Septoria and Stagonospora Diseases of Cereals - CIMMYT ...

wheat to increase grain size. The
six highest yielding lines derived
from this program outyielded their
breadwheat parent by 5-20% in
yield trials in Cd. Obregon, Mexico.
Shuttle breeding within
Mexico
Young and Frey (1994) provide
two factors that influence the
success of a shuttle program: a) the
use of a germplasm pool
encompassing genotypes with
broad adaptation, and b) the use of
selection environments eliciting
different responses from plant
types. They also state that the
wheat breeding program of N.E.
Borlaug met these conditions.
When Borlaug started the shuttle
breeding approach in 1945, his only
objective was to speed up breeding
for stem rust resistance. Since then,
segregating populations have been
shuttled 100 times between the two
environmentally contrasting sites
in Mexico, Cd. Obregon and Toluca
(Braun et al., 1992).
Some of the salient points of
this shuttle breeding program are:
• Cd. Obregon is situated at 28 o N
at 40 masl, in the sunny, fertile,
and irrigated Yaqui Valley of
Sonora. Wheats are planted in
November when temperatures
are low and harvested in April/
May when temperatures are
high. The yield potential of
location is high (±10 t/ha);
wheat diseases are limited to
only leaf rust, Karnal bunt, and
black point.
• The Toluca location is
characterized by high humidity
(precipitation: ±1000 mm). The
nursery is planted in May/June
when temperatures are high and
harvested in October when they
are low. High humidity causes
Historical Aspects and Future Challenges of an International Wheat Program 5
incidence of many diseases
including rust, septorias, BYD,
and fusarium.
An important result of shuttle
breeding was the selection of
photo-insensitive wheat genotypes.
Initially, selection for photoperiod
insensitivity was unconscious, but
only this trait permitted the wide
spread of the Mexican semidwarfs
(Borlaug, 1995). Today, this trait has
been incorporated into basically all
spring wheat cultivars grown
below 48 o latitude and is now also
spreading to wheat areas above 48 o
N (Worland et al., 1994).
Multi-locational testing
and wide adaptation
About 1500 sets of yield trials
and screening nurseries consisting
of around 4000 advanced bread
wheat lines are annually sent to
more than 200 locations. Multilocational
testing plays a key role in
identifying the best performing
entries for crossing. Since the
shuttle program permits two full
breeding cycles a year, it takes
around five to six years from
crossing to international
distribution of advanced lines to
cooperators. This “recurrent
selection program” ensures
continuous and rapid pyramiding
of desirable genes.
Ceccarelli (1989) pointed out
that the widespread cultivation of
some wheat cultivars should not be
taken as a demonstration of wide
adaptation, since a large fraction of
these areas are similar or made
similar by use of irrigation and/or
fertilizer. Therefore, the term wide
adaptation has been used mainly to
describe geographical rather than
environmental differences. If this is
true, the genotypic variation
should be considerably higher than
GxE interaction in ANOVAs of
CIMMYT trials. Braun et al. (1992)
showed that this is not the case.
When subsets of locations were
grouped by geographical and/or
environmental similarities, GxE
interaction was mostly greater than
the genotypic variance. The
environmental diversity of sites
where CIMMYT’s 21st
International Bread Wheat
Screening Nursery was grown and
the diversity among genotypes in
this nursery were demonstrated by
Bull et al. (1994). They classified
similarities among environments
by forming subsets of genotypes
from the total dataset and
comparing them with the
classification based on the
remaining genotypes. Using this
procedure they concluded that it
was not possible to form a stable
grouping of environments, because
little or no relationship existed
among them.
Conclusions drawn from trials
carried out on research stations are
always open to critics who argue
that these results do not necessarily
reflect conditions in farmers’ fields.
However, the wide acceptance of
CIMMYT germplasm by farmers in
MEs 1-5 does not support the view
that the wide adaptation of
CIMMYT germplasm is based on
geographical rather than
environmental differences.
Breeding for High Yield
Potential and Enhanced
Stability
Selection of segregating
populations and consequent yield
testing of advanced lines are
paramount for identifying high
yielding and input responsive
wheat genotypes. The increase in
yield potential of CIMMYT