Theoretical and Applied Genetics 46, 401-406 (1975) 9 by Springer-Verlag
1975
Breeding Implications from a Diallel Analysis in two Environments of Yield and Component Characters in Upland Cotton (Gossypiurn hirsuturn) H. E. Gridley Cotton
Research
Corporation,
Namulonge
Research
Station,
Kampala
(Uganda)
Summary. In a full diallel set of crosses involving nine Albar 51 lines the genetic control of yield and yield components were largely accounted for by additive effects, although non-additive effects were evident. Significant differences in correlation coefficients between sites and the interaction of genetic effects indicate the effect of the environment on development and gene action. The failure to achieve yield improvements appears to stem from a combination of testing at the early selection stage in a single environment and of a narrow genetic base. The implications for future breeding programmes are discussed.
I. Introduction BPA
is the present
commercial
land cotton in southern Church,
1968).
racters,
Uganda
Pedigree
that differ from
variety of African (Arnold,
selection
the first BPA
but no line has been obtained attempts
other characteristics
of successive
therefore,
issues
(Arnold,
Gridley
BPA
stocks
selected
equatorial to assess
lines.
regions
Innes,
genetic variability
within
1973).
made
on the genetic
characters,
in 1969
Environmental
of Africa
diallel set were
between
are extremely
parents
and
con-
a full dialnine pedi-
conditions
the effect of the environment
the genetic parameters,
Buganda,
issues
19711
trol of yield and its component
gree
1970,
and obtain information
was
seed
on the use of multiline seed
to increase
le set of crosses
cha-
the yield and
BPA
Innes and Gridley
and Busuulwa,
and lines
that is superior
to improve
been based
In an attempt
Costelloe
has produced
seed issue in some
to it in yield. Recent
have,
Up-
in the
variable,
on estimates F I hybrids
one at Mubuku,
Analysis
In the Hayman (1954) analysis, variation is divided into components which measure additive and non-additive genetic effects and reciprocal differences. Item 'a' refers to additive parental effects and "o' to non-additive effects. Where an additive-dominance model is adequate the latter can be partitioned into terms providing tests of mean dominance (b• of variation of dominance from array to array (b2) and of dominance deviation that is unique to each Fi (b3). If tests of the value of Wr (covariance) and Vr(variance) do not show the expected relationship, then non-additive effects can not be interpreted in terms of simple dominance. Item 'c' deA 474
b
5eoson
5 5eneroti0ns
1957
A (57)5 I
I
of the
Western
Region. 2. Materials
Genetic
and of
included in a rain fed trial at Namulonge,
and in an irrigated
Mubuku with two plants per stand in two row plots at a spacing of 0.9m X0.3m. The yield components, seed cotton per boll, loculi per boll, seeds per loculus, seeds per boll, seedweight and lint per seed were determined from fifteen boll samples at Namulonge and ten boll samples at Mubuku taken at the time of second pick. Ginning percentage and lint yield were estimated from the produce of whole plots.
and Methods
The nine parents used in the crossing programme, which included reciprocals, were the pedigree selections BPA 68, A(66)22, 29, 36, 102, 131, 134 and A(67)8, 12 from Albar 51 ; all nine selections stemmed from A ( 57 ) 5 (Innes and Jones, 1972). The derivation of these lines is shown in Fig. 1 where A(57)5 is the sixth generation inbred from A474, with single plant selection in each generation. The parents were therefore highly inbred and closely related. In 1970-71 parents, hybrids and reciprocals were grown in a 9 x 9 balanced lattice square with five replicates at Namulonge and Mubuku. Plot size was 4.3 m 2 in theNamulonge trial and 3.2 m e in the one at
I
1958
A(58)14
1959
A(59)Z8
1960
A(60) 6
1951
I
I t
/b A (51) 6
1966
BPA 65
1967
BPA 67
A(66)2Z 29 35 102 131 134
I
1968 BPA 68 F i g . 1. P e d i g r e e o f A l b a r 51 s e l e c t i o n s f u l l d i a l t e l set o f c r o s s e s
A(57} 8
12
used in the 9 x 9
402
H.E.
Table
I. Parent and hybrid performance (N) and Mubuku (M)
Character Seed
per boll (g)
Breeding
Implications
for the eight yield components
Parental mean
Site
cotton
Gridley:
increase F~ mean over parental mean
FI mean
4.8 • 0.08 5.3 -+ 0.05
4.9 -+0.02 5.4 -+0.02
2.8 ~ 1.9
Loculi
per boll
N M
4.2 • 4.3 •
0.02 0.02
4.2 • 4.3 -+0.01
Seeds
per loculus
N M
6.8 7.9
0.08 0.04
6.8 7.9
Seeds
per boll
N M
Seed weight (mg/seed)
N M
Lint p e r s e e d ( r a g )
N M
Lint y i e l d ( k g / h a ) Ginning percentage
28.7 • 0.37 33.7 + 0.34 119 108
+ •
1.1 1.0
0.5 -0.3
-+0.02 •
0.3 0.4
29.0 -+0.12 33.7 •
0.9 0.1
119 109
a Diallel Analysis
at Namulonge
N M
• +
from
r 0.30 -+ 0 . 4 0
0.5 1.2
4.8 + 0.05 5.2 • 0.05
4.9 • 5.3 + 0.02
1.2 1.7
N M
651.0 • 12.17 531.0 • 14.00
710.6 • 4.50 560.4 • 5.74
N M
3 1 . 2 -+ 0 . 1 5 32.2 • 0.09
31.3 +0.05 32.2 -+0.05
9.1 ~ 5.5 0.2 0.3
~,H~ p < O. O01 ~,~ P < 0 . 0 1 ~ p < 0.05
tects average maternal effects and item 'd' the reciprocal differences not ascribable to c. If there are differences between reciprocal crosses, then the mean squares for items c and d become the appropriate error terms for the a and b items, respectively (Wearden, 1964). A joint Hayman (1954) analysis of the data from the two sites was also carried out. The presence of an interaction of genetic effects with sites was then testedby comparing the item x sites with the item x blocks mean square. Adjusted data from the lattice square analysis were used to calculate the phenotypic correlation coefficients but for the Hayman (1954) analysis rows and columns of the lattice square were ignored. All statistical analyses were carried out at Rothamsted Experimental Station on an I.C.L. System 490 Computer using Genstat.
ces in sample
size which
was
9 in the case
of parents.
The
ulations
at both sites of seed
per boll, seeds erally number
pressed above
I indicate
that significant
as the percentage the average
increase
of the parents,
heterosis,
ex-
or high.
contributes
The
more
hybrid
boll with loculi
per boll were
This suggests
weight
the latter are highly
show
gen-
that total seed
to boll size than seed
although
populations
per
seeds
but only
in the two pop-
population
a positive
corre-
at Mubuku.
effect of increas-
ing boll size on lint yield. Correlations indicated
hybrid
in Table
and
lated with boll size in the hybrid
seed Data
cotton
loculus
and lint per seed,
positive
Heterosis
per
significant
significantly
by a bracket
differences 3. Results
72 for the hybrids correlations
in Tables
in degree
relationship.
are seed
for the former
between
2a and
of relationship
to a negative populations
different
2b and range to a switch
Examples
cotton per
and seed
sites are
weight
from
from
from
a
the
boll and lint per
and ginning
percen-
tage for the latter.
of the F I hybrids is shown
only by seed
Diallel Analysis
cotton per boll and lint yield at Namulonge. Analyses model, Correlations Phenotypic components given
were correlation for parents
in Tables
2a and
coefficients and hybrids
among
2b. The greater
cally significant in the hybrid
the eight yield
at the two sites are
populations
number
statisti-
reflects differen-
terms.
of variance are
shown
calculated The
using
relevant
for interaction
of lint yield, using Hayman's in Table
components
variance
the associated mean
squares
were
shown
( 1954 ) (VR)
interaction
and variance
in Table
calculated
ratios
block
of additive or non-additive
sites for lint yield are other
3. The
ratios
effects between
4. Results
similarly
and
for the are in-
H . E . Gridley: Breeding
Implications f r o m a Diallel Analysis
403
Table 2a. P h e n o t y p i c c o r r e l a t i o n c o e f f i c i e n t s b e t w e e n the eight y i e l d c o m p o n e n t s f o r p a r e n t s at N a m u l o n g e (upper
value)
Character
and Mubuku
(lower
value)
S e e d cotton
Loculi p e r
Seeds per
Seeds per
Seed
per boll
boll
loeulus
boll
weight
Lint p e r seed
Lint
yield
0.56
Loculi per boll
0.75+~
0 962
Seeds per loculus
-0.07 0.53
0.19
Seeds per boll
Seed weight
0 985 ~9+~
0.47
0 9 53
0 . 8 6 +~+~"
0 944 0.35
Lint p e r seed
0.06
0.37
-0.23
0.32
-0.31 0.43
-0.60 O. 10
-0.24 -0.23
-0.29 -0.57
-0.41 -0.47
Seed
Lint p e r
Lint
weight
seed
yield
Lint per seed
0.30+~-
Lint yield
O. 2949 O. 4 8 4 ~
- 0 . 5 0 +~+9+91+9 -0.50 +9+~+91e -0.12 l -0.13 l 1
in a summarised
form,
effects were
for lint per were
and
-0.09
appropriate.
Non-
agreement
of the ~oint regression
suggests
that simple
planation
of the non-additivity
E. However,
statistical tests des-
(1971)
using
None
for individual
Wr
and
Vr
an additive-domiof the three charac(a) the consistenbetweenthe blocks
and
re~
(c) the
0 9 56 +9+~*~]
- 0 . 2 8 +9 ]e~9 - 0 . 2 5 ~ ]~9~*~ 0.47+9~9'9 0 39 +~+91 0 58~9+Hq 0.48+9~
]~9
seed and lint yield at both
over arrays , (b) agreement Vr
0 53 +9~+~]
mation
ters satisfied all three tests, involving
on
0.30 0.00
for seed cotton per boll at
inbred,
Jinks
at both sites
at Namulonge.
0 5049~+~1r
0.19 -0.20
-0.17
with lint yield, in Table 5.
to investigate whether
of Wr
0.08
highly significant
evident
sites. As the parents
O. Z5~ 0.05
0.24
except seed weight
additive effects were
-0.23 +9 -0.20
-0.08 -0.18
O. 19 ~0.0 2
]~ 0 . 3 3 +~+9 ]
gression
O. 69+9~4~ J
]~9~9 -0.33 0.69~+~ ] -0. I0
Ginning percentage -0 902
cy of Wr-Vr
at Namulonge
Seeds per
-0.0 2 ] +9+9+9 - 0 . 2 0 O. 74~+9+~] -0.05
was
for hybrids
boll
Seed weight
model
different with the level of signif-
the eight yield components
loculus
0.80+~+~91~9~9 0.58 ~+~9 0 48~+~+91 0.70 ~*~+~
nance
they are significantly
0.34 0.59
-0.35 0.64
0.57
Loculi p e r S e e d s p e r
Seeds per boll
were performed
-0.27
boll
0.71~9~+914~9 0.10 0.30 +~ J -0.02
by Mather
I
0.591
Seed cotton
Seeds p e r l o c u l u s
cribed
indicates
-0.50]~
per boll 0.43 +9+9 0.36 +9~
and
O. 74+~ O. 73 +~
-0.50
Loculi per boll
Namulonge
-0.65
0.32
2b. Phenotypic correlation coefficients between (upper value) and Mubuku (lower value)
Additive
-0.07
-0.86J -0.53
Table
for all characters
-0.13 ]~
O. 11
Bracketing of two correlation coefficients icance shown at the top right hand corner
cluded
-0.06 -0.52
-0.33
Note:
Character
-0.30 -0.7049
O. 18
O. 11
Ginning percentage
0 . 8 8 +~+~"
0.40 -0.17
0.09
Lint y i e l d
0.85 ~+~
dominance
of the components
sociated
is not an adequate
of variation
significant
ex-
and does not allowthe
the statistically
with
slope with unity. This
DI, HI, H2,
significant
heterosis
estiF and
b I item is as-
for two
out of thl-ee
components. Differences were
between
evident for seed
Namulonge. reciprocals dications
For
arrays
of reciprocal
both characters
wei-e small. that hybrids
crosses
weight and ginning percentage
For
differences
the former
involving
BPA
at
between
there were
in-
68 as the female
404
H.E.
Gridley:
Breeding
Implications
from
a Diallel Analysis
Table 3. Hayman analysis of variance for lint yield in the 9 • 9 full diallel set of crosses at Namulonge and Mubuku Site
Namulonge
Mub~u
Item
DF
MS
VR
MS
VR
a b
8 36 1 8 27 8 28 32 144 4 32 108 32 112
30451.20 8732.29 154814.00 4016.80 4719.04 4297.32 4201.41 3491.70 3046.55 1709.97 2303.27 3316.29 1961.61 3709.00
8.72*** 2.87*** 90.54*** 1.74 ns 1.42ns 2.19 ns 1.13 ns
44844.80 9670.71 18112.11 11725.20 8749.33 10729.99 5218.23 10439.30 7061.76 1123.09 9036.86 6696.50 6699.00 8701.30
4.30** 1.37 ns 16.13" 1.30 ns 1.31ns 1.60 ns 0.60 ns
bl b~ b3 c d a • blocks b x blocks b~ • b l o c k s b2 • b l o c k s bs • b l o c k s c x blocks d • blocks
ns non-significant
Table 4. Analysis
Item
C
d
Table
for lint yield in the 9 • 9 full diallel set of crosses
Item • Sites
b~ b= b3
Note:
of item • site interactions
DF
MS
8 36 1 8 27 8 28
8165.00 6723.65 33779.51 8010.63 5340.26 8864.50 3314.04
Item • Blocks
VB
0.78ns 0.95ns 23.85** 0.89ns 0.80ns 1.32ns 0.38ns
Namulonge
Mubuku
Pooled
DF
MS
MS
DF
MS
32 144 4 32 108 32 112
3491.70 3046.55 1709.97 2303.27 3316.29 1961.62 3709.00
10439.30 7061.76 1123.09 9036.86 6696.50 6699.00 8701.30
64 288 8 64 216 64 224
6965.50 5054.16 1416.53 5670.07 5006.40 4330.31 6205.15
The variance ratios for all items except be are based square as tests showed them to be heterogeneous
5. Summary of Hayman analysis at Namulonge and Mubuku
Site
of variance
conservatively
for eight yield characters
Namulonge
Mubuku
on the greater
mean
in a 9 • 9 full diallel set of crosses
Interaction
Character Seed cotton per boll Loculi per boll Seeds per loculus Seeds per boll Seed weight
a*** a*** a*** a***
b~
Lint per seed
a***
b~
Lint yield Ginning percentage
a***
a***
d*
d*
b***
a*** a** a** a***
c**
d**
a*** a***
c**
d***
a***
a***
b~**
d* a***
d***
a***
b~
b~ b~ c*** d***
H.E.
Gridley:
parent
were
Breeding
Implications
contributing,
marginally
heavier
all hybrids
involving
slightly lower
than their reciprocals,
A(66)36
between
and seeds
at Mubuku.
Differences
nents appeared
by c, were
involving
Interactions
were
of variance calculated
ferences,
effects except
be-
ments~
seeds
who
of heterogeneity
square
have been
ratio was
only. Parts
or
due to scale dif-
but only for seed cotton per boll and ginning
percentage accounted
was
a significant
proportion
for by a linear regression
rental or array same
may
means,
be caused
them.
variable
of Wr
- Vr
on pa-
site.
material
The present
as the cause
Arnold
(1971)
interplay care
which
must
or si-
one or both of
found genes
de-
blight in Tanzanian
cot-
in certain environ-
variation
after transfer
of genotype
of
but can not dis-
in his review,
data serve
masked
indicates
to bacterial
(1969),
in expression
of previously
had found unexpected
quoted
authors
in phenotypically
to a new
non-
environment.
further to illustrate the dynamic and environment
and emphasise
be taken in interpreting
the
results from
a
single environment.
of the variance
in neither case for both at the
Walker
the interplay
and be non-transmis-
The diallel analysis
resistance
in develop-
from
by a change
ton stocks that are only expressed
the two sites the variance
all of these interactions
purely
and environment
tinguish between
evidence
mean
stem
or by the expression
termining
those switching
reflect a change
the latter two alternatives
if either of the
for all characters
using the larger
genes
lent genes.
is important.
there was
between
of the genotype
of a number
of additive or non-additive
Where
for seed
This may
sible, or it may
in correlation
sites, particularly
pattern.
and for all compo-
it is doubtful
evident
between
ment
at
on the coordinated
but the differences
crosses,
and ginning percentage
small
coefficients
genes
positive to negative,
different parents.
purposes
sites were
per loeulus.
evident
action of many
from
and ginning percentage
per loculus
sets of differences
tween
The yield of a crop plant depends
while
parent had
of reciprocal
to be due to the behaviour
practical
above
as the female
arrays
cotton per boll, seed weight Namulonge
405
in that six out of eight had
seeds
other than that measured
For
a Diallel Analysis
ginning percentages.
Differences
of crosses
from
In a breeding tion stages,
programme,
owing
testing at the early selec-
to limited
seed supplies,
stricted to a single environment ricultural produce
practices.
Selection
lines adapted
is often re-
employing
advanced
for improved
yield
to these conditions
agmay
but lacking the
4. Discussion stability required An understanding
of the genetic control of agronomical-
ly desirable
characters
ment
genetic variability
where
lection.
The
control
largely
determined
sites although interaction seeds
in crop
is exploited
of the eight yield by genes
non-additive
of genetic
per loculus
between
is necessary
shows
Wimble
having
has
a similar
that parental pattern
shown
shaped
environments, lations.
in this material
in press).
only to large climatic season
to season
cultivated variation
(Arnold
of soils and management, tal combinations. that under between
Arnold
such conditions
genotypes
reach
hybrids optima
factor and re-
will vary
in changes
in different
in dominance
re-
cotton is subject not from
ample,
reselection
produced sampling
ley,
1970,
was
year the environment grown
use of the environment
(St. Pierre,
thy of further evaluation
place to place and
may
selection
characteristically variability
Klinck
and
and might
prove
Short term multiline
(1975)
have
striking interactions for yield.
shown
Tanzania
(Lee,
occur
ly such
mixtures
if
although
advances
from
a programme
inbreeding,
which
by a rapid decrease
is in
1971).
inprovements
might
comprising
over
(Arnold,
1967).
aid is worbeneficial
stem
from
a number
related lines. In district trials they have
in Uganda
environmen-
and intensive
seed issues
yield advantages
in many
stem
each
material
Gauthier,
as a practical
accompanied
(Arnold,
with wide
by alternating
the segregating
be limited as the parents
of pedigree
Innes andGrid-
of barleys
in which
was
ex-
of the diallel cross has
in Canada
Such
resulting
and environment
The selection
achieved
For
yield advantage in trials but fail to maintain this
district trials (Arnold,
1971).
adaptation
1971) but also to a wide range
and Innes
in the parents
lines which exhibit a few environments,
in wide ranging
cultivation.
applied to the hybrid populations, and
surface
Fl'S and parents
In East Africa
vary
in the analy-
that parents
response
resulting
The
except
differences
emerged
at different levels of one environmental lations between
is
effects are also evident.
and Gridley,
(1973)
se-
effect at both
effects for all characters
sites. A similar
Knight
through
components
of additive
sis of three fibre quality characters (Innes,
improve-
for commercial
the mean
and Jackson,
of genotypes
against the environment
shown
small
of their component
Innes and Gridley,
Walton
the use of
of closely
1970)
1973).
lines
and
in
Theoretical-
will be better buffered
than a single genotype.
Although
406
H.E. Gridley: Breeding Implications from a Diallel Analysis
c o n c l u s i v e e v i d e n c e is l a c k i n g , the l a r g e a m o u n t of a d d i t i v e g e n e t i c v a r i a t i o n d e t e c t e d s u g g e s t s the f e a s i b i l i t y of s u c h an a p p r o a c h w h e r e a s e e d i s s u e i s g r o w n c o m m e r c i a l l y for s e v e r a l y e a r s . The y i e l d a d v a n t a g e of the F 1 h y b r i d s at both s i t e s i s not s u f f i c i e n t f o r the e c o n o m i c p r o d u c t i o n of h y b r i d s e e d , a s c r o s s i n g has to be a c h i e v e d by t i m e c o n s u m i n g m e t h ods of e m a s c u l a t i o n and hand p o l l i n a t i o n . S e v e r a l g e n e s c a u s i n g m a l e s t e r i l i t y h a v e b e e n found ( W e a v e r and A s h l e y , 1971) but no s a t i s f a c t o r y s y s t e m of r e s t o r i n g f e r t i l i t y has yet b e e n d i s c o v e r e d . Long t e r m i m p r o v e m e n t s in y i e l d will h a v e to s t e m f r o m s e l e c t i o n in g e n e t i c a l l y w i d e - b a s e d p o p u l a t i o n s s y n t h e s i s e d f r o m the c r o s s i n g of e x o t i c m a t e r i a l to l o c a l l y adapted s t o c k s . T h e s e r e s u l t s , and t h o s e on lint quality (Innes et a l . , in p r e s s ) ,
e m p h a s i s e the n e e d to t e s t
e a r l y g e n e r a t i o n s e l e c t i o n s o v e r as wide a r a n g e of e c o l o g i c a l c o n d i t i o n s as is p r a c t i c a b l e . The s y n t h e s i s of a r a n d o m m a t i n g pool of the b e s t l i n e s in the e a r l y s t a g e s of a p r o g r a m m e
might p r e s e r v e g e n e t i c v a r i a b i l i t y to
which a b r e e d e r c o u l d l a t e r r e t u r n .
Acknowledgement I would l i k e to thank D r . N . L . Innes f o r help in the d e s i g n of the e x p e r i m e n t and d i s c u s s i o n of the r e s u l t s and D r . R . H . W i m b l e for a s s i s t a n c e in the s t a t i s t i c a l a n a l y sis.
Literature A r n o l d , M . H . : Cotton i m p r o v e m e n t in E a s t A f r i c a . In: C r o p I m p r o v e m e n t in E a s t A f r i c a , ed. L e a k e y , C . L . A .
R e c e i v e d A p r i l 8, 1975 C o m m u n i c a t e d by R. R i l e y
C o m m o n w e a l t h A g r i c u l t u r a l B u r e a u x Techn. C o m m . 1_~9, 178-208 (1971) Arnold, M.H.; Costelloe, B.E.; Church, J.M.F.: BPA and SATU, Uganda's two n e w cotton v a r i e t i e s . Cotton G r o w . R e v . 45, 162-174 (1968) A r n o l d , M . H . ; Innes, N . L . : P l a n t B r e e d i n g . In: A g r i c u l t u r a l R e s e a r c h f o r D e v e l o p m e n t . The N a m u l o n g e C o n t r i b u t i o n . London: Cotton R e s e a r c h C o r p o r a t i o n and C a m b r i d g e U n i v e r s i t y P r e s s 1975 (in p r e s s ) A r n o l d , M . H . ; Innes, N . L . ; G r i d l e y , H . E . : Cotton B r e e d i n g . P r o g r . R e p . E x p . S t a s . , Uganda, 196970. Cotton R e s . C o r p . , London, 12-34 (1970) A r n o l d , M . H . ; Innes, N . L . ; G r i d l e y , H . E . : Cotton B r e e d i n g . P r o g r . R e p . E x p . S t a s . , Uganda, 1970-71. Cotton R e s . C o r p . London, 17-34 (1971) H a y m a n , B . I . : The a n a l y s i s of v a r i a n c e of d i a l l e l t a b l e s . B i o m e t r i c s 1._.O0, 235-244 (1954) Innes, N . L . ; G r i d l e y , H . E . ; B u s u u l w a , L . N . : Cotton B r e e d i n g . P r o g r . R e p . E x p . S t a s . , Uganda, 1971-72. Cotton R e s . C o r p . London, 13-28 (1973) Innes, N . L . ; W i m b l e , R . H . ; G r i d l e y , H . E . : E s t i m a t e s of g e n e t i c p a r a m e t e r s for lint quality in upland c o t t o n . T h e o r . a n d A p p l . G e n . , 46, 249-256 (1975) Innes, N . L . ; J o n e s , G . B . : A l l e n ; a s o u r c e of s u c c e s s f u l A f r i c a n cotton v a r i e t i e s . Cotton G r o w . R e v . 4_..99, 201215 (1972) Knight, R. : The r e l a t i o n b e t w e e n h y b r i d v i g o u r and g e n o t y p e - e n v i r o n m e n t i n t e r a c t i o n s . T h e o r e t . and Appl. Gen. 4..~3, 311-318 (1973) Lee, B . J . S . ; Walton, I . C . ; J a c k s o n , A . C . : B r e e d i n g and A g r o n o m y . P r o g r . R e p . E x p . S t a s . , Tanzania, 1972-73. Cotton R e s . C o r p . , London, 10-35 (1973) M a t h e r , K. ; J i n k s , J . L . : B i o m e t r i c a l G e n e t i c s ; 2nd E d . London: C h a p m a n and H a l l , 1971 St. P i e r r e , C . A . ; K l i n c k , H . R . ; G a u t h i e r , F . M . : E a r ly g e n e r a t i o n s e l e c t i o n u n d e r d i f f e r e n t e n v i r o n m e n t s as it i n f l u e n c e s a d a p t a t i o n in b a r l e y . Can. J . P l a n t S c i . 47, 507-517 (1967) W a l k e r , J . T . : S e l e c t i o n and q u a n t i t a t i v e c h a r a c t e r s in f i e l d c r o p s . B i o l . R e v . 4..~4, 207-243 (1969) W e a r d e n , S. : A l t e r n a t i v e a n a l y s e s of the d i a l l e l c r o s s . H e r e d i t y 1__.99, 669-680 (1964) W e a v e r , J . B . ; A s h l e y , T. : A n a l y s i s of a d o m i n a n t g e n e f o r m a l e s t e r i l i t y in Upland c o t t o n , G o s s y p i ~ h i z ~'suSw~ L. C r o p S c i . 1._!1, 596-598 (1971)
H. E . G r i d l e y Cotton B r e e d e r I n s t i t u t e of A g r i c u l t u r a l R e s e a r c h Samaru P . M . B . 1044 Zaria (Nigeria)