The Nipal branch is particularly interesting as showing a gradual change to a typo almost identical with Devanagari. The square Pali of Burmah is probably the nearest to the old Pali of any. The Burman itself being written with a heavy, slightly rounded character, painted with a brush, and a round, semicircular type for writing with a point, which may be taken as the distinctive appearance of most of the transgangetie alphabets. The Cambodian is contorted and intricate, and the Javanese crumpled. The Buddhists arrived hi Pegu at the beginning of the 5th c. A.D. There is a kind of scrawl used in Malabar which can hardly be placed, and in several instances the arrangement of Lenormant has been curtailed or altered. The best authority on Pali is the work of Burnouf and Lassen, corrected by latest essays. From Siam spreads north-east the great family of toned or musical languages, culminating in Chinese.
Buddhism made its way to Ceylon in the 3d c. B.C., and it is yet a question whether the last examples of the Dravidian family are really based on Pali or old Devanagari; they are certainly of that round type which is characteristic of the farther Indian alpha bets. The modern Devanagari deserves notice from its full alphabet, from its being the exponent of a whole horde of dialects, and from its expression of the farthest developed of all the Indo-German languages. It is the only language which still connects (except certain interrupting letters) a Nvhole•sentence, and accents and inwardly changes it as one concrete expression. But the question for all scholars is, Are there any texts what ever, old in date (before the 10th a. A.D.), or, if old in date, without palpable interpo lations, glosses, or subsequent precessions? The oldest Greek inscriptions, of Melos and Them, date from about 620 n.c. The old alphabets. besides this group of 23 letters, divide into another of 26, of two marked types—zEolian, or of Korkyra, and Doric. In many parts of Greece the Ionic, which may possibly show a derivation from secondary or Sidonian Phenician, replaced these older types, and the Athenians, adopting the characters for 8' and 6 aliout 400 n.c., make that variety the representative alphabet of classical Greek. The Italian alphabets have all an evident primitive air, and they differ from the Greek branch in that, like the Asi atic, they do not distinguish v from o. The dates of relics are about—Faliscan, middle of 4th c. B.C., the end, Sabellian the middle, Etruscan the first half of the 5th century. That the Etruscan does not belong with the Italic family is an inference which can hardly yet be considered proved, all this subject being made more difficult by the perfectly evident borrowing of all these types, one from another, and the perplexingway in which, while striving to make up signs not in the original Shemitic, each different dialect sets a different value on the very limited stock of new signs common to all. Everything points to the presence of the iypriot syllabary on the islands long before any alphabet through the Phenician arrived there, but no vestiges of any such alphabet are left; and the resemblances between that syllabary and earlier alphabets are explainable exactly as were those of Phenician and Egyptian. The Lycian alphabet, as we now have it, is very late, of the time of the satraps at time earliest; and it, like all the others, is distinctly Phenician. It may be believed that some 13 to 16 letter alpha
bet once existed; Ilnio universal tradition of antiquity was to that effect; and the incom plete alphabet of early Assyrian points out what the letters and signs would naturally be; but to find such an one we have to come down to the Norse runes, nearly 2,000 years later. The Asiatic systems are direct, but the others were at first reversed, then woven, only the survivals running direct like Greek and Latin. The late history of the two learned alphabets belongs rather to the subject of manuscript. It will be sufficient here to remind the reader that there was a cursive Greek as early as Alexandrian cultivation, and that Latin, already as much disfigured as Phenician in the careless scrawls on the walls of Pompeii, had also a distinct cursive as early as the 2d c. A.D. It is a mistake to speak of an Irish, Saxon, or Norse mediaeval alphabet. They are simply a beautiful variety of the usual early Lombardic letters; but the influence of the Greek scribes it evident in them, and they did not merge into the far uglier, angular, black-letter, The earliest rune futhark (like the word alpha-bet) is probably of about A.D. 300; it and all others show acquaintance with the Latin alphabet; but just as evidently, they are, on close inspection, not derived from that alone. The letters on the bracteate of , Scone]) are 23 (is there one lost ?) in number, an correspond closely enough with the names preserved by Ulfilas in his Meso-Gothic; they are, in fact, just the same as the Markman alphabet, and the old part of the Anglo-Saxon futhark. The old Norse futhark, on the contrary, though it does not appear until the 8th c., is much more primitive, and originally had 15 characters (staves). These were pointed to represent the needed sounds in the beginning of the 13th c. by king Waldemar, and then give way before Christianity and its prejudice in favor of the Latin alphabet While we see the west, or Teutonic system of runes, thus disappearing before an alphabet wholly inadequate to express the sounds of its dialects, the eastern or Slavonic took advantage or the (com paratively) brilliant ingenuity and learning of the Greek church, and, using as a basis a series of runes which unfortunately we do not possess entire, evolved the Moravian, or Cyrillic bukiviza (equals alphabet) in the 9th e., the Glagolitic, or southern Slavic, in the 13th century. The last is of, probably, 31 letters, and their forms are usually traced to a Coptic debasement of hieratic signs which equal their meaning in the original Slay. This is far-fetched. It would suffice that the Dalmatian monk Hieronymus, who invented them, had been a Copt by birth. The Cyrillic, parent of the modern Slav sys tems, shows the influence of the usual rounded Greek Byzantine character. It is neces sary' to say a few words of Russian, because it., with Devanagari and .Persian Neskhi, shows a distinct attempt at a phonetic system. It is the only system that eversucceeded in spelling for Europeans the harsh and convoluted sounds of the Caucasus; it is evident what small additions would perfect the system, and the additions, generated as they might he from the original shapes of the runes, would come in well with the present letters.