The beautiful alto-relievo in the Lateran Museum, Rome, representing an actor selecting a mask, contains a contrivance for reading a roll (fig. 12) which may have been usual in libraries and elsewhere, though I have not met with another instance of it. A vertical support attached to the table on which two masks and a MS. are lying, carries a desk with a rim along its lower edge and one of its sides. The roll is partially opened, the closed portion lying towards the left side of the desk, next the rim. The roll may be supposed to contain the actor's part[91]. It is much to be regretted that we have no definite information as to the way in which the great public libraries built by Augustus were fitted up; but I see no reason for supposing that their fittings differed from those of private libraries. [Illustration: Fig. 12. Desk to support a roll while it is being read.] When books (_codices_), of a shape similar to that with which modern librarians have to deal, had to be accommodated as well as rolls, it is manifest that rectangular spaces not more than a few inches wide would be singularly inconvenient. They were therefore discarded in favour of a press (_armarium_), a piece of furniture which would hold rolls (_volumina_) as well as books (_codices_), and was in fact, as I shall shew, used for both purposes. The word (_armarium_) occurs commonly in Cicero, and other writers of the best period, for a piece of furniture in which valuables of all kinds, and household gear, were stowed away; and Vitruvius[92] uses it for a book-case. A critic, he says, "produced from certain presses an infinite number of rolls." In later Latin writers--that is from the middle of the first century A.D.--no other word, speaking generally, occurs. The jurist Ulpian, who died A.D. 228, in a discussion as to what is comprised under the term _liber_, decides in favour of including all rolls (_volumina_) of whatever material, and then considers the question whether _codices_ come under the same category or not--thereby shewing that in his day both forms of books were in use. Again, when a library (_bibliotheca_) has been bequeathed, it is questioned whether the bequest includes merely the press or presses (_armarium vel armaria_), or the books as well[93]. The Ulpian Library, or rather Libraries, in Trajan's Forum, built about 114 A.D.[94], were fitted up with presses, as we learn from the passage in Vopiscus which I have already quoted; and when the ruins of the section of that library which stood next to the Quirinal Hill were excavated by the French, a very interesting trace of one of these presses was discovered. Nibby, the Roman antiquary, thus describes it: