ALPHABET 611 had done. The fact that X, found in the Latin and Faliscan alphabets, has the value of x, and not of ch, and that V as already mentioned, is found with the value of ch in Etruscan, shows that the common source of these five alphabets was a western, not an eastern Greek alphabet ; and the rounded form of C, and the peculiar L (V, not A) limit the choice to the Chalcidian family. The points in which the Latin differs from the Chalcidian alphabet of Cumje, from which it was probably derived through com mercial intercourse, lie (1.) In the application of the symbol van (F), to denote not the v but the / sound, which was probably strange to the Greeks. (2,) In allowing K to fall almost out of use it was employed only in abbreviations, such as the first letter of a prcenomen, as Kseso, or for Kalendse, &c. and employing C instead, which had of course in the present Greek alpha bet the power of g. This change may point to a time when the distinction of the sounds k and g was obliterated, to be afterwards restored. (3.) In the formation of the new symbol G i.e., C with a distinguishing line to mark the soft gutturals, when the want of a distinctive symbol was again felt. This was some time in the 3d century B.C.; but instead of replacing K for the hard guttural sound, they preferred to leave C in its old place, but with a new value, k instead of fj ; while the modified form G was inserted into the place of "T" (Z), which may have been taken by the Romans (as it certainly was found in the other Italian alphabets), but which fell out of use absolutely without any record. (4. ) In absence of the aspirates V, , and : these sounds were not natural to the Roman tongue, and there fore the symbols were never regularly received into their alphabet, though they were taken to represent numerals. Their forms, however, were much altered, and so in process of time they became confused with other letters : thus / denoted 50; but it came to be written _]_, and so naturally passed into the quite meaningless L : denoted 10, but being too cumbrous to write, the circle was dropped, and the cross (X) alone remained. A variant form of the same letter (0) seems to have originally represented 100, and either to have been shortened into the common form C, or C superseded it as being the first letter of centum. was taken for 1000, but for convenience of writing it was broken up into CIO, and this was the more easily done because the parts were characters in use; but this symbol also was replaced by M, the first letter of mille. It is probable that was simply divided, and the half of it (D) then stood for half of 1000, or 500; and half of X, ten, became V, five. Neither D nor V have any other propriety as symbols. (5.) In the addition, in the 1st century B.C., of the two symbols Y and Z after X (which had long been the last letter of the alphabet), to express the Greek sounds u and Z. In borrowed words these in earlier times had been roughly denoted by u and ss; but in Cicero s day greater precision was desired; and not being able to compound two characters of their own to denote the strange sound (as they did for the aspirates M, th, ph, formerly denoted only by k, t, and p or 6), they took sound and symbol together, so that &pvye<s appeared, not as Bruges, but as Phryges : Tpa-Tre^mys ceased to be tarpessita, and sona be came zona. f &c. The Latin alphabet agrees with the Chalcidian in the retention of koppa ( 9 ) ; the downward stroke became by degrees more oblique. This symbol had a much wider use in Latin than it had in any Greek language : it was needed to express a modified A:-sound which the Latins liked, wherein a slight w sound was heard after the L This sound was distasteful to the Greeks, and consequently they changed this kw (or qu) into p; so also did the other Italians (compare equos, LTTTTOS, JSpona, &c.); but the Romans liked it. and therefore, alone in Italy, kept the ? to denote it. It is true that the Q was generally followed by a written u, though not always in the older inscriptions ; but it was fully recognised that this u was not a real letter. It was only a symbol expressing further, and somewhat unnecessarily, the indistinct after-sound which made Q different from K ; it would have been more logical to have written Q alone, as was actually attempted under the empire, where we find on inscriptions forms such as qis, qidem, qaerella; but this never became general. The Latin and Chalcidian alphabets are again at one in not having the symbol M for s, differing in this respect from the alphabets of South Italy, and also from the Etruscan and Umbrian, which had both forms. Lastly, the Chal cidian (as we saw) had two forms for r, P and R; of these the Latin chose the last, and generally employed the first for p; though for that letter the genuine Greek form P also appears rarely. The Romans did not retain the Greek names for the characters of the alphabet. The vowels were known by their sounds only. The momentary sounds and h were denoted by their own sound followed by a vowel, as le, ce, de, ge, pe, and tc, but ka, ha ; q, as we saw, had sufficient vowel sound to float it ; on the other hand, the continuous consonants were preceded by the vowel, as ef, el, em, en, er, es ; x was called ix. The difference in the names of the consonants obviously was caused by their nature : momentary sounds are produced by a complete closure and opening of the organs required in each case ; when this opening is made, the organs are so placed as to form a vowel, which naturally is produced by the remnant of sound required for the consonant ; whereas a vowel cannot be produced before any one of these sounds without conscious effort : hence it was simpler to call k, ka, than to call it ok. But the continuous sounds are pronounced when the necessary organs only approximate more or less closely to each other; the channel through which the sound passes from, the larynx to the lips is never closed altogether, arid by reason of this slightly open position a certain amount of vowel sound tends to escape-just as the organs are drawing together to produce the consonant, and thus is heard before it ; but to sound a vowel after one of these consonants the organs must be intentionally put into the proper position. Thus, then, exactly the same principle the conscious or unconscious striving for ease of articula tion produces exactly opposite results in the case of the momentary and the continuous consonants. The same reason caused a different vowel to be employed for li and k from that which is used for the other letters. In sounding a the organs are in nearly the same position as in sounding these two gutturals, only a little more open ; whereas the position of e is more nearly that of all the other consonants. It must of course be remembered that a Roman, if he had wished to speak of his ABC, would not have said, as we do, a-bee-see, but ah-bay-kay. The arrangement of the letters of the alphabet has caused much ingenious speculation.- It has been more than once pointed out (as by Prof. Key, The Alphabet, p. 28) that there are certainly traces of regularity of arrangement. The three soft momentary sounds b, g, d, were placed together; and it is possible that p, k, t (if denoted by Pe, Koph, Tau), may have once been together, and sepa rated by later intrusions ; ?, m, n have an affinity more apparent than real, which was perpetuated by their meaningless designation as " liquids ;" still, the appearance
is sufficient to justify the idea that they may have been
