S. MARIA ANTIQUA.

23

cannot be his, it might well be a reconstruction by Hadrian of an arrange-
ment of such obvious convenience. Whoever may have been the origin-
ator of this extension of the Palace to the Forum, there was a time when
the site was laid out in a very different way. Below the floor of the
Hadrianic hall or vestibule there has come to light a large tank or piscina,
originally, and still in part, paved and lined with marble ; its sides treated
with a series of shallow recesses, alternately rectangular and curved. Set
at a very different angle from the later building, it was allowed to remain,
except where it interfered with the foundation walls of the latter. Its
length is determined by the remains of the flight of steps descending into
the water, which have been discovered between the wall and the first column
on the right of the peristyle, and exactly correspond to that which has
been preserved in the middle of the end near the entrance.1 The springs
for which this district is famous are sufficient to explain the origin of the
piscina, but of its history we are ignorant. It appears to belong to the first
century of the Christian era.
Such being the building which was converted into a church at some
time, as we believe, after the middle of the sixth century, a casual observer
might fancy that it was far from being well adapted for such a purpose. If
the peristyle, as might have been expected, was to form the atrium of the
church, the limited space of the tablinum was all that remained available
for the church proper. Yet if we are to believe that the earliest Christian
places of worship in Rome were private houses, the same conditions must
have occurred in the ecclesia domestica of pre-Constantinian days. It might
even be suggested that the tradition or survival of such arrangements may
have made the conversion of the present edifice more natural than we
should otherwise have thought. But if we are correct in the date which we
have assigned to the foundation of the church, viz. the period of the
Byzantine occupation of Rome, a more important and more probable
consideration presents itself. Superficially, as we have said, we might
fancy the building ill adapted for conversion into a church. To the
Byzantines of that age the very contrary would appear the fact, for its
plan is precisely that of the type of Byzantine churches which are dominated
by the central principle (Fig. 2).2 There is a narthex, there is the central

1 The original outline of the piscina is indicated on the plan by dotted lines.

2 I have to thank Cav. G. T. Rivoira for permission to reproduce the plan of S. Sophia at
Salonica from his Origini della Architettura Lombarda, Fig. 104, p. 70.