s time, the oldest road out
of Praeneste is that which is now the Via della Marcigliana, along which
were found the very early tombs. It is to be noted that these tombs
begin beyond the church of S. Rocco, which is a long distance below the
town. This distance however makes it certain that between S. Rocco and
the city, excavation will bring to light other and yet older tombs along
the road which leads up toward "l'antica porta S. Martino chiusa," and
also in all probability rows of graves will be found along the present
road to Cave. But the tombs give us the direction at least of the old
road.[76]

There is yet another old road which was lately discovered. It is about
three hundred yards below the city and near the road that cuts through
from Porta del Sole to the church of Madonna dell'Aquila.[77] This road
is made of polygonal stones of the limestone of the mountain, and hence
is older than any of the lava roads. It runs nearly parallel with the
Via degli Arconi, and takes a direction which would strike the Via
Praenestina where it crosses the Via Praenestina Nuova which runs past
Zagarolo. That is, the most ancient piece of road we have leads up to
the southeast corner of the town, but the oldest tombs point to a road
the direction of which was toward the southwest corner. However, all the
roads lead toward the southeast corner, where the old grade began that
went up above the arches, mentioned above, to a middle gate of the city.

The gate S. Francesco also is proved to be ancient because of the old
road that led from it. This road is identified by a deposit of ex voto
terracottas which were found at the edge of the road in a hole hollowed
out in the rocks.[78]

The two roads which were traveled the most were the ones that led toward
Rome. This is shown by the tombs on both sides of them,[79] and by the
discovery of a deposit of a great quantity of ex voto terracottas in
the angle between the two.[80]

THE WATER SUPPLY OF PRAENESTE.

In very early times there was a spring near the top of Monte Glicestro.
This is shown by a glance back at plate III, which indicates the
depression or cut in the hill, which from its shape and depth is clearly
not altogether natural and attributable to the effects of rain, but is
certainly the effect of a spring, the further and positive proof of the
existence of which is shown by the unnecessarily low dip made by the
wall of the citadel purposely to inclose the head of this de