The fifteen stades from Phœnix river to Thermopylæ is the only one of these distances which we can test at the present day. The fact that it is very fairly correct makes for the presumption that the others are near the truth.

Anticyra seems to have been on the road. Herodotus does not state that it was actually at the mouth of the Spercheios. The muddy, marshy shore of the gulf is not a place where a permanent dwelling would be established.

The detail which shows how very far the ancient line of road was up the present plain of the Spercheios is the statement of the distances between that river and the Dyras, and between the latter and the Melas. This last is identified, with all but complete certainty, with the Mavra-neria, or Black-water river. If this be so, the Dyras can be no other than the Gourgo-potamo, (Leake, “Northern Greece,” ii. 25). The fact that these streams were nearly two miles and a half apart at the points where the road crossed them, shows that it must have been necessary to make a great circuit in order to get round the then shore of the gulf and the marshes which no doubt bordered it. The mouth of the Spercheios must in those days have been several miles above the site of the modern bridge of Alamana, which is at the present day at least four or five miles from the present mouth of the river.

Even two hundred years after this time, the mouth of the river must have been at a considerable distance from the west gate of Thermopylæ. Had it been anywhere near the promontory which projects into the plain at that point, the Greek light cavalry force which was watching the river bank must have been cut off from the main army by the body of Gauls whom Brennus sent to cross the mouth of the Spercheios. Infantry might easily make their way over the low hills above this west gate; but I doubt whether cavalry would be able to do so (Paus. x. 20; 7, 8; 21; 1).

The rapidity with which the head of the Malian gulf has been filled up is very striking. This is chiefly due to the main river Spercheios. The amount of solid matter brought down by that stream when the water is high must be very large. The extraordinary length of the promontories formed by the deposit at its mouth shows it to be so. But as far as the plain in the neighbourhood of Thermopylæ is concerned, it seems to me, after examining the ground, quite certain that the little streams which come down from Œta have contributed more than half of the material which has accumulated in the neighbouring plain.

Though not strictly bearing on the events at Thermopylæ, the description of the site of the city of Trachis is of considerable importance from the point of view of the general topography of this region in after-times.

Herodotus tells us that it is five stades, i.e. somewhat more than half a mile from the Melas river. He also says that the plain is broadest at this point, and that the Asopos ravine lies south of the city (H. vii. 199).

Does he mean that Trachis was actually on the road? His language is doubtful; but I venture to think that he does. He is, as we have seen, under the impression that in traversing not merely the plain, but even the pass of Thermopylæ, he was travelling south, whereas, after passing the Melas, he must have been going nearly due east Therefore, when he speaks of the Asopos ravine as being south of Trachis, he means probably that it was the next noticeable point on the journey to Thermopylæ. It would therefore seem probable that the Melas, Trachis, and the Asopos ravine are three points on his road. The road evidently crossed the Melas high up the stream, and the natural way from it to Thermopylæ would be along the firm ground at the foot of the Trachinian cliffs proper. It would thus pass certainly within two hundred yards of the mouth of the ravine. The determinants then in his description of the position of Trachis are (1) the distance from the Melas; (2) the position of the ravine with regard to it, for which we must read “east” for “south.”

Leake’s view, which has largely (e.g. in Stein, Rawlinson, etc.), though not universally prevailed, is that Trachis was identical with the lower part of that Heraklea whose ruins are on the mountain at Sideroporto. To one who has seen this region Leake’s description is very difficult to understand; and in the diary of his journey the time of arrival at various places which he sets down are in certain respects so incomprehensible as to lead to the suspicion that there is something wrong in his calculation.

The diary reads as follows:⁠—

The total result is that Leake did about thirteen and a half miles over a rough and confused track, with, presumably, baggage mules in his train, in two and a quarter hours. I can only say this, that what he took thirty-nine minutes to do (9.22 to 10.04, with three minutes’ delay) took me fifty-five minutes of fast walking on two occasions, the distance being three and a half miles, and the track difficult, for the most part on the stony bed of the Asopos. Nor can I understand how Leake, or any one else, could, between 8.46, or 8.59 and 9.03, get from the Mavra-neria to the place in his map where he indicates the position of the tombs. The statement is quite incomprehensible. Leake’s map is hopelessly incorrect; but that does not wholly explain the difficulty. As to the argument from the tombs, I can only say that the whole face, not merely of the Trachinian cliffs proper, but even of the rocky hillside which lies east of the Asopos ravine, is honeycombed with them. A large number were discovered in the making of the new road. My very strong impression is that Leake has placed the Trachis of Herodotus several miles east of its true site, and that, though it is not far from the Heraklea at Sideroporto, which may or may not have been the Heraklea of the Peloponnesian war, yet that it was not in any sense a lower part of that town. I am disposed to think that the identification of the site with the remains on the steep, flat-topped hill three miles west of Sideroporto, near the hamlet of Konvelo, will, whenever the ruins are excavated, prove to be correct.

Strabo (428) gives the distance between Heraklea and Trachis as six stades. If any reliability be attached to the statement, it can only proceed from the assumption that after the establishment of the hill fortress by Sparta, the population of one of the towns which it was founded to protect either removed itself or was removed to the immediate neighbourhood of its protector.

The junction of the Asopos with the Spercheios is at the present day, as has been said, a few hundred yards east of the bridge of Alamana, off the west end of the west gate. In Herodotus’ time it entered the sea independently, and it seems from his description as if its mouth must have been near the point where it now joins the main river. In the first place he expressly tells us that the Phœnix river joins it. That river, so called, is to be identified with a little stream which issues from the rocks of the west gate, whose bed is of a ruddy brown colour, owing, no doubt, to its being impregnate with oxide of iron. This fact alone shows that the Asopos mouth must have been at some point off, that is, north of the west gate.

The historian also tells us that it entered the sea near village of Anthele, which he describes as having been between the Phœnix and Thermopylæ proper. Leake in his sketch map places Anthele on the great accumulation of débris which is brought down by the stream which issues from the great ravine just about half a mile west of the hot springs. The site is impossible. This heap is traversed in every direction by ever-shifting branches of the torrent, and anything built upon it would be soon carried away. Such a site, on what is practically a huge stream-bed, would also be excessively malarious. Herodotus describes the situation almost unmistakably when he says (vii 200), “There is a broad piece of land about it in which are the temple of Amphictyonid Demeter, and the seats of the Amphictyons, and a temple of Amphictyon himself.” This must be the fairly level piece of land just inside the west gate, under the old Turkish cavalry barracks. The latter stand at the extremity of a steep-sided shelf which projects from the hill which forms the west gate; and I am strongly inclined to think that one or both of the temples, and the seats of the Amphiktyons, were on this somewhat commanding site, looking down upon Anthele.

Coming to the actual pass itself, along the line which Herodotus followed, we enter it by the west gate. Herodotus says (vii. 176) that the pass is not narrowest at Thermopylæ itself, i.e. at the middle gate, but before Thermopylæ and behind it, meaning by the latter, as his next words show, the east gate. It will be seen that by using the words “before Thermopylæ,” of the west gate, he again indicates the route by which he visited the pass. THE WEST GATE. “In front,” he says, “by the Phœnix river, near the city of Anthele, is another place where only one waggon at a time can pass.” The road at the present day does not traverse the whole of the west gate. Crossing the bridge of Alamana, it arrives at the actual foot of the cliff at a point somewhat east of the stream which is identified with the Phœnix. In former days, coming from the west it would pass along the whole of the foot of the steep slope or cliff in which this low promontory of Œta ends towards the plain.

The total length of the passage, from the west end to the site of the Turkish barracks, which I am inclined to identify with that of the Amphiktyonid temples, is just under one mile,—about 1670 yards. To the right, or south of the line of road, the hill rises either very steeply, or in the form of a cliff, to a height of between two and three hundred feet; above which the slope is gentle. The total height attained by the promontory is not great, only varying between three and four hundred feet above the level of the plain. The road was in 480 artificially constructed at this point, as is indicated by Herodotus’ use of the word δέδμηται (H. vii. 200). It can only have been a few feet wide. At the present time, the ground between the left or north side of the road and the Spercheios is in winter time marshy throughout; and even in the month of August, 1899, the eastern portion of it, extending from the road to the river, was impassable, though there had been a long drought. In Herodotus’ time it would seem to have been for the most part a marsh between the cliffs and the Asopos river, though it may well be that the sea actually came up to the road towards the east end of the gate. No passage is possible between the road and the modern river; much less can it have been so in B.C. 480.

Leake speaks of the pass as “the false Thermopylæ,” and seems never to have dreamt of its being a possible site of the battle of 480. It would be quite indefensible to any force which was not large enough to occupy and defend the wide, and, to a light-armed man, easy passage of the hills above it. In some of the best known guide books the battle is described as having taken place in this gate. I do not know the authority for this statement. Apart from its indefensibility against superior numbers, Herodotus makes it quite clear that the battle took place near the hot springs in the middle gate.

On issuing from the west gate underneath the old barracks, the land on the right is fairly level, the hill-side retiring somewhat. This is, as I have already said, almost certainly the site of Anthele. On the left of the road the marshy land still continues for a space. But within a few hundred yards the road ascends in the form of a long gradual hill, to descend further in like manner to the great lime and sulphur deposits of the hot springs. This rise in the road is caused by the intervention of a mass of stream débris, which has poured out of a great chasm to the west of the hot springs and has extended far out into the plain. Its magnitude is enormous absolutely, and, relative to the mean magnitude of that torrent which has produced it, astounding. On issuing from the ravine it forms a fan-shaped mass, one mile in length from the chasm to the plain, with an extreme breadth of three quarters of a mile. Its apex is over three hundred feet in perpendicular height above its base. It is formed for the most part of large material, the finer stuff having gone down to form the new land of the plain. This is what Leake calls the plain of Anthele. East of this great chasm this mass of débris is edged by the great face of Kallidromos, which rises almost sheer to a height of three thousand feet above it. The hot springs. The eastern part of this great cliff overhangs the deposit at the hot springs, a white, glistening mass of lime impregnated with sulphur, which covers an area of ground five and a half furlongs in length from east to west, with an average breadth of one furlong; that is to say, about fifty-five acres of land. The hot springs rise in the side of Kallidromos on the edge of the great mass of débris which has been described; and the water from them is carried along an artificial channel, immediately at the foot of the steep side of the mountain, for more than half a mile. The stream first enters the baths, and then turns a mill, which may be called the upper mill; after this it crosses the road and is carried by a solid stone aqueduct to turn a second mill in the plain. The springs are very copious. In the artificial channel the water rushes along rapidly in a stream three feet broad by about eighteen inches deep, and is of that bright clear green which Pausanias describes. The Middle Gate. The original middle gate of the pass was undoubtedly under the steep hill-side between the baths and the upper mill. Herodotus makes that quite clear. Of the mountain overhanging it, he says that it is “untraversable and precipitous, lofty, extending to Œta” (vii. 176). Of the middle gate he says, “Again the entrance to Greece, through Trachis, is at its narrowest point half a plethron in width” (vii. 176). THE MIDDLE GATE. “To the east (really the north) of the road the sea and the marshes begin.” “There are also in this entrance warm baths, which the natives call the caldrons, and an altar of Herakles is built above them. A wall had been built at this passage, and in old days there were gates in it. The Phocians built the wall for fear of the Thessalians when these came from Thesprotis to settle in the Æolid territory which they now possess. In case of the Thessalians attempting to subjugate them the Phocians took this precaution, and at that time they turned the hot water on to the passage, so that deep channels might be formed in the ground, using every device to prevent the Thessalians invading the country. This old wall had been built a long time before, and the greater part of it had fallen down with age” (vii. 176). “This place is called by most of the Greeks Thermopylæ, but by the natives, and by those who dwell in the neighbourhood, Pylæ” (vii. 201). “The Greeks with Leonidas, as though going to meet their fate, now advanced into the broader parts of the neck much more than before. Hitherto they had held their station within the wall, and from this had gone forth to fight where the path was narrowest” (vii. 223). Herodotus says that the Spartans “retreated to the narrow part of the way, and, crossing the wall, came and took up their positions on the mound, all of them in a compact body, except the Thebans. The mound is in the pass where stands the stone lion in honour of Leonidas.” The Three Gates. The fact that this was regarded as “the gates” par excellence is shown by the passage in vii 201. The contour lines of the map of the pass as it is at the present day illustrate in a very remarkable way the description of it given by the ancient historian. The curves of the five-yard contour line are most noticeable. It is at the three gates of the pass as described by Herodotus that it approaches most closely to the hill slope. At the west and east gates it reaches the actual base of the mountain. At the middle gate it all but does so; but the accumulation of matter from the hot springs has somewhat pushed it back.

It cannot amount, of course, to more than a rough calculation, but I am disposed to believe that in the cases where stream débris has been pushed forward into the plain, the ten-yard contour, and elsewhere the five-yard contour, represent pretty closely the coast line as it was in the last few centuries before Christ.

In 480 there appears to have been deep water close to the road, for Herodotus in his account of the final attack of the Persians on the pass speaks of many of them falling into the sea and perishing (vii. 223). The truth of this detail is strongly supported by Pausanias’ description (x. 21, 4) of how the Athenian fleet two centuries later attacked the flank of the Gauls when they were assailing the middle gate. I am inclined to think that the deep water must have been at the base of the two mounds which stand at the east end of it. The marshes spoken of (vii. 176) would be those formed at the point where the hot springs reached the shore, probably several hundred yards west of where they now come down to the aqueduct.

The breadth of half a plethron, or fifty feet, is evidently to be applied to that part of the pass which extended from the first mound westward in the direction of the modern baths (vide Plan). South of this the mountain side rises at a very steep angle, and, apart from its steepness, is practically untraversable, owing to the nature of the rock-strewn slope, and the thick brushwood which covers it.

The question as to the exact position of the Phocian wall is a somewhat difficult one to solve. At the same time, I see strong reason to think that the existing remains of wall foundations on the neck by which the first mound is attached to the mountain side are relics of a wall which, if not identical with, is identical in site with the one which the Phocians built.

Of this wall I noted the following details. The general line of it is distinct. It can be traced seventy or eighty yards south from the summit of the mound along the neck of the ridge, and it seems to have been carried down the north slope of the mound towards what is now the sulphur deposit but which was in former times the sea. The detail of its plan is somewhat indistinct, largely owing to its having been very much cut about in constructing rifle pits for the expected defence of the pass in the recent war with Turkey. It seems to have been planned as a series of angle bastions projecting westwards, connected by curtains a few yards long.

It is from three to three feet six inches thick, and formed of stones, some of which are roughly squared, and have an average dimension of fifteen inches by ten.

The significance of its position will best be understood when the course of the ancient road through this part of the pass is described. There can be no question as to the line of it. Coming from the west across the great mass of débris formed by the stream from the great ravine, it must have passed close to the site of the present baths. THE PHOCIAN WALL. Thence it must have run along the foot of the mountain slope until it reached the point where the first mound projects into the plain. As the sea must at that time have washed the north foot of this mound, the road must have climbed the west slope of it, which is not peculiarly steep, and must have passed over the neck which connects it with the mountain. Hence it would be blocked by this wall, which runs along the neck, and which was probably carried higher up the mountain slope to the south. I could not find any traces of its south end; that is to say, it certainly went farther up the mountain side than I could trace it. I have no reasonable doubt we have here the line, if not the actual remnants, of the wall of 480. The position is a very strong one, such as a small body of men could defend against largely superior numbers.

It seems fairly clear that the Phocians used the water from the hot springs to damage the road between this mound and the modern baths. The stream itself is not, and I do not think can have ever been, of sufficient magnitude to form a serious obstacle in itself. The defence would simply consist in making the narrow passage awkward, if not impossible for cavalry.

This suggested position of the wall accords with the details of the last struggle given in vii. 225. The Greeks fall back from the broader part of the pass (probably near the baths) to the narrow part (between that and the mound). Then they cross the wall and take their position on the mound. Furthermore, the fact mentioned at the end of the chapter, that in the assault on the mound the Persians pulled down the wall, shows pretty clearly where the latter must have been, i.e. on the mound itself.

On this mound certain stone graves have recently been found. I am afraid the discovery was made when the rifle pits were being constructed. I could only get a vague account of their position and none of their contents, if any.

One word of warning. Viewed from the side of the road, the first mound looks like an integral part of the mountain slope, and the existence of the deep little valley which separates it almost completely from Kallidromos can only be ascertained by ascending the mound itself, or by going up the valley between the first and second mounds of which the little valley is a branch.

The line of wall which lies to the north of the present road among the bushes west of the aqueduct to the lower mill has been suggested as a remnant of the Phocian wall. The lime and sulphur deposit upon it, similar to that on the present aqueduct, shows that it carried at one time the stream from the hot springs to a mill which must have existed at its extremity.

The distance from the east end of the west gate to the middle gate at the first mound is one mile five furlongs. From the middle gate to the east gate is one mile three furlongs. Adding to this the length of the long passage of the west gate, which is nearly one mile, we find that the total length of the whole pass is four miles. Taking into account the necessary windings of the ancient road, the length of the pass was probably four and a half miles at least to one travelling through it.

After crossing the stream from the hot springs, the modern road is flat for a few hundred yards. At this point it passes on the right a salt spring, which forms a pond beneath the third mound.

After this the road begins a long ascent up the side of another fan-shaped mass of stream débris similar to that found west of the hot springs. In this case, however, the mass is the combined work of a number of small torrents which come down from the mountains and issue on the plain at a short distance from one another. The great line of cliff which overhangs the middle gate now recedes. It continues indeed, but high up the mountain side, and from its base a series of bastions and rounded ridges come down towards the plain, separated from one another by the deep ravines which the streams have cut.

As you ascend the hill on the road you pass, on your immediate right, four more mounds or hillocks. The last, or sixth mound, is at the top of the hill, and is a prominent object when viewed from the hot springs. Shortly after passing it the road begins to descend again; and about half a mile farther on you arrive at what was the east gate of the pass. Here the passage was narrow and short. It was formed by a somewhat lofty projecting ridge of the mountain slope, round the end of which the road passed. The five-yard contour makes a sudden bend inwards at this point. Herodotus tells us that, like the western gate, it afforded passage to but one waggon, and that it was near Alpenos (vii. 176). At the end of the same chapter he speaks of Alpenos as being “very near the road,” and as being the place from which the Greeks reckoned to supply their commissariat. THE EAST GATE AND WALL. Site of Alpenos. I think that the site of Alpenos is to be identified with the remains of a walled acropolis on a hill which stands out into the plain from the ὑπωρέη or lower slope of the mountain, about half a mile beyond the east gate. Between it and the east gate there must have been a bay of the sea, or harbour, to which, no doubt, the Greek provision ships went. Leake places Alpenos quite close to the middle gate (vide his map), but that is quite inconsistent with H. vii. 176. The incomprehensibility of the eastern part of his map of Thermopylæ makes one suspect that either he had never visited the east end of the pass, or, if he had, had done so very hurriedly.

From the east gate there runs up the ridge towards the high mountain the remains of a remarkable wall (vide map). I cannot find in history any trace or hint of its origin. It was evidently a defence against an enemy advancing from the westward; for the whole way along its west front there is a steep fall in the ground. It goes up the hill for half a mile in a south-west direction, and then, in order to keep along the ridge, it bends through an angle, and may be traced going S.S.E. to a point near the foot of the line of cliffs near the summit of the mountain. It would afford a very strong defensive position to a large force. I am inclined to think that it cannot be a mediæval work, because by mediæval times the sea must have retired from the east gate at its foot. The stones of it are of medium size. I fancy the outside courses were of squared material, while the interior was of unsquared stones set in mortar. Its width seemed to vary from eight to ten feet. The defence of the line of the wall would be a necessary adjunct to the defence of the east gate. It will be seen, therefore, that the latter did not afford a position where small numbers could have faced an enemy in great numerical superiority.

In entering upon the details of Herodotus’ account of the battle of Thermopylæ, it is necessary to bear in mind that it is a story written by a man who was indeed excellent as a topographer, but who seems to have absolutely lacked, in so far as can be seen, that practical or theoretical knowledge of war which alone could have enabled him to judge of the probable truth or falsehood of the various motives of events given in the sources, whether verbal or written, from which he drew his story. Special circumstances in connection with the battle made it more than likely that the real history of it would be officially concealed. The only people who knew the whole story of the events leading up to the great catastrophe were the authorities at Sparta. They can hardly have anticipated a disaster such as actually happened; but, after it had taken place, they had every interest in putting a complexion upon it such as would absolve them, in so far as possible, from blame. The result in Herodotus’ narrative is similar to what we find in other like passages in his history: on the pure question of facts he is reliable; but when he attempts to give the motives lying behind facts, he is only too apt to produce unreliable statements from his sources of information. He must not be judged too hardly in this matter. It is probable that neither he nor any other historian writing after 450 could have arrived at the truth concerning the motives underlying the events of 480. It is not merely a question of the absence of reliable records. He must have been influenced by the view of the past taken by the men of his own day; and there is only too much evidence in history of the extent to which the Greek permitted his imagination to take liberties with facts. The very caution, too, which he displays as an historian is evidently a prominent characteristic of the man,—a characteristic which would almost inevitably lead him to shun a collision with the views of those of his fellow-men with whom he was most intimately connected in life and sentiment. But his silence is sometimes eloquent. In the Thermopylæ narrative he mentions Leonidas’ demand for reinforcements: he states that the intention of sending them existed; but he never hints that even a show of carrying out the intention was ever made. The inadequacy of the reasons which he gives for the omission to send help,—the occurrence of the Carnean and Olympian festivals, and the alleged miscalculation made by the Spartans as to the time at which the Persians would arrive before the pass,—is patent on impartial examination. He himself clearly shows that the Greeks had information as to the movements of the Persian army; for he says that Leonidas was despatched to Thermopylæ “when they heard the Persian was in Pieria.” FACT AND MOTIVE. Leonidas and his band were then in good time. Nor was any such mistake made in despatching the fleet to Artemisium.

And when it is all over, what is the alleged cause of Leonidas’ death? It is represented as an act of self-sacrifice. An oracle had declared that one king must die, that Spartan liberty perish not. And so Leonidas immolated himself on the altar of Patriotism.

This is a motive which would directly appeal to the piety of the historian. But some contemporary sceptic might well have asked why, under the circumstances, the hero of a deed of such splendid self-sacrifice should have thought it right to involve therein the Thespian and Theban contingents. That Leonidas’ act was one of the most glorious recorded in history there is no reason to doubt: the question is as to the motive.

How can the motive be judged of at the present day? Herodotus himself supplies the apparatus of criticism by his narrative of the incidents which occurred at Thermopylæ, which, when statements of motive are excluded, is in nearly every single detail capable of verification on the spot at the present day. The modern world possesses in it a sound basis for an induction of the causes which produced the events. The motives alleged here and in other parts of his history of the Persian War are little more than hypotheses for the verification or disproof of which the genuine honesty of the historian often supplies the means.

Xerxes, on arriving at the Malian plain, pitched his camp in Trachinia, commanding the country as far as Trachis. The Persian encampment would seem to have stretched from the river Melas, or thereabouts, to the entrance of the west gate of the pass. The Greek encampment was in the middle gate, probably behind the first mound, either on the ridge beyond the valley, or on the stream débris over which the modern road passes. The numbers of the Greek army must have been about eight thousand, including the unstated numbers of the Opuntian Locrians. Of these, however, one thousand Phocians were set to guard the path of the Anopæa, more than three thousand feet above the gate. Leonidas was in command. H. vii. 204. Even before the battle brought him undying fame he seems to have been a man whose character excited great admiration. The three hundred men whom he took with him from Sparta were the royal bodyguard, picked men all, and fathers of families, so Herodotus says. The historian evidently implies that in so desperate a venture Leonidas considered it unadvisable to take with him men whose death would blot out their name from the land of the living. Is this detail quite consistent with the presumed intention to send reinforcements? It is possibly a graphic touch added by Spartan tradition after the event.

The Theban contingent was four hundred in number. Herodotus’ account of all relating to it is so full of disputable motives that it has been suggested that he followed a tradition which developed either under the influence of the bitterness of feeling called forth by the attitude of Thebes later on in the war, or of the violent enmity which sprang up between Athens and Thebes in after-time, and which was at its height at the time when Herodotus wrote. He represents Leonidas as taking the Thebans with him under a sort of compulsion, because he suspected them of a tendency to medize. There are several reasons urged for doubting this version of the story of the Theban participation in the campaign.

Plutarch’s statements, in brief, are as follows:⁠—

The objections to the Herodotean version are by no means decisive in character.

Two theories are possible: either that these Thebans were taken as hostages for their fellow-countrymen’s behaviour; or that they joined the expedition with a patriotism which stood the test of the extremest danger. Perhaps the most telling piece of evidence against them may seem to be the smallness of their contingent when compared with that of the minor town of Thespiæ. But here again the explanation may be that they shared the evident distrust with which the Northern States regarded the Peloponnesian effort at Thermopylæ, and were ready to fulfil their share of the obligation whenever they saw Sparta willing to undertake hers.

The evidence is not of a kind which renders it possible at the present day to decide between the truth of the respective versions. The motives which Herodotus attributes to Leonidas and to the Thebans at this time may not bear the stamp of conclusive truth; but they convey, when examined by the side of the contrary evidence, a stronger impression of truth than of falsehood.

Xerxes’ first measure on taking up his encampment was to send a scout to examine the Greek position.118 He could not see the whole Greek force, because those within the wall were hidden by it.119

Part, however, of the Spartan force was outside the wall, the men being engaged in exercise and in combing their hair. Having ascertained their numbers, the scout returned to Xerxes with his report. His description of what was going on surprised the king, who appealed to the Spartan Demaratos for an explanation. The latter told him that it was the national custom of the Spartans to adorn themselves before facing the danger of death in battle, and warned him that he must expect a stubborn resistance. H. vii. 210. The king, it is said, did not credit this, and allowed four days to pass “expecting ever that they would run away.” Xerxes’ motive for the delay was probably somewhat different from that given by Herodotus. THE FIRST ATTACK. The king thought, no doubt, that his fleet would within a few days force the passage of the Euripus, and that, the position of the Greeks being thus turned, they would have to retreat, and would thus save him the inevitable losses which must result from an attempt to force a pass so strong.

On the fifth day his impatience got the better of him, and he began a fierce assault with his Median and Kissian troops. After the Persians, the Medes enjoyed the greatest reputation in the army. In spite of the disparity of numbers, it was an unequal contest. Numbers were indeed of little avail, the passage being very narrow,—only fifty feet in width—and the Greeks having either flank protected by the mountain and the sea. But it was in respect to actual arms that the disparity was greatest. The Greek hoplite, with his heavy defensive armour and longer spear, was far more than a match for the Mede, who practically lacked defensive armour, and whose weapons of offence were of an inferior character.

The battle lasted all day; H. vii. 211. for when it became clear that the Medes could effect nothing, Xerxes sent his picked Persian troops, the Immortals, to the assault. Even these signally failed to make an impression on the Greek defence.

The excellence of the Lacedæmonian drill and discipline seems to have been remarkable. They employed tactics which very few troops in the world could employ in a narrow space in face of immensely superior numbers:⁠—an advance; then a pretended retreat, to draw the enemy on; and then once more a right-about-face to the attack. The object of this was, no doubt, to keep the enemy at close quarters. Xen. Anab. iii. 3, 9. The one thing they had to fear was that the lighter armed foe should confine himself to an attack with missiles, in which form of fighting they must have suffered severely, without being able to retaliate in adequate fashion.

The attack was repeated next day. The Persians hoped to win the pass by sheer weight of numbers, knowing that they could afford to lose ten men for every one of the enemy whom they placed hors de combat. But again the attack failed.

The king’s position at this time must have been very critical. Neither army nor fleet had forced a passage, nor had they been able to resume touch with one another. They maintained a communication, no doubt, by the coast road running along the north shore of the Malian gulf; but the transport of supplies to the army, whether by that route from the fleet, or from the Thessalian plain by the way of Thaumaki, must have been a work of such difficulty and magnitude that it is hardly possible to believe that it could have been adequately maintained.

The failure to reinforce the garrison at Thermopylæ was the most disastrous mistake which the Greeks made in the course of the war. Even with the ridiculously small force which was sent, the defence came very near being a success. Had an army only half the size of that which fought at Platæa been there, neither pass nor path could ever have been forced, and the fleet at Artemisium, or at least the Peloponnesian section of it, would have fought with a very different spirit to that which it actually displayed. Their stake at Thermopylæ was too small, and their stake at the Isthmus too great to render them aught but vacillating defenders of the channel. Had the stakes been reversed, had the safety of their compatriots at the pass been largely involved, the fleet would have been inspired with unanimous courage. And had it been so, the advance of the Persians, in spite of their superior numbers, might have been definitely checked. It is not probable, judging from the results of the fighting which did actually take place, that either side could have gained a decisive advantage over the other. Though the Greek fleet suffered severely in the final combat, the Persians were not successful in forcing the strait; and to them failure in both strait and pass would have meant failure of the expedition.

Men said in Greece in after time that the Persian was the rock on which he himself made shipwreck. That expressed but half the truth. It was true in the sense that the success in the war was not owing to good management on the part of the Greeks. THE PATH OF THE ANOPÆA. It was not true in the sense that the Persian’s mistakes were the main factor in his failure. The main factor was conspicuously the superiority of the Greek weapons in close fighting, a form of combat which the Persian as the assailant could not avoid. It was a lesson which the European Greek took long in learning. Aristagoras the Ionian had pointed it out to Kleomenes when he visited him at Sparta in the first year of the fifth century. And yet it was not until the early years of the next century that the Greeks of Europe came to appreciate to the full the effectiveness of their national panoply as compared with that of the Persian. The incidents of the strange adventure of the 10,000 which Xenophon recorded at last convinced them of the real nature and magnitude of their strength when matched with the great Empire of the East.

By Malian treachery and Phocian cowardice Xerxes was saved from the desperate position in which he found himself after the failure of the two days’ assault on the pass.

The tale runs thus in Herodotus:⁠—