Popular Science Monthly/Volume 48/December 1895/Prehistoric Engineering at Lake Copais
By JOHN DENISON CHAMPLIN.
HOMER, in his famous catalogue of the Greek and Trojan forces in the second book of the Iliad, enumerates more than twenty towns around Lake Copais which contributed collectively to the Greek fleet eighty ships, in each of which
"Were six score youths, Bœotia's noblest flower."
Map of Lake Copais and Surroundings
lies Platfea, its soil enriched with the blood of Mardonius and his Persians ; and not far away Leuctra, fatal to Sparta's power, and Cadmean Thebes, home of OCdipus and of Antigone, birthplace of Heracles and of Dionysus, where Amphion sang and the Epig- oni fought.
The Copaic basin itself and its surrounding hills are dotted with ruined cities. On the west shore of the lake Minyean Or- chonienus, from whose colony of lolcus sailed Jason and the Argonauts, still dominates the plain with its acropolis, its walls two miles in circuit. Its temple of the Graces, with its musical festivals, drew thither poets and singers from all the Hellenic world. Homer compares its wealth with that of the Egyptian Thebes, and so powerful was it that it held subject all the sur- rounding region until Heracles slew its king and made it vassal to Thebes. A little west of it is fatal Chgeronea, where Philip of Macedon rang the deathknell of Greece, and where, two and a half centuries later, Sulla overthrew Mithridates. Between it and Helicon lies Lebadea, where Croesus and Mardonius sought their fate from the oracle of Zeus Trophonius ; and hard by is Coronea, famous for its temple of the Itonian Artemis and the Pamboeotian festival. Near the lake is Tilphusium, with its fountain of Til- phusa, where blind Tiresias drank and died ; Alalcomenae, which claimed to be the birthplace of Athene ; Haliartus, under whose walls Ly Sander fell ; Onchestus, founded by Poseidon's son, meet- ing-place of the Amphictyonic Council ; Acrsephise, noted for its oracle of Apollo ; and Medeon, Copfe, Holmones, Hyettus, Hyle, Peteon, and Ocalea, each famous in ancient story, and most of which sent ships and troops to Troy.
With all these evidences of pre-Homeric prosperity, one is tempted to ask, What has changed the conditions in this once favored and still fertile land, which to-day supports but a few thousand souls in scattered villages and hamlets ? We find the answer in Strabo, who says : " The spot which the present Lake Copais occupies was formerly, it is said, dry ground, and was cul- tivated in various ways by the Orchomenians, who lived near it." This traditional account, about the only record of the prehistoric condition of the Copaic basin we possess, would seem to imply that it was kept dry artificially, and we find a partial explanation in other passages in which he describes certain subterraneous cav- erns and fissures through which the waters were carried off. " If the subterranean passages are stopped up, the waters of the lake increase so as to inundate and cover cities and whole districts, which become uncovered if the same or other passages are again opened." The memory of such a catastrophe, caused by the stop- page of the natural conduits, the result of seismic disturbances, as Strabo intimates, or from want of care in consequence of political
�� � disturbances, is embalmed doubtless in the tradition of the Ogygean Deluge, Ogygea being the original name of Bœotia. A similar trouble must have occurred about the time of Alexander the Great, who appears to have contemplated the reclaiming of the
Sustaining Wall of Masonry.
basin. Strabo says: "When the outlets were again obstructed. Crates, the miner, a man of Chalcis, began to clear away the obstructions, but desisted in consequence of the Bœotians being in a state of insurrection, although, as he himself says in the letter to Alexander, many places had been already drained."
These statements of Strabo would lead to the inference that the drainage of the basin by the ancients consisted only in keeping free from obstruction certain subterraneous passages through which the waters flowed to the sea; and this would probably have been the conclusion to-day but for the recent efforts of the Greek government to reclaim the submerged lands. These efforts, under the supervision of experienced engineers, have resulted in nearly draining the basin, and have led to the discovery of a complete ancient system of hydraulic works dating from so remote a period that all record or tradition of their construction has been lost. This system, so vast and comprehensive as to excite the wonder of modern engineers, taking into consideration the primitive appliances of the ancients, served to convert this now miasmatic basin into a fruitful plain, the home, a thousand years before our era, of a thriving and numerous population.
To give a clear conception of these ancient works and of the problems which the prehistoric engineers had to solve, it will be necessary to take a brief topographical survey of the region. Lake Copais is the receptacle of the drainage of the valley of the Cephissus and of the plain of Chæronea, which is watered by the Hyrcinns, Permessus, Olmeus, Lophis, and other streams that descend from Helicon. All these streams flow in on the south and west sides, where the shores of the lake are simply a continuation of the adjacent plains; but on the north, east, and southwest, where the waters would naturally find an outlet to the sea, the banks form steep, rocky shores.
At the southeast extremity the lake ends in the Bay of Carditza, which is inclosed in a fold of Mount Sphingium, an offshoot of Helicon, and at the northeast in the Bay of Topolias or Kephalari, inclosed in Mount Ptoum. A depression in the flank of Sphingium is called the Hill of Carditza, and behind this, between it and Mount Ptoum, is a smaller lake, Hylice or Hylicus (Likeri). Further east, near the seacoast, lies Mount Messapium, with another small lake, called Paralimni, between it and Mount Ptoum. A similar depression in Mount Ptoum, east of the Bay of Kephalari, is called the Hill of Kephalari. The Copaic basin is
Plan of the Wells.
thus a natural cul-de-sac, with no apparent outlet; but the pentup waters have worn fissures through the limestone rocks underlying the hills and formed for themselves, perhaps with some volcanic aid, as Strabo suggests, subterraneous outlets into the Euripus or channel of the sea between Bœotia and Eubœa. There are twenty-three of these subterraneous passages, locally called katabothra, but many of them unite underground and only four reach the surface on the east side of the hills. Of 2 14 THE POPULAR SCIENCE MONTHLY.
these four outlets, one passes southward under Mount Sphingium into Lake Hylice or Likeri, from which there is probably a sub- terraneous communication with Lake Paralimni and beyond it with the sea. The three others pass eastward under Mount Ptoum and its offshoots. The principal one, called the Katabothra of Bynia, with two openings on the lake side, passes under the Hill of Kephalari in a general direction from southwest to northeast, and opens on the east side of the hill in a large grotto about forty- five metres lower than its source, whence its waters flow through the deep ravine of the Valley of Larymna into the sea,
Strabo says that Lake Copais is three hundred and eighty stadia (about forty-seven miles) in circuit, but it differs greatly at different seasons, sometimes threatening to inundate the whole valley and sometimes forming only a series of fens overgrown with reeds the auletic or flute reeds of the ancients, from which Pan's pipes were made. Its bottom, which is ninety-five metres above the sea, is a nearly level plain with a slight incline toward the east and a little elevation in the center. Modern travelers, from the time of Sir George Wheler upward, have noted on both its north and its south shores the remains of ancient dykes, in some parts re-enforced with masonry. These dykes, in several places still used as roads, have generally been considered as ancient causeways, means of communication in times of flood between the towns on the banks ; but they are now shown to be parts of a sys- tem of drainage canals by means of which the superfluous waters of the basin were led to the katabothra under the hills.
The recession of the waters through the efforts of the present engineers has shown that there were three main canals through the entire length of the lake, branching at their western ends into subsidiary canals or feeders for collecting the various tributary waters. These main channels, which for convenience' sake we will call the north, middle, and south canals, are constructed partly of excavations and partly of a series of dykes or cause- ways, strengthened where necessary by walls of cyclopean ma- sonry. The north canal, the most carefully and solidly con- structed of the three, receives the waters of the Cephissus and conducts them into a common channel with those of the Melas, a stream which, rising near Orchomenus, is navigable almost from its source. After their junction the waters flow through a bed, formed on the north by the rocky shore of the lake and on the south by a massive embankment re-enforced by masonry, be- hind the island of Topolias, the site of ancient Copse. Thence the canal leaves the shore and, embanked on both sides, crosses the Bay of Kephalari and conducts its waters into the natural fissures under the mountain. This double embankment, though partly ruinous, is still plainly traceable.
�� � The middle canal, constructed to receive the waters of the Hyrcinus, traverses the center of the lake and connects near the east end with the south canal. Unlike the north canal it is made wholly by excavation, the earth thrown out forming its banks, and it is nowhere strengthened by masonry. The alluvial deposits from the stream have nearly obliterated it in parts, but its course is easily traceable and in some places, especially near the western end, the works exhibit formidable dimensions, showing in cross-section an excavation of more than a hundred metres to the lineal metre.
The south canal, for the reception of the waters of the Permessus, Olmeus, and other small streams from Helicon, starts from the southwest end of the lake, and, following the south bank
Profile following the Line of the Wells.
past Tilphusium and Haliartus, rounds the cliffs of Mount Sphingium, a little distance from the shore, and unites probably with the middle canal in the bay of Carditza, The two thus united finally join the north canal in the Bay of Kephalari, and the waters of all three are then conducted in a single channel across the bay and into the several katabothra at its foot. Although most of the first part of this canal is gone, parallel embankments, more than fifty metres apart, inclosing an excavation, are to be seen in places, occasionally strengthened with stone work. At the junction of the several canals especially, the work has been executed with great care, the dykes being sustained by Cyclopean walls as solid and probably as ancient as those at Mycæne and Tiryns. Although the main waters of the south canal were poured into the middle and finally into the north canal, all along its course part of their volume was deflected 2i6 THE POPULAR SCIENCE MONTHLY.
througli branch canals into the several katabothra along the east shore. Indeed, the details of the system are much more com- plicated than are indicated by this brief description, and com- prise, in addition to the main canals, many smaller subsidiary ones both for feeding and for draining them. When we take into consideration the difficulties attending excavation in so marshy a soil, and of transporting across it the heavy stones for the em- bankments, and note the immensity of the plan and the thorough- ness and solidity of its execution, we are moved to admiration for the engineers who conceived and built the great works which rendered this part of Boeotia habitable before the dawn of history.
The system involved, too, the clearing and the keeping open of the katabothra, which were liable to become obstructed and some- times to be entirely closed by the caving of the soil and rocks, and there are many evidences of ancient efforts to enlarge and deepen them. That these efforts were not always successful is proved by the traditions of early inundations, referred to before, caused probably by earthquakes, but which were attributed to the efforts of Hercules when he espoused the cause of his native Thebes against Orchomenus. To guard against the recurrence of a similar catastrophe, the ancient engineers planned several cut- tings and tunnels through the hills, which, if they had been car- ried to completion, would have rounded out the original design and accomplished what the Greek Government is to-day trying to effect the thorough reclamation of the basin and its protec- tion from any contingency of flood. On the southeast shore of the lake are vestiges of an immense cutting, thirty metres deep, through the Hill of Carditza toward Lake Hylicus, and beyond that traces of works to ponnect Hylicus with Paralimni, and the latter with the sea. Across the Hill of Carditza, too, are a series of excavated shafts marking the line of a tunnel through the hill constructed with an object similar to that of the cutting to con- vey the waters to the sea through the smaller lakes ; but the shafts are now filled up and there are no indications that the work was ever completed.
This route is the one adopted by the modern engineers, who, by a tunnel through the Hill of Carditza, not far from the line of the ancient tunnel, seek to carry the waters into Lake Hylicus, thence into Paralimni, and finally through another tunnel into the sea. There is also a plan to deflect a portion of the waters for use in irrigating the plain of Thebes.
A still more ambitious undertaking of the ancient engineers was an attempt to penetrate the Hill of Kephalari at the north- east end of the lake by a tunnel more than a mile and a quarter long. This hill, a depression on the flank of Mount Ptoum, has a maximum height of one hundred and forty-seven metres above
�� � the sea and fifty-two metres above the bottom of Lake Copais. Across this depression, from near the openings of the Katabothra of Bynia in the Bay of Kephalari, runs a line of ancient wells or shafts in a general direction from southwest to northeast, not in a straight line, but following the contour of the hill, ending on the east side not far from where the katabothra opens into the Valley of Larymna. There are sixteen of these wells, cut through the hard, gray limestone of which the mountain is composed, and carefully squared, with an average horizontal section of three to four metres. The first shaft, on the west side, is five hundred metres from the lake; the sixteenth, on the east side, two hundred and twenty-five metres from the opening. The wells are at an average distance from each other of about one hundred and sixty metres, and the whole distance from opening to opening is about twenty-four hundred metres.
These shafts are not mentioned by any ancient writer, but have been frequently described by modern travelers, notably by Forchhammer, who has given the most complete description
Longitudinal Section between Wells 15 and 16.
of them. The general conclusion in regard to their object was that they were designed to facilitate the clearing of the katabothra when, from caving or other causes, it had become clogged; but in 1846, M. Sauvage, who examined the shafts critically and cleared several of them, came to the conclusion that they were part of a tunnel scheme, and were sunk with the purpose of giving many points of attack to the workmen engaged in excavating the tunnel instead of a single one at each end. To the ancients, ignorant of the use of explosives, this was of great importance, for the cutting with hammer and chisel was arduous and slow. Even with these numerous shafts, which must themselves have been a difficult undertaking, the excavation of so long a tunnel would have cost the labor of many years. In 1882 several more were cleared and thoroughly examined — the first and the second on the west slope toward the lake and the thirteenth, fourteenth, fifteenth, and sixteenth on the east slope of the hill. The first and the sixteenth wells are each eighteen metres deep, while the sixth, at the summit of the hill, is sixty-six metres deep.
Sauvage concluded that the tunnel had been left unfinished, which later examinations have fully proved. The fact that the first and second wells contain water indicates that it had been completed on the lake side for at least five hundred metres. The exploration in 1882 of the thirteenth well, whose orifice is at an elevation of 107·68 metres, discovered, at a depth of 28·30 metres, a horizontal gallery, 1·60 metre wide and 1·65 metre in the axis.
Outlet of Katabothra of Bynia.
cut in each direction about six metres. At 2·15 metres below this was found a second gallery of the same section, cut in the same direction, and the shaft was excavated 270 metres farther down, probably for use as a drainage well, ending in a level bottom, its total depth being 36·50 metres. The fourteenth, fifteenth, and sixteenth wells, at decreasing altitudes, show a similar interior disposition, save that in the fourteenth the gallery has been but slightly advanced and there is no drainage well. In the fifteenth shaft, which has a total depth of 78·93 metres, the upper gallery is cut to a depth of five metres on the west and two metres on the east, and the lower one 10·30 metres on the west and 10·70 on the east, while the well is 4·70 metres deep. In the sixteenth well the PREHISTORIC ENGINEERING AT LAKE COFAIS. 219
per gallery is cut only two metres on the west and none on the east, the lower one seven metres on the west and 7'30 on the east, with a well of 4*20 metres in depth. In the last shaft the two galleries are three metres apart, and in the fifteenth only 2"15 me- tres, thus showing a tendency to diverge. This would seem to prove that there was no intention of ultimately uniting the two galleries by cutting away the rock between. It is more probable that the upper galleries were begun first, and that some consid- eration induced the engineers to change the grade and give a greater fall to the tunnel. The mean section of the lower gallery is about two metres, the fall from Copais to its mouth is O'Oll metre to the lineal metre, and its total length completed would have been about twenty-four hundred metres. The cutting of so long a tunnel through so hard a rock with the primitive means at the disposal of the ancients shows not only an audacity of plan and a persistent obstinacy in execution, but also a skill in the art of the engineer and the miner that would be no discredit even to the present age.
Who were the authors of these great works concerning which history is silent and which are themselves their only witnesses ? Perhaps this question will never be satisfactorily answered. Leake and others attribute all, the wells of Kephalari as well as the canals, to the Minyans, while some believe that Crates of Chalcis was responsible for the parts exhibiting the most engi- neering skill, and others ascribe them to some of the earlier Ro- man emperors. Curtius, in his Die Deichbauten der Minyer, a paper read before the Berlin Academy in 1893, carefully distin- guishes two distinct works and methods of work: (1) the utiliza- tion of the natural exutories toward which the waters were led by means of dykes and canals, and (2) the formation of an artificial emissary to draw ofl: either all the water or the excess of water from the lake. The first, grand and simple in design, he attrib- utes to the primitive or Homeric age ; the last, marked by careful calculation and executed with the skill of the practiced engineer, he ascribes to the age of Alexander, and presumably to Crates, the only name mentioned in connection with it. Unless the future shall bring to light some inscribed stone or other monu- ment which shall give us definite information concerning the promoter who planned or the engineer who executed these vast works, we shall have to accept the judgment of Curtius and give the credit of them to Crates, the miner of Chalcis.*
- I am indel^ted for much of the material in this article to two articles, by Michel L.
Kambanis, in the Bulletin de Correspondance Hellenique, published in 1892 and 1893, enti- tled Le Dessechement du Lac Copais par les Anciens ; and to an article by Dr. Alfred Philippson, in the Zeitschrift der Gesellschaft fiir Erdkunde zu Berlin, 1894, entitled Der Kopais-See in Griecheuland und seine Umgebung.