declensions of nouns. Without rashness we may take a step backward to the vowel-declensions of Umbrian, as follows: η When æ and œ have been corrupted into e, the dative sing. becomes the same in the three first declensions. In fact, the same holds of the dat. pl. For, ie in dat and acc. pl. has been replaced by ei, i, e, exactly as in the Latin acc. pl. turreis, turris, turres. If I were to print æ, œ, I should not deceive the reader, any more than in distinguishing e n, o w, in a Greek inscription which rejects n and w; but I should be open to the charge of ambitiously attempting to restore an older state of the language, while groping towards a knowledge of what is before us. I have, therefore, merely added grave and acute accents on e, writing è for æ and é for oe, which suffice to warn the reader to which declension a noun belongs. Also, I have admitted the circumflex as in the scheme above. It must be added, that -is for -œs -æs is sometimes found. To add a distinguishing accent to the -is is but consistent. The task of interpretation would be far easier if corruption of the vowel sounds alone troubled us. What completes confusion, the engraver, ad libitum, omits final m, and f of the accusative * Ui is corrupted into mere i. Compare modern Gr. vi. pl., and so often omits final s of gen. sing. or dat. pl. (or its equivalent r in the later dialect), that though this is not to be called ad libitum, and perhaps was carelessness, it is sufficiently frequent to involve uncertainties. I think it clear that the law of concord in nouns and adjectives was imperfectly established. An Umbrian probably reasoned like a Turk, that to say Owem sewacnem (ovem puram) or Anclaf esonaf (volucres pias) was superfluous. Why twice over denote that you mean the accus. ? Owem sewacne, or Owe sewacnem, will suffice: so will Anclaf esona, or Ancla esonaf. Out of this habit of alternate omission naturally springs that of total omission, which is worse in the later than in the earlier tables, where we find a state of things like that of Greece fifty years ago, in which it was an open question whether ἡ πόλι, τήν πόλι was more correct, or ἡ πόλις, TηU TÓMV. To aid readers, Lepsius often inserts m or f in brackets in his text; and, again I say, it saves notes: an important matter, where all effort is needed to hinder the notes from swallowing up the text. I have imitated him, by printing small letters (m, f, s) above the line, at least in the earlier tables. Afterwards I presume often that a reader can supply them of himself. I may add, that the inconsistent efforts at concord of the Locative case imply the laws of grammar to be unformed on this head. I have arranged the tables in what appears to me from internal evidence to be the order of their age. Ia. IIa. etc., denotes the front of Tables I. II. . . . and Ib. IIb. . . . their back. I do not know how to quit my pen without a few words to the persevering but almost solitary students of cuneoform inscriptions. I respectfully ask-Is it simply impossible to put before the public a transcription of their principal documents into a Roman character? Mathematical types give us letters modified by numerals; there is every facility for thus printing (somehow, if clumsily) every possible document that is truly alphabetical; and if all are not alphabetical, yet some are. Retired gentlemen from India, each acquainted with several different Indian languages, would soon multiply the students tenfold, if the inscriptions were but presented in an alphabet with which we are familiar. I am persuaded, that this is the thing needed to give a great impetus to the study, and promote even the perusal of the cuneoform character itself. For, those who will not encounter both difficulties at once, would be induced to have recourse to the originals, if they had already gained some insight and interest in the substance of the languages, by means of familiar types. Moreover, by practising for the third part of a century on the Arabic language, which abounds in consonants troublesome to us, I have satisfied myself that the problem of writing, as well as printing them, by easy modifications of our alphabet (without dots or accents) is very feasible: nor am I ready to believe that the ancient Persian or Assyrian can have any greater difficulties on this head than Arabic. THE IGUVINE TABLES. TABLES III. IV. VOLUNTARY SACRIFICES AT FEASTS AND PROCESSIONS. 1 Esono fuia herter sommè 2osditè sestentasiarum 3urnasiaru": Sacrum fiat ultro summæ prodita sextantariarum urnariarum: hontac Wocé promo" pehatu. 4Inoc ohturo ortès, pontîs inde Foco primum piato. Tunc auctorem éopraîs (et) pompis 5frater ostentôta, pore fratrom mersûs fust comnaclé. fratres proponunto, quisquis fratrum faustus fuerit communitati. Inoc ohtur wapere, 8 comnaclé sistu sacrem owem. Ohtur Tunc auctor [curiæ] (ac) communitati sistito sacram ovem. Auctor TABLES III. IV. 1. Esono, by indn. sacrum, religiosum; A.K.-The root is Son = Sna: Germ. Sühne, Versöhnen. So Snato, sacratus; Persontro, piatorius. Cmp. Va. 6, IV. 7. -May Lat. Sons = évayns ? 1. 8. Fuia, Fuja, Optative or Potential Mood. Cmp. -oín. Futu serves for Fito and Esto: thus Fu-v=Fi-0. 1 y. Her-ter-vol-tro, ultro. With -ter cmp. forti-ter. It recurs only IIa. 40: later Herte, -i, -ei; but Herifi, Vb. 6. For the root Her-vol-o, see on IIb. 10. 2. Osdita prodita, pronunciata. Ostentu ostendito, proponito, and Ditu =dato. Os = Lat. Obs, Ob; in sense, propalam. = 2 B. Sextantarius, epithet of an as in Pliny; weighing two ounces. In Va. 2, plenarius, of full weight, seems equivalent. 3. Urnasia, a coin; perhaps bearing an urn: cmp. cistophorus. The vow is voluntary; but to make it de certâ pecuniâ (Liv. 31, 9) the coin is defined.-B. Hontac (by cnx. and in IV. 32) inde; de hac pecuniâ. y. Foco, i.e. Lari? 4. Inoc is in Tables III. IV. I.; Enoc (Etr. Umb. character). in I. Va.; Inomec in III. IV. only; Enomec in Ib. Enom, Eno, replace them in VI. VII., but Eno is also in II. I. Inomec seems the most old-fashioned. -4 B. Ohtur, Ohtretie Va. 2; auctor, auctoritate; ht for ct: A.K. See note at Va. 2.-7. Ortès pontîs; kopraîs, Tоμжаîs. It is too tedious to tell, how was driven step by step to this, before I thought of the Greek words. I have long theorized that Pontifex means Pompifex, (as πέντε for πέμπε:) I now believe it. = 6. Mersûs Mersow (o)s; root Mers, Mers, fas. The Wia mersowa of 11 = via auguralis VIb. 52. With termination -owo, cmp. -oFo and -ivo. 7. Comnacle, Va. 15 is dat. sing. of a noun; which fixes the syntax here. Ib. 41, Comne = plebs, τὸ κοινόν. B. Waper, I confidently believed from this passage to be adjectival, and fancied I could identify it with åravт: yet its obvious, and only natural interprn. in VIa. 9-12 makes it to be a tall building. If it be a noun (which I hesitatingly 1 9deitu pontés dercantor. Inomec sacrem 10 owem ortâs, dicito: pompæ dedicantor. Tunc sacram ovem ἑορτῆς (et) pontes fratrom opetôta. pompæ fratrum procuranto. 11 Inomec wiam mersowam arwamen etôta: 12erac pir persclu Tunc viam faustam in arvum eunto: illac †quis ordine oretu sacrem owem. 13 Cletra fertôta, aitôta. Arwèn †adoleto sacram ovem. Lectos? ferunto, †disponunto. In arvo cletram 14amparitu: eruc esonom futu. Cletrè duplac 15prolectum apparato : illic sacrum fito. Lecto δίπλακα primom antentu. Inoc çihçera ententu; 16inoc cazi ferrime mùm imponito. Tunc cremia incendito; tunc [palos ferreos] antentu; isont ferehtrom 17 antentu; isont suffeṛaclom imponito: itidem feretrum imponito: itidem † sustentaculum antentu. Seplés 18 ahesnés trîs cazi astintu ferehtrom imponito. Singulis ahenis tribus †[palos] ava-stinato: feretrum etrés trîs 19ahesnés astintu; sufferaclom dowes ahesnés alteris tribus ahenis àva-stinato: sustentaculum duobus ahenis admit), it is in apposition to Comnacle, community, like "Senatus populusque,' and must express a more select body. I see nothing then so good as Curia. But etm. gives no support. 9. Dercantor, corrupt Latin; for De does not appear to be Umbrian; but in compn. Wen, We replaces it. See IV. 28. 10. Opetu obito, A.K. The vague sense procurato may evade the ill-omened cædito, jugulato, which indn. suggests. See V b. 9 on Opeter, curati, which I desire to explain purgati. It remains doubtful whether Op Lat. Ob, or whether Ope is a root akin to Latin Opis and Opera;-or even Op-petere be concealed here. The 3rd p. pl. in -tôta (=-ETWσAV) is peculiar to this table: elsewhere -tuto (=-TOVTWV) serves for 2nd and 3rd p. alike. 11, 13. Arwam-en: Arwè-n: see Appendix on Locative cases. 12. Pir, ignis (see 21) is surely here too poetical. Pis is quis; Pisher, quivis, VIb. 41; sopir, si quid, or siquis, VIb. 54. If Pir cannot be quis, may it not be contracted from Pisher? 12 6. Oretu=(ad)oleto, A.K. Urito is equally near. It recurs only IV. 30, and there seems to mean "fumigate." Our sacrificial fire is not yet lighted. The punctuation is not quite certain. 7. Persclo, ordo, in widest sense; from Persc, ordina, IIb. 32. Here, ordine, "in due course;" so VIb. 16, 36: elsewhere, Persclom, ritum, ceremoniam. 13. Cletra, KλThp? Ainλat seems to verify the sense but see whether IV. 24 opposes.-B. Aitota, "arrange"? See on Ib. 29. Does this imply Cletraf, pl. ? 14. Am-paritu, ap-parato? àvíoтalı? (Am=ava). In IIa. 42, Am-pari-hmu, perhaps avioral: but we have no test of these interprs. See IIa. 25 on Pur. 15. Ententu, by indn. incendito. Cmp. Anglo-S. tendan, (Germ. zünden, Engl. tinder), Gael. teine, and Welsh tan, fire. Ententu, Antentu from different roots are a paradox; but not worse than Discover and Recover; not so bad as Aperire, Deperire, Reperire, Experiri from four roots.-B. Cih-çera, by cnx. cremia: by analysis, crema-cula, See ceh in 21. 16-20. Antentu=intendito, in form; but by indn. imponito, as A.K. well render it. Anava, on and re; never I think in (intra). Thus Anstintu is, primâ facie, ava-stinato, fasten on, or above. Add Seplo, simplus, singulus, Ahesnés, ahenis; and you see cauldrons supported over the fire by frames of three sorts. Each of three cauldrons has its own Cazi. Lat. ferculum = feretrum; primâ facie, these explain Feraclo, |