By courtesy of the owner, I publish a standard weight found in Asia Minor and now in the Collection Breusch. It is a hexagonal plaque of lead, pierced at one angle for suspension by a cord, of maximum width 0·16 m. and weight 2·6 kg. Both faces are inscribed with letters in relief, within a raised border. On face (a) the letters are above and on either side of a relief showing the bust of a curly-headed man facing right and carrying a caduceus over his left shoulder—certainly Hermes Agoraios, patron of the market-place. The inscription on face (b) is upside down to that on face (a).

Face (a). Letters, c. 0·012; (See Plate I.)

The final is badly formed and looks more like

This inscription, dated in A.D. 2/3 and comprising a prose introduction and a set of verses cut in small letters (c. 1 cm.) below a relief of a man reclining on a couch, cup in hand, was found in the Sanctuary of Apollo at Cyrene in 1925 and published by G. Oliverio in 1933. Of the controversies that the monument has aroused, I am concerned only with the last line of the text=the last couplet of verses, where Oliverio's reading is difficult to translate. Various emendations have already been proposed, none of them wholly satisfying.

In considering whether or not this poem is genuine Tibullus we have only internal evidence to go on. Internal evidence is often not conclusive enough. It is difficult to assess, and the assessment usually involves a subjective element. The ideal is to reduce this subjective element to a minimum, to appeal to logic in the way a textual critic does when he chooses one MS reading rather than another or decides to admit an emendation of the text. The textual critic's choice of reading or the truth of his emendation cannot be proved in the way a madiematical theorem can be proved, but none the less we can know that the critic is right or wrong on grounds of probability, which in this field is the only available criterion.

I want now to point to the kind of internal evidence I shall appeal to: it is literary reminiscence, or imitation. Most often no deductions at all about priority can be drawn from parallel passages in different authors; one simply notes them and that is that. But sometimes a deduction about priority can be made. Under what conditions? And precisely how? To answer these questions I shall quote an example from English poetry, because it is particularly instructive to see ‘imitation’ at work in one's own language.

The following article forms the conclusion to my chapter in a projected work on the Temple of Hemithea at Kastabos. The sanctuary is described by Diodorus in Book v, 62—3, and lies about a thousand feet above sea-level on the north-west slopes of the Carian Chersonese, overlooking from the south-east the more easterly isthmus on the long Cnidian Peninsula. It is the first Greek sanctuary of any importance and the first peripteral temple to be explored in the Rhodian Incorporated Peraea (see Fraser and Bean, The Rhodian Peraea and Islands, pp. 123 ff.), of which the whole Chersonese formed part.

Our excavation, primarily salvage-work in remote country after a forest-fire had uncovered the remains, formed one stage in the mapping of ancient cities and demes of this neighbourhood by Professor John Cook and Professor George Bean. It established, from a stamped vase-handle and an inscription on a subsidiary building, that this was indeed, as surmised by Cook, the shrine of Hemithea, a local healing-goddess evidently, as Diodorus shows, of considerable local repute. We found that its greatness virtually began in the late fourth century, and was perhaps of fairly short duration.

The plays of Aeschylus contain many contradictions, and much recent criticism of the dramatic technique of this author has been directed at attempts to extract from these contradictions a unified picture of what the poet really intended us to understand in the case of each particular play. Within the last few years no less than four major articles have been devoted to giving a logically consistent interpretation of the Seven, and if one extends the range to, say, the last thirty years, one can find that all the plays of Aeschylus have been subjected to criticism of a similar kind.

All this may seem entirely praiseworthy; and yet, as we come to read more and more of these books and articles, the sensation may grow upon some of us that we are not so much learning about Aeschylus as witnessing the transactions of a private club: transactions conducted with admirable propriety, but according to a system of rules which has been left unrevised for too long, untouched even after the notorious affair of the Suppliants. Although the members of this club are willing to challenge each others' view-points, they are all agreed that no gentleman would venture to call in question the one great assumption that underlies all their discussions: namely that Aeschylus could not possibly have constructed plays in which such contradictions were deliberately intended; and only seldom can he be allowed to have contradicted himself by oversight.

Why Eunuchs? Primarily because they were important. No-one who has waded through the church histories of the fourth and fifth centuries or the numerous later Byzantine chronicles, or those lives of the saints which touch upon court life, can have failed to be struck by die frequent imputation that, in the Eastern Empire especially, the real power lay in the hands not of the emperor nor of his aristocrats, but of his chief eunuch; or alternatively that the corps of eunuchs as a group wielded considerable if not predominant power at court. Yet the eunuchs were barbarians by birth and slaves into the bargain. The purpose of this paper is to explain why eunuchs held so much power in the imperial and aristocratic society of Eastern Rome, to put this power in the context of the socio-political developments of the later Empire, and to analyse some of the social functions of this power.

Yet here, right at the beginning, the objection might be raised that we are faced with nothing but a problem in historiography. Eunuchs might have been to Byzantine historians nothing more than women and gods were to Herodotus, convenient personal pegs to hang historical causes on.