WW

W

I

WW?" .

L"

' I

‘ " (If

‘. I I II.
. 'I ’
, I

. I u'..

I .IKUII
3,

. .UIIahI DIE
1‘

O)
l

u .'
.mod.
1...
IL}
.

I .xwxukfi:

I .
I IVS? II;
V V r?‘ H. \ .Wovtu‘ III
dung; It... I I . r v
Nunn Juana» Lwhud 3.3 a... ! .flgmfin‘ .
Ilr Qua Nada- I I I I1)". I

r

I l . . 0 c f.’ 0"? V I .-
I o . I I)- a . I .
l I o .n. It; I

4. r :. .rJI mun}.

'vIIIII
- I I...

Q‘I

III‘II

I‘lI‘xn

III! IIIIII
‘Ir {.

I’II'.
-l I

”It"?

I I)!
III II III.
“HilfihlbII “.8”: I
g IIPIII'IIIfIIIVv‘I‘I’IIO‘IIIIIII'.I . II . I
III}: .Iglj. I‘ll“: ’I'IIIJu -vlll‘ul. I a I

IIQIIOI

JJIUB’II " I
1" ll 1' if

OI. .. {ION}!!! OloII

IIIIIIIII
o

IIII’III

I I ’IIDII,'I‘II‘II.I-.III

I I’Il llilcll‘l'I I I. l

{EI‘III II.- II .l\« D II

II?’ I'IIH’I ‘II

I

IWI;
III
‘I. MN”

,.'I .0
l I'IH’

v J I . or I

IrirIII
c l

I .1.

luv-
cl

LLLLL

I

II

.2.
up.)
. .344.-

III?

I
II

III-
I

’I:

1‘ II...

{I yy 1 ;;;;;;;;;
.. .II I. f" ......... I .

ma.

IIIIII .

III

I I c
. IIIIf’IIIII I

I
In-

I.

.IIIIIY

8‘.

If?

Irv?) ‘‘‘‘‘‘‘
I I I.IIII\'I

.I
I

III '"I
W

'Illl‘
.' L63

l
;h

' III’MI WI

IIIIII. I II I II I I I I

.I. IIILI. . I II I,I I I I, I IIII I I‘ll. I . .‘

..’I\I’JI . I . I . I I.. I I I I I
III'II I (I

 

I I- .UIIIIHI- -

In}, I»
fiIII
VI

Ill

4 I I. II
II II I. I . I I.l‘ I 'II

I ’1'.
1|le III

III-DP: -

'l

’II

   

IIIII III.HIII III- . IIIIIII

l’flI’lIWIII . rll I’v-

 

.l||.rr F

THES’S' MERARY

o , -,- _
Mick: ? E13: Erato

n:

University

 

This is to certify that the

dissertation entitled

THE PRIMARY SHOOT APICAL MERISTEM, YIELD COMPONENTS
AND YIELD OF BARLEY (HORDEUM VULGARE, L. EMEND. LAM.)
1. AN ASSESSMENT OF THE INVOLVEMENT OF

PROMOTIVE HORMONES
II. A MATHEMATICAmeggghnfi)R YIELD

 

James Benjamin Abaka Nhyte

has been accepted towards fulfillment
of the requirements for

Ph.D. degreeinCrOp Science

 

 

 

 

 

MS U is an Affirmative Action/Equal Opportunity Institution 0-12771

 

 

MSU

LIBRARIES
n.

 

 

RETURNING MATERIALS:
Place in book drop to
remove this checkout from
your record. FINES will
be charged if book is
returned after the date
stamped below.

 

 

 

 

 

THE PRIMARY SHOOT APICAL MERISTEM, YIELD COMPONENTS
AND YIELD OF BARLEY (HORDEUM VULGARE, L. EMEND. LAM.)

 

I. AN ASSESSMENT OF THE INVOLVEMENT OF
PROMOTIVE HORMONES
II. A MATHEMATICAL MODEL FOR YIELD

By
James Benjamin Abaka Nhyte

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements

for the degree of

DOCTOR OF PHILOSOPHY
Department of Crop and Soil Sciences

1981

.4 /'"‘.'/' ,.
L); x / ,

ABSTRACT

THE PRIMARY SHOOT APICAL MERISTEM, YIELD COMPONENTS AND
YIELD OF BARLEY (HORDEUM VULGARE, L. EMEND. LAM.)

 

I. AM ASSESSMENT OF THE INVOLVEMENT OF
PROMOTIVE HORMONES

II. A MATHEMATICAL MODEL FOR YIELD
By
James Benjamin Abaka Whyte

The study is in three parts. First, an assessment of
promotive hormones' involvement in the growth and develop-
ment of shoot apical meristem in relation with yield
components and yield of barley was made. Second, the use
of findings as the basis for reconceptualizing the
developmental allometry among the yield components and
yield. Third, assessment of the reconceptualized model
using data for twelve oat varieties planted between 1976
and l979, inclusive.

Twelve genotypes with differing yield components were
used in the experiments. Meristems were sampled at
transition, reproduction and elongation stages and measure-
ments-taken of their length, width and relative growth
rate. Additionally, yield, number of heads per unit area
(X), number of seeds per head (Y) and seed weight (2) were
determined. Hormone applications were done at transition

stage and two days after.

low
her
The
hor
tic

shc

to

We

fO
ef
Dr

ET

James Benjamin Abaka Nhyte

Genotypes with larger meristems had lower growth rate,
low X, high Y and bigger head size. Applications of the
hormones induced changes in the measured characteristics.
The magnitudes of the changes depended on the type of
hormone applied and genotype under consideration. Indica-
tions of existence of equilibrium among the hormones were
Shown.

The model involved the use of path coefficient analysis
to determine the effects of three hypothetically separable
independent environmental resources, E], E2 and E3, to
yield through their direct effect on X, Y and Z, respect-
ively. Two sets of environmental components (efficiencies)
were estimated from X, Y and Z.

Varietal constants were positive, highest for X
followed by Y and then Z. All varieties had similar
efficiencies in production of X and Y. Efficiencies for Z
production varied. All the varieties and seven of the
environments had significant yield predictions.

The relationship between the predicted and Observed

yields was linear and highly Significant.

IN MEMORY OF
MY GRANDMOTHER, NANA EFUA OTUA,
AND FATHER, JOHN KOFI ADOKO NHYTE.

SUbjEI
devot
I mou

study

0. Fe
for t

York

ment,

SUCCE

the u

York,

DTEpa

SUDDC

ACKNOWLEDGEMENTS

My interest in plant breeding, particularly on the
subject presented in this study, has been nurtured by the
devotion and the example of my mentor, Dr. J.E. Grafius.

I mourn his death which came before the completion of this
study.

I am indebted to Drs. D.D. Harpstead, M.w. Adams,

D. Penner and C. Pollard, members of my guidance committee
for their encouragement and constructive inputs into my
work as a graduate student.

Dr. R.L. Green, Dean of the College of Urban Develop-
ment, provided a great deal of support to make possible the
successful completion of my graduate studies. I am grateful.

My sincere appreciation goes to Mr. Dimon Wolfe for
the unqualified assistance he provided in my field research
work. Miss Patti Perkins' assistance with the typing and
preparation of the dissertation is acknowledged.

Finally, for their love, untiring encouragement, moral
support and the understanding they have always accorded me,

I am forever grateful to my FAMILY.

 

TABLE OF CONTENTS

35g;
LIST OF TABLES . . . . . . . . . . . . . . . . . . . v11
LIST OF FIGURES . . . . . . . . . . . . . . . . . . xv
INTRODUCTION . . . . . . . . . . . . . . . . . . . . 1
CHAPTER l: AN ASSESSMENT OF THE INVOLVEMENT OF
PROMOTIVE HORMONES . . . . . . . . . . . 3
Introduction . . . . . . . . . . . . . . . . . . 3
Literature Review . . . . . . . . . . . . . . . . 7
Theoretical prediction of the existence of
biochemical equilibrium in biological systems . 33
Materials and Methods . . . . . . . . . . . . . . 4o

Investigation of the levels and activities of

the promotive hormones and their effects on

growth and development of the primary shoot

apical meristem, yield components and yield

of barley . . . . . . . . . . . . . . . . . . . 40

A. Determination of the growth and develop-
ment of the primary shoot apical meristem . 40

B. Analysis of the levels and activities of
the endogenous promotive hormones . . . . . 42

C. Measurements of yield components and
yield of barley . . . . . . . . . . . . . . 43

Effects of manipulation of the endogenous

hormonal (promotive) levels on the growth and
development of the primary apical meristem,

yield components and yield of barley . . . . . 47

Results . . . . . . . . . . . . . . . . . . . . . 50
A. Relationship between the primary shoot

apical meristem, yield and components and
yield of barley . . . . . . . . . . . . . . 50

TABLE OF CONTENTS continued

B. The reaction of the primary shoot apical
meristem to applied hormones . . . . . . . 59

C. The reaction of yield components and yield
to applied hormones . . . . . . . . . . . 71

Discussion . . . . . . . . . . . . . . . . . . . llO
Summary and Conclusion . . . . . . . . . . . . . 127
Appendix .. . . . . . . . . . . . . . . . . . . . l3l
References . . . . . . . . . . . . . . . . . . . l4l
CHAPTER 2: A MATHEMATICAL MODEL FOR YIELD . . . . 154
Introduction . . . . . . . . . . . . . . . . . . 154
Theoretical Consideration . . . . . . . . . . . l56
Materials and Methods . . . . . . . . . . . . . . l64
Results . . . . . . . . . . . . . . . . . . . . . l68
Discussion . . . . . . . . . . . . . . . . . . . l92
Summary and Conclusion . . . . . . . . . . . . . 200
Appendix . . . . . . . . . . . . . . . . . . . . 204

References . . . . . . . . . . . . . . . . . . . 2l4

vi

TABLE

 

LIST OF TABLES

CHAPTER 1

1

Mean values for the maximum length (L) and
maximum width (ND) of the primary apical
meristems of barley in the transition
(Stage I), reproduction (Stage II) and
elongation (Stage IIIA and B) states of
development . . . . . . . .

Mean values for yield (N), number of seeds
per unit area (XY), head size (YZ), number
of heads per unit area (X), number of seeds
per head (Y), seed weight (Z), height (HT)
and heading (HD) of barley . . .

Correlation coefficients among the number
of heads per unit area (X), number of seeds
per head (Y), seed weight (Z), number of
seeds per unit area (XY), head size (YZ),
yield (N), height (HT), heading (HD)
maximum length (L) and maximum width (ND)
of the primary apical meristem of barley

in the transition (SI), reproduction ($11)
and elongation (5111) states of .develop-
ment . . .

Analysis of variance for the multiple
regression of the number of heads per unit
area (X), number of seeds per head (Y) and
head size (YZ), each as a dependent var-
iable on the maximum length (L), maximum
width (ND) and the relative growth rate (R)
of the primary apical meristem of barley

in the reproduction state of devel0pment

as the independent variables. . .

Multiple regression statistics for the
number of heads per unit area (X), number
of seeds per head (Y) and head size (YZ),
each as a dependent variable on the maximum
in length (L), maximum width (ND) and the
relative growth rate (R) of the primary
apical meristem of barley in the reprod-
uction state of development as the

vii

PAGE

51

52

53

58

LIST OF TABLES continued

TABLE
CHAPTER 1
5 independent variables. df = 8. Upper,

10

middle and lower values are the statistics
for X, Y and Y2, respectively .

Mean values for maximum length (L) and
maximum width (ND) of the primary apical
meristem of barley just before the transition
(S < I) and in the reproduction (511) states
of development .

Mean values for the percentage change over
control for the maximum length (4L) and
maximum width (AND) of the primary apical
meristem of barley in the reproduction

state of development due to the application

of gibberellin (6A3), cycocel (CCC), cytokinin
(Kinetin) and auxin (IAA) at the transition
stage of development . . . . . .

Mean values for the number of heads per unit
area (X), number of seeds per head (Y), seed
weight (Z), number of seeds per unit area
(XY), head size (YZ), yield (W), height (HT)
and heading (HD) of barley (control block)

Correlation coefficients among the number
of heads per unit area (X), number of seeds
per head (Y). seed weight (Z), number of
seeds per unit area (XY), head size (YZ),
yield (N), maximum length (L) and maximum
width (ND) of the primary apical meristem
of barley in the reproduction state of
development, height (HT) and heading (HD)
(control block) . . .

Correlation coefficients among the number
of heads per unit area (X), number of seeds
per head (Y), seed weight (Z), number of
seeds per unit area (XY), head size (YZ),
yield (N), maximum length (L) and maximum
width (ND) of the primary apical meristem
of barley in the reproduction state of

viii

PAGE

60

61

63

72

73

LIST OF TABLES continued

TABLE

 

CHAPTER 1

10

T)

12

13

development, height (HT) and heading (HD)
resulting from gibberellin (GA3) application
at the transition stage of development . . . . 77

Correlation coefficients among the number

of heads per unit area (X), number of seeds

per head (Y), seed weight (Z), number of

seeds per unit area (XY), head size (YZ),

yield (N), maximum length (L) and maximum

width (ND) of the primary apical meristem of
barley in the reproduction state of develop-

ment, height (HT) and heading (HD) resulting

from cycocel (CCC) application at the

transition stage of develOpment . . . . . . . 85

Correlation coefficients among the number of
heads per unit area (X), number of seeds

per head (Y), seed weight (Z), number of
seeds per unit area (XY), head size (YZ),
yield (N), maximum length (L) and maximum
width (ND) of the primary apical meristem of
barley in the reproduction state of develop-
ment, height (HT) and heading (HD) resulting
from cytokinin (Kinetin) application at the
transition stage of development . 92
Correlation coefficients among the number of

heads per unit area (X), number of seeds per

head (Y), seed weight (Z), number of seeds

per unit area (XY), head size (YZ), yield

(N), maximum length (L) and maximum width

(ND) of the primary apical meristem of barley

in the reproduction state of development,

height (HT) and heading (HD) resulting from

auxin (IAA) application at the transition

stage of development . . . . . . . . . . . . . 98

ix

LIST OF TABLES continued

TABLE

PAGE

CHAPTER 1

T4

15

)6

Correlation coefficients among the percentage
change over control for the number of heads
per unit area (AX), number of seeds per head
(AY), seed wei ht (AZ), number of seeds per
unit area (AXY), head size (AYZ), yield (AN),
maximum length (AL) and maximum width (AND)
of the primary apical meristem of barley in
the reproduction state of devel0pment due to
the application of gibberellin, cycocel,
cytokinin and auxin, respectively, at the
transition stage of development . . . . . . . 105

Correlation coefficients among the percentage
change over control for the number of heads
per unit area (AX), number of seeds per head
(AY), seed weight (AZ), number of seeds per
unit area (AXY), head size (AYZ), yield (AN),
maximum length (AL), and maximum width (AND)
of the primary apical meristem of barley in
the reproduction state of development due

to the application of gibberellin, cycocel,
cytokinin and auxin, respectively, for the
standard genotypes . . . . . . . . . . . . l08

Analysis of variance for the multiple
regression of the number of heads per unit
area (X), number of seeds per head (Y),
seed weight (2), head size (YZ), number of
seeds per unit area (XY) and yield (N) each
as a dependent variable on the maximum
length (L) and maximum width (ND) of the
primary apical meristem of barley in the
reproduction state of development as the
independent variables due to gibberellin,
cycoceT, cytokinin and auxin application

at the transition stage of development . . . l09

LIST OF TABLES continued

TABLE
APPENDIX 1
Al Mean percentage change over control for yield

A2

A3

A4

A5

(AN), number of seeds per unit area (AXY),
head size (AYZ), number of heads per unit
area (AX), number of seeds per head (AY) seed
weight (AZ), height (AHT) and heading (AHD)
of barley due to gibberellin application at
the transition stage of development

Mean percentage change over control for yield
(AN), number of seeds per unit area (AXY),
head size (AYZ), number of heads per unit
area (AX), number of seeds per head (AY),
seed weight (AZ), height (AHT) and heading
(AHD) of barley due to cycocel application at
the transition stage of development . .

Mean percentage change over control for
yield (AN), number of seeds per unit area
(AXY), head size (AYZ), number of heads per
unit area (AX), number of seeds per head
(AY), seed weight (AZ), height (AHT) and
heading (AHD) of barley due to cytokinin
application at the transition stage of
development .

Mean percentage change over control for
yield (AN), number of seeds per unit area
(AXY), head size (AYZ), number of heads per
unit area (AX), number of seeds per head
(AY), seed weight (AZ), height (AHT) and
heading (AHD) of barley due to auxin
application at the transition stage of
development . . .

Multiple regression statistics for the
number of heads per unit area (X), number
of seeds per head (Y), head size (Y2) and
yield (N), each as a dependent variable
on the maximum length (L) and maximum
width (ND) of the primary apical meristem
of barley in the reproduction state of
development as the independent variables.

xi

PAGE

T31

T32

T33

T34

LIST OF TABLES continued

TABLE

 

PAGE

APPENDIX T

A5

A6

A7

A8

The statistics are in the order of X, Y, Y2
and N, respectively (control block) . . . . . l35

Multiple regression statistics for the number

of seeds per head (Y), number of seeds per

unit area (XY), head size (YZ) and yield (N),

each as a dependent variable on the maximum

length (L) and maximum width of the primary

apical meristem of barley in the reproduction
state of development as the independent

variables. The statistics are in the order

of Y, XY, Y2 and N, respectively (cycocel

block) . . . . . . . . . . . . . . . . . . . l36

Multiple regression statistics for the number

of seeds per head (Y), number of seeds per

unit area (XY) and yield (N), each as a

dependent variable on the maximum length (L)

and maximum width (ND) of the primary apical
meristem of barley in the reproduction state

of development as the independent variables.

The statistics are in the order of Y, XY and

N, respectively (cytokinin block) . . . . . . 137

Multiple regression statistics for the number
of heads per unit area (X), number of seeds
per head (Y), number of seeds per unit area
(XY), head size (YZ) and yield (N), each as a
dependent variable, on the maximum length (L)
and maximum width (ND) of the primary apical
meristem of barley in the reproduction state
of development as the independent variables.
The statistics are in order of X, Y, XY, Y2

and N, respectively (auxin block) l38

xii

 

TABLE PAGE
CHAPTER 2

l Analysis of variance table for two-factor

mixed model study . . . . . . . . . . . . . .160
2 Names of oat varieties . . . . . . . . . . .165
3 Description of the thirteen environments . .166
4 Correlation coefficients among yield (N),

number of heads per unit area (X), number of

seeds per head (Y) and seed weight (Z) for

each of the varieties . . . . . . . . . . . .169
5 Path coefficients between the yield components

and yield for each of the varieties . . . . .170
6 Mean square values for yield (N), number of

heads per unit area (X), number of seeds per

head (Y) and seed weight (Z) of 12 varieties

tested over 13 environments . . . . .171
7 Arrangement of the varieties in the order of

increasing mean values for yield (N), number

of heads per unit area (X), number of seeds

per head (Y) and seed weight (Z) . . . . . .173
8 Arrangement of the environments in the order

of increasing mean values for yield (N),

number of heads per unit area (X), number

of seeds per head (Y) and seed weight (Z) . .174
9 Estimates of the varietal constant component

of the variety-environment interaction . . .182
10 Correlation coefficients among varietal mean

LIST OF TABLES

number of heads per unit area (XL), number

of seeds per head (Yi.), seed weight (Zi )

and estimates of the varietal constants

(Vi , V2, V3) of the variety- -environment
interaction . . . . . . . . . . . . . . . . .184

xiii

LIST OF TABLES continued

TABLE

11

12

Estimates of the environmental components
of the variety-environment interaction for
the varieties . . . . . . . . .

Estimates of the environmental components
of the variety-environment interaction for
the environments

APPENDIX 2

A1

A2

Analysis of variance for the prediction of
the observed yield (dependent variable) by
the three multiplicative terms of variety-
environment interaction (independent
variables) involving the production of the
number of heads per unit area (X), number
of seeds per head (Y) and seed weight (Z),
respectively . .

Multiple regression statistics for the
prediction of the observed yield (dependent
variable) by the three multiplicative terms
of variety-environment interaction
(independent variables) involving the
production of number of heads per unit area
(X), number of seeds per head (Y) and seed
weight (2), respectively . . . .

PAGE

185

187

204

205

FIGURE
CHAPTER

LIST OF FIGURES

l
Geometrical derivation of the diffusion

equation
Development of a standing wave
Diagram of barley shoot apex

Developmental allometry showing the influ-
ence of the primary apical meristem of barle
in the reproduction state of development

on the yield components . . .

Regression of the number of seeds per head
on the relative growth rate of the primary
apical meristem of barley in the reproductio
state of development . . . . . . . . .

Regression of head size on the relative
growth rate of the primary apical meristem
of barley in the reproduction state of
development. . . . . . . . . . . .

Percentage change in maximum length and
maximum width of the primary apical meristem
of barley in the reproduction state of
development due to gibberellin (GA3)
application at the transition stage of
development . . . . . . . .

Percentage change in maximum length and
maximum width of the primary apical meristem
of barley in the reproduction state of
development due to cycocel (CCC) application
at the transition stage of development

Percentage change in maximum length and
maximum width of the primary apical meristem
of barley in the reproduction state of
development due to cytokinin (Kinetin)
application at the transition stage of
development . . . . . . . . .

XV

35
39
42

Y
46

n
55

56

65

67

69

LIST OF FIGURES continued

FIGURE

10

11

12

13

14

15

16

17

18

Percentage change in maximum length and
maximum width of the primary apical meri-
stem of barley in the reproduction state
of development due to auxin (IAA)
application at the transition stage of
development

Regression of the number of seeds per head
on the number of heads per unit area of
barley (control block)

Comparison of regressions of number of
seeds per head on number of heads per unit
area of barley due to GA3, CCC, Kinetin
and IAA application at the transition stage
of development with a control . . .

Regression of the number of seeds per head
on the number of heads per unit area of
barley due to GA3 application at the
transition stage of development

Percentage change in yield, head size and
number of seeds per unit area of barley
due to GA3 application at the transition
stage of development

Percentage change in number of heads per
unit area, number of seeds per head and

seed weight of barley due to GA3 application

at the transition stage of development

Percentage change in heading and height of
barley due to GA3 application at the
transition stage of development

Regression of the number of seeds per head
on the number of heads per unit area of
barley due to CCC application at the
transition stage of development

Percentage change in yield, head size and
number of seeds per unit area of barley
due to CCC application at the transition
stage of development . . . . .

xvi

PAGE

70

75

76

79

81

83

84

87

88

LIST OF FIGURES continued

FIGURE

19

20

21

22

23

24

25

26

27

Percentage change in number of heads per
unit area, number of seeds per head and

seed weight of barley due to CCC application

at the transition stage of development

Percentage change in heading and height of
barley due to CCC application at the
transition stage of development

Regression of the number of seeds per head
on the number of heads per unit area of
barley due to kinetin application at the
transition stage of development

Percentage change in yield, head size and
number of seeds per unit area of barley due
to kinetin application at the transition
stage of development

Percentage change in number of heads per
unit area, number of seeds per head and
seed weight of barley due to kinetin appli-
cation at the transition stage of
development

Percentage change in heading and height of
barley due to kinetin application at the
transition stage of development

Regression of the number of seeds per head
on the number of heads per unit area of
barley due to IAA application at the
transition stage of development

Percentage change in yield, head size and
number of seeds per unit area of barley
due to IAA application at the transition
stage of development . . .

Percentage change in number of heads per
unit area, number of seeds per head and

seed weight of barley due to IAA application

at the transition stage of development

xvii

9O

91

93

94

95

97

99

101

102

LIST OF FIGURES continued
FIGURE PAGE

28 Percentage change in heading and height
of barley due to IAA application at the
transition stage of development . . . . . . . 104

29 Diagrammatic representation of sequence of
events during the growth and development
of the apical meristem . . . . . . . . . . . 122

30 Diagrammatic representation of the
sequential development of the yield
components and yield of barley . . . . . . . 125
APPENDIX 1
Al Developmental allometry showing the influence

of the primary apical meristem of barley in
the reproduction state of development on the
yield components . . . . . . . . . . . . . . 139

A2 Regression of the number of seeds per head
on the number of heads per unit area
(control block) with its 95% confidence belt
(variance of a predicted value) . . . . . . . 140

xviii

FIGURE

LIST OF FIGURES

CHAPTER 2

1

7

A causation diagram showing the developmental

relationships between yield and the yield
components . . . .

Yield response of Mackinaw, Orbit, Nright,
Cld 64, and Menominee to thirteen varying
environments in Michigan, 1976-1979

Response for the number of heads per unit
area of Mackinaw, Orbit, Nright, Cld 64 and
Menominee to thirteen varying environments
in Michigan, 1976-1979 . . .

Response for the number of seeds per head
of Mackinaw, Orbit, Nright, Cld 64 and
Menominee to thirteen varying environments
in Michigan, 1976-1979 . .

Seed weight response of Mackinaw, Orbit,
Nright, Cld 64 and Menominee to thirteen
varying environments in Michigan, 1976-
1979 . . . . .

Regression of predicted yield on observed
yield . . . . . .

Regression of residuals on predicted yield

APPENDIX 2

A1

A2

A3

Regression graphs for the yield response of
the twelve varieties to thirteen varying
environments in Michigan, 1976-1979

Regression graphs for the response for the
number of heads per unit area of the twelve
varieties to thirteen varying environments
in Michigan, 1976-1979 . . . . .

Regression graphs for the response for the
number of seeds per head of the twelve
varieties to thirteen varying environments
in Michigan, 1976-1979 . .

157

176

178

179

180

189

191

206

208

210

LIST OF FIGURES continued

FIGURE PAGE

A4 Regression graphs for the seed weight
response of the twelve varieties to thirteen
varying environments in Michigan, 1976-
1979 . . . . . . . . . . . . . . . . . . . . 212

XX

COT

of

VI:

co
lh
co
51
re
ti

he

INTRODUCTION

Yield is a complex trait and in barley, yield has three
components; the number of heads per unit area (X), the number
of seeds per head (Y) and the seed weight (Z). Changes in
yield are obtained through changes in one or more of the
components, however, simultaneous maximization of the yield
components for yield improvement has not been possible.

This is due to the negative associations among the yield
components and when these negative correlations are reduced,
significant improvements can be obtained. There has been
reports of yield improvements induced by the relaxation of
the significant negative correlation between the number of
heads per unit area and the number of seeds per head.

The level of number of heads produced per unit area has
a great effect on the sizes and numbers of the later-formed
organs owing to its direct association with the shoot apical
meristem size. This association can be modified by external
factors such as nutrients, light,water and temperature or
possibly through hormonal manipulations.

In altering the yield components and yield through the
manipulation of the hormonal levels, other plant character-
istics such as height, leaf area, growth rate, head size,

tillering,to mention but a few, change. Yield is either

reduced or increased, depending on the relationships among
the above mentioned plant characteristics with it. This
investigation was initiated to assess the involvement of
the promotive hormones in the development of the primary
shoot apical meristem, yield components and yield of barley
genotypes with differing yield components and yield. A
second objective was to determine the reactions of these
genotypes to application of some hormones.

It is documented that different genotypes perform
differently in various environments. The relationship between
the performances of the different genotypes in the various
environments and some measure of these environments
(environmental indices) is frequently linear or nearly so.
Using the concept of sequential development of the yield
components, the developmental allometry, and the proposition
that the yield components of cereal crops are produced at
different stages in the ontogeny of plants, a model is
presented for yield. This model provides an insight into
the genotype-environment interaction involved in the product-
ion of each component of yield toward the genotypic yield

performances.

All!

com;
numl
0rd.
hea
wei
as

com
its
wit

Ham

of
rear
1m;
0n

COL

CHAPTER 1
AN ASSESSMENT OF THE INVOLVEMENT OF PROMOTIVE HORMONES
INTRODUCTION

In barley, the complex trait, yield (N) has three
components: the number of heads per unit area (X), the
number of seeds per head (Y) and the seed weight (Z). The
order of development of the yield components are number of
heads per unit area to number of seeds per head to seed
weight. Geometrically, yield of barley can be expressed
as a volume of a rectangular parallelepiped with its
components as the edges. Yield changes when one or more of
its components change and the greatest change is obtained
with a change in its shortest edge (Grafius, 1956; 1964).
Hamid and Grafius (1978) showed that the earlier developed
yield components have a profound influence on the later
developed structures. Thus, the genetic control of yield is
indirectly channeled through its components, with the earlier
formed structures assuming the major part of the control.

The negative correlations among the yield components
of several crop plants as established by Adams (1967) has
rendered the maximization of X, Y and Z, simultaneously,
impossible. This has resulted in the imposition of a ceiling
on grain yield. Relaxation of these negative correlations

could result in great yield increases in such crop plants.

Grafius gt 11. (1976) reported such a relaxation in the X
and Y relationship in barley. A higher value of Y for a
given value of X was obtained, resulting in increased yield
of the unselected progeny over the better parent.

The develOpment in a plant of trait X at any one level
triggers a chain of reactions which affect all later formed
organs as shown by Sinnott's Law (1921). Sinnott stated
that, 'The size of any organ depends upon the size of the
growing point out of which it has been developed'. By virtue
of the direct association between X and meristem size, X
assumes a pivotal role in determining sizes of plant organs
formed later and eventually the determination of grain yield.

The conversion of the barley apical meristem from the
vegetative to the reproductive condition coincides with the
cessation of tiller and leaf bud development and the
initiation of the floral primordia and then kernel formation.
This dramatic switch exerts a direct effect on the relation-
ships among the yield components. The dependence of X on
meristem size and the effect that X has on organs formed
later in the ontogeny of the plant could be modified by
external factors (Aspinall, 1961; 1963; Cannel, 1969; Friend,
1965; Nardlaw, 1971) or internal factors (Leopold, 1949).

Improvement of crop productivity through plant breeding

has been mainly achieved through the manipulation of plant

cha
eff
ner
hm
M.
de
st
ia

al

I?

hc

characteristics to utilize environmental factors with greater
efficiency. Physiological and biochemical means of improve-
ment is a more recent objective and is being encouraged,
however, the frequency of achievements is quite slow. The
physiological and biochemical processes occurring during the
development of a plant is so integrated that an equilibrium
state is established at all times during growth, different-
iation and development. Changing the internal equilibrium
alters the final resulting product and the extent of this
alteration in relationship to grain yield is dependent on
the degree of association between the two.

Attempts have been made to increase yield through
hormonal application. Unfortunately, much of the reported
results are contradictory. In the quest for altered yield,
other plant characteristics such as height, leaf area, stem
diameter, growth rate, seed size, tillering, head size, etc.,
also change. Depending on the association between these
changes and yield, yield is either reduced or increased.

The following study seeks to:

1. Investigate the levels and activities of the
promotive hormones and to relate the observed differences to:

a. Growth and development
b. Yield components

c. Yield

sta
the
det

C0111

2. Synchronize the growth and development of the
standard genotypes with that of X969-3 (an outlier) through
the manipulation of their endogenous hormonal levels and to
determine the effect that such manipulation has on yield

components and yield.

LITERATURE REVIEW

All plants follow a developmental rhythm. Small grains
such as oats, barley, wheat, millet and sorghum start by
laying down tillers followed by floral initials, stem
elongation and cessation of tillering, pollination, filling
and maturation of kernels. The phases of tillering, floral
initiation and maturation extend over the ontogeny of the
plant and are directly related to the yield components.

Grafius (1969), Grafius and Thomas (1971) and Thomas
_t _l. (1971a, b, c) elaborated upon the concept of a
sequential developmental process of yield components. The
chronological developmental sequence of the yield components
of barley is number of heads per unit area (X) to number of
kernels per head (Y) to kernel weight (Z). Yield (N) is a
multiplicative product of the components, i.e. N=XYZ. Yield
is subject to change through change in one or more of its
components.

Yield components are determined at different stages in
the ontogeny of a plant (Rasmusson and Cannell, 1970) and
are differentially affected by variation in the environment
(Tai, 1975). This suggests that the three yield components
in barley are affected by independent environmental factors
during the same or different periods of the plant's develop-

ment. The formation of yield components in sequence results

the
com]
3011
ava
tra
inf
but
pre
are
0811
tra
cor
Har
do'
va

00

be

to

in a different relationship between a component trait and

the environmental resources. The develOpment of the first
component trait, i.e. number of heads per unit area, is
solely determined by genetics and the environmental resources
available during the early stage of growth. A component
trait which develops subsequent to others is not only
influenced by the resources available during its formation,
but also by the development and characteristics of its
predecessor. The mechanisms for controlling yield components
are thus increasingly complicated in the chronological
developmental sequence. Thomas _t _l. (1971a) proposed a
transformation to factor out the effect on yield of a
component trait appearing earlier in the development sequence.
Hamid and Grafius (1978) showed that the plant organs laid
down early in the sequence exert more genetic control over
variation in yield than traits laid down later in the
ontogeny.

Adams (1967) showed the existence of negative correlations
between the yield components of several crop plants.
Correlations among yield components may be due to genetic
linkages, pleiotropy or physiological develOpmental
relationships. These negative correlations have posed a
block to yield improvement of crop plants. The simultaneous

maximization of X, Y, and Z is prevented and as a result a

cei
tori
cm
of
of
yie
rel
hea
so

DO?

Dr

130

ceiling of yield is created. Relaxation of these negative
correlations can result in great yield increases of such
crop plants. The short statured hexaploid wheats, derivatives
of Norin 10 cultivar, outyield the standard wheats as a result
of the relaxation of the negative correlations between the
yield components. Grafius gt _1. (1976) reported the
relaxation of the negative correlation between the number of
heads per unit area and number of seeds per head in barley
so that a higher value of Y for a given value of X was
possible. This characteristic was inherited by some progeny
lines with resulting increased yields of its unselected
progeny over the best parent in one of the backcross
populations - after selfing several generations.

The analysis of crop yield entails the analysis of
plant growth. The attainment of the characteristic form
and function in a crop plant depends upon a chain of inter-
related events which are sequential in time, gene regulated
at critical sites and times and subject to modifying
influences of the environment. The events follow an
integrated pattern (Adams, 1967). Yield is an example of
integration in which the components of seed yield are to
some extent interdependent in their development.

The develOpment of organs in plants is controlled by
developmental allometry (Sinnott, 1921, 1960; Bonnett, 1964).

Each part and function is so closely related with the rest

that

am

EDC

Ce
81

Ol

10

that the whole plant develops in an orderly fashion to produce
a mature individual. Adams (1975) points out the phenomenon
of size and numbers as part of the overall allometry in a
plant. He showed significant relationships between number

of pods per plant vs. main stem node number and seed size

vs. leaf size in Phaseolus vulgaris. High yield potential

 

is achieved by a balance between 'factors of numbers'

(e.g. number of nodes) and factors of size (e.g. stem
diameter, leaf area). Size and number of appropriate
components of yield (N) may be more critical than the size
or number of the photosynthetic surfaces in causing differ-
ences in yield of genotypes (Evans and Dustone, 1970; Khan
and Tsunoda, 1970; Berdahl gt gt., 1972; Hamid and Grafius,
1978).

The primordia of organs evolve from meristems and the
central role of these structures has been pointed out by
Sinnott (1921). Sinnott stated that, ”The size of any given
organ depends upon the size of the growing point out of
which it has been developed". An important feature of
apical meristems of small cereal grain plants is that they
undergo both vegetative and reproductive phases of growth.
The vegetative growth period involves the formation of
tiller and leaf primordia. The change of meristem from the

vegetative to reproductive stage coincides with the cessation

611C
05‘.

de‘

th.

DE

11

of the development of tiller buds and leaves and the onset
of initiation of reproductive structures. The switch from
vegetative to reproductive phase exerts a direct effect on
the relationships among the yield components of small cereal
grain plants.

The level of X affects all later-formed organs (Hamid
and Grafius, 1978). By virtue of its direct negative
association with meristem size, X assumes a pivotal role in
determining sizes of plant organs and eventually the
determination of economic yield itself. The importance of
the number of heads per unit area (X) is demonstrated in a
path coefficient diagram developed in conformity with
Sinnott's Law and known developmental relationships (Hamid
and Grafius, 1978; Grafius, 1978).

The relationship between size of meristem and size of
plant organ was first recorded by Sinnott (1921). Since
then, others have noted this relationship for a wide range
of crops (Abbe gt gt., 1941; Stant, 1954; Aitken, 1967;
Quinby, 1970; Fisher, 1973; Blum, 1977; Nhyte, 1979). The
relationships among plant characteristics appears to be more
allometric than genetic. Genetic differences in leaf size
in barley do exist, but only minimal genetic variance will
be associated with variation between areas of leaves on the

same culm. Instead, the primary genetic variance will be

dSS

fro

to

V8

Le

01

12

associated with factors governing the size of the meristem
from which the culm, leaves and glumes have arisen.

Nhyte (1979) traced the origin of the relaxation of the
negative correlation between X and Y (Grafius gt gt., 1976)
to two major factors:

1. An initially broader meristem

2. A 4-day time lapse period between the end of
vegetative stage and the onset of the reproductive stage.
Lee gt gt. (1974), Nilliams (1975), Fisher (1973) and Blum
(1977) established similar relationships in detecting the
origin of the heterotic effect in the sorghum panicle.

Grafius (1978) introduced a dichotomy to the ideas of
Sinnott: allometry associated with organs arising from
either (a) the same meristem or (b) different meristems.

Allometric relationships among X, Y and Z might result
from competition rather than from the effects of common
origin. The stress matrix (Grafius, 1978) established be-
tween the components varies with the environment and the
gene pool. Linkage may be present but its effect is less
important if one assumes that genes for the components are
distributed throughout the chromosomes. Allometric
relationships among traits not arising from the same meristem
could also be brought about by the need for structural

balance and hormonal stimulation in addition to competition

pm
the

sep

COT
inc
tir

51';

th
111
as

te

13

for environmental resources (Adams, 1967; Hoen and Andrew
1959; Grafius, 1969).

Grafius (1978) proposed that, "Plasticity is inversely
proportional to ontogenetic proximity". Events arising from
the same meristem are harder to manipulate than those
separated in space and time. A second corollary states that
size of the organ and numbers of organs are negatively
correlated. Fowler and Rasmusson (1969) showed a diminishing
correlation between leaves borne on the same culm with
increase in distance between the leaves (both in space and
time of origin). Attempts to select for different leaf
sizes on the same culm were futile.

The dependence of X on meristem size and the effect
that X has on later-formed plant organs is quite intriguing.
This dependence could be modified by external factors such
as nutrients (Aspinall, 1961, 1963), water (Nardlaw, 1971),
temperature, light intensity and duration (Cannell, 1969;
Friend, 1965) or internally by hormone levels (Leopold, 1949).

Five distinct groups of compounds have been character-
ized as hormones. Auxins, gibberellins and cytokinins are
the promotive hormones while abscisic acid is regarded as
an inhibitory hormone. Ehtylene, however, acts both as a
promotory and an inhibitory hormone. The proportions of the

various hormones present appear to affect the growth rate and

14

subsequent differentiation pattern of a tissue in the complete
organism. The presence of both promotive and inhibitory
hormones permits a precise control of many developmental
activities.

Tata (1966) hypothesized that hormone action lies in
the control of the mechanisms by which enzymes are made in
the cell. The points at which they act include:

1. The genetic information in the cells which determine
their ultimate potential.

2. The transcription of messenger RNA (mRNA) from DNA
specifying the type of protein to be produced.

3. The machinery for protein synthesis involving the
ribosomes, transfer RNA (tRNA) and other regulatory substances.

4. Post-transcriptional processes.

Genes do exist in different states of activity and the
state of any particular gene is important for hormonal
activation or inactivation.

Evidence indicates that gibberellin is involved in
extension growth of plant tissues and has been cited as the
mechanism for bolting in rosette plants. Gibberellin treat-
ment of intact plants causes enhanced elongation of existing
internodal cells and also increases the number of cells
present in each internode principally as a result of an

increase in mitosis in the subapical region of the stem.

15

The magnitude of the stem elongation response to gibberellin
varies from species to species and from variety to variety
within a species. Jones and Phillips (1966) showed a direct
correlation between elongation rate of internodes of different
ages and the gibberellin contents of the same internodes.
Endogenous gibberellins are present in highest concentration
in those regions undergoing most rapid extension growth,
providing strong evidence that gibberellin is concerned in
normal control of stem extension growth. Radley (1970)

found that semi-dwarf wheats contain more endogenous
gibberellin-like activity than normal varieties, particularly
in their young stage. Application of gibberellic acid markedly
stimulates the growth of seedlings of normal varieties but

not the dwarf seedling. She, therefore, assumed that a

block to the utilization of gibberellin causes the accumula-
tion of the hormone.

The actual role of gibberellin in promoting stem
elongation is not known. Experimental evidence show that
nucleic acid metabolism is involved in the process. Lang
_t _t. (1967) showed that protein synthesis is required for
growth induced by gibberellic acid.

Gibberellic acid (GA3) causes an increase in activity
of two hormones required for the synthesis of lecithin, a

major component of cell membranes (Ho and Varner, 1974).

16

GA3 causes ggfggtg_synthesis of several hydrolytic
enzymes in barley aleurone layer. Since RNA synthesis is
required for this hormone effect, the synthesis of the
enzymes is dependent on the synthesis of their mRNA.

Higgins gt 1. (1976) showed a positive correlation between

the rate of a-amylase synthesis 1

vivo and the level of

 

translatable mRNA for a-amylase. The hormone is required
throughout the period of enzyme synthesis; its removal
causes the level of enzyme production to return to that of
the aleurone layers not treated with gibberellic acid.

a-amylase production is inhibited by inhibitors of
oxidative phosphorylation and protein synthesis. Abscisic
acid also antagonizes GA3-mediated hydrolase synthesis
through the prevention of the GA3 effect on poly(A)-RNA
synthesis it 11332:

The above evidences suggest that the expression of the
gibberellin effect may require the synthesis of enzyme
specific RNA molecules during transcription.

Cytokinins or plant cell division factors, are chemical
substances which have the ability to induce cytokinesis in
cultured plant tissues (Skoog gt gl., 1965). At the same
time, cytokinins evoke a number of physiological responses,
such as promotion of cell enlargement and delay of leaf

senescence, which do not involve cell division (Skoog gt al.,

17

1970; Kende, 1971). This suggests that the cytokinins might
not act directly to trigger the events of cytokinesis, but
rather that cytokinesis is a secondary result of some bio-
chemical role played by the hormone in plant cell metabolism.
Fosket and Short (1973) followed changes in protein
content and cell proliferation activity after a cytokinin-

requiring strain of cultured Glycine max tissue was transferred

 

to freshly prepared media which either contained or lacked
cytokinin. After two days, no further increase was observed
in the absence of cytokinin. Cell population attained during
the first six days was a function of the cytokinin concentra-
tion of the culture medium.

There is a temperature-dependent lag phase between
auxin application and the resulting growth response of oat
coleoptile sections (Rayle and Cleland, 1970; Nissl and
Zenk, 1969). This lag indicates that auxin does not act
directly on growth but on some process which later alters the

growth rate (Ray gt 1., 1962). The intermediate process is

sensitive to inhibitors of both protein and RNA synthesis

1., 1963), indicating the necessity of these

(Nooden gt
substances for auxin action. Auxins cause an increase in
the synthesis of RNA and protein (Key gt _t., 1964; Trewavas,

1968). This effect is counteracted by antiauxins.

Actinomycin D and S-fluouracil inhibit RNA synthesis

wit
Key
of

sut
the
of

grc

mar

Sy

18

without affecting auxin induced growth (Nooden, 1968; Lin and

Key, 1967; Ingle gt 1., 1965). Kinetic studies of the effect

of actinomycin D on growth show that there is a 'pile of
substances' on which auxin acts but this pile is absent in

the absence of auxin (Penny and Galston, 1966). Regardless

of when the inhibitor is applied, auxin application induces
growth before the effect of the transcription inhibitor is
manifested. Thus, mRNA specific for auxin induced growth
exist which is translated into protein upon application of
auxin. As the mRNA is used up, new mRNA is produced for auxin
induced growth to continue and it is this process which is
sensitive to actinomycin D.

Auxin brings about cell extension by increasing the cell
wall plasticity. This increase takes place in two stages
(Penny and Galston, 1966):

1. An early effect which involves loosening and break-
ing the physical bonds.

2. A later effect which is dependent on RNA and protein
synthesis.

The early effect is mediated at least in part by hydrogen
ion (H+) secretion (Hager gt _t., 1971; Rayle _t__t., 1972;
Rayle, 1973). Hager gt gt., (1971) stimulated growth through
acidification of the cell wall and concluded that auxin may

act by activating adenosine triphosphatase which pumps HT

into the cell wall. Jacobs and Ray (1976) reported a similar

and
gro
Ray
c011
two

01‘

at

aux
syr

60C

Ira

Dl‘t

Shf

am

19

increase in growth rate with auxin and externally applied H+
and also realized that before the initiation of the enhanced
growth rate, there was reduction in pH in both maize and pea.
Rayle gt gt. (1970) monitored the changes in cell wall
composition as a result of auxin application and proposed

two possible modes of action of H+ whether induced by auxin
or externally applied:

1. Chemical hydrolysis of cell wall polymers.

2. Activation of enzymes which are otherwise inactive
at the neutral pH.

At present, the major concept of the mode of action of
auxins is that they depress certain genes and induce mRNA
synthesis. This results in the synthesis of new enzymes,
and ultimately the cell wall, allowing cells to expand.
Induction of protein synthesis by a chemical effector via
transcription and translation of DNA is a time consuming
process and is preceded by a characteristic lag phase
(Branscomb gt _t., 1968). Nissl _t _t. (1969), however,
showed that the lag phase can be gradually shortened to zero
and the rate is hormone concentration and temperature
dependent.

Publications dealing with the relationship between

hormonal contents and activities, meristematic growth and

development, yield components and yield of cereals are lacking.

20

The develOpmental morphology of the shoot apex of
gramineous species is well documented (Bonnett, 1935, 1936,
1937, 1966; Sharman, 1947; Bremer-Reinders, 1958). The spike
is a branched system bearing fertile and infertile branches.
However, in barley, certain differences are noted when its
inflorescence are compared with other members of the graminae.
The main shoot axis does not differentiate into a terminal
spikelet, primary branches are terminated by an ovary and no
glume primordia are formed on the rachis.

In the mature cereal seed, there is already a well-
developed shoot with 3 or 4 leaf initials and an apical dome
enclosed within the coleoptile. As the plant grows, the
apical dome initiates acropetally a succession of primordia
of which the first three to ten form leaves. Each primordium
unit may later differentiate an elongated internode and an
axillary bud (tiller bud). The primordia subsequently
initiated on the main shoot develop to form floral parts.

In barley (Bonnett, 1966) and wheat (Barnard, 1964), as the
primordium develops the leaf initial is more or less complete-
ly suppressed and the axillary portion differentiates to form
single, many-flowered spikelets (in wheat) or three single
flowered spikelets (in barley). The transition from leaf

to floral development of the shoot apex is accompanied by

changes in the growth rate (Barnard, 1964; Nilliams, 1964,
1974).

21

Two distinct phases of growth exist in the rachis of the
barley inflorescence (Nicholls and May, 1963): 1. Formation
of primordia acropetally accompanied with little or no change
in the mean interprimordial distance. Cell number per
segment remains relatively constant. Cell length is maximum
at the end of this phase.

2. Cessation of primordia formation and elongation of
the apex due to an increase in rachis interprimordial length.
This is accompanied by increased cell division and the
appearance of stamen initials (Nicholls and May, 1963) or
awns (Aspinall, 1966; Kirby and Faris, 1970). Evidence is
provided (Nicholls and May, 1963; Paleg and Aspinall, 1964;
Kirby and Faris, 1970; Evans, 1971; Nicholls, 1974a) that
suggests that at least two factors are involved in this
growth phase: a factor produced in the inflorescence at
floret initial formation and a factor produced during long
days. Both of these factors must be present for rapid rachis
internode elongation to be initiated. If one of these is
missing for genetic or physiological reasons, then rachis
internode elongation is delayed.

Also evident during these phases are changes in the
apical dome. Increase in size of the apical dome occurs
during the vegetative growth period in barley (Fellipe and

Dale, 1973; Kirby, 1977) wheat (Nilliams, 1974) and maize

22

(Abbe and Phinney, 1951). Maximum dome size occurs when the
maximum number of floral primordium has been produced and
awns initiated, after which the size of the dome declines
(Kirby, 1974a). Most of the spikelets which are initiated
during the reduction process die (Kirby, 1977). Changes in
dome size appear to be related to the size and number of
primordia which are initiated and hence the number of grains
in the mature head. The developmentally most advanced spike-
let bud is found near the middle of the inflorescence
(Sharman 1947; Kirby, 1974a). Sharman (1947) speculated that
this pattern might be due to the fact either that each
successively initiated spikelet bud starts its development
earlier in the history of the node with which it is associated
or that it might develop at a greater rate. Kirby (1977)
showed that each successive primordium initiated had a
higher relative growth rate than the previous one in the
basal two-thirds of the apex. There was an increase in
length and diameter of each successive primordium initiated
until the middle of the head after which there was a decline
in diameter and volume. Differences in the growth rate and
in the inital sizes of the primordia on the floral apex det-
ermine primordium number and the gradient of grain size and
spikelet fertility which occurs in the mature head(Kirby,

1974a).

23

There is no detectable vascular system in the barley
inflorescence until after internode elongation has been
initiated (Nicholls and May, 1963; Kirby and Rymer, 1974).
This implies that the transport of nutrients from the base of
the inflorescence to the apical and lateral meristems is by
diffusion until the vascular system develops. Under these
conditions, the apical meristem ceases activity and dies
because the developing lateral spikelets compete more success-
fully for the available nutrients (Kirby and Faris, 1970).
The increase in the relative growth rate in the elongation
phase of the barley meristem coincides with the establishment
of vascular connections to the spikelet primordia. At this
stage it may be supposed that the young head is no longer at
a disadvantage in relation to the other organs of the plant.
The death of the terminal spikelets may be the result of a
high resistance to assimilate transport, leading to starva-
tion because of competition from the basal and mid spikelets
of the head which have an established vascular system (Kirby
and Rymer, 1974). Growth rate might be regulated hormonally
since a high concentration of gibberellin-like substances
coincides with the increased growth rate of the shoot apex
(Nicholls and May, 1964). Vascularization may be the
consequence of this growth rate rather than the cause of it

(Kirby and Rymer, 1974).

24

The initiation of internode elongation has been
correlated with peaks in activity of gibberellin-like
substances extracted from long-day and cold requiring
dicotyledonous rosette plants (Harada and Nitsch, 1959; Lang,
1960; Reinhard and Lang, 1961). It is also induced by
applying certain photoperiodic regimes or gibberellic acid
in a number of long-day plants (Sach gt gt., 1959; Baldev
and Lang, 1965; Jacqmard, 1968; Liu and Loy, 1976). Increased
growth of the apex and floral initiation have been observed
when GA3 is applied to plants growing in short days (Koller
gt gt., 1960; Evans, 1964, 1969; Hurd and Purvis, 1964).
Nicholls and May (1964) found a higher concentration of
gibberellin-like substances in the apex at the double ridge
stage in plants grown in 24-hr light periods compared with
plants grown in 8 hr conditions. This is compatible with
known biological properties of gibberellic acid, namely,
increased cell enlargement in treated plants (Brian, 1961)
and increased growth rates of gramineous apices following
gibberellic acid application (Purvis, 1960; Banbat and
Ochesanu, 1963). Nicholls and May (1964) suggest that at
the double ridge stage of development the supply of gibberellin-
like substances limits growth of the apex. At the spikelet

initial stage, however, a higher concentration of

gibberellin-like substances was observed. The difference in

25

concentration of gibberellin-like substances between apices
harvested at the stamen-initial stage and those harvested
later may reflect a difference in the rate of utilization.
There was a greater decrease in gibberellin concentration in
24-hr plants, with higher growth rate, than in 8-hr plants.

The growth and development of the barley spike is
inhibited by weekly applications of large amounts of GA3
(Paleg and Aspinall, 1958) or, if applied only once, abnormal
forms of development are observed (Kirby, 1971). Nicholls
(1974b) considered that the growth of the apex was dominated
by the meristematic activity of three regions:

1. The meristematic region of the apex that is concerned
with the elongation of the apical dome above the youngest
visible primordium.

2. The meristematic region of the single and double
ridge primordia that arise on the flanks of the shoot apex.

3. The internodal meristems in the main axis of the
young inflorescence which are involved in its elongation
(Nicholls and May, 1964).

A single application of gibberellic acid resulted in
enhanced rates of growth of the apex and subadjacent leaf
primordia for only a relatively short period in the life
cycle of the plant (Nicholls, 1978). The first morphological

response to the treatment was an increased dome length, which

 

wa

SU

At

26

was followed by increased rates of growth and development of
subadjacent leaf primordia and basal spikelet primordia.
After the period of enhancement, the growth rates of the
apices of the treated plants fell to values comparable to
those of the control plants. The enhanced growth rates may
have resulted either from an effect of GA3 on diffusivity

of nutrients or from an initial effect of GA3 on the shoot
apical meristem. However, any model of the role of GA3 in
the regulation of growth of barley apical meristem must take
into account the observed sequence of responses: the apical
dome in vegetative shoot apices and both the apical dome and
the upper ridges (rachille initials) in the first stages of
development of floral shoot apices. A response by the
remainder of the apical region follows.

Variations in the light environment have a profound
influence on apical development in barley (Aspinall and Paleg,
1963, 1964; Paleg and Aspinall, 1964) but also have many
other physiological and morphological effects on plant growth.
Some of these effects on vegetative growth are probably
mediated through variations in the supply of carbohydrates
particularly with changes in light intensity (Friend, Helson
and Fisher, 1962). However, many plant responses to light,
particularly variations in photoperiod or spectral content,

are derived from photomorphogenic pathways such as the

ce
Gr
da

19

01”
an

an

ce

or
to

th
Th

27

phytochrome system (Mohr, 1962). The growth of a plant in
one light environment will therefore be dependent upon the
interaction of several complex controlling mechanisms.
Considerable differences in photoperiodic responses of
cereal varieties are recorded (Takahashi and Yasuda, 1960;
Griffiths, 1961) and barley varieties have been considered
day-neutral (Takahashi and Yasuda, 1960), long-day (Guitards,
1960), or obligate long-day plants (Takahashi and Yasuda,
1960). In comparing the effects of light intensity and
photoperiod on 10 barley varieties, Aspinall (1966) found
that apical primordium production was linked with floral
organogenesis over all photoperiods. Generally, there was
an increase in time taken to reach double ridge initiation
and stamen initial stages of development through reduction
in photoperiod from 24-hr to 8-hr and omission of incandes-
cent light (spectral composition change). The effect was
greater on stamen initial production than on double ridges
formation, implying that the development following the
initiation of double ridges is more susceptible to control
by the availability of the products of photosynthesis than
is the initiation of double ridges itself. Night interrupt-
ions decreased, and omission of the far-red component from
the light source increased the time taken to initiate stamens.

Thus, stamen initiation is under photoperiodic as well as

28

photosynthetic control. Primordium production was lower the
shorter the photoperiod. The author concluded that the
relationship between apical development and internode
elongation is not fixed and that the physiological mechanism(s)
which initiates elongation may not be closely correlated with

a particular stage of apical development. Rachis internode
elongation occurs under short days (Nicholls, 1974a), perhaps
as a result of lack of gibberellin following floret initial
appearance (Nicholls, 1978).

The influence of soil moisture tension on plant growth
has been examined in many species (Stocker, 1960). Most
drought studies on cereals have been made on the later stages
of inflorescence development. Milthorpe (1950) and Amer and
Nilliams (1958) showed that immature tissues can tolerate
higher intensities of dehydration than more mature tissues
with a large proportion of vacuolated cells. Drought condi-
tions interrupt translocation of nutrients from the stem to
younger leaves of the shoot apex of tomato plants (Gates
1955a, b, 1957) and delay floral initiation and development
(Brown, 1953). In cereals there appears to be a stage
between spikelet differentiation and flowering which is
particularly sensitive to drought (Zavadskaja and Skazkin,
1960). Stress at this stage induces pollen sterility
(Novikov, 1952), a disruption of the reduction division stage

of meiosis (Zavadskaja and Skazkin, 1960) or disturbances

29

in spikelet differentiation (Novikov, 1954). The severity

of the effects depends on duration of stress, timing of the
stress period and environmental conditions during stress
(Russell, 1959). Aspinall, Nicholls and May (1964) found a
reduction in primordia formation, inflorescence development
and apical length with water stress. Generally, a stress
before stamen initiation is likely to increase tillering, and
marked reductions may occur in internode elongation and grain
numbers; after anthesis, effects on elongation and grain
number may still occur but grain size is severely decreased.
Soil moisture stress tends to affect the organs growing most
rapidly and the tendency for the effects of stress to be

more severe at the very beginning of a particular growth
process supports the notion (May and Milthorpe, 1962) that
cell division is likely to be the event most influenced.

The realization of the importance of temperature as a
regulator of flowering arose from studies of Gasner in 1918
on the flowering of cultivated cereals (Nareing and Phillips,
1970). Increasing the temperature from 10 to 30°C increases
the rate of shoot apex growth and primordium (leaf and floral)
production. Floral initiation occurs earlier (Friend gt gt.,
1963), as a result of either an increased rate of production

of flower—inducing substances or from an increased sensitivity

of the meristematic cells to a given level of flower inducing

3O

substance. This promotion of flowering by temperature in-
crease seem to be independent of the phytochrome system.
Head abnormalities in barley have been found with increasing
temperatures (Kirby, 1974b). The abnormalities obtained are
similar to shoot-apex of 2,4-D-treated plants (Leaf, 1959;
Luxova and Lux, 1964). Rapid changes in the concentrations
of growth substances within the plant in response to changes
in the environment are known (Osborne, 1972). Differential
responses of growth substances to changes in temperature
have been reported by Atkin and Barton (1972). Therefore,
the abnormalities may be due to temperature-sensitive changes
in the auxin metabolism of the plant.

The relative growth of total and leaf dry matter, and
lamina development are strongly influenced by plant population
(Kirby, 1967, 1969, 1973). These morphological changes come
about by plant population differences affecting initiation
and growth of the tillers, stem or leaf growth, and the
initiation and subsequent growth of the spikelet initials
at the shoot apex. The higher the population density, the
faster the growth rate. The shoot apex reaches the double
ridge stage earlier and this persists throughout the
development of the apex. Rate of primordium production is
little affected by density, but the duration is markedly
affected. Plants grown at high densities had a high

 

Far
hig
com
lea

of

C01?

the

01‘

or
de

”10

th.

911

31

concentration of gibberellins in their tissues (Kirby and
Faris, 1970). The enhancement of apex development by the
higher concentration of gibberellin results in earlier
competition for nutrients diffusing through the apex. This
leads to starvation of the tip of the apex and earlier death
of this region and consequently, fewer spikelets. Differences
in the light environment of the plants, either intensity or
composition, most probably brings about the differences in

the gibberellin content.

Tillering may be divided into three phases. The first
involves the initiation of tiller buds and is little affected
by the environment (Evans gt gt., 1964). The next phase
deals with the appearance of the tillers at an advanced
stage. Lastly, the fate of the developing tillers is
determined before their heads emerge (Kirby, 1967; Rawson
and Donald, 1969). Tiller bud initiation is little affected
by plant density, however, some of these buds did not grow
or grew and emerged from the subtending leaf sheath. The
growth rates of the tillers which grew were not affected by
density. Fewer buds developed at higher densities and the
morphology of the tillers produced was affected by density
(Kirby and Faris, 1972). This suggests that the growth of

the tiller bud may be controlled by levels of endogenous

gibberellin, while in the final stage, growth after emergence

fr
to

to

ax
an
On
or
ti
th

mi

th
in
co
ho
of

Dr

32

from the subtending leaf sheath, competition for light appears
to be the factor which determines whether a tiller survives
to produce a head.

A number of lateral and tiller buds are formed in the
axils of the lower leaves of the main stem during the growth
and development of the barley plant (Kirby and Faris, 1972).
Only a proportion of the tillers produced normally survive to
produce grains (Thorne, 1962). The death of some of these
tillers probably constitute a wastage of resources. Nhen
the tillers are developing they compete for assimilates and
minerals, intercept light and affect transpiration (Clifford
gt gt., 1973), thus reducing the size of these shoots and
their potential yield (Kirby and Jones, 1977). Nhen the
infertile tillers die, some of the minerals and carbon
compounds are translocated back to the other fertile shoots,
however, a proportion is not available to the surviving parts
of the plant (Rawson and Donald, 1969). Donald (1968) thus
proposed the 'uniculm' plant type to be the most suitable
plant model for the attainment of maximum yield.

Some experiments have been done to investigate the
proposition that non-headbearing tillers may be wasteful of
plant resources. Tiller removal affects the growth and final
size of the main shoot. A higher number of large sized

leaves emerge earlier on the main shoot of detillered plants.

 

Lar
the
nor
ti‘:
du1

COT

TH

Ea

he

so

33

Larger main shoots with heavier grains are produced. Nhen
the main shoot is removed, the coleoptile tiller produces
more leaves earlier and bear more grains than the coleoptile
tiller of the intact plant (Kirby and Jones, 1977). Thus,
during the initiation and early growth of the plant, tillers
compete with the main shoot for the limited resources, thus
reducing the size of the main shoot.

Nith ample water supply and severe detillering, only a
slight grain yield reduction is observed although the total
shoot dry weight is reduced. Nith the same amount of water
plants with few tillers tend to have greater grain yield,
higher water use efficiency and harvest index than plants
allowed to tiller freely. In effect, genotypes which produce
few tillers have a high rate of survival and are able to
achieve relatively high yields in drought conditions, and
their yield potentials are maintained under optimal conditions

(Jones and Kirby, 1977).

THEORETICAL PREDICTION OF THE EXISTENCE OF BIOCHEMICAL
EQUILIBRIUM IN BIOLOGICAL SYSTEMS.

The analysis used here is adapted from Smith's (1968)
mathematical presentation of diffusion along a tube of a
substance which ultimately induce a chemical change resulting

in changes in growth and development.

to
of

u

i l

34

Consider the flow of a substance in solution along a
tube with a cross sectional area of a. The concentration x
of this solution plotted against s, the distance along the
tube is shown in Figure l. The amount of the substance at
a point of the tube is ax, hence for a short length as of
the tube, the amount is axas. If A and B are the rates of
diffusion from left to right across the two faces of the
element, then A-B is the rate at which the substance is
increasing in 35.

Thus, dx

Edi}: (3x35) :1 aas _t = A'B. ooooooo (I)

The rate of diffusion across a surface is proportional
to the surface area of the tube and the concentration grad-
ient at right angles to the surface.

Hence, A = -au 1%?) at s and

w
I

- -au (31-?)61’. 5 +85

where u = constant, depending on temperature and the diffus-

ing substance. The minus sign occurs because when (%§) is

positive, diffusion would be from right to left and A and B

would be negative.

>
I
W
II
D.

-au [( )-at s - (g?) at s + as]

X
S

Q.

2
an [ as - g§% ]

D.

35

 

X
_ d dx _d2x
51 ope- ITS-(EST) -a—S‘2'
gt
ds

 
   

I as x slope

     

S as

Figure 1. Geometrical derivation of the diffusion
equation. (from J. M. Smith, 1968).

36

Substituting in equation (1), we get

0.

ds

This partial differential describes diffusion along a
tube. It indicates that if the graph of x against s is
concave upwards, then x is increasing at this point and vice
versa.

Consider the effects of diffusion and chemical reaction
simultaneously. Let the concentration of two chemical
substances A and B be X and Y. Suppose that the substrates
from which A and B can be synthesized and degraded, and also
the relevant enzymes are present. There will be some values
of concentrations X5 and Y5 for which there is a chemical
equilibrium. Ne can represent X and Y as; X = XE + x and
Y = YE + y

where x and y are measures of concentration departures
from their equilibrium values.

Assume our interest is in the changes in X and Y as

functions of time and distance along the tube, then,

when x = y = 0, dx 91 = 0
t dt

Nith small changes from the equilibrium

0.

x _ , d =
—f - ax + by, 3% cx + dy

37

where a, b, c and d are reaction rates.
If u and v are the rates of diffusion of A and B
respectively, and allowing for diffusion along the tube,

__ dzx

then dx
t ax + by + u 3:7

9.11
dt

2
cx + dy + v %E%

Ne can solve the equations algebraically, but it is more
interesting for the present purpose to solve them graphically.
If all values of x and y are zero for all values of s

' — .d_.x..= 9.x:
along the tube then at t1me t-O, dt dt 0

If this homogenous equilibrium is disturbed, the equili-
brium is restored through adjustments of the values of the
reaction rates a, b, c and d, and the diffusion rates u and
v, so that x and y tend to zero with increasing time.

Certain values of the constants render the equilibrium,
unstable. Thus, even when there is an initial homogenous
state with x = y = O, a standing wave of concentration of the
substances may arise from any small initial disturbance.

Some assumptions ought to be made at this point:

1. If the concentration of A rises above the equilibrium
level, rate of synthesis of both A and B rises. a and c are
positive.

2. If the concentration of B rises, it leads to

38

destruction of A. b is negative and d is assumed to be zero.
3. B diffuses faster than A. v > u.
Figure 2 a, b, c and d, show what happens when the homogenous
equilibrium is disturbed by a small departure in the concen-
tration of A. Fig. 2a, shows a disturbance in the equilibrium.
This leads to a further rise in A and B, but 8 diffuses out
further (Fig. 2b). At the point marked by the arrow, y is
positive and x is zero, leading to destruction of A (Fig. 2c).
This in turn, leads to the destruction of B so that a trough
develops on either side of the initial peak (Fig. 2d). These
troughs cause the development of other peaks and so on,
until a standing wave has been developed whose chemical
wavelengths depend on the constants defining the rates of
reaction and diffusion.
The wave pattern will be utilized to show the existence
of an equilibrium between the promotive hormones which

induce growth and development in an orderly manner.

39

a (aL

 

 

 

 

 

 

Figure 2. Development of a standing wave.
(from J. M. Smith, 1968).

MATERIALS AND METHODS

The same twelve genotypes of barley were used in all
the experiments. They were selected on the basis of
differing values in their yield components, namely number
of heads per unit area (X), number of seeds per head (Y)
and average seed weight (Z).

A rectangular-lattice design with four replications
was used. The plots were four-row plots 0.0254m apart
and 2.4m long, planted at a rate of 359 per plot. The study
was carried out at the Crops Science Research Farm in

Ingham county, Michigan. Planting date was May 5, 1980.

INVESTIGATION OF THE LEVELS AND ACTIVITIES OF THE PROMOTIVE
HORMONES AND THEIR EFFECTS ON GRONTH AND DEVELOPMENT OF
THE PRIMARY SHOOT APICAL MERISTEM, YIELD COMPONENTS AND
YIELD OF BARLEY
A. DETERMINATION OF THE GRONTH AND DEVELOPMENT OF THE
PRIMARY SHOOT APICAL MERISTEM
Meristems were sampled at three predetermined stages,
which were:
1. Transition, identified by the appearance of double
ridges.
2. Reproduction characterized by spikelet different-

iation.

4O

41

3. Elongation and further differentiation of meristems.

Before any sample was taken at each stage, seedlings
within each genotype were visually selected for similar
morphological characteristics from the outer two rows
(border rows) of the whole plot. One or two seedlings were
uprooted, and their meristems dissected to determine their
developmental stage. Five seedlings were then harvested
from the outer two rows. The portions of the main tillers
containing the meristems were preserved in a solution
containing 95% ethyl alcohol, water, glycerine and formalde-
hyde in proportions of 52%, 38%, 5%, and 5%, respectively, to
prevent structural changes. The five plants were selected
to represent the mean of each cultivar. The main culms
were used because they have a greater potential for product-
ion within a defined and finite environment than has any
other single tiller. The meristems were dissected and
measurements taken using a light microscope equipped with
a measuring ocular. Measurements taken include maximum
length (L) and width (N0) of the meristems (Fig. 3).

The relative growth rate (R) of meristems in the
reproduction state was determined using the procedure

outlined by Nhyte (1979)

42

Figure 3.
Diagram of barley
L shoot apex. The
measurements taken

of length (L) and
i ’ width (ND) are shown.

e-wu—al

8. ANALYSIS OF THE LEVELS AND ACTIVITIES OF THE ENDOGENOUS

PROMOTIVE HORMONES

A sizeable amount of sample was taken from each plot,
at each stage, and immediately frozen by placing them in a
chest containing dry ice. These were kept in a freezer at
-30°C until they were lyophilized. Meristems were dissected
from the lyophilized materials and attempts were made to
analyze them for their endogenous levels of gibberellin,
auxin and cytokinin. These determinations were futile, the
reason most likely being the insensivity of the methods
employed or the inadequate amounts of meristems obtained
from the lyophilized materials.

The object of the endogenous hormonal level determina-

tion was to find any existing equilibrium state between

43

the promotive hormones during the reproductive phase.
Although the utilized methodology did not reveal the
presence of the promotive hormones, a mathematical model
(Smith, 1968) will be used to establish the existence of
such an equilibrium. The author hopes that a more sensitive
procedure will be found in the future, to carry out such

types of experiments.

C. MEASUREMENTS OF YIELD COMPONENTS AND YIELD OF BARLEY

Measurements taken during the plant growth and at
maturity included height (HT) and heading date (HD) which
was calculated from April 30, 1980, when about 75% of plants
in each plot had headed. Data for the number of seeds per
head (V) were obtained from a random sample of twenty
heads per plot from the border rows just prior to harvest.
Yield (N) was obtained from harvesting the central two rows
of each plot. The average seed weight (Z) was calculated
from a 39 sample per plot using an electronic seed counter
to count the number of seeds within the sample. The number
of heads per 30cm of row was obtained by dividing one
sixteenth of the total weight of grain per plot by the
product of seeds per head and kernel weight in grams.

A number of plant characteristics were each used as a
dependent variable while measurements taken on the primary

shoot apical meristem in the reproduction state of development

44

were used as independent variables in a series of multiple
regression analyses. This was an attempt to break down
the yield components into their subcomponents at the
meristematic level.

Using the path coefficient analysis as outlined by
Nright (1921, 1934), the phenotypic correlation were parti-

tioned into their direct and indirect effects as follows:

°R(ND) = a

pRL = b

pL(wo) = C

pX(wo) = d 1 Cf

pXR = e + bf

pXL = f + cd + eb

pY(wD) = g + ci + jd + cfj

pYR = h + ej + bi + bfj

pYL = i + cg + fj + hb + jdc + jeb

pYX = j + gd + if + he + gcf + ibe + hbf + icd

pZ(ND) = k + nd + 09 + mc + ncf + oic + ojd + ojfc

pZR = l + mb + oh + ne + oje + nfb + oib + ojbf

pZL = m + lb + oi + nf + kc + ogc + ojf + ohb + nbe
+ ndc + ojeb + ojdc

pZX = n + kd + oj + mf + le + kcf + ohe + oif +

ogd + mbe + mcd + lbf + ohbf + oicd + oibe +

ogcf

45

pZY = o + nj + kg + mi + 1h + neh + nfi + ndg + kdj +
mfj + mcg + mbh + lej + nfcg + nfbh + mbej +

mcdj + lbfj + kcfj + nebi + ndci + kci + lbi

The rationale behind the diagram (Fig. 4) is based on
the developmental ontogeny of the plant and Sinnott's law
as presented by Hamid and Grafius (1978) and Grafius (1978).
Grafius (1978) showed a cause and effect relationship
between leaf size and seed size after considering that the
lemma and palea are modified floral leaves and hence larger
leaves should be associated with relatively larger sizes
of the lemma and palea, the determinants of seed size.
Since the relative sizes of the lemma and palea derive from
the same apical meristem, its substitution by the measured
characteristics of the apical meristem in the reproduction
state of development becomes more acceptable. A direct
connection is thus established between the meristematic
characteristics and number of seeds per head and the average
seed weight. Nicholls and May (1964) proposed two probable
factors contributing to the elongation of the apical
meristem during the reproductive stage of growth. The
major factor is the net increase in the number of primordia
produced and the minor factor being their elongation. I am,

therefore, inclined to represent the relationship between

Figure 4.

46

 

Developmental allometry showing the
influence of the primary apical meristem of
barley in the reproduction state of development
on the yield components: number of heads
per unit area (X), number of seeds per

head (Y), seed weight (Z); width (ND),
length (L) and relative growth rate (R) of
the primary apical meristem. Single
arrowed lines denote path coefficients and
double arrowed lines denote correlation
coefficients.

47

width and relative growth rate by a correlation instead

of path. However, both width and relative growth rate

are known to predict length, thus, the path representation.
The path coefficients can be obtained by solving the

simultaneous equations.

EFFECTS OF MANIPULATION OF THE ENDOGENOUS HORMONAL LPROMOTIVE)
LEVELS ON THE GROWTH AND DEVELOPMENT OF THE PRIMARY APICAL
MERISTEM, YIELD COMPONENTS AND YIELD OF BARLEY

This experiment consisted of five blocks, each planted
with four replications of the twelve cultivars. Each block
was randomly assigned to one of five treatments (control,
auxin, gibberellin, cytokinin and cycocel).

Solutions of auxin, cytokinin, cycocel, and gibberellin
were prepared, using indole acetic acid, kinetin, (2-chloro-
ethyl) dimethyl ammonium chloride and gibberellic acid (GA3),

6 4 4

with concentrations of 3 x 10' M, 10' M, 2.5 x 10' M and

10'4

M, respectively. Dimethyl sulfoxide (1%) was included
as a carrier in all solutions, and plants in the control
block were treated with the same concentration of dimethyl
sulfoxide.

Using the same procedure as in the first group of

experiments, meristems of the genotypes were sampled just

before they reached the transition stage. The plot(s) in

48

consideration were then sprayed with its respective hormone
solution thereafter.

Two sprayings of each plot were done. The first was
carried out when a genotype reached the transition stage and
the second done two days later. Sprayings were done in
the evening when air movement was minimal. Plants within
each plot were sprayed until the hormone solution started
dripping, thus allowing for possible root uptake of the
hormone especially with regard to the cytokinin application.
The site of biosynthesis of cytokinin is the roots (Kende,
1965) and cytokinin application experiments used mostly
soil drench (Bokhari and Youngner, 1971).

Meristems of genotypes reaching the reproduction stage
were sampled and preserved. Measurements were taken of the
maximum length and width of meristems sampled just prior to
the transition and at reproduction stages.

Other measurements taken during the plant growth and
at maturity included height (HT), heading date (HD) calcul-
ated fron April 30, 1981, when about 75% of plants in a plot
had headed, number of seeds per head (Y), number of heads
per 30 cm (X) and the average kernel weight (Z). Yield (N)
was obtained from harvesting the central two rows of each

plot.

49

Some analyses were done after transforming the data in
order to express the mean values as the mean percentage
change over the control. The formula used for the trans-

formation was:

AH - Ac

mean value for a plant characteristic

where AH
obtained after a hormone treatment.
Ac = mean value for a plant characteristic

obtained from the control plot.

Use was made of the mean percentage change over control
values for lengths of meristem to establish the existence
of an equilibrium between the promotive hormones and through
a reasonable extrapolation, comments were made about the

yield components.

RESULTS

A. RELATIONSHIP BETWEEN THE PRIMARY SHOOT APICAL MERISTEM,
YIELD COMPONENTS AND YIELD OF BARLEY.

The relative size of the primary shoot apical meristem
and its rate of transformation from the vegetative to
reproductive development state of the genotypes were signi-
ficantly different (Table l). Genotypes X969-3, 68-105-9
and 68-105-17 start with larger meristems but delay the
elongation process by four days. This delay results in the
formation of a larger sized meristem. They reached the
transition state four days in advance of the other genotypes.
Bowers barley (68-105-15) has been regarded as an outlier

(Grafius gt_ 1., 1976) but data of Nhyte (1979) and the

results of the development of meristems presented here do not
support the claim.

The mean values for the yield components (X, Y, Z), yield
(N) and the other plant characteristics measured are presented
in Table 2. There were significant differences between the
genotypes for the various characteristics.

Table 3 gives the correlation coefficients between
meristematic measurements, yield components and yield. The

length and width at stages 1 and 2 are positively and

significantly correlated with each other. Nidth at stage 3

50

51

 

 

 

 

 

 

 

eom. mpe.e Pmm. mum. m_m. e—e. mo.nuv omA
eAm. emm.e NRA. ANA. oeA. wee. Ame.uuA omA
mmm.m ooo.eN oom.P om~.m mum. ooo.m mpumo_-mo
omN.N oom.o~ oom.~ wee.e oow. m—e.m m1moplwe
Amm.m oom._m Mme.w oom.m com. mom.m apleopuwe
ooo.~ ooo.mm om_.w mmp.m mmm. mpm.~ m-¢o_-mo
mem.m ooo.- owm.P mem.m omm._ omm.e omm. mwo.m “Fumopiwe
mm¢.~ ooo.e~ wmo.~ mwm.m omm._ ooo.m o_~._ mow.m mimop-we
omm.~ mmw.mm mwm._ mmm.e mmw. wmm.m mm¥m<A
wmm.m ooo.- mme.w mmo.m mm“. www.m zomque
ooo.~ mmm.mp oww.w mnw.m mum. omN.N enmwmuoe
mwmAm www.mp wwm._ mwo.m om“. oom.~ ompm
mme.m m_¢.mm opp.m upe.w wwe._ wom.e mom. mFm.m Mumemx
mnm.m om~.em mpe.w mmm.m wmm. mme.~ mmmzom
AEEVAozv weave Aeavaozv AEEVA AEEVAozv AEEVA AEEVAozv Agave meexuecew
m AAA mwwem < AAA ammem AA amwem A meaem
.AceEee_e>ee
we meueum Am e:e.<AAA emepmw :ewuemcepe use AAA emeemv :ewueeeegeeg
.Aw emeemv :ewewmceep esp cw xepcee we mseemwees Aeewee ageswee
we» we Aazv spew: Eeswxes ece AAA gemcep Eeewxes esp sew mezpe> new: ._ mpeew

52

 

 

 

Am.A mm.m AN. me.e ee.¢ 4mm. we.eeA mm.ee AAo.uaA AAA
eA.A om.N AA. mm.e ee.m eAA. Ae.mmA em.ee Amo.uaA QmA
oo.e~ mm.mm oe.¢m ep.em oo.em me.w mm.mmmp mm.~_w ww-moA-me
om.mm oo.Ae om.om we.me ee.- mmm.w AN.wo¢_ om.eme m1mopume
oo.mm mm._e Pm._m um.mm mm.mm ewo._ ee.wwmw mm.mme mF1eowume
om.mm mw.mm mm.om mm.o¢ me.mm meN.A m¢.~mew om.wow m1eowuwe
mn.w~ om.mm om.mm nm.mn wo.mp mwm.~ ww.emm_ oo.~me w_-me_-me
mm.m~ mm.oe em.om wo.em mm.m_ eem.~ Ae.mee_ om.mm~ m-mo_-me
oo.e~ oo.oe F~.¢m mw.mm Nw.mm mwm._ mo.wemp mm.new mmxm<A
mm.NN mm.mm ¢_.mm eo.oe em.w_ meN._ ww.e~o_ oo.mom zomquo
em.em m~.Nm mm.wm em.mm me.mm mem.o em.w~__ mu.mom e-m_~-oe
m~.em oo.oe ce.om mm.oe mo.¢~ Aem._ Nw.mee_ mm.o_m omwm
mm.em m~.mm oe.m~ oo.e~ mm._~ oww.m em.¢mep mm.een m-memx
mN.NN om.mm e_.om e—.Nn mm.m_ eww.m mm.emew mN.ome mmmzom
Amxaevoz ACAAPI AmEAN AwA> vax AmAN> AwA>x Amvz mea>wezee

.AaAeaA eo AQIA meAemm; use AAIA egeAa; .ANA Agmwmz teem .A>A

ewe; gee meeem we geese: .Axv mega ewe: gee meme: we geese: .AN>V erm
ewe; .A>xv eege awe: Lee memem we geese: .sz ewmwz Lew meepe> cemz .N epeew

53

. —o.1wn_c¢

 

 

me. we .
om~.- wm—.- moo.n vn—.o ocm.n ¢¢v~m. mom. ~v~.u wom.u eoee.o com.: —o—. eowc.u mmv. a
mac. emmo. emwm. mov. omm. mmm.u New. —0m. mom. .wm. mcp. nmm. «mum.4__m-oz
Nap. mmm. ~oo. moo.- owo. mo—. @mw. m—m. wow. wow. _o~. mn~.1—__mua
canon. mmo. ~o-. wom.o «emww. owe. «New. wwm. va. —om. acoo.u_~m-oz
m—c.u amp. wn—.u .emom. mow. v_m. own. —wm. awe. omm.u __m14
iemnm. oo~.o .mo.u smp. mom. mom. moo.u Nov. uwc.u _m-oz
—om.- mom. mme. «mom. mme. vww. com. ~mn.- ~m14
owm.u enmm.1 moc.u mmv.u —vv.n mop. mo~.u o:
ecmms. mmm. mom. cwm. w—v. ~mm.: w:
mom. «omen. comm. own. Mac.: 3
emm. Om—. ceooo. eeouw.l N»
owe. «imam. mm—.1 >x
mm..- mmo. N
.cmmw.- >
___mpa3 __—m14 -mo3 __m14 ~m-c3 _m-A o: h: 3 N» >x N > x

 

.uce5ee_e>ee we meuoum AAA—my

neweemce—e ecu AAAmv ceAAeaeeseeg .AAmA cewu—meeeu ecu :_ xepgee we Eeumwcee _ee_ee mceewge wsu we Aazv sue.) E:E_xee
ecu AAA guano. Eaewxee .Aozv neweeen .szv ugowez .A3A e—ewx .Ava e~wm new: .A>xv omen awe: gee meeem we geese:

.ANA agape: teem .A>A eee: Lee meeem we geese: .AxA ease ewe: Lee meow; we Lease: use mceae muemweAwweee :e_ue—eggeu

.m m—aew

(”E

of

54

is significantly correlated with length and width at the
reproduction stage. Relative growth rate is significantly
and negatively correlated with length at the transition
stage, head size and number of seeds per head. Nidth at
the reproduction stage is significantly and positively
correlated with head size and height but negatively corre-
lated with number of heads per unit area. The relative
growth rate of the primary shoot apical meristem and number
of heads per unit area are negatively correlated with most
of the other measurements taken in the study but positively
correlated with each other. Both are negatively and
significantly correlated with head size and number of seeds
per head. A pattern is established: when the number of
heads per unit area increases, the relative growth rate
increases while head size decreases due to a decrease in the
number of seed per head. The relationship between the
relative growth rate with head size and number of seeds per
head are plotted in Figure 5 and 6 and the significant
regression line reaffirms the inverse relationship. A geno-
type with a higher number of heads per unit area will grow
at a higher rate, however, it will produce a smaller head
size and a lower number of seeds per head. Thus, a lower
rate of growth results in the production of a larger number

of seeds per head and eventually a bigger head size. Even

  
 

70
U
f0
0)
A:
S.
Q)
Q.
U)
13
Q)
Q)
q.
0
L 50
Q)
.0
E
3
Z
30
0
Figure 5.

55

O
_<
II

79.34 - 33.1OX
.4729*

 

.5 1.0
Relative growth rate

Regression of the number of seeds per
head on the relative growth rate of the
primary apical meristem of barley in the
reproduction state of development.

56

 

2.1

1.7
”a?
(D
N
‘0
f6
0)
:1:

1.3

.9

O .5 1.0
Relative growth rate

Figure 6. Regression of head size on the relative growth
rate of the primary apical meristem of barley
in the reproduction state of development.

57

though the above notion can be generally accepted, one has
to wonder whether the relationships arise because of the
number of heads per unit area associated with the system or
the number of heads per unit area cause such a relationship
to arise.

The mean square values for the multiple regression
involving length, width, and relative growth rate of the
primary shoot apical meristem as independent variables and
the dependent variables number of heads per unit area (X),
number of seeds per head (Y) and head size (YZ) are signifi-
cant, as shown in Table 4. The same regression using the
dependent variables average seed weight (Z), number of seeds
per unit area (XY) and yield (N) showed no significance.
The coefficient of determination (R2 = 0.5959, 0.7155 and
0.7912 for X, Y and Y2, respectively) indicate that the
variance in the dependent variables can be accounted for in
large part by variation in the three independent variables.

Some interesting observations found in the multiple
regression analysis include:

(1) Only the relative growth rate and width of the
meristem showed significance in the multiple regression
analysis presented.

(2) Nhile the width of the meristem has a positive

predictive value, the relative growth rate has a negative

58

we.
we.

vac}.

n. .4.

VI

 

Nwmn.o om—m.o mmmm.o

mweo. mom_.ww momm.~w

mee. Nome.mpm owom.mo
N> > x

m geggm

m :ewmmmemem

we eegeem

 

.meweewee>

peeeceeeecw ecu me aceseewe>ee we eueum :ewaeeeeeeee esp cw xeweee we

Eepmwses peewee aeeewge ecu we Amv mum;
guewz Eeswxee .AAV zumcep Eeewxes ecu :e

.AN>V erm ewe; ece A>v ewe; Lee meeem
meme; we geese: ecu we :ewmmegmee ewewuwes

zuzegm e>wue_eg ecu ece Aozv
eAeewee> neeeceeee e we seem
we geese: .Axv eege awe: gee

esp Lew eecewge> we mwmA—e=< .e eweew

59

predictive value for both head size and the number of seeds

per head. The reverse is the case with predicting the

number of heads per unit area. Nidth and relative growth

rates have negative and positive predictive value, respectively,
for the number of heads per unit area (Table 5).

The smaller the width of the primary shoot apical
meristem of a genotype, the higher the number of heads it will
produce per unit area and the higher its rate of growth.
Number of seeds borne per head will be small since width has
a positive correlation with Y. A similar picture is obtained
when the prediction of either head size or the number of
seeds per head is considered. The higher the relative growth
rate of the primary shoot apical meristem of a genotype,
the lower the number of seeds borne per head or the smaller
the head size it will have. The width of the meristem will
be small while number of heads produced per unit area will
be high since width has a significant negative relationship

with X.

B. THE REACTION OF THE PRIMARY SHOOT APICAL MERISTEMS TO
APPLIED HORMONES
Mean values for the maximum length and width of the
primary shoot apical meristem before the transitional and
at the reproduction stages of development are given in Table

6. The 'outliers' had larger sized apical meristems initially,

6O

 

 

 

ave. mm~.- woo. mom.- omm. .
Rpm. eew.- owe. mee.- mmm.wm-
wee. wmm. NNA. mom. emw.w m
emm. mew. Noe. eve. Nee._
mom. Pom. Nmo. «mm. opo.me
New. mme.- oeo. NmN.- mmm.nm- o3
om“. Nmo.i mew. meo.- one. .
mww. mmo.- mm“. mwo.- moo.wi
mum. mmm. New. «mm. Nem.— A
mega—me weewewwweeu we>eA musmwez ucewewwweeu e—eewge>
m :ewueweeeeu pceewwwcmwm epem :ewmmeemem Fewueea
N AeAAeeA
.»_e>wueeemee .N>
ece > .x eew mewpmwueum exp wee meewe> Lezew eee eweewe .eeee: umnwe
meweewee> pceeceeeecw ecu me geeseewe>ee we meepm :ewueeeeeeee we“ cw
xewgme eaAseumweee Peewee Ageswee ecu we Amy mum; sezeem e>wpepeg one
ece Aozv spew: Easwxes .AAV gumcew Eeswxes ecu :e epeewge> uceeceeee
e me some .AN>V ewwm ewe; ece A>v ewe; Lee meeem we geese: .Axv
eeee awe: gee meme; we geese: esp Lew mewemwueum cewmmmemeg ewewupez .m eweew

61

 

 

 

 

New. eee. eeA. eAm. AAe.nuA AAA
AAA. AAA. emA. eem. Ame.uuA omA
mAm.A mAm.m eew. mAe.A eA-moA-ee
eAm.A mem.m ANA. eme.A e-moA-ee
emm.A eom.e nee. mAe._ mA-eoA-ee
omA.A ewm.e eAe. eom.A m-eeA-ee
eem.A eNA.e who. mee.N AA-meA-ee
mee.A omw.e mew. mAN.N m-meA-ee
eom.A eem.e eme. ANA.A mA¥m<A
eAA.A eAN.e wee. mee.A zemxeAe
eoo.A mem.m emm. eme.A e-mA~-oe
mAm.A omA.e owe. oee.A emAm
eee.A Ame.e mew. mAN.~ m-meax
mem.A ORA.e ewe. mme.A mmezom
Aeerz AEEAA Aeerz AEEAA mAa>wezee
AA Ae<wm A v Aeewm

 

.pceEQewe>ee we meueum AAAmv :ewueeeegeee eze cw ecm
AAvmv eewuwmceeu esp eeewee emew xeweee we Emumwees Feewee xgeswge

wen we Aozv cuewz Eeswxee ece AAA seucew Eeswxee Lew meewe> :eez .e eweew

b1
Oi

01

of
me
of
ti
to

re

ho
De“
hi:
of

C01

whe

Set

of

Cdn

eXl!

62

but few differences were observed in the reproduction state
of development. A switch in rate of increase in the size
of the meristem is, however, shown.

Table 7 gives the mean percentage change over control,
of the maximum length and width of the primary shoot apical
meristems in the reproduction state, due to the applications
of gibberellin, cycocel, cytokinin and auxin at the transi-
tion stage of development. Significant differences exist
for both the length and width, suggesting differences in
reactions to the applied hormones.

These mean values were plotted as histograms to determine
how the reactions to hormone vary with the number of heads
per unit area (Figures 7-10). Genotypes are arranged in the
histograms in the order of increasing production of number
of heads per unit area. The first eight genotypes are
considered standard genotypes while the last four are 'outliers'.
Bowers is included in the set of outliers to determine
whether its reaction is more related to that set or to the
set of standard genotypes. Of the eight standard genotypes,
Dickson and 60-215-6 produce the least and the highest number
of heads per unit area, respectively. There is no signifi-
cant differences between the 'outliers' for the number of
heads produced per unit area while significant differences

exist between the standard genotypes for the same trait.

:.:E—.)flp: OIL. taCE. FCS‘CCL 50:: QCCGIL QC“.¢EQI\(¢QC 0“ (.CM‘. UQ~AFR~> talk-Q: 1N Qfiomifi

63

 

 

 

 

me.eA me.AA AAe.neA AAA

ee.NA AA.AA Aeo.neA AAA
AA.A AA.A AA.A A Ae.e Ne.e - Am.AA- AA.AN- AA-AAA-AA
em.“ ee.eA 4e.“ - Ne. - Ae.eA- A~.eA- ee.e~- Am.AN- e-mAA-ee
AA.AA AA.AA AA.AA ee.~ Am.em AA.AA AA.AA AA.A AA-AAA-AA
ee.eA eA.AA Am.AN AA.e - me.eA AA.A~ AA.Ae Ne.e e-eAA-eA
ee.~ ee.e ee.e me. AA.AA ee. mm.e AN.A AA-AAA-AA
Am.AA eN.e AN.AA AA.em AA.AA em.e AN.A Ne.Ae A-AAA-ee
AA.e - AA.AA- A mm.e - AA.AA AA.A AA.AA AA.e AAxeeA
AA.AA he. - AA.A AA.eA me.~A NA.A Ne.AA ee.e zeAXAAA
e~.e AN.AN AA.eA AA.NA AA.A Ae.AA AA.NA A~.Ae A-AAN-AA
Am. ee.A Ae.e - Ae.A - AA.eA ee.e - AN.NA- Ne.eA- AeAA
AA.e AA.A A~.AA ee.AA AN.NA Ae.AA AA.eA em.e e-AeAx
Ae.N AA.AA- AA.A - AA.AA- N~.A AN.N AA.AA ee.eA- meAzee
<<A ewAeeAg AAA e<A <<A eweeAAX AAA m<A AAA>we2AA

A: A A A

 

.uceseewe>ee we emeem :ewewmceee ecu we

A<<AV :wxee ece Acweecwxv cwcwxepze .Aueev weeeexe .Am<uv :wwwmeeeewm we
:ewueewpeee egg on wee aceseewe>ee we eeeem :ewpeeeeeeee we» cw xe—gee we
Eepmwgee weewee xeeswge ezu we wozev spew: Eeewxes ece “Aev spacey

Eeswxes ecu Lew wegwcee Ae>e emcece emeeceeeee ecu Lew meewe> ceez

.w mweeh

54

The lengths of the meristems at the reproduction stage
of the standard genotypes producing the least and the two
highest numbers of heads per unit area were increased,
while those for the other standard varieties were reduced
with gibberellin application. Genotypes 68-103—8 and
60-215-6 had the greatest decrease and increase, respectively,
in the lengths of their meristems. Of the outliers, the
length was reduced in Bowers and increased in the rest, with
68-105-9 having the greatest increase. Application of GA3
induced significant increase in the width of meristems of
Dickson and 60-215-6 but had no effect on the rest of the
standard genotypes. Nith the outliers, width was reduced
in Bowers but increased in 68-105-9 and X969-3. No
significant change was observed with 68-105-17 (Fig. 7).
Thus, the sizes of the apical meristem in the reproduction
state were increased in the outliers, with the exception of
Bowers, and in the standard genotypes producing the least
and highest number of heads per unit area. The other standard
genotypes and Bowers had a reduction in the sizes of their
apical meristems with gibberellin application.

Interesting trends worthy of note are the changes in
the length of the apical meristem due to gibberellin applica-
tion. Considering just the standard genotypes, there is an

increasing reduction in meristem length with increasing

UwHflptw“WwL®~: Cw QCCQIU WUQHCQULWQ

 

 

65

LENGTH

40

20

,—
o O
S.
4.:
C
O
00
..- -20
4-10
«34-1
E
01-1-1
4-10
V)
.A§§ -40
Len
QJGJ
ES.
:5
.A,A 60
"—
m3
05
:01
mu
m— 40
U-H
en
(1.)"-
Dis.-
mm
444-,
:(J 20
mm
US-
Ln!
(1).:
0.0
O
-20

 

DICKSON
68-103-18
68-103-8
68-104-19
LARKER
68-104-3
60-215-6
BONERS
68-105-9
68-105-17
X969-3

                    

GENOTYPES

 

Figure 7. Percentage change in maximum length and
maximum width of the primary apical meristem
of barley in the reproduction state of
development due to gibberellin (GA3) applica-
at the transition stage of development.

66

number of heads per unit area. However, after the greatest
length reduction is achieved, i.e., with 68-103-8, this
reduction starts to decrease and eventually the maximum length
starts to increase with further increase in the number of
heads per unit area. Little can be said about the trend
resulting in the changes in length of the apical meristems

of the outliers since the number of heads per unit area in
these genotypes are not significantly different from each
other.

Figure 8 depicts the effect of the application of
(2-chloroethyl) dimethyl ammonium chloride (CCC) on the length
and width of the primary apical meristem as compared to
control for the same genotypes. A similar trend in reaction
to GA3 is shown with the effect of CCC. The genotypes with
a relatively high reduction in length due to GA3 application
showed length reductions. These include 8130, 68-103-8 and
68-103-18 from the group of standard genotypes and Bowers
from the outlier group. The rest of the genotypes showed an
increase in length with CCC application. Generally, there
was a slight reduction in the reaction of the apical
meristem to CCC as compared to GA3 treated plants. The
-second highest tillering standard genotype, 68-104-3, had
the greatest change in reaction to CCC as compared to GA3,

i.e., from 5% to 30% increase in length. The width was

4O
20
O
A.
o
L
4—)
C
o -20
UL)
'1-
+30
«5+:
E
one -40
“U
ma)
"—0.
LU)
w w
E“ 50
CA:
.Pp
"—
m 3
CT!
CU“!
m U 40
AC"-
UH
U1
m-e
U'lL
m w 20
4444
EU
m e
UL
S-fU
0.1.: 0
QU
—20

Figure 8.

67

LENGTH

'1 11-11

 

 

 

 

GENOTYPES

Percentage of change in maximum length and
maximum width of the primary apical meristem
of barley in the reproduction state of
development due to cycocel (CCC) application
at the transition stage of development.

68

increased significantly in Dickson, 60-215-6, 68-105-9 and
X969-3, however, these increases were lower than that of the
gibberellin-treated plants.

The length of meristem were reduced in the reproduction
state for 8130, 68-103-8 and 68-103-18 from the application
of cytokinin (Kinetin) in comparison to the control. Only
X969-3 had a significant increase in length among the outliers
with the other three showing almost no change in length.

The highest increase was obtained from 68-104-3 followed by
68-104-19 then 60-215-6 (Fig. 9). Nith the exception of
Larker and Bowers, widths of meristems were either increased
or remained constant. The greatest increase in width was
obtained in 60-215-6 followed by 68-104-9 then 68-103-18 among
the standard genotypes. Nithout considering Larker, width
generally increased with increasing heads produced per unit
area from Kinetin application.

Nith auxin (IAA) application (Fig. 10), lengths of the
primary shoot apical meristems of all genotypes, with the
exception of 68-103-8 and 68-103-18, increased. The 104
lines had the highest increase. The width of the primary
apical meristem of Larker showed a decrease while that of
the rest of the genotypes showed an increase. Genotype
68-105-9 had the highest width increase with the rest of

the outliers showing a relatively small increase which is

4o

20

.—
o 0
S.
4.1
C
O
UU
'l— ‘20
4-10
04-)
E
(1)-H
+30
ma)
.,_Q_ '40
5.01
mm
EL
:5
,F+J 60
~1-
0.13
US
EU)
mu
c-r 40
OH
U1
(Ll-v-
U)!-
rum
4-1-9
: U 20
a)!!!
0&—
Ltd
(1).:
DAU 0
—20

Figure 9.

69

 

 

 

 

 

 

1 LEAALH.
' m
I
I
i WIDTH
1
I
1
I
(I) o3 l\
l—wf— Mk0 OW!—
Z III II II
C MMVKQ'LO mmmm
(f) OOOLuOr— EOOI
¥Ol_'r-l—¥l—N LLJr—r—C‘l
Uml'lmll 3||Lo
HFwCom<mo oooooci
D. mike I‘D-£0.4- 0. U). .m I‘D-“0.x;

 

GENOTYPES

Percentage change in maximum length and maximum
width of the primary apical meristem of barley

in the reproduction state of development due to
cytokinin (Kinetin) application at the transi-

tion stage of development.

4O
20
,—
O
s. 0
H
C
0
UL)
5e -20
64-3
E
Gal-OJ
4-10
mm
‘FQ _
L m 40
QJQJ
EL
CS
w—U
-P 60
(113
0'1
CU)
I'UU
.C‘r—
UH
m 40
G)"-
DDS-
0‘00)
UP
CU
w e 20
US-
3.0
0).:
0.0
O
-20

Figure 10.

7O

 

 

 

 

 

 

 

 

 

I LENGTH
I
I
q
i NIDTH
I
1 I I
I
CD Ch I\
v—wr— COLD 0‘!—
Z III 11 11
O mme-ocerm mmmm
U) OOOLIJOI— DCOOI
¥Or—r—r—¥i—N LIJF-r—m
umullrxil 311m
H—oooooo<ooo ooooocn
olmlololol—Jlkolfi Imlmlmli
GENOTYPES

Percentage change in maximum length and
maximum width of the primary apical meristem
of barley in the reproduction state of
development due to auxin (IAA) application at
the transition stage of development.

71

not significantly different from that of the control. The
relative width increases of both Dickson and 68-104-3 were

the same. A general pattern cannot easily be used to describe
the reactions of the genotypes to auxin application with
respect to their width changes, however, the length changes
show similar reaction patterns to the GA3, CCC and Kinetin

applications.

C. THE REACTION OF YIELD COMPONENTS AND YIELD T0 APPLIED

HORMONES.

The mean values for number of heads per unit area (X),
number of seeds per head (Y), average seed weight (Z),
number of seeds per unit area (XY), head size (YZ), yield (N),
height (HT) and heading (HD) for the control block are given
in Table 8. There were significant differences between the
genotypes for all the characteristics mentioned above.

Table 9 gives the correlation coefficients between the
plant characteristics measured (control block). X maintains
a negative correlation with the characteristics, though not
all the correlations were significant. X is significantly
and negatively correlated with head size (YZ), number of
seeds per head (Y), length (L) and width (N0) of the primary
shoot apical meristem at the reproduction stage. Thus, the
higher the number of heads produced per unit area by a

genotype, the smaller the head size, number of seeds per head

72

 

 

 

AA.A ee.~ AA.AA AAA. AA.AAA AA. AA.e AA.A AAA.nuA AAA
AA. AA.A AA.AA AAA. Ae.eAA AA. AN.A AA.A AAA.neA AAA
AA.AA AA.AA AA.AAA AAA.A AA.AAAA AA.AA AA.AA AA.AA AA-AAA-AA
AA.AN AA.AA AA.AAA AAA.A AA.AAAA AA.AA AA.AA AA.AA e-AAA-ee
AA.A~ AA.AA AA.AeA AAA.A AA.AAAA AA.AA AA.AA AA.A~ AA-AAA-AA
AN.A~ AA.AA A~.AAA AAA.A Ae.AAAA AA.AA AA.AA AN.AA A-AAA-eA
AA.AA AA.AA AA.AAA AAA.A ee.AANA AA.AN AA.AA AA.eA AA-AAA-AA
AA.AN AA.AA AA.AAA AAA.A AA.AAAA Ne.AN Ae.Ae Ae.eA A-AAA-ee
AA.A~ AA.AA A~.AAA AAA.A AA.AAAA AA.AA AA.AA Ae.AN AAXAAA
AA.AN AN.AA AA.AAA eAe.A AA.AAA AA.AA AA.AA AA.AA 2AAAAAA
AA.A~ AA.AN AA.Aee AAA.A AA.AAA AA.AA AA.AA AA.AA A-AAN-AA
AA.AN AA.AA AA.A~A AAA.A AA.AAAA AA.AA NA.AA AA.eA AAAA
AA.AN AA.AA AA.AAA ANA.A AA.ANNA A~.AA Ae.AA AA.AA A-AAAx
AA.A~ AA.AA AA.AAA AAA.~ AA.AAAA AA.AA AA.AA AA.AA AAAAAA
AAAAAAAA AAAAAI AAA: AAANA AAwa AAAAN AAAA AAAX AA>AA2AA
.AAAAAA

AALAAAAA AeAeeA AA AAAA AeAeeeA eee AAAA AAAAAA .AzA AAA.» .ANwA eeAA eeee
.A>xv eege ewe: gee Aeeem we geese: .ANV “sowez eeem .A>V eee; Lee
AeeeA we geese: .Axv eeee “we: Lee Aeee; we geese: use Lew Aeewe> :eez

.m wpneh

73

8. HA I

 

 

 

mo. W a e
Ame. —ee. wmm. wee. wee. mom. mmm.- owm. mue.- w:
mN_.- mmo.- omm.- Fem. Nm~.- wmm.- wmm. owe.- a:
emee. eemmu. «eopw. «Awe. mmp. econ. «New.. 03
Nme. «New. mum. «mp. *emm. empe.- A
moe. eemwm. wwo. cum. ow~.- 3
mam. mom.i ¥¥mpw. «xepm.i N>
mmm. mom. mm_.- >x
mmm.- com. N
xeppm.i >
a: a: A 3 N> >x N > x
.Axeewe —egpceuv onv mcweee; ece szv “some; .uceaee_e>ee we eeeem
:eweeeeeeeee ecu cw xeweee we.EequgeE Feewee Ageswge me» we Aezv spew:
Eeswxes ece AAA gumcew Seawxes .sz ewewx .AN>V erA eee: .A>xv eeee
“we: Lee AeeeA we geese: .ANV ucmwez eeeA .A>A eee: Lee AeeeA we geese:
.Axv eege ewe: Lee Aeee; we geese: exp meese Aucewewwweee :ewuepeegeu .m eweew

74

and length and width of meristem in the reproduction state.
Y is, however, significantly and positively correlated with
head size and length (L) and width (N0) of the primary shoot
apical meristem at the reproduction stage. The larger the
size of meristem at the reproduction stage, the larger the
head size, as shown by the significant correlation between
(L), (ND) and (YZ), and the higher the number of seeds
produced per head. Yield (N) is only significantly corre-
lated with number of seeds per unit area and width of
meristem at the reproductive state. Figure 11 shows the
regression of number of seeds per head (Y) on number of
fertile tillers per unit area (X) using mean values for the
standard genotypes. The variation in X accounts for about
92% (R2 = 0.9246) of the variation in Y.

Figure 12 shows the regression of number of seeds per
head on the number of heads per unit area for all
treatments, i.e. control, gibberellin, cycocel, cytokinin
and auxin using the mean values for the standard genotypes.
The points were eliminated to avoid confusion on the graph,
however, it will be presented presently. There were no
significant differences between the regression lines even
though changes in both number of seeds per head and number
of heads per unit area occurred.

Table 10 gives the correlation coefficients among the

Number of seeds per head

70
50
30
15
Figure 11.

 

75

o' v = 93.69 - 1.927x

= O.9246**

20 25 30
Number of heads per unit area
Regression of the number of seeds per

head on the number of heads per unit
area of barley (control block).

Number of seeds per head

  
    
   
     

50

3O

15

Figure 12.

76

Control:

W GA3:

CCC:

IAA:

20 25

Kinetin:

-<-<-<-<-<

93.69 - 1.927X
99.15 - 2.041X
103.11 - 2.171X
95.22 - 1.958X
103.99 - 2.329X

30

Number of heads per unit area

Comparison of regressions of number of
seeds per head on number of heads per unit
area 0f barley due to GA3,CCC, Kinetin and
IAA application at the transition stage of
development with a control.

77

Po. W 3 «¥

 

 

 

mo. W 3 ¥
wa. wom.i m—~.- ewmeN. ANNm. epmm. com. MNm. emm.- o:
mmm.u eem.a mmm.- Noe. mow.- meN.- Nwm. New.. A:
ewomm. mew. wee. NAP. Fem. woe. wNm.- e3
emw. mom. Nmo.- mew. mew. new.. A
«Ammm. *eovm. Noe. *{mmn. «mmm.i 3
eNme. awe. eeeNm. eemem.- N>
Nmo.- *Nme. erm.- >x
NNN. eem.- N
«*mmm.i >
o: o3 A 3 N> >x N > x
.uceEeewe>ee we
emeeA :ewuwmceep esp we :ewpeAAAQQe Am<uv ewppeeeeewm seew mcwuwemee
.erzv mcweee; ece Ahzv unmwe; .uceEeeFe>ee we eAeAA :ewpeeeeeeee
use cw xeweee we seumwees Peewee zeeswge use we Aozv nuewz Eeswxee
ece AAA zumcep Easwxes .sz e—ewx .AN>V erA eee; .A>xv eege “we:
Lee AeeeA we geese: .ANV usmwez emeA .A>A eem; Lee Aeeem we geese:
.Axv eeee awe: Lee Aeee; we geese: m3» mcese Aucewewwweee coweepeegeu .op e—eew

78

plant characteristics of the gibberellin applied block. X
is significantly and negatively correlated with Y, XY, YZ
and N, while Y is significantly and positively correlated
with XY, YZ and N. Z, however, has no significant relation-
ship with any of the plant characteristics measured. Heading
is significantly correlated with XY, YZ, Y and N. Thus,

the longer the heading date, the higher the values for XY,
YZ, Y and N. The maximum length and width of the meristem
in the reproduction state have no significant relationship
with the other plant characteristics but maintained the
significant correlation between themselves.

Figures l3, 17, 21 and 25 show the regressions of number
of seeds per head (Y) on the number of fertile tillers per
unit area due to hormonal application. The star and circle
points are for control and a hormone, respectively, and
each genotype has its values joined by the dashed line.

This technique is adopted to show the change that the
individual genotypes underwent as a result of hormonal
application.

Nith gibberellin application (Fig. 13), changes were
obtained in X for genotypes with the least and highest
number of heads per unit area, while changes in Y were
frequently observed with the outliers and with genotypes

producing medium number of heads per unit area of the group

UGO: Lwa mUmow we LQLE:E

Number of seeds per head

70

50

3O

15

Figure 13.

   

79

3;.
o”
_(

II

99.15 - 2.041X
' O.8964**

20 25 30
Number of heads per unit area

Regression of the number of seeds per head

on the number of heads per unit area of
barley due to GA3 application at the transi-
tion stage of development. (Stars and circles
represent the control and GA3 applied
characteristics, respectively, and individual
genotypes are joined by broken lines).

80

defined as standard genotypes. Associated with these
changes, were small changes in Y and X, respectively. Thus,
with gibberellin application, the obvious changes in X and

Y depended on the tillering capability of the genotype in
question.

There were significant differences between the genotypes
for the percentage change over control for most of the plant
characteristics due to gibberellin application (Table A1).
In both the standard genotypes and the outliers, genotypic
yield changes declined with increasing number of heads per
unit area. Dickson had the highest positive yield change
while 60-215-6 had a decrease in yield with gibberellin
application. Nith the outliers, Bowers and X969-3 had the
highest and least changes, respectively (Fig. 14).

The histograms of changes in YZ and XY were drawn to
determine whether the changes in N was due mainly to changes
in head size or number of seeds per unit area. The graph
of percentage change over control for YZ show that with
increasing number of heads per unit area, head size increases
to a point and start decreasing for the standard varieties
while the outliers had an increasing change with increasing
X. Changes in number of seeds per unit area, however,
decreased to about the control mean values with increase in

X. The outliers, however, showed an initial decrease and

20

-20
20

Percentage change in plant characteristics

with respect to control

81

 

1 HEAD SIZE

411147410
4 Lu

 

 

i SEEDS PER UNIT AREA

:1
:1

E

 

103-18
103-8
104-19
104-3
215—6
105-9
105-17

DICKSON
8130
68-
68-
68-
LARKER
68—
0-
ONERS
8-
8-
X969-3

(D
m
2
O
_‘
_<
'U
111
(I)

Figure 14. Percentage change in yield, head size and

number of seeds per unit area of barley due
to GA3 application at the transition stage
of development.

82

then an increase as fertile tillers production increased
(Fig. 14).

Changes in average seed weight showed no trend in the
standard varieties but an initial increase and then a decrease
was observed with the outliers, as fertile tiller production
increased (Fig. 15). The changes shown in X and Y due to
gibberellin application are similar to that shown in
Figure 13. Genotypes with the least and high X increased
the number of heads produced per unit area with slight
decreases in the number of seeds produced per head. Genotypes
with medium number of heads per unit area produced more seeds
per head with small changes in their X.

Heading and height (Fig. 16) were generally reduced in
all the genotypes but not at significant levels.

The correlation coefficients obtained from the CCC applied
block for the measured plant characteristics are given in
Table 11. X maintains its significant negative correlation
with Y and YZ while Y has significant positive correlations
with XY, Y2 and N. Nidth of meristem in the reproduction
state has significant relationships with Y, XY, YZ, N, L and
HT. Heading has a significant negative correlation with seed
size. Number of seeds per unit area is significantly
correlated with head size and yield.

The graph of number of seeds per head on number of fer-

tile tillers per unit area due to the application of CCC

83

 

 

 

 

HEADS PER UNIT AREA
20 i
10 -
o .
U1
Q
t” -10I
E
OJ
4.)
U
2 20 4 SEEDS PER HEAD
2
U
H 10. PH
C
(U
“'8 0 . I I , ,
1:44
w—C
O
3" -1o .
CO
(64-)
: SEED HEIGHT
U-H ______—
a 8 10
07C}.
MU)
E S
m 0
US
5.4-1
(1)-r-
9 3 —10
00 0'1 [\
I—WF- MED Chr—
2 III II ll
0 MMQQCQ'LD mmmm
V) OOOUJOI— dOOI
XOI—r—r—34u—N LIJr—r—O‘l
Umlllofill 3|I©
HFCOGDW<®O OGDCDCh
Iolmlkol‘oJkDIJI‘ol‘ol lml©1©l>:1__
GENOTYPES

Figure 15. Percentage change in number of heads per unit
area, number of seeds per head and seed weight
of barley due to GA3 application at the
transition stage of development.

Percentage change in plant character-
istics with respect to control

Figure 16.

84

1O HEADING

 

m C5 l\

r— 00 I'- (*3 KC on r-
2 I I I I I I I
O m m Q' 0: <1' LG (1) LO LO 0’)
V" O O 0 LL! O r— m 0 O l
3‘ l— I'— r— x I- N LL] l—' l_' 01
U I I I a: 3 I so
H w w CD < O w 0‘
Q £0 £0 $0 _.| d3 \0

       

     

GENOTYPES

 

Percentage change in heading and height of
barley due to GA3 application at the
transition stage of development.

Po. W 3 AA

 

 

AA. w A e

AAA.- AAA. AAA. AAA. AAA. AAA. AAAA.- AAA. AAA.- A:
.eAe.- AAA.- AAA.- AAA.- AAA.- AAA.- AAA.- AAA.- A:
«IAAA. eeAeN. .eAeA. eeeAA. AAA.- eeAeN. AAA.- A3

AAA. AAA. AAA. AAA.. AAA. AAA.- A

AAA.- aeAAA. AAA. eAAA. AAA.- 3
*mmm. coo. *eomm. *«wa.1 N>

A. AA..- 4AAA. AAA.- >x
8 AAA: AAA. N
#«mmw.i >
A: A3 A 2 NA Ax N > x

 

.pceEee_e>ee we emeuA :ewuwAceLe one we :ewueewweee Auuuv weeeexe
Eeew mewpwameg Aozv mcweee; ece szv uzmwe; .Aceseewe>ee we eueem
:ewueeeeeeee e3» cw xeweee we Eeumwees weewee Ageswge may we Aozv spew; Eeswxes
ece AAA sauce, Eeswxes .A3A e—ewx .AN>V eNAA eee; .A>xv eege Awe:
Lee AeeeA we genes: .ANV usmwez eeeA .A>v eee; Lee AeeeA we geese:
.Axv eege Awe: gee Aeee; we geese: egu mcese Aeeewewwweee :ewueweegeo .Aw eweew

86

is shown in Fig. 17. There was virtually no change in the
points for the genotypes with highest X. 8130 has the
greatest change in number of heads produced for CCC applica-
tion followed by Dickson. All other standard genotypes
plus the outliers increased their number of seeds produced
per head. The highest increase in Y was obtained with
68-103-18 from the standard genotypes and 68-105-17 from
the group of outliers. The increase in Y was accompanied
with almost no change in X.

There were no established trends for the percentage
change over control for yield, head size and number of
seeds per unit area (Fig. 18) although significant differences
exist between the genotypes (Table A2). Some genotypes had
increased yield while others were not changed substantially.
The changes in N were due to changes in YZ for 68-103-18
and XY for 8130 and Larker. The change in N for Bowers
is mainly due to its change in XY while that of the other
three outliers were due to YZ.

In determining the yield component prone to change with
cycocel application, it was observed that Dickson, 8130
and Bowers had relatively higher increases in X while in
68-103-18, 68-105-17, X969-3 and Larker increases in the
number of seeds produced per head were obtained. Average

seed weight was increased substantially in Dickson and

Number of seeds per head

70

 

15

Figure 17.

87

.<
ll

103.11 - 2.171X

O.8603**

20 25 30
Number of heads per unit area

Regression of the number of seeds per head

on the number of heads per unit area of barley
due to CCC application at the transition stage
of development. (Stars and circles represent
the control and GA3 applied characteristics,
respectively, and individual genotypes are
joined by broken lines).

20

10

I
_.a
O

I
N N
O O

—l
O

~10

Percentage change in plant characteristics

with respect to control

Figure 18.

88

 

HEAD SIZE

 

 

SEEDS PER UNIT AREA

 

 

 

a) m I\

F' m I'- W KO 0‘ r-
2 I I I I I I I
O 0") m 6’ O: Q' LO (OLD L0 or)
W O O 0 LL! 0 !— DCO O I
¥ 0 I'- l— t— M l— N “JF- I- 0‘
U m I I I m I I 3I I O
H I- m w m < m 0 om w 03
Q m 0 KO \D _| \O \0 m0 O X

GENOTYPES

 

Percentage change in yield, head size and
number of seeds per unit area of barley due
to CCC application at the transition stage
of development.

89

reduced in Larker (Fig. 19).

Heading was increased in 68-105-9 while height shows
not much change as compared to their controls (Fig. 20).

The correlation coefficients among plant characterist-
ics measured from the cytokinin applied experimental block
are given in Table 12. Length and width of meristem in
the reproduction state are both significantly correlated
with XY, N and HT while the width is additionally correlated
significantly with Y, Y2 and HD. Yield is significantly
correlated with XY and HD. X, however, is negatively and
significantly correlated with head size and the number of
seeds per head.

Cytokinin application tended to change number of heads
per unit area, generally, in most of the genotypes used in
the experiment. 68-103-18, 68-104-19, Larker and 68-105-17
had virtually no change in both X and Y (Fig. 21).

Only 60-215-6 and 68-105-17 had a reduction in yield
relative to the control while the rest had some form of an
increase in yield (Fig. 22). X969-3 had the biggest
increase in head size while 68-103-18 had the largest
increase in the number of seeds per unit area. The
manifestation of yield changes in terms of X, Y and Z,
varied among the genotypes (Fig. 23). 60-215-6 and X969-3

both had increased Y but decreased X while Dickson, 8130,

Percentage change in plant character-
istics with respect to control

90

 

 

 

 

20 . HEADS PER UNIT AREA
10.

U JJ
-10 ,

SEEDS PER HEAD

20
O
-10

SEED WEIGHT

 

10
0
~10
(X) OS I\
I— co !— m \O 0‘ r—
2 I I I I I I I
O m m V Q: Q‘ LO m LO Ln (‘0
(f) O O O LIJ C r— CZ C) O I
x O l— I— r— ¥ r- N LLI :— '— 0‘
Q m I I I CZ I I 3 I I KO
H r— 00 oo 00 < m C O 00 00 01
D co 0 to ID _I KO \O co KO 0 ><
1 I l n l n I I l I I l l

 

GENOTYPES

Figure 19. Percentage change in number of heads per

unit area, number of seeds per head and
seed weight of barley due to CCC application
at the transition stage of development.

91

mumomx
Nwimowuwo
mime—two
mmmzom

oimpmioo
mueowiwo

mmxm<A

 

AA-AAA-AA
m T AAATAA
m en. ATAATAA
w. “1.. AA;
ZAAAAAA

0 O O 0

weeecee e» peeeAeL new;
Aequweepeeeece ecewe cw emcege emepceegee

     

      

  

GENOTYPES

 

Percentage change in heading and height of
barley due to CCC application at the

transition stage of development.

Figure 20.

92

 

 

 

of...
mo.WA_«.
mN~.- «pm. «woeN. Ammo. owe. eeeNN. mo~.- woe. ooN.- o:
emNm.i «moo.- oN¢.- Now.- N—e.i omo.- mow.- Nmo.- w:
«Nmo. «oeo. «oom. «eooN. om¢.- eooo. wem.u o3
Acme. oom. NemmN. oo~.- mmm. o¢~.- A
omm. AANNo. NNo. oNe. omw.- 3
owm. mow.- AAPNo. eemoo.- N>
owe.- eon. om_.- >x
mmm.- ooo. N
eemoo.- >
o: o3 A 3 N> >x N > x
.Aceseewm>ee we eoeuA :ewuwmcegp esp ue.:ewueewweee Acwuecwxv :wcwxeuxe
segw ocwuwemee Aomo oeweee; ece .szv usowe; .uceseewe>ee we euepm
:ewueeeeeeee cw Aeweee we Eeamwges weewee Ageewge use we Ao3v spew: seswxes
ece .AAV Apocew seawxes .sz ewewa .AN>V eNAA eee; .A>xv eeee ewe:
Lee AeemA we geese: .ANV pcowez eeeA .A>v eee; Lee AeeeA we geese:
.Axv eege awe: Lee Aeee; we geese: ecu ocese Aucewewwweee :ewueweegeo .Nw eweew

Number of seeds per head

70

50
30
15
Figure 21.

 

93

95.22 - 1.958X

\~g Y =

' O.7708**

20 25 30
Number of heads per unit area

Regression of the number of seeds per head
on the number of heads per unit area of
barley due to kinetin application of the
transition stage of development. (Stars
and circles represent the control and GA
applied characteristics, respectively, and
individual genotypes are joined by broken
lines).

20

I
—l
O

-20
20

Percentage change in plant characteristics

with respect to control.

Figure 22.

94

 

 

 

 

 

 

11:13”

HEAD SIZE

SEEDS PER UNIT AREA

 

 

CO on r\

I— 00 r— M £0 0" r—
2 l I I I I I I
O m 0") <I‘ QC <r L!) (f) L0 1.0 m
(I) O O O LLI C) :— at O O I
>4 0 I— r— r- M r- N LLI :— r— 0‘
(A) m I I I o: I l 3 I I to
H r- co co 00 < CO C O CO oo 0‘
(2- ml OIQAmJ—Jnm-cl .m-kD-KO- >i

GENOTYPES

Percentage change in yield, head size and
number of seeds per unit area of barley

due to kinetin application at the transition
stage of development

95

20 HEAD PER UNIT AREA

 

 

20 SEEDS PER HEAD

 

 

lO

 

Percentage change in plant character-
istics with respect to control

 

I
_a
O

-17

104-19

104-3

215-6

105-9
3

(I)

1— oo
I I

m m

O O

l—' I'-

0: (I)
1.1.! or.
CD )4 LIJ on
m I I I a: I I 3 I I to
I— oo 00 CD < 00 O C CD 00 05
O on LO 0 to _I 0 O m to L0 ><

ICKSON
105

GENOTYPES

 

Figure 23. Percentage change in number of heads per
unit area, number of seeds per head and
seed weight of barley due to kinetin
application at the transition stage of
development.

96

68-104-3, Bowers and 68-105-9 showed the opposite effect.
Their number of heads per unit area were increased while
their number of seeds per head decreased. 68-103-18
increased its number of fertile tillers and number of seeds
per head produced. Even though there were slight changes
in the average seed weight produced, there was no
established trend and the changes were not significantly
different from the control. The number of days to heading
and height were not changed significantly from controls
(Fig. 24).

Significant differences exist between the genotypes
for the plant characteristics measured after the cytokinin
application (Table A3).

Correlation coefficients among plant characteristics
developed after auxin application are presented in Table 13.
Y, XY, YZ and N, are significantly and positively
intercorrelated with each other. X is significantly and
negatively correlated with Y and Y2. Length and width of
the primary apical meristem in the reproduction state are
significantly correlated with Y, XY, Y2 and N, but negatively
correlated with X.

Figure 25 shows the effect of auxin application on the
regression of Y on X. Changes in both X and Y are observed

but larger changes are observed in the number of seeds per

head.

97

muooox
Neumowiwo
mimowiwo
mmmzom

mum—Nico
mice—two
mm¥m<A
o—ieowuwo
wumopimo
mwumowimo
omwm
zomxooo

HEADING

 

wecpcee ee eeeeAeg new: Aequw
-Leeeeeege ucewe cw mocece eoeeceegeo

GENOTYPES

 

Percentage change in heading and height of

barley due to kinetin application at the

transition stage of development.

Figure 24.

98

—o. W a *A

 

 

 

AA.wA..
omm.- ooe. Nme. one. Pom. eooo. Ne~.- ome. Pom.- o:
NA¢.- owN.- mwm.- mop.- omm.- ooo.- om_.- ooF.- w:
*xmmm. ¥¥mmm. eepmw. «kmmm. —op. «¥wwm. *NON.1 oz
*emeN. eemmo. Aeomo. owo. «emoo. «woo.- A
«ANFN. «Aooo. Ame. aweo. wom.- 3
aooo. mom. eAeNo. «emoo.- N>
Nwo. eooo. oem.- >x
Nmo. mww.- N
«enmw.i >
o: o3 A 3 N> >x N > x
.AceEeewe>ee we eoeuA :equAceLA e3» ue.:ewueewweee A<<Av cwxze
segw ocwuwemeg Aozv oeweee; ece szv ucowe; .ucesee_e>ee we eueaA
coweeeeeeeee cw xeweee we EeAAwgee weewee zeeawee one we Aozv spew; Eeewxee
ece AAA geocew Eeswxes .A3A ewewx .AN>V e~wA eee; .A>xv eeee ewe:
Lee AeemA we Lease: .ANA pcowez eeeA .A>V eee; Lee AeeeA we geese:
.Axv eeee Awe: gee Aeee; we geese: ecu ocese Aecewewwweee :eweeweeeeo .mw eweew

Number of seeds per head

99

103.99 - 2.329x
7 o ‘\

’0
..<
N
II

O.8109**

 

15 20 25 30
Number of heads per unit area

Figure 25. Regression of the number of seeds per
head on the number of heads per unit area
of barley due to IAA application at the
transition stage of development. (Stars
and circles represent the control and GA3
applied characteristics, respectively, and
individual genotypes are joined by broken
lines).

TOO

Significant changes exist between the genotypes for
the percentage change over control on most of the measured
characteristics (Table A4). The yield changes of the
standard genotypes establish a trend when one considers all
but 68-103-8 and 68-104-19. Starting with Dickson, there
is an increase in the yield changes until genotype 68-103-18
(max. % yield increase) after which there is a reduction in
the percentage increase until a reduction in yield is
attained with the two genotypes producing high number of
heads per unit area. Auxin decreased the yield of Bowers
but had no effects on 68-105-9, 68-105-17 and X969-3 (Fig.
26). The trend in the changes in head size among the
standard genotypes are similar to that of the yield changes.
68-105-9, 68-105-17 and X969-3, however, had an increase
in head size. The number of seeds per unit area show a
decreasing trend with increasing number of heads per unit
area.

Auxin decreased the number of heads per unit area in
60-215-6, 68-105-9, 68-105-17 and X969-3. The other geno-
types were not affected significantly (Fig. 27). Changes
in the number of seeds per head were similar to those of
yield and head size. Number of seed per head for 68-105-9,
68-105-17 and X969-3 were significantly increased. The

average seed weight was generally reduced in almost all the

20
10
O
U)
U
3 —10
Z
S
U —20
E
m 20
.C
U
2 10
(U
ES
:3: 0
'r—C
O
(DU
30 ”IO
row:
.C
0+:
U
(DC)
35% 20
4-10)
CS-
0)
:35 10
0):!-
w 3
0
-10
Figure 26.

101

3%

i HEAD SIZE

2W

SEEDS PER UNIT AREA

 

 

 

E
in

I._l

 

Q) 01 l\

v- m r— ('0 Lo 0‘ '—
Z I l l l I | l
O m (*1 <1“ 0: <9“ 1.0 U) LO LD 0')
U7 0 O O LIJ O r— O: O O l
X 0 I— l— I-- M l— N LIJ !— r— 01
L) 0") I I I O: l I 3 I l \D
H I— 00 (D 00 < d) O 0 oo 00 Ch
D :2: NO to L0 _J ND ND an L0 RD ><

l n l I J 1 J j J j I I -
GENOTYPES

Percentage change in yield, head size and
number of seeds per unit area of barley due
to IAA application at the transition stage
of development.

102

20 HEADS PER UNIT AREA

 

10

 

I
._I
O

SEEDS PER HEAD

 

 

 

Percentage change in plant characteristics
with respect to control

I
:5
-19
-5-
—17

DICKSON
8130
68-103-18
68-103-8
68-104
LARKER
68-104-3
60-215
BONERS
68-105-9
68-105
X969-3

GENOTYPES

 

Figure 27. Percentage change in number of heads per
unit area, number of seeds per head and
seed weight of barley due to IAA application
at the transition stage of development.

103

genotypes with the greatest change observed in 60-215-6
followed by 68-104-19, X969-3 and B130. A clear case of
component compensation is shown by the outliers. This
phenomenon is also observed in some standard genotypes
while the others had no change in X with an increase in Y.

No significant changes were observed in the heading
and height graphs (Fig. 28).

The highest change in yield among the genotypes was
obtained from cycocel application followed by gibberellin,
auxin and cytokinin in that order. Table 14 gives the
correlation coefficients among the changes that were
observed in the measured plant characteristics. Generally,
a positive change in the number of heads per unit area is
associated with a negative change in the number of seeds
per head or vice versa for all the treatments. Changes in
head size are also substantially reduced in all treatments
but auxin. The correlation coefficient is low even though
it is negative (p=-.259). Changes in head size are highly
correlated with changes in Y in all treatments but cytokinin
(p=.221). XY changes correlate positively and significantly
with yield changes (all treatments), changes in X (all
treatments but auxin) and head size (auxin alone).

The type of change in the yield components or combina-

tions of them is dependent on the genotype and treatment

104

muooox
Npimownwo
onmowimo
mmmzoo

oimpmioo
mieowiwo
mmxm<A
opieowuwo
mimowumo
mpimowiwo
omwm
zomxooo

    

HEADING
HEIGHT

O 0 0 O 0 0
I I I I
. .

weeucee ea HeeeAeL AAA; AeweAwLe
-ueeeeze ucewe cw eocece eoeucmeeee

 

GENOTYPES

 

Percentage change in heading and height

Figure 28.

of barley due to IAA application at the

transition stage of development.

105

 

 

 

Table 14. Correlation coefficients among the percentage
change over control for the number of heads per
unit area (AX), number of seeds per head (AY).
seed weight (AZ), number of seeds per unit area
(AXY), head size (AYZ), yield (1M), maximum
length (AL) and maximum width (AND) of the
Primary apical meristem of barley in the reprod-
uction state of development due to the application
of gibberellin, cycocel, cytokinin and auxin,
respectively, at the transition stage of development.
AN AZ AYZ AL AND AX AXY

AZ .248
.048
.529
.542
AYZ .222 .544
.416 .172
.050 .060
.803** .383
A. -.590* .084 —.305
-.331 .295 -.082
-.267 .039 -.127
-.257 -.310 -.305
AND -.474 -.221 -.367 .783**
-.325 .529 .189 .744**
-.281 -.277 -.042 .200
-.115 .236 -.l79 .101
AX .519 -.225 -.7ll** -.133 -.053
.493 -.l35 -.570 -.282 -.531
.719** .366 -.652* -.l64 -.229
.362 .338 -.259 .075 .188
AXY .718** -.492 -.l94 -.582* -.245 .633*
.842** -.496 .251 -.451 -.563 .521
.8l9** .049 -.004 -.347 -.128 .611*
.829** -.017 .680* -.O67 -.273 .238
AV -.l4l .456 .834** .034 -.027 -.811**-.443
.160 .026 .833** -.100 .069 -.581* .118
-.567 -.707** .221 -.221 .134 -.548 -.212
.405 .006 .773** .050 -.O73 -.539 .463
* P g .05
** P g .01

106

under consideration. Genotypes producing low and high number
of heads per unit area are prone to changes in X while the
genotypes producing medium X have the least resistance to
changes in their number of seeds per head.

Nith auxin treatment, AX has high negative correlations
with AY. AXY has significant positive correlations with
AN, and AYZ. Generally, the correlations between XY with
X and Y is positive and negative, respectively or vice
versa, depending on the significance of the relationship
between X and Y. The positive correlations (not significant)
between AXY with AX and AV Show some relaxation between
A X and AY due to auxin application. A similar observation
is shown with the cycocel application even though the
correlation between AXY with AY is low (0 =.ll8). The
effects of gibberellin and cytokinin application follow
the normal trend, however, the correlation between AXY with
A Y is lower for the cytokinin treatment (0 =-.212) in
comparison with the gibberellin treatment (0 =-.4l4). The
above effects are also depicted in the correlation coeffi-
cients between AX and AY (p =-.8l4, -.581, -.548 and -.539
for the GA3, CCC, Kinetin and IAA treatments, respectively).

Removing the outliers causes the effect to show up
better. The correlation coefficient between AXY with AX

and AY are all positive, however, relaxation of the AX and

107

A Y relationship is shown in the cycocel (p =-.368) and
auxin (p =.204) treatments. The same relationship Show
significant negative correlation for both the gibberellin
(p =-.706**) and cytokinin (p =-.630**) treatments (Table
15).

Table 16 presents the mean square values for the
multiple regression analysis involving the length and width
of the primary apical meristem in the reproduction state
of development as the independent variables and plant
characteristics measured from GA3, CCC, Kinetin, IAA and
the control plots of the second group of experiments as
dependent variables. No significant prediction was obtained
for all characteristics resulting from the GA3 treatment.
Seed weight was not significant for the regressions in all
the treatments. Additionally, no significant relationships
were obtained for XY (control), X(CCC and Kinetin) and Y2
(Kinetin). The coefficient of determination for the
Significant characteristics were generally higher in the
IAA treatment, followed by CCC, Kinetin and Control in that
order. The width of the meristem is shown to be the more
important independent characteristic in the prediction of
the plant characteristics (Tables A5-A8) in all the

treatments.

108

 

 

 

Table 15. Correlation coefficients among the percentage
change over control for the number of heads per
unit area (AX), number of seeds per head (AY),
seed weight (AZ), number of seeds per unit area
(AXY), head size (AYZ), yield (AN), maximum
length (A1.) and maximum width (AND) of the
primary apical meristem of barley in the reproduct-
ion state of development dde to application of
gibberellin, cycocel, cytokinin and auxin,
respectively, for the standard genotypes.

AN AZ AYZ AL AND AX AXY
AZ .630*
.111
.303
.567
AYZ .397 .524
.451 -.054
-.087 -.249
.939** .579*
AL -.609* -.489 -.875**
-.450 .295 -.247
-.390 —.058 -.487
-.292 -.348 -.559
AND -.554 -.806** -.713** .784**
-.311 .448 .106 .850**
-.410 -.6l6* -.212 .212
-.303 .130 -.424 .068
AX .529 .105 -.566 .253 .132
.618* .107 -.410 -.298 -.477
.854** .424 -.585* -.101 -.246
.623* .328 .321 .430 .181
AXY .757** -.029 .092 -.395 -.065 .571
.840** -.444 .427 -.555 -.519 .508
.848** -.247 .042 -.377 -.092 .639*
.831** .016 .723** -.082 -.437 .573
AY .120 .038 .870** .768** -.389 -.706**.l47
.318 -.586* .838** -.372 -.l83 -.368 .598*
-.l90 -.699* .865** -.371 .157 -.630* .190
.780* .035 .834** -.426 —.595* .204 .904**
* P < .05

** P§.oi

109

 

 

 

Po. W a e*« mo. W a «* op. W a e
owoo. omNo. NooN. Nemo. Nemm. N3
mm.oooN momo. oo.omom A: mm.oN ow.o o Leggy
*xmo.—mwmp *«ecomo. #«xwm.mmwpo «eoo.omm «*@N.m¢ N cowmmmwmmx <<H
123..----11-11-1331---..---:NNTAA:----1------ ...... ------mm----------
me.mon A: wo.emmm A: Ne.oo A: o Leesm
eemm.wwwo eeeNo.momoe eoo.mmN N :ewAAeLoem :wue:A¥
--NNNNN---------ANAA1-MANAN----- ...... Naomi--- ............... -Am
mN.onw Nmoo. eo.eNmm A: ww.oo A: o Leggm
«eaeo.wNmm— «AoNee. «eamN.momNm «epo.—Ne N :ewAAegomm ooo
iiiuiiiiiiiiiiiuiu iiiiiiiiiiiiiiiiiiiiiiii I ..... i 11111111111111 iinuiiimm iiiiiiiiii
A: A: A: A: A: A: o :eLLN
N :ewAAeLoe: m<o
:ANNWA ......... NANA. .................. Emailmmmmn .............. 1M... ..........
oN.oow— ono. A: A: eo.oe N_.w_ o Leegm
eemo.moom «Aonm. «ewo.¢mN «om.me N :ewAAeeoe: wegu:eo
3 N» >x N > x we eegeem
.A:eE:ewe>ee we eoeeA :ewewmceeu e:e we .:ewueewweee :wxee
e:e :w:wxepxe .weeeexe .:wwwe:eeewo e» e:e Aeweewee> e:eeceeee:w e:e Ae
ucmsoo_w>we wo mpeum :owuuoeowamg 03p cw xmpwen wo EwumemE powwow hLeEwLQ mzp
we Ao3v :eewz seswxes e:e AAA :po:mw Eeswxea e:e :e eweewee> a:ee:eeee
e Ae :eee .A3v ewewx e:e A>xv eeee uwce Lee AeeeA we geese: .AN>V erA
eee: .ANV u:owe3 emeA .A>v eee: Lee AeeeA we geese: .Axv eeee aw:= Lee
Aeee: we geese: ecu we :ewAAegom: ewewawee e:e Lew we:ew:e> we Awmxwe:< .ow eweew

DISCUSSION

Tiller formation is one of the first developmental
processes that occur at the organ level. Once formed, the
growth and deve10pment of organs laid down later in the
plants' ontogeny are established. The higher the number of
tillers produced by a genotype. the smaller the size of
head it produces. Culm diameter, leaf size and number of
seeds per head are reduced (Grafius _t _l., 1976; Hamid and
Grafius, 1978, Nhyte 1979). The relative alteration in the
average seed weight is dependent largely on environmental
factors although the potential seed size is normally
determined by the genetics of the genotype in question
(Grafius, 1978).

The apparent effect of tillering on the growth and
development of organs produced later in the ontogeny of
the plant resides in the relationship that tillers have
with the shoot meristem. Meristems are localized regions
of mitotically active cells which are of diverse morphology,
a reflection of their mitotic activity, and location
(Sussex, 1963). Some function continuously throughout the
life of the plant (shoot apical meristem of annual plants),

others are persistent but seasonally intermittent (terminal

and lateral meristems of perennials) while others are

110

111

transitory and temporary such as the leaf apical meristem.
From the time of its initiation each meristem is stable,
however, changes occur within the meristem and its products.
Some of these changes are gradual, such as the ontogenetic
size change in a meristem. Others are sudden as in the
conversion of a vegetative shoot meristem into a flower
meristem.

In cereal plants, the above ground organs evolve from
the shoot apical meristem. The main shoot is developed
with the apical dome initiating acropetally a succession
of primordia to form its leaves. Each primordium unit later
differentiate an elongated internode and/or a tiller bud
and subsequently tillers and main shoot develop floral
primordia.

The results presented here and by Nhyte (1979) show
the nature of the relationship between tiller production
and the size of the primary apical shoot meristem. The
higher the number of tillers produced by a genotype, the
smaller is the size of the primary shoot apical meristem
it produces. Since organs arise from meristems, it is
reasonable to expect inter-organ associations to be highly
correlated. The correlation coefficients of stem diameter
with head size and number of seeds per head are .861** and
.794**, respectively (Nhyte, 1979) and culm diameter with

leaf area and head size are .806** and .798*, respectively

112

(Hamid, 1976). As pointed out by Adams (1975) and Grafius
(1978), high yield potentials are achieved through a
balance between factors of numbers (e.g., number of nodes,
number of tillers) and factors of size (e.g., stem diameter,
leaf area, head size, pod size). A balance is therefore
established between the number of tillers per unit area
with culm size, leaf area and head size through the
relationship that the number of tillers and meristem size
have with each other.

The early differences in time of differentiation and
rate of spikelet development are reflected in the mature
plant characteristics. There was a 4-day delay for the
transformation of the primary apical meristem from the
appearance of double ridges to the onset of spikelet
differentiation, among X969-3, 68-105-9, 68-105-17 and the
other genotypes. This delay allows for a larger sized
meristem to form. An extra surface is provided for the
development of additional whorls of seeds given the number
of tillers they produce per unit area. Lee gt gt. (1974),
Nilliams (1975) and Blum (1977) found that a delayed and
larger basal branch at the time of spikelet initiation
allowed for the formation of more spikelets, florets and
grains in sorghum. Although the development of the meristem

is delayed in the three genotypes, the difference in time

113

is compensated for by having greater rate of development
once the full set of floral primordia is established.
Equivalent heading dates are realized. Nhyte (1979) suspected
that the larger reproductive apex at the initiation of
elongation in X969-3 and the progeny lines with similar
behavior could be traced back to the larger vegetative apex.
In effect the gain in the number of seeds per head may be
established in the first developmental phase with all the
physiological and practical implications.

The path coefficients between the meristematic
characteristics at the reproductive state of growth with
the number of seeds per head and the average seed weight
confirm the above consideration (Figure Al). The width of
the apical meristem has its greatest influence on its
length and number of heads per unit area. Its relationship
to length (L) and number of heads per unit area (X) are,
respectively, positive and negative. Both length and
relative growth rate (R) are, however, positive determinants
of X. Nidth of the meristem (ND) has a positive influence
on the number of seeds per head (Y) while R, L and ND have
negative influence on Y. Y, L, R and ND are all negatively
related to seed weight (Z). X, Y and L are, however, more
important determinants of Z relative to R and ND.

The analysis of variance of the number of heads per

unit area, number of seeds per head and head size using the

114

maximum length, maximum width and the relative growth rate
at the reproductive state of development show that a
significant portion of the variation in the dependent
variables can be accounted for by the variation in the
three components. The coefficient of determination for X,
Y and YZ are .5959, .7156 and .7912, respectively. As
stated earlier, the salient feature of the multiple regres-
sion statistics show that the higher the number of tillers
produced by a genotype, the smaller will be the width of
the meristems produced, however, the relative growth rate
of meristem will be higher. Number of seeds borne per head
will be small since width has a positive correlation with Y.

From the above, one can deduce that a genotype with a
low growth rate produces a lower number of heads per unit
area. A further relaxation of growth rate encourages the
formation of organs (i.e., meristems) with larger width.
Since the width determines the length that the organ assumes,
an apical meristem with a large surface area is produced.
A higher number of floral initials are borne resulting in
the production of a higher number of seeds per head and
eventually, a bigger sized head. This confirms the fact
that sizes and numbers of plant organs are negatively
correlated (Grafius, 1978).

Leopold (1949) contended that the relationship between

the size of meristems and plant organs could be modified by

115

changing the internal hormonal content. This led to
research into attempts to change plant characteristics and
hopefully, yield through hormonal application. The
applications were done without much regard to the time best
suited for its effectiveness. There were applications
through seed soaking, spraying at one leaf stage and at
bolting, to mention but a few. Often a few genotypes
without varying yield components, if ever considered, were
used resulting in inconsistent results.

Much is realized when one considers the effects that
hormones have on the characteristics of meristems, the
determinants of organ size and numbers, when the applica-
tions are done at the right stage, given the objective of
the study. Nhyte (1979) showed that about 52% and 65% of
variation in X and Y, respectively, could be accounted for
by the variation in length, width and the relative growth
rate of the primary apical meristem at the reproductive
state of development. A similar determination is presented
in the data (60%, 71% and 79% of the variation in X, Y and
YZ, respectively, is accounted for by the same independent
variables).

The application of GA3, CCC, Kinetin and IAA induced
changes in both the maximum width and maximum length of

the primary apical meristem at the reproduction stage of

116

development. The magnitude of these changes depended on
the type of hormone applied and the genotype under
consideration. Basically, length measurements were changed
to a greater extent than the width measurements. GA3 induced
the greatest changes in the length, followed in decreasing
order by CCC, Kinetin and IAA. The length of meristems
of the genotypes with the least and highest X were increased
while those of the genotypes with medium X were decreased
in all the treatments. Only the least and highest tiller-
ing genotypes had significant width increases with the GA3
and CCC application. Kinetin and IAA however, were more
general in their induction of width changes among the geno-
types. The differences in reaction of the meristems to
the applied hormones is expected because at the stage of
development when the applications were carried out, the
maximum width of the genotype was about attained leaving
only the length to be subjected to greater changes.

Let us consider the meristem as the initial point at
which seedlings react to applied hormones. Since length
is more prone to change, any disturbance in the internal
equilibrium will be observed in the length changes, graphi-
cally shown as a wave pattern.

This relationship is shown by changes in length of

the apical meristem due to disturbances caused by the

117

applications of GA3, CCC, Kinetine and IAA. The establish-
ment of these wave patterns Show that the promotive hormones
are present in equilibrium with each other for normal growth
and development of the plant. Gibberellin has been docu-
mented as being the hormone involved in the increase in
length of barley meristem in the reproduction state. This
is not necessarily the totality since it is now being shown
that the other two promotive hormones are involved in the
growth and developmental processes. Any disturbance in

the hormonal equilibrium results in a series of changes
which is ultimately expressed in either an increase or a
decrease in the size (length) of the meristem. One is
cautioned to think that only the promotive hormones regulate
growth and development of the meristem, but that the
inhibitory hormones also play a large role in the regulatory
processes. Gibberellin, however, induces its effect on a
wider range of genotypes followed, in a decreasing order,

by cycocel, cytokinin and auxin. ‘The largest length
reduction was observed with 68-103-8 in all the treatments,
however, the largest length increase varied between

68-104-3 and 60-215-6, depending on the treatment. The
effectiveness of utilizing any of the promotive hormones

for inducing changes in the meristem of a genotype is thus

dependent on the genotype and the hormone in question. The

118

stage of development at which application is done is very
critical because induction of activity of the genetic
constituents does not occur haphazardly, but in an orderly
fashion coupled with the short range of time during which
hormones act. .

Some changes were observed in the outliers but because
of the insignificant differences in the number of heads
they produce per unit area, I am bound to view the changes
as genotype-specific.

Yield changes were observed as a result of application
of the promotive hormones. The changes were due to the
changes in meristematic sizes. Even though the meristematic
changes followed some wave pattern, not all the changes in
the yield components and yield followed the same trend.
However, the type of change also depended on the hormone
applied and the genotype in consideration.

The change in yield due to GA3 application decreased
with increasing tillering. A clear case of component
compensation is shown in changes observed in the number of
heads per unit area and the number of seeds per head.
Genotypes with the least and high number of heads per unit
area had the greatest changes in X while the genotypes
producing medium number of heads per unit area were more

prone to changing their head size. Changes in the average

119

seed weight was generally between +5 and -5%.

CCC induced yield changes followed a similar trend as
that of the GA3 application, hOwever, these changes are not
consistent. 8130 and Dickson had the greatest change in X
while the other genotypes had virtually no change. Most of
the changes in yield of the medium tiller producing
genotypes came from changes in the number of seeds per head.
The high tillering genotypes had no change in both X and Y.

Yield changes with cytokinin application were not
consistent with any trend, however, changes in X were
obtained with the high and low tillering genotypes while the
medium tillering genotypes had virtually no changes.
60-215-6 decreased its tillering and increase its head size.

Nith IAA application, there was an initial increase
in yield changes to a maximum (with 68-103-8) followed by
reductions with increasing tillering. 60-215-6, the highest
tillering genotype used in the study, had the greatest yield
reduction. The yield changes were due to changes in Y,
primarily. The changes induced with IAA application are
similar to that induced with CCC with a slight difference.
The major change in the yield were due to changes in Y,
however with auxin application, slight changes in X were
obtained in addition.

The length and width of the meristem were used as

independent variables in multiple regression analyses

120

involving X, Y, 2, XY, YZ and N as dependent variable for
each of the treatments. None of the predictions involving
the gibberellin application were significant. This 15

due to the lack of change in the width measurements in
comparison with the other treatments. Nidth has been shown
to be a more important component in the prediction of the
yield components.

One of the objectives of the study was to change the
number of seeds per head keeping the level of number of
heads per unit area constant. In addition to changing Y,
changes in X were observed in the data presented. These
changes were found to be negatively correlated, however,
the magnitude of the relationship is relaxed with the CCC
and IAA application.

The above observation raises the question whether the
significant negative correlation between X and Y is due to
the number of tillers associated with the system or X
causes such a relationship to arise. As it is generally
known, there is an initial well-developed shoot with 3 or 4
leaf initials and an apical dome enclosed within the
coleoptile of a mature seed. As the plant grows, a
succession of primordia are initiated, which later differ-
entiate an elongated internode and/or a tiller bud. The

tillers and main shoot later differentiate floral parts.

121

This led to the proposition of the sequential developmental
processes by Grafius (1969). Technically, one sees no flaw
with this reasoning, however, we have to caution ourselves
and critically examine the relationship.

Consider the flow chart shown in Figure 29. The
primary meristem differentiates a tiller bud which grows
into a tiller and later differentiates floral parts. The
question to ask at this point is when is the head size
(number of florets per head) really established? Two obvious
answers arise from the diagram.

1. At the point of differentiation of the tiller
meristem.

2. At a later developmental stage of growth.

The later answer is bound to be the dominant one because
signs of differentiation are depicted. The first answer
is favored for the following reasons:

1. Developmental ontogeny has both a space and time
reference and events arising from a common primary origin
can be removed in time and thus experience some relaxation
of the correlation as shown by Fowler and Rasmusson (1969).
The correlation between the area of leaves on the same culm
of barley diminished as the distance between leaves (both
in space and time of origin) increased. Attempts would

have been futile to select for different leaf sizes on the

122

Primary A tiller meri-
Meristem stem different-
iated from the

primary meristem Primary and
tiller meristem
differentiating maturation
floral parts of plant
which eventually
form seeds

 

 

 

Figure 29. Diagrammatic representation of sequence of events
during the growth and development of the apical
meristems.

same culm. This is so because they arise from the same
meristem and changing one factor leads to a similar change
in another factor in close proximity because of the high
positive correlation. Grafius (1978) proposed that
'Plasticity is inversely proportional to ontogenetic
proximity'. Events arising from the same meristem are
harder to manipulate than those separated in space and time.
One can accept not coming across any literature on the

relationship between individual tillers and their respective

123

head sizes but one can reason through the leaf area work
quoted above that a similar situation is bound to occur
between X and Y. The primary and tiller meristems differ-
entiate the floral parts and as a result changing either

X or Y will tend to change Y or X in the reverse manner.
Breeders are unable to use X or V as a breeding objective
since they arise from the same meristem and any determina-
tion to increase one yield component (either X or Y) will
be accompanied by a decrease in the other component (i.e.
Y or X). On the other hand, seed weight does not present
similar problems since developmentally, it is farther
removed in both space and time from the common origin of
the primary apical meristem; additionally, they arise from
different meristems.

2. The hormonal applications were done with the
objective of changing Y, however, changes in X were also
observed. The observed changes were negatively correlated.
Z was not affected by the changes in X and Y.

3. The influence of the length, width and relative
growth rate of the meristem on X and Y are almost the same
showing that the number of heads per unit area and the
number of seeds per head have a common origin at the
meristematic level.

4. The magnitude of the paths between X and Y with Z

are almost the same but with different signs. This is

124

interpretted as the result of the initial significant nega-
tive relationship established between X and Y.

5. This further reinforces my prediction (Nhyte, 1979)
that the larger reproductive apex at the initiation of
elongation could be traced back to a larger vegetative apex
and that any gain in the number of seeds per head may be
established in the first developmental phase with all the
physiological and practical implications.

6. The relative efficiencies for the production of
X and Y were similar while that for Z production varied as
will be shown in Chapter 2.

In no way is competition for nutrients discounted. It
strengthens the relationship between X and Y.

On the basis of the foregoing findings and reasonings,
we may reconceptualize, diagrammatically, the sequential
developmental process of yield components as shown in
Figure 30.

There has been some assertion that the sequence of
yield component formation overlap (Tai, 1975) but the
magnitude of this overlapping is traditionally thought to
be minimal.

A 95% confidence belt around the regression of Y on X
for the control block (Fig. 31) show that the hormonal

application induced certain changes in some standard

125

"1
I
I
7I
r.
'1

“Mi"
wfiflw

~
\I\
| 0......' ’ ' N,
O. ’
J”-

"mum
"Imam
mm,“

I

 

In another form:

X
0......
Y

or
XY—' Z—VH

Figure 30. Diagrammatic representation of the sequential

development of the yield components and yield
of barley.

varieties rendering them outliers. An 'outlier' is defined
as any genotype which does not fall within the 95% confid-
ence belt (variance of a predicted value) established around
the regression of Y on X for the control block (Fig. A2).

No single hormone induced changes great enough to render

all the genotypes as outliers. The yield component changes
varied among the genotypes which, by the definition, could

be described as outliers. Drastic positive changes in X

126

were attained in 8130 with GA3 and CCC treatments with
virtually no change in Y. 68-l03-8 (GA3 and CCC treatments),
68-l03-18(CCC treatment) changed basically their number of
seeds per unit area while 8130 and 68-104-19 (IAA treatment)
changed both their X and Y before attaining the outlier

status.

SUMMARY AND CONCLUSION

The present investigation was undertaken to assess the
involvement of the promotive hormones in the development of
the primary shoot apical meristem in relation with the
production of the yield components and yield of barley.

The results showed that the outliers reached the transi-
tion stage four days in advance of the standard varieties,
however, the elongation of their meristems were delayed by
four days. The sizes of the meristems of the outliers were
larger than those for the standard genotypes. A switch in
the rate of development was observed in the growth of
meristems. Genotypes with higher rate of growth at the
transition stage had a lower growth rate after the reprod-
uction stage of development.

The coefficient of determination (R2 = .5959, .7l56 and
.79l2 for X, Y and Y2, respectively) indicate that the
variance in the dependent variables X, Y and YZ, can be
accounted for in large part by variation in the meristematic
measurements used as the independent variables in the
multiple regression analyses performed.

Thus, it was thought that a genotype with a low growth
rate at the reproduction stage of development produced a
low number of heads per unit area. Further relaxation of

the growth rate encouraged the formation of organs, i.e.,

127

128

meristems with larger width. Since width determined the
length that the organ assumed, a larger sized meristem is
obtained. A higher number whorls of floral initials are
borne on the meristem resulting in the production of a
higher number of seeds per head and a bigger head size.

The determination of the endogenous promotive hormone
levels and activities was not successful due to the
inadequate amount of sample available and the insensitivity
of the method applied.

Application of GA3, CCC, Kinetin and IAA induced changes
in both the length and width of the primary apical meristems
at the reproduction stage of development. The lengths were
changed to a larger extent than the width, however, the
magnitude of these changes depended on the type of hormone
applied and the genotype in question.

Have patterns were shown by the histograms of percent-
age change over control for lengths of the primary shoot
apical meristem vrs. genotypes arranged in the order of in-
creasing number of heads per unit area in all treatments.
This indicated that the promotive hormones are in equilibrium
with each other for normal growth and development of the
plant. The involvement of the inhibitory hormones was not
discounted. The wave pattern was observed more in the

changes in lengths of the apical meristems. This was

129

expected because at the stage of development when the
applications were carried out, the maximum width of the
genotype was about attained leaving only the length to be
subjected to greater changes caused by the disturbances
created in the internal equilibrium.

Yield changes were observed as a result of application
of the promotive hormones. These changes were due to
changes in the sizes of the meristems. Even though the
meristematic changes followed some wave pattern, not all
the changes in the yield components and yield followed the
same trend. The type of change also depended on the hormone
applied and the genotype under consideration. Changes in
X or Y or both were induced in some standard genotypes to
render them 'outliers'. No single hormone induced changes
great enough to render all the genotypes as outliers.

Nidth was found to be the most important of the meristematic
measurements in the prediction of the yield components and
yield.

An interesting finding was the significant negative
correlations between the changes in the number of heads per
unit area and the changes in the number of seeds per head
in all the treatments. The intent here was to increase the
number of seeds per head keeping the number of heads per

unit area constant. Removing the outliers tended to show

130

an effect of relaxation between changes in X and Y in the
CCC and IAA treatments.

All the above findings prompted the raising of the
question as to whether the significant negative correlation
between X and Y is due to the number of fertile tillers
associated with the system of growth and development or X
causes such a relationship to arise as it is traditionally
observed. Evidence in favor of the negative correlation
between X and Y arising from the number of fertile tillers
associated with the system of growth and development was
advanced. This led to the proposition of reconceptualiza-
tion of the sequential developmental process of the yield

components and yield.

APPENDICES

l3l

 

 

 

mm.~ om.o Pw.m me.e_ eo.m_ em.¢_ mm.ep mm.m_ Awo.uuv om;
om._ Pw.e mm.w me.N_ we.ew om.o. cm.pw —w._w Amo.uuv om;
m~.o- om.m- mm.m ww.m we.m eo.__ mo.m m_.mw m—-mo_-mo
mm.m- mm.m- mm. FN.N mm._ wk.“ mw.w mm.m mumo_-mo
oo.¢- mm.p- mw.m mm.m em.m- mm.m me.~ om.e m_ueow-me
mm.m- mp.w- wo.m om.m m¢.N_ mw.m - om._ mm.e m-eopumm
ow.e- mm._- N_.m ow._ me.m- mm.m wm.m eo.e swamop-me
op.m- om.w- mm.m. no. w—.m- me.—— wo.m- ew.c mnmowumo
mm.m- mm.m- mm.c eo.~ m_.e mw.¢ um.w mm.m mmxm<e
mm._- mw.m- mo.e me.m wo.mm me.e - oo.N_ oo.m_ zomquo
Np.o- NN.m- om.o- om.m em.m em.__- em.m om.m - o-m_m-oe
—m.N- mm.m- em. me.P ep.ep PP.N om.m_ _N.ew omwm
mm.e- we.o- oe.~ mw.o_ wo.e- oe.mw mp.“ em.o_ mumemx
_e.~- mm._ mm._ em.¢ mm.e mm.m _o.m _m.o— mmuzom
e:e wze Ne >< xe N>< >x< 3e mma>wozmu

.uceseewe>ee we emepm

cewuwmcegu one we :ewueewpeee :__Feceeewm ea e:e xmwcee we Aozev mcweee;

e:e szev unmweg .ANev “sewez eeem .A><V ewe; Lee meeem we geese:

.Axev mace awe: Lee meme; we geese: .AN><V e~wm new; .A>xev mace awe:
Lee memem we geese: .Azev ewew» Lew Pecucee cm>e emcege emeuceegme com: .—< epeew

132

 

 

em.~ om.e .N.m Ne.ep so.m_ em.¢. ma.¢_ mm.m_ A_e.uav ems
ea._ _~.¢ mm.“ me.~_ N¢.¢P e¢.o_ em.__ _e.F_ Ame.uuv emu
o mm.N- om..- mm._m mo. - ee.m_ em.o~ _e.m_ e.-mo_-ee
“a. o mw.~- we.m we. me.~ so.e mm.m e-mo_-ee
oo._ o me.m- ea.e mm.m- om.~ _N.m Na. - m_-¢o_-me
mm. - om._- ¢N.P N¢.P we. - wa.~ Ne._ __.m m-¢o_-me
o mN.m- ee. o~.e_ mm.m- e_.w. we.m mo.o_ N_-mo_-ee
¢O.N- mw.m- ee.m Ne.m _m.m- oo.~_ Pm.N mm.m m-mo.-me
Rm. we._- om.m- mm.e_ mw.m mo.¢ P¢.¢_ mu.“ mu¥¢<e
mm... e we.“ ee.¢- um.e me.~ mm._ wm.a zemxeeo
~_.e e¢.m- mm. - me.m NN.P- em.e e5.“ N¢.~ e-m_~-oe
mm. -.m- e_.N- em._ eo.m_ em. - o¢.e~ ee.~. om_m
0 mm..- ow. .m.m_ me. - ¢~.¢. F¢.¢_ _m.m_ m-memx
mm. - oo.~ m~.N- um.e my.“ m_.¢ mm.e_ NR.__ mmmzom
are wza N4 >< x4 N>< >x< ,2 ma>wezwe

 

.uemEee_e>me we macaw

:ewpwmcegu e:e we :ewpeew_eee weeeexe e“ e:e xmwcee we Aozev mcweem;
e:e szev ucmwec .ANev «sawez ewwm .A><v wee; Lee mewem we geese:
.Axev emce “we: see meme: we geese: .AN><V mem ewe; .A>xev eege «we:

gee memem we geese: .A3Vv e_ewx Lew Pegucee Le>e emcese emeeceecee cam: .~< eweew

133

 

 

em.N om.e _N.m Ne.e_ eo.mp em.¢_ mm.¢_ mm.m_ A_e.uav awe
em.w _~.¢ em.“ me.~_ w¢.¢_ om.ew em.__ .m.__ Amo.uav ems
we._ me. - _N. - “P.m ow.“ mm.“ em.m_ m_.m_ e_-mo~-me

o e¢.N- mm.¢ em._ we. em.m No. - om.¢ e-mop-me
oo._ Ne. em. we. - mq.m No. ~o.m No.m m.-¢e.-ee
am. - om..- we.m mo.e- _¢.m NN.N- N¢.N m¢.e m-vo_-ee
mm. o om.¢- FN.N as. N..N- me. em.m- ~_-me_-ee
wo.~- we..- Ne.m mN.e- mm.e a¢.m- mm. - ep.N m-mo_-ee
mm. - m_.m- eN.e a“. we. eN.N om. “5.0 mm¥m<e
Ne._- me. - om.“ mo.e- mm.o_ me._ .e.¢ NN.__ zemxeeo
~_.e «N._ we. - se.m em.__- sm.m mm.N- om.m- e-mpm-ee
_m.~ mm.~- mm.¢ NN.m- om.mp em._- mm.“ we.__ empm
mo._- we. mm.P __.e_ Fo.ep- mo.N_ Ne. em.~ m-memx

o o _e. e¢.¢- e¢.ep mm.m- _m.m em.e mmmzom
are wzq Na >< x< N>< >x< 3e

 

.uceseewm>me we emeum

ce_uwm:egu e:e we :ewpeewweee cwcwxeaae e» e:e xmwgee we onev mcweee;
e:e Aw;ev pnmwmg .ANev ugmwmz eemm .A><v new; Lee meeem we geese:

.Axev mace “we: see meme; we geese: .AN>eV erm new; .A>xev ease awe:
gee memem we geese: .Azev ewww» Lew pecucee Lm>e emcese emegceugme new:

.m< mpnew

134

 

 

 

mm.~ o~.o Fw.m No.0, eo.m_ em.¢w mm.e_ mm.mw Apo.uuv om;
mm.w ww.e mm.w me.N— we.ew om.ow om.ww wm.w_ Amo.uav om;
eo._- mo. wo.~ mw.ww m_.~ o~.m— em.ww ww.e— mw-mow-ww
we. oe.m mo.w - wm.m m_.m mm.¢ ¢_.~ mo._ wlmowumo
o ow.m om.w . ma.m me.¢ ww.e ew.m mo. mwueowumo
mm. a c we. ee.m- wm.w mo.w om.e- _¢.¢ - m-eow-me
om.w mo. mw.m - mm.ow me.w me.o oo.~ ow.w - ww-mo_-me
No._- o om. - me.m m~.m m¢.m ow.m NN.m mumowumm
.m.~ o Fm. - oo.e mw._ _o.¢ me.m No.w mmxm<4
mm.w- me. me. we.m- mm._ ee.e em.m- me.~ - zomquo
~_.o ew.p ow.~_- Nw.~- mw.o mm.mw- pm.m- mw.ow- mum—N-oo
em.w mm._ No.o - Ne.ww me.m om.¢ mw.mw mm.m omwm
mm.w- mm.~ Nw.w - Nw.ww w¢.__- m_.ow om.m Pm._ . mumomx
o mm.w mw.m - m~.e- me. mo.w- ow.¢- mo.w . mmmzom
e:e wze Ne >< x4 N>< >x< 3e

.ucmEeewm>me we

mmeum :ewuwmcecu one we :ewpeewweem :wxze ea e:e xewcee we Ao;ev mcweem;

e:e szev ugmwe; .ANev uzmwmz eemm .Awev ewe; Lee mewem we geese:

.Axev eege “we: see meme; we geese: .ANwev mem new; .waev mega awe:
Lee meemm we geese: .Azev e_ewa cow wecucee ce>e mmeeze mmeuceecee cemz .e< eweew

135

 

«mm. NNo. —¢o. oww. opp.¢o_

 

mme. mum. «No. mom. wmo._
mmm. mmm. mmo. Nmm. mmm.em
mwm. Nwm.- omm. mmm.- wwm.m- o3
mmm. omo. «mm. mwo. wme.w
one. mom. mom. mum. mmw.
mme. emm. wee. mmm. owo.e
wmm. mmm.- mmm. mom.- www.mu A
meuewee e:ewewwweeu we>ee muzmwez ucewewwwmeo eweewce>
mm :ewuewecgeo e:eewwwcmwm epmm :ewmmegmem wewpgee
_eweeea

 

.Axeewm Pecueeuv >Pe>wueeemec .3 e:e N> .> .x we Loewe ecu cw

ewe mewumwueum mew .meweewge> uceeceeeecw ecu me e:eseewe>ee we eueem

:ewHeaeeweee ecu cw xewcee masseumwgee weewee xgeswce esp we Aozv spew:

Easwxes e:e Rev gumcew Ezawxee one so eweewge> acmeceeme e we seem

.Azev ewewx e:e AN><V erm new: .A><v new; Lee meeem we geese: .Axev
eece “we: Lee meee; we geese: ecu cew mewumwueum :ewmmecmeg e—ewuwaz .m< eweew

136

 

mmw. ewm. moo. mev.— omw.oee

 

 

mwm. wmo. wmo. «we.w omm.~
mmm. cow. woo. wwm.w mwo.mmm
mom. mmo. mwo. wmo.w www.me o3
omm. mmo.- wee. wmm.- mom.wm -
mmm. wmm.- omm. wee.- men. .
mam. oom.- www. weo.- owe.wmw-
com. mom.- wem. emm.- wme.m - 4
mmum—me acmwewwwmeu wm>mm mucowmz ucmwewwwmeu eweewee>
Na cewuewmeeeu uceewwwcmwm eumm :ewmmmemmm —ewpeem
wewpeea
.Axeewe wmeeeaev xwm>wpememme .3 e:e N> .>x
.> we emeee me» cw mee mewumwuepm mew .mmweewee> acmeememecw me» me
ucmseewm>ee we epeum :ewpeeeeeeme e:e cw xmweee we Emumweme weewee xeeswee men
we spew; Easwxee e:e Aev zumemw e:ewxes me» so mweewee> uemecmeme e we
seem .sz ewmwm e:e Ava mem eem: .wav emee awe: eme memmm we emesac
.va eem: eme memmm we eme53: me» eew mewumwueum :ewmmmemme mwewuwez .e< mweew

137

 

wme. «mm. mew. 0mm. mmw.wow

 

ovm. owe. Nee. wmm. mmm.mmm
cum. Fae. mmw. mmm. www.mm o3
Fmv. mom. mmm. cow. vom.m¢
mmo. Noe. wa. mom. wmm.mm
mmv. mmw. wee. wm—. ¢w~.¢ 3
mmeeeee e:ewewwweoe _m>me mesmwez ecmwuwwwmoe meeewee>
Na :ewuewmeeeu e:eewwwcmwm eumm :ewmmmeomm wewueee
eeweeee

 

.Axeewe cwewxepxev 3wm>wuememme .3 e:e >x.»

we emeee mew cw mee mewumwueum mew .mmweewee> e:eeememecw mew me gems

-eewm>me we meeum :ewuezeeeeme mcu cw xmweee we Emumwems weewee weeswee

men we Aezv spew: e:ewxes e:e Amy camcmw Eeswxes mew ce mweewee>

ucmeememe e we :eem .sz ewmwx e:e wav emee ewe: eme memmm we emee::
.va eem; eme memmm we eme52: mew eew mewamwueum :ewmmmemme mwewuwez .w< mweew

138

 

Nmm. wmm. omm. mom. www.mmw

 

nee. Nee. one. eme. eNN.e
Nee. New. nee. New. m_e.ee_
mee. oee. ewe. Nee. eem.em
eee. Nem.- eew. eme.- mee.m - a:
ewe. mom. ewe. mee. Neo._e
eee. eee. eee. oem. eew.
Nee. eee. eee. eee. mem._ee
eee. mee. eee. ewe. emm._e
Mme. ee~.- emm. eme.- ewe.m - e
meeeeee ecewewwweoe ee>ee meeewez eeeweweweee meeeweee
m cewuewmeeeu aceewwwcmwm eumm :eweememmm wewueee
N eeweeee

 

.Axuewe cwxzev >wm>wpmmeeme

.3 e:e .N> .wx .> .x we emeee cw mee eewuewuepe mew .emweewee>

uceecmeeecw me» me ucmseewm>me we mueue eewuezeeeeme mew cw xmweee we

Emuewems weewee xeeewee me» we Ao3v spew: Eeswxee e:e 53v zumcmw Eeswxee

me“ so .mweewee> “emeememe e we seem .A3v ewmwz e:e Ava m~wm eew;

.wav emee “we: eme memmm we eme53e .va eem: ewe eemmm we eme23: .Axv
emee ewe: eme eeeme we emeE== e:e eew eewaeweeue cewemmemme mwewpwez .w< mweew

Figure Al

139

-o899

 

Developmental allometry showing the
influence of the primary apical meristem of
barley in the reproduction state of development
on the yield components: number of heads
per unit area (X), number of seeds per

head (Y), seed weight (Z); width (ND),
length (L) and relative growth rate (R) of
the primary apical meristem. Single
arrowed lines denote path coefficients and
double arrowed lines denote correlation
coefficients.

Number of seeds per head

70

50
30

15
Figure A2.

140

      

93.69 - 1.927x

O
.<
ll

O.9246**

20 25 30

Number of heads per unit area

Regression of the number of seeds per head
on the number of heads per unit area
(control block) with its 95% confidence
belt (variance of a predicted value).

LIST OF REFERENCES

LIST OF REFERENCES

Abbe, E.C. and 8.0. Phinney. 1951. The growth of the shoot
apex in maize: external features. Am. J. Bot., 38:
737-744.

Abbe, E.C., L.F. Randolph, and J. Einset. 1941. The
developmental relationship between shoot apex and
growth pattern of leaf blade in diploid maize. Am. J.
Bot., 28: 778-782.

Adams, M.N. 1967. Basis of yield component compensation in
crop plants with special reference to field bean,
Phaseolus vulgaris. Crop Sci., 7: 505-510.

 

Adams, M.w. 1975. Plant architecture and yield in grain
legumes. In: Report of the TAC working group on the
biology of yield of grain legumes. TAC Secretariat,
FAO, Rome.

Aitken, Y. 1967. Leaf primordia formation in some agri-
cultural species. J. Anst. Inst. Agr. Sci. 33: 212-
214.

Amer, F.A. and W.T. Milthorpe. 1958. Drought resistance
in Pelargonium zonale. Ann. Bot., Lond., 22: 369-379.

Aspinall, D. 1961. The control of tillering in the barley
plant. I. The pattern of tillering and its relation
to nutrient supply. Aust. J. Biol. Sci., 14: 493-505.

Aspinall, D. 1963. The control of tillering in the barley
plant. II. The control of tiller-bud growth during
ear development. Aust. J. Biol. Sci. 16: 285—304.

Aspinall, D. 1966. Effects of daylength and light intensity
on growth of plant. IV. Genetically controlled varia-

tion in response to photoperiod. Aust. J. Biol. Sci.,
19: 517- 534.

Aspinall, D., P.8. Nicholls, and L.H. May. 1964. The
effect of soil moisture stress on the growth of barley.
I. Vegetative development and grain yield. Aust. J.
Agric. Res., 15: 729-745.

Aspinall, D. and L.G. Paleg. 1963. Effects of daylength
and light intensity on growth of barley. 1. Growth

141

142

and development of apex with fluorescent light source.
Bot. Gaz., 124: 429-437.

Asprinall, D. and L.G. Paleg. 1964. Effects of daylength
and light intensity on growth of barley. II. Vegeta-
tive deve10pment. Aust. J. Biol. Sci., 17: 807-822.

Atkiri, R.K. and G.E. Barton. 1971. Role of endogenous
growth substances in the effect of soil temperature
on shoot growth in the Graminae. A Rep. Grassland
Research Institute. pp. 51-52.

Balciev, B. and A. Lang. 1965. Control of flower formation
by growth retardants and gibberellin in Samolus parvi-
florus, a long day plant. Am. J. Bot., 52: 408-417.

 

BarWDat, I. and C. Ochesanu. 1963. Effect of gibberellin

on the variations of the growing point in winter wheat.
Naturwissenschaften, 50: 159.

Barwnard, C. 1964. Form and structure. In: Grasses and

grasslands. ed. C. Barnard, 269 pp.__Macmillan,
London.

Berwdahl, J.D., D.C. Rasmusson, and D.N. Moss. 1972. Effect
of leaf area on photosynthetic rate, light penetration
and grain yield in barley. Crop Sci., 12: 177-180.

Bann, A. 1977. Basis of heterosis in the Differentiating
Sorghum Panicle. Crop Sci., 17: 880-882.

&3khari, U.G. and v.8. Youngner. 1971. Effects of CCC on

tillering and flowering of uniculm barley. Crop Sci.,
11: 711-713.

Bonnett, O.T. 1935. The deve10pment of the barley spike.
J. Agr. Res., 451-457.

Bonnett, O.T. 1936. The development of the wheat spike.
J. Agr. Res., 53: 445-451.

Bonnett, O.T. 1937. The development of the oat spike. J.
Agr. Res., 54: 927-931.

Bonnett, O.T. 1964. Morphogenesis in the organism and the

expression of the complex traits. Crop Sci. 4: 500-
503.

143

Bonnett, O.T. 1966. Inflorescences of maize, wheat, rye,
barley, and oats: their initiation and development.
Univ. of Illinois College of Agric., Agric. Expt.
Station Bull. 721.

Branscomb, E.w. and R.N. Stuart. 1968. Induction lag as a
function of induction level. Biochem. Biophys. Res.
Com., 32: 731-738.

Bremer-Reinders, D.E. 1958. The early stages of develop-
ment in the rye spike. Acta. Bot. neer1., 7: 223-32.

Brian, P.w. 1961. The gibberellins. A new group of plant
hormones. Sci. Progr., 49: 1-16.

Brown, 0.3. 1953. The effects of irrigation on flower bud
development and fruiting in apricot. Proc. Amer. Soc.
Hort. Sci., 61: 119-124.

Cannell, R.Q. 1969. The tillering pattern in barley
varieties. I. Production, survival and contribution
to yield by component tillers. J. Agr. Sci. Camb.
72: 405-422.

Clifford, P.E., C. Marshall, and G.R. Sugar. 1973. The
reprical transfer of radiocarbon between a developing
tiller and its parent shoot in vegetative plants of
Lolium multiflorum Lam. Ann. Bot., 37: 777-785.

 

Donald, C.M. 1968. The breeding of crop ideotypes.
Euphytica 17, 385-403.

Evans, L.T. 1964. Inflorescence initiation in Lolium
temulentum L. V. The role of auxins and gibberellins.
Aust. J. Biol. Sci., 17: 10-23.

 

Evans, L.T. 1969. Inflorescence initiation in Lolium

temulentum L. XIII. The role of gibberellin. Aust.
J. Biol. Sci., 22: 773-786.

 

Evans, L.T. 1971. Flower induction and florigen concept.
Ann. Rev. P1. Physiol., 22: 365-394.

Evans, L.T. and R.L. Dustone. 1970. Some physiological
aspects of evolution in wheat. Aust. J. Biol. Sci.,
23: 725-741.

144

Evans, L.T., I.F. Nardlaw, and C.N. Williams. 1964.
Environmental control of growth. In: Grasses and
Grasslands. ed. C. Barnard, pp. 102-25, London:
Macmillan.

Felippe, G.M. and J.E. Dale. 1973. Effects of shading the
first leaf of barley plants on growth and carbon
nutrition of the stem apex. Ann. Bot., 37: 45-56.

Fisher, J.E. 1973. Developmental morphology of the
inflorescence and hexaploid wheat cultivars with and
without the cultivar Norin 10 in their ancestry.
Can. J. Plant Sci., 53: 7-15.

Fosket, D.E. and K.C. Short. 1973. The role of cytokinin
in the regulation of growth, DNA synthesis and cell

proliferation in cultured soybean tissues (Glycine max
var. Biloxi). Physiol. Plant., 28: 14-23.

Fowler, C.N. and D.C. Rasmusson. 1969. Leaf area relation-
ships and inheritance in barley. Crop Sci. 9: 729-731.

Friend, D.J.C. 1965. Tillering and leaf production in
wheat as affected by temperature and light intensity.
Can. J. Bot. 43: 1063-1076.

Friend, D.J.C., J.E. Fisher, and V.A. Helson. 1963. The
effect of light intensity and temperature on floral
initiation and inflorescence development of Marquis
wheat. Can. J. Bot., 41: 1663-1674.

Friend, D.J.C., V.A. Helson, and J.E. Fisher. 1962. The
rate of dry matter accumulation in Marquis wheat as
affected by temperature and light intensity. Can. J.
Bot., 40: 939-955.

Gates, C.T. 1955a. The response of young tomato plant to
a brief period of water shortage. I. The whole plant
and its principal parts. Aust. J. Biol. Sci., 8:
196—214.

Gates, C.T. 1955b. The response of the young tomato plant
to a brief period of water shortage. II. The individ-
ual leaves. Aust. J. Biol. Sci., 8: 215-230.

Gates, C.T. 1957. The response of the young tomato plant
to a brief period of water shortage. III. Drift in
nitrogen and phosphorus. Aust. J. Biol. Sci., 10:
125-146.

145

Grafius, J.E. 1956. Components of yield in oats. Geometri-
cal interpretation. Agron. J., 48: 419-423.

Grafius, J.E. 1964. A geometry of plant breeding. Crop
Sci., 4: 241-246.

Grafius, J.E. 1969. Stress: a necessary ingredient of
genotype by environment interaction. In: Barley
Genetics II. Proc. 2nd Intern. Barley Genet. Symp.,
Pullman, Washington St. Univ. Press, Pullman. pp. 346-
355.

Grafius, J.E. 1978. Multiple characters and correlated
response. Crop Sci., 18: 931-934. '

Grafius, J.E. and R.L. Thomas. 1971. The case for indirect
genetic control of sequential traits and the strategy
for deployment of environmental resources by the plant.
Heredity 27: 433-442.

Grafius, J.E., R.L. Thomas, and J. Barnard. 1976. The
effect of parental component complementation on yield
and components of yield in barley, Hordeum vulgare L.
Crop Sci., 16: 673-677.

 

Griffiths, D.J. 1961. The influence of different daylengths
on ear emergence and seed setting in oats. J. Agric.
Sci., 57: 279-285.

Guitard, A.A. 1960. The influence of variety, temperature
and stage of growth on the response of spring barley
to photoperid. Can. J. Plant Sci., 40: 65-80.

Hager, A., H. Menzel, and A. Krauss. 1971. Versuche und
hypothese zur primarwirkung des auxins beim streckungs-
wachstum. Plant, 100: 47-75.

Hamid, Z.A. 1976. Developmental allometry and its implica-
tion to grain yield in barley (Hordeum vulgare L. Emend.
Lam) Ph.D. Thesis, M.S.U., 54 pages.

 

Hamid, Z.A. and J.E. Grafius. 1978. Developmental allometry

and its implication to grain yield in barley. Crop Sci.
18: 83-86.

Harada, H. and J.P. Nitsch. 1959. Flower induction in

Japanese Chrysanthemums with gibberellic acid. Science,
129: 777-778.

146

Higgins, T.V., J.A. Zwar, and J.V. Jacobson. 1976.
Gibberellic acid enhances the level of translatable
mRNA for a-amylase in barley aleurone layer. Nature,
260: 166-169.

Ho, D.T. and J.E. Varner. 1974. Hormonal control of
messenger ribonucleic acid metabolism in barley aleurone
layers. Proc. Nat. Acad. Sci. U.S. 71: 4783-4786.

chen, K. and R.H. Andrew. 1959. Performance of corn hybrid
with various ratios of flint-dent germ plasm. Agron. J.
51: 451-454.

Htxrd, R.G. and D.N. Purvis. 1964. The effect of gibberellic
acid on the flowering of spring and winter rye. Ann.
Bot. (London)., 28: 137-151.

Irigle, J., J.L. Key, and R.E. Holm. 1965. Demonstration
and characterization of a DNA-like RNA in excised
plant tissue. J. Mol. Biol., 11: 730-746.

Jax:obs, M. and P.M. Ray. 1976. Rapid auxin-induced decrease
in free space pH and its relationship to auxin-induced
growth in maize and pea. Plant Physiol., 58: 203-209.

Jaczqmard, A. 1968. Early effects of gibberellic acid on
mitotic activity and DNA synthesis in the apical bud
of Rudbeckia bicolor. Physiol. Veg., 6: 409-416.

 

Jories, H.G. and E.J.M. Kirby. 1977. Effects of manipulation
of number of tillers and water supply on grain yield
in barley. J. Agric. Sci., Camb., 88: 391-397.

JCHies, R.L. and I.D.J. Phillips. 1966. Organs of gibberel-
lin synthesis in light-grown sunflower plants. Plant
Physiol., 41: 1381-1386.

Kende, H. 1955. Kinetin-like factors in the root exudate
of sunflowers. Proc. Nat. Acad. Sci. U.S. 53.
1302-1307.

KETlde, H. 1971. The cytokinins. Int. Rev. Cytol., 31: 301-
338.

K93’. J.L. and J.C. Shannon. 1964. Enhancement by auxin of
ribonucleic acid synthesis in excised soybean hypocotyl
tissue. Plant Physiol., 39: 360-364.

147

Khan, M.A. and S. Tsunoda. 1970. Evolutionary trends in
leaf photosynthesis and related leaf characters among
cultivated wheat species and its wild relatives. Jap.
J. Breeding. 20: 133-140.

Kirby, E.J.M. 1967. The effect of plant density upon the
growth and yield of barley. J. Agric. Sci., Camb.,
68: 317-324.

l<irby, E.J.M. 1969. The effects of daylength upon the
development and growth of wheat, barley and oats.
Fld. Crop Abstr., 22: 1-7.

l<irby, E.J.M. 1971. Abnormalities induced in barley ears
by gibberellic acid. J. Exp. Bot., 22: 411-419.

l<irby, E.J.M. 1973. The control of leaf and ear size in
barley. J. Expt. Bot., 24: 567-578.

l<irby, E.J.M. 1974a. Ear development in spring wheat. J.
Agr. Sci., Camb., 82: 437-447.

Kir~by, E.J.M. 1974b. Effect of temperature on ear abnorm-
alities in uniculm barley. J. Exp. Bot., 24: 935-947.

Kir‘by, E.J.M. 1977. The growth of the shoot apex and the
apical dome of barley during ear initiation. Ann.
Bot., 41: 1297-1308.

Kir~by, E.J.M. and D.G. Paris. 1970. Plant population
induced growth correlations in the barley plant main
shoot and possible hormonal mechanisms. J. Exp. Bot.,
21: 787-798.

Kir~by, E.J.M. and D.G. Faris. 1972. The effect of plant
density on tiller growth and morphology in barley.
J. Agr. Sci., Camb., 78: 281-288.

K1Wr‘by, E.J.M. and H.G. Jones. 1977. The relations between
the main shoot and tillers in barley plants. J. Agr.
Sci., Camb., 88: 381-389.

K1'T‘by, E.J.M. and J.L. Rymer. 1974. Development of the
vascular system in the ear of barley. Ann. Bot., 38:
565-574.

Killler, D., H.R. Highkin, and 0.H. Caso. 1960. Effects of
gibberellic acid on stem apices of vernalized grasses.
Am. J. Bot., 47: 518-524.

148

Lang, A. 1960. Gibberellin-like substances in photo-induced
and vegetative Hyoscyamus plants. Planta, 498-504.

 

Lang, A. and J. Nitsan. 1967. Relations among cell growth,
DNA synthesis and gibberellic acid. Ann. N.Y. Acad.
Sci., 144: 180-190.

l_eafe, E.L. 1959. Spraying of barley with 2,4-Dichlorophen-
oxyacetic acid. Nature, 183: 621-622.

L.ee, Kit-Nah, R.C. Lommasson, and J.D. Eastin. 1974.
Developmental studies on the panicle initiation in
sorghum. Crop Sci., 14: 80-84.

Leaopold, A.C. 1949. The control of tillering in grasses
by auxin. Amer. J. Bot. 36: 437-440.

l.in, C.Y. and J.L. Key. 1967. Dissociation and reassembly
of polyribosomes in relation to protein synthesis in
the soybean root. J. Mol. Biol., 26: 237-247.

L111, P.8.N. and J.B. Loy. 1976. Action of gibberellic acid
on cell proliferation in the subapical shoot meristem
of watermelon seedlings. Am. J. Bot., 63: 100-104.

141xova, M. and A. Lux. 1964. Character of morphoses caused
in the barley (Hordeum distichum L.) spike by 2-methy1-
4-chlorophenoxycetil acid (MCPA). Biologia Pl., 6:
258-264.

 

Ma)’. L.H. and F.L. Milthorpe. 1962. Drought resistance of
crop plants. Field Crop Abstr., 15(3): 1-9.

Mllthorpe, F.L. 1950. Changes in the drought resistance
of wheat seedlings during germination. Ann. Bot.,
Lond., 14: 79-89.

Motir, H. 1962. Primary effects of light on growth. Ann.
Rev. Plant Physiol., 13: 465-488.

N1°<2holls, P.B. 1974a. The effect of daylength on the
development of the barley inflorescence and the

endogenous gibberellin concentration. R. Soc. N.Z.
Bull., 12: 305-309.

N1°<:holls, P.B. 1974b. Interrelationship between meristem-
atic regions of developing inflorescences of four
cereal species. Ann. Bot. (London), 38: 827-837.

149

Nicholls, P.B. 1978. Response of barley shoot apices to
application of gibberellic acid and abscisic acid:
Initial response pattern. Aust. J. P1. Physiol., 5:
311-319.

Nicholls, P.B. and L.H. May. 1963. Studies on the growth
of the barley apex. I. Interrelationship between
primordia formation, apex length and spikelet
development. Aust. J. Biol. Sci., 16: 561-71.

Nicholls, P.B. and L.H. May. 1964. Studies on the growth
of the barley apex. II. On the initiation of internode
elongation in the inflorescence. Aust. J. Biol. Sci.,
17: 619-630. '

Nissl, D. and M.H. Zenk. 1969. Evidence against induction
of protein synthesis during auxin-induced initial
elongation of Avena coleoptiles. Planta, 89: 323-341.

Nooden, L.D. 1968. Studies on the role of RNA synthesis
in auxin induction of cell enlargement. Plant Physiol.
43: 140-150.

Nooden, L.D. and K.V. Thimann. 1963. Evidence for a
requirement for protein synthesis for auxin-induced
cell enlargement. Proc. Nat. Acad. Sci. U.S., 50:
194-200.

Novikov, V.P. 1952. The effect of a deficiency of water
in the soil at different stages of development in oats.
C.R. Acad. U.R.S.S., 82: 641-643.

Novikov, V.P. 1954. Development of the oat panicle under
different conditions of water supply. Bot. U. Zh.
39: 17-20.

Osborne, D.J. 1972. In crop processes in controlled
environments. Ed. A.R. Rees, K.E. Cockshull, D.N. Hand
and R.G. Hurd. Academic Press, London, pp. 251-64.

Paleg, L.G. and D. Aspinall. 1958. Inhibition of the
development of the barley spike by gibberellic acid.
Nature (London). 181: 1743-1744.

Paleg, L.G. and D. Aspinall. 1964. Effects of daylength
and light intensity on growth of barley. II. Influence

of incandescent light on apical development. Bot. Gaz.,
125: 149-155.

150

Penny, P. and A.N. Galston. 1966. The kinetics of inhibition
of auxin-induced growth in green pea stem segments by

actinomycin D and other substances. Amer. J. Bot.,
53: 1-7.

Purvis, D.N. 1960. Effect of gibberellin on flower initia-
tion and stem extension in Petkus winter rye. Nature,
(London), 185: 479.

Quinby, J.R. 1970. Leaf and panicle size of sorghum parents
and hybrids. Crop Sci. 10: 251-254.

Radley, M. 1970. Comparison of endogenous gibberellins
and response to applied gibberellin of some dwarf and
tall wheat cultivars. Planta, 92: 292-300.

Rasmusson, D.C. and R.Q. Cannell. 1970. Selection for

grain yield and components of yield in barley. Crop
Sci. 10: 51-54.

Rawson, H.M. and C.M. Donald. 1969. The absorption and
distribution of nitrogen after floret initiation in
wheat. Aust. J. Agric. Res., 20: 799-808.

Ray, P.M. and A.w. Ruesink. 1962. Kinetic experiments on
the nature of the growth mechanism in oat coleoptile
cell. Dev. Biol., 4: 377-397.

Rayle, D.L. 1973. Auxin induced hydrogen-ion secretion in
Avena coleoptiles and its implications. Planta, 114:
63-73.

Rayle, D.L. and R. Cleland. 1970. Enhancement of wall
loosening and elongation by acid solutions. Plant
Physiol., 46: 250-253.

Rayle, D.L. and R. Cleland. 1972. The in-vitro acid growth
response: relation to in-vivo growth responses and
auxin action. Planta, 104: 282-296.

Reinhard, E. and A. Lang. 1961. Natural gibberellins in
Hyoscyamus niger in relation to development especially
to flower formation. Plant Physiol. 36 (suppl): xii.

Russell, M.B. 1959. Plant responses to differences in soil
moisture. Soil Sci., 88: 179-83.

151

Sachs, R.H., C. Bretz, and A. Lang. 1959. Cell division
and gibberellic acid. Exp. Cell Res., 18: 230-244.

Sharman, B.C. 1947. The biology and deve10pmenta1
morphology of the shoot apex in the graminae. New
Phytol., 46: 20-34.

Sinnott, E.w. 1921. The relationship between body size
and organ size in plants. Amer. Naturalist. 55: 385-
403.

Sinnott, E.N. 1960. Plant morphogenesis. McGraw-Hill,
New York. 550 pp.

Skoog, F. and D.J. Armstrong. 1970. Cytokinins. Ann. Rev.
Pl. Physiol., 21: 359-384.

Skoog, F., F.M. Strong, and C.0. Miller. 1965. Cytokinins.
Science, 148: 532-533.

Smith, J.M. 1968. Mathematical ideas in biology. 152 pp.
Cambridge University Press, London.

Stant, M.Y. 1954. The shoot apex of some monocotyledons
III. Growth organization. Ann. Bot. n.s. 18: 441-447.

Stocker, 0. 1960. Physiological and morphological changes
in plants due to water deficiency. In Arid Zone
Research. XV Plant-water relationships in arid and
semi-arid conditions. (UNESCO).

Sussex, I.M. 1964. Meristems and differentiation. In:
Report of Symposium held June 3-5, 1963. Biology Dept.
Brookhaven National Laboratory, Upton, New York.

Tai, G.C.C. 1975. Analysis of genotype-environment
interactions based on the method of path coefficients
analysis. Can. J. Genet. Cytol., 17: 141-149.

Takahashi, R. and S. Yasuda. 1960. Varietal differences
in responses to photoperiod and temperature in barley.
Ber. Ohara. Inst. Landw. Biol., 11: 365-384.

Tata, J.R. 1966. Hormones and the synthesis and utilization
of ribonucleic acids. Progr. Nucleic Acid Res. Mol.
Biol. 5: 191-250.

152

Thomas, R.L., J.E. Grafius, and S.K. Hahn. 1971a. Genetic

analysis of correlated sequential characters. Heredity
26: 177-188.

Thomas, R.L., J.E. Grafius, and S.K. Hahn. 1971b. Trans-
formation of sequential quantitative characters.
Heredity. 26: 189-193.

Thomas, R.L., J.E. Grafius, and S.K. Hahn. 1971c. Stress:
an analysis of its source and influence. Heredity.
26: 423-432.

Thorne, G.N. 1962. Survival of tillers and distribution of
dry matter between ear and shoot of barley varieties.
Ann. Bot., 26: 37-54.

Trewavas, A. 1968. The effect of 3-indoly1acetic acid on
the levels of polysomes in etiolated pea tissue.
Phytochemistry, 7: 673-681.

Nardlaw, I.F. 1971. The early stages of grain development
in wheat in response to water stress in a single
variety. Bot. Rev. 34: 79-105.

Nareing, P.F. and I.D.J. Phillips. 1970. The control of
growth and differentiation in plants. 303 pp. Pergamon
Press, England.

Nhyte, J.B.A. 1979. Relationship between meristematic
characteristics, yield components and yield of barley
(Hordeum vulgare). M.S. Thesis, MSU.

Nilliams, R.F. 1964. The quantitative description of growth.
In Grasses and Grasslands, ed. C. Barnard. 269 pp.
Macmillan, London.

Nilliams, R.F. 1974. The shoot apex and leaf growth. 256 pp.
Cambridge University Press, London.

Nilliams, R.F. 1975. The shoot apex and leaf growth; a
study in quantitative biology. Cambridge Univ. Press.

Nright, S. 1921. Correlation and causation. J. Agr. Res.,
20: 557—585.

Wright, S. 1934. The method of path coefficients. Ann.
Math. Stat. 5: 161-215.

153

Zavadskaja, I.G. and F.D. Skazkin. 1960. Microsporogenesis
in barley with deficient soil moisture and application
of nitrogen at different developmental periods. CR.
Acad. Sci. U.R.S.S. 131: 692-694.

CHAPTER 2
A MATHEMATICAL MODEL FOR YIELD
INTRODUCTION

Performance of different genotypes vary in different
environments because of the existence of genotype-environment
interactions. The relationship between the performance of
different genotypes in various environments and some measure
of these environments is frequently linear or nearly so
(Finlay and Wilkinson, 1963; Perkins and Jinks, 1968; Breese,
1969; Baker, 1969).

Some articles are available which review and compare
different methods of genotype-environment analysis
(Freeman, 1973; M011 and Stuber, 1974; Hill, 1975). Differ-
ent methods have been proposed to measure the stability of
genotypes tested over a range of environments. These include
multivariate statistical analysis (Freeman, 1973; Hill,
1975), principal component analysis (Perkins, 1972) pattern

1., 1976), cluster analysis (Lin and

analysis (Byth gt
Thompson, 1975) sequential developmental analysis (Tai,
1975, 1979; Nelson, 1981) and independent measures of
environmental factors (Fripp, 1972; Perkins, 1972; Wood,
1976). Much attention is drawn to the factors which
contribute to the interaction of the genotype with the

environment.

154

155

Using the concept of sequential development of yield
components (Grafius, 1969; Grafius and Thomas, 1971;
Thomas gt al., 1971a, b, c), the developmental allometry
(Hamid and Grafius, 1978) in cereal crops, and the proposi-
tion that yield components of cereal crops are determined
at different stages in the ontogeny of plants (Rasmusson
and Cannell, 1970), Tai (1975, 1979) and Nelson (1981)
proposed a yield equation expressed in standard deviation
units. Tai's (1975) equation is composed of a mean genotype
effect, three multiplicative terms of the genotype-
environment interaction and an error deviate. The three
interaction factors are composed of three genotypic
components, each representing the efficiency of a genotype
to utilize a standard deviation unit input, in one of the
three environmental components towards the formation of the
final yield. Nelson (1981) proposed a modification of Tai's
equation by substituting the grand mean and grand standard
deviation (over genotypes and year sites) for Tai's
genotype-specific values. He, however, used the same type
of calculations to arrive at a similar yield equation.
Some erroneous reasoning is presented in the equations so I
intend to propose a new mathematical model for yield, still
based on the concept of sequential development of yield
components, in addition to the regression method of determin-

ing genotype-environment interaction.

THEORETICAL CONSIDERATION

The justification for this model is to offer a modified
form of the models proposed by Tai (1975, 1979) and Nelson
(1981). Use is made of the regression and path coefficient'
analytical methods for determining the genotype-environment
interaction.

The environmental resource is still assumed hypothet-
ically to be separable into three independent groups, each
used during the formation of one each of the yield components.
The cause-effect diagram for the developmental ontogeny of
cereal crop plants (Nelson, 1981) is presented in Figure 1.
The method of path coefficient analysis (Wright 1921, 1934)
is applied to the diagram to determine the relationship
between the three groups of resources and yield.

Let pXY, pXZ, pxw, pr, and pZW be the correlation
coefficients between the yield components and yield and
a], a2, a3, a4, a5, and a6 be their corresponding path

coefficients. Equating the two coefficients, we get:

va = a1
pxz = a2 1 a133
OYZ = a + 6162

3

pxw ‘ a4 + a135 T azae + a1335‘s

pvw = a5 + a1&4 + a335 + a1a235

02w aza4 + a3a5 + a1a3a4 + alaza5 ...... (1)

156

157

 

Figure 1. A causation diagram showing the developmental
relationships between yield and the yield
components: yield (N), number of heads per
unit area (X), number of seeds per head (Y)
and average seed weight (Z). The lower case
letters on the various single-arrow paths are
path coefficients. E1, E2 and E3 indicate
the independent groups of environmental resources.

158

The six path coefficients can be obtained by solving
the simultaneous equations. The path coefficients from
E], E2 and E3 to X, Y and Z, respectively, can be determined
using the regression analysis proposed by Finlay and
Wilkinson (1963), Breese (1969) and Baker (1969).

Suppose the performance of m genotypes is measured in
n environments then the number of heads per unit area Xijk
of the kth replicate of the ith genotype in the jth

environment can be represented by:

.......... (2)

Xijk=“ +91 +ej +dij +€ijk
where
u = the overall mean
91 = the mean effect of the ith genotype
ej = the mean effect of the jth environment
dij = the effect due to the interaction of the ith
genotype in the jth environment.
Eijk = normal random variable, mean 0, variance 02.

A mixed model is assumed with a fixed number of geno-
types and a random sample of environments with r replication.

In calculating the regression of X's of the separate
genotypes on the mean number of heads of all the genotypes,
the dij in equation 2 is regressed on ej thus;

dij = Biej + aij ............... (3)

Where Bi is the linear regression coefficient for the

ith genotype and aij is a deviation.

159

Using equation 3, equation 2 can be rewritten as

Xijk=u+gi+u +Bi)ej+aij+€ijk ...... (4)

The analysis of variance for the above equation is
given in Table 1.

Using the EMSZ, each of the genotype-environment
interaction terms can be compared with the residual
experimental error, and the heterogeneity of regression
further compared with the deviations. This is to determine
whether the regression accounts for a significantly large
part of the observed interaction, a procedure similar to
the joint regression analysis procedure.

A modification of Tai's (1975, 1979) and Nelson's
(1981) procedures for determination of the paths between
the independent environments and the yield components is
presented.

If one wishes to measure the amount of impact of one
variable on another, a standardized coefficient will serve
the purpose because the method of path coefficients is

essentially based on the degree of determination of the

variance of the dependent variable by others (Li, 1975).

2
P

all the environments

Let 0 be the variance in X of the genotypes within

and GE the environmental variance for X

 

2
Then, 0E proportion of population variance contributed
02 by the environment.

'0

'160

e
N

oflp

.uee e 3e eeueweee xwwwam me» em>e weueu e ea eeewme .x meme: eeueeee me wwwz

eewecm>=ee mew

D

N

DL+ D
N

0...... D

oce+
N

NUN'U

n
N

m use + e

N N

e
use ee 0
+N we
Afiwmm.“
an emeeeems
mexuecmovazN

.meaeeme weucme=eew>em ce eeweememme eemeww u me .ceee eewue—zeee

3.:
:se\ ...xe- xnwxeee

N N a
e e\.hwww 1 xnwxwww

N N N
... .a.

E:e\ x¥Ee\ . xw 1

Nu m w New n w
Nee + Ne :e\N xw 1 e\N x N w
NNeEe + No E:e\N...xN- Ee\N.w.x w
Mose + NuNee + No E:e\N...xw- ce\N..wx m

A... we
we emeememe
mezuecmuvpmzm

.xeeum wmeee emst eeueewlezu eew

II
x
a:

A—uevca eeeem

AN-:VAP-EvAeV:eweeeemme Eeew .>mo

P15 Any auwmcmmeememz
A—-:VA_-EV ucmsceew>cm x mexuecmu
_-c geeseeew>cm
_-E mexuecmu
we meeaem

m—eeu mecewee> we ewexwe=< ._ eweew

161

This measures the square of the path from E1 to X.

 

 

2
Thus, 0E = U2
o2 1
P
1/
P -2
U1 = 0%
2 .............. (5)
o
LPJ

 

 

The mode of production of the yield components as
presented in the earlier chapter allow similar analyses to
be used to determine the path coefficients between E2 and
E3 to Y and Z, respectively. Removing the effects of
variation of component traits appearing earlier in the
developmental sequence to yield, as presented by Thomas
gt al. (1971a) is not critical.

If W], r1, r2 and r3 represent yield and the three
independent groups of environmental resources measured in
standard deviation units, a relationship is obtained;

W = V].r1 + Vér2 + Vér3 + e

in which V1, V2 and V3 are the path coefficients from E],
E2 and E3 to yield (W), respectively,and e is the residual

effect. The three path coefficients can be solved as

follows:
v1 = U1 (a4 1 a1&5 T azae + 313335) = Ulp(XW)
v2 ‘ U2(a335 + a5)
V3 = U336 .......... (6)

162

In testing m genotypes over n environments, the yield of the
ith genotype in the jth environment can be expressed
generally as follows;

N - W.. X - X- . Y.. - Y.

 

 

 

 

13 = ij 1. 13 1
a + Vlirlj .L O
°w.. 0Xi. Yi.
' Z.. - Z. '
V2ir2j + 13 1. V3ir3j + e .......... (7)
C21

The above equation can be rewritten as:

w.. = a + K v + K v' r + K v

13 1 11’1j 2 21 2j 3 31’3j + e °°°°° (8)

where wij yield of ith genotype in jth environment
standardized over population mean and variance.
a = constant
K1 = genotypic number of heads per unit area (X)
standardized over genotypic mean and variance.
K2 = genotypic number of seeds per head (Y) standard-
ized over genotypic mean and variance.
K3 = genotypic seed weight (Z) standardized over
genotypic mean and variance.
All other variables have the same meaning as given
earlier.
The formula represents a new mathematical model for

the observed standardized yield, W of a genotype in an

ij’
environment. It is composed of a constant, three

163

multiplication terms of genotype-environment interaction
effect and an error deviate. The interaction effects are
made of the standardized genotypic yield components

(K1, K2, K3), constant genotypic components (V11, V21’ V3 )

i
and environmental components (r1j, ij’ r3j). Two forms

of environmental components can be obtained from X, Y and
Z. The first set of components gives the efficiency of a
genotype to use any of the three environmental components
to produce any of the yield components. The other set gives

the efficiency by which genotypes utilize the three

environmental components in any environment.

MATERIALS AND METHODS

Data from twelve varieties of oats (Table 2) grown
between the years 1976 and 1979, inclusive, will be used
in testing the proposed model. Thirteen sets of data
collected by the late Dr. John E. Grafius, were available
for the analysis in the present study. These were taken
as representing thirteen 'environments' and Table 3 gives
more information about them.

Planting was the same in all the environments. The
plots were four-row plots, 0.0254 m apart and 2.4 m long,
planted at a rate of 30 g per plot. Each plot was
replicated four times in any environment.

All the varieties were reasonably well adapted to
Michigan and exhibit a wide range of various traits.

Data for seeds per head (Y) were obtained from a random
sample of twenty heads per plot just prior to harvest. The
average seed weight (Z) was calculated from a 3 9 sample
per plot using an electronic seed counter to count the
number of seeds within the sample. The number of heads per
30 cm of row was obtained by dividing one sixteenth of the
total weight of grain per plot by the product of seeds per
head and kernel weight in grams.

Simple correlation coefficients between the yield

components and yield for each of the twelve varieties were

164

165

Table 2. Names of the oat varieties.

Menominee*
Korwood*
Mackinaw*
Orbit
Ausab1e*
Mariner*
Clintland 64 (Cld 64)
Dal

Portal
Wright
Garry

Noble

*Michigan Lines

Table 3.

Erivironment

1'10.

10
11
12
13

166

Site

Ingham County
Tuscola County
Kalamazoo County
Lenawee County
Ingham County
Tuscola County
Kalamazoo County
Lenawee County
Tuscola County
Kalamazoo County
Lenawee County
Tuscola County

Lenawee County

Year

1977
1977
1977
1977
1978
1978
1978
1978
1979
1979
1979
1976
1976

Description of the thirteen environments.

Planting
Date
March 14
March 13
March 13
March 11

March 19
March 16
March 23
April 6
April 5

167

calculated. The six correlation coefficients were used in
(l) to obtain estimates of the six path coefficients a1 to
a6 for each variety. The constant varietal components of
the genotype-environment interaction were calculated using
the path coefficients and the U's.

Varietal yield components and yield were regressed on
population yield components and yield to determine the
variation between yield components and yield under varying
environments for each of the varieties.

Using standardized varietal mean yield (over population
mean and standard deviation), standardized yield components
(over varietal mean and standard deviation) and the constant
varietal component of the genotype-environment (GE) inter-
action, the three environmental components of the GE
interactions of each of the thirteen environments were
estimated in standard deviation units by the method of least
SQUares. In determining the environmental component of the
GE interaction for each of the twelve genotypes, the constant
in equation was eliminated. This was done to force the
r‘egression line through the origin and to determine the

COefficient of determination of the GE interaction by the

three GE components of the equation.

RESULTS

The correlation coefficients among the yield components
and yield for each of the twelve varieties are shown in
Table 4. The correlations between the number of heads per
unit area (X) and yield (W) are positive and significant
(P i .01) for all the varieties. That between X and the
number of seeds per head (Y) are negative for all the
varieties but Noble and Mackinaw, however, none is signifi-
cant. Significant positive correlations exist between V
and seed weight (Z) for Mariner and Noble. W has a positive
and significant correlations with Y for six varieties
(Portal, Noble, Mariner, Dal, Cld 64, Mackinaw) and with
Z, for Korwood and Portal.

Table 5 presents the path coefficients between yield
components and yield. The number of heads per unit area
is the largest determinant of yield for all the varieties.
This is followed, in a decreasing order, by number of seeds
per head and seed weight. Wright has the largest a4 value
and the least a5 value. Garry and Menominee have the
highest a5 and a6 values, respectively. The least a4 and
a6 values were obtained for Portal and Orbit, respectively.

The mean square values (Table 6) show that significant

differences exist between the marginal means for varieties,

168

169

 

 

 

 

 

Table 4. Correlation coefficients among yield (W), number
of heads per unit area (X), number of seeds per
head (Y) and seed weight (Z) for each of the
varieties.

X vrs Y vrs Z_1£§

Variety Y Z W Z W W

Menominee -.211 .006 .801*** -.l99 .343 .143

Korwood -.196 .113 .824*** .456 .359 .499*

Wright -.416 .145 .851*** .409 .069 .202

Garry -.205 .336 .775*** .132 .422 .065

Portal -.127 .262 .723*** .375 .568** .583**

Noble .014 .354 .788*** .480* .593** .129

Ausable -.147 .024 .734*** .179 .383 .232

Mariner -.l35 .065 .725*** .490* .551** .391

Orbit -.261 .031 .838*** .313 .275 .192

Dal -.136 .379 .774*** .350 .478* .054

Cld 64 .130 .026 .864** .333 .591** .254

Mackinaw .111 .166 .872*** .008 .557** .216

* P i .10
** P i .05
*** P .01

170

 

 

Table 5. Path coefficients between the yield components
and yield for each of the varieties.

Variety a1 a2 a3 a4 a5 a5

Menominee .211 .038 -.207 .923 .588 .254
Korwood .196 .231 .501 .890 .453 .104
Wright .416 .030 .421 .058 .437 .164
Garry .205 .323 -.O66 .022 .632 .252
Portal .127 .315 .415 .758 .605 .155
Noble .014 .361 .485 .850 .487 .196
Ausable .147 .002 .179 .822 .567 .151
Mariner .135 .001 .490 .793 .580 .159
Orbit .261 .055 .327 .970 .509 .062
Dal .136 .337 .304 .933 .572 .212
CLD 64 .130 .070 .342 .810 .443 .127
Mackinaw .111 .167 -.011 .851 .461 .071

 

171

Po.Wm CI.

 

 

Nooooo. mm.oc Nw.e mw.wewm woe eeeem
wwmooooo. wemm.wow teem.w wwmo.mewo Nm— “emaceew>cu

+ mmwumwee>

e«mwoooo. eweo.NowN eemm.mwe wwom.meNewm Nw “emaceew>cu

wewmmooo. «eoo.Nwow «eN¢.No ewwm.amwem ww mmwuewee>

N w x 3 we meeeem

 

.mueme:eew>:m mp em>e emumep
mmwumwee> NP we ANV eeowez emmm e:e va eem; ewe memmm we emee::
.Axv emee “we: eme eeeme we emee:c .sz ewmwx eew mmewe> meeeem :eez .e mweew

172

environment and the interaction between varieties and
environment for all the characteristics under consideration.
Tables 7 and 8 give the mean values for yield, number of
heads per unit area, number of seeds per head and seed
weight for the varieties and environments, respectively,
arranged in a decreasing order. Cld 64, Mackinaw, Noble and
Cld 64 have the least mean values for W, X, Y and Z,
respectively. Menominee, Noble, Garry and Orbit have the
highest mean values for the above mentioned characteristics,
respectively. Generally, varieties with a low mean X values
had high Y values or vice versa. Menominee, the highest
yielding variety of the set used in the experiment, had
relatively high values for both number of heads per unit
area and number of seeds per head. It had medium seed
weight. Cld 64 had relatively low values for all the
measured characteristics.

Environments 13 and 4 were the worst in supporting the
production of the yield components and yield. Some forms
of negative associations are observed between the environ-
ments for the production of X and Y, however, these are
not as obvious as represented by the varietal mean values.
Environments 1 and 9 are examples. This is the result of
the low insignificant negative correlations between the

number of heads per unit area and the number of seeds per

173

 

 

 

 

 

 

 

emee. NN.ee em.m mm.Ne_3_o.uev ewe
mwoo. em.w_ wo.m ew.ww Amo.uav owe
enema. “weeo oo.wm xeeew we.w_ mweez Ne.m_e emcweecmz
owmmo. seewxeez Nm.eN weueee mm.e_ mmeweecmz Nm.mmm “weeo
ewNmo. mweeez< ew.eN mmcwEeeez N_.ew eweeo RN.mwm eeezeex
eeeeo. eeeez No.ee eoezeex eo.e_ eeeeez mN.eee eeeee
emomo. eeezeex mN.ew emeweez mo.mw eecweez om.mmm mweeee<
emomo. xeeeu ow._w mweeez< we.mw eeezeex No.mmm emcweez
moomo. mmcwEeemz me.oN gecwxeez wm.m_ wee Fm.mem gecwxeez
wmmNo. weueee mm.mo wee om.mw mweeee< oo.wmm mweez
omeo. wee ON.me Hemwez ON.m— em ewe No.eNm wewpeee
mmwNo. emeweez wm.we we ewe we.ew xeeew me.eme unmwez
memNo. uemwez Nw.Ne “weeo em.ew weueee Ne.mme wee
mmeo. em ewe ww._e mweez Nm.m— geewxeez mm.mee em e—u
mzwe> mexaecmo mewe> mexpecmw mzwe> meapecmw mswe> mexaecmw

:emz cemz :emz :emz

N > x 3

ewe memme we emee=e .Axv emee ewe: eme emww” wmmwwmemmemwuueeflwwzemwu

meswe> ceme mcwmemeeew we emeee men cw mmwumwee> me“ we acmemmceee< .e mweew

174

 

 

 

 

 

 

Table 8. Arrangement of the environments in the order of
increasing mean values for yield (W), number of
heads per unit area (X), number of seeds per
head (Y) and seed weight (Z).

W X Y Z
Env. Mean Env Mean Env. Mean Env. Mean
No. Value No. Value No. Value No. Value
13 258.88 13 10.34 13 54.77 4 .02899
4 321.10 4 10.71 1 64.72 2 .02900
12 452.79 12 12.21 9 65.35 13 .02948
11 456.63 11 12.93 4 66.62 3 .02973
8 530.40 10 14.67 11 67.38 12 .03047
6 575.02 8 15.34 8 70.61 6 .03085
10 580.25 6 16.15 6 72.93 8 .03095
9 603.48 7 17.27 2 73.06 5 .03132
1 611.13 5 17.68 5 74.81 7 .03146
2 636.79 9 18.04 3 76.42 10 .03155
5 651.98 1 18.75 12 77.39 1 .03186
3 676.92 3 19.17 10 80.22 9 .03235
7 704.31 2 19.17 7 82.48 11 .03307

LSD(«=.05)

77.76 3.01 11.56 .0019

LSD(«=.01)

102.33 3.96 15.22 .0024

 

175

head for all the varieties. Other contributing factors are
the differences in reaction of the varieties to differing
environments for the measured characteristics as shown in
the Appendix (Figures Al to A4).

Highly significant linear regressions exist between the
measured genotypic characteristics and their respective
environmental indices. An environmental index is the mean
for a characteristic of all the varieties within the
environment as used for the regression model for studying
genotype-environment interaction (Finlay and Wilkinson,
1963; Breese, 1969). The regressions accounted for most of
the variation in the variety-environment interaction for
the yield components and yield of all the varieties.
Although the regressions were highly significant (P i .01)
in almost all the determinations, there were relative

differences in the genotypic R2

values for seed weight.
Figure 2 shows the regressions of genotypic mean yield
on the environmental yield indices for five of the varieties
used in the study. Menominee's performance is above the
environmental mean yields while Cld 64's performance is
below. Wright performs well in the relatively poor growing
conditions and poorly in the good environment. The reverse

is the case with Mackinaw. It performs poorly in the poor

environment but better than average in the good environment.

Genotypic Mean Yield

Figure 2.

176

    
  

 

900
Menominee
Orbit
Mackinaw
700 "
Wright
Cld 64
500
Menominee: Y = 32.73+1.07X
300 Orbit: Y =-46.82+1.16X
,a’ Mackinaw: Y =-l38.65+l.26X
Wright: Y = 78.62+77X
Cld 64: Y =-28.25+.90X
100

200 400 600 800
Environmental Mean Yield
Yield response of Mackinaw, Orbit, Wright, Cld 64

and Menominee to thirteen varying environments in
Michigan, 1976-1979.

177

The performance of Orbit is above average in the good
environment, however, it has a mean yield equal to the
environment mean yield of the relatively poor growing
conditions.

The performances of Cld 64 and Mackinaw were below the
average genotypic performances while Menominee and Wright
had above average production of number of heads per unit
area in all the environments (Fig. 3). Orbit produced more
heads per unit area than the genotypic averages under the
good environmental conditions but produced less than the
average X in the poor environments.

The regressions of the genotypic mean Y values on the
environmental indices are shown in Figure 4. Menominee has
a better performance than average in all the environments
while Cld 64 and Orbit have poor performances in all
environments. Mackinaw produces a lesser number of seeds
per head than average in the poor environments while more
seeds per head than average are produced in the better
environments. Wright produces a higher and lower than
average Y in relatively poor and good environments, respect-
ively.

Orbit and Mackinaw have heavier seeds while Menominee,

Wright and Cld 64 have lighter seeds than average in all

environments (Fig. 5).

178

      

 

 

22 Orbit
Menominee
Wright
few 64
18 ‘ Mackinaw
CD
3
E
>
X
C 14 I
«5
w
E
.U
a
>,
.p
O
C
3 10 -
Menominee: Y = l 33+1.0X
Orbit: Y =-3.6l+1.27X
Mackinaw: Y =—3.11+l.06X
Wright: Y = .21+1.02X
Cld 64: Y = .02+.97X
1
6 I l
10 14 18

Environmental mean X value

Figure 3. Response for the number of heads per unit
area of Mackinaw, Orbit, Wright, Cld 64 and
Menominee to thirteen varying environments
in Michigan, 1976-1979.

179

100
Mackinaw
Menominee
g 80 ,o" 1d 54
'; '." Wright
>
>. ',v'.1"b1.t
C ’o"
g 0 /
Z ',
.2 5o ,.-"/ Menominee: Y = 9.05+.94x
g ,.-' Orbit: Y = 18.05+.63X
u ,ef’ Mackinaw: Y =-42.34+l.58X
2 Wright: Y = 27.17+.58X
g; Cld 54: Y ==4.80+1.02X

 

4O

50 6O 7O 80

Environmental Mean Y Value

Figure 4. Response for the number of seeds per head
of Mackinaw, Orbit, Wright, Cld 64 and
Menominee to thirteen varying environments
in Michigan, 1976-1979.

180

      

(x 10-3)"
40 Orbit
ackinaw
35

CD

3

g Menominee

N

c Wright

(G

G)

E

U Cld 64

'S. 30

>5

4.)

O

C

(3 Menominee: Y = .002+.92X
Orbit: Y = .003+l.09X
Mackinaw: Y = .003+l.09X
Wright: Y =-.OO4+1.05X
Cld 64: Y = .OO6+.73X

 

25 _3
28 3o 32 34(x 1o )

Environmental Mean Z Value
Figure 5. Seed weight response of Mackinaw, Orbit,

Wright, Cld 64 and Menominee to thirteen
varying environments in Michigan, 1976-1979.

181

One ought to be careful about these kinds of comparisons
since they seem to reflect adversely upon certain varieties.
We must remember that the particular sample of varieties
used in such analysis determines the mean and departures
from the mean slope and one variety may not actually be
worse than many other varieties if the sample were different.

Component compensation between X and Y is not shown to
any great extent in Figures 2 and 3. Performances of Orbit
and Wright in relationship to X and Y show some signs of
compensation. The signs of compensation shown by the two
varieties results from the relative high but insignificant
negative correlations between their X and Y. Orbit and
Mackinaw have the highest response to environment for seed
filling.

The path coefficients between E1 and X, E2 and Y and
E3 and Z are .7949, .5915 and .3849, respectively.
Environment thus play its largest role in the production of
heads followed by seeds per head and seed weight.

Estimates of the genotypic constants of the genotype
environment interaction are given in Table 9. The estimates
were positive for all the varieties. The constants were
highest for number of heads per unit area followed by
number of seeds per head and seed weight. v; values were

generally about twice the values of Vé. The V3 values were

182

Table 9. Estimates of the varietal constant component of
the variety-environment interaction.

 

 

 

Variety V1' V2' V3'
Menominee .6367 .3167 .0978
Korwood .6550 .2988 .0400
Wright .6765 .3006 .0651
Garry .6160 .3640 .0970
Portal .5747 .3959 .0697
Noble .6234 .3443 .0754
Ausable .5835 .3514 .0581
Mariner .5763 .3892 .0612
Orbit .6661 .3131 .0239
Dal .6153 .3765 .0816
Cld 64 .6868 .2877 .0489
Mackinaw .6932 .2722 .0273

Vl' = .7949 o(XW)
V2' = .5915 (a3a5 + as)
V3' = .3849 a5

183

characteristically variety-specific. They varied between
3.5% of v; (Orbit) to 15.7% of v; (Garry).

The correlation coefficients between the varietal mean
yield components and estimates of the varietal constants are
given in Table 10. v; has negative correlations with Vé and
V3, however, only its correlation with Vé is significant
(P i .01). An insignificant positive correlation exist
.

between V2 and V3 The genotypic mean seed weight has a

highly significant (P i .05) negative correlation with V3

values.

Table 11 gives estimates of the environmental components
of the genotype-environment interaction for the twelve
varieties. All the varieties have relatively similar
efficiencies in utilizing a standard unit of each of the
environments E] and E2 to produce their number of heads per
unit area and the number of seeds per head. Relative
efficiencies for the production of seed weight varies. Orbit
and Mackinaw have the highest efficiencies, 5.661 and 4.961,
respectively, in utilizing environment E3 to produce their

seed weight characteristics. Garry has the least efficiency

(1.916) in seed weight production through the utilization of E3.

The coefficient of determination for the multiple
regression analysis involving yield, as the dependent
variable, and the three multiplicative terms of genotype-

environment interaction, as independent variables, are

184

 

 

 

Table 10, Correlation coefficients among varietal mean
number of heads per unit area (Xi): number of
seeds per head (Yi); seed weight (21) and
estimates of the varietal constants (V1'. V2',
V3') of the variety-environment interaction.
x,- Y; 2,- v1' v2'

V1' -.120 -.420 .329

V2' .086 .355 -.382 -.936**

V3' .341 .433 -.600* -.441 .486
* P < 05

** P < .01

185

 

 

 

Table 11. Estimates of the environmental components of the
variety-environment interaction for the varieties.
Variety r1 r2 r3 R2
Menominee 1.491 1.736 2.643 .755**
Korwood 1.470 1.432 4.736 .959**
Garry 1.409 1.585 1.916 .949**
Portal 1.212 1.283 2.731 .965**
Noble 1.266 1.433 2.853 .990**
Ausable 1.381 1.837 3.529 .973**
Mariner 1.375 1.115 3.220 .983**
Orbit 1.478 1.420 5.661 .913**
Dal 1.189 1.104 2.634 .789**
Cld 64 1.024 1.516 2.530 .604**
Mackinaw 1.336 1.519 4.961 .981**
Wright 1.155 1.455 2.906 .794**
** P i .01

186

highly significant (P i .01). This shows that each genotype
has an unique developmental processes whereby the environments
E], E2 and E3 are utilized in the production of its yield
components and yield. The R2 values are obtained after
forcing the regression line to pass through the origin

(the constant in the yield equations was eliminated). Thus

a high proportion of the variations in the yields of a

variety within the tested environments is accounted for by

the variety's interaction with the environment.

Estimates of the environmental components of the
genotype-environment interaction of the thirteen environments
are presented in Table 12. Seven of the environments had
significant yield predictions for the genotypes. This is
interpreted as due to the similar behaviors of all the
varieties within these environments. The proportion of
variation among the yields of all the varieties in the seven
environments accounted for by the genotype-environment
interactions varied between about 55% (environment 6) and
82% (environment 5).

The r values presented in Table 12 give the requirements
of the varieties from E], E2 and E3 for the production of a
standard unit of each of the yield components X, Y, and 2,
respectively. The contribution of E], E2 and E3 to the
production of X, Y, and Z, respectively, and their signifi-

cance varied among the seven environments. E] was

Table 12.

187

Estimates of the environmental components of
the variety-environment interaction for the
environments.

 

 

 

Environment r1 r2 r3 R2
no.
1 .019*** 2.875* 7.904 .770***
2 .218 .018 .348 .034
3 .151 .685 5.252 .090
4 .782** .784 1.428 .417
5 .878*** 2.683 .537 .824***
6 .004** 1.562 9.412** .551*
7 .767 1.173 -8.661* .489
8 .283*** 1.236 3.361 .725**
9 .075*** 5.669*** 5.461** .737***
10 .721 .972 - .866 .235
11 .493** 2.173** 5.585 .685**
12 .329 1.191 7.003 .367
13 .667*** .889** 1.181 .788***
* P i .10
** P _ .05
*** P .01

188

significant in all the predictions. E2 was significant
for yield predictions in environments 1, 9, 11, and 13,
while E3 was significant in varietal yield predictions in
environments 6, 7, 9 and 11. Thus in environment 1, about
2, 3, and 8 units of E], E2 and E3, respectively, are
required for the production of standard units of X, Y, and
Z, respectively. The contributions of the genotype-environ-
ment interactions for production of X and Y are significant
in the prediction of yield of all the genotypes within
environment 1. Genotype-environment interaction for 2
production is not significant in the yield prediction. An
interpretation can now be given to the r values and the
significance of the genotype-environment for each of the
yield components in the prediction of yield.

Figure 6 shows the regression of the predicted yield
on the observed yield when the model presented in equation
8 was applied to the data for the twelve varieties in all

the thirteen environments. The R2

for the highly signifi-
cant (P i.°01) multiple regression equation is .8901. Thus
about 89% of the variation in the observed yields can be
accounted for by the predicted yield made primarily from
the genotype-environment interactions involved in the
production of the yield components. The analysis of

variance and multiple regression statistics for the regres-

sion are presented in Tables Al and A2, respectively. All

Predicted Yield (standard deviation units)

189

1.4 o a ,
o o.
o ' . ....
a... '0‘. .
’ . ..Q ...
0". no...
0 4 ° .-' . '5’: *-
0....’O . 0..
. O. O
o . ‘
. .'.oo.. .
o :13 . °.
0
-0.6 . 0‘0 .
O o .
o
t. w2= 1.318X+l.521Y+2.6782
. o 0 R = .8901***

 

-2.8 -1.8 -O.8 0.2 1.2

Observed Yield (standard deviation units)

Figure 6. Regression of predicted yield on observed
yield.

190

the independent characteristics were significant, however,
X had the largest predictive value followed, in a decreasing

order by Y and then 2.

The regression of the residuals on the predicted
yield show a horizontal band (Figure 7) showing no ambiquities

associated with the model.

Residuals (standard deviation units)

.6
.6
.6
-2.6
Figure 7.

191

 

-1.6 -O.6 0.4 1.4

Predicted Yield (standard deviation units)

Regression of residuals on predicted yield.

DISCUSSION

The basic assumption in the present analysis is that
three independent environmental components, E], E2 and E3
were involved in the production of the yield components X,
Y, and Z, respectively. Based on the concept of sequential
development of the yield components, the effect of each of
the environmental components on yield was determined for
each genotype and expressed as the genotypic constants
(Vi, Vé, V5). The constants are in effect the path
coefficients between the three independent environmental
components and yield.

Hamid and Grafius (1978) showed that plant organs
established earlier in the developmental ontogeny of small
cereal grains crops have a great effect on the later
developed organs. Thomas gt gt.(1971a) provided a method
for eliminating the variation of an earlier established
yield component on the development of later organs. The
adjusted yield component values so obtained are independent
of each other. Using these adjusted values for the
determination of the path coefficients between the yield
components and yield eliminates any dependency between them.

Tai (1975, 1979) and Nelson (1981), used similar methods

for the determination of the path coefficients between the

192

193

independent environmental components, E], E2, and E3 and
X, Y, Z, respectively. Their determination signifies that
X, the first developed yield component, is determined solely
by the environment. This is inflationary because the
variation within the performances of genotypes is not only
determined by the environment but, additionally, by
genotypic variation and the interaction between a genotype
and environment. Consideration to the three sources of
variations need to be taken into account in imposing
restrictions or making assumptions in the formulation of a
model for any plant characteristic. Using the method pro-
vided for the determination of the path coefficients
between E], E2, and E3 with X, Y and Z, respectively, the
author was able to remove the effects of the inflations
associated with the determinations used by Tai (1975) and
Nelson (1981). The effects of E1 is largest in the
production of its respective yield component (i.e., X)
followed by E2 and then E3 in the production of Y and 2,
respectively, as shown by their u values.

Another inherent error in the series of equation used
to determine the path coefficients from E2 and E3 to Y and
Z, respectively, is observed. The equations are valid if
one is dealing with only the yield components. Environmental

indices have been constructed using the mean values of all

194

genotypes in each environment as in the regression model
for genotype-environment interaction determinations.
Generally, much of the variability in the measured
characteristic (yield) of genotypes is accounted for by
the linear functions of the environmental indices. Using
the measurable traits (i.e., yield components) to depict
the independent environmental components (E2 and E3)
becomes erroneous because the relationships between the
yield components is not indicative of the correlation
between any of the environmental components. Different
genotypes behave differently in various environments. The
significant negative correlation coefficient between X and
Y between different genotypes in an environment was purely
physiologic as shown in Chapter 1. Additionally, the
effects of the independent environmental components on the
yield components is decreased with the later developed
components. Environments play their largest role in the
development of plant structures established earlier in the
ontogeny of small cereal grain crops. Plant structures
established later are genotypically controlled. Caution
has to be taken in calculating environmental effects as
presented by Tai (1975) and Nelson (1981).

Two measures of phenotypic stability of genotypes have

been used when genotypic characteristics are regressed on

195

their respective environmental indices. Since the mean
slope is given by the regression of the environmental
indices on itself, b=l. Thus genotypes with regression
coefficients significantly higher or lower than one are
regarded as having a low or high degree of stability,
respectively (Finlay and Wilkinson, 1963; Perkins and Jinks,
1968). Eberhart and Russel (1966) and Breese (1969)
proposed a second measure obtained by combining the
environment and genotype-environment terms in the analysis
of variance and testing for the heterogeneity (deviations
from regression) of the regressions. They realized that
regressing genotypic means on the environmental indices
leads to inexact tests of significance however, using the
deviations as a measure of phenotypic stability.

These determinations have been criticized on statisti-
cal grounds (Freeman and Perkins, 1971). Fripp (1972)
showed that biases introduced by the use of the indices does
not affect the ranking of the genotypes according to the
magnitude of their regression coefficients or to the
proportion of genotype-environment interaction accounted
for by the heterogeneity of these regressions when compared
with results of analysis of regression against various
independent but biological measures. Freeman and Perkins

(1971) outlined methods for characterizing environments

196

without using the same individuals to determine both the
environmental effects and genotype-environment interactions.

When the linear regression accounts for a significantly
large part of the genotype-environment interaction, reliable
predictions are obtained (Bucio-Alanis, Perkins and Jinks,
1969). If only a small portion of the variation is
accounted for, other forms of analysis are used because the
analysis will have no predictive value.

The genotypic stability for all the measured character-
istics can be determined using their respective regression
coefficients since significant portions of the genotype-
environment interactions are accounted for in the genotypes
for all the characteristics. These observations suggest
some genuine underlying linear relationships between the
performances of genotypes in different environments. The
dispersion of the regression lines for Z presents an
interesting observation which needs further enquiry.

The genotype-environment significantly accounted for
between 60% (Cld 64) and 99% (Noble) of the variation in
yield of the genotypes (after removing the constant from
the model). Addition of the constant increased the R2
values above .980 in all the genotypes. This shows that

each variety behaves similarly in all environments.

Estimates of the environmental components of the

197

genotype-environment interaction show that the genotypes
have virtually similar efficiencies in the production of
their number of heads per unit area and number of seeds per
head. The slight differences might be due to the additional
environment required for the transformation from the
vegetative to reproductive growth state and further
differentiation of the floral parts. Efficiencies for seed
weight production seem to be subjected to much environmental
effects as shown by the varying values for r3 among the
genotypes. Orbit and Mackinaw have the highest capacity
(5.661 and 4.961, respectively) in utilizing the variation
in the environment for their Z production while Garry has
the least capacity (1.916) for the same process. A genotype
is herein defined as stable when it produces a unit of the
yield components from a unit each of the independent
environmental components E], E2, and E3. Cld 64, Dal and
Garry are the most stable genotypes relative to the other
genotypes used in the study for the production of X, Y, and
Z, respectively. Menominee, Ausable and Orbit are the

least stable of the genotypes for the production of X, Y,
and Z, respectively. When all the yield components are
considered together, Garry emerges as the most stable among
the genotypes under test, the test having been conducted

in a given set of environments.

198

The coefficient of determination of the observed yield
by the predicted yield varied between 3.4% (environment 2)
and 82% (environment 5) for the thirteen environments.
Seven of the environments had significant yield predictions
for the genotypes. This is explained by the similarity of
behavior of the genotypes within those environments. The
independent environmental requirements for the production
of the yield components varied greatly from one environment
to another. In defining environmental stability as that
from which genotypes (behaving similarly) require one
standard unit of each of E], E2 and E3 to produce one
standard unit of X, Y, and Z, environment 13 is characterized
as being the most stable environment of the studied
environments. This environment is the worst of the lot.

Certain effects which are normally covered in the
linear regression method of determining genotype-environment
interactions are depicted when this multiple regression
model is used. Variation between genotypes exist in more
than one dimension.

The highly significant (P :_.01) linear relationship
between the predicted yield and the observed yield, the
horizontal band obtained by the regression of residuals on
the predicted yield and the magnitude of the coefficient

of determination (R2 = .8901) render the model useable for

199 :1

PI} T.

yield prediction. The basic phenomenon used in the formula-
tion of the model is the sequential development of the

yield components as outlined by Hamid and Grafius (1978)

and Grafius (1978).

 

SUMMARY AND CONCLUSION

A modified form of models proposed by Tai (1975, 1979)
and Nelson (1981) is offered making use of the regression
and path coefficient methods for determining the genotype-
environment interaction. Basically, the model involves the
use of path coefficient analysis (Wright, 1921, 1934) to
determine the effect of three hypothetically separable
independent groups of environmental resources to yield
through their direct effect on one each of the yield compon-
ents. The direct effects are calculated from the proportion
of environmental variance in the population variance for
each of the yield components.

The model is composed of a constant, three multiplicative
terms of genotype-environment interaction effect and an
error deviate. The interaction effects are made of the
standardized genotypic yield components, constant genotypic
components and environmental components. Two sets of
environmental components are obtained from X, Y and Z.

The first gives the efficiency of a genotype to use any of
the three environmental components to produce any of the
yield component. The other set gives the efficiency by
which genotypes utilize the three environmental components

in any environment.

200

201

Data from twelve varieties of oats grown between the
years 1976 and 1979, inclusive, was used in testing the
proposed model. There were thirteen environments in all.
The data included yield, number of heads per unit area,
number of seeds per head and seed weight. The necessary
analyses were performed to obtain values for the different
variables of the model from which the two sets of environ-
mental components were estimated. In determining the
environmental component of the genoquyenvironment interact-
ion for each of the twelve varieties. the constant in the
equation was eliminated.

The analysis of variance showed the existence of
significant differences between the marginal means for the
varieties, environments and their interaction for all the
characteristics. Highly significant linear regressions
exist between the measured varietal characteristics and
their respective environmental indices, however, there were

2 values for seed

relative differences in the varietal R
weight. Environment was detected as playing its largest
role in the production of number of heads per unit area
followed, in a decreasing order, by number of seeds per
head and then seed weight.

Estimates of the varietal constants were positive for

all the varieties. The highest group of values was

202

associated with the number of heads produced per unit area
followed by number of seeds per head and then seed weight.

All the varieties had relatively similar efficiencies
in utilizing a standard unit of each of the environments
E1 and E2 to produce their X and Y. Efficiencies for seed
weight production varied among the varieties, There was
evidence to show that each variety had unique developmental
processes whereby the environments E], E2 and E3 were
utilized in the production of its yield components and
yield.

Seven of the eleven environments tested had significant
yield predictions for the genotypes. This was interpretted
as the result of the overlapping of the unique developmental
processes possessed by the varieties. The contribution of
E], E2 and E3 to the production of X, Y and Z, respectively,
and their significance varied among the seven environments.

The regression of the predicted yield on the observed
yield using all the data was highly significant (P i .01).
The coefficient of variation for the multiple regression
equation was .8901.

A horizontal band was obtained when the residuals were
regressed on the predicted yield showing the absence of
ambiquities in the model.

Certain effects which are normally covered in the linear

203

regression method of determining genotype-environment
interactions are exposed using this model. Variation is

shown to exist in more than one dimension between genotypes.

APPENDICES

204

 

 

womm. n N3
coop. ova.mw wa eeeem
mooo.v owem.o_¢ wva.we Nwa.MNw m :ewmmmemmm
Nmmo.mmw mmw Feuew
mwm e m: mm we muezem

 

.xwm>wueeeeme .ANV pzmwmz emee e:e Awe eew; eme

memem we emee:: .Axv emee awe: eme eeem; we eeee=c we» we :ewueaeeee
meu mew>we>cw Aemweewee> acmecmemecwv cewpeeempcw u:mE:eew>:m
-xumwee> we mEeme m>wpeewwewpwee mmeep mep we Amweewee> acmecmemev
ewew» em>emeee me» we cewpewemee mew eew mecewee> we ewmx—e:<

.—< mweew

205

 

 

 

meow. meow. mooo.v ewew. mm0.N N
Nemw. emow. mooo.v eooe. wNm._ w
wemm. wam. mooo.v mNmN. mwm._ x
Foam. mmoo.- mooo.v Nem. Noo. . .peceu
meee_ee eeeweewweoe eeeeeewweewe eemwez eeewewwweee eeeewee>
m eewueweeeeu epem cewemeemmm
N eeweeee
.zwm>wuemeeme .ANV uemwmz eemm e:e
va eew; eme eemmm we emee:c .Axv emee ewe: eee eeeme we emee:: we eew
upeeeeee mew mcw>we>ew Ameweewee> acmecmemecwv :ewpueempew acmEceew>cm
izemwee> we meemp m>wuemwwewuwee mmeep men we Amweewee> uceeememev
ewmwx ee>emeee me» we :ewuewemee esp eew euwumwpeue :eweeeemme ewewuwzz .N< mweew

Figure A1. Regression equations and graphs for the y'i e1
response of the twelve varieties to thirteen
varying environments in Michigan, 1976-197 9 .
Varietal Varietal Regression R2
Letter Name Equation
A Menominee Y = 32.725+l.069X .9659**
B Orbit Y =-46.820+l.164x .9725**
C Korwood Y =-86.045+1.214X .9507**
D Mackinaw Y =-l38.653+l.257X .9684* *
E Garry Y = 51.298+.956X .9062**
F Ausable Y = 32.100+.969X .9783**
5 Portal Y =-5.505+.977x .9556**
H Noble Y = 19,240+.953x .8916**
I Mariner Y = 17.214+.985X .9773**
.1 Wright Y = 78.621+.769X .8957**
K Cld 54 Y =-28.247+.906X .8502*"‘
L Dal Y = 75.523+.773x .9411 **
** P < .01

206

Genotypic Mean\Yield

900 +

700 '

500 '

300 '

 

 

 

A
B,C
D
E,F
G
H
I
J,K
L

l I '

200 400 600 800

Environmental Mean Yield

208

Figure A2. Regression equations and graphs for the response
for the number of heads per unit area of the
twelve varieties to thirteen varying environ-
ments in Michigan, 1976-1979.

Varietal Varietal Regression R2

Letter Name Equation
A Menominee Y = 1.325+.999X .8840**
B Orbit Y =-3.607+1.266X .9457**
C Korwood Y =-3.709+l.244X .9309**
D Mackinaw Y =-3.105+1.055X .8824**
E Garry Y =-.039+.945X .8729**
F Ausable Y = .998+.919X .8211**
G Portal Y = 2.911+.748X .8736**
H Noble Y = 2.371+.985X .7328**
I Mariner Y = .704+.985X .9334**
J Wright Y = .214+l.018X .7680**
K Cld 64 Y = .024+.974X .7791**
L Dal Y = 1.919+.86OX .8974**

**P i .01

209

IF 0

, ’ Ir
HBACJ KLEG

 

221

I
8
.I:

14 -

m2Fe> x gem:

eweaueemw

 

22

18

14

10

Environmental Mean X Value

210

Figure A3. Regression equations and graphs for the response
for the number of seeds per head of the twelve
varieties to thirteen varying environments in
Michigan, 1976-1979.

Varietal Varietal Regression R2

Letter Name Equation
A Menominee Y = 9.051+.941X .6551**
8 Orbit Y = 18.093+.628X .5742**
C Korwood Y = 10.437+.910X .6982**
D Mackinaw Y =-42.339+l.582X .8693**
E Garry Y =-19.845+1.415X .8090**
F Ausable Y =-15.965+1.230X .7788**
G Portal Y =-23.420+1.402X .9268**
H Noble Y = 9.619+.732X .6074**
I Mariner Y = 10.29l+.897X .8904**
J Wright Y = 27.166+.576X .4326*
K Cld 64 Y =-4.799+l.015X .7007**
L Dal Y = 21.262+.683X .7682**

* P < 05

** P < .01

211

100

AFI

, ’ ,

G DCE KLJBH

8O
60

mzwe> > :emz eweXeecmw

 

40

Environmental Mean Y Value

 

212

Figure A4. Regression equations and graphs for the seed
weight response of the twelve varieties to
thirteen varying environments in Michigan,

1976-1979.
Varietal Varietal Regression R2
Letter Name Equation
A Menominee Y = .0018+.9l6OX .4652**
B Orbit Y = .0026+1.0927X .6607**
C Korwood Y =-.0158+l.4948X .7782**
D Mackinaw Y = .0026+1.0861X .6086**
E Garry Y = .0064+.7748X .5144**
F Ausable Y =-.0059+l.2317X .8405**
G Portal Y = .0084+.6771X .4455*
H Noble Y =-.OO39+1.1229X .5378**
I Mariner Y = .0058+.7498X .5378**
J Wright Y =-.OO38+1.0514X .7153**
K Cld 64 Y = .0057+.7263X .5720**
L Dal Y =-.001l+.9783X .6434**
* P i .05
** P < .01

 

213

 

 

 

(x 10'3)

4O

5
3

e:_e> N cemz ewexpeeeo

25

34 (x 10'3)

32

28

Environmental Mean 2 Value

LIST OF REFERENCES

LIST OF REFERENCES

Baker, R.J. 1969. Genotype-environment interactions in
yield of wheat. Can. J. Plant Sci., 49: 743-751.

Breese, E.L. 1969. The measurement and significance of
genotype-environment interaction in grasses. Heredity
27: 27-44.

Bucio-Alanis, L., J.M. Perkins, and J.L. Jinks. 1969.
Environmental and genotype-environmental components

of variability. V. Segregating generations. Heredity,
24: 115-127.

Byth, D.E., R.L. Eisemann and I.H. DeLacy. 1976. Two-way
pattern analysis of a large data set to evaluate
genotypic adaptation. Heredity, 2: 215-230.

Eberhart, S.A. and W.A. Russel. 1966. Stability parameters
for comparing varieties. Crop Sci., 6: 36-40.

Finlay, K.W. and G.N. Wilkinson. 1963. The analysis of
adaptation in a plant breeding programme. Aust. J.
Agric. Res., 14: 742-754.

Freeman, G.H. 1973. Statistical methods for the analysis

of genotype-environment interactions. Heredity. 31:
339-354.

Freeman, G.H. and J.M. Perkins. 1971. Environmental and
genotype-environmental components of variability.
VIII. Relations between genotypes grown in different
environments and measures of these environments.
Heredity, 27: 15-23.

Fripp; Y.J. 1972. Genotype-environmental interactions in
Schizophyllum commune. II. Assessing the environment.
Heredity, 28: 223-238.

Grafius, J.E. 1969. Stress: A necessary ingredient in
genotype by environment interaction. tg: Barley
Genetics II. Proc. 2nd Intern. Barley Genet. Symp;
Pullman, Washington St. Univ. Press; Pullman, pp.
346-355.

214

215

Grafius, J.E. 1978. Multiple characters and correlated
responses. Crop Sci., 18: 931-934.

Grafius, J.E. and R.L. Thomas. 1971. The case for indirect
genetic control of sequential traits and the strategy
for deployment of environmental resources by the plant.
Heredity, 27: 433-442.

Hamid, Z.A. and J.E. Grafius. 1978. Developmental allometry
and its implication to grain yield in barley. Crop
Sci., 18: 83—86.

Hill, J. 1975. Genotype-environment interactions - a
challenge for plant breeding. J. Agric. Sci., Camb.
85: 477-493.

Li, C.C. 1975. Path analysis: a primer. 347 pp. Boxwood
Press, Pacific Grove, California.

Lin, C.S. and 8. Thompson. 1975. An empirical method of
grouping genotypes based on a linear function of the
genotype-environment interaction. Heredity, 34: 255-
263.

Moll, R.H. and C.N. Stuber. 1974. Quantitative genetics
-empirical results relevant to plant breeding. Adv.
Agron., 26: 277-313.

Nelson, J.L. 1981. An analysis of selected breeding
approaches for oats and barley. Ph.D. Thesis, MSU,
117 pages.

Perkins, J.M. 1972. The principal component analysis of
genotype-environmental interactions and physical
measures of the environment. Heredity, 29: 51-70.

Perkins, J.M. and J.L. Jinks. 1968. Environmental and
genotype-environmental components of variability. III
Multiple lines and crosses. Heredity, 23: 339-356.

Rasmusson, D.C. and R.Q. Cannell. 1970. Selections for
grain yield in barley. Crop Sci., 10: 51-54.

Tai, G.C.C. 1975. Analysis of genotype-environment
interactions based on the method of path coefficient
analysis. Can. J. Genet. Cytol., 17: 141-149.

216

Tai, G.C.C. 1979. Analysis of genotype-environment
interactions of potato yield. Crop Sci., 19: 434-438.

Thomas, R.L., J.E. Grafius, and S.K. Hahn. 1971a. Genetic

analysis of correlated sequential characters. Heredity,
26: 177-188.

Thomas, R.L., J.E. Grafius, and S.K. Hahn. 1971b. Trans-
formation of sequential quantitative characters.
Heredity, 26: 189-193.

Thomas, R.L., J.E. Grafius, and S.K. Hahn. 1971c. Stress:
an analysis of its source and influence. Heredity,
26: 423-432.

Wood, J.T. 1976. The use of environmental variables in the
interpretation of genotype-environment interaction.
Heredity, 37: 1-7.

Wright, S. 1921. Correlation and causation. J. Agric. Res.,
20: 557-585.

Wright, S. 1934. The method of path coefficients. Ann.
Math. Stat., 5: 161-215.