57long-range type. The fi rst type prevents too early and untimely germination—an
important feature, pre-adapted to agriculture, particularly in view of the fact that seeds
were often kept under unsuitable storage conditions. The second type of dormancy
ensures in nature a temporal distribution of germination: it is invariably the larger grain
of the second fl oret in each spikelet that germinates in the fi rst year, while the smaller
grain of the fi rst fl oret germinates in the second year. However, since this dormancy is
induced by the palea, it was overcome by threshing (unpublished data).
The loss of self-propagation and of the temporal control of germination was fol-
lowed by the loss of self-protection. The wild forms have tightly-closed glumes
resulting in a “hulled” grain after threshing. Several primitive forms of domesticated
wheat species, e.g., T. monococcum subsp. monococcum, T. turgidum subsp. dicoc-
con , and T. aestivum subsp. spelta, macha, and vavilovii, retain this feature. Hence,
the appearance of naked kernels was the second most important step in domestica-
tion of tetraploid wheat after nonbrittle spikes. How did the free-threshing tetraploid
wheat evolve from dicoccon? Did it get its naked kernels directly, through mutations
in the genes that control glume stiffness (the Q factor and the Tenacious Glumes
( Tg ) locus) or, was there another intermediate step? Triticum turgidum ssp. parvi-
coccum, a tetraploid wheat “fossil” species (Kislev 1980 ) was relatively abundant in
the archeological record starting already 9000 years BP, but disappeared ~ 2000
years BP, had a compact spike and was free-threshing and thus probably already
contained the Q and tg mutations prior to durum (Feldman and Kislev 2007 ). This
raises the possibility that durum received these mutations from parvicoccum , rather
than evolving them independently from emmer wheat. Durum may thus derive from
hybridization between parvicoccum and dicoccon receiving the free-threshing trait
from parvicoccum and the large grains from dicoccon. The large grain of durum was
probably preferred to the small grains of parvicoccum that lead to the prominence
of durum as a tetraploid wheat and to the extinction of parvicoccum. Similarly, it
can be assumed that common wheat, T. aestivum subsp. aestivum , received these
mutations from tetraploid wheat but required additional mutation in the Tg gene of
genome D (Kerber and Rowland 1974 ).
The Q gene, located on the long arm of chromosome 5A, is one of the most
signifi cant domestication loci as it controls the free-threshing character and several
other domestication-related traits such as glume shape and tenacity, rachis fragility,
spike length and shape (square-head spike), plant height, and spike emergence time.
While the 5A homeoallele has the most signifi cant contribution, other homeoalleles
(on 5B and 5D) were also shown to be involved in the domestication traits (Zhang
et al. 2011 ). The Q gene is the only one that was so far characterized at the molecu-
lar level. Simons et al. ( 2006 ) isolated the Q gene and verifi ed its identity by analy-
sis of knockout mutants and transformation. The Q allele is more abundantly
transcribed than q , and the product of the two alleles differs for a single amino acid
at position 329, isoleucine in the Q protein and valine in the q protein. The q allele
is the more primitive, and the mutation that gave rise to Q occurred only once lead-
ing to the world’s domesticated wheat. Q is thought to be a major regulatory gene
for fl oral development (Muramatsu 1986 ). It encodes an AP2 -like transcription
factor that played an important role in the domestication of polyploid wheat.
(Chantret et al. 2005 ; Simons et al. 2006 ; Zhang et al. 2011 ).
2 Origin and Evolution of Wheat and Related Triticeae Species