ther with a considerable portion of a massive wall
of Vespasian's time. After a restoration by Caracalla the building came to
be called _Templum Sacrae Urbis_. It was first consecrated as a church by
pope Felix IV. (526-530), but he did little more than connect it with the
_Heroon Romuli_ (_ibid._ 5), and build the apse (_ibid._ 4).

[Illustration: Fig. 7. Plan of the Record-House of Vespasian, with the
adjoining structures.]

[Illustration: Fig. 8. Part of the internal wall of the Record-House of
Vespasian.

Reduced from a sketch taken in the 16th century by Pirro Ligorio.]

The whole building was mercilessly mutilated by pope Urban VIII. in 1632;
but fortunately a drawing of the interior had been made by Pirro Ligorio
in the second half of the sixteenth century, when the original treatment
of the walls was practically intact. I give a reduced copy of a small
portion of this drawing (fig. 8). As Lanciani says:

The walls were divided into three horizontal bands by
finely cut cornices. The upper band was occupied by the
windows; the lower was simply lined with marble slabs
covered by the bookcases ... which contained the ...
records ...; the middle one was incrusted with
tarsia-work of the rarest kinds of marble with panels
representing panoplies, the wolf with the infant
founders of Rome, and other allegorical scenes[58].

I explained at the beginning of this chapter that my subject is the care
of books, not books themselves; but, at the point which we have now
reached in regard to Roman libraries, it is necessary to make a few
remarks about their contents. It must be remembered, in the first place,
that those who fitted them up had to deal with rolls (_volumina_),
probably of papyrus, but possibly of parchment; and that a book, as we
understand the word, the Latin equivalent for which was _codex_, did not
come into general use until long after the Christian era. Some points
about these rolls require notice.

The length and the width of the roll depended on the taste or convenience
of the writer[59]. The contents were written in columns, the lines of
which ran parallel to the long dimension[60], and the reader, holding the
roll in both hands, rolled up the part he had finished with his left hand,
and unrolled the unread portion with his right. This way of dealing with
the roll is well shewn in the accompanying illustration (fig. 9) reduced
from a fresco at Pompeii[61]. In most ex