Agriculture Reference
In-Depth Information
landraces or populations are heterogeneous, pure-
line selections are often considerably different
than the original parental source.
The designation between mass selection and
pureline selection is often blurred in practice. An
example was provided by the release of 'Redland'
wheat (Schmidt et al., 1989). The parent cultivar,
Brule (Schmidt et al., 1983), was heterogeneous
for its reaction to stem rust (caused by
Puccinia
graminis
Pers.: Pers. f. sp.
tritici
Eriks. & E.
Henn.). Hence 100 lines were sampled from Brule
and their progeny grown in the fi eld for 2 years.
After selection for phenotypic and stem rust
resistance uniformity, 24 lines were composited
to form Redland. Hence, in this example, pure-
line selection was used to identify individual lines,
but to regain a portion of the heterogeneity of
Brule, 24 lines (as opposed to only one line in
traditional pureline breeding methods) were com-
bined to form Redland. As might be expected,
Redland was slightly different from Brule (4 cm
shorter and 0.5 day earlier, Schmidt et al., 1989).
As with mass selection, pureline selection is most
commonly used today to purify or select a better
line from an advanced line.
However, most breeding methods begin with
hybridizing two or more parent lines. Following
the hybridization, each heterozygous gene pair
will be reduced in heterozygosity by 50% for each
generation of inbreeding (selfi ng). It should
be remembered that wheat is a naturally self-
pollinated crop and that, with the rare exception
of natural outcrossing, the progeny are created by
selfi ng. With
y
heterozygous gene pairs, the
proportion of homozygous individuals after
w
generations of self-fertilization is expressed math-
ematically as
[(2
w
− 1)/2
w
]
y
. By the F
6
generation
of a biparental mating with 10 gene-pair differ-
ences, 73% of the lines will be homozygous for
some combination of the 10 gene-pairs. Allard
(1960) illustrated graphically the rates of fi xation
of loci as
y
and
w
change.
The plant breeder practices some form of selec-
tion during each generation of inbreeding until
satisfi ed that the selected materials will constitute
a distinct, uniform, stable improved cultivar
(required by many seed laws and certifi cation
standards). Various methods of selection during
inbreeding will be discussed. A key feature of
breeding is that different methods of selecting
(breeder selection or natural selection) may be
applied at various generations to meet objectives
and resources available. Breeder selection is an
active process in which the breeder selects the
type of plants or families that he or she wants.
Natural selection is done by nature, though careful
selection of environments by the breeder can
optimize natural selection.
Pedigree selection
Pedigree selection is a method of breeding in
which individual plants are selected from a segre-
gating population of known parents (Love 1927).
Hence breeder selection (sometimes referred to as
artifi cial selection) and natural selection occur in
every generation. An identity is assigned to indi-
vidual selections in each generation. The progeny
of an individual will be assigned an identity so its
progenitor is known. Documentation enables the
breeder to trace progeny-parent relationships to
the original hybridization of parents. The criteria
of selection may be phenotypic, genotypic, or a
combination of both. Family relationships may
also be used in the evaluation of data to make
selections. The pedigree method is labor-intensive
but provides the most genetic information about
a selection, as it is based upon parent-progeny
genetic analyses. For example in a segregating
population, if the progeny of a selected plant Z is
uniform for the trait, plant Z must have been
homozygous for the trait.
Pedigree selection also allows the breeder to
create a knowledge base using episodic traits
during the progeny evaluation to know the char-
acteristics of the selected new line. For example,
if an important disease occurs once every 3 years
during the progeny evaluation, the breeder will
know if the progeny family was uniform or seg-
regating for the trait (e.g., the parent plant was
homozygous or heterozygous for the genes con-
trolling the trait). Thus, once the selection history
has been created, the breeder can know the
response of the selected new line to various
