and on the Babylonian Lapidary Characters. 237 
sculptor for the syllable ma ; and I read the syllable which the Major expresses 
by nan as n.na, or simply na. 
From what has been already said on the root druj, it appears that No. 4 is 
the secondary form of No. 1, and not of No. 12. Its value is j(7), not zh(¢). 
These letters, being etymologically connected with g, No. 25, and kh, No. 27, 
and being transcribed by a Babylonian character, which certainly contains the 
consonant & or g, must certainly be 7. Neither z or zh are admissible for 
them. 
No. 11 is the secondary form of No. 3, m, not of No. 27, kh. It is m(w), 
not kh(w). ‘This appears fully from the inscription, in which Mudrdya, the 
country whose name begins with this character, is clearly identified with Egypt. 
This was the value which I myself had originally assigned to No. 11, but I was 
led astray by Westergaard’s positive statement, that the Median character, No. 28, 
which begins the corresponding word, was ag or k. I now see that he had 
no ground for that statement save Lassen’s hypothesis, that the country in ques- 
tion was not Egypt but Kurdistan. The value of that Median character will, of 
course, have to be corrected. 
No. 19 is used at Bisitun without an r after it, in a word which Major Raw- 
linson writes kufa. According to my principle, the first vowel should be gunaed. 
On the authority of this word, I must make No. 19 a primary letter, having the 
same relation to p and b as No. 27, kh, hasto kand g. Accordingly, I write it 
ph, making the above word képha. I still think, however, that p cannot precede 
r without an intervening a, and that if it should do so by analogy, it would become 
ph. Thus, parasdmi, “1 punish,” has for its participle phrasta, ‘* punished,” 
whence uphrastam, ‘well punished,” in IV. 38. The word which the Major 
reads pritd, and translates salvete, as if from a root pri, corresponding to the 
Sanscrit one, I take to be paritd, compounded of the preposition para and the 
root 2, eo; and I translate it “go forth.’ Now, if7 be so decided an aspirate 
that the Tenuwis p cannot precede it, as appears to be the case; and if, as also 
appears to be the case, it is never preceded by & (to avoid which conjunction of 
letters the Persians used kz for the Sanscrit k77), we may safely infer that No. 34 
was thr, not tr. The declension of pitd, which makes-in the genitive pithra 
with this letter, is strictly analogous to the change in the participle of parasdmi. 
Here, then, I restore Lassen’s value, and I think we may now safely regard this 
