When half is more than the whole: Wheat domestication syndrome reconsidered.

Two opposing models currently dominate Near Eastern plant domestication research. The core area-one event model depicts a knowledge-based, conscious, geographically centered, rapid single-event domestication, while the protracted-autonomous model emphasizes a noncentered, millennia-long process based on unconscious dynamics. The latter model relies, in part, on quantitative depictions of diachronic changes (in archaeological remains) in proportions of spikelet shattering to nonshattering, towards full dominance of the nonshattering (domesticated) phenotypes in cultivated cereal populations. Recent wild wheat genome assembly suggests that shattering and nonshattering spikelets may originate from the same (individual) genotype. Therefore, their proportions among archaeobotanical assemblages cannot reliably describe the presumed protracted-selection dynamics underlying wheat domestication. This calls for a reappraisal of the "domestication syndrome" concept associated with cereal domestication.