We've been thinking a lot about maize domestication lately and I'm happy to share some new research. People started moving maize around the Americas while domestication was still happening. science.sciencemag.org/content/362/64…
5,300 year old genomes from modern-day Mexico show domestication was still in progress, but people were growing maize in S. America from ~6500 bp onward, how do we reconcile that? cell.com/current-biolog… and pnas.org/content/113/49…
It looks like the earliest S. American maize was only semi-domesticated. The founder population had all the right building blocks, but needed more time and selection to fix the domestication syndrome. That's to say, maize populations became stratified early in the process. 
The southwestern Amazon was an ecological and cultural incubator for continuing the domestication process, probably because it was already a domestication hotspot. journals.plos.org/plosone/articl…
South American maize is *incredibly* diverse, with thousands of traditional landraces. New genomic data include a major lineage in lowland Brazil, which probably expanded out of the SW Amazon around 4kya and has been grown across Brazil ever since. (photo: Fabio Freitas)
There's also maize in eastern Brazil that REALLY looks like maize in the Andes. Linguistic and archaeological evidence points to people spreading this biodiversity across the continent around 1kya.
