of the pronunciation. It is therefore the mainstay of the
scholar in investigating or restoring the word-forms of the ancient
language. Greek transcriptions of Egyptian names and words are valuable
as evidence for the vocalization of Egyptian. Such are found from the
6th century B.C. in the inscription of Abu Simbel, from the 5th in
Herodotus, &c., and abound in Ptolemaic and later documents from the
beginning of the 3rd century B.C. onwards. At first sight they may seem
inaccurate, but on closer examination the Graecizing is seen to follow
definite rules, especially in the Ptolemaic period. A few cuneiform
transcriptions, reaching as far back as the XVIIIth Dynasty, give
valuable hints as to how Egyptian was pronounced in the 15th century
B.C. Coptic itself is of course quite inadequate to enable us to restore
Old Egyptian. In it the Old Egyptian verbal forms are mostly replaced by
periphrases; though the strong roots are often preserved entire, the
weaker consonants and the [Hebrew: ts] have largely or entirely
disappeared, so that the language appears as one of biliteral rather
than triliteral roots. Coptic is strongly impregnated with Greek words
adopted late; moreover, a certain number of Semitic loan-words flowed
into Egyptian at all ages, and especially from the 16th century B.C.
onwards, displacing earlier words. It is only by the most careful
scrutiny, or the exercise of the most piercing insight, that the
imperfectly spelled Egyptian has been made to yield up one grammatical
secret after another in the light brought to bear upon it from Coptic.
Demotic grammar ought soon to be thoroughly comprehensible in its forms,
and the study of Late Egyptian should not stand far behind that of
demotic. On the other hand, Middle Egyptian, and still more Old
Egyptian, which is separated from Middle Egyptian by a wide gap, will
perhaps always be to us little more than consonantal skeletons, the
flesh and blood of their vocalization being for the most part
irretrievably lost.[14]

In common with the Semitic languages, the Berber languages of North
Africa, and the Cushite languages of North-East Africa, Egyptian of all
periods possesses grammatical gender, expressing masculine and feminine.
Singularly few language groups have this peculiarity; and our own great
Indo-European group, which possesses it, is distinguished from those
above mentioned by having the neuter gender in addition. The
characteristic triliteral roots of al