Z the twenty-sixth letter of the English alphabet and
the last, although till recent times the alphabets used
by children terminated not with z but with &, or 🙵.
For & the English name is ampersand, i.e. “and per se
and,” though the Scottish name epershand, i.e. “Et, per se and,” is
more logical and also more clearly shows its origin to be the Latin
et, of which it is but the manuscript form. To the following of z
by & George Eliot refers when she makes Jacob Storey say, “He
thought it (z) had only been put to finish off th' alphabet like,
though ampusand would ha' done as well, for what he could
see.” Z is put at the end of the alphabet because it occupied
that position in the Latin alphabet. In early Latin the sound
represented by z passed into r, and consequently the symbol
became useless. It was therefore removed from the alphabet
and G (q.v.) put in its place.
In the 1st century B.C. it was,
like y, introduced again at the end, in order to represent more
precisely than was before possible the value of the Greek Ζ,
which had been previously spelt with s at the beginning and ss
in the middle of words: sona = ζώνη, “belt”; tarpessita =
τραπεζίτης, “banker.” The Greek form was a close copy of the
Phoenician symbol 
, and the Greek inscriptional form remained
in this shape throughout. The name of the Semitic symbol was
Zayin, but this name, for some unknown reason, was not adopted
by the Greeks, who called it Zeta. Whether, as seems most likely,
Zeta was the name of one of the other Semitic sibilants Zade
(Tzaddi) transferred to this by mistake, or whether the name is a
new one, made in imitation of Eta (η) and Theta (θ), is disputed.
The pronunciation of the Semitic letter was the voiced s, like
the ordinary use of z in English, as in zodiac, raze. It is
probable that in Greek there was a considerable variety of
pronunciation from dialect to dialect. In the earlier Greek of
Athens, North-west Greece and Lesbos the pronunciation seems
to have been zd, in Attic from the 4th century B.C. onwards it
seems to have been only a voiced s, and this also was probably
the pronunciation of the dialect from which Latin borrowed
its Greek words. In other dialects, as Elean and Cretan, the
symbol was apparently used for sounds resembling the English
voiced and unvoiced th (ð, þ). In the common dialect (κοινή)
which succeeded the older dialects, ζ became a voiced s, as
it remains in modern Greek. In Vulgar Latin the Greek Ζ
seems to have been pronounced as dy and later y; di being
found for z in words like baptidiare for baptizare, “baptize,”
while conversely z appears for di in forms like zaconus, zabulus,
for diaconus, “deacon,” diabulus, “devil.” Z also is often
written for the consonantal I (J) as in zunior for iunior, “younger”
(see Grandgent, Introduction to Vulgar Latin, §§ 272, 339).
Besides this, however, there was a more cultured pronunciation
of z as dz, which passed through French into Middle English.
Early English had used s alone for both the unvoiced and the
voiced sibilant; the Latin sound imported through French was
new and was not written with z but with g or i. The successive
changes can be well seen in the double forms from the same
original, jealous and zealous. Both of these come from a late
Latin zelosus, derived from the imported Greek ζῆλος. Much
the earlier form is jealous; its initial sound is the dz which in
later French is changed to z (voiced s). It is written gelows or
ielous by Wycliffe and his contemporaries, the form with i is
the ancestor of the modern form. The later word zealous was
borrowed after the French dz had become z. At the end of
words this z was pronounced ts as in the English assets, which
comes from a late Latin ad satis through an early French asez,
“enough.” With z also is frequently written zh, the voiced
form of sh, in azure, seizure. But it appears even more
frequently as s before u, and as si or ti before other vowels in
measure, decision, transition, &c., or in foreign words as g, as in
rouge. For the ȝ representing g and y in Scottish proper names
see under Y.
(P. Gi.)
