1801] 
427 
Literary Intelligence■ 
«rV = thousands 4 — while the tens and units have separate figures. 
Now to apply this to our inscriptions from Mathura, Manikyala and 
Kabul. The Mathura dates give XH ^ ^ and X • Bet us con- 
sider ^ as equivalent to the Arian letter ^ = h for hat = sat = 
100, then the first character x may be = = 4 and the date 
would be 4 hundreds plus 31 in the first case or 431, and 401 in the 
second case, by adopting Thomas’s Q for 30 — which I doubt. 
The figure 4 is represented indifferently by ch , or by chh — as ^ or 
— . In the Manikyala inscription the date is ^ which 
might be read as “ hundreds 4, plus 4, or 404. It is no matter which 
way the date is read—as by reading from the left it would be 4 plus 
4 hundreds. The Wardak date Q 3 y would be hundreds 3 plus 3 = 
303 which if of the Seleucidan era would be = 9 B.C. The day of 
the month, however, seems to include the same cipher r j. If this 
is the same character my new reading falls to the ground at once— 
but it is possible to read /J Jfc ^ = vrihiya 4. 
The whole subject is full of difficulty. In the Mathura dates it 
would be better perhaps to take the sloping character Q which agrees 
with the Kabul and Manikyala forms as the index for hundreds, but 
then the date would be >£0 in hundreds. 
One thing is certain =. in the Western Cave inscriptions, the units 
and tens are represented by independent cyphers — the hundreds 
and thousands by the unit cyphers with indices. Now as the Kabul 
and Mathura inscriptions are of about the same period, we ought to 
expect to find the same system of notation employed in them. 
I have a suspicion that the two Mathura dates of ’<X O ? 
and X ^ are the same, the two middle characters of the first being 
new exponents — must be an unit as it is used to number the 
day of the month. It is the figure 1 of the Satrap inscriptions of 
the Western Caves. If we might read xiyx as 4 ha 1, that 
is 4 hat aha anka 1 , = 4 hundreds + units 1 = 401 anka being 
taken for unit. The figure 7 is represented by in the Cave inscrip¬ 
tions. Thomas’s o> for 30 is a mistake, which he has adopted from 
Stevenson. His 1 for 30 may be correct—and if so, the Mathura 
