Unlike the worship of the Great Mother at Eleusis, the oracle of Delphi did not contain, to the Greek mind, any mystery. There were no unmentionable elements, no esoteric ritual. Philosophers felt free to discuss and explain it, geographers to describe it. Anyone who had the means might consult it freely, and in person hear the ravings of the sibyl. And yet the exact arrangement of the oracular shrine, the mechanics of its operation, seem almost as vague as those of Demeter's Telesterion.
The material remains are unfortunately very slight, in fact almost imperceptible today, and it is only by the illuminating studies of M. Courby, a decade ago, that we have anything to guide us other than the confused allusions of the literary sources.
M. Courby has shown1 that within the great temple, quite at one side, stood a little structure, the adyton, sanctuary of the oracle, very old in origin, even perhaps founded on prehistoric masonry. This edicule, he says, was of two stories, the upper room for the consultants of the oracle being built above an obscure crypt where, in the rocky floor, was supposed to be the mysterious fissure, the
That is all; for the arrangements and furniture of the edicule, we must rely on written statements which seem singularly contradictory and, whenever specific, are unfortunately late. The chances are, however, that in this place, perhaps the most venerated spot in Greece, all cult practices were maintained with great conservatism. And therefore, though the fertile Greek imagination was undoubtedly continually occupied with revising and reinterpreting its legends, the basic mechanism of the oracle may well have been the same at the beginning of the Christian era as in the fifth century before Christ. In that case, if we assume that a common foundation of fact underlies the often conflicting testimony, the very contradictions of our many sources may serve to point the truth.One of the earliest, and certainly one of the clearest descriptions which we have, is that of Strabo.2 He says: "The place where the oracle is delivered is Said to be a deep hollow antron, the entrance to which is not very wide. From it rises an exhalation that inspires a divine frenzy. Over the mouth (stomion) is placed a lofty tripod on which the Pythian priestess ascends to receive the exhalation, after which she gives the prophetic response."
Diodorus,3 after telling the legend of the discovery of the intoxicating vapor issuing from the rock, and how, to avoid accident, a priestess was chosen to be the sole medium of its inspiration, says: "And for her an apparatus was constructed, on which, climbing up in safety, she became inspired and prophesied ... and the apparatus had three feet, from which it was called a tripod."
Euripides4 also says that a maiden of Delphi has her seat upon a most holy tripod, whence she chants oracles uttered by Apollo.
Plutarch5 refers to the sacred vapor which formerly inspired the priestess, and Pollux6 remarks: "The cover upon the Delphic tripod on which the prophetess sits is called holmos." [202]
From this it is clear that at the beginning of our era the general belief-and probably the actual fact-was that the priestess sat upon, or used to sit upon, a high tripod, and there inhaled vapors coming from the ground, which threw her into a prophetic frenzy. The existence of a well known vase-painting found at Vulci7 (Fig. 1) representing Aegeus consulting Themis-by tradition one of the earliest prophetesses at Delphi-who is perched upon a tripod, indicates that this strange form of seat was at least as old as the fifth century B.C. There are also many representations of the god himself, instead of the Pythia, who was but his visible substitute, seated upon the tripod, and Euripides refers often to Apollo so enthroned.8
As for the source of time vapor, in the words of Suidas:9 "He (Apollo) chooses for prophesying, women frenzied by the breath of the Python"; Lucian10 says that "the serpent placed beneath the tripod (at Delphi) has a voice only because there is a serpent shining among the stars": the hymn to Apollo11 derives the Pytho from
, because the serpent slain by Apollo rotten in the sun and gave forth a putrid odor. These statements indicate that the Python, alive or dead, was supposed to be the source of the mephitic exhalation which rose beneath the tripod.In modern times the tripod itself has been a matter of much discussion, some believing that there were two at Delphi, one the mantic tripod of the Pythian priestess, the other the funeral tripod of Dionysos. For Clement of Alexandria12 thus relates a current myth: "The Titans, they who tore him to pieces, place a bowl (lebes) upon a tripod, and casting the limbs of Dionysos into it, boil them down; then piercing them with spits they hold them over Hephaistos ... Zeus time Thunderer discomforts the Titans and entrusts the limbs of Dionysos to his son Apollo, for burial. In obedience to Zeus, Apollo carries the mutilated corpse to Parnassus and lays it to rest." Arnobius13 repeats the story of the luring away of Dionysos by the Titans, in order to tear him to pieces and cook him in a pot; and of the Titans being cast into Tartarus by the thunderbolts of Zeus. It is quite possible that he copies directly from Clement, but he does not mention the entombment at Delphi. The Etymologicum Magnum14 says: "The Titans tore apart the limbs of Dionysos, cast them into a lebes and gave them to Apollo. This was set upon the tripod by the brother." That this savage legend was not purely a Christian concoction is attested by Tzetzes,15 who gives the following clear and definite statement: "Together with [203] Apollo. Dionysos was also worshipped in the innermost part of the temple at Delphi, as follows: The Titans cast into a lebes and gave to Apollo the limbs of his brother which they had torn apart, and he set it up upon the tripod, as Kallimnachus (c. 256 B.C.) says. And Euphorion (c. 235 B.C.) says they cast the divine Bacchus into fire above a bowl. We can be reasonably certain from this, that at least as far hack as the early third century it was believed that the bones of Dionysos lay in a bowl on a tripod in the Temple at Delphi. That this notion was widespread and persistent is shown by the definition of Servius;16 "Cortina, the place from which the oracle is given, because the heart (cor) of the seer is kept there."
But if we admit, beside the Pythia's throne within the edicule, a second tripod to hold the fragmentary bones of Dionysos, we are unable, logically, to stop there, but may have to admit a third. For Hyginus17 relates this parallel tale: "Python was a huge serpent, the son of Earth, who before Apollo's time, used to give answers from the oracle on Mount Parnassus ... Apollo ... killed Python with his arrows ... He cast his bones into a vessel and placed them in his temple and gave funeral games for him, the games which are called Pythian." And the uncritical Servius18 supports him thus: "In the same temple (at Delphi) there is a tripod with the bones and teeth of the Pythian serpent."
It is hardly reasonable to suppose that within time temple there weere two tripods set up by Apollo, each containing the bones of a slain divinity. And when we recall the probability that Dionysos, son of Semele, god of the grape, is only a later personification of the fertility serpent, son of Ge, god of growing grain, whose worship he had replaced at Delphi, the truth is evident. The story of how the one met his death, lured away as a child by the teacherous fire demons, has the ring of extreme antiquity. It parallels the tale of the infant Cretan Zeus tempted by playthings away from the protecting Kouretes, and may easily be related to the seasonal myth of the benevolent hero-god-Amphiaraos, Asklepios, and at Delphi, Python-destroyed by the fires of heaven. For our purposes, the common factor in the two stories is that there were believed to be sacred bones in the lebes of a tripod in the adyton.
The origin of this belief is suggested by Suidas,19 who says: "Near the city (Delphi) was a temple of Apollo, called Pytho. In this a brazen tripod was set up and above this was a basin which held the prophetic lots. Whenever those who were consulting the oracle asked a question some were drawn out, and the Pythia, employing them, or becoming possessed, declared that which Apollo revealed." Here we have a perfectly reasonable explanation of the presence of the "bones" in the lebes. Prophecy by the drawing of lots seems to have been of great antiquity in Greece. It was the method employed at Dodona, where an oracle existed from pre-Greek times, and there is a general probability that it was the method practiced at Delphi before the coming of Apollo, when Pytho and his two-phased mother, Ge-Themis, held the shrine. The more dramatic form of prophecy by inspired frenzy is often supposed to be an importation from the orient.20 If so, there can be little doubt that it quickly superseded the older method in its hold on the popular fancy, so it is not surprising [204] that in later days the presence of "bones" in the tripod basin should have given rise to explanatory legends, such as Greek wit was never at a loss to supply. In fact it may easily be that the tripod was originally set up solely for holding lots. For such a purpose it would be an ideal mechanism, while as a seat, it verges on the ridiculous. In any case there can be no doubt that the tripod which Suidas refers to as holding the lots, was the mantic tripod of the adyton at Delphi. Hence if the reputed bones of Python-Dionysos can be identified with these lots, there is no reason to assume a "tripod of Dionysos" distinct from that of the Pythia. Moreover, there is nowhere in ancient literature any reference to two or more tripods in the mantic shrine.
But it was probably not simply whim that led the Greeks to identify the bones in the tripod as those of Dionysos. For, quite aside from the legend of the Titans, there was a general belief that Dionysos was buried in the adyton. Plutarch,21 without mentioning the tripod, says that "the Delphians also believe that the remains of Dionysos are laid up with them, beside (or within-the word
is very varied in its meanings) the place of prophecy." Eusebius22 says that "meeting Perseus in battle he (Dionysos) was killed, as Deinarchos-the poet not the rhetorician-says. And it is still possible for anybody who wishes, to see his tomb in Delphi, near the golden Apollo," and Pausanias23 particularizes the locality by saying that the golden image of Apollo was in the innermost part of the temple, where few entered.The correctness of the quotation from Deinarchos is supported by Joannes Malala,24 who relates of Dionysos that he "fled from Lycurgus and, going to Delphi, died there. And the remains were placed there in a grave (soros) ... as the very wise Deinarchos wrote about Dionysos himself": and by Cyril,25 who says: "For Deinarchos, a poet not without fame, relating the deeds of Dionysos and all that he experienced in India, clearly tells how he killed Lycurgus and Actaeon, and was himself killed by Perseus and was laid to rest in Delphi near the Apollo called golden." Malala26 gives further evidence from an earlier authority than any yet encountered. He says that "Philochoros states in his exposition about Dionysos, that his tomb can be seen in Delphi, beside the golden Apollo and his grave
is thought to be a basis 
on which is inscribed, 'Here lies Dionysos the son of Semele, dead!' Similarly, the very wise Kephalion has also set forth the same things in his writings, on the same subject." How a bathron could be a soros is difficult to understand, but if we interpret the passage as a transformation of 
27 the sense becomes perfectly clear. "Under the monument (soros) there is to be seen a basis on which is written, etc." For his part Cyril28 gives a most peculiar versionof the same detail: "And Porphyry in speaking about Pythagoras says ... he sailed to Delphi in a ship and inscribed elegiac verses on the tomb of Apollo, in which he makes it clear that Apollo was the son of Silenus, that he was killed by the Python and laid to rest in what is called the tripod."A more tangled story than this last can hardly be imagined, and yet there is such a basic resemblance between it and that quoted from Philochoros, as to establish with [205] fair certainty that in the adyton near the golden statue of Apollo was some sort of soros, standing on a basis inscribed with elegiac verse to the effect that someone, son of someone, was buried there. Philochoros was a serious and careful antiquary; the few remaining fragments of his writings are among our most reliable authorities, and coupled with the quoted accord of Kephalion and the local belief which Plutarch mentions, there can be little doubt that the name on the basis in the fourth century B.C. was "Dionysos, son of Semele." By Porphyry's time the inscription may have become so obscure as to account for the extraordinary rendition given by Cyril.
But there is one point of considerable importance in Cyril's garbled tale. The inscription according to him said that the tripod was the resting place of the dead god. Probably the inscription on the basis said nothing of the sort, but merely used the phrase which Philochoros quotes "
." And that phrase "here lies" was thought to refer to the tripod rather than to the soros which stood upon the basis. For this to be the case, of course, it would be necessary for soros and basis to be in immediate juxtaposition with the tripod, but granted that, it is perfectly obvious how the prophetic "bones" which rattled when the frenzied Pythia set the tripod shaking came to be considered those of the god referred to in the nearby inscription.Strictly speaking, a soros is a tumulus, but that in the adyton at Delphi was obviously a small monument. A clue to its nature is given by Tatian,29 who says that "in the temenos of the son of Leto, there is a so-called Omphalos; now the Omphalos is the grave of Dionysos." And this statement is supported by the variant one of Varro,30 who remarks: "What the Delphians call the center of the earth is in the temple somewhat to the side, a kind of thesaurus which the Greeks call Omphalos, which they hold to be the tumulus of Python," and by that of Hesychius:31 "There also the serpent was shot with the bow, and the Omphalos of the earth is the grave of Python." Dionysos and Python are here again interchangeable names for the young fertility god whose grave was reverenced. A solution is thus offered for the controversy between Bouché-Leclercq,32 who maintains that the mantic frenzy of the Pythia was Dionysiac in origin, and that Dionysos must have preceded Apollo at Delphi, as otherwise he would never have been admitted there at all; and Dempsey,33 who disagrees, on the ground that the cult of Dionysos in Greece is altogether post-Apolline. These two views are easily reconciled by the hypothesis that there was a pre-Olympian, orgiastic, mantic cult at Delphi, where the annually dying son of the Earth-Mother (later confused with Hera) was worshipped as the serpent Pytho; and that afterwards, when the worship of the same or a cognate god with the name Dionysos, was introduced into Greece, the newcomer took over all the prerogatives of his older self which had not already been assumed by Apollo. The upright stone marking the chasm where the stricken serpent-boy was swallowed, became then the Omphalos above the grave of Dionysos.
This Omphalos, says Varro,34 was a sort of thesaurus, which we may take to mean a [206] "bee-hive" tomb, or more likely, as Karo35 has it, a savings bank in bee-hive form, which, of course, is in miniature just the form of the tumulus or soros that marks a hero's grave as represented in vase-paintings from very early times (Fig. 2). Of course it is possible that the statement of Hesychius does not refer to a monument at all, since, as Roscher36 has shown, the phrase "Omphalos of the earth" may very well have meant originally the stoma or cleft in the rock, and only later have been applied to the bee-hive shaped monument that stood there.37
So far, it appears clear that in the adyton there stood near to the golden statue of Apollo the so-called Omphalos upon an inscribed base, and close beside this the high tripod, in the basin of which were the prophetic lots, and on which the priestess sat to breathe the noxious vapors coming from below. On a number of vases Omphalos and tripod are thus represented, close together, though the value of such testimony is somewhat lessened by the fact that in other representations each stands alone. There is also a fine fifth century relief found at the Piraeus (Fig. 3),38 which shows Apollo seated on the tripod and at his feet the Omphalos flanked by two [207] regardant eagles. Roscher39 points out the great similarity between this representation of Omphalos and eagles and two other well cut reliefs of the same period40 -the only difference being that in both the others the Omphalos and eagles are on a low rectangular base (Fig. 4). He suggests, with good reason I think, that the three are intentional representations of the real monument as it appeared in the fifth century. It may be also, that the grouping of Omphalos and tripod in the first relief is a careful rendering of the actual relation-ship between the two.
Fortunately, the whole arrangement can claim to be more than a reasonable hypothesis, because of the support of material evidence almost miraculously preserved at Delphi. This consists primarily of a block of St. Elias limestone (Fig. 5)41 found in the western end of the temple. It was evidently a pavement block, and as M. Courby shows by a careful process of elimination, almost certainly came from the adyton. A channel cut in the top surface divides it roughly into two equal parts; on one stood some sort of a basis nearly square in plan, on the other a basis which would have been circular if a segment had not been removed to allow the [208] channel to pass between the two bases. Except where it was covered by the bases, the surface of the stone is encrusted with a heavy calcareous deposit, due undoubtedly to long continued wetting of the floor with the lime-impregnated water of Delphi. The channel was obviously to drain this water away from the vicinity of the round basis. Near the center of the area covered by the square basis a hole, roughly 0.15 m. square, has been cut straight through the block; the sides of the hole are not carefully dressed and the corners are rounded. It is three or four centimeters wider at the bottom than at the top (Fig. 6).42 Within the area covered by the circular basis are a number of cuttings, two for hook clamps to hold the block tightly to its neighbor, two small circular holes, and three rectangular ones grouped in a triangle about a fourth slightly larger rectangular hole. M. Courby considered this last cutting to be a lewis hole made for lifting the block, though for lifting by a single lewis, the lewis hole should be in the center of the block.
M. Courby points out that an inverted anathyrosis on one end indicates that the block has been reused, and may therefore have been reduced in size at some time. Even so the block would have had to have been of extraordinary size for the supposed lewis hole in question ever to have been in the center of it. Moreover, if the inverted anathyrosis indicates reuse at all,43 it indicates a second using when all the cuttings were on the under side, wherefore all and not merely the central one would belong to the period before reuse. The other three rectangular holes M. Courby considers to have been for dowels to hold the circular monument. He cites44 as evidence of the use of triple dowels in columns at Delphi a column base found in use as an altar in the church of St. Elias with a similar grouping of holes. But sections of columns before the fourth century are regularly joined to one another by single square empolia centrally placed; from the fourth century on, the use of iron dowels leaded into the stone becomes common, but where dowels are used there are never three, so far as I know; they are placed [209] relatively near the circumference, and the lower end of the dowel is always leaded in place by pour channels cut on the upper surface of the lower block. On the column base in question the three so-called dowel holes are rather near the center, there is lead around them but no sign of pour channels; there is a square central cutting for an empolion, and a single iron dowel actually in place on the lower side of the base. The Ionic capital45 which M. Courby thinks belonged, together with this base, to the inner order of the temple, shows a single central empolion hole and no sign of peripheral dowels at all. Since there are also three circular boles in the column base, which M. Courby attributes to a reuse, I see no reason why the triangular group of square holes should not themselves be due to a reuse, as sealings for some monument set upon the base after it had ceased to serve for a column.
With the stone from the floor of the edicule there is even less reason to believe that the triangular group of holes were for dowels to hold a circular monument, for not only are there no traces of pour channels, not only are the three holes placed far from the circumference of the circle, but they are not by any means concentric with that circumference.
On the other hand, the center of a circle drawn through the three is found to lie in the so-called lewis hole. Taking these four holes together it would seem obvious that they were for the attachment of the three legs and central columnar support of a metal tripod. The arrangement is not at all uncommon in Greek remains, and I am at a loss to see why, with his fertile imagination and clear insight, M. Courby has not suggested it here.46
A line drawn from one to the other of the two circular holes passes directly through the central rectangular one, and though they are not quite equidistant from this, they are very nearly so, much more nearly than they are equidistant from the center of the surrounding circle of the water-channel. M. Courby considers that these two round holes were for dowels of a later period than those he assigns to the rectangular ones. Again I am unable to follow him, for circular dowels, except as small pins, are a form quite unknown to me, and the relation of their position to the rectangular holes would indicate that all belonged to the same system. If the central group were for a tripod, I would surmise that the circular holes were for two upright rods or thin supports of some sort flanking the tripod at diametrically opposite points.
It can be seen (Fig. 6) that the hook clamps at the left of the stone have been roughly dug out for the sake of the metal, as was commonly done in mediaeval times. But from the other holes in the block the metal has been carefully removed without damage to the surrounding stone; evidently this was done in antiquity. It seems probable, therefore, that at some time the tripod was removed from the floor slab in order to raise it upon a circular base. This would account for the fact that the center of the group of holes does not correspond with that of the circular water-channel, the size of the circular base adopted requiring that the center be moved farther away from the adjacent square basis, which we may suppose to have been [210] already in place. Apparently the circular base was set tangent to the square one; but still later, the necessity for draining the floor caused the water-channel to be cut between them, and clipped a sector from the circle. The two hook clamps, I am at a loss to explain. One could hardly have found room for a proper hold in the adjoining block, within the circle of the base and water-channel, and yet it is difficult to imagine these two clamps set in the floor with no attempt to cover them. Since the method of removal indicates that they remained in place to a late time, it is possible that they date from some late reuse of the stone or from a time when a still larger base, covering the channel was set beneath the tripod. For the matter in hand, however, that point is of little or no concern.
Now a tripod of ample size-the space between the two round holes would permit a maximum of .60m. and the circle of the feet would require a minimum of .40 m. for the diameter of the lebes-set on a base within the place of prophecy, would almost certainly be the mantic tripod on which the Pythia sat. Perhaps from the holes on either side sprang the laurel boughs with real or metal foliage which the priestess in her frenzy is said to have shaken when she prophesied.47 And what more natural than that the Pythia, having filled a cup with the water of Kassotis from the crypt beneath the temple,48 should pour a libation of the sacred liquid round the tripod as she mounts and drinks? This would explain the water on the floor, which to judge from the channel was present chiefly around the circular base.
In fact, on the Vulci vase (Fig. 1), Themis, as she sits on the tripod, is shown with a sprig of laurel in one hand, and in the other, a phiale evidently for pouring libations, and a parallel representation on a Greco-Etruscan bronze cista (Fig. 7),49 shows Apollo holding a large branch of laurel in one hand and a phiale in the other as he sits upon a three-legged stool before an Omphalos covered with a network of fillets and surmounted by an indeterminate bird.
There is one argument, but only one so far as I can see, against the likelihood of the prophetic tripod having stood upon this slab. That is the express statement by Plutarch on five occasions,50 that one "goes down" to the oracle. The natural inference is that the prophetic utterance occurred in an underground chamber where the [211] tripod was set directly over a crevice in the native rock, the auditors seated in the edicule above, presumably hearing the sound of her ravings through some opening in the floor. M. Courby51 accepts this arrangement and so does Middleton.52 But the arrangement is not certain nor the objection fatal. For in the first place the upper story of the edicule may have been below the level of the temple floor, even though there was a crypt beneath it, and thus entering the edicule would be a "going down."
Probability supports this, for the temple is built on a slope which required foundations five meters high to bring the euthynteria on the south side to the level of the bed rock at the north, and the edicule, situated in the southern half of the temple would certainly not have its prehistoric floor more than three meters above the natural grade; even in the fifth-century Erechtheum the floor of the north porch is at a maximum height of 2 m. above the floor of the crypt. In successive rebuildings the floor of the edicule might have been raised, but conservatism surrounding so holy a spot would militate against this.
There is also material evidence for a lower floor level in the edicule than in the temple cella. M. Courby reproduces photographs of a wall block of the edicule, which he labels p.53 This block has disappeared since excavation, but photographs show a dowel hole in the upper surface and a dressing away of the top in a narrow band along the upper inner edge. This dressing is best explained by the hypothesis that the wall which stood upon the block was narrower than p, and the floor adjoining it lower, so that without the dressing a projecting corner would have appeared along the base line. Or it is possible that the cutting marks the level of the bottom rather than the top of the floor slabs, and that it was made so that these could be set close against the wall. I see no other good explanation. With the first and more probable hypothesis, if stone p was in situ when found, the floor of the edicule would have been slightly over one meter below the floor of the temple nave. With the second hypothesis the floor of the edicule would have been the thickness of a floor slab (about .44 m.) higher, but still at least two steps below the nave. One citation from Plutarch's Timoleon,54 supports this interpretation of "going down." He says that "going down into the manteion, from among the votive offerings suspended there, a fillet which had crowns and victories embroidered on it, slipped and fell directly upon the head of Timoleon." No private individual, not even Timoleon, would have been allowed to enter a secret and sacred underground cavern, and no normal college of priests would have hung votive offerings where they could not serve to stimulate the generosity of future visitors. The manteion which Timoleon went "down into" was in all probability the audience room of the edicule where consultants attended the utterances of the Pythia and where "ex voto" offerings would be hung.
Another passage of Plutarch shows definitely that the consultants were witnesses of the frenzy of the Pythia and therefore, that the tripod was not hidden in a secret cave. On one unfortunate occasion the priestess "went down against her will, but at the first words she uttered [212] showed she was possessed of an evil spirit.
Finally running toward the door to get out, she threw herself on the ground so that the pilgrims as well as the priests fled for fear."55 "Running toward the door to get out" gives much more the picture of a room above ground than of a subterranean cavern, and it is obvious that in this case pilgrims, as well as the priests, were present in the place of the Pythia's frenzy.
A second interpretation of the "going down" of the priestess would permit the mantic tripod to be in an upper room rather than in the crypt. This is, that the priestess first went down into the crypt for the water of Kassotis, supposed to flow there, and then coming up again mounted the tripod. Some such preliminary descent to the sacred spring is clearly implied in the passage cited from Pausanias,56 and in the words of Lucian:57 "Whenever his prophetess after drinking from the holy well, and chewing laurel and setting the tripod ashake-etc." It was probably a customary part of the ritual. Still, from the passage in Timoleon, I think it probable that Plutarch habitually used the phrase "going down," or "
," to mean simply "entering the edicule."If the prophetic tripod stood upon the circular basis, what was on the square one close by? We have mention of a golden statue of Apollo within the edicule and an Omphalos and a tomb of Dionysos; the last two were probably the same, and there is no record of any other monument there. Nothing is known in detail of the golden Apollo, but it can hardly be that any statue would be set close against the tripod. The Omphalos, however, by amazing luck has been preserved (Fig. 8), a small rough bee-hive shaped block of poros, .385 m. wide across the base by .287 m. high, which M. Courby58 has identified as the very ancient sacred stone of Delphi, said to mark the center of the Earth and also said to mark the grave of Python.
This Omphalos might be set on a low square base beside the circular one on which the tripod stood.59 The size of the rectangle outlined on the floor slab would give room on either side for the golden eagles close against the stone, as shown on the reliefs cited above (Figs. 3, 4). What then would be the purpose of the hole cut through the floor slab underneath? Obviously there would be no reason for such a hole beneath the base of the golden statue. But the Omphalos shows what is most remarkable, a hole about four centimeters square pierced through it from top to bottom.
M. Courby60 suggests that this may have been to hold a wooden shaft serving as a support for the golden eagles, or some other object, on top of the Omphalos. But such a purpose would not account for the proportions. A depth of .287 m. for a width of less than .04 m. quite removes it from the class of dowel holes or holes for empolia, while the fact that it goes clear through the stone at once recalls the hole cut through the pavement slab, and lends strong support to the hypothesis already developed on other grounds, that the Omphalos on a low square inscribed base stood on this particular slab, with a continuous axial hole worked through all three stones.
In the top of the hole in the Omphalos is wedged the iron blade of a knife (Fig. 8), [213] with two iron nails driven in between it and the stone. All this M. Courby thinks was an attempt, at a late period, to wedge more tightly the shaft that he reconstructs in the hole. Aside from the fact that a mast set in so deep a socket would hardly need wedging, the whole combination of blade and nails is too crude an arrangement to be conceivably the work of hands trying to repair or improve the Omphalos. It seems not the work of restorers but of vandals. I would suggest that the knife blade was inserted to extract a metal lining from the hole, and that having become stack fast there, the nails were used in a vain attempt to pry it out. The lining tube has gone; it must have been of bronze or other metal more precious than the knife blade.
What could be the purpose of this tube or pipe running from the crypt up through the floor and through the Omphalos? M. Courby, who places a tomb of Dionysos where I have placed the tripod, with an attendant altar above the hole, thinks that the
was to allow the blood of victims to flow down and impregnate the earth. The explanation is reasonable, though I know of no parallel example of a Greek altar with an internal drain; but if the prophetic tripod instead of a tomb were in the adjoining space, then the purpose of the hole must have been quite different. The idea suggests itself that the pipe served to convey the inspiring fumes to the feet of the Pythoness. There is no existing statement that the 
came through the Omphalos, but there is evidence that it arose beneath the tripod,61 and circumstantial evidence that the Omphalos stood close beside, and below.The "breath" is repeatedly said to have come from the
and the stoma and Omphalos are almost interchangeably indicated as the center of the earth. For both to hold this position, they could not be far apart; strictly speaking one would have to [214] be directly above the other.62 There is a curious remark of the Scholiast on Lucian;63 "the chasm is called the umbilicus of the earth because the air ascending from it and forming a link with heaven holds the earth suspended." The simile may explain why the pointed stone, with a column of vapor arising from it, was called the Omphalos.The ancients believed that the vapor which came forth beneath the tripod, issued from a hole in the earth. Excavation has shown there never was such a hole beneath the temple,64 and at Delphi volcanic gases or springs are quite unknown. Since the vapor is well attested, it must have been artificially created, as the various descriptions indicate. It was held to be mephitic, was called a "cold breath,"65 a perfume of the utmost fragrance.66 To pipe artificial vapor from a hole in solid rock is a difficult undertaking, but to pipe it through a hole in a floor built over an antron is not difficult at all. To one seeing the vapor rise from the floor and at the same time, through some open space, seeing that there was a hollow cave beneath, the effect would be quite convincing. To complete the evidence we have the holes through the pavement slab of the edicule and through the Omphalos.
These points force one to the conclusion that M. Courby has discovered not only the most sacred object in Delphi, the Omphalos of the earth, but that without recognizing it he has found also the slab on which both it and the other holy of holies, the prophetic tripod, stood in the adyton. As for the Delphic ritual, I suspect that when the Pythia went down into the crypt ostensibly to get the water of Kassotis, she actually lit a brazier there, and then coming up again and mounting upon the tripod, inhaled the fumes piped through the floor beneath her feet.
Just what the nature of the fumes was we cannot tell. Plutarch67 comments on the fact that instead of burning sweet-smelling incense, the Pythia used an ill-smelling combination of barley meal and laurel. Though unpleasant, barley smoke would hardly cause delirium, and laurel is quite harmless; possibly for her private inhalation, the priestess added some of the hemp seed used by the Scythians to produce intoxication;68 possibly also among the laurel leaves the Pythia chewed, were some of the highly poisonous oleander which resembles laurel, and is called rhododaphne by the Greeks to-day. The toxic element of this plant is a heart stimulant akin to digitalis in action.69 It does not by itself produce delirium, but what the result would be when mixed with fumes of hashish I cannot say, nor unless someone else conducts the experiment, do I ever expect to know.
, 23; cf. Schol. on Arist. Plut., 9, 39.
.
.
.
.
."
.
, 1909, p1. 8; cf. Roscher, op. cit., p. 86.
(Plato, Rep. IV, p. 427 c.), should be translated "seated above the omphalos," not "on the omphalos."
, to the "smoking omphalos" of Athens; this phrase has been variously interpreted. To my mind it signifies graphically the Acropolis (called the omphalos of Attica) with the smoke of sacrifice rising from it. The unusual expression may, however, have been suggested by a "smoking omphalos" at Delphi.