Edson 1947 TopographyofTherme

Notes on the Thracian Phoros
Author(s): Charles Edson

Source: Classical Philology, Vol. 42, No. 2 (Apr., 1947), pp. 88-105
Published by: The University of Chicago Press
Stable URL: http://www.jstor.org/stable/265819
Accessed: 08-07-2015 19:34 UTC
REFERENCES
Linked references are available on JSTOR for this article:
http://www.jstor.org/stable/265819?seq=1&cid=pdf-reference#references_tab_contents
You may need to log in to JSTOR to access the linked references.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://www.jstor.org/page/
info/about/policies/terms.jsp
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content
in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship.
For more information about JSTOR, please contact support@jstor.org.
The University of Chicago Press is collaborating with JSTOR to digitize, preserve and extend access to Classical Philology.
http://www.jstor.org
This content downloaded from 129.81.226.78 on Wed, 08 Jul 2015 19:34:39 UTC
All use subject to JSTOR Terms and Conditions
NOTES ON THE THRACIAN PHOROS
CHARLES EDSON
TI~ HE publication of the Athenian monarchy and on points in the historical

Tribute Lists' has marked an epoch geography of the Macedonian area.4
in the study of the fifth century,
ALvLaTat,AIveaTat,AlvecaTat
and it is hardly necessary here to add to
(Pp. 220-21, 464-65)
the praise which "this noble work"2has so
justly received. The aim of this paper is The editors rightly place Aineia on the
modest: it is primarily to re-examine the great promontory, Megalo Karaburnu.
existing evidence for the location of cer- To the evidence which they adduce may
tain cities which are entered in the Thra- be added Livy's statement that Aineia
cian panel of the tribute lists. The point of was 15 Roman miles from Thessalonica
view is that of a student of Macedonian, by sea.' The actual distance from Salonica
not Athenian, history; and hence the by sea to the modern fort at Megalo Kara-
places here considered are among those burnu is a little over 11 English miles.
which were, or are now thought to have This is decisive support for the proposed
been, in the sphere of interest of the Ar- location. Indeed, there are few sites in the
gead kings. Much of what follows may north Aegean whose locations can be so
seem nagging and picayune argument over accurately determined by literary evi-
minor points, but an investigation into dence alone.6
problems of ancient historical geography 4 Each city discussed below is listed in alphabetical
must necessarily be concerned with de- order with its ethnic in the form (or forms) which ap-
pear in the tribute lists. After the ethnic are added two
tail. In brief, it is my contention that the references to A TL, I; the flrst is to the pages of the
tribute lists offer no evidence for Athenian Register where the city's record Is displayed, the sec-
ond to the geographical discussion in the Gazetteer.
control of any place on the north coast of All map references, unless otherwise specifically stat-
the Thermaic Gulf north of Cape Aineia ed, are to the recent British General Staff Maps
(1: 100,000), GREECE. Each sheet is designated by its
during the fifth century or on the west number and name, e.g., E. 9, POLIYIROS. Because of
coast of the gulf before the outbreak their large scale and grid, which permits exact topo-
graphical reference, these sheets are of great value for
of the Peloponnesian War.3 The northern serious studies In the historical geography of main-
and western shores of the gulf were Mace- land Greece. Modern place names, with rare excep-
tions, are spelled as they appear on these maps. It is
donian territory. By contrast, the con- to be noted that many of the local place names in
sistent Athenian interest in the Strymon Macedonia have been changed. This can be a fruitful
source of error, since many of the new official names
region is notable. There has also been oc- are antique (or pseudo-antique) and have been given
casion to touch on aspects of the relations to localities often without any certainty that the an-
cient name corresponds at all to the modern site. To
between Athens and the Macedonian avoid possible confusion I have frequently used the
older form of the modern place name.
'B. D. Meritt, H. T. Wade-Gery, and M. F. s Livy xliv. 10. 7: ".... omissaque Thessalonicae
McGregor, The Athenian Tribute Li8ts. Vol. I. Cam-
oppugnatione Aeniam inde petunt. quindecim milia
bridge, Mass., 1939. Pp. xxxii +605, with 25 pls. and passuum ea urbs abest."
map. Cited hereafter as "ATL, I."
6The editors appropriately cite [Scymnus] 627-28:
2 The phrase is that of A. W. Gomme In A Histori- Kx&4aPr&rT2P &KpdlP 5i iiP scaXOVPh',V / Alvetav. Dionysius of
cal Commentary on Thucydide8, I (Oxford, 1945),
Halicarnassus Roman Antiquities i. 49. 4 also associ-
33, n. 1. ates Aineia with a promontory: ve@w 'AOpo&1rqs 6pv0avro
a So Gomme, ibid., p. 215. fri Tra &KpwTr7p1wP &S Kal wr6X&v
Atvetav &rtaav.
[CLASSICAL PHILOLOGr, XLII, APRIL, 1947]
NOTES ON THE THRACIAN "PHOROS" 89
However, this certain location has been 7rpcv in its meaning of "foreland," "head-
complicated with the editors' reliance on land," or "promontory," for the word can
Lycophron Alexandra 1236-37 as their mean simply a high mountain or peak, as
chief source:7 in Aeschylus (Frag. 12, ed. Smyth in the
6s[Aeneas] 7rpoa /uev 'Pa1K?)XOVoi'oO&et,uoAXcv,
"Loeb Classical Library"):
Ktovooi 7rap' aL7rvv7rpw-.va. Ha-y-yatov yap apyvp'XaTov
7rpcOV' aLOTpa7rts <
(rL,unrXt7n> 7reVKaEv aeTXas.
Kissos they equate with Megalo Kara-
burnu itself, abandoning the traditional Pangaeum is not a promontory. The
identification of Kissos with Mount Khor- steeply rising Mount Khortiatis, very
tiatis (Chortiach), southeast of Salonica.8 slightly less than 4,000 feet high,11is only
But this new identification cannot be rec- about 18 miles northeast of Megalo
onciled with other evidence. Xenophon Karaburnu. There is, then, no cogent
(Cynegeticus 11. 1) speaks of Kissos, a reason to abandon the identification of
home of big game, as a mountain "above Kissos with Khortiatis, which so well ful-
Macedonia" (KrTTOV T6V brip T'S MaKe8OVI- fils Xenophon's description. Lycophron,
as), a phrase which can never have been who is after all a poet, not a descriptive
used to describe Megalo Karaburnu.9 In geographer, can quite correctly speak of
this passage Xenophon associates Kissos Aineia as situated "beside the steep crag
only with such high mountains as Pangae- of Cissus."12
um, the Mysian Olympus, the Pindus, and By equating Kissos with Megalo Kara-
Nysa in Syria. In such company Megalo burnu, the editors find themselves forced
Karaburnu is simply out of place.10Fur- to take Rhaikelos as a synonym for Ai-
thermore, Lycophron by no means states neia.13 But Rhaikelos is the name of the
that Aeneas settled on Kissos. Ko-o-o-V7rap' region in which the settlement was made,
ai7rvv 7rp6va need only mean that Aeneas not the colony itself: O1KIEV, not OlKLUEL'.
settled beside, or near, or in the vicinity Obviously, the whole point of these lines
of, Kissos. We are not required to take of the Alexandra is the founding of Aineia
by Aeneas. Hence Lycophron cannot
7A TL, I, 465: "Perhaps the most exact indication,
when translated out of the riddling form, is in Lyco- mean that Aeneas founded Rhaikelos. In
phron 1236....... " In fact, the location of Aineia can be his usual style of calculated enigma, he
established without any recourse whatsoever to Ly-
cophron. Lycophron is Important only for Rhalkelos. mentions the region in which Aineia was
8 Oberhummer, "Kissos" (2) and (3) in RE, Vol. founded instead of spoiling the puzzle by
XI, col. 522, with references there cited.
9
calling a spade a spade.14 It has already
When Xenophon wrote, the territory of the Ar-
gead monarchy did not extend south of Anthemus, II D. 8, THESSALONIKI, grid P-026362, gives the
which is the valley of the modern Vasilikotikos (Vasi- summit of Khortiatls as 1,201 meters, thus 3,940
lika Deresi) River, southeast of Salonica and directly feet. The older British General Staff Map No. 2097,
south of Mount Khortiatis. And in the generation be- sheet "Mt. Athos," dated October, 1908, gives the
fore Philip II the Argead hold on Anthemus seems to height of "Chortiach" as 3,937 feet.
have been precarious (cf. Demosth. 6. 20). 12 The rendering of A. W. Mair in the "Loeb Clas-
10 The actual Cape Megalo Karaburnu proper does sical Library." It Is, of course, quite gratuitous to conI-
not rise to an elevation of over 54 meters (see D. 8, nect Kissos with Kithas (ATL, I, 465 and n. 2, 502,
THESSALONIKI, grid 0-7829). And one must proceed 540-41).
Inland from the cape for over 6 miles before coming to 13 A TL, I, 465, n. 2: "Lycophron is certainly speak-
a height of over 200 meters (E. 8, EPANOMI, grid ing of Aineia and Cape Aineia; if Rakelos and Kissos
0-910234, south of Kardhara). The highest point in are not proper synonyms for these, then his evidence
the whole area south of the Vasilikotikos Valley and is of no use for siting Rakelos and Kissos. We see no
east of the coast appears to be the height near Vavdhos reason, however, to doubt that they are proper syno-
(E. 9, POLIYIROs, grid P-194173), 938 meters; but this nyms."
is around 30 miles east of Megalo Karaburnu. The 14 So taken by A. W. Mair in the "Loeb Classical
gradually rising broken country east of the cape can- Library": "He shall first come to occupy Rhaecelus
not be mnadeinto a high mountain. by the steep crag of Cissus."
90 CHARLES EDSON
been shown that the identification of That Aineia was the foundation of Aeneas
Kissos with Megalo Karaburnu must be was the opinion of antiquity without ex-
abandoned. It is therefore apparent that ception, including the inhabitants of
the ancient name of Megalo Karaburnu Aineia, whose unique archaic tetradrachm
was Rhaikelos. Rhaikelos was the cape on bears on the obverse a representation of
which Aineia was situated. Aeneas carrying Anchises from Troy.'8 It
The supposed identity of Rhaikelos seems clear that the legend of the found-
with Aineia has led the editors to the con- ing of the town by Aeneas must have been
clusion that Aineia was founded by Pisis- well established by the time of the minting
tratus.15 The evidence is Aristotle (Ath. of the earliest coins.19In the early second
pol. 15. 2): KaL 7rpcLTOV /LEV OVVqJKLoE(Pisistra- century B.C. Aeneas was still worshiped at
tus) 7rEp r6v 0Ep,uaZoV Kolrov XCwpLov? Ka- Aineia as KTLcaTrTs.20
XELTaL 'PaLK-OXOS. But Aristotle says that It seems best to conclude that Aineia
the name of Pisistratus' foundation was represents a native place name having the
Rhaikelos, the name, understandably same (Thracian?) root that appears in the
enough, being taken from that of the ter- name of Ainos, the well-known city at the
ritory in which the settlement was locat- mouth of the Hebrus2' and whose sound
ed. Had it, in fact, been named Aineia by
Pisistratus, it is very odd that Aristotle neas. In i. 49. 4 he resumes his narrative. The flrst two
sentences of 49. 4 returil to and recapitulate the in-
does not say so, and even more striking formation already given in i. 47. 6, which, as we have
that the sources preserve no trace of such seen, is taken from Hellanicus. The next sentence (see
above, n. 6) describes the founding of Aineia by
a tradition, for Pisistratus was a figure in Aeneas. It is surely at least very probable that this,
whom ancient writers were interested. too, derives from Hellanicus. The next and concluding
sentence of i. 49. 4 is a short excursus on the movement
Hellanicus of Lesbos, who wrote about a of the inhabitants of Aineia to Thessalonica by Cas-
century after the death of Pisistratus and, sander.
Die anitiken Miitzen Nord-Griechen-
among other works, composed an Atthis 18 H. Gaebler,
lands, Vol. III, 2. Abt.: Die antiken Muinzen von Make-

which took the history of Athens down doniia untd Paionia (Berlin, 1935), pp. 20-21, No. 33
through the fifth century,"6 definitely on Tafel V at end. It is noteworthy that the legend on
the coins of the city down until after the middle of the
brought Aeneas and his followers to flfth century is AIN EA t, the geniitive of the city's
Krousis, the territory in which Aineia was name, which can also be taken as the name of the
founder in the nominative. The tetrobols down to
located, and very probably indeed attrib- ca. 425 B.C. have the head of Aeneas on the obverse.
uted the founding of Aineia to Aeneas.'7 It is only on the bronze coins of the fourth century
that the ethnic occurs (ibid., p. 22).
15 ATL, I, 465, nn. 1 and 2. 19 So Kohler, Berl. Sitzungsber., 1892, p. 345, n. 1:
16 FGrll, Vol. I, No. 4, Frags. 49,.170, and 172, pp. "Wenn die in dem Berliner Unicum bekannte Munze
120, 147-48. von Aineia mit dem Bilde des fliehenden Aineias ....
17 Ibid., Frag. 31, p. 117. Jacoby (FGrH, Vol. I, wegen des Stiles mit Recht in das sechste.Jahrhundert
pp. 444-45) has demonstrated that this long passage gesetzt wird, so muss die Stadt Aineia, deren Name
is an excerpt from Hellanicus. Hellanicus, as repro- Veranlassung zur Localisirung der Aineiassage in
duced by Dionysius (i. 47. 6), says that Aeneas crossed dieser Gegend gegeben hat, ialter sein als der Ovvoc-
the Hellespont to the nearest peninsuila extending Kcry6s des Peisistratos." The editors (A TL, I, 465,
from Europe, Pallene by name. Here the name n. 2) suiggest that this tetradrachm "was issued by
"Pallene" is extended to include the area north of the Peisistratos when he made a city of Rakelos."
isthmus of Potidaea. But the decisive point is that 20 Livy xl. 4. 9: "Proflciscuntur ab Thessalonica
Dionysius states that the territory where Aeneas Aeniam ad statum sacrificium, quod Aeneae conditori
landed was inhabited by a Thracian tribe allied to the cum magna caerimonia quotannis faciunt."
Trojans, KpooaloP KaX&oUfeVO. The territory was there-
fore Krousis, precisely the region at the northern end 21 Note that Vergil Aen. iii. 16-18 (apparently);
of which was Aineia (cf. Herod. vii. 123. 2). Dionysius Mela ii. 27; and Ammianus Marceellnus xxvii. 4. 13
(i. 48. 1), with the utmost formality, cites Hellanictis' make Aeneas found Ainos, as do the Scholia Vetera to
TpLKa as his auithority for this narrative. In the ex- Lycophron 1236. Conon (FGrH, Vol. I, No. 26, Frag. 1
cursuis runniing from i. 48. 1 througlh 49. 3, Dionysius [XLVII, p. 208) cani say that Ainos became the later
discusses different versiotns of the wanderinigs of Ae- namie of Aiincia.
suggested Aeneas to the Greeks and near-by Aineia, which in time gave its
hence, at an early date, gave rise to the name to the promontory.
legend of the city's foundation.22 The orig- Rhaikelos, then, was created by Pisis-
inal inhabitants of Aineia were probably tratus on the promontory of the same
progressively Hellenizing barbarians, like name during his second exile. Herodotus
their neighbors, the Bottiaeans, farther to (v. 94. 1) states that King Amyntas I
the south and east. offered Anthemus to Hippias after the
Pisistratus' settlement Rhaikelos, rath- latter's expulsion from Athens. Anthemus
er a fortified village or strong point than is the basin of the modern Vasilikotikos
a true polis,23 seems to have had a short River, just northeast of Megalo Kara-
and undistinguished life. It is not men- burnu and to the south of Khortiatis.28As
tioned by any of the historians, geogra- Busolt has seen,29there is a connection be-
phers, or orators and is absent from the tween this offer and Pisistratus' settle-
tribute lists, the Epidaurian list of theoro- ment at Rhaikelos. Indeed, the first con-
dokoi,24 and the great Delphian list of tact between Athens and the Argead
theorodokoi.25Aside from Aristotle, who monarchy is surely to be attributed to
had opportunities for very special knowl- Pisistratus himself.30 It was from Rhai-
edge of the Macedonian area,26and Ly- kelos as a base that Pisistratus set out for
cophron, a most learned professional anti- the lower Strymon and the Pangaeum re-
quary, Rhaikelos appears only in the gion, where he procured the money and
scholia to Lycophron and in Stephanus, the men which made his final return to
who clearly derives from the scholia. Both Athens possible.3' This was the first de-
the scholia and Stephanus call Rhaikelos cisive intervention of the resources of the
a polis, but their evidence on such a point Macedonian area into the history of
is far from decisive.27 Aristotle's state- Athens.
ment permits only a general location of For Anthemus see Hirschfeld, "Anthemus" (2),
28
Rhaikelos in the region of the Thermaic in RE, Vol. I, col. 2369; Geyer, RE, Vol. XIV, col.
655. Anthemus was originally a district or tribe, not a
Gulf; it is Lycophron who makes possible city (see Thuc. ii. 99. 6). Gomme's statement (op. cit.,
the specific localization. The Pisistratid p. 218)-"Thuc. ii. 99. 6 and 100. 4 suggest clearly
that Anthemus was on the northern side of the moun-
foundation was soon surpassed by the tains which separate Chalkidike from Mygdonia"-
22 J. N. Svoronos (L'Hellenisme primitif de la Ma- has no force. Amyntas would hardly have offered
cedoine [Athens and Paris, 19191, p. 145) makes the Hippias a region so far from the coast. Nor did the
interesting suggestion that the district Kpoavrabj or Argead kings possess the land to the north of the
Kpovrs, beginning at Aineia and running south along Khortiatis-Kholomon range until after the defeat of
the coast, may have been associated with Aeneas' Xerxes. The location of Rhaikelos is, in fact, excellent
wife, Creusa. confirmation for the traditional location of Anthemus
23 As against A TL, I, 465, n. 2: "Pisistratus, when in the Vasilikotikos Valley. Cf. Busolt, Griechische
he made a city of Rakelos (Aristotle, 'AO.roX., 15, 2). Geschichte, II (2d ed.; Gotha, 1895), 323, n. 4: "Es ist
.... Aristotle hardly says this. jetzt verstandlich, weshalb der makedonische Konig
24 IG, IV2, 1, 94/95. Amyntas I dem vertriebenen Hippias das benachbarte
25 A. Plassart, "Inscriptions de Delphes, la liste des Anthemus zum Wohnsitze anbot. Der durch den
th6orodoques," BCH, XLV (1921), 1-85. Column III, Synoikismos bei der dortigen Bevolkerung erlangte
11. 51-99, pp. 17-19, is an important source for the Einfluss der Peisistratiden sollte den Interessen Make-
historical geography of the north Aegean. doniens dienstbar gemacht werden."
29 Loc. cit.
26 Kohler, op. cit., p. 345: "Aristoteles hat wahr-
scheinlich uber den Aufenthalt des Peisistratos in 30 So Kohler, op. cit., p. 345; and F. Cornelius, Die
Anthemus [sic!] nach localen Traditionen berichtet, (lie Tyrannis in Athen (Munich, 1929), p. 45.
ihm von seiner Heimath her bekannt gewesen sind." 31 Aristotle Ath. pol. 15. 2: fKeZOeV [i.e., from Rhai-
27 Stephanus is often careless and sometimes nWakes kelosi U 7rapiXOev eh T0US
re fpf Hvyya&ovr6 T6rovs, 60ev xpu.sa-
cities out of tribes and regions. Thus, in the MIace- TLrOJ/APOS KaL OfTpaTrLCTar1AS /sOwff& Aepos, &&cv eis 'EprTp&ap
donian area, he makes poleis out of Bisaltia, Elinlia, iVUKaTW 7rAXLv eTOT T6Te TpW)TOP &vaow6araaOa& OLQ T7V &PXIV
Pieria, and Tymphaea (s.v. Tbpoi4). The scholia (to 7reXez. ..... Herodottus (i. 64. 1) mentions only the
1. 1238) can make a city out of Almopia. SL -y' 11011l.
92 CHARLES EDSON
A[Io]ov,Ataovis,haLoo, In fact, the suggested location raises in-

hato-bviot, hatovovEs, AIoUvtoI superable difficulties. Pieria was directly
(Pp. 222-23, 466-67; see also p. 489) under the rule of the Argead kings; in-
Because the Aisonioi appear in a special deed, it was in a very special sense the
rubric with Methone and AtKatoiroXZvrat
Macedonian homeland.36 Nothing is less
'EpErptov (A TL, I, 450) as paying only likely than that the kings would have per-
the aparkhe from the phoros for the initted Athens continuously to control so
year 429/8,32 the editors plausibly locate strategic a point. The only known city on
Haison somewhere on the shores of the the west coast of the Thermaic Gulf prop-
Thermaic Gulf. A more precise location is er37ever subject to Athens was Methone;
"tentatively" achieved by associating the first extant record of that city's trib-
Haison with the stream Aawv mentioned ute is in the aparkhe rubric of List 26,
by Plutarch (Aemilius Paulus 16. 5), to- 429/8, but the well-known Methone de-
gether with the AEViKOS, as flowing across crees (IG, J2, 57) show that it was subject
the ground on which the battle of Pydna to Athens in 430/29.38 It is absent from
was fought. The arguments in support of the full panels of 443/2 and 435/4.39 This
this thesis are given by the editors in their strongly suggests that Methone had been
discussion of the ancient geography of subject to the Macedonian monarchy and
Pieria.33Plutarch's AZuovthey equate with first became an ally of Athens about the
the modern Pelikas River which flows time of the outbreak of the war between
into the Mavroneri (now officially "Ai- Athens and Perdiccas II.40 Even after the
alliance with Athens, Methone's situation
son") about 7 kilometers from the coast.
was difficult. From IG, J2,57 (A TL, I, 162-
This is not the occasion to discuss the cor-
63, D3-5) it appears that, even in
rectness of this last identification.34 The
430/29, Perdiccas was exerting severe
essential point is that, by associating
Haison with Plutarch's stream Atowvand pressure on Methone, a pressure which
was still continuing in 426/5.4' The only
the latter with the Pelikas, the editors ob- 36 Pieria is the MaKeboviS -yi of Herod. vii. 127. 1.
tain a location for Haison at the modern F. Geyer (RE, Vol. XIV, cols. 649-50, and Makedonien
Katerini,35 thus in Pieria about 5 miles bis zur Thronbesteigung Philipps II [Munich, 1930],
p. 9) does not grasp the signiflcatnce of Herodotus' re-
from the coast at a point which controls nark.
both the main routes running south from 37 For Herakleion and Othoros see below, pp.
96 if. and 99.
Macedonia to Thessaly-that along the 38 Methone is restored with great probability by
coast over Herakleion and through Tempe the editors in the list of 432/1; here the payment is
and the alternative route over the Ayios preserved and is the same as that of 429/8, on which
only the aparkhe was paid. The editors restore the en-
Dhimitrios Pass to Olosson. tire aparkhe rubric in the list of 430/29, though no
32 Also restored for the year 430/29 (A TL, I, 149, traces are preserved.
List 25). 39 A TL, I, 338-39.
33 A TL, I, 489. 40 The fact that Methone, like Pydna, struck no
34 The site of the battle is to be sought farther coins during the fifth century (Gaebler, op. cit., II, 78-
noirtlh near the village Korinos (E. 7, KATERINI, grid 79) supports the view that it was not independent.
0-5509). The editors adopt the topographical conclu- 41 Note that Methone, unlike the site proposed for
sioIIs of J. Kromayer (Antike Schlachtfelder in Grie- Haison, was located on the coast and hence could re-
chenlerd, II [Berlin, 1907], 312-13), who makes the ceive direct support from Athens. In 430/29 Perdiccas
Mavroneri Plutarch's AeuKOs and the Peikas his Atmwv. was making it difficult for Methone to have access to
For a variety of reasons the Mavroneri must be the the sea, was interfering with the city's trade, and was
ancient Baphyras. moving troops across its territory (IG, I2, 57. 18-23).
35 A TL, I, 467. All the antiquities at Katerini are In 426/5 It became necessary for Athens to permit
Roman. None of them are in situ. Most of the In- Methone to import Pontic wheat and to free the city
scriptions found there are known to have been brought from all the obligations to which the other cities of the
from Dion. empire were subject, unless Methone were specifically
other Greek city ot the Pierian coast42was maic Gulf before 432/1 and then only of
Pydna, which never appears in the trib- Methone, which was at least 15 miles
ute lists and is known to have been subject north of the site suggested for Haison and,
to Macedon.43 And there are other con- unlike that site, was on the coast. But
siderations which definitely exclude the Haison's record in the tribute lists is very
proposed location of Haison. From Herak- complete." It paid in 451/0, 450/49,
leion (Platamona) to well north of Cape [448/7], [447/6], [446/5], 444/3, 443/2,
Atheridha the Pierian coast is simply an [442/1], 441/0, 440/39, 436/5, 435/4,
open beach. There are no harbors or eleva- 433/2, 432/1, and, in the aparkhe rubric
tions of any kind. All the points on this with Methone and Dikaia, in [430/29] and
coast are simple aKALXES, open roadsteads. 429/8. That the Macedonian kings for
Under such circumstances, without the over twenty years permitted Athens to
use of harbors or citadels, it would have control a town on the Pierian plain at a
been hardly possible for a naval power like point which could dominate the main
Athens permanently to control an inland land routes from Macedonia to the south
Pierian city. And no ancient source attests is highly improbable.
a city named ALowvin Pieria. To sum up: For the above reasons it seems best to
There is no evidence for any Athenian place Haison somewhere on the eastern
control along the west coast of the Ther- shore of the Thermaic Gulf. Indeed, one
may suggest that Haison may well be
named in the relevant decree (l1. 34-47). These last
provisions show that Athens had to make marked
identical with Akf?a,for variant forms of
coincessions to Methone to keep it in the empire at all. the terminations of ethnics and place
If Athens had such difficulties in the case of a coastal
city to which her fleet gave her ready access, it would
names do occur in the tribute lists. Haisa
hardly have been possible for her to maintain control is listed as one of the five cities of Krousis
of a town in the Pierian plain with no natural defenses
and some distance from the coast.
which appear only in a special rubric of
42 Gomme (op. cit., p. 214) calls Dion in Pieria a
the list of 434/3.45 Haisa is absent from
Greek city. I know of no evidence that Dion was a the full panels of 443/2 and 435/4, as well
Greek colony (see below, p. 97); nor is it correct to
speak of Dion as "the border town of Macedonia on
as from the full rubrics of 429/8. In all
the coast towards Thessaly" (ibid., n. 2). The site of these years Haison appears. Haison is not
Dion at the modern Malathria (E. 7, KATERINI, grid
T4795) is certain. In no strict sense can Dion be
present in the extant portion of the Thra-
called a coastal city. The ancient site at Malathria is cian panel of 434/3, which is largely,
over 3 miles from the nearest point on the coast. Cf.
Strabo (ed. Meineke) vii. Frag. 17: 57L 76 Aiov ) T6XLs
though not entirely, complete. Its pres-
OVIK iv TCo aytLaGXi ToIl OeplAalOOV K6Xi7ov ia-Thv fP Tact 6flwpeiats ToD ence in this panel, therefore, cannot be
OXbTOV, &XX' 5ov JrTT-&TeXef oa-Taovs. As so often in absolutely excluded. But the evidence as
the excerpts from Strabo as given by his epitoma-
tors, the text shows signs of disturbance and cor- it stands today permits the identification
ruption. The numeral is clearly corirupt or defec- of Haisa with Haison. If this be admitted,
tive, for there is no reason to suppose that the
Mavroneri has greatly extended the Pierian coast the localization of Haison is no longer a
through alluvial deposits since antiquity in this par- problem.46
ticular area. Livy's description (xliv. 6. 15: "nam cum
Olympi radices montis paulo plus quam mile passuum 44 The years inclosed in brackets indicate that the
ad mare relinquant spatium, cuius dimidium loci oc- town has been restored by the editors.
cupat ostium late restagnans Baphyri amnis, partem 46 A TL, I, 146, List 21, col. VI, 11.29-33. This ru-
planitiae aut lovis templum aut oppidum [i.e. Dion] bric is restored by the editors in the assessment of
tenet ....") does not actually refer to the distance 425/4: p. 156, A9, col. IV, 11. 83-85.
between Dion and the coast but to the space between 46 Down to 435/4 the payment of Haison, where
the lowest reaches of Mount Olympus and the sea.
preserved and as restored (there is a small epiphora in
43 Thuc. L. 137. 1: is b6,vav 21v 'AXet~A6pov. It was 436/5), had been 1,500 drachmas. The payment of the
still Macedonian in 432 (ibid. 61. 2-3). Pydna did not five towns of Krousis (the Tindaioi, Kithas, Smilla,
strike coins in the flfth century (Gaebler, op. cit., II, Gigonos, and Halsa) for 434/3 is 3,000 drachmas, pre-
105-6). cisely double the former payment of Haison. This sum
94 CHARLES EDSON
BEp-ya`oL tion of the site is as yet unknown,49 but

(Pp. 244-45, 474) there is now epigraphic evidence which
confirms Strabo. J. Roger has published
Berga is an admirable example of the
an inscription found at Philippi which
kind of historical information we owe to
commemorates the celebration of a sacri-
the tribute lists. Strabo (vii, Frag. 36)
fice in honor of Septimius Severus.50The
states that Berga was a K4-q located ap-
monument was erected by (11. 17-19) o'
proximately 200 stades to the north of
HTITETa7r\Lval [ ... . ]ot,'AapLavolroXEZTc|
|
Amphipolis among the Bisaltae (Ev 6E TOlS
[B]E[p]-yaoL, 2Ktjl/E'pTlOl, ratW'plol (Roger's
BtaaTaLs), thus to the west of the Stry-
text). My squeezes of this inscription
mon, over 20 miles inland.47[Scymnus] (11.
show in line 19 BEp-yaoL: the reading is cer-
653-54) agrees with Strabo in placing
tain. Thus in the early third century of
Berga KarTa /EOoyELOV.48 The precise loca-
our era Berga was associated with four
other communities in some kind of league,
also appears in the assessmenit of 425/4, where the probably religious.5' The Hadrianopolitai
editors persuasively restore the five names. In 434/3,
though the flve towns are listed separately, they pay and Skimbertioi seem to be otherwise un-
jointly. This probably indicates that we are not here known. Gazoros is mentioned by Ptolemy
confronted with true apotaxis (cf. A TL, I, T. 19 on
p. 573 [Antiphon, Frag. 55, ed. Thalleim] and U. as a town of the Edonians and Odomanti,
Kahrstedt, AJP, LVII [1936], 428 f.). I suggest that but his notorious inaccuracy makes this at
Haison was the name by which the flve communities,
none of which appears independently in the tribute best only a general indication.59 At the
lists, had hitherto been collectively desiginated. This beginning of line 18, Roger tentatively
separate listing and doubled assessment of the five
towns of Krousis lasted oinly for the year 434/3. Hai- suggests Zlpp&oL,the modern Serres: "la
son reappears in 433/2 and 432/1, paying a reduced pierre autoriserait la lecture, mais en sans
assessmenit of only 1,000 drachmas as against 1,500
before 434/3; in [430/29] and 429/8 Haison pays only exclure d'autres non plus absolument."53
the aparkhe on the 1,000 drachmas. Atheins, in view of My squeezes confirm Roger's suggestion:
the menacing tendencies in the Chalcidic area, aban-
doned the doubled assessment for the towns of Krousis I read [2]lppa`ol.54 It is to be presumed
and reduced their former collective tribute by one-
third; possibly in 430/29 and certainly in 429/8 the meters west-northwest of Nigrita (1:100,000, GREECE,
towns were granted the further and consider able privi- D. 9, NIGRITA, grid P-2876). Three inscriptions pre-
lege of paying onily the aparkhe. But in 425/4, confi- served at Kopach have recently been published by
dent of her position because of the victory at Pylos J. Roger (Rev. arch., XXIV [6th ser.; July-December,
(Wade-Gery and MVeritt, .4JP, LVII [1936], 394), 1945], 37-43); they contain no place names.
Athens could again assess them 3,000 drachmas, twice
49 Cassoni's attempt at location (BSA, XXIII
their original tribute. I am not sufficiently familiar
with the evidence of the tribute lists as a whole to be [1918-191, 33-35) is very attractive.
able to give any reason for the separate listing of the 50 BCHI, LXII (1938), 37-41, and P1. XIV.
towns of Krotisis in 434/3 aiid in the assessment of 1 Ibid., p. 41.
425/4. But the fact that these towns paid jointly and 52 iii. 12. 28. Gazoros is the new official name of the
were jointly assessed hardly suggests that the separate village Porna, which is about 5 kilometers due west of
listing of them indicates the dissolution, brought Nea Zikhna on the north side of the Angitis (Angitsa)
about by Athens, of a hitherto existing political union; basin (G. 10, DRAMA, grid P-6181). I do not kinow
collective payment of the tribute demanded common whether any archeological or epigraphic evidence has
action and mutuial arrangements. been discovered locally to support the new designa-
47 For the Bisaltae see (with caution) Oberliummer tiori. See Roger, op. cit., p. 39, n. 3.
in RE, Vol. III, cols. 499-500. 53 Op. cit., p. 41.
48Stephanus' location, if it may be dignifled as 54 The only other epigraphic possibility is [.]&oy7ya1o*.
such-7rAXLs Op4K7s 7rp6S Tjj XEppOVPW -is too vague to li-y-os in Chalcidice is geographically out of t.e
be taken seriouisly and should not have been used by question, and its correct ethnic is iy-YwOs.Stephanus
Svoronos (op. cit., p. 103) as evidence that Berga was gives this form but adds at the end of his article: Kal
on the west coast of the Strymonic Gulf! Ptolemy's 24y-yalos. This last form probably derives from Thucyd-
placing of Berga in Odomantike or Edonis (iii. 12. 28) ides (see A. B. West, AJP, LVIII [1937], 172). But
need not be taken literally. Berga is obviously the Stephanus' dishonesty as regards ethnics is notorious;
Bip7ri of Hierocles (Synecdemnts [ed. Burckhardt] cf. L. Robert, Etudes anatoliennes ("Etudes orien-
640. 6, p. 5). Bip-yq is now the official name of the vil- tales," Vol. V [Paris, 19371), p. 556: "On sait, ou l'on
lage Kopach (KouraqTi), which is sittuated about 9 kilo- devrait savoir, quEtienne de Byzance a ete, plus
NOTES ON TIE THRACIAN "PHOROS" 95
that the towns forming the pentapolis The inclusion of Berga in the Athenian
were in some proximity to each other. empire was an extension of Athens' power
Since the location of Sirrai is fixed with deep into territory recently added to the
certainty at Serres, Berga is also to be Argead kingdom by Alexander I. Were it
placed in the basin of the lower Strymon. not for the incontrovertible evidence of
Strabo's evidence is thus confirmed.55 the tribute lists, one would hardly have
The placing of Berga over 20 miles in- thought any such Athenian intervention
land on the west bank of the Strymon possible. But the lists are explicit. Berga
naturally raises the question of how paid tribute in 452/1, 451/0, and 447/6.
Athens could bring into her empire, and It is absent from the full panel of 443/2
continuously control, a point so far from but reappears again in 435/4, 433/2, and
the coast. Bisaltia was one of the regions 432/1. This reappearance is surely to be
acquired by the Argead monarchy after connected with the founding of Amphipo-
the collapse of Persian rule in the north lis by Athens in 437. The absence of Berga
Aegean, and Macedonian control was from the full panel of 430/29 is probably
maintained throughout the fifth century.56 to be motivated by the hostilities between
Athens and Perdiccas II. The editors re-
qu'un rassembleur d'eIthniques, un grand inventeur store the name of the town for the year
d'ethniques; un nom de lieu lui etant fourni par un
auteur, il forge aussitot un ethnique qui lui parait 429/8,57 but this is uncertain; [2ep/]alaoL is
vraisemblable.' equally possible. Thereafter no record is
S5 The register of the theorodokoi of the Asklepieion
at Epidaurus (IG, JV2, 1, 94. lb. 1. 19), to be dated preserved, but it is clear that Athens could
very close to 360 B.C., appears to list Berga between hardly have kept Berga under her control
Amphipolis and Tragila; but, as West has shown (op.
cit., p. 161), the cities are listed in the order in which after the liberation of Amphipolis by
they were visited by the Epidaurian theoroi. Brasidas in 424/3 and the resulting hos-
Berga's only famous son was the notorious Antiph-
anes, the Munchhausen of Hellenic antiquity.
tility of that city. Indeed, there are silver
Schmid (RE, Vol. I, cols. 2521-22) dates him "spates- and bronze coins of the turn from the fifth
tens im 3. Jhdt. v. Chr." But a fragment of Alexis-
fep-YaZos MXos-(see LSJ under fepyacw) places An-
to the fourth centuries which imitate those
tiphanes definitely in the fourth century. The theoro- of Thasos and bear the legends BEPF,
dokos in Berga is 'APrTLOiws and, as Hiller suggests in
his commentary, is very probably identical with the BEPFAI, and BEPFAIOY.58Given the
author. Antiphanes together with such flgures as dominant role of Thasos along this section
Asclepiades of Tragilos (FGrH, Vol. I, No. 12, pp. 166
and 484) and Zoilus of Amphipolis (ibid., Vol. IIC, of the northern Aegean coast at this time,
No. 71, pp. 103-4), is a part of the little flur- the adoption of Thasian coin types by
ry of literary activity which occurred in the Strymon
area about the middle of the fourth century. Wila- Berga is understandable.
mowitz (Hermes, XL 11905], 149-50) has shown the The fact of continued Athenian control
likelihood that Antiphanes had heard Plato. I have
not seen 0. Weinreich, "Antiphanes und Munch- of Berga during a large part of the fifth
hausen, das antike Ltigenmarlein von den gefrorenen century is certain. How was the control of
Worten und sein Fortleben in Abendland," Wiener
Siizungsber., 220, IV (1942), who also identifles the
a point so far in the interior maintained?
theorodokos of Berga with the author.
57 A TL, I, 150, List 26, col. II, 1. 32. The payment
56 Thucydides' evidence is formal (ii. 99. 6): fKpaT?q- is not preserved.
a5a Si Kall rc &X)v)vU OvGP3 ol MaKe6vMes OVTOm, & Kai vGv
TL lxovoL, rT6 Te AAve,.toD3TaKaL 1p7paTTWVaV Kal BBwaXTtav. Svoronos, op. cit., pp. 99-100; E. Babelon, Traite
58
Early coins of Alexander I are direct copies of late des monnaies grecques et romaines, Part II (Paris,
archaic Bisaltic pieces and thus indicate that the 1928), Vol. IV, cols. 794-98. It is surely unnecessary
king gained control of Bisaltia soon after the Persian to hold with Babelon that the form B EP FAI0 Ymust
defeat (see Svoronos, op. cit., p. 102; also Gaebler, op. be interpreted as an otherwise unattested personal
cit., Vol. II, Pls. XII and XXVIII [cf. Nos. 1 and 2 on name BepyaZos, which he takes to be that of a dynast.
the latter plate with the Bisaltic coins on P1. XII]). For Bip-y,7, Stephanus gives the variant Bep-y&ov,which
The silver mine near Lake Prasias which Alexander may possibly be a defective form of Bep-yalov. It is not
acquired after Xerxes' defeat (Herod. v. 17. 2) was in, apparent why H. Gaebler has excluded these pieces
or very near, Bisaltia. from his treatise on Macedolian coinis.
96 CHARLES EDSON
The tribute lists make it plain that Berga frontier of the monarchy,"3and a degree of
was dependent on Athens long before the political influence or control by Athens
creation of a strong, permanent Athenian could be permitted in this region which
base at Amphipolis, though, of course, might be thought dangerous farther west.
Eion and Argilos59 could be used, less That Berga later struck coins may indi-
effectively, for that purpose. The disaster cate a measure of laissez faire on the part
at Drabescus had shown once and for all of the kings, though it is perhaps more
the very real dangers of military opera- probable that the coinage dates from the
tions in the Strynion region. From Plu- period of Macedonian weakness after the
tarch (Pericles 11. 5) we learn that Peri- death of Archelaus.
cles sent one thousand cleruchs to "settle
'EpaKXELov
with" the Bisaltae. A cleruchy of one thou-
sand men could hardly install itself perma- (Pp. 278-79, 489)
nently among the Bisaltae by force of A Herakleion appears in the assessment
arms.60The date of this cleruchy cannot of 421 and is restored by the editors in
be determined with accuracy, and hence it that of 425/4 (here the payment only is
would be very hazardous to identify it preserved but is the same amount as that
with Berga, quite apart from the fact of 421). Herakleion is not listed elsewhere
that cleruchies did not normally pay in the extant portions of the tribute lists;
tribute. The cleruchy in Bisaltia must it is absent from the full panels of 443/2,
have been established with something like 435/4, 432/1, and 430/29. The editors
the co-operation, or passive acquiescence, identify the town with the Pierian Herak-
of the inhabitants and their suzerain, leion, which is certainly to be placed on
the Macedonian king. the height overlooking the sea just to the
Given the geographical factors in- north of the Platamona railroad station,
volved, it is most improbable that the the site of a magnificent Venetian cita-
Athenian control of Berga was accom- del.64No other Herakleion is known in the
plished by out-and-out military conquest Thracian district, and, though an argu-
and was maintained by force. Instead, ment from silence is usually hazardous,
diplomatic pressure, backed by the threat the editors' identification is to be accept-
of naval action, such as the blockade6'and ed, since it is normal practice in the trib-
raids62on the coasts of the Macedonian ute lists for homonymous cities to be dis-
homeland around the Thermaic Gulf, was tinguished by descriptive phrases.
responsible for this concession. Bisaltia, 63 From Thucydides we learn that the Odomaniti
after all, was on the then extreme eastern (v. 6. 2) and the Edonians (iv. 107. 3) were independ-
ent under their own kings (cf. [Scylax] 66: JeTa' 5 MaKf-
59 A TL, I, 232-33 and 469.
bovlav ZTpVJC.V VOTaO6S- OUTOS bplte* MaKe6Oviav Kal epqKvV).
60 Livy (i.e., Polybius) xlv. 30. 3, shows that even It was Philip II who flrst extended Macedonia east of
as late as 167 B.c. the Bisaltae still possessed a con- the Strymon.
siderable military reputation: "pars prima Bisaltas 64 E 7, KATERINI, grid T-543754. The location can
habet, fortissitnos viros."
be determined with certainty from Livy xliv. 8. 8-9. 2;
The colony to Brea (IG, 12, 45) has often been con- note particularly that Popilius made his attack "terra
nected with this cleruchy (see Gomme, op. cit., pp. marique .... et classis adpulsa ab litore stabat." For
373-75, for discussion and references). Jacoby the ancient remains see L. Heuzey, Le Mont Olympe
(FGrH, IID, 379-80), in his commentary on Theo- et l'Acarnanie (Paris, 1860), pp. 91-93. For a history
pompus Frag. 145, goes much too far in unequivocally of the site (incomplete as to antiquity) with a descrip-
placing Brea in Bisaltia. It is simply not known if the tion of the medieval fortiflcations and photographs,
colony at Brea had anything at all to do with the see 'Air. BaKaX67rouXos, "To K&f4TPO TOP IlXaTav46pa," MAKE-
cleruchy. AONIKA (Z&YypJpa,u,a HepLO8&K&V TfS 'ETacpetas MaKeOPLK&aP
61 E.g., Thuc. v. 83. 4. I (1940), 58-76; see also Oberhum-
Zwrovbev),eoawaovLKxn,
82 E.g., ibid. vi. 7. 3. mer's article in RE, Vol. VIII, cols. 499-500.
Herakleion is first mentioned in the ex- fifth century." On the contrary, it seems
tant literature, about 360 B.C., by [Scy- definite that the town was Macedonian-
lax] 66 as the "first city of Macedonia." indeed, that it was a Macedonian founda-
But there is other evidence which shows tion.69The first two cities of Macedonia as
that Herakleion existed in the fifth cen- one entered the kingdom from Thessaly
tury. It is mentioned in his letter to Philip via Tempe were Herakleion and Dion.70
II by Speusippus,65 who cites as his au- These are propagandist names: Heraklei-
thorities Herodotus and Damastes. As on is named after the heroic ancestor of
Herakleion does not appear in Herodotus, the Argead dynasty and Dion after the
Damastes remains as the source.66Damas- divine father of Makedon, the eponymous
tes mentioned Herakleion in connection forefather of the MaKE6Pves.7" These names
with Xerxes' invasion, and we must there- fit in perfectly with what is known of Ar-
fore conclude that Herakleion either exist- gead propaganda in the fifth century, a
ed at the time of the Persian Wars or at propaganda which culminated in Euripi-
least by the time Damastes completed his des' Archelaus, and strongly suggest that
history, at the latest well before the end Herakleion was a foundation of the dy-
of the fifth century. nasty. If so, there can be little question
The location is striking. The fortress that Herakleion was Macedonian in the
rests on a steep bluff, which falls away fifth century.
abruptly to the sea. It completely domi- When Brasidas in 424 was introduced
nates the important route from the Vale into Macedonia by the Perrhaebians
of Tempe into Pieria along the coast.67 (Thuc. iv. 78. 6), the first Macedonian
Both to the north and to the south of the city he reached was Dion. The fact that
citadel are open beaches on which triremes Dion is mentioned as Brasidas' point of
could land. An Athenian seizure of Herak- arrival in Macedonia very probably indi-
leion by a coup de main would have been cates that the route taken from Thessaly
perfectly feasible. The possession of He- was that from Olosson over the Ayios
rakleion would enable the Athenians to Dhimitrios Pass.72Now this route is very
control the more important of the two much more difficult than the alternate
chief routes from Macedonia south into and more important one through Tempe
Thessaly.68 69 Speusippus (op. cit., p. 8), in addressing Philip
Gomme (op. cit., p. 214) holds that II, can say that the Greeks encountered Alexander I
la-1 T(I iUJTepOv 'Hp&xXe&ov.Surely the implication here
Herakleion was "well to the south of the is that Herakleion at that time was Macedonian.
southern borders of Macedonia in the 70 So [Scylax] 66: A&w8e lPvetoD worajsoD Max&e6vfs el(O
WvoS, KaL K6Xros eEpjalot. 2pcjTi T&I; MaKe5optao 'HpluXeLov-
65 E. Bickermann and J. Sykutris, "Speusipps Brief ATop.The Delphian theorodokoi list (BCH, XLV [1921],
an Kinig Philipp," Berichte fiber die Verhandlungen der 17,11. 51-54) places Leibethroi between Heraklelon and
Sdchsischen Akademie der Wissenschaften, Phil.-hist. Dion, but this is because it was visited by the Delphi-
Kl., LXXX, No. 3 (1928), 8. an theoroi. Leibethroi seems not to have been on the
coastal route proper but on the east slopes of Olym-
66 FGrH, Vol. I, No. 5, Frag. 4, p. 154. pus. Oberhummer (RE, Vol. XII, col. 1858) has no
67 The Salonica-Larissa railroad is forced to tunnel real authority for placing it "am Fusse des Olympos"
through the elevation on which the citadel stands. (Livy xliv. 5. 12 does not prove that Leibethroi itself
The strategic importance of Herakleion in Macedoni- was on the plain); Heuzey (op. cit., pp. 93 ff.) puts
an hands is well demonstrated by the difficulty which Leibethroi near the modern Leptokaria (E. 7, KATERI-
the Romans had in breaking into Pieria during the NI, grid T-4780).
Third Macedonian War. 71 Hesiod Frag. 3, in "Loeb Classical Library":

68 o)[Thyia] 5'iwoKwa0pn&t1 Aye 7EParo1TepTLKepabpWp / vte 5bw M&y-
The more western route over the Haliacmon and
'pra MaKi56&ca 9'&ir&oX&p1qp'.
through the "Volustana" Pass, today the "Stena
Portas," to Olosson can hardly have been of much 72 So Gomme, op. cit., p. 215, but hardly the
importance to the Argead kings in the fifth century. "usual" route.
98 CHARLES EDSON
and along the coast over Herakleion. But tainly not relinquished Macedonian sover-
if, in 424, Herakleion was in Athenian eignty over the region,75and Thucydides
hands, as now appears from the editors' (ii. 99. 4) also speaks of the Argeadae as
very probable restoration in the assess- having conquered the land "beyond the
ment of that year,73then the choice of the Axius up to the Strymon which is called
more difficult inland route becomes un- Mygdonia." There can be little doubt that
derstandable. Herakleion in PieriA was Bormiskos was Macedonian. Certainly by
controlled by Athens in 421 and, almost 421 and very probably by 425/4 Athens
certainly, in 425/4. Athens acquired the had succeeded in detaching from Perdic-
town sometime after 430/29 in the course cas the cities Herakleion and Bormiskos,76
of her hostilities with Macedon. We owe strategic points on the main routes from
our knowledge of this fact to the tribute Macedonia to the south and to the east.77
lists.
It is interesting and significant to note 7 Note particularly Thuc. L. 58. 2: T7)S &v7UoD y Trjv
Muvyovlas .
that the record of BopMlfrKOS(A TL, I,
76 I do not believe that, on the basis of Thucydides'
248-49, 476) is identical with that of account (iv. 103. 1-2) of Brasidas' remarkable march
Herakleion. Bormiskos is missing from the from Arnai through the pass at Bormiskos and on to
Amphipolis, one would be justifled in concluding that
full panels of 443/2, 435/4, 432/1, and Bormiskos in the early winter of 424 was not under
430/29. It appears in the assessment of Atheniian control or at least allied to Athens. Brasidas
was desperately anxious to reach Amphipolis as soon
421/0 and is restored with great proba- as possible. He reached Bormiskos in the evening, fed
bility by the editors in that of 425/4 his men, andApushed on. Had Bormiskos been hostile
to him or even if it had contained an Athenian garri-
(here, too, the payment only is preserved). son, in view of his great objective it would inot have
Bormiskos' payment was also identical been worth his while to delay to take the place. Brasi-
das' stop at or near Bormiskos was certainly brief.
with that of Herakleion-1,000 drachmas. It is to be noted that Thucydides does not state that
Like Herakleion, Bormiskos also dominat- Brasidas entered or took the town; he merely says:
"And having come about evening to the deflle and
ed and controlled a very important land Bormiskos and having dined, he went forward during
route, that of the later Via Egnatia. The the night." AIny Athenian force there might have been
in Bormiskos would surely have been small and de-
town was located on the narrow pass signed only to lhold the place against any ordinary at-
leading from Lake Bolbe to the west coast tack. It would certainly not have been large enough
to dispute Brasidas' passage through the deflle. And
of the Strymonic Gulf, the Bogazi Rendi- the whole aim of Brasidas' operation was swiftness
nas, "Throat of Rendina."74 The location and secrecy. For these reasons I do not believe that
this passage of Thucydides throws any light at all on
is given with the utmost clarity by Thucyd- the question of whether Bormiskos was or was not
ides (iv. 103. 1). By controlling Bormis- Athenian in 424. Here the evidence of the tribute lists
becomes very relevant.
kos, the Athenians dominated the chief
7 The record of Tp&-Xos (or Tp&ayXos) is also identi-
route leading from Macedonia to the west cal with that of Herakleion and Bormiskos, save that
coast of the Strymonic Gulf and on its payment was one talent (A TL, I, 426-27, 556).
toward Amphipolis and the regions of the The site has been tentatively identifled with the mod-
ern village Aidhonokhorion, about 6 miles west-north-
lower Strymon. Though it is nowhere spe- west of Ampliipolis across the Strymon (D. 10,
grid P-5562). Its coinage (Gaebler, op.
cifically stated, it is very probable that RODHOLIVOS,
cit., IL, 131-32 and P1. XXIV) shows that Trailos had
Bormiskos had been Macedonian. Perdic- not been subject to Macedon before coming under
Atheinian control by 425/4.
cas had permitted the Chalcidians to cul- Argilos (A TL, I, 232-33, 469), located on the coast
tivate land in Mygdonia around Lake at Paliokastro about 2 miles west of the Strymon
mouth (D. 10, RODHOLIVOS, grid P-6354), appears to
Bolbe as long as the war with Athens have been more or less continuously under Athenian
lasted (Thuc. i. 58. 2), but he had cer- control. Macedonian territory, therefore, did not ex-
tend along the coast actually up to the mouth of the
73 ATL, I, 156, A9, col. IV, 1. 108. Strymon. The absence of Argilos from the full panels
74 D. 9, NIGRITA, grid P-4642-4943. of 435/4, the year after the founding of Amphipolis,
OOOPLOL, "OOOpOS Aloritae).81 It is surely best to regard the

(Pp. 358-59, 528, also 489) "Oloros" of the manuscripts as a simple
corruption of Aloros, and hence there can
The editors very tentatively suggest
be no connection with the Othoros of
that the Othorioi may be identical with
the tribute lists.
the "Oloros" mentioned by Pliny (NH
But even if Pliny's "Oloros" is retained
iv. 34) as, seemingly, on the west coast of
as a separate (though otherwise quite un-
the Thermaic Gulf. "Pliny's Oloros, or
attested) city and its identity with the
possibly Odoros, between Pydna and the
Othorioi of the tribute lists is assumed,
Haliacmon, ought not to be Aloros, which
there is no reason to take Pliny's apparent
he places elsewhere in the next sentence"
location of it at all seriously. This whole
(A TL, I, 489). This suggestion is based on
passage is filled with wrong, or at best
the assumption that Oloros (i.e., OAQPO)
hopelessly vague, locations: the Tymphaei
may be a corruption of OAQPO*, which,
placed between Heraclea Sintica and
in turn, is taken as a variant for "O6wpos.
Torone (35); Pyloros (surely the Sithoni-
But here, too, we are confronted with
an Piloros of the tribute lists) on the
Pliny's "known habits of contamina-
lower Axius (36); the Bottiaei near the
tion,"78 for example: Eordaea (34), Eor-
Odrysian Thracians (40); Olynthos near
denses (35); Mygdones (35), regio Myg-
Mount Pangaeum (42). Pliny simply can-
doniae (38); Arethusii (35), Arethusa (38);
not be used as a source for the location of
Apollonia (37), Apollonia (38) ;79 Heraclea Oloros.
Sintica (35), Heraclea (38).80 Following Othoros' record of payment is also
his usual practice Pliny has merely copied against the proposed location. It is known
out in a disorganized form the contents of to have paid in 443/2, 442/1, 441/0,
his (or his slaves') notebooks. Pliny is thus 436/5, 435/4, and 434/3 and appears in
quite capable of listing Aloros once and the assessment of 421. Othoros was thus
then naming it again in the next sentence subject to Athens for at least ten years
(particularly since in his notes the second before the acquisition of Methone, the
reference to the city was in the ethnic form only city on the west coast of the
Thermaic Gulf proper known to have been
and 432/1 very possibly reflects hostilities between
under Athenian control.82 For all the
Athens and Perdiccas. Before ca. 425/4 Athens seems above reasons it is best to consider the lo-
to have controlled no point on the west coast of the
Strymonic Gulf proper between Argilos and Acanthus
cation of Othoros as unknown. The ab-
except Stageira (ATL, I, 412-13, 550; see also Ober- sence of the town from the full rubrics of
hummer, RE, Zweite Reihe, Vol. III, col. 2125; the
city itself was about 5 miles from the coast). In the
429/8 (ATL, I, 359)-no record is pre-
assessment of 421/0 a IIoaibetov appears (A TL, I, 384- served from 433/2 to 430/29-very pos-
85, 541), whose payment, like that of Herakleion
and Bormiskos, is 1,000 drachmas. This the editors,
sibly indicates that it participated in the
citing Pliny (NH iv. 38) and Herodotus (vii. 115. 2), revolt of the Chalcidic cities, and this may
with great probability place on the peninsula which
eilds in Cape Elevthera (D. 10, RODHOLIVOS, grid
be an indication of the general area within
P-699266) and separates the Gulf of Ieriss6s from the which Othoros is to be sought.
Strymonic Gulf proper. But Posideion was surely
never Macedonian until the reign of Philip II. 81 The nominative plural masculinie form of the
ethnic strongly suggests that Pliny's source is one of
78 A TL, I, 546. the early imperial official surveys. For Pliny's meth-
ods see A. H. M. Jones, The Cities of the Eastern Roman
79 In both instances it is apparent from the con-
Provinces (Oxford, 1937), Appen. I, "Pliny," pp.
text that the Mygdonian Apollonia is intended. 491-96, and, particularly as regards the Macedonian
80The same Heraclea is named again under area, A. B. West, Class. Phil., XVIII (1923), 59ff.
Thrace (iv. 42)! 82 For HerakIeion see above. pl). 96-98.
100 CHARLESEDSON
2;,Ep,ua?ot, self in the tribute lists. Serme and Therme

2;Epi-Es, [XEp,uat]Es, 2;ipiuE
are quite distinct from each other.87 The
(Pp. 398-99, 546)
stem of Serme is probably that appearing
Serme appears in the lists for the years
also in Sermylia.88 The location of the
450/49, 448/7, 447/6, [446/5], 445/4,
town is unknown. Its absence from the full
444/3, [443/2], [442/1], 440/39, 439/8,
panel of 430/29 may be due to the fact
435/4, 434/3, 433/2, and 432/1. It is ab-
that it had by that year (the list for 431/0
sent from the full panel of 430/29 but ap-
is entirely missing) joined the revolting
pears in the assessment of 421 and can be
Chalcidic cities; this is a possible indica-
restored in the list for 429/8.83 Serme,
tion for its location.
then, was in Athenian hands for a consid-
Though the identification of Serme
erable period of time. The editors identify
the Serme of the tribute lists with Therme, with Therme must be abandoned, the
problem of the latter's location is of some
and from this identification there follows
importance. The editors place Therme at
the conclusion that Athens continuously
Sedhes (now officially "Thermi"), about 7
possessed a town on the northern or north-
miles southeast of Salonica,89 "whose
eastern coast of the Thermaic Gulf prop-
warm springs no doubt gave the town its
er,84that is to say, maintained an enclave
ancient name." There is a site of the
in Macedonian territory.
Hellenic period near Sedhes,90but the hot
No philological arguments are ad-
springs on which the editors rely are not
vanced to support this identification, nor
at Sedhes but at Loutra Sedhes, about 3
is there any positive evidence to support
miles southeast of Sedhes on the road to
it.85 efp,4,qor Osp,ua is the form of the name
Vasilika, squarely in the plain, the ancient
which invariably occurs in the literary
territory Anthemus.91 The site northeast
sources. If the editors' suggestion is to be
of Sedhes is over 21 miles from the nearest
entertained seriously, we must assume
point on the coast,92 and Loutra Sedhes
that the Attic chancery in the fifth cen-
87
tury employed a form of the town's name ceptedThe equation of Serme and Therme is not ac-
by Gomme (op. cit., p. 214).
quite different and distinct from that used 88 Possibly a Thracian root meaning "water" (RE,
by the authors. But the Thermaic Gulf Zweite Reihe, Vol. VI, col. 410).
89 D. 8, THESSALONIKI, grid 0-9433. Distance is
took its name from Therme,86and in IG, measured from the eastern wall of the old city of
12, 302. 68 there is the phrase: ev rOLOep- Salonica (see below, n. 107).
90 A. J. B. Wace, "The Mlounds of Macedonia,"
,ua'oL KAXlr[L. It is indeed unlikely that
BSA, XX (1913-14), 131, Cl: see also Anntals of
the chancery would employ an adjective Archeology and Anthropology, II (1909), 162. This is
derived from a form of the city's name apparently the site marked "Sermi," just northeast
of Sedhes on the British General Staff Map No. 2097,
which was so markedly different from the sheet "Mt. Athos," seemingly a learned guess.
name of the city used by the chancery it- 91 D. 8, THESSALONIKI, grid 0-9929. See A. Struck,
Makedonische Fahlrten, Vol. I: Chalkidike (Vienna and
83 [Zepp]aLm instead of [Bep]faoL (see above, p. 95
Leipzig, 1907), pp. 2--3. The modern town Sedhes (or
and n. 57).
"Thermi") is Struck's "Tschiftlik Sedes"; Loutra
84 Herod. vii. 123. 3 shows that the Thermaic Sedhes is his "Badeort Sedes mitten in der Ebene"
Gulf, properly so called, was thought to begin just (see the map at the end of the volume).
north of Aineia. This is probably the meaning of 92 Alluvial deposits of the Vasilikotikos River can-
Hecataeus (FGrH, Vol. I, No. 1, Frag. 146, p. 26):
not have extended the coast to any marked degree
Jv 5' abra J. e., x6Xrw] Opptl7 Wr6Ls 'EXXiev Op&tixwy, fv 5e Xa-
since the flfth century. Note the prehistoric toumba
X6)7Tptl WOXLS Op?)IKWY.
called "Gona," "a 500m a l'Ouest de la route entre
85 The &oodo0at of Thuc. ii. 29. 6 deflnitely implies Salonique et le Grand Karabouroun, a 1 kil. au Sud-
that Therme had previously been a Macedonian pos- Sud-Ouest de l'ancienne Ecole d'Agriculture," dis-
session. cussed by L. Rey, "Sites pr6historiques de la Mace-
86 Herod. vii. 121. 1: OepI U rfi
Tn rC OepdaW KOAXi doine," BCH, XL (1916), 271-78 and 285. The settle-
OIKT?71AEI, ijs
iS Kal 6dK6xos OUTOS TrJv hrwvVutar Xet ment was one of pile-dwellers on the edge of a marsh,
is over 6 miles inland. Granting that likos) River.94 Exactly what Herodotus
Therme received its name from hot means by HaWoVIK'
is difficult to determine,95
springs, it is very hazardous to place the but Crestonia was the western slopes of
ancient site solely by means of hot springs the Krusha Balkan (the ancient Dysoron
which exist today, for marked changes Oros) north of lakes Bolbe and Langadha
can, of course, have taken place during (now renamed "Kor6nia").96 The land
the last two thousand years. Indeed, there army under Xerxes' personal command
seems to be evidence for hot springs with- clearly proceeded through the basin of
in ancient Thessalonica itself.93 Lake Bolbe, the course of the later Via
Nor is the following argument, based on Egnatia,97 westward toward the Echei-
Herodotus vii. 121. 1, valid: "Xerxes or- doros River. If the army actually reached
dered the fleet and army to separate from the river before turning south to the sea,
Akanthos to Therme, since this gave the it came to the coast some distance west of
shortest march for the army. Sedes, and Salonica; much more probably, Xerxes
not Salonika, is the point where the army took the Derveni Pass, west of Lake
would first reach the coast" (italics mine). Langadha98 and so reached the coast al-
I can only interpret this as meaning that most precisely at the site of Salonica it-
the editors conceive the Persian army to self. Sedhes is not at all the point "at
have marched, more or less directly, from which the army would first reach the
Acanthus (Ieriss6s) to Sedhes, a most diffi- coast." Herodotus' narrative offers no
cult route along and over the mountains. support whatever to the identification of
This is incorrect. Xerxes (Herod. vii. 124), Sedhes as Therme. It is clear from Herod-
with the Persian land army, left Acanthus otus that Therme was chosen as a base
and moved through the interior (TV Ao0-0- for the Persian fleet99and the point where
'yatav) toward Therme. The route lay Xerxes and the army would reach the
through the Paeonian and Crestonian re- coast of the gulf and make contact with
gions toward the Echeidoros (now Gal- the fleet.100Therme, therefore, was located
directly on the coast, not inland-a con-
either by a lake or by the sea. This is very evidently clusion strongly supported by the fact
the toumba, D. 8, THESSALONIKI, grid 0-916314. Note
also the toumba at Tsair across the Vasilikotikos from that the gulf took its name from the city.
Gona (cf. BSA, XXIII [1918-19], 26-27). These Strabo (vii, Frag. 21) states that
toumbas show that there can have been very little, If
aiiy, extension of the coast since early antiquity. Therme was one of the twenty-six com-
93 Th. L. F. Tafel, De Thessalonica eiusque agro: 94 'EropebeTo 5&
5f TfS& HawOv&Kit KaL Kpi7omTv&KjS drLTOTcarajA&
dissertatio geographica (Berlin, 1839), pp. 12-13: 'Exel&Opov.
"Thermae urbis nomen thermis ipsius debetur. Ejus 96 It is certain, however, that he does not refer to
rei testimonium minime contemnendum extat apud
the region north of Macedonia, which was Paeonia
Anonymum Vaticanum, secull VII scriptorem, in par excellence. Generally speaking, it is apparent that
Actis Sanctorum ad VIII. Oct. sive S. Demetrii
he is referring to the area extending eastward from
(Bruxell. 1780) p. 94, cap. 94 [I substitute the later
the lower Axius to the lower Strymon (cf. Herod. v.
text of-Migne, Patr. Gr., CXVI (Paris, 1891), c. 15,
15-16; Thuc. ii. 99. 4).
col. 1184: as (Leontius) acdriKa KaTd T&s TWV KacL/4VW Ka/A&-
96 Herod. vii. 124 and 127. 2; see also
pas, &lAa Kal' TODT7&'Oepjs63' biUTWV ObKOV KaeX&P KaLLTep&KaO&paf Oberhurnmer,
AerT& Ka L TG fkeTe TWV 5rj0qo1WJ ipf.6XwA)VKaL TpO7rwv, &V?E- RE, Vol. XI, col. 1718; and Geyer, RE, Vol. XIV,
peV 1werfTO ov tKov Tr$ /ALp&TVp&, 5acP&,Xet(LKaTaKoU.Jas XepeLWJ
col. 655.
&aoTv ToD SiJoolov Habuit igitur ther-
XOVTpOD Kalr TOD rTaSOV]. 97 This was the route of the remnant of the Persian
mas quasdam, proprie dictas, ipsa urbs Thessalonica, army in its retreat after Plataea (Aeschylus Persae
a balneo publico, quo nulla civitas carebat, probe dis- 492-95).
tinguendas" (italics mine). See Tafel's further discus- 98 D. 8, THESSALONIKI, grid 0-9253.
sion. Oberhummer (RE, Zweite Reihe, Vol. V, col.
2392) is quite wrong in asserting that Tafel (p. 16) 99 vil. 121. 1: Tr& vaVT&K&' OTpaTr& TO-ojAe&ve
&v &Op#A.
distinguished Therme from Thessalonica. Oberhummer This is sufflcient to rule out Sedhes and Loutra Sedhes.
misses the whole point of Tafel's argument. 100 Ibid. vii. 124; 127. 1.
102 CHARLES EDSON
munities which formed the synoikismos of name of Thessalonica is an addition by the

Thessalonica. In another fragment (vii, epitomator.102 This is possible, but not
Frag. 24) he speaks of Thessalonica as proved. The scholia to Thucydides (i. 61.
7 7pOTIEOP OEp/L? EKaLXeTO. It is to be noted 2), whose value is admittedly not high,
that these two statements are not neces- have X vv3vOEoaXOOPLKI7 iorXis lr6XaL 0Ep,Atn
sarily in conflict; Cassander's new city CKaX6ETo.103 Malalas (vii. V. 80D [ed.
can have been founded on the site of Ther- Bonn, p. 190])has 0EaaaXovLK1v, Tf71'rpc.7tv
me.101The editors appear to suggest that XE'0O/LPEV?WKw'fl7V Oip/Ias [sic!1).104 Though
the assertion that Therme was the former these late sources are probably independ-
101 It is worth pointing out that the cities which ent of one another, it cannot be pretended
contributed to the population of the new foundation that their evidence is decisive. One may
did not, or did not all, cease to exist as poleis. Strabo
(Frag. 21) gives us the names of only six of the twenty- remark, however, that learned Byzantines
six towns participating in the creation of Thessaloni-
ca: Apollonia, Chalastra, Therme, Gareskos, Aineia,
may well have been interested in the ori-
and Kissos. Of these Apollonia (Livy xlv. 28. 8; Acts gins of one of the most important cities
17:1) and Aineia (see above, p. 88, n. 5; note urbs) of the eastern Empire.105 The literary
certainly continued to exist. Chalastra is mentioned
elsewhere by Strabo (vii, Frags. 20 and 23), but here sources, such as they are, associate Thes-
it is probable that Strabo is reproducing a pre-Hellen- salonica with Therme.
istic source. Gareskos is also mentioned by Strabo (vii,
Frag. 36) as apparently east of Lake Doirani, aild by In an important article, K. A. Romaios
Ptolemy (iii. 12. 22) as in 'Op1Xta, which agrees well has recently re-examined the whole ques-
enough with Strabo. Neither Gareskos nor Apollonia
was near Thessalonica. tion of Therme's location.106 The archeo-
Frag. 21 of Strabo is from the Vatican epitome, Frag. logical evidence assembled by Romaios
24 from the " Epitome edita. " As between the two, there
is,I suggest, cogent reason to hold that Frag. 24 repro- gives at least a clear indication as to the
duces the original text of Strabo more accurately than area wherein the site of Therme is to be
does Frag. 21. (I am not here discussing the
relative merits of the two epitomes but only of sought. Excavations carried out in 1930
these particular fragments.) Note that Frag. 21 by the University of Salonica on Mikro
has the late spelling 0eava)ovPUf a. Frag. 21 also
states that Thessalonica "is the metropolis of Karaburnu, the promontory which ex-
present-day Macedonia" (j 6f pn77p6srOXo; 7is vuv MaKe-
bovlat &7-0). In fact, the city did not receive the title 102 So the epitome of Strabo (vii, Frag. 22) has:
of metropolis until the reign of Decius, ca. A.D. 250 KTrpovKaXe7rat.The relative clause here is
MIIbpa,j VGV
(Gaebler, op. cit., 2. Abt., p. 130). But here the word clearly an addition of the Byzantine epitomator. But
surely meatns that Thessalonica was the seat of a the statement is correct (Diodorus xiii. 49. 2): 6 ,av
metropolitan bishopric (see 0. Tafrali, Thessaloniquze oUV 'ApXcaos . . .. oroXLopKjaas Tri IH&vap Kaa Kparjaao oETj4I-
des origines au xive siecle (Paris, 19191, p. 70; Oberhum- aepV a1TiV &r6 OaXiLXrTTO7scS The actual distance
eicoat arT&&a.
mer, RE, 2. Reihe, Vol. VI, col. 148). It is true froin 1Iakriyialos, the original site of Pydna, to Kitros
that Antipater of Thessalonica in an epigram (A P is slightly over 3 rmiles, which agrees remarkably well
ix. 428) celebrating the victory of L. Calpurniiis with Diodorus' round figure of 20 stadia. For Pydna
Piso, proconisul of Macedonia from 13 to 11 B.C., over see Geyer, RE, Vol. XIV, col. 668 and references there
the Bessi calls Thessalonica the "mother of all Mace- cited.
donia" (0fohaX0oViK77 A&jTrspi sra1nst.... MaKe6OV17S7), but 103 Cited by Gomme, op. cit., p. 213.
this expression must be dismissed as poetic hy- 104 The entries in the NVominaurbiiton m utata (Hier-
perbole; the silence of the inscriptions, coins, and lit- ocles Syotecdemus led. Burckhardt], App. I. 55: Ia. 15;
erary sources, save for Frag. 21 of Strabo here under III. 106) have no value.
consideration, is decisive. Note also that Frag. 21
speaks of the cities which contributed to the synoikis-
102 Stephanus under 0eIaaXOPIK-q has XTts &pa eKaXeLrTO
m0os as being destroyed (KaOeXiW), wlhile Frag. 24 men- 'AX!a.Meineke in his commenltary remarks: "Mirum est
tions, and rightly, only a movement of population memorari nomen urbis aliunde prorsus non cognitum,
(;efrKLO'e). It is evident tllat Frag. 21 shows clear reticeri autem longe frequentissimum Oipoa vel O6pA.. in
signs of Byzantine contamination and inaccuracy. R[ehdigeranol scriptum est aX&a i.e. AAAsive MA.
Such signs are not apparent in Frag. 24. These obser- corrigendum igitur ([Op]fAa. nimirum hoc quoque factum
vations do, I believe, suggest that Frag. 24 is deflnitely est quod plurimis locis factum iam vidimus et porro
the more trustworthy. It is therefore by no means un- videbimus, ut in libris Stephani primae nominum syl-
likely that in the latter the phrase e 7rp6TepOV 0ip,Xn
labae exciderint." How long must we wait for a criti-
EKaXeLTOis an accurate reproduction of Strabo's origi- cal text of Stephanus?
nal text. If such is the case, we have by no means con- 106K. A. 'PwcAaioi, "IIou fKELTO j 7raXaca Oipgji," MAKE-
temptible literary evidence for the equation of Thessa- AONIKA, I (1940), 1-7. The sketch map by Makaro-
lonica with Therme. nas facing p. 6 is particularly valuable.
tends into the Thermaic Gulf about 3 many architectural members in one place
miles (5 km.) south of the old city of Sa- indicates that they were not brought from
lonica,107brought to light significant ar- elsewhere for use as building materials. A
cheological finds. The cape was inhabited fragmentary Ionic capital of excellent
from the Iron Age (ca. 1000 B.c.) down to workmanship (fifth century) has also been
at least the fifth and fourth centuries and found near the University of Salonica.
perhaps throughout antiquity. There were These finds show beyond dispute that a
found remains of houses, undatable; city of the archaic and classical periods
graves of the fifth century and perhaps existed at or very near Salonica."'1This
older; sherds of the prehistoric and proto- can hardly have been other than Therme.
geometric periods; Rhodian and Chian Romaios, seeking to harmonize the
(so called "Naucratic") and Attic sherds. seemingly contradictory evidence for the
Among the finds was a remarkable Attic site of ancient Therme, suggests that
skyphos of about 450 and a large krater and Therme was the collective name for a
pelike of the period of the Peloponnesian whole series of villages grouped together
War. These finds were made on the north- in a political union which extended from
ern part of Mikro Karaburnu toward Sa- Salonica down to and including Sedhes.1"2
lonica. During the first World War an ex- But this view is hardly tenable. One may
tended necropolis was found on the north- grant that the territory of the city may
east side of the cape; the most recent have been large and have included a num-
graves were of the fourth century.108At ber of villages, but there was surely a cen-
the so-called "Megali Toumba" in the tral fortified point with the temples and
Kalamaria district just southeast of Sa- public buildings, as was normal for Greek
lonica, inland from Mikro Karaburnu, city-states. To the best of my knowledge,
Pelikides has found graves of the sixth nowhere in the northern Aegean area do
and fifth centuries; among the finds were we have any evidence, so far as Greek
Corinthian vases of the early sixth cen- cities are concerned, for the peculiar kind
tury and Attic black-figure ware.109There of community that Romaios suggests. All
is also important evidence from within the evidence indicates that Therme, very
Salonica itself. Two hundred meters west definitely a polis in the ordinary sense of
of the Administration Building (AwtKo- the word,"13 was located at, or very near,
between the latter and the Sara- Salonica.
1-r7pLOZ),
peion,"10have been found many architec- 1,I It is hardly necessary to point out the difflculties
tural members of the Ionic order, whose associated with excavation in a large modern city like
Salonica. In the present age it is probably naive to ex-
style and workmanship, according to Ro- press the hope that the antiquities and monuments of
maios, recall the Treasury of the Siphni- Salonica may receive something of the care which has
ans. Near by was found a small head from been expended on those of Athens. Salonica was one
of the great cities of the Byzantine world, and its
an archaic relief, dating from about the monuments, if properly displayed and preserved,
turn of the sixth to the fifth centuries. would in their own sphere attract enormous interest.
112 p. 4: alperat KOe bVOKOXIa, ZaY beX5eO6pV,
cit.,
Romaios holds that the presence of so op.
&Tr Kat Ol Tif6rOt 7rept7 T75 fOpetbTepOp IUx6v 7oi KcXiov bvh-yoVro
By "old city' I mean that portion of Salonica
107
eis TIrV aVTi KOLV6Tr7Ta i r6OX1V,TS)p 0p1p/V, av KaL 4WJ&KS
within the Byzantine city walls. Newer quarters of the etvpat va c)VoMi4orro IAA 9z'a j repta&O,repa TOVtK& o6/saTa.....
city extend south along the coast up to the base of P. 6: &XX'
i Gkplwq,orw,s vewripwei8ojuv, &rrTe)XfeTO fK 7roXXc2
.lVkro Karaburnu. 4LIpilO' KWpWV KaI *ETre&verO ets jAeytly&7v &TO1aapArb r6 Xies
108 See also BSA, XXIII (1918-19), 38-39. /AkXp' Tap TO67IcoTit KIcLT6r7v
Le OE5OLXOVLK?JS.
109 See also JHS, XLI (1921), 274. 113Therme is called a polis by Hecataeus (see
110For the Sarapeion see BCH, XLV (1921), 540- above, n. 84), Herodotus and [Scylax] 66. Aeschines
41, and Makaronas, MAKEAONIKA,I (1940), 464-65. (ii. 27) seems to refer to Therme as a xwp1ov, but this
104 CHARLES EDSON
The fact that Therme gave its name to quite distinct from Thessalonica, though
the gulf on the coast of which it was situ- it was surely in the immediate vicinity of
ated probably indicates that it was the the latter. But there is no certainty of
most important city on the Thermaic this; Procopius' Therma can have been
Gulf proper. It is also possible that promi- anywhere in Macedonia."15Indeed, the
nent physical features associated with its fact that Procopius lists Therma among
site may have caused ancient sailors to the places fortified by Justinian suggests,
name the gulf from the city. Cape Mikro rather, that it was not in the environs of
Karaburnu, the only important promon- Thessalonica, for, had it been, its inhabit-
tory along the northern coast of the gulf, ants could have taken refuge within that
can have been the cause of the name. It is city and the emperor would have been
to be noted that the site of the citadel of spared the expense of superfluous con-
Salonica, the Heptapyrgion, is the last struction."6
high point of land on the coast that one This investigation has not succeeded in
would observe from the sea in antiquity determining the precise location of an-
as one sailed from east to west, until cient Therme. It is clear, however, that
passing well beyond the mouth of the there is no longer any cogent reason to
Axius."14 place Therme at Sedhes. Therme is to be
Were it not for the fact that Procopius placed on the coast somewhere in the area
(De aedif.iv. 4. 3 [B 279])gives a O'p,LAain running from Mikro Karaburnu to and
his list of places in Macedonia fortified by including the old city of Salonica. Pro-
Justinian, I should be strongly inclined to copius notwithstanding, I regard the old
locate Therme on the site of Salonica. If city of Salonica as the most probable site.
Procopius' Therma is the Therme of the
tvos
archaic and classical periods, then there
can be no question that Therme was (Pp. 406-7, 548-49)
Sinos appears first in the idiotai rubric
has no weight as against the other evidence. It does,
for 434/3 and is restored by the editors
however, indicate that he thought of It as a speciflc for 433/2 (here the payment is preserved).
place, not as a group of villages extending over a fair-
ly large area.
It is absent from the full rubric of 429/8
but reappears again in 421/0, both in the
114 There is perhaps another Indication. Frags. 20
and 23 of Strabo (Book vii) speak of the Axius as list and in the assessment. In the list of
flowing into the sea between Chalastra and Therme. 421/0, Sinos appears with Tporoat and
Here Strabo is probably reproducing a pre-Hellenistic
source (Hecataeus?). The two passages show -that Ka,aKaL, both of which are known to have
Therme is thought of as the flrst Important city on the been Bottic cities."17 The land grant of
coast to the east of the mouth of the Axius. The arche-
ological evidence proves that an important Greek city Cassander, appropriately cited by the
of the archaic and classical periods did stand on or editors (Ditt., Syll., 332), states that the
very near the site of Salonica, and there is every reason
to conclude that this is the city to which Strabo's 115 There are, for example, hot springs near Langa-
source refers. It is hardly sound method to assume that dha and at the pass of Rendina.
the ancient city attested by the archeological flnds was 116 Tafel (op. cit., pp. 14-17) argues very effectively
not mentioned at all by Strabo (or his source) and that
the geographer silently passed on to a city located near against the identiflcation of Procopius' Therma with
Sedhes. Frag. 24 of Strabo begins: 6T& MeT& T7J' 'At&lv Therme.
iroiapu6v Oeo-raXoKl, wL6S, 7p6067poeP
71TA Opijn &aLxeTo. 117 IG, I2, 90, the treaty between Athens and the
In combination with Frags. 20 and 23, this strongly Bottiaeans. Tripoial and Kemakai together with
supports the conclusion that Thessalonica was founded Kalindola are listed at the end of the treaty. Enough
on the site of Therme. We may recall that Frag. 21 (see uninscribed surface is preserved below the text to
above, p. 102 and n. 101) says that Therme was one of make it, on the whole, improbable that any other
the towns forming the synoikismos. This Is, of course, cities save these three were listed as participating in
quite true, but, as stated, it Is very misleading. the alliance with Athens.
recipient held TOV aypov TOV EV Tnlt 2ltalc Meritt's discovery of a boundary stone
TpaPe
KaU TOV oiv7L
(rl (11. 4-6) and TOV reading BOTTlKOLS, of the turn from the
i2lrapTrc.XWL(11. 15-16). Trapezous seems fifth to the fourth centuries,'12 at Metokhi-
otherwise unknown, but Spartolos was on Zografou, 5 kilometers south of Vromo-
the chief city of Bottike. The land grant sirtis,122 which is a little over 9 miles
is decisive confirmation for the placing of north-northwest of Potidaea, gives at
Sinos in Bottic territory, that is, to the least a general indication of the southeast
northwest of Potidaea. The stele contain- frontier of Bottike.123 The site proposed by
ing Cassander's grant was found "sur the editors for Sinos is over 40 miles as a
l'emplacement precis de l'ancienne Poti- crow flies from Vromosirtis, and it is cer-
d6e, a l'entree de la Pallene.""118 The tain that at no time did Bottic territory
publication of the grant in Cassandreia is ever include the north coast of the Ther-
readily understandable if the estates were maic Gulf. The equation of Sinos with
not far distant. Sinos was a Bottic town. Sindos must be abandoned.124 Sinos is to
But the editors have unnecessarily be placed somewhere in the Bottic region
complicated the placing of Sinos by sug- of Chalcidice to the northwest of Poti-
gesting that "it may perhaps be identified daea.'25
with the IlISo&of Herodotus, VII, 123, 3." UNIVERSITY OF WISCONSIN
Ilerodotus places Sindos between Therme 121 AJA, XXVII (1923), 337 ff.; SEG, II, 408 (p.
and the Axius, thus directly on the north 71).
coast of the Thermaic Gulf. This region 122 E. 9, POLIYIROS, grid P-1202.
123 See A TL, I, 550.
was then Mygdonia (Herod. vii. 123. 3 124 Had Sinos in fact been located where the editors
and 127. 1),119which was separated from place it, the publication of Cassander's grant concern-
Bottike by Krousis and Anthemus.120 ing the estate b Tit 2Zwalcu would have been made
in Thessalonica rather than Cassandreia.
118 Duchesne and Bayet, Memoire sur une mission 125 It has not seemed necessary to discuss the loca-
au Mont Atho8 (Paris, 1876), p. 71. tion of Strepsa (see Gomme, op. cit., pp. 215-18).
119 See also vii. 124, where Herodotus describes the
Gomme's masterly discussion has show-n that the only
Echeidoros River as flowing through Mygdonia and passage in ancient literature which can be used for the
placing of Strepsa is Aeschines ii. 27, and tlhis pas-
reaching the sea at the swamp by the Axius.
sage does no more than put Strepsa somewhere in the
120 For Antliemus see above, p. 91 and n. 28. northwest frontier regions of Chalcidice.

Edson 1947 TopographyofTherme

Uploaded by

Copyright:

Available Formats

Edson 1947 TopographyofTherme

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Edson 1947 TopographyofTherme

Uploaded by

Copyright:

Available Formats

Notes on the Thracian Phoros

Author(s): Charles Edson

You may need to log in to JSTOR to access the linked references.

TI~ HE publication of the Athenian monarchy and on points in the historical

[CLASSICAL PHILOLOGr, XLII, APRIL, 1947]

lands, Vol. III, 2. Abt.: Die antiken Muinzen von Make-

A[Io]ov,Ataovis,haLoo, In fact, the suggested location raises in-

BEp-ya`oL tion of the site is as yet unknown,49 but

Third Macedonian War. 71 Hesiod Frag. 3, in "Loeb Classical Library":

OOOPLOL, "OOOpOS Aloritae).81 It is surely best to regard the

2;,Ep,ua?ot, self in the tribute lists. Serme and Therme

munities which formed the synoikismos of name of Thessalonica is an addition by the

You might also like