Summary

The purpose of this article is to describe the caves of Kephallinia and in particular the unusual cross island drainage system known as the 'sea-mills of Argostoli'.

I have put forward a theory for the mechanism causing the flow of sea water under the island from Argostoli to Sami and attempted to relate this to the water samples and other data collected when we were there.

Introduction

A group of N.P.C. members visited Kephallinia in the summer of 1980 for three weeks in the hope of finding new and fascinating caves.

Kephallinia is the largest of the Ionian Islands off the coast of Greece, about 200 miles north of Corfu and is the legendary Kingdom of Odysseuss.

The island is famous because of its sea mills where the sea continuously drains into a cave on the coast. The N.P.C. became interested in the island after reading a brief report by the R.R.C.P.C. Their visit was confined to a narrow area around the port at Sami. However, they thought the prospects for high level development looked extremely promising. This island consists entirely of cavernous limestone and the central peak rises to 1600m. The place seemed too good to be missed with a view to some serious caving and a good holiday.

List of Members

I.Crossley, K.Millington, M.Millington, J.Miriam, G.Smith, C.Purnell, R.Roebuck, R.Lonsdale, J.Lonsdale, S.Rhodes, A.Rhodes, D.Vosper, S.Clark.

We should like to thank the Sports Council and Cecil Cullingford for their kind assistance and support: also to Dr.R.Halliwell, University of Hull for the analysis of water samples.

The Sea Mills of Argostoli

This geological oddity known as the 'Sea Mills' of Argostoli has yet to be explained. The sea sinks into several fissures on a peninsula to the North of Argostoli and reappears two weeks later on the other side of the island in a series of brackish risings on the coastline at Sami. The phenomenon is further compounded when the height of the sink and the rising are compared. The rising is 1m. above sea level.

An early description and explanation was made by a Frenchman in 1900; he described it as one of the great unsolved mysteries of the world and went on to say ......

'.....millions upon millions of tons of water rush daily down into a bottomless pit and has been so running for millions of years; obviously it cannot merely connect one sea with another for, in the process of time, however vast the subterranean caves and hollows into which it pours they would ultimately fill up and the seas would stand at the same level ... as there is no scientific solution to the mystery it is permitted to anyone to rise, and mine is that the mighty deluge of water reaches internal regions where the temperature converts it to steam, which, rising up into the cavities, is condensed and forms rivers of fresh water, perhaps over the whole globe.

A less fanciful description was made in a report to the 'Times' by their 'Athens correspondent'.

The phenomenon, known as 'the sea mill of Argostoli', occurs on the coast north of Argostoli, the chief town on the island. The sea flows with great force into a shallow recess about 50 yards inland; the water feeds four small pools, and then disappears through two mysterious chasms at the edge of these pools. More than 30,000 tons of sea water vanish in this way every 24 hours. It is a geological oddity probably unique in the world.

Since its discovery many unsuccessful attempts have been made to determine where the water subsequently re-emerged. On one occasion in the last century British naval officers poured oil into the recess. It disappeared through the cracks, but no trace of the oil was ever found.

Unfortunately in 1953 the island was destroyed as a community when all the buildings were razed to the ground by an earthquake centred at Sami. The majority of the population moved away so the island is dotted with damaged and fallen buildings, overgrown paths and impenetrable, aggressive undergrowth. The earthquake reduced the volume of water sinking at the Argostoli mills to such an extent that they became uneconomic to operate.

Description of Known Caves

Argostoli 'Sea-Mills'

The 'sea-mills' are at two points on a headland at Argostoli; both have comparable flow rates and are about 200m. apart. The rate is constant at 0.16 cumecs and is unaffected by minor changes of sea level which occur in the Mediterranean. Both sinks have been used as mills and have concrete channels built around the mill race so it was easy to obtain accurate flow rate measurements.

The sea around Argostoli is shallow; around 5 to 10m. so it is possible that there are other sinks in the sea bed; however our discussions with the local museum staff indicate that there are none.

The water from the sinks reappears at Sami, and it has been proved that the water takes about 14 days to cover the 17km. underground.

Sami-Resurgences

At Sami, there are several resurgences over the coastline of about 1km. The three major salt water risings are above sea level and at the largest the water drives another tourist 'sea-mill', hence the curiosity, the water sinks at sea level and reappears approximately 1-2m. above sea level. The total volume of the coastal risings is about 1.5 cumecs. but there are numerous small risings below sea level. These can be seen from the coast on a calm day as small pressure domes. The volume of the sea risings is probably negligible compared to the major land resurgences.

The risings are all different in character, the most spectacular is 'Karavomylos' (1) where the water appears from a large flooded cave, flows into a man-made mill pond and flows out through a mill wheel into the sea. Approximately 300m. north of this point is another similar resurgence at the base of a cliff. This flows out over a 30m. front onto the beach.

To the south of Karavomylos there is a small rising in a field about 50m. from the shore. These are all salt water risings, however, about 100m. further south towards Sami the campsite has its own freshwater well. This is one of the few continuous sources of freshwater on the island but for some reason the two systems are separate.

The other caves are all in the lowlands behind Sami and fall into two classes:
i) salt water systems,
ii) fresh water systems.

Salt Water System

i) Melisani (2)

This is a large phreatic passage which has partially collapsed to form a chamber and a connection to the surface. The lake connects with the rising at Karavomylos and is inhabited by a distinctive type of eel. These can also be seen at the rising. The cave is a tourist attraction and boat trips can be made around the lake.

ii) Zervati (3)

This is another collapsed passage but not as impressive. The water is again full of fish and must also connect with the rising at Karavomylos. The French have dived the sump and found impressive stalactite formations in the flooded passages.

iii) Theodorous (4)

An impressive shaft with a 28m. pitch onto a rubble slope. At the bottom of the slope is a sump pool which contains the fish seen at Melisani. The hole is supposed to contain an unexploded German bomb and the bodies of several partisans.

iv) Elousa (5)

A large open depression with a small rubble slope at one corner, this ends in a sump pool.

v) Erobrousci (6)

This consists of two extremely unstable parallel shafts, both of which end in a lake. However, only one was surveyed to conclusion because of our fear of earthquakes; which occur at least once a week.

Fresh Water Systems

i) Angalaki (7)

This is the largest of the caves on the island. The entrance pitch is 30m. onto a large rubble heap. The cave has a small amount of horizontal development and two large sump pools. The entrance is caused by a roof collapse in a large phreatic passage and any continuation of the passage would have to be done by divers. The cave is well populated by pigeons and bats.

ii) Drongarati (8)

This is a show cave consisting of a large isolated, but well decorated chamber, there is no obvious exit and no connection with the salt water phreas.

Overall, the arrangement of the caves and the type of water lead to two general conclusions. The salt water from Argostoli is the same water which appears in a diluted form in the caves at Sami. Also, there are two separate water systems converging beneath the Sami valley. The salt water mixture from Argostoli and a fresh water phreas, probably from caves on Mount Eros. This generalised picture is considered in greater depth in the discussion of our field tests and water samples taken from various caves.

Geology of the Island

The island was completely surveyed by a B.P. geological team coordinated by Leeds University. This showed the island to consist almost completely of limestone. The limestone has an average dip of 30 degrees from Argostoli to Sami. There is a complex fault system around Argostoli and two large faults running along the East coast.

Sami is on the junction of two major faults and on the edge of the central limestone area.

Unfortunately, conditions in the field revealed that the map did not show the whole story; nearly all the rocks are extremely folded and smashed by earthquake action. Also the B.P. interpretation of limestone was rather broad. The quality of rock varied from a fine cement-like powder, through a loose conglomorate to the massive cavernous limestone we are used to in Yorkshire.

The recent geological history of the island has been dramatic. The whole island has been rotated from the horizontal by numerous earthquakes. Also the sea level has risen by about 100m. since the Wurm glaciation. This means that the established cave systems on the island were submerged about 20 thousand years ago and there has been insufficient time for a new network to develop. This has been confirmed by carbon dating of stalactites from the submerged passages of Zervati and Melisani.

Theoretical Models of the System

Several people have applied themselves to the problem of the seamills. Any explanation has to explain:
a) why there is a cross island connection of cave passages?
b) how the system developed?
c) what caused the flow of water to commence?
d) why does it flow?
e) what is the driving force?

Many of the possible explanations fail on one or more of the above tests. I have another possible explanation but like all the others it can only be disproven.

The development of the conditions for water to flow must be considered in stages this is illustrated in fig. 1.

Stage 1

The limestone on the island was horizontally bedded and a well developed cave system was in existence. fig. 1 (a).

Stage 2

As a result of earthquake activity the island was rotated and the normal water flow pattern in the caves was interrupted. This process continued until the majority of water flowed with the dip of the limestone to appear in coastal risings near Sami. fig. 1 (b).

Stage 3

The sea level rose throughout the Mediterranean after the Wurm glaciation. This submerged the majority of the caves and disrupted the now ancient cave passages. fig. 1 (c).

Stage 4

Following an earthquake the caves at Argostoli dropped below sea level and the salt water entered the old cave system at A. This would cause a difference in hydraulic pressure in the flooded cave system due to the different densities of water. This would cause a flow from A to S and the sea mills would have come into existence.

The Theory in Practice

There are few sources of fresh water on the island, however the theory depends upon a source of fresh water which acts as a reservoir to keep the system flowing. Fortunately, we did manage to find a suitable high level reservoir.

The '40 Wells'

Whilst we were looking for caves we found yet another curiosity. On the Argostoli side of the island, near Frangati there is a field approximately 20m. square which contains 40 wells. These were dug by the Italians about 300 years ago and despite the severe water shortage on the island have never been known to dry up. We talked to several locals around the area and confirmed that the wells are a constant source of water, filling to the ground level in winter and gradually falling throughout the rest of the year. The whole area is a large fertile plain completely enclosed by mountains about 500m. above sea level and appears to be a natural reservoir. The winter rains fill the reservoir to the surface level, any excess lying on the surface for several days in large pools.

Since all the water lies below the surface there is little evaporation loss and the amount of water used by the locals is negligible, so it must be assumed that the water gradually seeps away along the dip of limestone towards Sami. It could be possible that the reservoir is the driving force for the phenomena of the sea-mills.

Summary of Work Done

Whilst we were on the island we surveyed all the caves and took water samples where possible.

The samples were then taken in old film containers, two from each site. The samples were analysed for salinity when we arrived back in the United Kingdom.

* proven connection.

The decrease in salinity and the increase in volume are of the correct order of magnitude to indicate that there are no other components to the system.

A Brief Calculation of the Reservoir's Capacity

The catchment of the plateau is 50.3km² which would require a rainfall of 0.12m. per year. The average rainfall per year is 1m. Therefore, the rainfall is well in excess of the theoretical requirement. However, the analysis has assumed that the flow rate is constant and that it rains only once a year, but the figures are of comparable magnitude to indicate that this is the source of water at Sami.

R.I. Crossley.