stitute a series called the vowel series. letters be taken in their order. 1. Alef 3. Gimel 4. Daleth Let the first four of the vowel series. of series P. of series K. of series T. Herein the consonant of series B comes next to the letter of the vowel series; that of series K follows; and, in the last place comes the letter of series D. After this the order changes: daleth being followed by he of the vowel series. 5. He 7. Zayn 8. Keth 9. Teth of the vowel series. of series P. of series K. of series T. In this second sequence the relative positions of v, kh, and ť are the same in respect to each other, and the same in respect to the vowel series. The sequence itself is broken by the letter zayn, but it is remarkable that the principle of the sequence is the same. Series P follows the vowel, and series Tis farthest from it. After this the system becomes but fragmentary. Still, even now, pe, of series P, follows ayn; tau, of series D, is farthest from it; and koph, of series K, is intermediate. I am satisfied that we have in the Hebrew alphabet, and in all alphabets derived from it (consequently in the English), if not a system, the rudiments of a system, and that the system is of the sort indicated above; in other words, that the order of the alphabet is a circulating order. The earliest notification of this natural character, so disguised as to appear artificial, is to be found in the following extract from a work so often quoted already-Key's "Alphabet :" "But we are digressing too long from the question about the principle which governed the first arrangement of the Hebrew or old Greek alphabet, if principle there be. Though we cannot satisfactorily account for the whole order throughout the twenty-two letters, there are certainly traces of some regularity in the arrangement. We find first the simplest of the vowel sounds followed by the three medials, B, y, ; then another vowel, followed, with some irregularity indeed, by aspirates corresponding in order to the above consonants, vau, cheth, theth, no bad representatives of p, x, 0. Then again we have a vowel, followed soon after by three consonants related to each other, λ, μ, v. Soon after we find a fourth vowel o, and after it, with a little interruption it must be allowed, pi, koppa, tau. It cannot well be a mere accident that the several classes of labials, palatals, and dentals occur so nearly together in the different parts of the series, and always in the same order." We may now follow our alphabet into Italy. § 315. The Early Italian and Etruscan Alphabets.-This was taken from either the Phoenician itself, or the Phoenicoeid Greek. We know this from the presence of the letters F and Q-the former from Vau, the latter from Koppa. What more took place? Just what did in Greece. Where a letter was not wanted, it was either not taken in or allowed to drop out of the alphabet. Where a modification of its power was necessary, such modification was admitted. In this way the equivalent to Vau took the sound of F, rather than that of V. More changes of the kind will be found in the history of the Etruscan alphabet. In the Etruscan language, unless we are misled by the spelling, there was a remarkable preponderance of the sharp sounds as opposed to the flat, and a corresponding deficiency of the flat sounds as opposed to the sharp. Thus, there were the sounds of p, f, t, and s, in abundance, but not those of b, v, d, and z. Hence the equivalents to beth, daleth, and zayn, are wanting in the Etruscan; whilst that of vau is found with an altered power. An alphabet derived from the Etruscan would be poorer than one derived from the Greek, and if the language to which it applied contained the flat sounds of b, d, and z, the representation of that language in writing would run every chance of being but indifferently good; in other words, it would run every risk of being very bad. If the Etruscans wanted so many of the common elementary sounds, what did they have in their stead? anything or nothing? Was the sound-system absolutely poor, or was it only different from the Greek? It is not necessary to investigate these questions. It is enough to know that the signs for b and d, were wanting in the Etruscan alphabet—absolutely wanting. The sign for Vau was admitted its power being changed. Instead of = v it = ƒ. The effect of this we find in our own language at the present moment. But there was another sound which was also said to be wanting in Etruscan, viz. that of the Greek o-mikron. What was the effect of this? It might be one of two things. a. Either the sign which represented the Hebrew Ayn might be utterly absent from the Etruscan alphabet, or— b. It might be incorporated therein with an altered power. In other words, we might have either a case of ejection or a case of adaptation. I have little doubt but that the case is one of adaptation. If we look to any of the ordinary tables which give us an Etruscan alphabet, we find that, when we get to the place of the Hebrew Ayn, or the Greek Omikron, there is a blank, or hiatus-indicating that no equivalent thereof is to be found in the Etruscan. We note this, and proceed, until we come to Tau (t). Here the Phoenician and the Phonocoeid Greek alphabets end; but not the later Greek nor the Roman, nor yet the Etruscan. After T comes U-just as, in the later Greek, Upsilon follows Tau. But this is not all. Though, place for place, and power for power, this equivalent to the Greek v differs from the Hebrew Ayn, its form is the same. In considering the origin of Y, which he traces to V (which is only a modified U), Professor Key writes: "If we traced the Greek letter Y or still farther back, we should perhaps arrive at the opinion that it grew itself out of a carelesslywritten O. The Hebrew character which corresponds to O, viz. y, already exhibits the opening above, just as the Hebrew does, compared with the Greek . So, too, the English often write a capital O without joining the circle at the top. To these considerations may be added the fact that the Hebrew alphabet, which ended with a T, contains no other equivalent for the Greek Y: and again the Etruscans had but one character V, without any O." Now, I think that the identity of the Etruscan U with the Hebrew Ayn, form for form, is beyond question. If so, it may be doubted whether the current expressions of the differences between the Etruscan and Greek alphabets are strictly accurate. It may be doubted whether we may correctly say that the Etruscans had no equivalent to the Greek Omikron. They may easily have had no such sound. Whether, however, it is equally accurate to say that they had no such letter, is another matter. They certainly had a sign that originated out of the Hebrew Ayn; and this was what the Greeks had also. Hence, if I were writing on the antiquities of the Etruscan alphabet, I would rather, than say that they had no equivalent to the Greek o, write that they had such an equivalent, but that they sounded it somewhat differently (i. e. as u), and wrote it in a modified form, i. e. open at the top, and with an angle at the bottom, instead of making it circular (V as opposed to 0). If so, the Etruscan alphabet ends at t, and has no gap between the equivalents of Samekh (Xi) and Pe (Pi)—at least in its earliest stage. At a later period, when Roman influences may have acted on it, the case was, perhaps, different. of And this leads to the hypothetical explanation of the origin u; which is as follows: a. It was no invention of any Greek, but an original constituent of the early Phoenicoeid alphabet, in which, form for form, and place for place, it represented Ayn. b. But this it did in some peculiar dialect wherein o was sounded as u—a dialect which, pro tanto, agreed with the Etruscan, but which was not Etruscan, nor yet (necessarily) the dialect from which the Etruscans took their alphabet. c. The form of this equivalent of Ayn was angular rather than circular. Meanwhile, d. In another dialect wherein o retained its natural sound, the equivalent to Ayn was, in form, circular rather than angular, and its sound that of the Greek Omikron. Hence, there were (by hypothesis) two Greek dialects that adopted the Phoenician alphabet, one representing, by the equivalent to Ayn, the sound of o rather than that of u, the other that of u rather than o. Call these dialects U and O. In dialect O is it found that a new sign is wanting, a sign for a sound not identical with o, but allied to it; viz. that of u (whether sounded as in Italian or as in French). Such being the case, recourse is had to the allied dialect (U), which supplies the sign that afterwards takes the name Upsilon. Such is the author's hypothesis respecting the evolution of the Greek Upsilon and its equivalents-a hypothesis which arises out of the necessity of accounting for two antagonistic facts, viz. its similarity in form to the Hebrew Ayn, and its dissimilarity of power, place, and name. The importance of the investigation of the origin of the letter itself becomes. apparent when we consider the number of forms that it has given rise to. Out of the Greek Upsilon have grown the English u, v, w, and y; out of the Hebrew Ayn have grown not only u, v, w, and y, but o also. Returning to the more general questions connected with the Etruscan alphabet, we find that (to go no further into the list of its deficiencies) it was poorer than the Greek by the signs for b and d. Wanting these, did it want g (as in gun) also? We should ask this question. As b is to p, as v to f, as d to t, as z to s, so is g to k. With five, then, of the flat sounds being absent, it is by no means unlikely that the sixth was absent also. If so, there should be no Etruscan equivalent to Gimel or Gamma. But such is not the case. There is one. But, perhaps, this was sounded as Kaf or Kappa; just as Vau was sounded, not as v, but as f. There is, however, a sign for Kaf, or Kappa, as well; and this sign engenders a difficulty. It is a fact against the doctrine that the Etruscan equivalent to Gimel or Gamma = k. If it did, what was the use of Kappa? I think that this difficulty is got over by supposing that the Etruscan equivalent to Kappa was no ordinary k, but some guttural variation of it. If so, the original suggestion may stand, viz. that the third letter of the Etruscan alphabet was the sign for k. Still there is a complication. There is a sign for the Greek Khi, and this, unless either a superfluity or a mere Greek interpolation, was a guttural k—such a k as Kappa is supposed to have represented. I think, however, that this is an interpolation. If this be the case, there are no difficulties in the current and |