الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 95 |  |  |
| | | from | 95 | | |
Abstract
SUMMARY
Six Egyptian cotton cultivars namely; Giza 45. Giza 70.
Giza 66 , Giza 67 , Giza 69 • Giza 75. were crossed in a diallel
pattern and evaluated for heterosis.,.combining ability and
genotypic variation • AnalY8i~of genetic component controlling
the. variables. seed cottoti yield. boll weight. lint percent.
seed and lint indices, micronaire reading. hair weight. staple
and mean length and yarn strength was also conducted. The six
parents and their 15 FI’s plus their corresponding F
2
’s were
included in a randomized complete blocks experiment.
1· The analysis of variance revealed significant differences
among genotypes for all variables studied in F
1
and F
gen~~rations • 2
1· Estimates of mid-parent heterosis were significant for seed
cotton yield, lint percent, hair weight and mean length.
Meanwhile the average BP heterosis (useful heterosis for
. pl.nt breeder)w.as significant and positiveforseedC()ttonyield
Moreover. positive but insignificant BP heterosis was obtained
for fiber fineness (hair weight) and fiber length
(mean length). While then:mrlning ~~ showed no heterotic effects
over the average .BP heterosis. A considerable number of
hybrids exhi bited signi fican t MP heterosis. when t only a few
displayed significant BP heterosis. The parental cuI tivars
Giza 67 and Giza 70 were represented in moat cases which sh
showed BP heterosis for yield.
9. ThE! reciprocal crosses which exhibited true he~erosts.
aLnost showed heterotic effects, however, in some cases it did
n01: reach the significant level. Thus, the heterotic effects
are real.
4. Mean inbreeding depression effects showed insignificant
effects for all traits except for fiber length (mean length)
which gave highly significant value. The individual F2
means were almost close to the mean of their MP and Fl performance
for all traits. This suggested that additive and
do~inance effects are more important than those for epistasis
in this material.
~. Highly significant or significant estimates were calculated
for GCA and SCA mean squares in the F2 generation, while the
values of SCA mean squares were inconsistent in the F1 generation.
~. The GCA/SCA ratio of variance components indicated that additive
genetic variance was aproximately of greAter tmportaqce
for all traits.
1. The magnitude of GCA effects corresponded closely with the
rank of the parental means for all traits. This indicates
that selection of parents u’sed . in breeding to improve these
traits~be based largely on the phenotypic performance oc
the parents and also suggests the presence of a considerable
additive effects.
A· The reciprocal effects are significant for seed index •
lint index and all fiber property traits in both F
and
F2 generations. They are lower in magni tude compared to
GCA and SCA. Meanwhile, the FI hybrids almost showed significant
effects indicating that heterosis in most traits
is real. However. the results may indicate the possibility
of reciprocal effects in Egyptian cotton and the traits
expression in the FI hybrid may be the function of both
genotype and cytoplasm. Therefore, the studies point out to
the need for including reciprocal crosses in such biomettrical
studies.
9. Relatively high level of heterosis and SCA effects observed
for seed cotton yield and fiber fineness(hair weight) in
certain crosses suggested that some potential useful combinations
were present. It seems that for this material,
breeding procedures used should be that maXimize both additive
and dominance genetic variance.
IO.Estimates of D component were greater than Hi for lint percent,
lint index and hair weight in F1 generation. This
suggests that additive genetic variance is more important
for these traits in this material than dominance genetic
variance. On the other hand. for the other remaining
traits the dominance genetic variance is more i.portant •
Moreov4~r, the estimator hI also indicated that the parents
seemed to carry more dominant alleles than recessives for
these characters studied.
11. The average degree of dominance were in the overdominance
range for seed cotton yield, boll weight, seed index, microna:
ire reading and staple and mean length in the Fl’ In
the F2, seed index and micronaire reading showed complete
and overdominance, res pee tively , while seed cot ton yield,
boll weight and yarn strength were in the parential dom-
Lnance range.
12. The est ima tor of the average frequency of posi tive VB negative
alleles were lower than 0.25 suggesting that positive
and 11egative alleles were not equally distributed among the
parent LaI cuItivars. The parents seemed to carry more dominant
genes than recessive, as indicated by the positive
values of F component.
11. Estimates of the ratio dominant to recessive illleles in
the parents, Kn/KR ’ were greater than one for all traits
which indicates asymetry of the parents, this implies
excess of dominant genes in the parents for these traits.
14. The number of effective factors, k , of seed cotton yield
and boll weight were relatively high in both generations,
whereas the remaining traits showed fairly uniform or eratic
estimates over the two generations.
15. High negative correlation coefficients between the mean
value of (Wr + Vr) of each array averaged over blocks and
the corresponding parental means averaged over blocks ,
were calculated for boll weight, lint percent and lint
index in the F1 and seed cotton yield and micronaire
reading in the F2. This ind ica tes that the dominant a
alleles are positive in ’irection, i.e., operating in
the direction of heavier bolls, higher lint percent, lint
index, seed cotton yield and micronaire reading. Moreover,
the positive coefficient calculated for hair weight
may indicate that dominant alleles are negative in direction
and operating in the direction of finer fibers.
I
16. High narrow-sense heritability estimates were calculated
for hair weight, lint percent and lint index, whereas ,
reliable values were obtained for boll weight and micronaire
reading and relatively lower values were estimated
for the remaining traits, in the F1 generation. The heritability
values calculated in the ’2 were medium to low
for t he traits studied.
17. These results indicated that an effecting breeding method
within this material for improving seed cotton yield and
its ~elated variables along with higher fiber quality
would be tlBt leeding towarde the production of varieties
rather than hybrids. However, the information obtained
in regard to H and degree of dominance suggests that sub_
stantial potential exists for obtaining improvements in
yield, components of yield and fiber properties in this
material, through the use of breeding program utilizingF1hybrids.