In 1303, a powerful earthquake near Crete triggered an enormous tsunami that swept across the eastern Mediterranean. Medieval accounts reveal how the sea surged into cities, exposed the seabed and devastated coastal communities from Crete to Egypt and the Levant.
By Lorris Chevalier
The earthquake that struck the eastern Mediterranean on 8 August 1303, at approximately dawn in Crete, was one of the most destructive seismic events of the medieval Mediterranean. Yet the earthquake was only one element of the disaster: an enormous tsunami followed, striking coastlines across the eastern Mediterranean.
The event is particularly difficult to reconstruct because its effects were recorded across a vast geographical area and in several different linguistic and political traditions. Arabic, Greek, Byzantine, Venetian and Latin sources describe different aspects of the disaster, often concentrating on what happened within their own communities. As a result, earlier scholarship sometimes treated the earthquake and its consequences as a collection of separate local events rather than as parts of a single Mediterranean disaster.
The tsunami was a major disaster in its own right. Its waves struck Crete, Egypt and the Levant, causing destruction and deaths, while unusual movements of the sea were recorded as far away as the Adriatic. When these scattered accounts are brought together, they reveal the remarkable geographical scale of the catastrophe.
The earthquake across the Mediterranean
16th century map of the Eastern Mediterranean
The earthquake occurred at approximately 03:30 GMT on 8 August 1303, at dawn in Crete. Guidoboni and Comastri place its epicentral area in south-eastern Crete and assign a maximum intensity of approximately XI on the Mercalli–Cancani–Sieberg scale, with an epicentral intensity of around X. Their reconstruction draws on evidence from 41 localities and areas and approximately 90 references from 50 sources.
The scale of the event explains the unusually wide geographical distribution of reports. Strong shaking was experienced throughout Crete and across the eastern Mediterranean. The evidence is not equally reliable everywhere, however. Some sources use expressions that may mean simply that news of the earthquake had arrived from a distant region rather than that the earthquake itself had been strongly felt there.
This is particularly relevant to reports from Marrakesh, Tunis and Cádiz. The contemporary Arab historian Ibn al-Dawadari states that news arrived from places that had been ‘reached’ by the earthquake. Because of the enormous distances involved, Guidoboni and Comastri treated these locations cautiously, marking them as possibly affected rather than accepting them as definite points within the felt area. A similar expression is used for Constantinople, where another source suggests that only a small number of people actually felt the tremor.
Cyprus provides firmer evidence. At Leukosia, the earthquake was strongly felt according to a Cypriot chronicle. These distant observations help establish the enormous geographical reach of the seismic event, but they also demonstrate why historical earthquakes must be reconstructed by comparing sources rather than simply counting mentions.
The most remarkable evidence, however, comes not from the distribution of shaking but from the movement of the sea.
The tsunami across the Mediterranean
The tsunami generated by the earthquake was described as exceptionally large. Its effects were recorded in Crete, Egypt and Palestine, while unusual sea movements were also observed in the Adriatic. The geographical pattern is striking because it broadly resembles the distribution of the tsunami generated by the great 365 earthquake, also centred near Crete.
Yet the tsunami of 1303 has remained comparatively obscure. This may partly be because medieval sources recorded the event according to local concerns. In Crete, administrative documents concentrated on destroyed buildings, taxation and reconstruction. In Egypt and the Levant, Arabic historians described damage to ports, cities and religious buildings. Latin and Italian chronicles sometimes preserved only brief notices of extraordinary sea movements.
Taken together, however, these sources reveal a much larger phenomenon.
Crete: the sea sweeps into Candia
Crete from the Bodleian Libraries manuscript
The clearest evidence comes from Candia, modern Heraklion, the principal Venetian city of Crete.
According to several Italian chronicles, including Ricobaldo da Ferrara, the Annales Caesenates, the Annales Forolivienses and Bartolomeo della Pugliola, the sea suddenly swept into the city with enormous force. Buildings were destroyed and inhabitants were killed. The water then rapidly withdrew from the harbour, exposing parts of the beach before the sea returned.
This sequence — violent inundation followed by rapid withdrawal — is characteristic of a tsunami. It is one of the most important pieces of evidence for identifying the event as a tsunami rather than simply an unusually high tide or storm surge.
The withdrawal of the sea would also have created a dangerous illusion of safety. People might have moved towards the exposed harbour or shoreline, only to be confronted by the returning wave. Similar behaviour has been documented in many later tsunamis, making the medieval description particularly significant.
The tsunami therefore transformed the earthquake from a seismic disaster into a coastal catastrophe. Buildings already weakened by the earthquake were exposed to moving water, while people attempting to escape inland faced flooding and debris.
Acre: the sea exposes the seabed
14th century map of Acre – British Library Additional MS 27376 f. 190r
The effects at Acre, on the coast of present-day Israel, were equally dramatic.
The Arabic historian al-Nuwayri records that the sea flooded the shoreline as far as the Tower of Dayan. Objects that had previously been thrown into the sea during the 1291 siege of the city were reportedly found on the shore. People attempted to recover some of these objects, but a huge returning wave swept them away and drowned them.
Another account, preserved by al-Maqrizi, states that the sea withdrew by approximately two parasangs, around 12.8 kilometres, exposing a large part of the seabed before returning. The precise distance should be treated cautiously, as medieval measurements and descriptions cannot simply be converted into modern geographical measurements. Nevertheless, the account is important because it describes the classic sequence of a major tsunami: extreme withdrawal of the sea followed by catastrophic inundation.
The episode also illustrates the social danger created by tsunami waves. People were drawn towards the newly exposed seabed, perhaps seeing objects that had been hidden underwater for generations. When the water returned, those who had approached the shore were caught in the wave.
Alexandria: a port overwhelmed
The lighthouse as depicted in the Book of Wonders, a late 14th-century Arabic text – MS. Bodl. Or. 133 fol. 36a
The tsunami also struck Alexandria, one of the great ports of the medieval Mediterranean.
Arabic sources describe the sea first receding and then advancing towards the city, reaching the walls. Shops near the waterfront were submerged and destroyed, while food stored close to the shore was swept away. In the harbour, ships broke their moorings and boats were thrown against the rocks.
The importance of Alexandria lies not simply in the physical destruction recorded there, but in what the episode reveals about the economic consequences of a tsunami. Alexandria was a major commercial centre whose prosperity depended upon its harbour. A sudden displacement of seawater could therefore damage ships, warehouses, food supplies, quays and commercial infrastructure simultaneously. The destruction of harbour facilities would have disrupted trade precisely when the city was already dealing with the consequences of the earthquake.
The tsunami should consequently be understood not merely as a natural phenomenon but as an event capable of disrupting the economic infrastructure of an entire region.
The Nile and the landscape of Egypt
The effects were not confined to the Mediterranean coastline. At Minya, further south in Egypt, an anonymous source describes the waters of the Nile apparently separating and exposing the riverbed before coming together again. Landslides were also reported in the mountains around Minya.
These observations are particularly valuable because they demonstrate how unusual water movements and ground instability could extend well beyond the immediate coastline. Although the exact mechanism behind the Nile phenomenon remains difficult to establish, its inclusion in the historical record indicates how extraordinary the disturbance appeared to contemporary observers.
The earthquake therefore affected not only cities and ports but also rivers, mountains and the wider physical landscape.
The Adriatic: a distant sea disturbance
Late Medieval Portalan chart showing the Adriatic Sea
The tsunami’s effects may even have been detected in the Adriatic Sea.
Italian chronicles report an anomalous movement of the sea, although without significant damage. The distance from Crete means that the evidence must be treated carefully. Nevertheless, the reports are consistent with the possibility that the disturbance propagated across a considerable portion of the Mediterranean basin.
This is one reason why the 1303 event is so important for understanding medieval Mediterranean hazards. The disaster was not confined to a single coastline. It was a basin-wide phenomenon, although its intensity varied greatly from place to place.
How the disaster was recorded
The evidence for the tsunami is particularly interesting because Christian and Muslim sources did not necessarily preserve the same aspects of the event.
In the Arabic sources, considerable attention was given to Alexandria, Acre and Cairo, with descriptions of flooding, damaged buildings, ships and the movement of the sea. Italian and Venetian sources, by contrast, provide unusually detailed evidence for the destruction and subsequent reconstruction of Crete.
This difference is not simply a matter of historical survival. It reflects the priorities of the societies producing the records.
The Venetian government had a direct administrative interest in Crete. Candia was the centre of Venetian government on the island, and its fortifications, harbour, churches, warehouses and administrative buildings were essential to Venetian political and commercial power. Consequently, Venetian records continued to document the consequences of the earthquake for decades.
The tsunami, by contrast, often survived in narrative sources as a remarkable episode: the sea entered the city, withdrew, and returned with destructive force. Because such accounts were not always incorporated into administrative records, the tsunami has been easier to overlook when historians focus on the political and economic aftermath of the earthquake.
The human cost
BnF Français 28, fol. 66v
The precise death toll cannot be established. Contemporary sources refer to numerous victims in Crete, Alexandria, Cairo and Rhodes, but do not provide reliable overall figures. A much later source gives a figure of 4,000 deaths in Crete, although this number cannot be independently verified.
What is certain is that the earthquake and tsunami produced substantial mortality.
The combination of violent ground shaking and sudden coastal flooding would have been particularly destructive. People in buildings damaged by the earthquake were vulnerable to collapse, while those near the coast faced a second and very different danger: rapidly moving water carrying timber, masonry, boats and other debris.
The tsunami also complicated rescue and recovery. Ports were damaged, communications disrupted and food supplies destroyed. In other words, the disaster continued long after the initial shaking had stopped.
Crete after the disaster
The earthquake had particularly severe economic consequences in Crete because the island was at the centre of the disaster.
A letter from the Venetian Chancellery at Candia dated 28 August 1303, only three weeks after the earthquake, requested substantial financial assistance because tax revenues had fallen dramatically. The destruction of property and disruption of economic activity had immediately affected the Venetian government’s ability to collect revenue.
The disaster also contributed to political instability. The rural Greek population of Venetian-controlled Crete took advantage of the disorder to rebel against Venetian authority and began looting. The revolt was eventually suppressed by a military force led by Alessio Calergi.
The economic crisis continued for years. Soldiers and custodians of the island’s castles were reduced to severe poverty. Income from the sale of fish and bread declined sharply, while damaged bakeries led the authorities to exempt bakers from taxation for a year. Warehouses containing merchandise were destroyed, disrupting commercial activity and producing further economic losses.
The tsunami therefore formed part of a much broader chain of consequences: earthquake → tsunami → destruction of infrastructure → disruption of trade and taxation → political instability → prolonged reconstruction.
This sequence is important because it shows that the disaster cannot be measured simply by the number of buildings that collapsed.
Rebuilding Crete
17th century depiction of Heraklion / Candia – Wikimedia Commons
The Venetian response provides unusually detailed evidence for the long-term consequences of the disaster.
In September 1303, twenty-eight builders and carpenters, many accompanied by their sons, were sent from Venice to Candia for a year to undertake reconstruction work.
In March 1304, the Venetian Senate allocated funds for rebuilding the walls and towers of Crete. In 1307, further funds were assigned to repair San Marco at Candia, the town hall, the arsenal and the harbour fortifications, as well as to construct a castle at Hierapetra.
The rebuilding continued for decades.
In 1315, the Venetian government again addressed the reconstruction of San Marco and other buildings. Wood was imported from Venice specifically for the restoration of the church of San Tito. In 1333, the Senate approved plans for rebuilding the harbour at Candia. In 1336, funds were allocated to restore several castles considered essential to the defence of the island. Further expenditure followed in 1342 and 1362.
Remarkably, in 1363, sixty years after the earthquake, the Venetian Senate was still discussing repairs to fortifications damaged by the disaster.
The documentary record therefore reveals something that the dramatic tsunami narratives alone cannot show: the effects of the 1303 disaster lasted for generations.
Rebuilding Cairo and Alexandria
The damage was not restricted to Crete. In Cairo, several important religious buildings required restoration. Arabic historians record repairs to the mosques and minarets of al-Azhar, ‘Amr ibn al-‘As and al-Hakim, as well as other religious complexes.
At Alexandria, the lighthouse, city walls and other structures were repaired at the expense of the Sultan. The citadel and towers of Safad were also repaired in 1304.
The reconstruction itself created new economic pressures. Al-Maqrizi remarks that the demand for building materials in Cairo caused prices to rise dramatically. This is an important indication of how natural disasters can generate secondary economic effects: the destruction creates an urgent demand for labour and materials at precisely the moment when supply has been disrupted.
Dr Lorris Chevalier, who has a Ph.D. in medieval literature, is a historical advisor for movies, including The Last Duel and Napoleon. Click here to view his website.
In 1303, a powerful earthquake near Crete triggered an enormous tsunami that swept across the eastern Mediterranean. Medieval accounts reveal how the sea surged into cities, exposed the seabed and devastated coastal communities from Crete to Egypt and the Levant.
By Lorris Chevalier
The earthquake that struck the eastern Mediterranean on 8 August 1303, at approximately dawn in Crete, was one of the most destructive seismic events of the medieval Mediterranean. Yet the earthquake was only one element of the disaster: an enormous tsunami followed, striking coastlines across the eastern Mediterranean.
The event is particularly difficult to reconstruct because its effects were recorded across a vast geographical area and in several different linguistic and political traditions. Arabic, Greek, Byzantine, Venetian and Latin sources describe different aspects of the disaster, often concentrating on what happened within their own communities. As a result, earlier scholarship sometimes treated the earthquake and its consequences as a collection of separate local events rather than as parts of a single Mediterranean disaster.
The tsunami was a major disaster in its own right. Its waves struck Crete, Egypt and the Levant, causing destruction and deaths, while unusual movements of the sea were recorded as far away as the Adriatic. When these scattered accounts are brought together, they reveal the remarkable geographical scale of the catastrophe.
The earthquake across the Mediterranean
The earthquake occurred at approximately 03:30 GMT on 8 August 1303, at dawn in Crete. Guidoboni and Comastri place its epicentral area in south-eastern Crete and assign a maximum intensity of approximately XI on the Mercalli–Cancani–Sieberg scale, with an epicentral intensity of around X. Their reconstruction draws on evidence from 41 localities and areas and approximately 90 references from 50 sources.
The scale of the event explains the unusually wide geographical distribution of reports. Strong shaking was experienced throughout Crete and across the eastern Mediterranean. The evidence is not equally reliable everywhere, however. Some sources use expressions that may mean simply that news of the earthquake had arrived from a distant region rather than that the earthquake itself had been strongly felt there.
This is particularly relevant to reports from Marrakesh, Tunis and Cádiz. The contemporary Arab historian Ibn al-Dawadari states that news arrived from places that had been ‘reached’ by the earthquake. Because of the enormous distances involved, Guidoboni and Comastri treated these locations cautiously, marking them as possibly affected rather than accepting them as definite points within the felt area. A similar expression is used for Constantinople, where another source suggests that only a small number of people actually felt the tremor.
Cyprus provides firmer evidence. At Leukosia, the earthquake was strongly felt according to a Cypriot chronicle. These distant observations help establish the enormous geographical reach of the seismic event, but they also demonstrate why historical earthquakes must be reconstructed by comparing sources rather than simply counting mentions.
The most remarkable evidence, however, comes not from the distribution of shaking but from the movement of the sea.
The tsunami across the Mediterranean
The tsunami generated by the earthquake was described as exceptionally large. Its effects were recorded in Crete, Egypt and Palestine, while unusual sea movements were also observed in the Adriatic. The geographical pattern is striking because it broadly resembles the distribution of the tsunami generated by the great 365 earthquake, also centred near Crete.
Yet the tsunami of 1303 has remained comparatively obscure. This may partly be because medieval sources recorded the event according to local concerns. In Crete, administrative documents concentrated on destroyed buildings, taxation and reconstruction. In Egypt and the Levant, Arabic historians described damage to ports, cities and religious buildings. Latin and Italian chronicles sometimes preserved only brief notices of extraordinary sea movements.
Taken together, however, these sources reveal a much larger phenomenon.
Crete: the sea sweeps into Candia
The clearest evidence comes from Candia, modern Heraklion, the principal Venetian city of Crete.
According to several Italian chronicles, including Ricobaldo da Ferrara, the Annales Caesenates, the Annales Forolivienses and Bartolomeo della Pugliola, the sea suddenly swept into the city with enormous force. Buildings were destroyed and inhabitants were killed. The water then rapidly withdrew from the harbour, exposing parts of the beach before the sea returned.
This sequence — violent inundation followed by rapid withdrawal — is characteristic of a tsunami. It is one of the most important pieces of evidence for identifying the event as a tsunami rather than simply an unusually high tide or storm surge.
The withdrawal of the sea would also have created a dangerous illusion of safety. People might have moved towards the exposed harbour or shoreline, only to be confronted by the returning wave. Similar behaviour has been documented in many later tsunamis, making the medieval description particularly significant.
The tsunami therefore transformed the earthquake from a seismic disaster into a coastal catastrophe. Buildings already weakened by the earthquake were exposed to moving water, while people attempting to escape inland faced flooding and debris.
Acre: the sea exposes the seabed
The effects at Acre, on the coast of present-day Israel, were equally dramatic.
The Arabic historian al-Nuwayri records that the sea flooded the shoreline as far as the Tower of Dayan. Objects that had previously been thrown into the sea during the 1291 siege of the city were reportedly found on the shore. People attempted to recover some of these objects, but a huge returning wave swept them away and drowned them.
Another account, preserved by al-Maqrizi, states that the sea withdrew by approximately two parasangs, around 12.8 kilometres, exposing a large part of the seabed before returning. The precise distance should be treated cautiously, as medieval measurements and descriptions cannot simply be converted into modern geographical measurements. Nevertheless, the account is important because it describes the classic sequence of a major tsunami: extreme withdrawal of the sea followed by catastrophic inundation.
The episode also illustrates the social danger created by tsunami waves. People were drawn towards the newly exposed seabed, perhaps seeing objects that had been hidden underwater for generations. When the water returned, those who had approached the shore were caught in the wave.
Alexandria: a port overwhelmed
The tsunami also struck Alexandria, one of the great ports of the medieval Mediterranean.
Arabic sources describe the sea first receding and then advancing towards the city, reaching the walls. Shops near the waterfront were submerged and destroyed, while food stored close to the shore was swept away. In the harbour, ships broke their moorings and boats were thrown against the rocks.
The importance of Alexandria lies not simply in the physical destruction recorded there, but in what the episode reveals about the economic consequences of a tsunami. Alexandria was a major commercial centre whose prosperity depended upon its harbour. A sudden displacement of seawater could therefore damage ships, warehouses, food supplies, quays and commercial infrastructure simultaneously. The destruction of harbour facilities would have disrupted trade precisely when the city was already dealing with the consequences of the earthquake.
The tsunami should consequently be understood not merely as a natural phenomenon but as an event capable of disrupting the economic infrastructure of an entire region.
The Nile and the landscape of Egypt
The effects were not confined to the Mediterranean coastline. At Minya, further south in Egypt, an anonymous source describes the waters of the Nile apparently separating and exposing the riverbed before coming together again. Landslides were also reported in the mountains around Minya.
These observations are particularly valuable because they demonstrate how unusual water movements and ground instability could extend well beyond the immediate coastline. Although the exact mechanism behind the Nile phenomenon remains difficult to establish, its inclusion in the historical record indicates how extraordinary the disturbance appeared to contemporary observers.
The earthquake therefore affected not only cities and ports but also rivers, mountains and the wider physical landscape.
The Adriatic: a distant sea disturbance
The tsunami’s effects may even have been detected in the Adriatic Sea.
Italian chronicles report an anomalous movement of the sea, although without significant damage. The distance from Crete means that the evidence must be treated carefully. Nevertheless, the reports are consistent with the possibility that the disturbance propagated across a considerable portion of the Mediterranean basin.
This is one reason why the 1303 event is so important for understanding medieval Mediterranean hazards. The disaster was not confined to a single coastline. It was a basin-wide phenomenon, although its intensity varied greatly from place to place.
How the disaster was recorded
The evidence for the tsunami is particularly interesting because Christian and Muslim sources did not necessarily preserve the same aspects of the event.
In the Arabic sources, considerable attention was given to Alexandria, Acre and Cairo, with descriptions of flooding, damaged buildings, ships and the movement of the sea. Italian and Venetian sources, by contrast, provide unusually detailed evidence for the destruction and subsequent reconstruction of Crete.
This difference is not simply a matter of historical survival. It reflects the priorities of the societies producing the records.
The Venetian government had a direct administrative interest in Crete. Candia was the centre of Venetian government on the island, and its fortifications, harbour, churches, warehouses and administrative buildings were essential to Venetian political and commercial power. Consequently, Venetian records continued to document the consequences of the earthquake for decades.
The tsunami, by contrast, often survived in narrative sources as a remarkable episode: the sea entered the city, withdrew, and returned with destructive force. Because such accounts were not always incorporated into administrative records, the tsunami has been easier to overlook when historians focus on the political and economic aftermath of the earthquake.
The human cost
The precise death toll cannot be established. Contemporary sources refer to numerous victims in Crete, Alexandria, Cairo and Rhodes, but do not provide reliable overall figures. A much later source gives a figure of 4,000 deaths in Crete, although this number cannot be independently verified.
What is certain is that the earthquake and tsunami produced substantial mortality.
The combination of violent ground shaking and sudden coastal flooding would have been particularly destructive. People in buildings damaged by the earthquake were vulnerable to collapse, while those near the coast faced a second and very different danger: rapidly moving water carrying timber, masonry, boats and other debris.
The tsunami also complicated rescue and recovery. Ports were damaged, communications disrupted and food supplies destroyed. In other words, the disaster continued long after the initial shaking had stopped.
Crete after the disaster
The earthquake had particularly severe economic consequences in Crete because the island was at the centre of the disaster.
A letter from the Venetian Chancellery at Candia dated 28 August 1303, only three weeks after the earthquake, requested substantial financial assistance because tax revenues had fallen dramatically. The destruction of property and disruption of economic activity had immediately affected the Venetian government’s ability to collect revenue.
The disaster also contributed to political instability. The rural Greek population of Venetian-controlled Crete took advantage of the disorder to rebel against Venetian authority and began looting. The revolt was eventually suppressed by a military force led by Alessio Calergi.
The economic crisis continued for years. Soldiers and custodians of the island’s castles were reduced to severe poverty. Income from the sale of fish and bread declined sharply, while damaged bakeries led the authorities to exempt bakers from taxation for a year. Warehouses containing merchandise were destroyed, disrupting commercial activity and producing further economic losses.
The tsunami therefore formed part of a much broader chain of consequences: earthquake → tsunami → destruction of infrastructure → disruption of trade and taxation → political instability → prolonged reconstruction.
This sequence is important because it shows that the disaster cannot be measured simply by the number of buildings that collapsed.
Rebuilding Crete
The Venetian response provides unusually detailed evidence for the long-term consequences of the disaster.
In September 1303, twenty-eight builders and carpenters, many accompanied by their sons, were sent from Venice to Candia for a year to undertake reconstruction work.
In March 1304, the Venetian Senate allocated funds for rebuilding the walls and towers of Crete. In 1307, further funds were assigned to repair San Marco at Candia, the town hall, the arsenal and the harbour fortifications, as well as to construct a castle at Hierapetra.
The rebuilding continued for decades.
In 1315, the Venetian government again addressed the reconstruction of San Marco and other buildings. Wood was imported from Venice specifically for the restoration of the church of San Tito. In 1333, the Senate approved plans for rebuilding the harbour at Candia. In 1336, funds were allocated to restore several castles considered essential to the defence of the island. Further expenditure followed in 1342 and 1362.
Remarkably, in 1363, sixty years after the earthquake, the Venetian Senate was still discussing repairs to fortifications damaged by the disaster.
The documentary record therefore reveals something that the dramatic tsunami narratives alone cannot show: the effects of the 1303 disaster lasted for generations.
Rebuilding Cairo and Alexandria
The damage was not restricted to Crete. In Cairo, several important religious buildings required restoration. Arabic historians record repairs to the mosques and minarets of al-Azhar, ‘Amr ibn al-‘As and al-Hakim, as well as other religious complexes.
At Alexandria, the lighthouse, city walls and other structures were repaired at the expense of the Sultan. The citadel and towers of Safad were also repaired in 1304.
The reconstruction itself created new economic pressures. Al-Maqrizi remarks that the demand for building materials in Cairo caused prices to rise dramatically. This is an important indication of how natural disasters can generate secondary economic effects: the destruction creates an urgent demand for labour and materials at precisely the moment when supply has been disrupted.
Dr Lorris Chevalier, who has a Ph.D. in medieval literature, is a historical advisor for movies, including The Last Duel and Napoleon. Click here to view his website.
Click here to read more from Lorris Chevalier
Further Readings:
Guidoboni, E. and Comastri, A., ‘The large earthquake of 8 August 1303 in Crete: seismic scenario and tsunami in the Mediterranean area’, Journal of Seismology (1997)
Top Image: Walters Art Museum W.106.3R
Subscribe to Medievalverse
Related Posts