For Alexander's chief Macedonian silver mint I use here when necessary the traditional name of Amphipolis. This name is used with great reluctance, for I have no confidence that this city, rather than Pella or perhaps Aegae or Philippi, was the source of this enormous silver output. With no specific evidence supporting the claim of any other city, however, it seems preferable at least for the moment to retain the usual attribution to Amphipolis–but with no assurance that the coinage was in truth struck there. A second Macedonian silver mint, usually referred to as Pella, is treated here only rarely and peripherally. This study concerns itself only with the chief mint.
The Alexander tetradrachms' pattern, established long ago by Edward T. Newell,1 is of a number of successive groups, each of which includes from three to twelve different issues, i.e., coins with differing reverse markings. Within each group there is heavy obverse linkage among issues. Not every die is known in multiple issues, but with almost no exceptions every issue is obverse linked with at least one, and usually more than one, other issue in its group.2
Table 1 lists the groups and their constituent issues. Groups A through K are listed by Newell's letters as he published them in the Demanhur hoard.3 The next group, not present in that deposit, I have termed L. Groups after L are not included in this study.
The groups are listed in Newell's order, with the single exception that the minute group K is placed before J. Justification for this minor shift, as well as for its continued attribution to our mint, is given below.4 Within each group the issues are listed in the order given in Alexander, Martin Price's recently published monumental compendium of Alexander issues,5 although within each group any order is meaningless, as die linkage patterns show that the issues within each group must all have been struck more or less simultaneously.
Table 1 is organized by inscriptions and groups with the number of coins studied given for each group. The first column in the table gives Newell's group letters, joined by issue numbers (repeating for each group) assigned by the present author. Hesitant as I have been to introduce a new set of numbers into this subject, I have been convinced to do so by the unsatisfactory choices available for describing these issues, which so often form major components of hoards and provide the basis for dating those hoards. Müller issue numbers are incomplete and their order virtually meaningless. Alexander's issue numbers and Demanhur hoard coin numbers give only a rough indication of where in this vast Macedonian coinage the individual issues fall. A system which indicates the group (more important than the issue in any case) in addition to the specific issue should be far more descriptive than one which identifies only the issue and does not always accurately place that issue. Thus B8, E2, and G3, for example, provide more readily useful information than Demanhur's 247, 716, and 1,168, or Alexander's 32, 78, and 110.
The table's second column describes each issue's marking or markings (the primary marking preceding any secondary one, regardless of their positions on the coins). A bold P indicates that issues of Philip II's types are known with the same markings. These Philips are probably posthumous in the case of those similar to the Alexanders of group A. Those parallel to the later issues, from group I on, are decidedly so.6
The third column gives the plate numbers of examples of each issue. The fourth and fifth give the issue numbers in Alexander 7 and the initial Demanhur hoard coin numbers. Issues illustrated in Alexander are marked with an asterisk, and those with whose descriptions I differ are placed where I believe they belong but in parentheses. Finally, as an indication of their relative abundance or rarity, the numbers of examples studied from each issue are given. The numbers of obverse dies located and the estimated totals used, better indications of the original size of the groups, are given in Table 2.
The tetradrachms' types are
Obv.: Beardless head of Heracles r., wearing lion's skin headdress.
Rev.: AΛEΞANΔΡΟΥ (or BAΣIΛEΩΣ AΛEΞANΔΡΟΥ in groups G, H, I, K, and J). Zeus seated l., holding scepter and eagle.
|Issue||Markings||Plate||Alexander Issue||Initial Demanhur Coin No.||Examples Found|
|Group A, 250 coins|
|Group B, 212 coins|
|B5||Attic helmet||13||21*, 22*||243||11|
|B8||Caduceus (also in E)a||18||32*||247||8|
|Group C, 87 coins|
|C3||Grain ear||21||39, 39A||317||10|
|Group D, 216 coins|
|D6||Filleted caduceus M||32||65||426||1|
|Issue||Markings||Plate||Alexander Issue||Initial Demanhur coin No.||Examples Found|
|Group E, 605 coins|
|E9||Caduceus (also in B)c||49||99*||614||58|
|Group F, 224 coins|
|F2||Star in circle||51||103*||895||20|
|F5||Bow and quiver||55||106*||1014||41|
|AΛEΞANΔΡΟΥ BAΣIΛEΩΣ or BAΣIΛEΩΣ AΛEΞANΔΡΟΥ|
|Group G, 287 coins|
|G3||Bow and quiver||59||110*||1168||69|
|BAΣIΛEΩΣ AΛEΞANΔΡΟ Υ|
|Group H, 455 coins|
|Group I, 177 coins|
|Issue||Markings||Plate||Alexander Issue||Initial Demanhur coin No.||Examples Found|
|Group K,f 18 coins|
|K2||P||Λ, (or||72||421*, 425, 426||1582||10|
|Group J, 147 coins|
|Group L, 271 coins|
|LI||P||forked branch||90||(126*), 140||–||8|
|L2||filleted club||91||(127*), 128||–||5|
|L8||P||profile shield||101||136*, (137)||–||12|
Succeeding groups, all inscribed AΛEΞANΔΡΟΥ and struck before ca. 295 B.C.,8 were
not subjected to a die study. They include:
P Λ over race torch, and varying additional marking or markings;
P fulmen over I, and varying additional marking; and star, obelisk, and X
(varying positions), or star over obelisk, and varying additional marking or markings.
As has been noted, within each group it is clear that all issues must have been struck more or less simultaneously, and the die linkage is so complex that it is impossible to place the issues in any linear chronological order. Three typical clusters of coins are diagrammed in Figures 1–3. They come from group H, but similar clusters and die linkage are found in almost every group (e.g., note in Table 1 the obverse die used for six issues in group D). The clusters presented below are simplified. Another antler obverse, for instance, sharing a reverse die with the first coin listed but not linked by its obverse to any other symbol, is omitted. Brackets to the left and horizontal lines indicate obverse die identities, and brackets to the right, reverse die identies. All coins are illustrated on Plates 5–6.
Alexander Tetradrachms: Obverse Die Links within Group H
|Anter||Phrygian Cap||Macedonian Helmet||Trident||Tripod|
A further confirmation of the contemporaneity of issues within groups is provided by the obverse links between groups described in Chapter 3. Issues struck in linear sequence would tend to have one issue in a given group linked to one issue in another. Instead, especially among groups after A and B, the obverse dies forming links between groups were often employed for a great number of issues.
This small issue E1 is catalogued where Newell placed it in Demanhur. It is not clear whether he eventually knew the die link shown above (40 and 44), but in Reattrib. (p. 10, issue XXVII), he commented on the one specimen he knew from the issue that the coin's "obverse resembles the obverses of previous coins [of group D], while the reverse is almost identical in style and workmanship with the following [group E]." Once again, Newell's remarkable sense of style is clear, as there is now known a second obverse die in the issue, which was used also in group D: see Chapter 3, link 17. E1 might thus belong to either D or E, but is here left in its traditional place. In either case, groups D and E are joined by one known obverse die.
See issue B8 with note a, above.
"Antler," the accepted name for this symbol, is unsatisfactory. It often looks more like a ragged branch.
See p. 27, commentary on Alexander issues 118–19.
For discussion of the disputed placement and even attribution of group K, see pp. 49–50.
Reattrib., pp. 5–23, and Demanhur , pp. 26–32, 65–66, finalizing the classification presented in Reattrib.
The exceptions are very small issues in groups K and L (K3, K5, K6, L2, L9), whose markings make their group placements certain.
See above, n. 1.
Alexander, pp. 89–103, with the addition of some issues from p. 132.
See Chapters 4–6 for these late posthumous Philip reissues.
Alexander, pp. 89–103 and 132.
For these issues, see "Tetradrachms Amphipolis." Ehrhardt here also notes the posthumous Philip II issues which were struck in parallel with the Alexanders through those with fulmen over I. These Philip issues form Amphipolis group IV in Philippe . The final group, with star, obelisk, and X, may not belong to our mint. Price in Alexander (pp. 139–40) tentatively prefers an older attribution to Uranopolis, but an Amphipolis origin is most recently strongly defended by Thompson in "Cavalla," pp. 40–44.
Newell's coin numbers, as they are found on the ANS's coin boxes, cast cards, and photo file cards, are provisional working numbers only, and they encompass many numbers for which there seem to be no examples. When I finally consulted Newell's notebook (described in the introduction), no examples for the missing numbers appeared there either. Clearly he sometimes left runs of numbers unused available to be assigned to subsequently acquired specimens, and consequently his die numbers cannot be taken as cumulative and do not show the total numbers of obverse dies in the various groups. For example, in group I, his die numbers run from 660 through 723, for a total of 64 numbers. Three pairs of those numbers, however, were given to identical dies, for a loss of 3. Similarly, there are 13 numbers with no examples known (not in the trays and not mentioned in his notebooks), and I have found 8 additional dies. Instead of Newell's apparent total of 64 dies for group I, there seem to be only 56. Similar situations obtain in each group.
Table 2 shows the numbers of coins studied in the various groups and the numbers of obverse dies identified in each group. "Coins" include ANS coins (approximately half of all located), casts, illustrations in the ANS's photo file, or examples pictured in readily available publications. The number of obverse dies given for each group is reduced by 0.5 for each die shared with another group. The final column, the number of estimated dies, is the number arrived at by the useful equations published by G. F. Carter.9
Group E (605 coins, 193 dies known and 241 estimated) is clearly the largest group, but, if as seems probable, F and G should be combined into one group, then that resulting group would be a close rival (511 coins, 162.5 dies known and 203 estimated). Group L was also very large.
|Group||Coins||Obverse Dies||Coin/Die Ratios||Estimated Obv. Dies|
"A Simplified Method for Calculating the Original Number of Dies from Die Link Statistics," ANSMN 28 (1983), (1983), pp. 195-206, at p. 202. The total estimated dies are calculated from the total numbers of coins and dies, not by the addition of the estimated dies in the various groups.
|Alexander Issue||Troxell Issue|
|1, 4||A1||The prow on 1 faces r., on 4 1. The difference is significant, as the right-facing prow seems to appear on the very earliest coins of the issue. See pp. 87–89.|
|8, 9||A4||The fulmen is slanted on 8; on 9 it is vertical, large, and crude. Alexander's illustrated example of 9 perhaps shows a recut symbol. Other vertical fulmens are smaller and more neatly executed.|
|10, 11||A5||The rudder has tiller up on 10, down on 11.|
|21, 22||B5||The Attic helmet faces r. on 21, 1. on 22.|
|39, 39A||C3||The grain ear is vertical on 39, slanted on 39A.|
|110A||—||The issue is described with AΛEΞANΔΡΟΥ BAΣIΛEΩΣ, and with dolphin 1. in 1. field; the reference is to Reattrib., issue 40 (pl. 9, 8). The symbol however seems to be merely a degenerated cornucopia of group G (as indeed Newell suggested, p. 33, n. 39), cut over the Athena Promachus of that group. Issue 110A is a phantom.|
|116||J1||Issues 116-17A are wrongly placed here, between groups H and I. They are merely part of group J. Alexander even, exceptionally (p. 86), notes obverse links between 117A (J2) and 124 (J6), and between 117 (J3) and 124.|
|118, 119||I1||Alexander lists and illustrates two variations, and (actually as is clear from a cast at the ANS), of the usual monograms. See 85–66.|
|125||—||The issue is described with AΛEΞANΔΡΟΥ, and with wreath in 1. field and below throne. The reference is to Reattrib.'s issue LII-a, which there (p. 16) cites only Müller 548. Müller 548, however, has only the wreath, no , and issue 125 is apparently a phantom.|
|126||—||The coin is described as with and "oak(?)-branch," but a dot is visible on the illustrated example, joined to the bottom of the right vertical stroke of the . The illustrated example of 126 seems but one of many poorly executed examples of group L, and belongs instead in issue 140, below. Issue is a phantom.|
|127||—||The coin is described with and filleted club, but a dot is clearly visible just to the left of and below the right vertical stroke of the . The coin belongs in issue 128, so issue 127 is a Phantom.|
|134||—||The issue is described with AΛEΞANΔΡΟΥ, and with dolphin r. in l. field, and it is placed with the issues of group L (with ). The reference is to "Tetradrachms Amphipolis," issue 16, which cites as a parallel a Philip II issue (Müller 211), which might seem to suggest that the Alexander issue does belong at Amphipolis. The Philip issue is, however, decades earlier. See Philippe , Pella II.B, 410 ff. The present author strongly doubts that Alexander 134 was struck at Amphipolis.|
|135||[L3]||The wing described on the sole coin cited (here 93) would seem simply to be an aplustre, a symbol whose shape varies considerably. See 92–94.|
|137||[L8]||The cowrie shell described on 137 is almost certainly merely a degenerated profile shield as on 136.|
|140||L1||The issue, described with laurel branch and , cites Müller 561, whose symbol is pictured like the single straight upright laurel branch of issues J3 and J6. Two references are cited, the Aleppo 1893 hoard ( IGCH 1516), and "Tetradrachms Amphipolis." Newell's transcript of the Aleppo hoard coins, however, shows a forked branch as on issue L1. Citations in "Tetradrachms Amphipolis" reveal only coins as J6 ( Demanhur 1564 and Newell's list of the Kuft hoard) and L1 (Aleppo 1893 hoard, and Walcher de Mollhein 1061). As no coins with and straight laurel branch can be located, then, one can probably safely discount Müller's description and consider that Alexander issue 140 is equivalent to L1. Issue 126, described as with "oak(?)-branch" (perhaps a better description than "forked branch") also belongs in issue 140.|
|421, 425, 426||K2||The three issues seem but three variations in the secondary marking. Alexander has separated 421–27 ( Demanhur group K) from groups A–J and L and placed them at a different mint as the direct predecessors of the groups with or and bucranium or torch, etc. See Alexander, pp. 86–87. This separation seems incorrect in the light of the four die links now known between posthumous Philip II issues as group J and others as group K. See below, Chapter 6, links 14–17. Further, at least one obverse die link is known between group L and the Λ-bucranium Alexanders. See Chapter 3, link 22.|
|428||—||The issue is described withAΛEΞANΔΡΟΥ, and with A below the throne as the only marking. The reference given is "Tetradrachms Amphipolis," issue 5, which no doubt is derived in turn from a coin of this description at the ANS which was placed in its trays together with group K coins. Neither the coin's sole marking nor its style suggests any association with group K. I strongly doubt that the issue belongs at our mint.|
These smaller coins have received but one very brief study, by Newell in 1912.1 Table 3 presents the Alexander silver issues smaller than the tetradrachm: didrachms, drachms, triobols, diobols, and obols. All denominations have the obverse type of the tetradrachms, a beardless head of Heracles r., wearing lion's skin headdress. The various reverse types are noted after each denomination's heading in the table, and shown again in schematic form in Table 6, pp. 34–35. All coins are inscribed simply AΛEΞANΔΡΟΥ.
The first column in Table 3 gives the Newell tetradrachm group to which each issue belongs, and the specific tetradrachm issue number assigned in Chapter 1, if there is an exact correspondence. Some small coins' markings do not parallel any on the tetradrachms, but obverse links among the small coins securely place most of these non-parallel issues in group E, and the rest can be assigned with near certainty on other grounds.
The second column gives the coins' markings, and the third the plate reference for representative coins of the different issues. Virtually all known obverse dies are illustrated, the exceptions being the late issues with or arrow markings. Issue numbers in Alexander form the fourth column, and asterisks indicate the issues illustrated there. Where I differ on the reading of markings, the Alexander issue number is placed where I believe it belongs, but in parentheses. The fifth column gives the number of examples found in each issue. Brackets to left and right of the plate references indicate, as usual, obverse and reverse die links. All known die links between issues are shown. Issues of which I have seen no examples are shown in brackets, and are not counted among the examples located. The drachms, the commonest denomination, are divided between standing eagle reverse and seated Zeus reverse.
Table 4 summarizes the numbers of examples found of each denomination in each group. Table 5 shows the number of obverse dies located (shared dies reduce the number by 0.5), again for each denomination in each group. It is remarkable how close to 2:1 the coin to die ratio is for each denomination and for each group except group A.
|Corresponding Tetradrachm Issue||Markings||Plate||Alexander Issue||Examples Found|
|Rev.: Zeus seated 1.|
|Group B, 1 coin|
|Group C 14 coins|
|Corresponding Tetradrachm Issue||Markings||Plate||Alexander Issue||Examples Found|
|Group D, 8 coins|
|D7/8||Caduceus ( ? ?||140||68||5|
|Group E, 8 coins|
|A. Rev. : Eagle, head sometimes reverted, standing l. or r. on fulmen|
|Group A, 5 coins|
|Group B, 1 coin|
|Group C, 1 coin|
|Group D, 9 coins|
|Group E, 36 coins|
|E5||: eagle on club||157||85||4|
|eagle on thyrsus? or torch?||161||—||1|
|E-||No marking: eagle on caduceus;||164||144||1|
|E-||eagle on club;||165||145*||9|
|E-||eagle on thyrsus;||167||148||4|
|E-||eagle on torch||168||151||2|
|B. Rev. : Zeus seated 1.|
|Group E, 6 coins|
|Corresponding Tetradrachm Issue||Markings||Plate||Alexander Issue||Examples Found|
|Group E or F, 13 coins|
|Group F, 18 coins|
|Rev.: Eagle standing 1. or r. on fulmen|
|Group B, 2 coins|
|Group C, 2 coins|
|Group D, 1 coin|
|Group E, 24 coins|
|E-||No marking: eagle on club||193||146, (149)||7|
|E-||No marking L||195||150, 154||7|
|Rev.: Two eagles standing facing each other, on fulmen or exergue line|
|Group A, 1 coin|
|Group B, 7 coins|
|B6||Ivy leaf: in center; to right||198||25, (16)*||6|
|Corresponding Tetradrachm Issue||Markings||Plate||Alexander Issue||Examples Found|
|Group C, 2 coins b|
|Group D, 7 coins|
|Group E, 13 coins|
|E-||No marking: eagles on club;||206||147||3|
|E-||eagles on torch||207||152||1|
|Group A, 1 coin|
|Group B, 4 coins|
|Group C, 1 coin|
|Group D, 3 coins|
|Group E, 9 coins|
|Group||A||B||C||D||E||E or F||F||Total|
|Group||A||B||C||D||E||E or F||F||Total|
Table 6 summarizes the issues known of the small coins. The obverse type of all denominations is the same as the tetradrachms'. The reverse types are indicated in the table by the following abbreviations:
Z = Zeus seated, as on the tetradrachms
EH = Eagle standing r., usually on fulmen
ERH = Eagle standing r., head reverted, usually on fulmen
EL = Eagle standing 1., usually on fulmen
2E = Two eagles standing facing, on fulmen or exergue line
F = Fulmen
Issues in Alexander of which no specimens have been seen by me are shown in brackets.
|D7/8||Caduceus ( )||Z|
|E3||Cock or cock head||Z||ER|
|E5||eagle on club||ERH|
|E8||eagle on thyrsus? or torch?||ERH|
|E-||No marking: eagle on caduceus;||ERH|
|E-||eagle[s] on club;||ERH||ER||2E|
|E-||eagle on thyrsus;||ERH|
|E-||eagle[s] on torch||ERH||2E|
Alexander's sole reference is to Reattrib., p. 14, XXXIV. This cites "Imhoof-Blumer," which presumably is Monn. gr., p. 119, 25, a coin of 2.10 g with pentagram symbol. This coin is from an unidentified private collection and cannot be traced.
While this study was in page proof, Charles Hersh acquired a diobol with bow symbol corresponding to tetradrachm issue C6. The litte coin is from new dies. It is not illustrated, but it is included in Tables 4–6.
It has unfortunately not been possible to obtain a cast or photo of this coin, seen by Price in a private collection, but there seems no reason to doubt the issue.
Reattrib pp. 12–14 and 23.
It should hardly be necessary to state once again that these small coins, most with eagle as reverse type, are not subdivisions of the rare Alexander tetradrachms with eagle reverse.2 Those tetradrachms were struck to the old standard employed by Philip II, whereas the small coins are all of full Attic weight and most of their markings are clearly those of the Attic-weight tetradrachms of Chapter 1. The type of standing eagle with reverted head was simply an old Macedonian type continued by Alexander. It was used by Archelaus I, Amyntas III, and Perdiccas III,3 and the latter two, Alexander's grandfather and uncle, coupled with it the Heracles head obverse used by Alexander.
Unaware of the numerous obverse links now known between the many small coins without reverse symbols and those with symbols of group E, Alexander unfortunately has catalogued these no-symbol issues together with the eagle-reverse tetradrachms (while of course listing the symbol-bearing small coins together with the tetradrachms bearing their markings).4 All the small coins with eagle reverses can now, however, be associated with specific groups of the Attic weight tetradrachms. Together with the didrachms, which bear the tetradrachms' seated Zeus as reverse type, they are all simply subdivisions of the tetradrachms. It seems unnecessary to consider them a separate series struck "for local circulation" only.5
In groups A through D, the small coins' markings are exactly those of the tetradrachms, except for one drachm with filleted caduceus and which probably should be assigned to group D (the filleted caduceus occurs in both C and D, but only in D are monograms found). The drachms of groups A through D all have the standing eagle reverse type.
As just noted, the numerous obverse links within group E, diagrammed in both Table 3 and Table 6, allow the firm placement within that group of a number of anomalous issues of drachms, triobols, and diobols whose attribution has heretofore been uncertain. These coins have no regular issue markings and often show the eagle standing not on the standard fulmen, but on caduceus, club, thyrsus, or torch.
By any standard–number of issues, number of examples located, or number of obverse dies found–group E had the largest output of small coins. This is not surprising, as E was also the largest group of tetradrachms. In this group, too, the drachms with the usual imperial Alexander drachm reverse of seated Zeus first appear, with issue markings identical to those of some eagle-reverse coins in the group, and actually obverse linked to one other eagle-reverse issue.
A drachm issue with the simple marking has heretofore usually, and understandably, been associated with the Alexander tetradrachms of group L, which bear the same primary marking.6 The presence now of several examples of the issue in the Near East 1993 hoard,7 however, buried perhaps ca. 322 (several years earlier than the great Demanhur hoard interred before the striking of the tetradrachms of group L), shows that these drachms must be considerably earlier than tetradrachm group L, and the absence of the title requires a group prior to groups G–K/J.
Also in the Near East 1993 hoard were two drachms with laurel branch symbol, an issue previously unknown save for one example
1988 by Kamen Dimitrov. This was one of three Alexander drachms forming a small hoard discovered in 1976 at Calim, in Bulgaria.8
Dr. Dimitrov has kindly sent me not only a direct photo of a cast of the coin (175), but
also a translation of his relevant Bulgarian text:
... Calim, ca. 35 km. W. from Nicopolis ad Nestum. Three Alexander drachms are kept in the Historical Museum of Blagoevgrad. ... According to the
control marking . . . [the coin in question] corresponds to the issue of Demanhur 1563, [J1, with laurel
branch but with the omitted], Amphipolis 320–319.
At the same time the coin is struck from the same obverse die used for a specimen of an issue not represented in the Demanhur
hoard. ... [Sardes and Miletus, p. 87, 3 = 174].
The Sardes and Miletus issue cited, die linked with the Calim laurel branch coin, is the issue. The laurel branch issue's presence in the Near East 1993 hoard now shows that it too antedates 322/1 at the latest, and the absence of the title again indicates a group prior to groups G-K/J. No exact correspondences with any tetradrachms' markings exist for these two interesting issues, but the reverse variation and experimentation introduced in group E may in part explain their lack of correspondence. The obverses of these and laurel branch drachms are extremely similar to many tetradrachms of groups E and F (e.g., 40–56). Their reverse exergue lines, too, with one dotted exception, are formed by a simple line, an innovation which is known rarely among the group E tetradrachms, but which is common among those of group F.9 One of these groups then must be that to which these and laurel branch issues belong.
Another Zeus-reverse drachm issue with arrow symbol has long been known. The arrow, which again does not occur on the tetradrachms, could be considered as associated with group C's bow or with F's bow and quiver.10 But, as other Zeus-reverse drachms first appear in group E, these arrow-symbol drachms cannot be so early as group C. Again, the lack of the title rules out groups G-K/J. The obverse style of many arrow drachms, like that of the and laurel branch drachms just discussed, is very similar to tetradrachms of both groups E and F–but in the case of these arrow drachms, one iconographical detail allows a firm placement in group F. Just as on the group F tetradrachms, their exergue lines, instead of the normal dotted ones, are sometimes found as simple straight lines (177) or omitted altogether (179). And on at least one arrow drachm (178) the footstool is indicated by the slanting "short straight line (not to be confounded with an exergual line)" which is found only on the tetradrachms of group F.11 The arrow drachms can only belong to group F.
No small Alexander coins are known after group F. As will be seen below in Chapter 4, the revived tetradrachms of Philip II, many of whose markings parallel those of Alexander tetradrachms, start possibly as early as group I, and certainly by groups K and J, continuing through L and several subsequent groups. Philip II fractions accompany these Philip tetradrachms through those parallel with Alexander groups K and J–and then, as I shall argue in Chapter 5, probably are discontinued before the Philip group parallel to Alexander's group L.
Finally, following group L and the tetradrachms with bucranium and Λ, Thompson has deduced from the existence of a plated ancient Alexander imitation drachm with Λ and torch that there may have been genuine Alexander drachms with those markings also.12 If so, however, none have yet been discovered.
Thus the small coins were as follows.
Groups A-D: Alexanders, several denominations, drachms with eagle reverse
Group E: Alexanders, several denominations, drachms with both eagle and Zeus reverses
Group F: Alexanders, drachms, Zeus reverse
Groups G-H: –
Groups K-J and perhaps I: Philips. See Chapter 5.
E.g., BMC, pp. 165, 171–72, 176; SNGCop 505, 513–15, 522; SNGANS 94–96, 113.
Alexander 144–52, 154–55, 157. See Tables 3 and 6 and comments on 144–57, pp. 39–40. The association of the eagle-reverse bronzes of issues 158–62 with the eagle-reverse small silver coins is also quite uncertain.
Alexander , pp. 24, 88, and 103–4.
Chapter 8, hoard 7.
Chapter 8, hoard 11.
See pp. 91–92, and 53.
|Alexander Issue||Denom.||Corresponding Tetradrachm Issue|
|3A||ob.||[A1]||This is the coin seen by Price in a private collection.|
|16||2-ob.||(B6)||Described as with wreath between two eagles on reverse, the only coin cited actually has an ivy leaf (it is a die duplicate of several other specimens so marked, and the leaf is clear on Alexander 's illustration of 16). The coin belongs to group B's issue 25. No diobols with wreath are known to me.|
|25,||25 A 2–ob.||B6||The one coin known to me of issue 25A (199, with ivy leaf to right) is from the obverse of all five known examples of issue 25, with ivy leaf between two eagles (e.g., 198). Coin 199 is from the same die pair as Alexander's illustrated example of issue 54 and McClean 3509 (the symbol erroneously described as a bucranium), both with the eagle head of group D. The symbol of these last two coins has been cut over the ivy leaf of issue 25A. See also issue 54. Note the analogous recutting in the obols of groups B and D (issues 26 and 55).|
|26||ob.||B6||The ivy leaf on the single reverse die of all three known specimens has been recut to an eagle head on the two known specimens of group D's issue 55.|
|30||ob.||[B7]||Newell in Reattrib. mentions this issue, but I have found no examples. Possibly an ivy leaf was seen as grapes. Compare issue 26.|
|33||dr.||E9||Although listed in Alexander after the issues of group B, the shape of the issue's caduceus argues for a placement with issue 101 in group E.13 Issue 34, also with caduceus, is obverse linked with other E issues.|
|34||3–ob.||E9||Also listed after the group B issues, the issue belongs to group E. Of the four known examples, only three are in sufficiently good condition to allow die identification, and all three share their single obverse die with coins of group E's issues 82 and 149.|
|37||2–dr.||C1||The issue is perfectly valid. Note only that Lanz 48, 22 May 1989, 193, from the dies of the coin illustrated here (132), is erroneously described as bee on rose, and thus as a unique didrachm of Pella.|
|50||dr.||F–||See pp. 36–37 for the placement in group F.|
|53||3-ob.||(E7)||The issue is described with eagle head to right, but the sole known coin, at the ANS (188), seems on close examination to bear a crescent, with horns pointed downward–which is also the orientation of the same symbol in the exergue of a coin of issue 90 (189). Issue 53's flan and die sizes also accord far better with group E than with D, so that the coin probably belongs in issue 90. Issue 53 seems, at least from present knowledge, to be a phantom.|
|54||2–ob.||D1||The issue exists. See issues 25 and 25A for discussion of its recut symbol. SNGBerry 197, however, noted as an example, has not an eagle head but a horse head. Coin 203 clearly shows the horse's bridle. See issue 26 for discussion of the recut symbol.|
|68||2–dr.||D7/8||Price calls the monogram but its small size and condition on the known coins make it impossible to be certain whether it is or , or perhaps simply|
|69||dr.||D-||The caduceus is filleted.|
|88||3–ob.||[E6]||See p. 21, note a.|
|90||3–ob.||E7||See also issue 53.|
|94||2–dr.||E8||The sale catalogue reference cited in Alexander has a caduceus, not a bucranium.|
|94A||dr.||E8||The two citations refer to the same coin, and the Giessener (Gorny) coin number should be 221.|
|95, 96||dr.||E8||The symbol is vertical on 95 and horizontal on 96.|
|97||3–ob.||Two examples are listed. The Hague (now Leiden) coin must be an erroneous citation. J. P. A. van der Vijn has sent photographs of the cabinet's only Alexander triobol, and it is from the dies of the other examples of 146. The bucranium on the Hersh coin cited seems to be merely the final Υ of the inscription and the coin thus is part of issue 150. Issue 97 seems to be a phantom.|
|101||dr.||E9||Issues 33 and 101 both seem to belong to group E.|
|107||2–dr.||C2||The symbol is not the upright bow and quiver of group F (where the issue is placed) and G, but the simple quiver depicted in the slanting position of group C, where 107 shares its single obverse die with issue 45 (with group C's Pegasus forepart symbol).|
|141||dr.||E?F?||See p. 36 for the placement in group E or F.|
|146||3–ob.||E–||The object on which the eagle stands is not perfectly clear, but does appear to be a club on the three examples known, which are all from the same die pair. See also issue 149.|
|148||dr.||E–||The issue exists, but the eagle on the Weber coin cited (now at the ANS) seems to be standing on a club, not a thyrsus, and the coin thus belongs in issue 145.|
|149||3-ob.||E–||The eagle is described as standing on a "thyrsus( ?)" and the sole example cited is now in the ANS collection. In fact it is the coin cited under issue 146, with eagle on club. It is from the dies of the Hague (now Leiden) coin also cited under 146, and those of another example of 146 in the Hersh collection. No triobols with eagle on thyrsus have been found, and issue 149 appears to be a phantom.|
|150||3–ob.||E–||Some examples at least of issues 150, 154, 155, and 157 may be coins whose markings are off flan and which therefore belong elsewhere. The eagle stands to right on 150, to left on 154.|
|153||dr.||—||Whether one accepts Thompson's attribution of this issue with to Miletus (Miletus 28-31), or Price's to "Macedonia ('Amphipolis')" it does not belong at our mint.|
|154||3–ob.||E–||See comment at 150.|
|155||2–ob.||E–||See comment at 150.|
|156||ob.||—||The reverse type of the sole coin cited is an eagle standing left, head reverted, on an uncertain object. Dr. Price kindly confirmed that the coin's poor condition made recognition of a symbol, if any; or reading of any inscription impossible. All other known obols in this Macedonian coinage have a fulmen as reverse type, and nowhere here in any denomination is there known an eagle with reverted head standing left. Small coins of Amyntas III, however, bear precisely the types of issues 156, similarly oriented (e.g., SNGANS 94–96), and thus the coin cited as the only example of issue 156 is probably of that earlier king.|
|157||ob.||E–||See comment at 150.|
Reattrib., p. 17. See p. 92.
"Cavalla," p. 40 (discussion of hoard coin 17).
See p. 21, note a.
Obverse links provide by far the most important evidence for the order of the Alexander groups. These links, together with group A's use of symbols found in Philip II's coinage (immediately prior or perhaps for a time contemporary), the presence of the title BAΣIΛEΩΣ on five of the groups, and certain repetitions of reverse markings put all the groups into a firm order, with the one exception of the minute group K (whose placement will be discussed below). Some small confirmation of this order is provided by other types of evidence–hoards, stylistic considerations, and the small denominations of Chapter 2.
The 22 die links which have been discovered between the various Alexander groups are detailed on the following pages and summarized in Figure 4. Tetradrachms provide all but five: links 6 (drachms), 7 (diobols), 8 (obols), and 15-16 (didrachms). All coins known from these obverse dies shared by more than one group are described as a possible aid to future researchers. For the same reason, Newell's provisional tetradrachm obverse die numbers are also given, as the ANS's casts and photo file cards are marked with these numbers.
Further intra-group connections of the tetradrachms listed via reverse links are mentioned in the discussion following each die link in order to demonstrate further the complexity of the die linkage between issues within the groups and to show that the issues directly involved in the links between groups are often clearly contemporary with other issues in their groups. The reverses of the coins listed are described by Newell group letter, my issue number, and symbol, e.g., "A2, stern," while "same die" indicates that the reverse die is that of the immediately preceding coin.
The evidence is extremely incomplete or there would doubtless be more instances of links such as link 3, where a die was used for group B, then for A, and then for B again.
Group A with Group B
Link 1, tetradrachms, Newell obverse 25
A2, stern (215) ANS; ANS
B7, grapes (216) Toronto
B7, grapes (217) cast marked "Demanhur"; Naville 6, 28 .Jan. 1924, 721, same die
B7, grapes (218) Ball 6, 9 Feb. 1932, 167, same die
Breaks in the lion's mane commence on the two coins in stage 1, and become ever larger in succeeding stages.
Link 2, tetradrachms, Newell obverse 28
B7, grapes (219) formerly ANS = Reattrib., pl. 7, 12; ANS, same die; Oxford = SNGAshm 2538; Morgenthau 342, 26 Nov. 1934, 189, same die
A2, stern (220) ANS, stern cut over 219's grapes; ANS = Reattrib., pl. 7, 11, same recut die; Saroglos
A2, stern (221) ANS
The reverse die of 219 and 220 is the same but, when used for 220, group A's stern symbol had been cut over B's grapes. As noted above, Newell illustrated coins with stern and grapes in Reattrib. to show their obverse identity, but did not recognize the reverse identity and recutting at the time (his evidently subsequent ticket in an ANS coin's box, however, does describe the recutting).
In stage 2 slight deterioration has appeared around Heracles' mouth.
Link 3, tetradrachms, Newell obverse 47
B2, amphora (222) ANS
B2, amphora (224) Berlin, die of 222
A3, double heads (225) ANS; ANS
B2, amphora (226) ANS
In stage 1 there are no breaks in the dotted border at the top of the die, no break between Heracles' brow and the border, and no break in the field at the top of his nose. In stage 2 slight breaks have appeared in all three areas. In stage 3 the breaks in the border and at the brow are more pronounced, and the field behind the lion's mane is starting to deteriorate. Clearly at least some of A's double-head coins and B's amphora coins were struck simultaneously. The last coin listed, with amphora, is linked by a net of reverse and obverse dies to all seven of the other symbols of group B. All but one of these die links are found among coins in the ANS collection.
Link 4, tetradrachms, Newell obverse 40
A3, double heads (227) ANS = Reattrib., pl. 1, 8; cast marked "in trade, Cairo," same die; ANS; Knobloch FPL 33, Apr. 1968, 530, same die
Only in stage 2 are there die breaks at the corner of Heracles' mouth and on his neck below the lion's jaw. The first ANS coin in stage 2 is linked by its reverse die to another in the ANS collection, which is from the obverse die of a third there, from the B6 ivy leaf issue.
Link 5, tetradrachms, Newell obverse 52
B1, cantharus (229) ANS
A4, fulmen (230) ANS; Egger 10, 2 May 1912, 592. same die, not illustrated but a cast is at the ANS
B1, cantharus (231) Saroglos, die of 229; Coin Galleries FPL 5.3 (1961), C19 = Coin Galleries, FPL 4.3 (1963), C18, same die
In stage 2 only, breaks have occurred at the corner of Heracles' mouth, and in the lion's ear. The cantharus coins are linked by a net of reverse and obverse dies to five of the seven remaining symbols of group B (all but B7, grapes, and B4, stylis).
Group B with Group D
Link 6, drachms
B6, ivy leaf (232) Hersh = Giessener 58, 9 Apr. 1992, 234
See also links 7 and 8.
Link 7, diobols
B6, ivy leaf to right (235) St. Petersburg
The reverse die of the coins of group D is that of the St. Petersburg example of group B, but with the ivy leaf recut to eagle head. See also links 6 and 8.
Link 8, obols
D1, eagle head (238) Hersh, cut over 237's ivy leaf; Hersh, same recut die
The reverse die of all coins is the same, the ivy leaf having been recut to eagle head on the coins in group D. See also links 6 and 7. Also from this reverse die, in its first stage with ivy leaf, but from a different obverse die, are another ANS coin and a third coin in the Hersh collection (210).
Group C with Group D
Link 9, tetradrachms, Newell obverse 102
C2, quiver (239) ANS = Reattrib., pl. 3, 9; H. Schulman, 7 July 1970, 213, same die; ANS; Egger 40, 2 May 1912, 632, same die, not illustrated, but a cast is at the ANS
D1, eagle head (240) ANS; Weber 2082, same die; Reattrib., pl. 3, 10
A cast at the ANS (from link 10's obverse 117 and 240's reverse) associates obverses 110 and 117.
Link 10, tetradrachms, Newell obverse 117
C2, quiver (241) ANS
The die is associated with that of link 9.
Link 11, tetradrachms, Newell obverse 116 = 121
C2, quiver (243) Athens
C6, bow (244) Egger 40, 2 May 1912, part of non-illustrated lot 631, but a cast is at the ANS; ANS, same die; Gillette, same die D1, eagle head (245) ANS
In stage 2, a die break appears in the central row of the lion's locks, and the field just below the locks is breaking down. Newell obverses 116 = 121 and 105 (link 12) are both found in a group C cluster of ANS coins linked by a network of obverse and reverse identities. The cluster includes all the remaining three symbols of group C.
Link 12, tetradrachms, Newell obverse 105
C3, grain ear (246) ANS
C2, quiver (247) ANS; Berlin
C3, grain ear (248) ANS
C4, trident head (249) ANS
D5, star (250) Cambridge, Mass. = Dewing 1122
In stage 1, there is a small die break just to the left of and below Heracles' ear. In stage 2 this break has enlarged, and new breaks have appeared at Heracles' nose and at the angle of his chin and neck (this last break has been cut away on 249). The die is associated with that of link 11.
Link 13, tetradrachms, Newell obverse 107
C1, filleted caduceus (251) ANS; cast marked "Pozzi," same die
D1, eagle head (252) Cambridge, Mass. = Dewing 1117; ANS; cast marked "Mrs. Brett," same die
D1, eagle head (253) ANS, die of 252; Saroglos, same die; ANS
In stage 2, a die break beginning in the field at Heracles' brow has greatly enlarged. The first ANS coin (251) shares a reverse die with another ANS coin whose obverse was used also for coins of C4 (trident head) and C6 (bow).
Link 14, tetradrachms, Newell obverse 109
C5, Pegasus forepart (254) ANS
D2, Macedonian shield (255) ANS
Die breaks are present at Heracles' nose in both stages of the die, but only in stage 2 is there also a break in the hair at his brow and deterioration in the upper left field.
The reverse die of 255 is shared with another ANS coin whose obverse was used for five other issues of group D, namely, D1 (eagle head), D3 (club), D6 (filleted caduceus M), D8 (caduceus ), and D10 (club ) (see 26–27, 29, 32, 34 and 37) and with a third ANS coin whose obverse was used also for D4 (horse head).
Link 15, didrachms
C1, filleted caduceus (256) Hersh = Glendining, 7 Mar. 1957, 21; Lanz 48, 22 May 1989, 193, same die, but the symbol called bee on rose and the coin an unpublished didrachm of Pella
D5, star (257) ANS
Below the lower left lock of the lion's hair a small break appears only on the coins of group D.
Link 16, didrachms
C6, bow (260) ANS = Reattrib., pl. 15, 2
D4, horse head (261) Hersh = Giessener 60, 5 Oct. 1992, 114
D7, caduceus or possibly D8, caduceus or caduceus (262) Berlin
The last coin, 262, is extremely worn, but the obverse does seem to be that of the other coins.
Group D with Group E
Link 17, tetradrachms, Newell obverse 159
D5, star (263) cast marked "Case"; ANS; ANS, same die
D11, dolphin (264) ANS
E1, rose (265) Copenhagen = SNGCop 672
Either the die or the flan was defective when 265 was struck, as the type is missing in a large arc around the upper edge of the coin's obverse. The small E1, with rose, is known from but three coins and two obverse dies. One die, here, is shared with group D coins; the other, with another issue of group E (40, 44). The rose issue could thus belong with either group D or group E, but is here left where Newell placed it.1 In either case, an obverse link between D and E results.
Group E with Group F
Link 18, tetradrachms, Newell obverse 361
E3, cock (266) ANS; Parke-Bernet, 16 Oct. 1968, 23, same die; Grabow 14, 27 July 1939, 244, same die; ANS
E3, cock (267) ANS; Münz. u. Med. FPL 333, Apr. 1972, 11
F3, cornucopia (268) ANS
In stage 2, a dot just to the left of and below the lion's ear has enlarged, and another break has appeared to the left of and below the first one, between the second and third locks from the top in the outer row of the lion's mane. The reverse die of 268 is shared with another ANS coin whose obverse was used also for a coin of F5 (bow and quiver).
Group F with Group G
Link 19, tetradrachms, Newell obverse 427 = 490
F4, Athena Promachus (269) ANS = Reattrib., pl. 9, 3
G2, Athena Promachus (270) ANS = Reattrib., pl. 9, 4; Petsalis, same die
Group I with Group J
Link 20, tetradrachms, Newell obverse 681
In stage 2, the obverse has suffered general deterioration, and looks "softer," with breaks at Heracles' nose and to the right of his ear, and in the lion's locks.
Link 21, tetradrachms, Newell obverse 702
J1, grain ear (275) ANS
Group L with Λ-Bucranium Group
Link 22, tetradrachms, Newell obverse 896
Λ over bucranium in left field, E under throne (277) Saroglos; unidentified photo (278), same die
Although groups after L have not been examined in detail for this study, link 22 has come to my attention. Mando Oeconomides has verified that the Lamia hoard obverse and reverse casts are indeed of a single coin.
In Figure 4, solid brackets show tetradrachm links, and dashed brackets show links between smaller denominations. Brackets to the left indicate the 22 obverse links found between the Alexander groups, and those to the right show reverse links resulting from recutting of the reverse dies. Tetradrachms furnish 17 of the links and the remaining five are found among smaller denominations (which exist only in groups A through F). Arrows on the brackets show the order, when ascertainable, in which the dies were used. Numbers on the brackets are those of the links already described. Dotted brackets to the right indicate multiple identical reverse markings (groups F and G, J and L). As shown, groups G through K/J include the title BAΣIΛEΩΣ in their inscriptions.
See p. 22, note b.
Given the framework of obverse die links just detailed, other evidence does little more than confirm the order they provide. Still other observations are all perfectly consistent with the order in Figure 4 and will be discussed below in Chapter 9, in connection with the mint's absolute chronology'.
As Newell long ago wrote, the Kyparissia 1892/93 hoard, with its coins of groups A through D only, showed these four groups to be the earliest struck. Karditsa 1925 included coins of C through I, seven contiguous groups. Five hoards ending with group J are known. Of these, Akçakale 1958 contained every group except A and the small K, and Demanhur 1905 and Andritsaena ca. 1923 included every group, even K.2
Newell dealt with details of style and iconography, and the progression from group to group, at some length in Reattrib. His analyses cannot be improved by the present author, but such aspects as are relevant to absolute chronology, whether or not treated by him, will be discussed below in Chapter 9.
Not surprisingly, the present study of the small Alexander denominations only corroborates the group order already established, although it does provide the only actual die links known between groups A and B and the rest of the coinage. The eagle-reverse coins of various denominations are found only in A through E, and only in E do the Zeus-reverse drachms come in, which then are the only small coin struck in the following group F. No small coins of Alexander's types are known after group F.
See Chapters 8 and 9 for fuller discussion of these hoards.
Newell stated in Demanhur , without giving specific examples or illustrations beyond those few presented in Reattribution, that the tetradrachm groups were all bound in sequence by a series of obverse dies linking one group to the next: "... group 'A' will possess certain dies that were used in its production and then were continued in use, in a slightly more worn condition, for group 'B.' Group 'B,' in turn, will be found to possess certain obverse dies that had already been used for 'A,' and others that were later used for 'C,' and so forth."3 This account of the groups' linkage is somewhat of a simplification. Newell knew most of the links presented above. He apparently did not know the B-D or D-E links, and he evidently did not realize until after Reattribution's publication that at least some of group B was contemporary with group A.4 Further, no B-C links such as he suggests have been located.
At least since the publication of Reattribution, group A has been recognized as the first, because three of its symbols (prow, stern, and double heads) are the same as those found at the end of the lifetime or early posthumous coinage of Alexander's father, Philip II.5 And, although its shape is different in the two coinages, Le Rider has suggested that the rudder, which occurs rarely in Philip's issues, is a possible fourth symbol relating group A to Philip's coinage.6
Group B, repeatedly linked to A, should be next. But the first modification of Newell's order is that here some overlap between groups must be accepted, because of the links where an obverse die was used first for a coin or coins of group B before being used for group A (links 2, 3, and 5 above), and because of the unique recutting of a symbol of group B to one of group A (see link 2).
Groups C and D, linked by no fewer than eight obverse dies, are clearly contiguous. Group D would at first seem to have followed C, because, of the five shared obverse dies whose priority of use can be determined, all five were first used for group C. A complication is, however, introduced by links 6-8, where drachm, diobol, and obol obverses were used both for B and for D, the two smaller denominations having had their reverse symbols recut from one of group B to one of group D.
Because of the large number of obverse links between A and B and between C and D (a pattern which does not recur), and because of the newly recognized B and D links, it now seems probable that A and B were struck concurrently at two adjoining locations, followed by C and D at the same two respective locations (workshops? adjoining rooms? adjacent anvils?). If group C had chronologically separated B and D, all three groups emanating from the same workshop, it is hard to see why new dies should have been cut for C, while B's dies were preserved unused until returned to service, recut where necessary, for coins of group D. But certainty is not to be had, and no great violence can be done by leaving Groups A through D in their traditional order.
Following group D, successive obverse links, the introduction and abandonment of the title BAΣIΛEΩΣ, and similarities in reverse markings make the groups' order inescapable except for the position of the minute group K.
I have placed K in the tables before J, although a strict linear order is probably misleading. More interesting than the placement of K, however, is the question of its very attribution to our mint. Newell in Reattribution published only one issue of the group (K3, its largest) and assigned it to an uncertain mint of Macedonia, Thrace, or Asia Minor. By the time of Demanhur's publication, however, he had placed it, although without comment, at Amphipolis.7
Price has now argued against this attribution, considering group K (the Λ group) as the immediate predecessor of the Λ- or -bucranium and Λ- or -torch series–which he considered struck at Amphipolis. He posited that groups A–I, J, and L belong together, but without successors, at another mint, presumably Pella.8 I would not necessarily disagree with his suggestion that the mint for the huge output of groups A through L and their successors may have changed at some point. His suggestion of an introduction at Pella with a subsequent move to Amphipolis could possibly be true. But this study attemps to deal with numismatic evidence only, and that evidence seems at the very least to contradict the division at the particular point that Price suggests. Precisely because his monumental work will inevitably and deservedly become the standard reference for Alexander's coinage, I should like to respond here in some detail to Price's arguments.
First, he assumes that the title of BAΣIΛEΩΣ, once dropped (as it was in group L) would stay dropped, that there would be no brief recurrence. This is surely correct.
Second, he states that group J (the -group) follows directly on the symbol-only issues of groups A-I. This also seems correct, although not for the reasons he gives.9
Third, he says that group L (the -group) should follow directly on J for two reasons. One is that is an elaboration of : this is of course quite possible but not necessarily so. The second reason is the shared symbols between J and L, which is quite convincing.10
And, as group L first drops the title BAΣIΛEΩΣ,11 Price concludes that there would appear to be no room in the sequence for group K (the Λ-group), which bears the title. It then, he says, will have been the direct predecessor, but at another mint, of the Λ-bucranium and Λ-torch groups. His reasoning is tight and would be persuasive, but the separation of group K from our mint seems almost certainly impossible in the light of the four die links now known between the posthumous Philips analogous to group K and those analogous to group J. Moreover, any suggestion that dies might have been transferred from J at our mint to K as the initial group at another mint is ruled out by the observation that in the Philip link where priority of use can be determined, the die was used for coins of group K before being employed for coins of group J.12
Yet it remains quite true, as Price has pointed out, that K does not logically fit in the sequence either before or after J. The resolution is again provided by the study of the contemporary Philip groups, some analogous to J and K, some not, but all so tightly and intricately obverse linked that the only explanation seems to be that all were more or less contemporary.13 The tiny Alexander group K, if also struck concurrently with J, which would seem likely, then presents no problem. Price's sequence A through I to J to L is preserved, yet K being contemporary with J means that our mint need not be divided into two, at least at the spot Price proposes.
Newell in Demanhur placed group K after J, presumably because of the single die link which his tickets show that he knew between I and J. More recently, both Le Rider and Thompson have preferred to place K before J,14 but the disagreement is meaningless if K was contemporary with J. But because some order is inevitable in a serial listing, I have opted, despite the two I-J links, for K before J because of the more numerous shared dies among the analogous Philip II reissues. Another consideration is that after group I two markings rather than one identify the various issues and a primary marking for each group is accompanied by a varying secondary marking. Only in K is there inconsistency in the placement of the two markings, with the primary one either in the left field or below the throne and the secondary one in the other spot.15 In J and L, however, the placement is unvarying. Unfortunately then, the unavoidable strict linear order presented in the tables does not, in the case of group K, accurately represent reality.
The last group in this study, L, despite its superficial similarity to group J ( instead of , and the two groups' shared secondary markings), is a totally different outpouring from group J. No obverse links connect the two groups, and only one possible but quite doubtful link joins the analogous Philip groups 8 and 9.16 Several hoards contain coins of all or most groups down to and including J, but not L. Group L drops the title BAΣIΛEΩΣ present on the five preceding groups. And, while abundant small-denomination coins (of Philip's types) accompany groups K and J, none are known that are analogous to group L.17 may resemble –indeed may well be an elaboration of –but the two groups of coins are completely different.
Demanhur , pp. 65–6.
See discussion following, and comments on link 2, above.
Reattrib., p. 21; Philippe , Amphipolis group IIB.
Philippe, pp. 389–90.
Reattrib., p. 40, issue 62; Demanhur 1582.
Alexander, pp. 86-87, expanding on arguments previously given in his "On Attributing Alexanders–Some Cautionary Tales," in Greek Numismatics and Archaeology. Essays in Honor of Margaret Thompson , ed. O. Mørkholm and N. M. Waggoner (Wetteren, 1979), pp. 241–50, at 247–49.
He adduces obverse links between a coin with and laurel branch, and coins with crescent alone and with laurel branch alone. These latter two, however, are merely examples of a few rare, perhaps early or perhaps only poorly executed coins of group J. They are not part of a group of their own, nor are they connected to any earlier issues. See Chapter 1, issues J1 (grain ear alone, 3 coins and 2 reverses known), J2 (crescent alone, 3 coins and 2 reverses known), and J3 (laurel branch alone, 2 coins and 1 reverse known). Nevertheless, a firm tie between group J and earlier groups is provided by the two obverse dies now known to be shared by I and J. See links 20 and 21 above.
Price adduces four shared symbols: filleted caduceus, grain (or corn) ear, crescent, and laurel branch. Of these, only two (grain ear and crescent) seem to be shared. See the commentary at the end of Chapter 1 on Price's issues 127 ( and filleted caduceus" and 140 (" and laurel branch"). Nevertheless, among the Philip issues analogous to groups J and L there are four or possibly five common symbols. See p. 53, Table 7, groups 8 and 9. Therefore, again, group J does seem closely connected to L.
See Chapter 6, links 14–17 especially 14 and 17. Further, contrary to Price's assertion, Newell's trays, provisional die numbers, and notebook for both the Alexander and Philip series make it clear that his order was group J, K, L, Λ or bucranium, torch, Λ-torch. The use of the letter Λ is not limited in any case to group K and the Λ-bucranium and Λ-torch groups: it is found in Philip groups 5 and 6, contemporary with 8 (with ), and also in Philip group 9 (with ).
See Chapter 6.
Philippe, p. 397, n. 5; Sardes and Miletus , p. 88, n. 90.
See Chapter 6, link 18.
See Chapter 5.
No even reasonably satisfactory study of the Alexander coinage of Amphipolis can omit a study also of the late reissues of Philip II tetradrachms and smaller coins which parallel many of the posthumous Alexander issues. These tetradrachms' obverses depict a handsome head of Zeus, and their reverses bear the simple legend ΦIΛIΠΠΟΥ and a nude mounted horseman. A summary of the Philip tetradrachms whose markings correspond to those of the Alexanders of groups K, J, and L, and perhaps I, follows. These late Philip II reissues continue beyond those shown here, which end with those contemporary with Alexander group L.1
Table 7 summarizes the post-323 Philip issues through those analogous to Alexander group L. These late issues form Le Rider's Philippe Amphipolis silver group III.2 Le Rider gives only a brief overview of this group, not the thorough die study accorded Philip's earlier Amphipolis silver. Essentially he presents a list of issues to which a few corrections now seem justified. These are given at the end of this chapter.
The numbers assigned the Philip tetradrachm groups here are not Le Rider's (who gives none), but the present author's. Numbers were chosen rather than letters in order to differentiate the Philip groups from the Alexander groups. The order of the Philip groups here is for the most part that adopted by Le Rider, who remarks that his order is in many cases arbitrary.3 The only changes made here are that group 7, listed as two separate contiguous sections in Philippe , is presented as a single group and placed after rather than before group 6 because 5 and 6's secondary markings are largely identical. Also, the fractions of groups 2 through 6 (group 1 has none) normally bear both issue markings of their analogous tetradrachms, while 7's fractions, like those of group 8, bear only the secondary issue markings.4
The conclusion reached below in Chapter 6 will be that most if not all of these Philip groups were issued more or less simultaneously. The numbering of the groups has been adopted for convenience of reference, as we do not seem to be dealing here with a linear sequence of groups (see Figure 5 for the complicated die linkage among groups 2–8).5
Unlike the Alexander groups', the Philip groups' issue markings clearly show the internal coherence of each group. Many internal die links are known, some published in the summary in Philippe , and more in SNGANS. At least one example of each tetradrachm issue is illustrated here, on Plates 12–14, and a few internal die links are also shown which do not appear elsewhere.
Table 7 gives the issue markings found in each group, the plate locations of representative examples, Philippe plate references, initial SNGANS numbers, and the number of examples located for each issue. Regardless of their positions on the coins, the primary marking is given first, followed by the secondary one. When an issue has the same marking or markings as an Alexander issue, the Alexander issue's group letter and issue number are given in bold type, before the Philip issue's markings: e.g., K2before the first issue in group 7 indicates that this Philip issue has precisely the markings of Alexander group K, issue 3. Markings given in parentheses are known only in the Philip fractions and are included merely to fill out the issue list, as examples of such tetradrachms may well surface some day.
|Markings||Plate||Philippe Plate||Initial SNGANS Number||Example Found|
|Group 1, 1 coin|
|I1 ?||279||43, 1||–||1|
|Group 2, 9 coins|
|or ivy leaf||282||43, 10||572||3|
|( or star)|
|Group 3, 16 coins|
|amphora (club ?)||285||–||576||1|
|ivy leaf||286||44, 11||577||7|
|star||288||44, 5, 6||579||3|
|Group 4, 7 coins|
|Group 5, 13 coins|
|Causia E||295||44, 29||592||3|
|Causia M||296||44, 30||–||2|
|Causia T||297||44, 31||593||2|
|Causia, globule, A||298||44, 32||594||4|
|Causia, glouble, E||299||44, 33||–||1|
|(Causia, globule, M)|
|Causia, globule, T||300||–||–||1|
|Group 6, 45 coins|
|Wreath A||301||45, 24||600||7|
|Wreath E||302||45, 25||603||6|
|Wreath Λ||303||45, 22||606||6|
|Wreath M||304||45, 26||607||7|
|Wreath T||305||45, 27||610||14|
|Markings||Plate||Philippe Plate||Initial SNGANS Number||Example Found|
|Group 7, 72 coins|
|K2||Λ or||308||45, 5, 6||630||25|
|K3||Λ||310||45, 15, 16||643||16|
|Group 8, 93 coins|
|J4||grain eri||314||46, 3||667||27|
|forked branch||316||46, 5||683||6|
|profile shield||318||46, 6||688||6|
|Trident head||319||46, 1||691||2|
|trident head||320||46, 2||–||5|
|Macedonian shield||321||46, 7||692||4|
|Group 9, 235 coins|
|L1||forked branch||322||46, 11||–||6|
|L3||aplustre||323||46, 17, 118||738||86|
|L4||grain eri||326||46, 12||736||8|
|L7||dolphin||331||46, 9, 19||749||95|
|L8?b||profile shield||333||46, 13||–||4|
Group 1 may well be a phantom. One single tetradrachm is known, and the fractions which Le Rider places with it in Philippe solely on the basis of style would seem instead to belong with others with the same issue markings, which clearly belong in other groups.6 The tetradrachm's monogram may well be a variant of group 2's or , and perhaps the coin should be included in that group. Group 1's monogram also is identical to one variant in Alexander issue I1, and, as other Philip reissues repeat some markings of Alexander groups K, J, and L, it remains possible that the Philips commenced as early as Alexander group I.7
The composition of groups 2 through 7 is self-evident and the primary markings clearly show which coins and issues belong in each group. Groups 8 and 9, however, present problems. These are the coins with the primary marking or . The groups with these markings, both Philips and Alexanders, were for the most part poorly and often carelessly made, apparently in some haste. The two series in each king's strikings used many of the same secondary symbols, but are subject to being confused because of the similarity of the primary markings and , which differ only by a single dot. The correct attribution of an Alexander, even with a poorly or imperfectly executed letter or monogram, is simple because group J, with , included the title BAΣIΛEΩΣ in the inscription, while group L, with , did not. Among the Philip coins, however, the attribution depends solely upon whether the marking is or and, given the often poor workmanship involved, it can be virtually impossible to decide whether the presence or absence of the critical dot is intentional or accidental. Further, there exist numbers of barbaric imitations of the Philips, especially in these problematic groups 8 and 9 and in following groups also. Obvious imitations have been excluded from this study, but some may well not have been recognized. Some group 8 and 9 coins are possibly wrongly attributed in Table 7, but the overall picture should be approximately correct.
More important is the possible, but highly uncertain, die link between Philip groups 8 and 9 which results from taking a few coins at face value, that is, trusting that their markings are intentional and not the result of carelessness or accident. For discussion of the coins involved in these links, see Chapter 6, link 18, and p. 53, note b.
A coin with causia and Λ is reported in Philippe , p. 122 and p. 308, 281. The coin is not counted among the examples listed, as I have not seen it.
Although the issue markings are those of Alexander's L8, the Philip issue may well be a phantom. Four coins are known, from two die pairs. One die pair is illustrated here (333), but the reverse's general aspect is a bit odd (note in particular the unique orientation of the shield). This may be an ancient imitation, a common occurrence in group 9. Indeed, Newell marked an ANS cast from these dies as "Barbarian."
The other reverse die (Hunter, p. 291, 61, and the Paris coin, here 449) has a very fine, faint dot below the , very likely not made by the same tool which engraved the , and in the Hunter catalogue itself the marking is described as a simple . If the dot on this second reverse is a mere accident, the die would belong to group 8–which is made more likely by the fact that this reverse's accompanying obverse is found also in group 8's -crescent issue, forming the only possible die link between groups 8 and 9, see Chapter 6, link 18. It is worth noting that the Paris cast at the ANS had been placed by Newell with his casts of group 8, not 9.
Besides the coin illustrated, three other examples of the issue are cited in Philippe , p. 308, 717–19. Not seen by me, these three coins are not included in the count of examples located.
See "Tetradrachms Amphipolis" for a summary of later Alexander and Philip issues.
Philippe , p. 120–24.
Philippe , p. 120.
See Chapter 5.
See p. 69.
See pp. 63–64, comments on Philippe, pl. 43, 2–8.
The size of each group, as judged from the estimated number of obverse dies employed, seems to bear little relation to the number of issues in the group. Contrary to what one might at first assume from Philippe's treatment of these strikings, essentially a listing of issues, the sizes of the groups varied widely, from 3 to 56 estimated dies used for a given group. Table 8 shows the numbers of coins and of obverse dies located, the coin to die ratios, and, as in the similar table of Alexander tetradrachms above, Table 2,8 the estimated number of obverse dies employed for each group. All conventions are those of Table 2.
|Group||Coins||Obverse Dies||Coin Die Ratios||Estimated Obv. Dies|
See p. 70.
See p. 26.
Plate 46, 8, "aplustre and ." The issue may exist, but this particular coin does have a faint dot within the , and belongs to group 9's very large -aplustre issue. I am most grateful to Martin Price for a direct photograph of the coin (324) and an enlargement of the reverse. It is from the dies of Münz. u. Med. 13, 17 June 1954, 1096, and from the reverse of 325, both of which clearly show the . The obverse of 324 is not known elsewhere and 325's is known only in group 9: Myers, 11 May 1972, 18, aplustre; 329, wreath; and a cast at the ANS, dolphin.
Plate 46, 9, "dolphin and ." The ANS has a cast of this coin, which does seem to have a dot present, joined to the inner edge of the right perpendicular element of the . As the coin in question would be the only known example of the supposed -dolphin issue, it almost certainly is merely a poorly executed specimen of the extremely large -dolphin issue of group 9, where its poor, flat relief would be typical.
Plate 46, 12, "laurel branch and ." The coin would be the only known example of this supposed issue (note, however, the fractions with a horizontal, quite different branch).9 It seems more likely that the symbol of pl. 46, 12, is a poorly engraved grain ear, an issue not listed in Philippe , but of which several examples are known, e.g. 326. Ineptly engraved grain ears are common also on Alexanders with , e.g. 95–97.
See pp. 58 and 62.
By far the chief subdivision of the post-323 Philip reissues is a small coin with the head of Apollo wearing taenia on obverse1 and ΦIΛIΠΠΟΥ with a nude horseman on reverse. The denomination of these little pieces is unclear. As Le Rider points out, they are certainly too heavy to be considered tetrobols on the standard of the tetradrachm of the period (ca. 14.29–14.39 g), which would require a coin of, at most, 2.38–2.40 g. Nor are they heavy enough to be truly fifths of a tetradrachm (ca. 2.86–2.88), such as the fifths with the same types were in the lifetime coinage of Philip. Le Rider suggests that these fractions could pass at their period as tetrobols on the Attic standard, but on the whole prefers to regard them as fifths of the tetradrachm.2 Their correct denomination, however, being unclear, and Le Rider's persuasive "fifths of the tetradrachm" rather unwieldy, these coins will simply be called "fifths."
There are known also a few extremely rare "tenths" and several examples of what must be drachms on the Attic weight standard which belong with these abundant post-323 Philip fifths. These other denominations will be discussed briefly later in this chapter.3 A few corrections to Le Rider's small-coin listings are also given at the end of the chapter.
Table 9 presents the issues found of the fifths. The first column gives the issue's markings (primary marking before the secondary one, regardless of their position on the coins) and the second the plate location of a representative example or examples. Plate numbers in Philippe form the third column, and the fourth gives the issues' initial coin numbers in SNGANS. The last column gives the number of examples found of each issue. Brackets to the left of the plate references indicate obverse die links, those to the right, reverse links.
Some small issues cannot be definitely assigned to a particular group, namely those with the single markings of globule or amphora (group 2 or 3), and star (group 2, 3, or 4). The last issue listed, with simple straight laurel branch, can only probably be placed in group 8.4
|Markings||Plate||Philippe Plate||SNGANS Number||Examples Found|
|Group 2, 9 coins|
|or , star||339||—||574||2|
|Group 3, 22 coins|
|ivy leaf||340||44, 14||581||4|
|uncertain marking||348||44, 12||586||2|
|Group 2 or 3, 5 coins|
|Amphora||350||43, 6, 7||588||3|
|Group 4, 9 coins|
|353||44, 23, 24||—||7|
|Group 2, 3, or 4, 6 coins|
|Group 5, 13 coins|
|Causia A||355||44, 34||596||1|
|Causia E||356||44, 35||—||2|
|Causia Λ||357||45, 1, 2||597||4|
|Causia M||358||45, 3||—||1|
|Causia, globule, M||359a||—||598||1|
|Causia, globule, T||360b||45, 4||599||4|
|Group 6, 43 coins|
|Wreath A||361||45, 30||616||16|
|Wreath E||362||45, 31, 32||621||5|
|Wreath Λ||363||45, 29||622||7|
|Wreath M||364||45, 33||625||7|
|Wreath T||365||45, 34||628||8|
|Group 7, 50 coins|
|, or||366||45, 7–9||650||20|
|Group 8, 115 coins|
|Forked branch||379||46, 26–27||711||29|