In 1889 Hermann Dessau published a revolutionary article in which he suggested that the six authors to whom the various biographies of the Historia Augusta are ascribed were merely masks behind which lurked a single author who lived at the very end of the fourth century. Ninety years later, an attempt was made to prove Dessau's hypothesis of single authorship on objective grounds. Ian Marriott's article, ‘The Authorship of the Historia Augusta: Two Computer Studies’, presents the results of two statistical tests conducted on the text of the HA which appear to show conclusively that the work was composed by one author rather than six. Marriott's findings have been hailed by a number of scholars as settling once and for all the question of the authorship of the HA. While I agree with Marriott's conclusion, and I share the consensus view that this curious work is indeed the product of a single mind, I should like to raise some questions concerning the validity of the procedure adopted by Marriott and to suggest that the tests conducted by him are in fact not conclusive.

The inscription of Iconium published on p. 37, no. 19, gives us our first certain example of an east-Phrygian variety of the early Christian formula against violation of the tomb, which Professor Callander and I had already restored in a small group of inscriptions.

Sound chronology is fundamental in any attempt to trace the development of the Roman imperial administration. This is especially true for those officials whose careers lay in the sub-equestrian ranks of the administration, as the funerary inscriptions, on which most of our detailed knowledge of them depends, are extremely difficult to date. For the Augusti liberti the main dating criterion is the imperial nomen gentilicium in their full nomenclature, which fixes their date of manumission, normally at the age of thirty or soon after, to the period in which the manumitting emperor or emperors reigned; the terminus ad quem for holding a post in the administration is, therefore, not more than forty years after the death of the last emperor bearing a given nomen, and normally a post is not likely to have been held later than about twenty-five years after the emperor's death. In the numerous cases, however, where the nomen of an Imperial freedman is not recorded, but only his cognomen and status-indication are given, the problem is much more difficult. External dating criteria are usually lacking; we are forced to make the best use possible of internal evidence.

The publication of a second fourth-century epitaph from Laodicea Combusta, engraved on the tomb of two of its bishops, Severus and Eugenius, calls for fresh study, from a new angle, of the epitaph of the persecuted bishop Julius Eugenius of the same city, which was first published (with a provisional text) in the Expositor in 1908, and has already become the subject of a considerable literature. An opportunity therefore offers itself to present English readers with a complete and (I hope) final text of the bishop's epitaph. This is all the more necessary that an English version, inaccurate in almost every detail, has been incorporated in a text-book issued by the Cambridge University Press.