single one, the real and sole difference between both words consists in the one having a u or upsilon, and in the other having an e or epsilon; but in meaning the difference is this, annulus means "a ring," and annellus, a "LITTLE ring." Hence, when we admit all the letters in one of these words to be equal to all those of the other, with the exception of a single letter on either side (u and e), we are, if we find a difference in meaning between both words, to look to these two letters alone for that difference; for if both of them were, whilst meaning the round thing worn upon one's finger, written exactly alike, there could not exist any difference between them as to meaning, no more than there does between the words ring and ring in English, when taken in the same sense, that is, when meaning the ornament which females, and female gentlemen, are accustomed to wear on their fingers. Hence when we say that annellus is the diminutive of annulus, our words mean, when logically examined, that e is the diminutive of u. But is not epsilon the diminutive of Hta? It is. And the diminutive of epsilon also? And the diminutive of epsilon also. But how can this be? It happens from the two letters Hta and Ursilon being one and the same letter, as I have often had occasion to show. Thus from knowing and u to be radically the same, and that the latter is equal to an o, (as it is composed of its two parts, cc,) we see that = must be equal to an o also.

This knowledge explains at once several apparent

singularities in the orthography of words not hitherto to be accounted for. We now see why e is often pronounced in English like an o or a u. Thus the English word water is pronounced as if written wator or watur; and this is very correct, since the e in this word is equal to an o, which is in its turn equal to a u.

We laugh at ignorant persons for writing water wator or watur, but such is the exact pronunciation of this word; and it is as correct to write it with an o or a u as it is to do so with an e. Now, too, we can tell why the words shew and sew are pronounced as if they were show and sow, and that one of these words (shew) is written as often the one way as it is the other. Hence, when we see two words equal in meaning belonging to two different languages, and only differing from one another by the one having an o where the other has an e, we are to consider this as no difference at all. Thus the English word red, and the Danish word rod, which are equal in meaning, are also equal in letters, since e and o, whether long or short, make but one letter. When accounting for Dies and Dios, I knew very well that these words were radically the same, and that the root of both was Di, which went to prove that God and day made only one word. But I was not then aware that es in the one word, and os in the other, were also the same, namely, as os, another name for the sun or the Divinity, and consequently for the day. It has been a great loss to me, when analysing words,

not to have discovered sooner that an O and an E make but one letter.

The twenty-fourth and last letter of the Greek alphabet is the great O, which is formed thus, Ω w, and is named omega. In both these characters we have all that signifies what is double. The capital one, , becomes when turned up thus, 7, a Roman U, and this accounts for the latter being in several languages pronounced like a double 0. But though the letter U has the meaning of a double o, that is of a double one, since each of its two parts means one, still it should have never had the sound of a double o. And why so? Because when men wished in the beginning to signify in writing the double o, they could have had but three ways of doing so: the first and most simple of all was to write the double o itself, thus, 00; the second way was to write it thus, UO; and the third way was OU. The oo requires no explanation: every body can conceive that o and o put together ought to be equal to double o. As the character u signified what was double, from its being composed of a double one, men must have often employed UO instead of OO, since this also means double o, or double one. As to this form, OU, it cannot, as it is not the natural order, have been adopted but from the u having fallen behind o, just as words have fallen behind one another. Hence, when we give to the letter U the sound of 00, it does not arise from the U having such a sound, for this it could have never had, no more than i

could have had the sound of a single o; but it arises from this u having been anciently accompanied by an o, which it showed to be double; and afterwards, from the o having been wholly suppressed. Hence, in the two Greek characters 2 w, we have no o, but a Roman U, and a Saxon double u (W), so that the o, which must have one time accompanied them, is now wholly suppressed. I have often had occasion to wonder at the great antiquity of the form of French words; and, with regard to this double o, I have occasion to do so again. An instance cannot be found in this language in which the U represents the 00, this combination being generally indicated thus, ou; and sometimes by a long single ō. The French orthoepist has, of course, ever supposed that in the diphthong ou the u has assisted in giving to these two letters their peculiar sound; but it is not by any means heard in this instance, its use being only to signify that here the o is double, and that in the beginning this syllable must have been written oo, then WO, and afterwards ou; this again having, in other languages, as in Greek, Latin, and English, been often shortened to the single U, from the o having, with time, been dropped. But when, instead of oo, men began to use uo, did this make any difference in the pronunciation? In the beginning it could have made none whatever; but when, with time, men forgot why they had, in such a case, employed u, affected speakers must, in very civilised countries, have begun to sound the u before 0. Thus the Greek and Latin word duo

must have been first pronounced as if written doo, (the o, o, that is, the one, one,) the oo in this word having twice the length of a single o, in order to signify two. Here this English word two falls in my way very happily to verify what I have just stated. We see that this word is even still pronounced as it must have been at the time when written too; that is, before it had the w.

But what difference is there between U and W? None. For when U is analysed, it becomes iv; and from these two letters falling together, thus, w, we have w. Hence, in the two forms of omega, 2, w, we have only one letter, the U. But were it correct to write the word two, twoo? It were. For then the w and oo would stand in apposition to one another, and the analysed form and meaning of this word would be, it-w-oo, "the double thing (namely) oo;" that is, one one. Thus, in the Scottish dialect, this word is written twa, which, analysed, becomes it-w-o-i; and here o i, which has been contracted to d, is equal to oo in twoo. We have a similar instance in the English word twins, which, analysed, makes it-w-in-o (the double thing, one one). Here, in the analysis of the w, I have not put a vowel before it, since it is already iv, as we have just seen. The English

word twine furnishes another instance similar to those already given. It is, when analysed, it-w-i-in, that is "the double thing, one one;" twine or twyne being composed of two threads put together. The word twist has a similar meaning, and it is to be analysed thus, it-w-ist, "the double thing is it," "it is the double thing."