FIFTH AND FOURTH CENTURY GOLD COINS FROM THE THRACIAN COAST

A glance at the weights of these coins will surprise any one who expects a close correspondence between them and contemporary Attic issues, whether gold or silver. While one must be careful not to lay too much emphasis upon the weight of individual coins because of variations within a single issue, it must be remembered that gold coins adhere more closely to the norm than silver. 1 We shall see that the weight of many of the coins we are studying can be explained more easily by assuming that they belong to standards other than Attic.

As we happen to know the history of Thasos in greater detail than the history of the other three cities, its coins make a good point of departure. Professor Gardner says the Thasians "struck but little coin in the second half of the [fifth] century, that which they did strike following the standard of Athens" up to about 424. Then they made a breach with the past by adopting a new standard and new types. The standard they chose was probably Chian. 1

Leaving the question of date aside for a moment, let us consider the standard of the Thasian coins before the change. While didrachms not far from Attic weight were struck, 2 drachms are all of early date, so far as I have been able to discover. Moreover, their weight is such that it is scarcely correct to call them Attic. 1 A more important denomination is the so-called tetrobol weighing about 3.60 gm. 2 While it is impossible to fit an Attic didrachm and a coin of 3.60 gm into any normal scheme of coinage, there can be no doubt as to the value of such a piece in neighboring cities, for it is identical with the Abderitan-Phoenician drachm. It can not very well be called a Thasian tetrobol, for then we should expect to find staters weighing 10.80 gm to correspond, about two grammes higher than the contemporary Thasian staters, though possibly correct for the stater before it began to deteriorate. 3 What this coin may have been in the early fifth century when the stater of Thasos weighed about 9.25 gm 1 is immaterial to our discussion, but before 411 it was a drachm abroad and a pentobol at home, or wherever the Attic standard prevailed.

Since these pentobol-drachms had been for many years current in Thasos, the change in standard in 411 2 was very slight. Possibly the old staters had ceased to be struck shortly before the change of style, i. e., about 430, but of this we can not be certain. It is even possible that the reduced weight of the staters was due to the first Athenian monetary law, which I am inclined to think was passed early in the Peloponnesian War and required the use of the Attic standard for silver struck within the Empire. 3 In that case the last staters of the old type continued until the minting of silver was forbidden by Athens about 415.

There was a slight increase in the weight of the pentobol-drachms for some unknown reason 4 suffi- cient to bring them within the range of the Chian standard. 1 Whether we must assume on this account that the Chian standard had begun to exert an influence on the Thracian coast at this time, I am unable to say, but certainly if the assumption is necessary, we must date the change in 411 rather than in 424, for then there would have been an incentive on the part of Thasos to enter into closer relations with the opposite shore of the Aegean.

Ultimately, there was a more radical change when Thasos adopted tetradrachms of the new weight; but for the first few years after the change in type no staters were issued, whether of the new weight or the old. Thus the change from one standard to the other was gradual and almost imperceptible. The following table shows the development of the new standard, and the issues of various denominations.

It will be noted that about 370 there was a marked increase from about 14.75 gm to 15.05 gm in the average weight of the tetradrachm, 4 and a drop just as sudden in the standard about 356. The first change was made only a few years after the entrance of Thasos into the second Athenian confederacy; still the purpose is not clear, unless it was intended to place the stater of Thasos more nearly on a par with that of Aenus, or to facilitate exchange with Athens. A Thasian stater and a triobol would equal one Athenian tetradrachm. Nor is there a ready explanation for the decline of the standard about 356, but since it is contemporary with the advance of Philip into the Pangaean region and the adoption by him of the Phoenician standard, there may be some connec- tion between these events. The loss of the Thasian mainland colonies might help to explain the poverty which is apparent in the late pieces of reduced standard. The use of the didrachm in place of the tetradrachm is hard to explain unless we ascribe it to growing poverty.

At the time of the change of standard in 411 the bearded Dionysus became the obverse type for the tetradrachms, drachms, and quarter drachms. Half drachms used a youthful head in place of the bearded Dionysus. 1 The reverse of all denominations was characterized by the archer Heracles, kneeling, a type borrowed from an apotropaic relief which served to protect one of the city gates. 2 At this gate, Dionysus also stood guard. Thus the association of the two divinities which appears on the coins of Thasos was of long standing. Although the relief of the archer Heracles is much earlier than the rebuilding of the Thasian walls in 411 at the time of the revolt from Athens, still the adoption of the guardians of the city as coin types suggest a connection between the monetary reform and the strengthening of the city's defences.

On another group of quarter-drachms, not tabulated in the foregoing table, we find a kneeling satyr or silen holding a cantharus, copied likewise from a Thasian gate relief. 1 It is important to ascertain which series of quarter-drachms was the earlier. There are also two series of hemidrachms, one obviously contemporary with the quarterdrachms with kneeling satyr, for the smaller coin has on the reverse one amphora, the larger coin two amphoras. 2 The larger coin has for obverse type a Janiform satyr's head. Comparison of the Heracles hemidrachms with the staters and tetradrachms with the same type shows that they do not come at the beginning of the series. 3 They were struck apparently between about 390 and 370. Moreover, the style of the double-amphora hemidrachms is weak, and it points to a date pos- sibly about the middle of the fourth century, i. e., after the Heracles triobols ceased to appear.

While the Dionysus-Heracles quarter-drachms are difficult to compare with the larger denominations because of the smallness of the type, 1 still they seem to be somewhat earlier in date than the Heracles hemidrachms and contemporary with some of the early Dionysus-Heracles drachms. The lower limit for the issue of this type of quarter-drachms was possibly 395. It was followed almost immediately by the quarter-drachms with amphora reverse. The amphora series extended over a considerable number of years, as may be seen from the number of extant specimens and from the diversity of style shown in the various dies. Certain of the coins show very fine workmanship, Fig. 1, and belong to the best period of Thasian art. Others have the weakness of the Janiform silen found on the triobols.

Fig. 1

It is even possible that some of the earliest specimens were struck before 411 to accompany the larger coins with a silen kneeling, holding a nymph, not a cantharus, and that the dies were cut by the foreign artists who had produced the very fine coins which preceded the change in type. Since both the Heracles and the silen are types taken from Thasian gate reliefs, it is reasonable to suppose that the adoption of both types was more or less contemporaneous.

I have tabulated the weights of seventy-five quarterdrachms. They average .83 gm, and a frequency table gives a norm of about .85 gm.

These weights would be satisfactory as divisions of the drachm weighing 3.60 gm, for small denominations rarely weigh as much as the standard requires. They would not be extraordinarily low for divisions of the heavier drachm of new type, the norm for which is between 3.70 gm and 3.74 gm, according to a frequency table for seventy-nine specimens.

Thus the group of amphora quarter-drachms might have started as divisions of the lighter drachms and been continued as divisions of the slightly heavier Dionysus-Heracles coins. Nevertheless, when the Heracles-Dionysus types were used for the smaller denominations, there was apparently an increase of weight amounting to about .10 gm for the quarter drachm. I have come to this conclusion because of the uniformity of weights displayed by the few published trihemiobols of the Dionysus-Heracles series. The five specimens range from 1.01 gm to .90 gm, and they average .95 gm.

The narrow range in the weights of this group of quarter-drachms is in marked contrast with the diversity shown by the amphora series. It would seem as though the explanation for this fact lay in the changes of standard at Thasos during the time that the amphora pieces were used. This would be true whether the series began before 411 or after 390. Coins struck after about 356 would show a decline commensurate with the fall of the stater from about 15.20 gm to less than 14.00 gm.

This phenomenon is more apparent in the triobols. The Heracles series, struck before 370, shows a norm of about 1.81 gm (seven specimens ranging from 1.87 gm to 1.60 gm), while the double amphora coins show two areas of grouping, one about 1.50 gm, and the higher between 1.72 gm and 1.77 gm.

There is one peculiarity about the kneeling-satyr quarter-drachms, viz., the use of the omega in ΘAΣIΩN contemporaneously with the omicron (ΘAΣION) which appears on all Thasian coins of Heracles type. 1 An adequate explanation for this diversity of spelling is hard to find, but one may at least suggest that the omicron of ΘAΣION is a survival of the Parian alphabet which was once in general use in Thasos; and since the Parians interchanged the values of omicron and omega, the word ΘAΣION would be the equivalent of ΘAΣIΩN, the genitive plural of the ethnic. 1 It is true that the Parian alphabet was abandoned for Thasian inscriptions earlier in the fifth century, and official inscriptions contemporary with the first Heracles-Dionysus coins regularly used Ω and O for the letters omega and omicron. 2 But when one remembers that the new Thasian coinage (411 B. C.) inscribed ΘAΣION was the result of an anti-foreign (anti-Athenian) uprising, was accompanied by the renewal of close relations with Paros, and at the same time looked to sixth century reliefs for types, one may be permitted to express the opinion that some patriotic artist chose to revive the use of the ancestral alphabet, and that his example was followed in the following issues. Still, it is rash tobe dogmatic. The form ΘAΣION may be instead a neuter adjective, modifying some word like νόμισμα understood.

It should be noted in this connection that not all of the kneeling-satyr quarter-drachms use the omega. Two specimens, with incuses which seem to indicate a fifth century date (Plate VI, Nos. 50, 51), have the inscription ΩAΣION which appears on the contemporary kneeling-Heracles drachms. Because of these specimens there need be little hesitation about assigning the beginning of the series to a date not far from 411. After a short time the issue was replaced temporarily by kneeling-Heracles quarter drachms, and when the kneeling-satyr reappeared upon the quarter drachms, the omicron was finally displaced by an omega on this denomination.

At the time of the change of types in 411, Thasos began to issue gold coins with the kneeling Heracles on the reverse and Dionysus or a youthful head on the obverse. There were three denominations of these gold coins, and judged by type, symbols, legends, and weights, the extant coins are from several distinct issues probably extending over many years.

In the neighborhood of Thasos, gold was comparatively abundant and therefore cheap. There were mines on the island and the rich Pangaean region was readily accessible. 1 This proximity to gold mines would not necessarily have any effect upon the weight of the Thasian gold coins, but it would probably prevent them from being below standard and might even produce coins a little over weight as at Panticapaeum. 1 Quite probably Thasos was operating her own gold mines at this time. At any rate her large tribute of thirty talents indicates wealth; and after her revolt, her struggle for the control of Neapolis suggests a desire to increase or reëstablish her holdings in the mining region of the mainland. 2 Furthermore, Thasos, alone of the cities on the Thracian coast, coined gold year after year. Maroneia struck gold coins on two occasions, but there were eight different varieties of Thasian gold coins.

It is therefore clear that the coinage of gold at Thasos was of regular occurrence and that it lasted over a number of years.

Although Thasos made gold a part of her regular currency, it is possible that in the first place it may have been struck as money of necessity, perhaps during her struggle to subdue Neapolis. This would be the time most natural for the minting of gold coins. Money was needed to pay the troops of the Peloponnesian allies and to defray the large expenses incidental to rebuilding the walls destroyed by Cimon and to placing herself in readiness for the probable Athenian counter-attack. Thus, knowing that Thasos struck gold toward the end of the fifth century, we should expect to find that this began between 411 and 408 in the brief period of Thasian independence, when money was needed in large quantities and when the restraining hand of Athens was absent. Nor should we expect a city that had recently given up the Attic standard for silver to adopt it for gold, even though, as Professor Gardner says, "the mass of the currency in all the region at the time consisted of the silver coins of Athens". 1 Athens had not yet struck any gold coins of her own.

Nor do the weights of the Thasian gold coins bear out the theory that they were following Attic models. Of the three weights, only that of about 3.90 gm could be an Attic drachm, if it were minted in imitation of the Athenian coins; but as the norm for the Attic gold drachm was about 40 gm heavier than the heaviest extant Thasian piece, it seems incredible that any one would be willing to accept it at that valuation. 2 The coins of 2.78 gm might be considered Attic tetrobols, except that such a denomination, whether in gold or silver, was unknown at Athens, and that again it is slightly under weight. The division of the stater into thirds, tetrobols, was not an Attic device, and it fits in better with other standards and with antior non-Athenian influences.

The coin of 2.02 gm was perhaps originally about the weight of an Attic hemidrachm, but it might be considered a fourth of a daric stater. 3 It should be compared with a coin of Aenus of similar weight to be discussed later. As this was probably intended to pass as a quarter daric, the small coin of Thasos may have been valued at the same rate. Thus, on the whole, the weights of the Thasian gold coins bear out our previous conclusion that they were not minted with the Athenian example in mind. Certainly their weights can not be used to disprove our hypothesis.

It is more difficult to relate the coins to a single standard, but as issues of gold were only experimental, or perhaps money of necessity, 1 it may not be necessary to do that. One of the coins may have been minted to fill a specific need in one year, and the others may have been minted a few years later when conditions had changed and gold had become a recognized part of Thasian currency. In fact, a study of the coins themselves shows that they belong to different dates and different issues.

The heaviest of the three Thasian denominations in gold weighs about 3.90 gm. Although the gold drachms are uniformly heavier than the silver Chian drachms of ca. 3.70 gm, then used by Thasos, the difference is slight, 2 and there is no other standard to which they correspond even remotely. Athens in minting gold used her silver standard, and other cities would be inclined, to do the same in similar circumstances.

We must now consider the important question of dates.

It is difficult to be precise about the dates of the Thasian coins, whether gold or silver, but we know that during the Peloponnesian War changes of standard were frequent on the Thracian coast. Professor Gardner thinks that the chief landmark in the history of coinage in this region was the expedition of Brasidas. 1 While this is to a certain extent true of the district where Brasidas was active, one must remember that the adoption of the Macedonian variety of the Abderite-Phoenician standard in the Chalcidic peninsula began not later than the Chalcidic alliance with Perdiccas in 432, thus antedating the expedition of Brasidas, 2 and also that no city east of Amphipolis was affected. 3 There were changes at Abdera and Maroneia during the first years of the Peloponnesian War, but other forces were at work there for which Brasidas was in no way responsible. For the district eastward from Amphipolis, including Thasos and Neapolis, the chief landmarks are the growing popularity of the Attic standard about 430, the rise and fall of the Odrysian empire, and the revolt of 411–408. 4 Unfortunately these landmarks are not as clear as they might be, especially as chronological data.

But as the standard of Thasos was closely related to Attic in the period before the change, and as the new standard was Chian, not Abderite-Phoenician, we are in a position to ascribe the new monetary system of Thasos to forces that became active on the Thracian coast with the advent of Peloponnesian fleets in the last years of the war. The new system would be due largely to political considerations and it would show Thasian sympathy with the Peloponnesian side, indicating a stage in the long drawn-out struggle when Peloponnesian fleets were preponderant in the Aegean. 1 Furthermore, the Chian standard works in well with the Aeginetic standard, 2 the one to which the Peloponnesian sailors were accustomed and one almost identical with the standard recently adopted by Abdera. 3 It must be remembered also that Aenus about 412 changed from a light Attic standard to a form of Chian. 4 Finally, as the new type, the archer Heracles, was taken from an ancient basrelief over one of the gates of the city, the coinage must be associated with the newly-won independence of the city and the rebuilding of its walls. Heracles was restored to his ancient place, metaphorically if not literally, as guardian and deliverer (Σωτήρ) of the city. Thus the Thracian reforms can not possibly precede 411, and they probably occurred soon afterward.

The first minting of gold coins, which must be considered contemporary with the Thasian reforms, may be taken as an illustration of the pressing need for money at Thasos and of the measures taken by Thasos to finance her struggle with Athens and Neapolis. Thus to the need of filling the war chest may be ascribed this departure from the continental Greek custom of depending upon silver currency alone, and never was the Thasian treasury so much in need of filling as in the years following 411 when Thasos had to provide maintenance for the Peloponnesian fleet that was assisting her to regain control of the mining region on the mainland. Since the money was to be used for the payment of sailors who had become familiar with gold coins on the Asiatic coast, it was most natural for the Thasian moneyers to turn to the minting of gold.

The list of Thasian coins below makes no pretense of being complete. It is merely a list of pieces which I have used, supplemented by others which have come to my attention during the preparation of this monograph. The bracket { is used to indicate that coins were struck from the same dies; but no attempt has been made to examine all specimens for die combinations. Tetradrachms are given capital letters; didrachms and drachms, small letters; smaller denominations, Greek small letters. Plated and doubtful coins are given the letters x, y, z.

For convenience in establishing a chronological sequence for the gold coins, I have classified the issues of Thasos as follows:

Turning now to Maroneia, a city whose numismatic history is closely connected with Abdera, we find that it issued gold coins of two weights, ca. 4.00 gm and 3.24 gm (62 and 50 grains). 1 Up to about 430, as I shall try to show, Maroneia used the Phoenician standard that had been used in Abdera. Then for a few years, like Abdera, it seems to have been experimenting with its currency, for coins struck some time between 430 and 410 are of a nondescript weight used at Abdera, where it has been called Aeginetic. 2 During the years when the Thracian coast was in revolt from Athens, or possibly after the break-up of the empire in 405, Maroneia minted coins of another peculiar weight, sometimes named Light Attic. 3 When this period of experimentation was over, Maroneia, with Abdera, settled down to the use of the Persian standard, probably about 395.

Before discussing the problems presented by the gold coins of Maroneia, we should consider the evidence on which the foregoing outline of Maroneitan monetary history is based, and we must try to ascertain the significance of the changes in standard, the character of the various standards used by Maroneia, and their relationship with neighboring standards.

Our study must begin with the Maroneitan tetradrachms of Abderite-Phoenician weight, for the years when they were minted constitute a definite period in the history of the city. Up to the time when they were struck, Maroneia used didrachms (Pl. I, C), drachms, and smaller coins. Now the substitution of tetradrachms for didrachms 1 and the increased output of the Maroneitan mint indicate, so it would seem, prosperity due to an extensive foreign commerce. Quite likely, the growing commerce was with the interior, for trade with Athens or districts on the shores of the Aegean would have been better served by the use of the Attic standard.

Proof that Maroneia experienced a burst of prosperity in the second half of the century is to be found in the Attic quota lists. Up to 438 the tribute of Maroneia indicates no great wealth, but in 438–7 her tribute was raised from one and a half talents to ten, a sum that was paid at least until 432 or 431. By 427–6 Maroneia's tribute was reduced to three talents. 1

This sudden prosperity is almost certainly due to disturbances on the Thracian coast arising from the growth of the Odrysian empire. Just when Maroneia was prospering, Aenus was suffering, for its tribute was reduced from ten talents to four in 438–7. 2 After that date it is doubtful whether it paid any tribute at all, for its name does not appear on any of the later quota lists, some of which are complete for the Thracian district. Thus the loss suffered by Aenus was Maroneia's gain. But the Odrysian power was spreading westward, and the rich harvest of the first few years was not permanent. By this time Abdera also felt the influ- ence of these Thracian disturbances, for in 432 her tribute was lowered from fifteen talents to ten. 1 Lower than that it did not go in the extant lists.

Maroneia perhaps was the permanent gainer from the impoverishment of Aenus and Abdera, for its tribute never returned to the earlier sum of one and a half talents, and its coinage progresses uninterruptedly from one standard to another with a fairly plentiful supply of tetradrachms.

I think we may date the first minting of tetradrachms at Maroneia about ten or fifteen years before her tribute was raised to ten talents, that is, about 450. At first, coins of this denomination were struck in small quantities and at irregular intervals, but about the time Maroneia was paying her maximum tribute tetradrachms became plentiful. There is an appreciable gap, for example, between the coins No. 1 and No. 3 on Plate VII, but toward the end of the series the style changes imperceptibly from issue to issue. Frequently we have several specimens from the same dies, and occasionally we have visible proof that the dies received hard usage.

In order that we may have as complete a picture as possible of Maroneitan coinage in the last fifty years of the fifth century I have compiled the following tables. While my material is by no means exhaustive, it is full enough for the purpose.

Table of Weights for Abderite-Phoenician Tetradrachms

14.76 =	1
14.51–14.60 =	2
14.41–14.50 =	0
14.31–14.40 =	2
14.21–14.30 =	5
14.11–14.20 =	8
14.01–14.10 =	9
13.91–14.00 =	17
13.81–13.90 =	13
13.71–13.80 =	6
13.61–13.70 =	0
13.51–13.60 =	2
13.41–13.50 =	1
12.64–13.03 =	2
	68 1

Our table shows that the norm for this series was between 13.90 gm and 14.00 gm. The average of the high weight specimens of each issue is ca. 14.10 gm excluding the worn coin weighing 12.64 gm.

The magistrates, symbols, and coins of this series, together with the die combinations and weights, are listed below.

Serial No.	Die No.		Weight
	ΔΕΟΝϒΣ	Symbol: cantharus above horse. Legend Μ-Α-ΡΩΝ above cantharus.
1.	1–1	a. Benson 1909, 457	14.51

Serial No.	Die No.		Weight
ΕΠΙ	ΜΗΤΡΟΔΟΤΟ	Legend and symbol as above.
2.	2–2	Paris	14.00
3.	2–3	McClean 3952, ex Hirsch XIII, Rhousopoulos, 579	13.76
4.	3–4	a. Brussels	14.03
		b. Copenhagen	14.11
5.	3–5	a. Br. Mus. Cat. 11	13.77
		b. Bement 817 ex Egger XXXIX, 147	14.17
6.	3–6	a. Berlin	14.20
		b. Hirsch XXV, 175	14.00
7.	3–7	Benson 458	14.38
8.	3–8	Berlin 27	13.79
9.	4–9	Newell ex Na ville V, 1516 ex Hirsch XXXIII, 568 (XXI, 900)	13.92
	ΔΕΟΝɼΣ	Symbol as above. The legend ΜΑΡΩΝ is placed either above the cantharus, as above, or across the back of the horse.
10.	5–10	Berlin	13.82
11.	6–11	a. Hermitage	13.53
		b. Regling-Warren 493	14.16
		c. Berlin 28	13.73
12.	7–12	Pozzi 1042	13.03
12½.	7a–12a	Br. Mus. Cat. 10 (Fig. 4)	13.92

Fig. 4

Serial No.	Die No.		Weight
ΕΠΙ	ΜΗΤΡΟΔΟΤΟ	Symbol: eight-pointed star. The legend ΜΑΡΩΝ is below the horse's feet.
13.	8–13	a. McClean 3953 ex Soth. 1900, 234	14.24
		b. Pozzi 1043 ex Bachelor 1907, 69 (Anson, III, XII, 449)	13.83
14.	9–14	De Luynes 1772 (Imhoof-Blumer 22, Zeit f. Num. 1876, p. 274 ff.) 1	13.90
ΕΠΙ	ΜΗΤΡΟΔΟΤΟ	Symbol: crescent. Of the legend ΜΑΡ-ΩΝΙΤΕΩΝ, the first three letters are below the horse's head, the rest across its back.
15.	10–15	Berlin 26	13.44
	ΜΗΤΡΟΔΟΤΟΣ	Symbol: eight-pointed star. Legend as above.
16.	11–16	Empedocles ex H. Weber 2330, and Soth., Feb. 29, 1884, 438	14.10
17.	12–16	Br. Mus.	14.08

Serial No.	Die No.		Weight
ΕΠΙ	ΠΟΣΙΔΗΙΟ	Symbol: helmet. Legend ΜΑΡΩΝ above the horse's back.
18.	13–17	Soth. 1904, 182	14.00
19A.	14–18	Newell (Fig. 5)	13.91

Fig. 5

19.	14–19	Berlin (Imhoof-Blumer 23)	13.83
20.	14–20	Bement 822, ex Ratto 1912, 558, ex Egger, 1, 7, 1908, 308	13.95
21.	15–21	Br. Mus.
	ΜΗΤΡΟΔΟΤΟ	Symbol: crescent. Legend ΜΑΡ-ΩΝΙΤΕΩΝ. The first three letters are below the horse's head, the rest above its back.
22.	16–22	Naville XIII, 599 ex V, 1521	14.02
	ΜΗΤΡΟΔΟΤΟ	Symbol: wreath. Legend ΜΑΡΩΝΙΤΕΩΝ, ΜΑΡΩΝΗΙΤΕΩΝ, or ΜΑΡΩΝΕΙΤΕΩΝ, either divided as above, or placed in one or two lines above the horse's back.
23.	17–23	Br. Mus.	13.99
24.	18–24	a. Newell ex Naville V, 1519, ex Hirsch XXXIII, 569, ex Hirsch XXV, 176	13.98
		b. Copenhagen	13.99
25.	19–25	McClean 3954, ex O'Hagen, 364	13.99
26.	20–26	a. Berlin 29	14.12
		b. Br. Mus. Cat. 13	13.58
	ΜΗΤΡΟΔΟΤΟ	Symbol: pointed cap. No legend.
27.	21–27	Newell	13.84
	ΜΗΤΡΟΔΟΤΟ	Symbol: lyre. Legend ΜΑΡΩΝΙΤΕΩΝ in small letters
28.	22–28	Imhoof-Blumer 21	13.96
	ΜΗΤΡΟΔΟΤΟ	Symbol: owl. No legend.
29.	23–29	Br. Mus. Cat. 12	13.98
	ΜΗΤΡΟΔΟΤΟΣ	Symbol: bearded head. No legend.
30.	24–30	a. Morgan, ex Consul Weber, 1908, 901	13.86
		b. Hermitage	13.85
31.	24–31	Naville V, 1518, ex Pozzi 1044	14.32
32.	24–32	Berlin 30, Cf. Mionnet, Supp. II, 335, 317; Eckhel, Num. Vet. Anec., p. 57, Pl. V, I	14.15
	ΜΗΤΡΟΔΟΤΟΣ	Symbol: youthful head. ΜΑΡΩΝΙΤΕΩΝ in exergue.
34.	25–31	a. Br. Mus.	14.126
		b. Brussels	14.025
		c. Empedocles ex Bement 818, ex Hirsch XXXII, 417	13.82
		d. Hirsch XXV, 177, ex Soth. 1904, 184	14–57
		e. Sartiges 170	…
33.	25–32	McClean 3955	14.24
35.	25–33	a. Berlin 31	14.76
		b. Jameson 2017, ex H. Weber 2329	14.08
	ΜΗΤΡΟΦΩΝ	Symbol and legend as above.
36	25–34	Beatty (plugged)	13.89
	ΜΗΤΡΟΔΟΤΟΣ	Without symbol and legend.
37.	26–35	Hermitage	13.85
38.	27–36	Br. Mus. ex Bunbury 584 (Imhoof-Blumer 20; Anson, III, 450)	13.80
	ΜΗΤΡΟΔΟΤΟΣ	Symbol: wheel.
39.	28–36	Hermitage	12.64
	ΜΗΤΡΟΦΩΝ	Symbol: wheel.
40.	28–34	a. Berlin 32	14.05
		b. Bement 820	13.87
		Cf. Ivanoff, 14	14.00
41.	29–37	Br. Mus. Cat. 14	14.30
42.	30–38	Manchester	14.11
43.	31–39	Hague	13.90
44.	31–40	a. Naville XIII, 600, ex V, 1520, ex Soth. 1914, Schles. y Guz. 74, ex Merzbacher 1909, 2682	14.09
		b. Newell ex Hirsch XXI, 902
		c. Mionnet, I, 390, 170, Pl. 48, 6.	14.07
45.	32–41	a. Regling-Warren 492	13.94
		b. Munich	—
	ΕΠΙ ΑΘΗΝΕΩ	No symbol. Legend ΜΑΡΩΝΙΤΕΩΝ, of which the last three letters are retrograde.
46.	33–42	Berlin	13–95
	ΕΠΙ ΠϒΘΟΔΩΡΟ	No symbol. Legend ΜΑΡΩΝΙΤΕΩΝ in exergue.
47.	34–43	a. Berlin	13.90
		b. Bement 821, ex Fenerley Bey, 1912, 262, ex Egger 1908, 307	13.95
		c. Soth. May 9, 1904, 183 (Anson, III. 457)	14.25
		d. Jameson 1059	14.29
		e. Br. Mus. Cat. 14 α	13.71
	ΜΗΤΡΟΦΩΝ	Possible forgery Symbol: wheel. No legend. Fig. 6. Fig. 6
	a.	Berlin (Dannenberg)	12.96
	b.	Berlin (Prokesch Osten)	12.56
	c.	Berlin (Löbbecke)	13–53
	d.	Pozzi 1045 (Naville X, 473)	13.50
	e.	Bement 819	13.04
	f.	Hermitage	13.18
	g.	McClean 3956	13.95
	h.	Munich
	i.	Feuardent, 6, 9, 1913, 147

End Notes

1	In the table I have not included the weight of the Questionable Metrophon issue (p. 65, fig. 6).
1	Hereafter cited as Imhoof-Blumer with the number of the coin as given in his list.

Fig. 7

Fig. 8

In our list, we find forty-nine die combinations (one possibly spurious) and only seven mint officials, from which it is clear that magistrates could serve year after year. Just as the number of magistrates is no criterion of the length of the time during which the Phoenician weight was used, so I think the symbols must be rejected as a criterion, for apparently they could be repeated, as is shown by the cantharus on the coins of Deonys and Metrodotos. It is possible that the cantharus was a subsidiary symbol having no connection whatever with the magistrates whose names appear on the coins of Maroneia, until Metrodotos or one of his colleagues conceived the idea of distinguishing each issue by the use of a special mark, the mint official's symbol of the later coins. It is even possible that the star and the crescent of Metrodotos were each used for more than one term of office (?).

The practice followed at the end of the series is clear enough, for the symbol almost certainly changed every year, even though the same official was in office. Another fact is apparent from our study of the dies. They were often shared by officials using the same symbols; and when all the data is collected, it seems to indicate that frequently two men were in charge of the mint. For the period we are studying the evidence is as follows:

Because of the style of the earlier tetradrachms, I have assumed that Metrodotos and Deonys minted coins in more than one year, and it is certainly possible that the early coins with the cantharus symbol should be distributed over several years. For this group of coins seven obverse and twelve reverse dies are known, a number disproportionately large for one year's issue. Consequently I am inclined to believe that the frequently changing symbol was not introduced until several years had passed.

The long career of Metrodotos at the mint is as certain as it is perplexing. In what capacity did he serve? Was he technically a magistrate of Maroneia, or was he simply a professional mintmaster employed by the city? According to the list given above, he was in office eleven years out of fourteen, and there is no certainty that he did not issue coins in conjunction with Posideios, Pythodoros, and Athenes. Another striking fact is that the colleagues of Metrodotos changed frequently and that there are many years when no colleagues are known. Who were these colleagues? Political officers, or assistant mintmasters called in to assist when the mint was most active? Is it possible that they engraved the dies? At times a new name coincides with the introduction of changes in the style of the coins, as for example in the pieces of Posideios 1 , or in the unusual variety shown by the pieces of Metrophon and other late magistrates. Noe (Mende Hoard, p. 50) has shown that two artists were at work in Mende during this period, and as there are pieces in the Maroneitan series which cannot be brought into any logical scheme of chronological development, the same may be true of Maroneia, except that the second artist at Maroneia seems to have been a transient, so to speak, and to have changed from time to time.

Noe has called attention to the unique reverse Mende die which has no linear square about the grape vine, and to its resemblance to the unique Maroneia piece which shows the same peculiarity. This, by the way, is a coin of Posideios. The other specimens with the name Posideios differ from the contemporary coins of Metrodotos in having an inartistic graceless vine with four clusters. In the same way "certain pieces of Deonys do not fit into the normal scheme of Maroneitan coinage. One of them has ivy leaf punctuation which does not appear again until the next period, and the later Deonys pieces have the square vine. 1

The coins of Athenes and Pythodoros (Nos. 46 and 47) have no symbols, but it must be admitted that coins of Metrodotos show the same characteristic. The horse of Athenes is proudly stepping, not galloping, with right front foot raised high, a characteristic which appears on an interesting coin of Metrophon (Fig. 6) of which several specimens are extant. The horse of these pieces is so peculiar that its genuineness has been questioned, and to a suspicious style is added an unusual lightness which might be taken as confirming their spuriousness. On the other hand there is nothing suspicious about the fabric and the reverse is perfectly true to type. But it is to be remarked that the workmanship resembles a coin of Pythodoros (No. 47), although the latter is much better done.

One might regard the weight of the coins as decisive evidence against them, except that many of them fall within the range of the standard which followed immediately after the Phoenician. More- over, there is a coin of Metrodotos with wheel symbol of similar weight (No. 39) which shares dies with Phoenician tetradrachms. While this is a worn specimen, still it does not seem to have suffered enough to have brought its weight so far below the Phoenician norm, nor is it any more worn than others of the reduced weight which unmistakably belong to the new standard. It weighs 12.64. The Metrophon pieces which we are discussing weigh as follows:

13.95, 13.53, 13.50, 13.18, 13.04, 12.96, 12.56. They average about 13.20 gm, which is closer to the reduced standard than to the Phoenician which preceded. Other coins with the wheel symbol weigh 14.30, 14.11, 14.09, 14.07, 14.05, 14.00, 13.94, 13.90, and 13.87. Their average is thus 14.04. To show the divergence of standards, I list here the weights of all other specimens of Phoenician weight which are below 13.70.

When coins from similar, or identical, dies have an average weight lower than the average of the lightest pieces of the rest of the series, one must look for an explanation. If the coins are authentic, one must assume that the weights were lowered during the course of the year when the wheel was used as a symbol. In that case the specimens weighing 13.9513.50 might be the last of the old weight. In this connection, two other Metrophon pieces are to be cited, Nos. 49 and 50 of the reduced weight. 1 They both have the wheel and they are within the range of weights of the new standard. The second has the inscription ΜΑΡΩΝΙΤΩΝ beneath the horse. Except that it omits the epsilon, it resembles Nos. 34–36, and 47 in this respect, and these issues we have placed toward the end of the Phoenician series. Its weight is 11.77. The other, No. 49, is described as possibly a barbaric imitation. Not having seen the piece, I am unable to say whether it is more barbarous than the genuine Maroneitan pieces of this reduced weight, some of which are particularly crude, or whether it is another specimen of the suspected group of Metrophon pieces with the standing horse. No. 49 weighs 12.44.

Finally there is a Deonys piece with the wheel symbol, No. 51, undoubtedly genuine, of the reduced weight. Thus it is certain that the symbol was used by the mint in the period of the reduced weight; and secondly, the piece No. 50 seems to show that it was used by Metrophon immediately after the change of standards. Consequently, the weights of the suspicious Metrophon coins can not be used to discredit their authenticity.

During period II (stater of ca. 12.50 gm) the following conspectus shows a marked contrast to the practice of period I.

For the first issue (symbol) we have three names, the veteran Metrodotos 2 and two colleagues Deonys and Metrophon. This is the last appearance of Metrodotos and Metrophon at the mint. Deonys, presumably a different man from the mintmaster who first struck Phoenician tetradrachms fifteen or twenty years earlier, was responsible for two more issues. Then we have new names, year after year, and unsigned coins, many of which were made by unskilled workmen. It should be noted that the finest of the pieces of this period has the name Metrophanes and that Metrophanes' name appears on some of the finest pieces of period III.

In period III (staters of so-called "Light Attic Weight") there is no evidence to show that in any year two men were at the mint, and it is not easy to tell whether the three varieties of Metrophanes pieces were from one or several years. 3

In period IV, when Maroneia used coins of Persian weight, the evidence is conclusive that there were two officials at the mint in several years at least. I summarize it here. We begin with Athènes, the last of the officials of period III, whose colleague was Posideios, as is shown by community of symbols and of dies, both in staters and in triobols.

In this period the names are possibly those of officials who served annual terms, and it is probable that they might serve more than one term. It is also clear that two officials were occasionally in office at the same time.

From the foregoing analysis of Maroneitan magistrates and symbols, it is difficult to formulate a general rule which will apply to all four periods. Yet I am inclined to the view that the names on the coins are those of mint officials, possibly, though not certainly, concessionaires with whom the city had made contracts for the minting of coins. 1 Since the symbols vary from time to time and are used by two men simultaneously, one might conclude that they serve to fix the year of issue. Yet Noe's hypothesis (Metapontum, 31–41) that the symbol marks the issue and not the time of issue is equally satisfactory if one interprets the sharing of dies and symbols by two men as evidence that they had taken a contract in partnership to mint a specified amount of coin, for then the symbol would unite the part of the issue signed by moneyer A with the part signed by B. One might note in passing that the employment of two men simultaneously, though common, was apparently exceptional. But certainty as to the precise meaning of the symbol is out of the question, since contracts for specified amounts of coin would naturally have a time limit, and the contracts may well have been let annually.

At Maroneia the procedure may have been as follows. After the needs of the state had been determined, a contract was let for the coming year, specifying the number of pieces of each denomination to be struck. In some years no tetradrachms were struck, possibly no coins at all. In others, the need for staters may have been so great as to warrant giving the contract to two men, either jointly or separately. But such an hypothesis is not without its difficulties. Consequently, I propose it with hesitation.

Returning to our survey of Phoenician staters, we have seen that Metrodotos easily outstripped the other magistrates with his nine different symbols, counting the cantharus, but not the exceptional issue that had no symbol. Metrophon follows next with two symbols, both of which are used by Metrodotos. The other four magistrates have one symbol each, or none at all, even though one of them, Deonys, held office possibly on two occasions. We probably have specimens from at least thirteen issues, and if we date the first piece of this series about 450, we can date the end of it about 430 without difficulty.

If we take as a criterion of mint activity the number of dies of Metrodotos with the wreath as a symbol and of the succeeding issues, particularly the coins of Metrophon with wheel symbol, we must conclude that the years from about 440 to about 430 were busy ones for the mint. This activity coincides in point of time with the period of high tribute, 438 to about 432. Before 438, Maroneia had paid only one and a half talents. During the Peloponnesian War it paid three. In the ten or fifteen before the wreath issue Maroneia used about sixteen obverse and twenty-two reverse dies. In the few years that followed to about 430 it used eighteen obverse and twenty-one reverse dies, and the period of greatest activity comes at the very end. As it ought to synchronize approximately with the years when Maroneia was paying ten talents tribute, it will be difficult to extend our series beyond 430. While there is much uncertainty in the foregoing argument, nevertheless, since there are other reasons, as we shall see, for dating the end of the Phoenician period about 430, it can be taken as merely corroborative of a conclusion reached by other means.

Assuming for the moment that Maroneia ceased minting Phoenician tetradrachms about 430, we now turn to the standard that took its place. The weights of the twenty-four specimens known to me are distributed thus. I do not include the suspicious Metrophon pieces, which in types are like the Phoenician series, although the Metrodotos wheel coin of this weight and two Metrophon wheel pieces are included.

The mint officials and symbols found on the coins are these:

The following coins have MAPΩNITEΩN in place of the magistrate's name about the reverse square.

While this group of coins follows without a break the tetradrachms of Phoenician weight, both in style and fabric, there are peculiarities to be noted. Several specimens give merely the name of the city, not that of the mint official which is found regularly on coins of Persian and Phoenician weight. On the staters of about 16.20 gm, both names appear around the linear square of the reverse, and in the Phoenician series it was not uncommon for the name of the city to appear in part or in full on the obverse. On one of the Metrophon pieces of the new weight it so appears. There may be nothing significant about the absence of the offi- cial's name, but the fact is so unusual that it calls for comment and explanation if possible. If the names are actually those of magistrates, it may not be too far-fetched to ascribe the peculiarity to democratic reforms which denied to magistrates the glory of placing their names on coins where the name of the community belonged. Even where magistrates' names are given on coins of this series, they give the impression that the war had somehow affected the politics of the city. Instead of finding the same name year after year as in the preceding series, we have a rapid succession of magistrates in a manner characteristically democratic, and instead of having the old names like Metrophon and Metrodotos we seem to see new men in positions of responsibility, just as would have happened if the power of the aristocratic families had been destroyed by some democratic upheaval supported by Athens.

The vine of this type is less stocky and more twisted than on specimens of Phoenician weight, although McClean 3955, Rev. No. 32, is readily recognized as its forerunner, whereas the vines of the Persian weight coins have little in common with their fifth century predecessors. Some of our group resemble Phoenician tetradrachms in having two or three dots after the name. Some have more elaborate punctuation, an ivy leaf or astragal. Several, likewise, have either a dotted or linear circle about the obverse type, another characteristic of the Phoenician series. But while there are variations as to details among the specimens of our series, the general similarity to the preceding series is evident, and taken together they form a class by themselves both as to weight and style.

The heaviest coin I have discovered of this series weighs 12.95 gm, so close to coins which are unmistakably of the Phoenician period that it might have been considered a worn or light-weight Phoenician stater, were it not for its being a very wellpreserved and typical specimen of the new series with twisted vine and city name about the linear square on the reverse. 1 Since many others are badly worn, it is not at all strange that their peculiar weight went unnoticed. Yet the proportion of coins that fall within a narrow range of weights is very large. Of the twenty-three specimens, fourteen range between 12.19 gm and 12.51 gm, and since three of these specimens have duplicates of different weights, we can readily establish an approximate range for the series. For example, the heaviest of the three specimens of No. 63 weighs 12.50 gm, the lightest only 11.46 gm. Likewise the three specimens of No. 52 weigh 12.48 gm, 12.25 gm and 11.27 gm. Thus the light-weight coins of this group are hardly distinguishable by their weights from the Persian staters of Maroneia, the heaviest of which weighs 11.50 gm. 1 It should be noted here that reverse No. 3 is found joined with obverses 3 and 4. No. 51 also weighs 12.25 gm, thereby confirming our conclusion that the very light coin belongs to the series.

The specimens that rise above 12.51 gm are few. There are three specimens of No. 55, one of which weighs 12.25 gm the other 12.82 gm. An unpublished coin in the British Museum, No. 62 weighs 12.80 gm.

As for the names of the magistrates, Metrophon, Metrodotos, and Deonys are found on coins of the preceding period. Metrophanes appears again in the next period. The others appear only in this period. While Metrophon and Metrodotos were in office at the very end of the Phoenician period, Deonys is found on coins widely separate in point of time, for the first Phoenician tetradrachms, struck probably about 450, bear that name. Thus we may assume that the Deonys of our period is a younger man.

In one notable respect, our tables show a difference between the Phoenician period and its successor. The mint was much less active than it had been, as we may surely infer from the use of fewer dies and the scarcity of specimens now known. About half of the specimens and the same proportion of dies were not known before the discovery of a hoard about 1912. 1 Making allowances for losses, we could date the end of the series about 415, if we should take 430 as the approximate date for its beginning, and as Athens suppressed the coinage of silver in the subject cities about 415, the date has intrinsic probability. In that case the succeeding standard may have been introduced as a result of the disturbances on the Thracian coast between 411 and 408, instead of after the disruption of the Athenian Empire following Aegospotami. Nevertheless there is a crudeness about the last of the Athenian tridrachm staters which seems to indicate haste at the mint, and I am inclined to think that coinage was resumed for a short period between 411 and 408 by a rebellious Maroneia. It would then be suppressed when Athens reasserted her authority. When next Maroneia struck coins, soon after Aegospotami, the types and the weights make a clear-cut break with the past.

Having tentatively established a date for the end of our series, we can return to the date of its introduction. Here again the discovery of the hoard provides valuable clues, for it contained as many Abderitan as Maroneitan pieces. While some of the Abderitan coins are not described in the Berlin Corpus and new magistrate's for this period are found on coins of the hoard, we can use the Berlin Corpus as a guide. The Abderitan coins in the hoard follow the same standards as those of Maroneia, and as Strack has dated the introduction of the new standard (ca. 12.50 gm) at Abdera about 425, there can be no question that the Maroneitan series belongs to the period of the Peloponnesian War. But Strack has connected the change in standard at Abdera with the rise of the Odrysian power and the consequent reduction of the Abderite tribute, which he has dated erroneously in the first years of the war instead of in 432 or before. 1 Consequently, his sole concrete reason for choosing the approximate date 425 instead of 430 does not exist. Head, for stylistic reasons I presume, preferred the earlier date, and von Fritze states that this class of Abderitan coins, being directly influenced by the second bloom of Phidian art, dates in the last third of the century, and he accepts Head's date 430. 2

Although the standard is peculiar, I shall try to show that it is a variety of Attic and that it may be connected indirectly with the rise of the Odrysian power. If either of these hypotheses is correct, we ought to date the adoption of the new standard in the first years of the war when the position of Athens had been strengthened by the alliance with Sitalces. After the accession of Seuthes in 424 and the campaigns of Brasidas, the weakening of Athens would have militated against a voluntary adoption of the Attic standard. Nor would it be explicable through Odrysian influence, for Seuthes was antagonistic to Abdera. The approximate date 430 applies with equal force to Maroneia, for the evidence of the hoard, the style and fabric of the coins themselves, and historical probability all require us to give approximately the same dates to the Maroneitan and Abderitan coins of this denomination.

We must now consider the character of the standard and some explanation of its weight that will account for its adoption.

Strack has argued that the impoverishment of Abdera due to the growth of the Odrysian Empire was responsible for the reduction in the weight of its staters to about 12.50 gm, 1 but he was unaware of the fact that Maroneia also was using this standard during the Peloponnesian War. Professor Gardner, on the other hand, connects the use of the standard by Abdera, Aeginetic as he calls it, with the revolt of the Thracian coast between 411 and 408. 2 The sporadic appearance of the standard so far from its natural habitat would then be explained by the presence of Peloponnesian sailors in the rebellious cities. They could now be paid in a currency of familiar weight.

But the latter explanation is unsatisfactory in that it requires us to assume the adoption of the standard after 411 both at Maroneia and Abdera, whereas Maroneia probably adopted another standard almost immediately afterward. Furthermore, the long list of magistrates found on the so-called Aeginetic staters of Abdera indicates a period of about twenty-five years' duration. 1 As it was followed by the Persian standard about 400, 2 it is difficult to date the first use of the misnamed Aeginetic standard after 425. 1 Consequently, Peloponnesian sailors had nothing to do with the use of this peculiar weight in Maroneia and Abdera.

Nor is Strack's explanation any more satisfactory, for it ignores the question of exchange. Why should a city which had been using a standard obviously no longer suited to the needs of commerce—the Phoenician standard was gradually receding on the Thracian coast during this period— adopt another equally unsuited? If the standard was to be lowered merely because of growing poverty, why not lower it to the point where it would work well with the so-called Persian triobols which Abdera had been using since about 440? The new standard was neither suited for trade with countries where the Phoenician standard was used, nor with countries where the Persian stater formed the chief currency.

Moreover, it is wrong to think that the Persian standard was coming in on certain lines of trade, as has been stated on the evidence of the growing use of so-called Persian triobols. Since Persian triobols used by Aenus, Abdera, and Maroneia from about 440 were nothing but Attic tetrobols, as will be made clear later, the north-west Aegean was still dominated by the Attic standard, or to put it more accurately, up to about 411 the Attic standard was growing in use in this region, while the hitherto prevalent standards were declining. Thus we ought to explain the choice of the new weight in the same way that we explain the rejection of the old. The popularity of the Attic standard had made the use of the Phoenician stater unsatisfactory. It also dictated the choice of the new weight, an Attic tridrachm. In other words, Abdera and Maroneia, without raising the weight of their staters, had found a way to accommodate their coinage to that of Athens in a manner acceptable to a people accustomed to division and multiplication by three.

It is even doubtful whether poverty explains the failure of Abdera and Maroneia to adopt the Athenian form of the standard with its tetradrachm as the major denomination, for Maroneia was relatively more wealthy than she had been when she first struck tetradrachms; but compared with Abdera, whose tribute was ten talents a year in 430, Maroneia was poor. Furthermore, the poor Maroneia with her paltry three talents tribute was able after a short period of experimentation with the stater of about 12.50 gm to increase her stater to about 16.20 gm.

While the weight of the staters of about 12.50 gm rarely reaches that demanded by an Attic tridrachm, nevertheless there is apparently no other satisfactory explanation, especially as transplanted standards frequently were lower than their prototypes. Certainly the tendency was not upward. Thus, we have further reason for refusing to consider the Abderite stater of about 12.50 gm Aeginetic. It was as much too high for Aeginetic as it was low for Attic. The latter variation, however, is explicable.

It is difficult to tell whether the adoption of this peculiar variety of the Attic standard was due to the determination of Athens to make her standard supreme in the Aegean, or whether expediency was at the basis of its spread on the Thracian coast during the first years of the Peloponnesian War. Strack thinks that the uninterrupted Abderitan coin series proves that the Athenian monetary laws were never enforced there. 1 Nevertheless, an interruption such as the law would have entailed would be difficult to detect, if it were of short duration, as I think it was.

Although this is not the place to discuss the laws, it is necessary to ascertain their bearing on the problem of Maroneitan coinage, and in doing so, to determine their probable date and scope. 2 That there were two of them is clear from the second which makes a reference to a previous law that went by the name of Clearchos. The law of Clearchos imposed certain penalties for offenses the nature of which we can only conjecture. The second law forbids the cities of the Empire to strike money of silver and to use any but Athenian weights and measures. Now the date of the second law, since it forbids the coining of money, is of first importance to our discussion. Because of a resemblance in procedure to the assessment decree of 425–4, the most recent editors of the law have suggested that the decree was passed during the Archidamian War. 3 Before the recent discovery of a new fragment of the law, there was a tendency to date it about 415–4 on the strength of a passage in the Birds, 1040 f., which clearly refers to it. This play was produced in the spring of 414. It is difficult to believe that a law ten years old would have been the subject of an Aristophanic jest, and consequently despite the dative forms and the similarity with the assessment decree of 425–4, the later date seems preferable. The very fact that Aenus, Maroneia, Abdera, and Thasos were striking coins without any interference from Athens during the period from 425 on, ought to arouse skepticism with regard to the earlier date. It would require us to assume that the cities where coinage was still flourishing were not subject to the law, or in other words that the law applied only to the cities within the empire which had practically ceased to strike money at all. Moreover, the tetradrachms of Mende show that there was no prohibition of coinage before the time of the revolt, and the drachms struck at Mende not long after her recapture by Athens confirm the later date for the second law. (Cf. Noe, Mende Hoard, 53 ff.) 1

On the other hand, if we date the law in 415–4, the interruptions would be of short duration, from about 414 to 412, and again from about 408 to 405, for it goes without saying that cities like Abdera and Maroneia were not slow to resume coinage when the restraining hand of Athens was absent, first at the time of the general revolt of 412–1, and secondly when the empire finally disintegrated. Such interruptions would be difficult to detect even within a series, and if a break coincides with a change of types or of standards, it would be almost impossible to discover from the extant coins whether there had been any cessation of coinage at all.

Yet the unusual number of changes in type and standard that we have met with on the Thracian coast may be the result of the operation of the law. There was a change in both types and standards at Aenus, dated by Strack in 412, and the Thasian coinage was reformed at the time of its revolt. At Abdera the new standard appears not far from the turn of the century, and at Maroneia about ten or fifteen years after the adoption of the tridrachm Attic stater, this was superseded by a new standard with new types.

One can easily rearrange the coinage of Abdera to suit this hypothesis.

It is evident from the foregoing table that a date as late as 425 for the beginning of the new standard is out of the question. For some reasons it might be preferable to shift our dates upward a trifle so that we can assign the adoption of the Persian standard to the years of the revolt. I make this suggestion because in Strack's period VI, dated by him between 400 and 390, there is an abrupt change of types which might very well coincide with the reopening of the mint after a period of Attic suppression. We should then date the introduction of the tridrachm Attic stater just before the outbreak of the Peloponnesian War, and it might coincide in point of time with the adoption of an economic policy by Athens which found expression in the Megarean decree.

As for the law of Clearchos, nothing definite about it is known. Nevertheless, since the second law forbade the minting of silver, one may infer with some probability that the law of Clearchos did not prohibit coinage. 1 It is tempting to assume that its purpose was to prohibit the use of standards other than Attic by the mints of the empire. In that case it might be possible to date the law about 432 and to interpret the peculiar tridrachm standard in Abdera and Maroneia as an attempt at conformity. We have seen Thasos using a didrachm stater about this time, and the small change of Maroneia and Abdera was also based on the Attic standard.

Whatever the direct cause of the adoption of this variation of the Attic standard, it was in general accord with the monetary tendency of the region. Even the interior was not slow to see the advantages of a currency of the popular Attic weight. The Attic tetradrachms of Sparadocus and the didrachms of Seuthes furnish illustrations of this tendency.

It would take us too far afield to analyze the coinage of Sparadocus and to discuss the various theories which have been proposed to account for the types of his coins. 1 None is entirely satisfacfactory, and the matter needs further study. In my opinion Sparadocus, the brother of Sitalces, and the father of Sitalces' successor, Seuthes, lived about the middle of the fifth century. He probably struck coins as an independent or semi-independent ruler of one or more Paeonian tribes between Macedonia and the Odrysian Empire, attached possibly to the latter by an indefinite bond. His seat may have been near the Strymon River. He struck staters ranging from 16.20 gm to 17.00 gm, and fractions of the stater divided by four and twelve, drachms and diobols as they may well be called, although very light for the standard. The range of the drachms, 3.79 gm-3.95 gm, is not far distant from that of certain contemporary Neapolitan pieces, which may originally have been intended to serve as thirds of the local stater. Of the cities in this region that were committed to the Attic standard, only Acanthus struck quarter staters, and comparison shows that they were heavier than the pieces of Sparadocus. The diobol was more common, especially after its adoption by Aenus, Abdera, and Maroneia, and it was also more in accord with the local custom of division by three. But on the whole, the coins of Sparadocus show a certain eclecticism in denominations, as in types, which makes it difficult to explain the source from which his coinage was taken.

Seuthes, Odrysian King after 424, struck drachms (4.01 maximum) and didrachms (8.60 gm). It is difficult to see what purpose was served by the use of didrachms, unless Thasian and Neapolitan influence was responsible, yet the drachm seems rather to have come from eastern Thrace. Drachms of light Attic weight were struck at Aenus during the second half of the fifth century and in Selymbria about the time when the Odrysian power began to spread. The types of Seuthes, Thracian horseman on the obverse, and the unique inscription on the reverse, show clearly that he was copying no Greek models. When both types and weights are taken into consideration, it seems to be clear that Seuthes, and the semi-Odrysian Sparadocus as well, like the fourth century dynasts who used the mint of an inland town Cypsela, 1 had mints located somewhere in the interior. They would thus have been free to recruit die-engravers from the mints of neighboring Greek cities and to choose types to suit their economic needs and individual artistic preferences. The denominations struck by the two dynasts illustrate clearly how an old standard in new surroundings is modified by the economic needs and customs of those who adopt it. Both the didrachm and the diobol were denominations unknown to Athens in this period, though not unexampled on the Thracian coast. Thus the modification of the Attic standard (Attic tridrachm) that appears at Maroneia and Abdera, though more peculiar, finds its parallel in a neighboring coinage. Division and multiplication by three was apparently ingrained in the numismatic instincts of these towns. I know of no other explanation for the use of the tridrachm stater. That it was a tridrachm stater is evident also from the fact that the next change at Maroneia also resulted in the use of a tridrachm stater, this time of Persian weight.

Our study of the change from the Phoenician tetradrachms to Attic tridrachms would not be complete without some reference to the contemporary fractional currency of Maroneia. The tendency of the Thracian coast in the second half of the fifth century was away from drachms, i. e., quarter staters, whether of Attic or of Phoenician weight. Instead, nearly all of Maroneia's neighbors were using tetrobols of light Attic weight (Persian hemidrachms). Abdera, shortly before 430, had struck small coins of about 2.82 gm, 1 while those of Aenus were slightly lighter, 2.76 gm. 2 Dicaea, near Abdera, had also adopted this denomination. 3 With the tetrobols of about 2.80 gm, diobols were used, a denomination that makes up the bulk of the extant coins of Sparadocus, although both he and Seuthes, along with Aenus, struck coins of quarter stater weight.

It has been thought hitherto that the coins I have called tetrobols were Persian hemidrachms and indicated growing trade with Asia Minor. Several points are to be urged against this view of the origin of these coins on the Thracian coast. They were first adopted, so Strack thinks, at Abdera, 1 the farthest of the Thracian cities from Asia, and the weight there used showed the Persian standard in a high form. As a matter of fact fractional coins usually do not reach the weight demanded by the standard to which they belong. Thus while these small coins would be heavy as Persian hemidrachms, they would be about right for Attic tetrobols. Because the staters of Aenus followed the light Attic standard, Strack realized this was true for Aenus

A further reason for thinking the Abderitan coins were Attic tetrobols is the date at which they were struck. Minted as early as 440 (450 according to Strack) they would fit into the Attic standard, then extending its sway over the Aegean, and since they conform to the Thracian practice of dividing by three, their appearance at Abdera is natural. On the other hand, if they had been first minted toward the end of the century when the Attic standard was receding, the same coins would probably be Persian. So much depends upon circumstances.

In view of the popularity of the Attic tetrobols and diobols, one may ask whether Maroneia had conformed to the practice of her neighbors in this respect as well. Having seen no discussion of the question as to what denominations were being used in Maroneia during the last third of the century, I have tried to classify according to dates the chief fractional currency of Maroneia. Drachms of the old Phoenician standard which can be dated with any probability about 430 are very rare. Moreover, there was a reduction of these drachms until they reached the weight of a quarter of the Attic tridrachm stater. Whether coins weighing about 3.20 gm 1 continued to be minted under the new standard is very difficult to say. There are so few of them. Even the coins with the older type, ram's head, were scarcely higher in weight. 2

In types there is no break, however, between the last drachms weighing about 3.20 gm (Plate XVI, 161) and the first of the Attic tetrobols, ca. 2.80 gm (Plate XVI, 162a). They have the bunch of grapes and the fore part of a horse both before and after the reduction in weight, and except for weight, the first of the new weight is hardly distinguishable from the last of the old. 3 There are three possi- bilities as to the date of the last drachms of reduced Phoenician weight (ca. 3.20 gm), first, that they coincide in point of time, as they do in weight, with the reduced stater of ca. 12.50 gm; 1 second, that the Attic tetrobol (ca. 2.80 gm) was introduced with the Attic tridrachm. The third possibility would allow us to assume that the Attic tetrobol replaced the slightly heavier drachm in the middle of the period when the Attic tridrachm stater was being minted. In other words, Maroneia minted a few quarter staters of the Attic tridrachm standard and then discontinued the denomination in favor of one which would have a wider currency.

The first assumption has certain advantages. Staters and quarter staters would have provided a well balanced system, and there would have been no stater about four and a half times the weight of the chief smaller coin, as there was at Abdera and as there would have been at Maroneia if Attic tetrobols and tridrachms were minted contemporaneously. Furthermore, the introduction of the coin of about 2.80 gm at the same time as the Persian tridrachm, ca. 404, would be explicable, for the small coin, which we may now call a Persian hemidrachm, would be one-sixth as heavy as the stater.

There are reasons, however, for rejecting this solution of the problem. The coins weighing about 3.20 gm are as few as those weighing ca. 2.80 gm are plentiful, and it would seem as though the large number of issues of the latter make it necessary to assume that they were minted from about 430. There are two main varieties, those with magistrates' names in full, with shallow fourth century incuses, and those which have only abbreviations or no names at all.

The first class can be dated more or less precisely, for the names which are found on this group of triobols are in general the same as those which appear on Persian staters. A study of the Persian staters shows that Athenes and Posideios were the first magistrates to issue coins of this weight. Moreover, they share obverse dies and were possibly colleagues. They seem also to have shared obverse dies in the extant triobols, and these triobols judged by style and fabric, together with one of Metrodoros 1 (Hunter, p. 380, No. 7) are the first of the series to bear the full magistrates' names. Thus the first use of the full name on the smaller coins cannot be later than the adoption of the Persian stater. Moreover, a fairly large proportion of the magistrates who struck staters also struck triobols. They are Athenes, Posideios, Zeno, Noumenios, Eupolis, Metrodoros, and Heracleides. Only one name, Aristoles, cannot be identified with a magistrate known from the staters. Since this group of coins forms a homogeneous unit with the Persian staters, the triobols with and without initials must antedate the adoption of the Persian standard.

They have incuses which leave little question as to their fifth century date. Some of them have symbols together with initials, others initials alone, others merely the abbreviated name of the city. The coins which have neither symbol nor magistrate's name resemble some of the Attic tridrachm staters, and they are probably contemporary with them.

To the list of chief varieties which follows, I have appended references to representative coins, including those which are illustrated on Plate XVI.

The identification of the magistrates who struck these coins is for the most part impossible, but it is probable that HP is ΗΡΟΒΟΛΟΣ or ΗΡΟΦΙΛΟΣ, 1 and MHT either Metrodoros or Metrophanes. Possibly ΑΘΝ is Athenes. Except for ΗΡΟΒΟΛΟΣ, these are magistrates who struck Persian tridrachms. The other initials cannot be made to fit any known magistrate of the period 430–395. 2 When we add the issues of the unknown magistrates to the separate contemporary issues of staters, assuming that they were struck in years when no staters were issued, or that the staters have not been preserved, we reach a date not far from 430 for the introduction of this denomination. In other words it was probably contemporary both with the Attic tri- drachm stater and with the Persian tridrachm stater.

I have found weight records for ninety-one specimens of these coins, whereas of the quarter staters of about 3.20 gm (Plate XVI, 161) I know only five. 1 Thus the latter was comparatively unimportant, and if it was minted after the introduction of the Attic tridrachm stater, which is not impossible, its use was soon discontinued.

I append here a table showing the distribution of weights for the Attic tetrobols (Persian triobols) which do not have the name of the magistrate in full. 2

I have grouped them into approximately eleven issues. The average of eleven high weights is 2.79 gm, and their range is from 2.92 gm to 2.62 gm.

For the sake of comparison I give here the known issues of Persian triobols on which the name of the magistrates appear in full. Under each issue I list the heaviest specimen known to me and the piece illustrated on Plate XVI.

The average weight of the eight heaviest specimens is ca. 2.70 gm. As the norm of the Maroneitan stater of Persian weight was approximately 10.90 gm, there can be no doubt that the smaller coins were used as triobols of this standard. They are distinctly lighter than the tetrobols of the preceding period, as the following table shows.

A few specimens of a smaller denomination have been published. Some of them have a cluster of grapes as reverse type. Others have a tripod. In general they resemble the coins which I have called tetrobols. The following specimens are known to me.

Without inscription and with cluster of grapes.

With MAPΩN and cluster of grapes on reverse.

Letter H on obverse. Reverse as above.

MA and cluster of grapes on reverse. Symbol: ivy leaf.

MAPΩN and tripod on reverse.

Since these coins parallel the tetrobol series and are approximately half as heavy, they can be called diobols.

Two other pieces which are similar to the tetrobols, although of different weight, remain to be considered. One of them, a piece in the Hermitage weighing 1.85 gm, seems to be inscribed MHT on the obverse. The other (Spink's Circular, March 1914, 179), weighing 1.81 gm, has AΘN. On the reverse there is an amphora symbol and the letters MA. Thus these pieces, despite their weights, are probably contemporary with the tetrobols inscribed MHT and AΘN. I have already suggested that these abbreviations were used by Metrodorus, or Metrophanes, and Athenes, officials whose names appear on the heavy Persian tri-drachms. In confirmation of this hypothesis it may be noted that the coins weighing about 1.85 gm are approximately ninths of the Persian tridrachm stater.

In conclusion, it is not impossible that Maroneia struck quarter staters for a short time after the introduction of the Attic tridrachm stater. Then she introduced the Attic tetrobol and diobol which became Persian half and quarter obols when the Persian tridrachm stater was adopted. Finally, an experimental Persian diobol was minted, but in the fourth century when the Persian standard in its regular form was introduced, coins smaller than the half drachm were not issued.

Now that we have passed in review the silver coinage of Maroneia up to about 410, it is possible to turn our attention to a contemporary gold coin, B. M. C. Thrace, p. 233 (Num. Chron., III, p. 109), cf Pl. 1, A. There is a specimen better preserved in Paris, Pl. 1, B. This issue is very close in types to the Attic tridrachms. It has the same twisted vine characteristic of these coins, the name of the city about the linear square found on this series alone, and a horse almost identical with that of B. M. C. Thrace, p. 234, 14 β (Pl. x1, 61). It is possible too that the symbol above the horse, a bunch of grapes, not found elsewhere on Maroneitan coins, appears during this period, but even though it does not, 1 there can be no doubt as to the date of our gold piece. It belongs between about 430 and 410, and nowhere else.

In weight these coins are peculiar, for they do not fit into any ordinary standard. The better preserved of the two weighs 3.32 gm, while the other weighs only 3.14 gm. Certainly this is not an Attic tetrobol, being far too heavy for that non- existent Athenian coin. It is likewise too heavy for a third of a daric. Nor does it have a place in the Persian silver standard used in Maroneia after about 395.

It must be clear from the weight, that in minting gold, Maroneia used a standard derived from that of her silver coins. These gold pieces are quarter staters even though the Paris specimen is rather heavy, and though the majority of the silver staters are somewhat too low. Nevertheless, if we compare them with the heaviest specimen of Maroneia or Abdera, ca. 12.90 gm, the correspondence is sufficiently exact. The perfect Paris specimen is now a trifle too heavy and the worn British Museum piece is a trifle light.

Now that we have found Maroneia striking quarter staters in gold, possibly we should assume that the silver coins weighing about 3.20 gm were contemporary with it; but of this we cannot be certain. Moreover, when we come to relate the gold coin of Maroneia to the silver coins, tridrachms and tetrobols, we find that the three classes fitted precisely into one another and combined to form a most convenient monetary system.

With gold thirteen and a third times the value of silver, a gold coin would exchange for three and a third silver coins four times as heavy. Since the gold coin is a quarter stater in all probability, the exchange situation would have been not at all satisfactory if there had been silver quarter staters, but as the stater was divided by nine and four and a half, the gold coin equalled fifteen Attic tetrobols and thirty diobols of Attic standard. In other words, with a slight variation we have the Phoenician system of relating gold and silver coins, one gold coin for fifteen silver, rather than the Persian system where one gold piece exchanged for ten silver pieces. This is a system to which Maroneia must have been accustomed, since her early fifth century standard was Phoenician, and her first coins, didrachms and drachms, not tetradrachms. Thus a daric would have exchanged for fifteen didrachms. In this attempt to apply the old customs of exchange to a new standard, the peculiar weight of the gold coin may have originated. The standard for the gold was new, but the principle was as old as the Phoenician standard itself. 1

It should be noted also that this gold coin works in well with nearly every standard used in Maroneia. It equalled two and a half Attic tetradrachms. Possibly the heavy weight is due to a desire to make it equivalent to this number of standard Athenian coins. It also equals three staters of the old Phoenician standard. Nor does its usefulness cease here, for it could readily find a place in the Persian system. If we reckon the Attic tetrobol as a Persian hemidrachm, the exchange value would be seven and one-half Persian shekels. But as we have seen, the Attic tetrobol is a little heavy for a Persian hemidrachm and our gold coin is of full weight. Thus the exchange would be more nearly sixteen to one, four staters for one gold piece. The weight of the gold coin was most convenient.

It is possible too that the gold coinage was a result of the revolt of the Thracian coast from Athens between 411 and 408. Then the Peloponnesian sailors would have found the weight of the gold coin familiar, for it is not far different from the Aeginetic standard. Money was certainly needed at that time to pay the sailors, and while gold coins of Aeginetic weight were not known, they would doubtless be acceptable to the allied forces.

While it is possible that other causes were responsible for the first minting of gold at Maroneia, e. g., Odrysian intervention, we can hardly date this issue of gold much later than 410. Between 408 and 405 Athens was in control again, and Athens was not likely to allow free minting of gold in that period. After 404, possibly even as early as 411, we have the adoption of another standard by Maroneia, the so-called "Light Attic." To the stater of this weight I have given the name Persian tridrachm. Judging from the number of magistrates and types known, this series lasted about ten years to be followed by the Persian standard.

The following list of issues of the Persian tridrachm is based on Regling 1 with additions and a change of order necessitated by the fact that the first coins of Persian weight issued by Athenes imitate the reverse of his heavier staters. Thus the group of coins issued by Theodotos, with its single cluster of grapes on the reverse, is apparently the earlier.

Fig. 9

Analysis of these coins is difficult because of uncertainty as to the genuineness of many specimens. This series has been a fertile field for modern forgers, and Regling 1 has been inclined to consider spurious coins of Athenes and Erophilos, although he does not question the existence of an authentic prototype from which imitations of Erophilos' coins were made. 2 Eliminating these doubtful issues and two specimens of extraordinarily light weight, the coins group themselves as follows:

Thus the norm is not far from 16.20 gm. Nor would the distribution be greatly affected if we included the suspicious specimens. There would be eight weights grouped about 16.20 gm and six above 16.50 gm, and an average of the high weight specimens would be slightly above 16.60 gm. Without the suspicious specimens the average of three weights is 16.51. This standard has been dubbed Light Attic for want of a better name, but it is easy to see that it is very light Attic even if we take the highest of the weights given above.

Moreover, whenever the standard was adopted, whether in 411, when Maroneia must have been in revolt if she struck coins, or in 404 after the defeat of Athens, it is unlikely that Maroneia would have taken the Attic standard, for the Persian standard was either then supplanting, or on the point of supplanting, the Attic along the coast. It will be no surprise to learn that a coin weighing 16.60 gm was about six times as heavy as the small coins that Maroneia was then minting. We can now call them Persian hemidrachms. We have seen also that Athenes and Metrodoros probably struck diobols of Persian weight. Now the norm for the Persian staters which replaced the so-called light Attic in the fourth century was about 10.90 gm (average of high weights = 11.13 gm). The correspondence is as close as one could wish or expect. The so-called light Attic stater was a Persian tridrachm, just as the so-called Aeginetic stater was an Attic tridrachm. But this was only an experiment in an age and region of numismatic experiments, and the normal Persian standard soon was adopted both at Maroneia and Abdera.

With this Persian tridrachm stater went a gold coin, as before, about one quarter as heavy as the stater, 4.01 gm. 1 But in this case the standard is not new. Our gold coin was probably intended to pass for half a daric, a little light of course, although not so light as it would be for an Attic drachm. Consequently, whatever the date of this issue may have been, there is no need for considering that the Athenian issue of 406 was its forerunner as to weight.

Our coin, at the ratio of thirteen and a third to one, the usual ratio in Asia, would be worth twenty of the small coins now minted in Maroneia as Persian hemidrachms. Possibly the region called them drachms instead. While the gold coin would equal three and a third of the tridrachm staters, the exchange would be simple because of the custom of dividing by six. Of course the exchange with Persian staters would have been simple, but the type of our gold piece probably indicates that it was contemporary with the Persian tridrachm stater.

The obverse of the silver staters shows the unbearded head of Dionysus. On the gold coin we have the bearded head of this god. This differentiation of types between denominations is most natural and not at all unheard of. In fact, at Thasos the bearded god was used together with an unbearded head. 1 Since the god Dionysus replaces the horse as obverse type on the silver coins of Maroneia only in this period, we need not hesitate to date the gold piece about the turn of the century before the ordinary Persian standard was adopted there.

Our study of this issue of gold coins brings us again to the fortunes of the Odrysian empire, for it is at least possible that Maroneia's relations with the Odrysian princes were responsible either for one or both gold issues.

We know that the empire began to disintegrate in the time of Seuthes, 2 while Strack has shown that about 412 as a result of this disintegration Aenus revived and experienced a burst of prosperity such as she had not known for nearly thirty years. 1 The destruction of the Odrysian power was clearly of greater moment in the affairs of Aenus than the break-up of the Athenian empire. This was true to a certain extent at Maroneia also, for the abundant coinage of Maroneia indicates that the city prospered about 400. That Maroneia was more closely connected with the Odrysian dynasts than was Aenus can be seen from a study of the coins of some of these dynasts, struck either at the Maroneitan mint, as has been supposed, or more probably under its influence.

The chief cause for the decline of Odrysian power was the division of territory among princes presumably of royal blood and their almost continuous struggles for independence or supremacy. 2 Though it might be taken for granted that the Greek cities were not passive spectators of these struggles, since the success or failure of a prince might mean loss, or renewal, of trade or continued stagnation, yet it will be helpful to consider the scattered evidence and to see how the Greek cities acted as patrons for rival princes, possibly supplying them with money, allowing them the use of their mints, and serving as general headquarters for their expedi- tions against rival claimants. Besides the coins of Amadocus and Teres, which I shall discuss later, we have a fourth century coin of a certain Spokes, minted at, or under the influence of, Abdera, 1 when Maroneia and Abdera were probably the patrons of rival dynasts. This rivalry between Maroneia and Abdera is known from Diodorus and a scholion to Aristides, 2 which tells us of a joint Triballian and Maroneitan attack upon Abdera in alliance with the Thracians (Diod. XV, 36), and of the reconciliation of the kings of these two cities by Chabrias, in 375 perhaps. Of course, the cities had no kings, but it is easy to see that the scholiast confuses the royal protégés of the cities with kings actually exercising authority in them. Confirmation of the scholiast as to the general state of affairs, though not as to the facts of the year 375 for which he vouches, is to be found in the statement of Xenophon, 3 that Thrasybulus reconciled the two princes, Seuthes II and Amadocus, in the expectation that a settlement of their quarrels would cause the Greek cities of the Thracian coast to be more friendly toward Athens.

That Aenus was directly involved in Thracian affairs may be inferred from the bipennis which appears as a symbol on its coins just before 412, 1 about the time that Maesades, a petty prince whose territory extended from the Hebrus River to the Propontis, was driven into exile, and his territory was severed from the Odrysian empire. 2 This event opened up the land route to the Black Sea for Aenus, and assured her a new prosperity. 3 Later during the fourth century, when Cotys 4 (383–360/59) had consolidated his power, Aenus suffered economic decline, and two of her citizens murdered the Odrysian king, to avenge their father, as the story is told. This private feud is quite probably connected with political and economic rivalries, the details of which we do not know. Finally, in 342, the Greek cities along the coast, Aenus undoubtedly among them, called upon Philip for assistance against Cersobleptes, 1 one of Cotys' successors.

Thus it is clear enough that the Greek cities, to retain their independence, adopted the policy of fostering dissensions and disintegration within the Odrysian Empire. Too little is known about these struggles to warrant going into detail but I shall try to reconstruct as much of the general outline as concerns Maroneia.

We begin with Metocus, or Madocus, a prince, who, apparently with the friendly assistance of Maroneia, was able to keep and seize the throne after the death of Seuthes I. He is known as the Odrysian king in 405–4 (Diod. XIII, 105), and the territory over which he ruled directly was in the district inland from Maroneia, a kingdom consider- ably smaller than that of his predecessors. 1 A coin of Metocus shows by its type the friendly relations between him and Maroneia. 2 On the obverse is a bearded head, probably Dionysus, which suggests that it was struck when Maroneia was using the Persian tridrachm standard. In fact the coins of the Odrysian princes, Metocus, who is probably Amadocus, and Teres, show a progression in types parallel even as to details with that of Maroneia.

We can hardly doubt that Maroneia gave Metocus active support in return for commercial privileges, and knowing the incessant struggle necessary to gain and keep the Odrysian throne at this time, we must not minimize Maroneia's efforts. In as much as Metocus had a fairly firm hold on the throne about 400, it is probable that Maroneia rendered its most valuable assistance to Metocus when he gained the throne after the death of Seuthes I, in the last fifteen years of the century.

Maroneia was fortunate in the choice of her prince, for Metocus kept the throne for at least fifteen years and possibly even for twenty-five. He may have been succeeded by Amadocus, called Metocus by Diodorus, who was ruling about 389 when Thrasybulus acted as peace-maker between him and Seuthes, hoping thereby to bring about better relations between Athens and the Greek cities of the coast. 1

There is an element of confusion at this point, which makes it difficult to tell whether Amadocus and Metocus are variants of the same name, and if the names are to be considered identical, whether there were one or two princes of this name at the beginning of the fourth century. The evidence is as follows. A Metocus is known at the end of the fifth century from the writings of Xenophon. He struck silver coins under that name. At the beginning of the fourth century, Xenophon speaks of an Amadocus, and we have bronze coins with that name. Did the Metocus of the silver coins use a different spelling on his bronze issues? Diodorus, who calls the Amadocus of Xenophon by the name of Metocus, does not throw light on the problem except as he tends to show that there were two forms of the name.

While I am inclined to the view that Xenophon cannot be followed in distinguishing Metocus from Amadocus, I do not feel that the evidence is conclusive, or that the identification of Amadocus and Metocus should result in dating the bronze coins of Amadocus about the middle of the fourth century. 1

The date of the bronze coins of Amadocus is easily ascertainable by comparison with the coins of Maroneia. They show two varieties of reverse types, one the single cluster of grapes found on the early Persian tridrachms, almost identical in size with the bronze coins of Amadocus, and the vine with several clusters—five or seven. A mere description of the latter type gives no idea of the faithfulness of the reproduction of the vine with five clusters on the coins of Apelles (Pl. XIV, 102a), a fourth-century Maroneitan mint-official. Since the coins of Apelles are unique in the character of the vine on the reverse and exceptional in the absence of a symbol, they are not easy to place definitely within the series of Persian staters. Still, they seem to be early rather than late, i. e., much closer to 390 than to 350. That the bronze of Amadocus was also struck in the early fourth century is confirmed by the unusual thickness and the recht altem Aussehen of the coins. 1 But since the dates of the Odrysian dynasts are fixed by chance references in Xenophon, we can be more precise about the adoption of the two forms of the Persian standard at Maroneia.

Metocus was on the throne in 405–4, 2 and in 400. 3 His coins, as we have seen, are contemporary with the Persian tridrachm stater. Therefore, that was in use about 400. The coins with the name Amadocus are contemporary with both tridrachm and didrachm Persian staters. A man called Amadocus was on the throne in 389. Therefore, the tridrachm stater ceased to be issued between 400 and about 390—let us say, about 395. If it was in use ten years, it began as early as 404.

We can be more precise. Amadocus imitated the cluster of grapes on the Persian tridrachm stater, and this reverse type was apparently used on the first issues of this standard in Maroneia, Pl. XII, 65, 66a-c. The evidence is to be found in the coin of Athenes. The reverse of his first Persian staters, Pl. XIII, 76a, 77, are exact reproductions of his tridrachm, both as to inscription and details of the vine, Pl. XII, 74. They form the link between the two standards. Only one Maroneitan magistrate used the single cluster of grapes for a reverse type, and his coins come presumably at the beginning of the Persian tridrachm series. Thus, the name Amadocus was used on Odrysian bronze coinage from about 405 to about 386, the year when Hebryzelmis is known to have been on the throne. As Metocus was the Odrysian king in 405 and again in 400, the evidence of Xenophon and the coins is mutually corroborative. It points to the identification of Metocus and Amadocus.

While we are not told who was supporting Seuthes when Thrasybulus reconciled him with Amadocus in 389, the coinage of Amadocus shows where the latter obtained his support. But after Amadocus, the seat of Odrysian power passed to the east, as the coins of the later Odrysian monarchs show by their use of the type of Cypsela, an inland town on the Hebrus River. 1 Hebryzelmis, the first of them to use the mint of Cypsela, held a princi- pality in 386–5, 1 and this may or may not give us the lower limit of the reign of Amadocus. Whether Hebryzelmis was the interpreter of Seuthes II is uncertain. 2

There is one name that we have been unable to fit into the outline. A Teres also used the reverse type of Maroneia, somewhat similar to that which Amadocus took over from Apelles' coins. This identity of type between Maroneia and the Odrysian princes, taken in connection with the appearance of magistrates' names about the linear square on the reverse, has been interpreted as proof first that they were struck at Maroneia and secondly that the city at the time was subject to them. But neither conclusion is necessary.

Assuming for the moment that the Maroneitan mint was used, let us consider the question of overlordship here, for it has an important bearing on the date of the coins. In the first place, the use of the types of Greek cities was not uncommon in this region. Saratocus, Ketriporis, and Bergaeus —if we consider him a dynast rather than a town, as suggested by Svoronos—used Thasian types. They were not lords of Thasos. The coin of Spokes, minted in imitation of Abderitan coins, is more to the point, for it bears the name of a monetary official exactly as do the coins of Amadocus and Teres. Thus even though the Maroneitan mint may have been used, there would be no proof of overlordship. 1 Furthermore, so far as is known, the names of the monetary officials which appear on the coins of the Odrysian dynasts are not those of Maroneitan officials, of which we have a long list preserved.

When Wroth ascribed these coins to the dynasts Amadocus II, 359–351, who jointly with Cersobleptes succeeded Cotys, and Teres III, ca. 350, partly because neither Amadocus I nor Teres II is known to have possessed Maroneia, 2 the city in which he thought the coins were struck, while, as he said, Amadocus II ruled over a part of the coast between Maroneia and the Chersonese, I think he misconceived the relations between the Greek cities and the dynasts, forgetting that the Greek cities would undoubtedly find it profitable to subsidize friendly monarchs in order that they might preserve their independence against rival princes who might otherwise disturb the balance of power and gain control of their territory. Furthermore, the passage in Demosthenes 1 which suggested to Höck 2 and Wroth that Amadocus was ruler in Maroneia, implies rather that his authority did not extend to the territory of that city. Amadocus could prevent Philip from advancing beyond Maroneia toward the Chersonese, but he could not prevent him from entering the city, nor did he attempt it, so far as we know.

Thus even if the coins are rightly ascribed to Amadocus II, which I question, I think it probable that at the most the city loaned him the use of its mint as sponsor and ally, rather than as subject. 3 But the more we insist upon the actual use of the Maroneitan mint by Amadocus, the more emphasis we shall have to place on the details of the type which point to a date about 390, not to Amadocus II who ruled between 359 and 351.

But if the possibility is admitted that Maroneia was sponsor for Amadocus II, it must be admitted also for Amadocus I and Teres II. As for Teres III, there is no evidence whatever, aside from the coins that have been ascribed to him, that he had any connection with Maroneia. In fact, we cannot be at all sure of his existence, and although it would be rash to say that there was no Teres III, still Schaefer's conjecture, 1 based partly on a statement of Theopompus as to the longevity of an Odrysian king Teres, 2 that the Teres of Xenophon 3 lived to the ripe old age of ninety-two years and was the man mentioned in Philip's letter, 4 is as reasonable as Höck's hypotheses. 5 But even if Höck's conjectures are true, they cannot be used as evidence that the coins we are discussing were struck by Teres III after 351, when the coins themselves suggest a much earlier date. Nor did Höck ascribe these coins to Teres III.

Finally, while we know from an Athenian inscription, ca. 357, 6 that certain cities named in a part of the inscription now lost were tributary to the Odrysian princes, Berisades, Amadocus II, and Cersobleptes, we must infer that the obligations of these cities were limited chiefly to the payment of tribute. At the same time, the unnamed cities were apparently members of the Athenian confederacy, for the Thracian rulers gave assurances that Athenian interests would be respected and defended if members of the Athenian confederacy on the Thracian coast should attempt to rebel. As Maroneia was a member of the confederacy, it may have been among the unknown Greek citiestributary to the Odrysian empire about 357, i. e., while Amadocus II was ruling. But certainly it possessed complete local autonomy at that time. Hünerwadel has adduced further reasons for thinking that the Greek cities of this coast were not subject to the Odrysian rulers. 1 He has shown also how fertile a field the Odrysian dynastic struggles provided for outside intervention, 2 although he apparently did not recognize how clearly the evidence points to almost continuous intervention by the Greek cities closest at hand, the very states that are supposed to have lost their independence. They intervened to preserve it, recognizing the Odrysian rulers as their nominal overlords only when the latter were strong enough to collect the tribute which they considered their hereditary right.

Thus it is of little historical significance whether the coins of Amadocus and Teres were minted in the city of Maroneia, or whether the close connections between Maroneia and the Odrysian princes-resulted in the adoption of Maroneitan types and the use of Maroneitan engravers, as I am inclined to think.

Returning now to Teres, we find a reference in Xenophon to a petty dynast named Teres under Seuthes and Metocus ruling the Thracian Delta, a district near Byzantium. 1 Just what a ruler of this district had to do with Maroneia is not clear, but Höck's suggestion is possible, viz., that Teres was driven from his throne at the same time that the territory east of the Hebrus River belonging to Maesades revolted and was separated from the Odrysian Empire. 2 If so, he probably took refuge in Maroneia, then or later, and sought the support of Maroneia for his return. We know that Seuthes II, the son of Maesades, sought refuge with Metocus until he was old enough to undertake the reconquest of his father's territory, 3 and the fact that Seuthes' right hand man was a Maroneitan citizen suggests that during the time of Seuthes' visit with him Metocus lived at, or very near to, Maroneia. 4 The fortunes of Teres may have been simi- lar. At any rate, the coins of Teres are to be dated early in the fourth century about the same time as those of Amadocus.

It is now clear that Maroneia showed a direct and continuing interest in Odrysian affairs. I think it is also clear that much of her prosperity was due to the friendly assistance she had given to Metocus, the first Seuthes' successor. Probably during his struggles to gain the throne Maroneia struck the gold coins that have the obverse type found on the coins of Metocus. Possibly even her first issue of gold coins belongs to the first years of the reign of Metocus. Since it was almost a matter of life and death for the city that a friendly prince should rule in the hinterland, Maroneia undoubtedly devoted every resource of the state to securing her avenues of trade and prosperity. Money would have been needed, and as happened elsewhere in times of crisis, Maroneia may have resorted to the issue of gold coins to place and secure Metocus on his throne.

We have now two possible occasions when Maroneia may have needed to mint gold, the revolt of the Thracian coast from Athens and the dynastic rivalry after the death of Seuthes, one certainly before 408, the other uncertain because the date for the death of Seuthes is unknown. 1 While the discovery of a fitting occasion for the minting of gold does not prove that the gold was minted then, the close chronological agreement between the Maroneitan coinage and the Odrysian struggles, taken, together with the other points we have discussed, makes it very probable that one or both of the gold issues at Maroneia are connected somehow with Odrysian affairs. The first may or may not have been minted as an incident of the revolt against Athens, 1 but it undoubtedly antedates the adoption of the new standard about 404, thus coinciding very nearly in point of time with the revolt.

Maroneitan Staters of Persian Weight

The following catalogue of light Maroneitan staters (Persian standard) does not pretend to be exhaustive. It contains only such specimens as I have noted in the course of this study. I have attempted to arrange the issues in chronological sequence, but at best the order is merely tentative. To those specimens which I have seen or of which. I have seen casts or illustrations I have given numbers indicating obverse and reverse dies. Since an exhaustive study was not essential for my purpose, I have not attempted to get casts of many pieces listed below.

The obverse type is ordinarily a bridled horse left. The reverse contains a grape vine with four large clusters of grapes. When the horse faces right or is unbridled, I have indicated it in the following list. Likewise I have noted changes in the number of clusters on the reverse.

Serial No.	Die Nos.		Weight
		ΜΑΡΩ-ΝΙΤΩΝ-ΕΠΙΑ-ΘΗΝΕΩ No symbol.
76.	1–1	a. Berlin	10.28
		b. Copenhagen	10.92
77.	2–1	Berlin	10.96
78.	…	Hoffman Sale 230	…
	ΕΠΙ-ΑΘΗ-ΝΕΩ	Symbol: head of grain on reverse.
79.	3–2	a. Empedocles, ex Pozzi 1049, ex Hirsch XIII, 584	11.20
		b. Br. Mus. ex Weber 2333	11.29
80.	1–3	Commerce	…
	ΕΠΙΠ-ΟΣΙΔ-ΕΙΟϒ	Symbool: head of grain on reverse
81.	1–4	a. Berlin	10.85
		b. Br. Mus. Cat. 28	11.10
82.	3–5	Berlin	10.81
	ΕΠΙΚ-ΑΛΛ-ΙΚΡΑ-ΤΕΟΣ	Symbol: dog on obv. below horse.
84.	4–6	a. Newell	10.63
		b. Copenhagen	11.23
		c. Jameson 1061	11.07
		d. Mionnet I, 389, 166
85.	4–7	Berlin	10.54
86.	5–6	Vienna	10.50
87.	5–8	Munich, ex Helbing, 1913, 228, ex Hirsch XXI, 906	11.20
88.	6–9	Br. Mus. (Casson, fig. 14c)	11.21
89.	…	Hunter, p. 379, 3	10.80
	…	H rsch XII, 141
	EΠI-IKE-ΣIO	Symbol: thrysus on reverse.
90.	7–10	a. Vienna	10.57
		b. Hermitage	10.37
91.	8–11	Berlin	10.72
	EΠI-IKE-ΣIO	Symbol as above. Legend: MAPΩ on reverse below the horse.
92.	9–12	a. Hirsch XII, 139	….
		b. Hirsch XXI, 904	10.90
		c. Helbing, 1913, 227
		d. Br. Mus. Cat. 25	11.08
93.	….	Egger, Verzeichnis 49, 84	14.10 (sic)
	ΕΠΙ-ΙΚΕ-ΣΙΟ	Symbol as above. Legend: MAPΩ on reverse above horse.
94.	10–13	a. Copenhagen	10.90
		b. Br. Mus. (Guide 21, 5)	10.97
96.	…	Hirsch XIII, 583	10.59
	ΕΠΙΗΓΗΣ-ΑΓΟΡΕΩ	Symbol: eagle facing left on obverse above horse.
97.	11–14	McClean 3957 (Pl. 144, 9)	10.90
98.	11—15	Berlin	10.61
99.	….	Prokesch Osten, Num. Zeit. IV, p. 202, 11	9.76
	ΕΠΙΗΓΗΣ-ΑΓΟΡΕΩ	Symbol: flying eagle on obverse above horse.
100.	12–15	Vicomte de Sartiges 171, ex Hirsch XII, 138
101.	11a-15a	Br. Mus. Cat. 22	10.65
	ΕΠΙ-ΑΠΕ-ΛΛ-ΕΩ	No symbol.
102.	13–16	a. Newell	10.65
		b. Br. Mus. 20	10.30
		c. Br. Mus. ex Bunbury 592	10.83
		d. Berlin	11.12
		e. Berlin	10.95
		f. Hermitage	10.90
		g. Imhoof-Blumer 27	10.50
		h. Naville IV, 482 ex H. Weber 2334	10.53
		i. Pozzi, 1046	10.89
		j. Egger XLI, 265	10.43
	ΕΠΙΝΕΟΜΗΝΙΟ	No symbol.
103.	14a-17a	Mionnet, I, 389, 167
	ΕΠΙ-ΝΕ-ΟΜΗ-ΝΙΟ	Symbol: Bucranium on obverse above the horse.
104.	14–17	a. Copenhagen	11.41
		b. Berlin	11.30
		c. Pozzi, 1048	10.98
		d. Jameson 2018	11.03
		e. Naville XIII, 601	11.21
104A.	14b-17b	Br. Mus. Cat. 23 (Fig. 10)	10.89

Fig. 10

	ΕΠΙ-ΧΟΡ-ΗΓ-Ο	Symbol: fly on reverse.
105.	15–18	Berlin	11.33
106.	15–19	a. Berlin	10.58
		b. McClean 3961, P1. 144, 13	11.30
107.	16–20	Vienna (Fenerley Bey 264)	10.76
108.	17–21	Helbing 226	10.70
109.	18–22	Hirsch XVII, 672, ex Hirsch XII, 140
110.	18–23	Hirsch XXXII, 418	10.54
111.	19–24	a. Newell ex Hirsch XXV, 182, ex Hirsch XVII, 673	10.85
		b. Copenhagen	11.00
112.	….	Egger XLI, 264	….
113.	….	Collignon (Feuardant) 167
114.	….	Hirsch XXI (Weber) 905	10.63
115.	….	Br. Mus. Cat. 26	10.83
116.	….	Boston, Perkins 190	11.06
	ΕΠΙ-ΧΟΡ-ΗΓ-Οϒ	No symbol (Possibly barbaric imitations).
117.	20–25	a. Hermitage	10.10
		b. Hermitage	8.99
	ΕΠΙ-ΧΟΡ-ΗΓ-Ο	Symbols: fly on reverse, flying owl on obverse.
118.	….	Mionnet, Supp. II, 336, 817	….
	ΕΠΙΠ-ΟΛϒΑ-ΡΕΤΟ-ϒ	Symbols: scorpion on reverse, flying owl on obverse. Three clusters of grapes.
119.	21–26	Newell	10.46
120.	22–26	a. Berlin	10.95
		b. Br. Mus. Cat. 27	11.35
121.	21–27a	Mionnet 1, 389, 168
	ΕΠΙΠ-ΟΛϒΑ-ΡΕΤΟ-ϒ	Symbols: turtle on reverse, flying owl on obv. Three clusters of grapes.
122.	21–27	Vienna (Fenerley Bey, 263)	10.60
	ΕΠΙΠ-ΑΤΡ-ΟΚΛ-ΕΟΣ	Symbol: flying owl on obverse.
123.	21–28	Berlin	10.78
	ΕΠΙΠ-ΑΤΡ-ΟΚΛ-ΕΟΣ	Symbol: thunderbolt on obverse.
124.	23–28	a. Hermitage	10.70
		b. E. P. Robinson, Newport	10.69
		c. Copenhagen	10.60
	ΕΠΙ-ΜΗΤ-ΡΟΔ-ΩΡΟ	No symbol. Horse right.
125.	24–29	a. Berlin	10.91
		b. Empedocles ex Soth. (White King) 1909, 123	11.21
126.	25–30	Br. Mus.	11.01
127.	….	Hirsch XIII, 585	10.78
128.	….	Num. Chron. 1841, 109, 3	10.92
	ΕΠΙ-ΒΟϒΤΑ	Symbols on reverse: cantharus at left, thyrsus at right. Horse right.
130.	26–31	Empedocles	10.91
131.	26–32	Copenhagen	10.84
132.	26–32a	Mionnet, I, 390, 173
	ΕΠΙΜΗ-ΤΡΩΝΟΣ	Symbol on reverse: plow at left. ΜΑΡΩΝΙΤΩΝ below horse. Horse right.
133.	27–33	a. Empedocles, ex Weber 2335	10.92
		b. Vienna (Imhoof-Blumer 26)	11.02
		c. Berlin	10.77
	ΕΠΙΜΗ-ΤΡΩΝΟΣ	Symbol and horse as above. ΜΑΡΩ-ΝΙΤΩΝ above and below horse.
134.	28–34	Copenhagen	11.00
	ΕΠΙΜΗ-ΤΡΩΝΟΣ	No symbol. ΜΑ-ΡΩ-ΝΙ-ΤΩΝ about linear square on reverse; name of official on obverse above and below horse to r.
135.	29–35	Naville XIII, 603	10.92
	ΕΠΙΕ-ϒΞΙ-ΘΕΜ-ΙΟΣ	No symbol. Horse right without rein
136.	30–36	a. Berlin (Imhoof-Blumer 28)	10.46
		b. Vienna	10.05
		c. Hermitage	10.07
		d. Berlin	10.01
		e. De Luynes 1778	11.05
		f. Copenhagen	10.99
		g. Egger XLVI, 250 ex Hirsch, XXXII, 419	10.92
137.	30–37	Vienna	10.70
	ΕΠΙ-ΠΟΛ-ϒΝΙ-ΚΟϒ	Symbol: trident below horse on obverse.
138.	31–38	a. Berlin	11.15
		b. McClean 3958, P1. 144, 10	10.83
		c. Athens 957	10.83
139.	32–39	Copenhagen	10.55
140.	33–40	Naville XIII, 602	11.10
141.	….	Br. Mus. ex Bunbury 592	11.22
142.		Berlin (Imhoof-Blumer 29; forgery)	7–51
	ΕΠΙ-ΕΥΠ-ΟΛΙ-ΟΣ	Symbol as above.
143.	31–41	a. Berlin	10.90
		b. Munich
144.	32–42	Naville I, 1047	10.58
145.	….	Br. Mus. ex Montagu	10.54
	ΕΠ-ΖΗ-ΝΩ-ΝΟΣ	Symbol as above.
146.	33–43	Munich
147.	34–44	Vienna	11.08
148.	34–45	Berlin	10.62
149.	35–46	Br. Mus. Cat. 21	10.56
150.	35–47	a. Berlin	10.10
		b. Hermitage	10.55
		c. Hermitage	11.47
		d. Copenhagen	11.50
151.	35–48	Newell	10.65
152.	35–49	a. De Luynes 1773	10.20
		b. McClean 3959, Pl. 144, 11	10.58
153.	…	Berlin	10.52
154.	…	Soth., 1904 (Gentleman) 186	10.57
	ЕΠI-НРA-KΛEI-ΔΟϒ Symbol as above.
155.	34–50	McClean 3960, Pl. 144, 12	10.57
156.	35–50	a. Baldwin	10.22
		b. Br. Mus. (Num. Chron., 1924, 5)	10.98
157.	35–51	Hermitage	10.55
	ЕΠI-ZН-NΩ-NΟΣ MA in place of symbol on obverse.
158.	36–52	a. Ward 426A	10.48
		b. Egger XLV, 452	11.00
159.	37–53	Hermitage	10.54
	ЕΠI-ZН-NΩ-NΟΣ Letter A(?) in place of symbol on obverse.
160.	38–54	Athens 954	Plated)

Persian staters

11.50 =	1
11.40–10.49 =	2
11.30–11.39 =	4
11.20–11.29 =	8
11.10–11.19 =	4
11.00–11.09 =	12
10.90–10.99 =	19
10.80–10.89 =	11
10.70–10.79 =	8
10.60–10.69 =	10
10.50–10.59 =	19
10.40–10.49 =	4
10.30–10.39 =	2
10.20–10.29 =	3
10.10–10.19 =	2
10.00–10.09 =	3
8.99, 9.76 =	2
	144
Norm ca. 10.90 gm

Average of 18 high weights (one for each magistrate) = ca. 11.13 gm

End Notes

1

Since we have no ancient authority for the revolt of Maroneia between 411 and 408, we do not know that Maroneia joined her neighbors. The same is true of Aenus, whose coins we are about to discuss. Nevertheless, the revolt of these two cities is not at all improbable, for the revolt was widespread. We would have known nothing of the revolt of Abdera except for a statement, about its recovery.

As no gold coins of Abdera are extant, we pass to Aenus, the most easterly of the cities minting gold on the Thracian coast and the one most likely to show Asiatic influence. Its numismatic history is comparatively simple. About 480 it began to strike coins on a so-called light Attic standard similar to that used in the Thracian Chersonese at the beginning of the fifth century 1 and to that used later at Maroneia toward the end of the century.

The standard became gradually lighter until about 412, when the coins show a marked decrease in weight. 2 Henceforth Aenus struck coins which were almost identical with those of cities using the Chian standard. In fact, Professor Gardner calls the standard of Aenus Chian. 3 But it is doubtful whether it is any more Chian than Persian, even though the weight of the stater accords with the Chian standard. We have seen that the tetradrachms of the light Attic standard were tridrachms of Persian weight. This is as apparent in the coins of Aenus as in those of Maroneia, for Aenus after about 440 was striking small coins, called by Strack tetrobols, 1 which are nothing but Persian hemidrachms and weigh exactly one sixth as much as the light Attic tetradrachms. Twelfths however formed the great mass of the currency of Aenus during this period when Aenus and Abdera were suffering economic decline through the closing of their trade routes by the Odrysian empire. There are only two known issues of drachms. Thus of the currency of Aenus between 440 and 412, all but two issues of drachms find a place in the Persian system.

In lowering her standard to "Chian" weight, Aenus, so Strack suggests, suited her currency to the needs of foreign trade, by making her staters conform to a tetrobol of about 2.60 gm. This coin with the diobol of about 1.30 gm was a great favorite in and about the northern Aegean during the first half of the fourth century and it played a great role in the coinage of Aenus between 412 and 376. 2 This international fractional currency was minted in the Chersonese, at Parium, Lampsacus, Cius, Calchedon, Byzantium, Mesembria, Apollonia, Abdera, and Maroneia. 3

There is thus a close connection between Aenus and cities that were using the Persian standard toward the end of the fifth century, particularly Byzantium and Calchedon. These two important cities struck Persian drachms of 5.18–5.44 gm (80–84 grs.) just about the time that Aenus was accommodating her currency to the Persian standard, Calchedon about 411, Byzantium probably somewhat earlier. Calchedon had previously used Chian drachms of ca. 3.95 gm (61 grs.). 1 The two cities were minting Persian hemidrachms as well. 2 Thus as Aenus issued no Chian drachms after 412, and only a few drachms of light Attic weight in the years immediately preceding, 3 we are certainly justified in considering the tetradrachms after 412, and probably even a little before, Persian tridrachms.

Turning now to the gold piece of Aenus weighing 2.11 gm (32.6 grs.) we find that this coin comes as near to being a coin of Attic weight as any we have considered. Nevertheless it is apparently under weight. As the coin stands today, it is the least bit heavier, .01 gm, than the norm for a quarter daric, though well within the range of weights exhibited by extant darics. 4

While this is not the place to discuss the standard of the fourth century gold coins of the Greek cities of Asia Minor, it is worth noting that the long series of Lampsacene gold staters, which Professor Gardner thinks were minted with direct reference to Athenian gold coins and followed their standard, 1 fall far short of Attic weight. The heaviest weighs only 8.56 gm and the norm is clearly under 8.45 gm. 2 Even though we take 8.45 gm for the norm, the Lampsacene gold stater can not be considered anything but a daric stater raised by a very small amount. But whether we estimate the difference between the daric as 1½ centigrams with Miss Baldwin or as 5 centigrams, we are figuring to a nicety almost impracticable.

The early fourth or late fifth century coins of Abydos possibly show the same high form of the daric standard, likewise the Ephesian gold coins. Thus the existence of a standard slightly heavier, one might almost say infinitesimally heavier, than the daric seems to be proved. 3 Possibly the coin of Aenus is to be classed with the staters of Lampsacus as adhering to this standard.

Nevertheless, if it were not for the close relations between Aenus and the cities which revolted from Athens in 411, cities using the Persian standard in one form or another, Lampsacus, Byzantium, Chachedon, Cius, Abydus, and Abdera, the weight of our coin and its worn condition might allow us to agree with Professor Gardner as to its origin and standard. There are other reasons, however, for thinking that imitation of Athens had nothing to do with the minting of gold by Aenus. Whether Aenus revolted or not, the town would have needed funds to cooperate with the rebellious subjects of Maesades for the purpose of freeing the Hebrus valley and the route to the Black Sea from Odrysian overlordship, and since Greek cities outside of Asia Minor usually struck gold only when their other resources were exhausted, 1 it is not at all improbable that the coin was minted some time between about 415 and 408. Strack has shown that Aenus was poverty-stricken during the height of Odrysian power. 2

Almost immediately after the death of Seuthes (ca. 412) and the expulsion of the petty prince Maesades from the district east of the Hebrus River, the prosperity of Aenus began to revive. That Aenus was instrumental in some measure in the liberation of the Hebrus valley is suggested by the obverse type of the town Cypsela, which now being freed from Odrysian domination began to strike money. The obverse type of its first issue 1 imitates the contemporary bronze of Aenus struck just before 412. 2

The reverse of the Aenus gold piece 3 is striking enough to attract the attention of numismatists, a terminal figure of Hermes on a throne. That this was a representation of a cult statue is clear, but I think no one has suggested that the use of the cult statue as a coin type, a choice most unusual in the fifth century, was an emphatic way of calling upon their patron god to bring financial aid to his impoverished protégés. The type was used on three distinct occasions in the history of Aenus, and each time it synchronizes with some crisis in the affairs of the city. As a symbol it is found on tetradrachms of light Attic weight struck while Aenus, although still prosperous, was on the point of losing control of the trade route to Apollonia through the growth of the Odrysian empire under Sitalces. 4 In the fourth century, just before it was incorporated in the Macedonian empire, when the Odrysian princes had a second time destroyed the city's prosperity, the enthroned Hermes became the ordinary reverse type of the drachms of Aenus. 1 Thus the use of the type on the gold coin may point to the revolt of 411 or a determination on the part of Aenus to throw off the economic yoke of the Odrysian rulers. 2

Much depends upon the date of this coin. If it was minted between 365 and 341, as von Fritze and Strack after careful study of the coins have concluded, 1 what I have said must be modified to suit the political exigencies of the fourth century. Moreover, the discussion of the fifth century coin standards of Aenus is no longer apropos. But the evidence for a fourth century date is not wholly convincing. Much can be said for the preceding century.

If one looks only at the reverse, the unusual type, the inscription unabbreviated and running downward in the field to the right of the type, the staff of Hermes as a symbol identical with that of a fourth century drachm, 2 all of these point to a date contemporary with the last silver coins of Aenus struck about the middle of the fourth century. 3 But if one regards the obverse alone, there will be no hesitation about placing it before 412, as von Fritze recognized. The head of Hermes in profile, so similar to the types before 412 and so dissimilar to the full or three-quarter face on all later coins, the round petasos which had been replaced by a cap of different style by the middle of the fourth century, the character of the workmanship, all suggest a fifth century date. These points of difference, except for the character of workmanship, which is perhaps rather a matter of personal opinion, are unquestionable and absolute. There are no exceptions.

When we come to study more closely the details of the reverse, we find that the shallow incuse of the gold coin is found on numerous small silver coins dating before 412. 1 The staff of Hermes was used as a symbol on coins of two denominations between 440 and 412. 2 Furthermore Strack thinks that the staff of Hermes may not be analagous to the other symbols which he interprets as marks by which the magistrate in charge of the mint could be held responsible. 1 For an extraordinary issue of gold coins such marks of identification would not have been essential and thus the symbol of the magistrate could be replaced by the symbol of the god. On the bronze of the period before 412 the staff of Hermes appears as the obverse type with or without symbols in four of the five classes into which Strack has divided the early bronze coins.

While there is no example of an unabbreviated inscription on the coins before 412, it is quite clear that the peculiar nature of the type made a variation from the established custom possible. In most of the small coins with a goat for reverse type, the curling horns made the full inscription impossible. Likewise the type obviously called for an inscription above the goat, although once it is found below. On the gold coin there was no place for the inscription except at the side, and the unencumbered field made the full inscription possible. Then with the precedent set by the gold coin, the engravers of the next period, freed from this as well as from other conventions, usually gave the name of the city in full. But when one studies the coins of the last period, 365–341, it is evident at once that the throne would have permitted an inscription either above or at the side. 2 The engravers chose the side position in preference to the usual one above, not because the type required it, for it did not, but for some other reason. The reason, I think, is obvious. They were imitating the gold coin which they had taken for their model.

The bronze coins which have been tentatively assigned to the period from which the authors of the Berlin Corpus believe our gold coin to have come show how little dependence can be placed upon the form or location of the inscription. Of the three classes one has an abbreviated inscription, one has the full inscription at the side, and the other has it above. 1 The types, or in one case perhaps the inclination of the artist, were the determining factors. I have laid undue stress upon this matter of the inscription because it at first sight seems the surest evidence of a fourth century date for our gold coin, and because it is a criterion upon which Strack and von Fritze have relied.

But after all, the details of the reverse are not the best criteria, although helpful in arranging coins within a period or in determining whether a coin should be assigned to one period or another. The chief characteristics of the last silver coinage at Aenus, according to von Fritze, were the full face Hermes, the flat cap hanging over at the sides, the low relief, and the inferior workmanship. 2 To this list he adds the enthroned Hermes. If one compares the gold coin with the silver drachms of the last period 1 , the truly fine workmanship of the former will at once be apparent, while the slip-shod work of the latter will make it impossible to consider the two as contemporaneous products of the same mint, the enthroned Hermes notwithstanding.

The differences between the obverse of the gold coin and the obverses of the fourth century silver coins are hardly more marked than the resemblances between it and the small silver coins struck just before 412, as can be readily seen by a study of Strack's plates. In fact the obverse of the gold coin resembles no others in the whole period of the coinage of the city. Furthermore, the weight of this coin would almost make it necessary for us to assign it to the period of the light Attic standard before coins of Chian weight had been struck. 2

If the gold coin was struck just before 412, Aenus had a currency in two metals in which the gold coin was three-quarters as heavy as the contemporary tetrobols, or three halves as heavy as the more numerous diobols, a relationship between gold and silver that we have learned to expect. 1 Thus it is almost certain that the Asiatic ratio between gold and silver of thirteen and a third to one was prevalent at Aenus, 1 and that the gold coin was intended to pass for ten tetrobols, twenty diobols, five Persian drachms, or one and two thirds of the tetradrachms of the city, 2 an exchange that was perfectly simple because of the denominations struck at Aenus.

Of this coin of Aenus only this remains to be said, that in weight it is very close to the small coin of Thasos, the first issued by that city, and to a gold piece minted by Evagoras I of Salamis only a little later. 3

Returning now to the west, we come to Amphipolis on the Strymon, the possession of which Athens had been trying to regain since its rebellion in the time of Brasidas. This unchastised rebel is almost the last city one would expect to find following the example of Athens. Her silver coins, which began during the Peloponnesian War, follow the Phoenician standard of Abdera which became so popular in the Chalcidic peninsula at this time. 1 Nevertheless they are unique for this period, unless we equate them with the slightly heavier coins of Thasos and Neapolis. All of the other cities that adopted the Phoenician-Abderite standard followed the Macedonian practice of dividing the stater into six drachms, coins that are ordinarily called tetrobols. Amphipolis, however, divided the stater into four drachms. Only at the very end of her autonomous coinage do we find Amphipolitan tetrobols. 2

Thus the coinage of Amphipolis demands an explanation. It is possible, of course, that the adoption of the quarter stater instead of the sixth as a unit was due to the Athenian character of the population. But the majority of the population after 424 was at no time Athenian. Even the inscriptions on the coins show that the non-Athenian element was dominant, for except on the unique gold didrachm and on a few fourth century issues the legend reads ΑМΦІΠΟΛІТЕΩN, not ТΩN as it would if Attic engravers had made the coins. 1 In other respects Amphipolis did not run counter to the tendencies of the time and the region. Thus when we find the Amphipolitan coinage deviating from the custom that was almost universal where the Phoenician-Abderite standard was used during the last years of the fifth century and the first half of the fourth, we naturally search for the purpose of this deviation.

In the first place it should be noted that in the Amphipolitan coinage the staters seem to have outnumbered the drachms and other small coins. If there are only a few of these small coins, it would make comparatively little difference which divisional system is used. But one may ask where Amphipolis got the necessary change for transactions of small value. For the smallest transactions Amphipolis depended upon bronze coins, but bronze would be inconvenient for purchases beyond the value of a drachm.

When we compare the coinage of Thasos and its abundant issues of drachms with the coinage of Amphipolis, the absence of intermediate denominations is the more peculiar. Thus there are apparently two peculiarities in the Amphipolitan currency in need of explanation. If we work on the theory that the scarcity of drachms is somehow connected with the division of the stater into quarters, it may be possible to ascertain the cause of these irregularities. Since Amphipolis did not imitate her Chalcidic neighbors in coining sixths, she may have been imitating some other neighbor with whom her commercial relations were particularly close, possibly Macedon.

When Archelaus came to the throne in 413, the coinage of Macedon was identical with the form of the Phoenician standard adopted in the Chalcidic region during the Peloponnesian War. In fact, the Chalcidians could have got their standard from no other place. But Macedonian currency was not at all abundant. Archelaus, however, in his attempt to make Macedonia economically self-sufficing, minted coins in abundance and changed from the old standard to the Persian. It is still a puzzle just why he chose the Persian. Neither Abdera nor Maroneia could have had much contact with Macedonia at this time, and our study has showed that the normal form of the Persian standard was not adopted by Maroneia until after the death of Archelaus. The Persian tridrachm was used only after 404. At Abdera there may have been a few years, 411–408, when the Persian standard was used before the end of the War, but Macedonian imitation of Abdera is no more explicable on this account. Nevertheless, the fact of the standard being practically identical with Persian is unquestionable.

In the coinage of Archelaus there is one peculiarity. He struck no coins between staters and diobols. In fact, although he had changed the standard of Macedonian coins he did not change the divisional system, for, besides smaller divisions, he struck staters and sixths just as his predecessors had done, and unlike Amphipolis he minted about as many diobols as staters. Since the Macedonian diobol was an Amphipolitan hemidrachm, Amphipolis may perhaps have depended upon Macedonian diobols for small coins. Thus, while the staters of the two currencies were different, trade between Amphipolis and Macedonia was facilitated by the two to three ratio between the two standards and the dependence of Amphipolis upon the Macedonian diobol. It may be that there was an informal monetary understanding between the two states, or perhaps some sort of an entente by which Amphipolis became the port for neighboring parts of Macedonia. If the explanation given above is not the true one, one must admit that there was a connection between the coinage of Archelaus and Amphipolis.

This is more apparent when we study the obverse type of the two extant gold coins minted by Amphipolis, a young male head, with short curly hair, bound with a taenia. 1 Since the head is in profile, it is unlike the Apollo of the more famous Amphipolitan silver coins. In this and other respects it distinctly resembles coins of Archelaus, Aeropus, and Pausanias struck between 413 and 389. 2 While it proves nothing as to the connection between Amphipolis and these Macedonian kings, taken in connection with our previous conclusions, it is significant of the influence of Macedonia upon Amphipolis.

These gold coins weigh 8.59 gm and 4.17 gm. The smaller coin is about the same as the heavier of the two gold coins of Maroneia, while the larger is practically identical with the fourth century gold coins of the Chalcidic League. As the Amphipolitan triobol was a Macedonian diobol, the Amphipolitan stater would equal eight sixths of the Persian staters of Macedon. Thus gold coins of daric and half daric weights would readily fit into both local currencies. They would equal ten and five Macedonian staters, or sixty and thirty Macedonian diobols (Amphipolitan triobols). In other words, Amphipolis like Maroneia minted a gold coin that would exchange for fifteen drachms, fol- lowing a custom that was prevalent where the Phoenician standard was current. The gold didrachm, following the usual custom, can be equated with the silver tetradrachms on the basis of two-gold pieces for fifteen silver staters.

But while such an exchange value for the gold coin would be convenient in a place like Thasos- where drachms formed a large part of the currency, at Amphipolis with its scarcity of drachms, it might be a little awkward. Only if Amphipolis intended the gold for Macedonian trade would it be perfectly convenient. 1 There is one further possibility, viz., that the two gold denominations exchanged for thirty-two and sixteen Amphipolitan triobols (Macedonian diobols), and four or two staters. This would be more convenient in Amphipolis and not particularly inconvenient in Macedonia, for the "bulk of Macedonian currency consisted probably of diobols. But this hypothesis requires us to assume that the Amphipolitan gold coins used a standard slightly higher than the daric, one close to the Attic gold standard, and this assumption is apparently borne out by the weight of the gold didrachm. In other words, the weight of the gold coins might have been almost identical with Attic, because the silver standard and the gold-silver ratio demanded slightly heavier coins than the daric and half-daric.

The Chalcidian gold coinage furnishes a close parallel. In a recent discussion of this coinage I followed Professor Gardner in assuming that the standard was taken directly from Athens, 1 but I did not take into account the ease of exchange and the peculiarities of the Chalcidic monetary system. At thirteen and a third to one, a daric of full weight would equal eight staters of only 14.00 gm, but the Chalcidian silver stater, like the Amphipolitan, weighed from .25 gm to .50 gm more. While this is a slight inequality, it was probably sufficient to induce the Chalcidian moneyers to add to the daric weight enough to make the gold coins conform to the exchange value and pass for eight silver staters. Otherwise the gold stater would have equalled seven and a half silver staters or forty-five of the small coins of 2.40 gm. 1 But it would be much more convenient to have a gold' coin equal to forty-eight of these pieces. In some such way as this, either at Amphipolis or at the Olynthian mint, the heavy gold standard used by the Chalcidians must have arisen. Whatever name we give to this standard, I feel sure that neither the Amphipolitan nor the Chalcidian gold coins were Attic didrachms (or drachms) and it is also evident that the gold silver-ratio was not twelve to one when they were issued.

The date of the Amphipolitan issue or issues is easily determined by reference to the dated Macedonian coins of Archelaus and his two successors-where the same youthful head, bound with a taenia, is used for obverse type. Regling, who has made a careful study of the coins of Amphipolis, dates the drachm between 400 and 379, 2 while Head and Gardner have dated it about 400. 3 Since the youthful head is not used by Macedonian kings-after about 390, I see no reason for thinking that it was minted after that year. Jameson dates his gold stater about 400, and I am inclined to think that this is correct, even though the tripod symbol and the spelling АМΦІΠΟΛІТΩΝ, not АМΦІΠΟΛІТЕΩΝ, belong in Regling's second group, which he thinks began ca. 390. Regling was not acquainted with this piece. The drachm which has a bunch of grapes as a symbol and the ending ТЕΩΝ, Regling considers contemporary with the first and second groups of Chalcidian tetradrachms. If we date the first Chalcidian tetradrachms with Gaebler 1 in the last ten or fifteen years of the fifth century, there will be no difficulty in dating the transition between Groups I and II of Amphipolitan tetra- drachms about 400 and in making our gold coins contemporary with the silver issues of Macedonia which have a similar type, i.e., in dating them before 390.

Before discussing the first gold coins of Athens, it will be helpful to summarize the results of our study. Thasos with its coins of 2.02 gm and 2.78 gm, Aenus with its gold piece weighing 2.11 gm, and Maroneia striking coins of 4.01 gm, show the influence in one way or another of the Persian standard. They issued coins on the standard of the daric. Possibly the Amphipolitan pieces should be considered darics and half darics increased by a small amount. At Thasos even the type suggests imitation of the Persian Archers. Most of these coins, provided that the ratio between gold and silver was about thirteen and a third to one, were easily exchanged for the silver coins of their respective cities of issue. It is noteworthy that the Persian custom of minting a gold coin three fourths as heavy as a silver unit is found at Thasos, Maroneia, and Aenus. Then the gold pieces equalled ten silver pieces, except at Maroneia where no thirds of the Attic tridrachm stater were struck.

To the coins of Amphipolis the ratio of three to four does not apply, but the relation between the Persian and Phoenician standards was such that coins weighing a little more than darics and half-darics would fit into the Amphipolitan currency with little difficulty. Possibly we have here a modification of the custom at the basis of the Phoenician standard by which fifteen silver units exchange for two units of gold. Even the peculiar weight of the Maroneitan coins of about 3.20 gm is explicable on this basis. Finally the Thasian coins of about 3.90 gm are of the Chian standard, possibly slightly raised for the purpose of exchange.

It is difficult to prove that the ratio between gold and silver on the Thracian coast at the end of the fifth century was taken from Persia, as seems to me a priori probable, but there is one indication that gold was not undervalued and that the ratio between the two metals was not as low as it was in Athens in 406. There the ratio was probably twelve to one, an undervaluation of gold in comparison with Persia or even with Athens before the Peloponnesian War. The result of this undervaluation was that gold coins would almost immediately disappear from circulation into private hoards, the melting pot, or foreign hands. Gresham's Law was as true then as today, and we may take the good preservation of the Attic coins of 406 1 that escaped the melting pot as evidence, first, that the ratio between gold and silver on which this coinage was based was an artificial one that could not be maintained in the market place because it undervalued the gold, and secondly, that the gold coins would not leave Athens in great quantities except as bullion, even though they may not have been actually melted before leaving the city.

On the other hand, some of the coins we have studied show hard usage. 1 Applying Gresham's Law to them, and using the worn condition of the coins as a criterion of their popularity as a circulating medium, we are justified in concluding that they were not undervalued in terms of silver. Thus the Asiatic ratio, or one very close to it, must have prevailed, for if the ratio had been much lower than the Persian a profit could have been obtained by withdrawing them immediately from circulation and exporting them to Asia.

But since hoarding was an almost universal custom among the Greeks, Gresham's Law may have had nothing to do with the preservation of the extant coins, and my assumption that the gold coins of Athens were undervalued may be entirely unwarranted. In that case it might be necessary to explain how the cities of the Thracian coast could have based their coinage on the ratio of thirteen and a third to one when gold at Athens was worth only twelve times as much as silver. This is a question I have asked myself. There is only one answer. The Thracian coast was in close touch with Asia Minor, and through the Peloponnesians,even with Persia itself. Thus the Persian ratio would determine that of the Thracian cities. This answer, however, is not entirely satisfactory, for until we understand how the ratio of twelve to one came to prevail in Athens and how such divergent ratios can have existed side by side in Greece, the problem is only partially solved.

We know that shortly before the Peloponnesian War the value of gold in Athenian markets bore a close relation to the mint value in Asia. Gold was bought for the statue of Athena at the ratio of fourteen to one. 1 The difference between the mint ratio in Persia and the market value in Athens is therefore but five per cent., a difference that can be easily explained. As Athens was not a gold-producing country, the metal had to be imported at prices that would pay the cost of importation just as in the case of any other commodity, and as gold was a commodity in general demand, the world market would determine the basic price. One can hardly question that the Persian mint ratio was accepted as the base to which was added a differential equal to the cost of bringing it from Asia to Athens. For this charge, five per cent would be most moderate considering the risks and the large profits then obtained from commercial ventures.

What caused the fall in the price of gold in the Athenian market during the Peloponnesian War? That question has never been answered satisfactorily. Silver, not gold, was the measure of value in Athens. Hence a rise in the cost of living would tend only to decrease the value of silver in terms of what it would buy. A decrease in the purchasing power of silver would mean, ceteris paribus, an increase in the value of gold in terms of silver. This of course might be offset by an increase in the supply of gold, but we know of no such increase in Athens as would explain the depreciation of gold to twelve times its weight in silver. The gold darics poured into Greece by the Great King are hardly to the point, for they went to the enemies of Athens and therefore could not have affected the Athenian market. I doubt whether they affected other Hellenic markets to any great extent, for the sum of 5000 talents distributed throughout the Hellenic world over a considerable period of years 1 hardly compensated for the loss to Hellenic currency through the closing of the Attic mines during the last years of the Peloponnesian War. Though the annual production of these mines is unknown, it is probable that they produced at least 500 talents a year, or even more. Furthermore, just at the time when the mines were closed, the cost of living was increasing. Thus the higher prices would readily absorb the additional currency. But in any case this gold can not have come to Athens, for the balance of trade during the last years of the war must have been increasingly against Athens. In 406 the crisis came, and Athens had resort to two devices, the coinage of gold and the coinage of bronze. 1

The coinage of gold proves, as does that of bronze, that Athens was hard pressed for silver, not merely the state but also the people, for otherwise the gold could have been sold in the open market for what it would bring in silver. There was little silver to be had in the local market. Thus the adverse balance of trade, draining the country of silver as it did, acted exactly as an influx of gold would have done. Gold was no more plentiful than it had been, but it was perhaps relatively more plentiful. Thus from the local standpoint of Athens the ratio between gold and silver may actually have been twelve to one.

There are two flaws in this reasoning. The adverse balance of trade would have affected the supply of gold in the country as much as, or more than, it affected the supply of silver. Although gold was not money, it could be exported and exchanged for the necessities of life which Athens could not produce during the Decelean War. In the growing scarcity of money after the closing of the mines of Laurium, individuals and the state itself parted with treasures until Athens was almost stripped of valuables that would appeal to the foreign merchants from whom Athens bought grain and fish. Thus by 406 the state was reduced to the extremity of turning its gold statues into coin. Without doubt private citizens were no better situated, in which case it can be readily seen that the adverse balance of trade had affected both gold and silver alike and that gold relatively was no more plentiful than it had been before the war.

Secondly, it is incorrect to think of Athens as hermetically sealed and unaffected by outside markets. Even if there had been an influx of gold into Athens, or if gold had been relatively more plentiful than silver, so long as Athens was buying abroad more than she sold, the ratio would not be determined by local conditions but by conditions in the rather wide area with which Athens traded. So long as the huge territory in southern Russia and around the shores of the Black Sea was not satiated with gold, there would be a ready market for Athenian exports of the yellow metal. Here again the stabilizing effects of the Persian ratio would be felt.

Returning to our original question, why Athens adopted a mint ratio in 406 of twelve to one, we may answer it by saying that the conditions of 438–7 were exactly reversed. Athens was then a purchaser and had to import and pay the costs of import. Athens was now an exporter and she had to sell on such terms that the gold delivered would not be more expensive than if obtained elsewhere. The world market determined the price in Athens, just as truly as the world market determines the price of wheat on the farms of Dakota. Thus the ratio would be less than thirteen and a third to one, and the difference would be determined by the cost of export and the risks of the venture, which of course were not small at this time.

Thus the minting of gold by Athens on a twelve to one ratio does not prove that the ratio in continental Greece had reached that figure. We have been misled by our lack of knowledge of conditions outside of Attica, as is so likely to happen, into an unwarranted generalization. I think we have also been misled by our modern system of currency based on gold into thinking of gold as money, when in reality the money of continental Greece was silver, and gold was only a commodity like wheat and fish.

These gold coins, however, were never intended to serve as payments for imports. Otherwise there would have been no need of minting in small denominations. Staters would have been more convenient and more satisfactory for merchants-engaged in foreign commerce. Instead we have drachms, triobols, diobols, and obols, coins intended to take the place of the disappearing drachms and tetradrachms. For the smaller denominations bronze coins were minted for the first time, and judging from the scorn of Aristophanes, 1 we may conclude that the real value of a bronze coin was considerably less than the nominal value, in other words that this was token money. In fact some of it was actually silver-plated, a poor substitute for the vanishing silver currency. 2 Thus the gold and bronze were both minted for local circulation, to be used by the state for the equipment of the fleet, for the services of laborers reconditioning the triremes fitted out hastily and at great expense, 3 and for the many incidental expenses of small amounts. Perhaps some of the money was given to the generals before they set sail for the expenses of the campaign and the payment of the sailors.

In amount, this issue of gold money can scarcely have been very large. From the scholiast on Aristophanes, Frogs, 720, quoting Philochoros (F. H. G., I, 403, 120) τò (i.e., χρυσοῦν νόμισμα) ἐϰ τῶν χρυσῶν νιϰῶν, we may assume that the basis of the issue was gold from several statues (more than two at least) of Nike which Athens is known to have possessed during the period of the Peloponnesian War. Not all of the statues, however, went into the melting pot, as fourth-century treasure-records show. Each statue weighed about two talents, and if there were originally ten, the total silver value of the statues used for coinage would have amounted to about two hundred talents. 1 But Kolbe has argued recently, that no evidence exists for the number ten. 2 Thus the amount of money realized from the gold victories must remain in doubt.

Other treasures were melted at this time, as we learn from extant temple inventories. In the Pronaos, for example, everything was appropriated except a gold crown, 3 the only gold item kept there; and yet the value of the treasures taken from the Pronaos was probably less than five talents. While we cannot estimate the size of the gold issue, it is probably safe to say that it would have been exhausted almost immediately, whether it was used at home for the equipment of the fleet, or abroad for the pay of the sailors. Still the fact that Athenian generals were reduced to plundering the coast of Asia Minor and exacting contributions from unwilling allies in order to provide precarious pay for their crews 4 makes it probable that the greater part of these gold coins was spent in Athens before the fleet sailed.

Let us see whether we can trace them after they came into the hands of private citizens. Would they stay in Athens, or, if they did, would they remain in circulation? If they left Athens, where would they go, and how soon would they establish a reputation such as Attic silver enjoyed? In other words, when would imitations begin to make their appearance?

Two forces would be at work to drive the money both from circulation and from Athens, Gresham's law, and the need of money to pay for the imports of the city. As we have seen, the coins were so minted that they could compete with silver and the darics of Persia in distant markets. If we consider the gold merely as an exportable commodity, it would be more profitable to sell it at a greater price in silver than to make payments in silver coins. A merchant would then make a double profit on the transaction. He would have a cargo both ways. Thus the coins would rapidly disappear from circulation. They might also go into private hoards although not to such an appreciable extent. Yet if any one did have money to save, he would naturally choose to keep the silver to the debased bronze and the undervalued gold to both.

When and how did this money leave Athens? Assuming for the purpose of argument that some of it went with the fleet, it would never have reached the Thracian coast, for the fleet was active on the coast of Asia Minor and in the Hellespont for the next year, in fact until the final defeat of Athens at Aegospotami. If it left Athens in the course of trade, it would have gone principally no doubt to the food-producing regions around the Black Sea. Some of it may have gone to Thrace as well, but not until the following spring. The season of active commerce had practically ceased when the coins were minted about the end of June or the first of July. There was barely time for the fleet to sail before the Etesian Winds made navigation difficult Ships could return to Athens with the wind blowing from the north, but it was impossible for traders to sail northward after the Etesian Winds began. When they ceased the time for trading with the northern Aegean would be short. Thus in trade, the gold would first appear in northern markets in the spring of 405. In the late summer or early-autumn of that year came the final Athenian defeat. Thus we have less than six months for the Athenian, gold to become famous and to create emulation, a period entirely too short for the influence it is supposed to have exercised. Perhaps this period was too short for it to have left Athens in quantity, in which case it would disappear in Attica because of its excellence. After the darics of Persia entered Athens in the years following the war, Gresham's law would become operative as against them too, since the standard of the daric was slightly lower than that of the Athenian coins.

In conclusion, it is generally agreed that the disintegration of the Athenian empire beginning in 432, getting a further impetus through the activity of Brasidas in 425–4, and becoming more general after 412, resulted in the break-up of the monetary system imposed by Athens upon her subjects. 1 Toward the end of the Peloponnesian War, Spartan fleets were often on the coast of Thrace and the sailors must be paid with money with which they had been familiar. 2 Furthermore Persian power was growing and the Persian silver standard was "coming in on certain lines of trade" in the last decade of the century. 3 No one was imitating Athens, with the possible exception of Maroneia, Abdera, and the Odrysian princes, about 430. On the contrary, after 425–4, that year of misfortune to Athens, no city adopted the Attic standard for silver, and the majority of them rejected it certainly before the end of the century and most of them before 406. Furthermore the gold coins of Athens can not have been well known anywhere outside of Athens before the spring of the year in which the empire went to pieces, while the darics had been known long and favorably and had been distributed in quantities to the Greek allies of the Persian king in the very years when the gold coinage of the Thracian cities appeared. This would be just as true after 406 as before. Thus our explanation of the gold coinage of the Thracian coast is in perfect accord with our knowledge of political and commercial conditions there during the last years of the fifth century. Persia rather than Athens furnished the models. The gold Owls came too late to compete with the gold Archers.

End Notes