^{page 19 note 1} Klaproth, who had little favour for theories that did not originate with himself, rejects this idea as altogether improbable; he observes that it is impossible to trace any resemblance between the forms of the Semitic letters and the shapes of the objects of which they bear the names, and he conjectures accordingly, that the nomenclature was given as a mere sort of “memoria technica.” See Apercu de l'Origine des diverses Ecritures, p. 77.

^{page 20 note 1} I would employ the term Semitic, instead of Babylonian, but that that term itself is in an Ethnographical sense open to exception, and that Cuneiform writing moreover is opposed in all essentials of organization, direction, and extent, to what we know of the Semitic alphabets. I am inclined to think, also, that Scythic would be a more appropriate appellation than Median for the second class of Cuneiform writing.

^{page 20 note 2} I shall frequently have occasion to quote the opinions of M. Botta, and I shall always do so with respect. As I am, however, indebted for my acquaintance with those opinions entirely to private correspondence, I must apologize for their communication, should their present announcement anticipate M. Botta's own publications.

^{page 21 note 1} Dr. Pritchard, in his admirable researches into the Physical History of Mankind, has assigned the name of Syro-Arabian to the nations which are usually termed Semitic, observing (vol. III. p. 7,) that many of these nations are declared in the Patriarchal genealogies to have descended from Ham, and that it is evidently improper to apply to a whole groupe of nations an epithet, which derived from the Patriarch of one division, excludes all the rest. In real fact, if we must have a Patriarchal nomenclature, Hammite would be a more appropriate title than Semitic, for of the four sons of Ham, three at any rate, Cush, Mizraim, and Canaan founded nations of the Syro-Arabian groupe, while of the descendants of Shem, the Arabian Joctanides, the Aramæans, and the Assyrians were alone undoubtedly of the same Ethnographical family. In the Toldoth Beni-Noah, the majority of the Shemite nations will be found to be of the Arian family.

^{page 22 note 1} Babylon is too well known to require illustration. Erech was corrupted by the Greeks into Its true Chaldæan name was Warká, under which title it is described by the early Arabic geographers as the birth-place of Abraham, with an evident allusion to the Ur of the Chaldees. The ruins which still retain the name of Warká, are to be seen to the west of the Hye, near its point of confluence with the Euphrates, but they are now rarely accessible owing to the inundation of the surrounding country. Accad is in all probability a mistaken reading for Accar, the Hebrew d and r being nearly similar. The latter term, equivalent to the Greek was a generic title for a lofty embattled palace, and in this sense still applies to numerous ruins in Babylonia. The Accad or Accar of Genesis, I consider to be Akarkúf near Baghdad, which is called in ancient Oriental authors sometimes the hill of Nimrod, and sometimes the palace of Nebuchadnezzar. It it more difficult to identify Calneh; general opinion is in favour of Ctesiphon, but from the evidence of the bricks, I prefer the Chaldæan ruins of Kalwádha near Baghdad.

^{page 22 note 2} When the Sanhedrim and the Beresith Rabba were composed, (and they are among the earliest of the Talmudic writings, dating probably from the second or third century, A.D.) the Babylonian Jews were so convinced of the great ruin of Borsippa being the representative of the Tower of Babel, that they sought for derivations connected with the confusion of tongues to explain the name. If, therefore, the Borsippa of Berosus, Strabo, and Stephen (Barsita in Ptolemy, and Byrsia in Justin) can be shown to be identical with the present Birs-i-Nimrúd, the latter site will be determinately connected with the Tower of Babel. By the early Arabs (Beladheri, &c.,) the Birs is usually named the Sirh-un-Nimrúd, or Nimrod's palace, and in the Sidr of the Sabæans, Babel and Bursif are connected together. I have never found any reason for identifying the tower of Babel with the temple of Belus; the one was at Borsippa, the other at Babylon. See Bochart's Phaleg, col. 36, and Buxtorf's Talmudic Lexicon, in voce , Borsiph. Yákút's Mo'ejam el Baldán, in voce Burs, Ajummah, and Adami, Norberg's Liber, vol. I. p. 153.Google Scholar

^{page 22 note 3} See Genesis, ch. xi. v. 1.

^{page 22 note 4} Signior Mussabini's forthcoming work on Cuneatic writing will probably exhibit many specimens of this class of writing preserved in the Museums of Europe. I have at present before my eyes a very perfect relic of this class, which was lately disinterred from among the ruins of Cutha, the city of the Cutheans, who colonized Samaria. It consists of forty lines of writing, engraved on the two faces of a black, barrel-shaped stone; on one side the legend is as clear as if only recently inscribed; on the other it is a good deal mutilated.

^{page 23 note 1} The bricks at Susa are also stamped with inscriptions in the primitive Babylonian character.

^{page 23 note 2} See Journal of the Royal Geographical Society of London, vol. IX., p. 37.Google Scholar

^{page 23 note 2} Resen, or more properly Dasen, as it was written by the Septuagint, I place at Yássín Tappeh in the plain of Shahrizor, the original seat of the Dasíní Kurds; and Calah (the Halah of the captivity, in Isidore, and Halus in Tacitus,) I suppose to be identical with the Holwán of Syriac and Arabic History; the ruins of which are to be seen at Sir Pul-i-Zoliáb.

^{page 23 note 4} I have been favoured by Sir Gardiner Wilkinson with a copy of this curious relic. It merely contains the legend “Artaxerxes the Great King,” but the orthography is so barbarous that I have no hesitation in assigning it to the third king of that name in the Achæmenian line.

^{page 24 note 1} See Heeren's, Researches, vol. II. p. 322.Google Scholar This relic is usually called the slab of Michaux, it was found among the ruins of Ctesiphon; for particulars regarding it, see Ouseley's, Travels, vol. I. p. 422.Google Scholar

^{page 24 note 2} Rich observes, p. 187, that in excavating at the Kasr he found specimens of this writing on small pieces of baked clay of a darker and finer quality than the bricks, and the impression of one relic of the class he has published in plate 9,5, where however the writing appears to me to be of the Assyrian rather than the Babylonian type. Perhaps the specimens B and C, published by Grotefend in the sixth volume of the de l'Orient, Mines, p. 143, may be considered of the same class.Google Scholar

^{page 24 note 3} I presume the public to be too well acquainted with the recent excavations in the neighbourhood of Nineveh to need any detailed notice of Khorsabád. The ruins probably represent the palace of Evorita, to which Saracus, the last king of the lower Assyrian dynasty, retired on the approach of the confederate Medes and Babylonians; and the name of Sar'un, which attaches to the site in early Arab geography, appears to be identical with the Sarbena of Ptolemy and Sarbanna of the Peutingerian Table. See Eusebius, , I. 9. p. 25. Yákút's Mo'ejam el Baldán, in voce Sar'ún, Ptol., lib. 6. c. 1. and Peut. Tab. Seg. 10.Google Scholar

^{page 25 note 1} These Inscriptions, amounting in number to forty-two, were published In the Journal Asiatique, Ser, Tom. IX., No. 52, after copies taken by (be lamented Schultz.

^{page 25 note 2} For notices of these inscriptions, see Journal of the Royal Geographical Society, vol. X., Part 1., pages 12 and 21. I hare a correct copy of the mutilated writing at Dásh Tappeh, and as far as I am able to connect the fragments together, I judge the entire legend to hare been a mere transcript of one of the tablets at Van. The Kel-i-Shín pillar being enveloped with a thick covering of ice and snow at the period of my visit to Ushneï, I could only copy a few isolated letters on the eastern face of the slab, which are however certainly of the Assyrian type. I have since learnt that the western face of the pillar contains an inscription of equal length with that on the eastern face, and that in the fair season the whole of the writing is tolerably legible.Google Scholar

^{page 25 note 3} I first heard of this inscription from Mr. Robertson, an English engineer, whe visited the mines at Arghanah, in the spring of 1839. The tablet was subsequently examined by some Prussian officers who were with the Turkish army at die battle of Nezib; but I believe the entire legend never yet to have been copied.

^{page 26 note 1} See Rich's, Kurdistan, vol. II. pages 31 and 43. The Nineveh slabs are in the British Museum with Mr. Rich's other Oriental antiquities.Google Scholar

^{page 26 note 2} This place, the Larissa of Xenophon, is supposed by Bochart to represent Resen. See Phaleg, col. 237. I have no reason for identifying it with Rehoboth, beyond its evident antiquity, and the attribntion of Resen and Calah to other sites. The Arab geographers placed Rehoboth at Rahbeh on the Euphrates, the Talmudists at Borsippa, both of which positions being far beyond the confines of Assyria are obviously inadmissible. See Yákút's Mo'ejam, in voce Rahbeh, and Phaleg, loc. cit.

^{page 26 note 3} I was favoured with a fac-simile of this inscription by the Rev. Mr. Badger last year. The writing, which extends to six lines, is precisely similar to that at Khorsabád.

^{page 26 note 4} Rich has given a fac-simile of the writing on the Nimrúd bricks; Kurdistan, , vol. II. p. 130. On the Assyrian bricks each letter appears to have been separately impressed, and the writing covers the whole face of the brick, while at Babylon, the legends varying from three lines to seven, are formed in a parallelogram with a margin hi the centre of the brick, as if a framed stamp containing the inscription had been employed for the impression. At Susa, again, although the character is Babylonian, and the writing is divided by lines, the stamp is of a much larger size, covering in some instances the entire face of the brick.Google Scholar

^{page 26 note 5} The four primæval capitals of Assyria were, Nineveh, Rehoboth, Calah, and Resen, see Genesis, x. 11 and 12. For the identifications of the two latter, see note 3, page 23.Google Scholar

^{page 27 note 1} Late British Resident at Baghdad.

^{page 27 note 2} From an examination of a cast made from the monument by M. Bonomi, and placed in the British Museum, the character, so far as its mutilated condition will allow us to pronounce, appears to be of the Assyrian class, although some of the groupes approach more nearly to those Of the Ván Inscriptions.—ED.

^{page 28 note 1} I had the good fortune to be the first to draw Attention to the antiquities of Elymais, although, as I was indebted to oral information only for a knowledge of the ruins, the accounts were found by Mr. Layard and Baron de Bode when they visited the localities to be exaggerated, and in some instances incorrect. The Cuneiform Inscriptions of Shikaft-i-Salmán, were particularly mentioned in my Memoir on Susiana. See Journal of the Royal Geographical Society, vol IX. p. 84.Google Scholar

^{page 29 note 1} “ On Cuneatie Writing,” announced for publication by the Syro-Egyptian Society of London.

^{page 29 note 2} The direction of the Babylonian writing, from left to right, is another remarkable instance of departure from the usages of the Semitic nations. Professor Grotefend has pretended that a direction from right to left is the natural order of cursive writing among the Orientals who sit cross-legged, while the reverse is the natural order of inscriptions on stones; but the English translator of Heeren has well remarked, that this distinction is altogether fanciful. See Heeren's, Researches, vol. II. p. 348.Google Scholar Putting aside the doubtful questions of the origin of Cuneiform writing and the direction of the Himyaric inscriptions, we may consider it as a general, if not an universal rule, that Semitic languages in all ages have been written from right to left, while the contrary direction has been adopted in the writing of all nations of the Arian family.

^{page 31 note 1} Professor Grotefend alludes to several of these legends, (Heeren's, Researches, vol II. p. 345,) and Klaproth has given the copy and translation of a similar impression from a Babylonian brick in his Aperçu, p. 78.Google Scholar I have three specimens of the same character under my eyes at present, which are severally copied from Babylonian bricks, gems, and cylinders. I observe, that like the Cuneiform legends of the cylinders, the characters are reversed for the purpose of impression; the letters which run from left to right are evidently allied to the Phœnician, and according to Klaproth, are the earliest form extant of Semitic cursive writing.

^{page 31 note 2} Klaproth, in his usual dogmatic way, rejects this hypothesis as an absurd pretension which has nothing to support it beyond Rabbinical tradition. The squared Hebrew, he says, was borrowed from the Palmyrene, and its antiquity may be limited “presque avec certitude” to the fourth century of the Christian era. Aperçu, &c. p. 79. The only evidence I can find, however, that in any way supports this assertion is that of the coins of the Maccabees, which exhibit the same characters as the Asmonean medals in use before the captivity. That a squared character, however, approximating to the Hebrew more nearly than to any other Semitic writing, must have been employed, at least in the third century before our era, is shown by the Arianian and Parthian alphabets, of which the Western origin has been demonstrated by Professor Wilson, (see Antiqua, Ariana, pp. 260, 261,) and I have also met with Babylonian gems inscribed with Hebrew characters, of which the execution denoted a very high antiquity, as early probably as the age of Alexander. Gesenius would be better authority than Klaproth in a question of this sort, but I have not at present means of reference, either to the Geschichte der Hebräischen Sprache, or to the Scripturæ Lingiuæque Phœniciæ Monumenta.Google Scholar

^{page 31 note 3} Lib. iv. cap. 87.

^{page 31 note 4} Lib. ii. cap. 1. The passage is quoted from Ctesias, and refers to the inscriptions of Semiramis at Baghistan or Behistun

^{page 32 note 1} Democritus, the Greek philosopher, is said to have interpreted the inscriptions on the column of Acicarus, and to have incorporated their contents in his works on Babylonian Ethics. He was at Babylon in the commencement of the fourth century, B.C. See Clem. Alex. Stromata., edit, Sylburg lib. i. p. 303Google Scholar, and Clinton's, Fasti Hellenici, vol. II. p. 105.Google Scholar

^{page 32 note 2} M. Et. Quatremére appears to have especially treated on the ancient cursive writings of Assyria and Babylonia in his Memoir on the Nabathæans, Nouv. Journ. Asiat. tom. XV. page 244., sqq. I have never seen, however, this article, which is highly spoken of.Google Scholar

^{page 32 note 3} See Mémoire sur deux Inscriptions Cunéiformes, p. 176, sqq.

^{page 32 note 3} I allude to the Suez stone first copied by Gen. Dugua, and published in Denon's Voyage, pl. 124. See Heeren's, Researches, vol. II. p. 319;Google Scholar and Descrip. de l'Egypte, Antiquit. Mémoires, I. 3. p. 269.Google Scholar Lassen has described and translated the inscription in his last Memoir, page 81. See Zeitschrift, &c. vol. VI. No. 1.

^{page 32 note 5} See the Inscriptio Vasis ap. Caylus, tom. V. Tab. XXX., published and rendered by Grotefend in Heeren's Researches, vol. III. last plate. For the inscription on the Egyptian vase in the Library of St. Mark's, discovered by Sir G. Wilkinson, see chap. 5 of the present Memoir.

^{page 34 note 1} As in Armenian, Georgian, and Chinese.

^{page 34 note 1} DrPritchard, observes, in his Researches, vol. IV. page 385,Google Scholar that in the languages of High Asia, “the n and l are interchangeable consonants, as it has been abundantly proved by Dr. Schott.” This peculiarity of orthography appears to me in a remarkable way to connect the Scythic with the Semitic races. In Pehlevi the substitution of the n for the Persian r (which I consider to be the same as l) is the distinguishing characteristic of the language, and both in the ancient and modern dialects of Babylonia the same confusion is observable. We have thus Labynetus and Nabonidus, Nebuchadnezzar and Nabochodrossor, Rubil in Arabic for the Hebrew Reuben, and the modern Mendallijín for the ancient name of Band-i-Najín (the Mendalli of the Maps).

The substitution of these letters may sometimes be traced in the Arian languages, as in alius, alter, in Latin, for anyas, antar, in Sanskrit. Bána, in Kurdish, for the Persian Bálá, “above.” The of Dionysius for , &c. Eustathius, in his Commentary on verse 1143 of the Periegesis, says that the name was written indifferently and Peucela is Pali for the Sanskrit Pushkala.

^{page 35 note 1} See Pritchard on the Celtic Nations, chap. I. sect 2.

^{page 36 note 1} I allude to the Georgian and Armenian. Dr. Pritchard, in the 13th and 14th chapters of the 4th volume of his Researches, discusses in the most able manner the ethnographical relations of these languages. The Armenian he classes among the members of the Arian family, but he observes that its claim to affinity is remote, and that it has been but recently admitted; while of the Georgian speech he observes, “we may set it down as one by itself, unconnected or but distantly connected with any other idiom.”

^{page 36 note 1} Heeren considers the cognate origin of the Medes and Persians to be susceptible of direct proof; (see Researches, vol. I., p. 324;)Google Scholar and Dr. Pritchard, throughout his Asiatic Researches, employs the terms Median, Persian, and Medo-Persic, in an ethnographical sense, as almost synonymous.

^{page 36 note 2} For the etymologies of Frawartish and ‘Uwakhshatara the true native forms of and see the Vocabulary, in vocibus. M. Burnouf translates or Ajis-Daháka, as it is read in the Vendidad, “the biting snake.” See Nouv. Journ. Asiat. IV. series, tom. IV. No. 20, p. 498.Google Scholar

^{page 37 note 1} There can be little doubt but that the Cadusian war described by Ctesias as occurring during the reign of Artæus, refers to the Median revolt under Dejoces, and it is possible also that the Cadusian expedition of Artaxerxes, of which Plutarch gives an account, was conducted against the same people. In my Essay on Ecbatana, I have noticed this confusion between the Medes and Cadusians, (see Jour. R. G. Society, vol. X. part I, page 126;) and I shall have occasion to examine the subject more in detail in a future portion of the present Memoir. One of the chief cities of Media, I may add, is named Ghudhrush in the Inscriptions.Google Scholar

^{page 37 note 2} There are many notices in the classics which connect the ancient Median speech with the Scythic, but whether by Scythic in this case is meant Sclavonic or Tartarian, I doubt if we are in a position at present to decide. The Sarmatæ, or Sauromatæ, who spoke Scythian, although badly, were certainly a colony from Media. See Herod. IV. cap. 110–117Google Scholar; Diod. Sic. lib. II. cap. 3; and Plin. lib. VI. cap. 7. Again, Pomponius Mela connects the Sarmatæ with the Parthians, (lib. IV. cap. 3;) and of the Parthian language, Justin (lib. XLI. cap. 2,) says it was a connecting link between the Scythic and Median; while John of Malala (Edit. Dindorf, p. 26), quoting from some unknown passage of Herodotus, says, “the Parthians to the present day retain the dress, language, and laws of the Scythians.” Strabo's famous passage, p. 374, in which he connects the Medes, Persians, Bactrians, and Sogdians in one homo-glot family, refers apparently to a later period of Arian colonization, although, at the same time, it must be admitted that a quotation which he also gives from Nearchus of a nearly similar purport, as far as the Medes and Persians are concerned, can only apply to as early an age as that of Alexander's conquest.

^{page 38 note 1} This inscription, which refers to the various historical tablets executed by Darius in different parts of his empire, is of particular interest, as it records several names which I have not otherwise met with, and which, indeed, up to the present time, I have not been able satisfactorily to identify.

^{page 38 note 1} I need only mention the attribution by Ctesias of the name of Arsaces to Artaxerxes Mnemon, to show that the title was in use under the Achæmenians.

^{page 39 note 1} This inscription has been copied by Sir William Ouseley, Porter, and Rich. The best account of it may be found in Lassen's, last Memoir, Zeitschrift, vol. VI. p. 152.Google Scholar

^{page 39 note 2} For Lassen's translation of this inscription, see Zeitschrift, vol. VI. No. 1, p. 159, sqq. My own rendering, in chap. 5, will be found to be slightly different.Google Scholar

^{page 39 note 3} I do not think it necessary to specify the relative position of the tablets in all the various inscriptions of Persia. Instances will be found of the three different modes of collocation which I have here noticed, but the usual order of the series is from left to right, in the same direction as the writing.

^{page 40 note 1} This is the famous Geographical Inscription of Niebuhr, marked I., and copied from the outer face of the southern wall of the great platform. For Lassen's translation of the amended copy made by Westergaard, see Zeitschrift, &c, Vol. VI. No. 1, p. 42, sqq. The particular force, however, of the passages to which I allude, has escaped the Professor's observation.Google Scholar

^{page 40 note 2} Dr. Pritchard observes, in his Researches, &c, vol. IV. p. 382, “Another peculiarity prevails throughout the formation of the Turkish and Mongolian languages, in which certain consonants can only be pronounced in juxta-position with certain vowels.”Google Scholar

^{page 41 note 2} James Prinsep, who first decyphered this interesting character, was struck with its resemblance to the most archaic form of Greek, and he drew up accordingly a comparative table of the Pali and Sigæan alphabets, (see Journal of the Asiatic Society, vol. VI. p. 382). His theory, however, has found little favour with Palæographers, the names, powers, and series of the Greek letters, together with unanimous tradition, establishing, it is thought, beyond the reach of controversy, their Phœnician derivation. The extraordinary similarity, at the same time, between the forms of the letters, as they were used at a very early epoch by two branches of the Arian family, widely severed as were the Greeks and Indians, together with the common direction of the writing from left to right, in contradistinction to the Semitic usage, requires still, I think, to be explained. Prinsep's attribution of the Pali character to the fifth century B.C. is altogether assumptive. The reign of the Nandas, shortly before the Macedonian conquest, is the limit of its probable antiquity, although for the transcript of the Buddhist doctrines there must have been a cursive Arian prototype.Google Scholar

^{page 42 note 1} The very elaborate vocalic organization of the Zend indicates, I think, a comparatively recent era for the formation of its alphabet; while the disfigurement of authentic history affords an argument of equal weight against the possible antiquity of the composition of the Zend-Avesta. At the same time, there is strong evidence of the Magi having possessed from the remotest period, books which they ascribed to some proto-patriarch of the name of Zoroaster. These books are quoted by Plato (Pol., B. XXX). They were in the hands of the disciples of Prodicus, who flourished in the fifth century, B.C. (Clem. Alex., Edit. Sylburg, p. 304).Google Scholar They supplied Osthanes, who accompanied Xerxes in his Grecian expedition, with materials for his work on Magic. They were expounded and indexed by Hermippus (Plin., lib. XXX. cap. 1). I do not allude to the later extracts of Eusebius, Suidas, &c., or to the published Zoroastrian oracles, for their claims to antiquity are apocryphal; but notices of the fourth and fifth century, B.C. are certainly deserving of consideration.

^{page 42 note 2} Ezra, chap. VI. v. 2.

^{page 42 note 3} Daniel, chap. VI. v. 9.

^{page 42 note 4} Nehemiah, chap. II. v. 9.

^{page 42 note 5} Herod, lib. VII. cap. 100. Diod. Sic., lib. II. cap. 3. Herodotus in no passage expressly mentions the royal records, but he notices the muster-roll of the army of Xerxes. On this subject, see Heeren's, Researches, vol. I. p. 85, sqq.Google Scholar

^{page 43 note 1} These legends, I believe, up to the present time have never been decyphered. The coins are figured by Mionnet, vol. V, plate 21, and Gesenius speaks of the character as “specimens of the ancient Persian writing.” (Script. Ling. Phœn. mon., part I. p. 74.) By the generality of Numismatists they are referred to the cities of Sida and Celenderis, but M. Adrien do Longperrier, from the similarity of their emblems to those on the sculptures of Khoraabád, would refer them to Assyria, under the rule of the Achæmenians, (see Ninevé et Khorsabád, in the Revue Archtæologique of July 15, 1844.) It is probably to this writing that the apocryphal letter of Themistocles alludes, in speaking of the new Assyrian characters introduced by Darius, . Them. Epist., p. 117.Google Scholar

^{page 43 note 2} If Captain Cunningham be correct (and it is rarely safe to dissent from him) in attributing the bilingual coins, bearing the legend of “Amogha-bhutasa-maharajasa-rajnya kunandasa,” to Nanda Mahapadma, we have specimens of the Arianian Numismatic alphabet sometime anterior to the age of Alexander; and the lapidary character of Kapur-di giri, dating about a century after that era, will then be a derivative rather than a type. Judging from the mere forms of the letters, the rock inscriptions have all the appearance of superior antiquity, and Professor Wilson appears to consider the legends on the coins of Eucratides (B.C. 181) as the earliest specimens of the Numismatic Bactrian. See Journal of the Asiatic Society of Bengal, vol. VII. plates 2, 3, &c; and vol. X. p. 157.Google Scholar

^{page 43 note 3} For the Numismatic Baetrian, and the character which is found on jars, cylinders, &c., in the Buddhist topes, see Professor Wilson's Ariana Antiqua, passim, and particularly the chapter on the Arianian alphabet.

^{page 43 note 4} I cannot here enter upon the discussion of that controverted question, the origin of the Zend character, but I entertain a very strong suspicion that the alphabetical system is far more modern than is generally supposed. Klaproth would assign the character to an Indian source, (Aperçu, , &c., p. 68,Google Scholar) but all the letters with which he has brought it into comparison are very modern degradations of the old Pali type, and the resemblance, therefore, can but be accidental. (See Prinsep's, comparative Table, Journ. Asiat. Soc., vol. VII., p. 276, and plate 9 of Klaproth's Aperçu, &c.) It is worthy of remark that the early Sassanian Pehlevi is to all appearance a connecting link between the Zend and the Semitic type; for it is only through the Pehlevi that we can compare the Zend letters with the Hebrew. I must reserve this difficult subject for future examination.Google Scholar

^{page 44 note 1} The three varieties of Parthian to which I refer are, lstly. a very barbarous character, which is found on the tablets at Tang-i-Sulúk, near Bebahán; at Shimbor in the Bakhtiarí mountains, and in a cave near 'Amadíah. The first set of these inscriptions have been published by M. Boré, in the Journal Asiatique, after the Baron de Bode's copy; for transcripts of the others I am indebted to the kindness of Mr. Layard and M. de Laval; 2ndly. the character of the inscription at Holwan, which, in my Memoir on Susiana (Journ. Geo. Soc., vol. IX., p. 37)Google Scholar, I have named Pehlevi, but which, on further examination, I believe to be a very old type of the Parthian; and 3rdly. the character of the Parthian translation in the bilingual inscriptions of Ardeshír Babegán. De Sacy (Ant. de la Perse, pl. 1, A. No. 4) published, and attempted to read, a short specimen of this class, from Nakhsh-i-Rustam. Porter (Travels, vol. I., p. 512,Google Scholar pl. 15) contributed a long Parthian Inscription from a cave at Hajiabád, and I have also extensive legends in the same character, copied by myself, from the ruins of the five temples at Shahrizor. I doubt, however, if, as Professor Wilson observes (Ariana Antiqua, p. 261), the Parthian can be considered to form a stage in the transmutation of the Bactrian Pali to the Numismatic Sassanian; for the latter is the nearest to a pure Semitic type of any species of Persic Palæography, and must, therefore, I think, have been adopted direct from the banks of the Tigris. There appears, also, to have been a Parthian Numismatic character, which is figured on some Arsacidan coins published by Mionnet, vol. V. pl. 29, and which is also, I believe, noticed in the Memoir by SirOuseley, William, On some Medals and Gems, &c, &c, London, 1801,Google Scholar but I have never seen a specimen of this writing. The character on the Eastern coins of the Arsacides (Vonones, Undopherres or Gondophares, Pacorus, Arsaces, Orodes, &c.) is a mere degradation of the Numismatic Bactrian, and is sufficiently legible.

^{page 44 note 2} In the time of Ardeshír Babegán, the lapidary and numismatic Pehlevi were nearly identical, but in succeeding ages the latter became very essentially degraded, as has been clearly and elaborately shown in M. Adrien de Longperrier's Essai sur les Médailles des Rois Perses de la Dynastie Sassanide, Paris, 1840. At the same time the most degraded writing on the coins will be found to differ materially from the text of the Pehlevi books at present in the hands of the Indian Parsís, and I think, therefore, we may classify the character as lapidary, numismatic, and cursive. We may also trace the lapse of the degraded numismatic into the cursive character now in use, through the writing which is frequently found on the interior surface of jars and sepulchral urns disinterred in different parts of Persia. The short legend again, written on the breast of the king's horse, on the great tablet at Shápúr, appears to have been engraved while the Pehlevi was then in a state of transition, and I have impressions of several gems which still further facilitate a connexion between the modem and ancient characters. In the names of the Parsí witnesses attached to the copper Sasanam, which is at present in possession of the Syrian Christians of Malabar, we have probably an interesting specimen of the Pehlevi character, as it was carried to India by the first emigrants of the Zoroastrian faith, when they fled from the Arab army on its approach to Abilah, at the mouth of the Euphrates, and sought refuge at Sindán, a town on the coast of Guzerat, well known in Arab geography, but which, without this direct testimony of Hamzeh Isfahání, we should have some difficulty in recognizing in the St. John of the modern maps.

^{page 45 note 1} See Journal Asiat. Soc. of Bengal, vol. IX., p. 155.Google Scholar The same eminent numismatist, whose forthcoming work, “On the Coins of Alexander's Successors in the East,” is likely to throw a most valuable light on the subject of Arian Palæography, assigns, as I have before observed, the bilingual coins of Raja Kunanda to the elder Nanda, who preceded Chandragupta or Sandracottus, the contemporary of Seleucus Nicator, by two generations. If this appropriation be admitted, our earliest specimen of the Indian Pali will date from about B.C. 370.

^{page 45 note 2} See extract from the Index to the Káh-gyur, translated by de Körös, Csoma, in Journal Asiat. Soc., vol. VII. p. 282.Google Scholar The introduction of Tibetan into this list, is probably the natural vanity of a Buddhist writing in that language. Scythia, which is identified by its position beyond the Sita river, is called the Monkey Country, or the Gold Land.

^{page 46 note 1} The bilingual legends on the coins of Kunanda afford a striking proof that the direction of a writing depends on the alphabetic type, and not on the language. The inscriptions on the obverse and reverse of these coins are to the same effect, and in a common language, but they are written in a different character. The Bactrian Pali, which is allied to the Semitic alphabet, follows a direction from right to left, while the Indian Pali, which is of the Arian family, reads from left to right.

^{page 46 note 2} It can hardly be said that the Bustrophide Greek reads both ways, for the first line, which determines the direction, is from left to right, while the following line, in which the direction is reversed, is merely the complement, that was tacked on backwards to suit an agricultural conceit.

^{page 46 note 3} This short inscription, which merely contains the words, “I am Cyras the King, the Achæmenian,” is repeated several times on the ruins at Múrgháb, the remains, probably, of the tomb of Cyrus the Great.

^{page 46 note 4} I consider the mounds in the vicinity of the tomb at Múrgháb to be one of the most favourable spots for excavation in all Persia. If the site be really that of Pasargadæ, as there seems every reason to believe, we might hope to find among the monuments of Cyrus, which are doubtless buried in the vicinity, some record of the liberation of Persia from the Median yoke.

^{page 47 note 1} See Inscription marked B, in Lassen'a Zusammenstellung; Zeitechrift, vol. VI., No. 1, p. 170.Google Scholar

^{page 47 note 2} See Inscription marked H, in the same collection, p. 175. In chap. V. of the present Memoir, I have given amended translations of all these inscriptions.

^{page 47 note 3} Inscription marked I, in Lassen's, collection, p. 175.Google Scholar

^{page 47 note 4} This is the great Inscription of above 400 lines, which forms the principal subject of the present Memoir.

^{page 47 note 5} See Lassen's, collection of Inscriptions, p. 179.Google Scholar Inscription marked O. This is the record which was critically examined by M. Burnouf, in his Mémoire sur deux Inscriptions Cunéiformes, &c.

^{page 48 note 1} For Lassen's translation of this long inscription of 60 lines, which was very accurately copied in 1843 by Mr. Westergaard, see Zeitschrift, p. 81119.Google Scholar

^{page 48 note 2} On further consideration I am inclined to think this explanation somewhat fanciful. The sentence which introduces the Geographical list may be rendered, perhaps, “These are the nations which I have subdued, together with this state of Persia,” rather than “by means of this state of Persia,” which is the reading I have hitherto adopted. Lassen's, translation of the passage, Zeitschrift, p. 45, will, I have no doubt, be abandoned by the author when he peruses the present Memoir.Google Scholar

^{page 48 note 3} Lassen, in his Zusammenstellung, has eight inscriptions of Xerxes, copied from different parts of the ruins of Persepolis. There are also repetitions of several of these inscriptions upon other tablets at the same place.

^{page 49 note 1} The inscription of Ván is given by ProfessorLassen, , after Schultz's copy, Zeitschrift, p. 176.Google Scholar The transcript, which will be found in the fifth chapter of the present Memoir, and which follows the MS. of M. Boré, is more complete than. Schultz's copy, but is still defective in the last two lines.

^{page 49 note 2} I am perhaps hardly justified in this statement. There is, certainly, no autographic rock inscription of Darius Hystaspes which bears the title of Naqa (?); the term is found in connexion with the name of Darius upon the Suez stone, and upon a Babylonian cylinder, both of which relics are assigned to the son of Hystaspes; see Grotefend's, Neue Beiträge zur Erlauterung der Babylonischen Keilschrift, p. 34Google Scholar; and Lassen's, last Memoir, page 80;Google Scholar the honorary legend also of Artystone at Persepolis has the same title of Naqa in apparent allusion to Darius Hystaspes. If Naqa were a title first introduced by Xerxes, the three records above noticed must be referred to Darius Nothus, and I am unwilling, without further evidence, to adopt such an appropriation. In the inscriptions of Xerxes, the term Naqa and K'hsháyathiya are employed indifferently.

^{page 49 note 3} I allude to the terms A'urahya Mazdáha and Par'uwa-zanánám for A'uramazdáha and Par'uzanánám. The title, however, which is usually written Dhuriya-apiya, both by Xerxes and Darius, is found in the inscription of the latter king at Nakhsh-i-Rustam to be contracted in a single word Dhuriápiya, and the division, therefore, of compounds can hardly be admitted as a determinate effect of degradation in the language.

^{page 50 note 1} This inscription was, I believe, first published and imperfectly rendered by Professor Grotefend, in the Neue Beiträge, s. 13, from the copy in the British Museum. The text of the inscription was given entire in Rich's work, Plate XXIII., and it forms the principal subject of Lassen's, article, “ On some new Cuneatic Inscriptions,” in the Zeitschrift, &c., vol. III. No. 16, p. 442.Google Scholar Westergaard's duplicate, which was transcribed in 1843 (the slab having been previously laid bare and, I suppose, copied by Messrs. Flandin and Coste), and which amends the text of Rich in some important passages, has been published and translated by Lassen, , in his last Memoir, p. 159.Google Scholar

^{page 50 note 2} In one of these characters the two inscriptions vary. Rich has and Westergaard .

^{page 50 note 3} On the Egyptian vase in the Library of St. Mark's, which also dates from the age of Ochus, the orthography of the language is even more degraded than at Persepolis; for the name of Artak'hshatrá is there found to be corrupted to Ardak'hchâshcha.

^{page 51 note 1} I am aware that in thus suggesting a comparatively recent epoch for the reduction of the Zend language to its present form, I am treading on dangerous ground; but I am obliged to say that M. Burnouf's arguments have altogether failed to convince me that the Zend was immediately cognate with the Vedic Sanskrit, or indeed, that it was ever a spoken tongue. It is certain that the language of the inscriptions is not a derivation from the Zend; the two forms of speech may possibly have existed synchronously, one as a Demotic and the other as a Hieratic language; but in that case the disfigurement of historical names and the straining after artificial etymologies, which occur in every page of the Zend-Avesta, are points which will hardly admit of explanation. I shall recur to this subject in another place.

^{page 52 note 1} See De Sacy's, Antiquités de la Perse, p. 137.Google Scholar The author of the Mujmal-el-Tawárikh also states, on the authority of Hamzeh Isfahání (whose evidence is always valuable on the subject of Persian antiquities), that a Múbid having been invited to read the Pehlevi Inscriptions of Persepolis, interpreted the legends to signify that “King Jem had done so and so, on a certain day of a certain month.” Had the explanation referred to the inscriptions at Behistun, the coincidence would have been remarkable; for the various actions of Darius are there recorded according to their respective dates, but at Persepolis chronological annalism is not attempted. I must add, however, that the real Pehlevi Inscriptions of the Takht-i-Jemshíd are to the present day undecyphered.