Domestication of plants and animals is the major factor underlying human civilization. Cultivated wheats refer mainly to two types: the hexaploid bread wheat (Triticum aestivum) accounting for about 95% of world wheat production, and the tetraploid durum wheat (T. durum) accounting for the other 5%. T. aestivum derived from a cross between domesticated emmer T. dicoccum and the goat grass Aegilops tauschii, which most probably originated in the south or west of the Caspian Sea about 9,000 years ago. T. dicoccoides, the wild emmer wheat, is the progenitor of cultivated wheats, has the same genome formula as durum wheat and has contributed two genomes to bread wheat, and has played a core role to wheat domestication. This process of wheat domestication fits the gradual and multi-site model rather than the fast and single-site model. Domestication has genetically not only transformed the brittle rachis, tenacious glume and non-free threshability, but also modified yield and yield components in wheat. Wheat domestication is only involved in a limited number of chromosome regions, or domestication syndrome factors, though many relevant quantitative trait loci were detected. The available crop genome sequences and genome sequencing of wheat can transform today’s biology, dramatically advancing both theory and application of wheat domestication study. The nonrandom adaptive processes and complexes in T. dicoccoides and other wheat relatives could provide the basis for wheat improvement as single genes, QTLs, and interacting biochemical networks. Genome sequencing of diploid wild wheat, either T. urartu or Ae. tauschii, could be helpful for isolation of domestication syndrome factors and other relevant genes. The distinct adaptive complexes of T. dicoccoides to environmental stresses is of great importance for improvement of bread wheat.