FIN S:
mnrwmrim

RETUMWS LIBRARY MATERggS:
Place in book return to rulovt
charge from circulation record

4*
1...}: ....1

:.Ié’i“u“>-\

‘1‘ L‘ tam! '

 

 

 

RELATIONSHIP BETWEEN MERISTEMATIC CHARACTERISTICS,
YIELD COMPONENTS AND YIELD OF BARLEY (HORDEUM

VULGARE)

BY

James Benjamin Abaka Whyte

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
MASTER OF SCIENCE

Department of Crop and Soil Sciences

1979

DEDICATED
TO

MY PARENTS

ABSTRACT

RELATIONSHIP BETWEEN MERISTEMATIC CHARACTERISTICS, YIELD
COMPONENTS AND YIELD OF BARLEY (HORDEUM VULGARE)

 

BY

James Benjamin Abaka Whyte

Each organ is sequentially developed and although each
may be affected by environmental stress and different gene
systems, the phenotypic expression of each of the plant
organs is closely related to each other. This relation—
ship is brought about in higher plants by the nature of
the apical meristem because the size of any plant organ
depends on the size of the meristem from which it is
developed.

X969-3 followed a different pathway in development
than 3130, which had a develoPment pattern similar to
that of the other control varieties. x969-3 thus produced
a higher number of seeds per head for its level of X.

This property resulted from an initially broader based
meristem and a time lapse period between the vegetative
stage and the onset of the reproductive stage. Some lines
in the progeny inherited this property, together with
genes for higher x, contributed by B130. This resulted in
the production of a higher number of fertile tillers for

a given head size or vice versa.

Evidence is provided to show that 50 to 60% of the
relative size of x and Y can be determined by examination
of the meristem at Stage 3. The relative growth rate and
width of the meristem at Stage 3 have a positive and
negative relationship, respectively, in predicting the
number of fertile tillers per unit area. Size of the
meristem is most important in predicting the number of
seeds per head. The width of the meristem has a negative
relationship while the length of the meristem establishes
a positive relationship in the prediction of the number

of seeds per head.

ii

ACKNOWLEDGEMENTS

I wish to express my sincere gratitude to Dr. J. E.
Grafius, my major professor, for his encouragement and
guidance throughout the course of this work and during
the preparation of this manuscript.

Special thanks are also expressed to Dr. M. W. Adams
and Dr. R. E. Olien for serving in the guidance committee,
reviewing the manuscript and giving helpful suggestions.
James L. Nelson was particularly helpful in taking the
pictures used in this manuscript and I am very grateful
to him.

I also wish to thank all those who, at one time or
the other, helped me during the study.

I finally like to express my deep appreciation to my
parents and dear friend, Miliswa Sobukwe, for their

understanding and support.

iii

TABLE OF CONTENTS

LIST OF TABLES

LIST OF FIGURES

INTRODUCTION
LITERATURE REVIEW
MATERIALS AND METHODS
RESULTS

DISCUSSION

SUMMARY AND CONCLUSION
APPENDIX

REFERENCES

iv

PAGE
v-vii

viii

19
24
52
57
60
66

LIST OF TABLES

TABLE PAGE

 

1 Mean values for the maximum length (L)
and width (WD) of the primary apical
meristem at three different stages
(SII, SIII, SIV) relative to B130
(standard). 31

2 Mean values for maximum length (L) and
width (WD) of the primary apical meri-
stem at three different stages (SII,
SIII, SIV) relative to B130 (standard)
and the relative growth rate at Stage
III (GR). 34

3 Mean square values for maximum length
(L) and width (WD) of the primary
apical meristem at three different
stages (SII, SIII, SIV) relative to
B130 (standard). 35a

4 Mean values of selected lines with
peculiar meristematic growth pattern
and their parents for maximum length
(L) and width (WD) of the primary
apical meristem at three/four stages
(SII, SIII, SIV, SV) relative to B130
(standard) and the relative growth
rate at Stage III (GR). 36

5 Correlation coefficients of maximum
length (L) and width (WD) at three
different stages of the primary
apical meristem (SII, SIII, SIV)
relative to 8130 (standard) relative
growth rate (GR) at Staqe three and
stem diameter (SD). 42

6 Mean values for the number of fertile
tillers per 30 cm (X), number of seeds
per head (Y), average seed weight (Z),
grain yield per plot (W), number of
seeds per unit area (XY) and stem
diameter (SD). 43

LIST OF TABLES (continued)

 

TABLE

7

10

Correlation coefficients of maximum
length (L) and width (WD) at three
different stages of the primary apical
meristem (SII, SIII, SIV) relative to
B130 (standard), relative growth rate
at Stage III (GR), number of fertile
tillers per 30 cm (X), number of seeds
per head (Y), average seed weight (Z),
number of seeds per unit area (XY),
head size (YZ), stem diameter (SD) and
yield per unit area (W).

Analysis of variance on the multiple
regression of number of tillers per

30 cm (X) and number of seeds per head
(Y), each as a dependent variable on-

the length (L), width (WD) and relative

growth rate (GR) at Stage III.

Multiple regression statistics for
number of fertile tillers per 30 cm
(X) as dependent variable and length
(L), width (WD) and relative growth
rate (GR) at Stage III as independ-
ent variables.

Multiple regression statistics for
number of seeds per head (Y) as
dependent variable and length (L),
width (WD) and relative growth rate
(GR) at Stage III as independent
variables.

vi

PAGE

45

48

49

50

TABLE

Al

A2

A3

A4

A5

A6

PAGE

 

Mean values for head size (YZ), height

(HT) and yield in bushels per acre

(BU/AC) for the selected lines and

varieties, control and parents. 60

Correlation coefficients of stem dia-

meter (SD), number of seeds per head

(Y), head size (YZ), number of seeds

per unit area (XY), number of fertile

tillers per 30 cm (X) and grain yield

per plot (W). 61

Analysis of variance on the multiple
regression of yield (W) as dependent

variable on number of tillers per

30 cm (X), number of seeds per head

(Y) and average seed weight (Z). 62

Multiple regression statistics for

yield per unit area (W) as the de-

pendent variable and the number of

fertile tillers per 30 cm (X),

number of seeds per head (Y) and

average seed weight (Z) as independ-

ent variables. 63

Analysis of variance on the multiple
regression of number of seeds per

unit area (XY) on length (L), width

(WD) and relative growth rates (GR)

at Stage III. 64

Multiple regression statistics for
number of seeds per unit area (XY)
as dependent variable and length
(L), width (WD) and relative growth
rate (GR) at Stage III as independ-

ent variables. 65

vii

FIGURE

LIST OF FIGURES

Structural characteristics of barley
apical meristem at Stage I.

Structural characteristics of barley
apical meristem at Stage II.

Structural characteristics of barley
apical meristem at Stage III.

Structural characteristics of barley
apical meristem at Stage IV.

Graph of maximum length against
maximum width of apical meristem.

Meristematic
68-105-9 and

Meristematic
68-104-3 and

Meristematic
68-103-8 and

characteristics of
68-105-17 at Stage

characteristics of
68-104-19 at Stage

characteristics of
68-105-18 at Stage

lines
III.

lines
III.

lines
III.

Regression of number of seeds per
head on length of meristem at Stage III.

viii

PAGE

 

25

26

27

28

33

37

38

39

46

INTRODUCTION

Agronomists, as well as other agriculturalists, are
already confronted with the problem of providing food for
a world population that continues to grow at an accelerated
rate. Improvement of crop productivity through plant
breeding has been realized by induction and deduction
from empirical data. However, much advancement can
probably be obtained from the application of fundamental
concepts of plant growth and development.

Yield components were described as early as 1923 by
Engledow and Wadham, but the importance of yield compon-
ents as determinants of cereal yield was only fully
recognized when put in geometric context. Geometrically,
yield is expressed as a volume of a rectangular paralle-
piped with its components, number of tillers per unit
area (X), number of kernels per head (Y) and average
kernel weight (Z), as the edges. Yield is subject to
change through change in one or more of its components
and geometrically, the greatest yield change is obtained
with a change in its shortest edge. Heterotic effects
in yield result more from the interaction of yield
components instead of overdominance through loci inter-

action. Hamid and Grafius (1978) show that the earlier

developed organs have a profound influence on later formed
structures. The order of development of the yield
components of barley are number of tillers per unit area,
number of seeds per head and seed weight. Genetic control
of yield, W, is indirectly channeled through its compon-
ents, with the earlier formed structures assuming the
major part of the control.

The primordia of organs evolve from meristems and
the central role of this structure has been pointed out
by Sinnott (1921). He stated that, "The size of any
given organ depends upon the size of the growing point
out of which it has been developed." The interest in
using apical meristems of the shoot of flowering plants
in studies on morphogenesis may be questioned, since
the meristem is small and delicate, practically enclosed
in surrounding tissues and consequently very difficult
to handle experimentally. However, the conversion of
the meristem from the vegetative to the reproductive
condition is among the most dramatic examples of a switch-
ing of a developmental pathway. The vegetative growth
period involves the formation of tillers and leaf
primordia. The change of meristem from the vegetative
to reproductive stage coincides with the cessation of the

formation of tiller buds and leaf. The switching from

vegetative to reproductive phase exerts a direct effect
on the relationships between yield components.

The negative correlations between yield components
have posed a block to yield improvement of crOp plants.
In effect, these negative correlations prevent simultan-
eous maximization of X, Y and Z and thus impose a ceiling
on grain yield. Relaxation of these negative correlations
can result in great yield increases of such crop plants.
Short statured hexaploid wheats, derivatives of Norin 10
cultivar, outyield the standard wheats as a result of
relaxation of the negative correlations between the yield
components. Grafius gt El' (1976) reported the uncoupling
of X and Y in barley so that a higher value of Y for a
given value of X was possible. This characteristic was
carried over into the progeny with resulting increased
yields of unselected progeny.

In 1952, Watson stated that leaf size was the main
determinant of differences in yield of dry weight object-
ives of plant breeding. Thorne (1966) concluded that
grain yield of cereals was related to photosynthetic area
above the flag leaf node. Since grain filling is mainly
comprised of carbon derivatives synthesized during
photosynthesis, it would be logical to expect higher

rates of photosynthesis to result in higher yield. Up

to date, there is very meager evidence to show a direct
association between photosynthetic efficiency and
differences between cultivars in grain yield. On the
other hand, there is a great deal of evidence pointing
to the importance of sinks in increasing yield.

The following study was an attempt to relate the
relaxation of negative correlations between yield compon-

ents with events taking place in the primary meristem.

LITERATURE REVIEW

 

Considerable emphasis is currently being placed in
a number of breeding programs upon the improvement of
plant characteristics with an ultimate increase in grain
yield. The task of constructing higher yielding popula-

tions of barley (Hordeum vulgare L.) by combinations of

 

lines selected for their agronomic characteristics,
disease and insect resistance has been rewarding. Plant
breeders are now faced with the challenge of understanding
the mechanisms involved in the expression of yield to
plan their techniques for breeding higher yielding crops,
especially those that may have already reached a plateau.
In barley, the complex trait, yield (W), has three
components: the number of tillers per unit area (X),
the average number of kernels per head (Y) and the aver-
age kernel weight (Z). This biological phenomenon can
be expressed geometrically as the volume of a rectangular
parallepiped with its components X, Y and Z being the
edges. Yield can be changed by changing one or more of
the components. Grafius (1956, 1964) presented five
theorems to justify the geometric interpretation of
yield using its components.

Varieties which maintain their standing in different

plant communities either resist change or adjust favorably
to changes in environment. This buffering capability is
a result of physiological processes which are ultimately
expressed through their structural yield components. If
changes occur at random among the three dimensions of the
rectangular parallepiped, then the most stable configura-
tion is the cube. However, if changes are not random and
one edge is more resistant to change than the others,
then some other form of configuration might be more
resistant to changes in volume. Any negative change in
one edge is compensated for by a positive change in the
other edges. The greatest change in volume occurs with
changes in the shortest edge.

Removal of epistatic interactions due to the compon-
ents necessitates the use of geometric approach to
express yield. The question of overdominance as a genetic
basis for heterosis has raised some academic discussions.
The questions of multiplicative interaction between
components or edges of the geometric interpretation and
interaction at the locus level need to be solved. With
respect to yield, the effect does not appear to be
interlocus, but interaction between the edges of the
geometric figure.

A simple system was used by Grafius (1964) to show

this effect. "Let the various A1 to An loci interact with
the various Bl to En loci in pairs with summation of
effects between pairs, for example, AlBl + A282 + ... .
Aan where A1 and Bl represent loci, not alleles. As a
contrasting model, let the sum of all the A1 effects
interact with the sum of the B1 effects. It is apparent

that:

"MS
3’
H
"MS
(I!
H
;:

AlBl <

"MU
H

Epistasis in the classic sense shown on the left of
the inequality is less than the geometric interaction
between the gene systems affecting the individual yield
components.

Other forms of interaction are possible since gene
action in complex trait need not be restricted to any
one type. However, major heterotic effects can be
associated with the right hand side of (l).

Duarte and Adams (1963) showed that means of compon-

ents of leaf area in beans (Phaseolus vulgaris); leaflet

 

number and leaflet area, exhibited dominance effects.
However, the leaf area of the F1 exceeded the total for
both parents.

Representing the degree of dominance by a = H/D

where H and D are the non-additive and additive genetic

variances, respectively, they showed that leaflet number
exhibited complete dominance (a = 1) while leaflet size

showed partial dominance (0 < a < l). Heterosis (a > 1)
is observed in the product.

All plants follow a developmental rhythm. Small
grain plants such as oats, barley, wheat and millet 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 plants and are directly related to
the components X, Y and Z.

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
interrelated events which are sequential in time, gene
regulated at critical sites and times and subject to
modifying influences of the environment. The events do
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.

Bonnett (1964) describes morphogenesis as "The

development of the shape and arrangement of the parts of
the plant, the time and sequence of development of the
parts and the histology of the parts as they develop."
It is an epigenetic process; one condition leads to
another and does so in a channeled and controlled fashion.

In a series of papers, Grafius (1969), Grafius and
Thomas (1971) and Thomas gt_§l. (1971 a, b, c) presented
the concept of sequential developmental process of yield
components. The components of yield in the studies were
X, number of heads per unit area; Y, the number of seeds
per head; and Z, the average seed weight. The chronolog-
ical developmental sequence of the components is X to Y
to Z. Yield, W, is a multiplicative product of the
components, i.e. W=XYZ. A transformation technique was
given by Thomas gt El- (1971a) to remove the part of
variation of a component trait which is contributed by
the trait(s) which appears earlier in the development
sequence. Plant organs laid down early in the sequence
exert more genetic control over variation in W than
traits laid down later in the ontogeny.

Rasmusson and Cannel (1970) and Tai (1975) pointed
out that yield components in cereal crops are determined
at different stages in the ontogeny of the plant and

thus are differentially affected by variation in the

10

environment. This suggests that the three yield compon-
ents in cereals are affected by independent environmental
factors during the same or different periods of plant
development.

The formation of yield components in sequence results
in a different relationship between a component trait and
the environment resources. The development of the first
component trait is solely determined by the genetics and
the resources available during the early stage of growth.
A component trait whose development is subsequent to
others is not only influenced by the resources available
during its formation, but by the development and conforma-
tion of its predecessor. The mechanism for controlling
the formation of a yield component is thus increasingly
complicated in the chronological developmental sequence.

Correlations exist between yield components. These
correlations may be due to genetic linkages, pleiotropy
or to physiological developmental relationships. Adams
(1967) established the existence of negative correlations
between yield components of several crop plants,
attributing their occurrences to:

1. Possession of developmental plasticity in plants
which enable them to take alternate pathways to attain

their final adult forms. Variation in one component is

ll

compensated for by variation in another.

2. Developmental induction. Competition of two
plant structures for a common, limited nutrient supply
will tend to favor one structure over the other in the
amount received.

The concepts proposed by Sinnot (1921, 1960) and
Bonnett (1964) on the development of organs in plants as
controlled by developmental allometry needs special
consideration. Each part and function is so closely
related with the rest that the whole plant develops in
an orderly fashion toward the growth of a mature individ-
ual. Adams (1975) points out the phenomenon of size
and numbers as part of the overall allometry in a plant.
He showed the significant relationship between number
of pods per plant with main stem node number and that of

seed size and 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).

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 origin. A

second corollary states that size and numbers are

12

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).

Allometric relationships between X, Y and Z might
result more from competition than from the effects of
common origin. Grafius (1978) called this 'stress matrix'
because a correlation matrix is implied on correlations
between several traits. Since the correlations are
physiological, the stress matrix varies with the environ-
ment and the gene pool. Linkage may be present but its
effect can be reduced on the assumption that genes for
the components are distributed throughout the chromosomes.
Allometric relations between 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 for environmental resources
(Grafius, 1978). Adams (1967) reported a much reduced
correlation between yield components for space planted

versus solid stands for navy beans, Phaseolus vulgaris.

 

Hoen and Andrew (1969) reported near zero correlations
for the yield components in corn. Under closer spacing
and/or higher yields such correlations increased in

intensity (Grafius, 1969).

13

Grafius gt 31, (1976) reported the uncoupling of X
and Y in barley so that a higher value of Y for a given
value of X was possible. The parental materials showed
the uncoupling when Y was graphed against X. Variety
X969-3 proved to be an outlier in that a larger than
expected Y for a given X was observed. This characteristic
was carried over into the progeny with resulting increased
yields of the unselected progeny over the best parent in
one of the backcross populations - after selfing
several generations.

Hamid and Grafius (1978) developed the path coeffi-
cient diagram in conformity with Sinnott's Law and known
developmental relationships. Grafius (1978) updated the
pathway. The importance of the trait set early during
morphogenesis, namely the number of heads/area (X) is
demonstrated.

The plants' reaction to develop any one level of X
triggers a chain reaction affecting all latter formed
organs as shown by Sinnott's Law. By virtue of its
direct association with meristem size, X assumes a pivotal
role in determining sizes of plant organs and eventually
the determination of economic yield itself. This
dependence of X could be modified by external factors

such as nutrients (Aspinall, 1961, 1963), water (Wardlaw,

14

1971), temperature, light intensity and daylength
(Cannel, 1969; Friend, 1965) or internally by hormone
levels (Leopold, 1949). One must recognize also that
the initial control originates in the gene and the
magnitude of the stress matrix for the yield components
has a bearing on the phenotypic expression. It is an
intraplant response evoked by external factors in the
environment.

There is a general concensus that from 80-90% of
the carbohydrates in grain are obtained from C02 fixed
after anthesis. Many workers have stressed the importance
of photosynthesis in the upper leaf area and ear to grain
filling. Such physiological studies are necessary in
identifying the photosynthetic sites involved and their
relative contribution to post anthesis accumulation of
assimilates in the grain. However, the determination of
grain yield (W), specifically the observed differences
among a set of genotypes normally involved in a breeding
program, entails a more complex process and would rarely
be resolved merely by the relationship of photosynthetic
efficiency and W. In the first place, the expression of
the economic yield in cereals is the end product of
three major physiological processes, namely, accumulation

(of assimilates) translocation and storage. Any one

15

process could be limiting, and in the set of data presented
by Hamid and Grafius (1978) the relationship of average
leaf area to differences in grain yield was found to be
nonsignificant. Working with similar materials, Grafius
and Barnard (1976) attempted to relate leaf canopy,
integrated over time, to yield, but found no significant
relationship. Berdahl gt El; (1972) showed no consistent
yield advantage of small over large leaves or vice versa.
Evans and Dustone (1970) and Khan and Tsunoda (1970) have
observed in cereals that higher yielding cultivars had
lower photosynthetic rates.

Size and number of the appropriate components of
yield, W, may be more critical than the size or number of
the photosynthetic surfaces in causing differences in W
genotypes.

As reported by Grafius (1978), Sinnott redirected
allometric science in plants by discarding the then
current practice of trying to correlate the development
of plant organs with the growth and development of the
whole plant. Instead, he showed that if one related the
size of an organ with the size of the primary meristem
from which it arose, many of the ambiguities of the
earlier work disappeared.

Sinnott (1921) showed that the fundamental difference

16

between shapes of gourd Lagenaria vulgaris, Ser is K (the

 

regression coefficient of log length (Y) against width
(X) of the earliest ovary primordia). K equals to .8,
1.2 and 2.2 for bottle gourd, hercules club and snake
gourd, respectively. However, one can have different
shapes due to the length of development time and/or the
growth rate with the same K value.

Fruit shape is simply inherited, making the selection
of a new K value rather easy but an attempt to manipulate
the rate of growth of either width or length independently
is less likely to be productive.

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.

The size of the shoot apex is associated with the
size of leaves; species with smaller apices (e.g. ryegrass
and clover) have narrow leaf primordia and narrower leaves
than those with large apices (e.g. peas and maize)

(Aitken, 1967). He deduced that the association of the
width of the shoot apex may be an important limiting
factor to leaf size and hence, total leaf area. Maltzahn
(1957) detected that the primordia of flowers and leaves

are considerably longer in a large fruited type in a

17

comparative study of size differences in two strains of

Cucurbita pepo.

 

After observing a developmental relationship between
the shoot apex and leaf blade width in maize, Abbe gt_al.
(1941) concluded that it is possible to make a direct
comparison between the size of the shoot apex and the
width of the leaf blade from the earliest stages of
development to leaf 12.

Hybrid vigor has been suggested to operate early in
the embryonic growth period resulting in larger meristems
(Quinby, 1970). The relationship between plant charact-
eristics appear 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 associated
with factors governing the size of the meristem from
which the culm, leaves and glumes have arisen.

Publications dealing with the development of the
barley spike from germination to maturity are rare.
Bonnett (1935) dealt with the development of the barley
spike from the earliest stages to complete differentiation.
Fisher (1973) showed a marked difference between the

morphological development of the spike in short statured

18

hexaploid wheat (Triticum aestivum L.), a derivative of

 

Norin 10 cultivar and the standard hexaploid wheat. In
exploring the origin of the heterotic effect through a
comparative quantitative morphogenetical study of the
sorghum panicle, Blum (1977) realized that heterosis in
the number of grains per basal primary branch could be
traced to two major factors:

1. Larger reproductive apex that allowed development
of larger basal branch primordia.

2. A 4-day lapse between termination of the acro-
petal formation of branches and the onset of the basipetal
formation of spikelets. The time lapse was utilized in
the hybrid for the increase in the basal branch size prior
to spikelet initiation.

The control of the genesis of form in plants is thus
through three major physiological and structural factors;
size of the organ primordia, growth rate of the developing
organ and the physical constraints exerted by adjacent

primordia or organs.

19

MATERIALS AND METHODS

 

The material used was derived from two parental lines
of barley, X969-3 and B130. A straight cross and two
complementary backcrosses were made and grown to the
equivalent F4. Twenty random selections were made in
each progeny and allowed to self to the equivalent F9.

Six lines from each population plus 68-105-15 (recently
released as Bowers) were selected on the basis of contrast-
ing values for their yield components, namely number of
tillers per 30 cm (X), number of seeds per head (Y) and
average seed weight (Z). Two control varieties 60-215-6
and Larker (C.I. 10648) plus the two parents (X969-3 and
8130) were used in the experiment. Meristematic measure-
ments were taken on the selected lines and varieties
(parents and controls).

A lattice square design with four replications was
used. The plots were four-row plots 0.0254 m apart and
2.4 m long, planted at a rate of 35 g per plot. The
study was carried out in Tuscola county, Michigan.
Planting date was April 17, 1978.

Meristems were sampled in the following procedure:
Four predetermined stages were used as markers at which
meristems were to be sampled. The stages include:

1. Vegetative stage.

20

2. Transition stage showed by the appearance of
double ridges on the meristem (Figure I).

3. Reproductive stage characterized by spikelet
differentiation.

4. Elongation and further differentiation.

Owing to inaccurate timing, meristems were not sampled
for Stage I in the main experiment. The pictures of Stage
I in Figure I are from the previous year.

Before any sample was taken at each stage, seedlings
within each genotype were visually selected for similar
morphological characteristics from the outer two rows of
the whole plot. One or two seedlings were uprooted, and
their meristems dissected out to determine the develop-
mental stage. Five seedlings were then harvested from
the outer two rows, starting from the Blst day after
planting. Owing to the large number of seedlings involved,
and the rapidity with which they dehydrate, the portions
of main tillers containing the meristems were preserved
in a solution containing 95% ethyl alcohol, water,
glycerine and formaldehyde in proportions of 52%, 38%,

5% and 5%, respectively.

The main tillers were used because they have a

greater potential for production within a defined and

finite environment than has any other single tiller.

21

They have the principal benefit of the early water and
nutrient uptake by the seminal root system. The early
development of the main tiller gives a much longer inter-
val for the development of ear than in later culms and
they have far longer period to double ridge formation
(Rawson, 1967); and this may be a factor leading to a
greater number of spikelets. Similarly, the main culm
has a longer period in which to initiate florets.

The five plants were selected to represent the
mean of each line or variety. Subsequent sampling was
carried out at four day intervals.

The meristems were dissected out and measurements
taken using a light microscope equipped with a measuring
ocular. Measurements taken include maximum length (L)
and maximum width (WD) of the meristems.

The relative growth rate (GR) at Stage 3 was measured
as follows. A sample of 10 meristems of the same
relative size was obtained from each line. Counts were
made of meristems with characteristics similar to 8130
(used as standard). The number was expressed as a
fraction of the total sample and used as the estimate of
the relative growth rate. In making the readings, special
attention was paid to the stage of development as esti-

mated by growth of awns and to the expansion of the tip

22

 

 

 

 

of the meristem. Meristems in the vegetative stage (sim-
ilar to X969-3) still retain the cylindrical tip as shown
in Figure 3.

Other measurements taken during the plant growth and
at maturity include:

1. The average stem diameter (SD) was computed from
measurements taken at the base of the head. Use was made
of a simple and rapid technique suggested by Evans (1972).
A guage as illustrated above was carved out of a thin
resilient paper, the size of a standard credit card and
graduated markings in millimeters were line along its
inner edges. Rapid readings were obtained by inserting

the particular part of the plant in the guage.

23

2. Estimates of seeds per head (Y) were derived from
a random sample of twenty heads per plot preceding harvest.

3. The average seed weight (Z) was calculated from a
3 gm sample per plot using an electronic seed counter.

4. The number of tillers per 30 cm (X) was obtained
by dividing grain yield per 30 cm of row by the product of
seeds per head and the average seed weight.

5. The central two rows of the plot were harvested
for grain yield (W).

The correlation and regression analyses were performed
using mean values for the selected lines presented in
Tables 2, 6 and 7.

The relative contribution of the three yield compon-
ents, X, Y and Z on yield (W) was estimated using a
multiple regression equation with W as dependent variable
and X, Y and Z, as independent variables. The analysis
is not germaine to the objectives of this study, however,
it is reported in the appendix. The yield components, X,
Y, and XY were each used as a dependent variable while
meristematic measurements at Stage 3 were used as inde-
pendent variables in a series of multiple regression
equations. This was an attempt to break down the yield
components into their subcomponents at the meristematic

level.

24

RESULTS

The stages used as markers for meristematic measure-
ments are shown in Figures 1 through 4. Using B130 as
the standard, the following agronomic characteristics and
developmental processes are typical of the individual
stages.

Stage 1: Nearly all the leaves and leaf initials
that the main stem bears are present at this stage. The
leaves range from those fully differentiated at the base
of the meristem to leaf primordia just distinguishable as
ridges above. Stem development and elongation begin,
thus preparing for spike differentiation.

Stage 2: The first indication of spike differentia-
tion is the appearance of double ridges. The ridges are
shown as bumps on a rather smooth edge of the meristems
(Fig. l). The pair of ridges are nearly equal in size
initially. At a later time, the upper ridge of each
pair grows more rapidly and spikelets are formed from
them. The lower ridge probably becomes the internode of
the rachis (Bonnett, 1935). The ridges are more developed
in X969-3 than 3130.

Stage 3: Spikelet differentiation is indicated.
Several stages of spikelet development are shown on the

same spike. Spikelet initials are prominent in the

25

 

X969-3 B130

Figure I. Structural characteristics of barley apical
meristem at Stage I. Magnification = 80X

d - double ridge

26

xom n coflpmoflwflcmmz

.N mmmum um EmumHHmE HMOHQM mwaumn mo moaumfiuwuomumao Hmuouosuum

.m whomwm

wlmHNlom omam Mlmmmx

 

27

xow n GOAHMUAMHcmmS

.m wmmum um EmDmHHmE Havana wmfiuma mo woaumfluwuomuwso amusuosnum

wlmamlow omam Mlmmmx

.m wusmflm

 

28

 

Now n coaumoflmaommz

v ommum um Emumfiumfi HMOfimm >0HHMQ mo mofiumfluwuomumno Housuosuum

mImHNIow OMHm ml

.v muomflm

mwmx

 

29

central and basal portions while the ridges at the top
of the Spike show only evidences of spikelet different-
iation.

Differentiation of the first structure of the
spikelet, the lemma, takes place. Primordia of other
spikelet parts differentiate while the awn begins its
development as an outgrowth from the lemma. The internodes
of the rachis are very short at this stage. Spikelet
differentiation proceeds towards the apex, but the last
formed spikelets never complete their development. They
remain infertile and rudimentary.

Stage 4: There is further differentiation and
elongation of the spike to form the mature spike. The
degree of elongation determines the spike density
(Bonnett, 1935).

Table 1 gives the mean values for the meristematic
measurements taken on the varieties, parents and control.
Two discernible differences are obvious among the meristems
of the varieties. These include their relative sizes and
their rate of progress from the vegetative to reproductive
stage.

At Stage 2, both the maximum length and width for
X969-3 are largest. B130, Larker and 60-215-6 follow in

a decreasing order. However, at Stage 3, 60-215-6 has

30

the largest meristem followed by Larker, 8130 and X969-3.
The relative sizes of the meristem at Stage 4 follow the
same trend as that of Stage 3, with respect to the four
above mentioned varieties.

In comparing the rate of progress from the vegetative
to reproductive stages, regarding 8130 as possessing the
standard rate of progress within the gene pool, it was
found that X969-3 (low X) had a lower rate of progress
while 60-215-6 (high X) had a higher rate of development.
This is shown in Figure 3.

Determination of the rate of progress also showed that
differentiation and elongation of the awn coincides with
the development of the full set of spikelet initials on
the meristem as shown by 8130 and 60-215-6 (Fig. 3).

Figure 5 shows a graphical presentation of the above
mentioned differences. Two forms of growth pattern are
discernible from the graph of length against width.

1. The control varieties follow the same path
initially as 8130 but differ in their 'take off points'.
The take off point (TOP) is that point in development
when ratio of length to width increases greatly. 8130
and Larker take off from a smaller sized meristem than
60-215-6, the latter having a higher rate of development

before the TOP. The rate of elongation is higher in

31

.m momum an ma omam awn;
mflm>auommmmu .o.m cam m~.m .ma.~ mmmum an mum “wrung can mumawuom .mumomx

 

 

mmv.H mmm.n mam.o mmm.m mnm.o mmH.H mammmq

mmh.a www.ma Hmm.o mam.m 5mm.o hmm.o mlmamlow

mmm.a meo.v «we. mmw.H mmv.o mHN.H mmmx

mam.a hmm.h new. mmm.~ wam.o mmH.H omam

REEVB AEEVA AEEVB AEEVA AEEV3 AEEVA mfimz muucm
>Hm HHHm HHm

 

.Aoumocmuwv omam
ou m>aumamu A>Hm .HHHm .HHmv mmmmum pcmummmgo mans» um Emumhnme Hmoaam
mHmEHHm 038 no AQSV nupflz can any zumcma ESEflme on» How mo5am> com: .H manna

32

60-215-6 after the TOP.

2. X969-3, though it starts with a larger meristem
virtually skips the elongation process until after four
days (Stage 5). This delay results in a greater diameter
meristem which is followed by an increase in the elonga-
tion process to a rate higher than that of 60-215-6.

Other differences observed between the behavior of
X969-3 and the control varieties include the difference
in time to reach Stage 2. X969-3 reached Stage 2 four
days before the other control varieties, but maintained
its vegetative stage shape through Stage 4 before the
elongation process started.

Table 2 gives the mean values for the meristematic
measurements taken on the selected lines. There were
substantial differences in length and width at Stages 3
and 4 among the varieties and lines. These differences
are further shown by the variances in Table 3.

The mean values of selected lines, with a peculiar
apical meristematic growth pattern, and their parents are
given in Table 4. Lines 68-105-17 and 68-105-9 show a
similar growth pattern to parent X969—3 while lines
68-104-3 and 68-104-19 show a similar pattern to 8130.
The others are intermediate. Some start off with the

X969-3 growth pattern and end with the 8130 pattern of

(mm)

LENGTH

33

 

 

 

*
C
20 0 t
I
I I
15
10
5
0
O 5 10 15
WIDTH (mm)

Figure 5.

Maximum length against maximum width of meristem.

34

.m momum aw ma omam cmnz
.>Hm>fiuommmmu .o.m can mm.m .mh.m mommum ca mum meumq can mlmamlom .mlmwmx

 

 

 

 

mo~.H mno.v m o mmm. mm~.m mum. owo.a omumoaumo
mom.a ~mo.m m.o mom. Ham.~ New. N¢~.H maumoaamm
amm.a mmm.e H.o mmn. mam.a mam. Hmm.o sanmoaumm
mva.a mam.~a o.a mam. ~ma.~ Hmw. omH.H oaumoaumm
mno.a mmm.» m.o mmm. na~.~ awe. mam.a maumoaumw
mm~.H mmm.m m.o no». mmo.~ ova. AHH.H mumoaumm
mm>.H mmn.oa G.o hum. mwv.~ mmv. hmH.H mumoanmm
mam.a ham.m m.o Ham. aom.~ now. noH.H omuqoaumm
mum.a oom.oa >.o mam. mum.~ How. mmm.a manvoaumm
vom.H mmn.oa n.o awn. qu.~ new. mam.a mauvoaumm
omm.a nmm.ha m.o mmm. mam.~ use. mmH.H mauvoaumm
mmm.H mam.ma m.o Nam. nmm.~ omw. mm~.H oauvoaumm
nav.a mma.e e.o moo. 4mm.m «mm. mmH.H muvoaumm
mow.a mon.m m.o mam. mm>.m mmv. mmm.a maumoaumm
mmm.a mom.m 8.0 can. hoH.~ awe. omo.a haumoaumm
mam.H m~>.m o.o owe. mmm.a Nae. HFH.H maumoaumm
emm.H moo.m ¢.o man. mnH.~ one. H~N.H mumoaumm
man.a nam.m «.0 4mm. vmo.~ «H4. Hm~.H mumoaumm
ohm.a mon.aa H.o om». mnH.~ «ma. om~.H Humoaumm
leaves fiesta mo reeves fiesta reeves .equ 6882 Nuucm
>H mmmum HHH mmmum HH mmmum

 

.Aoumonmumv omam 0»
m>aumHoH A>Hm .HHHm .HHmv wwmmum ucmuowmap woman an Emumfinma Havana
humaflum on» no AQBV nupfl3 can Adv sumcma ESwamE on» How mosam> and: .m wanna

35

growth, and vice versa.

Generally, rates of development of the 104 group of
lines (X969-3 x 81302) were more related to 8130 (the
recurrent parent) while 103 lines (X969-3 x 8130) and the
105 lines (X969—32 x 8130) showed intermediacy between
8130 and X969-3.

Meristems from selected lines are shown in Figures 6,
7 and 8. Lines 68-105-9 and 68-105-17 show characteristics
similar to X969-3. They have a low number of tillers per
30 cm (X), but produced a higher than expected number of
seeds per head (Y). Their meristems in the vegetative
phase possess the characteristic cylindrical apex. Their
rate of development is low. With a high rate of develop—
ment, lines 68-104-3 and 68-104-19 have a high number of
tillers per 30 cm (X) and produce the expected number of
seeds per head (Y). Their awns are well differentiated
and the meristems have lost their cylindrical apex. Lines
68-103-8 and 68-105-18 produce a low number of tillers and
their expected number of seeds per head. Rate of develop-
ment is intermediate. About 50% of their meristems show
characteristics similar to X969-3 and 50% to 8130 (standard).
Meristem sizes are larger in lines 68-104-3 and 68-104-19
than in lines 68-105-9 and 68-105-17, while lines

68—103-8 and 68-105-18 have intermediate sizes between the

35a

Ho. Wm «k

 

 

 

 

 

moao. NHmm.v mmoo. came. mmoo. mmao. om Houum

««m~ma. *«mHmH.mm *«Hmmo. «aomam. mmoc. «above. mm mwfluoflum>

03 A 03 A Q3 A no mouoom
>Hw HHHm HHm

 

.Apnmocmumv omam
on m>fiumHmu A>Hm .HHHm .HHmv mommum ucmquMMp moms» um Ewumflume Hmowmm

humeflum on» mo AQSV nuofl3 paw Adv sumcma Eofiwxma uom mmsHm> mumoqm and: .m manna

36

.m mmmum 2H ma omam cm£3
.>Hm>fluommmmu .o.m can m~.m .mh.m mommum cw mum umxqu can oImHNIow .mImmmx

 

 

 

ovm. mmm.m mva. mow. FHH. Hmm. Ado. u mvamq
Hma. omm.~ HHH. mvm. mmo. oma. Ame. n mvomq
III III oam.a 5mm.n o.H nub.o mmm.~ mmm.o mma.a omam
omo.m OHH.- mmm.a mvo.¢ o.o ->.o mmm.a mwv.o mH~.H MImmmx
III III onw.a won.HH H.o omh.o obH.~ vm¢.o om~.H HImoHImo
III III mhw.a mmm.m m.o mmm.o th.N vmv.o mnw.a mHImoHIwo
III III «mm.a mmm.m o.o mom.o va.~ mhv.o ~¢~.H mHImoaImm
III III 5mm.a mmo.m v.0 mmh.o mna.m mmv.o HNN.H mImoHImm
III III mnm.a oom.oa h.o mvm.o th.m Hov.o mmm.a mHIgoaImm
III III 5H¢.H mmh.n m.o mom.o vmm.m mmm.o mmH.H mIvoaIwm
mmm.a mmm.o~ vmm.a mma.v H.o mmh.o mvm.a mmm.o Hmm.o hHImoaImm
onm.a oom.om mm~.H mmm.m m.o hwh.o mmo.m ov¢.o hHH.H mImoaImo
Asses fiesta reeves Aequ mo leaves Assaq Asst: Leach msmz muucm
>m >Hm HHHm HHm

.xmov HHH mmmum um mum“ auzoum m>aumamn ma» new Acumocmumv omam

ou m>flumHmH A>m .>Hm .HHHm .HHwV mommum H50m\mmuau um Emumflumfi Havana

wumefium man no AQZV nupfis can Aqv numcoa Edefixme MOM mucmumm “Hosp can
cumuumm nuzoum UHumEmumauoE umfiaoomm nufl3 mmcfla omuomamm mo mmoam> and: .v manna

37

xom n :ofiumoflwficmmz .HH mmmum um
hHImoaImw paw mlmoaImw mmcfla mo mowumfluwuomumno vaumfiwumflnmz

hHImOHImm mlmoalmm

.m mnsmflm

 

38

xom
malvoaImw cam Mlvoalwo mwcfla MO

mHIvOHImm

 

u :oflumoHMMcmmz .m mmmum um
mOAHmflumuomumno UflumeumemE

mlvoaImw

.n gunman

 

39

 

xom u aohumoHMHcmmz .m mmmum um
mHImOHImG Bum mumoaumm mmcaa mo mogumaumbomumao onumEmumaumz

mHImoalmw mlmoalmw

 

.m whomflm

40

two groups above.

Table 5 gives the correlation coefficients between the
meristematic measurements. The length and width at any one
stage are significantly correlated with each other. Sizes
of meristems at Stage 3 are correlated with sizes of
meristems at Stage 4 while a significant correlation exists
between growth rate and size at Stage 3.

The mean values for the yield components (X, Y, Z),
yield (W), number of seeds per unit area (XY) and stem
diameter are given in Table 6. There were significant
differences between the lines and varieties for the various
plant characteristics. The high number of tillers pro-
duced by some of the selected lines resulted from the
inheritance of the gene system for high tillering contri-
buted by the 8130 parent. Hamid and Grafius (1978) have
demonstrated a positive relationship between SD and Y.

The data here support the expected relationship between
SD and size of meristem (Table 5).

There is ample evidence for the relationships between
X, Y, Z and W in the published literature. There is,
however, only fragmentary evidence regarding the relation-
ship of the meristems to the components of yield and the
data will be examined from this standpoint. First and

crucial to the argument, I show that 50 to 60% of the

41

relative size of X and Y can be determined by examination
of the meristem at Stage 3.

Examination of the growing point at Stage 3 reveals
an important phase in the ontogeny of the genotypes. It
marks the end of tiller production and the initiation of
differentiation of florets. An indication of the potential
number of florets per head borne by a genotype is given by
the size of the meristem at this stage.

The time required to reach this stage is influenced
by the variety and the environment in which it is grown.
Different varieties vary in the time required from planting
to maturity, however, most of the differences are in the
early stages of growth. The time from heading to maturity
is the same in all varieties. Excesses or deficiencies
of the necessary environmental conditions may shorten or
lengthen the time period required to reach Stage 3.

With reference to Table 7, length at Stage 3 is
significantly and negatively correlated with number of
seeds per head and stem diameter. It, however, has only
a negative relationship with number of seeds per unit
area (XY) and yield. A genotype with greater length at
Stage 3 will produce a small stem diameter, small number
of seeds per head, etc.

Width at Stage 3 is positively correlated with

42

m0. v m k

 

 

Ho. v m ««

«mm¢.I samwm.l NHH. #MHm.I mma.l «amhm.l Om
va. «mom. «vmv. «ammn. 5mm. hwm. m
«know. 5mm. mmm. aamhm. *«mmo. v93
mwm. kmmm. *hmm. rammm. vmq
«same. vmo. mo3
vmv. wmg. mmq
«#0ou. NQ3

vQ3 vmfl MQ3 mmfl N93 qu

 

.Anmv Hmpoamflp Emum

paw moms» ommum an away mumu nusoum m>wumamn Apumpcmumv omam o» m>Hp
Inflow A>Hm .HHHm .HHmV Emumflume Havana mumawum 038 no mommum ucoquMHp
woman no ADBV nupfls can any numcma EoEfime mo mucmfloflmmmoo cowumaounou

.m magma

I+3

 

mn.~

 

 

:5 8.3: 3.3 cm.~ 58 :o. M town
HH.o mm.oHH hm.mo om.a om.v 0H.~ “mo. v Avomq
vm.H HN.HHm m~.h¢o om.vv mm.Hm mb.¢a oNImoaImw
mm.a m¢.mmm m~.mom oo.~v H¢.Nm m~.va mHImoalmm
ow.H mo.mmaa mh.bam oo.vv mo.mw mo.ha hHImoalmm
vm.a hm.mmoa m~.mv> on.m¢ mm.wm wm.ma mHImoaImm
mv.H hm.mmaa oo.omb om.Hv ~h.mm mh.ha mHImoaImw
cm.a Hm.mmoa om.omh om.m¢ wv.Hm mh.ma mImoaImw
m¢.H m¢.o~oa om.mhw on.av vh.am Hm.mH NImoaImm
ma.a hm.amm mh.wmm om.Hw hh.mm mm.- oNIvoaIwm
Hm.a no.omoa mh.aah om.o¢ Ho.¢m Hm.om malwoaImm
mm.a om.oHoH m~.mbm oo.mv mv.mm oa.na mHIvoHImm
m~.H hm.oam oo.aam oo.mv mm.mm ~m.ma mHIwoaImm
mN.H mo.a~oa mn.non om.m¢ Hm.~m mm.mH oaIonImo
¢H.H nm.onoa om.~mm om.mm mw.~v Ha.mm MIvoaIwm
H¢.H no.mmm mm.omm om.vv mv.qm mm.>H mHImoaImm
vm.H mm.mmoa om.wmh on.av mv.vm mm:mH hHImoalmm
m¢.H ou.mam mm.omo oo.mv mm.om no.ma mHImoaImm
mm.H mm.mmoa oo.m~h oo.vv Hm.mm av.ha mImoHImm
mm.a hm.Hmm mn.m~m om.mm hm.~m mm.mH MImoaImm
mm.H wh.mmoa oo.mmm ov.ow mH.mm om.wa HImoaIwm
mm.a mo.omoa mh.man ov.~v om.Hm ww.o~ mmmmgq
m~.H Hm.mmm mm.wmm o>.m¢ m¢.vm mm.vm mImHNIom
H¢.H om.mhoa oo.am> om.mv Hm.¢w Hm.ma mImomx
mm.H mm.moaa oo.mmm oo.mm mo.om .mw.mH omam
25$ 2:? EB: 32; :3» . 23x 6:82 Face
.Aomv umumamap Emum paw wav mono was: you mammm mo Hogan:
.sz Moan mom pamwa Gamma .ANV unmflmz comm momum>m .va pom: mom mommm
mo Hones: .Axv EU on non mumaaflu mawuumm mo Hones: man now mmoam> and: o magma

44

average seed weight and negatively correlated with the
number of tillers per 30 cm though not at high significant
levels.

The relative growth rate to Stage 3 is negatively
and significantly correlated with number of seeds per
head, stem diameter and head size, but positively corre-
lated with number of tillers per 30 cm. A genotype
with a relatively higher growth rate establishes a larger
number of tillers, however, stem diameter, head size and
number of seeds per head are reduced. The retardation
in growth rate allows for a bigger head size formation.
This is a natural phenomenon. Figure 9 shows the
regression of number of seeds per head on length at Stage
3. The regression is significant (P i .05).

The regression of the third stage measurements of
width, length and rate of development on dependent variables,
number of tillers per 30 cm (X) and number of seeds per
head are given in Table 8. The mean square values for
the regressions are highly significant (P i.'01) and the
coefficient of determination (R2 = .5196 and .6501 for X
and Y, respectively) indicate that the variance in the
dependent variables can be partially accounted for by the
variations in the three independent variables (L3, WD3,

GR). Tables 9 and 10 provide the statistics for the

45

Ho. v m *

 

 

 

mo. Ma .2.
vmo. a«mm¢.| **~>¢.I mvo. mmo.I «amHm.I «How. m
mH~.I ««vom.I Hmo. omo.I «mm.I mmo.I omo.I v93
mm~.I «amwm.I hmm. «mH.I «mN.I mH~.I mmo. vmq
moo.I «Ha. wbH.I NmH.I «vow. mma. Hmm.I mnz
mmm.I ««mam.I mvm. mom.I omo.I ««Hmv.I omH. mmq
>h~.I mmH.I moa.I nn~.I boo. omo. mh~.I mnz
Hom.I ««mhm.I mom. HhH.I H>~.I hmm.I NmH. mmq
3 am N» wx N w x

.ASV noun pan: Add onH» can Aomv Hmumfimac Emum

.ANMV mNHm 0mm: .wav mono was: mom mpmom mo Hones: .ANV unmwo3 comm

mommm>m .va pow: mom moowm mo Hogans .Axv EU on mom mumaawu wafluumm mo

Hosea: .Amwv HHH mmmum um mung nusoum m>flumamu .Aonmocmumv omam ou m>fiu

Imamu A>Hm .HHHm .HHmV Emumflnoe Havana mumfifium on» NO mommum quHoMMHp
owns» an Aazv zuow3 pom any numcoa Esaflxma mo muchOflmmmoo :oHumHmHHoo .h wanna

46

.HHH mmmum um Emumflume

mo naocma :0 com: Mom mammm mo quEsc mo coflmmmummm .m musmflm

n 03$ 3 REEL 595.. 23:55:22

 

- ma n~ _~ ._
_m .~ .~
.5
.— Um
w
0.
m
m.
.m
S
9
9
p
S
.. Mm
B
D.
.a

 

47

regression.

Aside from the degree of explained variances, there
are several other highly interesting observations. In the
case of number of seeds per head (Y), the b values are
negative, -15.29 and -6.52, for length and relative growth
rate, respectively, but positive 53.31, for width. The
effects of width and length are signficant (P i.'01) while
GR has no significant major effect in predicting Y. The
R2 delete values for WD, L and GR are 0.2951, 0.4638 and
0.5949, respectively. This shows that width is most import-
ant in predicting Y followed by length. Size of meristem
is thus most important in the prediction of number of
seeds per head.

The partial regression coefficient for width is
significant (P i .01) and negative (-22.35) in the predict-
ion of number of tillers. Relative growth rate and length
have a positive contribution to X, however, only the b

2 delete values

value for GR is significant (P i .05). R
are 0.1887, 0.3610 and 0.5024 for WD, GR and L, respect-
ively. Length of meristem is least important in determining
tiller production. Rate of development of width is more
important in tiller production. The higher the rate of

development of the meristem, the bigger the X value it

will have, however, width of meristem will be small.

48

8. Na 3.

 

 

Homm.o wmam.o mm
mnmm.ma mvmm.m ma Houum
«rmmmm.maa «aomwm.ha m coflmmonmmm
m x up wousom

mmumoqm com:

 

.HHH mmmum um Amwv oumn nu3oum m>HumHmu can Ao3v nupflz .Aqv numcma mnu co
manmflum> ucwpcwmwp m mm comm .ANV 6mm: Hmm mpmmm mo Hmnfida paw Axv EU on
mom mumaaflu mo Hones: mo coflmmmummu mHmHuHoE may no mochaum> mo mamaflwcd .m manna

49

 

 

 

oaom.o mmm¢.o Nvo.o hvmm.o m
hmma.o wmmm.oI moo.o Navn.OI mm:
vmom.o mmma.o mnv.o mmam.o mmq
mmumamo ucwfloflmmmoo Hm>wq
mm cowumHmHHoo Hmfluumm unmoHMHcmflm usmflmz mumm manmwnm>
.mmHQMwum> ucmpcommpcfl mm HHH mmmum um Amov mumn
nusoum m>flumamu new Ange sagas .Anv aumcma cam manmflum> ucmoammmc mm Axe
E0 om mom mumaaflu oawuumm mo “mason mom mowumflumum GOAmmwummH mamflpasz .m manna

50

 

 

mvmm.o mmwm.OI mvH.o even.oI m
Hmmm.o mmon.o Hoo.o hhmh.o ma:
mmmv.o emmm.oI mHo.o Hmmh.oI ma
mmumamo ucmHOAMMmoo Ho>mq

mm newuwamunoo Hafiuumm unmofimacmflm unmflmz muwm manmflum>

 

.mmHQ8flum> unmocwmmpcfl mm HHH mmmum um Amwv
ouch nusoum m>flumamu can Anzv suofla .AAV numcma can wanmaum> unopcmmmp
mm va town you mpmmm mo Hones: Mom mowumwumum c0flmmmumwu mamfluaoz

.OH OHQMB

51

Number of seeds borne will be small since width has a
positive significant relationship with Y.

This is expected in nature because a relatively low
growth rate allows for a low number of tillers. This
relaxed growth rate encourages the formation of organs
(i.e. meristems) with larger width. Since the width
determines the length that the organ assumes, a meristem
with a larger surface area is produced. A larger number
of floral initials are borne resulting in the production
of a larger number of seeds per head. This is in conform-
ity with the fact that sizes and numbers of plant organs
are negatively correlated (Grafius, 1978).

The development of plant organs in terms of their
gross size and number is closely interrelated. This
relationship exists even though plants organs are laid
down sequentially and each may be affected by different
modes of environmental stress. X has a negative influence
on stem diameter, number of seeds per head and average
seed weight. This influence is exerted through the
establishment of a meristem size at Stage 3. The rate
of establishment is genetically controlled and has a
direct relationship with the number of tillers coded for
by the plant. All other traits regarding size and numbers
are then evolved toward the attainment of the ultimate

character, grain yield.

52

DISCUSSION

 

Yield is determined by the number of fertile tillers
per unit area, number of seeds per head and average seed
weight.

Tiller production is one of the first developmental
processes at the organ level. It has a far reaching
effect on the growth and development of organs laid down
later in the plant's ontogeny. Generally, one finds a
negative correlation between number of tillers and number
of seeds per head and this correlation is largely
physiologic.

The morphological development of the primary culm of
barley from germination to pollination can be divided into
three phases as opposed to the two proposed by Bonnett
(1935). The phases can be determined approximately by
examining the stem and more accurately by examining the
apical meristem. During the first phase, the stem
internodes do not elongate, leaves and leaf initials
differentiate from the growing point, which remains smooth
in outline and increases in length. The apical meristem
increases in size during the second developmental phase,
double ridges appear, spikelet structures and tiller

initials differentiate. During the third phase, the

53

internodes of the stem elongate and further differentiation
of spikes occur to complete their develOpment in prepara-
tion for pollination. Development of the other fertile
tillers follow in rapid succession when the main stem
passes into the third phase.

The very early differences in the time of differen-
tiation and rate of spikelet development are reflected in
the mature plant characteristics. Although there were no
varietal differences in growth until Stage 2, differential
growth from just before and after Stage 3 became obvious.

There was a 4—day time lapse, for the transformation
of the apical meristem from the appearance of double
ridges to the onset of spikelet differentiation, between
X969-3 and the other control varieties. The progeny
which have a similar growth pattern to X969-3 possess this
pr0perty.

Spikelet differentiation is retarded in X969-3.

This retardation allows for a larger sized meristem forma-
tion. An extra surface area is provided for the
development of an additional number of seeds per tiller

by the variety. Similarly, Lee 33 31. (1974) and Williams
(1975) are of the opinion that a delayed and larger basal
branch at the time of spikelet initiation allows for the

formation of more spikelets, florets and grains in sorghum.

54

The interesting thing here is that although the develop-
ment of the meristem is delayed in X969-3 and in some of
the progeny, these plants are not necessarily later in
heading. Somehow or other the difference in time is made
up. Those with larger X tended to have smaller diameter
meristems as well as a faster rate of development. The
characteristic size of the meristem at Stage 3 is depend-
ent on the number of tillers coded for by the genotype.
Rate of development of the meristem at this stage is a
function of meristem size. The larger the meristem the
higher the rate of elongation. However, owing to the
4-day time lag X969-3 resulting in the larger sized
meristem, its rate of elongation is greater than that of
the control variety with the highest tillering within
this gene pool.

There is reason to suspect that the larger reproduct-
ive apex at T.O.P. in X969-3 traces back to a larger
vegetative apex. If so, then the gain in number of seeds
per head (Y), may be established in the first developmental
phase with all the physiological and practical implications.

This property possessed by variety X969-3 is under
direct genetic control because it is carried over into
its progeny. Grafius gt 31. (1976) reported the uncoupling

of X and Y in the same variety and showed that this

55

characteristic was carried over into the progeny with
resulting increased yield of the unselected progeny over
the best parent.

The production of a large number of seeds per head
can be traced to a large meristem size while tiller pro-
duction involves the relative growth rate of the meristem
and the width of the meristem. The greater the growth
rate, the larger the number of tillers produced. However,
the width of the meristem is small. Number of seeds
per head is thus a function of the number of fertile
tillers formed.

Plant development is programmed in its heredity which
interacts with the environment. This necessitates the
formation of the various morphological structures in some
integrated form and the control of the balance between
plant characteristics involving sizes and numbers is
manifested through the growth of the apical meristem.

The number of high yielding lines, originating from
the straight cross and the backcross to X969-3 parental
line, obtained from the few original randomly selected
populations is intriguing. Line 68-105-15 has recently
been released as Bowers because of its high yielding
potential in trials at many locations in Michigan. There

appears to be good reason to suggest that the meristematic

56

properties, instrumental in determing X and Y, are under

the control of a relatively few genes.

57

SUMMARY AND CONCLUSION

 

Four varieties with varying values in their yield
components, but having comparatively similar levels of
grain yield, in addition to 19 lines from the straight
and two complementary backcrosses between two of the
varieties, 8130 and X969—3, were used in the experiment.

The development of meristems between Stages 2 and
3 show a switch in the developmental pathway as shown by
the relative sizes of the meristems at the two stages.
Varieties with relatively large meristem sizes at Stage 2
developed relatively small sized meristems at Stage 3.

8130 followed the same developmental pathway as the
other control varieties initially, but differed in its
take off point. Its rate of development before the take
off point was intermediate between 60-215-6 (higher
tillering) and X969-3 (lower tillering), a property,
which is characteristic of the number of fertile tillers
it bears. Some lines produced high number of tillers
through the inheritance of the gene system for high tiller-
ing contributed by the 8130 parent. Their head sizes were
however, small.

X969-3 followed a different pathway in development

than the other varieties resulting in the production of

58

a higher number of seeds per head for its level of X. This
property results from an initially broader based meristem
and a time-lapse period between the vegetative stage and
the onset of the reproductive stage. Some lines in the
progeny inherited this property, together with genes for
higher X which resulted in a higher tiller number for a
given head size. Apparently, whenever these two occurred
together, we get more fertile tillers for a given head
size than would have been expected. In other words the
negative correlation between X and Y is somewhat relaxed.
The results also showed that variation in X and Y can
be significantly accounted for by the variation in size
and relative growth rate at Stage 3. Variation in width
at Stage 3 and the relative growth are most important in
predicting the number of fertile tillers per unit area while
width and length at Stage 3 are important in explaining the
variation in the number of seeds per head. The relative
growth rate has a positive effect in the prediction of the
number of fertile tillers per unit area. This results in the
formation of small diameter meristems with production of in-
creasing number of fertile tillers accounting for the nega-
tive relationship between the diameter of the meristem
and number of fertile tillers per unit area. Width

maintains the negative relationship in predicting

59

the number of seeds per head. Its significant positive
correlation with the length of the meristem establishes
a positive predictive value between the meristem length
and the number of seeds per head. With relaxed growth
rate, a high number of seeds per head is produced for
the number of fertile tillers borne.

From the number of lines with similar meristematic
characteristics as X969-3, it was proposed that meristem-

atic properties are controlled by few genes.

APPENDIX

60

 

 

-.HH hm.a m~.o Ado. w.mvoma
mw.m ~m.a mH.o Ame. v meomq
mm.mm m~.mm we.m omumoHImm
oa.mm om.mm mm.m maumoaumm
em.HHH oo.¢m Hm.~ eHImoHImm
mm.aoa om.mm Hm.m GHImoHImm
om.moa oo.mm am.~ maumoaumm
m~.mm m>.¢m mm.~ aumoaumm
me.~m oo.em em.m mImoaImm
«4.5e m~.em oG.H omeoaImm
mo.nm m>.mm Hm.m mHIvoHImm
Hm.~a om.mm om.~ mauaoanmm
em.mm om.mm ma.~ mauqoanmm
wm.mm oo.mm m~.~ oauvoaumm
mo.mm oo.ov oa.H muvoanmo
ma.mm oo.mm mv.m maumoanmm
am.mm ma.mm mm.~ Adamoaumm
em.mm m>.Hq No.~ maumoHImm
v¢.mm oo.a¢ mm.~ mumoaumo
mh.mm m~.em me.~ mumoaumm
He.om oo.mm am.m HumoHImm
vo.mm m~.mm ma.m mammaq
mm.Hm om.m~ om.H oumamuoo
He.am oo.mm ma.~ mummmx
am.mm mm.mm wm.~ omam
xo<\sm Agave: Aeneas mamz wanna

 

.mucmnmm can Houucoo .mmwumflnm> 6cm mmcfla wouumamm on» How AU¢\Dmv whom
mom mamnmon CH pHmH» can Aemv Macao: .Auwv onwm pom: How mwcam> com: .H4 manna

61

 

 

 

vmm. ««mmm. wma. Hmm. baa. mam. 3
«ovm. «cmam.l Hmm.l *«OHm.I «kmmm.l x
hmo.I N¢N.I mmo. wao. xx
«mmv. ««hmm. aaamm. Nw
HmH. omv. N
«avmh. %
x %x NM N N Om
.ABV uon mom pamflm cflmum can Axv EU om mom muwaafiu wafluumm mo
umnEdc .wav mono pas: mom mcmmm mo Hones: .ANMV omen com: .va pom:
you wpmwm mo Hones: .Aomv Hmuoamflp Emum mo mpcwHOAMMwoo coflpmHmuuoo .m¢ dance

62

HO. Ham «:1

 

 

Homh.vva ma Houum
«ammmm.ahawa m cofimmmnmom
mumzmm com: up mousom

 

.ANV usmflm3 comm mmmnm>m can AMV cow: mom
mowmm mo Hanan: .Axv EU om nod mamaawu mo Hogan: :0 manmfium> “coccmmmc
mm sz came» mo coflmmwummu mamfluflase on» no moccanm> mo mflmhamcd .m¢ manna

63

 

 

mono. momm.

 

mooo.ov wbmm. N
momm. mmom. mooo.ov mmm¢.H w
Homo. mmbm. mooo.ov omHN.H x
mmumamp ucmfloflmmmoo Hm>mq
mm coflucamunoo Acapumm ucm0flmwcoflm munmwmz oumm moanmwnm>

 

.mwanmflum> unoccmmmccfi
mm ANV uanmS comm wmmum>m can va cows mom mummm mo Hones:
.Axv EU om mom mnmaaflu mafluumm mo umnfidc map can mannanm> “accommmc

on» no ABC comm uflco mom camwm How moaumwumum cowmmoummu mamfluaoz .v¢ manna

64

 

mmmv.mo¢m ma uouum
hamv.aaom m coflmmmummm

 

mumsqm can: we mouoom

 

.HHH mmmum um
Amwv mmumu £u3onm 0>fiumeH can AQBV cuofl3 .AAV numcma co wav comm pass
mom mpmmm Mo “0355: m0 coflmmmnmmu mamwuaoe mnu co moccaum> mo mammamcg .md magma

65

 

 

ammo.o mmav.o hmo.o wamm.o m
mmvm.o mmoo.ol mam.o mmoo.ol mas
momo.o omvv.OI Hho.o mmNh.OI ma
mwpmamp mucmwowmmmoo mam>wq

mm GONHMHmHHou HMfluumm unmoflmwcmwm musmflw3 cumm mmHnmaHm>

 

.mmHQMHHm> unoccmmmocfl mm HHH mmmuw um Amuv much
nuzoum m>flumawn can Anzv nucw3 .AAV numcma can manmflum> unwpcmmmc mm
wav mono vac: mom mommm mo Hogans How mofiumaumum c0flmmwummu mamauazz

.m¢ OHQMB

10.

ll.

66

REFERENCES

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.
Amer. J. Bot. 28:778-782.

Adams, M. W. 1967. Basis of yield component compensa-
tion in crop plants with special reference to field 2
bean, Phaseolus vulgaris. Crop Sci. 7:505-510. l

 

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.

 

--—l(. _ -..-

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

$5...-4—‘a a. x
«

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.

Berdahl, J. D., D. C. Rasmusson and D. N. Moss. 1972.
Effect of leaf area on photosynthetic rate, light
penetration and grain yield in barley. Cr0p Sci.
12:177-180.

Blum, A. 1977. Basis of heterosis in the different-
iating sorghum panicle. Crop Sci. 17:880-882.

Bonnett, O. T. 1935. The development of the barley
spike. J. Agr. Res. 51(5):451-457.

Bonnett, O. T. 1964. Morphogenesis in the organism
and the expression of complex traits. Crop Sci.
4:500-503.

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

Duarte, R. and M. W. Adams. 1963. Component interact-
ion in relation to the expression of a complex
trait in field bean cross. Crop Sci. 3:185-186.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

67

Engledow, F. L. and S. M. Wadham. 1923. Investiga-
tions on yield in the cereals. I. J. Agr. Sci.
Camb. 13:390-439.

Evans, A. C. 1973. The quantitative analysis of
plant growth. Univ. of Calif. Press, Berkeley
and Los Angeles, 734 p.

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

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

Fowler, C. W. and D. C. Rasmusson. 1969. Leaf area
relationships 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.

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

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

Grafius, J. E. 1969. Stress: A necessary ingredient
in 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. (In press).

Grafius, J. E. and J. Barnard. 1976. The leaf canopy
as related to yield in barley, Hordeum vulgare L.
Agron. J. 68(2):398-402.

 

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

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

68

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.

 

Hamid, Z. A. and J. E. Grafius. 1978. Developmental
allometry and its implication to grain yield in
barley. CrOp Sci. 18:83-86.

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

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.

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

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

Maltzahn, K. 8. von. 1957. A study of size differ-
ences in two strains of Cucurbita pepo 1. Cross
size differences. Can. J. Bot. 809-830 pp.

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

Rasmusson, D. C. and R. Q. Cannell. 1970. Selections
for grain yield and components of yield in barley.
Crop Sci. 10:51-54.

Rawson, H. M. 1967. Competition within the wheat
plant and the plasticity of response. Thesis,
University of Adelaide.

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

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

36.

37.

38.

39.

40.

41.

42.

43.

69

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

Thomas, R. L., J. E. Grafius and S. K. Hahn. 1971a.
Genetic analysis of correlated sequential charact-
ers. Heredity 26:177-188.

Thomas, R. L., J. E. Grafius and S. K. Hahn. 1971b.
Transformation of sequential quantitative charact-
ers. Heredity. 26:189-193.

'u—“'

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

. _-_r___._ .... .A-__.....__..__
w .
2 ’ . -.A .

Thorne, A. N. 1966. Physiological aspects of grain
yield in cereals. IN The growth of cereals and i'
grasses. Butterworths London 88-105 pp.

Wardlaw, 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.

Watson, D. J. 1952. The physiological basis of
variation in yield. Adv. Agron. 4:101-145.

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

"I1.1111111111111171