521 
would be zm prefixed to the verb; but I am unable to say 
what this verbal root would be. It has been out of my power 
to see more than a very small proportion of the bilingual 
tablets in the Museum. It appears from what has been said 
that the z at the end of the Accadian mun, ‘I,’ is, like that 
at the end of in, ‘he,’ a termination peculiar to the Accadian 
language. The radical part of the pronoun is mu. The pas- 
sage of the digamma into m is admitted by all who have 
treated of itin Greek. ‘The Assyrians constantly confounded 
the sounds of w and m; and in Hebrew-the digamma was re- 
presented by © when it did not disappear in ¥, or pass into ». 
Of the last change there are instances that cannot be ques- 
tioned. It has been often remarked that }%> is the Greek 
Foiv-oc, the Latin vin-um, our own ‘wine.’ In like manner, 
mY, ‘a sea,’ was hwam. ‘The old Assyrian form was hwamat; 
see § 14 of my former paper on the Pronouns, where the femi- 
nine form of similar words is noted. As for 8, it is the most 
frequent representative of the digamma. It represents it in 
every case where it is a preformative; and in most cases, if 
not in every case, where it is a radical. The conversion of 
the digamma into is most remarkable in the word for ‘ wa- 
ter. The primitive word was wa, which sound was ex- 
pressed by a character intended to represent falling rain, YY 
More commonly this word was doubled, giving hwa.hwa. In 
Hebrew we have m6 for mahwa, and also may- and mdém-; 
the first digamma being always converted into m; while the 
second was sometimes contracted in the manner already de- 
scribed; sometimes changed into y, and sometimes into m. 
The Hebrew generally expressed this as a plural; and the 
primitive hwa-hwa is, in fact, a plural. The Indo-European 
nations generally adopted the double form for the noun; as 
in the Gothic ahwa, the Latin aqua, &.; while the simple 
form was used for the verb ‘to wet’ and its derivatives—t-w, 
J-Swp, u-dus, to we-t, wa-ter, &e. In amnis the second di- 
gamma is converted into an m. In avon, awen, &c., we have 
