308 ROMAN FRONTIER DEFENCES AND ROADS. [CHAP.

those of Egypt and of Asia. Variations such as these are hardly
reconcilable with the view that the whole Itinerary belongs to one
period ; indeed, a certain amount of accretion is only what we
should expect in the formation of such a document.

The contents of the Itinerary may be thus briefly summarised
First the roads throughout the north of Africa are

Its Contents. . -i-r r /. x/r

given, reaching from the extremity of Mauretania to
Alexandria. Next come those in Corsica, Sardinia, Sicily, and
part of Italy. Then follows the whole route from Rome to Hiera
Sycaminus, on the Nile to the southward of Pselcis, which was the
limit of the Roman empire in that direction : this passed by way
of Pannonia, Moesia, Thrace, Asia Minor, Syria, and Egypt From
Egypt again we are conducted back by the way of Syria, Armenia
Minor, and the Balkan peninsula, after which an excursus is
made into Italy. The concluding portion treats of the lines of
communication in the northern provinces, and westward through
Gaul and Spain, ending with Britain at the limit of the Wall of
Hadrian. The order here assigned to the countries traversed by
these roads is different from that usually found in ancient geo-
graphers, but it will be seen that it has a convenience of its own.
To students of ancient geography the chief value of such an
itinerary consists in its furnishing a more accurate knowledge of
the position of towns, especially in the interior of countries, than
would otherwise be obtainable.

Another and shorter Itinerary, which is usually regarded as
The forming a continuation of the one just mentioned,

is the Itinerarium Maritimum. In the first of the

Manttmvm.
method is pursued as in the Itinerarium Provinciarum, for it
gives the distances of the coast-towns from one another by sea^
and measures of the sea-transits (trajectus) from one country to
another. This is for the most part, though not entirely, confined
to the shores of the Mediterranean. As the distances are here
computed in stadia, it seems probable that it was originally the
work of a Greek, and that the information was drawn from Greek
sources, though it was subsequently modified. The second part,
which appears to be a fragment of an unfinished work, enumerates
in great detail the ports and roadsteads from the Portus Augusti