261wheat (Triticum durum Desf.) yield potential and its physiological determinants has
been widely studied in many reports (Motzo et al. 2005 ; Giunta et al. 2008 ; Royo
et al. 2007 ) and a genetic gain from 10 to 50 kg ha−^1 year−^1 has been recorded over
the last century in most countries and often associated with few key genes affecting
morpho-phenological traits, mainly Rht (Slafer et al. 1994 ). Some reports have sug-
gested that the improvement for yield potential may have detrimental implications
for yield stability in water-limited environments (Ceccarelli et al. 1991 ; Mohamamdi
and Amri 2013 ), since specific drought tolerant traits (i.e. flowering time, small flag
leaves, glaucousness, maintenance of green leaf area, plant height) have been con-
sidered inappropriate for achieving high yields under favorable conditions (Acevedo
et al. 1991 ). On the other hand, it has also been suggested that under mild drought
conditions, characterized by a wheat/barley grain yield between 2 and 5 t ha−^1 ,
selection for high-yield potential had frequently led to some yield improvements
under drought (Araus et al. 2002 ; Rizza et al. 2004 ). These results were achieved
through an empirical selection for high-yield potential and high-yield stability, with
the latter being attributed to a minimal genotype × environment (GE) interaction.
This implies that traits maximizing productivity normally expressed in the absence
of stress can still sustain a significant yield improvement under moderate drought
(Slafer et al. 2005 ; Tambussi et al. 2005 ).
3 Traits of Importance for Base-Broadening and Stress
Resistance
All traits of interest for crop productivity, adaptation and quality are subjected to
improvement with the appropriate combination of current and new introgressed
alleles. However, due to the genetic bottlenecks and the fact that the most of breed-
ing programs in the past usually exposed their segregating generations during selec-
tion to very specific conditions, this can result in genetic diversity of some specific
traits may be limited. For instance, International Wheat and Maize Improvement
Center (CIMMYT) has based their approach for wide adaptation on selection in
environments with high potential (Braun et al. 1997 ). This approach has been very
successful and has led to yield improvements under near-optimal conditions and
less limiting environments (Araus et al. 2002 ; Trethowan et al. 2002 ). However, due
to the intense selection under near-optimal conditions some alleles present in the
landraces and cultivars prior to the green revolution that could represent an adaptive
advantage under specific environmental conditions may have not been selected and
therefore in low frequency or even not present in the modern genetic pool. Moreover,
crop landraces and wild relatives have evolved during thousands of years under the
most diverse stress scenarios and therefore may carry novel specific genetic combi-
nations due to natural selection and founder effect. Alleles present in landraces and
wild relatives providing specific adaptation to environments under drought, heat,
salinity, and disease stress pressures are being tackled down and incorporated in the
modern gene pool.
Breeding and Genetic Enhancement of Dryland Crops