ALEXANDRIA. 367 
supposed reading of AIOKAHTIANON. But 
this introduction of the name of a Roman Empe- 
ror, without an epithet immediately preceding it, 
is unusual ; and when letters are thus to be added 
by conjecture, or in consequence of some imagi- 
nary resemblance, in the indistinct traces of the 
original legend, to the characters which have 
been substituted, every person is at liberty to 
make his own hypothesis ; provided only that a 
reading be produced which shall contain exactly 
the number of letters requisite to fill the vacant 
spaces upon the stone. For example, the per- 
pendicular line of the dotted K, as proposed in 
the paper read to the Society of Antiquaries ^ 
may with equal authority be written N . The wto 
lines of the A may also belong to A. The cross 
bar of the H may be the lower line of h., and the 
T may with equal probability be written p; and 
when this is granted, the reading becomes, evi- 
dently, AIONAAPI ANON. The use of AIOI, as 
an epithet, answering to divvs, so frequently 
bestowed upon Roman Emperors, and especially 
Vi^on Hadrian\ although authorised in this sense 
(3) See Dr. Raine's communication, as above. 
(4) Sic passim. "UtDivusHadrianus in quadam oratione ait," &c. 
(Ulpianu.1, lib. 50, Dig. tit, 15. de Censlbus, 8(c. ^-c.) IMP ' CAESARI • 
DIVI • H ADRI ANI • &c. Donii Inscript. /Intiq. ub Goroi. Claisis terlia, 
iVb.16. See also NosAl,\2,y!i{c. Florent. \1Z\. 
