cf. also Gr. Avxvoxata, and see Herod. ii. 62, 130.

Another interpretation may be suggested. ovsin, gen. and dat.  lovsni (lusni), is the Armenian name for the moon; and we find on a  patera or mirror (Lanzi, plate xii. No. 6) Diana called Losna. Now  thapna may be interpreted as meaning by itself ‘a lamp ;’ and tinskuil  might be taken in the general sense of deo-datus or numini-devotus. The  inscription would then be thus interpreted :—  thapna, lusni tinskvil, Athi = salthn.  lampas, Diane = numini devota, Atilii opus.

Cf. Pausan. lib. ii. 0. 22. ’Agsdor 8¢ nal viv Ers és Tov Bb0poy Karopévas AapwéSas Képy 77 Ajunrpos; i.e. to Hecate, the Infernal Diana. Pausa- 126 THE ARMENIAN ORIGIN OF THE ETRUSCANS. Tlen-asies. ‘This is found in the inscription previously  cited—  jreres thenases ever.  votum soror. Another form is tlenacheis. Asics, or acheis, may be referred  to the Arm, aféch, acc, afé, ‘pretium,’ a plural form; while  tlen may be explained from the Gael. dleas, ‘duty ;’ dligh, ‘to  owe ;’ which would give for t/en the meaning ‘ debitus,’ sup-  posing -en to be an adjectival termination engrafted on the  root @/-,in Etruse. ¢/-. Cf. Arm. fap, ‘heat ;’ ‘apean, © hot.’ This reduces the inscription .to— . freres tlenasies aver.  votum debitum-pretium soror (dat).

The votive offering (votum) was the due acknowledgment (debitum pretium) of a mercy received. Zlen-astes would nearly  correspond to the Latin expressions, digne grates, merita gratia,  debiti honores, merita dona.

We may now interpret the inscription on the statue of ‘the Orator.’ (Lanzi, v. ii. p. 468. Micali, Mon. Tav. xliv) :— Auless Metelis Ve Vestal klens . ken fleres teke sansl tenme  tuthines chiseliks.> The words requiring explanation here are klensi, ken, tenine,  tuthines, and chiseiks—Kilenst I take to be nearly identical in  sense with k/en, which has already been interpreted ‘pus.’ For the suffix -s:, compare Arm. Jays. and lagnit, ‘‘ broad ;’  bolor, ‘a circle,’ bolaréi, ‘round’—Ken seems nearly the same  as kehen, which we sometimes find in sepulchral .inseriptions  taking the place ‘of ‘eka, ‘ecce.? It may thus: becompared  with the Gael. cheana, ‘ certe, sane, en, ita,’ =Heb. den. But  nias is here speaking of a tomb or monument near Corinth; said to be:the  grave of Tantalus. This monument was supported by three figures of ~ Diana, Jupiter, and Minerva. Near it was the tomb of Pelasgus, the  son of Triopas, and the temple of Demeter Pelasgis -1]In the original, chtsviiks. I have before noticed the resatibiurice  between the Etruscan V (#’) and £. THE ARMENIAN ORIGIN OF THE ETRUSCANS. 127  kén is also found in Arm., though only in composition. From  the relative or, ‘who, which,’ are formed the adverbs or-kén,  or-bar, and or-pés, ‘as, for instance:’ and we find also, 2i-kén,  hi-bar, and hi-pés, signifying likewise ‘as,’. (or-4é" probably = ‘which-like,’ and i-kén =‘ this-like’). From this it is. plain  that 4én, bar, and pés; are very. nearly synonyms; and pé& (= Sansk. pega, ‘forma’) signifies ‘hke, such,’ and also ‘ so  that, as, when ;’ and therefore 4éz might have the force of wi,  quum, or tta——For tenine, the best Armenian analogies ‘are:  tant, ‘he brings,’ or ‘renders ;’ dud, ‘he places ;’ téné,. ‘he  solemnises;’ zéné, ‘he presents, dedicates, consecrates’ Tuthines appears to. be the gen..of tuthen, tuthina, or tuthine. If we resolve the nom. into ¢-uéhin, we get the common Arm,  termination -ovéhivn, corresponding to the Lat. -atzo ; and for  the root the Arm.: verb. ¢(a/), ‘d(are):’ whence we form  tovthivn, ‘datio, dwriyn.’ The actual Arm. form is rather dif-  ferent. The roat of ‘giving,’ ¢- or to, is first taken: then  the suffix -i¢ is added to form ¢ovid, gen. tovdi, ‘dator;’ and  then again the .suffix -ovthivn, to form: tovéovthivn, ‘ datio.’ Tuthines may be interpreted .‘.gratin, yaptroc,’ and might  signify either ‘thanks’ or ‘a mercy received.’ I shall take  the latter signification —The last. word, chiseliks; may be ex-  plained. from the Arm. yi§elich, ‘a memorial.’—The following  interpretation for the inscription is thus obtained :— Aulesi . Metalts Fe. Vestal klenst Aulus: Metellus, Veli filius, Vesia natus, pientissimus  ken fleres take sanal _tenine  ut votum ponit, libens reddit  tuthines chiseltke.  tise monumentum.

‘Aulus Metellus, the son of Velus and of Vesia, as he de-  voutly presents (this) votive offering, gladly brings a memorial  of mercy received.”! 1 Or ‘a memorial of gratitude.’ Compare ftenine tuthines chiseliks  with the Gr. votive expressions, d»¢0yxey xapwrripvor (Grater, p. Ixxy), or  xaprrhpia dvéOyxey (Muratori, p. lxxxix). 128 THE ARMENIAN ORIGIN OF THE ETRUSCANS. In this last inscription, fenine has been considered identical  with the Arm. fami. Now there are four conjugations. in Armenian, distinguished by the terminations of the infinitive, -el, -al, -ovl (ul), and -#/, the last having usually a passive  sense. Tenine might belong to the first, as 4anz does to the  fourth of these. The -im- in ten-in-e may be analogous to the Arm. -an-, which frequently occurs in verbs: e.g. liz-an-é=  liz-é = liz-ov = lez-ov, ‘he licks,’ got-an-ay = got-é, ‘he steals ;’  kher-an-ay = kher-i, ‘he insults.’ There is also kam-en-ay =  kam-i, ‘he wishes.” We meet with a similar form to ¢enine in  an inscription cut in the rock at Vulci (Micali, Mon. Ined. Tav. lix.). This form is kerinu. The inscription runs thus,  surrounding the sculptured figure of a man :—  cha suthik Velus Hupus klensi kerinu.

Here suthik appears like an Arm. diminutive in -2é of suth,  or sutht, ‘atomb.’ Cf. Arm. loys, gen. lovsoy, ‘light; lovsh, ‘little light.’ Kerimu probably means ‘ excavates’ or ‘ carves,’  and would be allied to the Arm. cherel, ‘to scrape ;’ grel, ‘to  write,’ t.c. ‘to engrave ;’ gir, ‘ ypaupua, xapaxrnp ;’ krel, ‘ to  hammer, to carve.’ Cf. ypdgw, graben, and grave. Thus the  meaning of the whole inscription would be— ‘Here Velus Evpus devoutly excavates a tomb.’ Kerimu may be most completely illustrated by comparing  the Arm.—/ovr, ‘carved, hammered ;? kovrch (plural form of  kovr), ‘image, statue ;’ Aré, ‘he hammers, he carves;’ kran, ‘a hammer, a pickaxe;’ /rané ‘ he hammers, he forges.’ The last inscription required for the exemplification of  votive terms is on a statue (Lanzi, 1. p. 455. Micali, Mon. Tav. xliii.) Velias Fanaknal thuf lthas' alpan lenache kien kecha tuthines  tlenacheis. | Lenache is the only word here uninterpreted. Now when  we compare— 1 In his text, but not in his plate, Lanzi erroneously gives this word as  has. | THE ARMENIAN ORIGIN OF THE -ETRUSCANS. 129  thuf lthas alpan lenache klen kecha  with thuplthas alpan turke.  and — jleres . . . turke klen kecha  which are found in the first and fourth of the group of votive  inscriptions in p. 118, we may see that Jenacke is probably a  verb and may be substituted for ¢turke, ‘ gives.’ If we resort - to the Arm., we get—  linel, ¢ esse, fieri, existere.’  etanil, * fieri.’  etanak, ‘modus, forma.’  etanakel, ‘ modulari.’  etanaké, ‘ modulatur.’ These analogies suggest: for /enache the sense, ‘ facessit,’ and  make it closely correspond to the Lat. faciendum curavit and  the Oscan upsannam deded. Compare also the Arm. suffix  etén, ‘made of :’ e.g. osketén, ‘made of gold’ (oski) ; arzathetén, ‘made of silver’ (arzath); and erkathetén, ‘made of iron’ (erkath).1 The inscription would then be rendered— Velas Fanaknal thuf thas alpan  lenaché klen Velie Fannacia-nate signum precis supplex facessit pia  kecha tuthines tlenachets. |  consecrat gratis debitum-pretium.

‘(This) memorial of the prayer of Velia, the daughter of Fannacia, she suppliantly produces (and) devoutly consecrates (as) the due price of a mercy received.” It may be instructive to collect together here, in con-  clusion, the various forms of dedication above considered, so  as to present a full list: of Etruscan votive expressions. Each  form would be completed by the name or designation of the 1 When we compare these names of metals with the Lat. awrwm, ar-  gentum, and ferrwm, and the Celtic or, aur, aour ; airgiod, arian,  argan ; iarunn, haiarn, houarn; we may perceive the respective degrees  of affinity, as far as these words are concerned, between the Celtic, Latin,  and Armenian. The Celtic is very near the Latin, the Armenian con-  siderably more remote.

3 Or ‘devoutly pays a due acknowledgment of gratitude.’ K 130 THE ARMENIAN ORIGIN OF THE ETRUSCANS.

donor. In four cases (2, 8, 7, 8) I have been obliged, for the  sake of comparison, to reverse the order in which two words  occur.

l. klenss fleres  teke  sansl tenine tuthines chiseliks. 2. alpan thuf lthas lenache kien kecha tuthines tlen-achevrs. 8. alpan thup lthas turke. 4. fleres . oe tlen-asves. 5. fleres  turke klen kecha.

6. Jleres-zek ... sansl.

7. Jler-thrke erske.

8. fleres _—turke.

9. turuke. The following would be the vocabulary of votive words :— Alpan ‘ supplex,’ = Germ. flehend. . . Chiseliks or Chisvitks, ‘ monumentum.’ Arm. o#6, ‘ fletus:’ -an, -ean, Arm.

adjectival terminations, The Etrus-  can possesses neither o nor 8. Arm. yiselich, ‘a memorial,’ the  plural form of yeh, of which the  dimin, would be yiselik, and its  plural form yiselikch, in the acc.  yiSeltks. The root here is y18, whence  is formed the infinitive yel, ‘to  remember,’ and the future partici-  ple yise, which appears in the  plural form in the noun yigelich. In a similar manner we have, from  the root ¢-, tal, ‘to give,’ and  talich, ‘a gift ;’ also /se/, ‘ to listen,’  and /selich, ‘ear, hearing, audience ;’ émpel, ‘to drink,’ émpelich, ‘ beve-  rage.’ But the Arm. forms de-  rived from the root ‘hat will most  clearly illustrate the supposed for-  mation of chiseliks from a root chis, = Arm, yi.

THE ARMENIAN ORIGIN OF THE ETRUSCANS, 131 Arm. Etruse.  khat, ‘ludus”’ . . . .« chose,  khatal, ‘ludere.’ |  khatali, ‘\udendus”  khatalich, ‘ladus,’ prop.

© ludenda.’  khatahhk, © ludus,’ dimin,  noun.  khataliks, ‘ludos” . . chiselike. For the affinity between chzs and  yig, compare the Arm. shovzel,  youzel, ‘to seek.’ The Arm. y is  aspirated. | Ereke, ‘sese offert? .. Arm. eresé, ‘sese offert:’ th. eres, —- - eresch, § facies.’ Fleres, ‘votaum, do- ( Arm. ovéerz, ‘donum;’ aters, ‘precis;’ -  num’........ eter, * fletus.’ Kana,‘ simulacrum’. . Gael. caon, ‘simulacrum.’ Kecha, ‘expiat, conse- \ Arm. chahé, ‘expiat ;’ chaké, ‘ solvit,’  crat, solvit? .... : Gael. glan ; Welsh glan, glain; Manx Klen ‘purus, pius,} glen; ‘pure, sincere, holy, righte- Klenst J pientissimus’ ous. Arm. getani, ‘fair, decent,  good.’——Arm. layn, layné, ‘broad.’ Lenache, ‘ facessit? .. Arm. etanaké, ‘modulatur;’ etanak, ‘modus, forma;’ etanil, ‘fieri;’ Lined, ‘ esse, fieri, existere.’ Lthas, ¢ precis, Aurie’?. Arm. atdthch, ‘ preces;’ atdthel, ‘ pre-  cari ;’ 2%z, ‘desiderium ;’ é#zal, ‘ de-  siderare. The Armenian, as I have  before observed, avoids the letter ?  as an initial; but we have fzali as  well as é?zah, ‘ desiderandus.’ Sansi, ‘libens’ .... Arm. Znzot, ‘ gaudens, libens.’ K 2 132 THE ARMENIAN ORIGIN OF THE ETRUSCANS. Teke, ‘ ponit, fert?. . . Arm, zgé, ‘ducit, fert,’? = Germ. zieht ; - zagé, ‘ rlere, rece,’ = Germ. zeugt. Tenine, ‘ fert, reddit’? . Arm. tant, ‘ fert, reddit, tenet.’ Thrke, ‘donum’. ... Arm. tovrck, ‘ donum,’ the plural form  of tour. - (Arm. tip, ‘typus;’ fophel, dophel,  thopel, ‘réwrav. Tip may be a  oe borrowed word, as ¢typus isin Latin ;  but the root of réwo¢ appears plainly  in Armenian. ( Tlen-, <debitum’. . Gael. dligh, ‘debe;’ dlighe, ‘lex, debi-  tum ;’ dleas, ‘ officium :’ th. d/-, in Etruse. ¢/-: -ean, Arm. adjectival Thuf ) ‘ signum, \ Thup tbro¢’ - termination. ~asies \ ‘pretium, Arm. aéé4, ‘pretium,’ a plural noun : -acheis akla.’ in the acc. the final -ck becomes -s. Osset. chas, chads, ichas, achos, ‘a  sum due;’ achza,.‘money.? Lapp.  dses, ‘merx.’, Gael. fiack, ‘value,  worth, debt.’ Another form of  azéch is artéch. . The th., as appears  from aréel, ‘valere, mereri,’ is az or  art, = Gr. a& (tog).

Turke, ‘dat, Swpsi.? .. Arm. tovrch, ‘donum.’ For the for-  mation of the verb from the noun,  see above, s. v. erske. It may have  been observed, how many Etruscan  verbs terminate in-%e. In seeking  the root, the 4 must frequently,  perhaps usually, be rejected, as well  as the e. The Lydian seems here  to resemble the Etruscan. (Cf. Lyd. Badoxe , ‘ %eOdaZe,’ with Arm.  waz-el, ‘to hasten, to run.’ Ante,  p- 39. THE ARMENIAN ORIGIN OF THE ETRUSCANS. 133 Tuthines, ‘ gratie, Arm. ¢(ad), ‘d(are) ;’ -ovthivn, ‘-atio ;’  xaptrog.” .. . } whence may .be .formed ¢ovthivn, ‘datio.? The actual Arm. form is  tovéovthivn, similarly formed from  tovié, gen. tovéi, ‘dator.’ The Arm. \ termination, -ovthivn, in Modern Arm., -ovthin, is so common as to  occur no less’ than three times in  the -Lord’s: Prayer—in archayov-  thivn, ‘kingdom ;’ in phorzovthivn, © temptation ;? and in zérovthivn, ‘power.’ To give another instance —the words ¢ and éh, ‘ essence,’  and éatan; ‘essential,’ all take this  termination, and produce the three  forms, éovthivn, éakovthivn, éakan- _ ovthiwn, ‘existence, substance.’ Ovthivn may be compared with our  termination, -ation, ‘ by which we.  recognise words of Lat. origin. Suthina and salthn, already noticed,  may be similar forms to ¢uthines,  but in the nom., not the gen.  ach, ‘sign um, statue, Arm. ‘den, sevak, ‘forma, figura.’  slxwy’ (7) -.-. ae The substantial correctness of the previous interpretations  may in great measure be confirmed by a comparison with Latin votive forms. I have therefore selected a number of  those which most usually occur, from Gruter and Muratori. By endeavouring to explain them from the Greek, a language  confessedly akin to the Latin, the degree of affinity between  the Armenian and Etruscan may at the same time be tested. It would be extremely difficult, if not impossible, to interpret  the following forms by the aid of the Greek vanguage alone,  if the Latin had been lost. 134 THE ARMENIAN ORIGIN OF THE ETRUSCANS.

Ka Voto P(osuit) Viotum) S(olvit) L(tbens) M (erito) Ez Voto Poswt ........ «Litbens) M(erito) Dedicavit Voto \ | |. D(onum) D(edit) L(ibens) M(erito) Suscepto J ° | Vot(um) Sol(vit) L(ibens) M(erito) V(otum) S(olvit) L(sbens) M(erito) Donum Posut ..... 006. ~Inbens Merito D(onum) P(osuit) .... 2a . L(tbens) M(erito) D(onum) Dledit). . . 1. eee TX ibens) M(ersto) Votum Solvit Inb(ero) Mun(ere) Votum Dat Inbens Merito V(otum) S(olvit) D(onum) D(edit) V (otum) S(olvit) Votum .....dthens...... Soluit Munus D(edit) Donum Dedit D(onum) P (osuit) Votum Retulit Fa Voto Fecté. .... 2.00 eee we tw es . Dedicavit Ea Voto Posuit Kez Voto Fecit Ez Voto Donum Posuit et. ce ccc ew eae oe ee oe DD. Fectt ch ow ee ww we we we ee ewe se ODD, Di) re oe ee « Consacravit Dedwcavit eb ow we eee es os ee es » Consecravit Fecit Gratias Agentes Posuerunt » & The Etruscan forms, as I have interpreted them from the Armenian, with a slight assistance from the Celtic, correspond,  it will be soon, closely to the Latin.) Disregarding the dif-  ference of tense, teke = posuit ; kecha = solvit, or else consecravit ;  lenache = feert ; turke = dedit or dedicavit ; and tenine = retulit, Sansl = libens ; fleres = votum or donum ; thuf lthas corresponds THE ARMENIAN ORIGIN OF THE ETRUSCANS, 135  to ez voto; and what the Latin expresses concisely by merito,  the Etruscan explains more at length by ¢len-asies, tuthines  tlen-acheis, or tuthines chiseliks ; tlen being = merito or meritus,  and asics = meritum. If we take the two fullest Etruscan forms,  and compare fleres teke sansl tenme tuthines chiseliks with |  donum posuit libens merito, the seventh of the Latin forms given  above, or thuf lihas alpan lenache klen kecha tuthines tlen-  acheis with ex voto posuit libens merito dedicavit, the second of  those forms, the close resemblance in sense will be at once ' apparent.( Nor is it merely the roots of the Etruscan words  which are Armenian : all the forms, with the exception of the  genitives in -as and -es, belong to the Armenian language. Some words, indeed, as sansl, chiseliks, and tuthin(es), if rightly  interpreted, exhibit in their construction very peculiar Ar-  menian affinities. : The following inscription (Gruter, p. xlvii.) may still  further illustrate the subject of votive expressions— Te precor' Alcide sacris invicte peractis Rite* tuis letus dona ferens meritis® Hec tibi nostra potest tenuis perferre camzna Nam grates dignas* tu potes efficere Same libens simulacra® tuis que munera® cilo Aris Urbanus dedicat’ ipse sacris. The inscription of Cervetri. The following inscription was found on asmall pot, made of  antique black ware, at Cervetri, the site of the Pelasgian town  of Agylla, which was said to have been afterwards taken by  the Etruscans, and called by them Care. The pot or cup,  which appears to hold rather less than a pint, may be seen in  the Mus. Etrusc. Vatic. pl. xcix.n. 7. The inscription would 1 Alpan. 2 3 Sansl tenine tuthines chiselike, klen kecha tuthines tlenacheis. 4 Tlenasies. 5 Kana, zek. © Fleres. 7 Turke, kecha. 136 THE ARMENIAN ORIGIN OF THE ETRUSCANS:  compose two hexameters, but the words are run together, so _ as to take the following form—  minikethumamimathumaramlisiaithipurenat  etheeraisicepanaminethunastavhelephu This inscription derives a peculiar interest from its being  considered by eminent philologists as a relic of the language of  the ancient Pelasgians, the nation whose name has had so great  an influence on the study of ethnology. There is little dif-  ference among the learned as to the division of the lines into  words, Dr. Donaldson (Varron. p. 167) reads the inscription—  mi nt kethuma mz mathu maram lisiat thipurenar  ethe erai sie epana mi nethu nastan helephu.  and Lepsius—  mi nt kethu ma mi mathu maram lisiai thipurenat  ethe erat sie epana minethu nastav helephu.

If we take the first line according to this latter reading,  dividing also maram into mar-am ; and the second line accord-  ing to Dr. Donaldson’s reading ; we shall obtain this couplet—  mi nt kethu ma mi mathu mar am lisiat thipurenat  ethe erat sve epana mi nethu nastav helephu.

Every word and form may here be considered as Armenian,  as will appear from the following analysis : Etruscan. Armenian.

me ....Nomes,*I” . . . . . . mech, ‘we.’ Acc. (2)ts, me” . ... . = (2)mez, Sus.’ The forms, mech, ‘ we,’ and (z) mez, ‘us,’ would be,  according to analogy, the plural of me, ‘I,’ and  me, ‘me,’ which thus exist implicitly in Arm. Dov, ‘thou,’ still makes dovch, ‘ye.’ So also  we have— é, ‘ existence, he is’ . . e-m,‘Iam’ .. . e-meh, “we are.’ Kurd. az, men, ‘I;’ me, ‘ me.’ ' ‘DHE ARMENIAN ORIGIN OF THE ETRUSCANS. 137 Etruscan, _ Armenian. Osset. dz, ‘1 ;’ mén, md, ‘ me.! Georg. me, ‘ L _ Welsh and Gael. mi, ‘I, me.’  mi... .. mi=Gr. wh= Lat. ne. ' ‘Welsh and Gael. ni, ‘ not. ? Pers. ‘mah, nah, Lith.  me. Obsset. ne, ma. ,  kethu . . . Nom. get, ‘ariver’ .. kath and sith, ‘a drop.’ Gen. getoy. hatha, Stha. Dat. getoy. kath. — thi. Abl. getoy. hathé. sthé. Instr. getow. kath. sthiv. Nom. hathn, ‘milk’ . . kith, ‘ milking.’ Gen. kathin.  hthoy.

Dat. hathin. hthoy.

Abl. kathing. - kthoy.

Instr. hathamb.  kthow. I take kethu to be most probably a gen. or instr.  case, and to signify ‘of,’ or ‘with water.’ Both getoy (geto) and getow (geté) would be-  come in Etruscan orthography fetu, as the Etruscans had no medials and no vowel o. Kthoy and kthow would in like manner become  kethu: kath (kathi) would become sathu or  katht: and shiv would become sethu or sethi, or  else chethu or chethi. Gael. cith, gith, ‘imber.’ Sansk. éut, ‘ stillare,  fundere, effundere,’ = Alb. cheth. Lat. gutta.  ma .... na, ‘but, however, rather, in fact.’ Arab. ammd, ‘but.’ Pers. magar, ‘but, unless, : moreover’ (gar, ‘if’). Osset. dmd, ama, ‘and.’ Me ore (As before, ‘ I’). 1 Among the European languages, the Lithuanian family exhibits  the closest affinity to the Armenian, Kurdish, and Ossetic, in the form  of the pronoun J. 138 THE ARMENIAN ORIGIN OF THE ETRUSCANS, Etruscan. Armenian.

mathu ... Nom. math, ‘syrup of grapes,  raisiné, defrutum” . . . . metr, ‘honey.’ Gen. Dat. mathoy.

Abl. ” meton.

Instr. mathow. | Mathoy and mathow become, in Etruscan ortho-  graphy, mathu.

mathovz, ‘the fruit of the arbutus,’ The termina-  tion resembles ¢hovz, ‘a fig.’ Gr. uéOv. Germ.  meth. Eng. mead. Welsh medd. Osset. miid, ‘honey.’ Sansk. madhu, ‘honey, wine, intoxi-  cating drink ;’ mad, ‘to be intoxicated, to re-  joice ;’? mada, ‘intoxication, madness.’ Pers,  may, mul, ‘wine.’ Gipsy mol, ‘wine’ Lyd.  pwdrak, ‘idoe olvov.? Arm. moli, ‘mad, in-  toxicated.’! .

Cf. Heb. debas, ‘honey, honey of grapes, t.e.,  must, or new wine boiled down to a third or  half? (Gr. &/nua, Lat. sapa, defrutum, Ital. musto  cotto). Gesen. & v. . Arm. mar, ‘a measure of liquids’-—‘ perpnrie,  firkin, (John ii. 6), ‘ Bdrog, z. ¢., 2, a measure’ (Luke xvi. 5). - Pers. mar, ‘measure, number.’ Gr. papic, ‘a measure containing six xoréAa’ (about three pints). 1 The connexion of ideas here may be still farther illustrated. Com-  pare bacca, Bacchus, bacchor, and the Gael. bach, ‘ to intoxicate,’ bachar, ‘an acorn, bachla, ‘a drinking cup;’ also uveo, uva, and uwvidus, and  the Gael. swbh, ‘a berry,’ subhag, ‘a raspberry,’ subhach, ‘merry ;’ and  dusredos with the Rhet-Rom. ampa, ‘a raspberry,’ and the Arm. ovmp, émpelich, ‘beverage.’ With bacca and bachar we might also compare _ She Arm. dbaklay, ‘a bean.’ Similarly we find the Lat. faba in the Gael. _.. * faob, ‘an acorn.’ THE ARMENIAN ORIGIN OF THE ETRUSCANS. 139 Etruscan.

hisiat . . Armenian. Alb. meré, ‘every liquid atid dry measure.’ Lith. méra, . \ ‘measure.’ Russ. mera, . Germ. mass, ‘ measure, pot, quart.’ Alb. marr, * to hold, to contain.’ Georg. marant, ‘a wine-cellar,’ = Arm. maran.

I interpret mar, in a general sense, ‘ measure, pot,  vas, xorbAn. The Gr. papic may be borrowed  from the Thracians, as may also perhaps the Lat. dolium, which appears the same as the Arm, doyl, Pers. dil, ‘a bucket.’ . Arm. em Pers. am ‘fT am,’ Alb. yam . Arm. lezov Lith. léuwis Arab. lisdn Heb. dagon  lezovt | - Arm.< lizovl  lizel ‘to lick.’ Lith. dééu Pers. Usidan Pers. Js, ‘licking.’ Insiai seems to be the dat. of disia, ‘a tongue,’ 4. €., ‘a licker;’ the root being found in the Pers. /és, or in the Arm, dz or lez, The de-  clension of disiat would resemble that of the Arm. archay, ‘aking,’ which makes archayi inthe  gen. and dat.; or of margaré, ‘a prophet,’ which  makes margeré. But the best parallels are  perhaps found in the declension of proper names :  as— ‘a tongue.’ 140 THE ARMENIAN ORIGIN OF THE ETRUSCANS. Etruscan. Armenian. Nom. Anania, ‘Ananias.’ Angtia, ‘England.’  er nt} Ananiag. Angtiay. Ioate Ananiav. Angtrav. Nom. Achayia, ‘Achaia.’ Hermés, ‘ Hermes.’  on nt} Aohayiay. Hermeay. Taste Achaytav. Hermeav. The gen. and dat. of Jezov (lezu) are lezovt.

thipurenai . thaph, ‘ ardour.’  tap, ‘ heat.’ i  tapean, ‘ burning, heated.’ Either of the first two words may give the root,  and the last word may give the meaning, of  thip-urenat ; but its termination must be ex-  plained from such Arm. words as the fol-  lowing :——  hayr, ‘ father.’  hayr-Orén, ‘ paternally.’  archay, ‘ king,’ { orokey Sede, ‘ royally.’  hamak, ‘entire, entirely’ (th. ham, hom, = bu(éc)).  ham-6rén, ‘entire, entirely.’  ham-6rini, gen. and dat. of ham-drén.  get, * beauty.’  get-a-ydbrén, ‘pretty. 6rén, ‘a law, a rule.’ . | 6rinak, ‘example, type, form’ (dimin. of éré).

yOrin-el, ‘to form, to shape.’ 6rin-akel, ‘to form, to represent.’ From these instances it may be seen, that the Arm.

suffix -orén or -yérén, when adjectival, which is  rarely the case, as it usually forms an adverb,  has the force of the Lat. -factus. We may thus _ form from the Arm. tap, ‘heat,’ the adj.— THE ARMENIAN ORIGIN OF THE ETRUSCANS, 141 Htruscan,  epana ... Armenian.

Nom. tapérén, ‘ tepefactus -a -um.’ Dat. tapérini, ‘ tepefacto -2 -0. - Tapérini becomes, in Etruscan orthography, tapu-  rini.

Compare also teraran, ‘a theatre;’ th. ¢es; Hayerén, * Armenian;’ th. Hay; phokharén, ‘ payment ;’ th. phokh.

Thipurenai would probably be a fem. adj agreeing  with lisiat. The Arm. has no indication of  genders; but in proper names, such as A¢henas, ‘Minerva,’ gen. and dat. Athenay ; Yowhanna, ‘Joanna,’ gen. and dat. Yowhannay ; we meet  with words declined like ¢hipurenai.  ethé, ‘if.’ Zend ethe, ‘ when.’ ‘joyous, gay.’ Ovrakh linel, ‘to be  khrakh merry, ev¢palvecOar’ (Luke Xv. 24),  ovrakh  erakhan  rakhgan. }. a banquet, a feast.? Cf. Gr. Eoavoe.  khrakhgan Kurd. tari, ‘a game, a sport.’ If erai be an oblique case like Zsiai, it may be  interpreted ‘of’ or ‘for joy.’ Era is indeclinable. But we might read, eth(e) era isie epana, regarding  the termination of ethe as short, and here elided,  and the terminations of era and isie as long, but’  shortened before vowels.

ize, ‘it may be;’ ¢,‘it is.’ Osset. sua or isua, ‘it  becomes.’ Germ. sez. Goth. styat. Sansk. sydt. Lat. sit.  eph, ‘ cooking!  ephel, ‘to cook.’ Nom. ephovmn, ‘ cooking.’ #3 142 THE ARMENIAN OBIGIN OF THE ETRUSCANS. Etruscan. Armenian. Gen. Dat. i pumas Abl. ephmandé. Instr. ephmamd. But, for the termination of epan(a), compare  kap, ‘a bond,’ sap-el, ‘to contract,’ hap-an, ‘a  strait ;’ gl-el, * to roll,’ gi-an, ‘a cylinder ;’ and see  just above, s. v. erat. Compare also chah-el, ‘to  expiate,’ chah-anay, ‘a priest.’ The Arm. prefers  to terminate words with -ay, instead of -a simply. We meet, again, with such equivalent terms as  tarphouomn and tarphanch (the plural form of  tarphan), ‘ cupido.’! Heb. aphak, ‘to cook.’ Gr. édrraw, hbw, dyov.

Lat. epulum, probably identical in h meaning with  epana.  mi... .. (As before; but here it seems to mean ‘me,’ not ‘T’).  nethu ... Nom. ivth, ‘substance... hetansvth, ‘ fluid  substance, liquor.’ Gen. Dat. \ nivthoy hetanivthoy. Abl. I suppose zethu to be a gen. or abl., and to signify ‘of? or ‘from liquor. Nethuns is the Etruscan  form of Neptunus. Nivthoy (niitho) would be written  in Etruscan, nwthu or nithu; although, as we have  in Arm., giv? and get, ‘a village,’ and ¢vt and ef, ‘oil,’ nivth would be very nearly neth.  nastav. .. Arm, nédeh or nédeh,* ‘stranger, foreigner, emi-  grant’ (Eévogc). 1 These words might be allied to Turan, the Etruscan name of Venus.

* Nastes was one of the two Carian leaders, JI. ii. 867, and C. Calidius  asta appears as a proper name in a Neapolitan inscription.—Donats,  p. 4. THE ARMENIAN ORIGIN OF THE ETRUSCANS., 143 Ktruscan,  helephu . Armenian. .

Pers. ndsid,..... ‘stranger, foreigner.’  nézt, ..... ‘stranger, foreigner.’ Arab | set -.... ‘stranger, foreigner, visi- —  tor, guest, banquet.’ Heb. nasa, ‘to migrate, to remove;’ naga, ‘to  err.’ oo | In Armenian poetry, »3deh would be written  nésdeh, the vowel é not being then, as usual, merely  understood.} . hetovi, ‘to pour out, texéw’ (Rev. xvi. 2).

zetovl, ‘to pour ;’ zetkh, drunken, dissolute.’?  hetov, ‘he pours out, he empties.’ The th. is het, ‘ pouring, flowing,’ found above  in hetanwth. Nethu, being qualified by helephu,  would acquire the meaning of setanwth instead of  nwtk. The formation of helephu from a root hel  may be thus illustrated from the Arm.— 36s, ‘ causing tremor,’ from which are derived—  s6s-aph-tl, ‘to tremble.’  thith-aph-el, ‘to shake’ (act.).  ded-ev-el, ‘to reel.’  khove-el ;  khove-aph-el § °° 8Y- Sarg  sarsaph Saréal  sarsil \ ‘to tremble.’ Lsaraaphit Similar forms aro— Sdtaphel, ' ‘to touch,’ and  kachavel, ‘to dance.’ ‘a trembling.’ 1 The same peculiarity seems to distinguish the inscription of Cervetri  from other inscriptions in Etruria. At least the customary deficiency of  vowels is not apparent. 3 Cf. Thrac. seira, ‘ olvos,’ and Gr. doehyéw, 144 THE ARMENIAN ORIGIN OF THE ETRUSCANS. Etruscan. Armenian. , These examples show how -api-,-av-, or -ev-, may  be inserted in Arm, between the root and the verbal  inflexion. Applying. this principle to hetov/, we  should obtain— :  hetaphov  hetavov: } ‘he empties,’ or ‘ pours out.’  hetevov With regard to the construction of helephu with  nethu, if the last word be taken as an ablative, it  may -be noticed that Aetov/ is used with an ablative  as well as an accusative. In Rev. xvi. 2, 3, 4,  eEéyee tiv ptaAny abvrov is rendered in the Arm.  version by ehet (z)skavarak ivr (acc.), and in 8, 10, 12, 17, by chet st skavaraké wrmé (abl.). Again,  in Acts 11.17, dyew ard rov wvebdparde pov, 1s  rendered by, het? yogvoy immé (abl.).

We should perhaps read the last two words of  the inscription, nasta Fhelephu, instead of aastav  helephu ; the digamma being introduced to avoid  the hiatus, or the elision of the short vowel.

_ f From the Arm. words which I have cited, we obtain, in  grammatical syntax, though in a foreign idiom, the following Armenian couplet. The orthography is Etruscan.  es mi ketu na es mathu mar em lezu tapean  ethe erah ize ephumn zis nithu nesteh helu. ) Or, adopting such modifications as I have shown to be war-  ranted by the Armenian language, i.e., writing me for es, ‘I,’  and zis, ‘me;’ ¢apurini, ‘tepefacte,’ for ¢apean, (indecl.), ‘tepide ;? and inserting -aph- between the root and termina-  tion of hel-u— ( (me) mt ketu na (me) mathu mar em lezui tap(urint)  ethe erah ize ephumn (me) nithu nesteh hel(aph)u.

Here then is an Armenian couplet, which scarcely differs at  all from the inscription of Cervetmi. Even the metre is but THE ARMENIAN ORIGIN OF THE ETRUSCANS. 145  little impaired. Yet such a resemblance would hardly be pos-  sible, unless the Armenian and Etruscan were dialects of the  same language, at least if the Armenian will give an appro-  priate sense for the Etruscan. } Bat this is the case; for the Armenian would lead to the following interpretation of the  inscription on the pot, which the recurrence of the word mi  shows to be speaking of itself, and which, as it belonged to a Bacchanalian people, may be expected to speak in accordance  with the national character— Arm. Me.mi ketu, na me mathu mar em lezur  tapurini : Etruse. Mi ni hethu, ma mi mathu mar am lsiai  thipurenat: Eng. 1 not ofwater,but I ofwine a pot am for thetongue  thirsty : Arm. the erah ize ephumn, me nithu  nesteh helaphu.

Etruse. Ethe erat sie epana, mi nethu nastav helephu.

Eng. If joyous be the feast, me of liquor the guest empties. Wine for dry tongues, not water, I contain : At joyous feasts the guests my liquor drain.’ Me vini haud lymphe plenum sitit arida lingua: Cum fervent epulz me totum combibit hospes.

In order to obtain this interpretation, the meaning of some  of the Arm. words bas been slightly modified. The case would  be analogous in Germ. and Eng. Let us, for instance, take  the supposed meaning of the inscription in Germ., and com-  pare the words with their kindred terms in Eng.— — ‘Ich bin nicht ein W asser-becher, aber ein Wein-becher fiir die  durstige zunge: I be not an water-beaker, (but) an wine-beaker for the  thirsty tongue : 1 If wo read ethe era isie epana, the resemblance to the Armenian will  be still closer. 2 So Gothe sings of the King of Thule and his goblet— Er leert ihn jeden Schmaus. L 146 THE ARMENIAN ORIGIN OF THE ETRUSCANS. Wenn lustig ist das Fest, so sduft der Gast meinen Saft.

When lusty is the feast, so sups the guest mine sap.

I may add two independent examples from Schiller’s elegiac .  couplets, in order to show that the Etruscan of Cervetri is  as near to the Armenian as the German is to the English. Jupiter to Hercules. Nicht aus. meinem Nektar hast du dir Gottheit getrunken: Not out (of) mine nectar hast thou thee Godhead (y)drunken: Deme Gétterkraft war's, die dir den Nektar errang. Thine God-craft was’t, that (to) thee the Nectar wrung. Votive Tablets. Was der Gott mich gelehrt, was mir durchs Leben What the God me (y)learned,! what me through?-the life  geholfen, | (y)holpen, Hang ich, dankbar und fromm, hier in dem Heiligthum auf. Hang I, thank(fal) and (devout), herein the halidom up. The termin. of dankbar is found in neighbour (nachbar). Thus dankbar is English nearly in the same manner as ¢hipu-  renai is Armenian, but not so obviously: as the corresponding English termin. 1s not so common as the Armenian.

Some modifications, which would not, however, affect the  question of affinity, might be suggested in the interpretation  of the inscription. Thus the Arm. would allow kethu to be  rendered ‘ milk,’ though the sense, ‘ water,’ seems to suit the  inscription better. If, again, we compare kethu, mathu, and  nethu with the forms of the Phryg. 3ééu, ‘water,’ the Macedon. BéSv, ‘air,’ the Sansk. madhu, or the Gr. péSv, we might be  inclined to consider those Etruscan words as in the nom. or  ace. case. If they be in the acc., then we should have to re-  gard maram as a transitive verb, of the second Arm. conjuga-  tion, like nztam, ‘ I contrive, I form,’ and signifying ‘ I con- 1 Chaucer has lered for ‘ taught,’ and we have still lore. * Chaucer writes thurgh. ‘  ee  a cm se a n THE ARMENIAN ORIGIN OF THE ETRUSCANS, 147  tain’ or ‘I dispense.’ Cf. Alb. marr, ‘I contain,’ and Germ. Jass and fassen. We should also have to interpret mi nethu  as ‘my liquor’ or ‘my contents,’ considering mz as equivalent  to ‘my’ or ‘of me,’ both rendered in Arm. by 2m, The Gael.  for ‘my’ is mo-; the Welsh, my; the Osset., ma; the Alb., om,  yim, or yem. If mi nethu be a nom., then helephu would have  a passive signification—the Arm. zefov is both active and -neuter—and zastav would be in the instr. case, and= Arm.  nidehiv, ‘by the stranger,’ 2.¢., ‘guest.’ Lisiar thipurenat  might also be connected, perhaps to the improvement of the  sense, with the second line instead of the first; and the in-  scription be thus given and translated— Mi ni kethu, ma mi mathu maram: lisiar thipurenai, Lithe erat ste epana, mi nethu nastav helepku. I do not contain water, but wine: on (his) thirsty tongue,  the guest pours out. When there is a joyous feast, my liquor 1% poured out by the  guest. Vinum non lympham teneo: me fervida lingua Haurit ubi dapibus letis interfuit hospes.

un (The inscription of Cervetri lends itself to the Armenian, the .  representative of the Thracian family of languages, with more  completeness and facility than any other inscription in Etruria, Now this inscription, as I noticed before, has been re-  garded by eminent scholars as Pelasgian rather than Etruscan. Were then the Pelasgians purer Thracians than the Etruscans ? It seems not improbable that they might have been so. The Pelasgian name of Cervetri was Agylla, in which we may  readily recognize the Arm. givt, z.¢., gyl, ‘a village.’ When  the Etruscans conquered it, they changed the name to Care,  which looks like the Welsh caer, though char signifies ‘ rock’  in Arm. What then would the Etruscan conquest of the Pe-  lasgian Agylla imply? Were the Etruscans, according to one  conjecture, the previously subdued Celtic Umbrians recovering L2 148 THE ARMENIAN ORIGIN OF THE ETRUSCANS.

their land? Many reasons seem to render this supposition in-  admissible. The E¢ruscans, for instance, are said to have con-  quered three hundred towns from the Umérians; and the Pelasgian Ravenna is said to have submitted to the Umbrians:  for protection against the Htruscans. Had then the Etruscans  so combined with the conquered Umbrians before the capture’  of Agylla as partly to Celticise their own speech? This may  have been the case ; though it would rather be expected that  the Celtic element found in Etruscan, and mainly, it is pro-  bable, derived from the Umbrians, would have been due to a gra-  dual infusion of later date after the complete conquest of Etruria  had been effected. But, even if the Etruscan language had  been thus partly Celticised at the time of the capture of Agylla, we have still to explain the distinction which was  made between the Etruscans and Pelasgians at that particular  time, and also the fact, due probably to the presence of Pelasgians at Agylla, and evinced by the inscription of Cervetri, that a dialect apparently not completely identical  with the Etruscan,’ but of a purer Thracian character,  existed at that place. The solution may perhaps be this. It  has been seen that a Celtic element seems to constitute the  difference between the Scythian and Proper Thracian languages,  the Celts having, as appears probable, passed into Europe to  the north of the Thracians. Among the northern Thracians,  those of Dacia, Pannonia, Noricum, and Rheetia, Celtic ele-  ments would also have penetrated. Of these northern Thracians  the Etruscans may have been a branch. The Pelasgians, on  the other hand, may have been southern or pure Thracians,  who passed into Italy at an earlier period than the northern Thracians or Etruscans. )  f ¥ 1-Yet I cannot assent to the opinion, that the language of the inscrip-  tion of Cervetri belongs to the same family as the Greek and Latin, and  to a different family from the Etruscan. Would not these be the cha-  racteristics of Oscan and Umbrian, instead of Pelasgian—of the lan-  guage of the Bantine and Eugubine tables, not of the inscription of Cervetri? : This last, again, is written like Etruscan: it has no medial  letters, and only the vowels, a, e, t, wu. 5 THE ARMENIAN ORIGIN OF THE ETRUSCANS. 149 Another mark of distinction may be discerned between the  language of the inscription of Cervetri, and that of the Etruscan vocabulary. At Cervetri we find the word xeth,  which is the Arm. nivth (myth). But in the vocabulary, the  same word takes the form nepos, which is the Alb. nepés. ‘Does this indicate a difference of dialect among the Thracians ;  such, for instance, as obtains among the Kymry and Gael, or  among the High and Low Germans? Did the Armenians and Pelasgians belong to one class, and the Etruscans and Illyrians  to the other? I have before called attention to the fact, which  may favour this last supposition, that a great part of the Al-  banians are still called Zoscans. A similar distinction might  also have helped to discriminate the Illyrians from the Proper ’ Thracians.

These distinctions would, however, be superficial, and the  language, or the dialects, of Etruria would be Thracian, or,  when defined by language, Armenian. And this seems to  render it difficult to accede to Mr. Rawlinson’s conclusion,  that the Etruscan language was ‘decidedly not even Indo- Germanic’ (v. ni. p. 541), especially when we perceive that  the Etruscan possesses Aryan forms of declension which are  deficient in Armenian. It is true that Mr. Rawlinson does not  consider the Armenian language as perfectly Indo-Germanic (v. 1. p. 652), regarding it, and probably with truth, as con-  taining some Turanian elements. Yet this does not, and  rightly does not, prevent him from considering the Armenians  as Indo-Germans, though he does not admit them to be either Phrygians or Thracians. Indeed the different members of, as I conceive, one race, the Thracian, are by Mr. Rawlinson  mostly separated from each other, and distributed into very  different families. The (later) Armenians are attached by him  to the Medo-Persian race (v. i. p. 676), which the further ad-  dition of the Cappadocians causes to extend from the Jaxartes  and the Indian frontier to the Halys. The Phrygians, Mysians, Lydians, and Carians are united with the Greeks ‘ 150 THE ARMENIAN ORIGIN OF THE ETRUSCANS. (ib.), and thus, it would appear, with the Latins also, so as to  form a second great race between the Halys and the Tiber. The Thracians form a third branch of the Indo-Germanic  stock (ib.): and the Etruscans are not only unconnected with  the Thracians, or Lydians, or Phrygians, or Armenians, but  also with all the members of the Indo-Germanic stock, in Europe and Asia. And this exclusion is the more remarkable,  as Mr. Rawlinson admits the Lycians, whose claim seems  much weaker, into the Indo-Germanie family, considering that  the Lycian language ‘presents on the whole characteristics  decidedly Indo-European’ (v. i. p. 668), while the Etruscan  language is.‘ decidedly not even Indo-Germanic.’ Mr. Raw-  linson’s reasons for this last conclusion seem to be (v. il. p. 541) * that it is impossible, even from the copious inscriptions  which remain (in Etruscan), to form a conjecture as to its  grammar, or do more than guess at the meaning of some half-  dozen words” This may be doubted; and, even if it were the  case, that is, if we knew substantially nothing of the Etruscan,  would it not be rather hasty to say that a language of which  we were almost entirely ignorant was decidedly not even Indo- Germanic? Suppose an Englishman were to draw a similar  conclusion with respect to Polish or Russian, in which, if  unacquainted with Slavonian, and possessed of only some in-  scriptions, he would very probably make out no more than Mr. Rawlinson -decides to be possible in the case of the Etruscan. In either of these instances, or in any other, does  the inability of an Indo-German to interpret a particular lan-  guage prove that language beyond question not to be Indo- Germanic ? ( Some of Mr. Rawlinson’s positions with respect to the Armenians seem also open to objection. The earlier Arme-  nians, he considers, were Turanians, who were succeeded by the  later Armenians, a tribe of Medo-Persian origin ; and although  the ethnic change by which an Indo-European thus succeeded  a Tatar preponderance in.Armenia was prior, as he believes, THE ARMENIAN ORIGIN OF THE ETRUSCANS. 151  to the time of Herodotus, yet the Indo-Germanic movement  which effected the change was probably no earlier than the  close of the seventh century B.c.—(v. i. pp. 652, 658). But is  this comparatively late. Indo-European movement, which must  have influenced Armenia from the east, consistent with the  fact of an Armenian and Indo-European dialect being spoken,  and probably ¢hen spoken, far away to the west of .Armenia, ‘by the side of the Tiber? At all events, the Aryanising of  x Armenia could hardly have been effected by a Medo-Persian  tribe. For the Armenian language is more nearly allied to  the Etruscan and the Phrygian than it is to the ancient Persian, either as represented by the inscriptions of Behistun  and elsewhere, or the Old Persian words collected by Bétticher  in his Arica. Nor, on the other hand, does it appear pro-  bable, as reported by Herodotus, that. the Armenians were  colonists from Phrygia. It is in Etruria, not in Phrygia, that  we find the language which most. closely resembles the Armenian. The difference between the Phrygian and Arme-  nian languages, and the distance between Armenia and Etruria, are reasons for throwing back to a time before the  memory of man the separation of the three peoples from one  another. The story of the Phrygian colony in Armenia can-  not therefore be received as historic, but must be considered as ‘an inference derived from the fact of the proximity of the two  kindred nations. Neither should I be inclined to consider Armenia as a late Thracian conquest at all, but rather as the  original seat of the whole Thracian race. * / The Perugian inscription. “To complete a survey of the Etruscan language, it may be  requisite to take some notice of the great Perugian inscription,  the only Etruscan inscription extant of any great length. It  is engraved on two contiguous sides of a block of stone, and  the words are, as will be seen, to a great extent run together. Micali gives it thus :— 152 THE ARMENIAN ORIGIN OF THE ETRUSCANS. 1. eulat. tanna. Jarezul velthinas ... 25 2. amevachriautn. velthinase atenazuk. ... 26 8. stlaafunas slelethkaru teneski ip .. . 27 4. tezanfuslert tesnsters a spelane ... 28 5. rasnesipaamahennaper thi fulumch . . 29 6. xiive/thinathurasaraspe vaspelthi ... 80 7. raskemulmleskulzwhien renethiest . .. 31 8. eskiepliulare akvelthina.. . 32 9. aulesi. velthinasarznalki akilune .... 33 10. ens. thil. thilskuna. kenu. e turunesk. .. . 34 1l. plk. feliklarthalsa/unes unezeazuk .. . 35 12. klenthunchulthe - 1. eneski. ath . 36 18. falas chiem/usle. velthina umiks. afu . . 37 14, Ainthakapemuniklet masu nas, penthn . . 38 15. naper srankzithifalstiz . a. amavelth . . 39 16. elthina. hut. naper. penezs ina. afun ... 40 17. masu. aknina. klel. afunavel thuruni. ein. . 41 18. ¢hinaslerzinia. intemame zeriunakch . . 42 19. r. knl. velthina. zia satene a. thilthunch . 48 20. tesne. eka. velthinathurasth ulthl. ichka . . 44 21. aurahelutesnerasnekei kechazwhuch . 45 22. tesnsteis rasneschimthsep rn 46 23. elthutaskunaa/unasena 24. hen. naper. kiknlhareutuse It will be readily seen that this inscription is of little use for  philological purposes. Not merely are the words run together,  but they are also frequently divided at the ends of the lines. Different readings of the inscription have consequently been  adopted, and whole or partial interpretations have been sug-  gested by the Italian antiquaries, by Campanari and Ver-  miglioli. Few such suggestions, however, seem very probable,  and many are inadmissible. I shall merely notice some parti-  cular points, availing myself of Dr. Donaldson’s opinion, that  the monument is a cippus conveying some land for funereal Ce ee Se THE ARMENIAN ORIGIN OF THE ETRUSCANS. 1538  purposes. One thing it seems allowable to take for granted,  that the inscription must contain several verbs, and those in  the third person.’ The beginning is full of proper names. Ja (1, 3), an  abbreviation of Lars, and also Lautn (2) and Afuna (8), we  know from other sources to be such. Velthina, as the inscrip-  tion shows, would be another proper name. s/, which  divides Lautn Velthinas and La. Afunas (2,8), may be com-  pared with the Arm. és¢, ‘according to, for.” Karutezan (8, 4)  is rendered conjecturally by Vermiglioli, ‘ proclamavit ;’ by Campanari, ‘ indixerunt.’? The persons, or some of them, pre-  viously mentioned in the inscription, may very probably ‘have declared’ something by it. Compare Arm. charozel, ‘to declare ;’ charozezin, ‘they have declared;’ charozezan, ‘they have been declared.’ Karutezan might also be con-  nected with the Arm. saroy?, ‘fixing, establishing,’ from which  is formed karovi-anel, ‘to fix.’ A regular verb, karov3-el,  of +" € That the Peragian inscription cannot be interpreted from the Arme-  nian would be no argument against the theory of a common origin for  the Etruscans and Armenians, even if the Etruscan had not taken up  some Celtic elements. A knowledge of Latin would not render pro- -  bable, or even possible, the interpretation of a page of Greek. There  would be a similar improbability in the case of English and German. > - Take, for instance, the first paragraph in Schiller’s Thirty Years’ Wa ‘Seit dem Anfang des Religionskriegs in Deutschland bis zum Miin-  sterischen Frieden ist in der politischen Welt Europens kaum etwas Grosses und Merkwiirdiges geschehen, woran die Reformation nicht den  vornehmsten Antheil gehabt hatte. Alle Weltbegebenheiten, welche  sich in diesem Zeitraum ereignen, schliessen sich an. die Glaubensver-  besserung an, wo sie nicht urspriinglich daraus erflossen, und jeder noch 80 grosse und noch so kleine Staat hat mehr oder weniger, mittelbarer  oder unmittelbarer, den Einfluss derselben empfunden.’ This passage is substantially English ; but, if we take away the words  which are originally neither English nor German, such as Religion,  polsttisch, Europen, Reformation, Staat, there is not very much that an Englishman would recognise, unless he were acquainted with German. Though such a word as Zeitraum is genuine English, and=¢ide-room,  yet he could hardly divine that it meant ‘period;’ nor would he be  likely to succeed much better with daraus, Wettbegebenheit, Anfang,  and several other words, which really exist, at least in their elements, in  his own language. 154 THE ARMENIAN ORIGIN OF THE ETRUSCANS.

would give karoviezan, ‘they (the dimensions of the ground)  have been fixed.? Compare also with sarut-ezan  trvpay, t.¢., Zrur-cay; and Welsh car-asant, Irish rockar-sat, ‘amaverunt’ (Zeuss, Gram. Celt. pp. 429, 497).

The next expression which deserves notice is éesnsteis rasnes (4,5). We find below, éeane (20), teeneraene (21), and, again,  besusteis rasnes (22). Campanari conjectures here, tesne (or,  as he reads it, tephne), ‘ten.’ The Arm. is fasn: the Sansk.  and Zend, dagan. If tesns=decem, then ¢ezs, it seems pro-  bable, = duo, and fesnstets = duodecum. It is remarkable that  we find xii. almost directly afterwards (6). Rasnes might be  explained from the Pers. rag, ‘a cubit,’ or from the Pers.  rasan, Arm. arasan, ‘acord;’ Alb. aréi#, ‘a measure contain-  ing three ells.’ Cf. Germ. é/after, ‘a cord, a fathom.’ The  dimensions of the burying ground in length and breadth may  be fixed by the words—  karutezan fuslert teasnsteis rasnes ipa ama hen naper XII.  duodenos cubitos, deinde simul XII. Thus, in a sepulchral inscription in Gruter (p. pcccxL), we  meet with the expression— In Fronte Pedes x11. In Agro Pedes x11.  and in another (p. pccccxLvi11)— In F. P. xxxvi. Retro P. xx.

I have just interpreted ipa (5, 27), ‘deinde,’ supposing it to  be = Arm. apa, ‘ then, afterwards, in the second place.’ Ihave  also rendered naper, ‘simul.’ It occurs four times (5, 15, 16, 24), preceded twice by hen (5, 24), and once by Aut (16). Naper may signify something like ‘thus, moreover, likewise,’  and be explained by combining the Arm. words, za-iér or  na-bar. We have in Arm.— 8a per, as. sapés, ° thus.’  da ‘this.’ = abr hipés} \ 6 as.’  na bar, ‘manner.”? hibar  nayapés, ‘ like him.’ 1 Hi seems here= Lat. hi-c. * Preserved in the plural form, barch, ‘manner.’ °¢ Gr. THE ARMENIAN ORIGIN OF THE ETRUSCANS. 155 | Na-thr or na-bar might = na-per, just as we have previously  formed the Arm. #a-imn=Phryg. vai phy. So, again, in Arm., naev (na, ‘this ;’ ev, ‘and’) signifies ‘moreover ;’ and  ovremn or na ovremn, ‘accordingly.’ As we meet with dué naper in the Perugian inscription (16),  so we also meet with Auth naper in-an inseription lately found  at Volterra. This inscription is engraved, like the Perugian,  on two contiguous faces of a block of stone; but each line  must be read completely across from face to face. I write in  capitals what are evidently proper names. One of them, Mestles, resembles MéoSAne, the name of one of the leaders of  the Mxonians in the Iliad (ii, 864).. The inscription runs  thus, as I copied it in 1857— Titesi Kale | si . - kina Ks Mes | tles  huth naper leskan  letm thui  arasa then ma  selaei tre ks .thenst me natha It may be worth while to analyse the first of these two parts  or sentences. Kina seems= Arm. fim, Gr. yuvf, and may be  compared with the /una of the Perugian inscription (10, 23). Thui seems to show the inscription to be sepulchral, as the  word is often found in epitaphs. I have explained it con-  jecturally (ante, p. 110), ‘ memoratur,’ from the Arm. thovi, ¢ it  appears, is counted.’ etm reminds us of Jethum, and also of  the Arm. /eth-2/, ‘to languish ;’ as well as of the Arm. atétch (plural form of até), ‘misfortune, calamity,’ atétali, ‘fatal,  funereal, sad’ (cf. ethalis). As the inscription is sepulchral,  leskan may be compared with the Arm. /egk, ‘a body’ (in com-  position), Kurd. lesc, Alb. dye&, Germ. Jeiche, ‘a corpse.’ We  find /eskul in the Perugian inscription (7) near another word,  tularu (8), which we may conclude from ¢tudar (ante, p. 111) to  be a sepulchral term. An Arm. verb, Jegka/, formed from /eék, 156 THE ARMENIAN ORIGIN OF THE ETRUSCANS.

would give /eskan for the 8 pers. plur. pres. ind., /egko? for the  pres. participle, and /eskeal for the past participle. Leshan  might also be an Arm. adjective formed from /eék, like  iskhan, ‘aruler,’ from 2s4h-el, ‘to rule.” Huth, in the Peru-  gian inscription Aut, may be explained from the Arm.  het, yet, yetoy, ‘after, then, subsequently.’ Cf. tr, yet, and Heb. od.

It does not appear difficult to perceive the probable con-  struction of the inscription on the sepulchral cippus of Volterra,  or even to interpret its general meaning. It seems to be of  this nature— Titest Kalesi, kina Ks Mestles Titus Calesius, (his) wife (the daughter of) Caia Mestles  huth naper, leskan. Letm  afterwards likewise, are buried (here). (Their) death  are dead.  thu. )  is commemorated. Arasa thenma, selaer treks, (A nominative), (A genitive, defining arasa thenma),  thenst me natha. ? me (the cyppus) provides. Cf. Arm. nitay, ‘ contrives ;’  nivthé, ‘forme,’ néthgé, ‘ procures.’ | Compare the following epitaphs in Gruter, pp. pccx11. and DCLXXVII.— Filii Posuerunt Memoriam Saturniano Potio Odzto Cum Compare Sua Valentiana Volusia. Corpus Hie Situm Est L. Clodii L. F. Rufini Fil. Clodii Pompeii Q. V. A. ix. M. vii. D. v. Sempronia C. F. Rufina Mater Fee. (Corpus hic situm est = Etruse. eka suthi nesl, p. 109). To return to the Perugian inscription. Zuki eneskt occurs THE ARMENIAN ORIGIN OF THE ETRUSCANS, 157  three times (7, 8; 26, 27; 35, 36). It may be a form of  comprehension. Cf. Arm. zoyg ‘ equally, together ;? zoyg, gen.,  dat., and abl., zovgi, ‘equal, alike, united, similar.’ The first  section of the inscription, the eight lines terminating with . tularu, seems to contain the names of the parties to the trans-  action commemorated by the czppus, the quantity of land con-  veyed, and a statement of the purpose to which the ground  was to be devoted.

If we make ep/ (8) a participle like es/ and sans/, and con-  nect it with the Arm. wép, ‘ frog,’ then zuki eneski epl tularu  might signify something like, ‘including the .....of the  said tomb.’ In 9, 10, we perceive the word Alensi, and in 12, klen, terms  with which we are already familiar (see ante, p. 1381). The same  may be said of muniklet (14), which we have seen before (p. 115) under the form munikleth. In 18, Velthina is a nom.,  to which kape and masu (14), the last followed by xaper, ‘also,’ may be the verbs; sape belonging to the Arm. -e/  conjugation, and masu to that in -ovl. In Arm. kapé means ‘he joins ;’ and if we unite /ape (14) with the previous word,  hintha, comparing hintha with the Arm. enth, ‘under,’ and  hinthakape with the Arm. enthadaté, ‘he suspects,’ and en-  thadré, ‘he subjects,’:we might render Ainthakape, ‘he sub-  joins,’ or ‘he enjoins.’ As we find the root mas in the Arm.  masn, * portion, allotment,’ and in the Alb. mas, ‘to measure,’  masu naper might signify ‘allots also.’ In 15, 16, 17, we  meet with velthina hut naper penezs masu, ‘Velthina then also  allots penezs ;’ this last word being in such a case an ace.  plur., which may be compared i” form with the Arm. (z)thizs,  the acc. plur. of ¢42z, ‘a span,’ and in meaning, not impro-  bably, with the Gael. peanas, “ poena.’ Cf. Gruter(p. pcccxxxv)— ... - ‘Si quis alienum corpus hic intulerit penam supra  scriptam inferet.’  and id. (p. DCCCKXV.) — _ ©Huic monumento manus qui intulerit dabit sestertios xx.’ ' 158 THE ARMENIAN ORIGIN OF THE ETRUSCANS, Aknina (17) and lerzinia (18) resemble, probably accidentally,  the Pehlvi aknin, ‘there,’ Pers. aknién, ‘now,’ and the Gael.  leirsinneach, ‘seeing.’ The termination of the mass of words,  kikndhareutuse (24) is not unlike the Arm. éovéé, ‘he fines,’ or ‘ punishes ;’ ¢ovés, ‘he recompenses’ or ‘ indemnifies :’ and sz/,  which is also found in 19, may be compared with the Arm.  gnel, ‘to buy,’ gnot, ‘buying.’ The section of the inscription,  beginning with 12, may perhaps contain such a statement of  penalties to be inflicted in case of a violation of the sepulchre,  or even its use or acquisition, as is frequently found in similar Latin inscriptions, as well as in those of Lycia.

In 28 we meet with spelane, in 80 with spe/, and in 22, 23,  with spel again. ‘To illustrate these forms, compare Arm. e/, ‘an ascent ;’ el, ‘he ascended ;’ elané, ‘he ascends:’ spand, ‘a  killing ;’ span, ‘ he killed ;’ spanané, ‘ he kills.’ We find these  last two words in a passage of the Armenian version of the Chronicle of Eusebius (Ed. Aucher, p. 327) in a manner very  like that in which spel and spelane occur in the Perugian in-  scription. The Etruscan, according to Dr. Donaldson’s divi-  sion of the words, would run thus—  chimth spel thutas kuna afunas ena hen  naper ki kn] hareutuse velthinas atena zuki eneski ipa spelane  thi(s) fulumchva spel ww ww tw ww we tl The Armenian is this— Zaridinos otompta span mayrn atéchsandri. tw Arideum Olympias iaéerfecitt mater Alexandn. Et  ena tireal § makedonatvoz ev zerkovs ordis. atéchsandri  illam imperantem Macedonibus et duos filios Alexandri  kasandros antipatreay spanané. Zam wcehn isk Cassander Antipatrs  occidit. Unum ipse quidem  span — interfecrt .

For a not improbable meaning of spel, cf. Arm. araspel, ‘ fable, apologue,’ a word compounded of araz, ‘an adage,’ arak, ‘fable, adage,’ or of their th., and of a word, not existing by THE ARMENIAN ORIGIN OF THE ETRUSCANS. 159  itself, spel, ‘a telling,’ =Teut. spel, preserved in our Gospel. Spelane might signify ‘dicit,’ and spel, ‘dixit,’ ‘ or ‘ dictum.’ Atena (26), which might be the nom. to epelane, is rather like  the Arm. atean, ‘a tribunal, a court of justice, a magistrate, a  senate,’ and may be compared with the atanisen of the Phrygian epitaph (ante, p. 34).

In 41 occurs ¢thuruni, in 6 and 20, ¢huras, an: inflexion of  thura, which is met with elsewhere (ante, p. 118). Cf. Arm,  archay, ‘king ;’ archovni, ‘royal;’ tér, ‘a lord;’ dtérovni, ‘Dominical :’ Argahovni, ‘ Arsacide, of the family of Arsaces.’ In 48 is found sechazt, though it is not quite certain that  there is a division of words after -z7. We have previously had  kecha, and in Dennis (v.i. p. 818) we find, ‘Laris Pumpus Arnthal klan kechase’ Such a variety of forms is particularly  instructive: to compare them with the Arm., we must take a  verb of the -a/ conjugation, like £hokal, ‘ to think,’ of which  the regular passive, if used, would be Ahokil. These verbs  would give us— Active. Passive. Ind. Ind.

Pres. Perf. Pres. Perf.  khokam. khokats, khokum, khokezay.  khokas. khokazer. khokis. khokezar.

(1) Ahokay. (a) khokaz. khoki. khokezav.

khokamch, khokazach., khokumch.  khokezach.  khokaych.  khokatich. khokich. khokezaych.

khokan, khokatin. khokwn. khokezan.

Subj. Subj.

Fat. Fut.  khokaytem, khokay2im.  khokayzes. khokayzis.

(2) khokayzé. (8) khokayzt.  khokayzemch. khokayzimeh.  khokayzéch. khokayzich.

khokayzen. khokay#in, 160 THE ARMENIAN ORIGIN OF THE ETRUSCANS. In khokay, khokayzé, and khokayzi, we have forms very like  kecha, kechase, and kechazi. Khoka? (a), and the fut ind.,  which would be shokasz#é, are also not unlike kechase.t The  meaning of kechazi, interpreted like kecha (ante, p. 1381), would  be ‘ shall have been paid,’ or ‘shall be paid.’ The conjugation  of the Albanian aorist subjunctive bears considerable resem-  blance to the Armenian future subjunctive. In plyak, ‘to  make old,’ it is thus conjugated :—  plyakia. or plyaktia. .  plyaks. plyakts. -  plyakté,  plyakium, plyakt&im.  plyakii. plyakt%.  plyakiine plyaktiiné. 1 T think the meaning of kechase is most likely ‘has expiated’ or ‘ has  paid,’ though the final vowel would point to the Arm. future rather than  the perfect. The inscription where it is found accompanies a procession  of departed souls and genii, of which a plate is given in Mrs. Gray’s Sepulchres of Etruria, but with the inscription not quite correctly  copied. The author seems to have been affected by the representation  with a profound interest, which her readers will probably find conta-  gious. She says (p. 211) of the person to whom the inscription would  refer: ‘This very handsome and noble looking youth is immediately  followed by a monstrous fiend, in whom we recognise the most frightful  development of the evil genius of Etruria—— One enormous claw was  pouncing upon the shoulder of the unfortunate youth, while the hammer,  the Etruscan badge of the angel of death, was raised aloft in the other. Behind him was the figure, lamentably defaced, of a female of surpassing  loveliness, and in her beautiful brow and eye the most intense anguish  was depicted. To her was attached an infernal guard, similar to the one  who had pounced upon the youth——The art of the painter had invested  these figures with the marks of individuality ; they must have been por-  traits; but whom did they represent, and why were they thus repre-  sented? What had they done, and why were they thus singled out, to  be handed down for two-and-twenty ages as the prey of demons, and  branded with the mark of reprobation?’ No light is thrown on this by  the inscription, which, like the speech of Loredano on the death of the Doge Foscari, is termbly brief—— Laris Pumpus Lars Pompeius Arnthal klan . Aruntia proles  kechase. expiant. ; solvit. Compare Dennis, v. 1. p. 309—314.

THE ARMENIAN ORIGIN OF THE ETRUSCANS. 161 In 18 is found /wsle, and in 4 fusleri. Cf. Pers. zan, ‘-yuvh ;’  zanrdé, ‘ yuval or yuvaika ;’ and also the Gael. rz, ‘ to,’ Arm.  ar,‘to.’ The Basque, again, forms the dative of nouns and  pronouns by postfixing -rz to the nominative.

Little more can be extracted from the Perugian inscription  in the way of evidence. The terminations in -/, which are  numerous in this as in other Etruscan inscriptions, I have  already compared with the Armenian (ante, p. 105), a language  of which they are equally characteristic. Some Etruscan  words, formed, like the Armenian, by excluding vowels, might  also be noticed: thus ackr aud dautn (2) might be compared  with Arm. phochr, ‘little,’ and otn, ‘foot.’ The terminations  of turunesk (34) and athumiks (36, 37) are also common in Armenian. On the whole, the result of the examination of  the Perugian inscription would not be sufficient to prove the Etruscan language to belong to the same family as the Armenian ; but it would at the same time fall in with such a  supposition, which is all that can be expected from a single  inscription of such length, and transmitted to us in such a  state as not even to admit of a complete trustworthy division  into words. If the Etruscan vocabulary, and the shorter Etruscan inscriptions, previously considered, exhibit decided Armenian affinities, the Perugian inscription will not tend to  shake, but rather to confirm, the Armenian character of the Etruscan. Geographical names in Etruria and other countries. We pass from the language of the Etruscans to a species of  evidence of the same kind, but of an independent nature, the  names of towns, rivers, &c., in Etruria. Now this evidence,  taken singly by itself, is frequently of very great weight, if  not decisive; as may be seen at once by the local terminations  comnion in our own country—Aam, ton, by, thorpe, castle,  chester, bury, ford, bridge, wich, don, stead, hill, field, wood, ley,  worth, bourn, stoke, holt, hurst, combe, dale, thwaite. These, M 162 THE ARMENIAN ORIGIN OF THE ETRUSCANS.

indeed, might be sufficient to prove that we were mainly of Teutonic origin; while the Celtic don and come, and above  all the names of rivers, such as the Thames, Tamar, Teme, Ouse, Avon, features of a country which are permanent, and not to  be made and destroyed like towns, would indicate who were  our predecessors in the land. Still more remarkable is the  confirmation of our history afforded by the Danish termination -by, t.e., ‘ village,’ which likewise deserves additional attention  here, as the presence of Thracians in Etruria will be indicated  by an exactly equivalent Armenian word.) Now the Danish  by is found in the West, as far south as Kirby in Cheshire, on  the estuary of the Dee; in the Centre, as far south as Rugby, Kirby, and Willoughby, all on the north-eastern edge of War-  wickshire; and in the East, as far south as Kirby on the Naze  in Essex. West of the Dove, and south-west of Rugby, such  terminations disappear, while in Northamptonshire and Leices-  tershire, to say nothing of more northern counties, they are  particularly abundant. Now, had our history been lost, and  had we been left to determine the extent of the Danish occu-  pancy of England by the aid of philology alone, we should  have come to a very accurate conclusion: for the English  districts yielded up to the Danes by Alfred, and which they  occupied, consisted of Northumbria, East Anglia, Essex, and  the north-east of Mercia, or the country of the Five Burghers,  so called from its five chief towns, Deréy, Nottingham, Lincoln, Stamford, and Leicester. Thus the presence of a single  geographical term would enable us to ascertain, at the distance  of nearly a thousand years, the extent to which an invading  race once occupied our country.

I have dwelt on this circumstance more particularly, because,  as I intimated, the Arm. sé, ‘village, dy,’ which forms the 1 By is Swedish as well as Danish. It is, indeed, the characteristic Scandinavian termination which corresponds to the Germ. -heim, the Eng. -ham, and the Frisian -wm. Compare Latham’s Germania of Lacitus, pp. 119—125. THE ARMENIAN ORIGIN OF THE ETRUSCANS. 163  most common. termination of Arm. names of places, seems to  correspond in Etruria and other parts of Italy to the Danish éy in England. Nor is this all. The Arm. és, or words  similarly allied to sn-e/, ‘to build,’ appear to be found in all  the intervening countries also, from Armenia to Italy. This  will be seen from the following list, in which I have compared Armenian words and names, not merely with the geographical  terms in Etruria, but. also with those in other countries, in  confirmation of what history and language alike tend to in-  form us, that there was one race, now represented by the Armenians, extending from Armenia to Italy and Rhetia. My Armenian names are mostly derived from Kiepert’s map. As I have, in consequence, not always been able to obtain  them in the Armenian orthography, some immaterial incon-  sistencies may be observed. Thus Kiepert writes Noraschén,  and I Norashén, he Lorri and I Lori, where. the Arm. would  be Noragén, and Lavri or Léri.4 But these discrepancies are  obviously of no importance. The names in the list, where the  locality is not indicated, are Etruscan..  Htruscan, &e. Armenian. (Sena... er vees nel, -‘to build, to construct, to Tur-sena or Cur-sena. make.’ - Vol-siniz. sén (gen. Sinz), ‘a habitation, a vil- Tar-quini. lage, an inhabited place.’ Sinaz, ‘a building.’ Cf. Phryg. - Synnas.

seneak, ‘lodging, chamber :’ -ak marks  a diminutive. Fel-sina * Gal. Cis eon, ‘a pillar’ Gr. xlwv. Ca-s ma P- Lith. séna, ‘a wall;’ sénys, ‘a build- . 3  ing. 1 §hén and Lori, from their signification, and the frequency of their  occurrence, would correspond to the English -ton and -ham (town and  home). Both are found in Etruria, while Germany only possesses -ham ‘ * Felsina was the Etruscan name of Bologna.

M2 164 THE ARMENIAN ORIGIN OF THE ETRUSCANS.. Etruscan, Sc.

Sar-sina Sena Gallica > Umb. | Fal-ginia Ri-cina—Picen,  nal Sean * Latiom, Mar-cina® Sinuessa } camp an.

Sa-s7na —Iapyg.

Tartu-saxa—Noric.

Senta Sinna, Chinna,  or Cinna Illyr.

Ful-senum ‘| Ol-einium San-dava® Singi-dava® Singi-dunum’— Mossia, Sane \ Dacia. Sani-ana—Thrace, Sanis or Sanaus — Synnas or Synnada  to Pisid. or Sinda | Phryg. Sin-ope—Paphlag. Phryg. Armenian.

Irish sunn, ‘ a fortification.’ The Arm. §i-el, ‘ to build,’ z.e. ‘ to  raise up,’ would be, apparently, the  same word as cio, cico, «lw, and xivéw,  and akin to the Sansk. ¢vz, ‘ crescere.’ In general, the Sansk. ¢ becomes s or  sh (8) in Arm. and« in Gr. So Gr.  xévoe = Arm. sin., and Gr. xéwy (kuv-) =Arm. govn. The same element  may thus be found in Tar-guini and Vol-sinii, in Mar-cina and Fel-sixa. It has been already seen that the Etruscans wrote both ver and sver,  tlenacheis and tlenasies. We find  also as proper names in Lanzi, vol. ii.,  vulsine (p. 885), velsinal (p. 869),  and velchinet (p, 466). Names of Towns and Villages. Senn.

Shin.

Nor-shén or Nora-shén (noratén, ‘newly built, new’). .

Hi-shén.

Bab-shén.

Hem-shin.

Gudra-shén.

Loma-shén, 2 The Volscian name of Tarracina.was Anzur, signifying perhaps ‘the defile.’ Cf. anxius, angustus, and dyxos. There is a defile at Ter-  yacina. 3 Mapxiva Tupsynrdr xriopa. Strabo. . ® The suffixes -dava and -dunum deserve notice, as the latter is a  common Celtic word, and the former would be the usual Dacian name  foratown. San-dava might be tautologous, like Hamp-ton.

THE ARMENIAN ORIGIN OF THE ETRUSCANS. 165 Htruscan, §c. Sinna Pardo-tena \ Galat. -| Singa Sacca-sena \ Cappad. Sana } Arm. (anc'.) Sinis Sinna— Mesopot. Armenian. Bashki-shé. Dada-shén. Arpa-shé. Ka-shin. Pirne-shin. Arda-shin. San-atis. Sana-hin (hin, old’). — Name of District.

Shikashén, 1. ¢. ‘red-built, red-  made,’ or simply, ‘red.’ So again  we have mezaién = mez, ‘great.’ Compare Por-sena, Ra-sena, Tyr-  senus. If we-had in Arm. the root  va, which appears in raise, rear,  rectus, we might form a word ragén, ‘high,’ and explain by its aid the Etrusc. Rasena, said to be the name  of a’nobleman. It might also be  the Etrusc. name of Pheatia, ‘the  high (land).’ It is a peculiarity of the Arm. that’  scarcely any words begin with r; but  we have rah, ‘a way,’ which, when  compared with ovti, ‘a way, and.  ovtit, ‘rectus,’ seems to contain the  required root ra. Hasena and Tyr-  senus might even be opposed to each  other, as divr.(dyr) signifies ‘a plain’  in Arm. There appears a similar  contrast in Sadint (Bret. sav, ‘a  height,’ Gr. alwewdc) and Latium (Welsh Jedd, ‘a plain’); and also in Samnium and Campania. 166 THE ARMENIAN ORIGIN OF THE ETRUSCANS. ° Etruscan, §c. Armenian, The suffix -sé has sometimes an  active sense, as in mekhenasén, ‘a  temple-builder ;’ ¢naiéx, ‘a house-  builder.’ Many have considered Tyrsenus to signify ‘tower-builder,’  though it does not appear a very  obvious designation for a people. Vol-aterres (Etruse. ) . ovttel, ‘to elevate;’ ovtit, ‘ rectus” Velathrt). The root is ovi. Fol-sinii. Ful-ginia—U mbr. Ful-sinum Ol-cinium ; Fei-sina—Gal, Cisp.. elch (plural form of eZ) ‘elevation ;’  l-atria—Epirus, el(anel), ‘to mount.’ Compare Higham, Hochhewm, Haute-  ville. Fal-erii. . 2... . wat, ‘ancient.’ Fal-eria—Picen. Compare Oldham, Civita Vecchia. Vol-aterra (Vel-athri). dth, Stharan (or avth, avtharan), Hadria—Picen. ‘lodging, dwelling.’ Atria—Venet. Gael. attreabh, ‘a dwelling.’ Ot-esia—Gal. Cisp.

El-atria—Epirus. ( Volaterra, built on } . ovitadir, ‘erect, perpendicular,’ (lit.  a Addo Uynrde, we- ‘high-placed’). Compare Mons  plxpnuvog warty. Vultur in Apulia. The Vulturnus  is in Gr. OvASupvoc.

Tar-quinti ...... dar )‘a height, an eminence, an Sar-sina—U mbr. gay? edge, a cape, @ promon- Tarra-cina—Latium.! sar tory.’  tayr, ‘a rock,’ 1 Compare with these names the frequently recurring form in the Eugubine Tables, ‘totar tarsinater trifor tarsinater tuscer naharcer  iabuscer nomner.’ The Zarsinates of these Tables might be the Tyrse- THE ARMENIAN ORIGIN OF THE ETRUSCANS, 167 Etruscan, Se. Armenian, ; The Italian names would be nearly  the same as our Clifton. Ce-sena—Gal. Cisp. . kay, kaych, ‘a stopping, a dwelling, a  seat, a place.’  kan, ‘clay, mud.’ Name of Place. Ka-shin. Cf. Hampton or Clayton. Veu, tg’ inbnrov | . weh, ‘high, great, noble.’  oxotéXov kal wih, ‘ditch, hollow.’ TEPPIPwyOe. os Vegia—Illyr.

nians, or rather perhaps the Sarsinates, who are distinguished by Poly-  bius from the Umbrians. Sarsina lay in the district called Sapinia Tribus. There may be an instance here of bilingual nomenclature. For sav signifies ‘ height’ in Breton, as sav does in Arm.; while tre, tref,  trefa, tregva, are common Kymric names for ‘an inhabited place,’ as  shén is in Arm. Thus the names, Sarsina, and Sapinia Tribus, would  be nearly identical in meaning, the one being apparently Tyrsenian and Armenian, and the other Umbrian and Celtic. So, in bilingual Bel-  gium, the capital of Hainault is called both Mons and Bergen. The  frequency in ancient Italy of the equivalent terms, tre and shén, and the  manner of their distribution, are worthy of note. The Celtic tre, tref,  tregva, treabh (Gael. for ‘tribe, people,’) is as common among the old Ttalian tribes in Umbria, Picenum, Sabina, and Samnium, as the Arm.  shén is among the intrusive Etruscans and Pelasgians. Thus we find— I quote from Cluverius :— Trea or Tretia.....+.. .. . Picenum. Treba or Trebl@... 1... 00s ZEqui. Trebia or Trev@ .....0008 Umbria (‘ pre verir treblanir.’ Eug. Tab.). Tribula or Trebula Balinensium . . Samnium or Campania. Tribula or Trebula Mutusca .... Sabina, Trebula Suffena..... pee eee ee Sabina. Tre-ventum.. 1... eee ee cceee Samnium (Welsh gwent, ‘a  plain;:’ Vente Icenorum, Belgarum, Silurum). Trievicum. . 0 ee ee ee ees Samnium (Welsh guwig, ‘ in-  closure, town ;’ Gael. fiog, ‘a wall;’ Lat. vicus). Tried. ww ew wwe . Apulia (said to have been  destroyed by Diomed). There was a town called Zyibola, belonging to Viriathus, in Spain 168 THE ARMENIAN ORIGIN OF THE ETRUSCANS. Htruscan, §e. Armenian. Lorium........ UWray or lavray, ‘a dwelling-place. Larissa—Pelasgian. Gael. Jar, ‘the site of a house.’ Laurium— Attica. In Modern Tuscany. Names of Towns and Villages. Doro (N.W. of Arezzo) Lori (2). Lari (BE. of Leghorn). Zar. Monte Loro (N. E. Lori or Lar.  of Florence). Lars—Osset. Larniano (S.W. of Poppi) ...... . learn, ‘a mountain.’ Lnerna(N.E. of Poppi). Volct ("OAktov). . . . ovtkh, ‘a ravine.’ Voleera—lIllyr." Name of District. Vulcetum—Lucan. Ovtka. Clusium (on a high hill). gtovkh, ‘ head, summit.’ Clu-ana—Picen. Gael. clog, ‘head.’ Pol. gtowa, ‘head.’ Lapp. luokka, ‘ collis.’ 3 (Appian, Iber.c. 62), and Con-é¢rebta was the ‘caput Celtiberorum.’ The  presence of the term ¢ve in Ancient Italy seems nearly to mark out the  region which the Thracian invaders failed to subdue. It is entirely  mountainous, the highest of the Apennines, ‘the Great Rock of Italy,’  standing nearly in its centre. Here the old Italian race would have  found refuge, like the Spaniards in the Asturian mountains and the Pyrenees, until in the course of time they succeeded in reconquering  the coasts and lowlands. Similarly, in our own country, as we pass into Wales, the Saxon ham, ton, and bury disappear, and tre, llan, and caer  take their place. But there are some places whose names begin with  tre, even in the modern Tuscany: Treppio, N.N.E. of Pistoja; a second Treppio, N. of Prato; Trebbio, N. of Florence; and another Zrebbio, N.N.E. of Arezzo: all about 12 miles from the cities mentioned. Such  names, however, are merely further instances of the same law of distribu-  tion; for all four places are in the highest Apennines, while Stena, Loro,  and Lari are in the lower, though hilly, country. Inthe Greek peninsula  we should find the same two races, the Thracian and Celtic; for the Pelasgians were probably Thracians, and the Greeks and Celts, as Dr. Donaldson holds (Pref. to Varron. p. ix.), ‘were scions ultimately of the  same (i.e. of one) stock.’ 1 Another Illyrian town was Pelva, which may be compared with the Peleva of Azerbijan (Atropatene), the Pelveren of what was anciently THE ARMENIAN ORIGIN OF THE ETRUSCANS. 169 Htruscan, &c. Armenian.  f Blea... ...... blovr, blrak, ‘bill, mound.’ Plera—Apul | , Name of Village. Bloor. Hasta... ce. hast, ‘strong,’ hastatel, ‘ to fortify. Gr. dorv. Pyrgos .. se. ees bovrgn, ‘a tower.’ Gr. wépyoc. Arab. burg, ‘a castle.’ | Germ. burg. Vetulonu (Etruse. Vet-  luna). ....6.5. ett, ‘a place.’ Roma—Lat. ..... ram, ‘vulgus, plebs.’ Romulea—Samn. ram, ‘an assembly, a troop.’ Romula—Pannon. The name of the Rammnes, one | iow. , of the three Roman tribes, was | Rhami-dava $ \ Dacia. said to be Etruscan. The name of Rhame—Thrace. Rome might perhaps be better con-  li In Modern Tuscany. nected with the Gael. rum, ‘a place,’ Romola (S.W. of Flo- or romho, ‘great.’ Cf. Gr. poun. |  rence) Remole (E.8.E. of Flo- ,  rence) . Cor-tona ...... . kar, ‘strength ;’ korow, ‘strong.’7—— Gyr-tona—Thessal. tovn, ‘ house, dwelling.’ Gor-tyntia—Maced. Gor-tyna—Crete. | | Cortona, &e1 ..... kertel, ‘to build, to make’ Cf. Gordium—Phryg. Tigranocerta. Gortys or Gortyna &krthel, ‘to form, to shape.’ —Arcad. _-gorzel, ‘to make.’ |  cherthovmn, ‘ wrolnpa.’ Commagene, and the Palu (Balov) of Armenia. Pala was a. town in Thrace. 1 Compare Gordona (Valteline) and Gorduno (Tessin). 170 THE ARMENIAN ORIGIN OF THE ETRUSCANS. Ktruscan, Sc. Armenian. Telamo (now Tala- . ( titm, ‘limus’ (réAya) ; th. tet(al), ‘to { mone’) { rain,’ Name of River. Ttmovt, t.e. ‘ muddy.’ Teleamo ....... teti, ‘a place;’ that, ‘ place, quarter. Gael. talamh, ‘earth, ground.’ Sansk.  tala, ‘solum, domus.’ Lat. fedlus. Name of Town. _ Thiln, probably the Thalina of Pto-  lemy. Aharna ....... akarn, ‘a castle.’ Acharna—Attica. Artena ... ees artevan, ‘summit, top.’ (d¢pde Artena—Lat. (6pove): Luke iv. 29)? Artenia—Venet. Gael. ardan, ‘aheight;’ ard, ‘ high.’ Lat. arduus. Gr. pony, apdtc. Sansk.  drdhva, ‘ altus.’ Su-ama.... ees wan, wanch, ‘house, dwelling, con- Clu-ana \ . vent.’ Cf. fanum. Picen ; Nov-ana avan, ‘ village.’ Vannia—Venet. Vannia—Rhet. Name of Town. Wan or Van. (Agylla........ git, gevt, get, ‘village, field.’ Acula, Aquila, or Aqui-  leia. Names of Places. Vir-acelum, Egil. The Arm. orthog. is Aki, the Synian, 4gyl. Cf. Acel-dama. 1 There is a Zalamona in the Valteline, a part of ancient Rheatia.

* The Latian Artena ‘is thought to have occupied the heights above Monte Fortino,’ and the Etruscan Artena to have stood at Buccea,  where ‘ there is a high and insulated point, which has all the appearance  of a citadel’ (Dennis, v. ii. p. 63, note). THE ARMENIAN ORIGIN OF THE ETRUSCANS.. 171 Htruscan, §e. Armenian. < Aquileia Gulli. Acelum \ Venet. Galli. Celia—Apul. Kili. Celeta—N oric. Esth. Ailla, ‘a village.’ There is a Cillium—Thrace. . Lithuanian village called Agila. Georg. adgili, ‘a place.’ Cales—Campan.. . . chatach, ‘a city;’) khat, ‘a lake,  shore, a marsh.’ Care... cen ee . char, Zayr, ‘rock, stone.’  kar, ‘strength.’ Names of Places. Kars. Kers. Keres. Chere. Vir-acelum ...... wer, werin, ‘ above.” Verentum wran, ‘tent, hut.’ Ferentum—Apul. Ferentinum—Lat. Verona— Venet. Virunum?—Noric. Capena........ kapan, kapanch, ‘a strait, a defile.’ Name of district, and of an Arm.  hill-fort in Cilicia, Kapan. Cosa or Cossa..... khov2? ‘a lodging.’ Cf. Eng. house. Cossa—Lucan. Name of village. | Kuash. Arretium . 2... oe Names of towne. :  Bréz. Aren, Araheza. 1 Compare the Assyrian Calah (Gen. xi. 11). * Compare Vrin in the Grisons. 172 THE ARMENIAN ORIGIN OF THE ETRUSCANS, Etruscan, &e. Armenian. Nepeta or Nepe... . . Name of Mountain. Napetia—Bruttium. Npat (Niphates), which might be  connected with zpatak, ‘ object,  mark, sight, aim.’ Cf. oxoma,  oxdreXog. Name of Plain. Npatakan. Perusia. . 5... Name of Village. Parush. Names of Districts. Perot Ormzdyeroé SpandaranyperoZ. Beroseh is a Kurdish town; Perozes  or Firouz, a Persian name, meaning ‘ victorious.’ Berta—Maced. .. . berd, ‘a castle.’ | Perta—Phryg. Pardo-sena—Galat. Ameria—Umbr. . . . amovr, ‘strong;’ amro?, ‘a fort.’ 10la— Sab. Ameriola—Sa Name of Town. Amaras, Compare Amras in the Tyrol. Gael. zmir, ‘a field.’ Hispellum—Umbr. . Name of Town and District. Ispir, Isper, or Sper. Atina—Lat..... . atean (gen. ateni), ‘a tribunal, a Atina—Lucan. court.’ Cf. Curia Rhetorum : also Etrusc. atena (p. 159), and Phryg.-A¢an-isen (p. 84). 1 Near Innsbruck. There are some places with remarkable names in  the vicinity of the Tyrolese capital——Rum, Thaur, Arzel, Vels, Vil, Pradel, Amras, Ampas, Axams, Pill, Vomp. Some are Celtic; as Rinn, Gael. rinn, ‘a promontory,’ Gr. Js: and Wiltau, the ancient Veldidena; Gael. aill, ‘noble, a cliff,’ fatll, ‘a cliff;’ didean, ‘a fort, a  sanctuary.’ : THE ARMENIAN ORIGIN OF THE ETRUSCANS. Etruscan, &c. Mesia Silva  e«oef8 @ Lacus Thrasimenus . . 173 Armenian,  motay. ‘a tamarisk.’ Cf, Lyd. pede, * 656n.’ Name of District. Thraki. Insula Igilium .. .. Atet, ‘an island.’ Gael. ighe, ‘an island.’ Fl. drno .... «+ arov, ‘a brook.’ Fl. Aro. | , Name of River. Arovn, Compare Eng. Arun, Heb. Fl. Cacma (Etrasce. Cerena) Fl. Cacinus (2), Brutt. Arnon.  sch, Sikn, ‘red, yellow? ~ Compare  flavus Tiber, flavus Mela, Xanthus, Red River, Rio Colorado, F\. Hel-  vinus in Picenum, Fl. Clanis or Glanis. glel, ‘to roll;’ glan, ‘a cylinder) i Les Fl. Clanis  or Glanis, Lat.  afterwards Cc ampan. Liris.

Fl. Clanis, Gla-  nis, Clanwus,  and Glanius,  also called Zi-  ternus, Cam  pan e Gurgures Montes Garganus Mons  si . \ pol. ' glan, ‘ what rolls,’ The name Jzris, when compared - with the Welsh dlyry, ‘ what glides or  flows, a stream,’ seems like a trans-  lation into Celtic of the Tyrsenian Glanis. ‘ But there was also a river ~ Glanis in Spain, so that the Gael. ‘clean,’ might be suggested as  explaining the name. In Switzerland  we have the river Glane in Freyburg,  and Glenner in the Grisons, There  is a Glan in Carinthia; a Glon in Bavaria; and a second Glan divides Rhenish Prussia from the Palatinate.  char, ‘a stone, a rock.’  karkar, ‘a block of stone.’  kharak, ‘ a rock,’ 174 THE ARMENIAN ORIGIN OF THE ETRUSCANS. Etruscan, §c. Armenian. _| Gargara Names of’ Places. Gergetha i Mysia Gurgur. Gargetins—Attica.  Gerger.

. Gargarlyk.

Gergeti—Osset.

Etruscan. Celtic, Populonia ...... Gael. pubull ; Welsh, pabel? ; ‘a tent, . a booth.’ Compare Peebles, Shields, Succoth.

Blera wc. eees Gael. dlar, ‘a field’ (appears in Blair - Athol). | Care... were Welsh caer, ‘a wall, a castle, a city.’ Umbria (district) . . . Gael. smer, ‘a ridge of land, a field.’ Fl. Umbro ...... Gael. amh, ‘the ocean ;’ amhainn, ‘a  river ;’ amar, ‘a trough, a channel.’ Cf. Arm. ovmp, ‘beverage ;’ émpel, ‘to drink ;’ amp, ‘a cloud ;’ aman, “a veasel:’? also Lat. amis, and Sansk. ambu, ‘ water,’ Fl. Tiber... 00 Gael. fobar ; Irish tobar, trobar, tidhir ; ‘a well, a fountain, a source, a  spring of water.’ Ttobar Seaghsa (seaghas, ‘a wood,’) was an old  name of the River Boyne. Another  name for the Tiber, and one said  to be more ancient, was the Aldula, In Arm., atéivr signifies ‘a foun-  tain, a spring, a source, a stream,’  and therefore = Irish ¢zd/ir. There  was another Al/juda-in Picenum. We have also the ‘A/bunea resonans’  at Tibur, now Tivoli, I have pre-  viously noticed how another river  eS OE Eee ee eT ee eee ee ee SS THE ARMENIAN ORIGIN OF THE ETRUSCANS, 175 Ktruscan, Celtic.  bore at one time the name Glanis (perhaps = Arm. glam), and at ano-  ther the name Zzris (apparently = Welsh fyry). The root of atéiwr  is ath, signifying ‘ discharging,  voiding, flowing.’ Cf. alvus, alveus, Alpheus ; and also Hive and the Swed. e//, ‘river.’ Two other fa-  miliar Celtic names of rivers in Italy deserve notice—the Zamar,  or Zamarus, in Samnium, and the Tyne, or Tinta, in Umbria, and Tinna in Picenum. - A second Pi- . centine river was the 7ruentus or | Durance (Druentia).

Insula Jwa...... Gael. i/bhinn, ‘a craggy mountain.’ Name of Island. Ulva, one of the Hebrides. The Pelasgians. L (; It. has been mentioned at the outset that the Etruscans,  according to the accounts of the ancients, were probably allied  to two nations, the Lydians and the Pelasgians. The investi-  gation of the first of these affinities, and of the consequences  deducible from it, has occupied us up to this point, and led to  the conclusion, involved in the supposition of the Lydian  affinity of the Etruscans, that one of the great families of the Aryan stock, a family whose branches spoke dialects akin to  the Armenian, now the only surviving dialect of that family,  extended in ancient times from Armenia to Etruria. If this  be true, and if the Etruscans, at the same time, were akin to  the Pelasgians, then the Pelasgians ought also to form part of  this family, and be in a similar manner allied to the Armenians, 176 THE ARMENIAN ORIGIN OF THE ETRUSCANS. Now, if the inscription of Cervetri be Pelasgian, this inference  might be considered as proved. If, however, the Pelasgian  character of this inscription be not admitted, then it will be  necessary to examine what few points, very few indeed, may be  relied upon as characteristic of the Pelasgians. These points  may indeed, I think, be limited to three. The Pelasgians  called cities by the name of Larissa: they possessed and  founded the oracle of Dodona: and their name, Pelasgz, as 16  does not appear to admit of explanation from the Greek, is  probably itself Pelasgian. To these three points I shall con-  fine my attention. » 1. Larissa. This was the ancient name of Nimroud, as Xenophon mentions. He found it in ruins, but said that it  had been once occupied by the Medes. In Armenia we find  towns called Lori or Lauri, and also Lar. In the Armenian  language, Jéray or Javray means ‘a dwelling.  Larel also  means ‘to set up;’ /earn, ‘a mountain;’ and /eruak, ‘an  eminence.’ We find a form like Larissa in the Armenian Arsisea, the native orthography of which is Argés, a name not  unlike the Thessalian Argissa. Salmyd-essus, Pan-issus, and Scot-ussa, are Thracian towns. In the countries near Armenia  there are also towns with names resembling Larissa, In the Ossetic country there is Lars. Near Trebisond is Laros. In Western Persia are Lar and Lour. JLarsa is the name of a  city in the Babylonian inscriptions.

Such terms are not, however, exclusively Oriental. In Gaelic, /arach signifies ‘habitation, the site of a house,’ aud  lar, ‘ground.’ Lapp. /aire and Swed. /er signify ‘clay. In  the north of Germany we find Goslar, Wetzlar, Fritzlar. Lahr and Lohr are also German towns.

2. Dodona. The primitive mode of divination at Dodona  was peculiar— ‘The manner in which the oracles were communicated was  originally extremely simple, by the rustling of an aged oak or  beech.’! 4 Ersch and Gruber, Encyc. s. v. Dodona.

THE ARMENIAN ORIGIN OF THE ETRUSCANS. 177 ‘The god revealed his will from the branches of the tree,  probably by the rustling of the wind, which sounds the priests . had to interpret:”! That the same mode of divination was anciently practised  in Armenia, appears from the following passage in the History  of Moses of Chorene (lib. i. c. 19). I give it in the Latin of  the Whistons. A certain Anusavanus is mentioned, ‘ qui Sosius appellatur, quippe qui pro eorum cerimoniis apud Armenaci’ cupressos in Armaviro consecratus fuerat ; quarum  cupressorum surculis (according to Aucher, /foliis), ramisque  seu leni sive violento vento agitatis, Armenii flamines ad  longum tempus in auguriis uti consueverunt.’ The Armavir here spoken of, the Armauria of Ptolemy,  was the most ancient capital of Armenia, ecclesiastical as well  as political, and was reported to have been founded by Armzis  the son of Armenak. The word which the Whistons have  translated ‘ cypress,’ 86s, is rendered ‘ poplar’ by the Armenian  writer, Avdall. Aucher translated it ‘ plane,’ but refers as a  synonym to satamakh, to which he assigns the meanings, ‘poplar, aspen, beech, elm.’ Sog, again, means ‘a pine,’ = Pol. sosna. The real meaning of sés would be ‘shaker,’ as is  evident from the words, sdésaphil, soskal, and sasanil, ‘ to shake’ (neuter). The Arm. has also théthaphel, dedevel, and tatanil, ‘to shake,’ in which the sibilants of sésaphil, &c., have become  hardened. Any tree might, in fact, be called ‘shaking’ or ‘waving, in Arm. ¢afan; and we might perhaps even derive  the name of Dodona from a similar root, and give it the appro-  priate sense, ‘ civoolpudXoe.’ 8. Pelasgus. This word is usually resolved into Pel-asgus.  e¢——___ SSS 1 Smith, Dict. Antig. s.v. Dodona. Cf. Odys. xiv. 327— Tov S és Awddvnv ddro Biuevar, Eppa Oeoto "Ex Spuds byixduora Adds Bovdhy éwaxovoat, 4 The eponymus of the Armenians, and, according to their traditions,  the son of Haik or Hayk (the Armenians call themselves Haych), the  son of Thorgom, the son of Thiras, the son of Gamer (Gomer), the son  ef Yabeth (Japheth). . , to, oo, N 178 THE ARMENIAN ORIGIN OF THE ETRUSCANS. ‘ Buttmann suggested long ago that the last two syllables were  an ethnical designation, connected with the name Asca-nius,  common in Phrygia, Lydia, and Bithynia, and with the name  of Asia iteelf.’! This root, as-, supposed to exist in 4s-7a,  might easily be the same as the Arm. root, az, which appears In az-g and az-n, both signifying ‘nation,’ and the latter  perhaps=Gr. &vog. Ascaniue might be compared with the Arm, azgayin; ‘ national, allied.’ For the second element of Pel-asg we thus get the Arm. azg, ‘ nation, race, family, tribe,  people,’ a word which I have previously supposed to enter into  the formation of the Etruscan Matulnask, the Matulnian gene  or familia.

There remains to determine Pel-. Now what race (az)  were the Pel-asgi? They were the old race, the Aborigines,  or ancient inhabitants, of Greece and the country to the north. This leads us to connect Pel- with the Arm. wat, ‘ ancient,  old,’ =Gr. waXd(atde),=Epirot wédtoc,=Alb. plyak? The  meaning, old-race, wat-azg, seems the best that can be given  to Pel-asgi. Strabo (p. 220) says of the Pelasgians, that it  was almost universally acknowledged, érix piv apyaiov re  pvAoy xara trav ‘EAXada wacay érerddace: and again (p. 827), of d& TleAXacyol ray rept riv ‘EAXada Suvacrevodvrwy  apxaérarot Afyovrat. Dionysius (lib. 1. c. 17) speaks of the Pelasgians as avrdéyovec, we of woAXolt rept avrwy Aéyoust. And Pausanias again, in the commencement of his Arcadica,  notices the Arcadian tradition, that Pelasgus was the first  that lived in that country: gaot 62 “Apxadse, we HeXaaydbe  yévorro tv rG yg ra’ry mpwrocg. Pausanias has considerable  difficulty in accommodating-this statement to the history of  his king Pelasgus: wolwy yap Gv xal ijpxyev 6 Wedacyic  avipwrwyv; but, if we interpret IleAaoyoe ‘the pel-asy’ or ‘old-race,’ then the Arcadian tradition is reduced to a simple  truism, With Virgil (2. viii. 600) the Pelasgi are veteres. 1 Varronianus, p. 39. 2 The Latin, it is to be observed, has no corresponding term to rada:és, THE ARMENIAN ORIGIN OF THE ETRUSCANS. 179 . Though the name Pedasgi, if its explanation from the Armenian were admitted, would thus be Thracian, yet it  would not necessarily follow, though it is probably true, that  every nation called Pelasgian was of Thracian origin: for  the Thracians may have ealled any aboriginal race Pelasgian,  whether of their own or of any other family. There would,  however, in any case, have been Thracians in the country to  give to the race the name Pelasgian.

The Arm. azg is not found, in that language, combined into  one word with wat ; but it does appear, so as to form words  like Pelasgus, in combination with several] adjectives.

1. With ayi, ‘other, different, but” Cf. a\do¢ and adda: Aylazg, ‘ different, various.’ - Aylazgi, ‘a foreigner, a stranger, different, adAoyerne (Luke ; xvii. 18), Eévog (Matt. xxv. 35), adAdrpiog’ (John x. 15).

2. With avtar or Otar, ‘a foreigner, distant, other” Cf.  abrap, arap, frepoc, and the name of the Illyrian Autariate : Otarazgi, ‘foreign, a foreigner. An equivalent term,  formed with azn, is dtarazn, = EreposSvhc. As Otarazgi = ‘ he-  terogeneous,’ so hamazgi = ‘homogeneous.’ 4. With dav, ‘ fine, good, better.’ Cf. Aywv, Agoroc: _ Lavazgi, ‘noble, of good family.’ Several Etruscan names begin with /au- or lav-. Thus Laukin or Lavkin is a woman’s name, appearing in the forms, Laukin, Lavkinal, Laukine, Lavkinasa. In Arm., lav-kin = ‘good-woman.’ Another Etruscan name is Lawin, which  might be compared with the Arm. /av-tovn, ‘ good-house,’ or ‘good-family.’? A third name is Lauchme, supposed to be the Etruscan form of Lucumo. ~ Possible extension of the Thracians to the west of Etruria. ( That the Thracians should have extended their settlements  beyond Etruria, especially during the time of the maritime N2 180 THE ARMENIAN ORIGIN OF THE ETRUSCANS.

power of the Tyrrhenians, is not impossible. } We know indeed  from Strabo (p. 225), that there were Tyrrhenians in Sardinia. But there are even some indications, though they may be few,  and undeserving that any great stress should be laid upon  them, of the existence of Thracians still fyrther to the west. It has already appeared (ante, p. 42), that(the Spanish words  pandero and pandorga would be allied to the Arm. phandirn  the Lyd. wravdotpiov, the Osset. fandur, and the Assyrian  wavooupa. ( The name of a glacier in the Pyrenees seems  also to be Armenian; while in the Alps, on the other hand,  such names are apparently German, Finnish, or Celtic, though  perhaps in one case Arabic, and brought from Asia by the Thracians. } The following list of names of glaciers, terms  which may be of importance in a question of ethnology, is  taken from an article in the Hncyclopedia Britannica, contri-  buted by our great authority on the subject of glaciers :— Tyrol, fern or firn. Simler (De Alp. p. 74,) rightly says  that jfirn means ‘old.’ It is the neve,  or old snow. The word is German,  being the Goth. fairnis, ‘ old,’ = Ang.- Sax. firne, fyrn,=Germ. firn, ‘of the  last year.’ (Cf. Diefenbach, s. v.). Firn is therefore not an ancient word  in the Alps, but one carried in by the Germans. It is remotely connected  with the Asiatic languages. airnis  has the sense of the Sansk. pérva, Kurd. gira, = Arm. parav; and the Germ. jirn, that of the Arm. herov, Sansk. parut, Gr. wfpvot, Osset. fare. Carinthia, dass (also Lapp. Aaisse, ‘mons altior, plerum-  written sees and que nive tectus.? Esth. kahho, ‘frost ;’  hkése, and used in kasse jia (jaa, ‘ice’), ‘ice formed by Salzburg). frost upon snow.’ Georg. giswa, ‘ frost.’ Vallais, diegno.. — Ital. dcanco, = Fr. blanc, = Span. blanco, THE ARMENIAN ORIGIN OF THE ETRUSCANS. 18] Italy (part of), ve-  dretto. (Add the Grisons, as ve-  dretta _ signifies ‘glacier’ in Rhzt-Rom.) Piedmont, ruize, Pyrenees, sernevtle. =Eng. and Germ. blank. O. Norse  blanka, ‘nitere” | Welsh Glanu, ‘to  manifest.’ Arm, pat, ‘frost, ice.’ Arm. eteamn, ‘ frost, rime.’ Compare with diegno the Val Blegno in Canton Tessin.

Gael. eithre, erdhre, eighre, eith, eigh, ‘ice.’ Compare Lat. vitrum, = Rhet- Rom. veider, and the Val Bedretto in Canton Tessin.

Another Rhet-Rom. word, besides  vedretta, 18 samada or samadra, ‘ hard-  frozen snow:’ and there is a mountain  called the Piz Kamadra at the origin of  the Val Blegno, the head of which valley  is called the Val Kamadra, as the head  of the valley of the Tessin or Ticino is  called the Val SBedretto. Cf. Kurd.  gemet, ‘ice;’? Arab. gamd, ‘ice, concre-  tion,’ gumid, ‘congealing.’ Gael. reoth, ‘to freeze.’ Lat. rigeo, Srigus. - Arm. sarn, ‘ice, frost;’ saril,. ‘to  freeze.’ Lith. szdlu, ‘ to freeze,’ szdina,  szarma, ‘hoar-frost.’ Pers. sard, ‘ cold.’ Serneilie appears to he foreign to Basque, Celtic, and Finnish, and would  probably be a diminutive, like another Pyrenean word, hourquette, ‘a pass,’ = Rhet-Rom. furcletta, both being obvi-  ously diminutives of /furca, like the Fr. JSourchette, Ital. forchetta, Span. hor-  quilla. In this last word, as in the  ital. sor-ella and the Fr. ad-eille, we  may recognise the termination of .sern- 182 THE ARMENIAN ORIGIN OF THE ETRUSCANS.

eille. The remainder, sern, certainly  appears = Arm. sarn.

It is rather singular to find in the Pyrenees a word ap-  parently allied to the Armenian. Yet there was anciently in  the Eastern Pyrenees a nation called Bebryces, whose name  appears also in Asia Minor as that of a Thracian people. These Pyrenean Bebryces seem to be distinguished from the Celts, Ligurians, and Iberians, just as the word sern-eille seems to be  neither Celtic, Finnish, nor Basque. Several classic authors  have mentioned the western Bebryces. -The writer who passes  under the name of Scymnus Chius, after noticing the Celts,  the Carthaginian settlers in Spain, and the Tartessians, thus  proceeds (v. 198) : | dr’ "IBnpeg ot TIpocsxetc. “Avw robrwy 8? xetvrae rwv rérwy BéBpuxeg. “Erera wapadadarrio xatw Alyveg Eyovrat, cal réAeg ‘EXAnvidsc, “Ac MacotXwwrat, cat Pwatic aTwxicay" Ilpwrn piv "Eurdépiov. ‘“Pddn 8% deurépa.

Steph. Byz. says— BePpbxwv Evy db60. 1d piv wpdg rey dvr év rp Aala.  ro 02 mapa toic "1 Bnpoty év rp Bupemy.

And Sil. Ital. (v. 417) : Pyrene celsa nimbosi verticis arce Divisos Celtis alte prospectat Iberos, Atque sterna tenet magnis divortia terris. Nomen Bebrycia duxere a virgine colles.

And Tzetzes in Lycoph. Cassandram : Alwy 8? Koxxetavoe rode NapBwvnciove BEBpuKac Aébyet, Ypaguy ovrwe’ rd waAar piv BeSpbxwy, viv d& NapBwvnolwy éori 76 [lupnvaioy dpo¢. (This does not show, as Tzetzes seems  to argue, that the Bebryces ever possessed the district of Narbonne).! 1 Bebriz is the name of a gladiator on a Ponfpeian bas-relief. THE ARMENIAN ORIGIN OF THE ETRUSCANS. 183 These quotations are giver. by Bouquet. If there were a Fhracian element in the population of Europe as far west as  the Eastern Pyrenees, we might perhaps expect to find such an  element in the country between those mountains and Etruria,  t.e., among the Ligurians. With respect to Liguria, however,  there is scarcely any evidence. Yet the Piedmontese autzn and  the Etruscan atesum, both signifying ‘a vine,’ seem allied. Dertona, again, might be compared with the Alb. dert-o2g, ‘TI  build :’ and there has been found (Lanzi, 1. p. 562) at Busca,  between Cuneo and Saluzzo, at the foot of the Alps, the fol-  lowing Etruscan inscription—  mi sutht Larthial Muthikus. The ‘locus Gargarius in finibus Arelatensium’ has, again,  an Armenian as well as a Mysian sound (anée, p. 173): and  the same may perhaps be said of the river Atax or Attagus (the Aude), a name which resembles the Arm. yatah, ‘a channel,’  and the Lith. attakas, ‘a small water-course.’ Conclusion. ( Thus then, not only in Etruria, but also in all the inter-  vening countries, and wherever the ancients have placed a Thracian  people, the language of Armenia is to be found. With regard to  thereligion and manners of the ancient Armenians, which might  afford subsidiary arguments, should any be requisite, in support  of the conclusion to which the extension of their language leads,  but little information can be given. It has already been noticed,  that the types of Etruscan sepulchres may be found in Armenia,  and also that the Dodonzan mode of divination was said to be  practised from a remote period in the same country.! Respect-  ing the religion of the ancient Armenians, as it exfsted in the  time of Strabo, we learn (p. 582) that, in common with the Medes, they practised the same worship as the Persians, but  were especially devoted to the goddess Anaittis, in whose loose  rites Strabo traced a resemblance to the manners of the Nee EE a 7” ees - e . 184 THE ARMENIAN ORIGIN OF THE ETRUSCANS. Lydians, as described by Herodotus. Axattis, who is usually  called Artemis or Diana, was properly the Bona Dea, like the Lydian Diana, who was also known, as I have before stated (p. 24), under the name of Anaiztis. She was also, probably,  the Assyrian Anata. Other Armenian deities were—Aramazd,  the same as the Persian Jupiter, Ormuzd; Wahagn, ‘ Hercules;’  and Spandaramet, ‘ Bacchus.’ This last name signifies ‘holy  origin,’ as interpreted from the Zend gpenta, ‘ sanctus,’ and Arm. armat, ‘origo.’ From this deity is derived the name  of the twelfth month of the Persians, asfendarmed, in Zend  gpenta drmaiti, and in Cappad. sondara, the name of the Cap-  padocian Hercules. (See Arica, p. 7). The old religion of the Armenians, it seems probable from such evidence as we can  collect, would have partaken of an Assyrio-Persian character. That such was the character of the religion, and also of the  arts of Etruria, seems sufficiently borne out by Etruscan  monuments. It is to prove such a resemblance, which he  extends to Egypt as well as to Asia, that the veteran Etruscan  archxologist, Micali, has devoted so much time and labour. He laid down this principle, with respect to Egypt, as early as 1810, in his [tala avant: 1 Romani ; followed it up, in 1832,  with an extension to Asia, in his Storia degli antichi popoli Italiani ; and finally, in the year 1844, published his Monu-  menti Inediti, in order completely to substantiate his position,  and to show, by comparing the Etruscan monuments with  those of ancient Persia and Egypt, that they were ‘sufficient  in themselves to demonstrate, that the civilization of Asia,  from an early date (di lunga mano) and in various forms,  exerted a preponderating influence upon that of Etruria’ (p. 5). Had the Assyrian, and also the Phrygian and Lycian, disco- ' veries been made when Micali wrote, he might have still  further illustrated his case. Some instances are noticed by Layard, Nineveh and Babylon, pp. 189, 190. The strange  animals, genii, sphinxes, and monsters, frequently winged and  sometimes double-bodied, which abound on Etruscan monu- THE ARMENIAN ORIGIN OF THE ETRUSCANS. 185  rnents, seem almost as if invented in Persia or Assyria.” The Eastern doctrine of two principles is also a marked feature in Etruscan designs ; and considerable resemblance may, I think,  be discerned between the early style of Etruscan sculpture and  that of Assyria and Babylonia; much more, indeed, than  between such works in. Etruria and Egypt. But these are  points which can only be fully elucidated by an actual eom-  parison of the different monuments in question, which may be  made with sufficient completeness, except in the case of Assyria  and Asia Minor, by the aid of the Monument: Inediti. Sir Charles Fellows’ Lycia and Asia Minor, and Steuart’s Monu-  ments of Lydia and Phrygia, should also be compared, especially  in their illustrations, with Micali and Dennis,} | Although the Asiatic character of the Etruscan institutions  seems thus well established by Micali and others from the  remains: which the Etruscans have left, yet the mode in which Micali chiefly explains that character, even in his latest work,  may admit of improvement. It is his opinion (p. 419) that  the bulk (22 geno) of the Etruscan nation was indigenous in Etruria (natio del lwogo, o altrimentt indigeno), and that their Asiatic manners, arts, and religious ordinances were communi-  cated to them mainly through the influence of a predominant .  sacerdotal caste (prepotente teocrazia), belonging to some of  those tribes which, ‘directing their movements continually  from the S.E. to the N.W., traversed in succession vast  regions from the mouths of the Ganges to the Atlantic Ocean,  oe eee and extended from place to place a Southern Asiatic  element as far as our (2. ¢. Italian) western countries’ (p. 4). But the evidence of language, so closely agreeing, and in so  many nations, with that of history, will lead to a simpler ex-  planation of the Asiatic customs and monuments of Etruria,  by exhibiting the Etruscan nation as a member of the western 1 The English reader may likewise consult an article on ‘ Etrurian Antiquities,’ in the Quart. Rev., yol. liv. See also +b., yol. Ixvii. p. 375. 186 THE ARMENIAN ORIGIN OF THE ETRUSOANS.

branch of the Asiatic Aryans; a branch which had its original  seat in Armenia, a country bordering on Assyria, Media, and Syria, and whose inhabitants were therefore in a position to  acquire those rites, arts, and practices, which we find exem-  plified in the monuments of Etruria. If the ancestors of the Etruscans came from Armenia, we need hardly seek any other  explanation of their national character and religion, as they  are laid open to us in the Etruscan remains, than what the  situation of their birthplace of itself affords. A race of Armenian origin was the most adapted of all the Aryans to  transport the rites and manners of the countries on the Tigris  and Euphrates to the banks of the Tiber and the Arno.! Neither is there anything improbable in the hypothesis, that  a single race once occupied the countries between Nineveh and Rome. It is merely assigning to the Thracians such an area  as the Indians or Persians still possess in Asia, and the Germans or Slavonians in Europe ; and such as we also know  was anciently occupied by the Celts, whose language is now  confined within limite of nearly the same extent as that of the Armenians.” é 1 According to the Armenian account of their own origin, their pro-  genitor Haik, the son of Torgomah, lived at first in the country of Shinar, in Mesopotamia, with his three sons, Armenak, Manavaz, and Kore. Cadmus and Armsis were sons of Armenak. Retiring from Mesopotamia to avoid the power of Belus, Haik came to Armenia, and  founded the town of Haikashén, ‘the dwelling of Haik.’ The inha-  bitants of Armenia, then in a rude state, submitted to his authority. Belus, invading Armenia, was defeated and slain by Haik; but the Armenians, after being long continually at war with the Assyrians, were  eventually conquered by Semiramis, and remained for several ages  under the supremacy and influence of Assyria, until the fall of that  empire with Sardanapalus.

8 Some observations of Zeuss are here deserving of attention, espe-  cially as they will show how I was to some extent anticipated in my  theory more than half a century ago by the author of the Mitkridates.

‘** Aus der nicht unbetriachtlichen Ausdehnung der Thraker von Ma-  kedonien bis nach Vorderasien und nérdlich bis an der Ister folgert Herodotus (v. 3): Opytcwr 52 Edvos péyoréy dori, werd ye "Ivdovs, xdvrwr  dvOpirwv, Aber hatte er gewusst, dass die Cappadoker, Syrer, Phéniker, Palistiner, Babylonier, Araber, alle eines Stammes seien, woran die THE ARMENIAN ORIGIN OF THE ETRUSCANS. 187 Alten, welche die Sprachen der Volker nicht beachteten, nicht gedacht  haben, oder hitte er einige Jahrhunderte spater geschrieben, und  erfahren, dass von den Katarakten des Isters oder von Kleinasien bis  nach Jerne und Thule ein Volk wohne von gleichen Sitten und gleicher Sprache, so stande diese Behaupting nicht in seinem Buche.” (Die Deutschen, p. 259.) ‘* Wenn einzelne illyrische Volker bei einigen Schriftstellern Thraker,  und umgekehrt thrakische Vélker Llyrier genannt werden, so kommt  dieses theils aus der spateren rémischen Provinzabtheilung, nach wel-  cher der Umfang dieser Namen willkiirlich erweitert oder verengert ist,  oder durch Irrung der Schriftsteller selbst, von denen dann eben so zu  urtheilen ist, wie von Florus, wenn er die Skordisken, deren Abstam-  mung doch bekannt ist, Thraker nennt, und von Appianus, dem dieselben Iilyrier heissen. Solche Irrthiimer kénnen freilich auch nur irriger Behauptungen Stiitzen sein, wie sie Adelung gedient haben, der die Sprachen der Volker im Siiden des Isters und selbst noch kleinasiati-  scher und italischer in einem ernzigen Sprachstamm, den er die thraki-  schen benannte, zusammenfasste, wie einmal Gelehrte die Sprachen der Nordvélker in den keltischen. Thunmann erinnert an zwei illyrische  mit dem thrakischen dava zusammengesetzte Ortsnamen, Thermidava  bei Scodra in Ptol., und Quimedava (cf. Arm. wém, gen. wimi, ‘a rock’)  in Dardania bei Proc. Dadurch sind aber die Illyrier noch nicht Thraker, so wenig wie die Ligier Kelten, wenn bei ihnen ein Ort Aovyl8ouvoy heisst”” (Jb. p: 250). It is probable that the Ilyrians were,  and were not, Thracians, in the same manner as the English are, and  are not, Germans. Where Adelung was in error seems to have been  in two points: he considered the Greco-Latins as forming with the Thraco-Illyrians one great race (Mith. v. ii. p. 330 et sgg.); and he  made the Etruscans Celts (p. 455) instead of Thracians, among which  last, I believe rightly, he reckoned the Pelasgians (p. 369). Of the con-  nexion between the Thracians and Armenians he speaks but doubtfully (vol. i. p. 410; ii. p. 422). APPENDIX.

As specimens of the Armenian language may be useful for  the purpose of comparison, I have subjoined the Armenian  version of the first five verses of the seventh chapter of the Acts of the Apostles, the traditional account of the origin  of the Armenians from Moses of Chorene, and the Lord’s Prayer. Acts yu. 1-5. 1. Asé (‘dicit”) chakanayapet(n), ethé ardarev ayd aydpés Etre 62 6 apxtepeve, si dpa ravra otrwc  t2é (‘an juste hoc ita sit’) : EXEL} 2. Ho na asé, arch etbarch ev harch, lovarovch inz: ‘O 88 %gn, dvdpec adeAgol «al warfpeg, dxobaare (uo).

astovaz phara? erevezav hér merovm (‘nostro’) abrahamov, 5 Gede rig O6Eng = HOSn re rarpi Hor "ABpacu  minéder (‘dum-adhuc’) 7s migagets é (‘erat’), minééev (‘ ante,’  ovre éy ry Mecomnoraplg, apy ij  lit, ‘dum-non-et’) dnakeZov%eal (‘xarotoag’) é& (‘erat’) zna  karouknoat avroy 4 kharan, ev asé na, évy Xappay, cal ele wpdc avbroy, 8. El yerkré chovmmd (‘ tua’), ev yazgé choommé "EEeASe ex rig yg cou, kat ék Tig cuyysvelag cov, (‘tuf’), ev ek (‘veni’) yerkir zor zovtiz chez;  kat Sedpo ele yiv iv av SelEw cot. 190 ° APPENDIX. 4. Yaynéam (‘in-eo-tempore’) eleal yerkré(n) chatdeazvot Tére gEeASa@v ee yg Xaddalwv  bnakezav *t kharan. evanti yet meraneloy hér (‘patris’)  karwKnoey ty Xappav’ xaxeiJev pera ro aroSaveiy roy waripa  nora (‘sui’), phokea? pandkhtezoy? zna =yerkri(s) yaysmik,  avroi, petepxioey (relegavit) atrév sig rny yy (sic) rabrny,  yoroum dovch ayém (‘ hoc-tempore’) dnakeal (‘ xarouxhoartec’)  sic Ny Ug =u KATOLKELTE. éch (‘ estis’) : 5. Ev of et mma sarangovthivn *t sma, ev o0é chayl Kat ovx Edwxev airy KAnpovopulay tv ary, ovd? Biya  mi otin, ev khostazav tal nma za 4 bnakevthivn, (@v) wodd¢’ cat érnyyelAaro Sovva airy abriy sic xardoxeotv  ev zavaki nora yet ora:  kal Tq ortpuart avrov per’ avrdy. Moses of Chorene, c. ix. * O¢  kameteal: Hayk(ay) hnazand lintl Bélay, yet -. Not having willed Haicus obedient to be to Belus, after  gnaneloy (z)ordt . ivr (z)Armenak 74 Babeléni, éov .  begetting son his Armenacus in Babylon, a journey ' ararea gnay yerkir(n) Araraday, or é& 't kotmans  having made goes to the land of Ararat, which is in the parts  hivsisoy, handerz ordvowch iwrowch ev dsteréch ev ordvod  ofthe north, with sons his and daughters and of sons  ordvowch, arambch zéravoréch, thovow. rbrev 800, ev aylowch  the sons, men mighty, in number about 300, and others ‘éndozgnéch ev  ekbch yarezelowch ’i na, ev bolor  home-born and strangers joined to him, and with the whole  atkhiv — ertheal bnakE 4 lernott mioom 74  band having come he dwells at mountain-foot one in APPENDIX, 191  dastawayri,  yoroum sakavch *1 mardkané yaragagoyn,  a@ plain tract, in which few out of mankind previously, Brovelo#(n) dadareal, bnakéin, (z2)ors hnazand wr  scattered having abided, were dwelling : whom obedient to him  arareal Hayk, &siné and toun bnakovthean  having made Haicus, he builds there a habitation of dwelling  kalovazoz ev tay "+ tarangovthivn Kadmeay  ordvoy  of possession and gives for au inheritance to Cadmus the son Armenakay: (ays ardarazoviané (z)angir hin asazeal  of Armenacus, (This confirms the unwritten old _ said  zroyzs :) Ev inchn khatay (asé ) aylow  story.) And he (Haicus) proceeds (says Maribas) with another  atkhiv(s) énd arevmovts hivsisoy, gay, bnaké *i barzravandak  band to the west of north, comes, dwellsin elevated  dasti miovm, ev anované (z)anovn lernadastaks(n) Harch,  plain one, and calls the name of the table-land ‘ Fathers,’  ays inchn astén bnakealch(s) azg "4 tann Thorgomay.  i.e. here having dwelt the race of the house of Torgomah: Sin€’ ev gevt mi, ev anovanéd ir anovn Haykatén :  he builds also village one, and calls its name Haicton.

THE LORD’S PRAYER. Hayr mer or yerkins, sour  ettds Father our which (art) in heaven(s), holy be Tlarep nuwv 6 éy Troi¢ ovpavoic ayiacthrw  anouon cho:  Ekeszé archayovthivn cho: Etizin  name thy. Come kingdom ___—ithy. Be (done)  ro Ovopa ov. "EASérw =) Bacrela cov. TNevnShrw  kamch cho, orpés yerkins, ev yerkri: (Z) haz  will(s) | thy, as in heaven(s), also in earth. Bread  rd SéAnud cov, we tv odpavy, kal il Tie yng. Tov aproy 192 APPENDIX.

mer  hanapazord tovr mez aysbr: tw thot mez  our continual give us this-day. And forgive as  yuwv tov triobaiov dog nuiv ohuepov. Kal agec nyuiv (2) partis mer,  orpés ev mech thotovmch  debts our, as _— also we forgive Ta OpetAjpata Huwv, we «kal tyeic agleuev  meroz# partapanaz: Eo mi tanir (z)mez 4 (to) our debtors. And not lead us _ into ToIc OpetAgraic = Hywv. Kat pn) eloevéyenc quag ele  phorzovthivn, ayl phrked  (z)mez 4  éaré(n):  temptation, but deliver us from _ evil.  weipagov, aAAad pvoat huac ard Tov movnpov. Zi —s cho €  archayovthivn, ev  xzérovthiun, ev For thine is  thekingdom, and_ the power, and "Orr cov torw 9 BaciAcia, Kal n Sbvamic, Kat  pharch, yaviteans: Amén:  the glory, for ages. Amen.  n Odéa, sic rovc aiwvag. “Aphy. CONJUGATIONS. ‘To be,’ Arm. Gr. Alb, Osset.  em sil yam din  e8 el ye da ; é gori —tié SLE UY, U, 18—Ye, yey, yes  emch topey yemt stdm—an  ech. éoré yin stuth—aythe  en  elal yane stiiy—anze} Sansk, . Rhat-Rom. lial, Lat.  asm sunt 80no sum  ast e178 $er es ~® There are two dialects in Ossetic, which produce these different forms.

Sansk.

* abhavam  abhavas  abhavat  abhavdéma  abhavata  abhavae Arm.  azem  wes  azé  azemch  azéch  azen APPENDIX.

Rhat-Rom: Ltal. Dat.  eu é est  essen siamo SumuUs €88€8 siete estis  ean $0n0 sunt Gr. Alb. Sansk.  hy ¥esé dsam  nC yesé dsts  hy 46 adstt  huey yesém dsma  ire yescté dsta  yoay 15né dsan Rhat-Rom. Ital. Lat.  fova era era eram  fovas eras ert eras  fova era era erat Jovan eravamo eramus  fovas eravate eratis Sovan erano erant Gr. Alb. Osset.

@ yom ayn—ayne 4¢ yes ays—ayse  a yet ayd—ayde WEY yemt aykkam—ayyane TE yint aykkath—ayyaythe  wae yené aykhoy—ayyonze  khet-Rom. Ital. Lat.  seig 82a 61m 8cias $it $18  seg 1a sit  se1an same. simus $E108 state sitis  selan 81€n0 sint 193 194 APPENDIX. ‘To say,’ Arm. Gr. Alb. Osset.  asem onl thom* zaghiin  ases dhc thoua zag hiis  asé onal thote® caghiiy  asemch gaply thom zaghim  aséch garé thor zaghuth  asen gaal thoné zaghiing Sansk. Rhat-Rom. Ital. Lat.  vaéms gg dico dico  vakit gis diet diets  vaktr gt dice dicit  vaémas sche diciamo dicwmus  vakatha schetts dite dicitis  bruvanti gin dicono dicunt ‘To go.’ Arm. Gr. Alb. Osset.  ertham Epxopat vete zaun  erthas E0XD vete zaus  erthay Epxerat vete Zauy  erthamch Epyduesa vemt zZauom  erthaych EPXEGIE vent zauth  erthan Epxovrat vené Zauné Sansk. - Rhat-Rom. Ital. Lat.  yam vom vado vaio  yas vas vat vadis  yar va va | vat  ydmas mein andiamo’ vadimus  yatha merts andate vaditis  yante van vanno vadunt 1 Allied to the defective Sansk..ah, ‘to say.’ 2 Also thomt, them, or themi. 4 = Lat. meamus. 5 Cf. Arm. énth-anal, ‘to go, to run,’ Germ. wand-eln. * Cf. Pers. Behist. thatiya, ‘ dicit.’ APPENDIX. “To believe.’ 195 Arm. Rhat-Rom, Ital. Lat.  karzem creig eredo credo  karzes cress cred. credis  karzé eret crede credit  karzemch cartein crediamo credimus  karzéch carteits credete credits  karzen crein credono credunt Trish Gr. Alb. Lapp.  creidim TIOTEVW bessorg jakkab  creidt WLITEDELC bessots jakkah  ereidid mlorever bessoyé jakka  creidam WLOTEVOMED bessoumé —s_ gakkebe  creidiad TLOTEVETE bessont jakkebet  crevdet miarevouat bessomé jakkeh DACIAN NAMES OF PLANTS. Towards the conclusion of his work (p. 807), Grimm gives  some various readings of these names from two MSS., B(yzan-  tine and N(eapolitan). They are— 6. ScovrvofZ. B. and N., an«ovrvoté.

7. MéZovAa. B. and N., ulZndAa.

9. Zoudarn. B., Covobarn: N., Covobarnp. 10. “Opma. B. and N., Sppea. 11. TovoAjra. B. and N., (deutlhches) yovoXjra. 12. ’Amagacéé. B. and N., amapccéké. 18. AoxeAa. N., yodsAa (was wol unrichtig). 14. Adxva. B., Saxtva: N., Sdxeva. 16. Korfara. B., xorhara. 17. Mavreta. B. and N., pavria. 18. HpowedovAa. B. and N., rpomediAd. 20. Acédna. N., duéAAEva. 196 APPENDIX. Q1. KuxwAlda. B., coaodsAda: N., xonxods . .

23. Koadapa. Scheint in N., xoadapa.

25. BovddAAa. B. and N., BovdasAa.

26. KapoziS\a. B. and N., xapwrlSra.

27. DiSopIeSer4. B., pIopSacSera: N., PSog@SeSeda. 28. [Ipodfopva. B. and N., wpociapva.

81. Tovrdorpa. B. and N., rpourpaarpa.

82. TIpiadfAa. B. and N., wpradida. Grimm also adds here three other names— 1. KaAdaplvSn, Adxoe revdsAa B., revderAa N. 2. ’Apdpaxov, Adxa Sovwdnda N. 8. Bpvwvla AEvKh, Adxor x.voufsorda N. These additions and various readings seem to bring out one  point with considerable clearness. When we find among names  of medicinal herbs, 8éA-era or d¢£A-Ascva, xorxo-d1d a, wpo7e- Sida, mpra-ShAa or wpea-dicAG, rev-dcAa or rev- Seda,  and Sovw-dA 4, we may infer with much probability that  there was a Dacian word very like the Arm. de? in form, and  which might have had the same sense, 7.¢., ‘herb, medicine,  poison.’ This word may. appear simply in dcéA-ea; or, if we  read SéAXsva, we may interpret it ‘poisonous,’ forming  an Arm. word detean, from det, as tapean, ‘hot,’ is formed  from tap, ‘heat.’ There is also the Arm. suffix etén, ‘made  of,’ and the verb din-el, ‘to be.’ The other Dacian names just  enumerated would be similar to the Arm. mkndet, ‘arsenic ;’  gakndet, ‘red beet;’ and shasnde?, ‘rhubarb.’ Grimm sug-  gests hukukskraut, ‘cuckoo-plant,’ as the interpretation of  xuxwAida, or, according to the new reading, xomoduAd. The Arm. for kukukskraut would be khkov-det: we have in Arm.,  kachavakhot, ‘marjoram,’ 2. ¢., ‘ partridge-plant,’ hachav-khot. The word shot, ‘ herb,’ the common equivalent in Arm. for the Germ. kraut, or the Eng. wort, I have already traced in the Dacian xor-fara: the v.1r., xor-fara, does not affect this ana-  logy ; and the v. r., avrap-oe&é, for anac-oc&é, leaves the proba- APPENDIX. 197  bility of ocEé being = Arm. séz, ‘grass,’ as it was before. If, how-  ever, coaAdua be read for cxoadaya, there would be no analogy  to the Arm. kotem, ‘ cress.’ But xoaXdya is only found in one MS., and there doubtfully (‘scheint’). Tpourpdorpa for  rouraorpa, again, would destroy the Arm. analogies presented  by ddovm, thovth, and thove— gourd,’ ‘ blackberry,’ ‘ fig’—  which would readily explain rovr-éorpa, ‘ colocynth’ or ‘ wild  gourd,’ where -aorpa would have the force of -as¢rwm and aster  in menthastrum and oleaster. Tpovrpacrpa does not seem  capable of explanation from any language.

In addition to the various readings, there are three new  names— 1. KaAaplvSn, Adxoe revdiAad. Supposing &Aa= Arm.  det, there remains rev-, of which the explanation is not  obvious. The Arm. for calamint is katovakhot, 1.¢., ‘cat-  herb,’ our catmint or cat’s herb: for mint in general the Arm.  is ananovkh. Grimm compares the Goth. thiuth, ‘ good,’ in  conjunction with the Goth. termination, -7/6. There appears  no affinity between revdcAa and revrAoy or cevrAov.

2.’Apdpaxoy, Aadxor Sovwdnrd. AndAd= Arm. det. ‘ Ori-  gan’ isin Arm, zovirak: 8ov-w and zov-irak may be derived  from a common root. Grimm notices the O. H. Germ. for ‘origan,’ dosto or tosto, as a word which would be like the Dacian, if SouwdnAd were changed into SovocdnAd.

8. Bovwvla Aeveh, Adxot ktvouordad. No name in Arm. A German name for ‘briony’ is hund’s riibe, ‘dog’s turnip.’ The first element of «vov-Bo.r\a might thus=Gr. xuvde (cf.  cuvopau(3n), = Gael. coin, = Lat. canis, = Sansk. gunas = Arm. San. For -Boird, see ante, in Dac., p. 78, No. 8, and ef. Gael. doi/, ‘ umbo,’ bolg, ‘bulga.’ ‘ Dog-radish’ would be in Arm., Sanabotk or snbotk. Cf. sanazovkn or snzovkn (dog-fish), ‘ashark ;’ Sukhatot (dog-grape), ‘the plant night-shade, sola-  num, But the Gael. coin-bhile (dog-tree), ‘ the dogberry-tree,  cornus,’ is the nearest name to xvovfoAd. Several names of  plants begin with ‘ dog-.’ INDEX OF ETRUSCAN WORDS. (The words gwen as Hiruscan by Greek or Latin authors are Achrun, 120.

Achs, 112, 114, 1165. Hear, 97.

Mteus, 97.

A galletor ? 97, Aivi), 104. Alpan, 122, 123, 128, 130. Am, 189, 145.

Andas, 97.

Ante, 97.

Antar, 97.

A(p)avenke, 114, 115. Aracus, 39, 97.

Arimus, 98.

Arse-verse, 98.

Atena, 152, 159. Atesum, 98.

Avil, 104, 111, 116, 117. Avils, 111, 114, 115, 116. Balteus, 98.

Burrus, 98.

Capra, 99.

Capys, 99.

Cassis, 99.

Celer, 99.

Chiseliks, 126, 127, 130. Dammnus, 99.

Druna, 99.

Eka, 109, 120, 152.

in Italics). Epana, 141, 145.

Epl, 152, 157.

Erai, or era, 141, 146. Erske, 120, 131.

Est, 152, 153.

Etera, 107.

Ethe, 141, 145. Falandum, 99. Februwm, 99.

Fleres, 118, 119, 122, 126, 131. Flerthrke, 119. Fuius, 107.

Gapus, 99.

Gants, 100.

Helephu, 143, 145. Hintha, 152, 157. Hinthiu, 110 (note). Hister, 100.

Hut, 152, 156, 157. Huth, 155, 156.

Ipa, 152, 154.

Itus, 100.

Kana, 118, 131. Kanthke, 115, 116. Kape, 152, 157. Karutezan, 152, 153. Keal, 116, 117. Kecha, 122, 128, 131. Kechase, 159, 160. — Nw et oe Oo TT ee egg om, Mg RR OT Em ee INDEX OF ETRUSCAN WORDS. Kechazi, 152, 159, 160.

Kehen, 110 (note).

Ken, 126, 127.

Kenaphe, 117, 118.

Kerinu, 128.

Kethu, 137, 145.

Kiklena, 113, 115.

Kina, 155.

Kis, 116, 117.

Kisum, 113, 115.

Kizi, 115, 116.

Klalum, 113, 115.

Klan, 106, 108.

Klen, 122, 128, 131, 152, 157.

Klensi, 126, 127, 131, 152, 157.

Knl, 152, 158.

Kuna, 152, 155.

Kver, 121.

Lena, 100.

Lanista, 100.

Leine, or line, 104, 106, 108, 110.

Lenache, 128, 129, 131.

Leskan, 155, 156.

Leskul, 152, 1565.

Letm, 155, 156.

Lisiai, 139, 145.

Lthas, 122, 123, 128, 131.

Lupu, 111, 115, 117.

Lupum, 114, 115.

Lusni, 123, 124.

Ma, 137, 145.

Mar, 138, 145.

Maau, 152, 157.

Mathu, 138, 145.

Mealchisk, 114, 115.

Meani, 115.

Mi, 107, 122, 136, 145, 183.

Muniklet, 152, 157.

Munikleth, 115, 116.

Nak, 120. Nanus, 100. Naper, 152, 164. Nastav, 142, 145. Nepos, 100. Nesl, 109. Nethu, 142, 145. Ni, 137, 145. Nupthzi, or nup’hzi, 115. Penezs, 152, 157. Puia, 107, 108. Puiak, 107, 108. Puiam, 113, 115. Puil, 107. Rasne, 152, 154. Rasnes, 152, 154. Ril, 104, 106. Sains, 110 (note). Sak, 107. Salthn, 123, 125. Sanal, 121, 126, 127, 131. Sech, 107, 112, 1165. Sek, 107. Sie, or isie, 141, 146. Spel, 152, 158. Spelane, 152, 158. Stalke, 117. Subulo, 100. Suthi, 108, 183. Suthik, 128. Suthina, 108 (note). Sver, 121. 199 Teke, 126, 127, 132, = theke. Tenine, 126, 127, 128, 132. Tesne, 152, 154. Tesnsteis, 152, 154. Thapna, 123, 124. Theke, 109. Thipurenai, 140, 145. Thues, 110 (note). Thuf, 122, 128, 132. Thui, 110, 155. 200 INDEX OF ETRUSCAN WORDS. Thup, 122, 132. Tinskvil, 123, 124. Tlen-acheis, 128, 132. Tlen-asies, 126, 132. Tular, 32, 111. Tularu, 152, 155, 157. Turke, 118, 119, 122, 132. Turuke, 118.

Tuthines, 126, 137, 128, 133. Via, 107.

Vorsus, 100.

Zek, 121, 133.

Zilachnke, 115, 116.

Zuki, 152, 156. THE END.

nc a te - cg ASRS . STANDARD EDITIONS PRINTED FOR . PARKER, SON, AND BOURN, 445, WEST STRAND, LONDON. History of Normandy and of Eng-  land. By Sir Francis Pareravs, Deputy Keeper of the Records. Octavo. Vols. I.and ITI, 21s. each.

History of England from the Fall of Wolsey to the Death of Elizabeth. By James ANTHONY Froupg. The Second Edition. Octavo. VolumesItoIV. 654s These Volumes complete the reign of Henry the Eighth. Vols. V. and VI. con-  taining the Reigns of Edward the Sixth and Mary. 28s.

The Pilgrim: a Dialogue on the Life  and Actions of King Henry the Eighth. By Wiitram THomas, Clerk of the Council  to Edward VI. Edited, with Notes, from  the Archives at Paris and Brussels, by J.A.Froupr. Octavo. 6s. 6d.

History of England during the Reign of George the Third. By WittIamM Massgy, M.P. Octavo. Vols. I., II., and IIT. 12s. each.

History of Trial by Jury. By WiLL1aM Forsytu, M.A. Octavo. 8s.

History of the Whig Administration  of 1880. By Joun Artnur Roxxzvucx, M.P. Octavo. Two Vols. 28s. The Spanish Conquest in America,  and its Relation to the History of Slavery  and to the Government of Colonies. By ArtHuR Hztps. Complete in Four Volumes. Octavo. Vols. I., II., 288.; Vol. IIT., 16s.; Vol IV., 16s. History of Civilization in England. By Hzney Tuomas Bucxrz. The First Volume, Octavo. Second Edition. 21s, The Second Volume, containing the His-  tory of Civilization in Spain and Scotland, Revolutions in English History. By Ropgrr Vauenuan, D.D. The First Volume, Revolutions of Race. Octavo. 15s.

The Second Volume, Revolutions én Religion, Octavo. 15s, Studies and Illustrations of the ‘Great Kebellion.” By J. Lanarton Sanwyorp. Octavo, 16s. The Holy City; Historical, Topo-  graphica , and Antiquarian Notices of  erusalem. y G. Writs, B.D. Second Edition, with Dlustrations and Additions, and a Plan of Jerusalem. Two Vols. £2 65s. History of the Holy Sepulchre. By ProFEssokR WILuI8s. Reprinted from WrttaM’s Holy City. With Illustrations, 8. Plan of Jerusalem, from the Ord-  nance Survey. With a Memoir. 93s.;  mounted on rollers, 188. The Roman Empire of the West: Four Lectures, by Richarp ConGREVB, M.A., late Fellow and Tutor of W College, Oxford. Post Octavo. 4s.

The Earliest Inhabitants of Ttaly. From Mommsen’s Roman History. By G. RoseErrson. Octavo. 2s. Claudius Ptolemy and the Nile. By WituraM DesporouaH Cooter. Octavo. Witha Map. 4s The Earth and Man; or, Phy-  sical Geogra hy in its Relation to the History o Mankind. From the Work of Guyot, with Notes and Copious Index, Cheap Edition, 2s.

Hellas: the Home, the History, the Literature, and the Arts of the Ancient Greeks. From the German of Jacoss, Foolscap Octavo. 4s. 6d. A History of the Literature of Greece. By Professor Mutter and Dr. DownaLps2N, from the Manuscripte of the  late K.O. Mutuger. The first half of the Translation by the Right Hon. Sir Gzorar CurRNngEWALL Lxrwis, Bart., M.P. The  remainder of the Translation, and the  completion of the Work according to  the Author’s plan, by JoHw WiLLi1aM Dowatpsonr, D.D. Octavo. Three Vols. 36s. The new portion separately, Two By Jouw Witi1uam Doxaxtpson, D.D. Varronianus; a Critical and Histo-  vical Introduction to the Ethnography of Ancient Italy, and the Philological Study  of the Latin Language. Third Edition. 16s, The New Cratylus; Contributions  towards a more accurate Knowledge of the Greek Language. Third Edition. Revised  throughout and considerably enlarged.

Homeric Ballads: the Greek Text,  with an English Translation in Verse, and Introduction and Notes. By Dr. Maainn. Small Octavo. 66. Modern Painting at Naples. By Lorp Narizrz. Foolscap Octavo. 4s. 6d, Principles of Imitative Art. By Groraxs Burizs, M.A. Post Octavo. 68.

From the German of BEcKER. Charicles: a Tale Illustrative of Private Life among the Ancient Greeks. N ow. fidition, collated and enlarged. 10s. 6d. Gallus; Roman Scenes of the Time  of Augustus. Second Edition, enlarged. With additional Illustrations. 12s.

By Witte Sriziine, M.P. Cloister Life of the Emperor Charles the Fifth. Third Edition. 8s.

A Long Vacation in Continental Picture Galleries. By T. W. JEx Buaxs, M.A. Foolscap Octavo. 3s. 6d. The Young Officer’s Companion. By Major-General Lord Dz Ros. Second Edition. 6a. Twelve Years of a Soldier’s Life in India. Extracts from Letters of Major Hopsoy,Commandant of Hodson’s Horse ; Edited by his Brother, the Rev. Gzorcr H. Hopsox, M.A. Third Edition, with Additions. 10s. 6d.

By Hazzis Prewpereast, Burrister-  at-Law. The Law relating to Officers in the Army. Revised Edition. 6s, 6d. The Law relating to Officers of the Navy. In Two Parts. 10s. 6d.

STANDARD EDITIONS PRINTED FOR By the Right Hon. Sir G. Cozrwzwaxh EWis, Bart., M.P. On Foreign Jurisdiction and _ the Extradition of Criminals. Octavo. 2s, 6d. An Enquiry into the Credibility of  the Early Roman History. Octavo. )  ols. . On the Use and Abuse of Certain Political Terms. Octavo. 9s. 6d. On the Methods of Observation and Reasoning in Politics. Octavo. Two Vole. 288. On the Influence of Authority in Matters of Opinion. Octavo. 10s. 6d.

George Canning and his Times. By Aveustus GraNvILLE STAPLETON, Octavo. 16s. Oxford Essays. By Members of the University, Four Volumes, 7s. 6d. each.

rd Cambridge Essays. By Members of  the University. Four Volumes, 7s, 6d.  eacn.

By the Author of ‘ Friends in Council.’ Friends in Council. A New Series. Two Volumes. Post Octavo. 14s. Friends in Council. First Series. New Edition. Two Volumes. Qs. Companions of my Solitude. Fifth Edition. 3s, 6d. ‘Essays written in the Intervals of Business. Seventh Edition. 2s. 6d. On Taxation: how it is raised and  how it is expended. By Lzowx Levi, Pro-  fessor of Commercial Law in King’s Col-  lege, London. Post Octavo. 7s. 6d, Man and his Dwelling Place. An Essay towards the Interpretation of Nature. I, Of Science. II. Of Philosophy. ITI. Of Religion. IV. Of Ethics. V. Dialogues. Small Octavo. 9, PARKER, SON, AND BOURN, WEST STRAND. By Jouw Sruart Miu. Considerations on Representative Government, Octavo. Dissertations and Discussions, Poli-  tical, Philosophical, and Historical. 24s. Thoughts on Parliamentary Reform. Second Edition, with Supplement. 1s. 6d. On Liberty. Second Edition. 7s. 6d. Principles of Political Economy. Fourth Edition. Two Volumes. 30s. A System of Logic, Ratiocinative  and Inductive. Fourth Edition. Two Volumes. 265s. By Avex. Barn, M.A., Examiner in Logic  and Moral Philosophy in the University of London. The Senses and the Intellect. Octavo. 165s. The Emotions and the Will: com-  pleting a Systematic Exposition of the  uman Mind, Octavo. 165s.

Dialogues on Divine Providence. By a Fellow of a College. Foolscap Octavo. 3s. 6d. God’s Acre; or, Historical Notices  relating to Churchyards. By Mrs. Srowsz. Post Octavo. 10s. 6d. Transactions of the National Asso-  ciation for the Promotion of Social Science, 1857. Octavo, 168. 1869, 16s. 1860. The Institutes of Justinian; with English Introduction, Translation, and Notes. By THomas C. Sawpanrs, M.A.,  late Fellow of Oriel College, Oxford. Octavo. 165s. Principles and Maxims of Jurispru-  dence. By J. G. PuirimoreE, Q.C., Header to the Four Inns of Court, Octavo, 8. Statutes relating tothe Ecclesiastical Institutions of EKugland, India, and the Colonies; with the Decisions thereon, B ARCHIBALD J. Stzruuns, M.A., F.B.S. Two Volumes, Royal Octavo. £3 3s. Charges on the Administration of  the Criminal Law, the ression of Crime, and the Reformation of Offenders. By Marruew Davenport Hix, Q.C., Recorder of Birmingham. Octavo. 16s, Remains of Bishop Copleston. With Reminiscences of his Life. By the Arch-  bishop of Dublin. With Portrait. Memoir of Bishop Copleston. By W. J. Cortestox, M.A. Octavo. 10s. 6d. Essays and Remains of the Rev. Ropert ALFRED VauGHaN. With a Memoir by R. Vauauay, D.D. Two Vols., with Portrait. 14s. English Life, Social and Domestic,  in the Nineteenth Century. Third Edition, Revised. 4s. 6d. Evelyn’s Life of Mrs. Godolphin ; Edited by the Bishop of Oxford. Third Edition, with Portrait, 66. The Merchant and the Friar; Truths and Fictions of the Middle Ages. An Historical Tale. By Sir Francis Paueravs. Second Edition. 3s. The Recreations of a Country Par-  son. Being a Selection from the Contri-  butions of A. K. H. B. to Fraser’s Maga-  sine, First Series. Second Edition. Crown Octavo. 9s. Second Series. Crown Octavo, By Wririam Grorar Crarx, M.A, Public Orator, Cambridge. Peloponnesus: Notes of Study and Travel. Octavo. With Maps. 10s. 6d, Gazpacho; or, Summer Months in Spain. New and Cheaper Edition, 6s. The Mediterranean : a Memoir, Phy-  sical, Historical, and Nautical. By Ad-  miral iral W. H. Surrn, F.B.8.,&0, Octavo.

6s. Tour in the Crimea, and other Countries adjacent to the Black Sea. By Lord Dz Ros. Crown Octavo, 4s. 6d. 4 STANDARD EDITIONS PRINTED FOR _A Manual of Geographical Science, Mathematical, Physical, Historical, and Descriptive. In Two Parts. Part I. comprises MATHEMATICAL GEOGRAPHY. By the late Professor M. O’ Brizy. PHYSICAL GEOGRAPHY. By D.T. Ansrep, M.A., F.R.8. CHARTOGRAPHY. By J. RB. Jacx-  son, F.R.8. THEORY OF DESCRIPTION AND GEOGRAPHICAL TERMINOLOGY. By the Rev. C. G. Nicoray. Part IT. contains ANCIENT GEOGRAPHY. By the Rev. W. L. Bevan. MARITIME DISCOVERY AND MODERN GEOGRAPHY. By the Rev. C. @. Nicouay. And a copious Index to the whole Work. Two closely-printed Volumes, Octavo, with  many Woodcuts, 25s. 6d. The Parts sepa-  rately. Part I., 10s. 6d.; Part IT., 15s. An Atlas of Physical and Historical Geography. Engraved by J. W. Lowry,  under the direction of Professor ANstED  and the Rev, C. G. Niconay. CONTENTS :— 1. Reference Map.—The World on Mer-  cator’s Projection. . 2. Meteorological Map of the World. 8. Relief Map of the World, showing the Elevations of the Earth’s Surface, 4. Phytographical Map, showing the Distribution of Plants in the World. Vertical Distribution of Plants and Animals, 5. Zoological Map, showing the Distri-  bution of Animals in the World. Ethnographical Map, showing the Dis-  tribution of the Races of Men. 6. Chart of Ancient and Modern Geo-  graphy and Geographical Discoveries. Imperial Folio, in a Wrapper, 5s. This Atlas was constructed with an  especial view to the above Manual, but  will be found a valuable companion to Works on Geography in General. The Military To ography of Conti-  nental Europe. From the French of M., Th. Lavallée. By Col. J. R. Jackson, F.R.S., &c. 8s. The Kingdom and People of Siam; With a Narrative of the Mission to that Country in 1855. By Sir Jonw Bowrrya, F.R.S., her Majesty's Plenipotentiary in China. Two Vols., with Illustrations and Map. 32s. A Year with the Turks. By Warinetor W. Sxxyra, M.A. With a Coloured Ethnographical Map by Lowry. Crown Octavo. 88. The Biographical History of Philo-  sophy, from its origin in Greece down to  the present day. By Gzorer Henny Lewes. Library Edition. Octavo. 168. Paley’s Evidences of Christianity. With Annotations by the ARCHBISHOP OF Dusirn. Octavo, 9s. Paley’s Moral Philosophy, with Annotations by Rrcnarp WHatgty, D.D., Archbishop of Dublin. Octavo. 7s. Bacon’s Essays, with Annotations  by Archbishop Waargiy. Fifth Edition. Octavo. 10s. 6d.

By Ricgarp CHENEVIX TRENCH, D.D., Dean of Westminster. A Select Glossary of English Words  used formerly in Senses different from  their present. Second Edition. 4. English, Past and Present. Fourth Edition. 40. Proverbs and their Lessons. Fifth Edition. 3s. On the Study of Words. Tenth Edition. 3s. 6d. - On Deficiencies in our English Dictionaries, Second Edition. Octavo. 3s.

State Papers and Correspondence,  illustrative of the State of Europe, from  the Revolution to the Accession of the House of Hanover; with Introduction, Notes, and Sketches. By Joun M. Kemaiz, M.A. Octavo. 16s.

On the Classification and Geogra-  phical Distribution of the Mammalia:  eing the Lecture on Sir Robert Reade’s Foundation, delivered before the Uni-  versity of Cambridge, 1859; with an Appendix on the Gorilla, and on the Extinction and Transmutation of Species. By Ricuarp Owen, F.R.8., Superin-  tendent of the Natural History Department  in the British Museum. Octavo. 6s. Leaves from the Note-Book of a Naturalist. By W. J. Bropzxir, F.R.8. Post Octavo. 10s. 6d. Familiar History of Birds. By Bishop StamLEy. Cheaper Edition. 3s. 6d.

PARKER, SON, AND BOURN, WEST STRAND. 5 By Witiram WueweE tt, D.D., F.RS., Master of Trinity Coll., Camb. History of the Inductive Sciences. Third Edition. Three Vols. 24s. History of Scientific Ideas: being  the First Part of a Newly Revised Edition  of the Philosophy of the Inductive Sciences. Small Octavo. Two Vols, 14s. Novum Organon Renovatum : being  the Second Part of a Newly Revise Edition of the Philosophy of the Inductive Sciences. Small Octavo. 7s. On the Philosophy of Discovery, Chapters Historical and Critical, being the  hird and concluding Part of the Revised Edition of the Philosophy of the Inductive Sciences. Qs. Indications of the Creator. Second Edition. 58. 6d. Elements of Morality ; including Polity. Two Vols. Third Edition 165s. Lectures on Systematic Morality. Octavo. 7s. 6d. Of a Liberal Education in General. Part I., 4s. 6d.; Part II, 3s. 6d.; Part III., 2s. On the Principles of English Uni-  versity Education. Octavo. 5s. Architectural Notes on German Churches. Third Edition. Octavo. 12s. By Many Rosegrs. Wild Animals; and the Regions  they Inhabit. Cheaper Edition. 2s. 6d. Domesticated Animals; with refer-  ence to Civilization, Cheaper Edition. 28. 6d.

By Emity SHIRRBFF. Intellectual Education, and its In-  fluence on the Character and Happiness of Women, Post Octavo. 10s. 6d. Why should we Learn? Short Lectures addressed to Schools, Foolscap Octavo. 2s.

A System of Surgery, Theoretical  and Practical, in Treatises by various Authors, arranged and edited by T. Howmes, M.A. Cantab, Assistant Surgeon  to the Hospital for Sick Children. Volume I,—General Pathology. Demy 8vo, £1 1s. Lectures on the Principles and Prac-  tice of Physic. By THomas Watson, M.D., Physician Extraordinary to the Queen. ourth Edition, revised. Two  olumes. Octavo, 34s.

By Hewny Gray, F.R.S., Lecturer on Anatomy at 8t. George’s Hospital. Anatomy, Descriptive and Surgical. With nearly 400 large Woodcuts, from  original Drawings, from Dissections made  by the Author and Dr, Carter. Royal Octavo, Second Edition. 28s. The Structure and Use of the Spleen. With 64 Illustrations. 15s.

Physiological Anatomy and Physio-  logy of Man. By Rosert Brnrury Topp, M.D., F.R.8., and WILiiaM’ Bowman, F.R.8S., of King’s College. With numerous Original ustrations, Two Volumes. £2. Manual of Human Microscopic Anatomy. By Aubert Kouiixer. With  numerous Illustrations. Octavo. 24s. On Spasm, Languor, and Palsy. B J. oe Winer, MD. Post Octave. 78. y By Grorer Jonnson, M.D., Physician  to King’s College Hospital. On the Diseases of the Kidney ; their Pathology, Diagnosis, and Treatment. Octavo. With Illustrations. 14s. On Epidemic Diarrhcea and Cholera;  their Pathology and Treatment. With a Record of Cases. Crown Octavo. 7s. 6d.

Lunacy and Lunatic Life: with Hints on Management. Small Octavo. 3s. 6d. On Medical Testimonyand Evidence  in Cases of Lunacy; with an Essay on the Conditions of Mental Soundness. By Tuomas Mayo, M.D., F.R.8., President  of the Royal College of Physicians. Fools-  cap Octavo. 3s. 6d. Diphtheria : its History and Treat-  ment, By E. Hzaptam GreENHow, M.D Fellow of the Royal Coilege of Physicians. Octavo. 78. 6d. A Dictionary of Materia Medica and Pharmacy. By WiLtiamM THoMas BRanpDeE, F.R.8. Ooctavo. 165s. Popular Physiology. ; . Tht Baio wey Yn re By Joun Tomes, F.R.S. Lectures on Dental Physiology and Sargery. Octavo. With 100 Ilustrations. 128. On the Use and Management of Artificial Teeth. With ustrations. 8s. 6d.

German Mineral Waters: and their  employment in certain Chronic Diseases. By Starismunpd Surao, M.D., Senior Physician of the German Hospital. Fools-  cap Octavo. 7s. 6d. The Influence of the Climates of Pau, and of various parts of Italy, &c.,  on Health and Disease. By A. Taytor, M.D. F-R.8.E. Second Edition, enlarged.

By Witulam ALLew Mriuzr, M.D., F.B.S., Professor of Chemistry, King’s College,  ndon, Elements of Chemistry, Theoretical]  and Practical. With numerous Illustra-  tions. Part I. Chemical Physics. Second Edition. 10s.6d. Part II. Inorganic Che-  mistry. Second Edition. 20s. Part III. Organic Chemistry. 20s. First Lines in Chemistry for Be-  inners. By Dr. ALBEHT J. Berways, .C.8., Lecturer on Chemistry at St, Mary’s Hospital. With Illustrations. 7s. The Chemistry of the Four Ancient Elements—Fire, Air, Earth, and Water :  an Essay founded upon Lectures delivered  before her Majesty the Queen. By THomas GairFirHs. Sccond Edition. 4s. 6d. Of the Plurality of Worlds. An Essay. Fifth Edition, 6s. Lectures on Astronomy, delivered  at King’s College, London. By Henry Mosgeizy, M.A., F.R.8., one of her Majesty’s Inspectors of Schools. Cheaper Edition, 3s. 6d. Recreations in Astronomy. By the Rev. Lawis TOMLINSON. Fourth Edition. 4s. 6d.

By J. Russzti Hip, Foreign Secretary . - of the Royal Astronomical Society  of London. The Comets: with an Account of Modern Discoveries, and a Table of all the Calculated Comets, from the Earliest Ages, Post Octavo. 5s. 6d. The Comet of 1556: on its antici-  ated Re-appearance, andon the Appre-  ension of Danger from Comets. Post Octavo. 2s. 6d. An Astronomical Vocabulary; an Explanation of all Terms in Use amongst Astronomers. Small Octavo. 1s.6d.

STANDARD EDITIONS PRINTED FOR Elements of Meteorology. By John Freperick DanrEgut, F.R.8., &. Two Volumes, With Charts and Plates. 32s. On the Nature of Thunder-storms ;  and on the Means of Protecting Buildings  and Shi ping inst the Effects of Light-  ning. By Sir W. Swow Haznzis, F.R.S. Octavo. 10s. €d. The British Paleozoic Rocks and Fossils. BY Professor SEDGWICK and Professor M‘Coy. Royal Quarto, with  numerous Plates. Two Vols. 42s. By Captain Lznpy, Director of the Practical Military College at Sunbury, late of  the French Staff. Elements of Fortification, Field  and Permanent. With 236 Woodcuts. Ts. 6d. The Principles of War; or, Ele-  mentary Treatise on the Higher Tactics  and Strategy, intended for the use of young Military Students. 6s. By Burigr Witiiams, C.E. Practical Geodesy; Chain Sur-  veying, Surveying Instruments, Levelling, Trigonometry, and Mining; Maritime, Estate, Parochial, and ilroad Sur-  veying. Third Edition revised. Octavo. 8s.8d. A Manual of Model-Drawing from Solid Forms; with a Popular View of Perspective; Shaded Engravings of the Models, and numerous Woodcuts. Octavo. 15s. This Manual is published under the Sanction of the Committee of Council on Education.

Readings in English Prose Litera-  ture; from the Works of the best English Writers; with Essays on English Lite-  rature. Fifth Edition, 3s. 6d. Readings in Poetry; from the Works of the best English Poets, with Specimens of the American Poets. Thir-  teenth Edition. 3s. 6d. Readings in Biography; a Selection  of the Lives of Eminent Men of all Nations. Fifth Edition. 3s. 6d. Readings in Science; Familiar Ex-  lanations of Appearances and Principles  in Natural Philosophy. Fourth Edition. 3a.

PARKER, SON, AND BOURN, WEST STRAND. Readings from Shakspeare ; for the Use of Young Persons. Edited the Author of Aids to Development, &c. Fools-  cap Octavo. 4s. 6d. Woman’s Rights and Duties, con-  sidered with reference to their Effects on Society and on her own Condition. By a  oman, Two Volumes, Post Octavo. 14s, Woman’s Mission. The Fourteenth Edition. 2s. 6g. Spiritual Songs for the Sundays and |  oly days throughout the Year. By Joun 8. B. Monsgxi1i, LL.D. Second Edition,  revised. 4s. 6d. By Coventry PaTMoRE. Faithful for Ever. Foolscap Octavo. 63. The Angel in the House. Part I. The Betrothal. Part II. The Espousals, Cheap Edition, in One Volume. 7s. 6d.

Songs for the Suffering. By Rev. THOMAS Davis, M.A. Foolscap Octavo. 4s. 6d. Cecil and Mary; or, Phases of Life  and Love. <A Missionary Poem. By JoszrH Epwarp JacxsoN. Foolscap Octavo. 4s. Orestes and the Avengers. An Hellenic Mystery. ‘Our eyes see all around in gloom or glow, Hues of their own, fresh borrowed of the  eart,’ Foolscap Octavo. 2s, 6d. Pinocchi, and other Poems. Octavo. 5s. Days and Hours, and other Poems. By Frarepurick Tennyson. Foolscap Oc-  tavo. 6s. Crown By the Rev. Coarizs KINGSLEY. Andromeda,andotherPoems. Second Edition. 5s. The Saint’s Tragedy : the True Story of Elizabeth of Hungary, Third Edition. 5s.

Oulita, the Serf; a Tragedy. By  the Author of Friends in Council. 68. King H the Second. An His-  total Drawe. 6s. en Nina Sforza. A Tragedy. By R. Zouem 8 teavenror. Phra Haition. 2s.

The Sea Spirit, and other Poems. By Lapy Lusuineronr. 4s. 6d. Arundines Cami, sive Musarum Can-  tabrigiensium Lusus Canori. Collegit  atque edidit Hxenaicus Drusy, M.A.

Fifth and cheaper Edition. 7s. 6d.

By Ricuarp CHENEVIX TRENCH. Calderon’s Life’s a Dream: with an Essay on his Life and Genius, 4s. 6d. Justin Martyr, and other Poems. Fourth Edition. 6s. Poems from EHastern Sources: Geno-  veva and other Poems. Second Edition. 5s. 6d. Blegiac Poems. Third Edition. 2s. 6d. By Evaas Atrexp Bownina. The Book of Psalms literally rendered  into English Verse. Small Octavo. 5s. The Complete Poems of Schiller, Attemp d in English Verse. Foolscap Octavo. The Poems of Goethe. Translated  in the original Metres. 7s. 6d. Translated by THropork Marrin. The Odes of Horace, with a Life  and Notes. Second Edition. 9s. Catullus. Translated into English Verse, with Life and Notes. Aladdin. A Dramatic Poem. B ApaM OEHLENSCHLAEGER. 58. , Correggio. A Tragedy. By Oxu- LENSCHLAEQGER. 33. King Rene’s Daughter: a Danish Lyrical Drama. By Hengiox Herrz. 2s. 6d. By the Author of The Heir of Redelyffe. Hopes and Fears; or, Scenes from  the Life ofa Spinater. Two Vols., Fools-  cap 8vo, 12s, The Heir of Redclyffe. Eleventh Edition. 68. Heartsease, or the Brother’s Wife. Sixth Edition. 6s, The Lances of Lynwood. Fourth Edition. 3s. The Little Duke. Cheap Edition. 1s, 6d. The Daisy Chain. Cheap Edition. Second Edition. One Volume. 6s. Dynevor Terrace. 68,