but the excavations could not
be made complete owing to the ground being used as a cemetery. One
pattern is purely geometrical; another has birds, dogs, hares, baskets
of flowers, and floral scrolls in octagons and squares set diagonally
between them; both marble and vitreous pastes are used, as well as gold
tesserae. Inscriptions were also found in letters of the third or
beginning of the fourth century: "***ore Felix hic crevisti hic Felix"
and "Cyriacus vivas." The former is held to prove that there was a
domestic basilica here at that period. The bottom of the wall was
painted with geometrical patterns imitating marble plating. The mosaic
runs right under the campanile. There is a door to the south, and two
pillars parallel to the face of the wall, and one to the left, opposite
the north angle. The upper building has a double row of bases of
columns, nine or ten in number, with an external wall 19 ft. 6 in. from
the present basilica, and with the western wall of the narthex level
with the present narthex, beneath the piazza. Antique fragments were
used in the foundations. The lower part of the wall of the existing
building is of the same materials and thickness, and probably of the
same date. The much simpler mosaic patterns of the floor are at the same
level both inside and outside--viz. 2 ft. 9 in. below the present
pavement. Near the round building in the north aisle a fish mosaic was
found on which the sarcophagus of Poppo stood. Signs of a
conflagration--fragments of charcoal, &c.--were also found on this
pavement. The colours used in the mosaics are white, blue, grey, palish
green-grey, yellow, brown, black, several blues and reds, and two
greens. The finest fragment has a figure of a peacock with tail
displayed, which was in the narthex in front of the door to the church,
and is now in the museum. On the pavement coins were found, most of
which belonged to the third and fourth centuries; but there were also
one Greek coin of Marcianopolis, two so-called Consular coins, one
Augustan, three of the second century, one Ostrogothic of Ravenna, and
several Aquileian of 1400. In the eighteenth century sarcophagi were
disinterred bearing fourth-century crosses, of an earlier date than
Attila, at all events.

There appears to have been a restoration in the sixth century, probably
under Narses; the use of super-abaci and the caps in the transept
suggest this. Perhaps the council of 557 may have had something to do
wi