Agriculture Reference
In-Depth Information
Qualitative trait selection
more sophisticated tests. For example, it is common
practice in most plant breeding schemes not to select
the early generation lines for quality traits which involve
either large quantities of produce, which provide only
crude estimates of worth with small samples or that are
expensive.
Obviously any character which is considered of high
importance should be selected for at the earliest stages
of a plant breeding scheme where greatest variation will
exist among families or populations but where the trial
designs and amount of material make selection effective.
Despite the simplicity of this statement, in practice it is
often completely ignored.
The characters which are evaluated at different stages
of a plant breeding scheme will be discussed in later
sections.
Characters whose expression shows a qualitative form
of inheritance can be easily selected for, provided that a
suitable screening method is available to determine the
presence or absence of the single gene in plants in seed-
propagated crops. If the expression of the qualitative
character is determined by a recessive allele, then a single
round of selection should ensure that all selected plants
are fixed for the particular trait. If the desirable allele is
completely dominant, several rounds of recurrent selec-
tion will be necessary to ensure that the character is
genetically fixed in selected plants.
Qualitative characters can often be selected relatively
quickly and using very small plots (sometimes, even
single plant plots) compared to quantitative inherited
traits. The ease of selecting for single gene traits has
resulted in these characters having high selection inten-
sity in the early generation stages where most genotypes
are evaluated and where it may only be possible to grow
small plots.
Selection for such qualitative expression can indeed
be a powerful tool in reducing the number of geno-
types selected in a plant breeding scheme, although it
should never be forgotten that it is often the quanti-
tatively inherited characters which add greatest value
to a new cultivar (i.e. yield, quality and durable plant
resistance). If early generation selection is to be carried
out for single gene traits, then the breeder must be sure
that this selection is not having an adverse effect on the
selected populations (i.e. no linkage between advanta-
geous qualitative traits and adverse quantitative traits
or any unwanted non-allelic interactions, or pleitropic
effects).
Positive and negative selection
Two forms of selection are said to be available to plant
breeders, positive and negative selection. It is difficult
to clearly define the difference between the two types
(and indeed, some wise and worldly breeders do not
distinguish between the two). In simple terms nega-
tive selection is where the very worst plants or families
are discarded while positive selection is where the very
best plants or families are selected. Perhaps the sim-
plest description would be related to the proportion of
plants that are selected from a population. If more than
50% of the original population is selected then this can
be considered negative selection. If less than 50% of
the population is retained then this would be positive
selection.
RESPONSE TO SELECTION
Quantitative trait selection
It has already been stated that selection will only be
successful if there is sufficient phenotypic variation and
that at least some of this variation is genetic in origin. It
should be of no surprise, therefore, that the response to
selection is related to heritability. Indeed consider the
equation:
Quantitatively inherited characters usually are more
difficult to evaluate due to the higher potential for
modification of expression by the environment. Greater
experimentation (replication or plot size) is necessary
to maximize selection response. As a result, many of
the quantitative traits are not positively selected for
in the early generation selection stages. Selection for
these characters is often delayed until the numbers of
genotypes which require testing are reduced and where
greater amounts of planting material are available for
h
2
where
X
1
is the mean phenotype of the selected geno-
types,
X
n
−
1
is the mean phenotype of the whole
population,
R
is the advance as a result of one round
X
1
−
X
n
−
1
=
=
σ
R
i
