Agriculture Reference
In-Depth Information
Table 7.5
Selection scheme used by Dr. Robert Zemetra in the soft white wheat breeding program at the University of Idaho.
Year
Number
Plot type
Characters assessed
of genotypes
1 - F
3
1 200 000
Single plants
Visual selection of plant types, plant height and stripe rust resistance
2 - F
4
20 000
Head rows
Visual selection of plant types, uniformity, yield, height, stripe rust
resistance, lodging resistance. Actual assessment of protein content and
kernel hardness
3 - F
5
500
Preliminary yield
trials, at one
location
Actual yield performance, stripe rust yield resistance, lodging, stand
establishment, heading date, test weight, dough viscosity, milling
quality, baking quality, protein content and kernel hardness
4 - F
6
80
Preliminary yield
trials at two
locations
Actual yield performance, stripe rust yield resistance, lodging stand
establishment, heading date, test weight, dough viscosity, milling
quality, baking quality, protein content and kernel hardness, yield
stability, Russian wheat aphid resistance, dwarf bunt, cephalosporium
stripe
5 - F
7
10
Advanced yield trials
at eight locations
Actual yield performance, stripe rust yield trials, resistance, lodging
stand, establishment, heading date, test locations weight, dough
viscosity, milling quality, baking quality, protein content and kernel
hardness, yield stability, Russian wheat aphid resistance, dwarf bunt,
cephalosporium stripe, foot rot, Hessian fly resistance and
winter-hardiness
6 - F
8
6
Advanced yield trials
at eight locations
Actual yield performance, stripe rust resistance, lodging, stand
establishment, heading date, test
Weight, dough viscosity, milling quality, baking quality, protein content
and kernel hardness, yield stability, Russian wheat aphid resistance,
dwarf bunt, cephalosporium stripe, foot rot, Hessian fly resistance and
winter-hardiness
of planting material are available, it may be impractical
to have very large plots and high replication of so many
different lines. Staff and land are not usually available
to carry out such large screens.
Early generation selection is therefore carried out on
small plots and most often on un-replicated plots. Even
when the test entries are not replicated it is possible to
increase the efficiency of testing by including a wide
range of control entries inter-spaced within the test
plots. Direct comparison can be made between the con-
trol lines (often existing cultivars) and those under test.
The control plots can be included more than once over
the whole trial area and from this an estimate of plot to
plot error variance can be obtained.
Several forms of analysis are available by which test
entries can be compared to, or adjusted to, adjacent
control lines. It should be noted that it is unlikely that
all controls within a trial will have equal performance
characters. Some controls will have higher yield, others
would have better disease resistance or high expression
in a single character, and this needs to be accounted for
if comparisons are to be made with test lines.
Visual assessment
The large number of genotypes which need to be
assessed in the early generations usually dictates that
most selection is by visual assessment. Visual assessment
of genotype performance is based on a mental image of
the desirable attributes (
ideotype
) that will constitute a
successful variety. Such assessments are therefore sim-
ilar to an informal selection index. The efficiency of
visual assessment can be influenced by breeders' expe-
rience and also the time taken over assessment. Visual
assessment has been proven to be more effective when
more than a single breeder is involved in assessment and
selection is carried out based on the average assessment
rating.
Throughout the season different traits can be visu-
ally assessed and then genotypes culled based on this
