Little need here be said of strictly historic times, the most characteristic feature of which is perhaps the general use of letters. By means of this most fruitful of human inventions, everything worth preserving was perpetuated, and thus all useful knowledge tended to become accumulative. It is no longer possible to say when or where the miracle was wrought by which the apparently multifarious sounds of fully-developed languages were exhaustively analysed and effectively expressed by a score or so of arbitrary signs. But a comparative study of the various writing-systems in use in different parts of the world has revealed the process by which the transition was gradually brought about from rude pictorial representations of objects to purely phonetical symbols. As is clearly shown by the "winter counts" of the North American aborigines, and by the prehistoric rock carvings in Upper Egypt, the first step was a _pictograph_, the actual figure, say, of a man, standing for a given man, and then for any man or human being. Then this figure, more or less reduced or conventionalised, served to indicate not only the term _man_, but the full sound _man_, as in the word _manifest_, and in the modern rebus. At this stage it becomes a _phonogram_, or _phonoglyph_, which, when further reduced beyond all recognition of its original form, may stand for the syllable _ma_ as in _ma-ny_, without any further reference either to the idea or the sound man. The phonogram has now become the symbol of a monosyllable, which is normally made up of two elements, a consonant and a vowel, as in the Devanagari, and other syllabic systems. Lastly, by dropping the second or vowel element the same symbol, further modified or not, becomes a _letter_ representing the sound _m_, that is, one of the few ultimate elements of articulate speech. A more or less complete set of such characters, thus worn down in form and meaning, will then be available for indicating more or less completely all the phonetic elements of any given language. It will be a true _alphabet_, the wonderful nature of which may be inferred from the fact that only two, or possibly three, such alphabetic systems are known with absolute certainty to have ever been independently evolved by human ingenuity[112]. From the above exposition we see how inevitably the Phoenician parent of nearly all late alphabets expressed at first the consonantal sounds only, so that the vowels or vowel marks are in all cases later developments, as in Hebrew, Syriac, Arabic, Greek, the Italic group, and the Runes.