….

uQoı x ARP, N. va A2. @ | CHICAGO JOURNALS AT OA AQ N KY KNOZ Author(s): Alexander H. Krappe Source: Classical Philology, Vol. 37, No. 4 (Oct., 1942), pp. 353-370 Published by: The University of Chicago Press Stable URL: http://www jstor.org/stable/264217 Accessed: 17/03/2009 12:35 Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://wwwjstor.org/page/info/about/policies/terms jsp.JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www jstor.org/action/show Publisher?publisherCode=ucpress. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We work with the scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that promotes the discovery and use of these resources. For more information about JSTOR, please contact support@jstor.org. CT The University of Chicago Press is collaborating with JSTOR to digitize, preserve and extend access to Classical Philology. http://www jstor.org

AΠ OΛ Λ Ω N KΥ KNOΣ Source: Classical Philology, Vol. 37, No. 4 (Oct., 1942), pp. 353-370 Published by: The University of Chicago Press Stable URL: http://www.jstor.org/stable/264217 Accessed: 17/03/2009 12:35 Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showPublisher?publisherCode=ucpress. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We work with the scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that promotes the discovery and use of these resources. For more information about JSTOR, please contact support@jstor.org. The University of Chicago Press is collaborating with JSTOR to digitize, preserve and extend access to Classical Philology. http://www.jstor.org

CLASSICAL Volume XXXVII PHILOLOGY OCTOBER 1942 Number 4 AIIOAAQN KTKNO0 ALEXANDER H. KRAPPE J7TYKNOSwas a king of the Ligurians who ruled at the mouth of the amber-producing Eridanos. He was a friend of Phaethon, and at the misfortune which befell that imprudent young man he was so stricken with grief that the gods transformed him into a swan. In this shape he chanted his dying song. This brief and somewhat pathetic story is probably best known from Vergil's lines (Aen. x. 189 ff.): Namqueferunt luctu CycnumPhaethontisamati populeasinter frondesumbramquesororum dum canit et maestummusa solaturamorem, canentemmolli plumaduxissesenectam linquentemterraset sideravoce sequentem. It is, however, much older, going back to Hesiod, whose account has been preserved by Hyginus (fab. 154). In Alexandrian times it was retold by Phanocles, on whose work Ovid drew for his own account.' The commentator on Vergil, Servius (ad Aen. x. 189), adds that Kyknos had been endowed with the gift of song by Apollon and that the gods, not satisfied with transforming him into a bird, further honored him by raising him to the nightly sky, where he forms the constellation ever since known by his name. Pausanias (x. 30. 3) is silent 1 Narr. ad Ovid. Met. ii. 367 (p. 797 [Stad.]); Lactant. Argum. in Ovid. Metam. 2; cf. A. Couat, Alexandrian Poetry under the First Three Ptolemies (London, 1931), p. 107; Karl Millenhoff, Deutsche Altertumskunde, I (Berlin, 1890), 218; A. B. Cook, Zeus, II (1925), 477. XXXVII, OCTOBER, [CLASICALPHILOLOGY,

ALEXANDER H. KRAPPE about Phaethon but relates that Kyknos, a musician, became king of the Ligurians inhabiting the Celtic country beyond the Eridanos and at his death was transformed into a swan by Apollon. The first problem requiring a solution is the reason responsible for the connection of Kyknos with Phaethon. The fact that Hyginus quotes Hesiod as his authority affords, of course, no guaranty that Hesiod was his only source,2and the association of the two heroes may possibly be later. Whether or no we believe it as early as Hesiod, it certainly is secondary. Our suspicion is aroused chiefly by the strange accumulation of transformations connected with Phaethon's fall: we are told first that Phaethon's sisters, the Heliades, were transformed into amber-exuding white poplars (XefKaLu);3 then Kyknos is metainto a swan. Even the motivation is the same: an act of morphosed divine mercy, since Phaethon's sisters, as well as his friend, are griefstricken and inconsolable at their loss. But we know that attraction has always been a powerful factor in the formation of cycles or clusters of myths, and the conclusion is probably not far wrong that the metamorphosis of the Heliades attracted that of Kyknos into the orbit of the Phaethon cycle. Still, to explain such an attraction, it ought to be shown that the two stories shared some other feature which facilitated the process, and this feature appears to be the geographical location of the two. The Heliades were transformed into poplars growing on the banks of the Eridanos. But the Eridanos was also the river famous for its singing swans, as we know from one of Lucian's witty dialogues.4 After inquiring into the amber-weeping poplars-with a negative resultLucian puts a second question to the boatmen, his informants: But how about your swans? At what time do they sing so melodiously, rangedalong the river, on this side and on that? People say, at all events, that they wereassociatesof Apollo,menwith the gift of song, who somewhere in these parts changed into birds, and for that reason do not forget their music, but still continueto sing. 2 Cf. C. Robert, Hermes, XVIII (1883), 436 f. 3 Ovid Metam. ii. 363 ff.; Eur. Hippol. 732 ff.; Apollon. Rhod. iv. 602 if.; Paus. i. 4. 1; Diod. Sic. v. 23. 2 ff.; cf. also Cook, op. cit., pp. 472 ff.; Apollodorus, The Library, ed. Frazer (London, 1921), II, 388 ff.; G. Knaack, QuaestionesPhaethonteae ("Philologische Untersuchungen," Vol. VIII [Berlin, 1886]), pp. 62 f. 4 On Amber 4-5; cf. also Pliny NH x. 63; Athen. ix. 393 D; Aelian Var. hist. i. 14;

Alas, the information elicited was even more disappointing: We are always on the water [saidthey] and have workedon the Eridanos since we were children,almost; now and then we see a few swans in the marshesby the river, and they have a very unmusicaland feeble croak; crows and daws are Sirensto them. As for the sweet song you speak of, we never heardit or even dreamtof it, so we wonderhow these stories about us got to your people. Lucian was, of course, quite right from his point of view. The river which he (following many others) took to be the Eridanos was the Po, and the only swans his informants knew were evidently specimens of the mute swan (Cygnus olor), whose croak is anything but melodious. The real Eridanos was, however, not (and never had been) the Po or any other river emptying into the Mediterranean basin, in which, besides, no amber is found. In prehistoric times the amber reaching the Mediterranean is known to have come from the North Sea coast, up the river Elbe, whence it followed the valleys of the Saale or Moldau, to cross the Alps at the Brenner Pass to the mouths of the Adige and the Po.6 This led to vague reports concerning a large river which emptied into the northern ocean near the amber isles.6 This river was known as the Eridanos.7 A few words about this mysterious river may be useful. As is generally known, the Greeks were shut off from the outer sea by the sea power of Carthage, which jealously guarded the secret of the northern tin and amber routes. As a result, the Eridanos came to be identified with the rivers through whose mouths the amber finally reached the Mediterranean, that is, the Po and the Rhone.8 Inasmuch as the older Cf. R. Hennig, Neue Jahrbicher f. d. kl. Altertum, XLIX (1922), 365; Von rdtselhaften Ltndern (Minchen, 1925), pp. 85 f.; H. Schaal, Vom Tauschhandelzum Welthandel (Leipzig, 1931), p. 19; E. Norden, Alt-Germanien (Leipzig-Berlin, 1934), p. 258, n. 3; Sir William Ridgeway, The Early Age of Greece,I (Cambridge, 1901), 182; for a good map cf. Cook, op. cit., P1. XXVI. The thesis of the otherwise excellent work of G.-E. Broche (Pythias le Massaliote [Paris, 1935], pp. 204 ff.), namely, that the protohistoric amber is of Baltic origin, is untenable in view of the results of a chemical analysis of this amber. 6Apollon. Rhod. iv. 505 f. 7 'EpaSavosis, of course, a good Greek or rather pre-Hellenic name, since it occurs in the old Attic genealogies (cf. Wilamowitz, Hermes, XVIII [1883], 427, n. 2). On the Attic river by that name cf. Strabo ix. 1. 19; Paus. i. 19. 6. 8 Miillenhoff, op. cit., I, 219 f.; A. Bertrand and S. Reinach, Les Celtesdans les vallees du P6 et du Danube (Paris, 1894), pp. 10 and 18. The strange phenomenon of earlier writers' being better informed than later ones, on matters geographical, is not unique. Familiar instances of this are seen in the false notion of the Caspian entertained by

accounts leave no doubt about the fact that the Eridanos was thought to empty into the outer ocean,9 these late identifications are, of course, inadmissible; and Karl Miillenhoff, in the last century, believed the Eridanos to have been the Rhine.10 There are grave objections to this identification, set forth by R. Hennig, who, following Olshausen,1 pointed out that the Eridanos was none other than the river Elbe, while the amber isles must be sought in the Frisian archipelago.12 This identification is now quite generally accepted. We thus find this northern river closely associated (1) with amber and (2) with swans. The question arises: Do these swans correspond to the description applied to Kyknos, whose musical endowments are stressed in the accounts outlined above, and to the expectations of Lucian? Antiquity had a good deal to say about the singing swan. From the account of Hecataeus of Abdera, who wrote toward the end of the fourth century of the pre-Christian era, we learn that there lived, on a certain island of the Far North, called Helioxoia, a people that worshiped Apollon in a round temple. The god used to come to this island at regular intervals, giving the signal for choral services at the spring equinox, when wild swans would come down from the Rhipaean Mountains, joining their notes harmoniously with the human voices.'3 The tradition of the singing swan is, however, much older. In Plato's Phaedo (84 E), Socrates asks Simmias whether he really thinks him inferior in prophetic power to the swans or less able than they to greet the unseen with a song. As early as the eighth century of the preChristian era the author of the Shield of Heracles, wrongly attributed to Hesiod, seems to speak of an island in the northwestern ocean characterized by a multitude of singing swans.l4 Then there is the hymn of Strabo and the opposite error in regard to the Indian Sea held by Ptolemy, as compared with the correct ideas on both subjects possessed by Herodotus. 9Herod. iii. 115. 10Op. cit., p. 221; cf. H. D'Arbois de Jubainville, Les premiers habitants de l'Europe (Paris, 1894), II, 28. 11Zeitschrift f. Ethnologie, XXII (1890), 270. 12 Neue Jahrbiicher, XLIX, 364 ff.; Von rdtselhaften Ldndern, pp. 82 if. 13 Diod. Sic. ii. 48; Aelian De nat. anim. xi. 1. 4 Vss. 315-17: o0 i Kar' aCbVTO KVKVOL &eprtLr6rat z/eya X' lrvov o' A' re7roXXot ViXOv rx aKpov tfWp' Trap&a6iLXObs eKXovkovTO.

Alcaeus telling how Apollon is given by his father Zeus a swan-drawn chariot and how the god bids these swans, instead of going to Delphi, as Zeus would have had him do, fly to the land of the Hyperboreans.15 In Euripides' Phaethon the bird of Apollon is described as singing a morning hymn on the waves of the Ocean that girds the earth.'6 The swan referred to is the whooper swan (C. musicus or ferus)a species entirely distinct from the mute swan (C. olor). The whooper swan is a migratory bird of virtually unvarying habits. A native of Iceland, Spitzbergen, eastern Lapland, and northern Russia, that is, of countries generally well within the Arctic Circle, it flies southward in autumn to return to its northern home in March or April. In its southward flight it visits the British Isles, western and central Europe, Italy, and the Balkans. In Germany it has been observed in the valley of the Elbe, particularly in the famous park of Worlitz in the former Duchy of Anhalt.17 The song of this swan has a strange melancholy sound, faintly resembling distant bells, sometimes loud enough to disturb in their sleep those who are not used to it. The bird was known in Greece and western Asia Minor in Homeric times, as is shown by a well-known passage of the Iliad (ii. 459 ff.); but on the whole it is much less common in the Mediterranean countries than the mute swan, and this accounts for the skepticism of Lucian. At all events the swans of the amber country were singing swans, and Kyknos was transformed into a bird of this species. The common occurrence of this swan along the amber coast of northwestern Germany no doubt accounts for the linking of the transformation of Kyknos to that of the Heliades. This result, it must be admitted, has been attained by inferences rather than on direct statements by existing documents. There is, however, one fact which strikingly confirms our deductions. Kyknos is said to have been a king of the Ligurians. But we know that Ligu15 'AXKalov MkX\T,ed. Edgar Lobel (Oxford, 1927), pp. 42 f.; cf. K. O. Muller, Die Dorier (Breslau, 1844), I, 271; Cook, op. cit., pp. 459 f. 16 Wilamowitz, op. cit., pp. 402 f. op. cit., I, 1 if.; O. Keller, Die antike Tierwelt, II (Leipzig, 1913), pp. 213 ff.; R. G. Wardlaw Ramsay, Guide to the Birds of Europe and North Africa (London, 17 Miillenhbff, 1923), p. 213; J. Rendel Harris, Bull. John Rylands Library, IX (1925), 375 ff.; Hennig, Historische (1929), 252. Zeitschrift, CXXXIX 17 f.; J. Hoops, Worter und Sachen, XII

rian tribes, prior to the advent of the Celts, i.e., in the first half of the first millennium before Christ, inhabited the North Sea coast between the Maas and the Elbe, as well as parts of Belgium and northern France.'8 The statement would thus appear to reflect geographical and ethnological data of a very early period, antedating the settlement of the Celts in northwestern Europe. It goes far to prove the essential correctness of our identification of the amber isles with the Frisian archipelago. The same fact would also seem to betray a very early authority and to strengthen Robert's view as to the correctness of Hyginus' citation. The statement of Theophrastus19to the effect that amber was found in Liguria can be explained only on the supposition that his source was some account based on a geographical and ethnographical map of Europe true for the eighth century of the pre-Christian era but certainly no longer true in the fourth. Pausanias is the only authority to ascribe the transformation of Kyknos into a whooper swan to the god Apollon. His statement cannot very well be separated from a passage in Servius' Vergil commentary, according to which it was Apollon who had conferred upon Kyknos the gift of song.20 Thus the question arises: What has Apollon to do in all this? It will not do, of course, to hold the story of Phaethon responsible: Phaethon was a son of Helios, not of Apollon, and there is nothing to indicate a possible identity of the two.21 There remain three other clues, viz., Apollon's connection with (1) the swan, (2) amber, and (3) the Frisian North Sea coast. Apollon is definitely connected with the whooper swan in the account of the Hyperborean land given by Hecataeus of Abdera, in the hymn of Alcaeus, and in Euripides' Phaethon, all referred to above, to say nothing of iconography (since in their outward appearance C. musicus and C. olor are hardly differentiated). 18 D'Arbois, op. cit., II, 205; Henri Hubert, The Rise of the Celts (London, 1934), p. 152. 19De lapid. 16 (ed. Schneider). 20 Op. et loc. cit.: "Fuit enim quidam Ligus Cycnus nomine dulcedine cantus ab Apolline donatus amator Phaethontis, qui cum eum fleret extinctum longo luctu in avem sui nominis versus est, qui postea ab Apolline inter sidera conlocatus est." Aeschylus (Sept. Theb. 859) appears to have been the first to identify Apollon with the sun.

AIhOAAQNKTKNO3 Apollon's connection with amber is no less certain. Apollonius Rhodius knew a Celtic tradition, according to which amber drops are the tears of the god when he had been banished by Zeus to the land of the Hyperboreans.22 In a study to be published shortly in Speculum it has been pointed out that the votive offerings sent to Delos by the Hyperboreans were probably the first fruits of the "amber season," which opened, as we learn from Pytheas of Massilia, quoted by Pliny,23 shortly after the equinoctial storms-a conjecture made nearly a century ago by F. G. Welcker.24 The center of the amber-fishing, according to Pytheas' account, was an island of the Frisian archipelago (Glaesariae) known as Abalus (or Abalum), a name which seems to survive in Habel and Appelland, two of the Halligen Islands off the west coast of Schleswig-Holstein, or in Avlum, on Thy, on the west coast of Jutland.26 It is difficult to separate this tradition from the report of Hecataeus of Abdera concerning the mysterious island of Helioxoia, where Apollon was worshiped in a round temple and where whooper swans, swooping down from the northern mountains, announced his presence. For it is certainly no accident that one of the most important stations in the southward migration of these birds is precisely the isle of Helgoland, at the mouth of the river Elbe, the ancient Eridanos. To quote a well-known bird specialist: Countlessmultitudesof all sorts of species are seen speedingtowardsall parts, and in all directions,in companiesgreat and small, solitary and in pairs. Indeed I have knowndays on which I have seen, far as the eye could reach,in all quartersof the sky, swarmsof these birds crossingeach other in all directions;and more astonishingstill, on looking upward, have beheld above me a teeming multitude, so thick that the highest swarmspresented the appearanceof scarcely discernibleclouds of dust..... Nowhere does the quick observanteye find rest. Suddenlyare heard-first faintly, then in 22Arg. iv. 611 ff. NH xxxvii. 35-36; cf. also iv. 94-95, 97, 103; xxxvii. 42; Diod. Sic. v. 23; D'Arbois de Jubainville, Revue celtique, XII (1891), 13 f., and Cours de littrature celtique, XII (Paris, 1902), 72 f.; D. Detlefsen, Die Entdeckungdes germanischenNorders im Altertum (Berlin, 1904), pp. 4 ff. 24 GriechischeGtterlehre (G6ttingen, 1857-62), II, 353 ff. As late as the sixth century of our era Cassiodorus (Var. v. 2) reports that a certain northern tribe had sent amber as an offering to the great Theodoric; cf. also Hennig, Historische Zeitschrift, CXXXIX, 26 Broche, op. cit., p. 219.

increasingloudness-sounds like distant trumpetblasts, and once moreour eyes are attractedupwards,wherea long chain of WhooperSwans,eighteen or twenty in number,in snow-whiteplumage,calmly pursuetheir way with measuredbeatingof theirwings.26 And again: This swan occurshere more or less numerouslyevery winter,being most frequentlyseen duringthe long-continuedfrost; on such occasionsit is not uncommonto see flightsof ten, twenty, or even muchlargernumbersmigrating high overheadin a long row one behindthe other, utteringloud trumpet calls as they pass along. There are two difficulties in the way of any attempt to identify Helgoland outright with the island of Hecataeus, viz., (1) the Greek writer asserts that the island is as large as Sicily, while Xenophon of Lampsacus, as quoted by Pliny,27 likewise speaks of the immense size of the amber island; (2) Plutarch, who also knows of a god exiled to an island in the northern ocean, called Kronos by him but doubtless identical with the exiled Apollon of Apollonius Rhodius,28states that on that island the sun disappears only for a short time from the horizon during thirty days in the year; but this would presuppose a higher latitude than that of Helgoland. As for the first difficulty, it is to be noted that Helgoland is known to have been much larger formerly, having been reduced in size by the action of the waves.29 It is also known that most data referring to distances in the north are greatly exaggerated, owing no doubt to the slow progress made by Mediterranean vessels in the northern seas with their strong tides.30 At all events, since there never has been an island of vast size anywhere in this region or even farther north, we must assume that the ancient authors erroneously transferred data referring to Britain31or Scandinavia (long considered an island)32 to the mysterious isle of the Hyperborean Apollon.33 This would also take care of the second difficulty. 26 H. 519. 27 NH Gatke, Heligoland: An Ornithological Observatory (London, 1895), pp. 21 f. and iv. 95; cf. Detlefsen, op. cit., p. 14. De facie in orbe lunae 26; cf. De defectu orac. 18. 29 G. Hergt, Die Nordlandfahrt des Pytheas (diss. Halle-Wittenberg, Hennig, Von rdtselhaften Ldndern, p. 90. 30 Detlefsen, op. cit., p. 6; Hennig, Von rdtselhaften Ldndern, p. 109. 28 Plutarch 1893), p. 34; 1aIt is noteworthy that Britain, like Sicily, was thought (not incorrectly) to be triangular (cf. Diod. Sic. v. 21). 32 Pliny NH iv. 27 and 30; Jord. 3. 33 Harris, op. cit., pp. 381 and 390 f.

Hecataeus of Abdera and Apollonius Rhodius are by no means the only or the earliest writers of antiquity who definitely associate Apollon with the far north. Pindar gives a glowing description of a Nordic land of bliss, an inaccessible island ruled over by Apollon: Neitherby ship norby land canstthou findthe wondrousroadto the tryst.....In the banquets and praises of that ing place of the Hyperboreans people Apollonchiefly rejoiceth.... neither sickness nor baneful eld mingleth amongthat chosenpeople;but, aloof from toil and conflict,they dwell afar from the wrath of Nemesis. .... 34 Strabo has preserved the fragment of a lost play of Sophocles dealing, presumably, with the rape of Oreithyia by Boreas: Let us thereforepass over .... what Sophocles,speakingof Oreithyiain one of his tragedies,says, that she, being snatched by the northwind,was carried "Overthe whole ocean, to the extremitiesof the earth, Even to the place wherenight receivethits birth, Wherethe oppositeside of the heavensis beheld, And whereis situated the ancient gardenof Phoebus."35 To this isle of bliss Apollon carried his favorite heroes. Thus Bacchylides (3. 53) relates how the god releases the good king Croesus from the burning pyre, wafting him to the Hyperboreans in reward for his piety.36 In Pindar's ode, referred to above, the inference is that King Hiero of Syracuse, to whom it is dedicated, will ultimately enjoy the same blissful fate.37 If these Hellenic traditions are founded on fact, we should expect that the island in question would have left an analogous reputation, or traces of such, in the traditions of the north. Such is indeed the case. Helgoland means nothing less than "holy island," a name fully justified, inasmuch as the island was the center of the cult of the Frisian god Foseti, who may be supposed to have taken the place of a Celtic predecessor, and was avoided by pirates down into the Middle 34 Pyth. x. 27 ff.; cf. also Wilamowitz, Pindaros (Berlin, 1922), pp. 127 f. 36vii. 3. 1; Nauck, Frag. trag. Gr., p. 333: 1ra'vr' br' raxara xOovos nriepre 7r6vo0v Tre7r'yaL obpavov 7' &vtaTTvXas VKTO6S 4ioiov re 7raXatC6KTcrov..... 36O. Crusius, Philologus, LVII (1898), 155 f.; Kirte, X (1907), 152. 37H. Jurenka, Philologus, LIX (1900), 313 ff. Archiv f. Religionswissenschaft,

Ages.38 In these circumstances the Greek form of the name given to the mysterious island at the time of Alexander the Great (when Teutonic tribes had undoubtedly reached the North Sea coast), Helioxoia, may very well resolve itself into corruptions of helig, "holy," and ey, "island," an etymology not unacceptable, perhaps, at least until a better shall have been found.39 The facts reviewed will leave no doubt about the very definite relations of Apollon with the amber coast of northwestern Germany. He shares with Kyknos his connection with the whooper swan, with amber, and with the North Sea coast and the Frisian archipelago, and his introduction into the transformation story of Kyknos would thus appear natural enough. What is striking, however, is the strange parallelism between the god and the hero; for, apart from the three features just mentioned, Kyknos shares with Apollon the gift of song, which the god is stated to have conferred upon his favorite. In other words, there is nothing (apart from the transformation proper) that Kyknos does not share with Apollon. This fact would seem to warrant the inference that Kyknos is merely the heroic form of Apollon; but this inference would imply that the god transformed himself into a whooper swan, that he was a swan himself; for, according to the well-known law first formulated, I believe, by Salomon Reinach, the shape assumed by a god in a metamorphosis story is merely the former animal form of the same god, and the myths of Zeus as a bull (in the rape of Europa), of Io as a cow, and of Artemis as a she-bear will readily come to mind. On the other hand, it is clear that such a divinity can have been conceived only in a country where the whooper swan was of common occurrence, which means that Hesiod's localization of the tale among the Ligurians of the North Sea coast of Germany or Holland presupposes the worship of Apollon Kyknos by those peoples. In other words, Apollon Kyknos is a northerner, whose home is the northwestern part of central Europe. Here one objection might be raised. If the foregoing inferences are justified, how are we to explain, on Greek soil, the connection of the swan with divinities other than Apollon? Thus in a well-known Pelo38 Adamus Brem. iv. 3. 39 Harris, op. cit., p. 382.

ponnesian tradition it is Zeus who assumes the shape of a swan in order to seduce Leda and to beget the Spartan Dioskouroi. To this objection we may point out that a number of Apollon's functions and epithets are occasionally found attributed to Zeus. Thus the Anatolian sun-god Syros or Lykos is not only the predecessor of Apollon Lykeios but also of Zeus Lykaios.40 The Nordic Kyknos may therefore very well have been identified, in the Peloponnesus, with the Olympian Zeus. This conjecture would seem to be borne out by the following rather significant circumstance. Leda, Dor. Lada, cannot be separated from Leto, Dor. Lato (whence the Lat. Latona). The word is generally thought to be the equivalent of Gr. yvvw, "woman," and the goddess would then be the Woman par excellence. But, by this very Leto, Zeus is the father of another pair of twin-gods, Apollon and Artemis, which would lead us back to the Swan Apollon. The relationship of Apollon and Artemis is known to be secondary; but the original and Indo-European, i.e., Nordic, character of the Spartan Dioskouroi has been known for some time, and their Sanskrit, Persian, Teutonic, and Celtic equivalents have been studied repeatedly.41 Da muss sich manchesRatsel losenDoch manchesRatsel kniipft sich auch, says Goethe, who knew what true research means. Several questions may, indeed, already have presented themselves to the mind of the reader. The strange idea of localizing an earthly paradise of the type described by Pindar and Sophocles in the dreary Frisian archipelago will certainly require some explanation. The ancients were quite aware of the true nature of amber, which is a resinous product.42 Thus the Collection of Marvelous Stories as40 Detailed proof of this has been presented in my study "Apollon," in Studi e materiali di storia delle religioni (Rome). On the Apolline character of the story of Zeus, Leda, and the Swan cf. M. Mayer, Die Giganten und Titanen in der antiken Sage und Kunst (Berlin, 1887), pp. 80, 141 f. 41 For a comprehensive bibliography cf. my book Mythologie universelle (Paris, 1930), p. 100; Acta philologica Scandinavica, VI (1931), 1-25. 42Cf. Arist. Meteor. iv. 10. 10, ed. Didot; Pliny NH xxxvii. 42; Tacitus Germ. 45; Martial iv. 59. 1 f.; Pauly-Wissowa, R.-E., Vol. III (1), col. 297.

cribed to Aristotle and written about 300 B.C. describes amber as a gum which liquefied from poplars near the Eridanos and which, being hardened into the consistency of stone, was collected by the natives and exported into Greece.43 What was less generally known was its fossil nature, though the geographer Philemon pointed this out.44 As a rule it was assumed that amber was exuded from living trees growing on islands of the northern ocean. Thus, according to Pliny, Mithridates reported that on the coast of Germany45there was an island called Serita, grown with cedars exuding resin. In other words, the islands near which amber was fished from the sea became, in the imagination of the Mediterranean peoples, islands overgrown with amber trees. The Mediterraneans were, however, neither the first nor the only ones thus to account for the valued product. The Germans were no less aware of the true nature of amber, as is sufficiently proved by MLG glar, "resin," as compared with OE glaer, "amber."46Knowing, then, that amber was not a marine product, they answered the question as to its origin by the assumption that it was the resin of trees growing on some island or continent beyond the sea, whence it was washed to the German shores. Since to the poor barbarians who inhabited northwestern Germany amber was about as valuable as gold is to us, this assumption quite naturally led to the belief in an island paradise, a sort of Eldorado. In other words, the mysterious northern isle owed its splendid reputation to the simple fact that it was an amber isle. If such is the case, one may reasonably expect this tradition to have left traces also in the north; for in the eyes of the barbarians who sold 43 Mir. ausc. 81. 44 Pliny NH xxxvii. 33: "Philemon fossile esse (electrum dixit)." To be sure, Mayhoff (in the Teubner ed. [1906]) adopts the reading "in Carmaniae litoribus insulam esse quam vocari Seritam, cedri genere silvosam"; but the compiler of the Index, s.v. Serita, locates the island in Germaniaelitore, accepting the reading of the Codex Leidensis (which reads germanicae), as did the translators of the English version published by the Bohn Classical Library. In Carmania (on the Persian Gulf) no amber is known to have been found. Mithridates Eupator-for he is the author referred to (cf. F. Susemihl, Geschichteder griechischen Literatur in der Alexandrinerzeit [Leipzig, 1891-92], II, 415)-was in a position, as ruler over what is now southern Russia, to have some information about northern Europe. 46 R. Much, Zeitschriftf. deutschesAltertum, LXI (1924), 101. 46 Ibid. xxxvii. 39.

AIIOAAN KTKN02 the precious substance the island was bound to have at least as much glamor as in those of the far-off Greeks. This conclusion is borne out by a number of facts. Certain Scandinavian documents, discussed in detail in my article to be published in Speculum, speak of a terrestrial paradise called Glaesisvellir or Glaesisvollum,names evidently derived from OG glez, "amber," Lat. glaesum. At the gates of Walholl is a grove named Glasir, i.e., evidently, a grove of amber trees. Celtic traditions know of an insula vitrea, OFr isle de voirre, which in the Middle Ages meant "glass island" but which originally meant "amber isle," owing to the semantic development of the word glaesum,"amber," in the Teutonic and Celtic languages: OG glez, OE glaer, "amber," MG, ME glass; Ir. glain, gloin, "glass," "crystal"; Welsh glein, glain, "gemma," "tessera." Lastly, the name of the amber island Abalus (or Abalum) mentioned by Pytheas of Massilia survives not only in the names of two Hallig islands, Habel and Appelland, and the name of Avlum in western Jutland but also in that of the Celtic Avallon, the island paradise to which King Arthur is wafted, according to Arthurian tradition, much as Croesus and Hiero are wafted, at least in the verses of Bacchylides and Pindar, to the island paradise of the Hyperborean Apollon. This island of Avallon, on the other hand, is identical with the insula vitrea, the ancient amber isle, which, in the Middle Ages, had become a "glass island." The report of an amber isle in the northern ocean might be regarded as a simple geographical myth which might be supposed to have reached Greece, as geographical myths usually reach distant countries, by the route of commerce-in this case the amber trade. Such a theory would fail to explain (1) the tradition of the votive offerings sent down by the Hyperboreans, i.e., the tribes of the amber coast, (2) the connection of the Hellenic Apollon with the whooper swan and the myth which formed the starting-point of this study, and (3) the consistent connection of the Hellenic Apollon with the north and with amber. All these peculiarities are explained quite satisfactorily by the assumption that the Hellenic Apollon himself had his home in northwestern Germany and that he was carried into the Balkans by some migratory northern tribe, which doubtless followed much the same route as the Celts of Brennus in the fourth century before Christ and Alaric's Goths in the fourth century after Christ. This would also ex-

plain the decidedly Nordic appearance of the god, which has been pointed out repeatedly.47 Many of the most bitter controversies concerning the nature and origin of Apollon could have been avoided, had it been realized that the Apollon of the classical epoch is as much of a compound as was the Hellenic nation itself. Thus Apollon Smintheus, the "Mouse Apollon," was recently shown to be a Mysian tribal divinity of east European origin, introduced into the Troad by the invading Mysians in pre-Homeric times and identified with Apollon, though by no means with Apollon only.48 Apollon Lykeios is, as Wilamowitz49rightly conjectured, an Anatolian and pre-Hellenic sun-god, called Syros, Lykos, or (in southern Italy) Soranus.50 Neither of these can reasonably be supposed to be responsible for the god's classical name, 'ArorXXWv, voc. "AwroXXov, There remains, then, the Delph. Dor. "A7reXXov.51 it the that was "Swan possibility Kyknos, Apollon," who provided the name for the compound divinity as we know it. It should be noted, in the first place, that the memory of the original identity of the Hellenic and the north European god would probably not have been preserved, as it was, down to Alexandrian times, had a similarity of their names not favored it. In the second place, the resemblance of the name of the amber isle, Abalus, Celt. Avallon, to the name of the Greek god, whose paradise and sacred property that isle was fabled to be, can hardly be a mere chance. So we are led to inquire into the meaning of that name. Abalus, as the MG names for the two Frisian islands of Habel and Appelland clearly indicate, meant the "apple isle." It might be noted, 47J. L. Myres, Who Were the Greeks? (Berkeley, Calif., 1930), pp. 193 and 568, n. 63; H. R. Hall, The Ancient History of the Near East (London, 1913), p. 520, n. 1; cf. also my Mythologie universelle, p. 262; Gilbert Murray, Four Stages of Greek Religion (New York, 1912), p. 69. 48 CP, XXXVI (1941), 133-41. 49 Hermes, XXXVIII (1903), 575-86; Greek Historical Writing and Apollo (Oxford, 1908), pp. 27 ff. 50Cf. a study of mine forthcoming in Studi e materiali di storia delle religioni (Rome) (cf. above, n. 40). 1 As E. Meyer (Geschichte des Altertums, I, Abt. II [Stuttgart-Berlin, 1909], 640) by Pulenida, i.e., by a borrowpointed out, the Lycians rendered the name 'A7roXXwvirpls ing from the Greek, which would be difficult to understand if the Anatolians had had a name of their own corresponding to the Hellenic 'A7r6XXwv.

in passing, that there also is an isle called Ebelo in the Danish waters, northeast of the Little Belt.52 Similarly, the Celtic Avallon is obviously connected with Welsh afal, Bret. Corn. aval, 0. Ir. abhall, from which O. Celt. *ab-allo-s may be inferred. The medieval Latin texts usually render it by Insula Pomorum. On the other hand, the Hellenic Apollon has curious connections with the apple tree and its fruit.53 There are, first, two cult titles of Apollon pointing to the apple (/i7Xov): Apollon MaXe&rws(from /iaX'a, "apple tree"), a formation parallel with Dionysos 2UVKE&arlS(from avKea, "fig"),54 and Apollon MaXoels,occurring in an inscription from Lesbos55and in a Thucydides scholion, which tells an amusing story about the origin of the name.56 In Lucian's Anacharsis 9 Solon explains that the prizes in athletic contests are: "At Olympia a wreath of wild olive, at the Isthmus one of pine, at Nemea of parsley, at Pytho some of the god's sacredapples." This statement is corroborated by a Delphian coin showing the apples on the victor's table.57 In the sacred grove of Apollon near Klazomenai, known as the Grynaean grove, there was an apple tree which on one occasion was the object of a dispute between the seers Mopsus and Calchas, as to which of the two could divine correctly the number of apples on that tree.58 Joining these facts to the consistent association of the Celtic Apollon with apples and apple orchards and adding the Greek statements which tacitly identify the Nordic and the Hellenic Apollon, we are led to conclude that the two are one and that Apollon was from the beginning the god of the apple tree. The vocative "ATroXXoO Delph. Dor. "A7reXXov represents the original form of the name, as has repeatedly been suspected,59and it is a cognate of the Celtic and Teutonic names 52Broche, op. cit., p. 211. 63Rendel Harris, The Ascent of Olympus (Manchester, 1917), pp. 36 ff.; cf. also B. F. C. Atkinson, Bull. John Rylands Library, VII (1922), 138-40; Cook, op. cit., 487 ff. 4 Ibid., pp. 37 f. 56IG, XII, 2, 484; cf. Harris, op. cit., pp. 37 and 42. 56Cf. Revue de philologie, I (1877), drawing on the AEaS,LKa of Hellanicus. 185; cf. also Steph. Byz., s.v. MaXX6ELS [sic], 57 Harris, op. cit., p. 39. 68 Myth. Vat., I, 194; Serv. in Verg. Ecl. 6. 72; cf. Harris, op. cit., p. 40. 59Cf. A. Nehring, Mitteilungen d. schlesischen Gesellschaft f. Volkskunde, XVIII (1916), 23; cf. also the Thessalian form "A7rXovv(F. Bechtel, Die griechischen Dialekte [Berlin, 1921], I, 172).

of the tree and its fruit. The god and the tree thus stand in exactly the same relation as the Gr. IlpKos (later obpKvs) and the oak (Lat. quercus).60 Let us now see what light philology may throw on the history of the apple. The word is common to the Celtic, Teutonic, and Slavonic languages; but the IE languages of southern Europe have different names for the fruit, though Lat. malum is probably derived from the Greek (Ion. Att. ,/zXov, Dor. tuaXov).61On the analogy of a considerable number of other fruit names derived from the Latin, it has been frequently assumed that the Celtic and Teutonic forms may go back to some Latin designation such as malum abellanum, from the town of Abella, now Avella Vecchia, in Campania,62which was famous for its fine apples, as may be seen from Vergil's verse (Aen. vii. 740): Et quos maliferaedespectantmoeniaAbellae. What militates against this derivation is the regularity with which the Celtic, Teutonic, and Lithuanian forms obey the sound laws-a rare phenomenon in loan-words. The borrowing would have to antedate the first Teutonic sound-shift63 and antedate by far the period when we may reasonably suppose the apples of Abella to have acquired their well-deserved reputation. On the other hand, the name of Abella is not the oldest name of the town, which is said to have originally been named Moera.64Accordingly, Schrader, who at first had shared the traditional view, later arrived at the conclusion that Abella is named after the fruit,65a view shared also byJohannes Hoops66and R. Much.67 60Revue archeologique,XXXVI (5th ser., 1932), 87 ff. 610. Schrader, Real-Lexikon, I2, 53; V. Hehn, Kulturpflanzen und Haustiere (Berlin, 1911), pp. 627 f. 62 Schrader, op. cit., I, 53; Hehn, op. cit., p. 627. The hypothesis is as old as Cormac's Glossary; cf. Hoops, Waldbdume und Kulturpflanzen im germanischen Altertum (Strassburg, 1905), p. 478. 63 Schrader, Prehistoric Antiquities of the Indo-Germanic Family (London, 1890), p. 276. 64Serv. in Verg. Aen. vii. 740; cf. J. Beloch, Campanien (Breslau, 1890), p. 411. Hubert (op. cit., p. 57) aptly suggests that the name of the town tends to show that the Italian peoples once had the word and subsequently lost it. 65Real-Lexikon, I, 53. 66Op. cit., pp. 477 ff. Zeitschrift f. 6sterreichische Gymnasien, XLVII

Abella is simply one of the many apple towns scattered all over the map of Europe, from the Fr. Avallon, the Eng. Appledore, the Dutch Apeldoorn, the Ger. Apolda, to the Slav. Jablonka in the Carpathian Mountains, and even in Palestine: Tappuiah, from Hebr. tappud4h, Arab. tuffah "apple."68 The evidence therefore points to the fruit's having been carried down into the southern peninsulas of Europe by invading northern tribes; and it is natural enough that the god of the apple tree, Apollon, accompanied the fruit. This accounts for the consistent association of the god with the fruit and with a garden or orchard of the type described in the verses of Pindar and Sophocles. There remains one problem. The amber isle of Apollon was called Abalus or Abalum, later Avallon. The trees credited with exuding amber are, however, not apple trees but poplars; the Heliades are transformed into white poplars, not into apple trees. To this still another difficulty must be added: the poplar is certainly no north or north-central European tree, as its names in the Teutonic and Slavonic languages suffice to show; Eng. poplar and Ger. Pappel are derived from Lat. populus, as are Pol. topola, Russ. topol', Croat topola, Czech topola. The tree was introduced into the countries north of the Alps in Roman times and even later.69 Thus the amber-exuding trees on the banks of the Eridanos cannot have been poplars. So we ask next: If not poplars, what were they? Light is thrown on the subject by a Croat word for "poplar," synonymous with topola, to wit, jablan,70a word evidently related to Russ. yabloko, Pol. jablko, Bulg. yabuka, Czech jablko, "apple." In the Middle Ages the Serbo-Croat word jablan still had the meaning of "apple tree."7 Now it is a well-known fact that similar shifts in the meaning of tree names have been noted frequently. An excellent modern example is the Ger. Buche, which generally denotes the beech (Fagus silvatica) but which also designates the hornbeam (Carpinus betulus), an alto68 For a partial list of these cf. Harris, 69H. M. Willkomm, JHS, XLV (1925), 236 if. Forstliche Flora von Deutschland und Oesterreich (Leipzig, 1887), pp. 519 f., 530; Hoops, op. cit., pp. 230 f. 70F. A. Bogadek, burgh, 1926), s.v. 71 E. Berneker, New English-Croatian and Croatian-English Dictionary (Pitts- Slavisches etymologisches Worterbuch (Heidelberg,

gether different tree. The MG F6hre, OHG forhe, is the common word for the Scotch pine (Pinus silvatica), but the word is related to Lat. quercus,the name of the Prussian-Lithuanian thunder-god, Perkunas, Gr. IIOpKos (later 4bpKvs); thus it was originally the name of the oak tree. The Eng. yew, Welsh yw (whence Fr. if), and OHG iwa (whence NHG Eibe) designate a coniferous tree known to botanists as Taxus; but the cognate O. Slav. jiva means "willow." Again, to Russ. grab, grabina, Czech habr, hrabr, and Pol. grab (Carpinus betulus) appear to correspond Gr. ypa&lov,"torch," originally, probably, "oak wood," and Umbr. Grabovius,a surname of the old Italian oak-god Jupiter.72 Variants of the Lat. carpinus are Lith. skirpstas, "elm," and skirpstuis, "beech."73 The poplar is a south European tree genus. The Illyrians, on first occupying the lands between the Danube and the Adria, are therefore likely to have encountered it for the first time. They did not coin a new term for it but transferred to it their old word denoting the apple tree, without, however, giving up the old meaning of the word. Subsequent generations, no longer aware of the extension in meaning, and, a fortiori, the Greeks, who seem to have learned the traditions of the amber isle from Illyrian intermediaries, were thus led into the error of assuming that the amber-exuding trees were poplars. But it is clear, on this showing, that originally they were apple trees. To sum up the outstanding results of this inquiry, let us say that to the "Mouse Apollon" and to the Anatolian sun-god must now be added a third component of the classical Apollon, fully as important as the other two: the "Swan Apollon," the god of the whooper swan, of amber, and of the apple tree-a divinity introduced into Greece by invading northern tribes hailing from the amber coast of northwestern Europe. PRINCETON,N.J. 72 N. Jokl, Worterund Sachen, XII 73 Ibid., p. 75. On shifts in the meaning of tree names cf. also H. Usener, Gotternamen (Bonn, 1896), p. 202.