Agriculture Reference
In-Depth Information
their genetic potential. In practice, selection differentials
are calculated separately for the buck and doe, and then
averaged. This is because fewer bucks are selected com-
pared to does resulting in greater selection pressure from
the bucks. A simple average of the selection differentials
of buck and doe is used because the male and female
groups each contribute equally to their offspring.
When selection is applied to a population, genetic prog-
ress occurs in the form of change in performance of indi-
vidual goats from one generation to the next. This change
(
Multiple Trait Selection
Economically important traits can be selected singly or
in combination. There are several ways of doing this.
With Tandem selection, one trait is selected for one or
more generations until the desired improvement has been
achieved, and then a second trait is selected until required
improvement has been achieved with it. This process can
be continued until all of the traits of interest are improved.
Tandem selection is useful for three or fewer traits that are
not correlated, otherwise the effi cacy of selection will be
reduced.
Another method of selection for multiple traits makes
use of Independent Culling levels for each trait of interest.
Threshold levels would be established for each trait, and
goats exceeding the threshold in the desired direction for
all traits would be selected for use as parents. The advan-
tage of this procedure is that selection can occur at various
stages, for example, during birth, weaning, and market
weight reducing, the number of goats that have to be
retained, before the fi nal stage of selection with the last
trait measured. This procedure is most effective when the
number of traits does not exceed three and the intensity of
selection is high. Both the Tandem and Independent
Culling level of selecting parents have an operational
advantage in that they can be implemented without
knowing genetic parameters for the traits or their exact
economic worth. It would be desirable, however, that the
choice of traits and the selection emphasis placed upon
them be based on knowledge of their economic worth and
expectation of genetic progress (heritability).
The most effi cient procedure in selecting for multiple
traits is the Selection Index. This is a function of pheno-
typic measures chosen according to the breeding objective
together with weights derived from genetic parameter esti-
mates and their economic values. The Selection Index is
given by Σ
X
i
b
i
, where
X
i
represents phenotypic traits (
i
)
such as fecundity, lean meat yield, milk production, etc.,
and
b
i
are coeffi cients derived from additive genetic vari-
ance and covariance of the measurement along with their
economic worth (for methodology see Cameron, 1997).
In the past, the high cost of computer processing and
slow turnaround time were disadvantages with index selec-
tion. Easy access to high-speed computers, however, has
made the Selection Index an important tool for prediction
of the breeding worth of multiple traits. One problem with
index selection is with situations where information is col-
lected on animals over a period of time. Here, it is impor-
tant to retain all of the animals until the evaluation is
complete. This may contribute to fi nancial loss from the
need to keep less productive animals until culling is
G
) depends upon the selection differentials of the
selected does (
X
-
SF
) and bucks (
X
-
SM
), as well as the herita-
bility of the trait. For simple selection on phenotypic per-
formance of goats, genetic progress per generation for each
Δ
Gh
XX
+
⎛
⎜
⎞
⎟
SF
SM
trait is predicted as
Δ
=
2
.
2
Selection can occur among individuals, family (full or
half siblings), or a combination of both. In a closed popula-
tion of goats, selection of individual goats based on their
phenotypic measurements is often referred to as mass
selection. In general, traits with medium to high heritabili-
ties, such as carcass measures and growth rate, are best
selected on the basis of performance of the individual goat.
Family selection becomes important when selecting for
traits that are low or moderate in heritability such as repro-
ductive traits. In this case, families are selected based on
the mean performance of a half- or full-sibling group. With
pedigree selection an individual goat is selected based
upon the presence of important ancestors or high-perform-
ing goats in its pedigree. This procedure is still used in
breeding champion horses and dogs to improve perfor-
mance. However, care must be taken not to place too much
emphasis on the presence of an ancestor with outstanding
performance in a pedigree when it is more distant than a
parent. A grandparent, for example, has only 25% of its
genes in common with its grand-progeny. Progeny testing
is common in many species and is valuable for sex-limited
traits such as milk production. However, this procedure
can be expensive because of the need to retain animals
until the superior individual is identifi ed, thus extending
the generation interval.
The choice of the method is dependent on the expected
annual genetic response to selection and the associated
costs. It must be noted that long-term genetic progress is
dependent on the infl uence of several factors. These
include reproductive rate, number of traits under selection,
heritability, genetic correlation, inbreeding and effective
population size, consistency of selection goals, uniform
test environment, generation interval, and genotype-
environment interaction.
