1. CLiist“: L‘izr‘lLZ‘swl NF Mai) H'SE 341335515 C-F ”LIN!"

' ERIN”! QEFFEEE-Hsiiéifi ES" 1.111.111: WL GARE (‘éiiLj E395?

D31% %1 Na Leewir"

me

 

Hrczc

  
 
 

Wag-'11 1- 4- . . k1,}.
L 1 E1 1:
"ll EFF-lg}. “(i 31L; "rt.

 

This is to certify that the
thesis entitled 1
C913" DEVICE} LICT RU) 'I‘FE it'JATU‘JE OF PLAT"? BEEF“?
i.11‘Fz?;-;«1;ze I}? 1 ”1.1110514. " ‘1 WILL.) Hz” :5":

presented by
Dirk F. Leeuwrik

has been accepted towards fulfillment
of the requirements for

1‘ g o
Jnlln— degree in _CT0 [.5 v0 1 en ce

(71/, C
K“; ‘ " @pum/

2

   

 

 

Major professor

Date ikz%vd\t‘JA-‘v’1 % L, / /;é Q)

J

0-169

 

 

. .N . 1
”uniiw

 

 

 

 

 

 

 

 

 

,‘ftm "
:. h Mo clam;

 

Maud
P
'- “11 number 9'

v
I
‘ l

.'
f._:tbu cm 'a 1-

‘- ' Mmocm 1"

hfiflh’J n1.
“urar'u‘.
Diffv'
1953211. 1" I
d tum-mm.

Dittye v-

 

 

-\s

.- - “I. “I reap-mo 1' '11

 

 

gym

?’//'-(_. -

ABSTRACT 1

CULH DEVELOPMENT AND THE NATURE OF HEIGHT DIFFERENCES

IN TRITICUM VULGARE (VILL.) HOST

 

by Dirk M. Leeuwrik

The nature of plant height differences was investigated
in three closely related, homozygous lines of wheat. Dataare
presented on total culm length, internode number and length,
cell number and length, and the effect of environment on
these characters. Development of the main culm and individual
internodes is described for the three varieties. Embryo
weights and dimensions of growing points and young seedlings were
determined.

Differences in culm length were generally due to internode
length, both within and between varieties. The contribution
of internode number was negligible.

Differences in length of comparable internodes of Genesee
and 1328 were due to cell number, whereas in 1518 and 1328 cell
size was responsible. Both were involved in the difference

between Genesee and 1518.

 

 

Dirk M. Leeuwrik

Height differences based on cell number appear to originate
during embryogenesis while those based on cell length arise
after germination. Thus, two independent developmental pro-
cesses are involved.

Although the varieties differed slightly in their
growth rates, the actual duration of extension growth appeared
to be the more important contributing factor to length
differences at maturity.

Compensatory mechanisms were evident at both cellular and
component levels. The variability encountered at these levels
was greater than that of the complex trait.

A strong positive relationship existed between variance
and height, especially in the tall variety, and there was
some evidence of differential homeostasis of homozygous
parents.

The greater variation and the compensation encountered
at the component and cellular levels lead to the conclusion
that the genetic study of height is not aided by the independent
study of its components. 0n the other hand, the strong associa-
tion of variability and height and the indication of differential
homeostasis probably due to the compensatory mechanisms at
component levels, tend to obscure genetic segregation d? the

complex trait.

 

 

CULK mm AND m NATURE OF PLANI‘ HEIGHT DIFFERENCES

IN TRITICUM VUIGARE (VILL.) HOST

33/

«‘IL
(1,,

Dirk quLeeuwrik

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR 0F PHIIDSOPHY

Department of Crop Science

1966

 

ACKNOWLEDGEMENTS

The author wishes to express his sincere appreciation to
Dr. E. H. Everson for his help and guidance throughout this
study.

Thanks are due also to Dr. C. Harrison and Dr. F.
Elliott for their advice on aspects of the study and assistance
in preparation of the manuscript; to Dr. 3. Dexter for help-
ful discussions and invaluable "bookkeeping"; and to Dr. D.
Marksrian for his critical appraisal of the manuscript.

The assistance of Dr. L. Mericle and Dr. J. Bukovac is
gratefully acknowledged.

The author also wishes to acknowledge the many hours
spent by his wife assisting with the many chores associated
with the study.

Appreciation is expressed to the Victorian Department
of Agriculture, Australia for granting the leave of absence

which made this experience possible.

11

 

TABLE OF CONTENTS

MRODUCT ION I I C l O O O O O I O O O C O O O O I O I C C 1
REVIEW OF LITERATURE. . . . . . . . . . . . . . . . . . . 2
MATERIAIS AND METHODS I I O I o O 0 I o I o 0 I o o I I o 13

RESULTS AND DISCUSSION. . . . . . . . . . . . . . . . . 22
Height Differences at Maturity
Components of Height
(a) Internode Number
(b) Internode Length
(1) Cell Number
(ii) Cell Length
The Effect of Environment on Height
The Effect of Environment on Components of Height
(a) Internode Length and Number
(b) Cell Length and Number
Culm Development
(1) Before Germination
(2) Germination to Three-Leaf Stage
(3) Three-Leaf Stage to Period of Rapid Elongation
(h) Period of Rapid Elongation Through Maximum Extension
Homeostatic Systems in Plant Height Expression

SWY o o s I I o o t I I o o o I I I I I I O O I s o I 58

LHEIRATIIRE CEED- 0 I I I o I I a o o I I I I O I I I o 0 6o

 

 

  

10.

ll.

 

Thble

 

LIST OF TABLES

 

variation in mean length (inches) of main culms of
Genesee, 1518 and 1328 with different environment .

Analysis of variance of internodal pattern of main
culls of Genesee, 1518 and 1328 under different
environmental conditions . . . . . . . . . . .
Analysis of variance of cell numbers in internodes
of plants of Genesee, 1518 and 1328 in Figure 2 (a).

Analysis of variance of cell numbers in internodes
of plants of Genesee, 1518 and 1328 in Figure 2 (b).

Analysis of variance of cell lengths in internodes
of plants of Genesee, 1518 and 1328 in Figure 2 (a).

Analysis of variance of cell lengths in internodes
of plants of Genesee, 1518 and 1328 in Figure 2 (b).

Analysis of variance of separate internodes to
remove the effect of internodes and variety x
internode interaction. . . . . . . . . . . . . . . .

Mean internode number and length and total culm
length of plants used for estimation of cell
lengths and numbers presented in Figure 2 (a).
Plants grown in flats. . . . . . . . . . . . . . . .

Fresh- and dry-weights of embryos excised from
kernels, in the late dough stage, of Genesee,
1518 and 1328. (gms. x 100,000). . . . . . . . . . .

Weight, volume and specific gravity of kernels
of Genesee, 1518 and 1328. . . . . . . . . . . . . .

Mean measurements in inches made of Various gross
dimensions of projected images of ten 2h-hour-old
seedlings of Genesee, 1518 and 1328, as indicated

in Figure 1+. I O O I D C O O I O O O C C 0 C 0 O 0

Means (cm.) and variances of eulm length of Genesee
and 1328 and the hybrid Genesee x 1328 . . . . . . .

Regression of variance on height in Genesee
and 1328 O O I O O O O O 0 O O O O I O O O O O O I 0

iv

Page

22

25

3O

30

32

32

33

33

38

39

hl

”m “'1.

 

  
 

 

"HIV W“

 

LIST OF FIGURES

Figure

Internode patterns of the main culms of

I 6909886, 1518 and 1328 I e I e e I e a e e e I

Cell length and number in four upper inter-
nodes of Genesee, 1518 and 1328. . . . . . . .

Diagramatic representation of a median
longisection of a 2h-hour-old seedling indica-
ting the various gross measurements made as
listed in Table 11 . . . . . . . . . . . . . .

Tracings of longisections of growing points
of Genesee and 1328 at the beginning of rapid
elonga tion 0 0 O I O O I O 0 O C O O O O O 0 0

Relative growth rates of main culms of
Genesee, 1518 and 1328 in growth chamber, as
expressed by total length. Plants were grown
in flats in peatmoss pots. . . . . . . . . . .

Approximate pattern of internode elongation of
Genesee, 1518 and 1328. Growth chamber 1963 .

Cell number of the peduncle of Genesee, 1518
and 1328 during development. . . . . . . . . .

23

29

AG

#3

#6

#8

51

 

 

  

MODUCTIOH _ ii '

Studies of plant height date back to the beginning of genetics

eels science; lendel used plant height as one of the heritable

traits in studies with peas. Cereals in general, and wheat in

particular, have received a fair share of the genetic studies of

plant height, but the picture has not been resolved completely.
Plant height is a quantitatively inherited character which,

although strongly influenced by environment, generally displays

a high heritability and is easily and simply measured. For this

reason it has been a popular subject for genetic studies designed ‘

to reconcile the behavior of quantitative traits with Mendelian ‘

principles. Plant height has been a useful trait for studies of 9 g

environment-genotype interactions, the relative contributions of .-.£

component characters to a "complex” trait and the genetic

behavior of such characters. There is a trend at present to

identify components of a quantitative trait biologically in

order to simplify genetic explanations. For size differences

those explanations must ultimately lie in the heredity of cell

 

size, number, shape and organization.
The data presented here come from a series of investigations
made to (a) more fully understand the nature of heritable height -

differences in Triticum vulgare (Vill.) Host, (b) to determine

   
  
 

 

where, when and how these differences first express themselves
in the development of the wheat plant and (c) to identify logical
components which could be studied separately in order to

elucidate the inheritance of height.

  

 

 

 
 
 

REVIEW OF LITERAIURE

nonmetyladonous stems have been studied for over a century.

 

lany of the older papers deal with detailed anatomy and nor-
pholegy with the main emphasis on systematic and taxonomic
implications. Duval, in 1869, described the anatomy of the stems, .3
leaves and rhizomes of Gramineae (l3). Guilland (31) published [
the comparative and developmental anatomy of some monocotyles
donous stems, not including those of grasses.

van Tieghem (63) distinguished between slow- and fast

growing stems, and terminal and intercalary growth. Ee identi- ;
fied the latter with fast growing stems and explained that inter- ..‘
nodes elongate while nodes do not. Drawing on some of Sachs' .#1
data (h9) he pointed out that the pattern of elongation of inter-' ‘ L?!

nodes was the same as for the entire stems. Athanassoff (13)
pointed out that the grass stem is a flowering stalk and he
therefore preferred the term culm.

At the turn of the century studies were conducted to relate
various culm characteristics to resistance to lodging. Papers
on this aspect have been reviewed by Percival (#6) and
Athanassoff (l3). Attention was mainly directed to vascular

organization and nodal anatomy, which seemed to be a natural out-

 

growth of earlier work by Andersson (10) and others.

   
  
  

Percival (#6) and Athanassoff (13) published the first com-
prehensive studies pertaining to stem development in wheat.

Percival described the cereal culm as erect, elastic, cylindrical,

 

 

 

furrowed, smooth or scabrid, the upper parts being green and
hence assimilatory. The majority of culms had six nodes but
the occurrence of five or seven was not infrequent. The
length of individual internodes increased from base to apex
with little exception.

In stem development, Percival explained, the hypocotyl
remained short and the second internode of the epicotyl
lifted the crown with terminal bud to just below the soil sur-
face. Hhen the crown reached its position the rhizome ceased
growth, internodes remained short and buds developed in the
exile of leaves .

When the plant entered the reproductive stage i.e. when
inflorescences originated at the apex of the branches, lateral
bud formation ceased. Elongation of the young stem was
effected by the apical meristem but also by the intercalary
meristems at the base of each internode, which continued to
lengthen each internode after the upper part of it had
ceased to grow.

Percival briefly described the anatomy of the culm and
presented some estimates of cell size. Although he indicated
the great variation in cell size which occurred, no attempt
was made to relate these to gross dimensions.

Athanassoff (13) has given an excellent and thorough

account of the growth and development of individual internodes

and culms of Triticum turgidum in terms of morphology,

 

 

 

 

 

 

anatomy and histology. Many of his general observations cone
firm those made by Percival. In addition he gave detailed
descriptions of the anatomical and histological changes
along the developing internode and stem.

The researches of Prat (h?) (#8) indicated that the
relationship between the internodes of some cereal and grass
culna can be expressed mathematically by the formula
I = Auk, I being the length of internode, n its rank and A
and k constants. Prat called this the biometrical gradation
of the internodes and showed that this relationship holds
fairly well for g. vulgare. There appeared to be a definite
association with maturity as, in general, the biometric
gradation increased when the life cycle of the plant became
shorter.

Probably the most significant contribution of Prat's
studies has been his demonstration of the effect of the
developing spike on culm development and his subsequent
development of a theory of the regulatory effect of histo-
chemical gradients in morphogenetic processes.

Johnson (37) investigated the effect of light intensity,
photoperiod and location on the length and internodal pattern
of five winter wheat varieties of different heights. All
varieties normally produced five above-ground, extended
internodes, irrespective of location or relative height.

Moisture stress reduced internode lengths.

 

 

l".

 

In a later paper Johnson (38) reported on culn morphology
and development in three varieties of winter wheat of different
heights. He concluded that differences in culm length were
determined by internode length rather than number. Differences
in internode length between varieties became progressively
‘greater in higher and longer internodes. Three internodes
elongated simultaneously according to a definite pattern of
sequential development, with the peduncle being slightly out
of phase with the others.

Whereas Johnson found that the major elongation of the

inflorescence of winter wheat was concurrent with that of

 

the third internode, Bennett (17), in cats, found that the
major increase in size of the panicle occurred during the
elongation of the peduncle.

Similar minor differences in development occur in early
stages of growth. For instance, the work of Avery (1h) and
McCall (1+2) show that the first expansive internodal growth
occurs in the first internode of oats, the second internode
in maize and in the third internode in wheat.

Kaufman (39) has stated that in rice, as in wheat, the
expansion of internodes occurs during the transition and re-
productive phase and occurs acropetally. Thus stem elongation
is slow during the vegetative stage but extremely rapid during

the reproductive phase. Rapid extension of relatively short

   
 

duration is the dominant feature of shoot development during

the latter stage.

 

 

 

Dimensions of plants and plant organs are the result of a
complexity of cell size, shape, number and organization
interrelationships. Sinnott ( 51+) (55) has written that
shape is primarily determined by the relative frequencies of
cell division in various planes and ultimate control of form
therefore resides in individual cells.

The question of whether an organism is the final result
of the behavior of an organized population of individual cells
or whether the cell as a unit is subordinate to the organism
has occupied research workers in biology for a long time.

Sachs (50) in 1893 discussed plant and organ size in relation

to cellular organization. He speculated as to the significance
of the apparent restriction of cell size to a specific amount of
nuclear material and cytoplasm and originated the concept of
"energids." Amelung (8) followed this with quantitative data
on mean cell size of different organs in various plants. He
tried to relate cell size to organ size and was surprised to
find the constancy of cell size in organs of different size

and purpose.

Sierp (53) summarized his research in the following manner:
(1) Cell size in any particular tissue of a plant species is
not constant. (2) Environmental influence on cell size is
great. (3) Despite this, cell size has, for any tissue, a
characteristic average value, which is heritable, (h) A

difference in plant- or organ size can be due to any of the

 

 

 

 

following: (a) reduction in cell size, (b) simultaneous
reduction in cell size and cell number, (c) reduction in
cell number and increase in cell size.

Brotherton and Bartlett (18), comparing etiolated-

with normal internodes of Phaseolus multiflorus, found

 

that the greater length of the etiolated internodes was due
to both cell elongation (66%) and cell division (3h%). Since
cell extension was decreased by light the latter represented
a very disturbing influence in inheritance studies of size,
because ultimately, any quantitative biological variation
must be due to variation in cell size, cell number or both.

Abbe ( 3) attempted to find the histological reason
for dwarfism in Zea mays. The main difference in length from
the normal types appeared to be due to differences in length I
of the lower internodes. Houghtaling (31+) presented data I
on cell size and number in four varieties of tomato and some [
F2 populations of crosses between them. She found that I
quantitative differences in cortical and pith tissues were
mainly due to cell number and concluded that cell number
was a heritable character.

That the picture is not always a simple one is shown by
Bindloss (15), who found that in a certain dwarf tomato the
cell size was actually greater than in normal varieties. She

found that tall varieties had a proportionally greater

 

 

neristel and were able to produce more cells. The cells in
the dwarf types, however, ceased dividing earlier and elongated
for a longer period. The trelationship of cell size and cell
number to stem length shifted throughout the ontogeny of the
plant and was genetically controlled.

Sharnan (52) has given a good account of leaf initiation
and internode formation during early ontogeny in maize.
Periclinal cell divisions proceed laterally from the point
of initiation of the leaf primordium around both sides of the
apical meristem until the margins nearly meet. Differential
rates and planes of cell division within the collar which is
thus formed produce a domelike structure which encloses the
shoot apex. In the lower half of the original collar, cells
begin to divide anticlinally forming a rib meristem with
which the first growth of the internode is initiated.

Abbe et al (1) put develoPmental aspects of the maize
shoot on a more quantitative basis. They demonstrated that
the shoot apex must attain a certain size before differentia-
tion of parts begins. For each general size increase of the
shoot apex with successive plastochrons there is a minimum and
a maximum size. Cell size approximates constancy throughout
growth of the apex and thus cell number increases in each
successive stage in a parallel fashion to external dimensions.

Nilson et al (ks) reporting on the relation of parenehyma

and epidermal cell length to internode length in winter wheat

‘

V _ ‘1 {$35!

 

demonstrated that in both cell types length differed
significantly between internodes. Whereas internode length
increased from base to apex of the plant, cell length changed
in a reverse manner. There was a direct relationship between
cell number and internode length. Epidermal and parenchyma
cell numbers increased in total number with increasing length
of eulm.

Cell length and number are delimited by differential
rates and duration of cell division and elongation, which were
studied by Kaufman et al (ho) in Azggg‘ggtizg. Concentrating
mainly on the development of epidermal tissue of the internode
next to the peduncle (P-l) they described intercalary meristem
activity, cell morphology, localization of "cytokinetic
activity" and differentiation pattern during internode elonga-
tion. They confirmed the findings of Athanassoff (l3) and
Prat (1.8) that growth and cellular differentiation are
basipetal.

Kaufman et al (ho) also showed that cell division is not
subJect to any rhythms but occurs continuously in the base
and peripheral portions of the internode, particularly in the

lower portionsof epidermis and outer ground-parenchyma.

Analysis of change and rates of change in cell number and cell

length in relation to stages of morphological development
showed that cell division and cell elongation are over-

lapping processes during early development.

‘

 

 

10

Ebiko (25) used plant height in Triticum as a growth
criterion in studying the inheritance of growth curves. His
study showed that growth rates are heritable and could lead
to genetically controlled differential height expression.

Attempts to simplify the analysis of inheritance of a
complex trait such as plant height by studying basic growth
rates, or by breaking it down into separate components, are
frequently useful because of the complexity of gene action.
Genetically, developmental processes may be independent but
at the phenotypic level they may interact (3o) (32). Sinnott
(SH) and Duarte (2h) have presented evidence of gene action
with a geometrically cumulative effect, i.e. their phenotypic
expressions are multiplicative in action.

The inheritance of plant height has been shown to
resolve itself into two parts: the fairly simply inherited
characteristics of dwarf vs. tall and the more complex,
continuously inherited height differences of a more subtle
nature.

The former has been shown to be mainly under digenic
(22) (26) (29) (61) or trisenic (27) (33) (35) (60) (6*)
control; moreover dwarf types have frequently been shown to

be unstable cytologically (29) (60) (65).

 

 

ll

Investigations into the inheritance of the more subtle
height differences have revealed heterosis (28) partial
dominance (20) (62) or no dominance (21) in the Fl,trans-
greasive segregation in the F2 and greater variability in
the F2 and F3 generations than in the parents (6), (19)

(20) (28) (#3) (57) (58) (62). It was generally concluded
that the mode of inheritance of plant height is Mendelian
but that the number of factors involved is large. Torrie

(62) established small, significant relationships between
plant height, straw strength and earliness in two crosses.

Carlson (19), in a number of crosses involving the
Japanese semidwarf,variety Norin 10, made the following
observations: (a) Since transgressive segregation occurred
in crosses of Norin 10 with taller varieties, Norin 10
must have some "plus" genes for height. (b) Some or all
the factors for height are linked and linkage is mainly in
the coupling phase. (c) Some or all the factors show dom-
inance, mainly in the direction of tallness. (d) Dominance
and environmental variance should not greatly impede selec-
tion for height i.e. heritability is high.

According to the results obtained by Allan et al (6)
the heritability of culm length in four winter wheat crosses
was high (85 to 92%) and the genetic factors involved, at

least in some of the crosses, few.

 

' x

 

 

 

Mgrkle. and Atkins (#3) also obtained high heritability
estimates of plant height in wheat. They showed that
changes in plant height are positively correlated with
changes in length of the third and fourth internodes.

Lebsock (kl) transferred genes for shortness from a
hexaploid Norin 10 derivative to tetraploid durum wheat

types. Short strawed progeny from the crosses stabilized

.7

at twenty eight chromosomes, but were apparently more sensi-
tive to environment with respect to reaching optimum yields.

F2 monosomic analyses carried out by Alan and ngel (5)
indicated that at least eleven chromosomes influenced culm
length in winter wheat hybrids incorporating Norin 10. No
definite location of dwarfing genes was arrived at; however,
monosomic analyses by Hurd and MoGinnis (35) and Hermsen (33)
indicated that in hexaploid wheat a locus for dwarfing occurs
on chromosomes 2A (XIII), ha (VIII) and 2D (xx).

The studies which will be reported here transgress the
fields of morphology, anatomy, histology and genetics of a
major economic crOp plant. The combined literature is con-
siderable and for this reason it has been impossible to pre-
sent an exhaustive literature review relevant to the problem.
Emphasis therefore has been placed on the selection of the

most important papers in each area.

 

 

MATERIALS AND METEDDS

The winter wheat variety Genesee and two homozygous
selections of the pedigree (Norin 10 x (Brewer x Yorkwin))
Geneseea, which will be designated 1518 and 1328, were used
in these studies.

Although of similar genetic background, under most
environmental regimes Genesee was tall, 1518 intermediate
and 1328 short.

Seeds of the three varieties were sown individually in
small peatmoss cups arranged in groups of seventy in wooden
flats, allowed to germinate in the greenhouse and then placed
in a vernalization chamber for eight weeks at hSOF. During
this period the young seedlings were provided with constant
low-intensity light from incandescent lamps. After vernali-
zation the seedlings, with peatmoss cups, were planted singly
in pots of varying diameter or left to develop and mature in
the flats. Whether in soil or in sandculture, plants received
applications of a complete nutrient solution (modified
Hoagland) twice a week. Plant numbers varied according to
space available and limitations of handling individual plants.
For investigations carried out in the growth chamber a cycle
of 16 hours light at 70°F. and 8 hours dark at 55—60%. was
found to be most satisfactory.

Statistical analysis of the results was carried out as
appropriate and statistically significant differences were
established using either t or F test. In the discussions to

follow plant height and culm length are taken to be synonymous.

13

 

 

1h

The studies were initiated with observations in the three
varieties on the gross morphology of the culms, height differ-
ences of mature culms and variations in height and culm
morphology with subtle environmental changes.

Small populations of the three varieties were grown
under the following conditions:

(a) Plants were grown in sandculture in the growth
chamber; to simulate conditions of closer spacing and increased
competition for rootspace they were planted simply in pots of
h inch, 3% inch and 3 inch diameter. Varieties were in a com-
pletely random arrangement.

(b) Plants were grown in soil in pots of h inch diameter
in the growth chamber until a few days before heading (early
boot stage), then transferred to a greenhouse kept at a rela-
tively high temperature to induce quick, sudden ripening. This
experiment was planned to observe, at the gross morphological
level, the effect of a short ripening period on the develop-
ment and final length of the peduncle, which under most con-
ditions contributes to the total length of the culm as
much as all other, lower internodes combined.

(c) Seedlings were kept in the vernalization chamber at
hsoF. for six months and then transferred to a greenhouse kept
at a relatively high temperature. The plants were left in
flats and, although they were watered every day, were not

provided with any nutrients.

 

 

 

15

(d) Plants were grown in flats under constant lighting at
a constant temperature of 55°F. At the appearance of the first
heads the plants were removed to a greenhouse kept at a higher
temperature.

(1) One half of each population was watered regularly
with complete nutrient solution.

(ii) The remaining half of each population was only
provided with water sufficient to keep the plants from wilting.

Experiments (c) and (d) were designed to accelerate
developmental processes within the plants and to reduce
actual growth in size to a minimum.

At maturity each plant was removed from its growing medium,
roots washed clean and trimmed back to the crown and all the
leaves and tillers carefully removed from the main culm. The
number and length of internodes of the main culm were recorded
for each plant. Where the internodes were short and difficult
to determine the culms were split longitudinally to facilitate
accurate measurement. Mean culm length and mean internode number
and length were calculated with their respective variances. The
observations of individual experiments were combined in an analysis
of variance in an effort to estimate the effect of envirqmment
on the internode pattern and total length of calms.

Having established some of the gross morphological differ-

ences in the mature culms of the three varieties under slightly

 

‘ and 2h-hour old seedlings. Differences between embryos were

16

different environmental conditions, the development of the culm
was studied next to relate differences in mature plant height
to differential varietal development. Large numbers of plants
of each variety were grown in flats in the growth chamber.
Five plants from each variety were randomly removed and sacri-
‘fieed at regular intervals to determine the developing inter-
node pattern and total culm length at various stages throughout
development. The plants were removed from the soil, the reots
washed and trimmed and the leaves and developing tiller buds
removed as far as possible. The remaining axis was then split
longitudinally to expose the developing main culm. All |
macroscopically discernible, extending internodes were counted
and measured and the height of the growing point determined. I
Growth curves Qf the culm and of individual internodes were ;
constructed from these measurements.

To ascertain if differences in size exist before any

appreciable growth has occurred, observations were made on embryos

established on a weight basis. Kernels were harvested in the
late dough stage from the heads of ten selected plants from

the field. A spikelet was removed from the middle of each spike,
care being taken to ensure that the spikelet came from the same

position on each spike. The two largest kernels were removed

 

 

1'?

from each spikelet and the embryos excised and weighed immediately.
The average weight of the two embryos was then calculated.

Embryos could be dissected from kernels in the late dough stage
with a minimum of damage and with far greater ease than from
mature, dry kernels. Microscopic examination revealed that the
embryos had fully differentiated and little or no difference in
size could be detected between embryos dissected from these
kernels and those dissected from mature imbibed kernels. The
embryos dissected from kernels in the late dough stage still

contained a high percentage of moisture and, owing to the very

small weight of individual embryos, moisture loss tended to

.increase the error of the fresh weight determinations. Reasonably

accurate results were obtained, however, when each embryo was

weighed immediately after dissection.

Observations were repeated using dryweight determinations. ;
Upper and lower spikelets were removed from selected heads leaving .
the eight central spikelets. Embryos were removed from the two :
largest kernels of each of these remaining spikelets, bulked i
in ten groups of 100 embryos for each variety and dried for
two days at approximately 106°C. before weighing. Average weights
and variances were calculated for the three varieties on both dry-
and fresh-weight basis.

To establish differences in size before the onset of extension
growth, 2h-hour old seedlings were removed from germinating kernels

and fixed in a solution of alcohol, water and formalin in the

 

 

 

/ 18

proportion of 60: h0:k. They were dehydrated in ethyl alcohol

and chloroform and embedded in parafin. Median longitudinal

 

sections of 10 micron thickness were cut and triple stained with

safranin, crystal violet and orange G. Images of sections of

ten seedlings from each variety were proJected on a screen with a

microproJector and measurements were made of various gross
dimensions.

The investigations thus far were designed to establish
(a) the relative contribution of internode number and length
to differences in mature culm length, (b) the influence of
environment on internode pattern and hence total length, (c)
differences in development of culms and individual internodes
of the three varieties leading to differences in length and
(d) the stage of development where size differences related
to height differences at maturity first become apparent.

Next, the relationship of cell length and cell number to
internode length and culm length was established for the
three varieties. Cell size and cell number were estimated in
plants grown in sandculture in h" pots and in soil in flats.
Here again the objective was not so much to study the effect
of a specific environmental factor on cell size and number as to
study the manner in which these two characteristics changed in
a slightly varying environment. Approximately 2 weeks after

general anthesis i.e. when plants had attained their final

 

 

 

19

mature height but were still green, four plants of each variety
were selected from the plant population in the flats and 6 plants
each from the sandculture. The main culms were carefully re-
moved from the plants and stripped of their leaves. After

the length of the internodes had been measured, a section 2cm.
long was removed from the middle of each internode for estima-
tion of cell size. Cell number was then calculated. Although
cell length along the mature internode varies slightly the cell
lengths in the middle of the internode give a reasonably accurate
estimate of the mean cell length for that internode. The tech-
nique that was first used to determine cell length in the
collected samples was to split the sections longitudinally,
clear the tissues in a mixture of phenolic and chloral hydrate
crystals 1 : l and then storing and mounting the cleared sec;‘
tions in lactic acid (17). Although the clearing technique was
excellent for studying cellular and vascular organization within
tissues, it was found to be faster to measure epidermal cells
by stripping off small segments of epidermal tissue from the
internodes and mounting them in a drop of aceto-carmine.
Eventually most observations on cell size were made this way.
According to Nilson et a1 (hS) parenchyma cells show less Varia-
tion and are more constant in demonstrating the changes in cell
size and number than epidermal and other cells. However, in the

above studies it was found that there was quite a variation

 

 

between successive layers of parenchymal tissue of the stem
and that it was difficult to select comparable files of cells
in the sample sections without bias. Epidermal tissue proved
to be quite satisfactory and far easier to observe. Moreover,
in developmental studies this tissue is far superior for de-
tailed observation of cell division, elongation, maturation and
differentiation. Epidermal long cells, three rows removed
from stomatal cell files were used for measurement. Ten
samples of 3 cells in succession, selected at random along the
‘2 cm. long portion were used to calculate the mean cell length
and hence the mean cell number pfr internode. Because of
variability of lower internodes only h upper internodes were
used for comparative studies.

Having established the relationship of cell length and cell
number to internode length and total culm length in the three
varieties, the next step was to study the histological
development of the internodes to relate final differences in
internode length to differences in rates and duration of cell
division and elongation. Populations of plants were grown in
the growth chamber in flats and samples of three plants removed
at intervals throughout development. The plants were split
longitudinally and the developing culm dissected out and stored

in Newcomer's solution (hh). The number and length of the

 

 

 

21

developing internodes was determined and small portions of the
vurious internodes mounted in aceto-carmine and studied micro-
scopically to ascertain cell division and elongation patterns
and cell lengths and numbers in the various regions of the
developing internode. The number of sample regions of each
internode that were studied varied and depended on what was
considered to be a reasonable and adequate sample to give a true
histological picture of the developing internode.

To conclude this study on plant height (or stem length)
an experiment was conducted to elucidate some of the factors
involved in the genetical control of plant height. Crosses
were made between 1328 (short) and Genesee (tall) to produce
populations of reciprocal F1, F2 and backcross plants. These
populations, together with those of the parents were grown in
the growth chamber in 3% inch pots. The plants were randomly
distributed on six tables and to equalize environmental condi-
tions as far as possible the tables were rotated every 2 days in
a clockwise direction and each table was rotated through 180
degrees, until the plants reached maturity. Observations made
on the populations included internode length, internode number

and total length of the main culm.

 

RESUHIS AND DISCUSSION

Height Differences at Maturitz

Data on mean culm length, mean internode length and inter-

node pattern have been summarized in Table l and Figure l

(a - h). Statistical analysis of these data is presented in

Table 2.

Averaged over all environments Genesee was tall, 1518
intermediate and 1328 short. The environmental influence on
height expression was far greater than that of the varietal
effect.

Table 1. Variation in mean length (inches) of main culms of
Genesee, 1518 and 1328 with different environment.

 

Environment Genesee 1 18 l 28 Mean
Short ripening 31.09 27.93 23.59 27.5h
period
h inch pots 27.30 25.57 22.58 25.15
3% inch pots 19.h5 17.70 16.1w 17.85
3 inch pots 21.12 19.95 17.1.0 19.1.9
55°F. High 18.95 19.25 16.611 18.28
nutr.
cont.
Low 15.07 13.16 10.1w 12.90
light. nutr.
6 months 15.31 13.1w 11.55 13.1+2
vernalization
Mean 21.18 19.56 16.95

 

22

 

 

 

 

 

 

 

 

 

23
Internode length
in inches
15 f
.I
I I
if
u
10 . - b) 4;
I
.I.“/
_I
_I
.I
.1
I" J
5 v 5 . I."
:L/
"4",!"
.—"/—
o F
c) d)
10. z
5 . /
_ _ij’#( Internode no.
figfi9” according to
,«M”' rank
0 l A A I J J l l J L
5 u 3 2 l O S u 3 2 1
Figure l. Internode patterns of the main culms of Genesee, 1518 and

1328.

a) Average of seven environments.
b) u.o inch pots. Sand culture.
c) 3.5 inch pots. Sand culture.
d) 3.0 inch pots. Sand culture

 

Internode length 2a
in inches
15 [

 

 

 

 

s) W h)

7

 

 

 

 

5 .
fl- ac. '
.r’llwv’”fiffi Internode no.
_#»j: .... ’" according to
o I. .......... ‘ L . I . rank .
5 Q 3 2 l 5 U 3 2 1

Figure 1 cont. Internode patterns of the main culms of Genesee, 1518 and 1328.
e) u.o inch pots. Soil. Short ripening period.
f) Seedlings kept 6 months at uSOF.
g) Continual light. Constant 55°F. High water and nutrition.
h) Continual light. Constant 55°F. Lew water and nutrition.

5;.

 

Imam r

 

 

 

Thble 2. Analysis of variance of internodal pattern of main

culms of Genesee, 1518 and 1328 under different
environmental conditions.

source 8.5. D.F. M.S. r.

Internodes 559.3867 1+ 139.866 600.208“

varieties 10.7029 2 5.351h 22.97**

Environment 111 .ohoo 6 18 . 5066 79 Jam

I x E 151.0%78 2% 6.2936 27.01***

I x V 0.785h 8 0.0981 NS.

v x E 1.5967 12 0.1330 us

I x v x E 11.1853 1+8 0.2330

Total 816.7th 101+ ‘

 

 

Components of Height

The length of the culm may be regarded as the product of
internode number and internode length. Each internode in turn
is a product of cell number and cell length. These component

characters were studied separately to determine their contri-

‘bution to culm length.

(a) Internode Number
The first three to four internodes formed during the vege-
tative stage of growth did not expand. These internodes were hard
to distinguish and did not contribute to the length of the
culm so they were not included in any of the observations made.
As the number of extended internodes depends on the lowest
internode to expand first, internodes have been numbered accord-
ing to rank from the top of the plant down, the peduncle being
designated internode l.

The number of extended internodes varied from four to six,

 

26

between and within varieties, but averaged five under most
environmental conditions. Some differential variation in the
internode number was observed. Genesee tended to have six
extended internodes more frequently than the two shorter
varieties and, similarly, the short variety 1328 frequently had
a proportion of plants with four extended internodes.

The statement that culm length differences were due to
differences in internode length therefore cannot be made
without qualification. Internode number did contribute to
culm length differences to a greater or lesser extent, depend-
ing on environment. As a rule, however, its contribution was
negligible.

(b) Internode Length

The length of the five extended internodes in the main
culm increased progressively from the base to the apex of the
plant, the uppermost internode or peduncle frequently being
almost as long as all the other internodes combined (Fig. 1).
This pattern did not vary with variety and appears to be
characteristic of grasses and cereals in general (13) (38)
(h6) (h8). Table 2 indicates that the variety x internode inter-
action was not significant, suggesting that the internodal
pattern is the same in all varieties. In all cases, the

‘variation due to internodes within plants was somewhat greater

than that due to plants indicating a certain amount of compensation

between internodes.

 

 

1'"

 

27

In the tall variety Genesee, the lower extended internodes
contributed proportionally more to total stem length than in the
shorter varieties. Averaging the internode measurements of
plants in the seven environments studied (Figure l (a)),

62.8% of the total stem length of Genesee was due to the upper
internodes whereas the three lower internodes contributed
37.21:. For 1518 and 1328 comparable figures were 6h.l1: and
35.9%, 69.1% and 30.9%.

Proportionally the difference between comparable inter-
nodes of the three varieties tended to increase down the culm.
However, as the internode lengths decreased sharply the contri-
bution to the difference in absolute terms was not so great.
The proportional increase in internode difference was not
evident under all conditions.

Maturity differences were very slight, but were observed
in a good many cases. Earliness was always correlated with
reduced height. The peduncle contributed progressively more
to total length in the shorter varieties. In 2; vulgare the
peduncle is frequently out of phase with the other internodes
especially in short and early varieties.

A definite association with maturity was observed by
Prat (k8) in both cereals and other grasses. He was able to
demonstrate that, as the plant became earlier in maturity,

the lover extending internodes became proportionally less and

less important in the length of the culm. In the very short lived

 

 

28

ephemeral types of grasses the cuhm is made up almost entirely
of peduncle.

Internode length contributes far more to plant height diff-
erences than internode number. Differences in total length of
culm between varieties are due to differences in length of all
individual internodes (Figure 1). Changes in internode length
must be due to alteration in cell number, cell size or both.

The evidence obtained from the observation of the three closely
related varieties indicated that indeed both are involved and,
moreover, that they are independent.

(1) Cell Number

Within each variety, cell number per internode increased
from the lower to upper internodes. Thus, the greater length
of each successive internode was due to a greater number of
cells. This was not a simple relationship because cell size
decreased in progressively higher internodes and so cell number
increased almost exponentially.

Figure 2 (a) and (b) shows the distribution of cell
length and number in the four upper internodes of each variety,
under two different environmental regimes. The great variability
which generally occurs is clearly demonstrated when these results

are compared.

 

Cell
length

uuo

320

200

 

nuyher

 

29

Gen.
1518
1328

 

Cell
number

1200

Figure 2 (a). Cell length and
number in four upper inter-
nodes of Genesee, 1518
and 1328 plants grown in
soil in l 3/u inch square
peatmoss pots in flats.

600

Internode rank from
O apex to base.

 

 

Cell
length

“00 ’

300 b

200

 

length

 

 

 

Cell
number

2000

Figure 2 (b). Cell length
and number in four upper
internodes of Genesee,
1518 and 1328 plants prown
in sandculture in u inch
pots.

‘ 1000

Internode rank from
apex to base.
0

 

 

30

Statistical analyses of cell number are presented in Tables
3 and h. Cell numbers are significantly different between
varieties and internodes. Because of the fairly great varia-
bility between plants the interaction varieties x internodes
only Just reaches significancef (Table h). The interaction is
due to the fact that the cell number of the Genesee and 1518
peduncle decreases more rapidly than that of 1328 (Figure 2
(b)). Although differences existed between all three varieties,
Genesee had a considerably greater cell number than 1518 or 1328.

Table 3. Analysis of variance of cell numbers in internodes of
plants of Genesee, 1518 and 1328 in Figure 2 (a).

 

 

 

 

Source S.S. D.F. M.S. F
varieties 230h8.0h17 2 1152A.0208 NS
Internodes 613h28h.2292 3 20hu761.u097 **
v x I 75125.9583 6 ,12520.9930 NS
subtotal 6232u58.2292 11

Within 361652.7500 36 100h5.9000
Total 659h110.9792 #7

Table h. Analysis of variance of cell numbers in internodes of
plants of Genesee, 1518 and 1328 in Figure 2 (b).

 

 

Source S.S. D.F. M.S. F
Varieties 629168.uh 2 31u58u.22 19.37**
Internodes 23h35365.0u 3 7811788.3h h81.18***
v x I ,91h07.3h 6 1623h.55 2.5h*
Subtotal 2h1619h0.82 11

Eithin 383291.17 60 6388.18

 

Total 2h5h523l.99 71

31

(11) Cell Length_

The change in cell length between internodes is not a
gradual one. Above each node cell length changes abruptly
but stays fairly constant within each internode, although at
maturity the lower cells of each internode are slightly larger
than those towards the top.

Cell length is inversely related to internode length
within each variety. Cell length decreased almost linearly
from the lowest internode up, except for the peduncle. In
this internode there was a sharply decreased reduction.in cell
size in all varieties (Figure 2 (a) and (b)). It is interesting to
compare this with the data obtained by Athenassoff (13) which
showed that cell length in.T:_turgidum was directly proportion-
al to internode length, except in the peduncle in which cells
were considerably smaller.

Each internode is a fairly distinct entity, the elements
of which are of a constant size at maturity. As a rule its
length, cell size and cell number are intermediate to its
neighboring internodes, but a great deal of variation occurs
and the remarkably constant eXpression of final height appears
to be the result of a compensating relationship among its com-
ponents. Statistical analysis of cell length data confirmed
that Genesee and 1328 have similar average cell length while

the average cell of 1518 was significantly longer (Tables 5

and 6).

32

Table 5. Analysis of variance of cell lengths in internodes
of plants of Genesee, 1518 and 1328 in Figure 2(a).

 

 

 

Source S.S. D.F. M.S. F
Varieties 167.3721 2 83.6860 3.u2*
Internodes 2526.hh02 3 8h2.1h67 3h,h3**
v x I 26.3183 6 n.3863

Subtotal 2720.1306 11 2h7.28h6

Within 1001.0055 _36 27.8057

Total 3721.1361 h7

 

Table 6. Analysis of variance of cell lengths in internodes
of plants of Genesee, 1518 and 1328 in Figure 2(b).

 

 

 

Source S.S. D.F. M.S. F
varieties 6655.66 2 3327.83 1.23 NS
Internodes 3u1316.69 3 113772.23 u5.21***
v x I 15098.98 6 2516.h9 49.33**
subtotal 363071-33 11

within 16177.29 60 269.62

Total 3792h8.62 71

 

In the analysis shown in Table 6, the great combined
effect of internodes and interaction variety x internodes ob-
scured the true differences between varieties. When the results
were reanalyzed with the effects of internodes and variety x
internodes removed (Table 7) only in the fourth internode did

the difference not reach significance.

33

 

 

 

 

Table 7. Analysis of variance of separate internodes to remove
the effect of internodes and variety x internode
interaction.

Source S.S. D.F. M.S. F.
Internode l Varieties 6226.5h 2 3113.27 36.76***
Within 1269.35 15 8h.62

Internode 2 Varieties 8739.9u 2 h369.97 23.20***
Within 2825221 15 188.35

Internode 3 Varieties 2901.63 2 1h50.8l 5.79*
Within 3758.2h l5 250.5h 4_

Internode h Varieties 3886.53 2 19u3.26 3.50NS
Within 832h.h5 15 55h.96

 

The results in Figure 2 (a) were obtained under conditions

of growth that produced internode lengths more variable than

normally encountered under more favorable conditions. 1518,

the intermediate variety, actually grew taller than Genesee. The
particular environment also affected the expression of internode
numbers (Table 8). It appears that in this instance the difference
in height between Genesee and 1328 was not only due to internode
length but also to the fact that 1328 consistently formed only

four internodes. Plants of Genesee had at least five internodes.

 

_Table 8. Mean internode number and length and total culm length
of plants used for estimation of cell lengths and numbers
presented in Figure 2 (8). Plants were in flats.

Variety Internode Total Internode
l 2 3 h 5 6 Length Number
Genesee 26.0 12.9 8.2 7.6 3.3 0.6 58.6 5.25
1518 26.5 13.8 10.6 8.6 h.1 --- 63.6 h.50
1328 2209 1100 8.]- ho? -"'- -"" M608 #000

 

3h

Height differences observed were not due to thennfliiplicative
effect of cell lengths and cell numbers, but to a variable and often
compensatory expression of internode number, cell number and
length. However, it would appear that cell number and length
characteristic for each variety, and for internodes within
each variety, have considerable stability.

Combining the data presented on cell number and size, and
keeping in mind the great variation encountered throughout
these studies, it can be stated that both cell number and cell
length contribute to plant height differences. Between Genesee
and 1328 it is mainly cell number, whereas between 1518 and
1328 it is cell size. Both are involved in the difference
between Genesee and 1518.

The Effect of Environment on Height

The tall variety appeared to be more influenced by environ-
ment than the two shorter ones. With decreasingly favorable
environment, the absolute height of all varieties decreased,
the height differences between the three lines were reduced,
but their relative positions remained the same, except on one
occasion when 1518 exceeded Genesee in height.

One exception to the reduction in culms length with in-
creasing stress occurred when plants were grown in 3 inch pots.
The mean length of all lines was greater than that observed in
the 3% inch pots, presumably a more favorable environment.
Plants in the 3% inch pots produced 2 to 3 tillers while those

in the 3 inch pots generally had only one culm. Possibly the lower light

35

intensity, owing to smaller spacing, and the greater nutritional
stress significantly reduced tillering at this point and all

the nutrition available went into the main culm.

The Effect of Environment on Components of Height

 

(a) Internode Length and Number

 

’ Statistical analysis of internode length shows that all
varieties are similarly affected by environment, which, within
varieties, has a differential effect on individual internodes.

It deserves special note that the interaction, internode x
environment, contributes more to the overall variance than
the effects of varieties and environment combined (Table 2).

The internode pattern of plants grown under optimum
conditions was nearly exponential. With increasing environmental
stress the pattern changed progressively from an exponential
to a linear one. Eventually, the upper internodes formed were
shorter than the lower ones (Figure l, b-h).

The lower internodes were more resistant to change with
environment than the upper ones. The internodal patterns as
plotted in Figure l (b-h) show this very clearly. These graphs,
called internode diagrams by Anderson and Schregardus (9) not only
show the change in length of successive internodes but also'
the rate of change of their respective lengths. This rate of
change, especially in the lower internodes, was less in the
shorter- than in the tall varieties. Environment greatly

influenced it, particularly in the upper internodes. Thus,

36

whereas genetic differences in length were evident in all inter-
nodes, environmental changes were reflected more in the upper
internodes.

Generally the upper internodes of Genesee were more affected
by environmental stress than those of the shorter varieties
(Figure 1). This accounted for the fact that, with increasingly
severe environment, the average difference in culm length
between varieties was reduced. The intermediate variety appeared
to be more variable in its behavior than the other two.

The length of an internode is determined by its sequential
position on the culm, i.e. it is probably influenced by the
internode above and below it, and by environment. Because the
develOpment of internodes is separated in time, minor fluctua-
tions in environment result in variation in internode pattern
of individual plants. The mean internode pattern, however, is
remarkably consistent and the general trend of environmental
effect on the final expression of internode- and culm length
is broadly predictable. Conditions of increasing stress and
competition result in progressively decreased length of internodes
and total length.

The effect of environment on internode number could not be
determined with certainty. It appeared that unfavorable condi-
tions tended to decrease the internode number. Observations
indicated, however, that this did not occur in all cases

(See Figure l (g)).

37

(b) Cell Length and Number

Environmentally induced variation tended to obscure the
differences in cell numbers between comparable internodes of
varieties. But, as in internode number, no general pattern
could be detected in environmental influence on cell number.

It has been reported that the decrease in culm length
as growing conditions move away from the optimum is due to
a reduction in cell size rather than cell number (66).
Observations made here indicated that both were affected.
Plants that were grown under less favorable conditions were
found to have smaller cell numbers, but cell size was actually
increased compared to those grown under more favorable conditions
(Figure 2 (a) and (b)). The greater height of the latter
plants, therefore, was actually due to increased cell numbers.

Culm Development

 

Primarily for convenience and simplicity, the developing
culm was studied in a number of more or less arbitrary growth
periods in order to determine where, when and how plant height
differences are first expressed and how they consolidate
throughout development. The periods of growth studied were
(1) before germination (2) germination to three-leaf stage
(3) three-leaf stage to period of rapid elongation (h) period of
rapid elongation through maximum extension. Growth periods (2)

and (3) were separated by vernalization treatment and thus the

38

critical transition from the vegetative to the reproductive phase

occurred there.

(1) Before Germination
Embryos of Genesee weighed significantly more than those of
1518 and 1328. No difference could be detected at all between
the latter two (Table 9).
Table 9. Fresh- and dry-weights of embryos excised from
kernels, in the late dough stage, of Genesee,

1518 and 1328. (gms. x 100,000)

Fresh weight S.D. Dry weight S.D.

Genesee 210 2.05 158 2.66
1518 19k 0.97 131 3.32
1328 193 0.95 133 3.82

 

Bigger seeds have bigger embryos as a rule. Hence the
greater "initial capital" frequently leads to increased size
at maturity. This has been established, although results do
not always agree. Bindloss (15) mentioned the conflicting
reports of the effect of seed weight on plant height and made
sure that both the tall and dwarf plants she studied had
similar seeds. Nevertheless, she found that height differ-
ences may be evident throughout development, or manifest
themselves only at a later stage of development. The seed
weight of Genesee, 1518 and 1328 decreased progressively in

direct relation to height, as did kernel volume and specific

39

gravity of the kernel (Table 10). Thus, there appeared to be a

positive correlation between height and seed weight of the three

varieties.

Table 10. Weight, volume and specific gravity of kernels
of Genesee, 1518 and 1328.

Mean Kernel weight Mean kerngl volume Spec. Grav.

 

‘__ mgs. ml. x 10 of kernel
Genesee 37.66 31.05 1.2133
1518 33.55 28.25 1.1881
1328 23.58 25.55 1.1581 _

 

The F value for varieties was significant at the .01 percent level._
Equal seed weight is no guarantee that plants start growth
with the same "initial capital.“ Correlation of seed weight and
plant height is affected by (a) the relative contribution of
embryo and endosperm to total seed weight and (b) the stage of
development that size (or height) differences are initiated. Bind-
loss found dwarf types with cells that matured earlier and elongated
for a longer period than those of tall types, resulting in dwarf
plants with fewer but larger cells. The difference between Genesee
and 1518 would appear to approach this situation. However, the
difference between these varieties was already evident in the
embryos, something which Bindloss attempted to avoid, although

she did not actually verify that this had been accomplished, other

than on the basis of seed weight.

no

(2) Germination to Three-Leaf stage

As representative of this period 2h-hour-old seedlings were
used. These seedlings had Just germinated and the three foliage
leaves, although present, had not yet unfolded. Estimates of
various gross dimensions of the young seedlings showed again
that 1518 and 1328 were similar, although the seedlings of
1518 were somewhat more slender. Genesee was larger than the
other two but did not appear to be more advanced in development
(Figure 3, Table 11). Owing to the difficulties involved in
obtaining comparable longisections for microscopic examination,
estimates of cell size and number showed a great deal of varia-

tion. However, the size of the meristematic cells in the apical

regions appeared to be the same for all three lines.

Figure 3. Diagramatic representation of a median longisection
of a 2h-hour-old seedling indicating the various
gross measurements made as listed in Table 11.

A

4A
N

#1

Table 11. Mean measurements in inches made of various gross
dimensions of projected images of ten 2h-hour-old
seedlings of Genesee, 1518 and 1328, as indicated
in Figure h.

A B c D E ‘F G ~s

Genesee 9.89 5.10 8.56 10.57 2.u7 3.u7 5.20 6.06

1518 8.87 t.07 6.66 8.1a 1.77 2.73 t.o3 b.83
1328 8.58 n.22 6.62 9.23 1.70 3.10 5.03 5.00
S.E. 0.hh5 0.121 0.268 0.283 0.18h 0.283 0.3h6 0.2h5

The F value for varieties was significant at the .01 percent
level for all measurements.

At this early stage, therefore, the difference between Genesee
on the one hand and 1518 and 1328 on the other appeared to be due
to cell number. Recalling that the difference in culm length
between Genesee and 1328 is due to cell number and that between
1518 and 1328 to cell length, it appears that height differences
based on cell number originate at some stage during embryogenesis
while the capacity for greater cell elongation is expressed at a
later stage of deve10pment, after germination. Thus, clearly,
two fundamentally different developmental processes must be in-

volved in causing the respective differences in height, making

apparent independence understandable.

#2

(3) Three-Leaf Stage to Period of Rapid Elongation

This relatively short period began when the vernalized
plants were brought into a favorable environment for growth
and ended with the beginning of rapid elongation. Although
growth was slight, differences in the height of the growing
.point became observable under a dissecting microscope as soon
as the plants had adapted themselves to their new environment,
after a few days at the most. During this period the first
evidence of greater cell size in the stem of 1518 became
visible. Varietal differences in extension were far greater
at this stage than any environmentally induced variations.
Genesee was at all times more advanced in development than
the two shorter varieties. Development was more rapid in all
varieties when kept at low temperature and low levels of
nutrition, but this was not so striking in Genesee as in the
shorter lines. Figure h indicates the differences in culm
extension and development that could be detected just prior
to the beginning of rapid elongation.

(h) Period of Rapid Elongation Through Maximum Extension

Growth curves of the three varieties and the relative
Chronological development of their internodes are presented
in Figures 5 and 6. It is evident that the difference in
culm length between the three varieties is manifested at a

very early stage.

In terms of relative growth during this period, the observed

Figure h. Tracings of projected images of longisections
of shoots of Genesee, 1518 and 1328 at the
beginning of rapid elongation.

(a) Genesee; high nutrition (b) Genesee; low nutrition
and moisture and moisture

(c) 1328; high nutrition (d) 1328; low nutrition
and moisture and moisture

A3

 

Ill")! . ill-If .

“5

differences in length of mature culms could be explained by
both differential duration and rates of growth. ‘Genesee
elongated over a longer period than either 1518 or 1328 but
its rate of growth was similar to that of 1518. The rate of
growth of 1328 however, was distinctly lower than that of the
other two.

Genesee did not always start elongation much earlier
than the other two varieties, but in such cases maturity
differences at the end were more pronounced with Genesee always
being the last to fully expand and mature. The total effect
appeared to be one of shifting the growth curve of Genesee
slightly to the right of 1518, (Figure 5), in such a way
that the curves of these two varieties cut across each other.
Especially under crowded conditions with high nutrition 1518
was frequently found to be taller in the early stages of
development. At a later stage, however, Genesee would over-
take it to mature at a greater length.

The pattern of internode elongation is basipetal, the
upper internode being the last to elongate. Elongation does
not begin until the plant has entered the reproductive phase
and flowering primordia are visible at the apex. The number
of internodes has therefore been determined at the time of first

elongation.

.moso monumwc aw mpmam cw czopw 0L0: mucmad

.moma nonemco Luzopm CH mama can mama .omwoco
ApnoEummpu ccflumnwfimcao> mo :0

mm

.cpwcma

. Hmuou

L W0 WEHDU CHmE “HO mwpmrH C >Qr waw
wpoaseoo 509m m>mcv uzoam o>m

om

mm ECXQ

Hwawx

cm

0: mm m SDKHL

u.“

      
  

0H
m

.Jlllllllllllllllll: 0

MS

I
I
In
a
n
n
III
IIII
IIIIIIII

mama

3mm. I.|.|.

.cmw mmma ..:..:

om
mama

wwwwfiwo

 

moccflw cw
newcofi
EH50

87

Three consecutive internodes elongate simultaneously at
any one time: the upper one starting, the middle one extending
rapidly and the lower internode just completing expansion. The
most rapid increase in total length takes place during the
elongation period of the first and second internodes from the
top. These two internodes combined make up a large percentage
of the total length at maturity and during their "grand phase"
of expansion the increase of the total length of the culm is
extremely rapid.

The growth curve of the entire stem is, in effect, the sum
total of the growth curves of its internodes. Differences in
the rates of growth of comparable internodes of the three
varieties were difficult to define. HOwever, both duration
and time of initiation of elongation of comparable internodes
differed greatly. The length advantage of Genesee is primarily
due to the fact that its lower extending internodes generally
elongate over a much longer period. The internodes of 1518
and 1328 elongate to an almost identical pattern except that
the lower internodes of 1518 elongate for a slightly longer
period (Figure 6).

The actual duration of extension appears to be the more
important contributing factor to mature length differences.
Johnson (38) studying the development of three wheat varieties
of different inherent height also found the difference to be

due to duration rather than rate of growth. This may explain

N8

.mmma LoLEmLo Luzoho .mmma mam mama .wmmmcow mo cemumwcoHo evocaoucw we camupmm oumEHxOLaa<
A.ucoEummhu cofipmNHHmcmm> mo coHpoHaeoo SCAM mzmcv

0: mm om . mm om ma OH m

I ‘ ‘.
IOIIIIIIIOCOIIIIIII

 

 

 

 

mmmifi ICI'.......

mama

.m whsxwu

 

1‘9

the definite association with maturity found here, by Johnson
(38), Prat (M8) and others. The results here, however,
indicate that differential growth rates are also involved.
These growth rates are heritable as was shown by Ebiko (25).

The data suggest that the greater culm length of Genesee
is due to a greater “initial capital" and longer duration of
growth. Ashby (ll) postulated that increases in growth due
to heterosis were no more than the maintenance of a greater
initial size caused by physiological stimulation during
embryogenesis. The greater embryo size of Genesee must also
be due to increased physiological activity during the deve10p-
ment of the embryo.

In trying to account for the heterotic effects, Ashby
could find no differences in growth rates so they must have
been due to greater duration of growth activities. Stein (56)
found that in reciprocal hybrids between inbreds of Zea maysz
scutellum deve10pment and leaf initiation appeared to start
earlier, giving hybrids an advantage. As no difference could
be detected between the rates of deve10pment in hybrids and
inbreds it seemed to be the longer duration of growth pro-
cesses that lead to eventual greater plant size. Since the
greater height of Genesee seems to be due, in no small part,
to longer periods of development of its internodes as compared
to those of the shorter varieties, it is easy to surmise,

considering this and the above evidence, that the larger size

50

of the Genesee embryo is possibly also due to a longer duration
of development. Duration of growth is therefore more important
than growth rate in the expression of culm length.

In Figure 7 the cell numbers of the developing peduncle
of Genesee, 1518 and 1328 are plotted against the length of
this internode at the various stages of development. This gives
some indication of the relative importance of cell division and
elongation in the extension of an internode and differences
which exist between varieties in this respect.
Homeostatic Systems in Plant Height Expression

The influence of environment on culm length is great.
Even greater variability exists at the internodal and cellular
levels due to environmental influence. For example, in one
of the experiments (Figure l (d)), the internodes of Genesee had
coefficients of variability of 19.9%, 13.1%, 20.3%, 18.8%
and 50.3%. The C.V. of total height, however, was only 10.6%.
These results are typical for all varieties under all environ-
mental conditions. Similarly, average cell size and cell number,
characteristic for each internode of a variety, are fairly con-
stant. However, a tremendous amount of variability exists
within internodes.

Considering the influence of light, temperature and other
factors on both cell division and cell elongation this is perhaps
not surprising (18). Variation is induced at this "micro level"

(16) and compensating mechanisms, which are due to nothing more

I lillllll‘

Sl

 

      

 

 

Cell
no.
Gen.
10 ’000 - . a- ' "" ' - ' -_,',:JSJSJMAMAMM-Immnma— - — --
1518 ...--""' 1328 1518
a". 1328
0....
5 ,000
.4 I ill I I
5 10 15 20 25

(Internode length in cm.)

Figure 7. Cell number of the peduncle of Genesee, 1518 and 1328
during deve10pment.

I‘ll}. I'll]. 11" (I l .
III}. I'll" 1'1

52

than the influence of one developing part on another become
evident. If the development of one part is altered by a change
in environment, presumably its correlative effect on that of
another is also changed giving a compensating effect. Compensa-
tory mechanisms are evident at both the cellular and component
levels.

This raises the question of how far down does one set the
micro- or component level in attempting to explain the heritable
behavior of a complex character. In research, the explanation
of any biological phenomemon is normally procured by breaking
it down into smaller and smaller components and by developing
a "micro-theory" (16). However, in heredity, the variation at
this microlevel may be so great that this, and the very large
number of interacting genetic and physiological components which
may be involved, lead to a situation where the complex, inte-
grated trait may be more stable and predictable in its behavior
than any of its individual components.

Plant height in cereals, although a complex character,
normally has a fairly high heritability. Its expression within
any one environment is stable owing to compensation of its
components, which themselves can display a great deal of varia-
tion. This would suggest that, from an applied point of view at

least, it may be more meaningful to remain at higher levels and

53

in more complex categories when manipulating quantitative
traits such as plant height in heredity. Studies, under
rigidly controlled environments, of components at the micro-
level may provide some of the answers as to basic mechanisms,
but from a practical plant breeding point of view the inter-
actions at higher levels are more important. Other complex
systems, however, may be amenable to the component approach.

In the final issue one is gnly concerned with inter-
actions:nucleus x cytOplasm, gene x gene, call x cell, cell
x external environment, and one can only arrive at sensible
explanations by specifying all known relative conditions.

In morphogenetic development of a quantitative trait this can
be exceedingly complex.

In crosses between Genesee and 1328, heterosis was evi-
dent in the F1. The variance of the segregating F2 population
was only little greater than that of the tall, homozygous parent.
Under field conditions, the variance of the F2 population was
actually smaller than that of the homozygous, tall parent.
Means and variances of culm length of Genesee, 1328 and their

hybrid are presented in Table 12.

5h

Table 12. Means (cm.) and variances of culm length of
Genesee and 1328 and the hybrid Genesee x 1328.

 

__5 Growth chamber Field 55
- 2 - Z
x 0— x CT'
Genesee 62.8h 30.58 Genesee 75.57 100.95
1328 52.26 22.73 1328 5h.27 16.79
F1 65.23 38.18 F2 62.57 55.01
F2 62.0u 36.66

 

It will be recalled from the results of earlier experi-
ments that the tall variety Genesee was less resistant to
environmental changes with respect to height than the short
1328. Other experiments in the growth chamber and in the
field confirmed this. It would thus appear that in the inter-
action with the micro-environment a homeostatic or buffering
element is present which masks the expression of quantitative
characters. This could take the form of compensatory inter-
action, which was evident at the component and at the cellular
levels.

Homeostasis is usually associated with heterozygosity,
the developmental paths of heterozygotes being better buffered
against environmental fluctuations than those of homozygotes
(23). Tebb and Thoday (59), however, found that heterozygosity
does not necessarily promote homeostasis. Moreover, others (h),

(7), (36), (51) have presented evidence that differential

55

buffering exisuaamongst homozygous genotypes. At first glance
it would therefore appear from the results in Table 12 that
differential buffering capacity may have been responsible for
differences in variability in the varieties, and that plants in
the F2 populations especially in the field segregate for these
properties of homeostasis and environmental interaction. One
must be careful in drawing conclusions, however, because most
of the difference in variability observed is due to variance

associated with height (Table 13).

Table 13. Regression of variance on height in Genesee and 1328.

 

 

Source DF S.S. M.S. F
Genesee Regression l 17h.2l l7h.2l 20.96**
Deviations from
‘_ regression 5 h1.56 8.31
1328 Regression l 80.62 80.62 h.93 NS
Deviations from
__ regression 5 81.79 16.36

 

Previously presented evidence showed that the greater
variability of Genesee was mainly associated with the greater sensi-
tivity of its upper internode to environmental fluctuations.

Table 13 indicates that when the variability dependent on plant
height has been allowed for, Genesee is actually less variable
than 1328 i.e. more homeostatic. This one would eXpect, as
Genesee is a commercial variety of long standing, probably more

highly adapted than 1328, a selection of recent breeding.

56

The great difficulty of obtaining and interpreting valid
data in inheritance studies of continuously variable traits is
emphasized by the above evidence. Differential homeostasis of
homozygous parents and heterozygous hybrid populations and
variability dependent on height will tend to obscure segregation.
In the particular case reported here both of the parental
lines were adapted lines, in harmony with their particular
environment, and had probably evolved an efficient buffering
system to maintain phenotypic stability. For full expression
of genotypic variation the F2 should probably have been grown
and observed outside their adaptive environment. Rigid control
of environmental conditions in the growth chamber only seemed to
enhance the buffering effect, most of the observed variation
being environmental.

Study of the character at the component- or even the
cellular level does not seem to be the answer, as the great
variability and compensatory interaction here leads one to the
conclusion that little can be gained from such studies.

Athanassoff (13) states in his concluding remarks: "Dans
le chaume de blé en question, il n'y a pas deux tranches qui
soient identique. Pourtant, chaque coupe a sea analogues,
plus bas et plus haut.” Herein lies the whole problem of
quantitative biological variation in heredity. 0n the one hand
the great constancy of phenotypic expression, on the other the
great variability, more often than not of'a compensatOry nature,

of all the characters associated with it. This constant

57

correlative interaction of character with character, component
with component and each with environment is a criterion of
biological phenomena and so prevalent in all studies of
morphogenesis, including this one.

The variation in characters, and in components of that
character, which occurred between plants are indicative of the
extraordinary variability that occurs with subtle differences
in environment, both in space and time, and probably also with
subtle differences in genotypes.

This has mitigated against the presentation of clear cut
results. However, the evidence presented here has led to a
somewhat better understanding of the nature of plant height
differences in cereals and difficulties involved in the genetic

study of this complex trait.

SUMMARY

The nature of height differences was investigated in three
closely related, homozygous lines of wheat: Genesee (tall), 1518
(intermediate) and 1328 (short).

Differences in culm length were generally due to inter-
node length, both within and between varieties. The contribu-
tion of internode number was negligible.

Differences in length of comparable internodes of Genesee
and 1328 were due to cell number, whereas in 1518 and 1328 cell
size was responsible. Both were involved in the difference
between Genesee and 1518.

It appeared that height differences based on cell number
originate during embryogenesis while those based on cell length
arise after germination. Thus, two independent deve10pmental
processes are involved.

Although the varieties differed slightly in their growth
rates, the actual duration of extension growth appeared to be
the more important contributing factor to mature length differ-
ences.

Compensatory mechanisms were evident at both cellular and
component levels. The variability encountered at these levels

was greater than that of the complex trait.

58

59

A strong positive relationship existed between variance and
height, especially in the tall variety, and there was some
evidence of differential homeostasis of homozygous parents.

Such properties tend to obscure genetic segregation.

ll.

12.

LITERATURE CITED

Abbe, E. C., B. 0. Phinney and D. F. Beer. The growth of
the shoot apex in maize. Am. J. Bot. 38:737-751. 1951.

Abbe, E. C. and O. L. Stein. The growth of the shoot apex
in maize. Embryogeny. Amer. J. Bot. hl:285-293. 195h.

Abbe, Lucy B. The histological background for dwarfism in
Zea mays. Proc. Am. Phil.Soc. 76:743-7u7. 1936.

Adams, M. W. and D. B. Shank. The relationship of hetero-
zygosity to homeostasis in maize hybrids.
Genetics uh: 777-786. 1959.

Allan, R. E. and 0. A. Vogel. F2 monosomic analysis of
culm length in wheat crosses involving semidwarf
Norin 10 - Brevor 14 and the Chinese spring series.
Crop Sci. 3:538-5h0. 1963.

Allan, R. E., 0. A. Vogel, J. R. Burleigh and C. J. Peterson,
Jnr. Inheritance of coleoptile length and its
association with culm length in four winter wheat
crosses. CrOp Sci. 1:328-332. 1961.

Allard, R. w. and A. D. Bradshaw. Implications of Geno-
type-Environmental interactions in applied plant
breeding. Crop Sc. #:503-508. 196%.

Amelung, E. fiber mittlere ZellengrBsse. Flora 77:176-207. 1893.

Anderson, E. and D. Schregardus. A method for recording and
analyzing variations in internode pattern. Ann.
Missouri Bot. Gard. 31:2hl-2h7. l9hh.

Andersson, S. Ueber die Entwickelung der primaren
Gefassbundel. Strange der Monokotylen. Review Bot.
Centralblatt 38:586-587. 1889.

Ashby, E. Studies in the inheritance of physiological
characters. I. A physiological investigation of
the nature of hybridvigour in maize. Ann. Bot.
uh: h57-h68. 1930.

Ashby, E. Studies in the inheritance of physiological
characters. II. Further experiments upon the basis
of hybrid vigour and upon the inheritance of efficiency
index and respiration rate in maize. Ann. Bot. 116:1007-
1032. 1932.

60

13.

1h.

15.

16.

17.

18.

19.

20.

21.

22.

23.

2h.

61

Athanassoff, Athanas D. L'Anatomie et la Maturation des
chaumes d'unipieddeBle’ collosse de Razgrad
(Triticum turgidum L.) Ann. des Sc. Nat., Bot. loe
series: 1-99. 1928.

Avery, G. 8., Jr. Comparative anatomy and morphology of
embryos and seedlings of maize,oats and wheat. Bot.

Gaz. 89: 1-39. 1930.

Bindloss, Elizabeth. A deve10pmenta1 analysis of cell
length as related to stem length. Amer. J. Bot.
29: 179-188. 19h2.

Bonner, J. T. Morphogenesis, an essay on development.
Atheneum. New York. 296pp. 1963.

Bonnett, 0. T. The oat plant: Its histology and
development. Bull. 672. Univ. of Illinois. 1961.

Brotherton, W. Jr. and H. H. Bartlett. Cell measurement
as an aid to the analysis of variation. Am. J.
Bot. 5:192-206. 1918.

Carlson, Irving T. Inheritance of plant height in wheat.
M.S. Thesis, Washington State College. 1952.

Clark, J. A. Segregation and correlated inheritance in
crosses between Kota and Hard Federation wheats
fer rust and drought‘resistance. J. Agric. Res.
29:1-h9. l92h.

Clark, J. A. and J. R. Hooker. .Segregation and correlated
inheritance in Marquis and Hard Federation crosses
with factors for yield and quality of spring wheat
in Montana. U.S.D.A. Bul. 1&03. 70 pp. 1926.

Clark, J. A. and K. S. Quisenberry. Inheritance of yield
and protein content in crosses of Marquis and Kota

spring wheats grown in Montana. J. Agric. Res.

Dobzhansky, Th. and H. Levene. Genetics of natural pop-
ulations. XXIV. Developmental homeostasis in
natural populations in Drosophila pseudoobscura.
Genetics h0:797-808. 1955. “'w

 

Duarte, R. and M. W. Adams. Component interaction in
relation to expression of a complex trait in a bean
cross. Crop Sc. 3:185-186. 1963.

25.

26.

27.

28.

29.

30.

31.

32.

33.

3t.

35-

36.

37.

62

Ebiko, K. Inheritance of growth curve. J. Am. Soc.
Agron. 30:558-563. 1938.

Everson, E. H., C. E. Muir and 0. A. V0ge1. Dwarfing
in Triticum vulgare (Vill.) Agron. Jour. h9:52lo 1957

Florell, V. H. and J. F. Martin. Inheritance of comple-
mentary dwarfing factors in wheat. J. Agric. Res.
53:151-160. 1936.

Freeman, G. F. The heredity of quantitative characters
in wheat. Genetics #:1-93. 1919.

Goulden, C. H. A genetic and cytological study of
dwarfing in wheat and oats. Minn. Agr. Expt.
St. Bull. 33:37. 1926.

Grafius, J. E. Components of yield in oats: A geometric
interpretation. Agron. J. h8:hl9-h23. 1956.

Guilland, A. Recherches sur l'anatomie comparee et 1e
developpement des tissus de la tige dans lea
Monocotylédones. Ann. Sc. Nat. Bot. 6e serie 5:

5-176. 1878.

Hayman, B. I. Heterosis and quantitative inheritance.
Heredity 15:32h-327. 1960.

Hermsen, J. G. Th. The localization of 2 genes for dwarfing
in the wheat variety Timstein by means of substitution
lines. Euphytica 12:126-129. 1963.

Houghtaling, H. Stem morphogenesis in Lycgpersicum:
A quantitative study in cell size and number in
the tomato. Bull. Torrey Bot. Club. 67:35-55. l9h0.

 

Hurd, E. A. and R. C. McGinnis. Note on the location of
genes for dwarfing in Redman wheat. Can. J. of
Plant Sci. 38:606. 1958.

Jinks, J. L. and K. Mather. Stability in development of
heterozygotes and homozygotes. Proc. Roy. Soc. B.
1h3z56l-578. 1955.

Johnson, V. A. Environmental factors affecting plant
height in winter wheat. Agron. J. h5:505-508. 1953.

38.

39-

ho.

hl.

h2.

h3.

nu.

2.5.

h6.

h7.

A8.

h9.

63

Johnson, V. A. Culm morphology and development in winter

Kaufman, Peter B. Quantitative studies on stem development
in rice (Oryza sativa L.). Indian Journal of
Agric. Sc. 30:233-2E9. 1960.

Kaufman, Peter B., Sandra J. Cassell and Paul A. Adams.
0n nature of intercalary growth and cellular
differentiation in internodes of Avena sativa.
Bot. Gaz. 126 (1): 1-13. 1965.

 

Lebsock, K. L. Transfer of Norin 10 genes for dwarfness
to Durum wheat. Cr0p Sc. 3:h50-h51. 1963.

McCall, M..A. Developmental anatomy and homologies in
wheat. J. Agric. Res. h8:283-321. l93h.

Merkle, 0. G. and I. M. Atkins. Inheritance of plant
height and stem rust resistance in wheat,
Triticum aestivum L. Crop Sc. u:h53-h5u. 196k.

 

Newcomer, E. H. A new cytological and histological
fixing fluid. Science 118:161. 1953.

Nilson, E. B., V. A. Johnson and C. 0. Gardner.
Parenchyma epidermal cell length in relation to
plant height and culm internode length in winter
wheat. Bot. Gaz. 119:38-83. 1957.

Percival, J. The wheat plant; a monograph. #63 pp. London.
1921.

Prat, Henri. A biometric study on the culm of cereals and
grasses. Can. J. of Research 10:563-570. 193A.

Prat, Henri. Recherches sur la structure et le mode de
croissance des chaumes. Ann. Des Sc. Nat. Bot.
106 ser. 17:81-1h5. 1935.

Sachs, J. Ueber den Einfluss des Lufttemperatur und des
tageslights auf die stundlichen und taglichen
Aenderungen des Langenwachsthums (Streckung)
der Internodien. Arbeiten des bot. Instituts in
wurzhurg I 99:193. 1872.

50.

51.

52.

53-

5h.

55..

57-

58.

59.

60.

61.

Sachs, J. Physiologische Notizen. VI. Ueber einigen
Beziehungen der specifischen Grésse der Pflanzen
zu ihrer Organisation. Flora 77:h9-8l. 1893.

Shank, D. B. and M. W. Adams. Environmental variability
within inbred lines and single crosses of maize.

Sharman, B. C. Developmental anatomy of the shoot of
Zea maysLW» Ann. Bot. 6:2h5-282. 19h2.

Sierp, H. Uber die Beziehungen Zwischen IndividuengrBsse,
Organgresse und Zellengrhsse, mit besondere
Berficksichtigung des erblichen Zwergwuchses.

Jahrb. wiss. Bot. 53:55-12h. 1913.

Sinnott, E. W. The relation of gene to character in
quantitative inheritance. Proc. Nat. Acad. Sci.

Sinnott, E. W. The cell-organ relationships in plant
organization. Growth (Suppl.) pp. 77-86. 1939.

Stein, Otto L. A comparison of embryonic growth rates
in two inbreds of gea mays L. and their reciprocal
hybrids. Growth 20: 37-50. 1956.

Stewart, G. and D. E. Heywood. Correlated inheritance
in a wheat cross between Federation and a hybrid
of Sevier x Dicklow. J. Agric. Res. 39:367-392.

1929.

Stewart, G. and D. G. Tingey. Transgressive and normal
segregation in a cross of Marquis x Federation
wheat. J. Amer. Soc. Agron. 20:620-63h. 1928.

Tebb, G. and Y. M. Thoday. Stability in development and
relational balance of X-chromosomes in Drosophila
mglanogastep, Nature l7h:llO9. l95h.

 

 

Thompson, W. P. The genetics and cytology of a dwarf wheat.
Roy. Soc. Canada Proc. and Trans. 22:335-3h8. 1928.

Tingey, D. C. Inheritance of dwarfing in wheat. Utah
Acad. Sci. 8:55-60. 1931.

62.

63.
6h.

65.

66.

65

Torrie, J. H. Inheritance studies of several qualitative
and quantitative characters in spring wheat crosses
between varieties relatively susceptible and resistant
to drought. Can. Jour. Res. C. lh:368-385. 1936.

van Tieghem, Ph. Traité de botanique. Paris. 1891.

Waldron, L. R. A study of dwarfness in wheat accompanied
by unexpected ratios. Genetics 9:212-2h6. 192k.

Watanabe, Y. Studies on the cytological instabilities of
common wheat. Jap. J. Breeding. #:67-76.
Oct. 195%.

Whiteside, A. G. 0. Effect of soil drought on wheat plants.