Some fifty miles north-east of Jerusalem, on the rugged plateau between the Jordan and the desert to the east, lies a ruined city whose striking remains have attracted the interest of hardy travellers ever since the beginning of the last century—Jerash, the ancient Gerasa, in Transjordania. This district was the Decapolis of the Romans, though its cities numbered more than ten. Since that time war and the centuries have obliterated all trace of many of them; but Jerash, with its temples and theatres, its paved and colonnaded market and streets, its almost intact walls and its many Christian churches, is still beautiful and imposing. None the less, it is only very recently that its full significance has begun to be realised. Heretofore it has been usual to assume that Jerash attained prosperity and importance only under the Antonines, beginning with Trajan's annexation of Arabia in A.D. 105–6, and his assignment of Jerash to that province. This opinion was supported by the fact that the inscriptions supply dates of the second century or later for most of the public buildings; but a reconsideration of the archaeological and epigraphic evidence shows that the present plan of the city and the present form of the public buildings are due merely to alteration and rebuilding under the Antonines. From the materials which we now have it is possible to reconstruct a good picture of the flourishing and active city of the first century—further, to show that this prosperity was itself not new, but firmly based on a long and continuous past.

I. Introduction.—The dating of building materials at Pompeii presents a problem which is in several ways peculiar. Although more than half the city has by now been uncovered and has provided a wealth of epigraphic material almost without parallel among Italian towns, yet the number of buildings which it is possible to date from such evidence can be counted on one hand, since for the most part the inscriptions concern events which are too trivial, or persons who are too obscure, to offer any indication of their date. Furthermore, the procedure adopted in the past for excavating the town has precluded all but the most meagre help being given by pottery. Till a few years ago, Pompeian excavators for the most part stopped digging as soon as the uppermost (and latest) level had been reached, and earlier buildings, of which the foundations presumably survive under the later floors, were left unexplored. Hence, the pottery which is most useful for purposes of dating (that occurring in association with structural foundations) is not forthcoming. Pompeii, in fact, has been too much a show-piece for visitors, too little a site to be scientifically explored.

‘Epistulae quoque ejus ad senatum extant, quas primum videtur ad paginas et formam memorialis libelli convertisse, cum antea consules et duces non nisi transversa charta scriptas mitterent.’ The form that Caesar used in writing his dispatches to the Senate is still a matter of dispute: the object of this note is to suggest that an old interpretation of this passage, to be found implicit in Lewis and Short's Latin Dictionary, s.v. ‘libellus,’ has more in its favour than recent critics have admitted and to claim for the papyrus codex, or rather for a crude precursor of that form, an antiquity reaching back to the middle of the first century B.C. The sentence in question occurs towards the end of the chapter, after the description of the literary works and before that of the private papers; consequently the position of the sentence does little to support Sir F. G. Kenyon's view that ‘the point of the passage must be that Caesar's dispatches to the Senate were in such a form that they could be enumerated among his literary works.’ The obvious meaning of the passage is that, of the contrasted methods described by ‘transversa charta’ and ‘paginas et formam memorialis libelli,’ Caesar was the first to adopt the latter in official correspondence.

In a very persuasive article in the Sitzungsberichte of the Berlin Academy (1931, p. 772 ff.) Wilcken has argued that Augustus in the third section of his Res Gestae enumerated only the expenses that he paid out of his patrimonium ,his privatum, and the manubiae, and that Mommsen, Hirschfeld, and Rostovtzeff were probably in error when they supposed that some of these payments derived from the Fiscus. To Wilcken's argument, which I accept in the main, I wish only to add some calculations of the amounts attributed to the various chests, in order to see how far his theory stands the test of practical application. And before doing this I wish also to call attention to some slight matters in his article that may need closer examination.

In three years' time, a hundred years will have passed since Luigi Rossini issued his series of plates illustrating Roman triumphal arches in Italy, delineations of interest not only for their accuracy and beauty, but also because Rossini made them the basis of the first detailed study of the Triumphal Arch as an artistic object in itself, treated apart from the surroundings which it embellished. All subsequent workers, Curtis, Nilsson, Loewy and Noack, took the same point of view: and the object of this study is to supplement their valuable research by reconstructing the surroundings of a group of arches connected with city gates, and seeing what light is shed by the relationship of the two structures upon the development of the decorative arch. It is not always possible to reconstruct the ancient grouping in full, nor can it be recovered so easily now as a century ago. But it will emerge that there is always something to be discovered which helps to determine the place of a given arch in the environment which it decorated.

In JRS xiii, 1923 (pp. 1–55), Dr. E. Nischer published a paper on ‘The Army Reforms of Diocletian and Constantine,’ in which, in opposition to the opinions of other scholars, including Mommsen, Seeck and Grosse, he attempted to prove that a proper appreciation of the reforms of the two Emperors could only be attained if a sharp distinction were drawn between their respective policies. According to this theory, Diocletian is the augmenter, Constantine the reformer, of the Roman military system, in the sense that the former doubled the number of existing legions, while the latter created the field-army of palatini and comitatenses by disbanding some of the frontier-legions and withdrawing detachments from others to form independent units in his new mobile army. In a note in JRS xv, 1925, Professor Norman Baynes suggested some general grounds for caution in the acceptance of Nischer's opinions, and it is the purpose of this paper to examine more closely the latter's arguments and, while rejecting part of his conclusions, to consider what inferences may safely be drawn from the evidence which we possess for the work of Diocletian and Constantine.

Two sharply contrasted views are held about the duration of Roman government in Britain and in the West generally. The traditional view, which I believe to be substantially correct, dates the abandonment of Britain early in the fifth century. The late Professor Bury put forward in 1920 a theory that the occupation was maintained down to A.D. 438. The theory was based primarily on a later dating of the Notitia Dignitatum, and on a mistaken estimate of the numismatic evidence.

In connexion with the preceding paper, a summary of the numismatic evidence as to the currency of Britain, in its relation to that of the Continent, till the establishment of the Saxon scheme, based on the silver penny, may be found useful. It is derived almost entirely from published material.

1. Greek and Roman coins of the third to first centuries B.C. are found sporadically in England: there is no sign that they were used for the purpose of currency, or related in any way to the British and Gaulish money.

2. Late Republican and Augustan coins occur fairly often, and their types are borrowed by British mints.