56
outcrossing events. While in its natural habitats this grass occupy poor, thin, rocky
soils, it responded well when transferred to richer habitats.
Domestication imposed a new evolutionary direction whereby traits that had the
greatest adaptive value in the cultivated fi eld were preferred. Selection pressures
have operated in a different, and sometimes, contradictory manner in cultivation and
in the wild. During the 10,000 years of wheat cultivation, the criteria for selection
varied from time to time and from place to place. Wheat cultivars have been devel-
oped through three main phases of selection: occasional and sometimes
nonintentional selection, exerted by the earliest farmers simply by the processes of
harvesting and planting; more deliberate selection by traditional farmers in poly-
morphic fi elds; and selection as part of scientifi cally planned modern breeding.
During the fi rst phase, a very signifi cant sequence of changes occurred in the
transition from the wild into the domesticated forms. Several wild characters that
had no pre-adaptive value for cultivation were selected against. These characters
included the wild seed dissemination, seed dormancy, and seed protection by the
tightly closed glumes.
Wild wheat has a brittle spike that upon maturity disarticulates into arrowhead-
shaped spikelets. While these seed-dispersal units facilitate self-burial in the soil
(hence protection during the long, dry summer and successful germination after the
fi rst rain), they must have proved a nuisance to the ancient farmer who had to collect
most of the spikelets from the ground or cut the culms before the grains matured. No
wonder, therefore, that those plants with brittle heads were selected against. Yet, it took
about a millennium or more until mutants for nonbrittleness appeared in the cultivated
fi elds and gradually became the dominant crops (Feldman and Kislev 2007 ).
Wild emmer was domesticated through the loss of spike fragility. This step was
probably a gradual process as suggested from both genetic and the archeological
evidence. The archeological record shows that in ancient sites where agriculture
was practiced, a mixture of fragile and nonfragile types were found, and it took 3–4
thousand years until the nonfragile spikes became prominent in farming units
(Kislev 1984 (in emmer wheat); Tanno and Willcox 2006 (in einkorn)). Two major
genes, brittle rachis 2 ( Br-A2 ) and brittle rachis 3 ( Br-A3 ) located on the short arms
of chromosomes 3A and 3B, respectively, control the brittleness of the rachis in
wild emmer (Levy and Feldman 1989a ; Watanabe and Ikebata 2000 ; Watanabe
et al. 2002 , 2005 ; Nalam et al. 2006 ; Millet et al. 2013 ); in addition, another locus
for spike brittleness was mapped to chromosome 2A (Peng et al. 2003a , b ; Peleg
et al. 2011 ). Comparative mapping analyses suggest that both Br-A2 and Br-A3 are
present in homoeologous regions on chromosomes 3A and 3B, respectively.
Furthermore, Br-A2 and Br-A3 from wheat and Btr1/Btr2 on chromosome 3H of
barley ( Hordeum vulgare L.) also are homoeologues suggesting that the location of
major determinants of the brittle rachis trait in these species has been conserved
(Nalam et al. 2006 ). The loss of fragility gave rise to the fi rst known domesticated
wheat, Triticum turgidum ssp. dicoccon , or domesticated emmer wheat, which is
grown to this day, albeit on a small scale (De Vita et al. 2006 ).
The second requirement for the newly domesticated wheat was uniform and rapid
germination. Wild wheat exhibits two types of dormancy: a post-harvest type and a
M. Feldman and A.A. Levy