ARISTEAS OF

PROCONNESUS
 y' t
va.vaL O OVTE 'ITE!:,OS'
\ ~' t1
u.uv K1' UpotS'
I ft

ES' y'ITEP,8opl.wv dywva. 8a.v.a.Td.V &S6v.

 ARISTEAS OF
PROCONNESUS
BY

J. D. P. BOLTON
FELLOW OP THE Q.UEEN'S COLLEGE
OXFORD

OXFORD
AT THE CLARENDON PRESS
 OXFORD
VKlffUITY .PUN
Groat Clamldon Street, Oxford OU 6DP
Oxford University Press ls a department of the University of <>xford.
It furthers the University's objective of excellence in research, scholarship.
and education by publishing worldwide in
Oxford New York
Athtns Aucldmd Bangkok Bogoul Buenos Aim C.lcutta
capeTown Chennai Dal"es5alaam Delhi Flortnce HongKong ktanbul
Karachi Kualal.umpur Madrid Melboumt MtxicoCity Mumbai
Nairobi Paris S!o Paulo Sinpport Tuipt!i Tokyo Toronto Wanaw
With associated companies in Berlin lbadan
Oxford is a registered trade mark of Oxrord University Press
in the UK and in certain other countries
Published in the United States
by Oxford University Press Inc., New York
Oxford University Press 1962
The moral rights of the author have been asserted
Database right Oxford Univenity Press (maker)
Special edition for 5andpiper Books Ud., 1999
All rights rtserved. No part of this publication may be reprodured,
stored in a rttrieval system, or transmitted. in any form or by any means,
Without the prior permission in writing of Oxford University Press.
or as expms!y permitted by law, or under terms agrted with the appropriate
reprographics rights organization. Enquiries concerning reproduction
outside the scope of the above should be sent to the Rights Department,
Oxford University Press, at the address above
You must not circulate this book in any other binding or cover
and you must impose this same condition on any acquirer
British Llbra,:y Cataloguing in Publication Data
Data available
ISBN 0-19-814332-X
1 3 5 7 9 10 8 6 4 2
Printed in Great Britain
on-acid free paper by
Bookcraf\ (Bath) Ltd.,
Midsomer Norton
 TO
MY PARENTS
 PREFACE
IN the following pages I have tried, as far as possible, to present
the reader with a developing argument; to advance only from
established positions. However, it has occasionally been neces-
sary to anticipate a conclusion, and anyway there are passages
where reference, either forward or back, may prove helpful.
Such cross-references within this book are indicated by numerals
in square brackets: to a chapter, by roman numerals; to a page,
by arabic. Arabic numerals preceded by the letter T refer to
the Greek and Latin texts given on pages 207-14. References
to other modem works are enclosed in round brackets, after
the name either of the work or of its author; an arabic numeral
indicating the page, a roman the volume (or, if off-set above the
line, a work, as shown in the Bibliography on pages 215-17,
where further details will be found).
The surviving fragments of Aristeas' Arimaspea will be found
numbered as such in the first five of the above-mentioned texts,
on pages 207-8. I have classified as fragments not only professed
quotations from the poem but also summary matter which
is specifically and solely ascribed by later authors to Aristeas or
to the Arimaspea. My ordering of the fragments, which may per-
haps at first appear haphazard, is in fact not so: the first gives
the broad outline of the poem's contents, the second the poet's
admission concerning the farthest people he reached and his
dependence on their account for the regions beyond, the third
a description of that people, the remainder their reports on their
neighbours.
It remains to say that, where I refer to quotations by ancient
authors of others whose work is otherwise lost to us, the reader
may find, under the quoted author's name in the IndexLocorum
on pages 2 19-47, a reference to his collected fragments, if such
a collection exists and is readily accessible. For example, on
page 51 there is a reference to a statement of Ephorus preserved
viii PREFACE
by 'Scymnus': if the reader looks in the index under 'Scymnus'
he will find a reference to the relevant lines of that author's
poem, but if he looks under Ephorus he will find opposite 51
the appropriate fragment number of Ephorus in Jacoby's
Fragmente tier griechischen Historiker.
Narrow as the subject of this work is, it quickly revealed to
me the breadth of my ignorance. I am conscious that the book
must contain many imperfections; but it would contain more,
had it not been for the friendly help and suggestions of col-
leagues, to all of whom I feel deeply grateful, and in particular
to Mr. Geoffrey Bownas, Mr. J. G. Griffith, Dr. J. L. Harley,
Professor David Hawkes, Professor Hugh Lloyd-Jones, Mr.
Edgar Lobel, Mr. L. D. Reynolds, and Mr.J. P. Sullivan. I owe
thanks too to the learned readers and staff of the Clarendon
Press, and to its Delegates for permission to reproduce Plates I
and III (b) ; similarly to the Trustees of the British Museum and to
Messrs. Thames and Hudson for permission to reproduce Plates
II and III(a) respectively. Finally, I must acknowledge a special
debt to two scholars, with whose views on Aristeas I sometimes
disagree: to Sir Maurice Bowra and Professor E. R. Dodds; for
had it not been for some remarks they made to me long ago this
book might never have been written.
J. D. P. B.
Oxford, October rg6r

I
I
i.:.l'.

I:I'
r~
;3
q
il
 CONTENTS
LIST OF PLATES X

I. FOUNDATIONS
II. KNOWLEDGE OF THE ARIMASPEA
IN ANTIQUITY 20
III. THE POEM'S DEBTORS 39
IV. TALES OF A TRAVELLER 74
V. THE TRAVELLER'S ROUTE 104
VI. THE POET u9
VII. ARISTEAS PYTHAGORICUS 142
VIII. CONCLUSION 176

NOTES 184
TEXTS AND FRAGMENTS 207

BIBLIOGRAPHY 215

INDEX LOCORUM 219

SUBJECT INDEX 248

 LIST OF PLATES
(at end)

1. MIRROR FROM KELER.MES, State Hermitage Museum,

Leningrad. From M. R. o s To v T z E FF, Iranians and Greeks in South
Russia (Clarendon Press)
II. MIRROR-HANDLE FROM ENKOMI. From A. s. MURRAY,
A, H. SMITH, and H. B, WALTERS, Excavations in Cyprus
(British Museum)
III, RATTLE FROM KELERMES, State Hermitage Museum,
Leningrad. a, From TAMARA TALBOT RICE, The SC)lthians
(Thames and Hudson), and b, From M. ROSTOVTZEFF,
op. cit.

MAPS
(at end)

1. Io's Journey: Aeschylus P. V. 707 ff.

II, Eurasia
 I

FOUNDATIONS

EVER since my childhood acquaintance with Alice in Wonderland

I have had a fondness for griffins. Yet even at that age I felt that
Carroll's amiable, ill-educated creature represented but shabbily
the dignity proper to a conflation of the King of Birds and the
King of Beasts, whose attributes should be grimness and fearful
power. Far more satisfying was it years later, skulking through
the solitudes of Galloway with the hunted Hannay of The Thirty-
Nine Steps, to hear the poetic innkeeper quote with ironic but
unconscious aptitude
As when a Gryfon through the Wilderness
With winged course ore Hill or moarie Dale,
Pursues the Arimaspian, who by stelth
Had from his wakeful custody purloind
The guarded Gold.

l read Milton after I read Buchan; and when I did come upon
the source of these lines, in the second book of Paradise Last, my
curiosity was properly aroused to know whence their author had
drawn the inspiration for his magnificent picture. The answer
was to provide another puzzle, to which this book suggests a
solution.
In that part of his work where he passes from a discussion of
the origin of the Scyths to a geographical description of their
south Russian home and the more northerly parts of Europe
and Asia, Herodotus states the following1 [T. r; r2; 2]:
Aristeas also, the son of Caystrobius, a native of Proconnesus [an
island in the Sea ofMarmara; Map II], says in the courseofhispoem
that, possessed by Apollo, he reached the Issedonians. Above them dwelt
the Arimaspi, men with one eye; still further, the gold-guarding
griffins; and beyond these, the Hypcrboreans, whose country extended
to the sea. Except the Hyperboreans, all these nations, beginning with
6860 n
II FOUNDATIONS
the Arimaspi, continually encroached on their neighbours. Hence
it came to pass that the Arimaspi gradually drove the Issedonians
from their country, while the Issedonians dispossessed the Scyths;
and the Scyths, pressing upon the Cimmerians, who dwelt on the
shores of the southern sea, forced them to leave their land.
The birthplace of Aristeas, the poet who sang of these things, I
have already mentioned. I will now relate a tale which I heard con-
cerning him at Proconnesus and Cyzicus [on the Asiatic mainland
nearby]. Aristeas, they said, who belonged to one of the noblest families
in the island, had entered one day a fuller's shop, when he suddenly
dropped down dead. Hereupon the fuller shut up his shop, and went
to tell Aristeas' kindred what had happened. The report of the death
had just spread through the town, when a certain Cyzicenian, lately
arrived from Artace (a seaport about five miles from Cyzicus], con-
tradicted the rumour, affirming that he had met Aristeas on the road
to Cyzicus, and had spoken with him. This man, therefore, strenuously
denied the rumour; the relations however proceeded to the fuller's
shop with all things necessary for the funeral, intending to carry the
body away. But on the shop being opened, no Aristeas was found,
either dead or alive. Six years afterwards he reappeared, they told me,
in Proconnesus, and composed the poem which the Greeks now know
as the Arimaspea, after which he disappeared a second time. This is the
tale current in the two cities above mentioned.
What follows I know to have happened to the Metapontines in
Italy two hundred and forty years after the second disappearance
of Aristeas, as I discovered by calculations I made at Proconnesus
and Metapontum. Aristeas then, as the Metapontines affirm, appeared
to them in their own country in person, and ordered them to set up
an altar in honour of Apollo, and to place near it a statue to be called
that of Aristeas the Proconnesian. Apollo, he told them, had honoured
them alone of the I taliotes with his presence; and he himself accom-
panied the god at the time, not however in his present form, but in the
shape of a raven. Having said so much he vanished. Then the Meta-
pontines sent to Delphi, and inquired of the god what they were to
make of this apparition. The priestess in reply bade them attend to
what the spectre said, 'for so it would go best with them'. Thus
advised, they did as they had been directed; and there is now a statue
bearing the name of Aristeas, close by the image of Apollo in the
market-place of Metapontum, with bay trees standing round it. But
enough has been said concerning Aristeas.
With regard to the regions which lie above the country whereof this
 FOUNDATIONS
\
?
3 I
portion of my history treats, there is no one who possesses any exact '!
knowledge. Not a single person can I find who professes to be ac-
quainted with them by actual observation. Even Aristeas, whom I
have just mentioned--even he did not claim in his poem to have got
further than the Issedonians, but on his own confession what he related
of the regions beyond was hearsay, being the account which the
Issedonians gave him of these countries.

At first sight, even though we discount the miraculous element

in it, Herodotus' story is startling enough : at least two and a half
centuries before his time, that is, not later than c. 675 B.c., a
Greek penetrated far into the hinterland beyond the Black Sea
and brought back with him an essentially true account of the
pressure of the nomadic peoples of those regions upon their
westerly neighbours. The fact that it contained this truth in-
dicates that the author of the Arimaspea either had himself made
the journey which he professed to have made, or was using infor-
mation gained from someone who had. There is too a notable
ring of honesty about the poet's confession that the limit of his
journey was the country of the lssedonians, and that his know-
ledge of the parts beyond depended upon hearsay.
If the poem is really to be dated in the seventh century we are
hardly entitled to assume that its contents were derivative-
for from whom, at that time when the Greeks were only just
beginning to establish a foothold on the northern coast of the
Black Sea, would they be derived? On the other hand it is hard
to believe that the first Greek penetration into Scythia was also
to be the deepest, and it is therefore natural to wish to bring the
date of the poem down to a time when the information it con-
tained could be ascribed to the settled intercourse between
native and Greek traders in the coastal marts, or when the
appearance of a Greek among the tribes of the interior would
not be so strange. This prompts the question, is there any indica-
tion that the Arimaspea was in fact composed in the seventh
century?
We must dismiss for the time being the prima fade evidence for
the early date, the 'two hundred and forty years' of Herodotus.
This is relevant to the dating of Aristeas himself, but it would
 FOUNDATIONS
only be relevant to the dating of the Arimaspea if we were sure
that Aristeas was its author; an assumption which we cannot
yet make, when not only has his authorship but his very existence
been denied by some scholars. I therefore confine myself here to
evidence related to the poem, and postpone consideration of its
authorship to a later chapter [VI]. If in the meantime I call its
poet Aristeas, it is without prejudice and merely as a convenience
to by-pass the phrase 'the author of the Arimaspea'.
The poem provides itself with an upper terminus by its mention
of the ousting of the Cimmerians from their homeland about the
Sea of Azov. Already towards the end of the eighth century these
had moved southwards past the Caucasus, and clashed with the
Assyrian empire. After a defeat in c. 679 by Esarhaddon they
turned west into Asia Minor, killed the Phrygian king Midas in
battle, and in the middle of the century sacked Sardis and
Magnesia; their power was then broken, under Scythian and
Assyrian attack, but Cimmerian settlements persisted for a longer
or shorter time at Sinope, Antandrus, and in Cappadocia (Minns
42 ; Rostovtzeff 1 36).
Whether these Cimmerians were in fact one people, who in-
vaded Asia Minor at one time from one direction, is doubtful;
but what matters for my purpose is certain enough-that
northern invaders whom the Greeks called Cimmerians were
active in Asia Minor from c. 680 till c. 630 B.c. This nearby
activity would have closely concerned the audience of a Pro-
connesian Greek poet, and have justified an explanation by him of
why the foreigners had left their own land.
If we could be sure that the reference to the Cimmerian migra-
tion in the poem was topical and not historical, we could date
the Arimaspea fairly confidently between c. 670 and c. 6!:w.
Aristeas may have spoken as if the Cimmerian emigration from
Scythia was completed, but the movement of peoples which
caused it was still continuing. The movement ceased after the
seventh century-the evidence of sixth-century graves shows
that by then the Scythian kingdom was well established and
settled in south Russia (Rostovtzeff 1 41 )-and was not renewed
till nearly 400 years later, by the Sarmatians (Vernadsky 74).
 FOUNDATIONS 5
Rawlinson understood Herodotus' present infinitives (E7TiTl8Ea6a.,
and Efw9JEa8a.,) of the Arimaspi, Issedonians, and Scyths (con-
trasted with the aorist eKtMrc'iv of the Cimmerians) to stand for
present indicatives in oratio recta [T. 1] ; but they may stand for
impenects, and I have emended Rawlinson's translation to meet
this possibility.
There can be little doubt that the Arimaspea was known to
Aleman, who mentioned the Issedonians in the form 'E(]C17JOOvs,i
Nevertheless this variant is for Schmid-Stahlin (i. 303) and for
Meuli (154), who favours a sixth-century date for the poem,
sufficient proof not only that Aleman did not hear of his 'Esse-
donians' from the Arimaspea, but that he did not know of the
existence of such a work at all. This is hardly reasonable. It is
true that in the one place where they are mentioned in the frag-
ments of the Arimaspea they are called 'la07J80{; but this cannot
be treated as evidence that the poem was unknown to anyone
who did not use the same form: that would disqualify Herodotus!
Aristeas could have used both' la07Joot and' laU7J86vs, metri gratia
-we should indeed expect to find such variants in an early poem.
It is not difficult to account for the slight change to 'EaC17JOOYs as
a mishearing of an outlandish name ;3 Aleman would more prob-
ably have heard the poem recited than have read it. Again, if he
did not get his know ledge of this remote tribe from the A.rimaspea,
whence did he get it? From some proto-Aristeas, or Ur-Arimaspea?
Finally, this, though it is the weightiest, is not the only indication
that Aleman drew on the poem [ 40 ; 43].
About Alcman's own dates we cannot, unfortunately, be cer-
tain. D. L. Page, in his circumspect discussion of the question in
his edition of the Partheneion ( 164), concludes that we are not
entitled to accept more than that he lived some time in the
seventh century. He may have survived into the sixth; but it
would not, I think, be held that his jloroit could be later than
c. 600. That the lower terminus for the Arimaspea must be about
this time is confirmed by archaeological evidence. In 1904 there
was discovered in a barrow at Kelermes, near the river Kuban
beneath the north-western spurs of the Caucasus, a silver mirror.
The back of its disk is divided radially into eight segments, in
6 FOUNDATIONS
each of which some picture is engraved: the 'Winged Artemis';
sphinxes ; lions, boars, and other creatures; and in one segment
two shaggy savages fighting with a griffin. It was made about
the beginning of the second quarter of the sixth century B.c., and
is the work of a Greek [89 ff.].
Here is our earliest example in Greek art of the 'grypomachy',
the battle between griffins and men, which was later to become
quite a popular theme. It seems to indicate that already by
c. 570 monsters who were opposed by human adversaries were
located in the Scythian region by report. For the artist of
the mirror these monsters were griffins; we may guess that
the two savages are meant to be Arimaspi, though they do
not appear to be one-eyed (they never are in fact so represented
in art).
There is of course the possibility that the artist was filling in
his panel with a subject which had no connexion with the
destination of the finished article-a conventional subject,
perhaps, like the sphinxes, or even one out of his own head :
a fantasy which itself was later to give rise to the rumours of
northern barbarians doing battle with griffins. I think this is
improbable. If the subject wei:e a conventional one in Greek art
we might expect to have some earlier example of it; yet we have
none. Nor was it a fantasy of the artist's own, for he had precedents
for his scheme in Phoenician art, where we find fights depicted
between men and griffins, the attitudes of the antagonists being
strongly reminiscent of those on the Kelermes mirror [87 ]. Why
should he have resuscitated this scheme? It is curious that the first
Greek example of the grypomachy which we have should vir-
tually coincide with the opening of the Scythian market. The
obvious explanation is that the artist had heard from some source
of men fighting griffins in the country for which his mirror was
destined.
Whether this source was Scythian or Greek, the identification
of the men's monstrous enemy with the griffin could only have
been Greek; for we have no reason to suppose that the early
Scythians were acquainted with this beast, whereas we know that
the early Greeks were [88]. Was it the artist himself who, hearing
 FOUNDATIONS 7

a Scythian story, equated the monsters therein with griffins?

Or did he receive it from another Greek, with the equation ready-
made (if indeed the whole story was not a Greek fabrication)?
lfwe knew that the artist dwelt in Asia Minor, I should have
no hesitation in concluding that his intermediary must have been
a Greek, and was likely to have been the Arimaspea; but he was
probably an inhabitant ofone of the recently established entrepots
on the north coast of the Black Sea: if so, the possibility of his
having got the story from his Scythian neighbours would have to
be considered. This possibility might seem to gain in attractive-
ness from the statement of Herodotus that the Scythians heard
of the one-eyed men and the gold-guarding griffins from the
Issedonians, and passed the account on to the Greeks (iv. 27:
oe .-KVO'ewv 71ei<;
1ra.pa't-'"" A "'' , ' , r
o, WV\Ot vevo.,Kap.ev, Kat ovo.a.,,op.ev , '
avrov,
uKvOurrl llp,a07rovs). But even if Herodotus can be trusted when
he says this (implying either that the Arimaspea was not the source
of the Greeks' belief in the Arimaspi, or that the Scythians rather
than the Issedonians were the informants of its author), his visit
to the Scythian coast took place more than a century after the
making of the Kelermes mirror, when relations between Scythians
and settlers were long-standing and close, and the trade-route to
the far interior well established-not a very good parallel for
conditions at the beginning of the sixth century.
I am inclined, therefore, to conclude that the grypomachy
motif was suggested to the artist from an outside sourer, and that
a Greek onc--the Arimaspea-and to regard it as buttressing the
evidence of Aleman for a lower terminus for the poem. This,
combined perhaps with the testimony of Aristeas himself that
migratory movement was still continuing in tht' Eurasian hinter-
land in his day, will date it with fair certainty between c. 670 and
the end of the seventh century, at latest.
I have deferred consideration of the surviving fragmrnts of the
Arimaspea so far because, in an investigation so best't bv pwbltms
and dubiety, the one firm rock is tht' notict' in Herodotus. On
this foundation we must build upwards and tl fr,,r) outwards,
and it seemed best to see what could be inltrred ;\bout the d.lle
of the poem from Herodotus and earlier l'vidtnce, and then to
8 FOUNDATIONS
inquire how far the professed remnants ofit square with the result
so obtained; for these remnants are preserved in late authors, and
their genuineness cannot be taken for granted.
Twelve lines are quoted in all, six by the Byzantine Tzetzes
(frr. 3-5), and six by the author, known as 'Longinus', of the
work On the Sublime, written probably in the first century of our
era (fr. 7) [T. 3; 5]:
fr. 3 'Iaarillol xalrr,au, dya>.>.6.evo, Tavafi<n
The Issedi glorying in their long hair
> 8
fir. 4 Ka.L .,,aa > av
I ,/.. <
pwrrovs EtVaL
I
Ka 8V1Tp 8EV o.ovpovs
< I I

' f1 \\ I > 8' \

rrpos opEw, 7TO/I/\OVS T Kai 1:a 11ot1s Kap'Ta p.axri'Ta,,
I \ I I

o.rpvnovs trrrro,ai, rroMpprivas 1ro>.vf301JTas.

And they say that there are men neighbouring them above towards
Boreas, numerous and very doughty warriors, rich in horses and possess-
ing many flocks and many herds of cattle.
',l..8-WlfkOV
f f. 5 O'i' \ \ O<,, V
o > "
EKaUTOS" EXE I
xaptEV'Tt fk'TW'Tr(t), I

xal-rr,a,{v) Acta,o,, 71'0.V'TWJJ a-r,f3apw,-a-ro, dvopwv.

Each has one eye in his comely forehead. They are shaggy with hairs,
toughest of all men.
&'.
1r. 7 8""'f''"'
av. T/f.1.'V Kai,TOV'TO
_ p.yaI.J..\~ I
't'PEULV T/f.l.E'TEprJUW.
11.vopes v8wp valova,v a.rro xBovos l.v rre.>ufyeau,
llva'TT/vol 'TWES law, ifxova, yap ifpya 1TOV'1JpO.,
~ ) > >1 I \
./, f
, ~, t1
o..a,- El' aaTpo,a,, 'f"VXT/V O EVI 1TOV'T(t) EXOVO'W,
~ 1TOV ,roMa 8ofo, ,f,l>.as 6.vd. XEipas :IxoVTES
o, \ I ~ > Q .. \ \ I
Etl](OVTaL arrAayxvo,a, KaKWS avapal\l\O/J,VOLaL.

This too we remark with great wonder: men dwell in the water,
far from the land in the midst of the sea. Unlucky wights they are, for
they suffer grievously, with their eyes on the stars but their life
amidst the waves. Assuredly, lifting up their hands to the gods, many
are the prayers which they roust make, with entrails sorely tossed.

Fragments 3-5, which are quoted by Tzetzes in uninterrupted

succession, though not a syntactic whole hang together in sense,
especially if we accept Hubmann's emendation of cf,a.a' for the
nonsensical urpas of the manuscripts in 4. 1 : the subject of tf,aul
may then be the Issedonians of fr. 3 (' I<rariool: another form of
'laario611,, according to Stephanus Byzantius under the latter
 FOUNDATIONS 9
name; also used by Zenothemis [67 f.]),and fr. 4may refer to the
Arimaspi as fr. 5 clearly does.
The subject-matter of fr. 7 presents more difficulty. Who are
these unlucky people whom it mentions? In the context in which
he quotes it 'Longinus' is discussing the treatment of the storm-
theme by poets, and compares the ways in which Homer and
Aratus do it: any tempest-tossed sailors, then, would seem to be
meant, but these would cause no surprise to a Greek, to whom
the hardships of a nautical life would be almost proverbial (cf.
vit. Hom. Herodotea 262 : va.ihm 7Tovro71opoi crrtrypfj Jva).lyKm,
~TT/, 1T'T'WKdC1LV a.l8vlncn {J[ov 8va{1)AOV EXOV7'S). Is Aristeas then not
speaking here in his own person, but ascribing these words to some
landsman to whom ships would be strange? This is the view of
Rhys Roberts in his edition of the treatise of 'Longinus' (218),
andofTournier (29), who thinks that men totally unversed in sea-
faring are expressing their astonishment at hearing of the boldness
of Greek sailors. The former would identify the speaker as an
Arimaspian, the latter as one of the untoiling Hyperboreans, and
see Aristeas as satirizing, through their mouths, his country-
men's way of life. However, we know that the author of the
Arimaspea did not reach the Arimaspians, much less the Hyper-
boreans; so Bowra (4), while accepting this general interpretation
of the fragment, suggests that it may express the view of the
Issedonians.
There are two slight indications that here it is not the poet
himself who professes to be speaking. First, Homer never uses the
particle~ except in the mouth of one of his characters (1. 5; J. D.
Denniston, The Greek Particles 279); but it would be dangerous to
assume that Homer's practice must also have been th.1t of a poet
in a different genre, who, relating or professing to relate his
own personal experiences, would be entitled to express personal
opinions (cf. Hes., Op. 333). Secondly, the use of the plural
~,iv and ~/1,ETEP'lJO'LV in 1. 1 : if this is really the work of a seventh-
century author these words are unlikely to refer to himself,
or to himself and his Greek audience, or to the Greeks generally
(Bowra 4 f.). 4 This argument has more weight, but it takes the
genuineness of the fragment for granted.
10 FOUNDATIONS
Regarding this thesis it must be remarked that the Issedonians,
a distant inland people, would know nothing about Greek, or any
other, seafaring in the seventh century B.c. In other words,
Aristeas must be making it up, and he must be making it up to get
a dig at his fellow-Greeks through the mouth of a barbarian. If
so, I find it hard to believe that these lines can be from the
seventh-century Arimaspea, they being appropriate neither to the
age nor to the faithful reporting for which the poem, as we shall see,
was remarkable.
There is another possibility: that sailors, or at any rate sailors
in the ordinary sense, are not meant, in spite of the context, which
is not in itself conclusive ('Longinus' merely says that the author
of this passage thought it would inspire terror) [26]. Thus Meuli
(155) thinks the sufferers are a sort oflake-dwellers, inhabitants of
houses raised on piles in the midst of the water (so too Schmid-
Stahlin i. 303). The author of Airs Waters Places (c. 15) says that
the dwellers in the marshland of the Phasis had houses just like
' l I
th1s-otlC7/fUlTa t
I\ I \ I > ~ ~,;, I
~v11wa 1<at 1<a11a.wa v Tots voaa, p,e.71xaVl).va
-but these or similar are hardly appropriate here: it would be
gross exaggeration to describe such as 'far from the land in the
midst of the sea', and their inmates as seasick ; and why, or how
(through the roof?), would they keep their eyes fixed on the stars
the while?
For Bethe (877) they are legendary sea-dwellers; similarly
Bowra (3 f.) has ingeniously suggested that they might be
Steganopods, web-footed men bobbing up and down in the waves.
This idea is commended by the fact that Aleman mentioned
Steganopods5 (Strabo 43), and, as has been seen, he appears to
have known the Arimaspea. Aristeas may have heard (from the
Issedonians again?) of the men-fish of old Chinese tale [194], or
a story like that told by a Russian traveller much later of the
Samoyeds on the coast near the mouth of the river Ob: that, in
the words ofHakluyt (i. 467), 'one moneth in the yeere they live
in the sea, and doe not come or dwell on the dry land for that
moneth'. Bowra is disinclined to press his suggestion, on the
ground that such creatures would hardly be called 'men' ; but
as Simias can call the Half-Dogs 'men' [68] perhaps it is not
 FOUNDATIONS II

impossible. A stronger objection to this, as to Meuli's interpreta-

tion, is the statement that they keep their eyes fixed on the stars-
comprehensible of sailors, but why should mermen do so?
Before examining the language of these fragments to see if we
can extract clues to their date a question arises: are they by the
same hand? More than half a millennium separates the preserver
of one group of verses, 'Longinus', from their putative author,
more than a millennium and a half the preserver of the other,
Tzetzes. In the case of the latter, a double act of faith is needed:
we must assume that he took the verses all from the same source,
and that he is quoting them correctly. His habit was to rely on his
memory, vast but not always accurate (Chil.i. 275 ff; cf. e.g. T. r3,
XaAKfov for yvacpdov).
A first impression, from the style of the two groups, is that they
are the product of one brain. This is somewhat confirmed by
metrical analysis. The two groups are clearly too small to justify
any firm pronouncements on this score, yet they are large enough
to yield some results. In such a case the argument from metre is
more trustworthy when it tends to establish unity of authorship
than the opposite. An author may have his idiosyncrasies (e.g.
Apollonius' liking for four-word lines), but is unlikely to display
them in almost every verse; thus if such an idiosyncrasy appeared
in one group but not in the other it could not be taken as evidence
of different authorship (and obviously we must know a great deal
of a poet's work before we can say that he always eschews certain
metrical phenomena, as for instance Callimachus is said to eschew
the bucolic diaeresis after a spondee).
By way of a check I have compared with these two groups a
number of six-line passages, syntactically complete, taken at
random from other authors: three from the Iliad, including one
from the (supposedly intrusive) Do[onea; three from the 04Jssey,
including one from the (supposedly intrusive) second }le~}ia;
three from Hesiod, including one from the (Hesiodic) Shield; two
Delphic oracles from Herodotus; and two passages each from
Callimachus, Apollonius, and Simias. 6
The most conspicuous differences between the Tzetzes and
'Longin us' groups are (a) in the number of spondees in the first
 12 FOUNDATIONS
four feet (thirteen examples to nine); (b) in the number of
trochaic caesurae in the first four feet (four to nine); (c) in the
use of diaeresis (four to seven). But judged on these three counts
alone the Dolonea was much closer to Iliad i than the latter to
Iliadxxii; and on counts (a) and (b) the discrepancy was as great
or greater between the two passages of the Works and Days ( (a)
five to nine, (b) seven to three), as also between those from
Apollonius ((a) eight to three, (b) six to three).
If attention is confined to trochaic caesurae in the first two feet
only, a difference is revealed between the two groups (none to
three) which is not paralleled elsewhere. But this is the only
substantial discrepancy. In the case of (a) above the difference is
more apparent than real, for both the Arimaspea groups are above
the over-all average (seven) in the freedom with which they admit
spondees in the first four feet. Likewise in their use of diaeresis
((c) above) they are both well below the average (eleven). In fact,
in all the phenomena examined with the exception of the trochaic
caesurae, both groups are found on the same side of the average
with each other. It is indeed remarkable that the Tzetzes group
again and again occupies the extreme wing: it shows the largest
number of spondees in the first four feet; of successive spondees
(two examples in the first two feet, and one example of three in
succession in the first three; equal as runners-up are the 'Longinus'
group and the Hecale, each with three examples of two successive
spondees) ; the least use of the trochaic caesura in the first, second,
and fourth feet (no examples at all); the least use of diaeresis;
and the lowest average number of words to the line. But, though
the 'Longinus' group is nearer to the average, in none of these
cases, with the exception of the trochaic caesura already men-
tioned, is it far away from its companion. I conclude, therefore,
that there are no adequate grounds on the score of metre for
believing that the two groups are by different hands, and some
grounds for believing that they are by the same hand.
There is one oddity of prosody : the lengthening in arsi of the
last syllable of xal711a, in xat711a, Mmo,. Homer frequently so
lengthens before certain words beginning with A, but Aaa,os is not
one of them (Il. i. 189 a'T?J81caaw Aaalo,a,). It is found later where
 FOUNDATIONS
one would not expect it (e.g. Theocr. xxii. I 2 I &.1rb >.ayovos; xxv.
246 v110 i\ayovos), but here Tzetzes himself may simply have
omitted the vv l.rf,EAKVU7'Kov, as elsewhere.
Accepting that the two groups are by the same author, let us
now see if the language of either helps to give him a date.
Fr. 5. 2 xalrncn(v) M.u,oi, ?10.VTWV C1T{1apcfJ'TO.TO av8pwv. l:nfJapos,
applied by Homer and Hesiod to human limbs and parts of
the body, is not found until much later applied to men themselves
(Ar. Thesm. 639). A somewhat similar extension is to be re-
marked in the use of Aa.u,os here: it refers not to a part of the body
but to the whole man, as ifhe were a shaggy-coated animal like
a ram (It. xxiv. 125 dts Mmos) (Bowra 8). So at least we may
take it, rather than as meaning 'with long, unkempt hair', when
we should have to go to Xenophon, I think, for the earliest
parallel use of the dative (e.g. Anab. ii. 6. 9 rfi <pwvfi -rpYJxvs;
CiJrOr
,h 11.
.. 3 6 OV'T
~ ' -raxvs
1TOC1tV ' OVTI!
" Xf.PCTLV ,axvpos. He11en-
1
' '

,I
'I
I
istic Greek preferred this construction to the 'accusative of res-
'I
pect'-Schwyzer-Debrunner, Griechische Grammatik ii. 168).
Fr. 7. 2 lv ?TEi\aywui. This epic dative plural is found in the
Odysso, (v. 335), Archilochus (fr. 12. 1), and the Homeric Hymns
(iii. 73; xxxiii. 15), but always in the phrase ru\os iv 1rAayaa,
(hence Empedocles' ai.a'Tos iv 'lTEAaycua,, fr. 105. 1 ). It occurs
unqualified in the Alexandrians (e.g. Call. H. iv. 36; Ap. Rhod.
iv. 240), a usage foreshadowed in Euripides (Or. 990 1Te.\ay1:a,
811:8[tppVC1),
Fr. 7. 3 lxovu, yo.p lpya 'Tl'OV'YJpa, The word 7TOVYJp6s, not found
in the Homeric epics or hymns, has in early Greek the sense of
'havmg. .
or mcumng. troubies' : H eracles was 1ToVYJpo-ra-ro,
' Kat'
11.purros (Hes. frr. 138, 139 R.; cf. Solon, fr. 15. if. oiloJ a.Kap
oti8eis 11li\E-rai (3po-r6s, d>..\a 1ToV'YJpo~ 11av-rf.s). Secondary seems
to be the sense of 'causing troubles', which develops into the
common Attic meaning of 'bad', pravus, with a strong moral tone
when applied to people. wEpya 1roVYJpa stands at the beginning of
this secondary development, as it were a bridge between the two
senses: when the adjective is first applied to things it must have
the connotation of 'having' (not of course 'incurring') 'troubles';
but things can only be said to have troubles in the sense of causing
 14- FOUNDATIONS
them to sentient beings (cf. Theognis 274). So in the so-called
'Homeric epigram' 14. 20 (vit. Hom. Herodotea 458} the poet prays
that the potters' workshop and its contents may be smashed by
the Centaurs and the potters themselves may groan when they see
the trouble they are being caused-a.&rol 8' ol.4,{ovn. <>ptf,o:ro
lpya 7TOV1Jprl. In this fragment of Aristeas it seems to be no more
than a periphrasis for 7t6vov., as Homer's n-ra lpya for -rlai. and
Empedocles' lpya pEv<J'Ta for fmJo:ra (II. xxiv. 213; fr. 121. 3). It
is a short step to regarding nuisances as bad, and, when the
nuisances are people, as morally bad. The non-moral sense of
1roV1Jp6. seems to have become so unusual by the fourth century
that Aristotle can misapprehend the line ov8Els- EKwv 110V7'/pOS'
oil8' d.1<.wv ,&.1<.ap (Victorius: axa.pw.--cf. [Plat.] 11Ept 8,1<.alov
374a)-of origin unknown-to mean 'no one is willingly wicked'
(E.N. iii. 5. 4) ; though the sense of 'causing troubles' recurs in
the following century in Callimachus (H. vi. 65 'Epval,c8ov,
TEVXE 1TOV'f/pd.) 7
Fr. 7. 5 .,,IJlaS'
,JJ\ ' ' xnpa. EXOVTES'. Th.IS ' re ff exive
ava A " ' use of,J,.!\
'f'""o
with parts of one's body, common in early epic, is found hardly,
if at all, in the Alexandrians (at Call. H. iii. 25, (1.1:ryrrJp E) 0')!1)-rl
r/,0..wv &:rrE8~KaTo yvlwv, it may not be reflexive, but mean 'dear
to me'; and at Ap. Rhod. i. 281 and ii. 710 it may mean 'loving').
But the early usage still occurs in the fifth century, in Empedocles
(fjr. I 15. 3 EVTE
1'
I > .\I ,J.
I ,JJ\ - I )
TiS' ap.1TllaKi!JU' .,,ovcp "f'IJla yv,a i'YJV!J
Fr. 7. 6 (11T,\a.yxvo,a, KO.KWS' ava{Ja.>J,,0'110,u,. An extraordinary
phrase: what sort of a dative is it? Hardly a locative, as one does
not pray in one's u71,\ay,,cva, which are usually the seat of the
passions or dispositions ; though Empedocles seems exceptionally
to connect them with ratiocination once (fr. 4. 3 yvwO, 8wr:r10lv-
TOS' (o,auCT'T]Olv-ros Diels) lv1 (J?T,\arxvo,a, ,Myow). Anyway a pre-
dicative participle with such a locative would be unusual. Bowra
{7 f.) suggests that this is a bold use of the instrumental dative:
'while prayer is normally made with words, in this case it is
made with cast-up bowels; (Aristeas') sea-folk are too miserable
to utter any words' : their prayers actually consist of 'physical
reactions of a painful and undignified character'. The expression
would then be very striking, and that our author is audacious
 FOUNDATIONS 15

in his use of language seems to me to appear in op,c.T' iv

a.crrp<>wL exov,n, though here the meaning is unmistakable.
Nevertheless I think myself that this is a 'dative of the accom-
panying circumstances', conjoined with a participle sometimes in
Homer (e.g. It. xi. 555 <i.1Tov6arp,v lf171 Ten716T, 8v.ij,), though the
construction occurs more frequently later, in Xenophon, both
with the perfect and the present participles: 1mrevs- -r,s- 1rp<>U1JAavVE
Kal &>.a luxvpws 1opwVTt Tij, Z1T1T<p (Hell. iv. 5. 7); 71vAt(eu8e eyKE-
xa.\ww~,s- TOtS i1T1TOtS" (Anab. vii. 7. 6); Tatppov Wpu'tTE KVKAo/
1TEp,\TTJV
\I'\ ""''
?TOAtV, TOtS" ~,
ev '1],taEat -
TWV ..,.
u-rpanw-rwv t.1-'
1TpoKav11f.UVOtS
\ " .'L - _.J. I ~ N < I > )' I
uvv 'TOtS O?TIID'S" 'TWV -ru.y,pEVOVTWV, Tois- 0 71,uea,v tiipya,.ocPots-
(Hell. v. 2. 4-a startling example).
This examination of linguistic usage has produced little positive
result. There are two indications of lateness, in the extended use
of rrnf.Jap&s- and the unqualified 1TEAayeaa,; perhaps a third in the
'dative of the accompanying circumstances' (if such it is)
qualified by a present participle in fr. 7. 6. Against these are the
earlier, non-moral sense of 1ToVTJpos and the reflexive ef,O..os; but
the former reappears in Callimachus, the latter in Empedocles,
and it is obviously easier to account for an early usage in a late
author than for a late usage in an early one. However, as we appear
to be dealing with a poet who uses conventional epic language in
a somewhat unconventional way, perhaps little weight can be
attached to these considerations. It remains to see if a study of the
literary relationships and style of the fragments adds any help.
It is notoriously difficult to date a work solely by its literary
relationships (that is, by similarities of phraseology with other
works), because it is rarely possible to be certain which is the
creditor and which the debtor. So here there is a similarity
between fr. 7. 4 'PVX!JV S' Jvi ?TOVT't) exova,v and Archilochus, fr. 2 I
ipvxds exoVTES Kvchwv lv a.yKaAais. This is so striking as to be
hardly ascribable to coincidence; yet it takes us nowhere. So with
other such similarities, mostly adduced by Tournier (22 ff.);
these therefore need not concern us, but there are two places
where it seemed that some result might be obtainable.
Literary borrowing is of two types : conscious and unconscious.
The former, when it is not simple plagiarism, tends to be cento-
 16 FOUNDATIONS
the conflation of words and phrases from another work. The latter
tends to be a process of fragmentation and digestion, in which
the borrowed matter is broken up and the words are dispersed over
a number of lines. Here, some word or idea that the author has
used causes him to recall subconsciously a passage from some
predecessor where the same word or idea appeared, and so
others from that same passage are drawn in the train of that re-
collection, suggested to him as he gropes for expression, and
reproduced anew. If we posit this as a rule of thumb-and it
could never be more-, we might say that conscious borrowing
implies repetition or compression of phrases, unconscious bor-
rowing implies dispersion.
Has this any application to fr. 4? At ll. ix. 154/296 we find
> N~<'
V I
o avapS' vaiovat ,,
7To,,vpp7JVES' '{3~
7TOAV oVTat.

This is obviously related to Aristeas'

acf,vnaus [7r7roia,, 7Ta.\&ppl')VaS' 7TOAv{3o&-ras,

and both to fr. 134 R. of the Hesiodic Catalogus:

Jan TI> 'E..01Tl'T] 'ITOAVA~ios ~8' Jv,\El,wv,
>,/. ,
a'f'VEL'T) >\ '8 Q I
fl-T)llOU]t KO. HllL7TO Eaa ,.,oEaaw
I\ '

EV o' av8pES' va.lova, 7TO.\tipp7JVES' 1ro.\v{3oih-at

7TOAAol d1mprfawi, cpO.\a Ovl]TWV dvOpclnrwv.

Here it would be tidy to see Aristeas as consciously borrowing

from the Iliad, and the Hesiodic poet as unconsciously 'digesting'
or dispersing Aristeas' line, so that the latter's ?uf,veiovs i.7T7roun
produces a<f,vei~ ,~>.oiai in the line before 7TOAVpp7JVES 7ro)i.vf3ov-rai.
Going farther, one might derive the 1ro'Mo{ and dv8pcfmwv of the
last line of the fragment of the Catalogus from the 7TO/l./l.ovs and
dv8p,f.17Tov, of the first two lines of this fragment of Aristeas. 8 How-
ever, inasmuch as the Hesiodic poet actually repeats verbatim
the line from the Iliad, whereas Aristeas does not, it seems just
as likely that the former was consciously borrowing from Homer,
and Aristeas laying the lines from the Catalogus under contribu-
tion. The case might be argued either way.
Perhaps le5s ambivalent is fr. 3
'la<7TJ6ot xal-rr,aiv dya,\MvoL 7ava.fjcn.
 FOUNDATIONS
This is clearly related, by exact metrical correspondence, to Il.
:xx. 222

A line of Xenophanes (fr. 3. 5),

avx<U\0,
' -' ' ](l1L7?10'W
' ay=p.EVo
, ..t, (Wilamow1tz:
' aya/111.0P,EV
' - " ' ') EtmpE7TEE!111V,
' '

which repeats xalT'!JUW ayalloEVO of Aristeas but breaks the

metrical correspondence with Homer, seems therefore to be the
last of the series. Ifso, the line of Aristeas could not be later than
c. 465 B.C.
More remarkable, however, is the use in fr. 4 of the language
of the Ionian geographers. In the first two lines,
Ka' yn,.u
.,_ __ , () '
av p=ovs; ivm~
Ka(J'vm,p (J V op,ovpovs;
' '
\ Q ' \\ , , 9'' ' ' '
7rpos; pOPEf<.tJ, 1TON\OVS TE Ka,\ Ell 11ovs; Kapa p.a)(1JTaS,

there are no less than four technicalities, if I may so call them:

&.vOpcfnrovs;, KafJV7rep8ev, J-01Jpovs;, 1TpOs; f:3opiw. These are illus-
trated from Hecataeus of Miletus and Herodotus : iv aJ 1T6;\,s;
'Ywmf ot s~ av8pw1TO eaO~Ta ~opiovmv oi7JV 1Tip 1Ia.~Aa:y6ves
(Hee., fr. 287); <f.v8pw1ro, 'Q-rrlm (fr. 299); 'Yrr,;p~oplwv oe Tripi
avOpcl>1Twv (Hdt. iv. 32; still in Dionysius Periegetes (656;
846)); Td. KaTl1'1Tep8e E~eye aKofi (Hdt. iv. 16 et saep. al.); Eeua.-
p7J8lwv 'ITpOs Bopiw olKiova, XEAt86vio, (Hee., fr. 100); t1.i/m18toia1
1TP6S e<nJfiplav Jovpiova, (X1:ppovfia,o,) (fr. 163; c frr. 108,

203, 204, 207).

It was the practice of Hecataeus sometimes, after naming a
''
people, to add a note on their manners (e.g. frr. 154, 28.4, 323,
I 335). Comparable is Aristagoras' exposition of the map of the
world to Cleomenes (Hdt. v. 49. 5-7)-and 11. 2-3 of this frag-
ment of the Arimaspea, which indeed reads like a rendering into
pedestrian verse of an extract ofHecataeus. As we have seen, the
ultimate source of the words 71'0Mpp7Jvas -rroAv/Joura.s was Homer
(if it was not from a common stock of epic formulae), but
the phrase might have been used as a ready-to-hand version
of, say, 'IT0Av1rp6fia1'0 in Hecataeus. If this were so, it would
account for the presence of Kap-ra, a word unknown to epic, and
M C

II
18 FOUNDATIONS
rarely used at all in dactylic verse (I note Callimachus, fr. 608,
and uncertain readings in Empedocles (fr. 4. 1) and another
fragment of Callimachus (75. 6)), but common in Ionic prose.
However, all these 'technicalities' can be paralleled piecemeal
from Homer, with the exception of 8ovpos (1r6>.is ep611wv
&.v(}pcfJ1TWV ll. XX, 2 I 7 ; Ka.(}6111:pee Xloio Od. iii. I 70; 11pos Boplao
1rrm N6'rov Od. xiii. 1 ro). In a poem more concerned with topo-
graphy one might expect them to be more concentrated, and
this very fact might have helped to cause their frequency in the
Ionian geographers, who would have drawn on the Arimaspea.
The style of these fragments is discussed with sympathetic dis-
cernment by Frankel and by Bowra. But however strikingly he
uses them, the author's choice of words seems to me tame: though
borrowing the language of epic, he avoids the grander and rarer
words10 and compound adjectives (except for the Homeric
1ro>.&pprivas 1ro>.vf]orn-as); repetitive: he uses lxi:iv three times in as
many lines in fr. 7 ;11 and tasteless: the choice of the epithet
xaplam is inept in a description of fierce and ugly savages (the
line is perhaps the somewhat unhappy child of a marriage of
7[ XVl.
.1., 798 alV\
~"' avopos
~ \ (} 00 KO.p'Y'f xapii:v TE /J,ETW1TOV WI"th H es,
I I I '

Theog. 143 ovvos o' drf,Ba>.os .laa'f) lvlKETO ETW1T'l)). u. A

similar tastelessness, of thought in this case, is shown in the
description of sea-sickness. In a high proportion of the lines the
thought is completed in the first half, the remainder of the verse
being filled with repetitious padding: fr. 7. r Oav' ~,tv i<a~
TOV'TO (lya cf,pi:a~v ~JJ,ETJP'(JC1V) 2 avSpr:s vSwp valova,v ((l1T(~ xeovos
> \ ~
EV 1TEl\0.')1UaL 3 OVl1"T7)VOL TVES e,aiv
I ) I I I H I)
O ("
exovai yap epya 1TO'JJ1'/pa
\

5 'I 1TOV 7TON\O. EOLUL ("'

.J: \ \ \ (} ~ I\ > \ ~ >I ) .,
y,L/\a, ava 'X,Etpa.S E'X.OV7'S EIJ'X.OVTa& , ,
fr. 4. 3 ,1/Jvewvs fo1roiat ('TToMpprivas 1To>.vfJ01fras). Though only in
the last case is the second half of the line filled with a phrase
which could be called formulaic, yet this phenomenon might be
an indication of oral composition; but it might argue no more
than lack of artistry. If these lines are in fact spurious, they must,
in my opinion, at least precede the Alexandrian era, for there is
no trace here of Alexandrian sophistication and polish.
To sum up the results of this examination: what appears to
be a reminiscence in Xenophanes is the strongest evidence that
 ,.
FOUNDATIONS 19 ii
:,,
these fragments were composed at the latest about 465 B.c. On
~
the other hand there are indications of a later date in the extended
use of CT'Tt{:Japos, the unqualified 1re>.&.yeuai, and perhaps the geo- '
'.,

graphical terms (together with KO.JYTa) in fr. 4. But Aristeas may

have been using the language of epic in a fresh and unconven-
tional (perhaps colloquial?) way (cf. Bowra 8), and it is not by
any means impossible that his poem helped to form the usages
of the Ionian geographers later. Much more compelling evidence
of spuriousness is fr. 7, ifit purports to give the description by an
Issedonian of the wretched life of Greek mariners. To use a
barbarian as a mouthpiece for a sort of burlesque of an impor-
tant Greek calling is out of keeping with what we know of the
Greece of the seventh century B.c., and with what we shall other-
wise discover about the nature of the Arimaspea.
I should not care to pronounce a final verdict, though I think
the balance of the evidence here considered is rather against
genuineness. But of this much I feel fairly certain: even if these
fragments are not the very words of Aristeas, frr. 3-5 are based
on him, through the mediation of Hecataeus or some Ionian
geographer (apart from the 'technicalities', the unusual but
warranted form 'la07Joot, which a forger would be unlikely to
invent, vindicates this). Their matter, then, at least will still
deserve attention.
 II

KNOWLEDGE OF THE ARIMASPEA IN

ANTIQUITY

IF the fragments discussed in the previous chapter are genuine

they are to be dated in the seventh century B.c. ; if spurious, in
the fourth or very early third, before poets had come to form
their taste according to those canons which are called Alexan-
drian. It is most improbable that a forged Arimaspea would be
produced while the genuine work was still extant, and it was
still extant in the time of Herodotus.
The fourth century was not a period when literary forgery
was rife. Heraclides Ponticus and Dionysius Metathemenus (or
Spintharus) wrote plays which they ascribed, the former 1 to
Thespis, the latter to Sophocles (D.L. v. 92). Yet such was not
common practice at that time, as we are told it was at the end of
the sixth and in the early fifth centuries, when Onomacritus and
the early Pythagoreans are charged with having perpetrated many
forgeries: Onomacritus in the name of Orpheus and of Musaeus;
Lysis and Hippasus in the name of Pythagoras; Pythagoras him-
self in the name of Orpheus (Suid. s. 'Oprf,Evs A.Ei{J71Bpwv; Clem.
Strom. i. r3r; D.L. viii. 7-8). Whatever we think of these charges
and the many others like them, compositions passing as those of
Orpheus were certainly current in the fifth century (cf. Eur.
Hipp. 953 f.; Ale. 966 ff.; I.A. 796 ff.), and must have been
written by someone. But the heyday of literary forgery dawned
with the Hellenistic era, when the eagerness of the Ptolemies and
the Attalids to acquire books for their new libraries led to a spate
offakes. 2
Literary forgeries can be divided into three classes: (i) Works
ascribed to fictitious or legendary or even divine characters,
such as Sanchuniathon, Chlonthachonthlus (Rohdeii 235),
Orpheus, and Hermes Trismegistus, some so-called but probably
 KNOWLEDGE OF THE ARIMASPEA IN ANTIQUITY 21

imaginary 'early Pythagoreans', and oracles purporting to ema-

nate from such figures as Abaris, Bacis, or a Sibyl. (ii) Works
ascribed to historical personages, but with new titles-new,
in the sense that there is no independent reason to suppose
that the people to whom they were ascribed ever in fact wrote
works thus entitled. These are very common. The motive for the
false ascription might be convention or respect, as when poems by
Homeridae or Hesiodei, or treatises by members of the philo-
sophical schools, were allowed to pass under the name of the
founder ('forgery' is hardly the proper word for these); or it
might be a desire to obtain a better reception and closer attention
for the work by giving it the cachet of a venerable name (this was
presumably always the motive for those in class (i)) ; or financial,
as in the cheating of the Hellenistic kings; or sometimes perhaps
even sheer devilment (Hippasus was said to have written the
.vcrriKos .:\oyos in order to bring its supposed author, Pythagoras,
into disrepute, and Dionysius seems to have passed his Partheno-
paeus off as Sophoclean for a joke. Heraclides himself when he
forged Thespis-perhaps only 'quotations' from him-must
have been either inventing evidence for his theories or hoping
to indulge in that purely aesthetic pleasure, found at times
among antiquaries as elsewhere, of 'getting away with it').
(iii) Works ascribed to historical personages, bearing titles known
to have belonged to actual works by those personages. This is
the category germane to our present study, but e.xamples of it
are extremely rare-not surprisingly, as a forger would have to
feel sure, before embarking on his task, that the genuine work
was utterly lost, so that there was no danger of its turning up
again. As a natural corollary of this most forgeries of this type
are relatively modern productions, e.g. that of Cicero's Con-
solatio, and the '1>0,vi1<id of Philo Byblius.3 In fact, I cannot
adduce an ancient example. If Hecataeus' I'ijs 1rEp{o8os, in the
fragments we possess, were a forgery of the third century B.c.,
as J. Wells argued, it would provide one; but this view is not
now generally accepted (Pearson 31 f[).
The chances are, therefore, that an Arimaspea was never forged
in a complete form : the most that is likely to have happened is
 a11 KNOWLEDGE OF THE ARIMASPEA IN ANTIQUITY
that some fourth-century author forged 'quotations' from it for
his own purposes, in the way I have suggested that Heraclides
may have forged 'quotations' from Thespis. This consideration
indicates another way to the solution of the question whether
the fragments are genuine; for if the Arimaspea can be shown to
have been still extant in the Hellenistic era or later, that will
be a strong argument for accepting their authenticity.
Clearly the first task here is to consider the contexts in which
the quotations are preserved, to see if 'Longinus' or Tzetzes in-
dicates whether his source is the poem itself. The answer of the
latter is explicit: they were merely a few lines of Aristeas that
he had happened to come across [T. 3: ov,TEp ath-os .Jv E'ITmv
lvlrvxov oMyoiS']. Yet even ifTzetzes himself had not told us this,
we should have been able to deduce it by means of a touchstone
which can be applied elsewhere too with significant results.
In the passage of the Chiliades in which the lines from the
Arimaspea appear Tzetzes is speaking of the Hyperboreans. They
are mentioned, he says, by Zenothemis and Pherenicus as well
as by Aristeas, and he quotes from these authors in support of his
statement verses which in fact do not appear to substantiate it:
in the lines of Pherenicus Hyperboreans do occur, in those of
Zenothemis and Aristeas they do not, though Issedonians and
Arimaspi do. The reason for this apparent anomaly is that
Tzetzes thought that the Arimaspi were Hyperboreans, following
a tradition that was long-standing but not derived from Aristeas
-for his Hyperboreans, as we know from Herodotus, were
separated from the Arimaspi by the griffins (and also, as will be
seen, by the Rhipaean mountains [39]).
This passage of Pherenicus (a poet of uncertain date, but
earlier than Athenaeus, who quotes him) appears again at
greater length in a scholiast on Pindar ( Ol. iii. 28) :
> "J.' (J>
al"'/', 'Y:m,pf3 opEwv, o, T l EC1X11Ta vatETaOVC1W
I tl tt f

V'f/'f w' }1,roAAwvo!i', d.'TTElp'YJTOi 1ro'M.oio.

TOU!i' iv If.pa. 1T(JOTtpwv Jf alt14TOS v.velovow
n"JVWV f3)11J,C1ToVTa!i' ~1rip (Voss: thr&) 8p6.ov al8p1]EJl'Ttl
v&.aaa.qBru Bopta.o yth)V i1pl.a.(17Tov lf.va1<Ta.

' , , and about the Hyperboreans, who inhabit the edge of the
 KNOWLEDGE OF THE ARIMASPEA IN ANTIQUITY
world close by the temple of Apollo, and know not war. They,
poets sing, sprung from the blood of the Titans of old settled
beyond the limpid course ofBoreas, [?begging] their acres of king
23

I
Arimaspus . .'. However the sense is to be completed (with
Aiuuohovs v.s. ?), here Hyperboreans and Arimaspi are closely
conjoined (a Scythian king Arimaspus is mentioned by Diodorus
Siculus (ii. 43) ; he is clearly a creature ofpoetic fancy). Stephanus
Byzantius calls the Arimaspi a 'Hyperborean tribe', but far
earlier than this they are Hyperboreans for Callimachus: those
who first carried the corn offerings to Delos (Hyperboreans,
according to Hdt. iv. 33) brought them 'from the fair-haired
Arimaspi' (H. iv. 291)-called Wvos Twv 'Y11ep/JopJwv by the
scholiast:
1TpW'Tal TO' Tab> 6',KQ.11 d1T<) ta.118w11 )1p,p.a.tnrWII
O&rrts T Ao[w T Ka, eva.low 'EKalm,
Bvya.Tlpes Boplao.

So too in Aetia, fr. 186. 8 'the sons of the Hyperboreans escort'

these offerings 'from the Rhipaean mount', and the men of
Dodona 'are the first of the Greeks to receive them from their
Arimaspian convoy' :

Though Callimachus is the earliest author who we can be cer-

tainS made the identification, Stephanus (s. 'Y11ep/36peo,) asserts
that it was made by Antimachus. Ifwe accept Stephanus' text-
and there is no adequate reason for emending to Ka).)..tp.axos
with Ruhnken-we still cannot tell which Antimachus is meant.
The poet of Colophon is perhaps the likeliest candidate, which
would take the confusion back another century and a half. The
'Homeric' Epigoni, in which the Hyperboreans were indeed men-
tioned (Hdt. iv. 32), was said to be by an Antimachus; but if this
were he of Teos6 and he were reallycontemporarywith the founda-
tion of Rome, he would, of course, be too early to know about
Arimaspi. Yet that the blurring of the distinction between the
124 KNOWLEDGE OF THE ARIMASPEA IN ANTIQUITY
Hyperboreans and their horse-keeping, nomadic neighbours
began very soon after the appearance of the Arimaspea is shown
by a fragment of the Hesiodic Catalogus, where they are called
'the well-horsed Hyperboreans' (['Y]1rep,8oplwv d.1l71'7Twv: fr. 49
T. = Pap. Ox. xi. 1358, fr. 2. 21) [189].
The amalgamation of the Arimaspi with the Hyperboreans
seems to indicate that men were forgetting the poem that first
had introduced the former to the Greeks. The supersession of the
tradition of the Arimaspea would have been helped by the interest
which Utopias held for fourth-century prose writers. The desire
to give these ideals a local habitation was one factor in the birth
of the 'philosophical romance'; but the Atlantis of Plato and the
Meropis of Theopompus (or his source) had their prototype in
the blessed Hyperboreans, whose location had long been fixed.
It is not therefore surprising that an author of this period wrote
an entire treatise on them.
This was Hecataeus of Abdera (.ft. second half of the fourth
century B.c.), who, in his TTEpl 'Y.,,Epfiopewv, let his imagination
run riot in an exuberance of detail. Not only did he describe the
customs of the people, hut he added a mass of topographical cir-
cumstance to give the ring of truth: the frozen sea of Amalcius;
the island ofElixoea inhabited by Hyperboreans called Caram-
bycae after a neighbouring river; another river, Parapanisus,
flowing into the northern ocean; the city of Cimmeris. Jacoby
(iii, note on 264 F I I) discusses the derivation of these names,
and justly remarks that Hecataeus' powers of invention were not
great. But it is interesting to find him subdividing the Hyper-
boreans, and giving names to the subdivisions. The Carambycae
constitute the only sample of these names we know from
Hecataeus; whether the Arimaspi were for him Hyperboreans
cannot be said, but to include their neighbours with the latter
would be an easy step in the process of provjding a detailed
Hyperborean ethnography, which perhaps Hecataeus started
and others developed (Hierocles spoke of H yperboreans called
TapKv11a'io,-named after Ta 14pKvv,a DPTJ (Arist. Meteor. i. 13) ?-
in whose country were the gold-guarding griffins: Steph. Byz. s.
TapKvvla).
 KNOWLEDGE OF THE ARIMASPEA IN ANTIQUITY
These considerations destroy the value of the entry in 'Suidas'
[T.u]: 'Aristeas: son of Democharis [204] or Caystrobius, from
25
I
Proconnesus, epic poet. Author of the hexametric poem called
Arimaspea ('Tct. J4.pi.d.cnma Ka.AoV.Eva l1rTJ); this is an account of
the Hyperborean Arimaspians, in three books. . . . He also
wrote a prose Theogo191, running to about a thousand lines (els
l1TTJ ,a)! At first sight this would seem to confirm the survival of
the Arimaspea into the Alexandrian era, its division into books
and cataloguing; but the statement that it was 'an account of
the Hyperborean Arimaspians' is enough to show the untrust-
worthiness of the testimony. It is guesswork: 7'0. J4ptp,a.cnma
KaAov&a l1T7J is an echo of Herodotus (T. 12: 7'ct. l1tEa 'Taiha
7'0. vvv {m' 'EM.,,vwv l1pi.au1Ta Ka.ATat), and the bald summary
of the contents a conjecture based on the title by someone to
whom the identification of the Arimaspi with the Hyperboreans
was the familiar tradition. 7 It may be added that we can discern
enough of the actual contents of the poem to be sure that it con-
tained much else besides an account of the Arimaspi; though this
account must have caught the imagination of early hearers
sufficiently to give a name to the whole work, as the slaying of
Dolon gave the name Dolonea to what is now the tenth book of
the Iliad.
Who was responsible for concocting the lie that 'Suidas' has
perpetuated? According to W. Cri:inert, following Hiller, it was
Lobon of Argos, an even mistier figure than Aristeas himself. We
know of his existence only from two mentions by Diogenes
Laertius, and anything that we say about him must be deduced
from such fragments as we are prepared to assign to him. He was
not, Gronert assures us, the evil-Ininded forger that Hiller had
depicted, peddling false scholarship in the hope that men would
take it for true, but a harmless joker, who would have doubled up
with laughter had he found later generations taking his inventions
seriously ( 128). Let us call the culprit Lobon, if that is more
satisfying than to call him x; but I cannot believe that he preceded
Callimachus ( I 30: 'adhibuit Lobonem praeter Hermippum
nemo Alexandrinorum, quod sciam. aspernatus est ut par est
Callimachus', &c.), for such ajeu d'esprit as his must presuppose
 (
of
,-.\
26 KNOWLEDGE OF THE ARIMASPEA IN ANTIQUITY . t,

not only the bibliographical activities of the Alexandrian scholars

(especially the division of longer works into 'books') but also
common knowledge of them. Against this, Cronert's main argu-
ment for an earlier date (Lobon's use of l1ros instead of
<rrlxos to mean a line of prose (127)) does not have sufficient
weight.
The importance of the extract of 'Suidas', then, is to prove the
opposite of what it appears to prove: it indicates that the
Arimaspea was not possessed by the cataloguers of the library at
Alexandria (at least, not by the earlier ones), thus confirming the
conclusion already foreshadowed by Callimachus' identification
of Arimaspi with Hyperboreans.
The fact that it was not extant in Alexandria in the third
century B.C. would not necessarily mean that it had disappeared
from human ken, though in view of the zeal with which the early
Ptolemies collected books for their library the probability would
be that it had. But the poem might have turned up later. Is there
any evidence that it did?
For an answer to this question it is natural to go first to our
other source of professed quotation from the Arimaspea, 'Lon-
ginus', but unfortunately we get little help from it. In the context
'Longinus' is remarking how the great poets pick out the most
striking circumstances in the situations they are describing, for
example Sappho describing the symptoms of one in love, and
Bomer the dangers of the storm-tossed. This last is contrasted
with the lines from the Arimaspea and the climax of a passage of
Aratus remarking the hardihood of sailors who brave the rough
sea at all seasons of the year (Phaen. 299). The comparison of
Homer with the others is introduced thus: 'It is, I fancy, much
in the same way that the poet in describing storms picks out the
most alarming circumstances. The author of the Arimaspea, to be
sure, thinks these lines awe-inspiring ... ' (trans. Hamilton Fyfe).
From this it appears that 'Longinus' imagined all three of his
examples to depict sailors in storms. I have already mentioned
the difficulties involved in thus interpreting these lines from the
Arimaspea [g] : if they are in fact about sailors they are probably
not genuine; but if they are not about sailors it would seem
 ..Ii'.
KNOWLEDGE OF THE A.RIMA.SPEA IN ANTIQUITY 27
that 'Longinus' did not know them in their context but only as
l'

a quotation. 8
Why did 'Longinus' not name Aristeas here, but merely speak
of 'the author of the Arimaspea'? He could hardly have been
ignorant of who the author was reputed to be, for he was well
acquainted with his Herodotus: had he forgotten-which would
suggest both that he got the lines at second-hand and that Aristeas
was not a well-known figure at that time? Or did he use this
fonn of words to express doubt of that authorship (in which case
we might rather have expected him to be more explicit, as he is
over the Hesiodic authorship of the Shield (9. 5))? Or is it just
a figure of contempt?
That the Arimaspea survived into Roman times seems so un-
likely on the evidence so far considered that it comes as something
of a shock to hear of a copy of Aristeas being picked up cheap in
the second century of our era. The event is related by the pur-
chaser, Aulus Gellius (T. 9] :

We had disembarked, on our way back from Greece to Italy, at

Brundisium, and were strolling through that famous port, which
Ennius has described by the unusual but exceedingly appropriate
epithet 'well-favoured' [praepetem: cf. vii. 6. 6), when we espied some
bundles of books for sale. I hurried eagerly across to them. They
proved to be all Greek books full of tales and marvels, unheard-of
things that passed belief, by old writers of no small authority-
Aristeas of Proconnesus, Isigonus of Nicaea, Ctesias, Onesicritus,
Philostephanus, Hegesias ; the very volumes were disfigured with the
mould of ages, and as filthy to the touch as they were to the sight.
Nevertheless I inquired their price: it was astonishingly low, and as
a result I bought a great quantity of books for a tiny outlay.
I spent the next two nights in running quickly through them, e.x-
cerpting and summarizing a number of curious items which Latin
authors have generally neglected ; and I have introduced them into
this work, that my readers may not, through ignorance, be caught
entirely at a disadvantage in discussions about such topics.

Tournier (52), taking this testimony at its face value, thought

that Gellius had lighted upon an actual copy of the Arimaspea.
28 KNOWLEDGE OF THE ARIMASPEA IN ANTIQUITY
Unfortunately it is not so. What manner of work it was is revealed
by Pliny (N.H. vii. 9-26), who anticipates the extracts of the
later writer in a fuller form, and, unlike him, assigns them to
their respective authors. I set the two texts side by side, italicizing
close similarities of language :

Pliny, .N.H. vii. g ff. A. Gell. ix. 4. 6 ff.

Esse Scytharurn genera et Scythas illos penitissimos, qui
quidcm plura, quae corporibus sub ipsis septentrionibus aetatem
humanis vescerentur, indicavimus ... agunt, corporibus hominum vesci
(ro) sed iuxta eos, qui sunt ad eiusque victus alimento vitam
septentrionem versi, haut procul ab ducere et d.v6pw1rocp&:yov, nomi-
ipso aquilonis exortu specuque nari;
eius dicto, quern Jocum Ges cli-
thron appellant, produntur Ari- item esse homines sub eadem
maspi, quos diximus, uno oculo in regione caeli unum oculum infrontis
Jronte media insignes. quibus ad- medio habentes, qui appellantur
sidue helium esse circa metalla Arimaspi, qua fuisse facie Cyclopas
cum grypis, ferarum volucri poetae ferunt;
genere, quale vulgo traditur,
eruente ex cuniculis aurum, mira
cupiditate et feris custodientibus et
Arimaspis rapientibus, multi, sed
maxime inlustres HERODOTUS et
ARISTEAS Proconnesius scribunt.
(II) super alios autem Anthropo-
phagos Scythas in quadam con-
valle magna Imavi montis regio alias item esse homines apud ean-
est quae vocatur Abarimon, in dem caeli plagam singulariae
qua silvestres vivunt homines velocitatis, vestigia pedum habentes
aversis post crura plantis, eximiae retro porrecta non, ut ceterorum
velocitatis, passim cum feris hominum, prosurn spectantia;
vagantes. 9 hos in alio non spirare
caelo ideoque ad finitimos reges
non pertrahi neque ad Alexan-
drum Magnum pertractos BAETON
itinerum eius mensor prodidit.
(12) priores Anthropophagos,
quos ad septentrionem esse
diximus, decem dierum itinere
supra Borysthenen amnem ossi-
bus humanorum capitum bibere
cutibusque cum capillo pro man-
telibus ante pectora uti rsraoNUs 70 praeterea traditum esse memo-
Nicaeensis. idem in Albania gigni ratumque in ultima quadam terra,
quosdam glauca oculorum acie, quaeAlbania dicitur, gigni homines,
 KNOWLEDGE OF THE ARIMASPEA IN ANTIQUITY rzg
a puentia statim canos, qui noctu qui in pueritia canescant et plus cer-
plus quam interdiu cernant.U idem nant oculis per 110ctem quam interdiu;
itinere dierum XIII supra Bory- item esse compertum et crediturn
sthmm Sauromatas tertio die cibum Sauromatas, qui ultra Borysthenen
capere sernper. 1z fluvium Ionge coltu1t, cibum capere
semper diebus tertiis, meclio abs-
tinere. (7) id etiarn in isdern libris
scriptum offendimus, quod postea
in Iibro quoque Plinii Secundi
naturalis historiae septirno legi,
(16) in eadem Africafamilias quas- esse quasdam in terra Africa hominum
dam e.ffascinantium ISIGONUS et NYM- f amilias voce atque lingua ejfa-
PHODORUS, quorum laudatione scinantium, (8) qui si impensiusfarte
intereant probata, arescant arbo- laudaoerint pulchras arbores, segetes
res, emoriantur irifantes, esse eiusdem laetiores, infantes amoeniores,
generis in Triballis et Jlryris adicit egregios equos, pecudes pastu
ISIGONUS, qui visu quoque effascinent atque cultu opimas, emoriantur re-
interimantque quos diutius intueantur, pente haec omnia nulli aliae
iratis praecipue oculis, quod eorum causae obnoxia. oculis quoque
malum facilius sentire puberes; exitialem f ascinationem fieri in
notahilius esse quod pupillas binas isdem libris scriptum est, tradi-
in oculis singulis habeant. turque esse homines in Illyriis, qui
interimant videndo quos diutius irati
viderint, eosque ipsos mares femi-
nasque, qui visu ita nocenti sunt,
pupillas in singulis oculis bmas
(22) in monte, cui nomen est habere.
Nulo, homines esse aversis plantis
octonosdigitosinsingulishabentes
auctor est MEGASTHENES; (23) in
rnultis autem montibus genus homi- (9) item esse in montibus terrae
num capitibus caniriis ferarum pelli- Indiae homines caninis capitibus et
bus velari, pro voce latratum edere, latrantibus, eosque vesci atium et
unguibus armatum venatu et aucupio ferarum venatibus;
vesci: horum supra centum viginti
rnilia fuisse prodente se CTESIAS
scribit, et in quadam gente Indiae
ferninas semel in vita parere geni- atque esse item alia aput ultirnas
tosque confestim canescere. idem orientis terras miracula homines,
hominum genus, qui Monocoli voca- qui 'monocoli' appellantur, singulis
I'. rentur, singulis cruribus, mirae pernici- cruribus saltuatim currentes vit1acis-
tatis ad saltum; eosdem Sciapodas simae pernicitatis;
vocari, quod in rnaiore aestu
humi iacentes resupini umbra se
pcdum protegant. 1 non longe eos
a Trogodytis abesse, rursusque ab
his occidentem versus quosdam sine quosdam etiam esse 11ullis cervicibus
cervice oculos in humeris habentes. oculos i1I h11meris habentes.
30 KNOWLEDGE OF THE ARIMASPEA IN ANTIQUITY
(25) MEGASTHEN'ES gentem inter
Nomadas Indos narium loco fora-
mina tantum habentem, anguium
more loripedem, vocari Sciratas.
ad extremos Jines JndiOI ob oriente ( I o) idem illi scriptores gentem esse
circa fontem Gangis Astomorum aiunt aput extrema lndiae, corporibus
gentem sine ore, corpore toto hirtam hirtis et avium ritu plumantibus,
vestiri frondium lanugine, kalitu nullo cibatu vescentem, sed spiritu
tantum viventem et adore, quern nari- jlorum naribus hausto victitantem;
hus trahant, nullum illis cihum, nul-
lumque potum, radicum tantum
florumque varios adores et silvestrium
malorum, quae secum portant
longiore itinere, ne desit olfactus;
graviore paulo odore haut diffi-
culter exanimari. (26) super hos
extrema in parte montium Tri- Pygm01osquoque haut longe ab his
spithami Pygll1lique narran tur, nasci, quorum qui longissimi sint
temas spithamas longitudine, hoe non longiores esse quam pedes duo et
est ternos dodrantes, non excedentes. quadrantem.

Clearly the source for both these passages was the same : the
authors named by Gellius in his introduction all appear among
those cited by Pliny (Onesicritus in this chapter (28); Philo-
stephanus and Hegesias 14 get a mention much later in the same
book (207)) ; the order in which the items succeed each other is
identical, as would be natural if both our authorities were sum-
marizing the same book; and the similarities of language are
remarkable-so remarkable as to suggest the suspicion that
Gellius was really using Pliny after all. But we have no right to
impute such flagrant dishonesty to him, and it is discounted by
the fact that he mentions his later discovery of the Plinian
passage (7). When he goes on to quote the older writer (13 ff.)
he makes proper acknowledgement. There is in fact actual proof
to the contrary, which also proves that the common source was
not a Latin work of any kind (lest there be those who would thus
explain the resemblances oflanguage) : in their accounts of that
delicate Indian tribe whose sole sustenance is the scent of flowers
Pliny and Gellius diverge uniquely on a point of fact, the former
saying that they clothe themselves in soft leafage (vestiri frondium
lanugine15), the latter that they are feathered like birds (avium
ritu plumantibus); a divergence which is most easily explicable on
 KNOWLEDGE OF THE ARIMASPEA IN ANTIQUITY 31
the assumption that one word in a Greek text has been confused
with another (e.g. ,,,,,.Di.ov with 1rlTa.'>..ov). This is reinforced by
the consideration that, though there is no disagreement about
the height of the Pygmies, it is expressed in different ways (ternos
dodrantes PI. : pedes duo et quadrantem Gell.-both interpreting
1'ptemlBa.fLO,). Finally, there are the Greek proper names, some-
times transliterated, sometimes not-Anthropophagi, Ges clith-
ron, Monocoli, &c.-and the hint of a Graecism in semper in the
expressions tertio die cibum capere semper / cibum capere semper diebus
tertiis {for Sil Tpl'T"f]'; 71lpa.s ad 0'TOWTCJ. v.s. ?) .
We may conclude, then, that the source we are seeking was
a Greek prose work concerning 1ra.pa.8ot0Aoyoveva 1rpt JOvwv: a
mere compilation from various authorities who were named, after
the manner of the extant 10"Topla., 8a.va0'a of Apollonius. The
terms of its date are Isigonus and Pliny: the latter half of the first
century B.c. or first half of the first century A.n. 16 There is clearly
no reason to assume that its author excerpted Aristeas at first-
hand: he may have owed this information to a succession of
previous writers. Whether there is here any testimony indepen-
dent of Herodotus is a question to be discussed later [64 f.; 93].
That there was in existence in the first century B.c. a prose ac-
count ofScythia passing under the name of Aristeas was suggested
by Hubmann (Tournier (32) disagrees with this). This view is
founded upon a passage of Dionysius of Halicarnassus [T. 1 o]:
All the historians who preceded Thucydides used some of these
stylistic devices. Now what sort of style was affected by the really
ancient authors, of whom nothing is known but their names, I am
unable to conjecture .. for in most cases their writings are no longer
extant, and those which are are not universally accepted as genuine;
instances are those of Cadmus of Miletus, Aristaeus the Proconnesian,
and the like.
The Arimaspea cannot possibly be meant here: of course Dionysius
is speaking of prose authors, as we should expect in one discussing
the antecedents of a prose author, and as is confirmed by his
mention of Cadmus. But what prose work of Aristeas could he
have had in mind? We hear of none such except the Theogony,
and that solely in the suspicious entry in 'Suidas'. It may never
 3~ KNOWLEDGE OF THE ARIMASPEA IN ANTIQUITY
have existed except as a title. Eudocia's statement [T. I 1] that
it was 'a fine piece of work' may be discounted as evidence of its
existence; it is her own editing of her source, Hesychius of
Miletus. 11 If it did exist it was certainly not genuine, for such an
ascription would only have been possible after Aristeas had
acquired the reputation of a Oeo'>o.oyos, like Abaris, Epimenides,
and Pherecydes, to whom also Theogonies were ascribed. If there
was in fact such a work, to which Dionysius is here referring, our
view of Lobon, or whoever he was, will have to be modified
accordingly: he was either an astonishingly industrious writer
if he composed all these forgeries himself, or a more truthful
reporter than has been credited if he found them already in
existence. But in fact Dionysius' meaning is ambiguous; it is not
clear whether he is instancing the writings of Cadmus and
Aristeas as works current at the time but of doubtful authenticity,
or as early prose works that had not survived. If the latter, then
he may have classed Aristeas as an early prose author because he
knew that he had been credited with a Theogo'!)l. 1s
Strabo, who says that Homer may have had the idea for his
Cyclops from Aristeas' description of the one-eyed Arimaspi,
probably owed this knowledge of the poem to Herodotus, of
whose language there seems to be a reminiscence in a later
reference to Aristeas as 'the poet of the so-called Arimaspea',
o 1TOL1Jrf/S 1'WV J.J.p,aa1rdw11 KaAoulvwv Jrrwv [T. 7; 25].
Only one more author, Pausanias in the second century A.D.,
gives the appearance of using the Arimaspea direct. Describing
the statue of Athena in the Parthenon he says that her helmet
had in its centre a representation of the Sphinx, flanked on both
sides with griffins. He goes on [T. 4] :
Aristeas of Proconnesus says in his poem that these griffins fight for
the gold with the Arimaspians who dwell beyond the Issedonians, and
that the gold which the griffins guard is produced by the earth. He
says, too, that the Arimaspians are all one-eyed men from birth, and
that the griffins are beasts like lions, but with the wings and beak of
an eagle. So much for the griffins (trans. Frazer).
These words have a Herodotean ring (especially the last, ypVTTwv
' I ~ >IL) f.Hdt'
P,EV 'TTEpi -rouawa '"P'fJUVW : C lV. 15. 4 n.pt<rTEW
A I I
P,EV JIVII
 KNOWLEDGE OF THE ARIMASPEA IN ANTIQUITY 33
I A
1tEp T0'1avra Eipr,a
t I ew;
36. I KO Tatn'a J"EV 'Y:7TEp,-,opEWV
Q I
'ITEpt
A ' I f

lp~u8w; and cf. ovocf,O&A,ovs 11a.VTas EK ')IEVE-rijs with Hdt. iv.

23. 2 116.v-res ,f,aAaKpo'i EK yEvefjs), and the source of much of the
passage could have been Herodotus. But it also comprises matter
which is not from him-the statement that the gold is the yield
of the earth, and the description of the griffins. However, it is
clear from Pausanias' remark in the course of a digression on the
Hyperboreans that he was not directly acquainted with the
Arimaspea [T. 6]: 'Aristaeus of Proconnesus, who also mentions
the Hyperboreans, may perhaps have learned something more
about them from the Issedonians, to whom he says in his epic
that he came' (trans. Frazer). The gist of all this is in Herodotus;
even the tell-tale guess is suggested by him (iv. 32: 'Y1repfiopiwv
~ I > 8
f I 't" '8 \ >I/ " I I >f ' -"\ \
O( 7TEpt Cl.JI pW1TWV O'IYTE 'n ""KV (U I\EYOV0" OVOEV OUTE TWES' (l./\/\0t
,... .t 't
'TWV 'TO.V77) 0K'T}J1,EVWV, :r1\ a.pa
.,, 'J,0'07/00VES.
I ~ I ' ~
WS O E')"W OOKEW, t ~:,
ovu f ~j ' ' '

OV'TOt Myovut ovUv. EAE')'OV yap a.v KUl J:1<v8a.,). An author who
knew the poem himself would not have served up this jejune
rehash of someone else.
Neither Pausanias' present (cpryalv, T. 4) nor Strabo's perfect
(iv8i8w1<Ev, T. 7) proves the contemporaneous existence of the
poem or acquaintance with it. There is a tendency to repeat the
tense of one's authority, a tendency that is natural, and some-
times perhaps a little disingenuous. This is illustrated well by
Pliny N.H. vii. 9 ff., which, as I have shown [go], is not derived
at first hand from the authors it quotes, in spite of appearing to
be so by the almost invariable use of the present (Agatharchides
scribit (14), refert Clitarchus (30), &c.).
In 1893 C. von Holzinger published, in ,Zeitschrift fiir die
iisterreir:hischen Gymnasien (xliv. 394), the text of an idyll by
Maximus Planudes, a fellow citizen of John Tzetzes though a
century later in date. Humorous in intent, this poem owes its
form and much of its phraseology to Theocritus. Thamyras opens
by complaining to another bumpkin, Cleodemus, that this is the
first time they have seen each other since they met in the house
of a mutual friend, Aristaeus, at a festival of Demeter (I. 8):
>f; f > > '
I >,t I 1 \ I
E:1, or ev, .eyapotmv .e1pur-rawu va,u?]CT'V
.. t , Al ~ , t1
'l')f.1,f:Vo, ,7epioi uT/fLYJ7pos E'Kpwowua.
6366 J)
 34 KNOWLEDGE OF THE .A.RIMASPEA IN ANTIQUITY
Cleodemus then relates a tale of troubles that have befallen him
recently. Having lost an ox he went to Aethra, the town of Zeus
(Js t:ycO..ov ..:hos Ai8p17v), to buy a replacement. This town was
situated on the peak of Olympus, 'near the clouds, veiled from
sight by the snow-flakes of Boreas' (l. 66: o.rxwEc,s x,ovEO'O't
KMV1TTOJ.LEVTJ f3opiao). When he arrived a festival to Zeus was
being celebrated with games and music; and in the midst of
the celebrations there came an Egyptian magician, a wild man
with bestial features, black all over, with straggling beard and
long, snaky locks darker than a raven's feathers (1. 78) :
I>\ I 1,l.l I> I I
A
7010'1 01; 'T<;fYITOfUVOtUW 'f'tlC"70 oatp.ovtos 'TIS
a.ypws &.vtJp 971pos ;xwv wa., 8r7POS O'ITW7M1V
I ,l.. \ft.I.\ \ I > I I
a.'1'""""1'7JS 1r1101<:a..ow,v a.a.vpoTti.po,u, Kopaxwv
oZ'ov Jx,8va.lnu, TtTawo,lvo,s 1rMov ovp'fjs
Ka.l u1<0A1fj, J.\t,ceuu, D,tuuolvo,s V'TT~p ci'i.wv,
p.7JK<;8a.vov TO ylvewv cbro UTOJLO.TWV 'ITpoLa.AAWV
' ,, )!,/, , , ' '8
Ka.t p.e11as ""'f'ea 1raVTa., 1rpouW'1Ta. XEpa.s Tti. 1ro as 'TE.

With him was a boy carrying a large sack on his back.

The magician addressed the assembled crowd, and said he
wanted to know what Zeus was doing at that moment. He then
produced from the sack two apples, one of which he threw into
the sky so that it disappeared. When it failed to return with news
of Zeus he threw its companion after it, and when this too failed
to return he threw the boy up. As the boy himself did not re-
appear, the magician told the crowd that, on his repute as a sooth-
saymg seer . 144: -r, E')'W
(] > \ >,/, ~ \
H '
a.,,f.VfJ'TJS T11t:vw
1/'j \ I >' ()"JS ,
,ea., aJ/'1'i: a11TJ I )

the gods were celebrating the marriage-feast of Ares and Aphro-

dite. As he said this the boy returned from heaven with the
apples in his pocket and holding a leg of chicken (I. 151:
ot:!i-rt:pfi o~ opvifJo-;; 1168a a.\ov), which he said Zeus had given him,
together with greetings to the magician, 'the best of Zeus' servants'
(I. 173: a.q,,11&>...wv Tov a.piaTov). Thamyras exclaims at the divine
powers of the magician : he must be a pupil and a scion of the
gods: ~ ,, 1. '''{3' ,,
oa,ov,os 'TtS e:KEtVOS' t:7111 Ka., t:'17'11 0110s aV1Jp,
afJa.VO.'TWV 'ITO.i8Evp,a. o[8a:ya n 0Vpavu/Jvwv1
8s 'TOta.irra TI.TEVXf" 9t:wv vu -rot t<17W r1'1Toppwf.
Cleodemus, however, has personal reasons for dissociating him-
 KNOWLEDGE OF THE AR/MASPEA IN ANTIQUITY 35
self from this praise ; for when the magician heard what he
wanted, he took a mouse, changed it into an ox, and sold it to
the bumpkin. But when it had been driven to Cleodemus' farm it
turned back into a preternaturally fierce mouse, which slew the
cat, invited in a multitude of other mice, at whose head it placed
itself, and proceeded to eat Cleodemus out of house and home.
To Thamyras' suggestion that they try to find the magician and
bring him to book, Cleodemus replies that he never wants to
meet him again, lest he tum Cleodemus himself into a beast, or
throw him into the sky. Thamyras then promises instead to in-
I troduce Cleodemus to a man who will sell him a 'wooden cat'-
I a mousetrap.
!
Little notice was taken of this trifle for forty years. Then, in 1936,
Th. Nissen pointed out that the original of Planudes' magician
appeared to be Aristeas, that 'trickster if ever there was one'
q
i! (T. 25: a.i,;,p 'Y07JS' et -rtS' ruoS'), That he is called an Egyptian
1;
i:
I
need deceive no one : Egypt was a conventional breeding-ground
i ofmagicians. His real identity is betrayed by his bestial appearance
and raven-black locks-a reminder that Aristeas himself was
a raven as well as a man. So too Aristeas was a great seer, .dVT,s
[x57], as is this magician. Again, though Planudes has not told us
his wizard's name, he has shown us that the Proconnesian was in
his mind as he wrote, for he calls the mutual friend of his two
bumpkins Aristaeus- a form under which Aristeas himself some-
times appears [ 205]. More: Aethra, the town of Zeus, hidden
beyond the snow-storms of Boreas, the north wind, is no market-
town on the top of Mount Olympus, but in fairyland-in the
land of the blessed Hyperboreans to be exact, which, as Pindar
says, 'cannot be approached by land or sea', but is situated in
the Olympian heaven, above the region of rain and snow (cf.
Od. vi. 42 ff.), accessible only to the disembodied soul. The pas-
sage of the boy into the sky, and his return therefrom holding a
chicken's leg, which proved he had indeed attended a banquet of
the gods, is a parody of a soul-journey into the Beyond ;just such
ajourney in the soul as Aristeas made to the Hyperboreans and
related in the Arimaspea, which Tzetzes had read not long before
Planudes wrote.
 36 KNOWLEDGE OF THE ARIMASPEA IN ANTIQUITY
That Planudes has here been much influenced by a tradition
about Aristeas the miracle-monger, whose soul visited the Hyper-
boreans (T. 20], is I think very likely. But Nissen goes too far
in suggesting that the Byzantine idyll may be a parody of the
Arimaspea. He was writing in the year following the publication of
K. Meuli's widely accepted suggestion that Aristeas was a Greek
shaman, who made journeys not in the body but in the spirit, and
the Arimaspea a 'shamanistic' poem, describing such a journey of
Aristeas' own soul. I hope to show that this view of the Arimaspea
cannot be sustained [132 ff.J. 19There may still have been even in
the fourteenth century popular oral traditions lingering on in
Byzantium about the old 'magician' from nearby Proconnesus
(cf. Nissen 299). Planudes may have known some such, or he may
have got his ideas from old literary sources; but there is no cause
to suppose that the Arimaspea itself figured among the latter. Had
it still been extant in the preceding century the case might have
been different, but Nissen is wrong is supposing Tzetzes to have
read it [ 22]. The idyll of Planudes preserves a reminiscence of
the poet but not of the poem.
There is another sphere besides the literary in which the
Arimaspea could be expected to have left an impression: that of
pictorial art. We have already seen that it probably did do so
in the case of the Kelermes grypomachy of the early sixth century
[6]. It remains, therefore, to review some other grypomachies of
classical art. 20
(i) Rhyton from Campania. A barbarian, right, clean-shaven,
bare-headed, long-haired, with one foot on a small rock and a
Scythian bow in one hand, with the other raises a sword to strike
a griffin of the classical type [89]. The griffin has sprung on to
the back of a horse facing the barbarian, and is burying its beak
in the horse's neck (Reinach i. 53). As the same scheme of a
griffin leaping on the back of its victim (this time an elk) is found
on a leather sheet of the fifth century B.C. from Pazirik in the
Altai (Rice 137), the theme is clearly both early and Greek in
origin, and may well be owed to the Arimaspea in the first place.
The barbarian has perhaps dismounted to steal the gold when his
horse is attacked (The enmity between the griffin and the horse,
 KNOWLEDGE OF THE ARIMASPEA IN ANTIQUITY 37
which carried the robber to safety, became proverbial-Verg. E.
viii. 27.)
(ii) Vase from Mylasa in Garia, Mounted barbarian, left,
bearded, wearing spotted 'tights', spears half-prancing griffin,
right. The terrain appears to be rough; to rear of horse's tail
is 'blasted' tree. Above griffin is an amorphous stone-like object
(Winter 376). The vase is a so-called pelike, which supplanted
the amphora in the fourth century B.c.; Winter takes it to be an
Attic export. The barbarian he calls a Scythian, probably rightly.
Is the stone-like object meant to be a nugget of the 'guarded
gold'?
(iii) Italiote warrior, left, mounted and wearing a helmet with
two horns, menaces with poised spear a half-prancing, classical,
and somewhat disgruntled-looking griffin, right (Reinach ii. 319).
The scheme is that of (ii). There are four odd objects, two round
and two square, above and below the combatants: formalized
nuggets?
(iv) Aryballus of early fourth century :B.c., by the Xenophantus
painter, depicting a hunt by the Persian king. Right and left
barbarians (some with 'Persian' names superscribed) attack two
half-prancing griffins, one classical, the other a strange horned
type, not unlike some fourth-century B.c. wooden griffins from
Pazirik (Rice, pl. 21). In the background are a date-palm and
silphium (Reinach i. 23; Minns 343). The vase comes from
Panticapaeum (Kerch), and either it is an Attic export or the
painter was an Athenian settler there.
(v) Vase of early fourth century B.c. Three Amazons fight
three half-prancing classical griffins (Reinach ii. 295 ; Tischbein
ii, pl. 9). Tischbein thinks (30) that this scene was suggested
by some tradition that the Amazons made an expedition against
the griffins to win the gold. (For another such fight between
Amazons and griffins cf. Reinach i. 492.)
Not a great deal can be surmised from these representations.
Just as the Ke1ermes grypomachy was a revival of a conwntional
scheme, though the revival was probably caused by the artist's
knowledge of the Arimaspea, so the later grypomachies, though
they differ schematically from that of Kelermes, tend to become
38 KNOWLEDGE OF THE ARIMASPEA IN ANTIQUITY
conventional. This is illustrated by all the examples I have
adduced, with the exception of (iv), which is a mere fantasy
of the artist, in which quite different elements, Persian (dress
and names), Libyan (silphium plant), and Scythian (griffins), are
jumbled together. It is easy to infer that (i) and (ii) are nearer
to the inspirational source, the Arimaspea, than (iii), but impos-
sible to say how near they are. The Amazons are interesting:
for though there is no evidence of a tradition such as Tischbein
postulates, the first artist to conjoin them with griffins must have
had some reason for doing so. The reason was, I suggest, that he
knew that some authority placed the Amazons in Scythia-in
griffin-country in other words; and we shall see reason to suppose
that Amazons were indeed located in Scythia in the Arimaspea
(5off.; 178].
The one considerable testimony that the Arimaspea survived
into the Hellenistic era is that of 'Longinus'. But even this is
not conclusive, and is, in my opinion, outweighed by the evidence
to the contrary. Except for a few quotations, and these of doubtful
authenticity, it had I think disappeared irrevocably before the
foundation of the Alexandrian library. The hope, then, would be
vain that we might recover any fragments of it from papyri; but
it is reasonable to assume that the poem had debtors who did not
always acknowledge their debt, and that borrowings from it were
handed down the ages, in literature as in art, after their originator
had been forgotten. It is in the identification of these that the
only way now lies to a fuller discernment of the contents of the
Arimaspea.
 III

THE POEM'S DEBTORS

OF clues to probable borrowings from the Arimaspea our initial

supply must be furnished by the information about its contents
provided by Herodotus. The Issedonians are crucial. No sub-
sequent author adds anything to his account ofthem; 1 he himself
states that they lay beyond the range of the regular Scythian
trade-route, though it was 'certain knowledge' that they in-
habited the country to the east of the limit oi this (iv. 25). No
one save Aristeas ever, as far as we know, claimed to have
reached them. Whether they were fact or fancy is at the
moment of no importance : if the latter, they were his fancy; such
account of them as the ancients possessed it is safe to credit to
the author of the Arimaspea.
But it was the Issedonians who told of the one-eyed Arimaspi
living beyond them, who stole the gold from the guardian
monsters identified in the poem with griffins. These, then, are our
primary clues : Issedonians, Arimaspi, and griffins (if gold-guard-
ing or given a geographical location) ; mention of any of these
will suggest the influence of the Arimaspea. Hyperboreans too \\~ll
merit attention, but here care will be necessary, as the concept
of the Hyperboreans may be older than Aristeas; they may be a
contribution of his own Greek world to the furniture of the poet's
mind, as was perhaps the appearance of the griffin (85 ff.]. Armed
with these clues we must try to add to the matter which Hero-
dotus expressly assigns to the poem.
The first addition can be made with confidence. The order in
which the peoples succeeded each other according to Aristeas
was: Scythians, Issedonians, Arimaspi, griffins, Hyperboreans,
sea. Now according to Herodotus' contemporary, Damastes of
Sigeum, 'beyond the Scythians dwell the lssedonians, beyond
them the Arimaspi, then there are the Rhipaean mountains,
 THE POEM'S DEBTORS
from which Boreas blows and where there is always snow. Be-
yond these mountains the country of the Hyperboreans reaches
down to the other sea' (&:vw .EKVIJwv 'luU'Y/86va!i olKEiv, TOVTwv
!>'
Q O.VWTEpw
' I r1pta(]1TOVS'
>A I avw
" riptJLO.U7TWV
!,\ 'A
OE - 'Ta\ 'Pt7Taia
- OfYTl,
" E1, 'l:
~ I /3 I ,.. I ~\ I ' \ J \ I f \ ~\
WV 'TOV opEav ,rvnv, xiova Of. JJ,'Y/7TO'TE av'Ta EK/\EL7TEW. V7TEp OE 'Ta op.,, \ II

TO.VT (V'
A /3 I e
l 7Tf.P opf.OVS Ka 7JKELV EtS 'T'Y)V ETEpav
I J \ t I 8-1\
Cll\auuav, fir. I : see
Jacoby's note). That this is from the Arimaspea can hardly be
doubted. The griffins are omitted, as palpable fable perhaps;
remarkable restraint on Damastes' part in that case, if Strabo's
criticism of him as an unreliable romancer is correct (47; cf.
684), which may be explicable by his use of a more sober inter-
mediary- perhaps Hecataeus (cf. Agathem. Geogr. i. I). But we
learn something new-that the poem told of the chilly, remote
mountain home of Boreas.
For Homer this wind dwells no farther afield than Thrace
(ll. ix. 5), in a mountain or where is not indicated. But the moun-
tain appears in the beautiful lines of Aleman (fr. 59) :
'Pt1ras /'Jpos d.v8lov ii>..q.,
VVKTOS e'Aalvas UTEpvov,

'mountain of Rhipe [or of the stormy blast, pi1ra,; ?] , blossoming

with woods, breast of black Night'. That this is no Thracian
Haemus but a distant range on the edge of the world towards the
north is shown by its association with Night: it is the 'Rhipae
buried in darkness', JvvJxia, 'Pi1ra,, of Sophocles (O.C. 1248). 2
Later, mathematical geographers were to deduce that the long
arctic darkness must be counterbalanced by equally long day-
light,3 but the northern expanses, which the sun seemed to shun,
were at first the special realm of cold, storm, and gloom to the
Greeks for that very reason. Undeserved as this reputation was
in fact (Minns 4 f.), it would be somewhat confirmed by the
bitter winters of the Russian steppes. It was a reputation not fully
enjoyed by Thrace, which, though cold, was not characterized
as particularly dark.
We have already seen that Aleman knew of the Issedonians,
and that it is difficult to escape the conclusion that he owed this
knowledge to the Arimaspea [5]. It is reasonable to suppose that,
 THE POEM'S DEBTORS 41
just as he took his Issedonians from that poem, so he borrowed
another feature from it, the far-off mountain of Boreas, en-
veloped in black wintry stonns.4
From the same source as Damastes used appears to derive the
reference in Hippocrates' Airs Waters Places 19 (a work generally
considered authentic) : Scythia
lies right close to the north and the Rhipaean mountains, from which
blows the north wind (o9ev o f3opE1]',; 1TVEt: cf. eg cLv 'T~JI {3oplav 7Tl'e,v
reported from Damastes). The sun comes nearest to it only at the
end of its course, when it reaches the summer solstice, and then it
warms it but slightly and for a short time. The winds blowing from
hot regions do not reach it, save rarely, and with little force; but from
the north there are constantly blowing winds that are chilled by snow,
ice, and many waters [or heavy rains], which, never leaving the
mountams (OtJOE?rO'TE
""' '"
Ta opea, EK/\Et1TH:
>\I Cxr.ovo.
Cl, '' I
fJ,1]7TOTE ' ' 'EK/\EL7THV
O.tJ'Ta \I
Damastes, and Eustath. ad Dion. P. 3 II), render them uninhabitable.
A thick fog envelops by day the plains upon which the Scythians
live, so that winter is perennial, while summer, which is but feeble,
lasts only a few days (trans. W. H. S. Jones, very slightly changed).
And so the tradition continued. Near the Arimaspi
lie the Rhipaean mountains and the region which, from the likeness
to feathers of the snow which is perpetually falling there, is called
Pterophoros; a part of the world which Nature has written off (pars
mundi damnata [cf. Luc. ix. 858] a rerum natura); it is buried in thick
murk, is unproductive except of cold, and holds the icy womb where
the north wind is conceived (neque in alio quam rigoris opm gelidisque
Aquilonis conceptaculis).
SoPJiny (N.H. iv. 88); andLucan (iv. 106ff.) likens to conditions
at the pole an unbroken pall of rain-clouds which obliterated
the Spanish day and joined night to night:
sic mundi pars ima iacet, quam zona nivalis
perpetuaeque premunt hiemes: non sidera caelo
ulla videt, sterili non quicquam frigore gignit
sed glacie medios signorum temperat ignes. 5

It is in the Airs Waters Places that we first meet the adjective

'Pmafos at first hand, later to be used by the Akxandrians and
to become more familiar to us through the Latin form. It was
 THE POEM'S DEBTORS
probably used too by Hecataeus of Abdera (Ael. JI.A. xi. 1 ),
and possibly by Damastes and Hecataeus of Miletus; but the
lines of Aleman and Sophocles indicate that the mountain was not
so called by Aristeas. To him it was perhaps 'the mountain of the
storm gusts' (ptTTwv opos u.s.) of Boreas-cf. ll. xv. 171; xix.
358 (p,mi Bopeao); if he gave it a more specific name I shall
suggest that it was other than Tct. 'P,1rata opri [118].
The earlier usage makes its last appearance in Aristotle, who
records two further opinions about this range: it was incred-
ibly high, and it was the source of a number of large rivers:
'Beneath the Bear itself beyond the farthest part of Scythia is
a range of mountains called the Rhipae : the stories told of their
size are too fanciful for credence, but they say that from them
the greatest number and, after the Istrus, the largest of
European nvers flow> ('V1T > aV7"11V
' {'' 'T'l)V
OE ' "
ap1<TOV < I
V7TEp -
'T'l)S > I
EOXa7't)S
'"' 8',as a,' KOJ\OV/J.f.Vai
.wKV -\ ' 'PA7Tat, 7TEp,' WV.. 'TOV- /.EYE'8ovs 1\taV
"' EWLV ' '
< \ I \ t () I { ' <I C,> .. C \ - \
0 I\EYOf'EV0 Aoyo, .v wons pEOVO't O ovv Ot 'll'I\ECM"O, Ka.,
I
f-LE"/ta'TOt f'E7a
\ \
7ov II T
La'Tpov 7WV -
=wv
_W\)
rrora.wv
- > -e c, .J.
EVTEV ev, ws .,.acnv:
Arist. Meteor. i. r3. 35ob7 (trans. H. D. P. Lee, very slightly
changed)). The immense height of this range was to provide
the sixth-century physicist Anaximenes with a screen to account
for the obscuration of the heavenly bodies as they moved laterally
about the earth from their setting to their rising. 6
Though Herodotus nowhere names the Rhipaean mountains,
he does speak of mountains which might well be these. At the
limit of the northern trade-route dwell the Argippaei, in the foot-
hills of a range lofty and impassable; it is inhabited, so these
people say, by goat-footed men, and beyond it are others who
sleep for six months of the year (iv. 23. 2; 25. 1). Now Herodotus
must have conceived of this range as lying to the north of the
Argippaei, for to the east of them were the Issedonians; if exten
ded, it would then lie to the north of the latter too, exactly where
geographers located the Rhipaean chain [117; 181 f.]; while its
great height would befit the fabulous Rhipae.
From the feathery snow that perpetually fell there the Rhi-
paean region was dubbed by someone Pterophoros (Pliny, I.e.).
The unmetrical form of this name is sufficient to show that Pliny
 THE POEM'S DEBTORS 43
was not here using Aristeas direct ; and the author of it (Heca-
taeus of Abdera ?) may have been inspired by no remoter source
than Herodotus, who says (iv. 7): 'Above, to the northward of
the farthest dwellers in Scythia,7 the country is said to be con-
cealed from sight and made impassable by reason of the feathers
which are shed abroad8 abundantly. The earth and air are alike
full of them, and it is this which prevents the eye from obtaining
any view of the region.'
Herodotus ascribes this information to the Scythians them-
selves, but it has that touch of the bizarre that suggests the
Arimaspea, and the suspicion that it may have originated there
finds some confirmation. First, its context: it occurs at the end
of what Herodotus calls the Scythians' own account of their
origin-they (or rather the so-called Royal Scyths, who regarded
themselves as pre-eminent, cf. iv. 20. r; 5. 4) were descended from
one Colaxais. This name does not occur elsewhere, but an adjec-
tive Ko>-.a{aio!: applied to a horse by Aleman (fr. I. 59) is
apparently coined from it to mean 'Scythian'. It seems not im-
probable that Aleman got his knowledge of this name, as of the
Issedonians and the Rhipaean mountains, from the Arimaspea;o
which need not, however, have contained all the genealogy and
outlandish proper names given in Hdt. iv. 5-7, but merely the
statement in 7. 2 that Colaxais divided his kingdom into three
among his sons. This statement is part of a tradition which
differs only slightly from that given by Herodotus immediately
before, in 5 ff. They are doublets, in fact: in the first version
the whole Scythian nation originates from Targitaus through his
three sons; the Royal Scyths, descended from the youngest son,
Colaxais, preserve the sacred gold which gave him his pre-
eminence. In the second version Cola.xais divides the whole of
Scythia among his three sons ; the sacred gold is preser\'ed in the
largest of these three kingdoms. It is noteworthy that Herodotus
remarks (iv. 13. 2) that Aristeas does not agree with the Scythians
in his account of the country (he means in his account of how its
inhabitants came to be there). This must mean that the first
version, in 5 ff., is not from the Arimaspea (for it treats the
Scythians as aboriginals, whereas the Arimaspea said they were
 THE POEM'S DEBTORS
immigrants), but need not preclude mention in the poem of
threefold rule among the Scyths by the descendants of their
common progenitor, Colaxais (as in the second version).
A second pointer to the Arimaspea is in iv. 3I. There Herodotus
expresses the opinion that by these feathers the Scythians and
their neighbours mean snow. So others beyond the Scythians spoke
of the 'feathers' : but how could Herodotus know this? From
Scythian report heard in a Black Sea mart? Possibly; but it is
also possible that these others are the Issedonians, and that the
feathers are another bit of Issedonian lore retailed by Aristeas.
If that is the case, Herodotus would be here making a tacit
assumption similar to the open assertion in iv. 27, where he says
that the Greeks got their knowledge of the Arimaspi from the
Issedonians by way of Scythian intermediaries (not, he implies,
by way of Aristeas).
Thirdly: when Herodotus, with a noteworthy fussiness, equates
the feathers with snow-flakes, he may seem to us to be labouring
the obvious. Yet strangely enough it does not seem to have been
a similitude that readily presented itself to the Greek mind, or
we might expect to meet it more often in the poets; but the only
place where I recall having met it is in the Prometheus Vinctus (993
>.evK01rTepo, vupa,)-significantly, I think, as there is evidence
that Aeschylus made use of the Arimaspea in this play. It is possible
that it did not occur to Hippocrates (I.e.), or his source, and that
the thick fog (~~P 1ro111J,) in which he says the Scythians live-
puzzling, as the truth is rather the reverse (cf. Minns 5)-is
another rationalization of the feathers which fill the air and
obscure the sight.
In the region of the falling feathers, then, are the Rhipaean
mountains and what Pliny calls the Aquilonis conceptacula, 'the
womb in which the north wind is conceived' [41 ]. The nature of
this conceptaculum is specified elsewhere by Pliny, in a passage
already discussed in another context [ 28; T. 8] :
That there are Scythian tribes, and those quite numerous, who
are cannibals I have already indicated .... Near those in the north,
and not far from the very starting-point of the north wind and the
cavern which is called 'the North Wind's Cave (haut procu ab ipso
 THE POEM'S DEBTOR5 45
aquilonir exortu specuque eius dicto )-the place named Ges clithron-are
said to be the Arimaspi, remarkable for having a single eye in the
middle of the forehead. They are always fighting for precious ore
with the griffins, winged animals whose appearance is well known,
which throw up gold when they make their holes [cf. xi. I 11 J..
Many writers relate this, but the most famous are Herodotus and
Aristeas of Proconnesus.

'The North Wind's Cave so-called' occurs again in (Plut.]

f{uv. 14. 5: the herb called Phrixa, to be found by the Tanais, is
particularly abundant near 'the cave of Boreas', -rrapa. Boplov To
1rpoaayopev6evov a.VTpov. Little reliance can be placed on the
writer of this treatise, who gives as his authority here 'Agathon
of Samos in the second book of the Scythica', an author and work
otherwise unknown; but at least the Cave ofBoreas is not hi; own
invention, and its situation 'by the Tanais' need not surprise us
when we recall that this river was thought by some to flow from
the Rhipaean mountains(e.g. Mela i. IIS; schol. ad Dion. P. 10).
The other name mentioned by Pliny, Ges clithron, presumably
means 'the entrance to the Earth's windpipe'-a sense of KAe,-
Opov for which the earliest testimony is the Hippocratic corpus,
and so it is perhaps not directly from the Arimaspea ;10 though
anatomical metaphors applied to geographical locations are old
enough (Od. i. 50 Vl)<1lp V &.<f,ipv711, 08, 'T' def,aAodITTL 8aAa.acrrys ).
The actual name may have been coined by the same brain that
produced Pterophoros, and like that may have summarized some
statement in the Arimaspea.
I have said that Aeschylus knew the Arimaspea. At least he shows
two of the primary symptoms-mention of griffins in a geo-
graphical location, and of Arimaspi; gold too is associated with
the latter. lo, instructed in the future course of her wanderings by
Prometheus, is told to avoid the griffins and the host of one-eyed
Arimaspi who dwell about 'the spring of Pluton's stream that
flows with gold' (P. V. 803 ff.):
o'i: z \ ,
5vITTo.ovs yap 71vos aKpay,ts KVl'as
I \ "'
I

yp01ras ef>oAagat, 7'0V 7' .ovvwrra. crrpa-rov

tA \ ( f:11 t
ft I
r1pi.a.a1rov t1T1To,-,a.ov , 01 xpvaopp11To1
, ~ , .I.'
OLKOVUW a..,,, va.a
~ n, ,
AOV'TWVOS 1Topav.
,
46 THE POEM'S DEBTORS
Th e 1our words ovvanra.
l'. ~ I
<J'Tpa:rov >A I
npia.<T11'ov ,
,.,, .R ,
..,,.o,.,aova are
almost a summary offrr. 4. 2 f. and 5. I of the Arimaspea-or the
latter might be an expansion of Aeschylus!
Because of this, and because Aeschylus' sketch of lo's route is
the earliest account of any length that we possess concerning
the Greek geography of the northerly parts of the world, I propose
to examine it in detail: if Aeschylus borrowed his griffins and
ArimaspifromAristeas he may have borrowed other things too. It
is indeed not easy to discern any rational scheme underlying this
account at all: it seems at first to be a mere jumble of geo-
graphical and mythological names, fitted together not only
capriciously but incoherently. Nevertheless I believe that sense
can be made out ofit, and in the exposition that follows I give my
own interpretation [Map I]. The clarification of Aeschylus'
geography is, of course, only half the present task : we have also
to identify borrowings if possible-always a difficult matter in
the case of genius, which does not copy but digests its sources. u
In the P. V. Prometheus is fettered to a craggy mountain some-
where to the west of Scythia and to the north of Greece. Aeschylus
does not, in this play at least, call this mountain the Caucasus;
it is called so by one Hypothesis, in accordance with the usual
tradition, but the warning later a pp ended to the same Hypothesis
should be heeded: 'Aeschylus does not say that the scene of
Prometheus' binding is the Caucasus, as it is in the common
account, but near the European margin of Ocean, as we can
infer from the speech with lo.' The question need not concern us
now whether the Caucasus was the scene of the Prometheus
Solutus, or whether Cicero in his translation of a portion of it
(fr. 193. 28) and Strabo (183) have been led into error by their
acceptance of the traditional scene of Prometheus' sufferings :
our present starting-point is sufficiently fixed by two bearings-
lo has reached Prometheus (82g-41) by way of Dodona, along
the Ionian sea (the Adriatic), and then has changed her north-
westerly course for an easterly one (1roJ.,p1rr>-.&.yK'TOt!: 8p6o,s 838) ;
when she leaves him she will approach the Palus Maeotis from
the west (707; 729-35). The first part of this agrees well enough
with the account of lo's wanderings in Apollodorus (ii. I. 3) : she
 THE POEM'S DEBTORS 47
reached the Adriatic (' l6v,os K6A1To,), then travelled through
IJlyria and crossed the Haemus range; but Apollodorus now takes
her over the Thracian, not the Cimmerian, Bosporus, adding
vaguely that she then 'went off into Scythia and the Cimmerian
land'. He appears to be conflating two separate versions of the
legend here.
When lo leaves Prometheus she will go east, and first come to
the Nomad Scyths, far-shooting archers who live in caravans,
whom she is to avoid by going along the shore (707-13):
'lrpcirrov Jv l118lv8 ~Alov 71'p6s aVToAaS'
'./,. \ ,..,, I I
crrptr.,,a.ua UaVT7'/II O'TECX aV7fpOTOVS' yva,
..... '8as "'... a't',S1/
"1KV '.l.'l: vo.aoaS',
"" o, 1r111CTa<;
\ ' ureyaS'
,

1T8a.pur.o, valovu' l:1r' w1ev1e;\o,s ;xo,s, 710

EK'1Jf36>..o,s To,occnv Jf17{'Tv{voc
'\IY 2\\>.!\ I '<'
o,s f.J.7) 7TE/\a1,Etv, U/\1\ <11\,0ToVOLS 1TOoas
XP'JJ-7TTOVUa fiaxta,<1w K1Tpo.v x06va.
We hear of these wagon-dwelling Scyths as early as the Hesiodic
Catalogus, where Phineus is driven by the Harpies I'>.awrocpa:ywv
' ya.,av
ES - a7TTJVaS'
'' ' " EXOVTWV
OtKt '' (fir. 54 R . = fir. 48T)Th'
.. e1r
position calls for little comment, except that Herodotus (iv. 19)
puts them nearer the Palus Maeotis, east of the Borysthenes. So
does Aeschylus himself in another passage (417 ff.); here they are
separated from the Palus Maeotis by the river Hybristes.
It is with the next people that lo will have on her left, the
inhospitable Chalybes, workers of iron (714), that a difficulty
arises;

Hecataeus (fr. i203) puts them on thesouthernshoreofthe Euxine,

north of Armenia, and this was the position assigned to them by
e.g. Herodotus (i. 28), Xenophon (Anab. iv. 5. 34; v. 5. 1),
'Scylax' (88), and Apollonius Rhodius (ii. 1001). I know of no
other author who puts them on the north side of the Black Sea,
though curiously they are called l8voS' 1:Kv8,Kov by Hesychius, and
by the scholiast on Apollonius Rhodius in several places (ad i.
13iu, 1323; ii. 375-where they are put by the Thermodon;
 THE POEM'S DEBTORS
and 378-where their neighbours the Tibareni are also called
l8vo, EKv81Kov; yet both the Thermodon and the Tibareni are
on the south shore of the Euxine).
Io's connexion with the Chalybes here may be a reminiscence
of the earlier version of her wanderings already mentioned,
whereby she crossed from Europe into Asia by the Thracian
Bosporus, and travelled through Asia Minor on her way to Egypt,
when she would have the Chalybes on her left. This earlier version
is recounted by Aeschylus himself in the Supplices (538 ff.), where
Io's route is through Phrygia, Mysia, Lydia, Cilicia, and Pam-
phylia. There is no mention of the Chalybes, however, and they
would indeed be a considerable way off her path.
Another explanation is possible. In several places Aeschylus
connects Scythia with iron: at P. V. 301 it is called 'the mother
of iron', a101Jpo~Twp; at Sept. 8 r 8 iron is called 'Scythian',
at 7!28 'an emigrant from Scythia' (Xd)..vftos l:Kv8wv a7TOIKOS).
Was this collocation merely a poetic one (it is appropriate
that the source of the metal of war should be a fierce people) ?
It would then be appropriate too to transfer the traditional
iron-workers, the Chalybes, to Scythia. Or was it not poetic
fancy but hard fact that Scythia was 'the mother of iron' ? The
description would be grounded on fact today, for Krivoy Rog
and Kerch are among the most impGrtant producers of iron ore;
but I do not know that we ever hear of exports of Scythian iron,
and in antiquity the discovery of bronze rather than of iron was
normally credited to the Scythians. A papyrus fragment (189)
of Hellanicus is thus restored by Wilamowitz: [<TW7Jp]a 8i 01T>-.a
1rpwTo; 'E'\\I
A ,
N\aviico; Ka'TaCJKEvaaaa ~I
8a,'.I.y,71a1v .:;,av,wvov "'BA
-"KV ,,
wv ovTa
~aa,Ma, 'Hellanicus is the first to say that a Scythian king
Saneunus discovered how to forge iron' ; 12 and Hesychius has
this gloss on Chalcidice: 'Scythian; iron-mines were first dis-
covered there' (Xa1'K181~ l:,cv81,c~ Tct. lTaMa Tov u,8~pov EK~,
1TpW'TOV vpe0wra). It is unfortunate that in the former every
letter of the crucial word except the last should be conjectural;
while in the latter a~pov may be a slip for xa>-.,cov- if Xa.Ai<,811<~
has not been confused with Xa.X.v~,~.
There is yet another possibility. The Chalybes, in their usual
 THE POEM'S DEBTORS 49
position south of the Pontus, are the neighbours of the Amazons
who dwell around the Thermodon; and when Aeschylus put his
Amazons about the Palus Maeotis (731) he may have transferred
their neighbours the Chalybes as well to keep them company.
Having passed the Chalybes lo will come to a river which
Aeschylus calls Hybristes.13 This she must not try to cross but
must follow to its source in the Caucasus, where, climbing the
lofty peaks, she will round it and head southwards. She will now
be in the country of the man-hating Amazons, who will be glad
to lead another woman to the next landmark, the Cimmerian
Bosporus:
~
'1)5tS' o,;,, <VQ
.i
'
r'P'G7TJV 7r07aov
ov' .I. s, '
.,,evowvvov,
I\ \ I > \
ov p:rJ '1Tpa<r[J,, ov yap e:v aros -rrepav,"{3 A

7Tpiv av 7TpOS' at}r-ov Kai)Kaaov .6>.I],, opwv

.,.
vy,,O'TOV, " 8a 'TT<rra.p,os
(V e:K.,,vaq.
,l. - f.LEVOS''
71.0
~J._ > )o ., t..., I ~ \ \
KpCYT.,,.,,wv aw avrwv. acrrpoye:,Tova, oe: XPTJ
kOp"'f'aS'
- ' - 'vm,p
. alV\OUaav ES'
fJ_!\\ , J1,E"'1JJ1,f3 pLVTJV
\
,,__ ,, (J " (J' ,,,
r-T/vcu Ke:l\t:V ov, e:v ;-1.a.,,ovwv a7-paToJ1r'
,t. I ) fi " I I
7J!:,LS' a-rvyavop ' aL <:7fUUKVpav 1TOT
> ,I..\ Jn '~ I))
KaTotKtovcnv a.y,i
A
l!::le:p.woovu , wa
Tpaxe:ta ?TOVTOV l:a.A.v8riuala yva8os,
EX8p6,e:vo, vain-r,cn, p.TJTpv,a. Yt:wv
a~a[ u' 081JY17UOVat Kat .oJ..' aup.EVWS'.
la6ov 8' J-rr' awats U'Tt:Yo-rropo,, Alp.VTJ, mJAa~s
K,..e:pt1<ov -jf[e:,s, ov
8paava1rAaY)(VWS aE XP1/ 730
At'TToiio-av av>.wv' EKTTEpriv Ma,wnKOV.
iO'Tat 6i 8V'Y)TOLS' e:laad A6yos .eya,
rrjS' afjS' ?TOpe:las, B6a-rropo, S' rnwvvp.os
KEKA~UETat. AnTOiiaa 8' EvpW'TTTJS' 1re8ov
,, "~ .,, ,..,
1J7TEtpov 'Y)s Ets no-taoa.. 735

You will come to the river Hybristes, which does not belie its name:
this do not cross-for its crossing is not easy-until you reach the
Caucasus-none other-, loftiest of ranges, where the river spouts
forth its might from the very brows of the mountain. In traversing
those peaks which neighbour the stars you must bend your path south-
wards, when you will come to the host of man-hating Amazons, who
one day will settle at Themiscyra by the Thermodon, where the step-
mother of ships, Salmydessus, arms the sea with jagged fangs and gives
evil entertaimnent to sailors. These Amazons will guide you on your
 50 THE POEM'S DEBTORS
way, and gladly, to where, right by the narrow gateway of the salt-
mere [the Palus Maeotis], is the Cimmerian isthmus. This you must
be bold to leave, and cross the Maeotian channel ; far flung and ever-
lasting will be the fame of your crossing, and from it men will call the
strait Oxford [Bosporos]. Now, having left the plains of Europe, you
will enter the Asian continent.

Here we are faced with the central problem of Aeschylus geo-

graphy: where is his Caucasus? If the Hybristes rises in the con-
ventional Caucasus, and lo heads south after surmounting its
peaks (721 ff.), then the Palus Maeotis and Cimmerian Bosporus
which she next has to cross will be on the east or even on the
south side of the Euxine-and she will cross from north to south
(or east to west),instead of from west to east as it appears she does
in 790 f.:
o-rav '7TEpa.crns pE'i9pov 1j'7f'Elpow /Jpov,
\ \ \ .J:\ - ., Q -
wpos aVToAas 'f'"oywwas 'IJ"f.OcrTL,.,ELS'

Kiessling (2 1 o) solves the problem ingeniously : Aeschylus'

Hybristes is a confusion of the Hypanis-Bug, which flows into the
Euxine to the west of the Palus Maeotis, with the Hypanis-Kuban
(or Anticites--Strabo 494), which rises in the Caucasus and
empties into the eastern side of the Palus Maeotis [Map I],
north of Colchis, where Aeschylus locates the Amazons (415).
This confusion has caused him to shift the Amazons west, and
place them on the far, or western, side of the Palus Maeotis.
For me this explanation lacks conviction because it is the sort
of mistake a scholar might make, as Alexander Polyhistor and
Pliny actually appear to do (Steph. Byz. s. "Y1ra.v,~; N.H. iv. 84),
but hardly a likely one for a poet who does not seem to have
briefed himself with up-to-date information: it postulates too
much topographical knowledge on the part of Aeschylus. It is
also perhaps worth remarking that the earliest extant reference
to the Hypanis-Kuban is in Aristotle (Hist. An. v. 552b 17). The
river is not mentioned by Herodotus, though the Hypanis-Bug is.
There is in fact no need to suppose that Aeschylus has trans-
ferred the Amazons at all, for there were Amazons to hand in the
 THE POEM'S DEBTORS
place in which he puts them, about the Palus Maeotis; and it is
these that are meant here. They are the Eavpop.a:ra, yvvmico-
Kpo:rot$.EVot of Ephorus ('Scymnus' 878-85), theAmazonsofMela
(i. u 6 ; cf. Plin. N.H. vi. I 9), whose presence in this district
Herodotus (iv. uo ff.) explains thus. After the Greeks had de
feated the Amazons in the battle of the Thermodon they put their
prisoners on three of their ships, and sailed for home. But the
Amazons rose and slaughtered their captors; then, having no
knowledge of seamanship, they were carried by wind and wave
to the west shore of the Palus Maeotis. Here they disembarked,
and soon made friends with some Scythian youths, who became
their husbands. The Amazons refused to join the Scythians'
families, on the ground that they would never get on with the
women-folk; instead they persuaded their husbands to migrate
with them.

I Crossing the Tanais [says Herodotus (iv. r 16)] they journeyed east-
wards a distance of three days' march from that stream, and again
t northward a distance of three days' march from the Palus Maeotis.

I
Here they came to the country where they now live, and took up their
abode in it. The women of the Sauromatae have continued from that
day to the present to observe their ancient customs, frequently hunt~
ing on horseback with their husbands, sometimes even unaccom-
panied ; in war taking the field ; and wearing the very same dress as
the men.

This is clearly a Greek aetiological myth, to account for certain

sociological characteristics of the Sauromatae, or perhaps more
properly of the southern neighbours of the Sauromatae by the
mouth of the Tanais and the eastern shore of the Palus Maeotis,
Iazamatae or Maeotians (Rostovtzeff 11 100 f.): they must be
evidence of kinship with the Amazons of Greek legend, who dwelt
on the south side of the Euxine (cf. Ebert 340). But Aeschylus, it
should be noted, knows nothing of the myth as it is given in
Herodotus; at least, he says the opposite: his Amazons will later
make their home by the Thermodon (724 f.; cf. Sall. Hist. iii
73). It seems as if Aeschylus were here drawing on someone who
had noticed the equality of the sexes in these tribes, and the
52 THE POEM'S DEBTORS
similarity of their women with the Amazons of Greek legend,
but who had preceded the time when the Greeks had worked
up the aetiological story given by Herodotus. These tribes are
considered to have been indigenous remnants, who survived the
Cimmerian and Scythianmigrations (Vernadsky 54); they would
then have been in much the same position in the seventh century
B.c. as in the fifth.
These northern Amazons are mentioned by Euripides (H.F.
408), and seem to be meant by Pindar: 'With this clear and
gratifying prophecy Apollo quickly drove away to Xanthus, the
well-horsed Amazons and the Ister' (OZ. viii. 46: clJ, {ipa 9E<>s
I.J.
aa.'f'a. "
Ei1ra,. ';i' I ()
'1 \ >A y, >I \ I
-,a.JI OJI 7l1TEt')'EV Km ri,a.,,ova.<; EVL'IT1T0Vo Ka, E<;
"Jcn-pov
J.:\avvwv).The Amazons of the Thermodon were not a favourite
haunt of Apollo; these must be the northern Amazons, and
be taken closely with the Ister-by which Pindar refers to
that place beloved of Phoebus, the land of the Hyperboreans,
where were the sources of the Ister ( 0 l. iii. r 3 ff.) .14 Finally, these
are the Amazons who are the antagonists of the griffins in
classical art [37].
If Aeschylus' Amazons are Sauromatae or kindred Maeotian
tribes, and lo approaches them from the north (722 ff.), then the
course of the Hybristes must be from north to south, and the
Caucasus, in which it has its source, cannot be the conventional
Caucasus. Its situation fits perfectly the traditional location of
the Rhipaean mountains, and if this identification is made the
worst stumbling-block to the introduction of some order into
Aeschylus' geography is removed.
Aeschylus' Caucasus shares another feature with the Rhipaean
mountains besides geographical position: immense height. It is
opwv vi/mJTos, 'loftiest of ranges'; its peaks 'neighbour the stars'-
compare the fabulous height of the Rhipaeans [ 42]. In the pas-
sage of the Meteorologica there quoted, Aristotle says that the
largest European rivers are reputed to have their source in the
Rhipaean range : Aeschylus puts the source of the Ister there, as
does Apollonius (Aesch., fr. 197 ap. schol. ad Ap. Rhod. iv. 284);
so too does Pindar, as I have just stated. The Tanais rises in the
Rhipaean mountains, according to Mela (i. I 15), Pliny (N.H.
 THE POEM'S DEBTORS 53
iv. 78), Lucan (iii. 272 f.), and Orosius (i. 2. 4); in the Rhipaean
Caucasus, according to Dionysius Periegetes (663 ff.):

TOU s ./frot '71'7/'Yal h lv oiJpeu, KavKau{ow,

T7JA08 opvpovaV' & 81 '1TAa'TrJS lv8a Kai MJa
laao.evos EKv81<0<1V emTpoxd.ei 1Te8lou:Tv

-on which comments Eustathius 'we must understand at this

point that the Caucasus [or the Caucasus in this context] is one
of (n) the most northerly sections of the Taurus, reaching down
to the neighbourhood of the Cronian sea [the Arctic Ocean], of
which a part is also the afore-mentioned [I. 315] Rhipaean
range' .15 The scholiast on 1. 1 o of Dionysius, quoting 1. 663 on the
Tanais rising in the Caucasus, says (what might, however, be a
mere guess) <'i.e.> the Rhipaeans; for some locate its source there'.
Those who thought that the Caucasus where the Tanais rose was
the conventional Caucasus had to postulate a course for the river
at first south to north from its rising, then turning south again to
the Palus Maeotis (Strabo 493)-though Herodotus had long be-
fore put its source somewhere in the far north (iv. 123; cf. 57).
Eustathius says that this northern Caucasus is the scene of
Prometheus' punishment (not true of the P. V. [46]), and is not to
be found in the map of the periegesis: -roi SJ -roiovrol' KavKaaov,
\a -npo'Y) ()'EWS' 7Tl\a'TTE'Tat avaaKOl\01TL<1fLOS',
1tEpt ov KaL O 'TOV
\< \I > \ I ( _\
OL 1TW\aLO
\

.~ iyKeiu6al cf,arn Tip rrjs- 7TEpt'Y)')'1)<TEWS 1T{vaK, (ad Dion. P. 663).

Whatever this means exactly (a map once attached to the
periegesis ofDionysius? or just 'the map of the world'?), it cannot
mean the conventional Caucasus. There appears to be a reference
to this Rhipaean Caucasus in [Plut.]fiuv. 5. 3, where Cronus is
said to have fled from Zeus to the Caucasus, which till then had
been called 'the lair of the North Wind' (Boplov Koln7) (and so
Cronus gave his name to the Arctic sea adjoining, To Kp&iwv
1TtAayo!.' ?) .16
A grave difficulty is the association of Arabians with the Cau-
casus in P. V. 420 :
J4pafJ[a,; T' apHOI' av(}o,;,
t .,. , I\ '\
v't'iKp7J.vov 01 1TOiHCJa

KavKcfoov 1rlAas vtoVTa1

 54 THE POEM'S DEBTORS
'and the warlike flower of Arabia, holders of the high-perched
fortress near the Caucasus'. It is a difficulty which has not yet
been satisfactorily resolved. The text as it stands is most improb-
able; yet no convincing emendation has been proposed. The
codices' {Jif,l,cp71.vov (J' is unmetrical, and was not the reading
before the eyes of the Medicean scholiast (:\d1m o ,cat). Of
1
attempts to emend l1paftlas 7 Burges's J4f1&.p,es most humours
my theory; but unfortunately Avars are not heard of until nine
centuries after Aeschylus, and in classical times Abaris is only
the name of an individual Hyperborean. I suppose that it is not
impossible that Aeschylus coined a tribe on the analogy of this
name 18 (the Hyperborean Abaris was known to Pindar, fr. 283),
but it would be very ill-chosen, as the Hyperboreans were the
reverse of warlike. Oddly enough Ovid (Met. v. 86) has a Cauca-
sius Abaris (again a personal name), but this is probably a mere
coincidence.
If, in spite of this difficulty, my location of the Caucasus of the
P. V. is correct, poor lo, as she wends her way up the western
bank of the Hybristes, will be marching straight into the teeth
of the north wind. Now there is a passage quoted by Galen as
from the P. V., though, as it is not found in our texts, Galen is
thought to have written lv llpo.7161:t S1:a,w771 in error for lv
llpo.7181:t Avo.lvq, (fr. 195) :

Paley tentatively inserted the passage in the P. V. between 791

and 792, to fill what appears to be a lacuna; there are, however,
three objections to such a context-(i) TTpwnCT'Ta h of the frag-
ment has no answering SJ (though of course the lacuna may have
contained more than these four lines); (ii) lo will hardly avoid
the harrying northern blasts simply by coasting along in the sea
(792)-as if the sea were immune from winds; (iii) it implies
that the north wind blows across south Russia only in a narrow
corridor-why should lo, if she is travelling west to east, have to
 THE POEM'S DEBTORS 55
take precautions against it at only one point? This last is a serious
difficulty for any interpretation except one-that lo is actually
nearing the birthplace of the north wind; and the fragment does
nt excellently between 720 and 721 :
-ijfe,s o' 'Yf1plrn7]V 7TOTa..clv oil if,ev8c!.rvv.ov,
&'I \ I tf,Q 't,. \
ov fl,'ff wepaar,s, ov yap ev,.,a.,.os 7TEpa.v,
7Tpiv civ 'TTp6s ail'rclv Ka6Kauov .6A11s, opwv
.,.
vy,,0'7'ov, wea TTO'ra.os.1.~,
cw EKy,VO'<[, .evo,. 720
ev8eiav ;f"TTE T1JV0E" KO. 7TpWTw'ra ,v
f3oped.8as .;;gei, 7rpcls 'fl'VOO.S, {v' evAa{3ov
R , ,y
,-poov KO.TIU')'!, OV'ra., :'I u ava.p7ra.ur, I " ' I
72oc
" ,
'.I. ,.,
OVO)(U/J,Ep<p 1Tf''f'''Y' O'VO"TpEy,O.t; a.VW
,!,J. ~!
KpoT.....,,w,V 1 ... :,
O.'IT
:t 't - '
aVTwV. 0.0"Tpo'}'ELTOVO.t;
I
0 XP"I\ 721
Kopv,f,a.r; tnrpf3&..ovuav ES p.EUT)JJ,{3p,W]v
flijva, KiAwllov
Do not cross the Hybristes [says Prometheus] until you reach its
source, in the Caucasus. Keep to this course [i.e. following the course
of the Hybristes, as implied in 718 f.] without wavering, and first of
all you will come to the blasts of the North Wind's breath, where
mind its hurtling roar, lest with its stormy whirl it spins you round
and sweeps you from the very brow of the mountain [i.e. in the very
act of accomplishing your goal of turning the river]. Then as you are
crossing those peaks that neighbour the stars [i.e. in order to round the
source of the Hybristes; note the present participle-Groeneboom's
thr,ppa>.oiiuav is misconceived] you must bend your way south ....

The i.v of the first line of the fragment has its answering~ in 721
(there would, of course, be no more adversative force here than
between 1Tp6nov lv and l:vOas o' a.cplfn in 707 and 709) ; and
there would be considerable point in telling lo to be particularly
careful when she is approaching the cavern in the mountain
whence Boreas issued, where his full force would be felt.
Now, under the guidance of the Amazons, lo will come to the
Cimmerian Bosporus, and must cross from Europe into Asia
(729-35), Herodotus states that some made the Tanais and the
Cimmerian Bosporus the boundary between the continents,
others the Colchian river Phasis; and expresses surprise at the
arbitrary division of a world naturally one (iv. 45). We may
share Herodotus' surprise on one score at least: the choice of the
56 THE POEM'S DEBTORS
Phasis as a boundary. This, to Herodotus and subsequently, was
the river known in modem times as the Rion, a third-rate stream
flowing into the easternmost recess of the Euxine. Yet it was an
older boundary-line than the Tanais, according to Agathemerus
(i. 3).
This seems very odd. Yet even odder is an author who names
both the Oimmerian Bosporus and the Phasis as the boundary of
Europe and Asia, and this not in separate works written at widely
different times but almost in the same breath: for Aeschylus him-
self at the beginning of the Prometheus Solutus says (fr. 191)
,rjj lv ot8vov x8ovas E?JpclmTJs
lyav 178' 14.ulas -rlpova tPauw

'where great Phasis, forming a double boundary, to Europe and

to Asia ... '. How is this to be accounted for? Is it just another
example of Aeschylus' wondrous ignorance of, or disregard for,
geography, or is it rather that our own ideas about the Phasis
need an overhaul?
It is obvious that the ideal dividing-line of continent from con-
tinent is a strait connecting sea and sea, such as the Hellespont,
Propontis, and Thracian Bosporus; or, even better, connecting
sea and Ocean, such as the Straits of Gibraltar. These were, of
course, well-known and accepted boundaries. When the Greeks
in their exploration to the north of the Euxine came upon the
Oimmerian Bosporus, it would seem clear to them that here was
another natural boundary between Europe and Asia, like the
Thracian Bosporus they knew. Then it would be discovered that
the Palus Maeotis was not a bay of Ocean, but a sea in its own
right. Enclosed, yes, but in the far north-eastern corner there was
a large channel (of what was later to be called the Tanais19):
was this a river, or did it lead through into Ocean, to complete
the division of the continents? Might this not be the Phasis,
famous in Argonautic legend? For in the earliest form of this
story the Argonauts sailed through the Phasis to the Ocean: in
other words, Phasis was less a river than a strait. Hesiod took his
Argonauts on their return journey through Phasis into Ocean,
and in this he was followed by Hecataeus of Miletus (schol.
 THE POEM'S DEBTORS 57
ad Ap. Rhod. iv. 284; 259). 20 Indeed, Mimnermus in the
seventh century B.c. put the city of Aeetes, the goal of the
Argonauts, on the brink of the eastern Ocean (fr. 11. 5 ff.):
I
A ''1J'Tao I\ '(J 'T WKl:OS
1TOI\W, 'TO '
' I 'H/\tOtO
\,
dK'TWES xpvulcp K1:la'Tm lv 6aM..cp
'DK1:a.voiJ ?TO.pa x1:i:Aos, i.'v' Cf'X'TO (JEWS l~awv

'the city of Aeetes, where the rays of the swift Sun rest in a
golden store-room, by the lip of Ocean, whither god-like Jason
went'. He presumably went through the Phasis to get there.
So the original Phasis, the legendary strait, would make an
ideal intercontinental boundary, and it can be seen how it might
at first have been thought of as an extension of the Palus Maeotis.
Later, after it had been discovered that the Phasis-Tanais was
only a river after all and when the Milesians had penetrated to
the easternmost comer of the Euxine and could sail no farther,
they considered that they had really come Js <Paatv, b-Oa vava,v
laxa.Tos 8p&fl-o,, 'to the Phasis, the farthest point a ship can reach',
and here was to be the permanent location of the Phasis, still
with its legendary reputation, which it could now ill support.
So later geographers, feeling the inadequacy of the Phasis-Rion
as an intercontinental boundary, sometimes spoke of this boun-
dary as 'the Phasis and the isthmus between the Pontus and the
Caspian'.
But the Phasis of Aeschylus, or of his authority, was not this
historic Phasis-Rion, but the older strait, conceived of as adjacent
to the Cimmerian Bosporus and continuing the latter's function
of cleaving the continents: so Aeschylus would be no more
inconsistent in speaking of both Phasis and Cimmerian Bosporus
as the boundary of Asia and Europe than he is when he speaks
of the Cimmerian and Thracian Bosporus as the boundary (cf.
Suppl. 544 ff.).
I suggest, therefore, that there were three stages in the history
of the Phasis: (i) it was a legendary strait towards the east con-
necting the inner sea with Ocean; (ii) it was identified with the
river which came to be called Tanais; (iii) it was later identified
with the most easterly river flowing into the Euxine, the Rion.
 THE POEM'S DEBTORS
(i) would account for its ever being selected as an intercon-
tinental boundary at all; (ii) for the Tanais being selected as an
intercontinental boundary; (iii) for the tiny Rion assuming such
disproportionate importance. When Agathemerus says that the
Phasis was an older boundary than the Tanais he is referring
to stages (i) and (ii), not to stage (iii)-though without know-
ing it: for him the Phasis was the Phasis-Rion. If Aeschylus,
or his authority here, were at stage (ii) it could also account for his
speaking of the Tanais as well as the Phasis as the boundary of
Europe and Asia in the P. Sol. (schol. ad Dion. P. 10; though
we cannot certainly infer from this that Aeschylus specified the
Tanais here by name).21
If this view is correct we should expect the Tanais sometimes
to exhibit another trace of its former identification with the
Phasis: that is, to have the character of a strait connecting
the Palus Maeotis with Ocean. And so it does. According to the
scholiast on Ap. Rhod. iv. 284 Scymnus stated that the Argonauts
sailed by way of the Tanais to Ocean and thence back into the
Mediterranean, adding by way of explanation (1rapeK/30>..w,,:ra,)
that they manhandled the Argo at some point overland until
they reached the sea. This sea is presumably the outer sea or
Ocean, and Scymnus may have felt that it was necessary to ex-
plain how the Argonauts could proceed by ship from the source
of the Tanais to the Ocean (cf. Timaeus ap. Diod. Sic. iv. 56);
but I suspect that no explanation was necessary in the original
version, in which the Tanais-Phasis actually debouched into the
Ocean. This would accord not only with the earliest (Hesiodic)
form of the legend known to us, but also with the latest: for in
the Argonautica of 'Orpheus' the heroes sail right through into
the Arctic Ocean by way of a river whose continuations or
branches are the Araxes, Phasis, Tanais, Thermodon, and
Saranges (749 f.; 1036-82) !22
Again, Eratosthenes spoke of those who represented the con-
tinents as islands divided from each other by the Tanais and the
Nile (Strabo 65 ; Hecataeus seems to have joined the Nile to the
Ocean (fr. 302)). These, then, conceived of the Tanais as a strait
(and that such a conception was current is confirmed by the fact
 THE POEM'S DEBTORS 59
that the strait of the Cimmerian Bosporus was sometimes called
Tanais [note 21]) ; one such was presumably Pytheas of Marseilles,
for it is the obvious explanation of his claim to have explored the
whole ocean seaboard from Gades to the Tanais-which would
be a way of claiming to have explored the whole northern coast-
line of Europe (Strabo 104: 7Taaav J7r{J..80, 71v 1Tapw1<eav'i-rw rl;!.
E{,pW1T7/!. a7TO I'a8Elpwv lw, Tavai8os )-whatever river it was whose
mouth he took to be the northern outlet of the Don (cf. Thomson
146).
Finally, this same idea appears to underlie the opinion that
the Palus Maeotis was an inlet of Ocean: cf. Lucan iii. 277 (and
other passages quoted by Housman ad loc.):
quaque, fretum torrens, Maeotidos egerit undas
Pontus, et Herculeis aufertur gloria metis,
Oceanumque negant solas admittere Gades
and where the Euxine drains the rushing waters of the Maeotian
Mere through the strait ; and thus men deny that Gades alone lets
in the Ocean, and the Pillars of Hercules are robbed of their boast
(trans. J. D. Duff).
This is perhaps the place to mention another topic which,
while not germane to Aeschylus and lo's wanderings, may be
germane to the main inquiry. I have just said that 'Orpheus' con-
nects up the Araxes, Phasis, Tanais, and Ocean. The Tanais is
said to be a branch of the Araxes by Hecataeus (of Abdera, prob-
ably [ note 2 I]) and by Aristotle; the latter states that the Araxes
itself flows into the Hyrcanian (i.e. Caspian) sea (.Meteor. i. 13).
According to Strabo (512 f.) the Araxes flows by one outlet into
the Caspian, and by several into the northern ocean; and Mela
says that the Caspian intrudes itself into the land at first like
a river through a long and narrow strait (iii. 38: 'mare Caspium
ut angusto ita longo etiam freto primum terras quasi fluvius
inrumpit'). Now wonder has often been expressed at the lack of
notice of the great river Volga before Ptolemy (who calls it
Rha); on the other hand, there are two passages in Herodotus
where the Volga is probably meant by the Araxes. In iv. 11 the
nomad Scyths, formerly dwelling in Asia, are said under warlike
pressure from the Massagetae to have moved across the Araxes
 60 THE POEM'S DEBTORS
into the Cimmerian land, their present home. Here the Araxes is
very plausibly identified as the Volga (Ebert 82); and in i. 201 f.,
where the Massagetae are placed 'towards the east, beyond the
Araxes, facing the Issedooian folk' (a.VTlov BJ 'laG"YJ86vwv a.v8pwv),
a description of the river is given which is in fact a conflation of
at least two, one of which may well be the Volga (How and Wells
i. 152), while Herodotus' 'towards the east, beyond the Araxes'
suggests a river with a north-south orientation. Mention of the
Issedonians is noteworthy, and How and Wells remark the poetic
language ('laG"YJ06v1:s avl>pEs), and think of the Arimaspea as a
possible source (which would carry with it the corollary that the
Massagetae were mentioned in the poem).
It seems to me that ifin the beginning the Araxes-Volga was
held to be an offshoot of the Phasis-Tanais we may have the
germ of two later stories: (i) that the Araxes bifurcated into
the Caspian and, through the Tanais, into the Palus Maeotis;
(ii) that the Araxes flowed both into the Caspian and into the
northern ocean. We may also have the germ of the persistent
theory of Hellenistic and Roman times that the Caspian was an
inlet of the outer sea (cf. Mela, I.e.). Herodotus perhaps says
nothing on this head because of his scepticism respecting an en-
circling Ocean or an outer sea in the northern hemisphere (cf.
ii. 23; iv. 45. 1; iii. I 15. 2. Outer sea in the southern hemisphere:
i. 202. 4). To illustrate my point I have shown the Araxes-Volga
on Map I.
To return to Aeschylus. Where the Phasis was, there would be
Colchians; hence the Colchis of the P. V. neighbours upon the
Palus Maeotis, and the Amazons can be called 'inhabitants of
the Colchian land', Ko>.xlBos yas lvoKO (415).
When Io has crossed into Asia her course will still be east
(791); east too-towards the eastern Ethiopians (807 ff.)-after
a lacuna of unknown length before 792, where she is swimming
again : through what sea it is perhaps vain to ask, but the Caspian
springs to the mind. Apollodorus is not specific: 'After travelling
to Scythia and the Cimmerian land, after covering vast tracts of
land and swimming wide stretches of sea in Europe and Asia,
lo reached Egypt' (ii. I. 3).
 THE POEM'S DEBTORS 61

Now she will come to the plain of Cisthene, the country of the
Gorgons, where dwell in darkness the Gorgons' aged sisters, the
three Graeae with their single communal eye and tooth. These
are to be avoided, as also are the griffins and the Arimaspians
who dwell by the stream of Pluton :
o-rav 1Tp&.ar,s 1ki8pov ~mdpow opov, 790
' ' \' ,/..\ ~ ., Q -
1rpo, aVTOl\a, 't'"oyw1Ta., 7)1\LOOTLpEL,

I
1TOV'TOV ~
1Tti.pwaa. ,/._ \ " Q
't'I\OLa,.,ov, "
OT
t
a.v tt
E<;;LKTJ
, r.
1TpOS OpyOVLO.
I
1T0CO.
,OI V _IU,
Il.Luvr{ll1Jo, ,va

al <l>op1<.l8ti.r; valovai, 87Jvaial K6pa.L

... I ,/.. \ 11 ) , I
Tpnr; KVKvo.op'l'oL, KOWOV o.. KT1)JLVO.L, 795
.ov680V'Tti.S, 8s ov8' '1)ALOr; 1rpoa8lpKti.TO.L
dKT'iOLV ov8' ~ VVKTEpo, JL~V'I) 1rOT,
I\ N t ,0 \,/._ 0 , I
1Tti.l\ar; 0 aoti.l\'f'O.L TWVO Tpti.t<; KO.T0.1TUpo,,
8paKOVn5p.a.>...>..oL I'opyOVE<; ppowaTvyE'i<;,
as 8V'TJ7'0S oti8ds la,8c1v etn 1TVOO.<;" Soo
~ I ,/._ I \I
'TOLOVTO JJ,EV aoi TOVTO 'l'povpiov IIEJIW,
cVJ..71v o' O.KOVaov OVUXti.pfj
'i:
osva-ro.ovs yap
I \ , 8Ewp{av
z 71vor;\ aKpaynr;
,. ,wvar; I

ypvrras ,f,v.>..ata,, 'TOV T .ovviima aTpa.TOV

,,.. , r R, , " 1
.11.p,.aaTTOV L1T7To,...a.ov ' OL )(pV(ToppV'TOY 805
~
OLKOVULV ' .I..' va.a
a.'l'L - II'I\OVTWVDr; I
1TOpou.
,

TOVTOLS au.~ 1TAa,E,

When you cross the current which divides the continents, to-
wards the fiery-faced, sun-trod Levant ... crossing the sea-surge,
till you reach the plain of Cisthene, home of the Gorgons, where
dwell the three daughters of Phorcys, aged swan-shaped maids,
possessing one eye in common and one tooth; them the sun never
gazes upon with his rays, nor the moon by night. Nearby are
their three sisters, the winged Gorgons with snaky hair, bane of
men, whom no mortal may look upon and still draw breath. Such
are they that beset this place. And hear of another sight that you
will see with woe: beware the sharp-beaked griffins, hounds of
Zeus that bark not, and the host of one-eyed Arimaspians, the
horsemen who live about the spring of Pluton's stream that flows
with gold. Go not near them.
Where was Cisthene? A Thracian mountain ('Suidas' and
 THE POEM'S DEBTORS
Harpocration, incorrectly glossing Isocr. Paneg. 153) will not do
here; and the Medicean scholiast, who says it is 'a city of Libya
or Ethiopia', is probably guessing, from the traditional location of
the Gorgons in the west of Libya. That Cisthene on the farthest
edge of the earth, which Cratinus mentions, is probably ours
(fr. 309 : Kav8&o' lll"i dpo.7'a yijs iif,s Kai K,a8~V1JS opos oi/,,).
More information is lacking. To Aeschylus it was in the east, and
that is all that can be said.
But what were the Gorgons doing here? The ancients usually
placed them in the west-with one notable exception. Pindar says
'under the guidance of Athena, the son of Danae once came to the
throng of the Blessed Folk [the Hyperboreans]; and he killed the
Gorgon' (Pyth. x. 45) ; on which the scholiast exclaims rather
pettishly:
What was Perseus doing, going to the Hyperboreans to behead the
Gorgon? The Hyperboreans live in the southern (7rpos 7'ots 'EpvfJpalois
.lpm [these are rather Hypemotians-cf. Hdt. iv. 36!]) and northern
parts, but the Gorgons either in Ethiopia, that is, towards the south-
east (7rpos avaro.\~v Kai ,<17JJJ,/3pla.v), according to some authorities
[who probably meant the western Ethiopians, in fact], or, as others
have it, on the western limits of Libya. That there are no Gorgons in
the north is clear: for no one says there are.

There is a further oddity about the Gorgons' sisters, the Phorcides

or Graeae, inasmuch as Aeschylus characterizes them uniquely
as 'swan-shaped' (KvKvo.op~o,). The answers to these two prob-
lems may be related, and the consideration of them must be post
poned for the present [IoI f.]. Here I would merely say that the
location of the Gorgons in the north-east is compatible both with
Perseus' slaughter of Medusa during his visit to the Hyperboreans
and with Prometheus' warning to lo to avoid them on her way
to the eastern Ethiopians; for we need not suppose that Aeschylus
thought of the disk of the earth as very large. 23
This last point makes me think also that perhaps too much
has sometimes been made of the position of the griffins here.
By moving them from Apollo's Hyperboreans and posting them
before Zeus' Ethiopians, and by calling these Apolline beasts
'hounds of Zeus', Aeschylus, it has been suggested, is deliberately
 THE POEM'S DEBTORS
extolling Zeus at the expense of Apollo (cf. Meuli 154). In fact,
they need be no nearer to the Ethiopians than they are to the
Hyperboreans; not on lo's route, but near enough to demand
caution lest she deviate from it. I do not feel well qualified to judge
whether the title 'hounds of Zeus' is religious propaganda or not;
but I do feel that this phrase is dramatically effective in the con-
text, whereas 'hounds of Phoebus' would have smacked of
pedantry. Anyway, Zeus has some proprietary interest in griffins,
which are half eagle-which is also called a 'hound of Zeus' in
this same play ( 1 02 1 f. : LI tos 1<vwv) !
Of the Arimaspi2 I merely note now that we have early con-
firmation of their equestrian habits (Z1r1To/3a..ova.), which are not
mentioned by either Herodotus or the fragments (fr. 4. 3 need
mean no more than that they owned horses) [198].
The final lap of Io's journey (807-15), alongside the river
Aethiops to the Papyrus mountains, thence by the banks of
the Nile to Egypt, does not concern us; but it is noteworthy
how briefly Aeschylus dismisses it, and with what jejuneness
of detail compared with the preceding narrative. This may be
significant: however much or little it had fed his imagination
up to this point, the Arimaspea would certainly not have helped
him here.
Aeschylus' use of some authority, and that an early one, for
his northern geography seems indicated by several things. There
is enough system, and enough accuracy, to argue a fairly good
source of information; yet the overall vagueness, the intrusion of
mythological elements, the dearth of geographical names suggest
that this source was earlier than the Ionian scientists. This
dearth seems to be a defect which Aeschylus attempts to repair
by his invention of descriptive names for his rivers Hybristes and
Pluton; as though his source had merely referred to the one as
a turbulent river hard to cross, 2 s to the other as a gold~bearing
river. The Sauromatae/Maeotian-Amazons have not yet been
snugly accommodated to the Amazons of Greek legend, as they
have in Herodotus (whose disembarkation of his Amazons on
the western side of the Palus Maeotis in the first instance may be
a sop to this earlier account). The strange situation of the
 THE POEM'S DEBTORS
Caucasus and the Phasis points to a time before the eastern
Eux.ine had been properly explored and those features dis-
covered which were finally to receive and keep these names.
(So far from being the first regions of the Black Sea coast to
become known to the Greeks, as might be inferred from the
common interpretation of the Argonautic legend, the east and
the north-east seem in fact to have been the last. 26) Finally, the
identification of this early authority as the Arimaspea is suggested
by Aeschylus' mention of Arimaspians, grHfins, and the lair of
the North Wind (for wherever fr. 195 should be placed I have no
doubt that it refers to this); also perhaps by the expression
AEVK01TTpos v,tpd.s.
Aeschylus makes no direct mention of the griffins' role of
guardians of gold, though there seems to be a side-glance at this
in his gold-bearing river about which the Arimaspi live. On the
way this gold is obtained Herodotus is non-committal: 'Evidently
by far the greatest quantity of gold is to be found in the northerly
parts of Europe. Its provenance I cannot confidently declare:
but the story is that the one-eyed Arimaspians filch it away from
the guardianship of the griffins' (iii. 116: 1rpos O apK'T'ov 'T'Tjs
' I
EvpW'TT1)S "
1TOIVl<p \ \ ,I. I >I " \ I
n TT/\rnJTOS xpvaos 'f'a,Vf.Tat WV, OKWS fl,EV ywo,vos,
' w ~ ' ' ~ \' ~\ \ ~ ~
OUK EXW OIJO TOIJTO aTpEKEWS Et7Tat, /\E"/ETat 0 V'TTEK TWV ypV1TWV
This is clearly from
ap1rd.{e,v J4.p,.aa1TO~S av8pas ,ovvoipOd.A,ovs).
the Arimaspea, and adds something to Herodotus' statement in his
summary of its contents that the griffins were the guardians of
the gold : the Arimaspians were the thieves ; and the phrase {,TrK
Twv ypV1Twv apmf{nv shows that in the Arimaspea the Arimaspi
stole the gold and did not win it by doing battle with the griffins
(this is confirmed by the 'doublet' of the gold-guarding ants
shortly to be considered, and later authors who imply otherwise
(Pausanias' .d.xwOa, and Pliny's bellum [T. 4; 8]) are inexact and
were perhaps misled by the grypomachies of artists). But ifwe
want more details of Aristeas' account we must search later
authorities.
The paradoxographer excerpted by Pliny, whose date was
about the turn of the era, states that the griffins throw up the gold
when they make their burrows (eruente ex cuniculis aurum); the
 THE POEM'S DEBTORS 65
authorities for this passage are said to be Aristeas, Herodotus,
and many other writers [28]. In spite of this plurality, we have
seen reason, from the mention of the starting-point of the North
Wind, and of the feather-filled region in the related passage of
Pliny [41], to accept the Arimaspea provisionally as the fountain-
head of all that we are told in this paragraph [T. 8]. Pausanias
reports Aristeas (though not at first hand) as saying that the
gold which the griffins guard is 'produced by the earth' ('rov Si
i(pVU<>V, 8v cpvAaaaovow ot ypiJm,.s, 0.VLEVat TYJV yriv) [32]. This
phrase, with its undertone of spontaneous production-as if the
gold sprang up from the earth like a crop-at first appears to
contradict the Plinian account ; yet that it is not obligatory to
assume any such contradiction may be shown by Plato, Crat.
403a: the god of the underworld gets his name Pluto from his
bestowal of wealth upon us, 'inasmuch as wealth is extruded from
the depths of the earth' (on EK Tfjs yi]s- KchwBt:v avieraL o1rAovros)
-though here not mineral but cereal wealth may be meant (c
Nilsson; 32). I think that all that Pausanias' expression is in-
tended to convey is that, where the griffins live, the gold is near
the surface of the earth-the topsoil is x.pvafn,.
That the gold was dug up by the griffins is also the assertion of
Aelian, using Ctesias (ft. end of the fifth century B.c.), in a most
detailed and colourful description of the creatures and their
antagonists.
I hear [writes Aelian] that the griffin of India has four legs like a
lion, with claws as strong as strong can be, themselves very like lion's
claws. It is winged, and the plumage on its back is reputed to be black,
on its breast red, while its wings are neither of these colours, but white.
Ctesias relates that its neck is bedecked with dark-blue feathers, that
it has the beak of an eagle, and a head such as the artists and sculptors
portray. Its eyes, he says, are fiery. It makes its nest in the mountains,
and, though it is quite impossible to take a fully-grown griffin, they do
sometimes capture the chicks. The Bactrians who are the Indians'
neighbours say that they guard the gold of those parts, which, when
they have dug it up, they weave into their nests, while the Indians
take the residue. The Indians themselves, however, deny that the
creatures intentionally guard the stuff: for, they say, grifiins have no
need of gold (and if they do say this they seem to me to be talking
d850 F
 66 THE POEM'S DEBTORS
sense); the truth being that on the gold-prospectors' approach they
take fright for their young, and so give battle to the intruders. They
never oppose a lion or an elephant, though they fight and easily over-
come any other animal.
The natives, out of respect for the mettle of these beasts, avoid
making their gold-hunting expeditions by day. They come in the
night-time, when they have most chance of avoiding discovery. Now
the place where the griffins live and where the gold is to be found is
a howling desert. Into this, waiting for some moonless night, come the
treasure-seekers in armed bodies of one or two thousand men, with
shovels and sacks, and dig. If the griffins do not notice them, they
reap a double reward, for their own lives are preserved and they
bring home their cargo, which those who have the peculiar art of the
goldsmith purify and render into a rich recompense for the dangers
they have undergone; but if they are caught in the act it is all over
with them. I am informed that they return home after an absence of
two or three years (N.A. iv. 27; cf. Ctes. Ind. 26 = Phot. Bihl. 46b).

This pleasing story looks uncomfortably like that which Hero-

dotus tells, on Persian authority, about the gold-guarding ants
(iii. 102 ff.). These inhabited the Bactrian desert, in shape like
Greek ants, but in size intermediate between foxes and dogs.
They had their homes in burrows, and in the process of making
these they extruded sand which contained gold. This the Indians
who bordered on the desert would gather in the following manner.
They would yoke together camels in groups of three, two males
with one female who had recently foaled; then, choosing the heat
of the day, when the ants retire underground for a nap, they
would approach, pack the sand into the saddlebags of the male
camels, and make off at top speed. The ants, SJl\elling the thieves,
would emerge and give chase ; and so fleet-footed were the insects
that were it not for the start the Indians gained while their
enemies were mustering, and the eagerness of the female camels
to get back to their young, not a man would reach home. A
similar tale is told by Megasthenes about a century and a half
later, perhaps from an Indian source, though his account has
reminiscences of, as well as additions to and sometimes disagree-
ments with, Herodotus (fr. 23 = Strabo 706; Arrian, Ind. 15).27
Clearly then it is Ctesias' griffins that are the interlopers, and we
 THE POEM'S DEBTORS
need suppose no more than that their author has let his ima-
gination (aided by some pictures of griffins?) work upon a con-
flation of three adjacent passages of Herodotus-iii. 102 ff. on
the gold-guarding ants and the Indians; iii. 116 on the gold-
guarding griffins and the Arimaspi ; and iii. 111 on the roe-like
birds who used cinnamon, coveted by the local Arabians, to build
their nests, which were plastered by means of mud against moun-
tain precipices (vEOC1C7WS' 7rpO(]'TTE7TAaul.va<; JK 7n/AOV 7Tpoc; &.7To-
'
KP"Jl'-VOl,(n "
opEC7: Cf A e1., 1.c. VEOTna<;
' OE
s:,, E'TTL
' 'TWV
- opwv
- 'TTOLEL'Tat A

. , [ - -J \ -~'
, , Ka.t EK TOVTOV SC, 'TOV XPVUOV 'TO.S' l(a.,\l,a<; V1T01TI\EKEL V ; an d
f ,, ()

xvii. 21).
The ripples of the ant-griffin confusion spread down anti-
quity. Nearchus said that the skins of these ants were like those
of leopards in appearance (Strabo 705.ftn.); Pausanias reproves
those who say that griffins have spots like leopards (viii. 2. 7).
According to comm. Bern. ad Luc. vii. 756 the ants dwell near the
Arimaspians: 'ARIMASPUs Scythiae populus. aput quos vpp.'f'/KE<;
sunt quaedam animalia formicis similia quae terras eradunt et
repertum aurum egerunt.' For Aelian they dwell near the
Arimaspians' neighbours, the Issedonians, and while carrying
out their duty of guarding the gold will go no farther than a cer-
tain river: 'The ants of India in guarding the gold will not cross
the river Campylinus. The neighbours of these ants are called,
and are, Issedonians' (.N.A. iii. 4: ot vpp.TJKES ol 'lv8tKOI TDV
I ,J. \ I t I t, I\{) , _ \ I U I\
xpvuov .,,vlla'TTOV'TES' ou,c a.v otEA OEV -rov K(J.l\ovEVov n.am,111vov
'ltOTaov.
I '[,
UC7"1/00VES
t, , t, '
OE <0L'> J
'TOt.l'TOtS' ~
<YVVOLKOVVTES 'TOL<; up'f/!,L
A , '

KaAouvral TE ,ea[ elcnv 2S). I take it that here the Issedonians are
the robbers, and that if they can pass the river with their spoils
the ants give up the chase (cf. the Gallic serpents who pursue the
mounted robber of their magic egg 'donec arceantur amnis
alicuius interventu' (Plin . .N.H. xxix. 52). Is it relevant that
supernatural beings often cannot cross water?).:9
If this is so, Aelian is, of course, carrying the confusion a stage
farther with his lssedonian gold-thieves. Yet the mention of the
nearby river is interesting, and may actually be a remote echo
of something in the Arimaspea itself. The Hellenistic writer Zeno-
themis, in a couplet of his Periplus preserved by Tzetzes (Ghil.
68 THE POEM'S DEBTORS
vii. 683) says: 'Marching with the Arimaspians is the great tribe
of the Issedonians belonging to Scythia ( ?) , by the springs of
the river. ' :
mfyxoprov 8' }:lp,.acmotu,(v) valEt lya tf,v>.ov
'[c,C17J8Wv J:Kv8[71s VO.a.at '7T<Lp 7TOTaofi

The form 'lu<nJSol is notable: though vouched for by Stephanus,

it only otherwise survives in fr. 3 of the Arimaspea [8 f.] ; also
va.ao,: in the P. V. the Arimaspi dwell about the 118..a of the river
Pluton [6 I]. Zenothemis probably gave the name of his river in the
next line. Aelian's Campylinus is a possible candidate; or is this
name only a distortion of another Asiatic river, the Campasus?
This latter features in a fragment of the Apollo of Simias
(ft. early third century B.c.), which is of such interest that I
quote it in full (Tz. Chit. vii. 695; Powell, Collectanea Alexandrina
109):
\ I
7"YJ"V'J'ETwv >.J.
o~ a'l'v1:,011
I
, 7Ttcp /3 op1:wv
<V f > I ~~
ava 071.ov,
TOLS 0~ Ka[ '7TOT' avae 7]pWS '1Tap1:8a{aaTO flEpCltil)S,
fv8a 8J Marraaylrm <TE) ()oC,v brt/3~Top1:; (7T'7Twv
valovrr,(v) .,.6go,a, '7TETTod}6us WKv/36,\0,ai(v),
()0'1TEaL6v TE 7TtcpL p6ov -/j.\v8ov d.1:v&.0,0 5
o; <9') al\a "\ o,av ~- ES '8 avaT7jV '!''=PE
1
.n.a.7raaov,
1T > a I I .J..
-~
vowp.
JK 8' lK671v JM,-raia, 1TEp2 x>.wpfia,v lp1:.vas
I t,/, I 1 ~ I ~ I
VIJUOVS, v.,,,Kooimv E7T7JPE'f'Eas OOVaKECTCTW,
~iKVvwv -r' Jv671aa ylvos 7TEpul.,a,ov dv8pwv,
-ro,; w.wv Ka 8'mrep 8JI VCTTPE'f'EWII
# 1 J..I
KVJIEO;
I
Kpas
'
IO
I .J. .,.J.. \~ I > I
TETPO'/' ya,.,..,,7Jl\'[ll1 7TEpLKpaTEEC1C1W Epvvos.
TWJI iv 8' WCTTE KVVWV vAaK,} 7TtAE, ov8l n TolyE
."\\ , ' R ~ ' ' \ ,~ I
a/\1\WV ayvwaaovaL ,-,poTWV 0110.aKI\VTOV aVOTJII

And passing through the rich land of the far-off Hyperboreans,

who once feasted the princely hero Perseus and where dwell the
Massagetae, riders of fleet steeds, whose trusty weapon is the swift-
sped arrow, I approached the wondrous stream of ever-flowing
Campasus, which rolls its waters to the divine immortal sea. And
my way lay by islands dark with green firs, 30 overgrown with lofty
reeds; and I remarked the monstrous race of men who are half
dog, upon whose supple necks is set a canine head armed with
powerful jaws. Like dogs too they bark, yet comprehend the articu-
late ( ?) speech of other men.
 THE POEM'S DEBTORS
This combines the marvels of the paradoxographer with the
orderly progress of the guide-book: we pass from the Hyper-
boreans by way of the Massagetae to the giant reeds and Dog-
men (of India ?).3 1 But the effect of speed is remarkable, as also
the apparent disregard of water-obstacles in reaching the islands
of I. 8 : who is this traveller who speaks in the first person, and
how was he travelling?
Antoninus Liberalis (Met. 20) tells the story of the Babylonian
Cleinis, who was so friendly with Apollo that he often accom-
panied the god on his visits to his temple among the Hyper-
boreans and witnessed the ass-sacrifices there (oov Tots 8ors
roVTo,s [Apollo and Artemis] a.tpt1<e-ro 1rpos 'TDV vaov TOV )b-6.>J.wvos
, 'Y.
10v32 EV
I ,QI
7rEpf'opeo,s \ fC:,
1<a, ewev t I
,epovpyovevas av-rcp~ -ras umnas
I \ ll I

1/v tvwv). Later, however, he got into trouble by trying to

sacrifice asses at home on his own account, and Apollo finally
turned the whole family into birds. The scholiast says that the
story is told by Boeus in the second book (of his Ornithogonia), and
by Simias in the Apollo. How much of it was related by Simias
wedo not know, but we may guess fairly confidently at the friend-
ship with Apollo and the visit to the Hyperboreans, for the ass-
sacrifice, which is only welcome to the god when performed by
the Hyperboreans, is fundamental to the story. It comes, of
course, from Pind. Pyth. x. 33 ff., and the second line of Simias'
fragment is inspired by the line almost preceding this (31: 1rap'
ok 1TO'Tf. llEpcu,vs JoalcmTo AaylTas). I approve, therefore, of the
view that Cleinis is the speaker here (cf. Powell), and believe
that he is giving an account of a ride in the god's chariotll on
the way back from the Hyperboreans. That the tale is not
mythology but fourth-century or Alexandrian romantic invention
is shown by the fact that there is no mention of Cleinis in earlier
literature, and that he is a Babylonian; but the inventor, in de-
picting his hero as 'seized by Apollo', cf,oip6>-.TJ1rTos (in a literal
sense-as carried to and from the Hyperboreans through the air in
the god's chariot3), and as describing the peoples and marvels
on his route, was following a tradition about Aristeas [ 121 f.].
The Campasus, then, is situated in the north-east near the
Massagetae, and is no doubt to be identified with the river called
 70 THE POEM'S DEBTORS
by Pliny Caspasus, which flows through the country of the Asiatic
Scyths, who comprise the Sacae, Massagetae, Essedonians [184],
Arimaspi, and others (N.H. vi. 51). I do not, however, think that
the alteration of KamJ,uov in I. 6 here to Karnrduov (cf. Powell)
on this account is justified, and would prefer to emend the text
of Pliny, if emendation is needed and it is not just a Inistake of
Pliny or his authority (by analogy with Caspius?). For the form
given by Tzetzes is supported by Campesus in Valerius Flaccus
(v. 593 al.). There it is the name of a hero, an ally of Aeetes, but
it is taken from the river in the way Valerius coins names for
a number of his barbarians.3s
There is something wrong with the text of the remainder of
" aAa o,av ' a'8ava'T7/v
' vowp os p B runek , os 8' ego;
0 0 0 O
1 6, os ' <- ES y,EpE,
,1. ' " ( ''

ofov Harberton), but we hardly need to follow Powell so far as to

adopt Bergk's Els ~a oiav OS de&.vaTOv ,f,l.pn iJowp. The alteration
of d8ava'T7/v to d8avaTov anyway introduces a difficulty of sense.
Most rivers flow into the sea, and we could guess for ourselves
that the Campasus does without being told so. But we should be
grateful for the information, which sea. And this information
d8ava'T7}V supplies : the immortal Ocean is meant. J6
For this reason I have tentatively shown Aeschylus' Pluton as
flowing into the Ocean [Map I] ; the river about whose upper
reaches the Arimaspi dwell may be the same as that about whose
upper reaches their Issedonian neighbours dwell (Zenothemis) :
the river which Aeschylus calls Pluton and some later writers
Campylinus or Campasus [but cf. 115 and I 18].
I turn now to a contemporary of Aeschylus. It may be that
Pindar knew something about Aristeas himself[ 127 ff.], but it is not
clear whether he knew the Arimaspea. In his extant work there is
no mention of griffins, Arimaspians, or Rhipaean mountains. Of
the customs of the Scythians there are two curiously intimate little
pictures : one, of the social ostracism incurred among the nomad
Scyths by him who has no wagon-home home (fr. 94. 4: vodotuu,
yap' EV JKV
'["' '8 at, aAa7'at
A1\ A <\ t ,L f
U7'pa7'WV os aa.,,oy,Op'tj7'0V
f ~
0KOV ) f )
OU 1Tf!1T0.7'0. ;

the other the grotesque fragment 192, about people who by day-
light profess a distaste for dead horseflesh but by night sneak out
and devour it avidly. But the only real indication of knowledge of
 THE POEM'S DEBTORS 71
the Arimaspea is Pindar's location of the Gorgons in the region of
the Hyperboreans [62]. Whether he had the journey of Aristeas
in mind when he wrote that one cannot reach the Hyperboreans
'by ship or by foot' (Pyth. x. 29) is a question interesting but
unanswerable.
In spite of thus placing them firmly in another world, Pindar
has quite a lot to say about the Hyperboreans. Theirs is an idyllic
life (Pyth. x. 31 ff.; fr. 272), passed in song and dance, feasting
and worship of their god Apollo. They know not disease or old
age, though their span of life is a thousand years [but cf. 99].
Toil, war, and Nemesis are strangers to this holy race:
voao, 8' oih-E yfjpac; ofJAOJLEVOV KKpaTa,
lepfl. yevef!. 1Tovwv 8J Ka! 1-'axiiv ,frep
~ , ,/.., I
o,KEo,a, 'f'vyOVTE<;
inrlp8,Kov Nlww.

There is no need to suppose any other source than his own

genius for Pindar's details; but it is worth remarking a similarity
of the just and sacrosanct Hyperboreans with the northern
Argippaei of Herodotus [ 42] : they too are unwarlike, inviolate,
and under divine protection (ToV'Tovs ovStS' d.SiKfo d.v/Jpc!mwv ([pot
yap Myovrat etvm), ov8e 'Tt &.p~iov 01T).ov (KTEa-rai), arbiters in their
neighbours' quarrels and a safe asylum for fugitives; trees provide
their food, trees their shelter (iv. 23. They appear in Latin writers
as Aremphaei or Arimphaei: Mela i. 117; Pliny, N.H. vi. 34 ).
In fact, the two characteristics of justice and vegetarianism
are ascribed to the northern barbarians by Horner (Il. xiii. 5:
ayavwv '11T7T'11Jl,Of\YWV
I A I\ \ OI >,1 Qt .., , > /)
'Yf\alCTO<paywv, .'i,-,tWV TE Otl\'.alOTOTWt' av pw-
Tl'wv). Probably for Homer they were just because they were
communists, as Strabo suggests (300 f.), not because they were
vegetarians ;37 but by the time of Aeschylus their justice seems to
be associated causally with their abstention from flesh: 'The
Scythians, eaters of mare's-milk cheese, <? are) a well-ordered
people' (fr. 198: ill' i1rm1.K'1JS' pwrrypEs Evvooi Ed/Ja,. Cf. Epho-
rus ap. Strab. 302). Hellanicus, a contemporary of Herodotus,
makes this connexion explicit in a claim for the Hyperbo-
reans themselves, that 'they practise righteousness by abstain-
ing from meat and living on fruit' ; and this shows a sign of
 THE POEM'S DEBTORS
deriving from Aristeas, as Hellanicus also says that the Hyper-
boreans 'live beyond the Rhipaean mountains' (fr. 187b: Tovs
~ , v
OE R ,
~ 7TEp,.,opE01JS"
'E.'', ,
NlaVtKOS" , , Ta, r1Tata
VTrep n "
op71
0KtV ,
urropEi
~
\>.II' ._/l
owaaKE=at ~'
oE
'
avTovs C' '
otKatoavV'T)V 77' ,L
1<p1:o.,,ayovnas, ~'\'\>
Cl/I/\
'
aKpo
OpVOLS" xpwlvovs), 38
The review of passages which by their dependence, direct or
indirect, on the Arimaspea may help to throw light upon its con-
tents is now almost complete. In the second chapter of Book I
of Dionysius' Ornithiaca we are told that the author 'has nothing
firm to report about the griffins, which they say live in the water-
meadows of the wealthy Arimaspae and dig up and collect gold
from the earth; the Arimaspae, so it is said, have one eye each
instead of two' (.77oev V7TEp ypv,rwv ua<f,es lunv El7TELV, OVS" cpaaw
Jv Tois TWV '7TAOVCJlwv :4.piaa1rwv EAEUt 7p1:<pop,lvovs xpvaov (.I( 7~S
>A , t'' ' , \ ~ I
y77c; ay1:ip1:w opvuuovras, rovs npin.cnras
-.. ) ' ' I '
oe -rovTovs ovxi ovo Tovs
\

.1.ea,'JM'vs,
o..,, i .. - ' ~,,,"
a/V\ "
eva eJ<a(]'l'ov "
EXEtv ,1. ' ) W e possess th'1s poem
't'auiv
only in a prose Byzantine paraphrase; the poet has been thought
with some reason to have been Dionysius Periegetes,39 which
would date him to the early second century A,D. From his scant
knowledge about the griffins and his use of the phrase 'they say'
it would seem that Dionysius did not know the Arimaspea at first
hand (the form }1.p,a.a"rrqs may be an inference by the paraphraser
from the accent of the genitive plural; but cf. Orph. Arg. 1063);
but the words d<f,Oa>.oos Eva EKaaTov exEw are so like Aristeas'
drf,8a11.ov 8' ev' EKaUTOS" EX' that one wonders if Dionysius had not
come upon the same fragment as Tzetzes (fr. 5. 1). This would
mean that what was only a quotation in the twelfth century was
still no more than a quotation in the second. As for the little
Dionysius tells us, we have heard most of it before, but the state-
ment that the Arimaspi lived in iA7J, 'water-meadows', is interest-
ing. The only other authority to say that the Arimaspi dwelt by
a riveris Aeschylus [6 r], and he was using theArimaspea apparently.
It looks, then, as ifDionysius, albeit indirectly, is preserving some
genuine lore from that poem. If so, and if fr. 5. 1 was associated
with it in Dionysius' source, weight will have been added to the
view that the fragments are authentic.
It remains to mention that griffins occurred in some Hesiodic
 THE POEM'S DEBTORS 73
poem, perhaps the third book of the Catalogus (see fr. 4-9 T ., and
F. Gisinger in R.M. lxxviii (1929), 319); anyway, the Medicean
scholiast on P. V. 830 claims that 'Hesiod was the first to portray
the monstrous gn'ffins, (1rpwTos <Hawoo,;
A , "' ' '
erepa.Tevau:ro '
Tovs
ypiJ1ras). Nothing can be built upon this claim. The scholiast may
have had his own views on the relative dating of 'Hesiod' and
Aristeas, or, more probably, have been ignorant of the Arimaspea.
Though Hyperboreans, Rhipaean mountains, griffins, and
Arimaspi (and even Essedonians [184-], in Lucan and Valerius
Flaccus) are known to Latin poets, they are no more than learned
stock-in-trade, and throw no further light on their ultimate source.
There are two possible exceptions, both early and isolated
references to the Rhipaean mountains as gold-bearing. The slave
of Lyconides in Plautus' Aulularia, when he has stolen the miser's
hoard, jubilantly cries out that he is wealthier than the griffins
who inhabit the golden mountains (701: 'picis40 divitiis, qui
aureosmontescolunt, egosolus supero'); and againEnnius speaks
of nuggets v.s. 'which ten one-eyed men dug up on the Rhipaean
peaks' ('<massas) . . . decem coclites quas montibus summis
Ripaeis fodere' Varro, L.L. vii. 71 41 ). There is perhaps nothing
more in the Plautine passage than hyperbole (Wagner compares
Stick. 30 'neque ille sibi mereat Persarum montis, qui esse aurei
perhibentur'), or in the Ennian than inexactitude; but there will
be cause to recall these later [97; 100].
It is quite uncertain-and indeed at present unknowable--
how far Hecataeus ofAbdera, in the fourth century B.C., was using,
or imitating, the Arimaspea in his book on the Hyperboreans. The
fragments of this work deal mainly with the Hyperboreans them-
selves, their customs and their country, whither the author of the
Arimaspea did not come. It is a pity that we do not know how
Hecataeus professed to have gained his information.Jacoby says
that we shall hardly be unfair to him in seeing him as a direct
successor to Aristeas. 42 It is possible that Hecataeus had in mind
a certain tradition about Aristeas, as I conjectured Simias may
have done [69]; but that we may be unfair to Aristeas by classing
him with a mere romancer like Hecataeus of Abdera will be
suggested by what follows.
 IV

TALES OF A TRAVELLER

IT is now time to make a tentative reconstruction of the contents

of the Arimaspea. The author stated in his poem that, possessed by
Apollo, he journeyed beyond the Scyths to a neighbouring people,
the long-haired Issedonians dwelling by a river. This was the
farthest point he reached, but the Issedonians (with whom he
presumably spent some time, if it is correct that he was absent
from home for six years) gave him an account of the regions
beyond. Therein was an immensely high mountain-range, covered
in darkness and snow (and wooded?), where was the home of a
wind in a cavern, a wind identified by Aristeas as Boreas, which
at home would have been the North Wind, but perhaps de-
scribed by his informants merely as a cold and powerful wind.
Visibility in these parts was further reduced by a rain of feathers
which perpetually filled the air. This mountain-range was not
called 'Rhipaean' until a later date; the Issedonians may have
given it a name which Aristeas rendered as 'Caucasus', Beyond
it lived righteous vegetarians, whose domain extended to the
outer sea; this folk Aristeas took to be the Hyperboreans, the
apparent goal of his journey.
The area between the mountains and the Issedonians was rich
in gold (or possibly the gold-bearing region was in the mountains
themselves). This was guarded, having been thrown up in the
process of making their lairs, by monsters, called by the poet
griffins, from which it was stolen by certain warlike neighbours
of the Issedonians, the Arimaspi. These were one-eyed, hairy
savages, equestrian nomads with a wealth of cattle, sheep, and
goats. They lived now about a gold-bearing river (perhaps flow-
ing into the northern ocean, and perhaps the same as the river
about which the Issedonians lived-though these are little more
than guesses), but they were not stationary, and the Issedonians
 TALES OF A TRAVELLER 75
themselves were compelled to shift their abode under the hostile
pressure of the Arimaspi. So it happened that the Issedonians
in their tum drove their neighbours on the other side, the Scyths,
from their territory; and this Scythian migration caused the
departure of the Cimmerians from their home on the northern
shores of the Black Sea.
The Issedonians may have told Aristeas other stories too: of
cannibals; of another kind of monster inhabiting the region of
darkness, which the poet equated with the Gorgons and Phorcides
of Greek legend; of giant reeds, and of dog-headed men who
understood the human tongue though they had not the use ofit;
perhaps also of some kind of mermen. And Aristeas must have
included in his work first-hand observations of the peoples
through whom he passed : possibly he told of the Scyths' working
iron, living in horse-drawn wagons, eating horse-flesh and mares' -
milk cheese, and how they claimed to be three nations descended
from the three sons of the patriarch Colaxes ; of a nation whose
women were so manly that he took them for Amazons; of the
difficulties of river-crossings, and of one broad river which he
thought was the Phasis of the Argonautic story, joining the inner
to the outer sea.
Was all this fact, or was it really nothing more than the work-
ings of some stay-at-home Greek's fancy? Were the very fantasies
mere creations of one poetic imagination, or were they genuine
bits of foreign folk-lore, obtained from a far-off people either
at first hand (as the poet professed), or at second hand through
some itinerant intermediary? I have anticipated my answer by
the phraseology of the above summary: I believe that the
Arimaspea comprised someone's first-hand observations of the
Eurasian hinterland, and items of Asiatic folk-lore interpreted,
not unnaturally, in terms that a Greek would understand.
At the outset one is predisposed in favour of Aristeas' good
faith by his confession that he got no farther than the Issedonians,
and that all he had to say of the parts beyond he owed to their re-
port. This is a mark either of extreme honesty or of cunning both
extraordinary and unnecessary: for why should a poet trouble
to disarm his audience thus? This favourable first impression is
 TALES OF A TRAVELLER
fortified by Aristeas' account of how the Scyths came to possess
their country to the north of the Black Sea : his description of
the migratory pressures of the nomads of the Eurasian steppes is
historically correct, and stands out in sharp contrast to what
might be called the official Greek and Scythian accounts of the
same thing, with their trappings of the marvellous and their
genealogy of the Scyths traced from Heracles or some mythical
royal ancestor (Hdt. iv. 5-10).
I have already said that Aristeas seems to have been the only
Greek who ever claimed to have reached the Issedonians. Hero-
dotus locates them beyond the terminus of the regular trade route,
yet he is able to give some account of their customs. It is reason-
able, therefore, to suppose that he owes this account to the
Arimaspea. This is what he says of them (iv. 26):
The Issedonians are said to have the following customs. When a
man's father dies, all the near relatives bring sheep; which are sacri-
ficed, and their flesh cut in pieces, while the body of their host's father
undergoes the like treatment. The two sorts of flesh are afterwards
mixed together, and the whole is served up at a banquet. The head
of the dead man is stripped bare, cleansed, and coated with gold
(KaTaxpvuo-Oa,: 'set in gold' Rawlinson). This they treat as something
precious and holy, to which they make important sacrifices every
(
# > .!\
year aTE ayw,.a-ri 8 .!\ > > \
XPEWV'TaL, va,as .Eya/\aS E7TE'TEtOVS E7T,TEI\EOl'TS' : 1t
I I I I ,

then becomes an ornament on which they pride themselves, and is

brought out year by year at the great festival which sons keep in honour
of their fathers' death' Rawlinson, which I think blunts the point).
A son performs this duty for his father, just as the Greeks observe the
Genesia [probably an aside by Herodotus himself]. In other respects
the Issedonians are reputed to be observers ofjustice, and their women
to have equal authority with the men.
The customs of the Issedonians, then, included the ceremonial
eating of the deceased father; the preservation ofhis skull by his
family; and equality of the sexes. It is as well to be clear about the
true significance of these, as the commentators arc sometimes
misleading.
In the first place we must be careful to distinguish the canni-
balism practised upon the bodies ofenemies or executed criminals,
ascribed by various authors to the Tibetans, the Tatars, and the
 TALES OF A TRAVELLER 77
Chinese (Yule i. 301, 311 ff.), from the cannibalism practised by
a man's kin upon him as an act of piety, as here. Such a custom,
sometimes anticipating nature by the slaying of the sick or aged
victim, was referred to the Massagetae, to their near neighbours
the Derbices, to the Indian Callatiae and Padaei, and even to the
Irish (Hdt. i. 216, cf. Strabo 513; Strabo 520, cf. Ael. V.H. iv. I;
Hdt.ili. 38, 99; Strabo 201). In more modem times it is attested for
tribes in India, the Arakan, Sumatra, Australia, and South Ame-
rica, as well as for the Samoyeds and Ostiaks (E.R.E. iii. 202 f.;
ix.465); so we cannot, with Hermann (2246), see here a possible
indication that the Issedonians were oflndo-Germanicstock. Thir-
teenth-century travellers said that it had been a Tibetan custom:
These people haue a strange or rather a miserable kinde of custome.
For when anie mans father deceaseth, he assembleth all his kindred,
and theyeatehim (Carpini) ... Men which were wont to eate the car-
kases of their deceased parents: that for pities sake, they might make
no other sepulchre for them, then their owne bowels. Howbeit of
late they haue left off this custome, because that thereby they became
abominable and odious vnto al other nations. Notwithstanding vnto
this day they make fine cups of the skuls of their parents, to the ende
that when they drinke out of them, they may amidst all their iollities
and delights call their dead parents to remembrance. This was tolde
mee by one that saw it (Rubruquis) (Beazley I 18, 232).
Ceremonial eating of the dead by Tibetan lamas is reported as a
fact even for the end of the last century by A. H. S. Landor, who
adds that if there are insufficient lamas available to perform the
rite the relatives expose the body to be devoured by animals or
time, having first themselves partaken of a morsel of the flesh
(ii. 68 ff.) ; 1 in this last we might perhaps see the atrophied token
of an earlier practice when the entire corpse was consumed.
It will be noticed that Rubruquis says that the Tibetans made
the skulls of their parents into drinking-cups. Their use of human
skulls for this purpose is vouched for elsewhere (et: Yule i. 31:2),
and Landor saw and sketched such a cup, which he says were to
be found in all the monasteries (ii. 71 f.). These were not family
possessions, but were used by the lamas in magical ceremonies,
and it may have been such a ceremony that Rubruquis's infor-
mant witnessed-and misunderstood.
 78 TALES OF A TRAVELLER
Herodotus states that the Scythians were wont to make drinking-
cups of human skulls, but they were the skulls of enemies, or of
kinsmen with whom they had quarrelled. They sawed off the
cranium, lined the outside with leather and the inside (if they
could afford it) with gold (luw8v KaTaXPvuwua~); it was then
produced at parties, and evoked suitable reminiscences from the
proud host (iv. 65). It is this custom of the Scythians to which
Plato refers in the Euthydemus (299). It is reported also of the Boii,
the Lombards, and other primitive peoples (E.R.E. vi. 535). But
there is no hint in Herodotus that this was the purpose for which
the /ssedonians preserved their fathers' skulls, and I believe Raw-
linson, How and Wells, and Hermann (2246) to be mistaken in
thinking that it was. J. A. MacCulloch is surely right in seeing it
as an example of the early and widespread practice of preserving
the heads of ancestors in order to maintain communion with
them or to retain their protection and good offices in return
for the honour paid them (E.R.E., I.e.). The skulls become cult
objects, the focus of annual sacrifices. The Issedonians, then,
went in for ancestor-worship, of which the respect shown to the
skull was one expression and the ceremonial eating of the body
another.
The Issedonian treatment of these skulls finds, I think, a re-
markable illustration in a recent discovery on the site of pre-
pottery neolithic Jericho. The inhabitants of this place often
detached the skulls of their dead; and a number of such skulls
has been recovered which had had the features restored in an
overcoat of plaster. The care with which this had been done, and
the fact that headless burials were found in successive levels of
excavation, indicate that the skulls were not mere trophies but
were those of venerated ancestors used for cult purposes. The sex
of these portrait-skulls was probably male in all cases; the radio-
carbon age of two samples comparatively high up in the pre-
pottery sequence was given as c. 6250 B.c. and c. 5850 B.c.
(Kathleen M. Kenyon and others in Antiquiry xxvii ( 1953), 105 f.;
xxviii (1954-), 198; xxx (1956), 196 f., 187). We can perhaps see
here a clue to the meaning of Herodotus' Karaxpvc,ovu,: the
inhabitants of Jericho reproduced the features of their dead in
 TALES OF A TRAVELLER 79
plaster, with which they coated the outside of the skull; the
Issedonians did the same, but their medium was gold.
It remains to note on the third characteristic of the Issedonians
recorded by Herodotus, the equality of their women with the
men, that Rawlinson goes too far in seeing in it a hint of matri-
archy. Such would be extraordinary in a society which treated its
dead fathers with religious veneration; Herodotus says only that
the women were woKpaTEffS To'io', a.vl>pa.u,; and How and Wells are
probably correct in seeing here no more than an allusion tci the
sharing by the women of activities which in other societies are
usually a masculine prerogative. The Issedonian ladies may have
been like those of the Hazaras of north Afghanistan, a pastoral
and agricultural Mongol tribe, of whom Josiah Harlan had the
following to say in 1840 (150; cf. 148):
The wife and husband amongst the Hazarrahs are inseparable in
public affairs. She sits with her husband, in the divan, dressed like him
and booted, ready for the chase or even a military foray! They would
not go upon a distant expedition, but in civil dissentions and in
border difficulties, to which their excursions are mostly confined, they
generally participate. In the chase both sexes use the fire-lock expertly
and accurately. They will gallop their unshod horses over a precipi-
tous deer path, regardless of danger, and bring down the game at full
speed. Females of the poor manage the household duties, assist in
tending flocks, bringing in thorns, carrying water, and all the hard,
laborious work. 2
These customs, though so far from being fantastic as to find
parallels in many ages and in many parts of the world, would
yet be so foreign to the mind of a seventh-century Greek that it is
easier to ascribe them to first-hand observation and to regard
them as the historical fact that they were supposed to be than as
the inventions of their recorder's imagination. They will then be
a further vindication of Aristeas' veracity.
The Issedonians were sheep-rearers; but it would be rash for
us, having in mind the example of the Budini [ 104], to conclude
from this that they were nomads of the steppes. I defer further
consideration of their possible location to the next chapter, and
turn now from the people themselves to an examination of the
tales Aristeas said they told him about the regions beyond them.
 So TALES OF A TRAVELLER
Were the griffins and the one-eyed Arimaspi who filched the
gold from them the creations of Aristeas' own fancy, or did the
Issedonians really tell him about them? The reader will recall
Herodotus' story about the outsize ants of the Bactrian desert,
how they cast up gold as they made their burrows, and how the
neighbouring Indians made forays to collect this [66]. The gold-
guarding ants and the Indian robbers must be doublets of the
gold-guarding griffins and theArimaspianrobbers, and this would
seem to prove that it was the same basic story, of monsters who
guarded and of men who stole the gold, that the Persians, whom
Herodotus names as his authorities for the ant version, heard
in the south and Aristeas heard farther north. The possibility
suggests itself, however, that the ant-Indian story is really a Greek
invention inspired by the griffin-Arimaspian story in the Arima-
spea; but fortunately this possibility can be ruled out. Herodotus
says that in order to evade the ants' pursuit the Indians would
take their camels yoked in groups of three, one member of
each trio being a female which had recently foaled ; her eager-
ness to get back to her young would give just that extra turn of
speed that was needed to out-distance the angry ants. Now
if the basic story were a genuine piece of folk-lore originating
from some central point and as it spread in different directions
assuming different locations and different actors appropriate
to the location, we should expect that the newly-foaled camels
in the southerly version would be answered by newly-foaled
mares in the northerly one. We are lucky in having the bit of
independent evidence needed to fulfil this expectation.
Speaking of the far north of Asia, Marco Polo says :
Still further north ... there is a region which bears the name of
DARKNESS, because neither sun nor moon nor stars appear, but it is
always as dark as with us in the twilight. The people have no king of
their own, nor are they subject to any foreigner, and live like beasts.
The Tartars however sometimes visit the country, and they do it in
this way. They enter the region riding mares that have foals, and these
foals they leave behind. After taking all the plunder that they can get
they find their way back by help of the mares, which are all eager to
get back to their foals, and find the way much better than their riders
could do (Yule, ii. 484).
 TALES OF A TRAVELLER
In his note on this passage Yule thinks that the story is probably
of great antiquity, as in a tale given by Rashiduddin the ruse is
said to have been used by the patriarch of the Turco-Tatars in an
incursion into the Kingdom ofDarkness. 3
That the story of the gold-guarding ants was not a Greek or
a Persian invention is further confirmed. It was known to ancient
India, for the Mahabharata calls gold paid as tribute 'ant-gold'
(Tam 107). The tale did not, however, originate in India but in
central Asia and came in with the gold, as has been shown by
Laufer, who ingeniously conjectured that it arose from a confu-
sion between the name of a Mongolian tribe (Shiraighol) and the
Mongolian word for ant (shirgol). This suggestion is treated with
reserve by Tarn, who states, however, that Laufer's view, that the
story travelled from Mongolia to India rather than the other way
about, has been proved by Darius' Susa inscription.
Laufer notes the mention of monstrous ants in the Mongolian
region by certain ancient Chinese authorities, and it is worth
expatiating on these:~ They occur in two works, Ch'u Tz'u and
Shan Hai Ching, both embodying the shamanistic lore of the
'Yangtze Valley culture'. From the former, Chao Hun ('The Sum-
mons of the Soul') is probably to be dated to the mid-third
century B.c.:
Oh soul come back! for the west holds many perils:
The Shifting Sands stretch on for a hundred leagues..
You will be swept into the Thunder's Chasm, and dashed into pieces,
unable to help yourself;
And even should you chance to escape from that, beyond is the
empty desert,
And red ants as huge as elephants, and wasps as big as gourds.s

The second book, Shan Hai Ching, is syncretic and did not
emerge in its present form much before the end of the first century
B.c. There is, however, good reason to believe that the parts to
be cited now and later were in existence two centuries before this;
and the matter they contain, being in the nature of folk-lore, was
presumably much earlier.
Hai Nei Pei Ching ('Within the Sea, north') .... To the north of the
K'un Lun Mountain is a man called Ta Hsing Po who holds a halberd
G860 0
 TALES OF A TRAVELLER
in his hand. East of him is the land of Ch'iian Feng. The body of
Erh Fu is to the east of Ta Hsing Po. The land of Ch'uan Feng is
also called the land of the Dog Jung. The appearance of these people
is like dogs... The land of the Kuei6 is to the north of the body of
Erh Fu. These people have the faces of men but only one eye... The
Ch 'iung-ch 'i is like a tiger with wings.. To the east of the Ch'iung-
ch'i are the giant wasps which look like wasps and the giant ants which
look like ants.7
It looks, then, as if central Asia is the starting-point for a story
which sent out radiations to China by an easterly, to India and
Persia by a southerly, and to Greece, through Aristeas, by a

I
westerly route.
Though Shan Hai Ching does not associate the ants with them,
it will be noticed that it lists in the same area men with one eye.
Such appear again in another passage, though this time located
beyond the northern sea.8
Hai Wei Pei Ching ('Beyond the Sea, north'). These are the things
from the north-east corner to the north-west corner. 9 The land of the
I
Shankless is to the east of the Long-legged. The people there have no
shanks. The Spirit of Chung Mountain is called Torch Dark. When
he opens his eyes it is day and when he shuts them it is night. When
he blows hard it is winter; when he blows soft it is summer. He neither
drinks nor eats nor rests. His breath makes the wind. His body is a
thousand li [three hundred miles] long. He lies to the east of the Shank-
'I
less. He has a man's face and a serpent's body, red in colour. He dwells
at the foot of Chung Mountain. The land of the One-eyed is to the
east. The people there have one eye set in the midst of their faces. The
land of Jou-Ii is to the east of the One-eyed. These people have one
hand and one foot. . , 1o
The One-eyed appear again in the same location in the encyclo-
pedia called Huai .Nan T ;:.u, compiled at the court of Liu An,
I
king of Huai Nan, who died in 122 B.c. 'There are thirty-six
lands beyond the seas... From the north-east to the north-west
are the Tip-toes, the Kou Ying, the Sunken-eyed, the No-bellies,
l
the Jou-Ii, the One-eyed .... 11
These strange peoples may be largely the product of Chinese l.
imagination; but some of them at least may be derived from the
folk-lore ofinner Asia, as the monstrous ants must have been. The

I
I
 TALES OF A TRAVELLER 83
Siberian Buriats in the region of Lake Baykal say that the Ruler
of the Dead has (like Odin) one eye in the middle ofhis forehead;
and the wealthy one-eyed Forest man of the Votiaks just west of
the Urals is said by Holmberg to have been borrowed from the
Tatars (479; 179. Cf. 181, 182). 12 Muchmoreremarkableisthe
following from A. N. Athanasiev (260) :
Herodotus speaks of a whole race of one-eyed people. The Ukranians
to this day preserve memories of that race; according to their stories
the edinookie dwell somewhere far away over the seas [cf. Shan Hai
Ching and Huai Nan Tzu]; the Tatars when campaigning used to sack
towns and villages, slay the old people and children, but carry off the
young men and women and sell the plumper ones to one-eyed people
who drove them off like sheep to their territory, fattened them up,
killed them and ate them (Lud. Ukrainski i. 352). The legend also lives
on with the Ural peoples (The Ural Peoples, the Life of the Ural Peoples
and the Kossaks, by I. Zheleznov, i. 87-89). 13
Here we have a blend of Arimaspians and Anthropophagi.
Just as Aristeas' gold-guarding griffins had their counterparts
in the gold-guarding ants on the borders oflndia, so did his one-
eyed Arimaspi have their counterparts in an Indian people, also
emanating, I suggest, from that common centre whence had ori-
ginated the guardians of the gold. These were described by
Megasthenes as having the ears of dogs, the eye in the centre of
the forehead, hair stiff like bristles, and shaggy breasts (Strabo
7I I : ovop.a.TOV<;
, .!'\ \ 1" \
TE (l/\/\0VS" WTO. /UV EXOVTO.S" KVVO<; EV /.LECTtp OE Tep
t,/ ' , , ~' ...

/J,ETW'Tl'<p 'TOV oef,Oa.>..ov, op8oxa.lTa.s, Aa.crlovs TO. 0'1'1/871). A charac-

teristic of the Dog-heads has clearly been grafted on to them, but
otherwise the similarity of this description with the fifth fragment
of Aristeas is striking. Whether there is any significance in the
fact that Megasthenes also mentioned in his account of India
Hyperboreans 'who lived for a thousand years' (Strabo, I.e.) will
be discussed later [98 ff.] ,1 4
Various theories have from time to time been put forward as
to the originals of the griffins and the Arimaspi. The most ridicu-
lous (cf. Tournier 12) are those that see in the latter a real tribe
whose other eye had become atrophied from being constantly
closed to aim a bow, or a mining people whose single eye was in
 84 TALES OF A TRAVELLER
fact a lamp strapped to the forehead. Hardly more plausible is the
view that the authentic griffin is the mild little baibak, a relation
of the jerboa inhabiting the steppe (Minns, Hennig). Far more
impressive is the explanation offered by Adolph Erman (ii. 87):
By comparing numbers of the bones of antediluvian pachyderms,
which are thrown up in such quantities on the shores of the Polar
Sea, all these people [the Samoyeds; Erman is dealing especially with
the Samoyeds in the region of the mouth of the Ob] have got so distinct
a notion of a colossal bird, that the compressed and sword-shaped
horns, for example, of the Rhinoceros teichorinus, are never called, among
the Russian promuishleniks and merchants, by any other name than
that of 'birds' claws'. The indigenous tribes, however, and the Yuka-
girs in particular, go much further, for they conceive that they find
the head of this mysterious bird, in the peculiarly vaulted cranium of
the same rhinoceros; its quills in the leg-bones of other pachyderms,
of which they usually make their quivers; but as to the bird itself, they
plainly state that their forefathers saw it and fought wondrous battles
with it: just as the mountain Samoyedes preserve to this day the tradi-
tion, that the mammoth still haunts the sea-shore, dwelling in the
recesses of the mountain and feeding on the dead.
These birds, then, are for Erman the original of Aristeas' griffins;
the Arimaspians he refers to the pre- or non-Samoyeds who
lived in subterranean dwellings and worked metal (the Samoyeds
had stories of how they had ventured to approach the entrances
to these caverns and had heard them within speaking an un-
intelligible language). Yet for all its attractiveness Erman's theory
must be discarded, if we are right in placing the starting-point
of the story in central Asia, among Turco-Tatars rather than
Samoyeds.
Scholars recently have followed a different tendency-to see
in the griffins and the Arimaspians figures from another world,
a world of the spirit, to which the shaman makes his journeys
in the soul. The former are to be identified with the dragons
which defend the Golden Mountain where God dwells, the latter
with evil one-eyed infernal beings; both known to the mythology
of Turco-Mongol Altaic peoples. This was the suggestion of
Alfoldi, followed by Meuli and Phillips. 1s Though primitive
men naturally tend to locate their spiritual world somewhere
 TALES OF A TRAVELLER 85

in this one, on the edges of the earth or beneath it, there is

yet an important distinction between this view and the earlier
ones: for if we accept it, and think that here Aristeas misunder-
stood his Issedonian informants, taking for fact some religious
myth, we are wasting our time if we try to equate the Arimaspi
with some historical tribe. Yet the Issedonians could hardly have
been forced to migrate by the pressure of infernal beings, and I
think that for them the Arimaspi were a real, flesh-and-blood
people, as real as were the Dog Jung and the Kuei of the Chinese
in spite of the prodigious shape ascribed to them. How they came
by their reputation of having only a single eye, whether this
feature was dubbed on to them by analogy with mythical demons
or was chance fancy born of that revulsion for the stranger of
alien race which lends him non-human characteristics, 16 cannot
be decided and does not matter.
If the Arimaspi were not entirely creations of the imagina-
:!
tion, their enemies from whom they were supposed to steal the
gold were. They were monsters of some sort, monsters described
in one version of the tradition as enormous ants, in the version of
Aristeas as griffins. To a Greek the word ypv,f,, 'griffin', would in
classical times denote a beast having the body of a lion and the
head and wings of an eagle, a very different thing from an ant,
however magnified. The attributes of the lion and the eagle are
sharp-sightedness, speed, and ferocity. The questions now at issue
are : what governed Aristeas' choice of the word ypt;r;, to denote
the creatures of the Issedonian tale? Was the word a translitera-
tion of some Issedonian word for the guardians of the gold, or
was it a name Aristeas knew already for a certain kind of monster?
How did the Issedonians themselves describe the creatures?
Actually as half lion and half eagle? Or rather as monsters with
the characteristics of speed, ferocity, and vigilance, whose physical
appearance was not clear to Aristeas, so leading him to make his
own identification with a monster he already knew?
To attempt an answer to these questions a survey must be made
of the history of the griffin in ancirnt art. 17 Three species may be
distinguished: the bird-griffin, with thr body of a lion and the
head of a bird, either winged or wingless; the snake-griffin, with
 86 TALES OF A TRAVELLER
the body of a lion (often covered with scales), snake's head, the
forefeet of a lion and the hindfeet bird's legs, with or without
wings, generally with a scorpion's tail; and the lion-griffin, with
lion's body (often scaly), lion's head and forefeet, bird's legs for
hindfeet, and bird's tail, with or without wings. We should not
expect a monster the principal component of which is leonine to
have originated in the more northerly parts of the world ; 18 and
the evidence points to the Middle East as its home. Prinz sum-
marizes its development thus: the bird-griffin started in Egypt,
whence between 2000 and 1500 B.c. it passed into the art of the
Hittites and of other Middle Eastern countries. The Hittite type
then in its turn affected certain griffin types of the Egyptian New
Kingdom, Assyria, and probably the Mycenaean-Cretan culture;
also derived from the Hittite is the Phoenician griffin. Lion- and
snake-griffins came from Babylonia, and influenced Hittite,
Assyrian, and Persian art. All these cultures regarded the griffin
both as a beast of prey and as a supernatural creature, but it might
also take on a variety of significances-in Egypt as representing
royalty and various gods; in Babylon and its environs and the
Mycenaean world as a divine animal. 19
In Egypt the bird-griffin only appears, and can be traced back
to prehistoric times (before 3300 B.c.). He is represented as a real
beast of the desert, to be hunted as men hunt the lion, while him-
self hunting other beasts. From the Middle Kingdom on we find
the griffin as a symbol of royalty, with the hawk's head of
Horus, and, congruously with this, as a supernatural creature.
In the New Kingdom there is a development in the artistic
treatment of the griffin which is owed to other Middle Eastern
countries, the principal feature of which is the loose curls on the
beast's head.
In Babylon the snake-griffin and the lion-griffin are found as
supernatural beings associated with various gods. From the time
of the dynasty of Ur onwards the lion-griffin is depicted in combat
with men and with other creatures; he is simply a fierce beast of
prey, and there is nothing symbolical here.
The typical Hittite griffin has the body of a lion, wings and
head ofa bird, on which is a sort ofcrest or comb in three sections.
 TALES OF A TRAVELLER
From the head depends a curling lock. It is here perhaps, in the
mountains of the Anatolian region, that the bird-griffin acquired
its eagle's head, in place of the falcon's or sparrow-hawk's of
the earlier Egyptian type (the eagle-headed type first appears
in Egypt in the time of the Hyksos: cf. Phillips 172 f.). The
Hittite griffin makes its first appearance in the first half of the
second millennium before Christ. It is essentially a beast of
prey, seizing lions or gazelles; no symbolical meaning can be
discerned in it, nor any connexion with the divine.
In the world of Crete and Mycenae only the bird(eagle)-griffin
is found. Typical are the curl-like decorations which not only
depend from the head but also sometimes surround the base of the
wings. The crest or comb also occurs. It is thus a direct relative
of the Hittite and later Egyptian griffin (the earliest example is
to be dated about 16oo B.c., which would not preclude at least
Hittite influence). It plays the part of a divine beast (e.g. held on
a lead by a god, or drawing a car in which ride deities, one of
whom may be the Magna Mater; and cf. J.H.S. xxi (1901), 158,
fig. 36), or of a beast of prey. It is on an object oflate Mycenaean
date, an ivory mirror-handle found at Enkomi in Cyprus, that
we come upon the first grypomachy proper, or fight of the bird-
griffin with a man [Plate II]. The theme was to become a common
one in Assyrian, Phoenician, and Persian art, but on the strength
ofthis mirror-handle Prinz is inclined to look to late-Mycenaean
Cyprus as the home of its first formal representation.
This is doubtful. The scene shows the figures in profile. The
warrior presents his right side to the viewer, and stabs with
downward sword-thrust in the chest the prancing griffin facing
him. The vigour of the man and the drooping wings of the dying
griffin are effectively portrayed. The beast is indeed of the
Mycenaean type, with eagle's head and curling locks which e.xtend
along the initial part of the wing; but the equipment of its
opponent alines him rather with the Philistines and Shardana,
the Peoples of the Sea who were defeated by Rameses III c. 1194
'
/, B.c., while the upward tilt of his face betrays Syrian influence (cf.
I
Dunbabin 36). The mirror then is probably of Phoenician work-
1,
,. manship, and an import rather than a product of Cyprus itself
88 TALES OF A TRAVELLER
(Lorimer 151, 200, 252, 392). In view of this it is perhaps better
to place the origin of the grypomachy theme farther east, and
to trace its inspiration to the combats of the Babylonian lion-
griffin with men.
Assyrian art knows all three types of griffin. Its snake- and
lion-griffins are closely related to the Babylonian counterparts,
its bird-griffins to the Hittite (the crest-like feature of the latter
has now become a proper crest or comb). There are several
representations of gods fighting or hunting griffins (mostly lion
headed).
This motif of combat between a god and the monster becomes
in Persian art (which confines itself to the type of the lion-griffin)
simply a combat between the monster and the Great King; other
wise the griffin's role is purely decorative.
Phoenician representations of the griffin follow closely the lines
of the late Mycenaean, and are also like the Hittite in many ways.
Egyptian influence is also evident; when for example the griffin
is depicted trampling an enemy. Phoenician grypomachies
generally follow the lines of the scene on the Enkomi mirror
handle (Perrot-Chipiez iii. 789, fig. 552 (here the monster's head
is turned away from its vanquisher, who delivers his blow under-
arm to the creature's belly, while grasping its locks; from Curium
in Cyprus); c 771, fig. 546. On a patera from Varvakeion a
grypomachy is depicted which is schematically quite different
from that on the Enkomi mirror-handle: 783, fig. 550).
In archaic Greek art the griffin is depicted in a calm, watchful
attitude, whether walking, sitting, or lying down. Its beak is
strongly hooked, occasionally continuing the curve of the skull in
an unbroken line, and almost always wide open and menacing.
The eagle's head always has long, pointed, upright ears; almost
always a lock hanging from the base of the ear to the neck; and
frequently an ornament extending back from the head, which
may end in stylized leafage or flower-shapes: a sort of crest. It
shares these characteristics with Egyptian, Mycenaean, Assyrian,
Hittite, and Phoenician types; but a specifically Greek develop-
ment is the upright knob above the eyes (cf. the horn of the lion-
and snake-griffin). Also purely Greek is the upward curve of the
 TALES OF A TRAVELLER 89
wings, a change in the tradition which can be paralleled from
Greek representations of Gorgons, Harpies, and Sphinxes. This
archaic Greek griffin appears ordinarily alone or in heraldic
pairs; if associated with other beasts it displays neither hostility
nor friendship towards them. Evidence for an early connexion
with Apollo is literary in the first place; the most important
example is the old statue of Apollo at Delos, which was flanked
by prancing griffins. Later it is associated also with Dionysus,
Artemis, and Nemesis; but the oldest and longest association is
with Apollo.
Classical Greek art removed all the 'unnatural' features from
the griffin-curling locks, ornamented crest and knob-except
for the pointed ears; and substituted the fin-like mane running
the length of the neck often as far as the beak (this was long to
be a regular feature of the griffin in art). The formal upward
curve of the wings too gave place to a more realistic representa-
tion.
We are now in a position to make a closer inspection of the
scenes on the Kelermes mirror (5 ff., and Plate I]. These are a
curious mixture of Greek and barbarian. In one panel are two
sitting sphinxes, heraldically arranged facing each other; their
wings have the typical upward curve of archaic Greek art.
Beneath them is a sitting griffin of the archaic Greek type, with
tall ears, gaping eagle's beak, upward-cuned wings, pendent
locks, and knob above the eyes. In the opposite panel are two
more sphinxes, heraldically arranged on either side of a thin
column surmounted by a pair of Ionic volutes. These sphinxes
are prancing, their heads are retorted, their wings lack the up-
ward curve. This motif of the heraldic pair is repeated in another
panel where two prancing lions face each other as if in combat,
in the panthers held one in each hand by the 'Winged Artemis',
and in the grypomachy panel. There is nothing particularly
Scythian in this motif, which has Mycenaean parallels; any more
than there is in the 'Winged Artemis', who was represented in the
art of Greece and the Mycenaean-Cretan world long before the
date of this mirror (Thompson 286 ff.; Nilsson; rg, 28; Childe
30). A theme popular in Scythian art, however-the slaying of
 TALES OF A TRAVELLER
r 90
a herbivorous animal by a carnivore-appears in the panel
showing a lion biting through the backbone of a bull. Even so,
the Scyt.l1ians borrowed this theme from their more southerly
neigh hours-the steppe is not prolific of carnivores (on a Phoeni-
cian silver plate from Caere in Etruria two lions attack a bull
with head retorted; a bird flies above, and there are palm-trees:
Perrot-Chipiez iii. 769, fig. 54'1-; and here the head of the bull
is not retorted in the way often favoured by the Scythian 'animal
style' (on the other hand the heads of the ibex and the wolf are,
as is the head of the griffin-but so is that of his Phoenician fore-
bear [88])). Much more extraordinary are the two panels contain-
ing as their principal features in the one case a lion in front of a
tree, in the other a bear just above whose back is a flying eagle.
These animals are 'non-heraldic', standing and presenting their
left side to the viewer, and are strongly reminiscent in their
attitude of a lion and other animals on one of the vases from
nearby Maikop on the north-western fringe of the Caucasus
(Rostovtzeff 1 23 and his pl. III; Minns 144). This impression is
enhanced by the representation of the tree on the Kelermes
mirror, which looks like an imitation of the trees on the Maikop
vase, and by the fact that on the latter a bird is perched upon the
lion's back in remarkable correspondence with the eagle above
the bear's back on the Kelermes mirror. Yet the Maikop vase is
to be dated in the third millennium B.c. ! Some of its features may
have been embodied in an artistic tradition that long endured
in the neighbourhood, and were known to the engraver of the
mirror (yet it is also possible that the ancestry of the latter in some
of its components may rather have been Phoenician).
The heraldic treatment is cleverly applied to the grypomachy
panel. The griffin is the archaic Greek type like his fellow on the
sphinx panel (the knob above the eyes is clear), though the
wings lack the upward curvature. He is prancing; his body faces
towards one of his opponents, while his head is retorted to face
the other, thus giving cohesion to the group. The prancing body
and the retorted head recall the Phoenician grypomachy already
mentioned [88]. The griffin's human adversaries face each other
across him, and appear to grapple with him with their bare
 TALES OF A TRAVELLER 91
hands. That they are savages of a primitive sort, hirsute 'wild
men', is clearly shown by their long, unkempt hair and beards
and the stippling that covers their bodies; they answer admirably
the description in fr. 5. 2 of Aristeas, 'shaggy with hairs, toughest
of all men'. 20 It is interesting that the lower part of this panel,
unlike all the others except that containing Artemis, is not filled
in with some other beast but, as it were the ground on which the
actors tread, with close-packed uniform squiggles: perhaps to
represent the desert habitat of the griffin.
It would appear, then, from the admixture of Greek and south
l
[I
11
Russian features in the scenes depicted on the Kelermes mirror
that it must be the work of a Greek craftsman living in one of the
, ,.
colonies on the northern Black Sea littoral. 21 That he included a
grypomachy not simply because it was one conventional form of
11 decoration but because he considered it particularly appropriate
j
tothedestinationofhiswork is, I think, proved by his substitution
of savages for the usual warrior antagonist of the griffin; though
he did not so far break with the artistic tradition of the grypo-
machy as to make them one-eyed.
The date of this mirror is about 575 B.c. It is not the only
archaeological evidence that the Greeks already connected the
griffin with Scythia; for also at Kelermes, in a burial dated by
Rostovtzeff to probably the second half of the sixth century, but
by other Russian authorities to the seventh (Rostovtzeffii 2i9;
Rice 154), was discovered an iron rattle surmounted by the head
of a griffin of the true archaic Greek type exemplified by a seventh-
century bronze found at Olympia (ill. Roscher i. 1766; cf.
Dunbabin 43). It has long upstanding ears, gaping beak strongly
hooked and continuing the curve of the skull in an unbroken line,
while above the eyes an atrophied knob can just be discerned in
the photographs [Plate III]. If Rostovtzeff's date for the burial
is right, this object must have been manufactured long before,
and even if it is itself Scythian work it is proof that the Greek
griffin had been introduced into Scythia very early in the sixth
century at the latest.
Once established there the griffin became a popular ornament
of the art of the European steppe (not true of the grypomachy,
 TALES OF A TRAVELLER
a fact which may indicate that it meant nothing to the natives;
on the other hand, the motif ofman against beast was not favoured
by Scythian artists). It remains to note its appearance also far
to the east, in the Altai mountains. There, at Pazirik (lat. 50 N.,
long. 86 E.), in burials dated from the fifth to the third centuries
B.a., both bird-griffins and lion-griffins have been found. The
latter no doubt made their way from Persia, but very remarkable
is a bird-griffin on a Persian pile carpet of the fifth century B.a.
from Pazirik (ill. Rice 138). It has the gaping beak and tall ears of
the Greek griffin, but is in an unusual attitude, standing looking
back over its shoulder. As the bird-griffin was not a Persian type
the carpet may have been home-made rather than imported,
and its maker may have owed this motif to Scythian influence.
The bird-griffin (though in this case with a bird's tail) is also
shown attacking an elk on a carpet from a site yet farther east,
a grave of the first century A.D. at Noin Ula in northern Mongolia
(Rice, pl. 29).
Ifone were to plot the incidence of the bird-griffin in its earliest-
known artistic examples one would get a roughly parabolic arc
on the map, starting in prehistoric Egypt, extending through
Mycenaean and pre-classical Greece to south European Russia,
thence to western Siberia and ending in Mongolia in the first
century of our era. It appeared in Greece long before Aristeas, in
south Russia soon after him, in Siberia long after him. It is hard to
escape the conclusion that Aristeas did not hear the description
of the physical shape of the griffin from the Issedonians, but took
the concept with him on his travels and dubbed it on to the gold-
guarding monsters of his Issedonian informants. His account of it
in the Arimaspea on his return home would then underlie its pro-
jection into Scythian art, in the first place through the mediation
of Greek craftsmen who knew the poem, and thence farther east
by a drift of artistic tradition across the steppe.
Though the Greeks of Aristeas' time certainly knew and often
used the griffin shape in art, and so presumably had a name for it,
we cannot unfortunately be sure at present that that name was
ypoif,. Ziegler finds the explanations of the word so far proposed
unconvincing-for instance, its derivation from the Indo-
 TALES OF A TRAVELLER 93
germanic root *grabh (etymologically unsound), or from yptmo~
(first in Xenophon :22 putting the cart before the horse?). He is
more sympathetic with those who would relate it to the Hebrew
kerub, though he maintains a proper scholarly reserve about this in
view of our ignorance of the nature of the original cherubim. He
sensibly remarks that, as the concept of the griffin was primarily
non-Greek, we should probably look for a non-Greek etymology
of the name. That it might be a transliteration by Aristeas into

II
Greek of an Issedonian word for their monsters, which he defined
in the terms of a monster already known to himself and his com-
patriots, is not to be precluded; obviously if he was introducing
a common noun unknown to his Greek public he would have to
define it, and this consideration might make us give the evidence
of Pausanias more weight than we should otherwise attach to it
[32; T. 4]. And it must not be forgotten that some early author
thought it necessary to give a description of the appearance of
a griffin (72 f.].
To sum up: I am inclined to believe that Aristeas heard from
the Issedonians of some sort of swift, fierce, and vigilant monster
which guarded the local gold; that the name they gave to this
sounded to him like ypvif,; that its attributes of speed, ferocity,
and vigilance reminded him of the lion-eagle monsters he had seen
in artistic representations at home; and that he accordingly de-
fined the word ypvif, in his poem in terms of this lion-eagle
monster.
The grounds are substantial for believing that the originals of
the one-eyed Arimaspi and their monstrous ad\'ersaries are to be
sought in the folk-lore of central Asia. Is the same true of the
mountain that neighboured them, where the wind Boreas had
his home in a cavern?
We must recall that in the 04Yss1y all the winds are gathered
together in a floating island under their ruler Aeolus, who re-
strains the adverse ones for Odysseus by tying them up in a bag
(x. I ff.). For Homer, then, they are not confined within a caver-
nous prison, as they are for Virgil (Am. i. 50 ff.), who is influenced
by the later tradition about the Cave of Boreas. This tradition
was certainly known in the fifth century B.c. to Sophocles, who
 94 TALES OF A TRAVELLER
says that Cleopatra, the daughter of Boreas, was brought up in
her father's cave (Ant. 983) :
'T1}Ae,r6po,s 8' .!v '5.VTpo,s
Tp&.,/>71 8vJ>..>..11uw iv 1ra-rpc{,a,.,
BopEa.s ap,t'lT'lTO'; &p867To8os {mlp ml.you.

Whatever the exact interpretation of this passage, the reference

to the North Wind's cave is plain enough; and it is tempting to
see in the 'sheer steep' ofl. 985 the Rhipaean precipices above
which that cave was situated. 23 Though we possess no earlier
first-hand mention of this cave, we have had good reason to infer
that it occurred in theArimaspea [44], so that it may quite well
have been Aristeas who first broke with the Homeric tradition of
the island of the winds, and located one of them in a cave on the
edge of the world. 2 4
The Issedonians again may have been his authorities for this.
The notion that winds have their habitation in caves is not only
originally un-Greek but unusual. Holmberg notes what he calls a
strange idea among the peoples of central Asia and Siberia: that
mountains are the home of the winds (457). The Yakuts about
the river Lena speak of the winds 'sleeping' in the mountains;
the Mongols call days of storm 'running-days', because then the
Mountain-spirit runs from mountain to mountain; 'the Goldes
believe the winds to come from caves in the mountains, where
the Wind spirit holds them captive'. These last dwell in the far
east of Siberia, but Holmberg quotes a similar belief among the
Lapps (and perhaps the Finns), that winds originate from chasms
in the fells. The inhabitants of a mountainous district would note
the force with which the wind blows down these natural funnels,
and this may be the germ of the idea. 2 s
However valuable their witness, Lapps and Goldes are far
removed from any possible location of the Issedonians. But the
belief is strikingly confirmed for the Altaic region of the Russian
steppe by Carpini :
Departing from hence, wee founde a certaine small sea, vpon the
shore whereof stands a little mountaine. In which mountaine is
reported to be a hole (quoddamforamen), from whence, in winter time
 TALES OF A TRAVELLER 95
such vehement tempests of winds doe issue, that traveilers can scarcely,
and with great danger passe by the same way. In sununer time, the
noise indeede of the winde is heard there, but it proceedeth gently out
of the hole (Beazley 134; 98).

The 'sea' here mentioned appears to be the lake called Ala

Kul. What Douglas Carruthers has to say about the tempestu-
ous gales of this vicinity is of such interest that I take the liberty
of quoting him extensively (411) [Map II]:
We had frequently heard of the terrors, dangers, and winds of the
Dzungarian Gate. We had read the records of such early travellers
as Carpini and Rubruck, both ofwhom26 mentioned that 'there blows
nearly continuously such a wind through this valley, that persons
cross it with great danger, lest the wind should carry them into the
sea'. We therefore approached this remarkable geological pheno-
menon both with interest and a certain amount of anxiety, for the
weather was unusually unsettled ....
Even at a distance, with the Dzungarian Gate lying before us-
unseen, we instinctively became aware of its presence; for when we
came within sight of Ala Kul, whilst crossing the open foothills of the
Barlik Range some twenty miles from the lake shore and some 2,000 ft.
above it, we were in absolute stillness ; and yet, strangely enough, the
waters of Ala Kul were tossed into waves--the white crests being
clearly seen with a glass, while even with the naked eye breakers could
be distinguished dashing on to the southern shore of an island in the
middle of the lake. Although we were becalmed, there was evidently
a gale blowing through the 'Gate', and as we approached nearer we
became at every step more keenly alive to the action of this wind-
trough. At night we heard a distant roar as the imprisoned winds of
the Dzungarian deserts escaped though this narrow defile. The only
night we camped on the very shore of the straits, the wind increased
to such a violence that our tents, though well protected in a valley,
were by the morning all blown away, for the wind swept in great
gusts over the hills, and the back eddies tore them down; the noise was
terrific, and sleep out of the question....
Fortunately we succeeded in crossing the depression in a nine hours'
trek without mishap, a strong head-wind being the only cause for
annoyance. Had there been rain or snow falling, travelling would
have been impossible, but the wind was luckily from the south, and
comparatively warm, the temperature at night only just touching
freezing-point. Only just in time did we escape from this home of the
96 TALES OF A TRAVELLER
winds, for the day after crossing the valley .. the wind swung round
to the north and swept cruelly through the gap, bringing with it hail
and snow. Had we then been journeying northwards the making of any
headway would have been out of the question, for neither man nor
beast could have faced the elements.... A bleak, inhospitable land-
scape now surrounded us, mountains, clad in fresh snow, showed up
here and there through breaks in the blurred atmosphere, and great
cloud-banks swept through the 'straits', as if rushing through some
gigantic funnel. ...
The natives relate the usual traditions as to the origin of the winds
in this locality. In the myths of Central Asia a 'hole in the mountain',
or 'an iron gate in a lake' is the usual explanation of the origin of
winds. In the case of which I am writing the island called Ala-tyube-
a small extinct volcano21 in Ala Kul-is made responsible for the
furious winds which sweep through the depression .... From autumn
to spring the prevailing wind is from the south-east. I think, however,
that the wind which causes havoc amongst the nomads, and kills
off men and flocks when caught unprotected, is this north wind when
it attains the velocity of what is called a 'buran'.
Humboldt also preserves the tradition, gleaned from Tatar
sources, not only of the violent winds issuing from the island in
Ala Kul, but of another 'Cave of the Winds' in this same region
of the Dzungarian Gate, some thirty miles south-south-east of Ala
Ku! (ii. 84 ff.; 49 r ff.). This latter was a very terrible place, of
unknown depth, from which issued, especially in winter, wind
storms of such force that they would tumble everything in their
path and hurl it into the lake.
That Carruthers, Humboldt, and Carpini are recording vir-
tually the same tradition there can be no doubt. Six hundred and
fifty years separate Carruthers from Carpini; it is still a long step
to carry it back another two thousand years before the latter, yet,
in view of the extraordinary persistence of folk-beliefs among
unsophisticated peoples, not impossible. 1s The tale of the wind
that irresistibly snatches away all before it recalls that fragment
of Aeschylus about the danger of being swept away by the blasts
of Boreas [54f.]. I suggest that we have found, if not the original
Rhipaean Mountain with its Cave of Boreas, yet at least a near
relation; and the general area, if not the precise spot, where it
was situated.
 TALES OF A TRAVELLER 97
There is indeed a notable difference between the home of the
wind in the accounts ofCarpini and Carruthers and the Rhipaean
range of the ancients : the former is a 'little mountaine' or a
'small volcano', the latter is immensely high. If, therefore, the
former is indeed the original of the latter it may have gained
altitude by conflation with some other mountain; and there is
a candidate for this last whose claims must be considered.
There is a comparatively early Altaic belief that the sky is
supported by a pillar at the north pole; and the majority of Cen-
tral Asiatic peoples tell of a mighty 'world-mountain', whose
summit, unattainable to man, is at the North Star, where dwells
the supreme god on his golden throne. Everywhere in Asia it is
assumed that this mountain is in the north of the world; the Mon-
gols call it Sumur or Sumer, the Buriats Sumbur. An Altaic tale
relates how at the creation God sat upon a 'golden mountain'
where sun and moon ever shone, which afterwards descended to
cover the earth. The underside of this (hollow?) mountain there-
fore forms our sky (the edges of which do not quite reach down
to the earth-thus allowing ingress and egress to the sun and
moon, presumably) (Holmberg 333; 341 ff.).
This mountain exemplifies certain traditions we have met con-
cerning the Rhipaean mountains: it is in the far north; it could
hardly be higher (most appropriate to it would be Aeschylus'
darpoyE{rwv and Aristotle's 7TEp, wi TOV i,uyi8ovs >.{m Eiah DL
AeyoEVOL ,\6yoi vBwOELS [ 49; 42]) ; and it is golden [73]. There
are, however, factors which prompt wariness in accepting that it
has contaminated the Rhipaean legend.
Three distinguishable conceptions underlie this 'world-moun-
tain' of Asiatic lore: (i) it is a giant mountain at the centre of the
earth's disk (i.e. at the terrestrial pole), whose summit touches the
sky; (ii) the mountain is in the sky; (iii) the mountain itself forms
the sky. The two latter conceptions are probably developments
from the first, and the first is obviously related to the 'world-
mountain' of ancient Hindu cosmology; a relationship which is
confirmed by the names Sumer, &c., by which the mountain is
called. The idea looks, then, not to be nati\'e to central A~ia, but
to be an importation from Indi,1; though it is difficult to say for
68~0 II
 98 TALES OF A TRAVELLER
certain that India was its original home (Holmberg 342). An
obvious vehicle for its importation would have been that lamaist
Buddhism which is widespread among Siberian and Mongolian
peoples today, but was of course non-existent in the seventh
century B.c. It did not reach the Ili (a river flowing into Lake
Balkhash) until the fourth century of our era (Humboldt i.
396).
If this were all, the 'world-mountain' need detain us no longer;
but unfortunately the 'blessed Hyperboreans', who dwell be-
yond the Rhipaean mountains, and whose first appearance in
Greek legend cannot be dated by other evidence before Aristeas,
demand otherwise: for a people very like them dwelt to the north
of the Indian mountain.
According to the geography of the Puranas, our earth is an
enormous continent surrounded by an ocean, beyond which are
six other continents, each surrounded by its ocean, arranged in
concentric rings about ours. In the centre of our earth (and so in
the centre of the whole system) rises to an immense height
Mount Meru, the seat of the gods. To the south and north of this
the continent is divided by six parallel mountain chains, three on
each side of Meru, the southernmost being the Himalaya. The
land mass to the south of the Himalaya is inhabited by the
Indians, among others, that to the north of the northernmost
range by the Uttarakurus. According to an alternative account
in the Mahabharata these last inhabit the northernmost continent
of four which extend towards the four cardinal points from the
central Mount Meru; but the crucial feature, that they are to the
north of the mountain, remains.
The classical descriptions of the Uttarakurus are in the Rama-
yana (iv. 43) and the Mahabharata (vi. 7). Their climate is tem-
perate, their land rich in precious stones and metals; they need
not labour, for trees provide their every desire-food, drink,
clothes, ornaments, and maidens. They are a pious and loving
people, who pass their lives in gaiety free from sickness and old
age, until they die at the age of 1 1 ,ooo years (or one thousand,
according to the more modest Buddhist estimate).2 9
This account is much like Pindar's of the Hyperboreans [71],
 TALES OF A TRAVELLER 99
but there is no need to postulate any common source to explain
a similarity of ideal bliss, which is so conceived pretty universally
among mankind. It is a pity that a detail which might be signifi-
cant, the thousand-year life-span of the Hyperboreans, cannot
be certainly claimed as Pindaric; for when Strabo (71 I) says that
Megasthenes in his account of India 'agrees with Pindar,
Simonides, and other story-tellers in his account of the Hyper-
boreans who live for a thousand years', the agreement may have
only been about their way, not their length, oflife-it may only
have been Megasthenes who mentioned the latter. For Megas-
thenes undoubtedly means the Uttarakurus here by 'Hyper-
boreans'; it was perhaps he who introduced the knowledge of
them to the Mediterranean world. He may have reported their
Indian name, for Pliny knows of 'Attacori' in the north of Asia
who lead a life like that of the Hyperboreans (.N.H. iv. 90; vi. 55,
where he says that a certain Amometus30 wrote a book about
them); they are the '0T7opoKoppai of Ptolemy (Geog. vi. 16. 5;
cf. vi. 16. 8 ; viii. 24. 7; oros. i.2 .45). Some knowledge of Mount
Meru itself (Mijpos) had reached Greece as early as c. 300 B.c.,
undoubtedly through Megasthenes or some companion ofAlexan-
der (Theophr. H.P. iv. 4. I; cf. Strabo 687).
Pliny and Ptolemy took the Uttarakurus for a real people: were
they entirely creatures of fantasy, or were the Indian stories
about them based upon a memory of an actual tribe? H. Jacobi
inclines to the latter view: their prototypes were a real people
living in the Himalaya (E.R.E. ii. 700a; cf. 699a). If this is true
it is decisive for our present investigation, for it would mean that
the Indian belief in a blessed people living to their north beyond
a lofty mountain was not an importation from central Asia but
had grown up in India itself-whencesoever they derived their
belief in the lofty Mount Meru. Indeed, it is safer to assume that
central Asiatic lore owes to India its cosmic mountain, together
with the cosmic ocean associated with it; for the idea of such an
ocean would not be likely to originate with an inland people
(Holmberg 345; 331). It is noteworthy that there seems to be
no trace of a northern Asiatic belief in a blessed people in the
far north, which here is rather the gloomy abode of the dead
 too TALES OF A TRAVELLER
(Holmberg 486) ; just as the Scythians knew nothing of any
'blessed Hyperboreans', according to Herodotus (iv. 32).
It appears, then, that it would be mistaken to identify the
Rhipaean mountains in any way with Mount :Meru, or the
Hyperboreans of the Arimaspea with the U ttarakurus. The original
of the Rhipaeans might be, quite generally, the western outliers
of the central Asiatic massif; or, if we must look for something
more specific, one of the high mountains in the region of the
Dzungarian Gate, such as the Altai, whose name means 'golden'
[73), and whose summit touches the Milky Way according to a
(modern) Chinese source (Humboldt i. 241; 230). It is worth
remarking that the Goldes of eastern Siberia, who believe that
the winds blow from mountain caves, are a branch of a race
which seems originally to have dwelt near the Altai, but is now
widespread over central and north Asia, the Middle East and
east Europe (Holmberg 299 f.).
I think that it is now safe to assume that Aristeas took with
him from Greece ready-made the concept of the happy race of
Hyperboreans, the favourites of Apollo, and that they were the
goal of his journey, which otherwise seems motiveless; just as he
took with him the concept of the griffin, which he attached to
monsters described to him by the Issedonians. But, such is the
respect that his veracity should now command, we must infer that
the Issedonians told him of a distant people with whom his pre-
conception of the Hyperboreans was not incongruous. That they
did tell him of such, and that he did not himself, having heard
of the starting-point ofBoreas from them, simply assume that the
Hyperboreans must dwell beyond it, is proved by the information
that this people 'reached to the sea' : the idea of an outer sea
would be foreign to a seventh-century Greek, for whom the earth
would be bounded by the river Oceanus. I incline strongly to the
opinion of those who hold thatAristeasheardfrom the Issedonians
something of the civilization of China. 31 To semi-savages living
hard and dangerously the ordered society of China might well
appear a model ofpeace,justice, and luxury; and to men whose
staple food was meat the agricultural Chinese, whose staple food
was cereal, could well appear as vegetarians. The claim of the
 TALES OF A TRAVELLER IOI

Chinese to be Aristeas' Hyperboreans is, I think, to be preferred

to that of the Argippaei [71], whose territory did not extend to
the sea, and for whom a place in the Arimaspea is not confirmed
by other evidence.
If I am correct, though the 'Hyperboreans' of whom Aristeas
heard from the Issedonians were the Chinese, the original Hyper-
boreans of the Greeks could, of course, have been nothing of the
sort. The origin of this fabled people has been the subject of
much dispute, into which there is fortunately no need to enter
here at length. It is sufficient to say that, if the interpretation of
'Hyperborean' as 'beyond Boreas' is mistaken, it was a mistake
already current in Aristeas' day, who almost certainly located
the Hyperboreans beyond Boreas. Personally, though I know that
the ancients were often wildly wrong in their etymologies, I see
no reason why they should be wrong in this case. 32
So strong evidence has now been produced that the Arimaspea
preserved genuine central Asiatic folk-lore that I think we are
entitled to reverse the process, and now to confirm for the poem
with some confidence our tentative claim to two items which are
not expressly attributed to it: the 'feathers' which the inhabitants
of the regions beyond Scythia said filled the air and covered the
ground to the north [ 42 ff.]; and the 'swan-shaped' Phorcides of
Aeschylus [62]. In the Chinese 'Annals of the Bamboo Books' we
are told that King Mu (?961-?906 n.c.)
in his expeditions to the north, travelled over the country of the moving
sands, for 1,000 le, and that of 'Heaps of Feathers', for 1,000 le. Then
he subdued the hordes of the K'euen, and returned to the east, with
their five kings as captives. Westwards, he pushed his expeditions to
where the green birds cast their feathers (the hill of San-\\"ei). On
these expeditions he travelled over 190,000 le. 33

So too central Asiatic tales tell of swan-maidens, ugly-they

have leaden eyes, hempen plaits, and yellow nails-and mur-
derous, who live in darkness. I accept E. R. Dodds's attractive
suggestion that these underlie Aeschylus' epithet (never before
satisfactorily explained) KVKi61-wpcpoi, 'swan-shaped', for the Phor-
cides, through the Arimaspea. 34
 10~ TALES OF A TRAVELLER
I venture to suggest a reason why Aristeas may have been led
to place the Phorcides and their sisters and neighbours, the Gor-
gons, where he did. If he heard from the Issedonians tales of
creatures ugly, malevolent, and endowed with the power offlight,
his thoughts might well have turned to the Greek Gorgons. But
there is another feature, common in Mongol and allied folk-lore,
which may have played its part: the ability ofsupernatural beings
to doff parts of their body. So swan-maidens can put aside their
skins; so too we hear of ogres who can take out their hearts and
secrete them elsewhere. If a Greek heard some such story of a
swan-maiden, of what would he be reminded more strongly than
the Phorcides, with their detachable and transferable eye and
tooth? That same process of Hellenizing barbarian story which
I have postulated in the case of griffins and Hyperboreans may
then have taken place here also, with the resultant location in the
Arimaspea of Gorgons and Phorcides in the north-east of the world,
which had its effect later on both Aeschylus and Pindar [62].
That Herodotus' Androphagi or Cannibals also had a place in
Aristeas' poem I have little doubt. They appear in Pliny in
association with Nomads (i.e. Scyths), Sauromatae, Issedonians,
Arimaspi, the Rhipaean mountains, and the region of 'feathers' ;
again, with the cave of Boreas, Arimaspi, and griffins (N.H. iv.
88; vii. 9 f.); and with Arimaspi and Melanchlaeni (Hdt.
iv. 20 al.) in a rather curious entry in Pollux, where we are told
that 'the Scythians, and of these especially the Androphagi and
Melanchlaeni and Arimaspi, use the bones of eagles and vultures
as flutes' (iv. 76: l:KJOa, ol, Kai aAiaTa TOVTWV J4vopo</,a:yoi Kai
MEAd')'XAawoi Kai J4piaa1rot, aETWV Kai yv1rwv OUT0S avA7JTtKWS
l1r1,lovaw). They are of the same order as the one-eyed Arimaspi,
with whom indeed we have found them blending in popular
tradition [83].
Further than this I am not prepared to venture. The goat-
footed people who inhabit the mountains to the north of the
Argippaei, and those others beyond them who sleep for six months
of the year (Hdt. iv. 25), have the flavour of Aristeas (the latter
are perhaps heard of again, as people who 'die' for six months of
the year, in the sixteenth century-Hakluyt ii. 330), but as I have
 TALES OF A TRAVELLER 103

said, I see no evidence that the Argippaei, said to be the authors

of these stories and at the end of the regular trade-route, were
mentioned in the Arimaspea. How dangerous it could be to assert
the authorship of Aristeas for a Scythian tale just because it is
quaint is illustrated by the Neuri, who became wolves once a year
and were driven out of their country by an incursion of snakes.
This is certainly quaint enough, but the latter event is said to
have happened only a generation before the invasion of Darius
(Hdt. iv. I 05) !
So too with the giant reeds and Dog-men [6gJ: though the
former are actually found, growing to a height of ten to fifteen
feet, in Dzungaria (Carruthers 437), and though the latter are
vouched for by the same ancient Chinese source that mentions
the One-eyed men and the giant ants [82], I am chary of accepting
them for the Arimaspea without further corroboration. It is true
that the Dog-men were known early in Greece, to 'Hesiod',
though where he put them is not certain; but Herodotus locates
them in the west of Libya : if the Arimaspea put them in the op-
posite part of the world it is strange that he says nothing of it
[192].
 V

THE TRAVELLER'S ROUTE

WHERE were the Issedonians of Aristeas? The question is a

peculiarly difficult one to answer, not only because of the scanty
information we have about them, but also because of the constant
movement and intermingling of the peoples of Russia since his
day. Before giving the views of some modern scholars I shall
survey such ancient testimony as we possess.
Aristeas himself placed them next to the Scyths, in the direction
of Boreas. His term 'Scyth' probably had a wider connotation
than that allowed to it by Herodotus, and he may have included
in it peoples whom the later author denies to be Scythian; for
instance the Androphagi and the Melanchlaeni (for the former,
compare Pliny, N.H. vii. 9 with Hdt. iv. r8; the latter were
called Scythian by Hecataeus of Miletus, 1 an appellation con-
tradicted by Herodotus (fr. 185; Hdt. iv. iw)). In Aristeas' day,
too, the Scythians may have extended over a wider area, being
still in the act of migrating. His Scythia, then, may have covered
much more than the Scythia of Herodotus. Aristeas too may have
been responsible for the statement that the Issedonians 'faced'
the Massagetae [60]; a statement so important for our present
purpose that it is a pity that we cannot be sure what it means.
As it is, it provides some sort of a check, inasmuch as any localiza-
tion of the Issedonians must take account of it and suggest how
it is to be interpreted.
In his fourth book Herodotus gives the following account. East
of the Tanais we leave Scythia and enter the country of the
Sauromatae, extending northward for approximately 350 miles
(fifteen days' journey; in iv. 101 he estimates a day's journey as
200 stades). It is entirely devoid of trees. Beyond this is a well-
wooded region inhabited by the lice-eating2 Budini (2I), a large
nomadic nation having red hair and blue eyes; also by the
Geloni, agriculturalists after whom is named a wooden town,
 1
';
THE TRAVELLER'S ROUTE 105

Gelonus. In the most thickly wooded part is a large lake

bordered by marsh, where are taken otters (lvuSptEs), beavers,
and 'animals with square faces'3 (108 f.).
Herodotus does not define the extent of this territory, but says
that beyond it (presumably still in a northerly sense) the traveller
will pass through uninhabited country for about 160 miles, after
which, taking a more easterly course (a.1roKAlvovn p,a.>J..ov 'Tl'pos
6.1TTJALWT"'JV ave.ov), he will reach the Thyssagetae and the I yrcae.

Both these are hunting peoples, the latter firing their arrows from
hideouts in trees, then chasing their (wounded) prey on horse-
back accompanied by their hounds. Inclining east yet again
t.-. > \I )
a7TOKI\LVOVTL you come upon a
f \ ~\ , ' \ \
( VTTEP OE TOV'TWV TO 1rpos T"'JV 7JW
group of Scyths. These Herodotus says seceded from the rest
and settled there (22); they may in fact have remained behind
when the main horde moved into the Pontic area (How and
Wells).
So far the way will have lain through rich plains, but hence-
forth the character of the ground becomes rough and stony
(ixp, fLV 0~ rijs TOlJTWV TWV EKvelwv XWtyrJS EC1'Tl ~ KO.Ta.AfXBEtcro.
~ ~ f I (:/ () I I <;,> > I I \ () f <;, > , I
1/'aCJo. 7TEOLO.S TE 'Y7J KO.L ,..,a vymos I TO O 0.1TO TOVTOV "' WOTjS T EOTL
1ml TPT/xl.o.) for some distance. Beyond this rough tract are the
Argippaei at the foot of lofty mountains. These Herodotus de-
scribes as large-chinned, snub-nosed, and bald from birth irre-
spective of sex. Though dressed like the Scythians they speak
a different language and have a different diet; for a tree grows
there about the size of a fig tree, called ponticum, producing fruit
like a bean, with a stone inside. The juice of this, which they call
aschy, they lap or mix with milk; the lees they make into cakes.
Their use of milk indicates partial nomadism, but they have not
many flocks because the pasturage is poor. They make their
homes under trees, which in winter they cover with white felt.
They are an unwarlike people, regarded as sacrosanct by their
neighbours, in whose quarrels they arbitrate and to whose
refugees they offer safe asylum (23).
So far the country is well known through Greek and Scythian
traders, who have to transact their business through interpreters
in seven languages (though Herodotus has only enumerated six
 106 THE TRAVELLER'S ROUTE
non-Scythian peoples on the route; perhaps the dialect of the
separate Scythian group also needed interpretation) (24). Of
the region to the north, which is cut off by an impassable moun-
tain range, nothing is known except wild stories told by the
Argippaei of men with goats' feet in the mountains themselves,
and of men beyond who sleep for six months of the year. But it is
known for certain that the territory to the east of the Argippaei
is inhabited by the Issedonians (25).
Our next informants are Mela and Pliny, writing 500 years
after Herodotus. The former repeats the Herodotean account of
the Issedonians' customs, but places them in Europe, west of the
Tanais between the Palus Maeotis and the Arimaspi, north of
whom are the griffins and the Rhipaean mountains where the
perpetual snowfall denies sight or passage (ii. I f. ; 9; I 3). Pliny
also places them in the vicinity of the Palus Maeotis in one
passage; but in another near the Colchians, and in yet another
north of the Jaxartes, where he well adds 'nee in alia parte
maior auctorum inconstantia, credo propter innumeras vagasque
gentes' (N.H. iv. 88; vi. 21; 50).
Ptolemy, in the second century of our era, mentions two towns
called Issedon in eastern Asia; which he distinguishes by the
epithets Scythica and Serica ; he places the Issedonians themselves
about the latter (Geog. vi. 16. 5 and 7; viii. 24. 3 and 5; cf. Amm.
Marc. xxiii. 6.66), Both towns are due north of the Himalaya on
his map, apparently in Chinese Turkestan (Sinkiang) ; modern
authorities differ about their exact situation (Thomson 310).
This completes the ancient testimony. I now propose to sum-
marize the views ofsome4 modern scholars [Map II].
1. Humboldt (i. 389 ff.): Moving in a generally northeasterly
direction from the Palus Maeotis, we shall pass according to Herodotus
through the Melanchlaeni, the Budini, the Thyssagetae, and the
Iyrcae (who have been mistaken for a Turkish people by some
scholars), and reach, towards the east, an isolated colony of Scyths
(perhaps middlemen in the gold and fur trade). Here the plains end
and we enter an extensive rough tract, rising into high mountains at
the foot of which are the Argippaei. Their physiognomy has led a
number of people to identify them as Mongols (Kalmuks), but the
Mongols did not penetrate so far west until the thirteenth century;
 THE TRAVELLER'S ROUTE 107
; ,

i they may have been Finnish. The name they gave to the juice of the
I
fruit produced by the ponticum (thought to be the bird-cherry, Prunus
padus) has suggested to Errnan that they were the Bashkirs, a Turkish
[Turco-Tatar]-speaking people of Finnish stock who now inhabit the
southern Urals [for the Bashkirs use the juice of the bird-cherry as
the Argippaei did their aschy, and call it atchui 5]. But we have no
historical notice of the presence of the Bashkirs in western Asia until
the tenth century of our era.
The regions beyond the Argippaei are unexplored because of high
mountains which bar the way. Now Herodotus in these chapters is
speaking not of one mountain range but two: this impassable range,
which clearly has a west-east orientation, is the Altai; but that in the
foothills of which live the Argippaei is the Ural. The route then crosses
the Urals from west to east; the Argippaei are to be placed on the
eastern slopes of these, and the Issedonians east of the Sea of Arai
and the upper Ishim, and north of the Jaxartes-Araxes (Syr), across
which they face the Massagetae to the south; perhaps between Kar-
karalinsk and Semipalatinsk. The Arimaspi would be on the northern
slopes of the Altai (the home of the goat-footed men), round the north-
western projection of which they would communicate with the
Issedonians; the griffins' gold would be the mineral resources of that
area, particularly rich in the Kuznetsk mountains. These considera-
tions put out of court a suggested location of the Issedonians about the
little river Iset in the Urals, or of the Arimaspi in the south of that
chain, rich though it is in gold.
It is strange that Herodotus appears to make no mention of the
Volga in his itinerary. This may be because he thought ofit as a con-
tinuation of the Don (which he says rises in the country of the Thyssa-
getae, iv. 123), across the narrow strip separating them [they are
now joined together here by a canal in !at. 48.5 N].

The weakest point in this reconstruction is undoubtedly the

attempt to force a reference to the Urals out of the oprn vi/rr}M. of
Hdt. iv. 23, and to a crossing of them in 3it>.0ovn 111, rpl)xl71s
xwpov 1ToAMv (such appears to be Humboldt's idea). The 'rough
and stony tract', being opposed to rr3ids r 1,fj Kal ~a8vya,os,
conveys the meaning of barrenness, not at all applicable to the
Urals. Also, no one would describe these inconsiderable hills as
'lofty': much of the chain consists oflow plateaux, only occasion-
ally rising to peaks over 4,000 feet. It would be very odd too for
 1o8 THE TRAVELLER'S ROUTE
Herodotus not to mention them specifically until after they had
been crossed; in fact, I do not think that his language can have
any other sense than that these 'lofty mountains' are distinct
from the 'rough and stony tract' he has just mentioned. Again,
the outliers of the Altai are some 700 miles from the Urals: it is
incredible that the Argippaei in the latter, with their stories of
goat-footed men and so on, could have been referring to the
former. The op1:a v,frr,">ia of c. 23 are clearly the same as the op1:a
v~ri">id. d.f3a-ra of c. 25, which Herodotus conceives as running from
west to east to the north of both Argippaei and Issedonians.
2. Minns (104 ff., mainly following Tomaschek): The Sauromatae
stretched up the Don and Volga nearly to Saratov and the for.est
region ; the territory of the Budini probably covered the lower reaches
of the Belaya and Kama. The Budini may be represented today by
the grey-eyed and reddish-haired Permiaks (and Ibn Fadhlan said of
the Bashkirs that they ate lice). The town of Gelonus perhaps stood
on the site of the modern Kazan, a trading-post of great antiquity.
The Thyssagetae would occupy the western slopes of the Urals from
Ufa to Orenburg; a hilly country would be consistent with the fact
that several rivers have their source there (iv. 123), and one of these,
the Chussovaya, may have the same root as Thyssagetae. Tomaschek
identifies them with the Voguls (and extends their territory to the
junction of the Iset and the Tobol]. Herodotus disregards the Urals,
whose incline is so gentle that they do not appear to the traveller as
mountains. The method of hunting of the I yrcae (the name is a Sarma-
tian fonn of Jugra) requires a country full but not over-full of trees;
the basins of the Tobol, Ishim, and Irtysh would fit. The 'separated'
Scyths [whom Tomaschek places from the Kulunda steppe north of
Semipalatinsk across the upper Irtysh to the Ob] were perhaps
remnants left behind in the original home of the Scyths (who were
Ugrian?). The rugged tract after passing these is the outliers of the
Altai, and the Argippaei (pure Mongols) would be in Dzungaria,
between the Altai and the Tien Shan. Their ponticum may be the bird-
cherry, but many other steppe berries are treated similarly by various
tribes. The 'trees' covered with felt under which they lived are a
description of the felt tent on a light wooden framework now univer-
sally used by Asian nomads. The Issedonians east, or rather south-
east, of the Argippaei could be Tibetans in the Tarim and Bulunggir
basin, where Ptolemy puts them (his Issedon Scythica is Ak-Su ?),
'opposite' the Massagetae to the west across the Pamir. The Arimaspi
 THE TRAVELLER'S ROUTE 109

are Huns in Mongolia, of whom the Argippaei would be a western out-

post. The Issedonians may have got gold from them, or perhaps from
the mountains to the south, above India. The Hyperboreans are either
purely imaginary, or may be an inkling of the Chinese; while the
Rhipaean mountains might be any of the many ranges in central Asia.

The 'separate' Scyths are very difficult in this location. Also

Herodotus quite definitely says that the Argippaei got their aschy
from the fruit of a tree, and there are few trees in the barren plains
ofDzungaria. He also says that they did without the felt covering
on their trees in summer; but at no season of the year do the
Asiatic nomads dispense with the covering of their yurts and
live in the framework! And the explanation of how the Isse-
donians are 'opposite' the Massagetae is very forced.
3. Westberg (183-7; 189-92): (a) Position of the lssedonians
according to Herodotus. The Sauromatae extend from just above the
mouth of the Don to Kamyshin. Herodotus says nothing of the extent
of the territory of the Budini, but it must have been considerable, as he
tells us that they were a large tribe. His account ofits features excludes
the treeless left bank of the Volga ; but the right bank in olden times
was wooded farther south than today-cf. the testimony of Ibn
Rosteh in the tenth century, who also speaks of open fields here (such
as the Geloni worked), but says that the principal products of the
locality were honey and furs; the extent of this region he puts at
seventeen days' journey both in breadth and length. This would
extend the domain of the Budini from Kamyshin to the north of
Syzran. The 'desert' to the north of the Budini which takes seven days
to cross corresponds to the Zhigulev hills, which e..xtend along the
right bank of the Volga to near its junction with the Kama, up which
the route is deflected eastwards. Here are to be located, in the bend of
the Volga and the lower reaches of the Kama, the Thyssagetae (not
farther east in the Urals-cf. the 7T&uls yij of c. 23). The account of
the Iyrcae's method of hunting shows that they lived almost in the
steppe. To the east of them were the 'separated' Scyths. This proves
that the trade-route did not follow the Kama to the middle Urals, but
kept an easterly direction along the Belaya to the high and thickly
wooded southern Urals, which is the rough, stony tract leading to the
Argippaei (Bashkirs?) at their tip. (Herodotus thought ofthe Urals as
running west to east because he identified them with the Rhipaeans.)
East of the Argippaei are the lssedonians, opposite the Ivlassagetae
 110 THE TRAVELLER'S ROUTE
across the Araxes:Jaxartes (Syr); whether Herodotus was right in so
placing them is another matter, but Tomaschek does too much
violence to his account in putting them in the Tarim basin in an
attempt to reconcile him with Ptolemy.
(h) Position ef the Issedonians according to Aristeas. Herodotus'
information concerning the peoples along this trade-route was derived
from oral reports; but that about the lssedonians (except perhaps for
their location east of the Argippaei), Arimaspians, griffins, and Hyper-
boreans from a written source---the poem of Aristeas (first half of
seventh century)-in spite of the impression given in c. 27 that he
may have got this too from the Scythians. Hence too he got the
etymology of Arimaspus as 'one-eyed', as is shown by [fr. 5] [?] (the
etymology of this as 'owners of docile steeds' or 'of wild steeds of
the steppe' is dubious; could it be ari ('noble')+ Maspii, a Persian tribe
(Hdt. i. 125) ?). Hdt. iii. 116 is pertinent here. According to this
passage the north produces most gold (which Herodotus connects with
the griffins and Arimaspi) ; we are not entitled to associate this
northern gold (the gold of the Urals) with that ofBaltistan and Tibet.
Indeed, if we compare the fragment of Damastes [39 ] where the
Rhipaean mountains are mentioned, but not the griffins, with Hdt.
iv. 13 [T. 1] where the griffins are mentioned, but not the Rhipaean
mountains, we may doubt if Aristeas had anything to say about the
gold-guarding griffins at all; for the fragment ofDamastes seems likely
to be closer to the text ofAris teas than is Herodotus, as being more con-
sistent and detailed. The griffins belong to India (Herodotus and
Ctesias), and it would be natural [for Herodotus, presumably, for
the Scythians could hardly know so much] to say that the northern
gold was got in the same way as the eastern gold.
The shifting of peoples in eastern Europe took place in or before
the time of Aristeas; there were no more changes between him and
Herodotus. Therefore, as Aristeas says that the Issedonians were the
next-door neighbours of the Scyths and had caused them to migrate,
his Issedonians are to be identified with the Sauromatae, whose
womenfolk were equal with the men (cf. what is said about the
Issedonians in c. 26) ; and it is tempting to see the Arimaspi in the
Massagetae.

There is a marked discrepancy between the scrupulous way

in which Westberg treats the evidence of Herodotus concern-
ing the trade-route to the Argippaei (except perhaps for the
glossing over of the 'separate' Scyths) and that in which he
 THE TRAVELLER'S ROUTE 111

treats the same author's testimony concerning the content of the

Arimaspea. We cannot thus over-ride Herodotus when he says
plainly that Aristeas put the gold-guarding griffins beyond the
Arimaspi; and that he did not insert this himself on the analogy
of the ant/Indian story [66] is proved by Aeschylus' association of
Arimaspi with gold and griffins in the Prometheus Vinctus. Also,
the identification of the Issedonians with the Sauromatae is pre-
cluded ifit is true that the latter were aboriginals, already occupy-
ing the country when the Scyths arrived [51 f.].
4. Hermann : The placing of the Issedonians in the far east, under
the influence of Ptolemy, is wrong. Only the evidence of Herodotus
and Aristeas is of any value, and this shows that they were lndo-
germanic nomads living east of the Urals about the lower Tobol and
the !set, which is obviously named after them (they appear too in
Ptolemy in this neighbourhood as Iastae [?]). The evidence of later
writers has been distorted by false associations.
Aristeas called them Issedi, but, as Aleman shows, the longer form
was soon in use. Hecataeus is the first to our knowledge to uselssedimes;
the form Essedones, it may be conjectured [why?], was given currency
by a younger Ionian geographer, Dionysius of Miletus. The Greeks
first heard of them from Aris teas (second half of seventh century), who
appears really to have visited them. He placed them north of the
Scyths and south of the Arimaspi and griffins. His placing of the
Rhipaean mountains north of these last merely reflected older
mythological ideas : they are the northern range which was invented
by the imagination of the old astronomers. 6 The same can be said of
the Hyperboreans.
The Arimaspea influenced poets (e.g. Aleman) and geographers.
Hecataeus undoubtedly [ ?] put the Arimaspi under the eastern part
of the Rhipaeans, which extended right through Europe, and beneath
them the Issedonians, as a subdivision of the Scythians. The view of
Kiessling, that Hdt. i. 201 [where Herodotus says that the Issedonians
'face' the Massagetae] derives from Hecataeus, is wrong; it was a
later modification of Hecataeus' work that made the Issedonians an
Asiatic people; for though Stephanus puts the Issedonians ofHecataeus
in Asia we know that for the Ionian cartographers the continental
boundary was the Phasis (but cf. 190!].
It was probably Dionysius of Miletus who more accurately defined
the country of the Issedonians, as a result of information about a
caravan route beyond the Tanais. Not one fragment of Dionysius
 112 THE TRAVELLER'S ROUTE
remains [!], but we know Herodotus used him for his account of
Scythia [?], as did a later historian (Theophanes of Mytilene ?), from
whom Pliny vi. 1 g and Mela i. 1 16 derive. These passages enable us
to determine what Dionysius had to say about the situation of the
Issedonians. He, like Aristeas and Hecataeus, put them north of
the Pontus but west of the Tanais; it was beyond the Tanais that the
tribes about which he had new information dwelt. The peoples west
of the Tanais were, from north to south, the Arimaspi, lssedonians,
Scyths; and east of the river Arimphaei, Thyssagetae, Tyrcae, Budini,
Sauromatae (when Mela and Pliny put the lssedonians west of the
Tanais they merely repeat Dionysius).
Herodotus, doubtless using information about the same caravan
route as Dionysius, moved the Issedonians over the Tanais to the far
north-east of the earth. Dionysius knew as his farthest people the
Arimphaei, with high mountains cutting off further knowledge;
Herodotus must have heard that east of them were the Issedonians,
so their neighbours the Arimaspi were shifted by him far away to the
east. The 'high mountains', in Herodotus north of the Argippaei
[= Arimphaei ?] , were probably originally a range running north to
south between the Argippaei and the Issedonians [why?]. As a result
of his new information about the way to these people, Herodotus was
forced to separate the Caspian Sea from the encompassing Ocean [in
which he did not believe [60] !]. This positioning of the Issedonians
accounts for their being 'opposite' (i.e. north of [ ?]) the Massagetae.
Later authors who misplaced the Issedonians in the region of the
Caucasus were misled by a confusion over Herodotus' Araxes (i. 202-
itself a confusion) with the Armenian Araxes. From Herodotus'
conjunction of the Issedonians and the Massagetae, and taking the
Araxes here as thejaxartes, Marinus of Tyre supposed there to be an
Issedon Scythica (Ptol. Geog. vi. 16. 7; cf. Amm. Marc. xxiii. 6. 66); his
Issedon Serica farther east is based on a misunderstanding of the
Itinerary of Maes (end of first century A.D.), whose knowledge of the
east Turkestan silk-route came not from travellers themselves but an
Iranian guide-book. It is possible with the help of c. 96 of the Han
Annals, which uses this same source, to reconstruct it exactly and see
where Marinus and Ptolemy followed, expanded, or altered it. From
this it seeins that Marinus thought the Massagetae were west of
Kashgar, and the inhabitants of the Tarim basin (who are not named
by the Han Annals or later Chinese authorities, and not being of
uniform stock probably never had a common name) to be Issedonians
also. It was Marinus, then, who put the Is.~edonians in eastTurkestan;
 THE TRAVELLER'S ROUTE 113
as this was a palpable misunderstanding, his authority is worth-
less.
Earlier discussion of the position ofthe Issedonians has been bedevilled
by an uncritical acceptance of the various ancient traditions. Toma-
schek's identification of them with Tibetans, unlikely enough in itself,
does not in the least fit the time to which Ptolemy's authority belonged;
and his whole inquiry suffers because he continually interprets
Herodotus' evidence in favour of Ptolemy. (Kiessling accounted for
the varying traditions about their position by postulating that the
Issedonians themselves moved about.)
The only authorities of value we have are in fact Aristeas and
Herodotus; and to them we must limit ourselves. Now as Aristeas'
route was from south to north, we might think we were concerned
with the prehistoric amber-route from the Baltic to the Black Sea [the
existence of which is unproved: P.-W. iii. 298 f.]; but as there are no
remarkable mountains on this route we must give Aristeas another
direction (no doubt the notion of a great northern range and Hyper-
boreans caused him to give his route a more northerly oriP.ntation
theoretically than in fact it had). In his time we must remember that
the Scyths stretched as far as the Volga; so our thoughts are turned to
that old trade-route mentioned by Herodotus and Dionysius. This
would follow the Volga north, until it turned north-east and east at
the junction with the Kama. Up to this point there would be, in
Aristeas' time, Scythians and the related Geloni. Dionysius did not
know of lssedonians on this trade-route; his knowledge stopped at
the Urals, in which he saw a branch of the Rhipaean mountains (and
therefore gave them a west-east direction). Herodotus is better in-
formed, and does not say that these are the Rhipaeans; he speaks only
of them as high and impassable, and puts the Argippaei at their foot,
the Issedonians to the east. We have now reached the eastern side of
the Urals, in the neighbourhood of the west Siberian steppe; here
were the Issedonians of both Aristeas and Herodotus, a placing which
is confirmed by the name of the river Iset (the riYer mentioned by
Zenothemis ?), where in the seventeenth century there was a district
called Issetia. Westberg's identification of the Issedonians of Aristeas
with the Sauromatae of Herodotus fails because the latter were east
of the Scyths, not in the path of Aristeas' northerly journey. We have
seen that there is no need to give any weight to Herodotus' conjunc-
tion of the Issedonians with the Massagetae in the neighbourhood of
the Aral Sea.
The Arimaspi would be on the old trade-l'Oute from the Tobol to
0160
 THE TRAVELLER'S ROUTE
Omsk and then up the Irtysh to the Altai, which with its rich gold
deposits provides the best scene for the griffins and the Rhipaean
mountains.
I concur with Hermann's dismissal of the evidence of Ptolemy,
if only because it is half a millennium later than that of Hero-
dotus, during which period there had been much movement of
the Eurasian tribes. But his cock-sure reasoning about the true
situation of the Issedonians, apart from being at times hard to
follow, I find quite unacceptable in its premisses. Useful though
it is to examine closely Herodotus' account of the peoples along
the Scythian trade-route, in order to try to discover where he
put the Issedonians, there is no ground for the assumption that
this was the route which Aristeas himself followed (Hermann
seems to imagine that the Scyths came from north Europe, when
he says that in Aristeas' day they would have extended as far as
the junction of the Volga and the Kama-in heavy forest!). And
even granting that the river called Iset preserves the name of the
Issedonians, nothing more can be inferred from it than that this
people lived at some time near it; just as the migrant tribes of
Rus (Ros) and As left their names in passing to rivers in south
Russia (Vernadsky 97; 148).
All these modern theories about the Scythian trade-route are
helpful, yet none is in my opinion entirely satisfactory. Though
Herodotus gives us little in the way of distances, he does tell us
enough about the characteristics of the terrain and the various
tribes along the way to show the sort of country through which it
passed. After the Sauromatae in the steppe, it traversed wood-
land mixed with pasture (Budini and Geloni), a wilderness (which
may have been uninhabited forest), then eastwards through
forested uplands-forested, because the inhabitants, the Thys-
sagetae, were hunters; uplands, because a number of rivers had
their source there. The forest must have been thinning out in the
country of their neighbours, the lyrcae. Another easterly in-
clination now brings us to a branch of the Scyths; that is, we are
out on the steppe again. A stony tract next leads us to the Argip-
paei (Turco-Tatars?) in a region where the trees are not too dense
to prohibit the rearing of flocks, at the foot of high mountains.
 THE TRAVELLER'S ROUTE
I would accept Westberg's exposition of the route as far as the
territory of the Thyssagetae, but this people I would spread over
both sides of the southern Ural in the area Belaya-Kama-Iset-
Tobol (cf. Tomaschek). The Iyrcae would then be in the semi-
steppe about the middle Ishim, and the 'separate' Scyths in the
steppe to the south-east. The road would now follow the northern
edge of the Kazak uplands, a region of low broken hills and
granite outcrops (Humboldt ii. 122-37)-the 'rough, stony tract'
of Herodotus-to the vicinity ofSemipalatinsk, across the middle
Irtysh to the semi-steppe between that and the Oh, on the
western skirts of the Altai (the Argippaei). I would place the
Issedonians not east but south and south-west of these, in
the steppe extending from the Irtysh to the head-waters of the
!shim, and perhaps farther towards the Syr, to 'face' the Mas-
sagetae beyond it7 (the Issedonians were a large tribe, according
to Zenothemis [68]). This violence I would excuse by supposing
that the Rhipaean range was not simply the Altai but the whole
western fringe of the central Asiatic massif; if these Rhipaeans
were imagined (as they were later) as being in the 'proper' place
for the home of Boreas, the north, and so slewed round to run
from west to east, carrying the peoples at their foot with them,
the Issedonians would be thought of as east of the Argippaei !
But how could Aristeas ever have supposed that his Issedonians
were in the north, if they were in fact in the same latitude as the
Palus Maeotis? The answer to this crucial question I believe to
be, quite simply, that he did not. Speculation on his geography
has been dogged, both in ancient and in modern times, by the
assumption that his Boreas, the only apparent clue we have to
the orientation of his journey, represents a cardinal point. But the
concept of fixed cardinal points is an advanced one, formed as
astronomical science and mathematical geography progressed.
Though the names of the winds were to be used for these points
by a more scientific age, it is anachronistic to suppose that this
could have taken place already in the seventh century B.c. The
germ of the idea is no doubt to be seen in the two pairs ofoppos-
ing winds in Od. v. 331 f., but even supposing that the east
and west winds were then defined by reference to heavenly
116 THE TRAVELLER'S ROUTE
phenomena-the rising and setting sun-(and Homer nowhere
implies this), it is still a long step to fixing east and west (for the
points at which the sun rises and sets vary through the year), and
an even longer step to fixing north and south. Homer's con-
ception, mythical though it is, that all the winds had a common
starting-point (the Aeolian isle) indicates that the association
of winds and 'points of the compass' had not yet got very far.s
In the first place the various winds would be identified, not
with the help of astronomy, but by the locality from which they
blew at any particular place, or by their effects.9 Thus Boreas is
a cold and powerful wind; and, for the Greeks on the western
seaboard of Asia Minor, blows from Thrace. But when Aristeas
had left Thrace far behind how could he identify Boreas? Only
by its characteristics of coldness and force; characteristics which
may be found also in the prevailing caster!), winds of south-east
Russia (E.B. xxiii 881b). It might be thus that Aristeas gave his
route an easterly direction, thinking that he was marching to-
wards the Hyperboreans beyond the source of Boreas; and that
source he might well have imagined he had found if he heard of
the bitter hurricanes which sweep through the Dzungarian Gate,
and of the Cave of the Winds there.
Here I must forestall a possible objection, that in fr. 4 Aristeas
himself says that the Arimaspi dwell beyond the Issedonians
Ka8V1Tp8 1tpo~ {3oplw, and that Ka8v1tp8 in itself means 'to the
north [of]'. That this is a misunderstanding of the word is suf-
ficiently demonstrated by Hdt. iv. I 74, TovTwv ot KaTvm,pOi 1tpo!;
VOTOV av.ov olKtOVut I'apa.aVT~. It seems to be based on an
interpretation of Od. iii. 169-72 which is not obligatory. Here
Menelaus comes upon Nestor's fleet at Lesbos, wondering
whether to round Chios by the north and west or by the east and
south:

Psyra is west of Chios, Mimas east; they might represent different

stages on each course, but they might (and I think do) rather
 THE TRAVELLER'S ROUTE

give exactitude to the prepositions: 'above' means not 'to the

north of'but 'on the far side of'; 'below', not 'to the south of' but
'on the near side of'-from the point of view of the poet and his
audience in Ionia. So too KafJv1repfJ in It. xxiv. 544 f. can have no
reference to the north (Achilles is describing the extent of the
kingdom of Priam):

oaaov A ea,-,o<;
,Q
I
avw, u I
1r1aKa.poc; '
eooc;, ev'To<; eepyn
Kal <Ppvyl'Y) Kallv-rrep(h Kal 'EM17a1roV'To<; chrelpwv.
Here KaBinr1:pfJE means 'inland' (and avw 'out to sea'?); the base-
line of reckoning, which gives the word its sense, may be either the
poet himself (as in the former example 10) or one of his characters
who is speaking (as perhaps in the latter). The Greeks seem to
have thought of a coastline as a sort of trough : you ascended from
it whether you went out to sea (dvam\dv) or inland (dva{Jatvnv).
So Ka8V1rEpfJe is used in topographical descriptions with reference
to an imaginary observer on the coast, and means simply 'up-
country' (cf. the use of {m/p with the genitive, e.g. in Hdt. iv.
170 f.U).
There is evidence enough that the ancients came to think of
the north of the earth as being higher than the south, but this
must have been the result of considerable scientific advance.
First had to come the discovery of the celestial north pole; then
would be asked, why is this pole not in the zenith (for the ques-
tioner would naturally assume that he was at the centre of
things)? The earliest answer to this question was, that the flat
disk of the earth was tilted towards the south; but this was
long after Aristeas' day, 12 and should not be connected with
the use of 1<afJv1rEpBE which I have been discussing.
Though such I believe to have been the state of affairs in
Aristeas' time, in the following two centuries, as a result of the
flowering of Ionian science, the quarters of the winds came to be
defined more and more exactly with the aid ofastronomy. 13 As a
result Boreas begins really to stand for the cardinal north; 1 and so
the Rhipaean mountains of Aristeas are shifted by the carto-
graphers from their true position in the east to the north.
There is, then, no objection on the score of direction to my
placing of the lssedonians; and that they were in that position
 118 THE TRAVELLER'S ROUTE
in Aristeas' time as well as in the time of Herodotus would be
probable, because there had been no major migration in between,
and is confirmed by the central Asiatic folk-lore which they
retailed to Aristeas. The region of the griffins, rich in gold, might
be the Altai and Kuznetsk mountains; but it might also be
located in the Tarbagatai mountains, between Ala Kul and Lake
Zaysan in the river Irtysh, on whose northern slopes are auri-
ferous streams whence in olden times gold was extracted (Hum-
boldt ii. 83; c iii. 501 ). Here, therefore, about the upper Irtysh
(Aeschylus' Pluton ?), I would put theone-eyedArimaspi, a branch
of the one-eyed Kuei (Mongols 15), who had pushed the Isse-
donians farther away from Mongolia and Dzungaria. I would
see the home ofBoreas in a legendary cave in the neighbourhood
of the Dzungarian Gate ; and the Rhipaean range in the succes-
sion of heights, forested on their lower slopes but often rising to
regions of perpetual snow, which form the western and north-
western extremity of the highlands of central Asia-perhaps
called Caucasus by Aristeas: it will not escape remark that later
the native informants of Alexander's expedition called the
southerly mountains of this group by a name which was inter-
preted as Caucasus. 16 (This may be the reason why Herodotus
does not name the 'Rhipaean' mountains [42] : because they were
called Caucasus by Aristeas, but he knew the Caucasus as the
historic Caucasus.)
Here, then, was the limit ofAristeas' journey. The way by which
he travelled there, however, can only be guessed. It was probably
through the 'corridor of the steppes', by which his Scythian
hosts must have come, and over which at that time they may still
have been strung out, rather than by the trade-route described
by Herodotus, except in its initial stages: he may have followed
the Tanais and the Volga (which he conflated into the Phasis ?) to
the vicinity of Saratov, and seen the Sauromatae or Maeotians,
whom he took for Amazons (the Sauromatae are called Scythian
by Hippocrates (Airs, &c. r7)). Had he followed the trade-route
throughout, through many peoples clearly non-Scythian, he
would hardly have said that the Issedonians were the Scythians'
neighbours. More we cannot say.
 VI
THE POET

ABOUT Aristeas himself fabulous tales were told from our earliest
testimony onwards. So far I have used his name as a convenience
for 'the author of the Arimaspea' : it is now time to examine the
evidence about him in detail, to review the theories of some
modern scholars, and to see if there is any good reason to discount
the ancient ascription of the Arimaspea to him. I start by repeating
the story in Herodotus [T. I 2].
According to a tale that Herodotus had heard at Proconnesus
and Cyzicus, Aristeas,
who belonged to one of the noblest families in the island, had entered
one day a fuller's shop, when he suddenly dropped down dead. Here-
upon the fuller shut up his shop, and went to tell Aristeas' kindred
what had happened. The report of the death had just spread through
the town, when a certain Cyzicenian, lately arrived from Artace, con-
tradicted the rumour, affirming that he had met Aristeas on the road
to Cyzicus, and had spoken with him. This man, therefore, strenuously
denied the rumour; the relations however proceeded to the fuller's shop
with all things necessary for the funeral, intending to carry the body
away. But on the shop being opened, no Aristeas was found, either
dead or alive (oVT TdJvd,rra OVT lwVTa ef,a{va8ai ltpLC1T11v). Six
years afterwards he reappeared, they told me, in Proconnesus (mi
ll~ J{Jllop.</) lTi ef,avlVTa av-rov lr; llpoKOVVTJCIOV), and composed the poem
which the Greeks now know as the Arimaspea, after which he dis-
appeared a second time (a.ef,aviu8fiva, To lldrEpoi). This is the tale
current in the two cities above mentioned.
What follows I know to have happened to the Metapontines in Italy
two [v.l. three] hundred and forty years after the second disappearance
of Aristeas, as I discovered by calculations I made at Proconnesus
and Metapontum (Td8 8 ol8a Mra1roVT{voun roia, l,, 'lraJ,.{TI avyKv
I \ \ ',,1._I ~ , I )1 I II I \
P"}CIO.JJTO. era. TTJV a.'f'a.VtaLV TTJV OEVTEPTJV p1C1TEW TWL TEC1C1paKOVTa KaL
8tTJKOalow, [v.l. TPLTJKOalow,], ws Jyw avfJa>..>.ooos EV IlpoKOW'tJCT'i'
u Kal MTa.1ToVTlcp dJpiu,cov). Aristeas then, as the l\Ietapontines affirm,
appeared to them in their own country in person (avrov i1p,arlr1v
 120 THE POET
<f,avlvra acp, Js 7'1]v xwp17v), and ordered them to set up an altar in
honour of Apollo, and to place near it a statue to be called that of
Aristeas the Proconnesian. Apollo, he told them, had honoured them
alone of the Italiotes with his presence (cf,avm yap a<pi TOV )1'1T6Mwva
'fro)..1WTWV JJ-OVVOtUt 0~ amfo8at Js 'T~V XWP'YJV) j and he himself
accompanied the god at the time, not however in his present form, but
in the shape of a raven. Having said so much he vanished (,;l'IT6vra
Tavra cicpav,a8fjva,). Then the Metapontines sent to Delphi, and in-
quired of the god what they were to make of this apparition. The
priestess in reply bade them attend to what the spectre said, 'for so
it would go best with them'. Thus advised, they did as they had been
directed ; and there is now a statue bearing the name of Aristeas, close
by the image of Apollo in the market-place ofMetapontum, with bay
trees standing round it.
In this, the earliest story we have about Aristeas, there are,
first, a miraculous disappearance-the dead body is animated
and rapt away by some supernatural agency (Apollo, one
supposes)-and, secondly, two supernatural reappearances. For
though, disregarding Aristeas' death in the fuller's shop, his first
disappearance from Proconnesus, followed six years later by his
reappearance there, would be perfectly compatible with ordinary
absence on travel, the words Herodotus uses about this re-
appearance are the same as those he uses to describe the second,
undoubtedly spectral, appearance to the Metapontines. The
Arimaspea, then, was composed by a ghost (or rather, a zombie!),
and the second disappearance too from Proconnesus we may
conclude was as mysterious as the first.
A century after Herodotus we meet a different story. In the
second chapter of the Historiae Mirabiles of Apollonius-a passage
probably derived from Theopompus 1-we read [T. 14]:
It is reported of Aristeas the Proconnesian that at the very time of his
death in a fuller's shop in Proconnesus many saw him in Sicily teaching
(yp&...aTa o,Ma-Kona). This kind of thing happened to him often,
and over a long period of time his appearances in Sicily became
quite frequent (m.JKv6Tcpov <f,av-ra(olvov) and generally known, so that
the Sicilians accorded a shrine and sacrifices to him as a hero (lcp&v
'I' "''
Ka8iopvuav-ro ,~ Kai'"8
avTqi '.
t vaav ws r,pwi")

On the face of it, all these Sicilian appearances are post mortem;
 I THE POET
there is no indication that the death in Proconnesus was not a
121

true death, and the phrase 'IToMiiKLS' ain-q, -rov TOL011Tov av.f3al-
voV7'oS' need only refer to the phantasmal manifestations, and not
imply a new 'death' on each occasion. It would, however, also be
compatible with the latter interpretation; and there was another
version of the Aristeas-legend in which his 'death' was but a
death-like trance, occasioned by the temporary absence ofhis soul
on disembodied peregrinations. 'Suidas' gives a report that the
soul ofAristeas could go forth from his body and return at will [T.
rr]; and it is this that Pliny means in a passage of proofs that
even death may be deceptive :
The soul of Hermotimus of Clazomenae would leave his body,
range abroad, and report distant happenings unknowable except to
an eye-witness, 2 while his body itself would be betwixt life and death.
Finally his enemies, the Cantharidae, bumt it, and so deprived the
soul on its return of its sheath. The soul of Aristeas too, about which
many a tall tale is told, was seen flying from his mouth in Proconnesus
1..
in the form of a raven.
,.
(N.H. vii. I 74: 'reperimus inter exempla Hermotimi Clazomenii
animam relicto corpore errare solitam vagamque e longinquo
multa adnuntiare, quae nisi a praesente nosci non possent, cor-
pore interim semianimi, donec cremato eo inimici, qui Can-
tharidae vocabantur, remeanti animaeveluti vaginam ademerint;
Aristeae etiam visam evolantem ex ore in Proconneso corvi
effigie, magna quae sequitur hanc fabulositate' [T. 15]). It looks
as if Pliny knew of stories about Aristeas similar to that which
he here tells about the psychic excursions ofHermotimus.
This form of the story finds a detailed exposition in two passages
of Maximus of Tyre, a sophist of the second century A.D.
According to the first [T. 19]
there was a man of Proconnesus whose body would lie alive, yes, but
with only the dimmest flicker of life and in a state very near to death
(lp:rrvovv ph, ID' ap.118pw, Kal lyyu-rara 8av<frov); while his soul
would issue from it and wander in the sky like a bird, surveying all
beneath, land, sea, rivers, cities, nations of mankind, and occurrences
and creatures of all sorts (1ml ,rnB~ara. i<a.l cf,Jrms rravrolas); then
returning into and raising up its body, which it treated like an
 1211 THE POET
instrument (,%,cnrep &pyd.vcp xpwp.EVTJ), it would relate the various
things it had seen and heard in various places.
I
.
The second passage concerns untaught sages : stories of inspira. i
tion or knowledge obtained by mysterious means, as of Hesiod
receiving the gift of poesy together with a laurel bough from the '~
Muses, of Minos receiving instruction in politics from Zeus in the
Idaean cave, and so on, allegorize innate genius. So too Aristeas'
edifying journeys in the spirit [T. 20]:
There was in Proconnesus a man who loved wisdom; his name was
Aristeas. Yet at first he was not credited with his wisdom, for he could
boast of none that had taught him it. So he thought of a device to
overcome men's disbelie He would tell how his soul would leave his
body and, flying up into the sky, would traverse the lands both Greek
and foreign, all islands, rivers, and mountains; how the limit of
his journey was the country of the Hyperboreans (yevlafJa, 8i -rijs
71'Ept71'o,\~aews a&rfi -rlp.a 7'1/V 'Y71'ep{3oplwv yf)v); how he thus obtained
a successive view of all usages customary and political, of varying
landscapes and climates, of inroads of the sea and outpourings of

rivers (.I.'l'VUE'S xwp,wv, /(al aepwv .e-ra,.,o
I I ' " IR ,\
as, Kat avaxvaEtS aAaTT'1js,
' ' I () .. \ I
I

Ka1 71'o-ra.wv rK{JoMs) ; and how his soul then had a much clearer view
of heaven than from below on earth (yevla8m 8l a&rfi i<a1 'Mjv -rov
ovpavofi Olav 7/'o,\u rijs vlpOev aa~UTEpav).

Aristeas' knowledge was not confined to geography and astro

nomy; it was not even confined to present time, but included the
future also, for he was an adept in prognostication, according to
Clement [T. 2 r]. Here may be another similarity with Hermoti-
mus, whose migrant soul not only brought back with it information
of distant events but during its wanderings appeared visibly in
various places and prophesied. So much is told by Apollonius
(Hist. Mir. 3), who supplements the account of Pliny already
quoted, possibly from the same ultimate source :
They say the soul ofHermotimus would wander away from his body
and absent itself for years on end, and, appearing now in one place,
now in another (?Ka-rd -r671'ovs y,vo.iv7Jv [leg. </>aivop.EV1JV ?]), would
foretell events such as floods and droughts, earthquakes, plagues, and
the Jike (op,/3pov, p.eyd.,\ov, Kai dvop.{Jp{a,, [n 8 ana,ovs 'TE Kai ,\otp.OUS
Kai 71'apa11>.~ma). After a certain time it would re-enter its body,
which had been lying inert, as into a sheath (l,\VTpav: there can be
 THE POET 123

little doubt that this is the word translated by Pliny's vagina), and
arouse it.
(There follows the story of the burning of the body by Hermo-
timus' enemies, after suborning his wife: this explains why no
woman is allowed to enter the temple which the Clazomenians
erected in honour of Hermotimus ([Epov airrov Ka8opvTa,-cf.
the phrase which Apollonius used of the Sicilians and Aristeas
[ 120]). 3
So many coincidences have now been noticed between the
stories about Hermotimus and those about Aristeas that I think
we are entitled to postulate one more to account for Clement's
statement: the discarnate soul of Aristeas too had prophetic in-

I
sight. This mantic capacity of Aristeas throws light on a rather
odd tale in Athenaeus, apparently from Theopompus (though
Jacoby has his doubts about this (ii. 389)) [T. r8]:
Philomelus gave the golden laurel wreath, which the Lampsacenes
had dedicated [at Delphi], to a Thessalian dancing-girl called Phar-
salia. As soon as this Pharsalia entered the market-place at Metapon-
tum a voice issued from the bronze laurel which the Metapontines set
up at the time of the visitation (lm817.Lav) of Aris teas the Proconnesian,
when he said he had come from the Hyperboreans, and the seers in
the market-place went mad and tore her to pieces. When they later
inquired the reason for this [from Delphi, presumably] they found that
!j the cause of her destruction had been the wreath belonging to the god.

The suggestion may be that the prophet Aristeas had the power
to impart the man tic gift to a clientele of hangers-on around the
monument of his visit to Metapontum, and on occasion to possess
them with a more than mantic frenzy, as he himself had once
been possessed by Apollo. On the other hand, the voice that issued
from the laurel may have been the voice of the god himself, just
as the Homeric Hymn to Apollo says that he gave oracles from
a laurel at Delphi (iii. 396; cf. Parke i. 3).
This is perhaps the place to record one further mention of
Aristeas, though its relation to the traditions so far given is not
readily apparent. It is in Gregory of Nazianzus [T. 23]:
Let us leave this childishness to such Greek worthies as Empedocles,
Aristaeus, Empedotimus, Trophonius, and the wretched mob of like
1q THE POET
kidney. The first won divinity for himself, or so he thought, by jumping
down the Sicilian crater, but was given away by his own sandal,
which the fire cast out ... while the others, through this same disease
of self-importance, hid themselves away in cells (aMroi~ -riulv) but
were found out, and earned more contempt from their exposure than
they had honour from their deceit.
The views of modern scholars about Aristeas may be divided
into three groups. First, there are those who hold that he was a
mythical personage (and therefore could not be the author of the
Arimaspea) ; second, those who hold that he was historical, but not
the author of the Arimaspea; third, those who hold that he was the
author of the Arimaspea (and therefore historical). Of the first
group, Crusius is perhaps the most outspoken (Roscher i. 2814) :
the poetic fantasy called Arimaspea originated among the Ionian
colonies of the Sea of Marmara and came to be ascribed to the
miracle-monger (Wundermann) Aristeas of Proconnesus, who has
been recognized as a mythical figure of the cult of Apollo, like
Olen and Abaris. It would seem that Bethe thinks similarly (876) :
Aristeas was a Wundermann like Abaris and Epimenides, given a
local habitation in Proconnesus and Metapontum; the probable
date of the Arimaspea, wherein Aristeas appears to have been
named as author, is the sixth century, an era to which is well
suited that combination of fantasy and religion which he, like
Abaris, Salmoxis, and Hermotimus, exemplifies. To Diels Aristeas
was a fabulous traveller so well known in antiquity that it was
as natural to ascribe a fantastic travel-poem to him as it was to
ascribe cosmogonies to Orpheus or cookery-books to Apicius (21).
For Rhys Carpenter he is but another type of the 'bear-hero',
identical with Aristaeus and Salmoxis (162 f.). I am not sure
how much historicity Jacoby would grant to the Wundermann
whose works he rejects as spurious (i. 519) ;4 or whether Guthrie
regards him as an entirely legendary figure (he is one of the
'strange servants' of Apollo, 'figures of legend'; but elsewhere
'other figures also' besides Aristeas (Abaris, Hermotimus,
Epimenides) 'emerge from the twilight between myth and his-
tory' ( I 93; r 95)). I hope I do them no great injustice by including
them in my first group.
 THE POET 125

Representatives of the second group are Rohde, who, though

seeing no reason to doubt the historicity of Aristeas, yet is non-
committal about his authorship of the Arimaspea (i 329 f.; ii 186);
and Hennig (7 I), who sees a gap of at least two centuries between
Aristeas and the author of the Arimaspea-if indeed the former
lived at all, for Hennig circumspectly will not go beyond admit-
ting the possibility of this (he follows Bethe in dating the poem in
the sixth century, and the 'better tradition' in reading 7p,71Koal-
owi in Hdt. iv. 15. 1).
The third group follows ancient tradition in allowing not only
that Aristeas existed but even that he composed the poem with
which he was credited. This view was expressed a century ago by
Tournier (4 ff.), but went out of fashion with a later and more
sophisticated generation of scholars. Today Tournier does not
lack supporters; but whereas he also thought that Aristeas was
a genuine explorer, there is a disposition among these modern
scholars to doubt whether he actually made the journey which
he professed to have made in his poem.
Years ago Rohde, contrasting the account in Herodotus of
Aristeas' death and disappearance from Proconnesus with the
other traditions given above, concluded that Herodotus had con-
flated two incompatible versions: according to one, Aristeas did
not die but was translated, body and soul: according to the other,
Aristeas 'died' not once but often, when his soul left his body
temporarily to wander abroad. This latter, preserved in the non-
Herodotean sources, is yet really the more primitive and original
version (i 328 f.). In 1935 this argument was carried a stage
farther by K. Meuli: drawing attention to the many similarities
between the stories about Aristeas and modern knowledge of
Siberian shamanism-which, he infers from Hdt. iv. 73 ff., was
not unknown to the ancient Scythians, whose regions were the
scene of Aristeas' travels-he suggested that Aristeas was a sort
of Greek shaman, and the Arimaspea a 'shamanistic' poem.
Meuli's theory, cogently argued and supported with a wealth
of learning, has been enthusiastically received and may be con-
sidered orthodox today. Among English scholars it has been
accepted by, for instance, Cornford (eh. 6), Dodds (eh. 5), Phillips
 THE POET
(176), and Bowra (2). The similarities between Aristeas and the
shamans are certainly impressive. The soul of the shaman leaves
its body behind and makes long ecstatic journeys to places both
real and imaginary (cf. Phillips 176) : so did the soul of Aristeas.
During these journeys it meets with fantastic adventures: so did
Aristeas, and related them in the Arimaspea as the shaman re-
counts his adventures to his audience. The shaman can appear
in two places at once, as did Aristeas; or his soul issues from him
as a bird, as Piiny tells us the soul of Aristeas issued from his
mouth in the shape of a raven (old Siberian myth even called
the prototype of shamans 'the Great Raven'-Meuli 158). The
shaman is an adept in divination: so was Aristeas. When we
consider further that the scene of Aristeas' travels was the very
country of these medicine-men, and that he himself stated that he
made his journeys 'possessed by Apollo', that is, in some un-
usual state, the case seems pretty conclusively proved. There are,
however, some awkward difficulties, shortly to be discussed.
I agree with Rohde that there are inadequate grounds for
doubting that Aristeas was an historical character. Though there
is some evidence of confusion between him and Aristaeus the son
of Apollo and Cyrenc, he does not show signs of being a 'faded
god' or hero. Antiquity provided him with a birth-place that
was only founded in comparatively recent times, with a father
not otherwise known but with a convincingly human name [ 131],
and even with a date-or rather, dates, for our two authorities
on this subject disagree.
According to 'Suidas' [T. Il], Aristeas was a contemporary of
Croesus and Cyrus, flourishing in the fiftieth Olympiad (580/77):
'
')'E'')'OVE' 'IT ~
Ka-ra' n.po,aov Ka,' nvpov,
'IT- '"' v' [71 ' F ; oyoor,
'0'11v1r,aa, ' ~' V
(i.e. 748/5, surely from a confusion of N and H); vr]' Rohde
(548/5), accepted by Meuli]. I take it that the combination of the
more precise dating of the Olympiad with the vaguer 'in the time
of Croesus and Cyrus' means that the former was Aristeas'
ftoruit. 5 If this dating is right, Aristeas could not have been
the author of the seventh-century Arimaspea; but reasons will
be adduced to show that it may be part of an unreliable tradi-
tion [ 172].
 THE POET 127
Herodotus, as we have seen, says that Aristeas' appearance at
Metapontum occurred two hundred and forty, or according to
another reading three hundred and forty, years after his second
disappearance from Proconnesus. The higher number, in spite
of its respectable manuscript support, cannot be fitted into any
scheme; it will not fit inside even the most extreme termini con-
ceivably possible, the foundation-date of Proconnesus (at earliest
c. 685, cf. Burn 132) and the year of Herodotus' death (at latest
42 5, cf. P.-W. suppl. 2, 232). The lesser figure, which is supported
by Origen (Gels. iii. 26) and Tzetzes [T. 13]-to which must be
added the interval between Aristeas' second and his first dis-
appearance from Proconnesus: not less than seven years-can
just be squeezed between these termini, but with uncomfor-
tably little elbow-room; and it would make the appearance of
the ghost of Aristeas to the Metapontines coincident with the
last part of Herodotus' own life--indeed, when he was living next
door, at Thurii, in which case it is almost incredible that he
should not have remarked how recent it was. Anyway the
whole story smacks of an earlier age. Rohde concludes th2.t
Herodotus has made an error of calculation.6
Some account of a mysterious appearance, or appearances, of
Aristeas may have been known to Pindar. This is an inference
from a remark ofOrigen; and even ifit is justified we still cannot
decide (as yet-papyri may give us the answer) whether this was,
or included, the Metapontine appearance; if it did, of course
even Herodotus' lower figure would be quite impossible. Origen
quotes Celsus [T. 1 7] : 'Then take Aristeas of Proconnesus, who
disappeared so strangely and again reappeared to the sight of
men, and long afterwards was a visitant in many places bringing
marvellous reports: Apollo ordered the Metapontines to regard
him as divine, but no one now worships Aristcas.' He then adds
that Celsus 'probably got the story from Pindar and Herodotus'.
It must be borne in mind that, though Origen thought that
Celsus was using Herodotus and Pindar here, this does not mean
that he was necessarily right; but it must mean that something
in Pindar, as well as in Herodotus, afforded reasonable ground
for his deduction. Knowledge of Herodotus' story alone could
 nB THE POET
account for Celsus' first words, 'who disappeared so strangely and
again reappeared' , a'f'avu7
>,I. 8VTa -re OV'TWS"
I a <'
oai,ovs
I I; > (}
Es av pw7Twv
I

[the death and disappearance in the fuller's shop] ,ea! av8,s

J.vapyws cf,avEVTa [ the reappearance in Proconnesus six years laterJ;
'ITOAAoi's varepo" xp611ois, 'long afterwards', too fits with Herodotus'
long period between the second disappearance and the Meta-
pontine reappearance. But with 1T0Maxov rijs ol,covEvrJS, 'in many
places',7 we meet something non-Herodotean. We have indeed
heard of a number of other appearances of Aristeas in Sicily
[120]; and Plutarch tells how at the moment of his death in the
fuller's shop he was met upon the road to Groton in south Italy
[T. 16]. But the latter certainly, and the former possibly, are con-
temporaneous with a real or putative death of the man, not 'long
afterwards'. Also, the evidence of Plutarch here must be treated
with great reserve, for it looks too like a distortion of Herodotus'
account [cf. 201]. Reasons will be given too to suspect that the
Sicilian appearances are a late invention [ r 69].
The next words, 8avaa'TtL ayye{AaJJ-ra, 'bringing marvellous
reports', could be Herodotean, and refer to Aristeas' statement
that Apo1lo had visited the Metapontines and that he himself
had accompanied the god in the shape of a raven; but, when
one recalls that Pliny uses adnuntiare (i.e. ciyylX\ew) of the mar-
vellous clairvoyant reports that Hermotimus gave [B!I), one
wonders whether Celsus did not know, and here refer to, similar
reports given by Aristeas.
Th e C1ause 'TOU n.'ITOI\J\WJJOS
M I\\ ' ''
'1TtGIC7/'f'aV'TOS
'TOtS ~,,. I
1V1E'Ta1TOV'TWOLS '
(II
81:wJJ o{pq. vleiv TOv J1.piaTlav, 'Apollo ordered the Metapontines
to regard Aristeas as divine', may be non-Herodotean, or it may
be an inference from Herodotus' words. According to these
Aristeas told the Metapontines to set up an altar to Apollo and a
statue to himself; and the Delphic oracle confirmed this behest.
There is no explicit statement here that a cult of Aristeas was thus
to be established, though the passage might so be taken. He is
indeed said to have been given heroic honours by the Sicilians
[ I 20], and divine or semi-divine rank is indicated by Theopompus'
calling his appearance at Metapontum an Jm87Jta. [ 1 23; cf. Celsus'
J.m87/1aaVTa here], a word particularly used of divine visitations;
 THE POET 129
but in both these cases the evidence is suspect, showing signs of
fourth-century working-over and touching-up (if not making-
up).
If Aristeas was an historical personage such heroization, espe-
cially as early as the fifth century, would present peculiarities.
He might have qualified as the minister of Apollo or the apostle
ofhis cult (cf. Farnell ii 53 ff.) ; but canonization usually centred on,
and depended upon the existence of, the supposed grave or relics
of the hero. In a parallel case to that of Aristeas, that of the
athlete Cleomedes, who vanished mysteriously in 492 B.c., his
cult was established at his home, Astypalaea (Plut. Rom. 28;
Paus. vi. g. 6 ff.), so that Proconnesus rather than Metapontum
would have been the natural place for a cult of Aristeas.8 How-
ever, I suspect that divine honours were never paid to Aristeas
in fact, but were a fiction of some fourth-century author with
Pythagorean leanings. Pindar (fr. 127) and Empedocles (fr. 146)
are evidence that for early Pythagoreanism the highest incarna-
tion was to be a man of outstanding strength or wisdom, a king,
or a seer. The next stage was escape from the flesh and graduation
to divinity, as 7Jpwi; (Pindar) or (Je6i; (Empedocles) (c the
'Orphic' tablet of fourth/third century B.c. from Petelia, Diels-
Kranz i. 15. 31 ). Such cults of hero-philosophers are indicated in
the fourth century for Pythagoras and Anaxagoras (Arist. Rhet.
ii. 23) ;9 the idea is persistently expressed by Plato (e.g. Phaedr.
248 f.; c Boyance 268 ff.), and it was, I suggest, some pupil of
Plato (aided no doubt by the Herodotean account) who imented
the story of the bestowal of divine honours upon Aristeas by the
Metapontines at the bidding of Apollo (that is, of course, Delphi,
the normal dispenser of heroic status); honours which Aristeas
now merited as the philosopher [T. 20] and seer [T. 21] he had
become. This argument would carry with it the story of the
heroization by the Siceliotes too: there I think the important
point is not so much that Aristeas appeared supernaturally but
that he appeared as a teacher-though it is certainly strange that
he should be described as teaching /ettas (ypa.aTa S,3aaKoVTa,
usually of a schoolmaster! 10). The use of the words 80<; by
Celsus and 8vLV (for lvayl,f:Lv, cf. Hdt. ii. 44. 5) by Theopompus
K
 130 THE POET
[ I 20] with reference to the man Aristeas also seems to me to indicate
a later rather than an earlier date: at least post-Empedoclean,
and so post-Pindaric (quite possibly fourth-century: after Em-
pedocles' translation in the story by Heraclides Ponticus, Pausanias
tells his friends that they must sacrifice to the vanished man as
to one who has become a god, OvEw avrcp 8,v KaOam:pEi ')IE')lov6n
8Ecj, (D.L. viii. 68) ).
If this is so there could have been nothing in Pindar about
Aristeas' divinity, nor could Celsus in fact have had Pindar and
Herodotus as his sources here. The only logical basis for Origen's
inference that Pindar as well as Herodotus underlay Celsus'
words must have been the phrase 1To.V.axoii rijs olKovJV7Js E1TL
871~aavra, for which alone Herodotus provides no warrant:
Origen may have known that Pindar told of a number of ghostly
appearances of Aristeas similar to, and perhaps including, that
at Metapontum. This is the most that can be said, and even so
it may be pressing logic too far. It is noteworthy that Origen, in
support of his statement that Celsus probably got the story from
Pindar and Herodotus, proceeds to quote Hdt. iv. 14 f. verbatim,
but of Pindar not one word. It is not impossible that Origen was
really thinking of (or Celsus himself appealing to?} Pythian ix.
63 ff., about the bestowal of divinity on the child of Apollo and
Cyrene, 'who will be called Zeus and holy Apollo, . . and by
others Aristaeus', and that there was in fact nowhere in Pindar
any mention of Aristeas of Proconnesus. (This would not be the
only case of confusion between these two.)
An attempt has been made to save Herodotus' arithmetic at the
expense of his text. E. Schwyzer proposed to read in iv. 15. 1:
TO.BE 8t ol8a M<iTa?Twrlvoia, Tofo, Jv 'fra'At-n auyKvp~aas [for auyKu-
, ] ~ ,.4
P'YJUllVTll fl-ETa\ T'YJV
\ 'J..I
a.,..av,aw \
T'YJV I I t1 I
oEUTEp7JV nptCTTEW ET'1C1t T'1C1C1'1pa-
~
KOVTa Km' ot7JKOC1,o,u,, I <
WS > \
E')IW au f]-atVIOtvo,
\\ I >n
EV p0KOVV7JC1([)
I
TE

Kat METa?Tovrf.q, dJp,aKov, 'what follows I know because I fell in

with the Metapontines two hundred and forty years after the
second disappearance of Aristeas . .'. The effect of this then is
to date the Metapontine visit of Herodotus himself, not of
Aristeas, two hundred and forty years after the second dis-
appearance of the latter; no date is then given for the ghost's
 THE POET
appearance at Metapontum, which can now be moved back into
the decently distant past; while, assuming what is probably true,
that Herodotus was in Metapontum some time in the thirties,
Aristeas can be one of the first colonists of Proconnesus two
hundred and forty-seven years before. Also, one can see how
Herodotus might have calculated the time of his own presence
in Metapontum relative to Aristeas' disappearance: by genera-
tions, which he might have heard recounted in Proconnesus,
plus the time between his own visits to Proconnesus and to Meta-
pontum (this would work out at seven generations, or two hundred
and thirty-three years on his reckoning (ii. 142), plus seven years).
But the emendation fails to convince me on two grounds, lin-
guistic and artistic. First, MTa1ToVTlvotat O'V)'KVP17aa, should mean
'having come upon the Metapontines accidentally', whereas we
may be certain that Herodotus did not find himself in Meta-
pontum quite by chance. Secondly, there is a point in giving the
time between Aristeas' disappearance and reappearance, but
none whatever in giving that between Aristeas' disappearance
from Proconnesus and Herodotus' own visit to Metapontum; it
would be a most curious era to choose for his dating. 11
It seems, then, that attempts to defend even the lower figure
in the Herodotean manuscripts as historically accurate must be
abandoned. The most that we can deduce is that for Herodotus
Aristeas was a very remote figure, who would suit well enough
as one of the earliest colonists of Proconnesus.
The name of Aristeas' father, Caystrobius, appears to be of the
same class as Cephisodotus, Cephisodorus, &c. (cf. Nilsson ii ro):
the parents of Caystrobius in their desire for a child had prayed
for fertility to the river-god Cayster, and in gratitude for his
answer to their prayer had called their son 'he who owes his life
to the Cayster' (the form of the name is e..xactly paralleled by
Hermobius, who came from Temnus on the Hermus (Cic. Fl. 43)).
A river-god is naturally a local god, and so this Caystrobius must
have been born, or at least conceived, in lonia. If the father of
Aristeas was the first of his family to bear this name, it is possible
to reason thus: when Herodotus says that Aristeas 'belonged
to one of the noblest families in the island' (he was Twv aaTwv
 THE POET
ov8tvoS' ylvoS' {mo8tlU'TtpoS') he must mean that his family was one
of the original founder-families of the colony: that either Caystro-
bius or his father, or (less likely perhaps in view of Herodotus'
ylvoS') Aristeas himself had shared in the original foundation.
Even if this is right, all the suggested dates for Aristeas could still
be accommodated-even Rohde's emendation of the Olympiad
in 'Suidas'; but this last would now be rendered extremely un-
likely, for it would necessitate a series oflast-minute occurrences:
the latest possible foundation-date for Proconnesus ( c. 650:
Strabo 590 and 587); Caystrobius born the same year or a year
or so before; Aristeas born when his father was verging on
senility.12 The probabilities would point to a birth-date for
Aristeas between 680 and 620; in other words, we should arrive
at a likely date for Aristeas which matches remarkably the date
we found, by quite independent arguments, to be the likely one
for the poem which was ascribed to him. I have no qualms about
accepting this ascription.
A seventh-century date for the Arimaspea and its author poses
a problem for the 'shamanistic' theory. We have seen that the
poem contained matter the accuracy of which could only be
owed to real knowledge of central Asia. How did Aristeas come
to be so well informed? From some genuine Greek traveller or
Scythian trader? But the foundation of the earliest Greek entrepots
on the north side of the Black Sea does not appear to be much
before 600 B.c. (Olbia founded c. 610: Cook 71 f.; 76 ff.; 82);
and though there is some support for the view of Burn (135)
that there had been two or three generations of exploration and
commerce before this (Roebuck I 18 f.), Greek trade with the
deep Scythian hinterland could not yet have been established.
And it is uneconomical to postulate an 'Ur-Aristeas'-an un-
known Greek traveller to do in fact what Aristeas himself claimed
to have done-without sufficient cause. But if Aristeas made the
journey in the flesh and himself collected the information an
extraordinary state of affairs follows : we have to believe that he
returned to his home-town, where his absence of years would be
well known, went into a shamanistic trance, and then pretended
that during this he had been journeying in the spirit: surely an
 THE POET 133
unnecessarily elaborate as well as patently obviously deceit. No
wonder that Meuli was ready to accept Rohde's emendation of
'Suidas'' Olympiad as near the truth, and so bring Aristeas down
to a time when the flow of Scythian trade to the Black Sea
colonies could be supposed to have carried with it knowledge of
the far hinterland of Asia!
I hope to have shown that objections to a sixth-century date are
manifold and grave. Even if I have not succeeded, obstacles
remain to an easy acquiescence in the 'shamanistic' theory. Hero-
dotus states that he heard the story ofAristeas' first disappearance
and reappearance from the Proconnesians and Cy;:icenes. It is
a reasonable conjecture that the Cyzicene contribution was the
man of Cyzicus who met the 'dead' Aristeas on the mainland,
and that Herodotus himself has combined this with a Procon-
nesian account--of the corpse which vanished from the fuller's
shop--in a manner dramatically effective though chrono-
logically absurd, for logic demands that the Cyzicene had left
the mainland before the death, ifhe was to arrive in Proconnesus,
a good four hours' voyage from Artace, just after it had been
announced. This is all we need suppose, and there is no cause to
follow Meuli in seeing in this episode proof of a much more
deep-seated tampering with, and rationalizing of, an original
trance-journey of Aristeas the shaman. 1 J Also, if Aristeas was
lying all the time in Proconnesus 'shamanizing', his soul would
be flying through the air, perhaps even in the form of a bird, and
it is difficult to see how the Cw:.icenes
., would come to have an
interest in the story at all.
An even more awkward absurdity is involved in regarding the
Arimaspea as a shamanistic poem. We are told emphatically and
unambiguously by Herodotus that in it Aristeas claimed only to
have got as far as the Issedonians, and admitted that what he had
to tell of the regions beyond was hearsay. But whoever heard of a
shaman's soul 'breaking down' in mid-flight and confessing that
henceforth it had to depend on the reports of the people among
whom it had made its forced landing? I find this so ridiculous
(the more so as it would be quite unnecessary) as to be incredible.
Once again we are confronted with that old dichotomy in the
134 THE POET
tradition: according to one form ofit Aristeas reached the Hyper-
boreans, according to the other he reached only the Issedonians.
In the former he travels in the soul (Maximus [122]); but the
latter derives from the Arimaspea itself. Attempts to paper over
this dichotomy have so far not been satisfactory. Dodds (141)
says 'whether Aristeas' journey was made in the flesh or in the
spirit is not altogether clear' ; Phillips (176) suggests that he went
part of the way in the flesh and the rest in the spirit; Bowra
(2), that he claimed to have travelled both in the flesh and in
the spirit, and that Herodotus chose to emphasize one claim
and Maximus the other.
Yet it will be remembered that Aristeas in his poem said that
he had made his journey 'possessed by Apollo', rendered by
Herodotus as ef,o,f36>.a.1TTos ya,o.evos. What are we to make of
this? Does it offer support to the view that Aristeas was a shaman?
It offers, I think, support to this extent, that it could mean 'filled
with prophetic insight' ; and the mantic knowledge which Apollo
bestows upon his priest may be of contemporaneous events, and
is not necessarily confined to the future (cf. Hdt. i. 47). But such
a meaning seems barely possible in this context: the seer shares
the god's knowledge in the moment of possession, or the god
imparts it through the mouth of the seer, but there is no question
of the seer's soul going forth and as it were fetching the knowledge
back with it. Herodotus' language then, 'Aristeas said that,
possessed by Apollo, he reached the Issedonians', is either very
inexact or puts this interpretation out of court.
Moreover, I have yet to find characterized as ef,o,/30>.711TTos one
whose soul wanders while his body lies in a trance. The word is
never applied to the doyen of such, Hermotimus (for whose status
as an Apolline figure anyway the evidence is tenuous in the ex-
treme14) ; and the fairly numerous others of whom similar stories
came to be told [148 ff.] are never described as 'possessed' by
any god when they undergo this separation of soul and body. In
fact, such uses as I have been able to find of this adjective,
and of related adjectives and verbs, have with one exception
implied inspiration or frenzy, and the activi[v of speech at least.
In Lycophron's Alexandra Cassandra calls herself 17 ef,o,/Jo'A71TM"os
 THE POET 135
XEAwc6v (1460); and the slave, reporting her prophecies at the
beginning of the poem, says (3)
OV ')lttp 'IJCTVXOS K6p'TJ
J,\v<TE XPTJ<Taw, WS' 71'p{v, al6Aov cn-6a,
..!,\,,,a(T71'ETOII XEaO'U
WV\
' 7tap./U'Y'TJ""''1'
/JOTJV
8a.t/,V7Jif,aywv if,otPa(EII EiC ,\111.wv thra

'for not quietly as of old did the maiden loose the varied voice
of her oracles, but poured forth a weird confused cry, and uttered
wild words from her bay-chewing mouth' (trans. A. W. Mair).
Cassandra's psychic disturbance, and the use of the verb cf,o,{Ja-
'Ew, may here have been suggested by the literary tradition of the
frenzied inspiration of the Pythia; but that such manifestations
were expected of a prophet by the man-in-the-street too is in-
dicated by the counterfeit seer in Diodorus Siculus (xxxiv and
xxxv. 2. 6), who, to give authority to his utterances, even went so
far as to breathe fire in accompaniment with them, and, raving
so, would foretell the future (S,&. nvoS' :11xavqs; 71fJp f.1.ETa. -rwos-
'8OIJO'tauov- KatI y,11oya
Ell .1.\ I
ota
<:, \ '.I.'
TOV- OTO.aTo, 'T}y,tH,
I
Ka.t OVTW
~
Ta
{AAoV'Ta a1ri;q,olf3atEv15).
In these places derivatives of Phoehus have been used to denote
the physical accompaniments of divine inspiration; but they are
also used to suggest divine inspiration for the form or content of a
man's utterances. So, after the high-handedness of Pompey and
Caesar and the passage of the latter's agrarian law in 59 B.c.,
Cato, according to Plutarch, foretold in the Senate the mis-
fortunes of Rome cZa1rEp brt1Tvov, ,m1 cf,01{36:\ry1TTos; (Pomp. 48).
Here no doubt it was primarily the content of Cato's speech which
earned the epithets-his prophecies seemed god-given because
they turned out true. But implied too may be the loftiness of the
form in which they were expressed, as in three passages of the de
sublimitale: Demosthenes enunciated his famous 'Marathon oath'
as if suddenly inspired, Ka.8a.1rEp e.71'VE110-8ElS' Jfalrf,VT]s- v1ro 8EofJ
Kal ofovlil ef,o,fJ6A1J1TTO~ yEVoi;vaS' ( r6, 2)-surely a case in point
that there is nothing so conducive to lofty style as heartfelt passion,
'which breathes forth our words as if in the frenzy of divine
inspiration, and gives them a sort of sanctity', wcnrEp v1ro avlaf:
 THE POET
\
'TWOS Kat\ 7TVE1!J,ttlTOS
I '
EV 8 . . . EK7TVEOV
OVO'Ul.<TTtKWS ' I ~
Kat owvei .,,o,
.,J.. {Ja,.ov
' 'T '

-rovs ,\6yovS' (8. 4) ; if they model themselves on such as this even

those who are little prone to inspiration may draw upon and
share something of the divine spirit of their great exemplars:
OVTWS' dm~ ri/s 'TWV apxalwv J.tE'/aAocf,vtaS' elS' -rcls- 'TWV {17,\oJVTwv
9 f ,/, \ r , \ t .-. I , I I .J.. I
EKEtVOVS' 'l'vxaS' WS' a7TO LEpwv a'TO/J,tWV a1roppotat TtvES 't'EPOVTat,
f,,/..' 'f'
v't' , I
WV E7Tt7TVEO.EVOt Kai\ oir /L?]\ Atav
\ I ,J.. Q
't'o,,..aa'TtKOt\ ,... t I 8
Tep ETEpwv O'VVEV 01)
EyE8Ei ( 13, 2).
CJtwCJi

On the other hand, yet another element of divine inspiration,

fluency of composition, is emphasized by Strabo (675) when he
relates how Diogenes of Tarsus, being set a subject, could throw
off a poem on it 'as if inspired' (Kai 1roi~.aTa wa1rep d1recf,0{{3a~e
Te8EU17JS' {mo8laews).
In all these examples of Apolline possession, literal or meta-
phorical, the simultaneous activity of speech at least is implied,
as I have said. The subject has the god within him, who does the
talking: he is ev8eos Kai eKcf,pwv, the god has displaced his own
mind; meaning, not that his soul or mind is away from his body
wandering ecstatically, but that it is for the moment ofno account,
thrust aside by another power which has taken control of the man.
The one exception that I mentioned above, where not activity
but serene contemplation is deemed to accompany the fact of
possession, indeed brings out the meaning of cf,o,{J6,\171rToS' as
lv8Eos-he whom Apollo has seized has the god within him:
Plotinus tells us that we must view Beauty not as something
external but as something within ourselves, 'as if one possessed
by Apollo or some Muse were to contemplate the deity within
himself, were he capable ofsuch a thing' (Enn. v. 8. 10: w0'7Tep E'tns
, , - , .1. a f\ .. , , "" , , , _ ..
1)7T0 8EOV Ka'Taaxe 8ELS' 'l'oL,-.ol\?71T'TOS' '1] V7TO 'TWOS' lYlOVCJ7]S' EV av'Tcp av
- 'TOI)- 8<:OV- 'T'l]V (J eav, Et ovva.,v
7TOtOL'TO
1 1
'I> I " '=XOL Ell ' - 8<:ov {J\
> aVT<p
1 1
I\E?TEtv
)

Seizure by other deities too brings an unaccustomed activity

of some sort to the possessed. 'Does it strike you as it strikes me,
my friend,' asks Socrates of Phaedrus in that sequestered beauty-
spot by the Ilissus, 'that a god has done something to me?' 'You
certainly have an unwanted fluency, Socrates.' 'Then hear me in
silence. We really are on holy ground, so don't be surprised if
during my discourse I am perhaps seized by the nymphs (vvtf,6-
 THE POET 137
~1111-ros); for already I am almost talking in dithyrambs' (Phaedr.
2 g8c).
Socrates expects seizure by the nymphs because they are
appropriate to the setting; though the consequences he foresees
are more appropriate to seizure by Apollo. Many Greeks, ancient
or modern, would not share Socrates' serene humour in such
a situation, for seizure by the nymphs is no laughing matter (cf.
Lawson 142 ff.).
Hermias, in commenting on this passage of the Phaedrus, says
elal 8t Kal 7TavoATJ7TTOt Kai fJ,TJTpoft.'f}?TTOt Kat Kopvf3avna,ot ( xo5).
Now we have an account in Plutarch, derived from Posidonius,
of a spurious f.L'fJTp6>.TJ1TTos, who pretended to be seized by the
Sicilian goddesses called 'the Mothers', which is worth quoting
because it shows us what symptoms were expected to accompany
such a seizure (Marc. 20; the translation is Langhorne's in the
main):

There is in Sicily a town called Enguium, not large, indeed, but very
ancient, and celebrated for the appearance of the goddesses called the
Mothers This town was strongly inclined to favour the Cartha-
ginians; but Nicias, one of its principal inhabitants, endeavoured to
persuade them to go over to the Romans, declaring his sentiments freely
in their public assemblies, and proving that his opposers consulted
not their true interests. These men, fearing his authority and the in-
fluence of his character, resolved to carry him off and put him in the
hands of the Carthaginians. Nicias, apprised ofit, took measures for his
security, without seeming to do so. He publicly gave out unbecom-
ing speeches against the Mothers, as if he disbelieved and made light of
the received opinion concerning the presence of these goddesses there.
Meantime, his enemies rejoiced that he himself furnished them with
sufficient reasons for the worst they could do to him. On the day which
they had fixed for seizing him, there happened to be an assembly of
the people, and Nicias was in the midst of them, treating about some
public business. But on a sudden he threw himself upon the ground,
in the midst of his discourses, and after having lain there some time
without speaking, as if he had been in a trance (def,rj,m ils TIJV yijv
\ \ \ ~ ~- \ r 'f' t I f' I \ t \ It
TO UWJJ,a, Ka, JJ,LKpOV o,aA,1TWV, OtOV 'KOS, 'l)UVXLaS OVV 1ffl"M),;;L 1'0'0
lV1Js KTA.), he lifted up his head, and turning it round, began to speak
with a feeble trembling voice, which he raised by degrees: and when
he saw the whole assembly struck dumb with horror, he threw off his
mantle, tore his vest in pieces, and ran half:naked to one of the doors
138 THE POET
of the theatre, crying out that he was driven by the Mothers. From
a scruple of religion no one durst touch or stop him ; all, therefore,
making way, he ran towards the city gates, omitting neither sound nor
gesture befitting one that was heaven-struck and distracted (8a,.011w11T1
Kett 1rapa<ppovofiv,n). His wife, who was in the secret, and assisted in the
stratagem, took her children, and went and prostrated herself as a
supplicant before the altar of the goddesses. Then pretending that she
was going to seek her husband, who was wandering about in the fields,
she met with no opposition, but got safe out of the town; and so both
of them escaped to Marcellus at Syracuse.
Here we notice the sequence of events: the victim is behaving
normally when he suddenly falls to the ground, where he lies for a
short time bereft of his faculties; then springs into wild activity
and is compelled to roam about. Nicias' fraud was clearly well
conceived and executed (cf. Plutarch's ofov ElK&s). A possible
parallel can be discerned here with Herodotus' account of how
Aristeas walked, apparently normal, into the fuller's shop, fell,
apparently suddenly, into a state that simulated death, and later
was gone from the shop; there is no parallel at all with the version
of Maximus and others of the cataleptic Aristeas whose soul
wandered. Such a state, as I have said, is not described by the
ancients as seizure or possession.
This brings me to an illustration given by Rohde (i 328, on
the ekstasis of Aristeas). Rohde believed that the version of Maxi-
mus was in origin earlier and truer than that of Herodotus ; and
that, when in the Arimaspea Aristeas said that he reached the
Issedonians <f,o,f3&.\a'TT'ro, y.v61uvos, this at least meant 'in some
strange way impossible for other men, i.e. in Apolline ecstasy
(cf... EV EK<1'TClUE a1ro<f,o,fJcI,evos, P. Mag. Par., p. 63 Wess.)'.
Now the words in question may not refer to the way but to the
state in which the poet travelled; and for the assertion that it was
impossible for other men there is not a shred ofjustification. The
parallel for 'Apolline ecstasy' (in Rohde's sense of cataleptic
trance) from the magical papyrus looks at first impressive in spite
of its lateness; but it looks less so when quoted in full: eav8t Ka1
I 1" \' .- '~ \ ""'8 ""' t J \I t
ovos !}', Km E')'){f:LP!}', 'Ta V1TO 'TOV EOV EtpTJEva, I\E')'ELS ws El'
't

EK<rrtluL a1ro<poi{JcfJevos ( = iv. 736 of P. Graec. Mag. ed. Prei-

sendanz). This ecstasy too supposes the accompanying activity
 THE POET 139
of speech, and J.1rorpoif3dJEvo1: has no sense that we have not
already considered in its related derivatives: it refers here to the
'fine frenzy' displayed by those whom Phoebus inspires.
Rohde's assumption that ekstasis originally involved the idea
of the soul's departure from the body has been challenged (cf.
Dodds 94 f.) : the word is used of mental disturbance of varying
degrees, but not in passages where one would expect it if Rohde
were right. In the question under discussion we must be more
exact: we must distinguish between three conditions: that of
possession, in which the subject is unwontedly active in some
respect or respects; that of hypnotic or mediumistic trance, in
which the subject is quiescent but palpably alive; and that of
catalepsy, where the subject is so bereft of all faculties, even to
the cessation of breathing and pulse, that he appears to be dead-
though after revival he tells of perceptions made, and knowledge
gained, 'in the spirit',
All three of these are authenticated psychic phenomena; the
last-named perhaps rarer than the others. Tyrrell gives some
interesting modern examples ofit ( 149 ff.), of which I quote one:
Dr. Wiltse, of Skiddy, Kansas .. , lay ill with typhoid and sub-
normal temperature and pulse, felt himself to be dying, and said
goodbye to his family and friends. He managed to straighten his legs
and arrange his arms over his breast, and sank into utter uncon-
sciousness. Dr. S. H. Raynes, the only physician present, said that he
passed four hours without pulse or perceptible heart-beat. He was
thought to be dead and the church bell was tolled. The doctor, how-
ever, thought he could perceive occasionally a very slight gasp; he
thrust a needle into the flesh but got no response. The percipient,
however, says that at last he again came into a state of conscious
existence and found he was still in the body, 'but the body and I no
longer had any interests in common'. He seemed to be getting out of
his body, rocking to and fro and breaking connection with the body's
tissues. He seemed to feel and hear the snapping of innumerable small
cords and then, he says, 'I began slowly to retreat from the feet to-
ward the head, as a rubber cord shortens.' Presently he felt he was in
the head and then emerging through the sutures of the skull. 'I re-
collect distinctly', he continues, 'how I appeared to myself something
like a jellyfish as regards colour and form. As I emerged I saw two
ladies sitting at my head. I measured the distance between the head
140 THE POET
of my cot and the knees of the lady opposite the head and concluded
there was room for me to stand, but felt considerable embarrassment
as I reflected that I was about to emerge naked before her.... As I
emerged from the head I floated up and down and laterally like a soap-
bubble attached to the bowl of a pipe until I at last broke loose from
the body and fell lightly to the floor, where I slowly rose and expanded
into the full stature of a man. I seemed to be translucent, of a bluish
cast and perfectly naked.' The percipient fled towards the door, but
when he got there, found himselfsuddenly clothed, and his elbow came
in contact with one of two men standing in the doorway (they were
actually there) and to his surprise went through him without en-
countering any resistance. He tried to attract the attention of his
friends but without success, so he walked out of the door and into the
street. He says, 'I never saw that street more distinctly than I saw it
then. I took note of the redness of the soil and of the washes the rain
had made.' (There had been heavy rain and the roads were marked
by rain-washes.) He then noticed that he was attached by means of
a small cord, like a spider's web, to his body in the house, the cord
coming from his shoulders. He then seemed to be propelled, as ifby a
pair of hands, and found himself on a roadway, while below him was a
scene of mountain and forest which looked very like the local scenery.
After various experiences he came to rocks blocking the road which
he tried to climb round. But at that moment a black cloud descended
on him; and he opened his eyes to find himself back on his sick-
bed.

Clearly the Aristeas of Maximus' tradition is a cataleptic,

accurately portrayed ; there is no confusion with the symptoms of
possession. Ancient authors are not always so careful to keep the
two things apart; a notable example of such 'contamination' is to
be found at the end of the first book of Lucan, where a frenzied
Roman matron, possessed by Apollo, speaks as if her soul has left
her body and is flying over the face of the earth, viewing the
lands beneath ; yet at the same time her body is tearing through
the streets of Rome (674 ff.).
I do not doubt that the events in the fuller's shop, as recounted
by Herodotus, preceding as they do Aristeas' departure on his
journey, constitute that 'seizure by Apollo' which the poet him-
self said he had undergone when he made his travels. This seizure
was possession, not catalepsy, and it perhaps ushered in a sense of
 THE POET

superhuman strength and confidence, like that imparted by

Apollo to his devotees at Hylae:
In the territory of Magnesia, on the river Lethaeus, there is a place
called Hylae, where is a grotto consecrated to Apollo. There is nothing
very wonderful in the size of the grotto, but the image of Apollo is
very old, and it imparts strength equal to any labour. Men sacred to
the god leap down precipices and high rocks, tear exceedingly lofty
trees from their roots, and walk with their burdens along the narrowest
paths [mountain-paths, I take it] (Paus. x. 32. 6, trans. Frazer). 16
There is some evidence that Apollo was a popular god in the
environs of Proconnesus. He was said to be much worshipped by
the natives in the neighbourhood of Cyzicus (Strabo 551).
Cyzicus itself was named after a son of Apollo, and the god was
somehow particularly closely connected with that city (Aelius Aris-
u.des xxvn.
5 : 'TaS'
' Jl-EV
' yap
\ W\l\aS'
_\ \ 1TOMtS'
I\ ~ ' A ,
oLa 'TWV OLKLGTWV CfKLGEV OVS'
A

, I . \. ~ I , t, \ >
07TEG'TEI.I\EV EKaa.axoae, 'TOIJ'T7/S' OE K 'TOV
..,. '8I
EV EOS
' \ I '
av'TOS' y1ayovev 01.J.:L

OTIJS). It seems to me not impossible that Aristeas himselfmay have

been a devotee or priest of Apollo, who, hearing perhaps of some
exploratory venture to the northern shores of the Black Sea (and
there may well have been such even as early as the mid-seventh
century [132)), felt possessed by his god and impelled to visit
Apollo's paradise among the H yperboreans, who could not be far
away if Boreas blew from Thrace. The act of divine possession,"
followed by an abrupt departure from Proconnesus, would not
need to be relayed through many mouths before it became a death
and strange disappearance. Given this, the story of the acquaintance
who met Aristeas near Cyzicus presents no difficulty, once we elimi-
nate the preternatural element ofspeed; and this I have done [ I 33]:
the acquaintance met Aristeas en route on the mainland (Cyzicus
may have been the starting place for the expedition), and some time
later heard the rumour that he was dead by that time. This made a
good story, and was preserved in Cyzicus even to Herodotus' day.
I have given my reasons for believing that the tradition about
Aristeas represented by Maximus is not, after all, the truer and
more primitive, as has been held by Rohde and others. It now
remains to examine this tradition more closely, to sec if we can
establish when and how it arose.
 VII

ARISTEAS PYTHAGORICUS

IN that unreliable catalogue of early Pythagoreans which Iam-

blichus gives at the end of his Vita Pythagorica (267) there occurs
the name of 'Aristeas the Metapontine'. Tournier suggests (4.5)
that this personage may be the Aristaeus, son of Damophon, who
is said to have married Pythagoras' widow Theano and succeeded
him as head of the school (Iambi. V.P. 265, where admittedly he
is called a Crotoniate). But the untrustworthiness of this evidence
is plain from its anachronistic talk of an early Pythagorean
'school' with a headship, which dates it at earliest to the fourth
century B.c.-probably indeed later, to the Hellenistic 'Peripa-
tetics' with their interest in ,f,,>..oa&,f,wv 8,a.ooxa.l; the mention of
Theano the wife of Pythagoras also betrays the influence of the
'Pythagorean romance' of later times.' We do not hear of this
Aristaeus until nearly a millennium has passed since his supposed
lifetime, and though a work 11ep, apovla.s was current under his
name (an extract, duly written in pseudo-Doric, is given by
Stobaeus (i. 176), and Iamblichus appears to refer to the book
(Theo[. Arithm. 41)), it is merely another example of a neo-
Pythagorean forgery ascribed to a putative early Pythagorean. 2
I myself do not believe that Aristaeus, the Crotoniate successor of
Pythagoras, ever existed, but that he is the ghost of a ghost : a
phantom reflection of Aristeas of Proconnesus, who appeared at
Metapontum and so is called a Metapontine in Iamblichus' list. 3
If Aristeas was supposed to have been a Pythagorean it would
account for the interest he held for later Pythagoreans, who,
according to Iamblichus [T. 22], not only believed the tales told
about Aristeas, Abaris, and the like, but in many ways tried to
emulate them. Iamblichus has just been speaking of the divine
nature of the arts of divination, av-nK~, and ritual, -reAeanK+-
almost 'magic'-and implies that in Aristeas and Abaris we have
two experts in these arts: he emphasizes them as religious teachers,
 ARISTEAS PTTHAGDRICUS 143

8110,;\&yoi. It was their skill in divination (carried out in a certain

way, i.e. through the trance?), and perhaps apotropaic magic,
which these later Pythagoreans cultivated, as they were said
to have been cultivated in the circle of the Master himself:

It is still remembered how infallibly Pythagoras prophesied earth-

quakes, how swiftly he averted plagues, how he quieted tempests and
hailstorms, and how he smoothed the waves of river and sea to ease his
friends' crossing. Such powers, we hear, were shared too and often
exercised by Empedocles, Epimenides, and Abaris. Their own poems
are clear enough evidence of this; besides their nicknames---Em-
pedocles the Galestayer (d>.e~t5.ve.os-), Epimenides the Purifier (Ka.flap
7'lls), and Abaris the Sky-traveller (a.Wpo{3t5.77JS-), so-called because, the
story went (apa.), riding on the arrow of the Hyperborean Apollo
which had been given him he traversed rivers, seas, and steeps, in
a manner 'walking the air' (a.Epo{JaTwv 7p6Trov TLva). Pythagoras, some
have thought, had a similar experience (07Tep TTETTav81.va.,) when
he conversed with the societies at Metapontum and Tauromenium on
the same day (Porph. Vita Pythagorae 29; cf. Iambi. V.P. 135 ; Levy
I 90 ff.).

I have quoted Porphyry's words at length here, not only to

illustrate the magical feats which Pythagoras and his strange
friends were supposed to perform, but to draw attention to two
more cases of bilocation like those of Hermotimus and Aristeas.
It is clear from Porphyry's language that someone had suggested
that the tale of the Hyperborean Abaris flying about the world
on his arrow allegorized the ability of his soul to leave his body
and travel about; and had likened this to an appearance of
Pythagoras in places 200 miles apart, and separated by sea, on
the same day-this was another '.journey in the soul'. Such a
suggestion about Abaris could not have been made before the
fourth century B.c., for the tale of his riding his arrow is no earlier
than that (it is from Heraclides Ponticus [158)). Tauromenium
too was not founded until the beginning of that century, at least
roo years after the death of Pythagoras; but this seems to be
merely a distortion, deliberate or otherwise, of another story,
that Pythagoras appeared simultaneously at Metapontum and
Groton (Aristotle ap. Apoll. Afir. 6; cf. Levyigg). There may be a
144 ARISTEAS P2"THAGORICUS
reference to a trance journey in a remark ascribed to Pythagoras,
'when out of the body I have heard a musical scale being
sounded'; the Music of the Spheres must here be meant, but the
bodiless state in which Pythagoras had this experience may have
been between bodily incarnations rather than a trance (schol.
Ambros. ad Hom. Od. i. 371: lgw yevo.evos TOV aw.aTOS d/(1/Coa
l.e>.ovs a.p.ovlas; cf. Levy ii 148. Voss (57) assigns this saying
to a work of Heraclides Ponticus, but this is no more than a
likely guess).
A further connexion of Aristeas with Pythagoras and early
Pythagoreans, imaginary or real, is implied in that curious
passage of Gregory of Nazianzus [r23], where his name is
associated with that ofEmpedocles, who tried to gain credence for
his own apotheosis by destroying all relics of his body and so dis-
appearing from human ken, but was betrayed by the preservation
of his sandal; and of Trophonius and others not mentioned, who
tried to win some supernatural reputation by hiding away in cells
(do1frois). This latter ruse is illuminated by a story in Herodotus
of the Thracian Salmoxis, 'a servant of Pythagoras', according
to some, who, after his emancipation and return home,
had a chamber built, in which from time to time he received and
feasted all the principal Thracians, using the occasion to teach them
that neither he, nor they, his boon companions, nor any of their
posterity would ever perish, but that they would all go to a place
where they would live for aye in the enjoyment of every conceivable
good. While he was acting in this way, and holding this kind of dis-
course, he was constructing an apartment underground, into which,
when it was completed, he withdrew, vanishing suddenly from the
eyes of the Thracians, who greatly regretted his loss, and mourned
over him as one dead. He meanwhile abode in his secret chamber
three full years, after which he came forth from his concealment, and
showed himself once more to his countrymen, who were thus brought
to believe in the truth of what he had taught them (iv. 95).

We hear of a similar trick played by Pythagoras himself on the

Italiotes-but on no earlier or better authority than that of
Hermippus (ap. D.L. viii. 41). 4 That the two stories are related is
obvious, but which gave rise to the other is not so clear; probably
 ARISTE.1S PrTHAGORICUS 145

the Salmoxis story is indeed the earlier, based on certain similari-

ties between the teaching and methods of Pythagoras and the
Thracian cult ofSalmoxis (as is well argued by Morrison ( I 39 ff.)),
and then adapted by ill-wishers to Pythagoras himse1 Some such
tale about more than one personage appears to have been known
to Sophocles, for he makes Orestes say (El. 62 ff.)
7]01]
yap el/3ov ,,.o.MaKLS' Kat TOVS aocpovs
>.6ytp /J,UT1JV 9vnaKoVTas el9', o-rav /36.ovs
eMwcrw aJ9,s, EKTeT(7JVTaL ,,.Mov.
'Before this I have marked more than one occasion when clever
men have died by false report; then, when they come home again,
they are henceforth respected more than ever.' Orestes plans
to put about a report that he has been killed, to lull his enemies'
suspicions; he has disappeared, and the presumption of his death
must be encouraged. The stories about Salmoxis and Pythagoras
just mentioned would provide a good parallel to his present case;
so would the Herodotean ('anti-shamanistic') story of Aristeas;
the 'shamanistic' tradition of Maximus and others, none at
all,5
Gregory's words are the only inkling that the disappearance
and reappearance of Aristeas may have been explained by some-
one as a withdrawal into a cellar, after the manner of Salmoxis
and Pythagoras. This would have been a handy bit of support for
Rhys Carpenter's theory that Salmoxis, Trophonius, andAristeas-
Aristaeus, to name only a few, were originally hibernating bears
(II2 ff.); but unfortunately Empedotimus, who is charged by
Gregory with the self-same fraud as Trophonius and Aristaeus,
was almost certainly a character invented by Heraclides Ponti-
cus, so could hardly have been a bear (cf. Rohde i 330; Bidez,
eh. vii. I may add that the earliest authority we know to have
given the famous story ofEmpedoclcs' leap into Etna and his be-
trayal by his bronze sandal is this same Heraclides (D.L. viii.
67 ff.)). I rather think that Gregory is somewhat indiscriminately
lumping together here a number of 'supernatural' men with
certain features in common, the lowest common denominator
being a connexion ofsomcsortwithPythagorasor Pythagoreanism
6358 L
146 ARISTEAS PrTHAGORICUS
(Trophonius might qualify for inclusion by having his oracle
underground, but also his consultants were said to have visions
ofa Pythagorean type, cf. Plut. gen. Socr. 21 f.).
In the last chapter I suggested that the bestowal of divine
honours upon Aristeas may have originated in the fancy of some
fourth-century author with Pythagorean leanings [129], and al-
ready in this chapter we have met pointers to the same period, in
the indications that Heraclides may in some way underlie the
Gregorian passage, and in the 'journeys in the soul' of Abaris and
Pythagoras (where again there was cause to mention the name of
Heraclides). Such journeys in the soul are, of course, the central
theme of the non-Herodotean or 'shamanistic' tradition about
Aristeas : I therefore propose now to examine more closely the
history of this notion among the Greeks.
The edifying flight of the soul and its keener perception when
separated from the body was a topic dear to the Hellenistic age
(cf. especially Festugiere ii. 441 ff.). Pindar perhaps had already
spoken of the far-reaching flight of mind: '[His] intellect flies
everywhere, in Pindar's words "beneath the earth" and measuring
the earth's surface, "and above the heaven" observing the stars,
and nosing out the nature of all things everywhere' (Plat. Theaet.
I 73e: ~ 8J 8uf.vota . 1raJ1'7'axfi 1Tlrrat 1<ard. fl{v8apov "rus Tfi
yiis {ml.vep8<" 1<d l-rrt1rt:8a ')l<W,erpoiJaa, "oi3pavoiJ (}' mrep" darpo-
vo,oiJaa, 1<al TTiaav mf,117"{/ ,f,iJow Jpwvw,lll't)) ;6 like Empcdocles'
God, who is 'holy mind alone . traversing the entire universe
with swift thoughts' (fr. I 34. 4; ,f,p~v l<p~ . ,f,poVTlat 1<6a,ov
a:rraJl'7'a 1<arataaouaa Bofjaw). This faculty is enhanced in propor-
tion as the soul is freed from the encumbrance of an importunate
body, hence the philosopher will welcome death eagerly, as
offering to his liberated soul a clear sight of universal truth. This
note was struck for later ages by Plato (Phaedo 62-68, especially
66e) : 'For if we cannot have pure knowledge of anything in com-
pany with the body there are only two possibilities : knowledge is
obtainable by the dead or not at all ; for then and not till then
will the soul be on its own, untrammelled by the body.'
Pindar's theme is taken up in the proem of the pseudo-Aristote-
lian de mundo, in praise of philosophy: 'Since it was not possible
 ARISTEAS PTTH.<CORICUS ,,/
to reach the heavens in the body, the soul at least, by means ot
Philosophy and under the leadership of Intellect, has made the
crossing and sojourned there, and in the understanding has
annihilated the greatest distances in space; easily recognizing its
kin, apprehending the divine-for the soul sees with divine eyes-
1
and proclaiming it to men' (39ra: bmoiJ yap oJx ot6v 'T ,iv 'Tlf
, , ' ' >~ , 8 , t .... .,. ' \ ~
uwa'TL ELS' 'TOV ovpavwv a.,,LKfiU at 'T01Tov 'TJ yovv .,,vX'TJ oLa
<f;,>.ouocf,{as, >.a{JofJua ~y,q.1,6va 'TOV vouv, br.patGiifJri Ka ol[01777u1:v
KaL\ 'TO.\ 1TllfiLU'TOV
\ - ' \ \ ,,
allll'f}llCJ.JV , .J. ~
ay,EU'TW'Ta -
'TOLS' '
'T01TOS ~ ~ '
'T'(/ oLaVOLC?, UVV-
J..'
fi'l'OP'TJUfi, t_~I
P<fOLWS, '7'
OLa,, 'Ta\ uvyyeV7J...., y,,wp,uaua,
I
KO.L\ 8' ,/.,
ncp yn)X'TJS
oaT 'TO. 8E,a. KaTa.Aa.{3oiVT), 'TOtS 'Tfi av8p,fnro,s 1rpo<{n/'TfiVOVUa)-
a theme to be often repeated by later writers.
The variation in the Phaedo too is used by Cicero (from Posi-
donius ?) in T.D. i. 44 f.: that love of beauty, which compels us to
philosophize, will be really gratified after death. Those who have
travelled to the Propontis or the Pillars of Hercules think they
f have seen something: 'Can we imagine what a sight it will be
when we can take in the whole earth at a glance, its position,

I shape, and circumference, its habitable zones and its zones un-
inhabitable by reason of heat or cold?' So too Seneca pictures
Cremutius Cordus welcoming his grandson to heaven and teach-
ing him the secrets of the universe he sees around and below
him, from the fund of true knowledge now, no longer from guess-
work ('ille nepotem suum ... adplicat sibi nova luce gaudentem
et vicinorum siderum meatus docet, nee ex coniectura sed
omnium ex vero peritus in arcana naturae libens ducit ... et in
profunda terrarum permittere aciem <iubet); iuvat enim ex alto
relicta respicere' (Dial. vi. 25). Cf. Luc. ix. r-14).
All this is not unlike what Maximus says of Aristeas, that his
soul in its flight through the air over lands Grecian and barbarian
saw all the features of earth and the customs of its inhabitants,
while having a clearer view of the celestial bodies than we have
here below [122]. There is, however, this difference to be noted
between Maximus' Aristeas and the other passages: in the latter,
from the de mundo onwards, the thought is never far away of the
pettiness of the things of earth, as contrasted with the view of the
cosmos (Festugiere calls it a contrast ofa moral order (ii. 446 ff.)).
148 ARISTEAS PTTHAGORICUS

This is absent from the former, a fact which may indicate a pre-
Hellenistic source.'
There is of course another difference. In Pindar and the
pseudo-Aristotle the metaphorical flight of the mind (ouivoia,
vous) is meant, and in Cicero and Seneca the actual flight of the
understanding soul after death has released it from the anchor of
the body. But Aristeas' soul bears his understanding mind be-
tween the face of earth and heaven before this final separation
from the body: he anticipates death.
This notion too is popular in Hellenistic times. For an under-
standing of its history two texts are of major importance; both of
them fragments of the work On Sleep of Aristotle's pupil Clearchus
of Soli (frr. 7 and 8). One relates a kind of hypnotic experiment
carried out on a boy by means of a i/Jvxov'AKos paf3oos, or 'magic
wand', which persuaded Aristotle, who witnessed it (according to
Clearchus), 'that in truth the soul is a separate entity from the
body, that it enters into the body and uses it as a sort of hostel'
(ws apa xwp,,Tat TOV adJ,aTOS <~ r{;vX1J> KaL ci.JS XPfiTat at)T<p ofov
Ka-raywylcp). Striking the child with his wand, the operator 'drew
the soul forth, and seeming to lead it by means of the wand away
from the body demonstrated that the latter, remaining motionless
and uninjured (i.e. not reacting to the injuries?], was as insens-
ible to pricks as a corpse' (T7JV rf;vx~v lg,t'AKVCTEV ICO.L otov aywv
t,) ) ""' I .,.. I > I 'Jo I~ l:. \ "" \
Ot O.VT7JS 7T0ppw TOV crw,a-ros O.KWY/TOV EVEOt1, TO crw,a Kat

>
d/J'Aa{3Js crcp,6,Evov dvaicr8rirEtv 7Tpos ('rd.s 7TA'YJYd.S TWV yva7TTOVTWV
op,owv difivxcp). After the soul had been returned to its body it
recounted all that had happened (d.7MJyytAv iKa.a-ra.). [ Cf. 139 f.]
Such a temporary separation of soul from body might happen
spontaneously. Speaking of men of old who died and lived again
-Aristeas, Hermotimus, Epimenides, for example [T. 24]-
Proclus (our intermediary for this as for the previous fragment)
quotes Clearchus8 again: Cleonymus the Athenian, on the death
of a dear friend, pined away so that he seemed to die (<Wv.,7aas
i>..i7Torf;vxricrv). Two days later, just as his funeral was about to
take place, he revived and told all that he had seen and heard
when absent from the body. He said that his soul, freed, as it
were, from its corporeal chains, was raised heavenwards and saw
 ARISTEAS PYTHAGORICUS 149

upon the earth beneath 'places of different shapes and colours,

and rivers that no man could look upon' (dp8efrrav i'.mJp yfjs iSei:v
' ' ,.... ~ ... , ' ... , '
'T01TOVS' ev atrT'[/ 1TaV'Toaa1rovs Kai -ro,s ux:r,f.l,aO't Kai 'TDLC: xpw.aaw Kai
\ \

ptEvp.aTa 1T0Ta,wv 6.7Tp60'01T'Ta dvBpdmois). Finally he reached a

place sacred to Hestia, where were indescribable female demons;
there he met another, a Syracusan, who was in the same condition
as himself. A mysterious voice told them to stay quiet and watch
what went on, and they saw 'souls being judged and punished and
purified one after the other under the supervision of the Furies'
,/, ,.. ' ,.. \ I \ I \ , 9
J I , ,
( 'f'IIXWV EKt KOI\Q0'1S' TE KaL KptaEt, Kat 'TQS ae, Ka atpo.eva, KQL

Tas ToVTwv lmuKo11'ovs E-J,ev{Sas). They were then ordered to

return to earth; they arranged to look each other up if possible,
and later did meet and recognize each other in the flesh.
This story, or rather the father of it, pretty clearly underlies
that in a work of Cornelius Labeo ( ?de dis animalibus): 'Two men
died on the same day and met at a sort of crossroads; then were
ordered to return to their bodies. They made a pact that they
would be friends for the future, and so they were until they
actually did die' (Aug. G.D. xxii. 28: 'Labeo enim <licit, duos uno
die fuisse defunctos et occurrisse invicem in quodam cornpito,
deinde ad corpora iussos fuisse remeare et constituisse inter se
amicos essc victuros, atque ita esse factum donec postea rnore-
rentur'). I mention this for an interesting variation: the 'place
sacred to Hestia' of Clearchus is here a crossroads (compitum
quoddam, apparently a rendering of -rpfoSos); the significance of
this will appear shortly.
The experience of Cleonymus strongly reminds us also of those
ofTimarchus in the oracle ofTrophonius and ofThcspesius as
a result of a concussion (Plut. gen. Socr. 21 f.; ser. num. z,ind. 22).
The souls of both ascended temporarily from their bodies9 and
witnessed the fate of the dead in 'the Hades in the air'. The
information they brought back was eschatological only; but in
Cicero's Somnium Scipionis, in the sixth book of the de republica,
Scipio Aemilianus seems to ascend in sleep to the region of the
Milky Way, where the spirit of his grandfather, the great
Africanus, drives home a religious and moral lesson by means of
a cosmical discourse: the smallness of the earth, and hence the
 ARISTEAS PrTHAGORJCUS
pettiness of earthly ambition, are thrown into relief by the majesty
and harmony of the celestial spheres, and the beauty of the
Galaxy, where dwell the souls of the virtuous--above all, of
virtuous statesmen. 'The very earth now seemed so small to me',
says Aemilianus (16), 'that I quite resented our empire, which
covers no more than a speck ofit.' For small as the earth is, it is
only a small part of it which constitutes our world (olKavlvri):
'the world which you inhabit', explains Africanus, 'is nothing
but an islet embraced by the sea which you call the Atlantic, or
the Great Sea, or the Ocean; grand names, but you can see how
mean it really is. Yet even within this world you know, has the
fame of you or any other Roman ever overleaped the Caucasus
you see there, or the Ganges here?'
Undoubtedly the first-parent of all these visions, in their
eschatological aspect, is the myth of Er in the tenth book of
Plato's Republic; in their physical and geographical aspect,
perhaps the notion of the all-pervading understanding wedded
to portions of the myth of Er (e.g. 615a TriS o' av JK 'TOV ovpavov
dnra8Elas o,riyEfo8a, Kai Bea<; d.rixa.vovs 'TO KllAAOS'; cf. Phaedr. 246e
ff.) and of that in the Phaedo-especially where Socrates describes
how the earth would appear 'if one were to view it from aloft'
(ooh: at -ns avw8av Baijrra). One seems to hear in Clearchus'
words concermng the 'places on the earth's surface of different
shapes and colours', -ro1ro, lv rfi yfi 1rav-ro8a1rai Kai -rofs crxflau,
Kai -rofs xpclJ.au,v, an echo of Phaedo 10gb, 'there are dotted about
the earth's surface many hollows of different shapes and sizes',
dvat yap 1ravraxfi 7TEpi rrJV yfjv 7TOAAa Koi.:\a 7TaVTOOaTra Kai 'Td.S'
lUas Kat -ra. .Eye8ri, and I I ob-the earth is 'marked off in varie-
gated colours', 1ra,KlATJ, xpw.auw 8m>..w1..evri ; his 'rivers that no
man could look upon' remind one of the mysterious rivers of
legend for which Socrates finds a geographical location ( 1 12e ff.) ;
while the 'place sacred to Hestia' attended by female demons
(oa,.611,a, Svv&.a,s-) is the 'demonic place' ('r611os om.611,as) of
Rep. 614c, where the judgement of the dead takes place.
The trail takes us no farther back than Plato. The fact that
Plato calls Er a Pamphylian, the son of Armenius, suggests that
the ultimate source of such trance visions is to be sought in Asia
 ARIS1"EAS PrrHAGORJCUS
Minor or the Middle East; but confirmation is lacking, and is not
provided (Clemen, for example, (133) is mistaken here) by the
fact that Zoroaster replaces the name of Er in some tradi-
tions [ 159].
But between Plato and Clearchus two developments have
occurred: the geographical and eschatological features of the
visions have been conflated (a process which would have been
helped by the cosmological disquisition in Rep. 6r6b ff.), and the
place of the souls' punishment has been raised from under the
earth (as it is in Rep. 614d al.) to the air (in Rep. 6r4c the righteous
souls only traverse the heaven). This is implied in Clearchus-
Cleonymus could not have witnessed the punishment of souls
from his station in the sky had it taken place beneath the earth-
and is explicit in Plutarch.
Now Servius auct. ad Geor. i. 34 has this: 'Varro says he had
read that a Syracusan called Empedotimus had his eyes opened
with a more than human vision by some divine power, and saw
among other things three gates and three roads : one by the sign
of Scorpio, which was the way Hercules was reported to have
ascended to heaven; another by the path between [or through
the boundary between] Leo and Cancer; and the third between
Aquarius and Pisces' ('Varro ait se legisse Empedotimo cuidam
Syracusano a quadam potestate divina mortalem aspectum de-
tersum, eumque inter cetera tres portas vidisse tresque vias:
unam ad signum Scorpionis, qua Hercules ad deos isse diceretur;
alteram per limitem qui est inter Leonem et Cancrum; tertiam
esseinter Aquarium et Pisces'). Proclus adds to our information:
once when out hunting Empedotimus found himself alone in a
deserted spot and saw an appearance of Pluto and Persephone.
He was dazzled by the light that encircled the gods, and through
it saw all the truth about the fate of souls as if witnessing a drama
(in Plat. remp. ii. r rg: Ka-raA.atf,Oijvat ev V'TTO TOV t/,wTOS TOV
J J \ \ e I ~ - 0 \ 0 > > ~ ~ \ \
1T(pi0(OVTOS KVKIICf) TOVS E'OVS', we,v Ol' o, av'TOV 11'ao-av Tr)V 11'p,
TWV ifroxwv ~710e,av J.,, lv UV'TO'IT'TOLS' 8eaau11
1
). The story is told
10
by Heraclides Ponticus, says Proclus.
Little of the truth about the dead which was revealed to Em-
pedotimus in his vision has been preserved for us. Philoponus
152 ARJSTEAS P'l'THAGORICUS

reports that Empedotimus called the Milky Way 'the path of

the souls who traverse the Hades in the sky' (ad Arist. Meteor.
. 8 : '1'7JO't
1, .l. ' yap
' '
KEtVOS ooov ~
'"' tvat ,/, ~
'l'VXWV TO' ya/\a
_,,
TWV ' ,t
TOV L
n.tv,,v
-rov Jv o~pavcp 8ta1ropevohwv); Olympiodorus, that Empedo-
timus said that the realm of Pluto [Hades] comprised all be-
neath the sphere of the sun (in Plat. Phaed. 238). 11
Empedotimus was the author of a work on natural philosophy
(m:pt 4,uatKf/s a.Kpoaai;ws), according to 'Suidas', who quotes
Julian as saying that 'we believe Empedotimus and Pythagoras,
and what Heraclides Ponticus says on their authority' (s. 'E1re-
S6nos: f/efs 8~ 'E1re80-rlcp Kat Ilv8ay6pq. 7TtUTetfoVTes, ols TE
tKe'i8ev Aaf3wv 'HpaKAel871s cl lioVTtKos l.4,71). But it is clear, as
Rohde says (i 330), that he is, in fact, no more than a fictitious
character in a dialogue of Heraclides. His very name is merely
a conflation of Empedocles and Hermotimus (cf. Bidez 55). The
belief that souls inhabit the Milky Way is elsewhere ascribed
to Heraclides himself, and to Pythagoras (Stob. i. 378; Nu-
menius ap. Prod. in Plat. remp. ii. 129). From this, and from
Julian's words, we may deduce that Pythagoras too featured in
some way in this same dialogue (cf. Voss 60). 'Suidas' statement
that Empedotimus himself wrote a book is probably a mis-
taken deduction from references to 'the discourse of Empedoti-
mus [in the dialogue]', cl 'E1re80Tlou A6yos, and the like (so
Rohde). 12
The influence of this work of Heraclides was considerable. It
explains the transfer of the site of Hades from beneath the earth to
the air which recurs in authors from Clearchus onwards. The 'tres
portae tresque viae' which Empedotimus saw underlie the title
of Varro's satire Tpw8l77Js Tp,1r6>..,os, 1ri;pt aprijs K'T'IJO'Ews. The
-rplo8os is the compitum quoddam of Labeo, where Cleonymus met
the Syracusan Lysias in Clearchus' story; this infernal road junc-
tion appears first in the myth of the Gorgias (524a) ,13 but that Hera-
elides is behind Clearchus here is indicated by his location of
Hades in the air. 14 Wehrli has gathered over 200 mentions
of him by many authors, Latin as well as Greek-Varro clearly
knew his work well, and the four fragments which Cicero has
transmitted are no measure of his considerable acquaintance
 ARISTEAS PTTIIAGORICUS 153
with Heraclides' writings (the de republica, for instance, is very
probably a sample of a Heraclidean type of dialogue: Voss 23 f.
Cf. fr. 27 a-f W.).
Even so, Empedotimus differs from his successors (better per-
haps, his descendants), Cleonymus, Timarchus, and Thespesius,
inasmuch as they witnessed the fate of the dead in the spirit-
thefr souls visited the scene-whereas Empedotimus witnessed it
in the body, not by any transportation but clairvoyantly; as
Proclus says, 'Some have seen these things [the next world] in
the body, as we are told Empedotimus did ;1s others in the spirit
(a'.vev aw.a-ros), like the Athenian Cleonymus' (in Plat. remp. ii.
122). Nor do we know that his lesson in eschatology included
any geography, as Scipio's did; indeed, it is hard to see how it
could have done, for such a panorama of terrestrial features
demands that the observer be suspended aloft, as Empedotimus
was not.
Nevertheless Heraclides also treated at length that phenomenon
of catalepsy of which Aristeas appears to have been so remarkable
an example. Pliny, discussing men who have 'died and lived
again' (N.H. vii. 173 ff., perhaps directly from some Roman
authority-Varro or Messalla Rufus?), enumerates among others
Hermotimus, Aristeas, and Epimenides; then adds that women
are particularly liable to 'this affliction' (huic morbo) : 'relevant
here is that work of Heraclides, well-known among the Greeks,
about the woman who was revived after being dead for seven
days' (she might provide evidence of the possibilityofresurrection
to convince those sceptics who will not accept the story of Er,
says Origen (Gels. ii. r 6)).
We know a certain amount about the setting of this disquisition
ofHeraclides. It seems to have been in dialogue form, part of, if
not identical with, the work On Diseases (Voss 69). Its preamble
featured Pythagoras conversing with Leon, tyrant of Sicyon, on
the subject of philosophers (D.L. i. 12; cf. Cic. T.D. v. 8 f.).
After the manner ofHeraclides (Voss 23), this preamble was only
loosely connected with the main part of the book, in which
Empedocles related to his friend Pausanias, at an al fresco ban-
quet, the case of the Cataleptic Woman, whom he had cured

_,,-_..,,.
154 ARISTEAS Pf'THAGORICUS

after she had been for thirty days without pulse or breath (-rpw.-
KOV'Ta ~ipas cfovovv Kal a.<1g>VK-rov -ro awp,a). Later on in the night
when Empedocles had described this case, one guest saw a 'light
from heaven, and heard a great voice calling Empedocles', who
in the morning had disappeared-translated to the gods, ap-
parently, wherefore Pausanias told his fellows that they must now
pay divine honours to Empedocles (as happened, we recall, to
Pythagoras, Hermotimus, and Aristeas [ 129; 123]). A more
sceptical and naturalistic explanation of the disappearance was
also expressed in the dialogue-by another character ?-which
Pausanias disputed: that Empedocles had jumped down the
crater of Etna (D.L. viii. 60-71).
Because of this feat with the cataleptic, Heraclides called
Empedocles both a healer and a seer (Kai l71-rpov Kai A.VTw), con-
firming these titles also from some lines of Empedocles' own,
according to Diogenes (fr.112. 1-11). In these lines Empedocles
says that men come to him demanding oracles and remedies
against disease, which certainly implies that he is both healer
and seer; but the poetic form l'YJ-rpos which Heraclides used in-
dicates that he also quoted some other passage of Empedocles
in which it actually occurred-probably fr. 146, where the final
incarnation of the soul before deification is said to be as a seer,
a poet, a healer, and a champion of men:
Els 8~ TEAos .d.VTE,s TE Ka, i5.vo1r6>.o, Ka., l11-rpo,
Ka, '7Tp6.oi ri.v8pdJ1Tot<1w J.mx8ovlot<1t 1rEAoJ1Ta.t,
JI EV /3'110.<1TOV<1t 8EOt' 'Tt/J,T}<1t 'f'pt<1To,.
8 ava A A ,I. I

This would fit excellently with the dramatic ending ofHeraclides'

book, where Empedocles is translated to heaven.
We can readily see how his success with the cataleptic woman
would attest Empedocles' skill as a doctor; but how would it
reveal him as a seer also? Did the woman on her recovery impart
to him knowledge she had preternaturally acquired when in her
deathlike state? At that time she would have been especially
capableofclairvoyanceand prophecy; for,just as on the point of
its separation from the body at death the soul has prophetic in-
sight, 16 so in the deathlike trance of catalepsy it has the same
 ARISTEAS PTTHAGORICUS 155

power. This is well illustrated by a fragment of Aristotle preserved

in an Arabic source :
Aristotle tells of the Greek king whose soul was caught up in ecstasy,
and who for many days remained neither alive nor dead. When he
came to himself, he told the bystanders ofvarious things in the invisible
world, and related what he had seen-souls, forms, and angels; he
gave the proofs of this by foretelling to all his acquaintances how long
each of them would live. All he had said was put to the proof, and no
one exceeded the span of life that he had assigned. He prophesied too
that after a year a chasm would open in the country of Elis, and after
two years a flood would occur in another place; and everything hap-
pened as he had said. 17 Aristotle asserts that the reason of this was
that his soul had acquired this knowledge just because it had been
near to leaving his body and had been in a certain way separated from
it, and so had seen what it had seen (Ross, Aristotle Translated xii. 23).

Galen refers several times to this work of Heraclides. One

passage is of particular interest to the present investigation (viii.
414 = fr. 79 W.): women afflicted with one kind of hysteria lie
bereft of sense and movement, with only the faintest pulse or
none at all; 'this kind, which is described in Heraclides' book, is
very puzzling. There, a woman is said to have passed into a
state where neither respiration nor pulse was manifest, only dis-
tinguishable from death inasmuch as there was a trace of warmth
about the middle portion of the body.... The doctors present
disputed whether she were not actually dead' (i-yt1 oe BEaaa.EVo;
1TOAAas yvvai'Ka; VO'TEplKO.S', 7-WtiS' ev ava,aB~rovs rE aa Kat
, , I ) ~I I \ I t I" ,I..\
aKIV1frOVS' KE,Eva,, avoporarov r Ka, vcporaTov EXovaa<; a.,.,vyov
\ \A > .J..I .J.. I \ i' I \ \
'T} Kai 7TaVTEI\WS' aa.,.,vKrov<; .,,awoEvas, TJ J,Llf.V ovv TTPWTTJ /\E/\E-
";'/.1.EV'TJ S,aq,opa. Kara. ro Toii !IovnKoii 'HpaKAEl8ov yEypal.vov
R Q\ I , I ,, \ \ \ ff I \ I ' ,,
,.,,,.,I\LOV a1rop,av EXE' 7TO/ll\'T}V 01rws yt)'l'Erm. AEyErm yap a1rvot<; rE
\ I/ J.. ' I f I/ /J
Km aa.,.,vKros EKHV'TJ 71 av pw1ros yEyovEvm, rwv J'EKpwv EVL ovcp
I ""' ..., t- \ I

S,aAA<l'Trovaa, TCf) ~paxEtav XHV 8EpCYT')Ta Kara 'Tfl. iaa l.pr,

I
A I R I)\ # ' lJ A I~
rov aw.aros. E1TL'}'Eypa1rra, yovv ro l"'l""'v aTTvovs npaK Etoov,
A '

Ka~ ,~'MJULV </,'(J yEyoviva, Tots 11apoiiaw larpofs, d ~1rw re8V1fKEV).

The words KEtfLEVllS' dp.v8poraTOV lxovaas aq,vyov AEyEra,
I/ I ...
a1rvovs yEyovEva,, TWV vEKpwv EJJI ov'fl o,U/1/\arrovaa
,.. t\ I \;'~\\!
are cunously
like those of Maxim us on Aristeas [ 121), ro fL~J' awa eKuro ;1rvovv
 ARISTEAS PTTHAGORICUS
iv, ~ av8pws Ka, t1yy&ra:ra Oav&:rov. There is also a similarity
betweenPlutarch's explanation of the Cretan Epimenides' ability
to maintain life by means of the plant called o..\,os- 'because
nature can nourish the vital spark and preserve the organism on a
quite minute supply ofheat' (Jae. in arb. lun. 25. 94oc)-and what
Galen says of Heraclides' cataleptic: in those in whom an excess
of heat agglomerates about heart or lungs the breathing is fast,
but in those in whom the temperature has been lowered the
opposite occurs, even to the extent of simulating death (fr. 80 W.:
, o,s
Ell " OE
<" a7TEy,VKTa,
' ',/, 'TO 8Epov,' 71 Evavna,
' ' WCT'TE
" Ka,' 'TtCTW 71071
"<> 'TEI\EWS
\ '
"~ t. ' \ \ '
E005EII aTTOI\WI\Ella.,,
\ ~ n rU \ ""
Ka., 7111 0 ovnKOS n.paK11Eio71s aTTVOVV Eypay,EY
I ,/,

Tj8e E1=lv). Epimenides might indeed be called the prince

of cataleptics, having slept solidly for fifty-seven years with-
out physical ill-effects, during which time he claimed to have
been in the presence of the gods, to have heard divine discourse,
and to have witnessed Truth and Righteousness (Max. Tyr.
,,
x. I : ovap ".I.- evrvxew
e..,,, 1
, avros eeo,,
A ,- Km. eewv
, - l\oyots-
', Ka,, w,:11 ~, eEt(!-,
Kat UK?)), It should come as no surprise now to learn that
Epimenides' soul also, like those of Hermotimus and Aristeas,
could leave and return to his body at will (Suid. s. 'Em.evl871s).
I suggest that the dissertation of Heraclidcs TTept rfjs o.TTvov dis-
cussed the physical causation of a phenomenon which also had an
occult side to it: that of the deathlike catalepsy wherein the soul
seems to acquire paranormal knowledge of events distant in time
or place; that it centred round Empedocles' female patient, but
that strange stories about other people were adduced as instances
of the same phenomenon-Hcrmotimus, Epimenides who was
said to have slept so long, Aristeas who travelled so far and
appeared elsewhere after his 'death', perhaps Abaris the Hyper-
borean seer who was also like Epimenides in that he too journeyed
in the soul and had some means of maintaining life without eating
(Hdt. iv. 36), and Pythagoras himself, who could remember his
soul's experiences out of the body. All these figures, including
Empedocles, come to be connected with Pythagoras, and the
ability to pass at will into a prophetic trance comes to be treated
as a particularly Pythagorean aptitude; for which perhaps
Heraclidcs may be responsible too.
 ARISTEAS PYTHAGORICUS 157

It may be felt that my suggestion that Aristeas was mentioned

in the Trept rij, ll.1r11ov is too speculative. However, he almost cer-
tainly found a place in another work of Heraclides, perhaps the
book On Oracular Utterances (1ript XPYJaci.w) or that On Oracles (1r1:pt
XPTJUT7Jplwv). I havelong thought that a passage of Clement [T. 21]
was Heraclidean, and am glad to see that Wehrli is also of this
opinion, though he assigns it to one of the works On the Soul (fr.
go): 'Prognostication was practised by the great Pythagoras, by
Abaris the Hyperborean, by Aristeas of Proconnesus, by Epi-
menides the Cretan, he who went to Sparta, by Zoroaster the
Mede and Empedocles of Acragas and Phormio the Spartan;
also to be sure by Polyaratus of Thasos and Empedotimus of
Syracuse and especially by Socrates the Athenian' (here Wehrli
ends the fragment; there follows a quotation from the Theages,
and a reference to a method of divination used by a Phocian t)Tant
called Execestus on the authority of Aristotle in The Phocian Con-
stitution 18).
The appearance of Empedotimus immediately points to Hera-
elides as the ultimate source of this passage, and we have already
seen that Pythagoras and Empedocles played a part in his writ-
ings. About Polyaratus I can discover nothing more, and the
mention of Socrates is of no help; but an examination of the other
names yields interesting results.
Hcraclides believed, or spoke as ifhe believed, that the Hyper-
boreans really existed, and that in his own day. Plutarch records
that Heraclides claimed to have heard a report from the west that
a Hyperborean army had captured a Greek city called Rome,
which was situated somewhere near the Great Sea. 'But it
would not surprise me', continues Plutarch, 'that Heraclides,
inventor of fairy tales as he is (u0di87J Ka11r>.aaarla11 ovra), had
grafted this nonsense about Hyperboreans and the Great Sea on
to a true account of the capture of Rome [by the Gauls]'
(Cam. 22). 19 In at least two of Heraclides' books a particular
Hyperborean, Abaris, appeared or was mentioned: in that On
Righteousness (m,pt 811<a1oat5V7J,), and in that Concerning the Sayillgs
ascribed to Abaris (1rcp1 T<.ul' el, :4/3apt11 d1a<f,epoi11w11. 2 Frr. 51c;
74, 75 W.).
 ARISTEAS PYTHAGORICUS
Pindar is the first author we know to have mentioned Abaris.
He said that he came to Greece in the time ofCroesus (fr. 283);
Herodotus, that he traversed the whole earth carrying an arrow
and eating nothing (iv. 36) ; Plato, that he was a purveyor of
spells (l1rq,Sa{: Charm. 158b). A contemporary of Heraclides, the
orator Lycurgus, provides more information in his speech Against
Menesaechmus: as a result of a famine among the Hyperboreans
Abaris came and served Apollo [at Delphi, presumably], and,
obtaining mantic power from him, went round Greece prophesy-
ing, having as his badge the god's weapon [the arrow] (fr. 85:
, - , ,
MJ.LOV YEVOJ.LVOV EV TOtS
'Y7TEpf3opEo,s
~ , /\"8,WV O, ~AR , ,
n,.,ap,s J.Lt0'8WTVO'
T<p >A I\\
- h'TT0/\1\WVL, Ka' a (jwv
' XP'YJO'J.LOVS
' 7Tap ' avTov,
' - <TVP,,.,OAOV
, R \ "
XWV TO'
A I \ \ 'E'" ,., Th e
/3 /\OS
I\
'TOV- "'17TOI\I\WVOS I
7TPL?JH ,
V -
T?) 1\1\aOL I
,aVTVOJ.LVOS )

arrow here seems to have been meant as a guarantee of the
genuinely Apolline quality of Abaris' divination; but Heraclides
makes it do a real job of work. For him it is a huge weapon, the
personal property of Apollo, on which Abaris comes riding, like
a witch on a broomstick (fr. 51 W.), to discourse with Pythagoras
in the presence of Phalaris on a most appropriate subject for a
Hyperborean, Justice. 21 It has been suggested that Heraclides in
fact preserves the original version in which the arrow was the
carrier and not the carried, and that Herodotus has rationalized
the story ;22 but the evidence of Lycurgus, which is independent
of Herodotus, is against this. 23
It seems, then, that Heraclides, just as he improved on the
capture of Rome by the Gauls by adding romantic touches about
Hyperboreans and so on, in the same manner has improved on
earlier stories of Abaris and his arrow; perhaps (somewhere, if not
in the 7Tpi S,Kawc1VV1JS) suggesting that Abaris' transportation
on the arrow symbolized his ability to make trance journeys in
the soul [ I 43]-at any rate this suggestion could not have been
made before Heraclides, if he invented the tale that the arrow
carried Abaris ;24 and associating him for the first time with
Pythagoras, thus causing his inclusion in the catalogue of early
Pythagoreans given by Iamblichus [ 142], and also the legend that
he was a teacher of Pythagoras: for the latter was a disciple not
only of Pherecydes and Hermodamas in Samas, but later of
 ARISTEAS PrTHAGORICUS 159

Abaris and 'Zaratas the Magus', according to a scholiast on Plato

(in remp. 6oob: vi. 360 Hermann).
This Zaratas is the same as the Iranian religious teacher more
commonly called Zoroaster, also mentioned in the passage of
Clement under consideration. His name was apparently known
to Greek learning as early as the fifth century B.c. (to Xanthus
of Lydia); it was certainly known in the fourth, to the early
Academy and the Lyceum; and, most pertinently for the present
investigation, Heraclides Ponticus wrote a work entitled Zoroaster
(Plut. Col. 14 = fr. 68 W.; c Voss 61).
We have no fragments specifically from this book, but it is
fortunately possible to make important deductions about it with
considerable assurance. In it Pythagoras visited Babylon and
there sat at the feet of Zoroaster, by whom he was purified and
taught not only to maintain this purity by abstention from beans
but also truths about the soul and about natural philosophy. 25
Whether in this dialogue Zoroaster also claimed to have made
journeys in the soul and to have had previous incarnations can
only be surmised; but the former claim was certainly current
by the early third century B.c., for Colotes, a pupil of Epicurus,
found grounds for asserting that the original hero of the myth in
the tenth book of the Republc was not Er but Zoroaster (Prod.
in Plat. remp. ii. rng),26 and later apocryphal works of Zoroaster
appeared in which he personally professed to supernatural
knowledge gained in the next world by the process of 'death and
resurrection' (Bidez-Cumont ii. 158).
As to the claim to previous incarnations, there is no direct
evidence that any such was made on behalf of Zoroaster, 2 7 but
there are discrepancies in the later Greek tradition about him that
could be explained on the assumption that somewhere it was
made. It was asserted by some authority that he was a Greek and
a Proconnesian. A scholiast on the Platonic Alcibiades I states that
'some people made Zoroaster a Greek, others the son of one of
the invaders from the continent beyond the Great Sea; and they
say that he was taught all wisdom by the good deity' (122a
(Bidez-Cumont ii. 23) : ol Jv "E:V..1]va, ot St TWV lK -rijS' v1r,p '"}V
/J,fi')'U)<1]V {J&).auuav ~1T{pov wp1)VWV 1TatlJa tpaut, 'TTU.Oa.V 'TE uotp{av
160 AR!STEAS P1"THAGORJCUS

JK rou &yaOou ~atovo1; lKaOEiv). Pliny preserves a tradition that

there was a Zoroaster of Proconnesus living in the sixth century
(N.H. xxx. 8 : 'diligentiores paulo ante hunc [Ostanes, who accom-
panied Xerxes in his invasion of Greece] ponunt Zoroastren
alium Proconnensium'). It has been suggested that this was
Aristeas himself, a conjecture which does not commend itself
to Bidez-Cumont (i. 23). I should explain it rather as a con-
fused reminiscence of some assertion that Zoroaster (or rather
Zaratas) was a reincarnation of Aristeas.
There is, of course, a glaring chronological inconsistency be-
tween the teacher of Pythagoras, who for Heraclides was a con-
temporary of Phalaris, and the Persian magus who lived 6,000
years before Plato (as stated by the scholiast on the Alcibiades
just quoted; cf. Arist. ap. D.L. i pr. 1 f.; Plin. N.H. xxx. 3) ; but
there is the further curiosity that, although the work of Hera-
elides was entitled Zoroaster, the sage therein who instructed
Pythagoras at Babylon seems to have been called <,aratas or
<,aratus (cf. the passages of Aristoxenus and Porphyry referred to
in note 25). Now a Zaratas living in the sixth century B.c. is
much closer in name and date to the founder of Mazdaism
(Pehlevi <,aratu!t, c. 600 B.c. : Bidez-Cumont i. 37 ; Henning
35 ff.) than a Zoroaster living in the sixty-fourth; perhaps
Zwpocf.<TTprw, 'Purely-astral', really was an etymological fiction
based on Zap&.ras by someone trying to make a significant name
for 'the first astronomer'. I advance this suggestion with great
hesitation (cf. Bidez-Cumont i. 6); anyway, two forms of the
name may have been known to Heraclides, who then in his
dialogue made Zaratas, the teacher of Pythagoras, a later in-
carnation both of Aristeas of Proconncsus and of the Zoroaster
who had lived 5,800 years before (Zaratus is distinguished from
Zoroaster by Pliny, N.H. xxx. 5). 28
We have already seen another name in Clement's list, that of
Epimenides the Cretan, associated with Pythagoras [ 143]. He
accompanied Pythagoras into the Idaean Cave when the latter
was returning from his visit to the Magi (D.L. viii. 3), and is
called a pupil of Pythagoras by Iamblichus ( V.P. 222) in a story,
appended to the resume of the Heraclidean 1rEpt 81KmoavV7Js, where
 ARISTEAS PrTl/AGORICUS

Epimenides is said to have caused his would-be assassins to go

mad and slay each other (perhaps suggested by the tale of Cadmus
and the Sparti). We have also seen that he was a cataleptic, who
learned divine truth when in this state [ r 56]. We only need to know
that he was reincarnated for him to conform admirably to the
pattern ; and such a claim does seem to be made for him in a
passage of Diogenes Laertius (i. r 14) : Epimenides said that he
had first been Aeacus, had foretold to the Spartans their capture
by the Arcadians, and had often 'lived again'. 29 This would
explain a chronological problem in the legends about Epi-
menides. He is said to have visited Athens at the beginning of the
fifth century by Plato (Laws i. 642d), a century earlier by
Plutarch (Sol. I 2 ; cf. Arist. Ath. Pol. I ; Suid. s. 'Em.ev{871s; D.L.
i. z zo). There is no difficulty about this for a man who lived to
be more than I 50 (so as early as Xenophanes: D.L. i. I I r); but
he would have to be much older than that to be able to prophesy
to the Spartans their capture by the Arcadians, as Diogenes says
he did. The only known incident to which this prophecy could
refer is the capture of the Spartans by the Tegeans (Hdt. i. 66), 30
put by Pausanias in the reign of Charilaus in the ninth century
(iii. 7. 3 al.), which could not have happened in Epimenides' life-
time even if we accept the Cretan version of his longevity, that
he died aged 299 ! In fact, it must have happened in an earlier
incarnation, if this was the distasteful prophecy for which the
Spartans are said to have put Epimenides to death (Paus. ii. 2 I. 3 :
7Tp6 0~ TOV vaov -rfj, l1871vas 'EmEvloov Myova,v Elva, -rac{>ov.
AaKEOatovlovs yti.p 'TTOAe~aavm, 'TTp<>s Kvwolov, EAe,v {wna
'E7T&J.LEVto71v,
,., 11a \ /3'ovras oE
.,, a1roKTuvai,
' A <> '
own ,J..'
o't'tow ,
ov,c
a,o,a ,
eavrEv-
I

ETo). This view would account for the way in which Diogenes'
mention of the prediction is sandwiched between Epimenides'
claim to have been Aeacusand his claim to have often 'lived again'
(cf. note 29).
About Phormio, the remaining name in Clement's list, Pau-
sanias tells the following (iii. 16. 2 f.). He acquired a house in
Sparta where once had dwelt the Dioscuri. These came to him
in the likeness offoreigners; they said they had come from Cyrene,
and asked him to put them up in the room that had been their
 ARISTEAS P'YTHAGORICUS

favourite when they were on earth. Phormio made them free of

the rest of the house, but would not grant their request for that
particular room, as it was his daughter's. The next day not only
the strangers but the girl and all her things had vanished, and in
her room instead were found two images of the Dioscuri and some
silphium on a table (c Plut. non posse suaviter vivi 22. 1103a).
This is one story about Phormio. But there is another, from
Theopompus, which explains his inclusion with Aristeas and the
rest in the Clementine passage ('Suidas', s. <l>opLwv):
He was a Crotoniate,3 1 and was wounded at the battle of the Sagra.
As the wound proved difficult to heal, he received an oracular bidding
to go to Sparta ; there the first person to invite him to dinner would
cure him. On his arrival at Sparta he had just dismounted from his
chariot, when a young man extended such an invitation. After dinner
his host asked him the reason for his journey, and, when he heard about
the oracle, took some scrapings from his spear (dTTofuaa, -rov 8&pa-ro,)
and placed them on the wound. After Phormio left, he was about to
ascend his chariot, as he thought, when he found himself grasping the
handle of his own front door back in Groton. On another occasion he
was celebrating the Theoxenia, when the Dioscuri [who were presum-
ably his guests at the sacral entertainment] invited him to go to Cyrene
to meet Battus; and he stood up holding a stalk of silphium (Ka,
dvlcn-TJ -r lxwv atA~lov 1eav.\&v).

Here, not only does Phormio undergo a miraculous transporta-

tion in the body from Sparta to Groton, but there is in the last
sentence an obscurely worded reference to a transportation in
the spirit, a soul-journey. He is invited to meet King Battus of
Cyrene because he is another seer (he is presumably the Battus
who is said by Clement, immediately before the passage under
discussion [T. 21], to have compiled a treatise known as the
Mantic Art of the Argonautic diviner Mopsus. The original Battus
who founded Cyrene introduced the rites of the Dioscuri there
(schol. ad Pind. Pyth. v. I o) ). Silphium was the principal export of
Cyrene (cf. Plaut. Rud. 629 ff.; Hdt. iv. 169; connected with
Battus, Ar. Pl. 925); when Phormio got up out of his trance
(avl0"1'1)) he was holding a stalk ofit, proof that he had not merely
been dreaming. The story can be nicely paralleled from Jacobus
 ARISTEAS PYTHAGORICUS

de Voragine: a lady who was devoted to the Virgin was unable

owing to the absence of her chaplain to hear mass on the festival
of the Purification. She entered her chapel and fell into a trance
before the altar, wherein she found herself in a magnificent
church ('subito in excessum mcntis facta videbatur sibi se in qua-
dam ecclesia pulcherrima etspeciosa collocatam csse'), and seemed
to hear mass celebrated by Christ Himself in the presence of the
Virgin. At the end she refused to surrender to an acolyte the
candle she had been given ; there was a somewhat unseemly
struggle, the candle broke in two, and she came to herself in her
own chapel holding half of it ('ad hanc igitur vehementem fracti-
onem illa subito ad se rediit, et se iuxta altare ubi se posuerat
inveniens cereum fractum in manu sua repperit' (LegendaAurea 37
-de Purijicatione B.V.M.)).n
There can be little doubt that the first Phormio story, given
by Pausanias, is more primitive in form than the second, given
by Theopompus. It embodies a theme common in folk-lore, in
which a god in human guise visits some mortal and makes a
request which is refused; the mortal is punished for his refusal.
Here, the images and the silphium are the signatures, as it were,
of the Dioscuri, left to inform Phormio who his guests really were.
In the second story elements of the first are taken and adapted to
an entirely new setting. The bodily transportation of Phormio's
daughter becomes a bodily transportation of Phormio himself
(like the bodily transportation of Aristeas from the fuller's shop,
of Empedocles in the 1rpt rij, am,ov, and of Cleomedes from
the chest in Astypalaea) ; Phormio's unwitting entertainment of
the Dioscuri becomes a deliberate entertainment at the Theoxenia;
Cyrene, the provenance of the Dioscuri in the first story, becomes
the goal of a trance journey in the second, because at Cyrene
dwells the seer Battus; the silphium, evidence of the guests'
identity in the first story, becomes in the second evidence that
Phormio's soul has actually made the journey.
This is complicated enough to suggest that the second story
is a literary adaptation of the first; which is confirmed by the
fact that the author has complicated it still further by introducing
elements of extraneous legends. The oracular instruction about
164 ARISTEAS P1'THAGOR!CUS
who should be Phormio's healer and the method by which the
cure was effected are a variation on the story, treated by
Sophocles and Euripides, ofTelephus, whose wound, inflicted by
Achilles, festered incurably till he was told by Apollo that the
dealer of the wound would heal it; as happened when Achilles
was persuaded to scrape some rust from his spear on to it (d1ro[v-
aavTos rijs ,),{as TOY l&v: Apollod. Epit. iii l 7-20, with Frazer's
note). Clearly, then, theyoungmaninSparta who invites Phormio
to dinner is the warrior who wounded him in the battle of the
Sagra, and as clearly he is one of the Dioscuri, who helped the
Locrians to victory over the Crotoniates at the Sagra by fighting
at their side, according to Trogus (Justin xx. 3)-a romantic
elaboration of what may quite well have been an historical fact:
that Sparta, on being asked for help against Groton by the
Locrians, said she would send them the Dioscuri, who then
travelled back with the Locrian ambassadors symbolically (cere-
monial couches were prepared for them on the ship: Diod. Sic.
viii. 32). There is yet another variant of the T elephus theme con-
nected with the battle of the Sagra, to be found in Pausanias
(iii. 19. II ff.) and Gonon (fr. r. 18 = Phot. Bibl. 133b):
a Grotoniate called Leonymus or Autoleon was wounded in the
chest or thigh by Ajax, who was fighting for the Locrians. As the
wound would not heal he consulted Delphi, and was sent to
Leuke Nesos, a sort oflsland of the Blest in the Euxine, where he
was cured by the ghost of Ajax, met other heroes, and was given
a message for Stesichorus by Helen. This looks to me like another
literary invention, related to the second Phormio story indeed,
but as cousin rather than as progenitor.
The personages in the Clementine passage are now seen to have
much in common. Abaris, Aristeas, Epimenides, Zoroaster, Em-
pedocles, and Empedotimus were all associated with Pythagoras
(and it will not escape attention that Phormio too lived in Groton,
where Pythagoras himself taught for twenty years); Pythagoras,
Empedocles, Epimenides, also possibly Aristeas and Zoroaster,
were reincarnated; Pythagoras, Abaris, Aristeas, Epimenides,
Zoroaster, and Phormio all had trance experiences, 'journeys in
the soul', while Empedotimus had a revelation of a similar nature
 ARISTEAS PTTHAGORJCUS 165

'in the body', and Empedocles experimented with the cataleptic

trance and may have obtained such revelations from his subject
at second hand; finally, Pythagoras, Empedocles, Zoroaster,
Abaris, and Empedotimus are known to have featured in dialogues
by Heraclides Ponticus, and Empedotimus is an invention of
Heraclides. Most of the tales about these people we have seen to
be current by the end of the fourth century B.c., some could not
have been earlier than that century, and indeed the whole theme
of knowledge 'gained in the soul' while the body is left as if dead
appears to be derived in the first place from the Platonic myth of
Er. All these similarities are best explained as the work of one
adapter or editor living in the fourth century, and the weight of
evidence points to Heraclides as the man.
If this argumentation is correct, the stories I have been dis-
cussing, and especially the two versions of the Phormio story, pro-
vide us with a good sample of the way Heraclides went to work,
more extensive than, though not dissimilar from, the samples of
his treatment of Abaris and his arrow or the Gallic capture of
Rome. He would take some strange tale and use elements there-
from to construct an even stranger, shaping it to conform to
a pattern of association with Pythagoras, reincarnation, trance
journeyings of the soul, divination and supernaturally acquired
knowledge, intercourse with the gods, who 'have a concern for
human affairs' (fr. 75 W.); the bizarre is enhanced by the intro-
duction of magi, Sibyls (cf. frr. 130, 131 W., and note 25),
Hyperboreans, outer continents, and so on; and other tales and
earlier literature are laid under contribution. An example of this
is his treatment ofEmpedoclesin the,rph-i]sa'.mou [ 153 f.]: much
of it can be seen to be developed from things in Empedocles' own
writings--his friend Pausanias with whom he discourses is the
addressee of the 1Tp, </,vCTws (fr. 1) ; the cataleptic woman herself
may have been suggested by the promise that Empedocles makes
to Pausanias, that after studying his teachings he will be able to
bring the dead back to life (fr. 111. 9) ; the translation of Empedo-
cles to heaven is based on what he himself says about the ultimate
lot of the good man (fr. 146); while the translation and its cir-
cumstances, the heavenly light and the great voice 33 summoning
166 ARISTEAS Pl'THAGORICUS
Empedocles, are surely borrowed from Sophocles' account of the
supernatural summons and disappearance of the hero Oedipus
at Colonus (O.C. 1621-65), and from Euripides (Bacch. 1078):
JK 3' alelpo; tf,wi1 nr, WS' .Jv elKa.aai
.d ,6vvaos, av{36aaV' wveavt/l,s-,
V \ t "' ) \ ,
I , It
ayw TOV vp.as Kap. Tap.a 'T opy,a

ytAwv .,.,()lp.evov aAM np.wp,fo()J v,v.

Kat TaiJ{)' /f.p.' ~yopfiVE Kat 'Tl'pOS O~pavov
Ka, ya,av EO"T'1/P"" </>ws a,p.vov 'Tl'vp6s.
And there came a voice from heaven-Dionysus, at a guess-
crying: 'Maenads, I have brought you the man who mocked
you and me and our rites: now punish him.' And with these
words a pillar of divine fire was set between earth and sky.
It must have been just such a command to punish impiety that
the mysterious voice gave from the bronze laurel in the market-
place at Metapontum when the girl Pharsalia came wearing the
golden wreath stolen from Delphi by Philomelus [ r23] ; and the
fate that befell her at the hands of the maddened seers was
exactly that of Pentheus at the hands of the maddened Bacchae.
This tale must have been invented between 356 B.c., when Philo-
melus rifled the Delphic treasures (cf. Diod. Sic. xvi 30), and
Theopompus, who is our earliest authority for it. It would then
fit Heraclides chronologically, and it shows several Heraclidean
idiosyncrasies: Aristeas and the Hypcrboreans, the seers, the
divine intervention and the threatening voice, the distorted bor-
rowing from Hdt. iv. 15-Herodotus' 'statue dedicated toAristeas
and surrounded by laurels' has become a bronze laurel ; and
Aristeas did not say in the Herodotean account that he had come
from the Hyperboreans. We even have the account, quite pos-
sibly historical, of which it could be a typically Heraclidean
elaboration. This we owe to Plutarch, who says that the golden
wreath was a dedication of the Cnidians (not the Lampsacenes),
and that a mob of young men, excited with cupidity for the pre-
cious object, tore Pharsalia to pieces in their struggles for it near
the temple of Apollo in Metapontum (Pyth. orac. B).The more
fanciful version handed down by Theopompus could well have
appeared in Heraclides' book ,rept VaE/3dar;, from which is
 ARISTEAS PTTHAGORICUS
preserved an example of divine vengeance wreaked upon the
sacrilegious (fr. 46 W.).
Another personage relevant to an investigation of the Aristeas
saga appeared in the work of Heraclides. This was Aristaeus,
the son of Apollo and Cyrene. Cicero says: 'We have it that the
Ceans are wont to observe carefully the rising of Sirius, and to
deduce therefrom the healthfulness of the ensuing season, as Hera-
elides Ponticus writes. For if the star presents a dim and misty
appearance the atmosphere must be dense and torpid, so that its
effluence (adspi,ratio) will be oppressive and unhealthy; but ifit
shines brightly it is a sign that the atmosphere is rarefied and
clean, and so salubrious' (Div. i. 130 = fr. 1,p W.). In chapter
9. 2 of the excerpts de rebus publicis, ascribed by the codices to
a Heraclides, 34 is the following fragment (the chapter is about
Ceos) : 'They say that Aristaeus learnt the care offlocks and herds
from the Nymphs, and beekeeping from the Brisae; when plants
and animals were dying, by reason of the etesian winds blowing
[ 1rvEi:v AB, 8Et1rE,v (supra scripto 1TVEi:v) C, AEt1rEiv Schneidewin] . . .'35
This provides a link between the Ciceronian passage and a
scholium on Apollonius Rhodius (ii. 498) :
When the Dog Star was oppressing the Cyclades with its heat and
for long causing drought and embarrassment, the Ceans as the result
of an oracle summoned Aristaeus the son of Apollo and Cyrene from
Phthia. He, accompanied by some Arcadians, came to Ceos and
founded the temple of Zeus lkmaios, in order to cause rain; and he
appeased the Dog Star. And he ordered the Ceans to observe the rising
of the Dog Star each year under arms and to sacrifice to it. It was as
a result of this that the etesian winds blow, and cool the earth in the
summer heat [cf. Call. Aet., fr. 75. 32-37]. So Greece was freed from
the drought.
(Apollonius himself says much the same in ii. 516-27: the oracle
mentioned by the scholiast was given by Apollo, though the poet
does not say that it was the Ceans specifically who summoned
Aristaeus; the latter left Phthia at the bidding of Apollo and
settled in Ceos-he too, then, had received an oracle (519).)
This account of Aristaeus can be expanded yet further from
Diodorus (iv. 82):
They say that Aristaeus, after the death of Actaeon, visited his father's
168 ARJSTEAS PTTHAGORICUS
oracle, where Apollo foretold to him his migration to Ceos and the
honours he would receive from the Ceans. He went to the island, and as
Greece was suffering from a plague p.o,p.ov: Apollonius (ii. 519) says
they summoned Aristaeus Aotp.ofi dA~frJ'ipa] he sacrificed on behalf of
all the Greeks. 36 As this took place at the rising of Sirius, when the
etesian winds happened to be blowing, the plague ended .. , [here
ensues what appears to be an aside by Diodorus]. They say that
Aristaeus left offspring in Ceos and returned to Libya to his
mother, then sailed for Sardinia, which island, captivated by its
beauty, he reduced from wilderness to cultivation. There he begot
two children, Charmus and Polycarpus. Afterwards he visited other
islands and spent some time in Sicily, where he was inspired by the
fruitfulness of the island to give the natives the benefit of his own dis-
coveries [the arts of beekeeping, &c.]. For this reason they say that
Aristaeus receives divine honours from the Sicilians above all others,
especially from those who harvest the olive. Finally they tell how he
went to Dionysusl7 in Thrace, was initiated into his mysteries, and
passing his time with the god learnt much of utility from him. After
dwelling some time in the neighbourhood of Mount Haemus he dis-
appeared (a,f,avrov yevla8at), and has earned worship not only there
among the barbarians but also in Greece. 38
This I suspect to be all Heraclidean: the association with, and
instruction by, divine beings (cf. Zoroaster [ 159], Epimenides
[156], and Phormio's friendship with the Dioscuri [162]); the
consultation of the oracle (Heraclides was fond of oracles, cf.
frr. 46b, 50, 65, 130, 136, 138 W.); the ambivalent attitude to the
miraculous, which is sometimes accepted, sometimes explained
away (the prayer to avert the drought or plague was successful
because it coincided with the blowing of the etesian winds; this
looks as if it were a scientific explanation of a tradition in which
the winds blew as a result of Aristaeus' supplications. Compare
the method of prognostication from the look of the Dog Star
in Cicero, the varied theories of Empedocles' end, the scientific
explanation of catalepsy and of Epimenides' ability to do with-
out food),
The several references here to the conferment of divine honours
on Aristaeus as a teacher of men are also remarkable, and recall
what we have heard of Pythagoras, Empedocles, Hermotimus.
and Aristeas himself. In fact, there are two further points of
 ARISTEAS PTTHAGORICUS 169
contact with traditions about Aristeas: the mysterious final dis-
appearance of Aristaeus, and the special reverence in which he
was held by the Sicilians. There is good evidence for this last in
Cicero (Verr. u. iv. 128), and it accounts for the curious story that
the Proconnesian was worshipped by the Sicilians [ 1 29 f.]. This
is not, I think, a confusion of Aristeas with Aristaeus, so much as
deliberate adaptation or contamination of two different traditions
in the manner ofHeraclides: on the foundation that the Sicilians
give heroic honours to Aristaeus because he helped farmers has
been erected the fiction that they give heroic honours toAristeas the
Proconnesian because he helped philosophers-or more properly,
perhaps, a philosopher, Pythagoras, to whom Aristeas imparted,
or with whom he discussed, divine knowledge gained 'in the
soul'. 39 (Though we know of 'Pythagoreanism' in Sicily-at
Acragas-as early as the first quarter of the fifth century B.c.,
there is no evidence that the Master himself was ever there: but
the scene of Pythagoras' discussion with Abaris in the 1rep,8,Kato-
aVY1)s of Heraclides was the court of Phalaris, tyrant of Acragas.)
It is by a similar postulate that I would account for the ap-
pearance of 'Aristeas of Metapontum' in the catalogue of early
Pythagoreans ( I 42). It was in Heraclides' manner to use the same
characters in different works: Pythagoras featured not only in
the proem to the 7rep, Tfis a.1rvov but also in the Zoroaster, the 1rep,
81Kaioa&V1Js, and with Empedotimus; Abaris, in the 1rep, 8,Kaio-
a&117Js, and he must at least have been mentioned in the 1rep,
Twv elS' "A{Jap,v dvaif,Epohwv, as was Aristeas not only in one
of the works on oracles but very possibly also in the 1rep, Tf/S'
a.1rvov; his name occurred too in yet another book, which
we have not before considered-probably On the Date ef Homer
and Hesiod.
The Christian Father, Tatian, is concerned to show [T. 27)
that Moses was 'not only older than Homer, but older than all
the writers that were before Homer: older than Linus, Philammon,
Thamyris, Amphion, Musaeus, Orpheus, Demodocus, Phemius,
the Sibyl, Epimenides of Crete, he who went to Sparta, Aristaeus
of Proconnesus, who wrote the Arimaspea, Asbolus the Centaur,
Isatis, Drymon, Euclus the Cyprian, Horns the Samian, and
 ARISTEAS PTTHAGORICUS
Pronapides the Athenian'. 40 The Heraclidean provenance of this
list is immediately indicated by the characterization of Epimeni-
des: it was that earlier incarnation of Epimenides [161], for the
contemporary of Pythagoras was certainly not older than Homer,
And this indication is reinforced by the fact that Heraclides
elsewhere spoke of the Sibyl (as older than Orpheus-fr. 130
W.), and of Linus, Philammon, Thamyris, Amphion, Demodocus,
and Phemius as early musicians, to whom he actually assigns
not only historicity but works (fr. 157W.): to Philammon a 'Birth
of Leto and of Artemis and Apollo',41 to Thamyris a 'Titano-
machia', to Demodocus a 'Sack of Troy' and 'Marriage of
Aphrodite and Hephaestus' (obviously suggested by Od. viii.
492 ff. and 266 ff.), to Phemius a 'Homecoming of Agamemnon's
Army' (from Od. i. 326 f.).4Z Heraclides may have dated Aristeas
early because of the chronology ofHerodotus (iv. 15), or because
he thought that the one-eyed Arimaspi were the model for
Homer's Cyclopes, which seems to have been the reason why
someone made Aristeas a teacher of Homer [T. 7; 26). 43
But if Aristeas wrote the Arimaspea before Homer's time he
could not have conversed in the flesh with Pythagoras, who for
Heraclides, as we have seen, was a contemporary of Phalaris. He
must, then, have done so either in a later incarnation, or in the
spirit, as a ghostly manifestation. I believe that it was the latter, an
elaboration by Heraclides ofthe Herodotean tale ofthe appearance
of Aristeas' ghost in Italy centuries after his disappearance; that
not only did Heraclides speak of appearances of Aristeas in Sicily
but also in Metapontum (as in Herodotus; hence Aristeas is
called a Metapontine in Iamblichus' list) ; that he claimed to
have come from Apollo's country, the land of the Hyperboreans
(in Herodotus he said that he was the companion of Apollo, and
he had told of the Hyperboreans in his poem) ; that he imparted
to Pythagoras knowledge of things terrestrial and celestial which
he had gained when his soul was freed from its body [ 121 f.] ; and
that the Metapontines set up a bronze laurel in commemoration of
this visitation [ I 23] (in Herodotus it had been a statue of Aristeas
surrounded by real laurels). My argument is spun from such
slender threads that it is unwarrantedly audacious to conjecture
 ARISTEAS PYTHAGORICUS
in which dialogue this interview may have occurred; nevertheless,
it would suit quite well one of the works On the Soul, in which we
know Heraclides was at pains to prove the actual existence of
the Hyperboreans (fr. 102 W.).
I would venture farther and put it that we possess a fragment
of this discourse of Aristeas. There is in 'Suidas' this curious entry
(s. "A/3apis): 'The Avars were utterly wiped out by the Bulgars.
These Avars drove out the Sabinores, having themselves been
compelled to emigrate by tribes inhabiting the shores of the
Ocean but driven out of their country by a fog bred of the Ocean
overflowing and by the appearance of a multitude of griffins,
about which there was a saying [oracle?] that they would not stop
until they had devoured the entire human race. Under pressure
from these terrors they assaulted their neighbours, who, proving
weaker than the invaders, had to emigrate' (on ToVs 'AfJap,s ot
'\
Bov"yapoi ' '
1<aTa1<paTos NO').'
apo'Y)V OL"A/3'
'YJ'f'av,aav. OTL <' '1:'\
.t1 ap,, ovToL Es'Y)"aaav
~
"'a /3' I
tvwpas, ETavacrrai I
,'EVO/UVOL f
V'ITOI '()
E
...,
vwv ) I
OLKOVVTWV \
/UV \
T'Y)V
A > I ' ~ \ I ) \ I ~ \ \ tt , I
1TapWKEaVLTLV aKT'Y)V, T'Y)V
xwpav U1TOI\L1TOVTWV OLU TO E~ avaxvuew<;
OE
A ,,... A < \ AS, A() > ,/. ' , \ A \ \
TOV ~~KEaVOV oixl\WOES yivoevov, Kat 'YPV'ITWV OE 1Tl\7l OS ava'f'al'EV'
orrep ~v ,\6yos [leg ..Myiov ?] ~ 1rp6Tepov 1ravaaaOa, 1rplv ~ {3opa.v
A
1TOLTJUaL I
TO TWV a'v8pw1rwv
A , '
yevor;. 0 \ 0 \ < \
OLO - 0
OTJ V1TO TWVOE ',\
E
,
avvoevot
TWV
A A ,\ , > 'tl- \ \ \ A > ' \' ,
OELVWV TOLS 1T TJULOXWPOLS EVE,.,U/\1\0J/' KaL TWV (1TLOVTWV oviaTWTE-
This is clearly a
pwv ov-rwv oi ~v icf,o8ov ucf,,aTo.evo, eTavfoTav-ro).
weird distortion of Herodotus' summary of the account given in
the Arimaspea about the migratory pressures of the central Asiatic
peoples ([T. I] : a.lei TO tat TrA'Y)aLoxwpo,aL imTt8a8ai--cf. -rofs
1TATJaioxdJpois in 'Suidas'). There it was not 'the inhabitants of the
shores of the Ocean' (i.e. the Hyperboreans, 'whose country ex-
tended to the sea') who began the movement as here, but the
inland Arimaspi; there the griffins were not a mob that suddenly
appeared and threatened the destruction of mankind but the
permanent guardians of the gold of those parts. The fog and the
inroads of the Ocean (avaxvui, TOI) 'Q1avoii) are an importation
(Heraclides knew about the Atlantic tides: fr. I I 7 W.); it is note-
worthy that one of the things that the soul of Ari5teas saw in its
flight between heaven and earth in the account of Maximus
was 'inroads of the sea' (avaxvan, 8aAaTTl)S' [H!2]).H
 ARISTEAS PrTHAGORICUS
This seems, then, to be from Aristeas, but it is not the Aristeas
of the Arimaspea. I think that we have here another Heraclidean
adaptation: he has worked up hints given by the author of the
Arimaspea as reported in Herodotus' summary, and put them into
the mouth of Aristeas in his own work. But why should this fan-
tasy come to be mixed in with the history of the Avars in 'Suidas''
article? Because the Avars have the same name as Abaris the
Hyperborean, and Heraclides' Aristeas was giving the reason why
Abaris was coming to Greece: for help against a scourge that was
already causing his countrymen to move, and threatened to be
universal ('they say that on the occasion of a world-wide plague
Greeks and barbarians consulted Apollo, who bade the Athenians
pray on behalf of all [cf. 168, and note 36]. Many peoples sent
ambassadors to them, among whom came Abaris from the Hyper-
boreans', Harpocr. s. "Af3ap,s). A memory of this, I suggest, has
caused the muddle in 'Suidas' (the relevant passage is not in
Priscus (Muller, F.H.G. iv. 104 f.)).
It is, then, in the Aristeas ofHeraclides that we should see the
origin of Aristeas the Pythagorean, the seer, traveller 'in the soul',
and religious teacher, now well fitted to have a Theogony ascribed
to him [31 f.]; it is the source of that dichotomy in the tradition
about the limit of Aristcas' travels [ 133 f.], and accounts for the
date given by 'Suidas' for Aristeas, the fiftieth Olympiad [ I 26):
that is, in the time of Phalaris (who was 'tyrant of the whole
of Sicily in the fifty-second Olympiad', according to 'Suidas', s.
<Pcf.>.ap,s), with whom, according to Hcraclides, Pythagoras was
contemporaneous [169).
What manner of man was the author of these curiosities?
Heraclides was born in Heraclea Pontica some time between
388 and 373. From there he moved to Athens, in 364 or earlier,
and joined the Academy (Voss 8 ff.). Voss disbelieves Diogenes'
statement (v. 86) that Heraclides also heard the Pythagoreans,
but on insufficient grounds; there is no impossibility in it, for a
younger contemporary, Aristoxenus, was a pupil of Xenophilus,
one of the 'last of the Pythagoreans', at Athens. In 338 Heraclides
was nearly elected to the headship of the Academy in succession
to Speusippus; after this narrow defeat he return~d to Heraclea,
 ARISTEAS PYTHAGORICUS 173

where he was still alive about 315 (Voss 18 ). There is a tradition

that made him a Peripatetic also, and Sotion said that he was
a disciple of Aristotle, but this can hardly mean that he was a
member of the Lyceum (Voss 13 ). He must indeed have known
Aristotle well when they were both pupils of Plato (there was
little difference in their ages), but I think that his characteriza-
tion as a Peripatetic is based on no more than the wide extent
of his interests, in the way of that school.
He was said to be very fat, gentle, and dignified in appearance;
and Diogenes has a queer tale to tell about his end (v. 91).
Heraclea was suffering from a famine, and the people consulted
Delphi for a remedy. Heraclides bribed the envoys to produce
a forged oracle, to the effect that the city must crown him with
a gold crown and give him heroic honours after his death; but as
he was being crowned in the theatre he was seized with a fit. The
author of this story was the liar Hermippus, to whom perhaps
should also be ascribed what looks like a sequel (preserved
through Demetrius Magnes and Hippobotus: D.L. v. go; cf.
H.P., fr. I 7 W.): when on the point of death, Heraclides
ordered a confidant to substitute a snake for his corpse upon
the bier, whose appearance would lead the onlookers to suppose
that the dead man had become a hero (cf. Cumont 394); but
the fraud was discovered. Here it looks as if Heraclides' own pro-
pensities-his interest in oracles, the punishment for impiety,
the conferment of divine honours upon men, even his suggestion
of deception in the attainment of the latter (cf. Empedocles)-
have been turned against him: the fiction has been built up out
of hints gleaned from his own work, even as he himself had
built up such fictions out of hints gleaned from the work of others.
Even the forged oracle in this tale may be an extension of the
reputation which Heraclides had for the perpetration of literary
forgeries. For not only did he invent works whicl1 he ascribed to
fictitious personages [170], but he forged 'plays ofThespis' [:w],
and at least two oracles which he quotes are highly suspect. s
We have already seen reason to believe that the fragments of the
Arimaspea, if not genuine, are fourth-century forgeries [20J: in
Heraclides we may have their author.
174- ARISTEAS PrTHAGORICUS
It is clear that any evidence about anything that can be traced
to the authority of Heraclides must be treated with the greatest
reserve. It is also clear that it cannot be peremptorily dismissed
as sheer invention, for the elements out of which he moulded his
fancies were often provided by earlier literature or tradition. The
difficulty is to distinguish these elements from the peculiar twist
that he has given them. We are not, for instance, entitled to con-
clude that Aristeas had no connexion with Pythagoras before
Heraclides; indeed, there is reason to suppose the contrary.
A reader of Herodotus' story about the appearance of Aristeas'
ghost at Metapontum must wonder about both the message the
apparition gave and the choice of messenger. Why did he say
that Apollo visited the Metapontines alone of the Italiotes, and
why should it be Aristeas who was selected to announce this, an
antique poet from an island on the other side of the Greek world,
having no connexion with Metapontum? It is a type of story to
appeal to the Pythagorean mind4 6-the still-living soul of Aristeas
reincarnated in the raven and accompanying Apollo, the favourite
god of Pythagoreanism. 47 It will not be forgotten that Pythagoras
ended his days at Metapontum, after he had left Groton ; and at
the latter place he had been dubbed, according to Aristotle
(ap. Ael. V.H. ii. 26), 'the Hyperborean Apollo'. Why Hyper-
borean? We may have here an important clue: for the Hyper-
boreans, the special people of Apollo, figured in the Arimaspea.
What is more, they were the prototypes of an important side of
the Pythagorean way of life, for Hellanicus, in a fragment which
has signs of deriving from Aristeas' poem, said that they 'prac-
tised righteousness by abstaining from meat and living on fruit'
[71 f.]. It is, then, possible that the answer to all these questions
is this: that Pythagoras, who particularly reverenced Apollo, the
god of music, was also specially interested in the Arimaspea, whose
author had clearly been a favourite of the same god, having been
possessed by him. In the poem it was stated that a feature of the
righteousness of the Hyperboreans was their vegetarianism; it
would follow that vegetarianism must particularly commend
itself to their god, and so Pythagoras adopted it.48 From his
support of his tenet by appeal to this authority, and from his
I'
!..
/
AR/STEAS PTTHAGORICVS 175

demeanour as of one more than human (cf. Aristotle ap. Iambl.

V;P. 31), he came to be called by the Crotoniates 'the Hyper-
borean Apollo'-whether seriously or in jest does not matter: it
was taken seriously by his disciples. After the move to Metapon-
tum from Groton it occurred to them to try to win countenance
for their sect by introducing the idea there that their leader was
the embodiment of Apollo, who thus favoured the Metapontines
alone with his presence. The introduction was engineered by
means of a 'supernatural revelation'; and what better instrument
could be chosen to make such a revelation than that poet who
had been beloved by both Apollo and Pythagoras, who had him-
self sung of the Hyperboreans, Aristeas of Proconnesus ?49
 VIII
CONCLUSION

IN the survey of possible debtors to the Arimaspea I said that

primary clues would be provided by what Herodotus tells us of
the poem's contents: mention in other authors of one or more of
the lssedonians, Arimaspi, griffins, and possibly Hyperboreans,
would make a significant context. In this way, by interlocking
one with another, a number of passages were brought under
review. Moreover, besides establishing an upper terminus for
the poem (the Cimmerian invasion of Asia Minor), the resume
of Herodotus also revealed an important fact about the poet: his
veracity. He gave an historically true account of the westerly
pressure of the peoples of the steppe, and his one-eyed men and
gold-guarding monsters were actual creatures of central Asiatic
folk-lore. Claims, then, for borrowings from the Arimaspea should
fulfil two requirements to be acceptable: their matter should be
in, or at least near, a significant context in the sense outlined
above, and should reflect historical truth about, or the actual
folk-lore of, the Eurasian and central Asiatic regions. It remains
to summarize the results of these tests.
The Rhipaean mountains. Damastes repeats Aristeas' order of
peoples from the Scyths to the outer sea, omitting the griffins.
Between the Arimaspi and the Hyperboreans he places the
Rhipaean mountains, always snow-bound, from which Boreas
begins his course. They were incredibly high, according to
Aristotle. Pliny puts this range near the Arimaspi, in the region
called 'Featherful' (Pterophoros) from the perpetual snows; the
North Wind has his origin from a cave there which bears his
name. According to the pseudo-Plutarch the so-called Cave of
Boreas was by the river Tanais. Caves of the Winds appear in the
folk-lore of northern Asia, and there was a famous one near the
Dzungarian Gate.
There are indications that this range was not originally called
 CONCLUSION 177

'Rhipaean' but 'Caucasus'. The pseudo-Plutarch says that the

Caucasus was at first called 'the Lair of Boreas'; Aeschylus, in a
significant context, puts his-immensely high-Caucasus in the
place normally assigned to the Rhipaeans; later geographers
sometimes identified the Rhipaeans with the Caucasus, or held
that the former were a continuation of the latter; the source of the
Tanais was located in both.
The Region of Feathers. This, associated by Pliny with the
Rhipaean mountains and Arimaspi, was reported by the Scyths
and their neighbours (Issedonians ?) as hindering sight and
progress, according to Herodotus. As it is also put by an early
Chinese source in central or northern Asia it must have been a
piece oflocal folk-lore. The fact that it was impossible to see there
may underlie descriptions of the Rhipaeans as murky or benighted
(e.g. Sophocles, Pliny), and account for the strange mis-statement
of Hippocrates, that Scythia is perpetually fog-bound (in a
passage where he speaks of the Rhipaeans in language like that of
Damastes).
The Arimaspi. They are said by Herodotus to steal gold from
the griffins. For Aeschylus they are one-eyed horsemen living by
an auriferous river; Dionysius also places them by a river, in
company with the griffins. In art the opponents of the griffins
are sometimes depicted as horsemen. Robbers who seize their loot
and flee on horseback occur in old central Asiatic story, and the
Arimaspi might have been some of the first Mongols to learn to
ride. Gold is found in plenty in some parts of central Asia-even
in rivers, though Aeschylus' Pluton might only 'flow with gold'
because it was used for washing the ore.
The griffins. Pausanias actually says that the Arimaspea described
them as a mixture oflion and eagle. As such they were a Mediter-
ranean conceit, though the old Chinese tell of a 'winged tiger'
in the neighbourhood of the one-eyed Kuei. Central Asiatic tales
ascribed to monstrous ants the role of Aristeas' griffins; they
apparently extruded the gold as they made their holes, as the
griffins are said to do by Pliny (Dionysius says they dug it up).
Gorgons and Graeae. Aeschylus, in a significant context, locates
them strangely, apparently in the north-east of the world (Pindar
686& N
178 CONCLUSION
puts the Gorgons near the Hyperboreans): they live in darkness,
and the Graeae are 'swan-shaped'. Central Asiatic lore knows
of fierce and ugly 'swan-maidens', who live in Stygian darkness
and can fly.
Amazons. In the same context, Aeschylus places Amazons oddly,
by the Sea of Azov. Amazons sometimes replace male warriors in
artistic grypomachies. Actual sociological features of aboriginal
tribes could explain their location in Scythia.
Cannibals figure in significant contexts in Pliny and others and
are mentioned among Scythian peoples by Herodotus. The one-
eyed men themselves are said to be cannibals in a Ukrainian tale
about the Tatars.
The Issedonians. These were not on the north-easterly trade-
route, and Aristeas is the only Greek we know who claimed to
have reached them. It may be assumed therefore that Herodotus'
account of their customs comes from the Arimaspea. This account
need not have been an invention; in fact it is very unlikely to have
been, for, strange as Issedonian manners would have appeared
to a Greek, they all have historical parallels. Herodotus' state-
ment that they 'faced' the Massagetae may not be from the poem
(there is no reason to suppose that his other statement, that they
were to the east of the Argippaei, was), though his poetic lan-
guage here may point to it. According to Zenothemis (supported
perhaps by a muddled remark of Aelian), they lived by a river.
Aleman spoke of the Issedonians (as well as the Rhipaeans,
dark and forested); he, therefore, together with the Kelermes
grypomachy, gives us a lower terminus for the Arimaspea.
One of the doublet stories which Herodotus tells about the
Scythian prince Colaxais and his two brothers may come from the
Arimaspea, as Aleman also seems to have known the name
(Colaxes ?) . (It may be remarked that this type of folk-tale, of the
successful youngest brother and his two unsuccessful elder
brothers, though common among some other peoples was foreign
to ancient Greece.)
I would claim the Phasis-Tanais and the vegetarianism of the
H yperboreans for the Arimaspea on other grounds : the former
because Aeschylus so identified them in the Prometheus trilogy,
 CONCLUSION 1 79
which seems to have contained so much from this early poem (and
such an identification could only have been made very early) ; the
latter, by linking a fragment ofHellanicus with the use made of
Aristeas by the first Pythagoreans. There may be more matter
from the Arimaspea lurking in other passages I have discussed in
the course of this book ; but ifso, I do not see how it can be isolated.
I end with an hypothesis which embodies my main conclusions
and, I think, 'saves the phenomena' economically. I beg the
reader to forgive the categorical tone of this, which is not intended
to reflect the certainty of revelation.
Aristeas the son ofCaystrobius flourished in Proconnesus in the
third quarter of the seventh century before Christ. He was de-
voted to Apollo, with a fervour more commonly felt by the initi-
ates of Dionysus for their god. At this time the Greeks were
reconnoitring the west and north coasts of the Black Sea, 'beyond'
the north wind Boreas, whose home was popularly thought to be
in Thrace. It must have seemed that they were penetrating the
regions of the blessed Hyperboreans, the favourites of Apollo,
and when news of such an expedition from Cyzicus reached Pro-
connesus Aristeas felt impelled to join it in order to find d1.e
country of his god. This urge was so compulsive that it took
the form of a psychosomatic disturbance which to his fellow
citizens seemed to be a death and resurrection, and to Aristeas
himself to constitute actual possession by Apollo.
Aristeas' goal proved to be more distant than he had expected.
The home ofBoreas was not in Thrace after all; but, undaunted,
Aristeas landed, perhaps in the neighbourhood of the Sea of
Azov, and headed inland, determined to find it. But now he had
no bearings, and no way of identifying Boreas save by its force
and bitterness; so the course of his slow progress came to be
directed into the teeth of the east rather than of the north wind.
This reorientation would have been aided by the fact that the
Scythian tribes, on whose hospitality he must have depended,
were themselves still strung out over the Eurasian steppe in the
act of migration, and would naturally pass him back along the
way they themselves had come. As he went, he seemed to
recognize features of legend that he had heard at home--the
180 CONCLUSION
Phasis of Argonautic story in the Tanais, Amazons in the mascu-
line ladies of the Sauromatae. His Scythian hosts also told him of
the patriarch from whom they were descended.
At last, in the region of the head waters of the Ishim, he passed
from the Scythians to another people, the Issedonians. Among
them he sojourned for some time, and from them he learned news
of what he sought. They told him that beyond them were lofty
storm-wrapt mountains-the western outliers of the central
Asiatic highlands-which they called by a name Aristeas inter-
preted as 'Caucasus', whence a furious wind issued from a cavern.
This cavern he took to be the source of the 'blasts of Boreas',
beyond which the Hyperboreans must live their idyllic life; and
when his informants also told him of the settled civilization of
the cereal-eating Chinese, whose territory extended from beyond
the mountains down to the sea, he thought they were speaking of
these very Hyperboreans.
Almost within sight of success, however, he heard also of
dangers and terrors ahead. The lssedonians said that they had
been pushed away from the mountains by the Arimaspi, whom
they described as horsemen having only one eye. These people
now dwelt about an auriferous river, and purloined the gold of
those parts from the guardianship of swift and fierce monsters.
Whether the name 'griffin' given to these was Issedonian or
not, their shape was Greek, conferred upon them by Aristeas
himself from his memory of half-leonine, half-aquiline monsters
he had seen depicted by artists at home. Similarly also, when he
was told of other monsters living in the region of darkness,
hideous and murderous 'swan-maidens' who could doff their
feathers and become hags, Aristeas identified them with the
Gorgons and Graeae of Greek legend. The list of horrors was
capped with tales of cannibals (though these may have been
Scythian rather than Issedonian lore) and a land where ever~
falling feathers hampered sight and progress. Aristeas decided
to turn back.
Of the Issedonians themselves he noted that they were a
nomadic (or semi-nomadic) people, whose women did a man's
job, and who venerated their fathers to the extent of eating them
 CONCLUSION 181

when dead and preserving their skulls, encased in gold, as cult

objects. He may have added that their (southern) neighbours
were the Massagetae.
On his return to Proconnesus more than six years after he had
set out, Aristeas related his experiences in a hexameter poem,
which Greeks later called the Arimaspea. In this he faithfully dis-
tinguished his own first-hand witness from hearsay, and also
reported accurately the westerly drive of the peoples of the
Eurasian steppe which had already caused the Cimmerian in-
vasion of Asia Minor. 1 After this he disappeared a second time
from Proconnesus, perhaps on another expedition to the land of
the Hyperboreans, from which there was to be no homecoming;
it is pleasing to fancy that this time his persistence was re-
warded, and that he ended his days in China!
The poem quickly contributed to literature, art, and geography.
Through it Scythia became known as the home of griffins, and
Greek artists used them, and their rivalry with the savage Ari-
maspi, as a decorative motif for wares destined for the Scythian
market; whence native Scythian artists then borrowed the griffin.
Aristeas' identification of the Argonautic Phasis with the Tanais
did not hold for long, but explorers who knew the Arimaspea and
were looking for the lofty range of Caucasus in that area thought
they had found it in the high mountains bordering the eastern
shores of the Euxine, which as a result were to bear the name
thereafter.
A Hesiodic poet portrayed the griffin, after the Arimaspea;
and from it Aleman learned of the gloomy, wooded moun-
tain of the storm-wind, of the Issedonians, and of the Scy-
thian Colaxes. The Scythian Amazons reappear in Pindar,
Aeschylus, and Euripides, and a later Greek attempt to account
for their presence there is preserved by Herodotus. It is these
Amazons who sometimes replace male warriors in the grypo-
machies of vase-painters. Aristeas' unorthodox location of the
Gorgons in the neighbourhood of the Hyperboreans is repeated
by Pindar and Aeschylus. His Caucasus, home of Boreas, which
was not heard of again by the Greeks until the time of Alexander,
suffered a dichotomy at the hands of Ionian geographers, on the
 CONCLUSION

one hand being mistaken for the range which still bears its name,
and on the other being slewed round to the proper position of
Boreas, the cardinal north, to form a non-existent 'Rhipaean'
range spanning northern Europe and Asia from west to east. The
effect of this can still be seen in Aeschylus' curious position for
his Caucasus, in the pseudo-Plutarch's statement that the Cauca-
sus was once called 'the Lair of Boreas', and in ancient theories
that the Rhipaeans were a spur of the Caucasus.
If 'Longinus' has preserved for us a genuine fragment of the
Arimaspea, Aristeas also told of a sort of mermen leading a
miserable existence in the waves of the sea; but the authenticity
of this as of the other professed quotations from the poem is still
debatable. The work was yet current in the fifth century B.c., but
may soon have disappeared: it does not seem to have been known
to the Alexandrian librarians, and Latin authorities, though they
know of Aristeas, do not name his poem or even give any hint
that he composed a poem at all. We need not suppose that anyone
ever produced another, spurious, Arimaspea in its entirety: the
fragments we possess may be no more than faked 'quotations'
from it in a prose work-perhaps by Heraclides Ponticus in the
fourth century n.c., who practised literary forgery and was in-
terested in Aristeas, Hyperboreans, and the bizarre.
Even so, at least those fragments preserved by Tzetzes may
embody matter derived from the genuine Arimaspea through the
mediation of Ionian geographers; and it is probably thanks to
such intermediaries that Hellenistic and Roman writers occasion-
ally reproduce matter from the authentic Arimaspea that is not in
Herodotus' summary.
The poem was known to another devotee of Apollo, Pythagoras
of Samos. From it he learned that Apollo's righteous Hyper-
boreans were vegetarians, and so adopted as a religious tenet
a practice so acceptable to his god, and earned himself the nick-
name of 'the Hyperborean Apollo'. After his enforced move from
Groton to Metapontum, his followers, to win esteem for their
master's teachings in the new place, put about the story of
a divine revelation that the Metapontines had been especially
favoured by a visitation of the god, and chose as the vehicle for
 CONCLUSION
this revelation the ghost of Aristeas, Apollo's possessed and
Pythagoras' own authority for one of his most striking injunc-
tions.
This early connexion of Pythagoras with Aristeas was ex-
ploited further in the fourth-century Academy, by Heraclides
Ponticus, who moulded a number of earlier misty 'mystery-men'
to a set pattern of friendship with Pythagoras, wonder-working,
reincarnation, divination, and trance. In the trance state, when
their bodies lay as if dead while their souls wandered forth, these
people gained preternatural knowledge not only of far parts of
the earth but of divine truths. The inspiration for this idea of
Heraclides came from his master Plato, but its particular applica-
tion to Aristeas was no doubt suggested to him by Herodotus'
story of Aristeas' apparent death, resurrection, and subsequent
ghostly reappearance far from Proconnesus.
It was under the influence of the popular Heraclides that for
later generations Aristeas became a 'theologian', and the e..'Ctra-
ordinary journey he had actually made in the body became a
mythical journey that he made in the spirit; and his association,
whether as ghost or reincarnation, with Pythagoras in the same
author's imagination caused his date to be given in the sixth
century n.c. The reputation of the old wizard of Proconnesus
lingered on in nearby Byzantium until the fourteenth century; it
may still linger somewhere today, connected as ever with strange
things, for the ship which in 1938 fished up a living coelacanth,
as it were from the Mesozoic era, was called Aristea.

I
j
I

II
 NOTES (pp. 1-14)
CHAPTER I
1. The translations of Herodotus in this book are based on that by
Rawlinson (Everyman ed.).
2. Stephanus Byzantius, s. 'lcr<n)OOV*S, who may owe this information to
the work of Alexander Polyhistor (first century u.c.) 1r,;pl TWV 1rap'
l1>.K.6.v, 'TO'lrLKWS lp7J.lvwv (or lu-rop7J.lvwv)-cf. s. l1pa~ai, l1uuos.
3. If the form used by Aleman were really l1u,;oov,;s or l1uu,;oov,;s (vid.
Meineke's app. crit. to Stephanus), or 'Euu,;oov,;s (Schneidewin, Coniectanea
Critica 28), this theory could hardly be maintained. But it may in fact have
been l1u<n)OOV<tS (Lobel in The Bodleian Quarterly Record iv ( I 923), 48: Pap.
Ox. 1611 appears to give }fo<n)8wv as an alternative form of' lu<n)Owv, cf.
Herodian, 1r<tpl .0V1Jpo11s Mfws a' 917). It may be just worth noting that the
places apropos of which Stephanus mentioned the work of Alexander named
in note 2 both begin with A, and the other strange places which Stephanus
says occurred in Aleman also begin with A-l1vvlxwpov and J!p11/3/3a. There
are few other references to Aleman in Stephanus; perhaps for some reason
he knew only that part of Alexander's book that dealt with names beginning
with A. The Romans spell this people with an initial E (Val. FI. vi. 750
perhaps excepted), the second syllable being long in the poets.
4. I am not completely convinced that "17.fv in Od. i. 10 (-rwv a..o8Ev ye,
80:f, 8Jya.,.,;p Llt.6s, l1rJ Kai ~11-fv) is not the equivalent of .o, in I. I. The Kal
in I. 10 might be taken (i) with TWV (=the companions of Odysseus)-
'tell us [poet and audience] of these men too [as well as of Odysseus]': but
this seems an incredible postponement; (ii) with ~.,v-'tell us too [as well
as other poets and audiences] of these things (Twv)': but a.11-09ev-'taking
up the tale from any point you like'-implies that there was a corpus of the
adventures of Odysseus already in existence, and the poet could hardly thus
assume that his audience was ignorant of them; (iii) with "17.fv-'tell me too
[as well as other poets who have sung of Odysseus] of these things': this
seems to me the most natural, but the lack of certain parallels counsels
caution.
5. I wish I could be sure that these curious people were not the same as
the Sciapods, who lay on the ground and shaded their bodies from the sun
with their single huge feet as with a parasol (Ctesias ap. Harpocr. s. EKufao-
8Es j Plin. N.H. vii. 23).
6. ll. i. 53-58, x. 465-70, xxii. 331-6; Od. i. 126--31, xix. 4-g, xxiv. 57-62;
Hcs. Op. 83-88, 618-23, Scut. 261-6; Call. H. iv 28-33, fr. 260. 62-67; Ap.
Rhod. i. 569-74, iv. 1318-23; Simias, fr. 1. 1-6, fr. 6; Delphic oracles ap.
Hdt. vi. 86 y 2. 1-6, vii. 140. 2. 1-6.
7. Epicharmus lv 'Hpa.KAEt -r<Fi 1rapa. IPo>.tp (fr. 7. 2) has the earlier sense:
"
oio.ai o~, 011ons
>t- \ ~ \
<tKwv '
1roV7Jpos ovo't'' aTav
,, ,,
EXWV. so too Acsch., fr. 401 S1dg
.
wick, though the fragment is suspect (the sense of I. 1, {011s 1rov11 pas 8&.vaToS
alpETwupos, may have been rather that of fr. 90, f3lov 1roV7Jpoii OavaToS
 NOTES (pp. 16-23)
f!iJKAE(<rrEpos but glossed by II. 2 f. to give it the meaning of, for example, Eur.
Hee. 377 f.). The meaning of 'grievom' assigned to the word at Ar. Pl. 352
(roVTi 'TTOIIT/pov <{,alvE-ra, -ro <{,op-r(ov) ill suits the context, which requires
something like 'here's a load of mischief', 'a bad lot' (cf. Eur. /. T. 1306).
8. As it happens, we are able to witness the digestive process applied to
the very fragment of the Catalogus under consideration, in the account of the
lm'TTW).'Ju,s of the herds of Augeas in the 'HpaKMjs ).Eov-roc/>ovos ([Theocr.]
xxv. 85 ff.). Here we meet O.'TTft['fiGlwv in I. 100, 'TTOAvppTJVES in I. 117,
a.cpvEtov .~).o,s in I. 119, and d).m63GGtv in 1. 13 I.
g. It is odd that nearby are two words recurring in fr. 4-avOpw'TTwv and
a.cf>vf!l6-ra-ros (//, xx. 217; 220). Does this mean that frr. 3 and 4 were in fact
closely conjoined? Dion. P. 953 is also related to II. xx. 222.
JO. Hesychius has a gloss 'TTWKaAtTat g')palve-ra,. ~ &.v-rt TOV t71-rEi-ra,.
)tp,a.,..ias. Onions ( Origins of European Thought 31) suggests that this may be
the Proconnesian or the grammarian: the fragments of the former (if they
are his) do not show any propensity for rare words, and why should Hesychius
quote this unmetrical form? The latter seems more likely (for him cf. Varro,
L.L. x. 75).
11. Frankel (320, n. 6) thinks that the repetition might be deliberate; but
what the point could then be is obscure.
12. It may be evidence of oral composition: cf. Dodds on 'decayed'
epithets, in Platnauer's Fifty rears of Classical Scholarship 14 f., 34. If so, it
would argue for the genuineness of the fragment.

CHAPTER II
I. The authority who reports Heraclides' fraud was a contemporary,
Aristoxenus, a malicious man (cf. fr. 67), who, however, may here be telling
the truth [VII].
2. Galen, in Hippocr. de nat. horn. (xv. 109) (where we should read
[~] 'TTpt -ros tmypo.rf>as?). Juba II was also duped: David. scholl. Aristot. 28a
Brandis, quoted by Mullach, Philosophorum Graetorum Fragmenta i. 411.
3. Consolatio: no. 283 N. (fr.phil. ix M. forfragmentsofthegenuinework).
See J. A. Farrer, Literary Forgeries 5 ff. <Po,vtKtKa.: Farrer 191. If the Cu/ex
extant in the first century A.D. was not the poem we now possess we should
have another example; but this is quite uncertain.
4. My supplementation was published by C. A. Trypanis in his Loeb
edition of the fragments of Callimachus (98 f.).
5. Pfeiffer in his edition ofCallimachus (on fr. 186. 12) regards it as uncer-
tain whether Callimachus did in fact identify the two peoples. But surely H.
iv. 291 ff. proves it: there either the 'daughters of Boreas' must be Arimas-
pian, or the Arimaspi must stand for the Hyperboreans. The former would
be hard to maintain in the face of the names and parentage of the girls (cf.
Hdt. iv. 35 (Opis); [Plato], Axiochus 371a (Opis and Hecaerge); Serv. auct.
ad Am. xi. 532; Hee. Abd. ap. Diod. Sic. ii. 47. 7 (descendants ofBoreas
held hereditary sway over the Hyp('rboreans)).
6. Not stated by the scholiast on Ar. Pax 1270, as Schmid-Stahlin ass('rt
(i. 205).
I 186 NOTES (pp. 25-32)
7. Meuli ( I 56 f.) thinks more highly of the value of this tradition: for him
it is evidence that the Arimaspi are one-eyed infernal beings of Altaic-
Turkic and Mongol myth, beings, like the Hyperboreans,not of this world:
'offenbar ist es sehr gute gelehrte 'Oberlieferung, wenn Antimachos, Kal-
limachos und andere die Arimaspen den Hyperboreern gleichsetzen, und
filr die endgilltige Deutung der Arimaspen wird man dies beachten milssen';
griffins ( = dragons defending the Golden Mountain of God) and Arimas-
pians are genuine products of Scythian folk-lore, 'die nicht in goldsand-
haltigen Wilsten oder in Gebirgen mit Goldgruben wohnen, sondern in
einem mythischenJenseits, wie die Hyperboreer; einemJenseits, in das der
Schamane zu reisen pfiegte'.
That we are dealing here with traces of genuine 'Scythian' folk-lore I
have no doubt; but I doubt very much if the Hyperborean-Arimaspian
tradition goes any way to proving the supernatural status of the Arimaspi
in that folk-lore. The important point for my present purpose is that this
tradition does not derive from (the 'shamanizing' !) Aristeas (who plants his
Arimaspi very firmly on this earth, calling them avOp=oi (fr. 4. 1; av8pes in
fr. 1), who press upon their neighbours the Issedonians).
8. It has been suggested that de subl. 10. 5f. is from Posidonius' avyl(ptats
'Op.17pou l(ai }tp&:rou (Schmid-Stiihlin ii. 477, n. 4). lfso, and if 10. 4 (which
is closely connected with 10. 5-cf. Op.ev ya.p 'Jl"Ot1JO'a~ 0 8e ~op.'f/pOS)
is carried with it, it merely takes our problem one stage farther back.
9. Abominable Snowmen? In Pliny's time Imaus ('Snowy', cf. N.H. vi.
64: this is correct) was the name given to the easternmost portion of the
Himalayas (cf. Strab. 519). Everest expeditions of recent years have heard
and told of the swiftness of the Snowman, and the reversal of his feet (he
has thus acquired a common characteristic of supernatural beings in the
East)! These silvestres homines are the o.ypwt o.v6pw1rot of Megasthenes ap.
Strab. 711.
10. Cf. Hdt. iv. 64 f.; Anon. Vat. 49.
11. Cf. Antonius Diogenes (508 = Phot. Bibi. 109b).
12. Anon. Vat. 50.
13. A similar story in Fife not so long ago (Coun!, Folklore vii. 11) !
14. On the strength of this chapter of Gellius a predilection for the
marvellous has been ascribed to the Magnesian Hegesias (e.g. in Schmid-
Stahlin ii. 207; Smith, Dictionary of Greek and Roman Biography ii. 368). It may
be so, but I see that an Agesias of Megara is quoted by the Vatican paradoxo-
grapher (1). Is this really our man?
15. Cf. Sen. Ep. 90. 16.
16. It could underlie the whole of the second chapter of N.H. vii, in spite
of the references therein to Cicero and Varro (13; 18; cf. 19): Isigonus used
Varro (Schmid-Stlihlin ii. 420).
17. Eudocia (or whoever the author of the Violarium was) may have
been led to alter Ka, KaTaAoy&~v 8oyovlav ds l'TfYJ p. to 1(0.! 8toyovla.v 1r&vu
l(aAws through puzzlement at the conjunction of l(a"TaAoy&8'1/v, 'in prose',
with l'll"'f/, a word usually used of hexameter lines (for its use of prose lines see
Cronert 127).
18. I agree with Jacoby (i. 519): 'dall Dionys noch eine prosaschrift des
 NOTES (pp. 36-41)
Aristeas besal3, ist nicht zu entnehmen.' It is not clear from Clement's
' '
M1\'f/UO.yopov ' Kl\'1'V
ya.p ' .,. 0' n '
poKOVV'fjUIOS B',wv, OS'
KO.L, 'TO., .nao.ov
u 10
'TOV 1Ta./latov /J,E'TypaipEV K<pallatov.VOS (Strom. vi. 26. 8) whether the
abridgement of Cadmus by Bion was a frank summary or an intended
forgery. In the context it looks as if some dishonesty is imputed to Bion; if
so, it need only mean that Dionysius here is giving one example from each
of his two categories-writings by Cadmus have survived but are of doubt-
ful authenticity, those of Aristeas have perished.
19. Nissen (299) thinks that Gell. ix. 4 is evidence that there were yet
other books of marvels besides the Arimaspea passing under the name of
Aristeas; but the only evidence that Aristeas may have been the author of
anything other than the Arimaspea and the (improbable) Theogany is the
dubious T. 28 (and what the 'spurious poems' of Peisander there men-
tioned may have been about we have no idea).
20. The list does not profess to be exhaustive.

CHAPTER III
1. The only exceptions are fr. 3 of Aristeas himself and a line of Zeno-
themis which may well be based on something in the Arimaspea [68]. Mela
(ii. 9; 13) compounds Hdt. iv. 26 and 65.
2. L. & S. 8 are surely wrong in preferring pma,
here, and interpreting
'[the quarter of] the night storms'. SeeJebb's note.
3. Cf. Plin. N.H. iv. 89; Ant. Diog. (5 IO = Phot. Bibi. 11ob-11a). Both
contain errors which should not be imputed to the scientists, represented by,
for example, Cleomcdes i. 7; Geminus, El. astr. 6 (Geminus here gives the
explanation of Od. x. 82--86 by Crates of Mallus: Homer places his Laestry-
gonians near the Arctic circle where~ llvms ,rapaKE'i'Ta, -rfi ava-rolli). That
Homer shows here some inkling ('folk-memory'?) of conditions in the far
north has been accepted by some modern scholars (Minns 436f.); but
difficulties remain, and it may be that Laestrygonia is a purely imaginary
country enjoying a perpetual twilight).
4. The possibility can almost be discounted that Alcman's fragment refers
to some mountain in Arcadia, near the Rhipe of JI. ii. 6o6. This place was
unknown to the ancient commentators (Strab. 388; Paus. viii. 25. 12; cf.
Page 122). The scholiast on O.C. 1248, who quotes the lines of Aleman
and presumably knew their context, does not take them so, though he
must be mistaken in placing the mountain in the far west (a deduction
from Jvvvxtat).
5. Either both poles or the north must be meant. The ancient notion that
the north pole was higher than the south is well known; but it would not be
natural for a Roman to confine to the south pole a description applying
equally well to either. Ima here has the sense of extrema: cf. Varr. Atac., fr.
14. 1 f. 'at quinque aetheriis zonis accingitur orbis, ac vastant imas hiemes
mediamque calores'; and Lucan's own use of in.fima at viii. 464. lacet, while
playing upon ima, has a metaphorical sense, 'lies torpid and useless'. I say
this because modern scholars have tended to follow Burmann in under-
standing this passage to refer to the south pole.
188 NOTES (pp. 42-46)
6. Not all scholars are agreed on this. Some hold that Anaximenes con-
ceived of the whole plane of the earth's surface as sloping from N. to S. This
can only be reconciled with Anaximenes' other theory, that the earth floats
on air like a lid (Diels-Kranz i. 94. 24 '1TL'1TWa.Tltw), if he thought of the
earth as a cross-section of a cylinder with the upper surface bevelled. Such
an interpretation is obligatory if Anaximenes really held that the motion of
the heavenly bodies is lateral in one plane, like that of a millstone, &c. (ibid.
i. 92. 16 ff.; 93. 20 f.). But the analogies may be a misunderstanding by
later doxographers; for the visible paths of the heavenly bodies appear to
be far from lateral, except to dwellers in very high latitudes. I agree with
Kiessling that for Anaximenes the heavenly bodies describe their orbits in
two planes-in a more vertical one above the horizon and in a lateral one
below, or more properly along, it. According to Kiessling the sun is masked
at night by the high outer rim of the earth's disk, though he allows that
Anaximenes may have reinforced his argument by calling in the aid of the
Rhipaean mountains. Though agreeing with much of Kiessling's impressive
exposition, I think that he, in common with the supporters of the 'sloping
earth' interpretation, credits Anaximenes with more sophistication than he
merits. See Zeller-Mondolfo, La Filosofia dei Greci I. ii. 219 f.; P.-W. iA.
848-50. The (or some) Epicureans appear to have suggested obscuration
of the sun by a northern mountain range as the (or more probably, after their
manner, one possible) explanation of night; Anaximenes need have been
no subtler (Avien. Or. Mar. 644-73; the sense seems clear in spite of an
unfortunate lacuna in the text. The passage is discussed by Kiessling,
P.-W. iA. 852 ff. Kiessling {ibid. 850 ff.) would also ascribe a similar view
to Heraclitus (fr. 120)-unconvincingly, I think).
7. TU. 3i KaT!mEp9E '1Tpos {Jophiv O.flE/J.OV TWV tnrEpalKWV rijs XWPTJS.
How and Wells seem to me to be wrong in treating tnrpalKwv as possessive--
'lands that lie above to the north, of those who dwell in the upper parts
of the country'. I take the tnrepotKot to be the same as the '7TEplo,Kot of
c. 31.
tnro
8. '1TTEpwv KExv.evwv. The verb combines the ideas of both falling
and heaping up.
9. This was suggested by A. A. Blakeway (C. M. Bowra, Greek Lyric
Poetry 66).
Aristeas' form of the name might have beenKo,\atos (Hee. Mi!. said there
was a branch of the Colchians called Kopatol-Steph. Byz. s.v.), but perhaps
Ko>ta67s is more likely: a Scythian prince Colaxes appears in Valerius Flaccus
(vi. 48 ff.), descended from an lxwva /J.~0'1Tap9EJIOS like Scythes, who corre-
sponds to Colaxais in the Greek version of the Scythians' descent given
by Herodotus (iv. 8-rn). Either Ko>.atos or KoMEas could give Alcman's
Ko,\aEaios (Schwyzer, Griechische Grammatik i. 467).
10. The phrase recurs, in a different sense, in Eur. (Phoen. 1058).
11. Cf. Wilamowitz, Aischylos-Interpretationen 156 f. (I take it that Wilamo-
witz means we must not imagine Aeschylus 'reaching for' his Arimaspea (or
any other authority) when he d;ci~e~ to write of 'Scyt~ian' geography,
rather than as Meuh ( 154, n. 2): be1 A1sch. Prom. 803 ff. 1st nach Wilamo-
witz nicht Aristeas als Quelle anzunehmen'.)
 NOTES (pp. 48-54)
12. '1Tp{jyro,; is surely a mistake in the papyrus for 11pw,-ov-'H. says that
Saneunus was the first to discover', rather than 'H. was the first to say'. It
looks like an extract from one of those lists of discoverers or inventors
popular from Heraclides Ponticus on. Scythians as inventors of bronze: see
Jacoby's note on this fragment of Hellanicus. Jacoby says that 'Scythian
iron' is not unusual, though he refers only to Sept. 817 (= 818 Murray).
His qualms because Herodotus (i. 215) says there was no iron in the
country of the Massagetae seem unnecessary: it was true, but Scythia was
far away.
I 3. I follow Bothe here in printing with a capital: cf. the other named
rivers Pluton (806), Aethiops (809), Nile (812). And ov ,pEv8wvvp.ov loses all
point if we are not told what the name is to which it refers (cf. 85). This
punning figure was liked by the Greeks (cf. Arist. Rhet. ii. 23); uf3pun71v
would tum it into a riddle! So the scholiast took it, and provided an
answer: 'TOY 14.p~v, '1Tapo. 'TO apo.<1<1ELV Ka& 11xerv'TO. JOJp.a'Ta aV'TOV ccf. Strab.
531; Steph. Byz. s. 14.p&.g..,,). That he was guessing is indicated by his pre-
fixing the article; but ifhe really read 14.p&.g'l)v it would be interesting (the
Araxes flows into the Tanais and rises in Parnassus (i.e. Paropamisus = Indian
Caucasus), according to Aristotle, Meteor. i. 13).
14. The epithet EM1T7rov,; has a Scythian or northerly connotation: cf.
the 'Y11Epf36pw, evm110, of the Calalogus [23 f.] (Hyperboreans are equated
with Scythians by the sixth-century Ionian Ananius (fr. 4. 3)).
15. VO'l)T0V 8e viiv, <{,aut, KavKaaov 'TfJ-fip.o. .,., 'TOV 11postp'l)fJ,VOV Tavpov
Q ,
tJOpEtO'Ta.'TOV, 11.p,\ 'T'l)V
' K pov,a.v
I
aV'l)KOV
, - 8.!.'
W\a.<1<1a.v, ov?' fJ,E(YTJ
I
Ka,\ Ta\ 11po\
'TOVTWV ypa.</,IVTa 'Pma,a op71. It is not clear whether 00 refers to KavKa-
uov or Ta.vpov: the latter seems indicated by Kal (the Rhipaeans as well as
the Caucasus are a far-northern spur of the Taurus-hence also 'Tf'-7Jp.a. -r,).
Yet cf. schol. ad Dion. P. 666: -rov KavKauov '"P -n,v Kpovi.a.v BaJ..aauav
inro-rl8Ta,, Kai pTJ aV'ToV T(i. 'P,11aLa ~oUAETa.1. Elva,, 1eal T0v KalJ,caaov
-r.fj.a. Elva, Toii Ta.vpov (= Eustathius?); Mela i. J09; Pliny, N.H. vi. 15
(the Rhipaeans are an extension of the Caucasus).
16. In the codices of Dion. Hal. (Ant. Rom. i. 61. 1) Atlas is said to have
been an Arcadian king who dwelt 1rEpl TO AEyop.EVOV Kav,afuwv opos. Is this
something more than a mere misreading (of Ka.vKwv,ov? so C.Jacoby), and
should it be connected in some way with the 'Hyperborean Atlas' (Apollod.
ii. 5. 11)?
17. For example, 14.{Jaulas Freshfield (supposed to mean 'land of the
Abasci', a Scythian tribe north of Colchis (Arrian, Periplus Ponti 1 1) : but
the name 14.fJaula is not vouched for, and the Abasci appear to be late);
Xo).vf3la.,; Schiltz (not vouched for, and paleographically unlikely); 14.plas
Heimsoth (unmetrical?). W. Ridgeway'sjustification of 14.paf3i.a,; from Xen.
Cyrop. vii. 4 and 5; viii. 6; and Plaut. T rin. 933 ff. ( Transactions of the Cam-
bridge Philological Socie~ ii (1881-2), 179 f.) does not convince me: it is not
surprising that Xenophon knew of the small district called Arabia on the
border of Commagene (it is presumably the Arabia of Pliny, N.H. v. 85 f.;
vi. 129), but it would be surprising if Aeschylus had known of it; and I
find it hard to treat the Plautine passage as a serious piece of geography
(yet Professor Lloyd-Jones has pointed out to me that the Medicean
190 NOTES (pp. 54-58)
scholiast's metrical comment on P. V. 436 (1rws- T~v :4pa{Jlav KavKauru
crovcf,K,uev ;) may come from Comedy, which would make the reading of o;r
text, if not authentic, at least very early). I note a river Arabis near
Colchis ('Scylax', Periplus 83) !
18. Burges himself appears to have favoured this view (see Paley on P. V.
421). Avars are supported by D. W. Freshfield in The Academy, 15 (Jan.-June,
1879), 566 (contradicted by A. R. Fairfield, ibid. 16, 33; further corre-
spondence between the protagonists, ibid. 68, 105).
19. The modem Don enters the sea through a multiplicity of channels
in a rapidly growing delta (Minns 21): in classical times it was 8laToos-
(Anon. Periplus Ponti Euxini 49 = 'Scymnus' 871 ff.; Pliny, N.H. vi. 19.
According to Strabo (493) these mouths were sixty stades apart; seven,
according to Artemidorus (ap. schol. ad Dion. P. 14)-presumably t' in
error for~'). Anyway, the arm of the sea, some forty miles long and fifteen
broad, through which the Don is approached, would condition explorers'
minds to the idea that they were entering a strait.
20. See Jacoby on Hee., fr. 18.
21. A quite different history for the intercontinental boundary is suggested
by Jacoby (on Hee., fr. 195): (i) it is the Colchian Phasis of the Argonautic
legend (the Phasis-Rion), leading to the Ocean; (ii) later it is moved north
to the Hypanis-Kuban and the Cimmerian Bosporus (the latter is also called
Tanais by some authorities); (iii) finally it is moved yet farther north to the
Don, which is then called Tanais. Whereas Hecataeus meant (i) in his
Histories he meant (ii) in his geographical work. This theory solves some
anomalies in Hecataeus-it explains why certain tribes near the Caucasus
are said to be, some in Asia, some in Europe; and why Phanagoria and
environs are in Asia but the Dandarii, just to the north, in Europe. Yet
there are still difficulties. The only ancient author to call the (or a!) Hypanis
the boundary between Europe and Asia is Cornelius Gallus (a/J. Vib. Seq.
11); Jacoby says this is 'aus gelehrter hellenistischer quelle' (this is the only
certain remnant of Gallus we have !)-it is more likely to be a mere slip. And
though Hecataeus put the Dandarii in Europe he put a tribe to the north
of these, and to the south-east of the Tanais mouth, in Asia (cf.Jacoby on fr.
216) ! The whole question is complicated by the fact that our sources for
Hecataeus' views are sometimes palpably inaccurate in referring a place to
Europe or Asia (cf. Jacoby on fr. 193). But it is clear that for Hecataeus
the Phasis-Rion was not the intercontinental boundary; for myself, I am
inclined to think the Phasis-Tanais was, as for Aeschylus.
(This is not a fragment of Hecataeus of Miletus but of Hecataeus of
ws
Abdera, if Roper's (most attractive) emendation of 'EKa.Taws- teiponeis-t
o
to elf' T1),os- is accepted (cf. Dicls-Kranz ii. 240. It was accepted later by
Jacoby himself: 264 F 13). I do not understand Jacoby's argumentation
when he says on the authorship that 'die ableitung des Tanais aus dem Araxes
kann nur ein sehr alter autor gegeben haben'. This may (or may not) be
true of the original author of the idea, but it in no way helps towards choosing
between the earlier and the later Hecataeus here. Why should this author
be any older than Aristotle, who says exactly the same?)
22. Antonius Diogenes also appears to have taken his heroes through
 NOTES (pp. 62-66) 191
the Tanais into the northern ocean: Sul Toii Jl6V7'ov Ka., a.110 rijs Ka.Ta
Ka.a,rla.v Kai 'YpKavla.v IJaAa.uO'f/S 1rpos Ta 'P,1ra.'ia KaAoOp.Va ()(YTJ Ka.i
'T'OV Tav&,llos 'ITO'T'ap.oii 'T'llS lK{Jo>.as ruf,,ypho,, l7'a. llr.a 7'0 'ITOAV 'T'OV t/rvxovs
~L TOV 1:KvlJtKoV lmOTpa.cpiV7'S WKa.v6v (507 = Phot. Bihl. t09a).
23. A general contraction, or underestimation, of distances may explain
the curious error in 725 f., where Salmydessus is spoken of as near Themiscyra
(about five hundred miles away!). Aeschylus' mistake is anyway less gross
than Lucan's at viii. 540 ('vada testantur iunctas Aegyptia Syrtes').
24. If the position of the Phasis suggested above is right, the Arimaspi
would not be far from it. There is a tradition that they helped the storm-
bound Argonauts (Steph. Byz. s. Etlpyfrat). (The Ariaspi got the name
Evpyira, from confusion with Arimaspi- or vice versa (cf. Q. Curt. vii. 3.
1)? See Tomaschek in P.-W. ii. 826 f.)
25. Jacoby (i. 353) states unequivocally that the name Hybristes conceals
the Tanais. This would fit the ancient reputation, and the ancient etymology,
ofTanais (cf. Mela i. I 15; Tanais llu:t 'T'o 7'Ta.phws piv, Eustathius et schol.
ad Dion. P. 14; cf. [Plut.],fiuv. 14), but not the account of Aeschylus, whose
Hybristes must empty, not into the Palus Maeotis, but into the Eux.ine to
the west of it. This is one requirement which any identification of the
Hybristes must fulfil; another is, that it should have the Maeotian-Amazons
on its eastern bank. If the Maeotians/Sauromatae did in pre-classical times
in fact dwell to the west of the Tanais, as may be indicated in Herodotus'
story of the Amazons being washed ashore on the west side of the Palus
Maeotis, we could perhaps identify the Hybristes with the Borysthenes (or
the Hypanis-Bug) without more ado; yet perhaps it is a mistake to try to
identify it with any one river: a conflation of the Tanais with the Borysthenes
or the Hypanis would be preferable, as it would enable us to account for
Aeschylus' self-contradiction in putting the Scythians on the western shore
of the Pal us Maeotis in 417 ff., when in 709 ff. they are away to the west of
the Hybristes. There is no need to suppose that the process of conflation (if
such there was) was conscious: it may have been inadvertent, and aided by
the fact that Aeschylus' authority gave no names to the rivers he de-
scribed.
26. For dating the Pontic colonies I follow R. M. Cook. Archaeological
discoveries (at present to seek) may, of course, necessitate the abandonment
of this position.
Rhys Carpenter (59) thinks that there was nothing to take the Greeks
into the Black Sea before the Cimmerian invasion of Asia Minor, and
therefore that the localization of the Argonautic legend could not have
been much earlier than mid-seventh century. Indeed, in an article entitled
'The Greek Penetration of the Black Sea' (A.J.A. lii ( 1948), I ff.) he argues
that it would have been impossible for ships to enter the EILxine against
the current of the Thracian Bosporus before the invention of the penteconter
(which he suggests took place c. 700): 'The years just before or just after
680 .c. [Carpenter's approximate foundation-date for Chalcedon] must be
our choice for the sensational event which was to become so mighty in
legend-the first passing of a Greek ship into the Black Sea' (9).
27. Mcgasthenes' ants inhabit a mountain plateau in Dardistan. The
 NOTES (pp. 67-72)
Indians distract the insects with lumps of meat, and apparently use yokes
of horses, not camels, in their forays (Muller compares Dio Chrys. 35. 24.
Is his reading of l'IT'ITo,vylo,, an emendation?). Herodotean reminiscence;
the ants are as big as or bigger than foxes; perhaps the meat-bait was sug-
gested by Hdt. iii. 1I I. 3. Pliny's account of the ants (.N.H. xi. 111) is mainly
a blend of Megasthenes and Herodotus.
28. Jacobs'sinsertion of ol before TOV xpvudv <f,vAdT-rov-re, I think obscures
the sense: oil,c civ l>u0,80,1av requires a protasis. I have not followed Hercher
in showing a lacuna before ,ca.Aoiill'Tat, but have supplied ol before -rov-ro,,,
considering this to be but a specimen of Aelian's fatuous style (cf. ix. 36
/J e, ouo,
opv, /J-=a'T'T'7JS
\' ,__ ,1. ,ea,'
EVTl""f'o, e,u,
, 'Y
,ea,' vo.,.,ov-ra, )

29. Griffins in India, ants in Ethiopia: Philostr. vit. Ap. vi. 1; cf. iii. 48.
30. V..alo.,u, [l>.da,u, Brunck] '11'pix>.wpfjuiv lpv.vd.s Tz.
31. The reeds are first heard of from Herodotus, the Indian Dog-men from
Ctesias (Dog-men in Libya already in Hdt. iv. 191; where Hesiod and
Aeschylus (Strab. 299) put them is not certain): see references by May-
hoff on Pliny, N.H. vii. 21 and 23, to which add (on the reeds) Strab. 710-11;
Varr. Atac., fr. 20 Baehrens; Marcianus, Periplus maris exteri i. 44 (G.G.M.
i. 537). On Dog-men see also Yule ii. 311; 228.
32. TJv Frankel: Toii cod. For Apollo's temple among the Hyperboreans
cf. Her. Pont., fr. 51 W.; Paus. x. 5. 9 (cf. Arist., fr. 3; Strab. 421; Philostr.
vit. Ap. vi. 10f.; Plut. Pyth. orac. 17); Hee. Abd. ap. Diod. Sic.ii. 47. 2; Iambi.
V.P. 91.
33. Cf. Alcaeus' Hymn to Apollo (fr. 307). Because I think the narrator was
air-borne I have not followed Wilamowitz in emending I. 5 7r<apl ( = 'in the
neighbourhood of '; cf. 'TTEpl , .. v~uovs in 7 f.) to 1Tapd.. Apollo carried off
Cyrene thus (Pind. Pyth. ix. 5 ff.), and also perhaps Stratonice, a daughter
of Porthaon, according to a new fragment of the Catalogus to be published
by Lobel.
34. It could be Cleinis' soul in ecstasy, but there is no hint of this in
Antoninus.
35. For example, v. 584 Choaspes; 596 Iaxartes; vi. 193 Strymon; 20 I Tyres;
220 Rhyndacus; 618 Hebrus.
36. If correct, this would rule out the identification of the Campasus
with the Cambyses, which flowed into the river Cyrus or the Caspian.
37. I am inclined to date the beginning of Il. xiii to about the middle of
the seventh century-the time of the earliest probings of Greek exploration
to the west and north-west of the Euxine: the Scythians' dietary habits are
known, but not the name 'Scythian'. It may be conjectured that the author
of these lines (as of Il. ix. 5) lived on the Asiatic coast of the Propontis-the
natural base for such exploration (this is confirmed by his apparent know-
ledge that the highest point of Samothrace is visible from the Troad
(Carpenter 36)), But there seems to me no reason to assume indebtedness
to the Arimaspea here.
38. From Clem. Alex. His subsequent words, To~s Jta,coVTaTfitS oOTot
ltw 'TTVAWV a'.yoll'T<aS' a.,f,avl{ovuw, may not be from Hellanicus; note the
reversion to direct speech. They are perhaps a muddle in Clement's own
mind of the Hyperboreans' suicide before reaching senility (Pliny, .N.H. iv.
 NOTES (pp. 72--82) 193
89; Mela iii. 37), of stories of the Massagetae who kill off their old folk
(Hdt. i. 216. 2), and of something like Mimnermus' at yap a-rp vovuwv
'Tfi Kai apya),l.wv p.<iALOwvlwv te,r,KoVTal'TT/ o,pa Klxo Oava-rov (fr. 6).
39. Schmid-Stahlin ii. 678; but cf. P.-W. v. 925.
40. So Lindsay: an divitiis pici? (pici divitiis Non. But both the Plautine
codices and Nonius have the same word order.) 'Picas veteres esse voluerunt,
quos Graeci grypes appellant' Nonius 152; but see J. 0. Thomson in C.R.
70 ( 1956), 3.
41. The Loeb editors of Ennius and Varro have been unnecessarily per-
verse about this fragment: Warmington (Remains of Old Latin i. 392) thinks
decem qualifies the unnamed antecedent of quas; though Varro, who pre-
sumably knew what it was, clearly did not. The reference is presumably
to nuggets not only of rare quality but of ex;ceptional size. Kent, in his edition
of Varro's L.L. (i. 328), takes fodere to mean infodere, so giving the verb
an unusual sense and the Arimaspians an unusual role.
42. iii (comm.), 53: 'man tut H. schwerlich unrecht, wenn man ihn
unmittelbar hinter Aristeas stellt, der cf,oif36Aap:rr-ros yEvop.Evo, von den
Skythen zu den apollinischen Hyperboreern gelangt [not so!] ... es ist ein
unterschied mehr der zeit als des literarischen ylvo,, wenn er seinen reiseweg
- die zwischenhaltung eines "landsmannes" als vermittler seiner kenntnisse
(Rohde Rom. 2 238 [? 228]) ist nach den parallelen unwahrscheinlich-
genau beschreibt' [how is it known that Aris teas did not describe his journey
in detail?]. Hecataeus, then, visited the Hyperboreans in person; and Jacoby
ingeniously reconstructs his route: from Teos by Phanagoria and the Tanais
to the Caspian [not necessary], thence to the north-eastern ocean and along
its coast.

CHAPTER IV
1. I am not sure how much weight should be given to this testimony.
Hue, who was in Lhasa 1845/6, says that the bodies of the dead were cut
up and fed to the dogs, in the 'most complimentary' form of burial (ii. 251 f.).
2. Landor (i. 55) remarks on the general superiority of Tibetan women
to the men, and says that 'the ladies of the Forbidden Land seem to have it
all their own way'. Hue notices the free and active life led by the women of
Tibet (ii. 1 76).
3. This observation of Marco has also been noticed by How and Wells
(on Hdt. iii. 105).
4. All that I have to say about these Chinese sources (including the
translations) I owe to the kindness of Professor D. Hawkes.
5. Chao Hun. Ch'u Tz't,, 9/4b in the SSu Pu Ts'ung K'an edition (Hawkes
104).
6. Both the Dog Jung and the Kuei were real people against whom the
Chinese made war.
7. Sha11 Hai Ching, S.P. T.K. ed. B/55a-56a. ls the 'winged tiger' that
neighbours the ants significant for the griffin story?
8. 'Sea' in these contl"Xts appears to connote something like the Greek
Oceanus.
oar,6 0
19-4 NOTES (pp. 82-91)
9. In the context it appears that this is an error for 'north-west to north-
east'.
rn. Shan Hai Ching, B/43a-43b,
11. Huai Nan T;:, S.P. T.K. ed. 4/7b.
12. A one-eyed giant features in a thirteenth/fourteenth-century Mon-
golian story like that of Odysseus and Polyphemus. But variants of this tale
are so widespread that it has little evidential value for my purpose. See
Frazer's Loeb ed. ofApollodorus, ii. 404 ff. (452 ff. for the Mongolian legend),
13. I am indebted to Mr. L. D. Reynolds for the translation of this
passage.
14. An account of an illustrated Chinese 'Book of Monsters' is given by
de Mely in Revue archeologique iii ( 1897), 367 ff. It notices one-eyed men;
dog-headed men, the males of whom barked only and could not understand
speech, though the females could speak and understood Chinese; pygmies,
who are attacked by cranes; men with horses' legs; wild men with hairy
bodies and pendulous breasts (Abominable Snowmen?); long-lived men;
and men with fishes' bodies (these can be traced back to the fourth century
B,C,: one of the 'Heavenly Questions' (T'ien Wen) is 'where does the man-
fish live?'-immediately after a reference to long-lived men (Hawkes 49).
The pygmies and cranes are strange. They are put in India by Megas-
thenes (Strab. 711 ). They appear of course in Il. iii. 3 ff. (For geranomachies
in Greek art of the seventh century B.c. see Dunbabin 79; 87.)
15. Gnomon9 (1933),567f.; Meuli 156f. [185f.]; Phillips 174. It appears
to be accepted also by Dodds (141) and Bowra (1). Ft.rtwangler's view is
rather different: the griffins are treasure-guarding dragons, their Arimaspian
enemies are goblins (Roscher i. 1768).
16. Cf. Yule 311 f.; 228 (on Dog-heads).
17. This survey is based on the admirable disquisitions of Prinz and
Ziegler (art. 'Gryps' in P.-W. vii), where references to illustrated examples
will be found. See also Furtwangler in Roscher (Gryps: illustrated);
Leibovitch in Bulletin de l'Institut d'Egypte xxv (1942/3), 183ff. (illustrated);
Phillips 172.
18. That lions were known in neolithic times as far north as !at. 46 is
proved by a clay model of one that was unearthed at Schipenitz (ref. in
note 19).
19. There is no room in this scheme for the 'griffin' which was discovered
on a late neolithic site (perhaps c. 2400-1700 B.c.) at Schipenitz (Bukowina)
(Gordon Childe in Journal of the Royal Anthropological Institute !iii (1923),
281; 283). The finds here included some painted pottery with animal
figures. Most of these, though stylized, are not monstrous; but one, a
silhouette in profile of a couchant four-legged creature with animal's
muzzle, prick-ears, tail curved over body, and sharp claws on the hind feet,
has also a distinct forward-tilted hump rising from the shoulders, which
might be meant for wings. Childe calls this a griffin; and, while not ruling
out the possibility that the wing-like projection may be due to accident or
to a later stain, is himself convinced by close examination that it was part
of the original figure.
20. Is it because the one-eyed Arimaspians were shaggy all over that
 NOTES (pp. 91-101) 1 95
later poets and artists so represented the one-eyed Cyclopes (e.g. Theocr.
xi. 50; Ov. Met. xiii. 846 f.; Philostr. Jmag. ii. 18. 3: optoS' T K(lt 8tVOS'
, I ' ) I ) ()' ~ I
'Y'Ypa.1TT4t XT'TJV fLEV ava<1tWV op 'rJV K<lt\ 41"'1'"'"'1''/
, ~ \ , _ ~ I
1Tt"TVOS OtK'rJV,
I ~ \I' ,/.. I )I:'.'
I ) Q "" "" I I
KpxapoVS' OE V1TOy,aG11Wll OOOl/Ta!; K ,-,opou TOV 'YVWV <1TEpVOV TE
Cf. Call H lll.
... 76)?.
1
Kat 'Y0.<1TEpa KO.t' TO' EtS' ' OVVX<l
1KOV 1\0.UtoS'
.J: '' 1TO.VTO.;
'
21. Dr. Paul Jacobsthal expressed the opinion to me, in 1953, that the
artist was an Ionian living in a Greek colony in south Russia; he dated
the mirror to c. 575 B.c. Otherwise, he said, he agreed in the main with
Payne (Necroct>rinthia 231).
22. But brlyp=os in Hdt. ii. 76.
23. Thus attaching op0wo3os inrep mfyov to the verb-rpo.ef,'r/, with the other
adverbial phrases, rather than to ap,t7T7ToS, whatever that means (perhaps
'stabled with the horses' of the wind-god? The passage is discussed by Lloyd-
Jones in C.Q.. li ( 1957), 24ff., who does not, however, consider this possibility.
If this phrase is to go with ap,t1T1TOS' then no particular 110.yos need be meant,
as far as I can see). Ilo.yos means 'a precipitous excrescence' ('hill' is insuf-
ficiently precise); sometimes the 'precipice' itself (Hes. Scut. 439), and this
might be its sense here, the cave being then imagined as situated in a
mountain side above a precipice.
24. The Cave of Boreas appears again in Callimachus (H. iv. 62 ff.,
where note the odd epithet in I. 65, l-rrra..vxov f3oplao 7Tapa. <11TE0,), who
located it not in the Rhipaean mountains but in the Haemus range. So too
Apollonius Rhodius (i. 211 ff.) puts Boreas' home in north Thrace, near
the 'Sarpedonian rock'-also known to Sophocles (fr. 637: ,j.eis 3' [Cleo-
patra speaking?] iv a.v-rpo,s [Bergk: a<1Tpo,s] lv/Ja l:af'1TTJ8wv 1TETpa) and
Simonides (schol. ad Ap. Rhod. i. 212). It could be that Boreas' cave was
first located in Haemus and then moved to the Rhipaeans; but this would
make the belief Greek, and if so why do the other Greek winds not live in
caves? I think that the cave was brought to Haemus from the Rhipaeans,
under the influence of Homer's 'Thracian Boreas'.
25. The same story was told in Britain, of a mighty wind issuing from
the Peak Cavern in Derbyshire (Gervase of Tilbury, Otia lmperialia 24; 117
Liebrecht; cf. Camden's Britannia 495 (trans. Gibson; London, 1695)).
26. Rubruck only in fact-if even he? Cf. Humboldt 1i. 501 [
27. But see Atkinson 472.
28. The Spirit of Chung Mountain, who blows hard and soft to cause
winter and summer (82] may help to take the tradition back to the third
century B.c, (or earlier); and a Chinese poem of about that date mentions
a 'Cave of the Winds' (location not stated) (Hawkes 79).
29. E.R.E. ii. 689 ff.; Rohde ii 233 f.
30. Himselfa 'blameless' Hyperborean? The name is clearly a pseudonym.
3 I. Apparently first suggested by Gladisch (in Roscher i. 2829), and
followed by Tomaschek (764 ff.), Minns (ll3 f.), Casson (C.R. xx.xiv (1920),
3), G. F. Hudson (Europe a,id Chilla 27 ff.), and Sykes (Quest for Catha_y 19 f.).
32. I do not see why Farnell (i iv. 100) calls it a 'foolish etymology'
(Dodds (162) thinks it is probably the right one after all). References to
discussions of the problem will be found in Guthrie (78 f.) and D. L. Page,
Sappho and Alcaeus 251. The modern etymologies are two: ( 1) 'Y7Ttpf36peo1 is
 NOTES (p. 101)
a vulgar corruption of inrlp{Jopo, which may have been a north Greek form
of inrl~po,, which is supposed to mean 'carriers over' and to be connected
with trEp<pEpEES, the name by which Herodotus says the escort was called
which accompanied the annual gifts sent by the Hyperboreans to Delos via
the Adriatic, Dodona, Euboea, and Tenos (Pausanias much later gives a
different route, through Sinope). This would make the legend of the Hyper-
boreans Delian, for no such gifts were sent to Delphi. But (a) there appears
to have been a Delphic version of the Hyperborean legend as early as c.
6oo .c. (our evidence for this-Alcaeus-is much earlier thl\n for the Delian);
this has to be dismissed as 'rivalry with Delos'; (b) why should gifts sent
from the north of Greece to Delos follow such a roundabout route? (2) The
root of the word is a pre-Greek word for 'mountain' (the Slavonic gara
is adduced, though its relevance to seventh-century Greece is obscure); and
Bora was the name of a mountain in Macedonia, according to Livy. This
would make the legend of the Hyperboreans Delphic (they dwelt 'beyond
the mountain' from Delphi). Then what about the Hyperborean offerings
to Delos? Where did they come from, and why were none sent to
Delphi?
Yet these offerings seem to have been historical. C. T. Seltman ingeniously
suggests (in C.Q.. xxii (1928), 155 ff.) that they really emanated from a
Graeco-Getic [?] offshoot of the Milesian colony oflstrus (founded 656 B.c.),
descendants of settlers who had moved well inland up the Danube valley.
The route by which the offerings were sent according to Herodotus (iv. 33)
may have been cut in the fourth century u.c. by Celtic eastward migration,
and so diverted to that given by Pausanias (i. 31. 2, where Arimaspians and
Issedonians are inserted to push the Hyperboreans farther away). Connexion
of the Delphic Apollo with the Hyperboreans was due to deliberate Delphian
invention, to rival Delos.
The theory is an attractive one, so long as it is understood to apply only
to the source of the 'Hyperborean' offerings, and not to account for the
origin of the Hyperboreans themselves. If it is right, the offerings could not
have started before the second half of the seventh century; more probably
later, as the foundation-date for Istrus which Seltman accepts may well be
too early, and the archaeological evidence he mentions from the Danube
valley is not earlier than the sixth century. This would make his account
of the beginning of the Delphic-Hyperborean connexion hard to maintain.
(This Delphic connexion seems to me to indicate that the Hyperboreans
existed in the Greek imagination before their offerings.)
Our earliest mention of the Hyperboreans (apart from Aristeas) is
probably in a fragment of the Hesiodic Catalagus (before 600 n.c.? cf.
Schmid-Stahlin 1. i. 269). There they are called iiJ,1r1ro,, and so are clearly
merged with the northern nomads [189]. There is no certainty about even
the approximate date of the Homeric Hymn to Dionysus (vii), which Guthrie
(O.C.D. s. 'Hyperboreans') says contains the first mention of them (28 f.).
I would note that here they appear to be located overseas, though this could
not be pressed; they are at least 'at the back of beyond',
I am disposed to incline to the view that Page says the evidence favours:
that in origin the Hyperboreans are not a real people but a folk-memory
 NOTES (pp, 101-116) 197
-'the earlier settlement of an Apollo who spread with his worshippers
southward over the Greek mainland'.
33. J. Legge, The Chinese Classics m. i. 151. Some of this work, but not all,
belongs to the sixth century A.D,; but Mr. Geoffrey Bownas tells me that
this particular note is probably to be dated to the end of the fourth century
u.c. Cf. Hennig 23 ff.; Alfoldi, Gnomon ix ( 1933), 566.
34. Dodds 162. Baring-Gould is reminded by these swan-maidens of the
Phorcides of the P.V. (Curious Myths ojthe Middle Ages 568).

CHAPTER V
1, Hecataeus may have been following Aristeas here; but it should be
noted that he called the Issedonians 'Scythians' also (fr. 193), which Aristeas
did not.
2. tp8tp<Yrpaylovot. To save the reputation of the Budini How and Wells
wish to render this 'eat fir-cones'.
3. These are a puzzle. Elks and reindeer have been proposed, but neither
of these is aquatic. Rawlinson's seals are surely impossible? Minns (105) on
second thoughts would see otters in the fhipla TtETpaywvo11p6uW1Ta, and take
tvvopw, as water-snakes; I suppose people who ate lice might also hunt water-
snakes ! How and Wells suggest that the square-faced animals were mink.
4. Representative of the principal schools of thought. There seems no
point in trying to be exhaustive here. References to the ideas of other
authorities will be found in the works summarized.
5. Erman 211 on the Bashkirs and the bird-cherry; 29i on aschy: 'the
Bashkirs denominate every acid, especially the sour juices which they mix
with their milk, by a word which is written atchui in Russian, and call the
cherries themselves tchia, a word formed, probably, from the same root',
'the word for sour, among the Tatars of Kasan, is a/chi. The Sclavonian is
quite different.' The word in fact is Turkish (dci or aksr Minns mg).
6. [42], and Kiessling in P.-W. iA 846 ff. But there is no reason to suppose
that this notion was as early as Aristeas.
7. This would not mean that the whole area was inhabited by Issedonians
but that it was the region over which the horde moved its flocks for pasture.
8. A seventh-century Greek in a strange land could not have known his
bearing in relation to his home at all accurately, and must have found it
exceedingly difficult to maintain a course in one direction consistently over
a long period. Though sailors already knew how to get a rough bearing
from Ursa Major (Od. v. 277), the science of navigation by the stars was
rudimentary. This may have been the chief reason why the early Greek
mariners tended to coast and avoid the open sea if possible (that they had
no fear of the open sea if they were sure of making a correct landfall is
indicated by the fact that the regular trade-route to India lay across the
Indian Ocean, once the regularity of the monsoons had been discovered
(Plin. N.H. vi. too ff.)). The stars would be even less use to an explorer
travelling in the daytime over featureless plains.
(The other constellations mentioned in Od. v. 271 ff., Bootes and the
Pleiads, would not help Odysseus much in finding his way, though they
 . ~

198 NOTES (pp. 116-118)

might tell him the time (the Pleiads were of more importance to farmers).
It is surprising how unobservant the Greeks were for long about the heavens,
until they learned more from the peoples of the Middle East in the fourth
century B.c. It does not appear that they realized the superiority of the
Little over the Great Bear for orientation until then (or even then: cf. Arat.
Phaen. 36--44). Before then too the only constellations our authors name
are the Great Bear, Bootes, Orion, the Pleiads, and the Hyades (the mention
of Aquila at Rhesus 530 rather points to a post-Euripidean date for the play).
There is no first-hand evidence that they had any knowledge of the planets
either until the end of the fifth century, in Xenophon ( Mem. iv. 7. 5), apart
from the obvious Venus. The names of the planets are even later: Plat. Tim,
38 d; Epin. 986 e ff. ; Seneca (N. Q.. vii. 3. 2) says 'Eudoxus primus ab
Aegypto hos motus [sc. of the planets] in Graeciam transtulit'.)
9. Cf. the brief Peripatetic treatise, Ventorum situs et cognomina.
10. As also at Od. xiv. 403 f., if'the Syrian island' is Syros and 'Ortygia'
Delos (east of Syros). But this is uncertain (cf. Kirk and Raven, The Pre-
socratic Philosophers 52 ff.).
11. iJ1rEp Kp~T'TJS' at Od. xiv. 300 may mean 'to the south of Crete' ('above
Crete' from the point of view of the Ionian poet).
, 2. The 'tilted disk' theory is ascribed to Anaxagoras, Leucippus, and
Democritus in the fifth century (Diels-Kranz ii ..5 34; 22. I 1; 71. 17; 78.
12; 107. 4; cf. Hdt. i. 142. 2); it may have been held by Anaximenes in the
sixth [ 188]. It persisted even after the discovery of the sphericity of the
earth. Cf. the Pythagorean view quoted by Aristotle, cael. ii. 285b; Virg.
G. i. 240 f.; Macrob. S.S. ii. 5. 22.
13. A beginning is being made with a1TTJALWTTJ<:, an Eastern Ionic word;
the consummation of the process is to be seen in Arist. Meteor. ii. 6.
14. Though not exactly: the name given to the true north wind was
more often Aparctias (Arist. Meteor. ii. 6; [Arist.], de mundo 394b29-/3opla,
is also a generic name for winds 0.1T(~ apKTOV, 394h20; Pliny, N.H. ii. 119;
Favorinus ap. Aul. Gell. ii. 22; Vegetius iv. 38; Lydus, mens. iv. 76).
15. The name 'Arimaspian' may be Mongol (iiriim diik means 'one-eyed'
among the Mongols, according to Laufer 452). The etymology is uncertain;
if known, it need only tell us what language the /ssedonians spoke, of course.
We hear of a king of the lazyges calledBa.vall<t<nros (Dio Cass. Ixxi. 16), and
an lberianJ4a,a.C11Tos (/.G. xiv.1374). These are late (second century A.o.),
but if anything weigh against a Mongolian origin for J4pia.C11ToS',
If the Arimaspi were Mongols, they must have been among the first of
their race to learn to ride the horse (cf. C. W. Bishop in A11tiquiv, xiv ( 1940),
314; Phillips 173, and further in Artibus Asiae xx (1957), 160 f.). Mongols
are not hairy, but the Arimaspi might have got their reputation for hairiness
from their garments: '[the Tatars'] shoubes or gownes are hayrie on the
outside, and open behinde, with tailes hanging downe to their hammes'
(Carpini (Beazley 109)).
16. The modern name, Hindu Kush, was used c. 1332 by Ibn Batuta,
who explains it as (Persian) 'Hindu killer' (E.B. xiii 514b). Kush might be
related in some way to Caucasus (cf. Thomson 126), but even so whether
as progenitor flr descendant cannot be told,
 NOTES (pp. 120-131) 1 99
CHAPTER VI
1. c. 1 quotes Theopompus by name; c. 5 has two stories about Pherecydes
which are branded as falsifications of Theopompus' making by Porphyry
(ap. Euseb. P.E. x. 3. 6 ff.).
2. Cf. Plut. gen. Socr. 23. 592c-d (where he is called Hermodorus, a psycho-
logical mistake of Plutarch, presumably, caused by the fact that 'Hermo-
dorus'' soul remembered its experiences in and out of its various bodies by
the gift of Hermes (D.L. viii. 4 f.), rather than a copyist's error as Rohde
thought (i 331, n. !12)).
3. Was Hermotimus originally a local hero-prophet (cf. Amphiaraus
and Orpheus), and were his enemies, the Cantharidae, the clan of a rival
prophet (cf. the Cragalidae and Delphi, Parke i. 342 f.)?
4. i. 519. I do not understand his remark 'die schriften der wundermanner
Abaris und Aristeas sind wahrscheinlich nur erfindung Lobons'.
5. See Rohde's art.' I'lyovii in den Biographica des Suidas' (KleineSchriften
i. 114 ff. = R.M. xxxiii (1878), 161 ff.). Rohde thinks that the coupling of
Cyrus with Croesus must indicate the capture of Sardis by the former, and
that the variants in the date of the Olympiad are to be explained by the
postulate that one tradition has preserved the first figure of 117/', the other
the second (Kl. Sehr. i. 136). This is plausible, but a less likely corruption than
the one I suggest.
'Suidas' normally uses yc!yov in the sense of 1]1<.a,ii, much more rarely
in the sense of cyrn1179'7 (ibid. 177).
6. i 329. The earliest date for the Metapontine appearance Rohde
makes to be 434, but I would not allow some of his premisses-for example,
he postulates the Cimmerian invasion as a terminus post quern (thus assuming
the identity of Aristcas with the author of the Arimaspea that Herodotus
knew). In ii 186, n. 1, on the other hand, Rohde moves the date of the
Metapontine appearance back some sixteen years, to c. 450, and appears
to feel no qualms about this; but here he takes no account of the Cimmerian
invasion (though the poem still features in his train of reasoning), and any-
way makes the foundation-date of Proconnesus too early.
7. One naturally takes this to mean 'in mattY places', though I suppose
strictly it need mean only 'more than one' (c Eur. He[. 587f.). Even so, the
testimony would be non-Herodotean.
8. Cf. Nilsson 1 104. In the case of the honours paid by the Abderites to
Timesius (Hdt. i. 168) it is not clear whether his grave was there or not (Jt-
>..aa8el~ could mean that his followers were driven out, but he himself was
killed).
9. Cf. Boyance ~34 Boyance thinks it was probably heroization, and in
Pythagoras' case widespread in Magna Graecia, not confined to the place
where he ended his days.
10. Has the tradition passed through the hands of some joker like Her-
mippus (cf. his account of the death of Pythagoras (D.L. viii. 40), and of
the works of Zoroaster (Pliny, N.H. xxx. 4))? Hermippus may have been
guying Heraclides Ponticus in particular (cf. Levy i 39 ff.).
11, Meuli (154, n. 2) also rejects Schwyzer's emendation, but for a bad
 1
'
200 NOTES (pp. 132-141)
reason: 'd"IC worte wr . eyw
' ' UV/Jc{3-O.IVIOfLEVOf
\\ I '
EV IIp0KOVVYJC1W
I
TE Ka,
Mc;-ro.1rov-rlqJ evptcrov verbieten wohl diese Anderung; mit ' (7 a.) Jv
MeTa1ToV'Tlq> kann doch H. nicht seinen eigenen Aufenthalt, sondern nur die
br18rila des Aristeas gemeint haben.' uv,f1aM6,evos does not here mean
'comparing' but 'calculating'; not Ta iv Me-ro.1ToVTlqJ but -ra freo. is to be
understood as its object (cf. ii. 31). The emendation was published in
Philologische Wochenschrift ( 1922), 528.
12. Proconnesus may not have been founded from Miletus direct, but
from one of the Milesian colonies on the east coast of the Propontis, perhaps
through fear of the oncoming Cimmerians (e.g. from Cyzicus, itself founded
c. 675?--cf. J.H.S. Iv. 146). But this is sheer speculation. For the little that
is known of the history of Proconnesus see F. W. Hasluck in J.H.S. xxix
(1909), 6 ff.
13. 157 f. Herodotus was only too pleased to have everything as rational
as possible, 'selbst . wenn ... die wunderbare Schnelligkeit der Reise van
Prokonnesos nach Artake nicht wegzubringen war . . . An Stelle dieser
urspriinglicheren Sagen haben die Griechen von Prokonnesos und Kyzikos
eine Erzahlung gesetzt, die den Wundermann seine Reise ganz leibhaftig,
unter mi:iglichster Ausschaltung alles Obernatiirlichen, ausfiihren liess'. But
it is not so much the speed of Aristeas' journey from Proconnesus to the
mainland as the speed of the Cyzicene's in the reverse direction which is
really anomalous.
14. The story from Heraclides Ponticus (ap. D.L. viii. 5) about Hermoti
mus identifying Euphorbus' shield in the temple of Apollo at Branchidae,
and himself being a previous incarnation of Pythagoras, is insufficient
to establish that Hermotimus had any strong or traditional connexion
with Apollo (though there was a temple of Apollo in the Clazomenian
territory-Strab. 645). Hermotimus is not numbered among Pythagoras'
incarnations in the lists of Dicaearchus and Clearchus (Aul. Gell. iv. 11. 14);
that of Hermippus is probably derived from Heraclides (cf. Levy i 40; so
also Tert. anim. 28; Porph. V.P. 45).
15. Cf. xxxi. 10: Ll'1],~Tpws .; <1'a.AYJpVS a80.1TEp 'XP'1JC1[.L'{J8wv
EVC1Toxwr TOV'TOVS TOIJS .\oyovs 0.1T01T,f>ol/30.KEV. It should be said that in
neither of these passages can we be sure how far the words are those of
Diodorus or of his excerptor; not that it matters a great deal as we have the
use of rf,ot{3a,nv by Lycophron.
16. Pausanias does not call these men rf,ot/JoA'YJ1TTOt, but the fact that
they were iEpoi T<p Oe<p suggests it. If so, Apollo's possession of his devotees
(Frazer's suggestion that they may have been scapegoats seems to me
unhappy) at Hylae and his possession of Aristeas have features akin to
Dionysiac frenzy: superhuman physical feats (cf. Eur. Bacch. 755-8) and
compulsive wandering.
Plutarch (E ap. DeljJh. 20) preserves a tantalizing bit of information:
certain Thessalian priests on days of ill-omen withdrew into the solitudes
and were said to be 'ruled by Apollo' (rf,ot{3ovo.etcr8o.t). Plutarch guesses
that this means they kept themselves pure; but it is an odd word, and one
would like to know more.
17. Carpenter (161) discounts theories that Aristeas was an epileptic.
 NOTES (pp. 142-149) 201

CHAPTER Vll
1. Theano, the wife of Pythagoras, was probably an invention of the
fourth century B.C. Our earliest authority for her is Hermesianax (fr. 7. 85 f.
Powell (Collectanea Alexandrina 100)), who flourished in the early third; the
line ascribed to Empedocles by Hippobotus ap. D.L. viii. 43 is probably
spurious (cf. fr. 155). The name may have been suggested by that of the
wife of the mythical king Metapontus (Theano perhaps as early as Euripides:
Hyg. Fab. 186); or by the Thracian Theano of I!. xi. 224 al. (a number of
the names of 'early Pythagoreans' in Iamblichus' catalogue are taken from
Homer).
2. There are other mathematical works ascribed to an Aristaeus, who
may have been a real person. His Solid Loci was used by Euclid (Heath,
Greek Mathematics i. 438); and if he is the same Aristaeus who wrote a Com-
parisun ef the Five Figures he would probably have flourished in the fourth
century B.C. (after Theaetetus; cf. ibid. i. 162).
3. Hippasus is listed as a Sybarite, whereas he is a Metapontine or a
Crotoniate elsewhere (D.L. viii. 84; Iambi. V.P. 81); Timaeus, elsewhere
a Locrian (e.g. Plat. Tim. 20; [Arist.) mir. ausc. 178), is here a Crotoniate
or a Parian. That Aristeas should be classed as a Metapontine therefore
causes me no misgiving (especially as it can be explained [ 170]); nor that
he should then become a Crotoniate, in view of the way in which Croton
and Metapontum, from being both in6mately connected with Pythagoras,
were interchanged (cf. the confusion about Hippasus already mentioned;
Diels-Kranz i. 102.16 (contrast 20 f.); Plut. Rom. 28 [128]).
4. Cf. Levy i 37 ff. Hermippus' story was known to the schol. ad Soph.
El. 62 and to Tertullian (anim. 28). The latter calls the olKluKos in which
Pythagoras lay hid by Gregory's word, aqyta.
5. I am unable to agree with Dodds about this. Nor are the stories about
Hermotimus and Abaris in point here, as far as I can see (Dodds 141 ).
6. In the tradition of the wandering poetic Sfvo,a may be the souls of
the vaporous poets which Trygaeus met in the sky grubbing among the
clouds for matter (Ar., Pax 827 ff.). This passage implies the notion ofsoul-
journeys during life; the poets themselves were perhaps asleep! Cf. Pind.,
fr. 116, where Dodds (104) infers that soul-journeying is meant.
7. The idea of the pettiness of mundane affairs is already present in
Plato (e.g. Theaet. 173e).
8. Who 1<YToplav nva. 'TOUlVTl)V 1rpwTOS 1rapa8e8wKV Bav.aalav. I suspect
1rpwTos: that it cannot here have a temporal sense is shown (a) by the tense
of the verb, (b) by the existence of the Platonic precedent, the myth of Er,
on which indeed Proclus is here commenting!
g. It is noteworthy that Timarchus' soul escaped from his body through
the sutures of his skull (Tijs Kfcpa):ij, a.a. tf;6cf,'I' 1rpacnmr6vn 1r>.11yd0'1),,
'TO.S' pacpa.s Sia<YTauas p.8,eva, 'HJV tf;vxfiv), like that of Dr. Wiltse [139]. A
similar sensation experienced at the onset of catalepsy is recorded in R. C.
Johnson's Watcher on the Hills (51). According to the Tibetan Book efthe Dead
(ed. W. Y. Evan-Wentz (1949), 91 f.), the most desirable outlet for the
202 NOTES (pp. 151-157)
soul of the dying is through the sutures of the skull. The idea looks as if it
may be connected with a belief that the soul is introduced into the embryo
through the fontanelle.
10. And is nicely parodied by Lucian (Philops. 22 ff. Cf. Wehrli on H.P.,
fr. 92).
11. From which it might be inferred that for Empedotimus the circle
of the Milky Way is at the same distance from the earth as the circle of the
sun-it is the boundary of Pluto's realm, Hades. This would accord with
a Pythagorean belief that the Milky Way marks a former orbit of the sun
(Arist. Meteor. 1. 8. 34514). For Cicero (rep. vi. 16) the Galaxy is, and for
Manilius (i. 758 ff.) it may be, the seat of the souls of the blessed; for Statius
(Silv. i. 2. 51 ff.), the seat of the gods themselves. I do not know whether
there is any significance in the fact that the Milky Way intersects the Zodiac
in the sign of Scorpio (Man. i. 6go), where was the way by which Hercules
went to the gods [151].
o
12. Cf. Arist. de anima i. 3. 10, Ttp,aios cpvrr,o>i.oyEi '7'1/V ipvx},v Kwerv
.,-o uwp,a /C'TA., where he means oll)vl:rwv Jv T.alq, (36e). Cf. Ammonius,
tk interpr. 133, 16 Busse ( = Diels-Kranz i. 223. 12).
13. It became an eschatological commonplace, apparently: Porph. ap.
Stob. i. 447. 5, ~ ,\eyop,EVTJ Kat vo.,to.EV7/ Tt~v Jv ':4,8ov .,-plo8os.
14. And Clearchus, fr. 10 may be a parody of Heraclides (D.L. viii.
4 f.).
15. This, in conjunction with inremp. ii. 119 [151],disproves Levy'sintcr-
pretation ofVarro's mortakm aspectum detersus (Ii 155, n. 2). Cf. Wehrli's note
on H.P., fr. 91.
16. A belief ascribed to Pythagoras himself by Diodorus (xviii. 1. 1),
which need mean no more than that it was held by Pythagoreans. It was
accepted by Aristotle (ap. Sext. Emp. 1Tpos cpvuKovs i. 20 f. = Ross, Works
of Aristotle Translated xii. 84: cf. Jaeger, Aristotle 161 f.). It was a piece of
popular lore (cf. It. xvi. 843 ff.; Plat. AJ,ol. 39c; Harland and Wilkinson,
Lancashire Folk-lore 104), taken up by savants looking for proof of the soul's
divinity.
17. Cf. Arist. Eud., fr. 1 (Aristotle Translated xii. 16). b the flood that which
overwhelmed Helice and Bura (Hcraclides Pont., fr. 46 W.)?
18. Though the fact that the Tlleages is quoted might have some sig-
nificance. Heraclides has been suggested as the author of this dialogue
(A. E. Taylor, Platos 534). Even the reference to Execestus could be from
Heraclides. He and Aristotle were exact contemporaries, and must have
known each other well in the Academy. This makes it impossible to pro-
nounce on their Ii terary relationship.
19. Once Gauls have become Hyperboreans it is a short and logical step
to see the Rhipaean mountains in the Alps: Steph. Byz. s. 'Y1rEp/36pE01
(Protarchus); Posidonius ap. schol. ad Ap. Rhod. ii. 675 (this scholiast
also tells us that Hee. Abd. said that the Hyperboreans existed in his time).
20. This seems to have been a separate work from the 'TTEpL 61Ka.10UV"')S
(or o1Tpos J1{3apw >..oyos Prod. in Tim. 141d = fr. 36 Voss), of which Abaris
may have been an alternative title (perhaps fr. 73 W. should rather be
assigned to the 1rEpl 01Ka10UVVT}S; but cf. fr. 75 W. with Iambi. V.P. 218).
 NOTES (pp. 158-159) 203

wEpt -rwv .,,.pas :14pap,v &.vaef,EpO/J,EVWV naturally means 'concerning the say-
ings [or writings] ascribed to Abaris', (not 'remarks addressed to A.', which
Voss (57) thinks possible): it may have been about his XfY'1<TJJ,Ol (Ap. mir. 4
(Theopompus); schol. ad Ar. Eq. 729; Lycurgus, fr. 85; Suid. s. J!f3ap,,).
2 z. Accepting the convincing thesis of Boyance (ReVU4 des Etudes ancunnes
x:xxvi. 321 ff.) that Iambi. V.P. 2 15-2 z is a summary of the 1TEpl o,i<aLOCTWTJS';
I am sure that it is a summary of some Heraclidean dialogue, but not so
sure which one,
22. Cf. Dodds 161, n. 33; Meuli z6o.
23. The function of the arrow in Aristotle's account (Iambi. V.P. 140--3,
if it is from Aristotle; Ross accepts it, Aristotle Translated 135) seems to have
been that of a guide: whether Abaris rode on it is not clear. Anyway, see
note 18 above.
24. Of the nicknames given to the 'pupils of Pythagoras' Abaris, Epi-
menides, and Empedocles [ 143], the last, in the form 1<wAvaavla,, was
known to Timaeus (D.L. viii. 6o), a younger contemporary and (hostile) user
of Heraclides (the most improbable tradition that Empedocles was a pupil
of Pythagoras also comes from Tim. (D.L. viii. 54), perhaps from the same
source as i<wA.vuavla,). If this trio of nicknames were really coeval, as is
suggested by their conjunction in the later authorities, Heraclides must
almost certainly have been their inventor.
25. See B1dez-Cumont i. 8o--84; ii. 63-66. Aristoxenus (fr. 13) was using
Heraclides. Lydus (mens. iv. 42) provides the link whereby we can draw on
Antonius Diogenes (507 ff. = Phot. Bihl. 109 ff.) to e.xtend our knowledge
of Heraclides' ,Zoroaster: what Dercyllis is said in Photius' epitome to have
heard from Astracus about Pythagoras and Mnesarchus is given at greater
length by Porphyry (V.P. 10-14). Antonius claimed to have authority for
his fantasies in ancient writers; unfortunately Photius does not preserve
their names, but Heraclides was clearly one. Noteworthy also is the location
of the Sibyl in the moon in Antonius, with which should be compared a
passage in Plutarch, very Heraclidean in its interest in oracles and its pro-
posal of a 'scientific' explanation of them (Pyth. orac. g. 39c-d; cf. ser. num.
uind. 112. 566d; and, for the proximity of the moon to the island of the
Hyp<'rboreans, Hee. Abel. ap. Diod. Sic. ii. 47).
Wehrli's assignation offrr. 69 and 70 to the Zoroaster cannot be supported:
the original of tlw magus who claimed at the court of Gelo to ha\'C circum-
navigated Africa was surely Sataspes (Hdt. iv. 62). not Zoroaster--cven
though the historical Sataspes was unsuccessful.
26. o( i ~v oU] -rOv T Hpd <paa,v Etva, TDv 1TaTJpa. ToV l-'08ot, ToUSc: 1ravTOs,
>\\' z '
1t11a wpoa<TTfY11V, Kat\ w;-
' TOvoE
_"':'~ 'TOV" ' o,oaTOS'
t I I I \ ..J.J..
}'Kfl/LEl'o11 1"1)'' ypa'f''I''
J1<00w1<aaiv, W<T'11'Ep l<Cl< 0 'Em1<0Vp10S' Ko,\wl"'f)S'- I take it that Proclus
means that some texts of the Platonic myth had actually been editro with
the name of Zoroaster substituted for that of Er, and is not referring to the
apocryphal work of Zoroaster quotro by Clement (Strom. v. 103), as Bidez-
Cumont think (ii. 160).
27. The real Zoroaster could mver haw made such a claim himsdf:
metempsychosis was not a tent't of true l\fozdaism (Bidez-Cumont i. 28).
If Aristens was anywhere said to h:ivt had ,1 munber of incarnations,
202 NOTES (pp. 151-157)
soul of the dying is through the sutures of the skull. The idea looks as if it
may be connected with a belief that the soul is introduced into the embryo
through the fontanelle.
10. And is nicely parodied by Lucian (Philops. 22 ff. Cf. Wehrli on H.P.,
fr. 92).
11. From which it might be inferred that for Empedotimus the circle
of the Milky Way is at the same distance from the earth as the circle of the
sun-it is the boundary of Pluto's realm, Hades. This would accord with
a Pythagorean belief that the Milky Way marks a former orbit of the sun
(Arist. Meteor. 1. 8. 34514). For Cicero (rep. vi. 16) the Galaxy is, and for
Manilius (i. 758 ff.) it may be, the seat of the souls of the blessed; for Statius
(Silv. i. 2. 51 ff.), the seat of the gods themselves. I do not know whether
there is any significance in the fact that the Milky Way intersects the Zodiac
in the sign of Scorpio (Man. i. 690), where was the way by which Hercules
went to the gods [151].
o
12. Cf. Arist. de anima i. 3. 10, Tl.ar.os ,f,vu,o>..oyi ~v y,vxqv KLViv
TO aw.a KTA,, where he means ollM.Twv lv T,.alq, (36e). Cf. Ammonius,
de interpr. 133, 16 Busse(= Diels--Kranz i. 223. 12).
13. It became an eschatological commonplace, apparently: Porph. ap.
Stob. i. 447. 5, ~ A"J'O/J,V1J Ka.I vo.i,o.lV1} 'TlOV EV ".A,aov Tploaos.
14. And Clearchus, fr. 10 may be a parody of Heraclides (D.L. viii.
4 f.).
15. This, in conjunction with in remp. ii. 119 [ 151], disproves Levy's inter-
pretation ofVarro's mortalem aspectum detersus (ii 155, n. 2). Cf. Wehrli's note
on H.P., fr. 91.
16. A belief ascribed to Pythagoras himself by Diodorus (x.viii. 1. 1),
which need mean no more than that it was held by Pythagoreans. It was
accepted by Aristotle (ap. Sex.t. Emp. wpos ,f,vaiKovs i. 20 f. = Ross, Works
of Aristotle Translated xii. 84: cf. Jaeger, Aristotle 161 f.). It was a piece of
popular lore (cf. It. xvi. 843 ff.; Plat. Apol. 39c; Harland and Wilkinson,
Lancashire Folk-lore 104), taken up by savants looking for proof of the soul's
divinity.
17. Cf. Arist. Eud., fr. 1 (Aristotle Translated xii. 16). Is the flood that which
overwhelmed Helice and Bura (Heraclides Pont., fr. 46 W.)?
18. Though the fact that the Theages is quoted might have some sig-
nificance. Heraclides has been suggested as the author of this dialogue
(A. E. Taylor, Plato 5 534). Even the reference to Execestus could be from
Heraclides. He and Aristotle were ex.act contemporaries, and must have
known each other well in the Academy. This makes it impossible to pro-
nounce on their literary relationship.
19. Once Gauls have become Hyperboreans it is a short and logical step
to see the Rhipaean mountains in the Alps: Steph. Byz. s. 'Ywp{16pot
(Protarchus); Posidonius ap. schol. ad Ap. Rhod. ii. 675 (this scholiast
also tells us that Hee. Abd. said that the Hyperboreans ex.isted in his time).
20. This seems to have been a separate work from the wept a,KatOO'VV'l)S
(or owpos '1{1a.pw >..6yos Prod. in Tim. 141d = fr. 36 Voss), of which Abaris
may have been an alternative title (perhaps fr. 73 W. should rather be
assigned to the wep2 OtKar.oO'VVl'JS; but cf. fr. 75 W. with Iambl. V.P. 218).
 NOTES (pp. 158-159) 203

1TEp, Twv 1rpos '4/Ja.p,v o.va.cpEpoi.vwv naturally means 'concerning the say-
ings [or writings] ascribed to Abaris', (not 'remarks addressed to A.', which
Voss (57) thinks possible): it may have been about his XfYTIUf.1,0[ (Ap. mir. 4
(Theopompus); schol. ad Ar. Eq. 729; Lycurgus, fr. 85; Suid. s. :?1/Ja.p,s).
21. Accepting the convincing thesis of Boyance (ReV1JB des Etudes anciennes
xxxvi. 321 ff.) that Iambi. V.P. 215-21 is a summary of the 1rp, o,KaLOUWl)Si
I am sure that it is a summary of some Heraclidean dialogue, but not so
sure which one.
22. Cf. Dodds 161, n. 33; Meuli 160.
23. The function of the arrow in Aristotle's account (Iambi. V.P. 140-3,
if it is from Aristotle; Ross accepts it, Aristotle Translated 135) seems to have
been that of a guide: whether Abaris rode on it is not clear. Anyway, see
note 18 above.
24. Of the nicknames given to the 'pupils of Pythagoras' Abaris, Epi-
menides, and Empedocles [143], the last, in the form Kw>..vaave.a.s, was
known to Timaeus (D.L. viii. 6o),a younger contemporary and (hostile) user
of Heraclides (the most improbable tradition that Empedocles was a pupil
of Pythagoras also comes from Tim. (D.L. viii. 54), perhaps from the same
source as Kw>..vaa.ve.as). If this trio of nicknames were really coeval, as is
suggested by their conjunction in the later authorities, Heraclides must
almost certainly have been their inventor.
25. See Bidez-Cumont i. 8o-84; ii. 63-66. Aristoxenus (fr. 13) was using
Heraclides. Lydus (mens. iv. 42) provides the link whereby we can draw on
Antonius Diogenes (507 ff. = Phot. Bibi. 109 ff.) to extend our knowledge
of Heraclides' Zoroaster: what Dercyllis is said in Photius' epitome to have
heard from Astraeus about Pythagoras and Mnesarchus is given at greater
length by Porphyry (V.P. 10-14). Antonius claimed to have authority for
his fantasies in ancient writers; unfortunately Photius does not preserve
their names, but Heraclides was clearly one. Noteworthy also is the location
of the Sibyl in the moon in Antonius, with which should be compared a
passage in Plutarch, very Heraclidean in its interest in oracles and its pro-
posal of a 'scientific' explanation of them (Pyth. orac. 9. 39c-<i; cf. ser. num.
vind. 22. 566d; and, for the proximity of the moon to the island of the
Hyperboreans, Hee. Abd. ap. Diod. Sic. ii. 47).
Wehrli's assignation of frr. 69 and 70 to the ,Zoroaster cannot be supported:
the original of the magus who claimed at the court of Gelo to have circum-
navigated Africa was surely Sataspes (Hdt. iv. 62), not Zoroaster--even
though the historical Sataspes was unsuccessful.
26. o[ l .Jv ov] 'TOV Hpd. cpa.a,v Elva., 'TOV 1TO.'Ttpa 'TOV .J/Jov 'TOVOE 1Tavros,
_, \\ \ z
a.t\l\CL
I
wpoaaTp'TJV, .~~
Ka.,\ ws TOvo TOV ovo.aTos
f I,.. , l -.J.. \
'
y1<nevou 'M')V YP"'l''T/v'

lKOEOWKaaw, wawep Ka., J 'EmKoOpEtoS KoAW'M')S, I take it that Proclus

means that some texts of the Platonic myth had actually been edited with
the name of Zoroaster substituted for that of Er, and is not referring to the
apocryphal work of Zoroaster quoted by Clement (Strom. v. rn3), as Bidez-
Cumont think (ii. 160).
27. The real Zoroaster could never have made such a claim himself:
metempsychosis was not a tenet of true Mazdaism (Bidez-Cumont i. 28).
If Aristeas was anywhere said to have had a number of incarnations,
 NOTES (pp. 160-16'.i)
Democharis (25] might have been the father of one of them; but 'Suidas'
[Hesychius'] doubt about his subject's paternity may be ascribable to simple
confusion with some other Aristeas.
28. It must surely have been this antique Zoroaster who was descended
from 'one of the invaders from the continent beyond the Great Sea'. This
continent was originally a passing fancy of Plato (Tim. 24a), later taken up
and developed by someone on lines suggested by the accounts of the
inhabitants of Atlantis in the Timaeus and the Critias, and blended with
the myth of the discourse of Silenus to Midas mentioned by Aristotle in the
Eudemus (Ross, Aristotle Translated xii. 18 f.) (cf. Xen. Anab. i. 2. 13). The
author of this development may have been Theopompus (ap. Ael. V.H.
iii. 18 = Jacoby ii B. 551 f.), but it has the look to me of a Heraclidean fable.
This continent was 'beyond the Great Sea' (schol. ad Alcib.---cf. H.P., fr. 102
W.(157]), [gw ro&rov Tov Koa.ov, inhabited by a sort of super-Hyperbo-
reans (Theopompus).
I see that Bidez-Cumont (i. 14) suggest Heraclides as possibly the author
of the tradition that made Zoroaster the descendant of one of the immigrants
from this outer continent. The tradition would perhaps be at least as early
as Hermippus if I am right in emending Arnobius, adv. nat. i. 52 (Bidez-
Cumont ii. 15) 'age veniat, quaeso, per igneam zonam Magus inferiore
[interiore cod.] ah orbe Zoroastres, Hermippo ut assentiamur auctori'. I
find Bidez-Cumont's explanation of this passage, after Kroll, quite uncon-
vincing: the ignea .;ona is naturally the 3taKEKa.v.lV1J, and Zoroaster (or
his progenitor) may have come from a part of the great continent in the
southern hemisphere (often thought of as lower than the northern [117]).
Arnobius would be sneering at an impossible story: no one could travel
through the uninhabitable torrid zone!
29. AEYETa. 3E ws Ka., 1rpwTOS' a.OTOY Ala.KOY Myoi, Ka.! Aa.KE3a.,p.oYlois
'
1rpoEL1TOL ' V1T
rriv 'A '~ a./\(J)(1LV
.n.pKaowY \ o~ ' ,\,\' '
1rpo<11TOLT/ 7IYO.L 'TE 1TO 0.KLS' a.va.,..E,..LWKEYa.L.
Q Q '

The last word could merely refer to 'returning to life' after catalepsy (cf.
liva.{3wvs of Er in rep. x. 614b), but the ref. to Aeacus can only be under-
stood as meaning a previous incarnation. The text appears to be corrupt;
perhaps we should read 1TPWTOY [so Casaubon] avrov Ala.KOY .\lyo, <yeyo
vlva.,) (for the Greek cf. i. 25, a.vTos 3l </nJaw fl,OV'I/PT/ auTov ')IEyovlva.t).
Cf. Dodds 143.
30. Diogenes (i. 115) says it was at Orchomenus that they were taken;
either this is an event otherwise unknown, or it is a slip, perhaps of Theo-
pompus, who knew of the prophecy. KP71al here also seems to be a mistake
(for Aa.KE3a.,.ovlo,s: see note 29 above).
31. He is called a Crotoniate also by Cratinus in his Troplzonius (Meineke,
F.C.G .ii. 143; I agree with Meineke that there can be no doubt that he is
the same as the Phorrnio called oAaKwv by Pausanias and Clement: see his
useful discussion, ibid. 1227 ff.). Kock and Meineke assign this play to the
elder Cratinus, rather than to the younger (fourth century). Such slight
evidence as there is seems to me to suggest the latter: fr. 1 M. (221 K.), about
abstention from the Tply,\71, TpvywY, and fJ,EAal'Oupos, looks like a dig at one
form of Pythagorcanism, for which these creatures were tabu (D.L. viii. 33; cf.
the Hippocratic treatise 1ripl lpfj; vovO'ov 2; Arist. ap. D.L. viii. 19); and
 NOTES (pp. 163-173) 205

Pythagoreans were a favourite butt of the Middle Comedy (cf. Diels--Kranz

i. 479f.).
32. For a curious account of a modern 'journey in the soul' confirmed by
an apport see Nandor Fodor, On th4 Trail of the Poltergeist 91 ff.
33. A mysterious voice from heaven addresses Epimenides in D.L. i.
115 (from Theopompus). Cf. Cleonymus and Thespesius [148 f.]. (But
Bidez thinks Heraclides did not invent the Empedoclean translation: Vie
d'Empldocle 35 ff.)
34. Mi.iller (F.H.G. ii. 197 ff.) discusses various theories about its author-
ship. He is inclined to accept much of it (including this chapter} as ulti-
mately Heraclidean.
35. Cf. Empcdocles-D.L. viii. 60: also derived from Heraclides?
36. So the pious Aeacus (a previous incarnation of Epimenides, it will
be remembered) prayed on behalf of all the Greeks for delivery from a
drought (Isocr. Euag. 14 f.; Clem Strom. vi ( = Euseb. P.E. x. 2. 8); Diod.
Sic. iv. 61. 2). So too the Athenians pray on behalf of all the world for de-
livery from a universal plague (Harpocr. s. '1fJap1!:; cf. schol. ad Ar. Eq.
729).
37. Dionysus Brisaeus? Cf. Macrob. Sal. i. 18. 9; Et. magn. s. Bp10'aLOS'.
The Brisae taught Aristaeus beekeeping [167]. (On the name Brisae see D.
Detschew, Die thrakischen Sprachresle 89 f.)
38. Cf. also a fragment of Sallust (ap. Serv. ad Geor. i. 14); Nigidius
Figulus ap. schol. in Germ. Arat. 287; Paus. x. 17. 3 f.
39. Such a blending of the two personages might help to explain the fact
that later authorities sometimes call Aristeas Aristaeus (Pausanias actually
uses both forms [T. 4; 6]).
40. Asbolus the Centaur (surely a figure to appeal to Heraclides!) is
the '1.af3o>.os- olwvurrij, of the Hesiodic Shield ( 185) ; Euclus the Cyprian was
said to have prophesied that Homer would be born in Cyprus, and forecast
the Persian war (Paus. x. 24. 3; 14. 6) ; Pronapides was supposed to have
been a goodly poet and teacher of Homer (Diod. Sic. iii. 67. 5).
41. A7JTOVS' Tc Kai .'4.pTl.,oos- Kai )foo>.>.wvos yl,aw 017>.waat EV
.lA<:a,. ug. A117ous- TOKOl' [ )1pTl.tOOS' Kai 'A1ro>.Awvo5' jlVEO'IV]?
42. In the assignation of works to such characters as these we can see
the beginning of the process carried farther by Lobon and others later [25].
43. Euphorion made Homer a contemporary of Gyges (Clem. Strom.
i. 2 I = Jacoby ii. 579): because of a supposed connexion with Aristeas?
44. But Maximus uses the phrase elsewhere (x.wi. 1 a: diaxvO'EIS" 0aAa.TT1JS"
~ 'Q\I " Q\I)
1TOTa.wv EK,.,ol\aS, atpwv /LETa,.,o"as .
45. ( 1) The elegiac couplet with the pentameter first about the lovers
Chariton and Melanippus who conspired against Phalaris (fr. 65 W. Cf.
Ael. V.H.ii.4; Suid. s. cJ>&>.ap,,; Oenomaus ap. Euseb. P.E. v. 35. 3). It seems
to be based on the fourth-century Pythagorean tale about Phinteas and
Damon and the tyrant Dionysius (Iambi. V.P. 233 ff., from .\ristoxenus).
Phalaris' unwonted leniency to his would-be assassins earned him from
Apollo a postponement of his fate for two years-an idea suggested by
Hdt. i. 91. (2) fr. 138 W.: apparently an order to Xerxes to abstain from
sacrilege, which is a eonthtion of Hes. Op. ~7, Empedocle~, fr. 141, and
206 NOTES (pp. 174-181)
perhaps an oracle given to Jason of Pherae in 370 B.c. (cf. Parke in A.J.P.
lxi (1940), 78 f. Parke thinks that the Heraclidean oracle is genuine).
46. er. Tournier 45; Meuli 159.
47. er. Empedocles, frr. 134; 140. Apollo is frequently associated with
Pythagoras in later tradition.
48. It would be safer to say 'was confirmed in his adoption of it'; but,
whatever psychological causes we may believe to underlie the practice of
religious vegetarianism (Dodds 154; 175), the fact that there is no first-hand
evidence for this practice in Greece before Pythagoras-quite the reverse-
indicates that he introduced it, and prompts the question, Why?
49. My theory would imply that the authorship of Aristeas was evident
from the poem itself. This is not in the least improbable: he may have been
addressed in the vocative by other personages in it.

CHAPTER VIII
1. Though the Cimmerians of the Otfyss~ (xi. 13 ff.) have moved from
the world of fact to that offancy, and are not, I think, to be located anywhere
on this earth, the poet had heard from somewhere that the real Cimmerians
had dwelt in a 'region of darkness'-the north [40]. I believe that it was
Aristeas who first told the Greeks whence their Cimmerian invaders had
come (having himself been told by the Scyths): criticizing the theory that
the Cimmerian raids in Asia Minor gave an impulse to Greek colonization
in the Black Sea, Cook (79) says 'it would be remarkable if at that critical
time the Ionians had tried to plant colonies in the hornets' nest'; indeed it
would-if they knew where the nest was! But how could they know?
 TEXTS AND FRAGMENTS
I, HERODOTUS iv. I 3
[Arimaspeorum fr. 1] lcfn, Sl J.1p,crrh}c; &Kaiicrrpo{3tov aV7Jpllpo-
' ' ~ , I 8 , >lUv
T -~ ' ,1._ /3 I \
KOVVTJUiOc;, 1TOiWV 1Ta, aTTiKU a, <; 1vOJIOS y,vt Ol\a:1rroc;
' >T - S : I <:,\ t I >A \ # <:,
'YVOJ',IIOS, lUv,,.,ovwv OE VTTEpotKEflJI r1pi.aaTTOlJS avopas .ovv-
-"'8al\,ouc;'
..,.,, '\ t \ <:, \ I I
VTTEp OE TOt/'TWV TOVS XPVUvy,vl\aJ<aS ypUTTaS, TOV'TWV <, \
OE
\ A,/.,.(\

TOVS\ 'Y1TEp/3opEOlJS Ka7'7}KOJ/'Tas ETTL 9.1,

I ,
W\auuav. TOVTOVS I .,. 1TaJl'Ta<;
WV , , \

\ \ 'Y:1TEP/3opEwv
1Tl\7]V I > /: I
apc,;all'Twv >A
r1pia<11Twv > \
a,n
Tota, \ I
1Tl\7JU1oxwpo,a1
1
E1TiTI
' 8EU8at, Kat' VTTO
' ' Ell
' r1ptp,a<11TWV
'A t: 8'EEa80.t EK
E~W ' 7'7/S
XWP7J<;
'
'I<1<171Dovas, V1TO DE 'Ia<171S6vwv l:Kv8as, K,Eptov, Si oldoJl'Ta<; l1rl
I o~ \
< \ '<'I8 y > \ \ '
T'[) VOTL?J W\4UU?J VTTO "1KlJ EWJI TrtE'>Of-LEJ/OVS EKl\lTTHJ/ 7'7)11 XWP7JII.
I I

2, HERODOTUS iv. 16
r. ] ><, \ \ ><, \ >A I >\ I 1 ,
[ ff, 2 OlJOE yap OlJOE r1p1crrE7JS, TOV 1TEp OMYlf' 7TpOTEpov TOVTWJI
JLV7/f-L7JV l1ro1Ev71v, ov8i oo-ros 1rpoaw-rtpw 'la<17186vwv aVTO<; Jv
# ' .J.,.. , I 8 ~\\ \ ' I 8 #\
TOIUI ETTEUI 1TOIEWV E'f'11UE a1TIKEa a,, al\l\a Ta KaTV7rEp E El\t:yE
' ,I.). , , "' \ \'
aK"'lJ, 'f""S lUU7Joovas e111a1 -rou, -rav-ra I\EyoVTas.

3. TzETZES, Chiliades vii. 676-g; 686-g2

\ t ,h
KaL O -vEp11LKO<; 'Y'la, 1TEpt
I I 6'.-, ' fy:1TEp/3opewv,
I

waTTep Kat o ' z7)110'8 Et,, Ka.,'>A
oov r1ptcrrea,,
'
'A I
0t r1ptUTEac; .,..,J.,j,_
0t av,yvc;, V ..
0t TOV.-. n.avcrrpo /3'LOV,

001TEp aVTOS iv 1TEUIJI lvfrvxov oA{yo,c;

Kat''A I
r1ptcrrEas <>',I.
OE ' TOtS
'f'7JaLV Ell A '
r1p,.aa-rrnois
[fr. 3] ,,lUU7)oOi
.,, xai7?Ja1V
' '"'
ayal\l\Of-L110L .
TaVaTJOL
[fr. 4] ,ca2 turpast av8pcfmouc; ElVat Ka(}wep8Ev 6.ovpous
\ /3 opEw,
1rpos ' , 8''I\OUc;
\\ ' TE Kat\ Ea
1TOI\I\OVc; ' 1mp-ra
' '
.aX7rTas,
acf,vnovs (1TTTotcn, 1TOA'OpP7Jvas 1T0Av/3ov-rac;.
,.l.(J \ \ .,, # , # ' '
[fi 5] o'f' al\JJ,OV O
r. Ell EKacrros EXEi xaptEVTL f-LETW1Tlf''
xalT'[]ut(v) Aaaw,, 1TaVTWJI UT1/3apw-ra-ro1 av8pwv.

4. PAUSANIAS i. 24. 6
r. 6] ' \ ypV11as
, A I II
[ 1r TOV'TO!Jc; TOVS Ell TOtc; E7TEalJ/ r1p1crrEas O poKOJl-
inia,os- cfxt:aea, 7TEp2 TOV xpuuofJ 'P"f/<1111 )1p,a.UTTots TOLS inrEp
208 TEXTS AND FRAGMENTS
,r C:,I \ C:,\ I <\ ,I_ \I A > I \
LC1C17JOOVWV" 'TOV 01; xpvaov, ov .,,vl\atJCTOVC1tv o, YPV7TS, av,1:va, '7'1/V
A
yriv ELVaL
c:, \
0
14ptp,a.a1rovs
\ \
fJ.EV
,I c:,
avopa<; fJ.OVO'l'Oo.l\fJ.OVS
,JJl_!\ I
7r0.VTO.S 1(
'

A A C:,I 11 I ,, ' I \ C:,1" 1

,YEVTTJS, ypv1ta.s 01: 1111pia. 1\1:ovm 1:tKUUfUVO., '1TT1:pa. oE EXELV Ka.,
C1'TOp.o. aeToii. Ka.l ypV1rwV Jv 1Tip, Toaaih-a. 1:lp~u8w.

5. 'LoNGINus', de sublimitate 10. 4

~ \ \ \),4 I I ) .,. ,I ~ I
o ,uv yap Ta npip.a.u1rEta 1ro,11aas 1:Kova oieTai 01:wa
[fr. 7] Bail' r,fv Ka.1 Touro .lya cf,p1:ulv ff.J.t!rlpr,uw.
C:,
avopES ~C:,
vowp I
vawvaw ''(
a1ro XJ ' '1:v 1T1\ay1:uu1
ovos \I

~I I I , " \ ,, I
OVUTTJVO' 'TLVES 1:,uw, EXOVcn yap Epya 1TOV1Jpa,
,, > , JI ./.... \ ~> > \ I tl
op.a.T 1:v aurpo,a,, "l'"X1JV o EVt 1TOV'T<f' Exovutv.
'7 1TOV 1TOAAa 81:0,u, cf,Dias dva x1:,pas lXoVTES

wxovrm \ I
C11Tl\ayxvo,a, KaKws ava{J-(J.1\1\01:vo,u,.
\\ I A '

6. PAUSANIAS V, 7. 9
A A c:,, TT , , , , v Q
npw-raws OE O upOKOVV1JUW<;, /J,111}/J,1/V yap 7TOL1JUUTO i1T1:pl'o-
l \ ~- I JI \ \. I \ > ,.. I "
p1:WV Kai OVTO<;, Taxa aV Ti '/TI\EOV 1Tpi O.V'TWV 1T7TVUfJ,EVOS Et1}
KO.I
\ >T ~ I , do ~,I. (J ..J.. \ ,
I .... tl
1rapa l.UC17JDOVWV, ES ovs ....,,,Kf.(1 ai .,,11aiv f.V Tois f.1Tf.UW.

7 STRABO 21
, C:,\ \ \
TCI.XCI. OE Ka.L 'TOVS fJ.OVOP,P,aTOVS i>..VKI\W'ITO.S l:K T'l}S ,<,KV IK1JS
, Vf \ ' A '(1 e A

t I I
tUTop,a.s fJ,ETf.V1JVOXE" 'TOIOV'TOVS yap rivas TOVS" np,aa1TOVS I I \ 'A I cf,(1.0'IV,
A,,t , ,.,,., ,,t,
ous EV TOL<; npiaa?THOtS f.1T<1LV 1:VOEoWKEV r1.p,ar1:os O pOKOV
n
Y>]UIOS,

8. PLINY, Naturalis Historia vii. 10

sed iuxta eos, qui sunt ad septentrionem versi, haut procul
ab ipso aquilonis exortu specuque eius dicto, quern locum Ges
clithron appellant, produntur Arimaspi, quos diximus, uno
oculo in fronte media insignes. quibus adsidue helium esse
circa metalla cum grypis, ferarum volucri genere, quale vulgo
traditur, eruente ex cuniculis aurum, mira cupiditate et feris
custodientibus et Arimaspis rapientibus, multi, sed maxime
inlustres Herodotus et Aristeas Proconnesius scribunt.
g. AuLus GELLIUS ix. 4. 1-4
cum e Graecia in ltaliam rediremus et Brundisium iremus
egressique e navi in terram in portu illo inclito spatiaremur...
fasces librorum venalium expositos vidimus. atque ego avide
 TEXTS AND FRAG1\'1ENTS
statim pergo ad libros. erant autem isti omnes libri Graeci
miraculorum fabularumque pleni, res inauditae, incredulae,
scriptores veteres non parvae auctoritatis: Aristeas Pro-
connesius et Isigonus Nicaeensis et Ctesias et Onesicritus et
Philostephanus et Hegesias; ipsa autem volumina ex diutino
situ squalebant et habitu aspectuque taetro erant.
1 o. DIONYSIUS HALICARNASSEUS, Thucydides 23
H
yap maaw.,,oVTal TWV '7TI\HOVWV al ypacf,a,I .exp,
OV'T
t ~ , y ,... \ ,' ~ -
TWV Ka (}'
~.as xp6vwv ovB' al o,aaw{6.evat 1rapa 1TUUtv CiJS El<t;.{vwv ooaa, TWV
dvSpWv 7Ti<rretfov-rat lv al~ cla,v ai TE Kc1.8ov TOD M,Ar;alov KaL
'ApcaTalov Tov IlpotKOYVrJalov Kai TWv 1rapa1rll'fJ!1lwv TovTois.

1 1. 'SumAS'
.r1p1UTeas,
>,1 ' A '
'-'TJ/J,Oxapwos <'
TJ~ n.avaTpOl'IOV,
U HQI n '
po,KOWYJUIO<;, E1TO- '
, ' ,A ' ' I v ,, ~\ " I .-.
'7TOtoS" Ta .r1p,.aa1re1a Kal\OVeva 1TTJ E!JTI OE taTop,a TWV

'Y1repf3oplwv 'Ap,aa1rwv, {31{3>.ta y'. TOVTOV cf,aal 7'1JV if;vx~v, OTaV

e'{3OVI\ETO,
I\ 'l:: I
Esteva, I
Kai1 e1rav,eva,
' I'\
1TaMv. I
yeyove ~ \
oe 1",T "
Ka:ra\ n.powov Ka,I
Kvpov, 'O,\v.mao, v'. eypaif;e OE OV'TO<; Kal KaTaAoya.017v Beoyovlav,
' ,,
HS E7TTJ a. 1

Cf. Euooc1A, Violarium clvii

'AptUTlas Llri,oxapioos ~ KavaTpo{3fuv, IlpoKOVVTJULOS, E1T01TOIO;;.
UVVTE8TJKE Ta 'Ap1.a.<11Te1a KaAov.eva E1TTJ (EU'TI OE ZUTopla TWV
'Y1TEp{3opEWV 'Ap1aa1TWV, /31{3,\{a TpLa), eypatpe Kat i9Eoyovav 1TO.VV
KaAws.

12. HERODOTUS iv. 14-15

o
Kai o8ev ,ev -qv 'AptU'TE'tJS Taura 1TOl~aas, e i.p71Ka TOV OE 1TEpt
aVTOV 7JKOvov .\6yov ev IlpoKOVV7J<11p Kai Kv,LK<p, .\lfw. l4.p1aTl17v
yap Myovac, EOVTa TWV llU'TWV ovoevos ylvos imooelanpov' EUEABoVTa
.I..!.
ES KVa.,,.,,ov V
n I
poKOYVrJU<p ,
a1TO e .
ave,v, KatI TOV
I .J.'
Kva.,,ea \ ,
KaTaKl\7JL-
aaVTa TO epya~p,ov oixea8ai ayy1aMoVTa Tora, 1Tp0<1'1)KOVa, Tlf)
VKpif,. EUK0au.lvov OE 7)01) TOV ,\6yov ava. 7'1/V 1TOAIV CiJS n8vewc;
, ,,1 , , .i. fJ , ,, , ..
"'TJ o .r1p1UTTJS, ES a,.,,.,,,u autas 701a, Aryova, a1rtKVEeu8ai avopa
Kv~IKlJVOV ,jfKoVTa ;g 'Apa.KTJS 1r6AtOS', cf,avra UVV'TVXiv 'TE ol
LOV'TI
'I
l:1TI
, \ KV\,IKOV
Y'
Kai' <; , ' I
l\oyOVS' , 8
a1T1KU
:,
at. Kat 'TOV'TOV
~
ev
0950 P
210 TEXTS AND FRAGMENTS
, I '6'._ .f:l. I \ ~ \ I ,.. ,... , \ \
EVTE'T<lP,1/WS' (11-"'l'L0'/"'0.TEEtv I TOVS' OE 7Tp00'T)KOVTO.S' Tlp VeKp<p E1TI TO
,..~ I .,. tl \ I ~- t ) I
KV""t'7JWV 1T<lpHva, EXOVTas Ta 7Tpou,yvpa WS' ava,PTJUO/J.EVOVS'.
, 0 , ,:,, A , , ,, a. A ,, rA .L , ,,
<lVOIXC7EVTOS' OE 'TOV 0UC7Jp.a'TOS OV'TE TEuVEW'Ta OV'TE ":,WV'Ta 'l'atvEO'C7ill
>,I
np10'TE1JV, eT<l
I \ <, \
oe <Q<, I
e,.,oo,up II
ETU" ,l._ I
.,,._.vEVTo. > -
0.V'TOV ES
> np0KOVV1}0'0V I

A
'1T011JU<ll 'Ta E'TTEa 'Ta vw V1T
\ ,, \ A 'E'\\
IV\1}VWV
I >,I
np1.au7TEa
I _\ I
K<II\EET(ll,
/ 8' '.l.-
1TOl1JC1<1VTa E a.,,...v,au7JV<11 1ro11u,s
QA
TO
\ (> /
oeVTepov. TaVTa .ev ill A \ t /\ 1'
av-,a,
\ I I(> > '[. __ '1
(> \ . I -,S, U" I A
11eyova,, Ta.OE OE OLOO. lflE'T0.1TOV'TWOU1' TOLO'L EV 'T!J.1\1,1} UlJ)IKVP7JC1<lV'Ta
\ \ , .J..,! \ C, I 'A I JI I \
f'E'TO. '1"1]V a.,,...v,aw '1"1]V OEVTEPTJV npiO'TEW ETEUI TEC1C1EpaKOVTa Kai
<, I C > \ ,,,f:/.- \\ I > TT I \ 11
OITJKOUIOIO'I, WS' eyw
av,..,.,al\l\Op.VOS' EV HpOKOVV'TJO'Cf> TE Ka, mETa-
' a II " 7 .J.
7TOVT'</J evp,aKOV. lVlE'T0.1TOV'TLVOL '1'0.0'L 0.V'TOV
> \ I .J. I ,1.
ptC1TE1JV 'l'avev-ra ']'f''
14
' \ I \A Q ''"'" ,,:,, LI
ES' '1"1]V XWPTJV KEI\EVC1a ,.,w.ov n'TTOIV\WVOS' wpvuauua, Ka, np1C1TEW
I ,,,i
AnpoKOVV1JC1LOV E1TWVV/J,L1)V
TOV
I ) I '1
EXOVT<l
) ~
avoptaV'Ta 'TT<lp <lVTOV 0'7"1JC1<1L'
I ) _) \ ..._

.l..!
't""va,
'
I
yap .l.
a'I',
\
\
TOV 'A I\\ - -
.tl-7TOI\I\WV(l
I \ , I t t1
'I'Ta/\W'TEWV
-- \ - I
l) C' ,.. 11\
Jl.OVVOLO'L
I

11A
v,,
~

-1
,
(l7TLKEC111aL
I
I Q

~I
ES' '1"1]V XWPTJV, Kai aVTOS' OL E1TEO'VCl 0 vw EWY nptC1TE1JS'' TOTE 01:,
rt rt ""'.ll ,., 1 I i:. \ \ \ ' I "" ',J.
OTE EL7TE'TO Tf./) C7Ef./), Wo.L Kopa~. 1((1 TOY p.Y El'TTOVTa Tav-ra a'l'av,-
.J. I II \ I > ,I ~../.,. \ I ,/, \
a 811vm, u'l'eas OE lVlET(l'TTOY'TIVOI IIE")'OVO'L ES' ~E,"f"'US' 7TEJJ.'1'0.VTO.S' TOV
A (> \

LI ,
OEOV
, A q
E1TE1pwTav O TI TO .,,...u.a TOV a.Y PW'TTOV E7]. '1"1JY oE
, .l..! A B , ,, , <>' nVB',71v
.J.I
CT'l'eas KEIIEVEW
\ / '(j (j .J.I
7Tfl EO' a, Tf./) 'l'auaTI 1
A 8 I
7TEI oEvo,u,
C,\
0 aEWOV
II

, o .I.' <> t , \I, ,

CTVVOICTECJ' a,. Ka.L O''j'EO.S oeso.1/ovs TO.V'TO. 7TOL'lJO'O,L E7TITE/\EO,, KOL vuv
A A , A

EC17"1JKE av8pws E1Twvul7Jv exwv 14ptCTTW 1tap' av-r{j, 'Tij, ay&AaT1

A t,I ~ ~ . .
TOV n7TOI\I\WYOS', 'TTEPI~ OE av-rov oa'f'vai ECJ'Taa,' TO
7 C C, \ t I t, I ,l. t A I 8E\ >i
a.ya11a
\ >
EV

rfi ayopf/ r8p!Y1'0.I,

13. TzETZES, Ghiliades ii. 723-35

o14p,uTla.s JUV ul.os ~v TOV Ka.iia'1'po{3lov,
Tip ylve, IIpoiKoviii]aw, TWV EV)'EVWV Kat 7Tpmwv.
.. \ I <,, l) I I I I
OVTO, xa11KE'</J 1TapEl<TOVS' C7V'1JO'KEI Kai '11'L7T'TEI YEKVS',
\ I <,> tQ I C \ \ t A I \ a
KIIEIO'<l, Q EVC1WS' 0 XO.IIKEVS' EKEUIO 'f'O XO.IIKEIOY
'f'ot, CTU")'YEYEa1 TO oewov MyE1 'f'OV l4p1<1'f'Ea.
OL oE
f C, \ 8paoVTES
I t,;,
oovpms 7T<lVTES' 1TpOS TO xal\KEioV,
~ I \ \ \

' I/:.
aY015aVTES' t~\
OVOEV 1'
evpov, ovt VEKW
I '~'""
OVOE \,WV'Ta..
\ <, < \ ' .J. I _!\ t >A I
/J.ETO. O E7TTO. XPOVOVS 'f'<lYEIS' 'TT<II\IV O nptCTTEaS',
W \ >,I I \ I ,!,l.
E'TT'1] TO. np1p.a.C11TELa I\Eyo.eva. avyyp....,,E,.
, _!\
Kat 'TT<II\W ....,,aYL":,ETal TO oev-repov KO.I
.?.l. 'Y , <> , , ,
8V?IUKEI.
KaiI .eTa\ 8,aKOCTLa
I 8'IS' e11<oa, Ta ,I \ W
E'T'1]

J.cp' 'Hpo8oTOV ylyovE, Kat 'TT<lAW ave~&.V1/,

WC11TEp <pTJaiY 'HpoOOTOS' el o' GA7J8.ls oil ol8a..
 TEXTS ANO FRAGMENTS 211

Cf. Chiliades iv. 521 f.

KaL .:4purria, /1.,a.,
~ >,.f I f .J. t f K, .,
,.., QI
o r1pta-rEas o ao'l'os o 'TOU aua-rpol'wu,
' 8"1JUKEIV
<SC. 'T<p a'ITO ,
~ 'TE Kai\ ~'TJ"
Y" Ey,poVEI
',I. I .ya
'
>
14. APOLLONIUS, Mirabilia 2
>A I \ S,\rT r I " ,J. I a
r1pta-rEaV OE IGTOpH'Tat 'TOV 11pOKOV1l'YJUIOV EV 'TWt yva'l'E"t' 'T'Y/>

llpoKowfiaou 'TEAEV'T?)UaVTa V rfi a?nfj ~ipq. KaL V wpq.

.E1KEA~ {m;, 'ITOAAwv 8wprJ9fiva1 ypa..aTa 8100.aKOVTa. 88v 7ro,\-
, I - ~.... , R I \ .,J. .... I
I\UKI; UVT<p 'TOU 'TOIOVTOU au.l'awoVTO<; Kat 'ITEp..,,avou; y,yvop.EVOU

Sul ?ToMwv TWV KaL 7TVKVO'TEpov iv rfi .EiKEAUf ,f,aVTa,a.01ou ol

.EtKEAOL l.Ep6v TE Ka8wpvaaVTO ai!T[i, KUL Wuaav W<; ifpw1.

r 5. PLINY, Natural is Historia vii. r 74

reperimus inter exempla (sc. uti de homine ne morti quid em
debeat credi) ... Aristeae etiam (sc. animam) visam evolan-
tem ex ore in Proconneso corvi effigie, magna quae sequitur
hanc fabulositate.

16. PLUTARCH, Romulus 28

EOLKE .a,oJv Tawa [ the mysterious disappearance of Romulus,

his subsequent reappearance and deification] Tots u,f,' 'EM17vwv

'ITEpl J4p,a-riou 'Tau llpoKoV1l'YJalou Kal filEo.~3ou; TOU J4=a-
TE
, '
/\UIEWS .u ()011oyou.Evo1s.
\ ' r1p1a-rEaV
'" ' [,LEV
' yap
'
l' 'TIVI KJJU'F'</'
,I.~'

'TEAEurijaal ,f,aa, KaL 'TO aw.a J.LETt6VTwV ai!'Tov 'TWV ef,l\wv d,f,mJ;;
8at IIEYELV
OIXU \ S, I '8 \ >l:, > S, 1 >
OE 'TLVUS EU US E5 U7T007J(,LlUS 7JKOVTUS EVTVXHV
f "

llp,a-req. ti}V '111 Kp6rwvos 'ITOpEVoevtp.

17. ORIGEN, contra Celsum iii. 26

t3wEv 3J Kal cl P,ETa ra.wa Myn ci Ki,\aos, '1Tapan8i.Evos a'ITo
i:,
t
LGTOpLWV ,..
'ITO.paoo,.a I~
KaL\ Ka ()' aura
t \ \
P,EV )
a'ITtaTOIS I )
EOLKOTO. U'IT t )I )
0.UTOV A

~ \ > > I fl > \ "" \ Ii:, > ""' \ ,.. I \

OE OUK a.ma-rouEva oaov YE Em T"{/ /\Er,, (L avrou. Kat 'ITPWTOV ')'E Ta

'ITEPL' ''TOV l lpoKOV"TJULOV I ,A

r1p1aTav,I
1TEp1\ _ 'OU
f:
ravra ,..,.J..
'f'1JUlV" UEL'T 1'>

r1pLaTEaV
'" ' J.LEV
' 'TOV' Hp0KOV1l'YJC7LOV
rT ' a'f"'v,a
,1._ EVTa 'TE OVTWS oai.o11ws
()' ff ., '

>i; > () , \ ..() > a ,t._ I \ ,\\ a N

E1, av fJW'ITWV KaL au IS 1:vapyws 'f"'vEVTa KO.I 'ITO /IOtS vaTEPOl'

xpovot, 1ro.:\,\a.xou rfis olKovh"1], E1TL87J~aaVTa KaL 8auaard

, I\
ayy.111aVTa, Kat' TOU" .-.'ITOi\1\(tJl'OS
,. ,, \ , ,.,.
E7TIC7/(7/'l'aVTOS '7"015"
" MT07TOl'TLJ'OIS I
212 TEXTS AND FRAGMENTS
, 0EWV
EV - p,otp<f, VEfLEW, ' .rtpUTEaV,
TOIi >A I 'Y ()EOv."
TOUTOV OUoES ET 110/J,<,,EL A '"' ' ,,

"
eotKE "'' El,\1/'l'eva,
oe J., ' t<rrop,av
71111 , ' Illivoa.pov
a.1ro "'I Ka' 'Hpooorov
"'I [cf. 2 7
"fLJ" /<;- '
'TJ n.poooros .ev Ka. 'll'8 >,/, <:' fj'
w apos a.,,evoew TTapa ao, vo,,.,,ovra, ... ;] A \ \

[Origen then repeats T. 12 verbatim].

Cf. AENEAS GAzAEus, Theophrastus lxxxv. 993 Migne [a
careless compilation from Origen, it would appear]
lI{voapos 0~ 0 eri{3afos Kat 'Hpoooros J4.>..,Kapvaaad1c; )1.plCIT!a
, n poKOllll'YJGLOV
.1.. , TOV
'/'a.a, ' ELGEI\
\8'ovTa napa' TOIi KVa'i'Ea n poKOVV1JU'fl,
.1..' El/ '
lKei Ka~ Tf.Ovdvai Kat a.cf,av,aOIVTa Kv{U<TJVoi'c; c/mvEpwc; 8,a.\iyea9ai
<;-
Kat1 otaKouw,s I "
vaTEpov >f
1:T1:at KO. TEaaapaKoVTa I \
1:v >JT<J.11(!,
\ f mera- .I
, 'J.8A I I
7TOvrivo,c; o'i' TJVa,, Kat K1\VEV I
1:avTov vaia,s
\
n,av, Ka, rov
C ' () A \ \

A I\\ I I >fJ. >,f I\\ I I i: I

.rt7TOI\I\WVCL aVVE7TETO yap, E'/'TJ, T'f> .c:!.7TOI\I\WV TOTE KOpO.c,,, V77apxwv
A

< - >,f I , \ AA" I ,,,,/, A \,L_ I

0 VVV n.pc.<rrVS' KO.t 'l"OUS lfJETa.7TOVTtl/Ot1S 7!'1-"'f'O.VTO.S ES L.H:l\'f"'US.
, A u ,, , , , , A _ A

epwrav TOV ri.11'01\1\WVa f. XfYY/ TCf r.ip,aTet 7TE6ea 8ai Ka TT)V vuiav
, , , a, n
xf'Tiaat WS 7T{)0f1,El/0S a.,ewov Jara.i. Kat 1/UV O.VOptOS Cl'TTJKV
,Jf I\\
' '
E7TWVVJJ.tf!, t,,f
ri.punevs, ,
-,rap ' aVT<p t ...,
To/- ' _I\.
ay<J.l\fJ,O.TL TOU""' .N.7TOIV\WVOS. Kat'
.t. ,
ws eo,s a,..,..yorEpo,s
' () A
71 (J vata
.. I
KOWTJ\ vo,,.,,ercu.,y

18. ATHENAEUS xiii. 605c-d

<Papaa>..lq, -rfi <9maa.>..to, dpxriarp{o, odcpv17S' aTlcpavov xpvaow
<P,,\&,.,,,\os EOw,ce, Aa.ipaK'fJVWV a.va.871,a. avrfJ 17 <Papaa>..ta iv
ME-ra7TOVTU[J {J7T() TWV EV rfi ayopij. fJ,O.VTEWV, yeVOfJ,1ITJS cpwvfjc; EK
'T7}S' od.r:pllTJ, -rfjs XO.AKfjs' ~JI larryaav MET0.7TOVTUIO Kara rrjv )1.p,arla
-n , , " , ., , , _,_
TOV pOKOllll'YJGtOV emori,iav, OT E'i'TJG'EJ/ E!, 1: 'Y a , TTapayEyo-
11'EPtJOPEWV
1
VEva,, e
WS' I
-raxiara "..J..() ,
TJ S'w.,,\
TTJ" >
ayopav \ 't Q \ ....
Er'a11ovaa, '
t:avwv.,.

I - I ~ lfl-~>,-
yevoevwv 'T'WV ,aVTEWV oEC17Ta.av,, VTT a.vrwv. Ka. TWV <tV pw'Tl'WV ' " " ' ' ( ) I
"
va-repov 1 j' I
ava.,,71roVVTWV TTJII
I > f
atTtav < '()
evp1: <:' \
1/ ota I
"TOIi TOV- 8EOVA CJTE.,,a.VOV
IJ.

avr1PTJIV1J.

rg. MAXIMUs TvRrns x. 2 f.

llp0Kov71alcp &.v8pt T() ,ev aw,a EKEtTO EJJ,TTVOVV Iv, d.\,\' d,vopwc;
Ka\ eyyVTaTa
> I I
avaT01J' e
TJ< oe
S:, 1 ,/, > <;-
'f'VX'YJI EKovaa. I > \
-rou awaros, E7TAavii-ro ev
A -

- '() f <;- I \ () f \
T<fJ a, p, opvi oc; oKT)V, 1rav-ra U'Tl'O'Tl'ra EW/J,EVlJ, ')l'YJV, 1<at a.M:rrav,
>f () I I <I- ()

"A- > ~ ....

Ka\ 7TOTa.ovs, ,
Kat' '1TOAES,
I'\
KO.\ Evv,, avopwv, Kai' TTa ()71,a-ra,
,
KmI
,J.I I \ "'() l ~ I \ ..., \,
'i'vue,c; 1TaVTOtaS'' Kai a.v is Etaavo.E117J To aw,a Kat avaaT,Jaaaa,
WG'ITEP opyavcp xpw.EVTJ, m71yetTO C1TTO. l"''
" f f "
01/ Tf. KO.\ TJKOVGEV,
A >f "
1rap'
aAAOLS' aAAa.
 TEXTS AND FRAGMENTS 213

20. MAXIMUS TYRIUS xxxviii. 3c-f

, I
E)'EVE'TO Ka. EV
\ ' npoKOVVTJU<.p avr,p I , \ J..- \ I -.J..
y,WlOU"'f'OS' ovo.a np,<rrms.
1' 'A ,

,.,I.'
' -
7111,UTe,-ro <'\ '
oE aVTcp ovx, a"'f'1,0, 'Ta\ 11pW1"a, own 710EVa aV'T'TJ,
A ' \ A <' ' <' / , A

<' "'-' - \
omaUKat!OV .~.I.
11po"'f'EPEV. ~
\ ovv
1rpos i
W/ \ 'TWV a'v8pW1TWV
l 1'1JV ' ' '
a1r,u-nav A

JEvpEV ,\6yov E,PaUKEV 7'7]V ipvx!JV atn"~ KO.'TaAi'TTOVUaV 1'0 aw.a,

&.va1rTiiaav d,8v 'TOV al81.pos, 1rEpi1roMjua, 7'1/V yf]v 7'1JV 'EAAcf.8a Kal
\ /3' Q I I I \ I \ N I (J
1'1JV ap,-,apov, Km vr,uovs 1rauas, Km 11o-raovs, Kai OfYYJ yevEa a,
8, rijs 11ep,1roA17aEws at'n-fi Tl.p.a 7'7]V 'Y1TEp/3opl.wv yf]v l1101TTwaa,
~ \ I tt.::, I \ 11/l_. \ _ I \ .J.. I I \
OE 1TaV'Ta E~ '/> vo.a1,0, Ka 7/v '/ '110/11.TIKa, Kai y,VUES' xwpiwv, KO.
''
aEpwv p,ETa ol\a<;, Km avaxvaeis f3 \ '
8'-a11a1'T7J,,
\' \,
Kai\ r.o-rawv EKfJOl\as
Q \ ' ' A ,

,_{J "\ 8'

' \ \ 8 .I.L
yevEuva, oe avrn Km 1'1JV 'TOV ovpavov eav 1TOI\V 1'1JS vep EV ua-raTE-
A \ \ ' A , A A '

pav.

2 1. CLEMENS ALEXANDRINUS, Stromata i. 21

11poyvwaei ' "'' n e '
oe Km v ayopas o eyas 11poaaveixev au, .t:t '"/3apis
' ' -
TE O
,,.
Q,
1rep,-,ope,os
i Ka,, .t:tp,UTea<;
'" ,
0

poKOVVTJUIOS /TTlp,EVov,,s TE 0 n , ,"'- -~ .
v ' Oa'TIS
n.P")S, ' "-'1TaP1"7JV
ELS "' ' .?.1.'
""f'KE'TO, Ka'Zwpoaa'TP")S
' O'M-"'-
7/00S' 'EP,'1rE-

8oKAfis TE J ;4Kpayav-rivos ical <Poplwv J AaKWV, va, P,T}V lioAv-

cf.pa'TOS 6 @cf.cnos 'E1re86-n6, TE 6 .EvpaKovaws br{ re 'TOVTois
'[' I f >,f/l_ - I\
"-'WKpa1'1JS o nv,1vmos al\u,-ra.

22. !AMBLICHUS, Vita Pythagorica 138

Ka, TOV"T6 ye 1rcf.v-res ol IIv8ay6pnm Jw, lxovai 1r1aTc1micw,,
ofov 1Tep2 i1p1a7afov TOV IIpo1<0V7Jafov Kal 14{3cf.p,8o, rou 'Y1rep-
f3 opeov Ta
I \
v 8Ol\oyoveva,
\ I I \ \ '
Kat oaa a/\1\0. ro,av-ra /\E)'ETai. 1raa, yap - \ ' -

marevova, 'TOLS TOLOV'TOLS, 'TTOAAa 8, KO.L avrol 'TTELpwv-rai.

23. GREGORIUS NAZIANZENUS, Oratio iv (adversus lulianum i) 59
- ev
TaVTa , 1TO.L\,ETWUO.V
r, 1Tap EKEWOLS'
, 'EP,7TEOOK/\ELS
" , , KO.L, .t:tptaraw,
'" -
KO.L' 'E/L'TTEOOTLo,
<;, I ,
TtvES KO.L\ ,,. .I. '
i po'f'wv,o, KO.L' 'TOLOVTWV
' 8VUTVXWV -
' ( } ' ~ , , - ' " ' ' - - ,
apt os WV O P,EV TOI, .(JIK1\Ll<OLS' 1<pa7TJpatv 0.VTOV Ewaas, WS'' O '

cpETO Tip - 'f'll\'TO.Ttp
.1. ' ' aav8-''
al\lp KO.TE71vv'81/ 1rapa' 'TOV- 1rvpos '
, Q (JI '
EKfJpaa EV'TL o, OE O.OV'TOIS -na,v EaVTOVS' Ej'Kpvy,aV'TES V1TO
~ \ ,~ I \ ' ' , '.I. t \

- aV'T'TJ,
1'1JS' > - 11oaov
' .I. \ '
KaLI 'f',l\av-nas, > EMYX
e,r '\ 0'EV'TES', ov' al\
,-\.,\ov EK
' -
1'1J,
KAo'TTfis lT,~871aav ~ l1< TOV ~ Aa(}Efv Ka0vf3p{a81/aav.

24. PRocLus, in Platonis rem publicam ii. 113

\ I IJ.1
KCI.L yap e'I' 71.wv TWES' 11071 Kai
< -~ \ 8 ' <;,_t;
>
a1ro avetv Eo~av
I
Km V7J.auw
,
21.1- TEXTS AND FRAGMENTS
eve-re'(J11aav Ka\ ave,..,twaav
> ' ,QI \ ,, .J..(J C' \ t tJ,,!.
Ka, w'f' 11aav o, ev eyKav,1evm Tois
I
p.V1jp.a.<1tv, <:, \
0t OE Ka\ E'f' EC1TWTES' Ka (J0.7TEp
I
VI/ <>,I.>
5W. KO.\ E7T
' ' 'TWV L\ -
- 1TUl\<U
, , ...
y1:yovo-rwv u1-ropoVV7'at Ka, np,a-reas o poKOVV7Jaws Ka P/JhT'/LOS
, 'A , t n , , 'E ,
o' nAa1:,o,ev,os
rn. r ' Ka' 'Em,ev,071s
,1:, rr ' e-ra' (J'
o' .n.P1'JS ' -ro,s
ava-rov ev r-
~waw
'
yEvo,evo,.

25. STRABO 589

' -o'ev [Proconnesus] eC1Ttv
EV'TEV ' .t1pta-rEas
A ' o 1ro,717"1'Js
' -rwv
- .t1p1p,a.
A
I \ I ' ,.. ) \ f tl _W\ '\
<1'TrEIWV Kal\OV/J,EVWV E7TWV > O.V1jp ')'OT)S H 'TtS 0,/\I\OS

26. STRABO 639

'TIVS i>E o,SaaKaAov 'O,~pov 'TOV'T'OV [Creophylus] <f,aaw, ol 3'
OV 'TOV'TOV
' - ~\\>
(U\I\ n.p1C17'Eav
A ' 'TOV
' npoKOVV1)<10V,
'
Cf. EusTATHIUS, ad Il. ii. 730
'TIVES i>E Kat 8,SaaKaAOV 'O,~pov 'TOV Kpec!Jcf,v>..ov et7TOV' e-repo, 3
}1p1C1'Tafov -rov IlpoKo~a,ov, ciJs KaL -raV'T'a oyewypdcf,os [Strabo]
la-ropet.

27. EusEBIUs, Praeparatio Evangelica x. 11. 27 (= TATIANUS, ad

Graecos 41)
, ~, ... ,, ,
'TO oe vvv EXOV' <11TEVC1'TEOV /J,E'TO. 7T0.<17/S aKp,,.,nas u......,,71v11:,EIV ~,
, , , n , -~ ,y ,!IL">

ovx
o/J,'f/POV ,ovov 1rpEa,.,v-repos EC17'tV O mW<17/S, E'T OE Kai 'TWV 1rpo
> I I ,fl I I , t l(,f' "' tl ~ \ \ ,.. \

O.V'TOV avyypacf,lwv, Alvov, <P,M,,ovos, ea:op,Sos, :4,cf,lovos,

'Opcf,lws. Movaalov, Ll11,oS6Kov, <P.,,iov, .E,flv>J.,,,s, 'Em,evloov
TOV- K P1'JTOS',
' " 1S e,s
OC17' ' 'T1jV
' .411TaP'T1'JV
" ' ',/.'
a'f'IKETO, np1C1'TaLOV
A ' 'TOV
rr: pGKOVV1j<JLOV,
I
KO.I\ 'TOV- 'Ta\ np,,aa1r,a
A I ,-,!,/,
uvyy,,..,_.,,av-ros, >,f ,QI\
na,.,ol\OV 'TE

TOV-K,EVTavpov,
, Kai\ ,iaa-r,oos,
T , (> ' TE Kai' E"VKI\OV
Llpv,wvos
, \ -rov- n.V'Trptov,
1T '

Kat "Qpov -rov .Ea,lov, Kat Ilpova1tt8ov TOV i1871valov.

?28. 'SumAS'' s. Ilelaavopos IlEWWVOS

, ,:, , , - u , , , a a,, a' " 8, , u
7TOL'f/p.a.-ra oE av-rov npaKAELa, ev ,.,,,.,I\LOS' ,., Ean e -ra npa
\'
Kl\f!OVS' "
epya' "(Ja 7rpW7'0S'
EV - u \ p01rat\OV
npO.KI\E ,, .\ 7TEp1Tf'(JE1(E, 'TO.' oe
I>'

,!\ \
atl/\0.
-
'TWV I
7TOt1j/J,0.'TWV VO'8 (1 0.V'TOV
> - .. I: I)' I
001,a1:,eTO., ')IEVOp,EVa t
V'ITOI ,!\ \
'TE atll\WV
[This form of the name is used by
Kat l1p,aTlws Tov TrOt'f/'Toii.
Aeneas of Gaza, T. 17. The 'Pythagorean Aristeus' of Clau-
dianus Mamertus (Stat. An. ii. 7) may be Aristaeus rather than
Aristeas.]
 BIBLIOGRAPHY

A.J.A.: American Journal of Archaeology (Norwood, Mass., 1897-),

A.J.P.: American Journal of Philology (Baltimore, 188o-).
C.A.F.: Cmnicorwn AtticOTUm Fragmenta (ed. Kock; Leipzig, 188o-8).
C.Q.: Classical Quarter!, (London, 1907-).
C.R.: Classical Reuiew (London, 1887-).
E.B.: E,u;yclopudia Britannica (nth ed.; Cambridge, 1910).
E.R.E.: Encyclopaedia of Religion and Ethics (Edinburgh, 1go8-26).
F.C.G.: Fragmenta Comicorum Graecorum (ed. Meineke; Berlin, 1839-41),
F.H.G.: Fragmenta Historicorum Graecorum (ed. Millier; Paris, 1841-70).
G.G.M.: Geographi Graeci Minores (ed. Muller; Paris, 1855-61).
I.G.: /nscriptiones Graecae (Berlin, 1873-).
J.H.S.: Journal of Hellenic Studus (London, 188o-).
O.C.D.: Oxford Classical Dictionary (Oxford, 1949).
P.-W.: Paulys Real-Encyclopiidu der classischen Altertumswissmschaft (2nd ed.;
Stuttgart, 1894-).
R.M.: Rluinisches Museum fur Phi/ologu (Frankfurt a. M., 1842- ).

ATHANASIEV, A. N.: Russian Folk Tales viii (Moscow, 1863).

ATKINSON, T. W.: Oriental and Wtstem Siberia (New York, 1865).
BEAZLEY, C. R.: The Texts and Versions of John de Piano Carpini and William de
Rubruquis (ed. for the Hakluyt Society; London, 1903).
BETHE, E.: 'Aristeas' (in P.-W. ii).
BmEZ, J.: Eos (Brussels, 1945).
- - and CUMONT, F.: Mages helllnisis (Paris, 1938).
BoWRA, C. M.: 'A Fragment of the Arimaspea' (in C.Q. xlix (1956)).
BovANCE, P. : Le Culte des muses chez les philosophes grecs (Paris, I 93 7).
BURN, A. R.: 'Dates in Early Greek History' (in J.H.S. Iv ( 1935)).
CARPENTER, RHYS: Folk Tale, Fiction and Saga in the Hmneric Epics (Berkeley,
Cal., 1956).
CARRUTHERS, D.: Unknown Mongolia (London, 1913\.
CH1LDE, V. G.: The Dawn of European Civilization (London, 1927).
CLEMEN, C.: Religionsgeschichtliche Erkliirung des Neuen Testaments (Gicssen,
1909).
COOK, R. M.: 'Ionia and Greece, 8o<r6oo B.c.' (in J.H.S. !xvi ( 1946)).
CORNFORD, F. M.: Prindpium Sapimtiae (Cambridge, 1952).
CaciNERT, W.: 'De Lobone Argivo' (in Xcf.p,-rff F. Leo dargebracht; Berlin,
1911).
CuMONT, F.: Recherches sur le syrnbolismefuniraire des Rmnains (Paris, 1942).
DIELS, H.: Parrnenides (Berlin, 1897).
- - and KRANZ, W.: Du Fragmente der Vorsokratiker (Berlin, 1951).
Dooo.~, E. R. : The Grteks and the Irrational (Berkeley, Cal., 195 1) .
DUNBABIN, T. J.: The Greeks and their Eastmi Neighbours (London, 1957).
1116 BIBLIOGRAPHY
EBERT, M.: Sudrussland im Altertum (Bonn, 1921).
ERMAN, A.: Travels in Siberia (trans. Cooley; London, 1848),
FARNELL, L. R.: i Cults of the Greek Ci/)1 States (Oxford, 1896-1909).
- - ii Greek Hero Cults and Ideas of Immortali/)1 (Oxford, 1921).
FESTUGIERE, A.-J.: La Revelation d'Hermes Trismigiste (Paris, 1944-54).
FRANKEL, H.: Dichtung und Philosophie des fri.ihen Griechentums (New York,
1951).
GUTHRIE, W. K. C.: The Greeks and their Gods (London, 1950).
HAKLVYT, R.: V(!)'tlges (Everyman ed.; London, 1907).
HARLAN,J.: Central Asia (ed. F. E. Ross; London, 1939).
HAWKES, D.: Ch'u Tz'11, (Oxford, 1959).
HENNIG, R.: Terrae Incognitae (Leiden, 1944).
HENNING, W. B.: Zoroaster: Politician or Witch-Doctor? (Oxford, 1951).
HERMANN, A.: 'Issedoi' (in P.-W. ix).
HoLMBERG, U.: Fmno-Ugric and Siberian Mythology (vol. iv in The Mythofugy
of All Races, ed. MacCulloch; Boston, 1927).
How, W.W., and WELLS, J.: A Commenta7J' on Herodotus (Oxford, 1912).
Hue and GABET: Trmiels in Tarla'J', Thibet and China (trans. Hazlitt, ed.
Pelliot; London, 1929).
HUMBOLDT, A. VON: Asie centrale (Paris, 1843).
JACOBY, F.: Die Fragmente der griechischen Historiker (Berlin, 1923-).
KIESSLING, M.: 'Hypanis' (in P.-W. ix).
LANDOR, A. H. S.: In the Forbidden Land (London, 1898).
LAUFER, B.: 'Die Sage von den goldgrabenden Ameisen' (in T'oung Pao ix
(1908)).
LAWSON, J. C.: Modem Greek Folklore and Ancient Greek Religion (Cambridge,
1910).
Livv, I.: i Recherc'll4s sur Les sources de la legende de Pythagore (Paris, 1926).
_ _ii La Ligemk de Pythagore de Grice en Palestin (Paris, 1927).
LoRJMER, H. L.: Homer and the Monuments (London, I 950).
MEuLI, K.: 'Scythica' (in Hermes lxx (1935)).
M1NNs, E. H.: StzPthians and Greeks (Cambridge, 1913).
MoRRtsoN,j. H.: 'Pythagoras ofSamos' (in C.Q.xlix (1956)).
NILSSON, M. P.: 1 A Hislo7J' <if Greek Religion (Oxford, 1956).
- - " Greek Popular Religion (New York, 1940).
N1ssEN, T.: 'Die Aristeas-Legende im Idyll des Planudes' (in By.;:antinische
-<:,eitschrift xxxvi ( 1936)).
PAGE, D. L.: History and the Homeric Iliad (Berkeley, Cal., 1959).
PARKE, H. W., and WoRMELL, D. E.W.: The Delphic Oracle (Oxford, 1956).
PEARSON, L.: Ionian Historians (Oxford, 1939).
PERROT, G., and CHJPIEZ, C.: Histoire de l'art dans l'antiquite (Paris, 1882-
1914).
PHILLIPS, E. D.: 'The Legend of Aristeas' (in Artibus Asiae xviii (1955)).
REINACH, S.: Repertoire des vases peints grecs et itrusqias (Paris, 1899-1900).
RICE, T. T.: The S'i)lthians (London, 1957).
RoEllUCK, C.: Ionian Trade and Coloni,i;ation (New York, 1959).
RonDE, E.: i Psyche (trans. Hillis; London, 1925).
- - ii Der griechische Roman (Leipzig, 1900).
 BIBLIOGRAPHY 217

RoscHER, W. H.: L&.ikon der g,uchischen und romischen Mythologie (Leipzig,

1884-1937).
RoSTOVTZEFF, M.: i Iranians and Greeks in South Russia (Oxford, 1922).
- - ii Skythien und der Bosporus (Berlin, 1931).
SCHMID, W., and STAHLIN, O.: Geschichte der griechischen Literatur (Munich,
1920).
TARN, W.W.: The Greeks i11 Bactria and India (Cambridge, 1951).
THOMPSON, M. S.: 'The Asiatic or Winged Artemis' (in J.H.S. xxix ( 1909)).
THOMSON, J. O.: History of Ancient Geography (Cambridge, 1948).
TISCHBEIN, J. H. W.: Collecti011 of Engravings from Ancient Vases (Naples,
1791-5).
ToMASCHEK, W.: 'Kritik der altesten Nachrichten iiber den skythischen
Norden' (in Sitzungsberichte der Wiener Akademim der Wissenscheften cxvi
and cxvii ( 1888--g) ) .
TouRNIER, E.: De Aristea Proconnesio et Arimaspeo Poemale (Paris, 1863).
TYRRELL, G. N. M.: Apparitions (London, 1953).
VERNADSKY, G.: Ancient Russia (New Haven, Conn., 1943).
Voss, 0.: De Heraclidis Pontici Vita et Scriptis (Rostock, 1896).
WESTBERG, F.: 'Zur Topographic des Herodot' (in Klio (1904)).
WINTER, F.: 'Vase aus Mylasa' (in Mitteilungen des kaiserlich deutschen archiio-
logischen lnstituts (alhenische Abteilung), 1887).
YuLE, H., and CORDIER, H.: The Book of ser Marco Polo (London, 1929).
 INDEX LOCOR UM
AELIAN Hercher
de Natura Animaliwn
iii. 4 67, 68
iv. 27 65, 67
ix. 36 192
xi. I 42
xvii.21 67
Varia Historia
ii. 4 205
26 174
iii. 18 204
iv. I 77
AENEAS GAZAEUS Migne
Theophrastus
hocx.v. 993 T. 17
AEsCHYLUS
Prometheus Vinctus Murray
85 189
301 48
415 50, 60
417 ff. 47, 191
420 53
707 46, 55
707-13 47
709 55, 191
714-16 47
717-35 49
720 f. 55
721 97
722 f. 52
724 f. 51
725 f. 191
729-35 46, 55
731 49
790 f. 50
79 1 60
791 f. 54
frl, IOI
795
803 ff. 45, 68, 72, 188
805 63
806 189
807 If. 60, 63
l!!lO INDEX LOCORUM
Aeschylus: Prometheus Vinctus {cont.)
809 189
8111 189
829-41 46
993 44, 64
101!1 63
Hypotheses 46
Septem contra Thebas
728
818
Supplices
538 ff.
544 ff.
fragmenta
191 56, 58
193. 28 46
195 54, 64, 96
197 52
198 71
Sidgwick
90
401
AGATHEMERUS Muller
Geographiru ltifomiatw (G.G.M.)
i. I 40
3 56, 58
ALCAEUS
Lobel-Page
fragmenta
307
ALCMAN
ap.Steph.Byz.s.'lucni86ves 5 Meineke
ap. Strab. 43 10
fragmenta
Diehl
!. 59 43 (Anth. fyr. Gr. 3)
59 40
AMM!ANUS MARCELLINUS
Gardthausen
Res Gestru
xxiii. 6. 66 106, 112

AMMONIUS
In Aristotelis De Interpretatione
Busse
133. 17 1!02
ANANIUS
fragmenta Diehl
(Anth. {yr. Gr.3 )
4 3 189
 INDEX LOCORUM 221

ANONYMI
Oxyrhynchus Papyri xiii Grenfell-Hunt
1611 184
Papyri Graccac Magicac Prciscndanz
iv. 736 138
Paradoxographus Vaticanus Rohdii Keller
(Rerum naturalium
scriptores Graeci
minores (Teubner))
186
49 186
50 186
Periplus Ponti Euxini Mi.ill er
49 190 (G.G.M.)
.ANTONINUS LlBERALIS Martin
Metamorphoses (Mytlwgraphi Graeci
ii (Teubner))
20 69, 192

ANfONlUS DIOGENES Hirschig

de incredibilibus (Erotici Scriptores
507 190 f., 203 (Firmin-Didot))
508 186
510 187
APOLLODORUS Frazer
Bibliotheca
ii. ,. 3 46, 6o
5. II 189
Epitoma
iii. 17-20
APOLLONIUS Keller
Historiae mirabiles (Rerum naturalium
scriptores Graeci
minores (Teubner))
199
2 T. 14; 120, 129
3 122
4 203
5 199
6 143
APOLLONIUS Rnomus Seaton
Argonautica
i. 211 ff.
281
569-74
111111 INDEX LOCORUM
Apollonius Rhodius: Argonautica (cont.)
ii. 516-117 167
519 168
710 14
1001 47
iv. 1140 13
1318-23 184
.ARATUS Maass
Phaerwmena
36-44 198
299 116
ARCHILOCHUS Diehl
fragmenta (Anth. ryr. Gr,3)
12, I 13
21 15
ARISTEAS Bolton
Arimaspeorum fragmenta
5, 112, I IO, 134, 138,
171, 186
2 1 33
3 5, 8, 16, 19, 68, 185,
187
4 8, 16, 17, 18, 19, 46, 63,
I 16, 185, 186
5 8, 13, 19, 46, 72, 91, I 10
6 32, 64, 65, 93
7 8, 9, 13, 14, 15, 18, 19,
26
ARISTIDES Keil
Orationes
xxvii. 5 141
ARISTOPHANES Hall-Geldart
Pax
8117 ff. 201
Plutus
352 185
925 1611
Thesmophoria.,usae
639 13
ARISTOTLE
de Anima Ross
i. 3 202
de Caelo
Allan
ii. 285b 198
 INDEX LOCORUM 223

Aristotle (cont.)
Ethica Nicomachea Bywater
iii. 5. 4 14
Historia Animalium Dittmcyer
v. 552b17 50
Meteorologica Fobes
i. 8 202
13 24, 42, 52, 59, 97, 189
ii. 6 198
de Republica Atheniensium Kenyon
I 161
Rhetorica Ross
ii. 23 129, 189
fragmenta Rose (Teubner)
3 192
6 16o
10 202
34 16o
37 202
44 204
191 143, 174
192 175
194 205
Eudemi fragmenta Ross
JI 155 (Works ef Aristotle
translated xii)
[ARISTOTLE]
de mirabilibus auscultationibus Apelt
178 201
de Mwido Lorimer
391 146, 148
394b 198

ARISTOXENUS
Wehrli
fragmenta
13 203
31 205
67 185
Marchesi
A.RNOBIUS
adversus Nationes
i. 52 204
Hercher-Eberhard
ARRIAN
Indica
15 66
Periplus Ponti
II 189
224 INDEX LOCORUM
ARTEMIDORUS geographus
ap. schol. ad Dion. P. 14 190 Ben1hardy
ATHENAEUS Meineke
Deiprwsophistae
xiii. 605c-d T. 18; 123, 128
AUGUSTINE Dornbart
de Ciuitate Dei
xxii. 28 149
AULUS GELLIUS Hosius
Noctes Atticae
ii. 22 198
iv. 11. 14 200
vii. 6. 6 27
ix. 4. 1-4 T. 9; 27, 187
6 ff. 28
7 30
13 ff. 30
AVIENUS Berthelot
Ora maritima
644-73 188
CALLIMACHUS Pfeiffer
Hymni
iii. 25 14
76 195
iv. 28-33 184
36 13
62 ff. 195
291 23, 185
vi. 65 14
fragrnenta
75. 6 18
32-37 167
186. 8-12 23
12 185
260. 62-67 184
608 18
CELSUS philosophus
ap. Orig. c. Gels. iii. 26 127 Koetschau
CICERO
pro Fiacco
Clark
43 131
in Verrem
II. iv. 128 Peterson
169
de Divirwtione.
i. 130 Muller
1G7
 INDEX LOCORUM !2!25

Cicero (cont.)
T usculanae Dispulationes
i. 44 f. 147, 148
v. 8 f. 153
de &puhlica
vi. 16 149, 202
fragmenta philosophica
ix 185
[CICERO] Nobbe
no. 283 (Consolatio) 185
CLEARCHUS Wehrli
fragmenta
7 148
8 148, 151
10 200, 202
CLEMENS ALEXANDRINUS Stahlin
Stromata
i. 117. 9 205
131 20
133. 2 T. 21; 122, 129, 157,
162, 164
v. 103 203
vi. 26. 8 187
28 205
CLEOMEDES Ziegkr
de Motu Corporum ctUlestium
i. 7 187
Co NON Jacoby
fragmenta (F. gr. Hist.)
I. 18 164
CRATINUS senior Kock
fragmenta (C.A.F.)
221 204
39 62
Trophonii fragmenta Meineke
204 (F.C.G.)
9 204
Jacoby
CTESIAS
(F. gr. Hist.)
fragmenta
45 26 65
6o 184
Jacoby
DAMASTES (F. gr. Hist.)
fragmenta
39, 41, I 10
6~60
Q.
 1

226 INDEX LOCORUM

DICAEARCHUS Wehrli
fragmenta
36 200
Dm CAssrus Boissevain
Historiae Romanae
boci. 16 198
D10 CHRYSOSTOMUS von Arnim
Orationes
xxxv. 24 192
DIODORUS SICULUS Muller
Biblwtheca
ii. 43 23
47 185, 192, 203
iii. 67. 5 205
iv. 56 58
61. 2 205
82 167
viii. 32 164
xvi. 30 166
xviii. l. l 202
xxxi. 10 200
xxxiv & xxxv. 2. 6 135
DIOGENES LAERTIUS Hicks
Vitae Philosoplwrum
i. pr. I f. 160
12 153
25 204
IIO 161
I II 161
ll4 161
!15 204, 205
v.86 172
go 173
91 173
92 20
viii. 3 16o
4f. 199,202
7 f. 20
rg 205
33 201, 204
40 199, 200
41 144
54 203
60 203, 205
67 ff. 145
68 130
84 201
 INDEX LOCORUM 1127

D10NYSIUS HALICARNASSEUS
d, Thucydide Usener-Radennacher
23 T. 10; 31
Antiguitates Romanae Jacoby
i, 61. I 189
DIONYSIUS J'ERIEGETES Bernhardy
Orbis Descriplio
315 53
656 17
663-5 53
846 17
953 185
DIONYSIUS (?PERIEGETES) Lchrs--Diibner
Omithiaca (Poet. bucol. et
i. 2 72 dida&t. (Finnin-Didot))
EMPEDOCLES Diels-Kranz
fragmenta (F. Vors. 6 )
I 165
4 I 18
3 14
105, I 13
Ill. 9 165
112 154
115. 3 14
121, 3 14
134 146, 206
135 201
140 206
141 205
146 Hlg, 154, 165
ENNIUS Warmington
Saturarum fragmenta (&mains of Old
115 73 Latini)
EPHORUS Jacoby
fragmenta (F. gr. Hist.)
42 71
16o 51
EPICHAR.MUS Diels-Kranz
fragmenta (F. Vors.6 )
7 2 184
ERATOSTHENES
ap. Strab. 65 58 Meineke
ETYMOLOOICUM MAGNUM Gaisford
Bp,<10.io, 1105
228 INDEX LOCORUM
EuooCIA Flach
Violariwn
clvii T. II; 32

EUPHORION
ap. Clem. Strom. Stahlin
i. 117.9 205
EURIPIDES Murray
Alcestis
966 ff. 20
Bacchae
755-8 200
1078-83 166
Hecuba
377 f. 185
Helena
587 f. 199
Hercules Furens
4o8 52
Hippolytus
953 f. 20
lphigenia Aulidensis
796 ff. 20
lphigenia Tawica
1306
Orestes
990 13
Phoenissae
1058 188
Rhesus
530 198
EusEBIUS Dindorf
Praeparatio evangelica
v. 35. 3 205
x. 2. 12 205
3. 6 ff. 199
II. 27 T. 27; 16g
EUSTATH!US
in Dion. P. Commentarii Bemhardy
14 191
311 41
663 53
in Hom. II. Commentarii
Leipzig ccl., 1825-5o
ii. 730 T. 26
FAVORINUS
ap. Aul. Gell.
ii. 22
Hosius
rg8
 INDEX LOCORUM 229
GALEN Kuhn
opera
viii. 414 155
xv. 109 185
GALLUS Baehrens
fragmentum p. 336 190 (F. poet. Rom.)
GEMINUS Manitius
Elementa Astronomiae
6 187
GREGORIUS NAZIANZENUS Paris ed., 1840
Orationes
iv (adv. lulianum i) 59 T. 23; 123, 144
HARPOCRATION Bekker
Lexicon
71.fJap,s; 172, 205
KinihJVYJ 62
l:Kta:TTo0!, 184
HECATAEUS ABDERITES Jacoby
fragmenta (F. gr. Hist.)
7 185, 192, 203
8 24
II 24
12 42
13 59, 190
14 24
HECATAEUS MILESIUS Jacoby
fragmenta (F. gr. Hist.)
100 17
108 17
154 17
163 17
185 104
193 197
195 190
203 17, 47
204 17
207 17
210 188
216 190
284 17
287 17
299 17
302 58
323 17
17
335
1130 INDEX LOCORUM
HELLANICUS Jacoby
fragmenta (F. gr. Hist.)
187b 71
18g 4,13

HERACLIDES PoNTICUS
fragmenta Wehrli
17 173
117a-f 153
46 167, 1102
46b 168
50 168
51 158, 192
51c 157
65 168, 205
68 159
69 203
70 203
73 202
74 157
75 157, 165, 203
79 155
Bo 156
85 145
89 200
go 157
91 202
92 152, 202
93 151
97 152
102 171, 204
117 171
130 165, 168, 170
131 165
136 168
138 168, 205
141 167
157 170
Voss
36 202

HERACLIDES (?PoNTicus)
Muller
de Rebus Publicis
(F.H,G.)
9. 2 167
l-IERACLITUS
fragmenta Diels-Kranz
120 (F. Vors, 6 )
188
'I!

INDEX LOCORUM 231

HERMESIANAX Powell
fragmcnta (Coll. Alex.)
7 85 f. 201
HERMIAS Couvrcur
in Platonis Phaedrum
I05 137
HERMIPPUS Muller
fragmcnta (F.H.G.)
23 144, 199
8o 204
HERODIAN Lentz
1TEp, .ovlJpovs MtEws
i. 9. 17 184
HERODOTUS Hudc
Historiae
i. 28 47
47 134
66 161
91 205
125 IIO
142. 2 198
168 199
201 f. 6o, 104, II I
202 6o, 112
215 18g
216 77, 193
ii. 23 6o
31 200
44 129
76 195
142 131
iii. 38 77
99 77
102 ff. 66, 67, 8o
II I 67, 192
115 6o
116 64, 67, 110
iv. 5 ff. 43, 76
8-10 188
II 59
T. 1; 5, 43,110, 171
13 T.1;T.2;T.12; I ff.
13-16
T. 12; 119, 130, 133
14 f.
T. 12; 25, 32, 125, 127,
15 130, 166, 170
16 T. 2; 17, 133
232 INDEX LOCORUM
Herodotus: Historiae (cont.)
18 104
19 47
20 43, 102, 104
21 104
22 105
23 33, 42, 71, 105, 107,
108, 109
24 105
25 39, 42, 102, 106, 108
26 76, 110, 187
27 7, 44, 110
31 44
32 17, 23, 33, 100
33 23, 196
35 185
36 33, 62, 156, 158
45 55, 6o
57 53
62 203
64 f. 186
65 78, 187
73 ff. 125
95 144
IOI 104
105 103
108 f. 104
I 10 ff. 51, 63
123 53, 107, 108
169 162
170 f. 117
174 116
191 192
v. 49 5-7 17
vi. 86)12 184
vii. 140. 2 184
HESJOI>
Opera et Dies
Rzach
27 205
83-88 184
333 9
618-23 184
Scutum Herculis
185 205
261-6 184
439 195
Tluogonia
143 18
 INDEX LOCORUM 233
Hesiod (cont.)
fragmenta
54 47
62 103
63-64 56
134 16
138-g 13
Catalogi fragmenta Traversa
4-8 47
49 24, 73

liESYCHIUS ALEXANDRINUS Schmidt

Lexicon
1TEVKaATa, 185
XaAKiS,tj 48
X&)..v{Js 47

HIEROCLES paradoxographus Muller

fragmenta (F.H.G.)
3 24

HIPl'OBOTUS
ap. D.L. viii. 43 201 Hicks

lilPl'OCRATES Jones
de Aere Aquis et Locis
15 IO

17 118
19 41, 44
de M orbo sacro
2 205

Holl!ER
Iliad Monro-Allen
i. 53-58 184
189 12
ii. 606 187
iii. 3 ff. 194
ix. 5 40, 192
154 16
296 16
x. 465-70 184
xi. 224 201
555 15
xiii. I ff. 192
5 71
xv. 171 42
:i34 INDEX LOCORUM
Homer: Iliad (cont.)
xvi. 7g8 18
843ff. 202
xix. 358 42
XX. 217 18, 185
220 185
222 17, 185
xxii. 331-6 184
xxiv. 125 13
213 14
544 f. 117
Odyssey Allen
i. 10 184
50 45
126-31 184
326 f. 170
iii. 169-72 116
170 18
v. 271 ff. 197
277 197
331 f. 115
335 13
vi. 42 ff. 35
viii. 266 ff. 170
492 ff. 170
X, I ff. 93, 116
82-86 187
xi. 13 ff. 206
xiii. 110 18
xiv. 300 198
403 f. 198
xix. 4-g 184
xxiv. 57-62 184
Hymni
iii. 73 13
396 123
vii. 28 f. 196
xxxiii. 15 13
ap. vit. Hom. Hdt.
262
9
458 18
HYGINUS
Fabul~ Rose
186 201
IACOBUS DE VORAGINE
Legmda aurea Grasse
37 2 163
..

INDEX LOCORUM 235

!AMBLICHUS
tk Vila Pytluzgorua Deubner
31 175
81 201
91 192
135 f. 143
138 T.22;142
140--3 203
215-21 203
218 203
222 16o
233 ff. 205
265 142
267 142, 158, 169, 170
TMologumena Arithmeticae de Falco
41 142
INSCRIPTIONES
!.G. xiv. 1374 1g8 Kaibel
!SOCRATES Benseler-Blass
Evagoras
14 f. 205
Panegyricus
153 62
IULIANUS
ap. Suid. s. 'E:,rea&n/W~ 152 Adler
luSTINUS Ruhl
Epitoma
xx. 3 164
'LoNG1Nus' Pridard
de Sublimitate
8.4 135
9. 5 27
IO. 4 T. 5; 8, 9, 26, 38, 186
5 f. 186
13. 2 136
16. 2 135
LU CAN Housman
de Bello ciuili
i. 674 ff. 140
iii. 272 f. 53
277-g 59
28o 73
iv. 106 ff. 41
viii. 464 187
540 191
ix. 1-14 147
858 41
236 INDEX LOCORUM
LUCIAN Fritzschc
Philopseudes
22 ff. 202
LYCO~HRON Mair
Alexandra
3-6 135
146o 134
LYCURGUS Blass
fragmenta
85
LYDUS Wunsch
de Mensibus
iv. 42 203
76 1g8
MACROBIUS Jahn
Saturnalia
i. 18. 9 205
in Somnium Scipionis Commentarii
ii. 5. 22 198
MANILIUS Housman
Astronomica
i. 690 202
758 ff. 202
MARCIANUS Muller
Periplus Maris exteri (G.G.M,)
i. 44 192
MAXIMUS PLANUDES von Holzinger
(Cleodemus)
8 f. 33
66 34
78 ff. 34
144 34
151 34
173 34
MAXIMUS TYRIUS Hobein
Philosophumena
X.I 156
2 f. T. 19; 121, 155, 171
xxvi. 1a 205
xxxviii. 3 c-f T. 20; 36, 122, 129, 134, 147
MEGAST!l.ENES Jacoby
fragmenta (F. gr. Hist.)
23 66
27 b 83, 99, 186, 194
 INDEX LOCORUM 237
MELA r'rick
Chorographia
i. !09 189
115 45, 52, 191
116 51, 112
117 71
ii. I f. !06
9 w6, 187
!06, 187
iii. 37 193
38 59, 6o
MIMNERMUS Diehl
fragmenta (Anth. [yr. Gr. 3)
6 193
JI. 5-7 57
NEARCHUS Jacoby
fragmenta (F. gr. Hist.)
8b 67

NIGIDIUS FmuLUS
ap. schol. in Germ Arat. Eyssenhardt
287 205
NONIUS Muller
Compendiosa Doctrina
152 193
NUMEN!US
ap. Procl. in Plat. remp. Kroll
ii. 129 152

0ENOMAUS
ap. Euseb. P.E. Dindorf
v. 35 3 205

0LYMPI0D0RUS Norvin
in Platonis Phoedonem Commenta,ia
238 152
0RIGEN
Koetschau
contra Celsum
ii. 16
iii. 26
Zangemeister
0ROS1US
Historiae adversum Paga11os
i. 2, 4 53
45 99
238 INDEX LOCORUM
'ORPHEUS' Dottin
Argonautica
74g-5o 58, 59
1036-82 58, 59
1063 72
Ovm Merkel-Ehwald
Metamorphoses
v.86 54
xiii. 846 f. 195
PAUSANIAS Schubart
Descriptio Graeciae
i. 24. 6 T. 4; 32, 33, 64, 65, 93,
205
31. 2 196
ii. 21. 3 161
iii. 7. 3 161
16. 2 f. 161
19. 11 ff. 164
v. 7. 9 T. 6; 33, 205
vi. 9. 6 ff. 129
viii. 2. 7 67
25. 12 187
x. 5. 9 192
14. 6 205
17. 3 205
24. 3 205
32. 6 141
PHERENICUS
ap. schol. ad. Drachmann
Pind. Ol. iii. 28 22
PHILOPONUS Hayduck
in Arist. Meteor. i Commentarium
117. II 151 f.
PHILOSTRATUS Conybeare
Vita Apolwnii
iii. 48 192
Vi, I 192
lO f. 192
PHILOSTRATUS irmior Fairbanks
Imagines
ii. 18. 3 195
PHOTIUS Bekker
Bihlwtheca
46 b 66
109 a 191, 203
109 b 186
 INDEX LOCORUM 239
Photius: Bibliotheca (cont.)
IIO bf. 187
133 b 164
., PINDAR Bowra
-~ Olympia
L~ iii. 13 ff. 52
'1-
,f viii. 46 52

~
Pythia
ix. 5 ff. 192
63 ff. 130
x.29 35, 7 1
31 ff. 69, 71
45 62, 71
fragmenta
94. 4 f. 70
II6 201
127 129
192 70
272 71, 98
283 54, 158
302 146, 148
PI.Aro Burnet
Apologia
39c 202
Charm ides
158 b 158
Cratylus
403a 65
Epinomis
986 e ff. 198
Euthy<kmus
299 78
Gorgias
524a 152
Leges
642d 161
Phaedo
62-68 146
109b 150
110b 150
II2 e ff. 150
Phaedrus
238c 136
426 e ff. 150
248 r. 129
Res puhlica
614 ff. 150 f.
614b 204
240
Plato (cont.)
Thttutetus
i73e
INDEX LOCORUM

146, 201
l
.l,,
)

Timaeus ;j
24a 204
38d 198
[PLATO] Burnet
Axiochus
371 a 185
de iusto
374a 14
Pu.urns Lindsay
Aulularia
701 f. 73
Rudms
629 ff. 162
Stichus
30 73
Trinummus
933 ff. 189
PLINY Mayhoff
Naturalis Historia
ii. 119 198
iv. 78 53
84 50
88 41, 42, 44, 65, 102, 106
89 187, 192
90 99
v. 85 f. 189
vi. 15 189
19 51, I 12, 190
21 106
34 f. 71
50 106
51 70
55 99
64 186
100 ff. 197
129 189
vii. 9-26 28, 33, 44, I 02
9 104
10
T. 8; 44 f., 64 f.
21 192
23 184, 192
28 30
173 ff. 153
174 T, 15; 121, 128
207 30
I INDEX LOCORUM 241

I
.\
Pliny: Natura/is Historia (cont.)
xi. I II
xxix. 52
xxx. 3
4
45, 192
67
16o
199
5 16o
8 16o

PLOTINUS Brehier
Enneadu
v. 8. 10 136

PLUTARCH
adversus Coloten Bernadakis
14 159
de E apud Delphos
20 200
de Facie in Orbe Lunae
25 156
de Gmio Socratis
21 f. 146, 149, 151
23 199
non posse suaviter vivi secundum Epicurum
22 162
de Pythiae Oraculis
8 166
9 203
17 192
de sera Numinis Vindicta
22 149, 151, 203
Vita Camilli Sintcnis
22 157
Vita Marcelli
20 137
Vita Pompeii
48 135
Vita Romuli
28 T. 16; 128, 129, 201
Vita Solonis
12 16!
Bcrnadakis
[PLUTAR.CH]
de Fluviis
5.3 53
14 45, 19 1
Bcthe
POLLUX
Onomasticon
iv. 76 102
R
6S~6
242 INDEX LOCORUM
PORPHYRY
Vita Pythagorae Nauck
10-14 203
29 143
45 200
ap. Euseb. P.E. Dindorf
x.464d 199
ap. Stob. i. 447 5 202 Wachsmuth

Posmomus Jacoby
fragmenta (F. gr. Hist.)
!03 202

PRJSCUS PANITES Muller

fragmenta (F.H.G.)
30
PROCLUS
in Platonis Rem publicam Commentarii Kroll
ii. !09 159
113 T. 24; 148
119 151, 202
122 153
129 152
in Platonis Timaeum Commentarii Diehl
ii. 8 202
PROTARCHwS
ap. Steph. Byz. s. 'Y1Tp/36p0, 202 Meineke

PTOLEMY Nobbe
Geographia
vi. 16. 5 99, I06
16. 7 rn6, 112
16. 8 99
viii. 24. 3 & 5 I06
24. 7 99
PYTHEAS
ap. Strab. 104 58 Mdneke
QUINTUS CuRTIUS Foss
de Gestis Alexandri Magni
vii. 3 I 191
SALLUST
Maurenbrecher
Historiarum Reliquiae
ii. 6
iii. 73
I
l
j
SCHOLIASTAE
INDEX LOCORUM

ad Acsch. P.V. 436 190 Wecklein

437 54
743 18g
819 62
830 73
ad Ap. Rhod. i. 212 195 Wellauer
1321 47
1323 47
ii. 375 47
378 4,8
4g8 167
675 202
57
52, 57, 58
ad Ar. Eq. 729 203, 205 Dilbner
Pac. 1270 185
(David) in Arist.
Cat. 28a 185 Brandis
ad Dion. P. 10 45, 53, 58 Bcrnhardy
14 190, 191
666 18g
ad Germ. Arat. 287 205 Eyssenhardt
ad Hom. Od. i. 371 144 Dindorf
(Comm. Bern.) ad Luc.
vii. 756 67 Usener
ad Pind. Pyth. v. 10 162 Drachmann
x. 72 62
ad Plat. remp. 6oob 158 Stallbaum
ad [Plat.] Alcib. I 122a 159, 16o
ad Soph. El. 62 201 Elmsley
o.c. 1248 187
'SCYLAX' Muller
Periplus (G.G.M.)
83 190
88 47
'ScvMNus' Millier
Orbis Descriptio (G.G.M.)
871 ff. 190
878-85 51
SENECA
Hermes
Dialogi
vi. 25 Haase
Epistulae morales
go. 16 Gercke
.Naturales Quaestiorus
vii. 3. 2 198
R2
6358
 INDEX LOCORUM
SERVIUS Thilo-Hagcn
in Vergilii Carmina Commentarii
ad Aen. xi. 532 185
ad Geor. i.14 205
34 151
SEXTUS EMPIRICUS Mutschmann
adversus Mathematicos
ix. 20 f. 202
S1M1AS Powell
fragmcnta (Coll. Alex.)
I 10, 68, 184
6 184
SIMONIDES Bcrgk
fragmenta (Poet. {yr. Gr.4)
3 195
SOLON Diehl
fragmcnta (Anth. {yr. Gr,3)
15. I 13
SOPHOCLES Pearson
Antigone
983-5 94
Electra
62-64 145
Oedipus Coloneus
1248 40
1621-65 166
fragmcnta
637 195
STATIUS Phillimore
Silvae
i. 2, 51 ff. 202
STEPHANUS BYZANTIUS Meineke
Ethnica
14.pd[ai 184
14.pdt,,s 189
14.p,.a.U7Tol 23
14.<1<16s 184
Evttpylra., 191
'Jc,C1'1/8611tts 5, 8, 68, 184
Kop~ol 188
Ta.pKVVla. 24
y'IT0.11!; 50
'Y'1Tttp{36ptto, 23, 202
STOBAEUS
Wachsmuth-Hense
Anthologium
i. 176 142
378 152
447 5 202
 INDEX LOCORUM 245
STRABO Meineke
Geographica
21 T. ?; 32, 33, 170
43 10
47 40
65 58
ro4 59
183 46
201 77
299 192
300 f. 71
302 71
388 187
421 192
493 53, 190
494 50
512 f. 59
513 77
5 19 186
520 77
531 189
551 141
587 132
589 T. 25; 32, 35
590 132
639 T. 26; 170
645 200
675 136
684 40
687 99
705 67
706 66
710 f. 192
711 83, 99, 186, 194
'SUIDAS' (THE SUDA) Adler
71{1ap,s 171, 203
.'Apurrlas T. 11; 25,121,126
'Ep.1ria&n.os 152
'Em.ivl871s 156, 161
K,a8~V7J 61
'Op<f,iv, AHMBpwv 20
Ililaav8po, lIElawvos T.28; 187
172, 205
<Pa>.apis
<Pop.lwv 162
Schwartz
TATIAN
Oratio ad Graecos
T. 27; 169
41
246 INDEX LOCORUM
TERTULLIAN
de Anima Lindner
28 200, 201
THEOCRITUS
Idyllia Gow
xi. 50 195
xxii. 121 13
XXV, 85 ff. 185
246 13
THEOGNIS
Diehl
Elegi (Anth. (yr. Gr.l) '1l
274 14
THEOPHRASTUS
Hort
Historia Plantamm
iv. 4 I '
99 I'.
'
,.
THEOPOMPUS
Jacoby
fragmenta (F. gr. Hist.)
69 205
248 123, 128
392 162
TIMAEUS
Jacoby
fragmenta
(F. gr. Hist.)
30 203
85 58
TZETZES
Kiessling
Chiliades
i. 275 II
ii. 723-35 T. 13; 11, 127
iv. 521-2 T. 13
vii. 676-g & 686- T. 3; 8, 22
92
683 9, 67
695 ff. 68
V ALERIUs FLAccus
Argonautica Kramer
v. 584 192
593 70
596 192
vi. {8 ff. 188
193 192
201 192
220 192
618 192
750 73, 184
 INDEX LOCORUM 247
VARRO ATACINUS Baehrens
fragmenta (F. poet. Rom.)
14 I f. 187
20 192
VARRO JlEATlNUS Kent
de Lingua latina
vii. 71 73
x. 75 185
VEGETIUS Lang
Epitoma Rei militaris
iv. 38 198
Vrnms SEQUESTER Oberlin
de Fluminibus etc.
II 190
VIRGIL Hirtzel
Aeneid
i. 50 ff. 93
Eclogues
viii. 27 37
Georgies
i. 240 f. 198
XANTHUS Jacoby
fragmenta (F. gr. Hist.)
32 159
XENOPHANES Diels-Kranz
fragmenta (F. Vors. 6 )
3.5 17
20 161
XENOPHON
Anabasis Gemoll
i. 2. 13 204
ii. 6. 9 13
iv. 5 34 47
V. 5 I 47
vii. 7. 6 15
Hug
Cyropaedia
ii. 3. 6 13
vii. 4 & 5 189
viii. 6 189
Marchant
Hellenica
iv. 5. 7 15
v. 2. 4 15
Memorabilia
iv. 7 5 198
ZENOTHEMlS
9, 22, 67, 113, 115, 187 Kiessling
ap, Tz. Ghil. vii. 683
 SUBJECT INDEX

Abarimon, 28. Amphion, 16g, 170.

Abaris, 54, 124, 156, 157 f., 164 f., Anaxagoras, 198; cult of, 129.
172,201; oracles of, 21, 157, 158, Anaximenes, 42, 188, 198.
203; and Pythagoras, 142 f., 156, ancestor-worship, 76 ff.
158, 164, 169, 203; Theogony of, Androphagi, 102, 104; see also
32, 199. Anthropophagi.
Abasci, 189. Antandrus, 4.
Abominable Snowmen, 28, 186, 194. Anthropophagi, 28, 31, 44, 75, 83;
Academy, 159, 172, 183, 202. see also Androphagi.
Achilles, 164. Anticites, river, 50; see also Kuban.
Acragas, 16g. Antimachus, 23.
Adriatic Sea, 46 f., 196. ants, gold-guarding, 64, 66 f 8o f.,
Aeacus, 161, 204, 205. 111, 177, 191; monstrous, 81 f.,
Aeetes, 57, 70. 85, 103, 177, 193.
Aeolus, 93, II6; see also Winds, Aparctias, 198.
Cave of. Apicius, 124.
Aeschylus, and Arimaspea, 45 ff., 111, Apollo, 2, 120, 124, 126, 127, 128,
181, 182. 129, 130, 141, 158, 164, 166, 167,
Aethiops, river, 63, 189. 172, 174, 182, 200, 205, 206; and
Agathon, of Samos, 45. griffins, 62, 89; and Hyper-
Agesias, of Megara, 186. boreans, 23, 52, 62 f., 71, 100,
Ajax, 164. 141, 143, 170, 174f., 179, 182,
Ak-Su, 108. 192, 196 f.; possession by, 1, 69,
Ala Ku!, 95 f., I 18. 74,123,126, 134ff., 14of., 179,
Aleman, and Arimaspea, 5, 7, 40 f., 183, 192, 200; seealsorf,01{J6)."f/1TTOS.
43,111,181,184. Apollonius Rhodius, 11.
Alexander, the Great, 99, 118, 181 ; apport, 34, 35, 162 f., 205.
Polyhistor, 184. Arabia(n}, 53 f., 67, 18g f.
Alexandria(n), 18, 20, 25 f., 38, 41, Arabis, river, 190.
69, 182. Arakan, 77.
Alice in Wonderland, 1. Arai Sea, 107, 113.
Alps, and Rhipaean Mountains, Araxes, river, 58, 59 f., 107, uo,
202. l 12, 189; see also Syr; Volga.
Altai Mountains, 36, 92, 94, 100, Arcadia(n}, 161, 167, 187, 189.
107 f., 108, I 14, I 15, I 18. Arctic Ocean, 53, 58, 59, 60, 74, 84,
Amalcius, sea, 24. 191.
Amazons, and griffins, 37, 38, 52; Aremphaei, 71; see also Argippaei.
location of, 38, 49, 50 ff., 6o, 63, Argippaei, 42, 71, 101, 102 f., 105,
75, 118, 178, 18o, 181, 191. 106, 107 f., 108 f., 109f.,112, l 13,
America, South, 77. I l 4, [ I 5, l 78,
Amometus, 99. Argonauts, 56 f., 58, 64, 75, 162,
Amphiaraus, 199. 18o, 181, 190, 191.
 SUBJECT INDEX 249
Ariaspi, 191. 179, 183; ancient stories about, 2,
Aridaeus, see Thespesius. 25, 32, 35 f., 69, 70 f., 73, 119 ff.,
Arimaspae, 72 ; see also Arimaspi. 133 f., 138, 140, 143, 145, 147 f.,
Arimaspea, contents of, 1 ff., 4 f., 148, 153, 154, 156, 157, 163, 164,
25 f., 28, 31, 32, 33, 38, 39 ff., 166, 167, 168., 170, 174f., 183,
43 ff., 45, 6o, 63, 64 f., 67 ff., 204; modern theories about, 36,
71 f., 72, 74 f., 76, 82, 92, 94, 98, 111, 124 ff., 132 ff., 145, 146, 160,
1or f., 102, 103, 110, 111, 134, 179 ff., 200; see also Aristaeus;
138, 171 f., 174, 176 ff., 199; date Aristeus.
of, 3 ff., 10, 19, 40 f., 43, 110, Aristeus, 212, 214; see also Aristeas.
124f., 132,170,176,178; influ- Aristotle, 157, 173, 190, 202.
ence of, 7, 18, 24, 26, 31, 33, 35 f., Aristoxenus, 172, 185.
36 ff., 39 ff., 70 f., 73, 80, 111, Artace, 2, 119, 133, 200.
181 ff., 187, 188, 192; name, 2, Artemis, 69, 89; 'Winged', 6, 89.
25, 32, 169, 181; nature of, 35 f., As, 114.
125 f., 132 ff.; quotations from, 5, Asbolus, 16g, 205.
7 ff., 20 ff., 27, 46, 63, 68, 72, 91, aschy, 105, 107, 109, 197.
1Io, 116, 173, r82, 185; ditto Asia, 48, 50, 56, 75, Bo, 82, 84, 93,
analysed, 11 ff., 185. 94, 96, 97 ff., 100, 101, 106, 107,
Arimaspi, 67, 68, 73, 107, 111, 112, 108, 109, 111, 115, 116, 118,
113, n6, 171,176,177,191, r93, 132 f., 150 f., 171, 176, 177, 178,
196; in art, 6; description of, r, 180, 181, 182, 190, 191, 192, 206.
8 f., !.!8, 32, 39, 45, 61, 63, 64, 72, Assyria(n}, 4, 86 ff.
74, 83, 170, 177, 180, 194 f.; ety- Astomi, 30.
mology, 110, 198; Hyperborean, Astypa!aea, 129, 163.
22 ff., 25, 185, 186; Scythian, 23, Athena, 32.
67, 70, 102 ; originals of, So, 83 ff., Athens, Epimenides at, 161 ; Hera-
93, !08 f., 1 IO, 118, 186, 194. elides Ponticus at, 172 ; and world-
Arimaspus, king, 23. wide plague, 205.
A rim phaei, 71, 112 ; see also Argi p- Atlantic Ocean, 150, I 71,
paei. Atlantis, 24, 204.
Aristaeus, 35, 123 f., 124, 126, 130, Atlas, among Hyperboreans, 189.
145, 167 ff.; of Proconnesus, 31, Attacori, 99; see also Uttarakuru.
33, 16g, 205; Pythagorean, 142, Attalids, 20.
201, 214; s11 also Aristeas. Australia, 77.
Arist1a, ship, 183. Autoleon, 164.
Aristeas, grammarian, 185; of Pro- Avars, 54, 1 71 f., 1go; see also
connesus, and Arimaspea, 1 ff., 9 f., Abaris.
22, 25, 27 f., 31, 32, 39, 45, 64 f., Azov, Sea of, 4, 178, 179; see also
7 I f., 76, 79, 8o, 82, 92 f., 94, Palus Maeotis.
1oof., 102, rn4, 11of., 113,114,
I 16, I 19, 124 f., 170, 172, 176, Babylonia(n), 69, 86, 88, 159, 160.
178, 193, 197, 199, 206; other Bacis, 21.
works of, 25, 31 f., 186 f.; date of, Bactria(n), 65 f., Bo.
3 f., 73, 110, 111, 117, 118, 126 ff., baihak, 84.
131 ff., 169 f., 172, 179, 183; form Balkhash, lake, 98.
of name, 35, 130, 205, 214; and Baltic Sea, r 13.
Pythagoras, 142, 144, 16.f, 1H f., Baltistan, 11o.
250 SUBJECT INDEX
Bamboo Books, 101. cannibalism, 76 f., 83, 178, 18o f.;
Bashkin, 107, 108, 109, 197. see also Androphagi; Anthropo.
Battus, 162, 163. phagi.
Belaya, river, 1o8, 109, 115. Cantharidae, 121, 199.
bilocation, 120, 126, 143. Carambycae, 24.
Bion, of Proconnesus, 187. cardinal points, 115 ff., 182, 197 f.
bird-soul, 126, 133. Carpini, John de Piano, 77, 94 ,
Black Sea, 3, 7, 47, 64, 75, 76, 91, 96, 97
I 13, 132 f., 141, 179, 191, 206; Caspasus, river, 70.
see also Euxine; Fontus. Caspian Sea, 59, 6o, 112, 192, 193.
Boeus, 69. Cassandra, 134 f.
Boii, 78. catalepsy, 138, 139 f., 153 ff., 161,
Bora, mountain, 196. 165, 168, 201, 204; see also soul-
Boreas, 8, 23, 34, 35, 40, 41, 42, 94, journey; trance.
100, 104, 115, 116, 117, 141, 177, Cato, the younger, 135.
179, 18o f., 182, 185; Cave of, 28, Caucasus Mountains, 4, 5, 46, 49,
41, 44 f., 55, 74, 93 ff., 102, 118, 54,64,90, 112,150,181, 18g, 190,
176, 18o, 195; see also North 198; and Rhipaeans, 52 f., 54 f.,
Wind; Winds, Cave of. 74, 177, 18o, 181 f., 189; see also
13orysthencs, river, (Dnieper), 28, Rhipaean Mountains.
29, 47, 191. Cayster, river, 131.
Bosporus, Cimmerian, 47, 49 f., Caystrobius, 1, 25, 131 f., 179.
55 ff., 59, 1go; Thracian, 4 7, 48, Ccos, 167 f.
56, 57, 191. Chalcidice, 48.
Branchidae, 200. Chalybes, 4 7 ff.
Brisae, 167, 205. Chao Hun, 81.
bronze, 48, 189. Charilaus, 161.
Buddhism, 98. Chariton, and Melanippus, 205.
Budini, 79, 104, I 06, t 08, t 09, 112, cherubim, 93.
114,197. Chinese, 100, 101, 112, 177, 193,
Bug, river, 50; see also Hypanis. 195; cannibals, 77; and Hyper-
Bulgars, 171. boreans, 100 f., rn9, 180; on
Bulunggir, river, 108. monstrous peoples, 10, 81 f., 103,
Bura, 202. 194.
Buriats, 83, 97. Ch'iung-ch'i, 82; see also tiger,
Byzantium, 36, 72, 183, winged.
Chlonthachonthlus, 20.
Ch'u T;;,'u, 81.
Cadmus, hero, 161 ; of Miletus, 31, Ch'Uan Feng, 82; see also Dog Jung.
32, 187. Chung Mountain, spirit of, 82, 195.
Caere, plate from, go. Chussovaya, river, 108.
Callatiae, 77. Cicero, Consolatio of, 21.
Callimachus, II, 25. Cimmeria(n), 2, 4, 47, 52, 75, 176,
Cambyses, river, 192, 181, 191,199,200; in Odyssey, 206.
Campania, grypomachy from, 36. Cimmeris, city, 24.
Campasus, river, 68 ff., 192. Cisthene, 61 f.
Campesus, 70. Clazomenae, 121, 123, 200.
Campylinus, river, 67 f. Cleinis, 69, 192.
 SUBJECT INDEX 251
Cleomedes, of Astypalaea, 129, 163. of Miletus, 111 f., 113; of Syra.
Cleonymus, 148 f., 152, 153, 205. cuse, 205.
Cleopatra, Boread, 94. Dionysus, 8g, 166, 168, 179, 200,
Cnidians, 166. 205; Hymn to, 196.
coelacanth, 183. Dioscuri, 161 ff., 168.
Colaxais, 43 f., 178, 188. divination, 35, 122, 123, 126, 128,
Colaxes, 75, 178, 181, 188; see also 134 ff., 139, 142 f., 154 , 156,
Colaxais. 157, 158, 159, 162, 163, 165, 166,
Colchis, 50, 60, 106, 189, 1go. 169, 172, 183.
Colotes, 159. Dnieper, river, see Borysthenes.
constellations, 197 f. Dodona, 23, 46, 196.
continent, outer, 159, 165, 204. Dog-heads, 29, 75, 83, 103, 192,
Cornelius Labeo, 149, 152. 194; see also Dog Jung; Half-
Cragalidae, 199. Dogs.
cranes, and pygmies, 194. Dog Jung, 82, 85, 193; see also
Crete, griffin in art of, 86 f. Ch'iian Feng.
Croesus, 126, 158, 199. Don, river, 59, 107, 1o8, 109, 190;
Cronian Sea, 53 ; see also Arctic see also Tanais.
Ocean. Drymon, 16g.
Cronus, 53. Dzungaria, 103, 108 f., 118.
Croton, 128, 142, 143, 162, 164, Dzungarian Gate, 95 f., 100, 116,
174 f., 182,201. II8, 176.
Ctesias, 27, 65 ff., 110.
Curium, grypomachy from, 88. edinookie, 83.
Cyclops, 28, 32, 170, 195. Egypt, 35, 48, 6o, 63, 86 ff.
Cyprus, 87, 205. Elis, 155.
Cyrene,ny,:nph, 126,130,167,192; Elixoea, Hyperborean island, 24.
place, 161, 162, 163. Empedocles, 123 f., 130, 143, 144,
Cyrus, king, 126, 199; river, 192. 145, 152, 153 f., 154, 157, 163,
Cyzicus, 2, I 19, 133, 141, 179, 200. 164, 165 f., 168, 173, 203, 205.
Empedotimus, 123 f., 145, 151 ff.,
Damon, and Phinteas, 205. 157, 164f., 169,202.
Dandarii, 1go. Enguium, 137 f.
Danube, river, 196; see also Istrus. Enkomi, grypomachy from, 87 f.
Dardistan, 191. Epicureans, 188.
Darius, J03. Epigoni, 23.
'Darkness', 'Kingdom of', 80 f., 180. Epimenides, 32, 124, 143, 148, 153,
Delos, 23, 196, 198. 156, 157, 160 f., 164, 168, 169,
Delphi, 2, 120, 123, 128, 129, 158, 170, 203, 205.
164, 166, 173, 196, 199. Er, 150 f., 153, 165, 201; and
Democharis, 25, 204. Zoroaster, 159, 203.
Democritus, 198. Esarhaddon, 4.
Dernodocus, 169, 170. Essedonians, 5, 70, 73, 111, 184; see
Demosthenes, and 'Marathon Oath', also Issedonians.
135. Ethiopia(n), 60, 62 f., 192.
Derbices, 77. Etna, Empedodes and, 124, 154.
Diogenes, of Tarsus, 136. Euboea, and Hyperborean gifts,
Dionysius, Metathemenus, 20, 21 ; 196.
252 SUBJECT INDEX
Euclus, of Cyprus, 169, 205. 1, 28, 32 f., 39, 45, 61, 65 ff.,
Euphorbus, 200. 72 f., 85 f., 92 f., 177, 18o; ety
Euripides, 164, 181. mology, 85, 92 f., 18o; location
Europe, 48, 50, 56, 59, 64, rn6, 110, of, 1, 6 f., 22, 24, 28, 39, 40, 45,
111,114,182, 190. 61 ff., 65f., 72, 73, 74, 91, 102,
Euxine, 47, 50, 57, 59, 64, 164, 181, 107, 110 f., 114, 118, 171, 176,
191, 192; see also Black Sea; 177, 181, 192; origin of, 83 ff., 93,
Pontus. 186, 194; and other monsters,
'evil eye', 29. 83 ff., 93, 186, r 94; see also
Execestus, of Phocis, 157,202. grypomachy.
grypomachy, 6, 36 ff., 52, 64, 87 ff.,
178, 181.
'Feathers', 'Region of', 42 ff., 65,
Gyges, 205.
74, 101, 102, 176, 180; see also
Pterophoros.
Fife, Sciapods in, 186. Hades, located in the air, 149, 151 f.,
Finns, 94, I 07. 202.
forgeries, literary, 19, 20 ff., 25,
Haemus, Mount, 40, 47, 168, 195.
31 f., 142, 159, 173, 182, 185, 187, Hakluyt, rn, !02.
1 99
Half-Dogs, rn, 68 f., 82 ; see also
Dog-heads.
Gades, 59. Han Annals, 112.
Ganges, river, 150. Harpies, 4 7, 89.
Gaul, 202; serpents of, and magic Hazaras, 79.
egg, 67; and Rome, 157, 158, 165. Hecataeus, of Abdera, 24, 43, 73,
Geloni, 104 f., mg, 113, I 14. 190; of Miletus, 17, 21, 40, 111,
Gelonus, town, 104 f., 108. 190.
geographers, ancient, 17 f., 19, 40, Hegesias, 27, 30, 186.
42, 63, 111, 117, 177, 181 f. Helen, 164.
geranomachy, 194. Helice, 202.
Gervase of Tilbury, 195. Hellespont, 56.
Ges clithron, 28, 31, 45; see also Heraclea Pontica, 172.
Boreas, Cave of. Heracles, 76, 151.
Gibraltar, Straits of, 56; see also Heraclides Ponticus, 144, 151 f.,
Hercules, Pillars of. 157, 167, 169, 170 f., 189, 199,
gold, in Eurasia, 45, 61, 64 ff., 107, 203,205; on catalepsy, 153 ff.; on
109, 110, 114, 118, 177, 180; early 'miracle-men', 143, 145,
guarded, 1, 32, 37, 45, 64, 65 ff., 146, 152, 159, 160, 165, 183, 200,
74, 8o f., 85, 110 f., 171, 176, 177, 202; life and personality of, 172 f.;
18o; sacred, 43. and literary forgeries, 20, 21, 182,
'Golden Mountain', 73, 84, 97, 186. 206; literary usages of, 153, 158,
GoJdes, 94, JOO, 165 ff., 168, 169, 172, 173, 174,
Gorgons, 61 f., 70, 75, 89, 102, 182, 183, 203, 204; popularity of,
177 f., 180, 181. 152 f., 183.
Graeae, 61, 62, 177 f., 180; see also Hercules, Pillars of, 59, 147; see also
Phorcides. Gibraltar, Straits of.
griffins, in art, 6, 32, 52, 67, 85 ff., Hermes Trismcgistus, 20.
177, 180, 181, 194; description of, Hermippus, 25, 173, 199, 201.
 SUBJECT INDEX
Hcrmobius, 131. Iastae, 11 1
Hermodamas, teacher of Pytha- Iazamatae, 51 ; see also Maeotians;
goras, 158. Sauromatae.
Hermodorus, 199. Ibn Batuta, 198.
Hermotimus, 121, 122 f., 124, 134, Ibn Fadhlan, 108.
143, 148, 152, 153, 154, 156, 168, Ibn Rosteh, 109.
199, 200, 201. Idaean Cave, 122, 16o.
Hermus, river, 131. Iii, river, 98.
'heroization', 120, 123 f., r 26, 127 ff., Imaus, Mount, 28, 186; see also
146, 154, 168 f., 173, 199. Himalaya.
Hesiod, 13, 122; on griffins, 73. India(n), 65 f., 69, 77, 80, 81, 82, 83,
Hcsiodei, 2r, 73, 181. 97 ff., 109, I 10 f., 192, 194, 197
Hcstia, I 49 f. inter-continental boundary, 55 ff.,
Himalaya, mountains, 98, 99, 106, 11 1, 190; see also Phasis; Tanais.
186; see also lmaus. lo, journey of, in Prometheus Virn:tus,
Hindu Kush, mountains, 198. 45 ff.; in other accounts, 46 f., 48,
Hippasus, 20, 21, 201. 60.
Hittites, and griffins, 86 ff. lonia(n), 17f., 19,63, 111,117,124,
Homer, 9, 13, 26, 170, 187, 201, 131, 195.
205; and Aristeas, 32, 169, 170; Irish, 77.
winds in, 93 f., 1 1 5 f. iron, 47 f., 75, 18g.
Homcridae, 21. Irtysh, river, 108, 114, 115, 118; see
horse, 36 f., 63, 177, 192, 198. also Pluton.
Horus, deity, 86; of Samos, 169. Isatis, 169.
Huai .Nan TCJ, 82, 83. Iset, river, 107, 108, 111, 113, 115.
Huns, rn9. Ishim, river, 107, zo8, 115, z8o; see
Hybristcs, river, 47, 49, 50, 52, 54 f., also Pluton.
63, t 89, 19 t ; su also Araxes. Isigonus, 27, 31.
Hylae, 141,200. 'Island of the Blest', 164.
Hypanis, river, 50, 190, 191; see also Issedi, 5, 8, I 9, 68, I I 1 ; see also
Kuban. Issedonians.
Hyperboreans,and Abaris, 158,172; Issedon, town, 106, 108, 112.
and Apollo, 22 f., 52, 62 f., 69, 71, Issedonians, form of name, 1 1 1, 184;
100,141,170,174,179,192, 196f.; in Arirnaspea, 1 ff., 7, 8, 9 f., 22,
and Arimaspi, 22 ff., 25, 185, 186; 32, 39, 40 f., 44, 60, 67 f., 74 f., 76,
and Atlas, 189; and Gorgons, 6'2, 85, 92 f., 94, 100 f., 102, 133, 138,
71, 178, 181; and Scythians, 176, 177, 178, 180, 181, 186, 197;
189; adjacence of moon to, ,203; customs of, 76 ff., 106, 178, 180 f.;
Aristeas reaches, 122, 123, 134 1 location of, 1, 42, 60, 67, 105 ff.,
181, 193; in Arimaspea, 1, 9 1 22, 178, 180, 197, 198; and Hyper-
33, 39 f., 71 f., 74, 98, 100, 110, borean gifts, 196; see also Esse-
116, 171, 176, 180; way of life of, donians; Issedi.
23, 54, 68, 71 f., 73, 98 ff., 178 f., Issetia, 113.
182, 192, 195; Heraclides Ponti- Istrus, river, 42, 52; colony, 196.
cus, and, 165, 166, 171, 182; Italiote, 2, 37, 120, 144 f., 174.
theories about, 35, IO 1, 109, 111, Italy, I l?8, I 70.
157, 186, 195 If., 202. Iyrcae, 105, 106, 108, tog, 114, 115;
Hypernotians, 62. see also Tyrcae.
254 SUBJECT INDEX
Jason, 57. magician, 34 ff., 183.
Jaxartes, river, 106, 107, 1 w, 112; Magna Graecia, 199.
see also Syr. Magna Mater, 87.
Jericho, ancestor-worship in, 78. Magnesia, 4, 141.
Jugra, 108. magus, 159, 16o, 165, 203, 204.
Mahabharata, 81, 98.
Kalmuks, !06. Maikop, go.
Kama, river, 108, w9, 113, 114, mammoth, 84.
II5. Marco Polo, 80.
Kamyshin, 109. Marmara, Sea of, 1, 124; see also
Karkaralinsk, I 07. Propontis.
Kashgar, 112. Maspii, 11o.
Ka8U1TEp8E, geographical sense of, Massagetae, 59 f., 68 ff., 77, to4,
II6 f. 107, !08 f., !09 f., IIO, I II, 112,
Kazak uplands, I I 5. 113, II5, 178, 181,189,193.
Kazan, 108, I 97. matriarchy, 79, 178.
Kelermes, mirror from, 5 f., 7, 36, Mazdaism, 160, 203.
37, 89 ff., I 78. Medusa, 62.
Kerch, 37, 48. Melanchlaeni, 102, 104, 106.
K'euen, 101. Melanippus, and Chariton, 205.
Krivoy Rog, 48. mermen, 10 f., 75, 182, 194.
Kuban, river, 5, 50; see also Hypanis. Meru, Mount, 98 ff.
Kuei, 82, 85, II8, 177, 193. Messalla Rufus, 153.
Kulunda steppe, 108. Metapontum, 2, 119 f., 123, 124,
K'un Lun Mountain, 81. 127, 128, 129, 130, 131, 142, 143,
Kuznetsk mountains, 107, 118. 166, 169, 170, 174f., 182, 199,
201.
Laestrygonians, 187. Metapontus, king, 201.
Lampsacenes, 123. Midas, 4, 204.
Lapps, 94. migration, 1 f., 3, 4 f., 52, 59 f.,
laurel, 2, 120, 122, 123, 166, 170. 74 f., 76, 105, 110, I II, Il4, 118,
Leon, of Sicyon, I 53. 171, 176, 179, 180, 181, 196.
Leonymus, I 64. Milesian, 57, 196, 200.
Leucippus, 198. Milky Way, IOO, 149 f., 152, 202.
Leuke Nesos, 164. Minos, 122.
Libya, 62, w3, 192. Mongol(ia), 81, 84, 92, 94, 97, 98,
Linus, 169, 170. 102,106,108, rng, 118,177,186,
Lobon, 25 f., 32, 199, 205. 194, 198.
Locrians, 164. Monocoli, 29, 31 ; see also Sciapods.
Lombards, 78. monstrous peoples, 1, 1o f., 28 ff.,
Lyceum, 159, 173. 32, 42, 68 f., 82 f., 85, !02, 103,
Lysias, of Syracuse, 152. 106, 107, !08, 177, 194.
Lysis, Pythagorean, 20. Mopsus, 162.
'Mothers', the, seizure by, 137 f.
Maeotians, 51f., 63,118,191; see Mu, king, 101.
also Iazamatae; Sauromatae. Musaeus, 20, 169.
Maes, Itinerary of, 112. Muses, 122, 136.
magic, 142 f., 148. Music of the Spheres', 144, 150.
 SUBJECT INDEX 255
M ycenac, griffin in art of, 86 ff. Pazirik, 36, 37, 92.
Mylasa, grypomachy from, 37. Peak Cavern, 195; see also Winds,
Cave of.
Nemesis, and griffin, 89. Peisander, of Cameirus, 187.
Ncuri, 103. Pentheus, 166.
Nicias, of Enguium, 137 f. Permiaks, 108.
Nile, river, 58, 63, 189. Persephone, 151.
Noin Ula, 92. Perseus, 62, 68 f.
North Wind, 41, 53, 54 f., 64, 65, Persia(n), 37, 38, 66, 8o f., 82, 86 ff.,
1 79; see also Boreas. 92, 110; war, forecast, 205.
nymphs, and Aristaeus, 167; seizure Phalaris, 158, 16o, 16g, 170, 172.
by, 136 f. Phanagoria, ,go, 193.
Pharsalia, dancing-girl, 123, 166.
Ob, river, 10, 84, 108, 115. Phasis, 55 ff., 59, 6o, 64, 75, 111,
Occan(us), 46, 56 f., 58, 59, 70, 100, 118, 178 f., 18o, 181, ,go, 191;
112, 150, 171, 190, 191, 193; see lake-dwellings in, 1 o; see also
also Arctic Ocean; sea, outer. intercontinental boundary; Rion;
Odin, 83. Tanais.
Odysseus, 93, 194. Phemius, 16g, 170.
Oedipus, translation of, 166. Pherecydes, 32, 158, 199.
Olbia, foundation of, 132. Philammon, 169, 170.
Olen, 124. Philistines, 87.
Olympia, griffin from, 91. Philo Byblius, 21.
Olympus, Mount, 34, 35. Philomelus, 123, 166.
Omsk, 114. Philostephanus, 27, 30.
one-eyed men, 73, 82 f., 103, I 76, Phineus, 4 7.
177, 178, 194; see also Arimaspi. Phinteas, and Damon, 205.
Onesicritus, 27, 30. Phoenida, griffin in art of, 6, 86 ff.
Onomacritus, 20. </,o,fJ6>.:r17rToS, 69, 134 ff., 200.
Orestes, 145. Phorcides, 62, 75, 101 f., 197; see
Orpheus, 20, 124, 169, 170, 199. also Graeae.
Ostanes, 160, Phormio, of Croton or Sparta, 157,
Ostiaks, 77. 161 ff., 168, 204.
'OTTopo1<6ppm, 99; see also U ttara- Phrixa, 45.
kuru. Pindar, and Arimaspea, 70 f., 181 ;
and Aristeas, 70 f., 127 ff.
Padaei, 77. planets, 198.
Palus Maeotis, 4 7, 49, 50 f., 56, 58, Plato, 24, 165, 173, 183, 201.
59, 106, I 15, 191; see also Azov, Pluto, 65, 151, 152, 202.
Sea of. Pluton, river, 45, 61, 63, 68, 70, 118,
Pamir, mountains, 108. 177, 189; see also lrtysh; !shim.
Panticapaeum, 37; see also Kerch. poles, 187.
Papyrus Mountains, 63. Polyaratus, of Thasos, 15 7.
Paradise Lost, 1. Polyphemus, 194.
Parapanisus, river, 24. ponticum, 105, 107, 108.
Pamassus, Mount, Asiatic, 18g. Pontus, 49, 1 B!, 191; set also Black
Paropamisus, mountain, 189. Sea; Euxine.
Pausanias, of Gela, 153 f., 165. Posidonius, 186.
 SUBJECT INDEX
Proconnesus, t f., 4, 36, 119 f., 97 ff., 109, 111, 113, 114, u5
120 f., 122, 124, 127, 128, 129, 117, 118, 202; see also Caucasus;
131 f., 133, 141, 179, 181, 183, Rhipae.
199, 200; Zoroaster from, 159 f. Rhipe, 40, 187.
Prometheus, instructs lo, 45 ff.; Rion, river, 56 ff., 190; see also Phasis.
scene of punishment of, 46 f., 53. roe, 67.
Promethms Solutus, 46, 54, 56, 58. Rome, captured by Hyperboreans,
Pronapides, 170, 205. 157, 158, 165.
Propontis, 56, 147,192,200; see also Ros, u4.
Marmara, Sea of. Rubruquis, William de, 77, 95.
prunus padus (bird-cherry), 107, 108, Rus, u4,
197. Russia(n), 1, 91, 104, 114, 195.
Pterophoros, 41, 42, 45, 176, 177;
see also Feathers, Region of. Sabinores, 171.
Ptolemies, 20, 26. Sacae, 70.
Puranas, 98. Sagra, river, battle of, 162, 164.
pygmies, 30, 31, 194. Salmoxis, 124, 144 f., 145.
Pythagoras, 142, 143 f., 144 f., 152, Salmydessus, 49, 191.
153, 157, 158 f., 159, 16o, 164 f., Samoyeds, 10, 77, 84.
169, 170, 172, 174 f., 182 f., 200, Sanchuniathon, 20.
201, 203, 206; cult of, 129, 154, Saneunus, 4-8, 189.
168, 199; and literary forgeries, San-Wei, hill, 101.
20, 21. Saranges, river, 58.
Pythagorean(s), 129, 142 f., 144, Saratov, 108, 118.
146, 156, 158, 169, 172, 174f., Sardinia, 168.
179, 182 f., 198, 201, 202, 205; Sardis, 4, 199.
and literary forgeries, 20, 21, 142. Sarmatian(s), 4, w8.
Pythagoreanism, 145 f., 169, 174, 'Sarpedonian Rock', 195,
204. Sataspes, and circumnavigation of
Pythia, 135. Africa, 203.
Sauromatae, 29, 51 f., 63, 102, I 04,
Ramayana, 98. 108, 109, 110 f., 112, 113, 114,
Rashiduddin, 81. 118, 18o, 19i; see also Iazamatae;
raven, 2, 35, 120,121,126,128,174; Maeotians.
see also bird-soul. Sciapods, 29, 184; see also Fife; Mono-
reeds, giant, 68 f., 75, 103, 192. coli.
reincarnation, 159 f., 161, 164, 165, Scipio, Aemilianus, 149 f., 153 i
I 70, 174, 183, 200, 203, 204, Africanus, 149 f.
Rha, river, 59; see also Volga. Sciratae, 30.
rhinoceros, 84. Scythia(n), 1 f., 3, 4, 6 f., 28, 31, 37,
Rhipae, 40, 42; see also Rhipaean 39, 41, 42, 43 f., 47, 48, 51, 52,
Mountains. 59 f., 68, 70, 71, 74 f., 76, 89 f.,
Rhipaean Mountains, 23, 45, 102, 91, 92, 100, 102, 103, 104, 105,
178, 188, 195; in Arimaspea, 22, 106, 108 f., IIO, 111, 112, 113 f.,
39 ff., 44, 72, 74, 93 ff., 98, 114, 115, 118, HIS, 132f., 176,
110, 176 f., 181; and Caucasus, 177, 178, 179, 18o, 181, 186, 188,
52 ff., 74, 118, 176f., 182, 189; 189, 191, 192, 197, 206.
gold-bearing, 73, 74; originals of, sea, outer or Great, 1, 39, 40, 74, 75,
 SUBJECT INDEX
100, 150, 157, 159, 176, 180, 204; 6o, !04, 1o6, II I, 112, II8, 176,
see also Ocean. 177, 178 f., 18o, 181, 189, 190,
seer, see divination. 191, 193; see also Bosporus, Cim-
Semipalatinsk, 107, 108, 115. merian; Don; Phasis.
sexes, equality of, 51 f., 76, 79, II o, Tarbagatai Mountains, 118.
178, 180 f., 193. Tarim, river, 1o8, 110, 112.
shaman, 36, 84, 125 f., 132 ff., 145, Tatars, 76, 80 f., 83, 84, 107, 114,
146, 186. 178, 197, 198.
Shan Hai Ching, 81 f., 83. Tauromenium, 143.
Shardana, 87. Taurus, mountains, 53, 18g.
Siberia(n), 92, 94, 98, 100, u3, Tegeans, defeat Spartans, 161.
125 f. Telephus, 164.
Sibyl, 21, 165, 16g, 170,203. Temnus, 131.
Sicily, 120, 123, 124, 128 f., 137 f., Tenos, 196.
168, 16g, 170, 172. Teos, 193.
silphium, 37, 38, 162 f. Thamyris, 169, 170.
Sinkiang, 1 o6. Theages, 157, 202.
Sinope, 4, 196. Theano, 142, 201.
Sirius, 167 f. Themiscyra, 49, 191.
skull, as crockery, 28, 77 f.; as cult- Theogony, of Aristeas, 25, 31 f., 172,
object, 76, 78 f., 181; soul escapes 187.
through, 201 f. 8oAoyos, 32, 142 f., 172, 183.
Socrates, 136 f., 150, 157. Theopompus, 24, 166, 199, 204.
Sophocles, 20, 164. Theoxenia, 162, 163.
soul-journey, 35 f., 69, 121 ff., 125, Thennodon, river, 47 f., 49, 51, 52,
126, 133 f., 136, 138, 139 f., 58.
143 f., 146 ff., 153, 156, 158, Thespesius, 149, 153, 205.
162 f., 164, 165, 170, 171, 172, Thespis, 20, 21, 173.
183, 19'2, 201, 205. Thrace, 40, 61, I 16, 141, 144 f., 168,
Sparta(n), 157, 161 f., 164, 169. 179, 195, 20!.
Sparti, 161. Thurii, 127.
sphinx, 6, 32, 89. Thyssagetae-, 105, rn6, 107, rn8,
Spintharus, 20. 109, I 12, I 14, 115,
Steganopods, 10; see also mermen. Tibareni, 48.
Stesichorus, 164. Tibet(an), 76 f., 1o8, 110; Book ef
Stratonice, 192. the Dead, 201 f.
Sumatra, 77. Tien Shan, rn8.
Sumbur, see Sumer. tiger, winged, 82, 177, 193.
Sumer, Mount, 97; see aiso Meru. Timarchus, 149, 153, 201.
Sumur, see Sumer. Timesius, 199.
swan-maidens, 62, 101 f., 178, 180, Titans, 23.
196. Tobol, river, 108,111,113,115.
Syr, river, 107, 110, 115; see also trade-route, 3, 7, 39, 42, 76, !03,
Jaxartes. !05 f,, 1o8, 109, I 10, 111 f.,
Syria, 87. 113 ff., 118, 132 f., 178, 197.
Syzran, rng. trance, 121, 139, 143, 144, 150 f.,
156, 158, 162, 163, 164, 165, 183;
Tanais, l'iver, 45, 51, 52 f., 55 ff., 59, see also catalepsy; soulJoumey.
 SUBJECT INDEX
translation, of Empedocles, I 54, wasps, monstrous, 81 f.
165 f., 205; of Oedipus, 166. Wiltse, Dr., 139 f., 201.
Trophonius, 123 f., 144, 145, 149. Winds, Cave of, 94 ff., 100, 116,
Turkestan, Chinese, 106, 112. 118, 176, 180, 195; see also Boreas,
Tyrcae, II 2 ; see also I yrcae, Cave of.
Tzetzes, John, II, 13.
Xanthus, river, 52.
Ugrian, 108. 'Xenophantus painter', 37.
Ukrainian(s), 83, 178. Xenophilus, 172.
Ural Mountains, 83, 107 f., 109,
110,111, "3, 115. Yakuts, 94.
Uttarakuru, 98 ff. Yukagirs, 84.
yurt, 108 f.
Varro, of Reate, 151 f., 153.
Varvakeion, grypomachy from, 88. Zaratas, 159 f.; see also Zoroaster.
vegetarianism, 71 f., 74, 100, 174, Zaratus, 16o; see also Zoroaster.
178 f., 182 f., 206. Zaysan, lake, 118.
visions, of next world, 148 ff. Zeus, 34, 45, 53, 61, 62, 122, 130,
Voguls, 108. 167.
Volga, river, 59 f., 107, 108, 109, Zhigulev hills, 1 og.
113, 114, 118; see also Araxes. Zoroaster, 151, 157, 159 f., 164 f.,
Votiaks, 83. I 68, I 99, 203, 204,
I. ~IIRROR FRO~! KELER:\!ES
II. :\!IRROR-IL\:\DLE FRO:\J E:\K0:\11
(/
/,
Ill R\l'J'LE FRO:\! kELER\lE"
'
'"'
~ c:::]steppe
Semi -desut or desert

500
 OTHER TITLES IN THIS HARDBACK REPRINT PROGRAMME FROM
SANDPIPER BOOKS LTD (LONDON) AND POWELLS BOOKS (CHICAGO)

ISBN Author Tide

0--19-
8143567 ALFOLDIA. The Conversion of Constanrme and Pagan Rome
6286409 ANDERSON George K. The Literature of the Anglo-Saxons
8219601 ARNOLD Benjamin German Knighthood
8228813 BARTLETf & MacKAY Medieva1 Frontier Societies
8111010 BETHURUM Dorothy Homilies of\Vulfstan
8142765 BOLLING G. M. E.tternal E\'idence for 1nterpolatim1 in Homer
814332X BOLTONJD.P Aristeas of Proconnesus
9240132 BOYLAN Patrick Thoth, the Hermes of Egypt
81!4222 BROOKS Kenneth R. Andreas and the Fates of the Apostles
8203543 BULL Marcus Knightlr Piety & Lay Response to the First Crusade
8216785 BUTLER Alfred]. Arab Conquest of Egypt
8148046 CAMERON Alan Circus Factions
8148054 CAMERON Alan Porphyrius the Charioteer
8148348 CAMPBELL J.B. The Emperor and the Roman Arm~ 3 J BC to 235 AO
826643X CHADWICK Henry Prlscitlian of Avila
826447X CHADWICK Henry Boethius
8219393 COWDREY H.EJ. The Age of Abbot DeS1derius
8148992 DAVIESM Sophocles: Trachm1ae
82S301X DOWNERL. Leges Hennc1 Pnmi
814346X DRO'- KE Peter ~tedienl Latin and the Rise of European LoYc--Lvric
8142749 Dt:1'BABI:S. T.J. The Western Greeks
8154372 F.-\ULK:'\ER R.0. The Ancient Egyptian Pyramid Texts
8221541 FL\NAGAN Mane Therese Irish Society, Anglo--'iorman Settlers., .\nsc\"ln Km~htp
8143109 FRAEN KEL Edward Horace
8201540 GOLDBERG PJP \\"omen, \\'ark and Life Cycle in a \ledieval Economy
8140215 GOTTSCHALK H.B. Heradides of Pontus
8266162 HANSON R.PC. Samr Patnck
8224354 HARRISS G.L. Kmg, Parliament and Pubhc Finance in \ ledie,al England to 1369
8581114 HEATH Sir Thomas Aristarchus of Samas
2115480 HENRY Blanche British Bot:m1caland Horucultural Literature before 1800
8140#\ HOLLIS c\.S. Callimachus. Hccale
8212968 HOLLISTER C \\arren \nglo-Saxon \libt;in Jnstnut1ons
8219523 HOUSLEY Norman The Italian Crusades
8223129 HURN \RD Naomi The Kmg's Pardon for Homicide- bt!forc .ID B07
8140-l-01 HUTCHINSON G.0. 1-Jdlemstic Pocrr,
9240140 JOACHIM H.H. .\nstotle. On Commg-to-be an<l P;issing-;1\\.1,
9240094 JONESA.HM Cities of rhe Eastern Roman Prm,nces
8142560 JONESA.Hl\l The Greek Cit,
8218354 JONES Michael Ducal Brittan, I )64--1 J9<l
8271484 KNOX & PELCZ\'NSt-1 HegeJ's Political \\-nt1ngs
8225253 LE PATOURELJohn The Norman Empire
8212720 LENNARD Reginald Rural England !08~113;
8212321 LEVISON\V En~land ;nd the Continent m the Sth 1..-entun
8148224 LIEBESCHUETZJH.\V G. Continuttv and Change in Roman Religion
8141378 LOBEL Edgar & PAGE Sir Denys Pocra.-um l..c-sbiorum fr:unnenta
9240159 LOEWE.A. The Benc,enran Scrtpt ~
8241445 LUKASIEWICZ.Jan Anstotle's S,l1ogist1c
8152442 MAAS P & TR\'PANISC.A Sancti Romani \ leloJ1 C.anttca
8142684 ,\,IARSDEN E.\\ Greek and Roman \rtillcn-Histork:il
8142692 MARSDEN E. W Greet.. anJ Roman \rrlllcn-TrchniC:11
8148178 MATTHEWS John \\'estern Aristrn:ra'-'1es anJ lmperi:il Court .\D ,\6-4-4!~
9240205 MAVROGORD,\TO John Dtgenes \li.ntes.
8223447 McFARLANF. KJl. Lmc.1strian h.ings and Lollard knights
8226578 McFARLANE K.B. The ,ob1ht\' ofl ..1.ter \lc:dic:,al Fn)!l.mt.l
9240205 MAVROGADOJohn D1gcnes -\kntcs
8148100 MEIGGS Russell Roman O~ti,1
8148402 MEIGGS Russell Trees and Timber in the \ndent .\leJ11c:rranlan \\'orkl
8142641 MILLERJ. Innes l'hc SpK'C 'J'raJtofthc. Roman Empire
8147813 MOORHEAD John Thco<.kric 111 ltal,
8264259 MOORMAN John .\ l-liston of the Fnmd~L.111 Order
9240213 MYRESJL, Hermfot;1s The Father ofHiston'
8219512 OIIOLENSK YIlim,tri Six \hzan1ine Pot{raits
OWEN Al .. The F'an\ous Oru1Js
8116020
PALMER, L.R. The lntcrpretationof\lycenacan Greek Tc\:ts
8131445
PFEIFFERR. 1listor} ofCl.\...sil'.11 Schof:lrship (,ol I)
814.1427
PFEIFFER Rudolf Uisron of Classical St.holarship 1:l00-1850
8143648
8111649 PHEIFERJ.D. Old English Glosses in the Epinal-Erfurt Glossary
8142277 PICKARD-CAMBRIDGE A.\V Oithyramb Tragedy and Comedy
8269765 PLATER&WHITE Grammar of the Vulgate
8213891 PLUMMER Charles Lives oflrish Saints (2 vols)
82069SX POWICKE Michael Military Obligation in Medieval England
8269684 PO\VICKE Sir Maurice Stephen Langton
821460X POWICKE Sir Maurice The Christian Life in the Middle Ages
8225369 PRAWERJoshua Crusader Institutions
8225571 PRAWERJoshua The History of The Jews in the Latin Kingdom ofJerusalem
8143249 RABYF.J.E. A History ofChristian Latin Poetry
8143257 RABYF.J.i. AHistory of Secular Latin Poetry in the Middle Ages (2 vols)
8214316 RASHDALL & POWICKE The Universities of Europe in the Middle Ages (3 vols)
8154488 REYMOND E.A.E& BARNSJ.W.B. Four Martyrdoms from the Pierpont Morgan Coptic Codices
8148380 RICKMAN Geoffrey The Corn Supply of Ancient Rome
8141076 ROSS Sir Da,id Aristotle: Metaphysics (2 vols)
8141092 ROSS S1r Da,id Aristotle: Physics
8142307 ROSTOVJ'ZEFF M. Social and Economic History of the Hellenistic World, 3 vols.
8142315 ROSTOVTZEFF M. Social and Economic History of the Roman Empire, 2 vols.
8264178 RUNOMAN Sir Steven The Eastern Schism
814833X SALMON J.B. Wealthy Corinth
8171587 SALZMAN L.F Budding in England Down to 1540
8218362 SAYERS Jane E. Papal Judges Delegate in the Province of Canterbury l l 9&-1254
8221657 SOIEIN Sylvia Fideles Crucis
8148135 SHERWIN WHITE A.N. The Roman Citizenship
9240167 SINGER Charles Galen: On Anatomical Procedures
8113927 SISAM, Kenneth Studies m the History of Old English Literature
8642040 SOUTER Alexander A Glossary of Later Latin to 600 AD
8270011 SOUTER Alexander Earliest Latin Commentaries on the Epistles of St Paul
8222254 SOUTHERN R.W. Eadmer Life of St. Anselm
8251408 SQUIBB G. The High Court of Chivalry
8212011 STEVENSON & WHITELOCK Asser's Life of King Alfred
8212011 SWEET Henry A Second Anglo-Saxon Reader-Archaic and Dialcctkal
8148259 SYME Sir Ronald History in Ovid
8143273 SYME Sir Ronald Tacitus (2 vols)
8200951 THOMPSON Sally Women Religious
8201745 WALKER Simon The Lancastrian Affinity 1361-1399
8161115 WELLESZ Egon A History oflly1,antinc Mu,ic and Hymnography
8140185 WESTM.L. Greek Metre
8141696 WESTM.L. Hesiod. Thcogony
8148542 WESTM.L. The Orphic Poerns
8140053 WESTM.L. Hc,iod: Works & Days
8152663 WESTM.L Iambi et F:lcgi Grnec,
9240221 WHEELWRIGHT Philip Heraclitu,
822799X WHITBYM&M The I listory ufThcuphylucr Sirnocatta
8206186 WILLIAMSON, E. W Lcne"ofO,bert of Clare
8208103 WILSON FP 1
Plague: in Shakc'-1>cArc !. I.ondun
8114877 WOOLF Rosemary The English Religious Lyric in the Mi<ldlc Ages
8119224 WRIGHT Jo>eph Grammar of 1hc Go1hlC J.aniuugc