^{page 2 note 1} The true spelling of this word is restored from Russian authorities; the S was adopted by the Germans, to suit their pronunciation, in the same way as the Zend language is made by them Send. In the present instance, an obvious eonfusion is avoided by the adoption of Z.

^{page 4 note 1} Strabo, lib. XI. cap. xiii. sec. 3, 6.

^{page 7 note 1} The references are made to the columns at Behistun; those in small Roman numerals refer to the minor Inscriptions, which are not generally engraved here.

^{page 10 note 1} These suffixes are not added indiscriminately, but the termination follows words whose last syllable begins with a liquid or semi-vowel, while follows those beginning with any other consonant; we have thus Assura-fa, Arbaya-fa, Yauna-fa, &c, for Assyrians, Arabians, and Ionians; and Vata-pa, Sakka-pa, Markus-pa, &c, for Medians, Sacse, and Margians. This is not inconsistent with the principle of vocalic harmony as prevalent in the Manchu, Mongol, and Turkish languages, in all which certain consonants always affect certain vowels.

^{page 14 note 1} In this and similar cases I am compelled to anticipate. I hope to be able to show that the Ostiak, Zyrianian, and other Ugrian tongues ara allied to the languages under investigation; but in the meantime the statement in the text must be taken quantum valeat. The transcription, at all events, is left with the rowel e, the sign of doubt or ignorance.

^{page 16 note 1} We find always put for this Inscription; as instead of .

^{page 19 note 1} See Rask's Icelandic Grammar, p. 21. Stockholm, 1818. 8vo.

^{page 20 note 1} Ostiak Grammar, p. 7. St. Petersburg, 1849. 8vo. I use the word Tartar in its widest signification, without wishing for one moment to insist upon its propriety; I mean nothing more than the “Allophylian” of Dr. Pricbard.

^{page 22 note 1} The second letter in the name of this province at Naksh-i-Kustam is , as appears from Professor Westergaard's Note, page 293; he thought he saw , but having So present for the form, he wrote it ku.

^{page 23 note 1} As a general rule, when I transcribe the sentences into separate words, I put a grave accent over the compound syllable, making tàs. In the other case, as , I Put >, writing täs This is obviously unnecessary where the syllables are separated, and is not done except in cases where ambiguity is thereby avoided.

^{page 34 note 1} Zeitschrift der Deutschen morgenländischen Gesellschaft, Vol. V. p. 162Google Scholar.

^{page 43 note 1} This word is written at Naksh-i-Rustam, line 23, The variation may be accidental, for in other characters, such as ar, the quadruple wedges of Behistun are replaced by triplets at Naksh-i-Rustam. But the character under consideration is usually made in the inscription found there (see lines 43, 46, 47). This circumstance indicates an affinity of sound in the two characters.

^{page 62 note 1} E.g., vä-ne, “in the night;” tel-nä “in the winter. See Gabelentz, p. 245. The Turkish language has traces of such case-ending, as yazin, “in the summer;” qyshin, “in the winter;” euylèn, “at midday.”

^{page 62 note 2} Works on Tartar and Ugrian languages:— Versuch Über die Tatarischen Sprachen, von Dr. Wilhelm Schott. Berlin, 1836.

De Affixis Personalibus Linguarum Altaicarum Dissertatio. Conscripsit Dr. M. Alexandér Castrén. Helsingfors, 1850.

Versuch einer mordwinischen Grammatik, von H. C. v. d. Gabelentz. In the Zeitschrift für die Kunde dea Morgenlandes. Göttingen, 1839.

^{page 64 note 1} These may be more alike in sound than here represented; the syllable I have called ki is not found in any proper name, and the vowel is inserted merely on conjecture; ikki is much more frequent than ikka.

^{page 66 note 1} See also Castrén, De Affixis personalibus, &c.: “Prsecipue tertiæ personæ affixum cornpluribus in linguis cognatis, ut videtur ad nativam suam indolem exuendara, articulique definiti, quo hse linguae generatim careut, vim sensim suscipiendam nititur,” page 11.

^{page 71 note 1} Apin looks like a nominative case in vi. 16, yupa apin marris, “that they kept;” but it may be “that to themselves they kepf.”

^{page 81 note 1} This is in entire accordance with Ugrian usage. See (for the Mordwin)

^{page 90 note 1} See Wiedemann's Grammar, p. 144.

page 106 note * We now come to the part which corresponds with the second Persian column.

page 107 note * A paragraph mentioning the frequent participial signification of neuter verbs terminating in ka and fa, such as ivaka and pafatifa, when unaccompanied by the pronoun ir, was inadvertently omitted in p. 88.

page 132 note * Some doubt may attach to this explanation, arising from the emendation in Notes, p. x, where we find jadanautuva instead of danautuva.

page 146 note * For convenience of reference, a list of all the small inscriptions, whether with or without Scythie versions, is here given, in the order followed by Colonel Rawlinson, together with the letters by which they are designated in Lassen's and Westergaard's Memoirs; a few additional inscriptions have been since found. In all, whenever they are referred to in this memoir, the numerical order is followed. The detached inscriptions at Behistun are referred to by Rawlinson's letters. The lithographed independent inscription is made L.

page 148 note * The paging refers to Westergaard's Memoir in English, published in the “Mémoires de la Société Royale des Antiquaires du Nord,” Copenhagen, 1844Google Scholar.

page 153 note * From M. Oppert's Treatise. I have not De Saulcy's paper at hand.

page 154 note * Since writing the above, I have seen some MS. notes of Colonel Rawlinson on the Babylonian alphabet (which I must have seen before, but had forgotten), where he gives good reasons for translating the first words by “marble hall.” Perhaps the whole may therefore now definitively receive this version— “marble hall [or arch], built for the family of Darius the King.”

page 158 note * The lithographed copies are reduced facsimiles, made from casts, with which I was supplied by the kindness of W. Kenneth Loftus, Esq., who discovered these inscriptions. They were not received until long after the Memoir was concluded.

page 159 note * I should have expected niyastiyam here, but there are certainly the distinct traces of and on the fragment.

page 161 note * See Gesenius. Phœn. Monum., 1837, p. 115.

page 184 note 1 A close examination of the paper cast in a bright light has satisfied me that the space after the first character, which I supposed unlettered, is occupied by the letter t, the word is, therefore, certainly titrasra, and the lithograph and transcript must be corrected accordingly.

^{page 200 note 1} Marco Polo's balu, in Kanbalu (City of the Khan). The word now used by the Mongols is Balgasun, in the Dictionaries. I know not if balu be also in use.

^{page 201 note 1} An examination of the Persian text, which I had not seen when the ahove was written, shows the correctness of Colonel Rawlinson's explanation; but the adoption of a Persian pronoun is at least singular.

^{page 208 note 1} Colonel Rawlinson, in his Vocabulary, withdraws the chitá of 1. 28, which he had restored in his Anal) sis, though not on the engraved plate, on the plea of want of room; but there is a defective space in the preceding line, between pasava and hya, where the word might have stood; and a comparison with the Scythic II. 20 will show that it almost certainly was there.