One of the most enigmatic features in the Maltese landscape remains the existence of cart-ruts. These paired, parallel tracks can be found in about one hundred and fifty sites across the islands. Particularly striking examples are to be seen near the Dingli Cliffs. It is theorised that these tracks were worn into the soft limestone by the runners of sleds used by the Bronze Age inhabitants of Malta and Gozo as a means of transport. Some of these ‘ruts’ run straight over the sea cliffs and along the bottom of the sea-bed!

Gozo has four good sets of the mysterious cart-ruts, and a few minor ones. At Tan-Nemes, well east of Qala, a pair runs out onto a spur overlooking the Fliegu ta’ Għawdex. At Ta’ Tingi, several fine pairs can be found in the olive grove west of the pumping station, just south of Xewkija. An extensive system covers the Ta’ Ċenċ plateau, best seen between the hotel and the cliffs, and running near, but not to, Borġ l-Imramma. They continue for nearly a kilometre to the east. An impressive pair zigzags down the slope from the San Lawrenz quarries to end on the cliffs above the Azure Widow at Dwejra



These parallel channels, sometimes in single pairs but often found in groups are cut in rock and occur in different patterns. There are instances when a single pair branches into two or more separate tracks. Cases when a group of cart-ruts open up in a fan like manner are also recorded. In the latter case the multiplication of cart-ruts was to accommodate several vehicles at once, as we shall see later on in this study. At other instances the presence of two or more parallel cuts is due to the introduction of a new pair to substitute an earlier one which had deteriorated. In such cases the pairs often overlap each other and do not run close to each other without touching. Shunting occurred when it was necessary for traffic to pass in different directions and cross over a common spot.

The distance between the channels varies, but the average width is 1.4 metres centre to centre. The channels are U or V shaped in section and have an average depth of about 8 to 15cms but there are rare instances when a depth of 60 cm is recorded.

Different hypotheses were proposed regarding the type of vehicle used to travel across the cart-ruts. These include the cart with wheels, the sledge with fixed runners and the slide car with two sliding poles having stone runners, fixed at the bottom. The vehicle is thought to have been pulled by a draft animal or man, and even both, where the action of pulling and pushing occurred at the same time.

The theory that the tracks were deliberately carved and subsequently deepened by the continual passage of vehicles is more plausible than the idea that the formation of the channels was initiated by the wear provided by the vehicles. It is argued that since at some sites there is a number of cart-ruts which occur close to each other, it does therefore not make any sense to dig so many channels close to each other. This argument, is incorrect as we shall prove later on in this paper. On other sites there is only the presence of one pair of cart-ruts which runs for a hundred metres and more and it should be impossible for the traffic to coincidentally follow always the same track. If one compares ancient cart-ruts with tracks on rocks made by modern carts, one can easily distinguish between naturally and deliberately made channels.

Cart-ruts are also found in Sicily, Sardinia, Italy, Greece, Southern France and Cyrenaica but are most abundantly found in the Maltese islands.

The first recorded reference to cart-ruts was made by Gian Francesco Abela in 1647 who suggested that they were used to transport stones from quarries to the sea for exportation to Africa during the Arab rule in Malta . Temi Zammit suggested that the cart-ruts were used during the Temple Period to carry soil from the valleys up to the hill tops in order to make fields along the slopes and on the plateaux . Parker and Rubinstein stated that the ruts were man made channels for transporting stones and soil to build terraces. Still owing to lack of evidence they admit that the mystery of the cart-ruts remains speculative. They dated the ruts to the Punic Period . Gracie thought that the cart-ruts were a means of communication between settlements . Barreca had the same idea about the cart-ruts at Monte Sirai in Sardinia and explained them as being a road system between the Phoenician - Punic settlement and the Nuragic hamlet situated nearby . Other theories, that cart-ruts were used as irrigation channels for the distribution of water from springs up to the agricultural terraces , that the carts at "Misrah Ghar il-Kbir" were used by decorated ceremonial carts which were pulled slowly across the terrain , or that ruts served as trails for the carrying of megalithic blocks in order to build the temples, have been suggested. In all the above instances no direct evidence has yet been associated with any of these theories.

We absolutely agree with Gracie who states that a great proportion of cart-ruts disappear under cultivation. Others disappeared naturally through weathering . A large number disappeared due to the over increasing urban developments. We have seen cart-ruts heading straight under the buildings foundations. Other fell victims to the quarrying industry, a process which we believe had been going on for a very long period of time.

The dating of cart-ruts is mainly based on the argument that some pairs are cut by Punic tombs. For this reason it is suggested that by this time the cart-ruts had already served their purpose. Owing to this fact, which is valid for only a very small percentage of the total number of cart-ruts found in the Maltese Islands, their dating was placed by David Trump in the Later Bronze Age . He strengthens his arguments by saying that in few sites, ruts clearly run up to the entrances of Bronze Age villages. Although one cannot exclude the possibility that in certain cases cart-ruts were communication tracks between settlements, still one cannot generalise and apply in a few examples to the whole system on the Maltese Islands. On the other hand, one can argue that tombs were only cut where the cart-ruts in these particular sites were abandoned and this does not necessary mean that the other remaining cart-ruts did not remain in use or even so other cart-ruts were being developed. More over a tomb could have been easily filled up and the short gap did not necessarily hinder the vehicle from maintaining its movement along the same track. In these particular instances the tombs and the cart-ruts could be associated rather than disassociated .

Bonanno contested the claim that the cart-ruts are earlier than Punic basing his argument that 'Temi Zammit, its originator, by the term Punic meant anything from the 7th century B.C. to the 3rd century A.D, as far as the tombs are concerned' .

Trump states that some cart-ruts pass near quarries but to our surprise he further states that there exists no associations between cart-ruts and quarries in Malta . On the other hand, Bonanno emphasises the fact that 'cart-ruts very frequently, almost invariably are associated with quarries' . We absolutely agree with Bonanno's suggestion and we have found absolute evidence to prove it.



One of the most complicated networks of cart-ruts is found at "Misrah Ghar, il-Kbir"' limits of Rabat. The site was nicknamed as "Clapham Junction" by David Trump after the complex railway tracks of a London station. The cart-ruts seem to be present everywhere and run in all different directions, covering an area of about 8 hectares or more. Still when one takes the whole network into account and, transfers the information into a full plan, as we did (Fig. 1), one can fully understand the real function of what has up to now been termed as the most enigmatic features found in the Maltese landscape.

The cart-ruts at "Misrah Ghar il-Kbir" are not only possibly related to quarries, and do not only pass close by them but they are fully associated with them! As we shall see in the following survey, they run straight into the quarries, connecting the latter with one another, throwing light on how a vast quarrying industry functioned during the ancient periods.

We surveyed more than of couple hundred different sites bearing cart-ruts on the rocky coralline and sometimes globigerina terrain in Malta and Gozo. We have a record of 31 different sites where cart-ruts are clearly associated with quarries.

Still we have never came across a more clear evidence as is the case at "Misrah Ghar il-Kbir" where a number of cart-ruts run straight into two quarries ( A and B on Fig: 1, Fig. 2 and 3), which lie in close proximity to one another, joining them via their parallel channels. This example provides absolute evidence that cart-ruts and quarries are fully related with one another and as we will prove below leaves no question unanswered regarding their behaviour, function, and possibly, their dating


A number of cart-ruts run parallel to each other towards the South-West direction. They cover a distance of about 200 metres. Two pairs along the main route change direction and branch towards the North-West leading straight into the centre of a quarry (marked A on Fig. 1, Fig. 2 and 3). The quarry has rectangular cuts, which indicate clearly the exact locations from where the large ashlar blocks were extracted. Along all its perimeter, the quarry is about 0.9 to 1.0 metres deep and forms a quasi-rectangular basin within the upper coralline limestone. It has only one opening, to allow the passageway for the cart-ruts which stop abruptly almost at its inner perimeter. The blocks of stone were loaded on some sort of vehicle supported with wheels or sliding poles within the channels and pushed or pulled out from the quarry into the main route and transported to the road for the construction of buildings in ancient times.

The channels of the cart-ruts within the quarry are approximately 0.45 metres deep. One may argue that the ruts were cut by and are therefore older than the quarry. This is definitely not the case. The surrounding terrain all over the area is approximately of the same level except for the cavity of the quarry which is circa 1.0m deep. It is therefore impossible, considering the depth of the cart-ruts (circa 0.45 metres), to have the latter with an original overall depth of 1.35 metres. It is beyond imagination for a vehicle to be constructed with wheels having a diameter of 2.7 metres. From the above evidence one concludes that the cart-ruts are contemporary to the quarry.

The main route continues further along the South-West direction for another 75 metres and finishes straight into a larger quarry (marked B on Fig. 1, Fig. 2). The quarry, similar to the previous one, is made of rectangular cuts with sometimes an ashlar block still attached to its bed (Fig. 4).

Further towards the West, the network becomes denser. A large group of parallel cart-ruts run, this time along the North-West, South-East axis. Surface quarrying consisting of the cutting of irregular and regular blocks took place over a vast area but is most evident in areas C, D and E (Fig. 1). The ruts, run parallel for a distance of about 200 meters. On the western side of this parallel network, the ruts are hidden by a stretch of longitudinal fields. Still the main stream continues to run for a further 100 metres towards the same direction, but converges inwards until they meet a number of intersecting ruts which emerge in the form of segments. It is important for one to note that the long formation of cart-ruts is intersected three times, at the Northernmost end, at the centre and at the Southernmost end of the network (marked X,Y,Z respectively on Fig.1). Although the association between cart-ruts and quarries is not as evident in this area as in the case of quarries A and B, we arrived at these possible conclusions regarding this particular area.

There is evidence of extensive quarrying over a vast area which is evident via the deteriorated rock surface made up of pointed spurs of rock running in straight lines, and the quasi -rectangular pans in the coralline surface. These are probably the results of the negative cuts left from extracted blocks of stone or even the removal of parts of cart-ruts which were no longer in use and which made ready available semi-regular blocks which could easily be extracted. The resultant rock surface, due to the continuous erosion caused by rain water and other natural elements took the present shape. The overall pattern of the rock surface in some area can be very easily misinterpreted as natural and not the result of human intervention. Still if one scrutinises the rock surface very carefully, one can spot short stretches of channels, being remnants of cart-ruts in these particular areas.

The interesting cart-ruts situated at X, Y and Z constituted three stations from where the blocks which were carried along the main stream in a number of vehicles were diverted to the road or main path. One pair of the converging cart-ruts turns towards the South-West direction and runs for about 400 metres, meeting another pair and together they lead to another large quarry (H on Fig.1).

Two other small quarries having rectangular cuts (G and F on Fig.1, Fig. 5), are situated in an earlier and larger quarry. Today this large quarry is partly hidden, underlying rubble walls and the surrounding fields but the path lying North of quarry G exposes parts of it. Quarry works were still probably conducted in the larger quarry alongside the small quarries. A rut having only one channel, the other channel probably hidden under the adjacent rubble wall runs all along the quarry. It is connected to the main network via the long cart-rut leading to quarry H, but this time it has both its channels exposed.


The cart-ruts at "Misrah Ghar il-Kbir" constitute an area of intensive quarrying works where irregular and regular blocks were cut from the surface and carried to the road along different directions in a number of vehicles to be transported away for the erection of building and any other structures.

Quarries A and B and the interlinking cart-ruts are an absolute evidence that both features are fully associated with one another. Due to the size of the detached regular blocks and the negative cuts still evident in quarries, which vary in the range of 1m to 1.5 metres in length by circa 0.5m in width and circa 0.5m in height, we attempt to date the whole network as Phoenician - Roman. During the Phoenician, Punic and Roman periods, the use of ashlar blocks of the quoted size was very much common. We do not exclude the possibility that the area was used as a quarrying site also before the Phoenician period since there are signs of surface quarrying of irregular blocks. Still one should consider the fact that irregular blocks of stone were also used during the Phoenician and Roman periods. We also consider the possibility from other evidence we have recorded and which we intend to publish in the future that the site may have continued to be used for quarrying in the post-Roman times. In the dating of cart-ruts and quarries it is important for one not to be rigid and constrain oneself in dating the system to one period only. Cart-ruts and quarries could have been a system which saw its birth in the Neolithic Period and continued to be used until the Classical and possibly later periods


The cart-ruts at "Misrah Ghar il-Kbir" have been a great attraction to both tourists and locals alike. Yet the visitor, on arriving on site is welcomed by a massive intervention of human activities, which do not help him to understand the surrounding landscape but leave him in a more perplexing situation as to what these rock-cut channels really are.

The encroaching agricultural activities within the site have been going on for ages. The fields enclosed within the rubble walls have expanded all around the site with the consequence of covering a substantial part of the area. A number of cart-ruts lead straight under the rubble walls and fields. Moreover the cavities formed by the quarries within the area have provided adequate ready made basins for the planting of crops and other products and were consequently used to make fields.

This procedure has made the relation that exists between cart-ruts and quarries very difficult to perceive. It is the eastern part of the network, tightly enclosed between fields, that have saved the situation and provided the necessary evidence.

The intervention of agriculture, with all its conglomeration of rubble walls, terraces and fields has been accepted as part of the Maltese landscape. Apart from being a necessity for the subsistence of the Maltese economy, is a process which has been going on for ages, and part of the terraces may date back to the Classical and to the Phoenician periods, maybe even earlier. Moreover the system of terracing, rubble walls and fields help in preserving what lies under them as evident from the numerous sites which have come to light since the 19th Century to the present time. Still one must also consider the fact that the reclamation of land, and the transporting of soil from one side to another hinders and distorts the stratigraphical layers which are extremely important in archaeology.

Ironically enough, quarrying in Malta has also been a major industry since ancient times, as is evident from the site of "Misrah Ghar il-Kbir". But here one has to define the impact that is left by modern quarrying which is absolutely different than that left by our ancient ancestors in the making of their surface quarries. As their name implies these quarries are only one metre to some metres deep and do not consist of large craters which mercilessly amputate large portions of the landscape, insensitive to all that lies in the ground and to the impact they leave to the surrounding landscape and to the whole environment.

Towards the South-West, large quarries, more than 25 metres deep are extending towards the archaeological site destroying all the fields and the valuable information that lie stored in them. These large quarries are not only destroying the archaeological data but are ruining the surrounding environment. The rocky terrain overflowing with flora, the rubble walls and the fields, not to mention the rural dwellings within the area are being erased from the landscape and replaced by huge cavities, machinery and large mounds of stock piles. The value of this archaeological site is diminishing continuously with each cubic metre of rock cut by modern quarrying.

The threat of modern buildings within and near the site is also helping in destroying the archaeological data and the environment as evident from Fig. 6. A pair of cart-ruts runs straight under the modern road and building. The foundations of the buildings required the levelling of the bedrock with the consequence that any rock cut features were obliterated.

Vandalism along with other malicious human interventions contribute to lessening the importance of an archaeological site. The area around "Misrah Ghar il-Kbir" has been, for a long time an attraction to hunters and bird-trappers. The construction of bird-hides is a common feature around the area. This involves the transportation of soil and earth which are deposited on the rocky terrain over large areas, consequently covering the rock-cut features and distorting the archaeological value of the site. To make matters worse the display board was stolen and the large boulder on which it was fixed was broken and turned over.

All in all, the archaeological site at "Misrah Ghar il-Kbir" is in a continuous threat of being at a slow but constant rate defaced and consequently erased from our National Heritage.

The site in question is legally protected by the Antiquities (Protection) Act, 1925. However this Act does not specify the limits of protection. Despite the protection provided by this Act, quarrying activities, reclamation and other incompatible development and land-use persisted till recent times. In April 1998 the site was fully scheduled by the Planning Authority as part of an Area of Archaeological Importance, including a protective buffer zone establishing specific limits. This scheduling brought about a number of legal appeals by the quarry owners who contested the merits of the site and the extend of the buffer zone. After evaluation, it was decided that both merits and the extent of protection were justified.

Due to its importance, the site has remained a main tourist attraction. This article gives a full comprehensive view of the site with the aim of making both tourists and locals fully aware of its great archaeological value and the important evidence it holds with the understanding of the real function and behaviour of cart-ruts.