Agriculture Reference
In-Depth Information
they have a suffi cient number of environments
(e.g., location-years) to adequately represent the
target environment. In the authors' experience, at
least 30 environments (location-years) over at
least 2 years are needed to estimate line
performance.
In reviewing line performance data across envi-
ronments, the most troublesome statistical effect
is G × E interaction caused by crossover interac-
tions (e.g., Line A is better than Line B in some
environments and Line B is better than Line A in
other environments; Baker 1988; Crossa et al.,
1993; Russell et al., 2003). At fi rst glance having
lines perform better in one set of environments
than in other environments seems nothing more
than making a recommendation for different
target environment, which is routinely done.
However, the diffi culty arises when a location in
different years (hence different environments)
clusters with a different target environment. For
example, a drought-stressed location might be
clustered with low-yielding environments,
whereas the same location with more rainfall will
be clustered with medium- or higher-yielding
environments. While it is important to under-
stand the underlying environmental causes of the
G × E interaction (Cullis et al., 1996b; Basford
and Cooper 1998), it remains largely impossible
to predict future environmental conditions.
Perhaps the importance of understanding the
environment and developing lines for broad adap-
tation is best exemplifi ed by the shuttle breeding
program of CIMMYT (Ortiz et al., 2007), which
has global responsibilities for maize and wheat
improvement. As such, CIMMYT has parti-
tioned the world into megaenvironments and,
within those megaenvironments, has clustered
similar environments that share common attri-
butes such as high night temperatures. The
megaenvironments and the subclusters within
megaenvironments are useful for directing germ-
plasm exchange and evaluation, and they reduce
the problems of crossover interactions within
subclusters.
Initially the shuttle breeding program for wheat
improvement was designed for rapid generation
advance, but by using two diverse locations in
Mexico, the program also selected for lines which
were photoperiod insensitive and were exposed to
greatly different diseases and soil types. These
lines could be widely grown and became the wheat
cultivars of the Green Revolution. As the impor-
tance of the international effort grew, so did the
need to understand their international target
environments and G × E interactions. It also
highlighted that the original shuttle breeding
effort needed fi ne-tuning. Ortiz et al. (2007) pro-
posed an updated and expanded shuttle breeding
program, which includes a global effort to breed
wheat cultivars that are evaluated and can be suc-
cessfully grown in “hot spot” regions for various
biotic and abiotic stresses. Coupled with the
extensive international testing program are
sophisticated statistical analyses and physiological
assays to effi ciently characterize their germplasm.
Due to its broad adaptation and excellent disease
resistance, the CIMMYT germplasm is viewed as
a global genetic resource and treasure.
BREEDING HYBRID WHEAT
This section will provide a brief update on the
review by Edwards (2001). While hybrid wheat
research was largely conducted in the private
sector but has been reduced in developed coun-
tries, hybrid crops and especially hybrid cereals
that are normally self-pollinated continue to be
researched in developing and emerging econo-
mies (e.g., hybrid rice,
Oryza sativa
L., in Asia,
Food and Agriculture Organization 2004; and
hybrid wheat in India, Matuschke et al., 2007). In
these countries, the adoption has been spectacular
for rice (15,000,000 ha in China and 200,000 ha
in India in 2001-2002) and less so for wheat
(25,000 ha in India in 2005). The economic ben-
efi ts of hybrid wheat were found to be farm-size
neutral, with small landholders benefi ting as well
as or better than larger landholders (Matuschke
et al., 2007). Access to information and credit
were critical to the adoption of hybrid wheat tech-
nology. The key for developing or emerging
economies in developing hybrid wheat is that they
seem to be more patient with the needed long-
term investment that will successfully bring
hybrid wheat to market.
