204 Rev. Epwarp Hincxs on the Number, Names, and Powers 
language ; and, when it was so, it was unfortunately by Frenchmen that the equi- 
valents of its letters were taken down, who, we know, cannot pronounce ch or J, 
but substitute for them si and zh. It is quite plain that x expressed the soft 
sound corresponding to 6; the former is a softening down of v, as the latter is of 
kK. From the analogy of our own languarge, we should, therefore, expect that 
their values would be 7 and ch, which is what the hard Anglo-Saxon g and c have 
become in a multitude of words. Accordingly, the French report is that x ex- 
pressed the Italian g before 7 (i.e. 7), or the French 7 (i.e. zh), which the re- 
porter could not distinguish. Of & he reported that it had the same sound as wy, 
or sh. It is not an admissible supposition that the powers of these letters were 
identical ; the conclusion, then, is to my mind irresistible, that 6 had the value 
of ch, and of course x that of the English 7. It appears from what has been 
said, that in the third or fourth century, when the Coptic alphabet was formed, 
the Greek « before z, and similar vowels, had undergone the same softening as 
C in the two first syllables of Cicero has undergone among the Italians. It is, 
therefore, by no means improbable that it had undergone this change in the lat- 
ter part of the second century, when Galen wrote ; and that the sound of the 
long serpent indicated by his transcription was ch and not k. 
(2). The long serpent is used as a representative of ¥ in the name of Sidon, 
Tr. III. 5. This Hebrew letter is, I think, always represented in transcriptions 
of foreign words by the present character, or by C 2, never by the purse or by 
any of its known homophones; but there is one instance, that of ¥0, in which 
one of these homophones, the semicircle, which I have shown to be implied in 
fig. 78, has been represented by ¥ in Hebrew. In the Roman age this character 
is used as initial in the name of Titus, on the obelisk of Domitian, in the Piazza 
Navona, at Rome. This is the chief ground for attributing to it the value T ; 
and it certainly appears as decisive an argument for refusing to it the value K, as 
the transcription «idx is for refusing to it the value T. The following considerations 
will, however, show that this transcription is by no means conclusive in favour of 
T, as against CH or TS. In the Gnostic papyrus, at Leyden, attributed to the 
second century after Christ, but perhaps written in the first, 7e occurs only twice, 
and is in both cases represented by the Enchorial characters corresponding to 
t 2,82, and 12, that is, by és’. In vil. 8 waorivé is transcribed by mastsinks ; 
