XIV.] THE ANTONINE ITINERARY. 307
the reign of that emperor. At the same time in its main features
it is hardly later than Constantine's era. Thus Cirta in Numidia,
which at that time became Constantina, and Ostudizum in Thrace
and Antaradus in Phoenicia, which thenceforward were called Nice
and Constantia, here appear under their earlier names j and, what
is still more important, Constantinople is not treated as the
starting-point or terminus of roads in the same way as Rome is,
though this was subsequently the case. In one passage, where the
distances on the route between Sirmium and Nicomedia are being
computed, the great city on the Bosporus, which was necessarily
passed on the way from the one to the other, is not even noticed.
Elsewhere it is introduced under the name of Byzantium, that of
Constantinopolis being added by a kter hand. Mannert, indeed,
maintained, that the date of this document was not earlier than
364 A.D., because Mesopotamia is unnoticed in it, and that country
first ceased to be a Roman province in that year, when it was
ceded by the emperor Jovian to Sapor, king of Persia1- This
omission, however, is equally well explained by supposing that
this part of the Itinerary was either lost, or intentionally removed
after Mesopotamia had passed out of the hands of the Romans.
At the same time, though we may approve the conclusion which
has been stated above as to the approximate date of the bulk of
the work as we now possess it, points are not
..... r i ,- r i L- Notacom-
wantmg which intimate a plurality of authorship pieteiy Homo-
and difference in date of composition in certain
parts. Thus in some of the lists the distinctive
character of the halting-places, according as they were colonies, or
garrisons, or villages, is stated, while in others this is not done.
A similar irregularity is noticeable in respect of the insertion or
omission, at the end of the description of a certain route, of the
total of the number of miles which it contains. Again, the same
name, when it recurs in different places, is apt to be spelt in
different ways. In the account of the roads through Sicily we
meet with the entry, Item a Catina Agrigenium mansionibus nunc
institutis; here the form of expression seems to suggest that it is
a later insertion. Finally, the notice of the route through Thrace
is introduced out of its natural position, being placed between
1 Maonertf Xntrod* to Tabula Pcutingerfa*&i p. 7.
20—a