318 forms as those "most in use." Of course it is at once evident that
this alphabet cannot represent elementary sounds. The Assyrian
characters do, in fact, correspond, not to letters, according to our
notion of letters, but to syllables. These syllables are either mere
vowel sounds, such as we represent by our vowels and diphthongs, or such
sounds accompanied by one or two consonants.

The vowels are not very numerous. The Assyrians recognize three only as
fundamental--_a, i_, and _u_. Besides these they have the diphthongs
_ai_, nearly equivalent to _e_, and _au_, nearly equivalent to _o_. The
vowels _i_ and _u_ have also the powers, respectively, of _y_ and _v_.

[Illustration: Partial PAGE 172]

From these sounds, combined with the simple vowels, comes the Assyrian
syllabarium, to which, and not to the consonants themselves, the
characters were assigned. In the first place, each consonant being
capable of two combinations with each simple vowel, could give birth
naturally to six simple syllables, each of which would be in the
Assyrian system represented by a character. Six characters, for
instance, entirely different from one another, represented _pa, pi, pu,
ap, ip, up_; six others, _ka, ki, ke, ak, ik, uk_; six others again,
_ta, ti, tu, at, it, ut_.

If this rule were carried out in every case, the sixteen consonant
sounds would, it is evident, produce ninety-six characters. The actual
number, however, formed in this way, is only seventy-five. Since these
are seven of the consonants which only combine with the vowels in one
way. Thus we have _ba, bi, bu_, but not _ab, ib, ub; ga, qi, gu_, but
not _ay, iq,ug_; and so on. The sounds regarded as capable of only one