Later Maya city-states

Two powers succeeded one another in the north, and then the north fell apart. Chichén Itzá, a city of causeways and cenotes with unmistakable central Mexican sculpture, dominated the peninsula from roughly the tenth century until about AD 1100. Mayapán, a walled city of perhaps 15,000 people, held the hegemony from the thirteenth century until a revolt in 1441 destroyed its ruling house. After that the peninsula broke into a patchwork of provinces that never again acknowledged one master, and it was in that condition that the Spaniards found it. In the highlands, the K'iche' and the Kaqchikel built militarised kingdoms with capitals at Q'umarkaj and Iximche', extracted tribute from conquered towns, and fought each other until Pedro de Alvarado arrived in 1524.

The Postclassic is where the oldest surviving Maya books were made and where the last independent Maya state survived. It is also the period in which the Maya stopped being the only actors in their own story: Mexican merchants, Mexican gods, Mexican military orders and finally Spanish soldiers all entered the peninsula, and the question of how much of Postclassic Maya culture was "Toltec" has been argued for more than a century. The answers matter beyond Yucatán, because the Spanish conquest of the Maya took longer than any other campaign in the Americas — from the first landfall in 1517 to the storming of Nojpetén in 1697, a span of 180 years.

At a glance

What did the Maya world look like when the southern cities fell?

The collapse that ended the Classic period was not an event but a long unravelling, and it happened at different speeds in different places. The last dated monument at Palenque was carved in AD 799; at Copán the last royal stela belongs to AD 822. Within two or three generations the great lowland capitals of the central Petén had stopped building, stopped carving dates, and lost most of their populations. Tikal, Calakmul and their rivals had been entangled in a system of marriage alliances, proxy wars and prestige rivalry that had no way of winding down gracefully, and when the largest centres faltered the smaller ones that depended on them went with them quickly.

The causes are still argued, and the argument has become more sophisticated rather than more settled. Warfare is visible in the archaeology — burned palaces, dismantled monuments, fortifications at places like Aguateca and Punta de Chimino. So is environmental stress: sediment cores from Lake Chichancanab in Yucatán, analysed by David Hodell and colleagues in 1995, showed a series of severe droughts in the ninth and tenth centuries, and later work has refined the picture into repeated multi-year dry spells rather than one terminal catastrophe. Deforestation, soil exhaustion and the sheer cost of maintaining monumental courts all contributed. The honest summary is that no single cause explains a collapse that took a century and a half and varied from valley to valley; what failed was a political system, not a people.

The collapse was also regional. The northern Maya lowlands did not fall. The Puuc cities of western Yucatán were entering their greatest building phase exactly as Palenque and Copán went quiet, and the north-coast settlements that would feed Chichén Itzá were growing. What collapsed was the particular concentration of power in the central Petén, and with it the ideological claim that a single great city could be the centre of the Maya world. The Classic Maya city-states had been a system of rivals with a shared court culture; the Postclassic produced something more commercial, more militarised and more widely distributed.

There is also the question of who moved where. Population did not vanish; it shifted. Refugees from the Petén appear to have gone north into Yucatán and south into the highlands, carrying ceramic habits, ritual practices and probably genealogies with them. The Itzá, who gave Chichén Itzá its name, are the clearest case of a group whose identity was remembered in both the northern capital and the Petén lakes four centuries later. That migration is one reason Postclassic Maya culture looks simultaneously continuous and transformed: the same calendar, the same gods, the same ballgame, arranged in a new political frame.

Why did the Puuc cities flourish while the south emptied?

The Puuc is a low range of hills south of Mérida, and its cities — Uxmal above all, with Sayil, Labná, Kabah, Xlapak and a dozen lesser centres — produced the most refined architecture the Maya ever built. The style takes its name from the Maya word pu'uk, "hill", and its hallmarks are a plain lower wall of carefully fitted limestone veneer topped by an elaborate frieze: geometric latticework, step-and-fret patterns, rows of Chaac masks with their long curling noses, and occasionally human figures. The effect depends on restraint. Nothing is painted, nothing is modelled in stucco; the ornament is cut into the stone and read in raking light.

Uxmal is the masterpiece and the best dated. The Nunnery Quadrangle, four palace buildings around a courtyard, was built between about 900 and 1000, and painted dates of AD 906–907 survive on vault stones, which anchors the whole Puuc florescence. The Governor's Palace, on its massive terrace, is one of the longest single façades in Mesoamerica; the Pyramid of the Magician, 27.6 m high, was rebuilt several times over earlier structures, so that its final form masks a longer sequence. There is a small ballcourt, a House of the Turtles, a Great Pyramid, and a sculpted phallus (a much-photographed curiosity) that says something about how nineteenth-century explorers chose what to notice.

The engineering was as impressive as the masonry, and less visible. The Puuc has no rivers and no cenotes in most of its area; the water table lies far below the surface. The cities depended on chultuns, bottle-shaped cisterns cut into the limestone to catch and store rain, and on aguadas, natural depressions enlarged and lined to hold water through the dry season. A lidar survey published by William Ringle, Tomás Gallareta Negrón and colleagues in 2021 mapped 237 square kilometres of the region and found terracing on hillsides — the first unequivocal evidence of it in northern Yucatán — together with quarries, lime kilns and a settlement density of about 42 basal structures per square kilometre. That is high for the Maya lowlands, though the survey also found the population nucleated rather than spread evenly, and no defensive works at all.

Then it stopped. Major construction at Uxmal ceased after about AD 1000, and the Puuc cities shrank through the eleventh century. The water system that made them possible was also their vulnerability: a run of dry years emptied the cisterns, and there was no river to fall back on. The political initiative passed east to Chichén Itzá, which had cenotes, a coast, and access to the salt pans. Uxmal was not abandoned outright — colonial tradition keeps it as a member of the Mayapán confederation and its ruling line, the Tutul Xiu, later governed from Maní — but it never again built at scale.

Why did Chichén Itzá become the dominant power?

Chichén Itzá sits in the dry eastern half of Yucatán, between Valladolid and Mérida, and its name is usually translated as "at the mouth of the well of the Itzá". Two great cenotes lie within the site, and a third nearby; in a landscape without rivers, that is the difference between a village and a city. Settlement began in the Late Classic, and the earliest architecture — the complex now called Chichén Viejo, where the Temple of the Initial Series carries one of the site's oldest dated lintels — belongs to the eighth and ninth centuries. The city that everyone means by Chichén Itzá, the one with the great plaza and the colonnades, was built mostly between about AD 900 and 1100.

Its monuments are on an imperial scale. El Castillo, the pyramid the Spaniards named after a castle, rises about 24 m to the top of its temple, with stairways of 91 steps on four sides plus the platform that makes 365; the effect is a calendar in stone, and at the equinoxes the afternoon light throws a shadow down the northern balustrade that the site now sells as a serpent. The Great Ballcourt is the largest in Mesoamerica, roughly 165 by 68 m, with temples at either end and a relief programme of decapitation and ballgame ritual. The Temple of the Warriors stands behind rows of square columns — the "Group of the Thousand Columns" — with a Chac Mool at its summit and serpent columns at the door. There is a skull rack, a circular observatory known as El Caracol, an ossuary pyramid, and more than eighty causeways, sacbeob, tying the complexes together.

The economy behind it was coastal and commercial. Chichén controlled the north coast through ports, above all Isla Cerritos, a small island anchorage off the Yucatán shore where excavation produced quantities of obsidian, basalt, ceramics and imported goods. Salt from the northern salinas was one of the great bulk commodities of Mesoamerica, and whoever controlled the pans controlled a trade that reached the Gulf coast and the highlands. Green Pachuca obsidian from central Mexico, copper bells from West Mexico, jade and gold all turn up at Chichén, which is what a capital of a trading system looks like in the ground.

Population estimates vary widely. Britannica gives 35,000 at its peak; other scholars have argued for 50,000 or more in the city and its immediate hinterland, which would make it one of the largest cities in the Americas at the time. What is not disputed is that the boom ended. Construction stopped around 1100, the monumental core was partly burned and looted at some point, and by the thirteenth century the city was largely abandoned as a political centre, though pilgrims kept coming to the cenote. The old story that Mayapán conquered it has little archaeological support: Mayapán's rise comes after Chichén's decline, not before it.

Was Chichén Itzá a Toltec city?

Walk into the Temple of the Warriors and you could be at Tula. There are serpent columns, Atlantean figures carrying altars on their heads, Chac Mools reclining with a bowl on the belly, a skull rack, benches with processions of warriors, and a feathered-serpent cult attached to the name Kukulkán, the Yucatec form of Quetzalcóatl. Nineteenth- and twentieth-century scholars read this as evidence of conquest: Toltec warriors, or Toltec-led groups from the Gulf coast, took Yucatán around AD 1000, installed a dynasty at Chichén, and imposed central Mexican political and religious forms. UNESCO's own description of the site still speaks of a fusion of Maya and Toltec civilisations, and the phrase "Maya-Toltec" is hard to dislodge from guidebooks.

The invasion model has problems. Tula's own chronology is not securely earlier than Chichén's great building phase, which makes the direction of influence uncertain at best. The supposedly Toltec traits are widely distributed in Late Classic Mesoamerica — chacmools appear at places with no Mexican connection, and serpent columns are a pan-regional fashion. Nor is there a clear horizon of destruction at Chichén that would mark a conquest. An alternative reading, argued forcefully by William Ringle, George Bey and Tomás Gallareta Negrón in 1998, sees the central Mexican forms as evidence of a religious and political movement that Chichén itself helped to create: a cult of the feathered serpent and of warrior orders that spread through the Maya lowlands and beyond, with Maya cities as producers rather than consumers of the style.

The safest position is that the sculpture is real and the ethnicity is not recoverable. "Toltec" in Mesoamerican usage could mean an inhabitant of Tula, but it could also mean something like "cultured" or "urban" — a title of prestige rather than a passport. The Toltecs of the Mexican plateau were certainly a model that later peoples, including the Aztecs, looked back to, much as Teotihuacán had been before them; whether soldiers from Tula governed Chichén is another question, and the evidence does not settle it.

A related argument concerns how Chichén was governed. In the 1990s Linda Schele, David Freidel and others proposed that the city was run by a multepal, a joint council of ruling lineages, rather than by a single king — an inference from the absence of a clear royal monument programme and from the multiplicity of colonnaded "palace" structures. That reading was influential and is still repeated, but it has been strongly criticised, and the current trend in Maya epigraphy has moved back towards individual rulers at Chichén. Colonial tradition, for what it is worth, remembered three brother lords who came from the west and ruled together in peace for some years before one died — a memory that fits either model.

What did the Sacred Cenote mean, and what did the dredging find?

The Cenote Sagrado is a near-circular collapse, about 60 m across, with sheer walls dropping some 27 m to the water. To the Maya it was a doorway: the surface of the earth was a crocodile's back, the sky was held up by four Bacabs, and the water below was the entrance to the underworld and the home of Chaac, the rain god. Offerings went in for centuries, and not only at Chichén. Colonial accounts describe droughts being met by processions to the well, with human victims thrown in and, if they survived the night, hauled out and questioned for prophecy — an origin story that the chronicles also attach, pointedly, to the rise of Hunac Ceel, the lord who later destroyed Chichén according to tradition.

The archaeological record is more prosaic and more revealing. Edward Herbert Thompson, the American consul in Yucatán, bought the hacienda that included the ruins in 1894 and dredged the cenote between 1904 and 1910, recovering jade, gold, copper, shell, wood, copal incense, pottery and human bone. The human remains show injuries consistent with sacrifice. Thompson shipped much of the material to the Peabody Museum at Harvard, and in 1926 the Mexican government charged him with theft and seized the hacienda; the Mexican Supreme Court eventually ruled against him, and he never returned to Yucatán. Later expeditions in the 1960s and after, supervised by the Instituto Nacional de Antropología e Historia, added considerably to the assemblage.

What the objects show is a shrine of regional and even international reach. There is gold work of Central American and Caribbean origin, jade from the Motagua valley, copal from the forests, and pottery from several regions — the accumulation of centuries of pilgrimage rather than the rubbish of one city. That is the most useful thing the cenote tells us about Postclassic politics: Chichén's sacred prestige outlived its political power by centuries, and later lords in Yucatán still wanted its blessing. The well was still being visited when the Spaniards arrived, and it remains the reason the site is a place of ritual as well as tourism.

What was the League of Mayapán?

Mayapán lies about 40 km south-east of Mérida, near the town of Telchaquillo, and it is the only Postclassic Maya capital that can be walked around in an afternoon and understood as a whole. The city is enclosed by a stone wall 9.1 km long, roughly oval with a pointed north-east corner, pierced by twelve gates, seven of them with vaulted entrances. Inside are about 4.2 square kilometres and more than 4,000 structures, most of them small residential platforms packed around courtyards and cenotes; the Carnegie archaeologist A. L. Smith estimated 10,000–12,000 people within the wall, and later survey work outside it has pushed the total to 15,000–17,000. That makes it the largest known Postclassic Maya city.

Its ceremonial core copies Chichén. There is a Castillo, colonnaded halls, serpent columns, and a cult of Kukulkán, all built at a smaller scale and in a poorer masonry. The imitation is the point: Mayapán was asserting succession to Chichén's authority. Excavations since 1996, reported by Susan Milbrath and Carlos Peraza Lope, have also shown that the city was cosmopolitan in its own right, with art motifs drawn from central Mexico and Oaxaca and ceramics and architecture shared with the east coast and Petén. Milbrath and Peraza Lope argue that Mayapán's foundation may predate the conventional date of 1263 — the end of a Katun 13 Ahau in the colonial chronicles — and that its earliest buildings are contemporary with Chichén's last.

The "League of Mayapán" is a modern phrase for a relationship the sources describe without using the word. Colonial tradition, above all Landa, says that the lords of Mayapán allied with the Tutul Xiu, who had come from the south, and that Uxmal and Chichén were at various times part of the same arrangement. Whether this amounted to a confederation with a shared foreign policy, or simply the habitual deference of lesser lords to a strong capital, is disputed. What is clear is that Mayapán extracted tribute, that its ruling family, the Cocom, grew rich on it, and that the city's wealth rested on a trade route running from the Gulf coast through the peninsula to the east coast and Petén.

Landa's account of the end is a political morality tale. The Cocom lord, greedy for wealth, invited Mexican garrisons from Tabasco and Xicalango into the city, promising to hand it over; he oppressed the commoners and made slaves of them. A later Cocom did the same again. The lords then gathered behind the Tutul Xiu and killed the Cocom and all his sons but one who was away, sacked his houses, and took his cacao orchards in compensation. Faction fighting continued, he says, until the city had stood more than 500 years and the lords abandoned it, each going back to his own country — a figure that would put the end in the eleventh century and shows how schematic his chronology is. The date usually given, 1441, comes from katun reckoning: the revolt falls in a Katun 8 Ahau, and the colonial chronicles date the fall of Chichén by the same katun name, which has confused readers ever since.

The archaeology supports a violent end in the right period. Burned roofs and floors have been dated to the mid-fifteenth century; a mass grave in the main plaza and a burial shaft with bodies covered in ash, some still with flint knives among the bones, point to killing inside the city rather than a siege. Drought and economic strain in the decades before the revolt probably made the Cocom's exactions intolerable. What is certain is that Mayapán was not reoccupied as a capital. Its population dispersed, and the chronicles say the lords took the books of their sciences with them when they went.

What did the Maya do after Mayapán fell?

The answer, for about eighty years, was fight each other. Yucatán fragmented into roughly sixteen provinces, the kuchkabaloob, each with its own lord and capital: Maní under the Xiu in the south-west, Sotuta under the surviving Cocom, Hocaba, Ah Canul around Calkiní, Ceh Pech in the north, Cupul around Valladolid, Cochuah in the south-east, Chikinchel on the salt coast, and others along the eastern shore. Landa, writing in the 1560s, lists hurricanes, plagues and wars among the calamities of the century before the conquest. There was no hegemon, no supreme court, and no mechanism for settling disputes except war and marriage.

The material culture of these provinces is thinner than Mayapán's but not poor. The east coast flourished: Tulum, built on 12-metre cliffs above the Caribbean, is a walled port with a small Castillo, a Temple of the Frescoes and a Temple of the Descending God, built and rebuilt between about 1200 and 1450, and still inhabited when the Spanish first sighted it in 1518. Cozumel was a pilgrimage island; Xelhá, Xcaret and El Meco were ports; Chetumal was a trading town at the head of a bay. Canoes moved salt, cotton, cacao, obsidian, copper and slaves along the coast, and the same routes connected Yucatán to the Gulf coast entrepôts and to the Petén lakes.

Inland, the provinces were agrarian and fiercely particularist. Landa describes a society of lords, priests, merchants and farmers, with slaves at the bottom, in which each town had its own chief and the great lords lived in Maní, Sotuta, Valladolid and a handful of other seats. Justice was local, warfare was endemic, and the religious calendar was still kept. It was this fragmentation that the Spaniards exploited: Francisco de Montejo the Younger allied with the Xiu of Maní against the Cocom of Sotuta, and the conquest of Yucatán was achieved by playing the provinces against each other rather than by defeating a single enemy.

The highland story ran in parallel. There too the Classic cities had declined, and there too new powers emerged in the eleventh and twelfth centuries, drawing on Mexican models, trade and migration. The difference is that the highlands had denser populations, better soils and more valuable export goods — obsidian, jade, cacao from the Pacific slope, salt from the interior — and the states that grew there were correspondingly larger and more centralised than the Yucatec provinces.

How did the highland kingdoms work?

The K'iche' kingdom, with its capital at Q'umarkaj (the Spaniards called it Utatlán), was the strongest state in the Guatemalan highlands in the fifteenth century. Its own histories, preserved in the Popol Vuh and the Título de Totonicapán, describe ancestors coming from a place called Tulan Zuyua in the west, crossing the sea, and receiving the gods before founding a series of capitals. Those narratives are charter myths as much as chronicles: they were written down in the sixteenth century, in Latin script, by nobles with land titles to defend. Archaeology shows a real movement of people and ceramic traditions from the Gulf coast and the Maya lowlands into the highlands in the Late Classic and after, but it does not match the migration stories in detail.

Q'umarkaj was founded around 1400 by a ruler called Q'uq'umatz, "feathered serpent", on a plateau north of the earlier capital at Pismachi', surrounded by deep ravines. It was a compact royal city of temples, palaces and ballcourts, dominated by the temple of Tohil, the patron god whose worship the K'iche' carried with them. The kingdom's ruling structure was composite. Power was shared between the ajpop, the "lord of the mat", and the ajpop k'am ja, a second title of near-equal rank, and below them stood the heads of the great lineages — the Kaweq, Nijaib and Ajaw — each with its own territory and retainers. The K'iche' expansion is associated above all with K'iq'ab, who ruled from about 1425 to 1475 and pushed the kingdom's frontiers onto the Pacific slope and into the Rabinal valley, and made tributaries of the towns along the way.

The Kaqchikel had been junior partners in that expansion. Their own annals describe service in K'iche' wars, then a growing estrangement as Kaqchikel power increased. Around 1470 — the date is debated, with some scholars preferring as late as 1485 — the Kaqchikel left Q'umarkaj and founded Iximche' on a defensible ridge almost surrounded by ravines, in what is now Chimaltenango. The break was followed by decades of war between the two peoples, which is the single most important fact about the highlands when the Spaniards arrived: the two strongest states were enemies, and both were willing to use foreigners against the other.

The other highland powers were smaller but real. The Mam held the western highlands around Zaculeu, near Huehuetenango, and had been the K'iche's chief rivals before K'iq'ab broke them. The Tz'utujil held the Lake Atitlán basin, with their own capital and a substantial population. The Q'eqchi' and the Poqomchi' occupied the wetter country to the north. All of these were Maya states with Maya institutions: they kept the 260-day calendar, worshipped a pantheon headed by sky and rain deities, built ballcourts, and organised labour through lineage obligations. What distinguished them from the Yucatec provinces was scale and density — the highlands held a large peasant population packed into a small area of terraced and irrigated land.

What did they trade, and who carried it?

Landa's brief chapter on "industry, trade and money" is one of the most valuable passages in the whole Relación, because it describes an economy rather than a ritual. The occupation to which the Maya were most inclined, he wrote, was that of merchants, who carried salt, cloth and slaves to the land of Ulúa and Tabasco and traded them for cacao and stone beads — their money. Cacao beans served as small change; strings of red shell beads served as larger denomination and as jewellery; and slaves could be bought with either. Credit existed, debts were paid courteously, and — Landa notes with some surprise — without usury.

That passage maps onto a real network. Salt came from the north-coast salinas and from interior springs; it was the one bulky necessity that the highlands and the Petén could not produce for themselves, and salt cakes circulated as trade goods as well as condiment. Cacao came from the hot Pacific slope of Soconusco and from the Tabasco lowlands; it did not grow well in Yucatán, which made it a permanent import and, in the Aztec tribute lists, a strategic one. Cotton cloth, obsidian from the highland sources at El Chayal and Ixtepeque, jade from the Motagua, copper bells and axe-shaped copper currency from Michoacán and the west of Mexico, feathers, and slaves made up the rest of the cargo.

The carriers were professionals. On the Gulf coast, Chontal Maya merchants — often called Putun in the literature — ran canoe fleets between the Laguna de Términos, Xicalango and the Yucatán ports, and their descendants were still trading when Cortés crossed the region in 1525. In the highlands, the K'iche' and Kaqchikel extracted tribute in cacao, cotton, salt and precious feathers from subject towns, and their nobles traded on their own account. From the north came the Aztec Triple Alliance's professional merchants, the pochteca, who operated at Xicalango and the Gulf entrepôts, dealt in cacao, feathers, jade and slaves, and doubled as intelligence-gatherers for the Mexica state. They did not rule Yucatán, but their demand shaped what Yucatán produced.

Two consequences follow. First, the Postclassic Maya economy was monetised to a degree the Classic economy was not: cacao beans and shell beads are money, and money implies prices, credit and markets. Second, the political map of Yucatán after 1441 is best read as a map of trade routes. The provinces that mattered — Chikinchel with its salt, Chetumal with its bay, Cupul with its interior roads, Maní with its cacao and its access to the Gulf — were the ones that controlled a segment of the route. When the Spanish cut those routes, the provinces lost the revenue that sustained them, and the political system collapsed faster than the conquest alone would suggest.

How did the Spanish conquest end the Maya world?

The conquest of the Maya was not one campaign but a series of them, spread over two centuries. The first contact came in 1517, when Francisco Hernández de Córdoba's expedition was blown west from Cuba and fought a battle at Champotón; Juan de Grijalva followed in 1518; Cortés touched the coast in 1519 on his way to Mexico. Two Spaniards from an earlier shipwreck, Gerónimo de Aguilar and Gonzalo Guerrero, were living among the Maya — Aguilar became Cortés's interpreter, Guerrero fought for the Maya and refused to leave. None of this amounted to conquest; it amounted to reconnaissance and the transmission of smallpox, which reached Yucatán in the 1520s and killed a large share of a population that had no immunity.

The highlands fell first, and quickly. Pedro de Alvarado left Mexico in December 1523 with Spanish cavalry, infantry and a large force of Tlaxcalan and other Mexican allies. He defeated the K'iche' army in the Quetzaltenango valley in February 1524, in fighting that killed Tecún Umán, one of the four lords of Q'umarkaj; the Kaqchikel annals give 20 February as the date, though the sources disagree about whether he died at El Pinar or at Xequikel later the same day, and Alvarado's own report says only that a captain-general was killed. Alvarado then entered Q'umarkaj, was received by the surviving lords, seized the ajpop Oxib-Keh and the ajpop k'am ja Beleheb-Tzy, burned them, and burned the city. Iximche' submitted and supplied auxiliaries; Zaculeu, the Mam capital, was taken in 1525 after a siege; the Tz'utujil of the Atitlán basin were brought under Spanish control in the same years. The Kaqchikel, having seen what alliance cost, rose in 1524 and were not finally defeated until 1530.

Yucatán took much longer. Francisco de Montejo's first attempt, from 1527, failed; a second, in the 1530s, also failed, and the Spaniards withdrew. The third campaign, from 1540, succeeded because Montejo's son, acting for the Spanish crown, allied with the Xiu of Maní against the Cocom of Sotuta, and because the provinces were exhausted by their own wars. Valladolid and Mérida were founded in the early 1540s, and in 1546 the eastern provinces rose in a coordinated revolt that was put down in a single battle. Even then the north-west was governed through Maya lords, under a Spanish veneer, for the rest of the century. In 1562 Diego de Landa, then head of the Franciscan mission in Yucatán, held an auto de fé at Maní and burned 27 Maya books, by his own count, along with thousands of images — an act that destroyed most of the surviving Maya literary heritage and, incidentally, preserved the memory of it in his own Relación.

The last kingdom held out in the Petén. The Itzá, who had given Chichén Itzá its name and whose own tradition traced them back to that city, had built their capital at Nojpetén, on an island in Lake Petén Itzá; Cortés passed through in 1525 and left a lame horse that the Itzá treated as a god. Franciscan missions in 1618 and again in the 1620s were expelled, one of them violently. It took a military expedition under Martín de Ursúa y Arizmendi, with 235 Spanish soldiers, 120 Maya labourers and a large contingent of Yucatec allies, to bring it down: Nojpetén was stormed on 10 March 1697 from an oar-powered assault boat, and the last independent Maya state ceased to exist. The Itzá kingdom of Petén had outlasted Tenochtitlán by 176 years.

Common myths about the Later Maya city-states

Timeline of the Later Maya city-states

What the map shows next

At AD 900 on the map the drawn shape is the northern lowlands at the moment the southern capitals were emptying: Chichén Itzá is building at scale, the Puuc cities are still inhabited, and the peninsula has not yet been unified by anyone. The highland kingdoms described above lie outside this polygon, further south and west, and the map does not draw them.

At 1250 on the map the same ground belongs to Mayapán, and by 1441 on the map it has fragmented into provinces that the map can only show as one shape. At 1524 on the map the northern polity is still drawn, although its real authority had ended eighty years earlier and the Spaniards were then burning Q'umarkaj in the highlands. The successor in the forests is the Itzá kingdom of Petén, drawn separately from 1525, which is where the story continues until 1697.