Navigation – Plan du site

Wetlands in the river delta plains: evolution, values and functions during the Roman times. The coastal landscape close to the Garigliano river mouth

Les zones humides dans les deltas des rivières de plaine : évolution, valeurs et fonctions à l’époque romaine. La zone côtière autour de l’estuaire de la rivière Garigliano
Kevin Ferrari, Piero Bellotti, Pier Luigi Dall'Aglio, Lina Davoli, Mazzanti Marta et Paola Torri
p. 273-283


La plaine côtière du fleuve Garigliano, à la frontière entre le Latium et la Campanie en Italie, étaient caractérisée jusqu’à une période récente par une mosaïque de zones humides et de terrains bien drainés, ces derniers étant localisés dans les secteurs d’anciennes plages. Ces conditions géomorphologiques et hydrologiques ont sensiblement conditionné le peuplement et la vie économique. La spécificité de la plaine du Garigliano, et particulièrement l’embouchure du fleuve du même nom, est d’avoir été décrite de façon détaillée par les auteurs grecs et latins. Nous avons donc pu confronter les informations tirées de l’interprétation de ces textes avec les données géologiques, archéologiques et paléobotaniques disponibles. Nous avons ainsi montré que les marais de l’embouchure du Garigliano, bien qu’apparaissant de prime abord comme des lieux marginaux, avaient toutefois un rôle spécifique dans la vie romaine, tant d’un point de vue des pratiques religieuses que de la vie économique.

Haut de page

Texte intégral

Wetlands in the Roman times: coasts and river delta plains

1The modern landscape in several regions (like Emilia Romagna or Veneto in Italy) is characterized still now by a regularity that originates from the Roman centuriation, the most famous system of assignation (limitatio) that involves the achievement of infrastructures (roads and channels), the geometric division of the land into squares with sides of 20 actus and the reclamation of wet zones.

2Even though the centuriation and its geometric regularity involved the creation of a cultivated and agricultural landscape, it didn’t eliminate the existence of wetlands, marshes, woods and grazing lands that continued to exist. In the Greek and Latin imagination wetlands were marginal and dangerous zones in contrast with the civilized world, where mysterious events could happen and the reclamation of lands was linked to heroic acts (Locchi, 2010). Nevertheless we have to reject the idea of a clear contraposition between wetlands and anthropized lands as it exist in the modern ideology and, at the same time, the idea of a Roman world based on the reclamation of new lands in contraposition to the pre-Roman and Medieval world where the nature prevailed (Traina, 1988 p. 15-18). Into the centuriated landscape there were in fact wet zones and marshes as we can gather from the theoretical texts of the gromatici and form the analysis of epigraphic and literary sources.

3The Corpus Agrimensorum Romanorum contains some indications about the existence of lands with a different status compared with the usual land assigned to the settlers. This is the case of the subsecivum that was literally «the land that was cut out by a line» (Front. De Agr. Qual. 2, 16-17 Th.: Subsiciuum est, quod a subsecante linea nomen accepit [subseciuum]). There were, according to Frontinus, two different types of subsecivae: firstly this term indicates the lands out of the centuriated area, secondly it concerns a portion of territory inside the centuriated area but that was not apt to be assigned. The subsecivum could correspond to wetlands, woods or areas influenced by river floods. A picture in the Corpus Agrimensorum (Fig. 38 Th. - fig.1) shows the representation of a subsecivum like an independent area inside the greed of the centuriated territory and it characterizes this zone with typical elements of wetlands (plants or aquatic birds). This image concerns a passage of the text that talks about some areas interested by floods (zones where we can find wetlands). Hyginus Gromaticus reports that woods and grazing lands could occur also inside the cultivated and assigned territory (Hyg. De Lim. Const. 159, 18-21; 160, 1-4 Th.), as another picture reports (Fig. 122 Th. – fig. 2). A similar situation is proved also by literary and epigraphic sources. For instance Tacitus recounts that some veterans received infertile lands because of the presence of wetlands (Tac. Ann. I, 17, 3) and Varro talks about the possibility to inherit plots near swamps (Varro. De Re Rust. 1, 12, 2). An epigraph concerning the territory of Parma reports that a wetland near the city was reclaimed by Caius Preconius Ventilius Magnus and transformed into a garden for the augustales (Susini, 1960 p. 153-155). As we can see several wetlands weren’t reclaimed and so the Roman territory was characterized by the coexistence of a geometric and cultivated landscape and a natural and not organized environment both around both inside it.

4The coast and the river delta plains were particularly rich in wetlands, lagoons, marshes and lakes. Cicero reports the occurrence of a lake near the Ombrone delta (Cic., Pro Mil., 27, 74 ) and Pliny the Younger talks about the presence of noxious territories in Tuscany probably due to swamps along the coastline (Plin. Ep. 5, 6, 2). Strabo (V, 2, 8) says that Cosa was built near a lagoon. Southward there is the Latin coast. This area is fertile except for some marshes near Ardea, Antium and Lanuvium, the Pontine Marshes, the area of Setia and the surroundings of Terracina (Strabo. V, 3, 6). In front of this city there is a big swamp (Strabo. V, 3, 6). Also a picture in the Corpus Agrimensorum shows the proximity of wetlands to the city and its centuriated territory (fig. 3 – Fig. 92 Th. – the picture shows a lake that we have to consider like a representation of the sea, but over the centuriated area there is a zone characterized by the word paludes). The Circaeum is like an island between sea and marshes (Strabo. V, 3, 6) and all around there are woods and grazing lands (Strabo. V, 3, 5). Plutarch, Appian, Cicero and other Latin historians make mention of marshes near the Garigliano delta. Also along the Campanian coast there were other wetlands, like the Acherusia marsh (Strabo V, 4, 5).

Figure – Fig. 38 Th. (fig. 39 La)

Figure  – Fig. 38 Th. (fig. 39 La)

Figure - Fig. 122 Th.

Figure  - Fig. 122 Th.

Figure – Fig. 92 Th. (Marshes near Terracina – Paludes)

Figure  – Fig. 92 Th. (Marshes near Terracina – Paludes)

5The occurrence of lagoons, lakes and swamps is a typical phenomenon of the river delta plains. As a matter of fact these zones are subject to a continuous evolution of their physical geography because of the constant interactions between marine and fluvial dynamics: the river mouth shifts, the lagoons evolve into lakes or marshes and the coast tends to move forward. The Tyrrhenian river deltas belong to the category of “cuspate delta” and their characteristics and general evolutions are well known (Bellotti, 2000). Since the sea level stabilized near the current level at about 6000 years BP, the rivers started to fill large bays creating alluvial plains. Their sandy sediments created some barriers isolating lagoons that were filled until the river mouth joined with the barrier. From that moment on the coast moved forward initially with beach ridges parallel to the coast, successively with the creation of a cuspate delta. The lagoons became lakes isolated from the sea because of the continuous development of beach ridges and were gradually transformed into marshes and wetlands by fluvial floods. This process took place over a long time. In the maps of the 17th or 18th century the coast is still characterized by the occurrence of several lakes and wetlands and only in the last century these zones were completely reclaimed. Despite these evolution phases are very similar, each delta plain has distinctive characteristics depending on the volumetric flow rate of the river, the geologic background, the weather trend.

6The Garigliano River Delta Plain is a very interesting case of study because of the abundance of sources. We have detailed literary references to the marshes that could be compared with geological and pollen data and could help us to understand functions and values of these wetlands and their relationship with the settlements, the Roman city and the organization of the territory. There is also a picture in the Corpus Agrimensori Romanorum that represents the territory of Minturnae (fig. 4 – Fig. 89 Th.) This image contains several elements that can’t be easily interpreted (also a big lake near the river mouth). In our opinion the picture illustrates accurately the passage of Hyginus Gromaticus concerning the land division of the ager Minturnensis (Hyg. Grom. De Lim. Const. 142, 2-19 Th.), but it doesn’t contain any element relating to the wetlands. For this reason we won’t take this picture into account in the comparison between environmental and literary data, but we will spend a specific paragraph later on.

Figure 4 – Fig. 89 Th. (territory of Minturnae)

Figure 4 – Fig. 89 Th. (territory of Minturnae)

7This geoarchaeological research could be considered part of a larger study coordinated by the Soprintendenza per I Beni Culturali del Lazio that long since is working on the palaeoenvironmental evolution of the Garigliano delta plain with a view to conservation and promotion of the archaeological heritage (Bellini, 1998; Bellini, 2007; Bellini, Trigona, Matullo, 2011). Our methodology is based on a multidisciplinary approach that combines the typical analysis from archaeology, topography, geomorphology and palynology (Ferrari et al. 2012; Ferrari et al. 2013).

The Garigliano River Delta Plain

Geographic and Historic context

8The Garigliano Delta Plain is located between Latium and Campania within a graben generated by anti-Apennine faults. During Tyrrhenian time, the coastal zone was characterized by beach ridges bordering a wide bay. The Garigliano River locally eroded the ancient beach ridge during the subsequent glacial phase. After the last glacial maximum, the sea level rose approaching the present sea level about 6 ka BP and new beach-dune ridges developed westward of the older ones (fig. 5). Two systems of beach ridges with a NW-SE trend, referring to the Euthyrrenian (inner system) and to Holocene (outer system), border a depressed area, separated by the Garigliano channel, where a wetland developed during the last 6 ka. In this depression the area to the north of the Garigliano is narrower than in the south. Some traces of palaeo-channels indicate an old River Garigliano mouth located in the southern area. The Garigliano River built up a lightly cuspate wave-dominated delta after the postglacial. Historical maps show that these depressions were partially submerged until the 18th century and today they are kept dry by land reclamation infrastructures (channels or water pumps).

Figure 5 – Geomorfological units and settlements near the Garigliano River mouth.

Figure 5 – Geomorfological units and settlements near the Garigliano River mouth.

9The first settlement near the coast was a village on the top of the rocky promontory of Monte d’Argento (2 km north of the River Mouth) that was in use from the Middle Bronze Age to the Iron Age (Guidi, 1991 p. 20-30; Alessandri, 2007 p. 158-160). The Late Bronze Age is characterized by the occurrence of little settlements along the Thyrreninan beach ridge but they were probably abandoned since the end of the Bronze Age (Ferrari et al. 2012). In the Iron Age the region was inhabited by the Aurunci. They were settled in villages on the top of calcareous terraces in the hinterland (known as oppida) from where they could control the plain and the sea (Talamo, 1987). The coast was uninhabited except for the oppidum of Monte d’Argento and a place of worship on the right bank of the Garigliano near the river mouth, between the sea and the wetlands (the sanctuary of Marica – Mingazzini, 1938; D’Urso, 1985; Talamo 1987, p. 67-96; Bellini, 2002; Andreani, 2003; Livi, 2006 p. 105-113). The archaeological survey didn’t find any other site along the coast.

10The Roman colonization of this region started in the last quarter of the 4th century BC as reaction to the support of the Aurunci to the Samnites during the Second Samnite War. The Romans destroyed the most important cities of the Aurunci and they deduced the Latin colony of Sessa Aurunca in the inland (313 BC) and the two Roman colonies of Minturnae and Sinuessa on the coast (296 BC). Above all that they realized a new road system centered on the Appian Way and they assigned plots of land to the settlers. Minturnae was a maritime colony located on the Euthyrrenian beach ridge, were the Appian Way crossed the river, and it was separated from the sea by a wetland and the Holocene beach ridge dunes. Even though originally these territories were not so attractive (Liv. X, 21, 7-10), Minturnae became very soon a prosperous city thanks to its harbor and trades (Guidobaldi, Pesando, 1989; Bellini, 1998; Bellini, 2000; Mesolella, 2000). The countryside was reach in settlements and farms whose products were exported in the Mediterranean basin thanks to the fluvial harbor and the landing place near the sanctuary of Marica.

11The 3rd century AD marked the beginning of the decline of the region: several settlements disappeared and also the city itself shows traces of decadence (Arthur, 1989). Minturnae was abandoned in the course of the 6th century AD and the population moved to the nearby hills where the modern Minturno is situated. The coast was not totally uninhabited. Between the 881 and the 915 AD the ruins of the Roman city and the area close to the river mouth were interested by the occurrence of a Sacacen ribat (Rossillo, 1985). After the Saracens were defeated the only settlements near the coast were the village of Castrum Argenti on the summit of the Monte d’Argento promontory (Torre, 1988; Torre, 1998; Ciarrocchi, Torre, 2003) and a tower south of the Garigliano mouth to control the coast against the pirates (Di Biasio, 1994 p. 73-79). After this period the coast remained unpopulated and became more and more marshy as we can gather from the toponyms Pantano di Sessa and Pantano di Traetto.

The wetlands near Minturnae

General Evolution

12The maps from 16th to 18th century, the toponyms, the analysis of the aerial photographs show the occurrence of wetlands, swamps or lakes between the shoreline and the Euthyrrenian beach ridge both on the north both on the south of the river. Recent palaeo-environmental studies allow us to reconstruct the evolution of the wetlands since their origins until this time (for a preliminary publication of geological data and pollen analysis see Ferrari et al. 2012 and Ferrari et al. 2013).

13The analysis of several cores shows that at about 8-6 ka BP the river delta plain was affected by the presence of a bay/estuary in the southern area while further north a marsh developed (fig. 6). In effect the sedimentary sequence in the northern palaeo lagoon is characterized by a peaty bottom member 5 m thick with remains of woody plants, while the southern palaeo lagoon sedimentary sequence has a two meters thick basal silty member with rare peaty intercalations. After the last glacial maximum, the sea level rose approaching the present sea level about 6 ka BP and the fluvial sediments were sufficient for more continuous development of the beach ridge. The basal member of the southern core is followed upwards (about 2,55 m deep) by a peaty member, 1.5 m thick, deposited between 5.5 and 3.5 ka BP showing that also the southern area was isolated and the brackish environment changed into a fresh-water marsh.

Figure 6 – Sedimentary sequence of the cores

Figure 6 – Sedimentary sequence of the cores

14Both stratigraphy and pollen analysis show a significant environmental change. At 1,65 m deep (corresponding to 3 ka BP) the stratigraphy shows a passage from soft peat to silty-sandy level with rare mollusc fragments highlighting a fresh-water environment and in correspondence with this passage we can record a drastic decrease in hygro-hydrophyte plants. Above this phase there is a remarkable expansion of riparian trees and the freshwater aquatics reappear. Probably river overflows altered the environmental features of the marshes. This situation continued until recent times. The wetlands between the Holocene and the Euthyrrenian beach ridge became more and more marshy, even though in the modern age they dried out at times and were utilized for pasture.

Before the Roman time

15During the Bronze Age the zone behind the Holocene beach ridge was characterized by the occurrence of marshes not influenced by fluvial or marine sediments. These wetlands were reach in vegetation as we can gather from the peaty member and both mollusc fragments both pollens confirm that we are in front of an euthrofic freshwater basin. We don’t have many information about the exploitation of these marshes during the Bronze Age. The most important settlement was a village located on the rocky promontory of Monte d’Argento. It was on the coast but isolated from the nearby environment by steep scarps. The archaeological survey found several sites with probably Late Bronze Age pottery all around the inner border of the ancient marshes but located on the Euthyrrenian beach ridge that was higher and well drained. They are characterized by little fragments of pottery spread over small areas without traces of structures. The position of these settlements led us to think that they were strictly linked to the wetlands. As other sites in Tuscany and Latium probably they were periodically inhabited for activities like harvest wood, fishing, clay extraction (Pacciarelli, 2000 p. 170-176; Angle, Belardelli, 2007 p. 765-766.; Alessandri, 2007 p. 211). At that time the wetlands near the Garigliano river mouth could not be used for saltworks because they were characterized by a fresh-water environment.

16The passage from the Bronze Age to the Iron Age correspond to a transformation of the wetlands. Probably the river started to overflow into the two basins and the peaty sediments were replaced by silts, even though the pollen analysis shows that the environment was still dominated by aquatic plants and trees like Salix and Alnus. The basins were occupied by little lakes with fresh and shallow water but reach in vegetation. This situation correspond to the environment during the Iron Age, but we can hypothesize that in the Roman Republican period it was very similar. The settlements with Late Bronze Age pottery were abandoned and the coast was uninhabited save the village on the summit of Monte d’Argento and a sanctuary active from the 7th century BC near the river mouth. This was an important place of worship for the Aurunci with the function of interface between the inland and the Mediterranean trades, between land and water and between the underworld and the human world. The emporic sanctuary assumed the role of place of trade and cultural exchange (Bellini, 2002). The venerated goddess was Marica, whose name is associated to marshy waters (its root is *mari-/*mori- and concerns a standing body of water inland or coastal - De Simone, 1996 p. 66). She had a chthonian trait and she mediated between the human and the natural world by means of the religion (Cerchiai, 1999).

17The sanctuary was monumentalized since the beginning of the 6th century, even though the most ancient remains dated to the last quarter of the 6th century (Livi, 2006 p. 109-111). The temple was oriented toward the marshes and probably it had or a Hellenistic plan (Laforgia, 1992; Rescigno, 1993). The cult of Marica continued also in the Roman Period.

The Roman Period: nature and function of the wetlands

18We have more information about the nature of the wetlands near Minturnae during the Roman times. In addition to the palaeoenvironmental analysis, we have some detailed descriptions of the zone in the literary sources. The most significant text is the Life of Marius by Plutarch, but also Cicero (Cic. Ad Quir. VIII, 20; Cic. Pro Sestio, XXII, 50; Cic. Post Redditum ad Quirites, VIII, 20; Cic. In Pis. XIX, 43) and Appian (App. B.C., I, 7, 61-62) report interesting information. Also Livy talked about the Marius’ escape and descried the swamps near Minturnae, but the book that concerns this episode is lost. We can infer what Livy said by analyzing other texts that used his book like their source (Val. Max. VIII, 3; Juv. Sat. X, 276-278; Solin. 27, 40; Firm. Mat. I, 7, 37; Ps. Aur. Vict. 67, 4, 5; Oros. Hist. V, 19, 7; Vell. Pat. II, XIX 2, 3; Luc. II, 70-72. For a detailed analysis of the literary traditions see also Bang, 1910; Andreani, 2003). All these authors give a similar description of the wetlands, with a lot of references useful to reconstruct what were their values and functions and what was the ancient landscape.

19Plutarch talks about the wetlands using the words «λίμνη» and «λoς» that usually concern swamps, marshes and coastal lakes. The first term focus especially on the occurrence of water and on the demarcation between land and water. The second term focus on the combination of water and land and concerns both the water both the habitat and the vegetation (Traina, 1988 p. 54-57). The Latin authors refer to the wetlands using the word «palus», that indicates usually a place with stagnant water. Varro defines palus as a place with shallow water but spread over a quite vast area (Varro, De Lingua Latina,5,4,26: Palus paululum aquae in altitudinem et palam latius diffusae ). Only Lucan describe the territory near the Garigliano river mouth with the term «stagna», that indicates the occurrence of ponds or bodies of standing water (Luc. II, 70-72). According to Greek and Latin authors the coastal zone near the river mouth is characterized by a marshy environment with little lakes or ponds rich in vegetation, where land and water could blend. Plutarch talks about mud (βορβόρου) and muddy water (τελματδες). Velleius Paterculus recounts that when Marius was extracted from the marshy waters he was covered by mud (nudus ac limo abruptus - Vell. Pat. II, XIX, 2- 3). Also Lucan (II, 70) refers to the «stagna» with the adjective «limosa» that is equivalent to «muddy» or «silty». Probably the Garigliano overflew periodically into the basins with its silty sediments as we can gather from the sedimentary sequence. The expression «το Λίριος ποταμο διάχυσιν λιμνώδη λαμβάνοντος» used by Plutarch indicates in effect that sometimes the Garigliano (ancient Liris) could reach the wetlands with its waters.

20The wetlands were characterized by a small amount of water as to depth (λη βαθέα). The analysis of the sedimentary sequence shows that in Roman times the northern basin could be deep at most about 1,5 m. Nevertheless there was enough water to allow Caius Marius to plunge into the marshes as reported by Cicero (In Pis. XIX, 43: demersum in Minturnensium paludibus); Velleius Paterculus (II, XIX, 2-3 oculis tantum modo ac naribus aminentibus, extractus arundineto circa paludem Maricae, in qua […] abdiderat); Lucan (II, 70: caput abdidi), Orosius (Hist. V, 19, 7: in Minturnensium paludibus sese abdidit) and Plutarch (Mar., 38, 2 :καθκεν αυτν ες τν λίμνην). The wetlands were also rich in vegetation. Plutarch (Mar., 37, 6: τν τε καλάμων πολλος) and Velleius Paterculus (II, XIX, 2-3: extractus arundineto) allude to the occurrence of grooves of reeds and Lucan says that Marius covered himself with ulva, a typical aquatic plant. In effect also the pollen analysis shows that aquatic plants were abundant in this period (fig. 7). As well as these type of plants, there were also several trees. Strabo (V, 3, 6) says that near the river mouth there is a holy wood (ες λσος ερν). Also Martial (Martial. XIII, 83) refers to the sanctuary of Marica with the expression «silva Maricae» and Lucan (Lucan. II, 424-425) talks about the same zone as a shaded place (umbrosae Liris per regna Maricae). Moreover Plutarch and Appian recount that Marius could hide his body with twigs and leaves, typical elements of a place rich in vegetation. Among the arboreal pollens the dominant taxa are Salix, Quercus deciduous and Alnus glutinosa type. The diffusion of the Salix was so relevant that it is attested also by the ancient authors. For instance Cicero (Leg. Agr. 2.14.36) talks about «salicta ad Minturnas». Also the quercus has a literary attestation. The Latin poet Claudianus (Paneg. in Probum et Olyb., 259) in fact, mentions querceta Maricae. As we can see, the sedimentary sequence and the pollen analysis correspond to the literary description of the wetlands near Minturnae.

21The wetlands had the right traits to become places of worship because they are places where natural elements dominated the landscape. In effect the wetlands near the Garigliano mouth had also a holy trait since the Iron Age. The area on the right bank of the river, between the shoreline and the marshes, was consecrated to the goddess Marica. The cult of Marica was pre-roman, but it continued still for a long time with similar characteristics. Many Latin authors refers to the occurrence of this sanctuary and its temple. According to the chthonian trait of Marica, the offerings were probably thrown into the waters of the marsh (Mingazzini, 1938). Plutarch confirms this aspect of the cult. The Greek historian (Mar., 39, 4) connotes the holy wood of Marica as a place from which nothing was permitted to be carried out that had ever been carried in. Marica was probably assimilated to Artemis, another goddess associated with the uncivilized, natural and savage world (Cerchiai, 1999 p. 236-237; Bellini, 2002; Livi, 2006 p. 108). She is usually related also to the value of asylum or inviolable space (Cerchiai, 1999, p. 236). A statue of Artemis was found into the river in front of the temple (Bellini, 2002, p. 67). In addition to Marica, the Romans venerated also Aphrodite. This cult is attested both by literary (Serv. Ad Aen., VII, 47: Dicunt alii Maricam Venerem intellegi debere, cuius fuit sacellum iuxta Maricam,in quo erat scriptum ΠΟΝΤΙΗ ΑΦΡΟΔΙΤΗ) both by epigraphic sources (a statue of Aphrodite Pontia was found in the Garigliano river – Bellini, 2002, p. 67). All these cults testify also the role of landing place of the sanctuary. In effect it was a transitional place between the human and the natural world, but also a refuge for seamen and a trading place.

22The wetlands were marginal places, but not only from a religious point of view. Their savage and uninhabited nature makes them good places also for people who live on the fringes of society. When Caius Marius was trying to leave Italy in 88 BC, he passed through the territory of Minturnae. When he was in danger because his enemies found him, he took refuge in the marshes. In fact the wetlands were inaccessible places: Plutarch says that they lack of any path (Mar. 37, 4: νοδίαις) and Appian refers to unfrequented roads (B.C., I, 7, 62: δος τριβες). Moreover the marshes were rich in vegetation and were characterized by several depressions and hollows (Plut. Mar., 37, 6: ν χωρί κοίλ παρ τν ποταμν). According to Appian (App. BC., IV, 4, 28), in 43 BC another man tried to seek refuge in the marshes of Marica. As Marius, also Sextus Quintilius Varus hide himself into the vegetation, but its hiding-place was revealed by the agitation of the reeds. Varus was discovered because the inhabitants of Minturnae were mounting over the marsh in search of robbers. This passage shows that the presence of robbers and exiles in marginal places like wetlands was very frequent and recurring. For instance, near Cuma there is a forest extended over numerous acres where the admirals of Sextus Pompeius assembled their gangs of pirates (Strabo, 5, 4, 4). It was probably because of these reasons that Columella compares the danger of marshes and wetlands to the danger of military roads (De Re Rustica 1.5.2).

23Even though wetlands were marginal places, they were in some way integrated into the Roman economy. Several ancient authors talk about the resources available in swamps and marshes and also the wetlands near Minturnae were exploited. A significant proof could be found in the text of Plutarch (Mar. 37, 5). According to the Greek historians, in fact, Marius met in the marsh the hut of a man that gave him help. This man was a λιμνουργος that literally means “a person who got his living from the wetland/water”. This passage shows that the area was not completely uninhabited (there was a little building) and that there were persons who exploited the wetlands. The exploitation concerned especially the vegetation even though we can suppose that also fishing was practiced. Appian reports that Marius rushed to the boat of an old fisherman (B.C., I, 7, 62), but he doesn’t say if the boat was anchored in the river (probably more) or in the marshes. So, it is not possible to use this information as a proof of sure fishing activities in the wetlands..

24Among the plants attested by pollen analysis and literary sources we can find several types that were someway used. The reeds were useful for buildings (Plin. N.H., 16, 156), paper (Plin. N.H., 16, 157), musical instruments (Plin. N.H., 16, 164). Moreover the Umbri obtained salt by boiling the ash of reeds (Plin. N.H., 31, 83). A passage of the Metamorphosis of Ovidius (VI, 344-345) confirms that the gathering of reeds, osiers and bulrushes were usual activities in wetlands.

25Several trees could be used for economic pursuits, for instance to produce stakes (Plin. N.H. 17, 151). The Alnus, a taxon attested with abundance in the marhes of Minturnae, was employed for making hollow pipes for the conveyance of water, because, when buried in the earth, it lasts for many year (Plin. N.H. 16,224). Minturnae had probably a boatyard near the river mouth, as we can gather from the existence of an acrhitectus navalis and a societas piacriorum and the wood was sure a necessary element to produce boats (Bellini, Trigona, Matullo, 2011 p. 567-568).

26The Salix is the most attested taxon. According to Cato (De Agr. 1: de omnibus agris optimoque loco iugera agri centum […] tertio salictum) the willow fields are at the third place in the rank of the best type of fields. Pliny enumerates several ways to use willows (Plin. N.H. 16, 174). They could be used as perches for the wine, for the purpose of tying or for the manufacture of wicker-work. The willows could be useful in the construction of articles of luxury or were exploited for weaving panniers, and many other utensils employed in agriculture. It’s no coincidence that according to Cato Minturnae was famous for its manufacture of agricultural tools (De Agr. 135: Calibus et Minturnis cuculliones, ferramenta, falces, palas, ligones, secures, ornamenta, murices, catellas).

27The pollen diagram (fig. 7) shows that the values of Salix decrease at about 95 cm deep, but return to higher values in the upper part. At the same depth, we can register the occurrence of Myriphyllum Alterniflorum. This plant is a light demanding species that grows in water with low mineral content. It is currently used as a biomarker for its sensitivity to "minerals" in solution. Immediately above and below this level, Myriophyllum verticillatum type is also present, which is less sensitive in mineral content than the previous one. We can suppose that this phase corresponds to the Imperial period (approximately I century AD), when Romans reinforced the river banks (structures along the river banks are signaled by Ruegg, 1995 p. 32-33, 132; Arata, 1993 p. 159-160; Andreani 2003 p. 202) and developed the sanctuary of Marica with new structures (Mingazzini 1938; Andreani 2003). In fact, by these actions, they could probably prevent the water of the Garigliano, rich in calcium carbonate, from overflowing into the basin. If this hypothesis is right, we have to think that the reduction of the number of willows took place during the Roman Imperial time, probably because of natural and climatic causes, probably because of the exploitation of this plant for economic pursuits.

Figure 7 – Pollen diagram

Figure 7 – Pollen diagram

28The description of the Greek and Latin authors refers to the late Republican period and finds several correspondences with the geoarchaeological data. During the Roman times the landscape changes. Myrtus reappears. At this time the area is covered by oak-dominated woodland, made up of diverse trees (e.g. Corylus, Ulmus, Ostrya/Carpinus orientalis, Carpinus betulus), by evergreen communities, probably more abundant along the coastal belt, and by riparian trees and freshwater aquatics related to the presence of marshlands. On the contrary the number of willows decrease. The two basins were probably preserved from the river overflows and were characterized by limpid and oligotrophic water (Ferrari et al. 2013).

  • 1

29As we said before, a picture in the Corpus Agrimensorum Romanorum represents the territory of Minturnae. We have 2 versions of the picture that differ only for little details (the codex Palatinus 1564, the codex Gudianus of the 9th century): we can see the Liris river (Garigliano) passing through the city and flowing into a big lake. Behind the city there are some mountains, on the right side of the river there are 3 monuments (a statue, a hexagonal monument, another monument similar to a tomb) and the words agri adsignati per professiones, on the left of the river there is a grid with the expression assignatio nova. According to recent studies the lake would represent the swamp or lagoon described by Plutarch and the three monuments on the right could be interpreted as the amphitheatre (the hexagon), the sanctuary of Marica (the statue) and another monument not identifiable. In this way the picture would be a schematic but realistic representation of the ager Minturnensis (Bellini, 2000 p. 21-22). Chouquer, that analyses this image in a text published on internet1, thinks that the lake represents the sea. Nevertheless it is open to different interpretations and he shows a photo with a palaeoenvironmental reconstruction similar to the ancient picture.

30In our opinion the picture doesn’t have any correspondence with the roman environment, but is only a schematic representation of the data concerning the centuriation (about the relationship between the picture, the Hyginus’ text – Hyg. Grom. De Lim. Const. 142, 2-19 Thul. – and the passage of the Liber ColoniariumLib. Col. 235, 12-14 Lach. – see Chouquer et al., 1987). The unique geographic elements are the mountains, the river and the city. The lake, as in the picture of Terracina, symbolizes the sea. The grid on the left is the symbol of the adsignatio nova with limites and the three monuments are the signs of a reorganization of the preexistent land division without tracing new limites (in other pictures of the Corpus Agrimensorum Romanorum these signs indicate division of lands without limites es. Lach, fig. 210). For these reasons this picture doesn’t contain any environmental data concerning the wetlands.

31Sometimes coastal lagoons and ponds could be used as saltworks. The epigraphic sources attested the occurrence of socii salinatores at Minturnae and this information could led to think that also the wetlands near the Garigliano mouth were used to obtain salt. The analysis of pollens and mollusk fragments indicates that in the Roman times we were in front of a fresh-water environment. Also Plutarch describes the marshes as places dominated by silty water and vegetation and this description is scarcely compatible with the occurrence of saltwarks. Probably we have to think that the socii salinatores were linked to salt trading activities (or something similar) rather than salt extraction. Moreover it is not so clear if the terms themselves “salina” and “salinatores” refer to the occurrence of salwarks (Traina, 1992). Unfortunately this problem is not yet solved and the question is still open.

32According to recent studies (Andreani, 2006) the area corresponding to the wetlands was reclaimed and assigned to the settlers. By analyzing the aerial photographs, the researcher identifies some traces that were interpreted as ancient limites of a centuriation, dated probably to the period after the victory of Sulla. The geoarchaeolocal data allow us to reject this hypothesis. Both the sedimentary sequence both the pollen analysis show a freshwater and marshy environment and there are no traces of cultivated plants. Also the literary sources describe a similar environment both in 88 BC both in 43 BC. In the Roman times the hollowed area between the Holocene and the Euthyrrenian beach ridge was occupied by water and was not reclaimed. Crop marks and soil marks identified by Andreani should be attributed to a more recent period. The unique traces of centuriation didn’t interest the coast and could be find in the inland (Chouquer et al. 1987 p. 169-180).

33The situation changes again since the 3rd century AD. Several settlements attested in the previous period near the coast were abandoned. The disappearance of Myriophyllum Alterniflorum and later also of the Myriophyllum Verticellatum indicates that probably the Garigiano, with its waters rich in carbonate, restarted to overflow periodically into the basins and to fill them with its sediments. The landscape became uninhabited and more and more marshy. The wetlands were sometimes dry and used for pasture, as we can gather from the increase of Cichorieae and wild Poaceae with the good presence of coprophilous fungal spores.


34As we have seen before, the abundance of data allows us to reconstruct the evolution of the wetlands. The correlation between geoarchaeological data and literary information led us to understand what was the landscape in Roman times and what were the value and the function of the marshes. In the time of Marius the coast was characterized by the occurrence of swamps and little lakes rich in vegetation. We can hypothesize a little transformation during the Imperial period. The Salix decreased, the water became more limpid and oligotrophic, the Garigliano was banked, even though an aquatic environment is still attested. The marshes were very close to the city, but they were not reclaimed. Men could not control the wetlands with their laws and order. Therefore they were places where it was possible to contact the nature and the underworld thanks to the ritual mediation of the religion. They were also ideal places for robbers and exiles because they were outside the civil world. Nevertheless the marshes offered economic resources and they were someway exploited. In conclusion we can say that, even when the Romans could not reclaim wetlands, they didn’t exclude marshes and swamps from their life, but included them from an economic and religious point of view.

Haut de page


ALESSANDRI L. 2007, L’occupazione costiera protostorica del Lazio centromeridionale, Oxford, BAR (BAR International Series, 1592), 241 p.

ANDREANI M., 2003, Sul santuario di Marica alla foce del Garigliano, in QUILICI L., GIGLI QUILICI S. (eds.), Santuari e luoghi di culto nell’Italia antica, Roma, L’Erma di Bretschneider (Atlante Tematico di Topografia Antica, 12), p. 177-207.

ANDREANI M., 2006, Tracce di centuriazione alla foce del Garigliano, Agri Centuriati, 3, p. 59-72.

Angle M., Belardelli C., 2007, Insediamenti artigianali nel Lazio meridionale durante la tarda età del bronzo, Atti della 40. Riunione scientifica : Strategie di insediamento fra Lazio e Campania in età preistorica e protostorica (Roma, Napoli, Pompei, 30 novembre-3 dicembre 2005: dedicati ad Amilcare Bietti), p. 763-774.

ARATA F.P., 1993, Indagini subacquee nell’alveo del Garigliano, presso Minturnae, Archeologia subacquea, I, p. 159-162.

ARTHUR P., 1989, Assetto territoriale ed insediamento fra tardo antico ed alto medioevo nel bacino del Garigliano, in COARELLI, F. 1989 (ed). Minturnae, Roma, NER, p. 183-191.

BANG M., 1910. Marius in Minturnae. Klio, X, p. 178-191.

BELLINI G.R., 1998, Minturnae: 296 a.C. – 44 a.C. Dalla deduzione della colonia alla morte di Cesare. In BELLINI G.R., 1998 (ed.), Monete dal Garigliano. III. La città ed il porto (296-44 a.C.): monete romane repubblicane (fino alla morte di Cesare), Milano, Ennerre (Minturnae Antiquarium – Monete dal Garigliano III), p. 9-15.

BELLINI G.R. 2000, Minturnae: trasformazioni e società tra la tarda repubblica e gli Antonini, in BELLINI G.R., 2000 (ed.), Monete dal Garigliano. V : 1.Minturnae: trasformazioni e societa tra la tarda repubblica e gli Antonini; 2. Committenza degli edifici pubblici di Minturnae tra la tarda eta repubblicana e il periodo antonino; 3. monete dal Garigliano (68-98 d.C.), l'interregno, la dinastia Flavia e il regno di Nerva. Milano, Ennerre (Minturnae Antiquarium – Monete dal Garigliano V), p. 9-23.

BELLINI G.R., 2002, Il santuario emporico di Marica alla foce del Garigliano, in Il Lazio regione di Roma: Palestrina, Museo archeologico nazionale 12 luglio - 10 settembre 2002, p. 67.

BELLINI G.R., 2007. Minturnae porto del Mediterraneo, Romula, 6, p. 7-28.

BELLINI G.R., TRIGONA S.L., MATULLO G., 2011, Minturnae. Il Garigliano, in GHINI G. (ed.), Lazio e Sabina, 7: Settimo Incontro di studi sul Lazio e la Sabina (Atti del convegno: Roma 9-11 marzo 2010), Roma, p. 563-574.

Bellotti P., 2000, Il modello morfo-sedimentario dei maggiori delta tirrenici italiani, Bollettino Società Geologica Italiana, 119, p. 777-792.

CERCHIAI L., 1999, Appunti sui culti di Marica e Mefite, Ocnus, 7, p. 235-241.

CHOUQUER G., CLAVEL-LEVEQUE M., FAVORY F., VALLAT J.P., 1987, Structures agraires en Italie Centro-méridionale. Cadastres et paysages ruraux, Roma, Ecole française de Rome. (Collection de l’Ecole française de Rome, 100), 414 p.

Ciarrocchi B., Torre P. 2003, Reperti vitrei dallo scavo di Monte d’Argento (Minturno, LT): un panorama tipologico dal tardoantico al basso medioevo, in PICCOLI C, SOGLIANI F. (ed.), Il vetro in Italia meridionale e insulare: settime giornate nazionali di studio Comitato nazionale AIHV. (Atti del secondo Convegno multidisciplinare, Napoli, 5-6-7 dicembre 2001), Napoli, De Frede, p. 185-196.

D’URSO M.T. 1985, Il tempio della dea Marica alla foce del Garigliano. Volume Primo. La Storia, Scauri, Archeoclub Minturnae, 143 p..

Di Biasio A., 1994. Il passo del Garigliano nella storia d’Italia. Il ponte di Luigi Giura, Marina di Minturno, Caramanica, 295 p.

FERRARI K., BELLOTTI P., Dall’Aglio P.L., Davoli L., Mazzanti M., Torri P., 2012, Environment and settlements near the Garigliano river mouth: history of an evolving landscape, in Bertoncello F., Braemer F. (Eds), Variabilités environnementales, mutations sociales. Nature, intensités, échelles et temporalités des changements, Antibes, Editions APDCA, p. 309-321.

Ferrari K., Dall’Aglio P.L., Bellotti P., Davoli L., Di Bella L., Esu D., Torri P., Mazzanti Bandini M., 2013, Holocene landscape evolution at the Garigliano River mouth, Annali di Botanica, 3, p. 191-198.

GUIDI A., 1991, Alcune osservazioni sul popolamento protostorico tra il Golfo di Gaeta e gli Aurunci, Latium, 8, p. 5-31.

GUIDOBALDI M.P., PESANDO F., 1989, La colonia civium romanorum, in COARELLI F. (ed). Minturnae, Roma, NER, p. 35-66.

LAFORGIA E., 1992, Nuove osservazioni sul tempio di Marica, Annali di Archeologia e Storia Antica. Istituto universitario orientale. Dipartimento di studi del mondo classico e del Mediterraneo antico 14, p. 69-76.

LIVI V., 2006, Religious locales in the territory of Minturnae: aspects of Romanization, in Schultz C.E., Harvey P.B., Religion in republican Italy, Cambridge, University Press (Yale classical studies, 33), p. 90-116.

LOCCHI A., 2010, Le acque insidiose. Laghi e paludi nello scenario mitico greco, in DI GIUSEPPE H., SERLORENZI M., (edts.), I riti del costruire nelle acque violate. Atti del Convegno Internazionale. Roma, Palazzo Massimo 12-14 giugno 2008. Roma, Scienze e Lettere, p. 69-90.

MESOLELLA G., 2000, Considerazioni sulla committenza degli edifici pubblici di Minturnae tra la tarda età repubblicana e il periodo antonino, in BELLINI G., (ed.) 2000, Monete dal Garigliano. V : 1.Minturnae: trasformazioni e societa tra la tarda repubblica e gli Antonini; 2. Committenza degli edifici pubblici di Minturnae tra la tarda eta repubblicana e il periodo antonino; 3. monete dal Garigliano (68-98 d.C.), l'interregno, la dinastia Flavia e il regno di Nerva., Milano, Ennerre (Minturnae Antiquarium – Monete dal Garigliano V), p. 25-47.

MINGAZZINI P., 1938, Il santuario della dea Marica alle foci del Garigliano., Milano: Hoepli (Monumenti antichi pubblicati per cura della R. Accademia dei Lincei, 37), col. 693-984.

PACCIARELLI M., 2000, Dal villaggio alla città: la svolta protourbana del 1000 a. C. nell'Italia tirrenica, Firenze, All’insegna del Giglio (Grandi contesti e problemi della protostoria italiana, 4), 311 p.

RESCIGNO C., 1993, L’edificio arcaico del santuario di Marica alle foci del Garigliano: le terrecotte architettoniche. Annali di Archeologia e Storia Antica. Istituto universitario orientale. Dipartimento di studi del mondo classico e del Mediterraneo antico, 15, p. 85-108.

Rossillo O., 1985, La colonia musulmana del Garigliano. Scauri, Archeoclub di Minturnae, 67 p.

RUEGG S. D., 1995, Underwater investigations at Roman Minturnae : Liris-Garigliano river Goteborg, S. Dominic Ruegg and Paul Astroms (Studies in Mediterranean archaeology, 119), 2 vol., 147 and 221 p.

SUSINI , G. 1960, Note di epigrafia parmense, Epigraphica, 22, p. 152-170.

TALAMO P., 1987, L’area aurunca nel quadro dell’Italia centromeridionale. Testimonianze archeologiche di età arcaica, Oxford, BAR (BAR, 5), 191 p.

TORRE P., 1988, Monte d’Argento: indagini preliminari. Archeologia Laziale, IX, p. 432-440.

TORRE P., 1998, Il rinvenimento di ceramiche invetriate e smaltate con motivi decorativi nell’insediamento di Monte d’Argento, in DE MINICIS, E. (ed.), Le ceramiche di Roma e del Lazio in età medievale e moderna, III (Atti del III Convegno di Studi – Roma 19-20 aprile 1996), Roma, Kappa, p. 183-206.

TRAINA G., 1988, Paludi e bonifiche del mondo antico. Saggio di archeologia geografica, Roma, L’Erma di Bretschneider, 159 p.

TRAINA G., 1992, Sale e saline nel Mediterraneo antico, Parola del Passato, 47, p. 363-378.

Haut de page



Haut de page

Table des illustrations

Titre Figure – Fig. 38 Th. (fig. 39 La)
Fichier image/jpeg, 292k
Titre Figure - Fig. 122 Th.
Fichier image/jpeg, 276k
Titre Figure – Fig. 92 Th. (Marshes near Terracina – Paludes)
Fichier image/jpeg, 288k
Titre Figure 4 – Fig. 89 Th. (territory of Minturnae)
Fichier image/jpeg, 152k
Titre Figure 5 – Geomorfological units and settlements near the Garigliano River mouth.
Fichier image/jpeg, 792k
Titre Figure 6 – Sedimentary sequence of the cores
Fichier image/jpeg, 568k
Titre Figure 7 – Pollen diagram
Fichier image/jpeg, 143k
Haut de page

Pour citer cet article

Référence papier

Kevin Ferrari, Piero Bellotti, Pier Luigi Dall'Aglio, Lina Davoli, Mazzanti Marta et Paola Torri, « Wetlands in the river delta plains: evolution, values and functions during the Roman times. The coastal landscape close to the Garigliano river mouth », Géocarrefour, vol. 88/4 | 2013, 273-283.

Référence électronique

Kevin Ferrari, Piero Bellotti, Pier Luigi Dall'Aglio, Lina Davoli, Mazzanti Marta et Paola Torri, « Wetlands in the river delta plains: evolution, values and functions during the Roman times. The coastal landscape close to the Garigliano river mouth », Géocarrefour [En ligne], vol. 88/4 | 2013, mis en ligne le , consulté le 21 mai 2019. URL : ; DOI : 10.4000/geocarrefour.9237

Haut de page


Kevin Ferrari

Dipartimento di Storia, Culture, Civiltà – Università di Bologna

Piero Bellotti

Dipartimento di Scienza della Terra – Università di Roma

Pier Luigi Dall'Aglio

Dipartimento di Storia, Culture, Civiltà – Università di Bologna

Lina Davoli

Dipartimento di Scienza della Terra – Università di Roma

Mazzanti Marta

Laboratorio di Palinologia e Paleobotanica- Dipartimento di Scienze della Vita – Univ. di Modena e Reggio Emilia

Paola Torri

Laboratorio di Palinologia e Paleobotanica- Dipartimento di Scienze della Vita – Univ. di Modena e Reggio Emilia

Haut de page

Droits d’auteur

© Géocarrefour

Haut de page
  • OpenEdition Journals