Reproduced with permission of the copyright owner.  Further reproduction prohibited without permission.

The Corycian nymphs and the bee maidens of the Homeric Hymn to Hermes
Larson, Jennifer
Greek, Roman and Byzantine Studies; Winter 1995; 36, 4; ProQuest
pg. 341

The Corycian Nrmphs and 
the Bee Maidens 0 the Homeric 

Hymn to Hermes 

Jennifer Larson 

X 
TIlE END OF TIlE Homeric Hymn to Hermes (552-66), 

Apollo bestows upon Hermes certain oracular bee 
maidens as a consolation prize for his refusal to let the 

younger god share in his own mantic privileges. 1 These three 
sisters live "beneath a ridge of Parnassus" and teach a form of 
divination that Apollo praised as a boy: 

o"Ejlvru "(up 'tLVE~ dO"t lCUO"tYVTl'ytat "(E"{uutm 
7tUp9EvOt rolCElnO"W U,,(UAAOjlEVUt 7t'tEpU"(Eo"O"t 
'tPEt~· lCU'tU Of lCpU'to~ 7tE7tUAa"(jlEvm aAqmu AEulCU 

555 OilCta vanUouO"LV U7tO 7t'tUXt I1UPV11O"o'io 
jlUV'tEl11~ U7tUVEu9E OtOUO"lCaAOt ~V E7tt ~ouo"t 
7tat~ E't' E<J)V jlEA£'t11O"U' 7tU't1,P O' EjlO~ OUlC UAE"(t~EV. 
EV'tEU9EV 01, E7tEt'tU 7tO'tcOjlEVUt aAAo'tE (lAAn 
lC1lptU ~60"lCoV'tat lCUt 'tE lCpatVouO"tv flCaO"'ta. 

560 ui. 0' O'tE jlfV 9uioXHV E0110Utal ilEAL XAropov 
7tPO<pPOVE~ E9EAOUO"LV ciA119Et11V U)'OPEUEtV' 
ilv O· ci7tovooqH0"9oXH 9Erov 1'!OEtav Eor08Tjv 
'l'EUOOV'tUl oTj E1tEt'tU Ol' UAATJArov OovEouO"al. 
'tu~ 'tOl E1tEt'tU Otorolll, O"U 0' U'tPElCE~ EPEElVroV 

565 O"1,V UU'tOU CPPEVU 'tEp7tE, lCat Ei ~po'tOV (lVopa OUEl11~ 
7tOAA.O:lCl 0"1l~ OjlCPll~ E1tUlCOUO"E'tUt ut lCE 'tuxnO"t. 

Their identity is an old puzzle that once appeared to be resolved 
by Hermann's conjecture that they were the Thriae, a trio of 
sisters personifying the mantic pebbles used at Delphi.2 Her-

1 Text of the Hymn: T. W. ALLEN, W. R. HALLIDAY, and E. E. SIKES, edd., 
The Homeric Hymns 2 (Oxford 1936: hereafter 'AHS'). 

2 According to Philochorus (FGrHist 328FI95), "nymphs inhabited Parnas­
sus, the nurses of Apollo, three, called thriai, after whom the mantic pebbles 
are called thriai and prophesying, thriasthai." Cf Suda and Hsch. S.'U. eptai; 

341 



Reproduced with permission of the copyright owner.  Further reproduction prohibited without permission.

342 NYMPHS AND BEE MAIDENS 

mann's 1806 identification of the bee maidens as the Thriae 
became widely accerted by editors and critics, though the 
actual emendation 0 O£l1vat or 110tpat (552) to Spta.t did not 
have a parallel success. 3 In 1979 Scheinberg published a detailed 
reexamination of the problem, concluding that the Bee Maidens 
could not, and probably should not be equated with any other 
group of mythological sisters. She pointed out the similarities 
between the Bee Maidens and several other groups of kouro­
trophic sisters-the Charites, various nymphs, the daughters of 
Cadmus, etc. In fact, she argued, these groups are so numerous 
that to single out one particular group as the Bee Maidens' 
equivalent would be ill-advised. o4 Given her arguments, the sim­
ilarities of the Thriae to the Bee Maidens are not so striking, and 
there are further objections: the Thriae have nothing to do with 
either bees or Hermes; and the Hymn mentions nothing about 
mantic pebbles, the main attribute of the Thriae. 

Despite some compelling arguments against the Thriae hy­
pothesis, Scheinberg did not sufficiently examine the possibility 
that the Bee Maidens might be identified with the other group 
of sisters located on Mt Parnassus, the Corycian Nymphs. Al­
though many groups of kourotrophic sisters existed, all having 
similar properties, only the Bee Maidens, the Thriae, and the 
Corcyrian Nymphs are located on Parnassus. 5 Furthermore, 
Scheinberg did not take into account the excavations of the 

Callim. Ap. 45 with l:; Pherecydes FGrHist 3F49. Steph. Byz. S.'li. Spia, Etym. 
Magn. S.'li. Sp'ia, and Zen. 5.75 have the Attic variant legend that Athena dis­
covered or was given the mantic pebbles; see U. von Wilamowitz-Moellen­
dorff, Der Glaube der Hellenen I (Berlin 1931) 379ff. 

3 Among the many works that explain the Bee Maidens as the Thriai are 
AHS 346; S. Eitrem, -Hermes," RE 8.1 (1912) 784. For the early reception of 
Hermann's emendation see M. FEYEL, -l:MHNAI: Etude sur Ie v. 552 de 
I'H ymne Homerique a Hermes," RA 25 (1946: hereafter 'Feyel') 7-11; dissen­
ters from the Thriai hypothesis: P. AMANDRY, La mantique apollinienne (Paris 
1950: 'Amandry') 62; M. Delcourt, ['Oracle de Delphes (Paris 1955) 77; and, 
more recently, F. Casso la, Inni Omend (Milan 1975) 542f, as well as Feyel. 

4 S SCHEINBERG, -The Bee Maidens of the Homeric Hymn to Hermes," 
HSCP 83 (1979: hereafter 'Scheinberg') 14. 

S Location is always an important factor in cult identity, whether that of 
gods, heroic figures, or nymphs. The Corycian Nymphs have been briefly 
suggested as possible candidates by J. Fontenrose, Python: A Study of Delphic 
Myth and its Origins (Berkeley 1959) 42M; P. Amandry, in Antre II (n.6 
infra). To avoid confusion I use an upper-case N when referring to the 
Corycian Nymphs or other specific groups, and lower case when speaking of 
nymphs in general. 



Reproduced with permission of the copyright owner.  Further reproduction prohibited without permission.

JENNIFER LARSON 343 

Corycian cave, which have brought new evidence to the 
debate. 6 

Dionysus 

I take Scheinberg's reason for discounting the Corycian 
nymphs to be their association with Dionysus. She describes 
these nymphs as the mythical equivalents of the Thyiads, or 
Delphic and Attic maenads "in the entourage of Dionysus at 
Delphi" (12). In fact, there is little evidence for this characteri­
zation. Amandry emphatically argues (Antre II 400 n.9) that the 
Corycian nymphs have no link at all to Dionysus before the 
Hellenistic period. The truth appears to be somewhere be­
tween these opposing views. Two pieces of evidence from 
tragedy juxtapose Dionysus and the Corycian nymphs. 7 The 
first is the well-known opening of Aesch. Eumenides (1-26), 
where the Pythia lists the deities of Delphi: the 'first generation' 
deities Ge, Themis, Phoebe; then the latter-day occupants 
Apollo, Athena, the Corycian Nymphs, Bromius, the springs of 
(the river) Pleistus, Poseidon, Zeus. The text (Page [1972]) runs 
as follows: 

OE~CO OE vUllcpw;, Ev9a Kcop\>1c1.S 1tE'tpa 
KOlA", CPlAopvlS, OalllOVlcoV avaa'tPO<Jl1l. 
~polllOS ExEl 'tOY X&pov, ouo' allV111l0v& 

25 E~ ot'tE ~aKxalS EO'tpa'tT,"fTlOEV 9EOS 
Aayro OlK"V TIEv9Et Ka'tappa'l'as Ilopav. 

There seems to be no consensus among editors and translators 
whether the placement of Bromius after the nymphs implies a 
relationship. The issue turns on whether tov X&pov means 
Parnassus and the general site of Delphi or, specifically, the 
Corycian cave. As the Pythia refers repeatedly in the speech to 

6 L'Antre carycien II (=BCH Suppl. 9 [paris 1984: hereafter 'Antre II') esp. 
411; see also the preliminary reports by Amandry and J.-P. Michaud, BCH 95 
(1971) 771-76; Amandry, BCH 96 (1972) 906-09. 

7 Aristonous Paean in Apoll. 3sff (3rd c. B.C.) puts the Corycian Nymphs 
between Poseidon and Bromius; Philodamus Scarphaeus Paean in Dian. 
20-23 speaks of "Delphian korai" accompanying Dionysus in the glens of 
Parnassus; it is unclear whether the reference is to the Nymphs or the 
Thyiads. See Powell, Call. Alex. 162-70. Nonnus (Dian. 9.287) speaks of 
Corycian Bacchants and Macrobius (Sat. 1.18.3ff) mentions "Bacchic caverns 
on Parnassus." 



Reproduced with permission of the copyright owner.  Further reproduction prohibited without permission.

344 NYMPHS AND BEE MAIDENS 

"this place," i.e., Delphi, the former interpretation is less 
forced. 8 The Corycian nymrhs were said (Ap. Rhod. Argon. 
2.711) to be the daughters 0 Pleistus, and Bromius' appearance 
between the Nymphs and Pleistus is suggestive but not 
definitive. 

The other passage, Soph. Ant. 1126£f, cited by Scheinberg but 
not Amandry, lists the many places where Dionysus holds 
sway. Eleusis and Thebes are mentioned in turn, and finally 
Parnassus, where Krop{l1cuU O''tEtxouO't l3a1(xH5E~. 9 Here at last is 
evidence that the Corycian Nymphs were thought to partici­
pate in the biennial Dionysiac celebrations on the mountain. 
This does not show, however, that Dionysus was worshiped in 
the Corycian cave, nor that the Corycian Nymphs were 
merely mythical counterrarts of the Thyiads. The Nymphs are 
year-round residents 0 Parnassus who welcome Dionysus' 
arrival; they are his neighbors. Thyiads were said to honor 
Apollo on Parnassus (Paus. 10.32.7); why should the Corycian 
Nymphs not honor Dionysus? In view of Dionysus' links to 
nymphs in other contexts (e.g. his nurses), it would be 
surprising if the Corycian Nymphs remained completely 
pristine of associations with him. The location of both the 
Thyiads and the Nymphs' cave on the mountain would 
encourage such a rapprochement. Pausanias (10.32.7) casually 
juxtaposes discussions of the Corycian Nymphs and Dionysus' 
Thyiads because they both belong to the mountain slopes, but 
gives no sign that they are related and no hint of Dionysus' 
worship in the cave. In fact, he remarks that the cave is well 

8 'tTtvl)e ya'iav (line 11), XcOpa~ 'tijol)E (line 16); cf 'tol)' ... fUlV'tE'iov (lines 3f). 
A. J. Podlecki, Aeschylus Eumenides (Warminster 1989), comments on lines 
22f that -the Corycian grotto ... was a well-known haunt of the Nymphs, 
who are often associated in art and cult with both Apollo and Dionysus." A. 
Sommerstein, Aeschylus, Eumenides (Cambridge 1989), makes no mention of 
any association between the Nymphs and Dionysus, and in fact comments 
that the order of the gods in the Pythia's speech ·seems almost random." 
Translators are divided on whether E~ ot't£ of line 25 is temporal (Podlecki) 
or spatial (Lattimore). Podlecki (131) translates: -Bromios too inhabits the 
place (as I am fully aware) from the time when the god commanded an army 
of Bacchants and wove a net of death for Pentheus as for a hare." 

9 -Nymphs of Parnassus" undoubtedly refers here to the Corycian 
Nymphs, in view of their rank in the invocation of the priestess in Eumen­
ides. Notice that they are identified by location alone. Similarly we can un­
der stand the ·Parnassian nymphs" of Hypothesis C in I: Pind. Pyth. as the 
Corycian Nymphs. They brought gifts to Apollo after his combat with the 
dragoness and the Corycian Nymphs are said to have encouraged him with 
their cries of ie, ie paian: Ap. Rhod. Argon. 2.711. 



Reproduced with permission of the copyright owner.  Further reproduction prohibited without permission.

JENNIFER LARSON 345 

below the heights at which the celebrations took place, and it 
would be difficult to reach these heights from the site of the 
cave (Paus. loco cit.). Furthermore, the excavation of the cave 
revealed no evidence of a link to Dionysus. None of the cave 
dedications mention the god. Lolling's reading "Thyiads" on an 
inscription from the cave is very uncertain. lO Thus, any 
"rapports a voisinage"ll between Dionysus and the Corycian 
Nymphs are secondary to the Nymphs' association with 
Apollo and other gods, as we shall see. Certainly there is not 
sufficient evidence to characterize the Nymphs as part of the 
Dionysiac entourage at Delphi.12 

Apollo 

The Corycian Nymphs are part of an ancient tradition of 
nymph-worship in spring-bearing caves that has nothing to do 
with Dionysus. 13 They are associated first of all with Apollo's 
youthful deeds; secondly, I argue, with Hermes; and finally 
with Pan.14 At Ap. Rhod. Argon. 2.711££ Orpheus' song to 
Apollo tells how the god slew the monster Delphyne beneath a 
rocky ridge of Parnassus, and how the Corycian Nymphs, 
daughters of Pleistus, encouraged him with their cries of ie, ie 

10 J.-Y. Empereur cites two completely different readings of the inscription, 
the relevant portion of which is extremely worn. Lolling's reading probably 
depended on the literary evidence discussed here. See Antre II 344f no. 8. On 
the Thyiad ritual of Delphic and Attic women at Parnassus see A. Henrichs, 
·Greek Maenadism from Olympias to Messalina," RSCP 82 (1978) 152-55. 

11 See Amandry, in Antre II 400 n.9. Although -Thyiads" can be nymphs, 
as in Alem. fro 63 Page rnaiads, lampads and thyiads"), it does not follow 
that the Corycian Nymphs must be Thyiads. 

12 On nymphs as companions of Dionysus see T. Carpenter, Dionysian 
Imagery in Archaic Art (Oxford 1986) 79-85, who emphasizes the need to 

distinguish as far as possible between 'nymph' and 'maenad'. In a similar vein 
see now G. Hedreen, ·Silens, Nymphs and Maenads," IRS 114 (1994) 47-69. 

13 For a list of nymph cults in caves see Amandry, in Antre II 403-11. 
14 These are the deities most often associated wih nymphs in cult (along 

with the generalized river-god Achelous). That the Corycian Nymphs belong 
to the theogonic stratum of Delphic myth is supported by Ovid's account of 
the flood myth: Deucalion and Pyrrha, landing on Parnassus, worship the 
Corycian Nymphs, the numina mantis, and Themis (Met. 1.313-21). 



Reproduced with permission of the copyright owner.  Further reproduction prohibited without permission.

346 NYMPHS AND BEE MAIDENS 

paian. Similary, Hypothesis C of L Pind. Pyth. says that the 
Parnassian Nymphs brought gifts to Apollo after his victory. 15 

Both the Bee Maidens and the Nymphs had connections with 
Apollo's early life. The Hymn states that the Bee Maidens were 
his teachers (OlOclm((xAol) when he was a youth tending cattle. 
Notice that the association mentioned is in neither case the 
cliche of "nurses" (tPOCPOl) to an infant god (as is the case with 
the Thriae). Instead, both Nymphs and Bee Maidens contribute 
to Apollo's assumption of his adult powers, through slaying the 
dragon and learning the mantic art respectively.16 The Hymn 
makes it clear that the form of divination he learned from the 
Bee Maidens and practiced as a youth was a simple method that 
might be used by nomioi (556£): J.l.cxvn:il1~ a1tclVEU8E OlOcla­
lCCXAol, fly £1tt J3<>uat1tcxt~ £t' £rov J.l.EAEtl1acx. It is not coincidental 
that just as he practiced this art while tending cattle, so he passes 
it to Hermes along with the whip for herdingY The two gifts 
given by Apollo are linked through this rustic association. Both 
Apollo and Hermes were gods associated with livestock; both 
consorted with nymphs, the denizens of the wild, remote 
places where shepherds work, and both were linked to divina­
tion; Hermes had to do with lucky finds and with lesser 
methods of divination such as cleromancy. The Hymn, then, 
asserts the prestige of Apollo, who is represented as ceding his 
humbler interests to Hermes in favor of music and the 
exclusive rights to disseminate the will of Zeus. 18 

15 Wilamowitz (supra n.2: 379ff) correctly identified these Parnassian 
Nymphs with the Bee Maidens of the Homeric Hymn on the basis of their 
shared location, though he did not make the further connection with the 
Corycian Nymphs, but instead accepted Hermann's view that the Thriai are 
the Bee Maidens. He also attempted to bring into the equation the leukai 
korai of the oracle at Cic. Div. 1.37.81 (EJ!ot J!£A.tl<J£t 'tau'ta lCat A.£,\)lCa'i~ 
lCOpatt;); this cannot be accepted with certainty. 

16 It is indeed possible that divination using bees was part of early Delphic 
tradition. The second temple of Apollo was supposed to have been built by 
bees and birds of wax and feathers (Paus. 10.5.5); Pindar (Pyth. 4.61) calls the 
Pythia "Delphic Bee," and so on. See Scheinberg and C. Sourvinou-Inwood, 
"The Myth of the First Temple at Delphi," CQ NS. 29 (1979) 231-51. 

17 Hermes and the nymphs are especially associated in the context of live­
stock and herdsmen: Eumaeus' sacrifice to Hermes and the nymphs at Od. 
14,432ff; Semon. fro 20 West. 

18 Likewise Apollo's remark about this method of divination, 1tariJp O' EJ!ot; 
OUlC aA.E'Yt~£v. does not so much place the Bee Maidens "outside the main­
stream of prophetic technique" (Scheinberg 10), as it emphasizes their humble 
status. On the Delphic point of view of the author see R. JANKO, Homer, 
Hesiod and the Hymns (Cambridge 1982: hereafter 'Janko') 149; AHS 345. 



Reproduced with permission of the copyright owner.  Further reproduction prohibited without permission.

JENNIFER LARSON 347 

In the Hymn the Bee Maidens are said to practice their form 
of divination "apart" from Apollo (cl1t<xv£u8£, 556). As Schein­
berg notes (10), in Homeric diction this adverb denotes 
geographical rather than metaphorical distance: "Whether in 
this case it means 'apart from me' and hence 'not at Delphi,' or 
'apart from the temple at Delphi,' or 'apart from Olympus' 
where Apollo and Hermes are conversing, or even 'apart from 
the haunts of men,' the adverb helps isolate the sisters from the 
mature Delphic Apollo." Why should the word not indicate the 
physical separation between the sites of Delphi and the 
Corycian cave, a distance of about seven miles? There is a 
further physical resemblance between the abodes of the two 
groups of sisters-the cave. The Bee Maidens are specifically 
said to reside U1tO 1t'tUXt IIapvT1oolo ("under a ridge of Parnas­
sus," 555). More will be said of the cave later, but for now I 
merely mention that the cave is the normal cult site for nymphs 
(and is not particularly associated with the many other groups 
of kourotrophic sisters listed by Scheinberg). 

Thus we are looking for maidens with links to both Apollo 
and Hermes, located "under a ridge" of Parnassus, yet not at 
the site of Delphi itself, and in connection with whom a humble 
form of divination is practiced. The Corycian Nymphs fit this 
description perfectly. The excavations of the Corycian cave 
revealed 25,000 astragali, which were almost certainly used for 
divination. 19 The cave seems to have been visited not by the 

19 For divination with astragali in sanctuaries see r Pind. Pyth. 4.337f and 
Paus. 7.25.6, where astragali are used (and large numbers accumulate) in a cave 
sanctuary of Heracles at Bura in Achaea. Amandry, Antre II 376 n.60, collects 
citations of astragali at several sanctuaries, to which should be added Ephesus: 
D. G. Hogarth et al., The Archaic Artemisia: British Museum Excavations at 
Ephesus (London 1908) 190f; and the Leokoreion in Athens: H. A. 
Thompson, • Athens Faces Adversity," Hesperia 50 (1981) 348. Artemidorus 
(2.69) characterizes the use of astragali as an inferior method of divination 
akin to palm-reading. Astragali often appeared on coins, perhaps as symbols 
of local oracles; early examples include coins of Himera (482-472: Head, Hist. 
Num. 144) and Cierium-Arne in Thessaliotis (400-344: A. Moustaka, Kulte 
und My then auf Thessalischen Munzen [Wiirzburg 1983] 120, PI. 3; Head 
293). A local nymph may be shown using the astragali or figures may cast 
them before a cult statue: see W. R. Halliday, Greek Divination: A Study of 
its Methods and Principles (London 1913) 205-16. Amandry (Antre II 409) 
mentions as a possible alternative explanation that the astragali were dedi­
cated by young people before marriage; but the Corycian cave, unlike other 
caves of the nymphs, does not seem to have been used for this purpose; the 
miniature loutrophoroi common at other sites are here missing. Astragali 
were, however, placed in graves, particularly but not limited to those of young 



Reproduced with permission of the copyright owner.  Further reproduction prohibited without permission.

348 NYMPHS AND BEE MAIDENS 

richer clients of Delphi, but by the humble, probably with a 
large admixture of locals. The ceramics in the cave are numer­
ous but quite poor in quality. The evaluation of A. Jacquemin is 
striking when read in the context of our passage in the Hymn to 
Hermes: "on a l'impression d'avoir devant soi les rebuts de 
Corinthe et d' Athenes, encore assez bons pour les paysans et 
les bergers de Phocide et les humbles divinites de l' Antre ... la 
clientele de l'oracle d'Apollon et celIe de l'Antre n'etaient pas 
absolument identiques!" (Antre II 153ff). 

The version of Hermes' birth and exploits given by Apollo­
dorus is important for our argument. It is not derived directly 
from the Homeric Hymn, but contains many similar elements 
pointing to a common source: Hermes' birth in the cave, the 
theft of Apollo's cattle, Apollo's search for the culprit, and the 
exchange of gifts at the end. 20 In Apollodorus' version, Hermes 
gives the lyre in exchange for the cattle, and makes a further 
trade of his shepherd's pipes for Apollo's golden wand that he 
owned when he herded cattle, as well as divination by pebbles 
('tllv Bux 'trov '!rilcprov JluvnKilv). The two versions are strikingly 
similar, but instead of divination through the Bee Maidens, 
Hermes learns a form of cleromancy, "divination by pebbles." 
Below I shall suggest that these two seemingly disparate 
accounts can be reconciled. 

Hermes 

The next task is to demonstrate that Hermes is associated with 
the Corycian Nymphs and their cave. Amandry discusses a 
relief belonging to the second half of the fourth century B.C., 
found at Delphi.21 The lower and larger register of the relief 
shows a scene of sacrifice beside a stylized mountain. The 
upper register shows the deities of Parnassus: three seated 

people; their significance in this case is unknown. See the citations for Myrine, 
Cyme, Samos, Rhodes, and Locri in 1. Robert, "Les epigrammes satiriques de 
Lucilius sur les athletes. Parodie et realities," in A. E. Raubischek, ed., L' Epi­
gramme grecque (=Entretiens Hardt 14 [Vandreuvres 1969] 229ff with notes. 

20 See Feyel (9) and AHS (271) for differences between the two accounts, 
which show that Apollodorus is not derived from the Homeric Hymn. His 
source may have been Soph. lchneutai: AHS 271. 

21 See M. A. Zagdoun, FdD IV.6 (1977) 32-36 no. 8; Amandry, in Antre II 
398ff. 



Reproduced with permission of the copyright owner.  Further reproduction prohibited without permission.

JENNIFER LARSON 349 

nymphs wearing poloi; beside them the figures of Apollo (with 
his lyre) and Hermes, plus another unidentified figure. 22 There 
can be little doubt that the Corycian Nymphs are the three 
maidens in this relief; the only other possiblity is the Muses, but 
they are not as well attested at Delphi for the Classical period as 
the Corycian Nymphs. 23 Also found in the cave was a fragment 
of an Attic nymph relief of the standard type, showing three 
nymphs led in the dance by Hermes. 24 

Before the Hellenistic period, Hermes is more frequently 
associated with nymphs in cult, literature, and art than any other 
Olympian deity, including Artemis.2s From the battle of Mara­
thon (490) through the fourth century, Pan's worship spread 
from Arcadia to other areas including Attica and Delphi. Pan 
quickly came to fill the position vis-a.-vis the nymphs that had 
earlier been Hermes'.26 The Corycian cave became the cave of 
Pan and the Nymphs, though the earliest dedications and 
literary allusions mention only the NymphsY It is not unlikely 

22 The third figure cannot be conclusively identified as either male or fe­
male. Amandry argues persuasively for the river god Pleistus, as the figure 
seems to be holding a horn; on the horn see R. M. Gais, "Some Problems of 
River-God Iconography," AJA 82 (1978) 355-70; Kritzas (n.23 infra) identifies 
the figure as Agathe Tyche. 

23 The Muses are considered and rejected by C. Kritzas, "Muses delphiques 
a Argos," in L'Antre corycien I (=BCH Suppl. 7 [Paris 1981) 208. Hermes also 
has links to Parnassus through his cattle-stealing son or protege Autolycus, 
who is said to live on Parnassus (ad. 19.392-466; Paus. 8.4.6). 

24 See J. Marcade, in Antre II 314f. On reliefs of the nymphs see C. Ed­
wards, Greek Votive Reliefs to Pan and the Nymphs (diss.New York Univer­
sity 1985); R. Feubel, Die attische Nymphenreliefs und ihre Vorbilder (diss. 
Heidelberg 1935); W. Fuchs, "Attische Nymphenreliefs," AM 77 (1962) 242-
49, figs. 64-69. 

25 The association of the nymphs with Artemis belongs primarily to epic 
and with a few exceptions, such as the Laconian Caryatids, does not extend 
to cult. Hermes and nymphs: Hymn. Hom. Yen. 262f; Hymn. Hom. Pan 27ff; 
Semon. fro 20 West; Hes. fro 150.31 M.-W.; Eumaeus' sacrifice to Hermes and 
the nymphs: ad. 14.435ff; Ar. Thesm. 977; Soph. aT 1104; etc. For black­
figure vases see P. Zanker, Wandel der Hermesgestalt in der attischen Vasen­
maleri (Bonn 1965) 19-24, esp. 21; 56-59; PI. 1-2; for inscriptions: IG II/UF 
2934, 4278, 4546, 4728. In nymph reliefs Hermes is usually shown as nympha­
getes; Hermes is part of the original composition and Pan is added to the 
rocks above or is squeezed in to one side: P. Borgeaud, The Cult of Pan in 
Ancient Greece, tr. K. Atlass and J. Redfield (Chicago 1988) 158f. 

26 On Pan's introduction to Attica see Borgeaud (supra n.25) 133ff. 
27 Amandry, in Antre II 398. Pan is not mentioned among the Parnassian 

gods in Aesch. Eum. 1-28. For a terra cotta circle dance of Pan and the 
nymphs, probably marking the introduction of Pan's cult to the cave, see A. 



Reproduced with permission of the copyright owner.  Further reproduction prohibited without permission.

350 NYMPHS AND BEE MAIDENS 

that the cave may have been sacred at first to Hermes and later 
passed into the sphere of Pan, who in many ways succeeded 
Hermes as the companion of the nymphs. 

The date of the Hymn to Hermes is unknown, but scholars 
seem to agree that it is one of the latest hymns, belonging 
perhaps to the late sixth century, though some studies place it as 
late as the fifth.28 Cult activity at the Corycian cave begins in the 
seventh, though the period of greatest use according to the 
excavators was 550-350. 29 Thus the Hymn appears to coincide 
with the first great popularity of the cave. In the context of the 
present theory I would suggest that the section of the Hymn 
describing the Bee Maidens antedates the advent of Pan at the 
cave, which came sometime after the battle of Marathon. Given 
the debate on whether the last lines are a "late addition," I shall 
assume, with the majority of recent scholarship, that they are 
contemporary with the rest of the poem Qanko 133). 

Cleromancy 

The presence of the astragali in the Parnassian cave is compat­
ible with a cult of Hermes, a god associated with games of 
chance, lucky windfalls, and cleromancy.30 Astragalus-oracles of 
Hermes are well-attested at several sites in Asia Minor, though 
these are all considerably later than our passage. 31 Hermes also 

Pasquier, "Pan et les nymphes a l'antre corycien," BCH Supp!. 4 (1977) 365-
87, esp. 383. 

28 Scholarship on the date of the Hymn summarized in Janko 133-50. 
29 Jacquernin, in Antre II 154; cf Amandry, Antre II 396£. 
30 On the astragali see supra n.19; on Hermes, good and bad luck, and cler­

omancy, see A. Bouche-Leclercq, Histoire de la divination dans l'antiquite 
(Paris 1879) 190-97; Eitrem (supra n.3) 783f; Hermes is associated with the 
first lot drawn or thrown; also the first portion given: Hsch. S.v. 'Epj.iO\i KAi1~; 
Suda S.v. lCA.i1POC; 'Epjloo; Pollux 6.55 Bethe; l: Ar. Pax 3M£. 

31 Inquirers at these oracles threw sets of astragali, then consulted metrical 
inscriptions listing the interpretations of their throws. The presiding gods were 
usually Apollo and Hermes. This particular system, however, seems to have 
been unique to Asia Minor, and the attested examples (from Pamphylia, 
Pisidia, Phrygia, and Lycia) date to the second century. If an astragalus oracle 
existed at the Corycian cave, it must have been a much simpler affair. For the 
sites in Asia Minor see the early work of T. Hopfner, "Astragalomanteia," RE 
Supp!. IV (1924) 51-55; F. Heinevetter, Wurfel- und Buchstabenorakel in Grie­
chenland und Kleinasien (diss.Breslau 1912); and the discussion in W. R. 



Reproduced with permission of the copyright owner.  Further reproduction prohibited without permission.

JENNIFER LARSON 351 

had an oracle at Pharae in Achaea (Paus. 7.22.2), which was con­
sulted by burning incense and filling with oil the lamps attached 
to his statue, then whispering the question in the god's ear. The 
answer was the first phrase heard after making the petition. This 
type of divination is very old indeed (el Odysseus and the 
women at the mill, Od. 20.100). The absence of a priestly inter­
mediary gives it (like the use of astragali) both a popular appeal 
and a lower status. 

It might be argued that because mantic pebbles continued to 
be used at Delphi, it makes little sense for Apollo to have 
'handed over' a form of c1eromancy to Hermes (Feyel 10). On 
the other hand, it is possible to distinguish between c1eromancy 
performed by the Pythia, i.e., under Apollo's influence, and the 
do-it-yourself methods practiced by 'pebble-tossers'.32 The 
Pythia might draw lots under Apollo's influence 33 or she might 
interpret pebbles agitated in a tray or jar, as in some later 
accounts. 34 But in no case does she 'toss'. The solitary user, 
without the aid of a priest or seer, consults both mantic pebbles 
and astragali by "tossing" (ballein). 35 The notion of this method's 
inferiority might also have been reinforced by its similarity to 
gaming with dice and astragali-another activity in the sphere of 
Hermes. 

Halliday (supra n.19) 213 with n.3. See now C. Naour, Tyriaion en Cabalide 
(Zutphen 1980) 22-36, where the oracle was ceded from Apollo to Hermes, as 
in our Hymn; J. Nolle, -Epigraphica Varia," ZPE 48 (1982) 274-82 (5 E G 
XXXII 1313). 

32 The pebble-tossers (eptOjX>Aot) were the subject of a derisive proverb, ltOA­
Act eptO~6Act, ltaupot OE 'tE llaV'tlE~ aVOPE~ (Steph. Byz. S.7/. epia). For the dis­
tinctions between ballein, piptein, and anairein, see Amandry 26 n.15. 

33 Hence the term anairein; see Scheinberg 9 with n.36; Nilsson, GGR3 171; 
Amandry 25 n.lo 

34 See Amandry 27-36, 63£; Scheinberg 9; Suda S.7/. nuero; Fontenrose (supra 
n.6) 426-33, and The Delphic Oracle: Its Responses and Operations with a 
Catalogue of Responses (Berkeley 1978) 219-23, who argues unconvincingly 
against all use of lots at Delphi. 

35 Thus Athena is said to have Ktossed" pebbles (thriai) into the Thriasian 
plain: Etym. Magn. S.7/. epia. KThria" is a generalized term for lot: thriai, 
sometimes -leaves," can also be drawn (Apollodorus, FGrHist 244F153; Hsch. 
S.7/. epta~Etv). Artemidorus (3.1) equates 1C1l~Ot (dice), (hnpayaAOt (knuckle­
bones), and IIf11q)(lt (pebbles) in the context of gaming, presumably because they 
are all thrown. 



Reproduced with permission of the copyright owner.  Further reproduction prohibited without permission.

352 NYMPHS AND BEE MAIDENS 

Bees 

The main objection to the theory I have outlined is that the 
Corycian Nymphs are never, so far as we know, referred to as 
'bees' .36 This may not, however, be a serious problem. The con­
nection of nymphs in general with bees is much stronger than 
usually supposed, and in several instances bees and nymphs are 
interchangeable. From the time of Homer, bees are said to issue 
from clefts in the rocks (Ii. 2.87ff, 12.167ff). Their hives require 
sheltered spots with access to water. Caves are the ideal homes 
for bees because they often contain springs-hence their selec­
tion as the standard cult site of the nymphs. The cave of the 
nymphs on Ithaca actually contained a hive of bees (0 d. 
13.103), and we must assume that such installations in other 
caves, whether of wild or domesticated bees, were not 
unusua1.37 

In the stories of Zeus' infancy in a cave, he is often said to 
have been fed on honey either by nymphs, bees, or both. 38 

Scholiasts to Pindar preserve Mnaseas' fourth-century account 
of how certain nymphs discovered honey and taught human­
kind, along with its use, to refrain from cannibalism, to eat 
fruits, and to wear clothing. The nymph Melissa gave her name 
to the insect, and this explains why "those who devote them­
selves to holy things" are called "melissai," for without nymphs, 

36 This is an important reason for ruling out the Thria~ who likewise have 
no connection with Hermes; but the Corycian Nymphs, like other nymphs 
worshiped in caves, have strong associations with both Hermes and bees. The 
Thria~ though referred to by Philochorus (supra n.2) as nymphs, are trans­
parently personifications of the mantic pebbles used at Delphi and probably 
have no separate existence in cult. 

37 The literature on bees in Greco-Roman antiquity is extensive. See A. B. 
Cook, -The Bee in Greek Mythology," JHS 15 (1895) 1-24; H. Ransome, The 
Sacred Bee in Ancient Times and Folklore (London 1937); L. Bodson, Hiera 
Zoia: Contribution a l'etude de la place de l'animal dans la religion grecque 
ancienne (Brussels 1978); M. Davies and J. Kathirithamby, Greek Insects (Ox­
ford 1986) 47-73; also of interest: M. Fraser, Beekeeping in Antiquity 2 (Lon­
don 1951); E. Crane, The Archaeology of Beekeeping (London 1983). 

38 Bees: Rhea gave birth in a cave sacred to bees, who nourished the infant 
Zeus with honey: Boio, in Ant. Lib. Met. 19; bees nourish the infant Zeus: 
Verg. G. 4.149ff; Servo in Aen. 3.104; Lactant. Plac. in Stat. Theb. 7.784 (=Myth. 
Vat. 1.1 04, 2.16); nymphs: nymphs feed honey to Zeus: Diod. 5.70.3; the 
nymphs Adrasteia and Ida, daughters of Melissaeus: Apollod. BibL 1.1.7; Zeus 
cared for by Dictaean Meliai and bees: Callim. jov. 46ff; Nilsson, GGR3 320f; 
on Zeus' birth and early life in caves see P. Faure, Fonctions des cavernes 
critoises (Paris 1964) 110-26. 



Reproduced with permission of the copyright owner.  Further reproduction prohibited without permission.

JENNIFER LARSON 353 

continues the scholiast, no rite of Demeter is conducted, nor is 
any marriage accomplished. Thus the scholiast assimilates the 
various meanings of melissa: "bee, lO "priestess," and "nymph." 39 

The bee may be a messenger or proxy of the nymph. In the 
story of Rhoecus (Charon of Lampsacus, FGrHist 262F 12), a 
nymph agrees to sleep with him after he spares her tree. She 
says that a bee will act as the messenger when the time comes 
to fulfill her promise. But when the bee arrives, Rhoecus 
shouts at it and incurs the wrath of the nymph, who blinds him. 
The other usual place for beehives was the hollow of a tree, 
which coincides of course with the other main locus of nymph 
lore.40 These connections were recognized in the ancient name 
of "nymph lO for the larval bee. 41 Bee and nymph are both para­
digms for different aspects of ideal wifehood: the nymph is a 
model of the bride's beauty and sexual allure; the bee repre­
sents the ideal wife's industry and sobriety. Both, however, 
unlike the human wife, retain their independence and both can 
visit punishment on men when they are displeased. 42 

J9 LPind. Pyth. 4.106. Hesychius glosses BplOat as W~t. The word is 
etymologically related to (iAh:'tEtv ("to remove honeycombs"). Brisai too are 
nymphs who taught about apiculture: Etym. Magn. S.'V. (ipioat; Arist. fro 511 
Rose; for parallel roles of bee and nymph in the Aristaeus myths see Verg. C. 
4.317; M. Detienne, -The Myth of Honeyed Orpheus," in id. et aI., Myth, 
Religion, and Society, ed. R. L. Gordon (Cambridge 1981) 102f, esp. 60-63; on 
bee-nymphs see further Hsch. S.'V. OPOO£/lVUlO£;;' al VUflCj)at !Cat al fl£Al't'tat; 
Cook (supra n.37) 15ff, 19. 

40 For the bee in the hollow tree see Hes. Gp. 233, Th. 594 with West's 
notes; Strab. 2.1.14, C73. 

41 Arist. HA 551b, 555a, etc.; Photo S.V. VUflCj)Cll; Pollux 2.147 (bees when their 
wings are grown); Plin. HN 11.48. See G. Fernandez, Nombres de insectos en 
griego antiguo (Madrid 1959) 208-12; J. Winkler, -Gardens of the Nymphs: 
Public and Private in Sappho's Lyrics, H in H. P. Foley, ed., Reflections of 
Women in Antiquity (New York 1981) 78. The folktale of the poet or prophet 
whose lips are moistened with honey by the bees is widespread, but it also 
has many intersections with nymph lore. In Plato's case the visitation by the 
bees is said to have happened when his parents took him to Mt Hymettus to 
sacrifice to the nymphs and Apollo Nomios (Olympiodorus in Ale. 2.24-29; 
other sources in Riginos infra). This story probably refers to the cave of the 
nymphs at Vari on Hymettus: see A. S. Riginos, Platonica: The Anecdotes 
concerning the Life and Writings of Plato (Leiden 1976) 20f; G. Fowden, 
-City and Mountain in Late Roman Attica," ]HS 108 (1988) 48-59, esp. 57. In 
Neoplatonic theory the bee-nymph equation was taken for granted, as in 
Porphyry's discussion of the nymphs and bees as souls. On the bee-nymph­
soul intersection see esp. Davies and Kathirithamby (supra n.37) 64f; Cook 
(supra n.37) 1M; Fernandez 208-12. 

42 Semonides' Bee woman: fro 83ff West; Xen. Gee. 7.32ff; Pluto Mor. 144D. 



Reproduced with permission of the copyright owner.  Further reproduction prohibited without permission.

354 NYMPHS AND BEE MAIDENS 

These associations make it at least plausible that we under­
stand the bee maidens of the Hymn as nymphs. The Corycian 
Nymphs did have one rather beelike quality: they were said by 
the fifth and fourth-century dithyrambic poet Philoxenus to 
inhabit '"golden-roofed chambers" (auto\. yap ~ha TIapvaO'O'ou 
XPuO'opo<p<ov NUJ,UpEc.oV £to'c.o 9aA,6.llc.ov), and the cave was said to 
appear golden. 43 Words derived from £p£<pc.o are often used to 
describe caves; they also are appropriate to beehives, as Hes. 
Th. 594, EV O'IlTtVEO'O't Ka'tT}pE<p£EO'O't IlEA,tO'O'at KT}<PTlvac;; ~Kc.ocrt, 
598, E1tT}pE<pEac;; Ka'ta O'lllJ3A.oUC;;, and Arist. H A 624a6, where 
o po<pTt is used to describe the inner ceiling of the beehive to 
which the bees first attach their combs. 44 Ancient (and many 
modern Greek) beehives were terra cotta vessels shaped like 
cylinders with one end closed and rounded; the analogy to the 
cave is obvious (see Crane [supra n.37] 45-50). 

Bees and Divination 

Another issue is the exact nature of the Bee Maidens' method 
of divination, for they are said to tell the truth when given 
honey, but without it they '"lie as they swarm in and out" (562f). 
Scheinberg envisions priestesses who drink an intoxicating 
beverage made from honey and prophesy under its influence. 
This scenario, though it has an excellent Indo-European pedi­
gree, is not without problems. First, there is little or no evi­
dence in the Greek world for divination under the influence of 
intoxicants. The Pythia is never said to drink an intoxicating 
substance, and her chewing of laurel leaves, if true, had no 

43 In Page, Poet. Mel Gr. 430. Philoxenus is quoted by Antig. Car. Mir. 127 
(p.31ff Keller), who comments, ·the Delphians say that on Parnassus at cer­
tain times the Corycian cave appears golden." Cf Boio's story of Zeus' birth 
cave, sacred to bees (Ant. Lib. Met. 19), where a fiery light comes from the 
cave at certain times of the year. Cf Fontenrose (supra n.5: 431), who inter­
prets the Philoxenus passage rather loosely to mean either that the Nymphs 
wear golden raiment or that the cave contains a golden treasure. The descrip­
tion of their cave as a hollow rock by the Pythia (Aesch. Eum. 22: 1tEtpa 
lOOtAtt) is also suggestive of the hollow rocks from which bees issue at Il 2.87: 
1tEtPTl; EX 'YAaql\lpTj;; 12.167: lCOlA.oV oo~. 

44 £1tT}P£ql1\; of overhanging rocks: Od. 10.131, 12.59; of overhanging rocks in 
a cave: Ap. Rhod. Argon. 2.736; xatTlP£ql1\;. the ·vaulted cave": Od. 13.349; of 
a cave embowered with laurels: Od. 9.183. (Jif.l~Ao;, used at Hes. Th. 598 for 
"hive," means a hole in the rock used as a hive at Ap. Rhod. Argon. 1.880. 



Reproduced with permission of the copyright owner.  Further reproduction prohibited without permission.

JENNIFER LARSON 355 

pharmacological effect. 45 Maenads were under the influence of 
wine, but they were not noted for prophecy.46 In fact, the evi­
dence for the brewing of mead in the archaic and classical 
periods is scanty, though the Greeks seem to have been well 
enough acquainted with 'hydromel' and other such mixturesY 

Secondly, Scheinberg assumes that the Bee Maidens are 
human prophetesses like the Pythia. Presumably they would be 
attached to a cult of Hermes, for the oll'Pi] given in the oracles 
will be his (566). This too is problematic. Although priestesses 
called "bees" are known, they are attested with certainty only in 
the cult of Demeter. Possibly also they were connected with 
Artemis Ephesia. 48 But nowhere are they, or any other pries­
tesses for that matter, attached to Hermes. The "Bee Maidens" 
of the Hymn are actually described in sharp detail not as meta­
phorical bees but as real insects. They have wings and they fly; 
they rage (clearly the buzzing of an excited hive); they swarm in 
and out. Why should we not assume that the bees described are 
real bees residing in an oracular cave of the Nymphs? A normal 
libation for both the nymphs and Hermes is honey.49 Perhaps 
the visitor made a libation to the bees/nymphs and consulted 

45 The trance was self-induced: W. Burkert, Greek Religion, tr. J. Laffin 
(Cambridge [Mass.] 1985) 116; H.-W. Parke and D. E. W. Wormell, The 
Delphic Oracle I (Oxford 1956) 39. 

46 Despite Eur. Bacch. 298-301. For the question of Dionysiac influences on 
the Pythia see Fontenrose (supra n.34) 207. 

47 Use of the term 'hydromel' is problematic, because it can refer to either a 
mixture of honey and water, or the same product fermented, or mead proper: 
PI in. HN 22.24; see Fraser (supra n.37) 142-46. 

48 On Bee priestesses see Pind. fro 158 SnelP; Hsch. S.V. ~tcrcrat; Porph. De 
Antr. Nymph. 18; P.Oxy. XV 1802 xi.29-35; Callim. Ap. 11 Off; 1: Pind. Pyth. 
4.106c; Servo in Aen. 1.430; Lactant. Div. Inst. 1.22; Page, GLP no. 91.408f (3Td 
C. B.C.). Priests at Ephesus were called" drones" (Etym. Magn. S.v. 'EcroiIv; 
Callim. Jov. 66; Paus. 8.13.1) and Melissonomoi are mentioned at Aesch. 
Hiereai fro 87 Nauck2, but this does not constitute sufficient evidence for bee 
priestesses at Ephesus: see Bodson (supra n.37) 38-43; cf Sourvinou-Inwood 
(supra n.16) 231-51. Pindar called the Pythia the "Delphic Bee (Pyth. 4.61), 
but it unclear whether this is a cult title. 

49 Wineless libations to the nymphs were made in Attica and Olympia: 1: 
Soph. OC 100; Paus. 5.15.10; on VTppaAta in Attica see A. Henrichs, "The 
Sobriety of Oedipus: Sophocles OC 100 Misunderstood," HSCP 87 (1983) 57. 
Bronze amphoras filled with honey were found in the shrine of the Nymphe 
at Paestum: B. Neutscb, TAS NYNPHAS EMI HIARON: zum unterirdis­
chen Heiligtum von Paestum," AbhHeid (1958) 16 with n.18. Hermes is 
satisfied with libations of honey and milk from shepherds: Anth. Pal 9.72. 



Reproduced with permission of the copyright owner.  Further reproduction prohibited without permission.

356 NYMPHS AND BEE MAIDENS 

the oracle, afterwards leaving his dice in the cave.50 If one tried 
to use the dice without making the customary libation, how­
ever, one could not expect a true answer. 51 Certainly bees 
would have been attracted to such a libation, for honey is the 
food of adult bees. 52 The discovery of such an offering might 
well have made the hive 'rage' with excitement. Probably too 
much emphasis has been placed on the use of eUlro, leading 
scholars to believe that a process analogous to that experienced 
by maenads is described. 53 But even if the word has these over­
tones, they might be metaphorical here. The same goes for the 
description of the Bee Maidens as having heads covered with 
barley meal-an apt description of the appearance of worker 
bees, whose bodies are covered with stiff hairs for the purpose 
of collecting pollen; especially the foreparts would be covered 
with what appeared to be flour. 54 Thus, most past explanations 
of the Bee Maidens have taken as metaphorical precisely those 
elements that I would consider literal, and vice versa. The 
theory I have offered, though it is not satisfactory in all 

50 Strictly speaking, dice (kuboi) should be distinguished from knuckle­
bones (astragali). The only functional difference, however, is that astragali 
have four usable sides; dice have six. Both kuboi and astragali were found in 
the Corycian cave, though the astragali were by far the more numerous. 

51 It is also possible that the movements of the bees were used as oracular 
indicators. The text certainly seems to suggest such a method, and the possibil­
ity occurred to Amandry (63 n.l). This would not explain the presence of the 
astragali, but they could have been used somehow in conjunction with the 
bees, or they could represent a method that replaced the early use of bees. 
Oracular bees are not unknown: a swarm of bees led men to the site of Tro­
phonius' oracle (also in a chasm or cave: Paus. 9.40.1 f). This may be especially 
relevant in view of the Hymn's apparent Boeotian authorship: see AHS 174f; 
Janko 143, 149. For prophecies derived from the buzzing of bees see F. Jacoby, 
FGrHist Ills Supp!. 560. Bees could foretell the weather: Arist. HA 627blO; 
Plin. HN 11.20; Ae!. NA 1.11; Verg. G. 4.91, etc. 

52 The ancients knew that bees could be fed in the winter with solutions of 
honey and water or wine: Varro Rust. 3.16. 

53 On this verb see Amandry 63 n.1. It is used in Homeric diction to 
describe rushing and rapid movement: e.g. a flying spear (ll. 11.180), a river 
(21.234), and a storm (Hes. Th. 874). 

54 This explanation was first put forward by Ilgen (1796): see AHS 347. The 
barley meal has been interpreted as an allusion to the ritual use of meal by 
kanephoroi or to alphitomancy: Amandry 61. The description of the Bee 
Maidens as "aged maidens" or "hags" in AHS and H. G. Evelyn-White, 
Hesiod, the Homeric Hymns and Homeriea (Loeb ed.: Cambridge [Mass.] 
1939) 403 n. is unjustified. Parthenos refers not to virginity but to youth (ef 
Scheinberg 5) and is also used to denote nymphs: e.g. Ibycus fro 286 Campbell 
(1991). 



Reproduced with permission of the copyright owner.  Further reproduction prohibited without permission.

JENNIFER LARSON 357 

respects, at least is consistent with what we know about Greek 
divination and about the cult of Hermes.55 

KENT STATE UNIVERSITY 

February, 1996 

55 The fact remains that cleromancy, whether of pebbles or astragali, is not 
mentioned in the Hymn. This is not necessarily a problem, for we know from 
Delphi that more than one method of divination might be used at a given site 
and Hermes has independent connections with cleromancy. The objection 
carries more weight against the Thriai than against the Corycian Nymphs, as 
the Thriai are personifications of the pebbles; see Scheinberg 11. 

I wish to thank the anonymous referee and the Kent State Unversity 
Research Council.