The Nature of the Currency. During the last few years several hoards of tetradrachms struck at Alexandria under the Roman emperors, have come into my hands from different sources in Egypt: and the statisfics as to the composition of these hoards which I have been able to collect provide some material of interest as bearing on the activity of the Alexandrian mint at different periods.

It should be premised that, from the time of the conquest of Egypt by Augustus to the monetary reform of Diocletian, the tetradrachm was the most important coin ordinarily circulating in Egypt. It was nominally of silver, but actually very debased: the earliest examples struck under Roman rule, in 21 A.D., contain about 50 percent. of silver; the latest in 296 AD., less than1 per cent. The deterioration in fineness, which proceeded at varying rates through this period, was associated with a diminution in size and weight: the first coins are about 1 inch in diameter and weigh on an average about 180 grains-the last about 3/4 inch and 9o grains. No other silver, or nominal silver, was struck at Alexandria, except for an issue of didrachms under Claudius, which are extremely rare: and no external silver coinage circulated in the country. A certain amount of Roman gold was brought in, but was probably used mainly by the Roman officials: there is no trace of its occurrence in the records of ordinary commercial transactions. Bronze coins of lower denominations than the tetradrachm were issued regularly till about 180 A.D.: after that date the depreciation of the tetradrachm appears to have caused the abandonment of a bronze currency. In consequence of, the importance of the tetradrachm, it was the coin usually hoarded in Egypt: it is rare to find large hoards of bronze, or any considerable admixture of bronze with tetradrachms. And, as the tetradrachms were always dated, bearing the regnal year of the emperor under whom they were issued, these hoards shew, not only the approximate date when they were formed, but also what length of time coins remained in circulation.

The Hoards. The following brief description of the hoards included in the tables, will serve at the same time to indicate certain points which affect their value for statistical purposes. Hoards i, ii, and iv were found in the Fayum by Messrs. Hogarth, Grenfell, and Hunt in 1895-6 the two first at Umm-el-Atl, the other at Kom Ushim. I contributed an account of the coins to the volume on Fayum Towns published by the Graeco-Roman Branch of the Egypt Exploration Fund in 1900 (see pp. 64 ff.). These three hoards are the only ones of which I can say with certainty that they reached me just as they were found, without any loss or addition.

Hoards iii, v, vi, vii, and viii were obtained by Messrs. Currelly and Frost at Tell-el-Maskhut? in 1905. These five are probably “ uncontaminated ”: they were purchased from the native finders, without the intervention of any dealers, and there was every appearance, in the condition of the coins, that those in each lot had been found together. It is possible that any one of the lots may only be a part of an original hoard, since, if a discovery of coins were made, and more than one man was aware of it, the find would probably be divided; but the division would take the form of a haphazard separation of the whole into shares without any selection of individual coins, so that the ratio of the numbers of coins belonging to different years would not be seriously affected in the respective parts as compared with the entire hoard.

Hoard xiv I purchased in Cairo from an Arab dealer: it was originally in two lots. I described this hoard in the Archiv für Papyrusforschung, ii, p. 529, where my reasons for treating the two lots together may be found. So far as I could judge, this hoard was practically “uncontaminated.”

The remaining six hoards have come to me through Signor Dattari of Cairo; and, as they have passed through several hands, and have lost most of their history, I cannot feel at all certain how far the coins which reached me represent what were originally found. In no case, except that of hoard x, was there any internal evidence of confusion: the coins were, in each instance, in generally similar conditions of preservation, and ran in fairly even distribution; but, as the former holders of the hoards may have chosen out some specimens of the rarer types, such diminution of the numbers by selection would decrease the value of the results derived from them for statistical purposes.

Hoard x appeared to have been made up from two distinct lots: the coins of the first and second centuries in it may have belonged to one lot, and in fact, when they reached me, there were many instances in which coins ranging from Claudius to Commodus were corroded together; but I did not find any third-century coins united to any of an earlier date.

In spite of this possibility of “contamination,” the total number of coins of each year found is of considerable interest, as shewing the relative sizes of the issues in different years; though the comparison can only be made, with any degree of fairness, between neighbouring years which are covered by an equal number of hoards.

Variations in Minting. I had hoped that it would be possible to carry the comparison further, and, by determining the rate of wastage, to construct a table shewing the original proportion of the coinage belonging to each year. But, after a careful examination of the statistics with Professor Petrie, I conclude that the present evidence is insufficient for this purpose, though some approximate results may be obtained. The chief difficulties are set forth in the following paragraphs.

It is fairly clear that, during the first two centuries of our era, there was nothing like a uniform rate of coinage of tetradrachms at Alexandria. None were issued from the time of the Roman conquest, in 30 B.c., till 20 A.D.; a few were struck under Tiberius, but mysteriously disappeared from circulation: then, in the second to the sixth years of Claudius, fairly large numbers appeared, and again, after a lapse of ten years, in the third to the sixth years of Nero; two years followed without any silver coinage, and then, after a very small issue in the ninth year of Nero, came an enormous activity of the mint, which died away eight years later in the third year of Vespasian. In the following thirty years there are only seven of which any tetradrachms are known, and the issues of all but two of these years must have been small, as specimens are rare. Thereafter, for seventy years, we find examples dated in every year except two; then comes another gap of ten years, broken by one issue only; then, during the thirteen years of the sole reign of Commodus, there are coins of every year, and in some cases large numbers are found. In the reigns of Septimius Severus and his sons, covering a quarter of a century, tetradrachms occur belonging to most years, but in all instances they are extremely rare, and very few can have been struck. With the second year of Elagabalus fairly large issues begin once more, and thereafter, till the reform of the coinage under Diocletian in 296, there was a mintage of every year, except 251-2 and 252-3, though the number of coins put into circulation must have varied considerably from year to year.

Wastage of Currency. As no uniform rate can be postulated, the alternative way of defining the wastage would be to compare several hoards covering a fairly long period, but ending at different points during the period. The hoards described here, however, do not give very good data for this purpose; and, so far as I can ascertain, the majority of similar hoards found in Egypt share the same characteristics. That is, most of them seem to have been buried within certain very limited periods: the coins found in hoards usually end with the middle of the reign of Marcus Aurelius, the early years of Aurelian, or the end of the issues of Diocletian. These dates mark lines of general disturbance in Egypt-the “ Bucolic war “ and revolt of Avidius Cassius, the Palmyrene invasion, and the usurpation of Domitius - when it would naturally occur that large quantities of treasure would be buried, part of which the owners would never return to recover. But this narrow limitation of the hoarding periods is unfortunate for our present purposes.

It happens that these times of disturbance were almost coincident with times of debasement of the coinage--there may indeed have been some causal connexion between the two. The size and fineness of the Alexandrian tetradrachm persistently diminished under Roman rule; but the most sudden and marked depreciations were at the end of the reign of Marcus Aurelius and in the tenth year of Gallienus, while the issue of tetradrachms ceased entirely during the revolt of Domitius. A man hoarding coins would probably select, from amongst those that came into his hands, the specimens of most intrinsic value, and keep these, passing back into circulation baser pieces which were nominally worth as much. Hence any hoards formed shortly after a sudden depreciation of the coinage would tend to be composed mainly of the older issues, to the exclusion of the new; and thus the composition of such a hoard would not be a fair index of the actual circulation at the time of its formation.

It is, however, possible to obtain from these statistics some evidence as regards wastage during the years from 230 to 280 A.D. Professor Petrie has kindly investigated the six hoards which cover the greater part of this half-century, and finds that the period of half-waste lies between 15 and 22 years, the mean being 18 years. Taking this result, he states the following as the proportions of the original issues left in successive five-year groups:-

But, while this rate of wastage seems reasonable within the period from which it is calculated, it does not suit the figures of the first century. If the wastage of the coinage of Nero had been at this rate, the issues of his later years would appear to have been almost incredibly large: thus, in hoard ii, buried after 165 A.D., out of 4344 coins, 2380 belong to the years 63 to 68 ; and yet, according to the wastage table, only one-fiftieth of the original coinage of these years should have been in circulation when the hoard was buried. That the number of coins of Nero present in this hoard is not an isolated accident, due to some such cause as the man who formed it having come on an earlier hoard, is shewn by the approximately similar proportion of these coins in other hoards covering the same period. Some explanation may be found in the probable tendency, noted above, to reserve the oldest and best coins for a hoard. But it is most likely that the wastage of tetradrachms in the first century was actually less than in the third: for one thing, the earlier coins were larger, which would make them less liable to casual loss than the smaller later ones; also, in the first century there was a considerable amount of bronze currency in circulation, whereas in the third the tetradrachm was the chief medium of exchange, and was the only official coin issued, so that the comparative liability of waste of the later tetradrachms would be much greater. That tetradrachms were actually lost more frequently during the third century than during the first appears from the statistics of the coins found in the excavations of Drs. Grenfell and Hunt at Oxyrhynchus, which I gave in the Numismatic Chronicle (1908, p. 303). These coins all came from the rubbish-mounds of the ancient town, and represent the ordinary losses of daily life; and the summaries for the three centuries are:—

So far, therefore, as wastage is determined by casual loss, the wastage of the third-century tetradrachms would appear to have been far greater than that of the first. There are, of course, many other factors in wastage; but it is fairly clear that the rate discovered for the third century cannot be taken as even approximately applicable for the first; and it remains for further evidence to be adduced in order to shew how fast the first-century tetradrachms wasted.

Irregularities of Hoarding. In the comparison of the issues of different years, the totals of several hoards are a safer guide than any single hoard, as the proportions in any individual case may be affected by accidental circumstances. For instance, in hoard xv, the figures are certainly abnormal at two points. The man who collected this hoard seems to have had a special fondness for the coins issued jointly by Aurelian and Vaballathus, or by Vaballathus alone,- possibly he was a Palmyrene,- and he accumulated 262 of these, thus swelling the proportionate numbers for the first and second years of Aurelian in his hoard far beyond those of any other. Then, in the eighth year of Diocletian, he seems to have secured a consignment of coins fresh from the mint, and to have put them away promptly: the evidence for this is that a large number of the specimens of this year in the hoard are quite unworn, and further present several instances of fairly long series of coins struck from the same dies; among the 117 examples of this date, there are series of 12, 16, 18, and 22 coins from the same obverse and reverse dies, which would hardly have been found together if they had passed into general circulation; and, if the hoarder added to his deposit in this year a special lot of freshly struck coins, instead of such as casually came into his hands, the result might again be a disturbance of the proportionate total for this year as compared with adjacent ones.

The Billon and Bronze. There may be similar disturbing factors, not so readily discoverable, in other cases; but, subject to this possibility, the totals given against each year furnish some index to the activity of the Alexandrian mint as regards the issue of tetradrachms. It may be worth while to repeat here two facts which I have previously pointed out (Fayurn Towns, p. 68): firstly, that when this mint was busy striking tetradrachms, comparatively little bronze was coined; the chief issues of bronze in the first century A.D. were in the reigns of Augustus, Claudius (latter part), Vespasian (latter part), and Domitian, at which periods little or no billon was minted; and in the second century, while there was a steady, but not large, output of billon from the reign of Trajan to that of Marcus Aurelius, there was a very considerable amount of bronze issued; and, secondly, that, as a rule, the larger the coinage of tetradrachms in any year, the smaller was the number of distinct types used. The latter point deserves emphasizing, as an historical argument in regard to the recognition of an emperor at Alexandria has lately been founded on the fact that several different types of his coinage of a particular year are known to exist in different collections, from which it is assumed that he must have issued a large number of coins in this year and have been recognised for a considerable part of it. This argument is quite unsound, in view of the principle just stated.

The Tables. It seems desirable to publish these tables, in spite of the imperfect nature of the conclusions, as the material will be of service to any one who can secure more hoards and work them out in a similar manner, especially if thereby the rate of wastage in the first century can be determined and the gap between the second and third centuries be bridged. I have entered all the years from the beginning of the reign of Claudius, whether there are any coins belonging to them or not, for convenience of reference, placing a mark o in front of any year of which no tetradrachms are known to exist. I have omitted the earlier years, the 7th, 11th, 14th, and 18th to 23rd inclusive, of Tiberius, when tetradrachms were struck, to save space, as no specimens of any of these years occurred in any of these hoards: indeed, so far as I have observed, and my observation is confirmed by the much wider experience of Signor Dattari, tetradrachms of Tiberius are never found associated in hoards with those of later reigns. The dates are given by regnal years and years A.D., which are not coincident: the regnal years of Roman emperors were reckoned in Egypt on the local kalendar, the year of which began on August 29th; and any fraction of a year from the accession of an emperor to the following August 28th was counted as his first regnal year.

To complete the record, it should be noted that some of the hoards comprised a few coins which are not given in the table, as follows:—

Hoard ii. 2 Ptolemaic; 1 bronze of Antoninus Pius ; 75 tetradrachms of Nero of doubtful dates (ie. coins so misstruck that the date, or an essential part of it, is off the flan).
Hoard x. 2 Ptolemaic bronze ; 37 illegible tetradrachms.
Hoard xi. 5 Ptolemaic bronze ; 31 illegible tedradrachms.
Hoard xii. 78 illegible tetradrachms.
Hoard xiii. 2 Ptolemaic silver; 1 bronze of Claudius ; 62 illegible tetradrachms; 20 of doubtful dates.
Hoard xiv. 4 illegible tetradrachms.
Hoard xv. 2 Ptolemaic bronze ; 4 illegible tetradrachms.