The objects here described were seen at Kutahia in June, 1924; too often, owing to their situation, no photograph could be taken by us, and in such cases we have tried as best we could to represent them by measured drawings; the interest and variety of their designs makes it desirable that before they disappear they should all be adequately photographed. For the notes on nos. 128, 161–168 Buckler is responsible, for those on nos. 125, 150–155 Calder, and for the remainder Cox; but we are all three indebted to each other and especially to Dr. Josef Keil, who placed the resources both of the Austrian Archaeological Institute and of his own learning at the disposal of Cox during the winters of 1924–1925 and 1925–1926.

In the course of last summer the York Roman Excavations Committee invited me to direct some excavations at the east corner of the Roman fortress as a preliminary to more extensive work in the future. It was supposed that the remains of a bastion—similar to the Multangular Tower in the Museum Gardens—might be found under the mound upon which the city wall is built. Before exploring that possibility, however, we decided (1) to see what evidence would be given by a section through the north-east rampart close to the east corner; (2) taking advantage of the fact that a yard off Bedern was available for excavation, to supplement our first section by cutting a trench across the south-east defences where they have parted company with the later mound, city wall and moat, and where, therefore, one might hope to get a profile of the Roman ditch; and (3), guided by the results so obtained, to examine the east corner for traces of the rounded turn and internal angle-tower of the pre-bastion type of fortification. It was after those evidences had been secured that we proposed, if there was still time, to trench outside the corner and prove (or disprove) the existence of the supposed bastion.

The name Central Etruria is applied in this article to the region which is, roughly speaking, bounded by the river Mignone on the south, by the Ciminian Hills and the Plain of Viterbo on the east, by the Lake of Bolsena and the River Fiora on the north, and by the Roman Maremma on the west (fig. 1). Even now it is sparsely populated and covered with scrub; and the peculiar conformation of the ground has from the earliest times imposed on its scattered inhabitants special conditions of choice both for the position of their cities and for the type of their sepulchres.

The present aspect of this vast region is very strange. It may be compared to the Roman Campagna, but to a Campagna the monotonous desolation of which is relieved by frequent ravines, carving the landscape into fantastic and romantic pictures.

In 1921 Mr. Léon Homo published an elaborate study in the Revue Historique on ‘Les privilèges administratifs du Sénat romain sous l'Empire et leur disparition graduelle au cours du IIIe siècle. I have not seen any critical consideration of that study: in the following note I desire to discuss the conclusions of Mr. Homo so far as they are concerned with the edict of Gallienus and its application down to the military reorganisation of Diocletian. Mr. Homo's treatment of the problem is based upon his view of the historical value of the Historia Augusta: for a detailed statement of that view the student should consult his earlier paper on ‘La grande crise de l'an 238 ap. J.-C. et le problème de l'Histoire Auguste.’ His position he has summarised thus: the Historia is not a vast falsification dating from the end of the fourth or the beginning of the fifth century: ‘Non. Il s'agit bien d'un recueil authentique de biographies rédigé sous la dynastie dioclétiano-constantinienne, et, par consequent, l'historien du IIIe siècle n'a pas le droit de la rejeter systématiquement. Mais deux réserves capitales sont nécessaires. Tout d'abord, l'Histoire Auguste abonde en anachronismes, dus à l'information médiocre et au manque d'esprit critique de ses auteurs.

In attempting to estimate the rate of the Imperial Post in the first two centuries, two points must be borne in mind. The first of these is that the object chiefly aimed at was not speed, but the certainty of arrival within a reasonable and calculable time. It was as an instrument of government, and a means of centralising the administration of the Empire, that Augustus instituted the Imperial Post. This is clear enough from the whole scheme of his administrative system, and is accepted by ancient writers generally, though expressed most clearly by Procopius, in the period when the Byzantine government had demonstrated the full use which might be made of the institution. ‘The earlier Emperors, in order to obtain information as quickly as possible regarding the movements of the enemy in any quarter, sedition or unforeseen accidents in individual cities, and the actions of the governors or other persons in all parts of the Empire, and also in order that the annual tributes might be sent up without danger or delay, had established a rapid service of public couriers throughout their dominions according to the following system. As a day's journey for an active man they fixed eight ‘stages,’ or sometimes fewer, but as a general rule not less than five. In every stage there were forty horses and a number of grooms in proportion. The couriers appointed for the work, by making use of relays of excellent horses, when engaged in the duties I have mentioned, often covered in a single day, by this means, as great a distance as they would otherwise have covered in ten.’

In the absence of a full and satisfactory record in literature of the reign of Hadrian, we turn with more than ordinary expectation to his large and varied coinage. We are at once confronted with a formidable difficulty: most of Hadrian's coins bear no other sign of date than COS · III—that is to say, they can only be dated to the long period from A.D. 119 to 138. This problem, then, must be faced at the outset. To save space, we will at once set out the system of dating we shall follow, adding only the few notes absolutely required to justify it:

A.D. 117. Coins dated COS.

A.D. 118. Coins dated COS · II

A.D. 119–138. Coins dated COS · III

The drawings (plates XII–XVIII) which accompany this article were prepared during my tenure of the Rome Scholarship at the British School at Rome; and, while they may be sufficient for the architect or the layman to obtain a better realisation of the original state of the monument, it has been thought that an attempt should be made to discuss, in greater detail than it is possible to present on the drawings, the various considerations and data upon which an essay in the restoration of an ancient building must necessarily be based. To this end the following article has been prepared.

I wish to acknowledge, with many thanks, the valuable help and advice which I have received from Dr. Thomas Ashby, late Director of the British School at Rome, especially in connexion with the first section of the article, and also the suggestive criticisms which I received from Dr. Ashby and Prof. Ch. Hülsen during the preparation of my drawings.

(1) In connexion with the two Antonine types of Britannia with the legionary standard, described in my paper on ‘Britannia on Roman Coins of the Second Century A.D.’ (J.R.S. xiv, p. 149 and note 2, plate xxiv, 5, 6), it should have been mentioned that there are two other provinces, besides Britannia, which are represented with this attribute instead of the usual vexillum. Both the Dacia of Hadrian (Cohen, Les monnaies de l'empire romain, 2, 11, pp. 150, 151, nos. 526–533) and her Trajanic predecessor (ibid. p. 31, nos. 125–127) hold a legionary standard, the former in her right hand, the latter in her left, and the same attribute appears in the Dacia Felix of Traianus Decius (Cohen, 2, V, p. 188, nos. 30–36). Secondly, Traianus Decius' Pannoniae types (ibid. pp. 193–4, nos. 79–89) have the legionary standard, contrasting in this respect with the Pannonia type of Aelius Verus, where the Province holds a vexillum (ibid. pp. 260–1, nos. 24–33). It is possible that the representation of the legionary standard, as opposed to the vexillum, in the case of these three provinces—Britannia, Dacia and Pannonia—may have had some special significance.

In the article referred to above Sir William Ramsay has published with introductory remarks, but without photographs or measurements, several inscriptions from Pisidian Antioch which he copied in 1914, 1924 and 1925. Of most of these I also independently made copies, squeezes, and photographs in the summer of 1924. There are various points in regard to which I am able to supplement or to correct Sir William's account, as follows.

It is usually held that Hippo Regius, after successfully resisting a siege by the Vandals in 430–1 A.D., was abandoned by its inhabitants and burned by the invaders, and that it was afterwards reoccupied by the Romans. There are, however, certain difficulties involved in the acceptance of this view, which is based on the following passage in the Life of St. Augustine by Possidius: ‘Of the innumerable churches he [St. Augustine] saw only three survive, namely those of Carthage, Hippo and Cirta, which by God's favor were not demolished. These cities too still stand, protected by human and divine aid, although after Augustine's death the city of Hippo, abandoned by its inhabitants, was burned by the enemy.’

In The Journal of Roman Studies (vol. xiv, pp. 257–8), Mr. E. C. Howard published ‘A Note on the Vandal Occupation of Hippo Regius’, in which he suggests that ‘Hippo,’ in the passage quoted above, refers not to Hippo Regius, now Bona, but to Hippo Diarrhytus (or Zarytus), now Bizerta. That his suggestion is impossible will be shown below. But first, in order to facilitate an understanding of the whole question, it will be necessary to review briefly certain portions of the history of the period, to state the problem, and to outline Mr. Howard's arguments.