American maize in the Canary Islands: the Columbian exchange measured in bone chemistry
The Canaries are usually a waypoint in this story - the last watering stop before the Atlantic. Isotopes in 66 skeletons from Gran Canaria turn the islands into the measuring instrument instead. Before the conquest the diet was Mediterranean barley and wheat; by the eighteenth century American maize supplied about two thirds of the carbon in island childhoods, and the granary returns agree. The same record holds the people the change was built on.
Why this wins its question: Everyone writes that American crops transformed Old World diets; almost nobody can say by how much, or when. Here it is quantified in one place with two independent instruments: collagen carbon values moving from about -19.4 per mil before the conquest to -13.8 per mil by the eighteenth century, C4 foods rising from nothing to 41 and then 64 per cent of childhood intake, and granary returns showing maize at 42,554 bushels in 1789 and overtaking wheat by 1813. The Canary Islands are the hinge of the Atlantic exchange and are usually skipped in one line. This object also refuses to leave the ledger half-written: the enslaved North African and sub-Saharan workers of the Finca Clavijo sugar mills are in the same isotopic series, and the ancient-genome evidence puts the indigenous share of modern Canarian ancestry at 17.8 per cent.
Claims
Every assertion below is bound to registered sources and carries its own confidence. Weight them; do not treat the page as uniformly authoritative.
The measurement is direct. A 2025 study analysed 66 human skeletal samples and 42 animal bones from Gran Canaria dated to the sixteenth to eighteenth centuries for carbon and nitrogen stable isotopes, and compared them with 53 individuals of the indigenous period, sixth to fifteenth centuries.
The pre-conquest baseline is a C3 Mediterranean diet: the indigenous population relied on a short set of C3 cereals, especially barley and wheat, with collagen carbon values averaging about -19.4 per mil.
The colonial series then moves, and the direction is American. Mean collagen carbon values run about -17.8 per mil in the sixteenth century (15 individuals), about -17.3 per mil across the sixteenth to eighteenth centuries (23 individuals) and about -13.8 per mil in the eighteenth century (28 individuals); the calculated proportions of C4 resources consumed during childhood rise from 41 per cent to 43 per cent and then to 64 per cent.
The written record moves in step. Maize cultivation on Gran Canaria gradually replaced wheat, with production reaching 42,554 bushels by 1789; by 1813 maize at 98,708 bushels had passed wheat at 57,527. Two independent kinds of evidence - the granary counts and the chemistry of the people eating it - describe the same substitution.
Maize did not arrive alone. The Kingdom of Castile conquered the archipelago in the late fifteenth century and introduced new livestock - cattle, horses, donkeys - and agricultural techniques, particularly animal-driven ploughs; by the eighteenth century elevated nitrogen values indicate intensive manuring, with an offset between herbivores and local forage pointing to the use of local animal manure as fertiliser. The crop transfer travelled inside a change of farming system.
The same cemeteries record who the new economy ran on. The sixteenth-century burial ground of Finca Clavijo holds the remains of North African and sub-Saharan people who were made to labour as enslaved workers in the sugar mills from the late fifteenth century onward; their low rate of dental caries, 15.6 per cent, suggests they consumed sugarcane byproducts rather than the C4 crops directly. The sugar that made the islands profitable is in the same isotopic record as the maize that fed them.
Diet also separated by station. Burial sites differ isotopically, with the population of the Convento de San Francisco showing a more diverse diet than other urban centres, which the authors read as different social backgrounds. A staple shift measured across a whole island still did not reach everyone in the same way.
The people whose diet forms the baseline were not a small prehistoric footnote. Genome-wide data from 40 newly sequenced ancient individuals, with nine previously published, from sites on all seven islands and dated between the third and sixteenth centuries, confirm an indigenous population of North African Berber origin - their ancestry modelled as largely Late Neolithic Moroccan, with the basal North African Y-chromosome lineage E-M183 predominant among the men - who first settled the archipelago between about the second and fifth centuries.
What the conquest did to them is legible in living genomes. Present-day Canarians are modelled as about 79.7 per cent Spanish, 17.8 per cent indigenous and 2.6 per cent sub-Saharan African. The dietary transformation this object documents happened on islands whose original population had been reduced to a minority share of the ancestry of their own descendants.
The indigenous islanders were already living close to the edge of viability before Castile arrived, which is part of why the collapse went so far. Effective population sizes estimated from the ancient genomes are small across the archipelago - about 411 overall, from 460 on Gran Canaria down to 75 on Fuerteventura - and El Hierro passed through a strong bottleneck between the seventh and twelfth centuries, well before European contact. Conquest fell on populations that genetic drift had already thinned.
The authors' own framing is globalisation rather than empire: the transition reflects broader agricultural innovation and is driven by transatlantic trade networks linking Europe, Africa and America. That reading is theirs, it is well supported by their data, and it is worth noting that the three continents named entered those networks on very different terms - one shipping crops, one shipping people.