Title: Comparisons of seed weight and seedling characteristics of diploid and autotetraploid red clover
CITATION PDF VIEWER THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00100830/00001
 Material Information
Title: Comparisons of seed weight and seedling characteristics of diploid and autotetraploid red clover
Physical Description: Book
Language: English
Creator: Furuya, Hideto, 1977-
Publisher: University of Florida
Place of Publication: Gainesville Fla
Gainesville, Fla
Publication Date: 2001
Copyright Date: 2001
 Subjects
Subject: Agronomy thesis, M.S   ( lcsh )
Dissertations, Academic -- Agronomy -- UF   ( lcsh )
Genre: government publication (state, provincial, terriorial, dependent)   ( marcgt )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )
 Notes
Summary: ABSTRACT: Seedling establishment is a critical aspect associated with the profitability of annual forage production in Florida. The objective of this research was to evaluate the effect of chromosome doubling on seed production and seedling characteristics of red clover (Trifolium pratense L.), a cool-season forage legume adapted to winter-spring conditions in north Florida. Autotetraploid (4x) populations were produced by nitrous oxide treatment within the diploid (2x) cv. 'Cherokee.' Four clones of 2x and 4x seedlings from each of eight crosses were grown in the field to increase seed with the 2x and 4x populations physically isolated. Seeds were harvested from individual clones, and seed number and weight of each genotype were determined at each ploidy level. The mean seed number per plant for the 2x and 4x populations was 806 and 26, respectively. Mean seed weight of the 4x was approximately 1.5 times that of the 2x. There was a ploidy level x cross interaction effect, indicating that superior seed-producing genotypes at the 2x level were different from those at the 4x level. Thereafter, seeds from all genotypes within each ploidy level were composited into two populations. The 2x and 4x seeds were germinated in growth pouches at 12, 20, and 28 degrees C in the dark to compare seedling growth of 2x and 4x populations. The seedling characteristics measured were total length, hypocotyl length, root length, crown diameter, hook diameter, and middle diameter of the seedling between the hook and crown. The characteristics were measured at four dates in each temperature. The 4x had greater means for all seedling characters at any temperature. Especially, the 4x hypocotyl mean was longer than the 2x.
Summary: ABSTRACT (cont.): The increases in diameter may indicate greater emerging forces of the 4x. The effect of seed weight on seedling growth at both ploidy levels was tested at a constant temperature of 20 degrees C using regression analysis. Seed weight was shown to influence all the growth characters measured. The influence of seed weight was generally the same regardless of ploidy level. Thus, the greater seedling response means of 4x were primarily due to heavier individual seed weights. The effect of various chemical additions and combinations (Hoagland solution, indole-3-acetic acid, kinetin, and sucrose) in the germination medium and presoaking treatments was studied to determine their effect on seedling growth at both ploidy levels. Among the chemicals tested, only sucrose had a positive effect on root development. These experiments indicated that the 4x seeds were heavier and the 4x seedlings were longer in length and thicker in diameter. This would suggest that, under field conditions, the use of 4x seeds would probably result in superior establishment. However, the 4x seed fertility in the first generation was low, limiting practical use unless greater fertility can be obtained in successive generations.
Thesis: Thesis (M.S.)--University of Florida, 2001.
Bibliography: Includes bibliographical references (p. 75-79).
System Details: System requirements: World Wide Web browser and PDF reader.
System Details: Mode of access: World Wide Web.
Statement of Responsibility: by Hideto Furuya.
General Note: Title from first page of PDF file.
General Note: Document formatted into pages; contains xiii, 80 p.
General Note: Vita.
 Record Information
Bibliographic ID: UF00100830
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 48991993
alephbibnum - 002763578
notis - ANP1600

Downloads

This item has the following downloads:

fullmaster2 ( PDF )


Full Text











COMPARISONS OF SEED WEIGHT AND SEEDLING CHARACTERISTICS OF
DIPLOID AND AUTOTETRAPLOID RED CLOVER













By

HIDETO FI RUTYA


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA


2001

































Copyright 2001

by

Hideto Furuya





























This thesis is dedicated to my grandmother, Morimoto Matsuyo (1907-), who has farmed
all of her life.


with respect,
















ACKNOWLEDGMENTS

I would like to express sincere appreciation to Dr. K.H. Quesenberry, who was the

chair of my committee and directed my research program.

Dr. P.M. Lyrene served as my external member and horticultural sciences minor

representative. His way of thinking, consistent friendliness, and supportive attitude greatly

encouraged me during the progress of research. Particularly I thank him for letting me

participate in a part of his blueberry breeding program.

Most of what has been done in this paper could have not been accomplished

without Dr. P.L. Pfahler. He generously provided equipment and expertise to conduct the

experiments. We argued about all subjects, but the exchange will become an unforgettable

memory. He generously provided me with a research assistantship during my last term,

which greatly enriched my experience in the entire program. Together with Dr. R.D.

Barnett, my involvement in the small grain breeding program was also a valuable

experience.

I am impressed by the deep thoughts and critical thinking of Dr. T.R. Sinclair. His

method of conducting research and making hypotheses will be a valuable tool in my

future. Always, I wish I could have more time to spend with him.

It was a pleasure meeting Dr. L.E. Sollenberger. His reputation extended to the

University of Kentucky. Even though he was in Australia during most of my program, he









was always the first one to respond to my e-mail questions. Involvement in lunch

discussions with his lab. group was also memorable. I appreciated the opportunity to

present my research results to his congenial circle. When I obtain a professional position, I

would like it to be associated with an organization like his.

Some other names should be mentioned. Erika. R. Henderson and Judith M.

Mullaney really initiated the research. The technical advice from Dr. Francisco C.

Krzyzanowski, visiting scientist in the seed technology laboratory, was also critical at the

initial stage of my work. I enjoyed very much talking to Roger A. Haring. We never know

which strangers become friends. I also appreciate the friendship with Yoana C. Newman,

Eastonce T. Gwata, and Liana Jank, residents of our graduate room and their families.

Activities with Eric R. Ostmark and the Aikido Club were my only social life. I sincerely

hope the best for them in the future. I also appreciate the help from the CIRCA assistants.

They were the most valuable key to the presentation of this thesis.

I acknowledge the contributions of my father and mother, Furuya Toshihiko and

Tomoe. Even though we never met each other during this period, their support and

encouragement was the only reason I could continue my education and conduct the

research necessary for this thesis. They never understood what I was doing, but their

commitment was unending. It is now my turn to support them in any way.

This research was designed to improve red clover establishment in Florida. I would

be very grateful if these research results could provide an improvement of forage

production worldwide as well as contribute knowledge to academia.















TABLE OF CONTENTS

Page


ACKNOW LEDGM ENTS ................................................ iv

LIST OF TABLES ........ ............................................ ix

ABSTRACT ........ ..... ................................ ........ xii

CHAPTERS

1 INTRODUCTION ...................................................1

General ....... ............ ...... .. ................. ... ......... 1
Influence of Climate in Southeastern USA ................................ 2
Red Clover ....... .................................................. 3
Forage Use ........ ............................................. 4
Mode of Pollination ............... .............................. 4
Nitrogen Fixation ................................................. 5
Red Clover in Florida .................................................. 5
Polyploidy ........................................................... 7
Methods of Chromosome Doubling ................ ................... 10
Research Justification and Objectives ................................... 11

2 THE EFFECT OF PLOIDY LEVELS ON SEED NUMBER AND WEIGHT ...... 13

In tro du ctio n . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Literature Review ..................................... ................ 13
Materials and Methods ............................................. 15
Production of Tetraploid Plants ................ ................... 15
Field Study ..................... ......................... 16
Results ...................................................... 17
Seed Number .......... ................................. ......... 17
Mean Seed Weight ................................................ 17
Discussion and Conclusions ............... .......................... 17



vi











3 THE EFFECT OF TEMPERATURE ON DIPLOID AND TETRAPLOID RED
CLOVER SEEDLING GROW TH .......................................... 23

Intro du action . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2 3
Literature Review ..................................................... 23
Seedling Characteristics of Clovers .......... ......................... 23
Temperature Effect on Seedling Growth ............................ 24
M materials and M ethods .............................................. . 27
Results ...................................................... 28
Evaluation at 28C ......... ......... .................. ......... 28
Length measurement ........................................ 28
Diameter measurement ................ ................... 29
Evaluation at 20C ......... ......... .................. ......... 29
Length measurement ........................................ 29
Diameter measurement ................ ................... 30
Evaluation at 12C ......... ......... .................. ......... 30
Length measurement ........................................ 30
Diameter measurement ................ ................... 31
Discussion and Conclusions ............................................. 31


4 THE EFFECT OF SEED WEIGHT ON SEEDLING GROWTH OF DIPLOID AND
TETRAPLOID RED CLOVER .............. .................. 47

Introduction .............. ................. ............... ... .47
Literature Review ..................................................... 47
Materials and Methods ................... ............................ 50
Results ....................................................... 51
Discussion and Conclusions ............................................. 52

5 THE EFFECT OF VARIOUS CHEMICALS ON DIPLOID-TETRAPLOID RED
CLOVER SEEDLING GROW TH .................................... 57

Introduction ...................................... .................... 57
Materials and Methods ................... ............................ 58
Results ........ .................................. 60
Hypocotyl Length ................................................. 60
Root Length .......... ............................... 61
Crown Diameter . ................. ......................... 62
Middle Diameter ................. ......................... 62
Hook Diameter .................. ............................ 62
Discussion and Conclusions ............................................ 63










6 SUMMARY AND CONCLUSIONS ......... ....................... 72

REFERENCES ......... ....................... .......... ....... 75

BIOGRAPHICAL SKETCH ...................................... 80















LIST OF TABLES


Table Page


2-1. Mean squares and significance levels from the analyses of variance of seed number
plant-' (transformed) and seed weight ................................. 20

2-2. Mean seed number plant-' from each cross at each ploidy level. The log
transformed mean is given after each mean in parentheses for statistical
comparisons ...................................................21

2-3. Mean seed weight from each cross at each ploidy level ................... 22

3-1. The crosses and the estimated seed numbers composite to develop the diploid
(2x) and tetraploid (4x) populations used in this and subsequent studies ...... 34

3-2. Mean squares and significance levels from the analyses of variance of total seedling
length (TL), hypocotyl length (HL), and root length (RL) at 28C ........... 35

3-2. Means of total seedling length (TL), hypocotyl length (HL), and root length (RL)
at each ploidy level (PL) at 28 C on each of four days after germination
initiation .................................................... 36

3-4. Mean squares and significance levels from the analyses of variance of crown
diameter (CD), middle diameter (MD), and hook diameter (HD) at 28 C ..... 37

3-5. Means of crown diameter (CD), middle diameter (MD), and hook diameter (HD)
at each ploidy level (PL) at 28 C on each of four days after germination
initiation ......... .............................................. 38

3-6. Mean squares and significance levels from the analyses of variance of total seedling
length (TL), hypocotyl length (HL), and root length (RL) at 20C ........... 39

3-7. Means of total seedling length (TL), hypocotyl length (HL), and root length (RL)
at each ploidy level (PL) at 20C on each of four days after germination
initiation ........ ............................................... 40









3-8. Mean squares and significance levels from the analyses of variance of crown
diameter (CD), middle diameter (MD), and hook diameter (HD) at 20 C ..... 41

3-9. Means of crown diameter (CD), middle diameter (MD), and hook diameter (HD),
at each ploidy level (PL) at 20C on each of four days after germination
initiation .................................................. 42

3-10. Mean squares and significance levels from the analyses of variance of total seedling
length (TL), hypocotyl length (HL), and root length (RL) at 12C ........... 43

3-11. Means of total seedling length (TL), hypocotyl length (HL), and root length (RL)
at each ploidy level (PL) at 12C on each of four days after germination
initiation ..................................................... 44

3-12. Mean squares and significance levels from the analyses of variance of crown
diameter (CD), middle diameter (MD), and hook diameter (HD) at 12C ..... 45

3-13. Means of crown diameter (CD), middle diameter (MD), and hook diameter (HD)
at each ploidy level (PL) at 12C on each of four days after germination
initiation ..................................................... 46

4-1. The seed weight mean, standard error (SE), standard deviation (SD), maximum
and minimum values, normality test, and total sample number of diploid (2X) and
tetraploid (4X) populations ..................................... . 54

4-2. Linear regression analyses [L (mm) = a (mm) + b (mm mg-1) SW (mg), where seed
weight (SW) is the independent variable] between SW and the various characters
at each date and ploidy level. The data reported are the mean, regression equation,
and standard error (SE) of the b-value for total seedling length (TL), hypocotyl
length (HL) and root length (RL) at each ploidy level (PL) on each day ....... 55

4-3. Linear regression analyses [L (.im) = a (.im) + b (.im mg-1) SW (mg), where seed
weight (SW) is the independent variable] between SW and the various characters
at each date and ploidy level. The data reported are the mean, regression equation,
and standard error (SE) of the b-value for crown diameter (CD), middle diameter
(MD), and hook diameter (HD) at each ploidy level (PL) on each day ........ 56

5-1. Mean squares and significance levels from the analyses of variance of hypocotyl
length (HL), root length (RL), crown diameter (CD), middle diameter (MD), and
hook diam eter (H D ) ............... ................... ..........66









5-2. Means of hypocotyl length (HL) in various chemical treatments at both ploidy
levels and between media and presoaked treatments 10 days after germination
initiation. HS = Hoagland solution. (-) indicates no HS added. (+) indicates HS
added .................... ............................... 67

5-3. Means of root length (RL) in various chemical treatments at both ploidy levels and
between media and presoaked treatments 10 days after germination initiation. HS
= Hoagland solution. (-) indicates no HS added. (+) indicates HS added ...... 68

5-4. Means of crown diameter (CD) in various chemical treatments at both ploidy levels
and between media and presoaked treatments 10 days after germination initiation.
HS = Hoagland solution. (-) indicates no HS added. (+) indicates HS added ... 69

5-5. Means of middle diameter (MD) in various chemical treatments at both ploidy
levels and between media and presoaked treatments 10 days after germination
initiation. HS = Hoagland solution. (-) indicates no HS added. (+) indicates HS
added .................... ............................... 70

5-6. Means of hook diameter (HD) in various chemical treatments at both ploidy levels
and between media and presoaked treatments 10 days after germination initiation.
HS = Hoagland solution. (-) indicates no HS added. (+) indicates HS added ... 71














Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

COMPARISONS OF SEED WEIGHT AND SEEDLING CHARACTERISTICS OF
DIPLOID AND AUTOTETRAPLOID RED CLOVER

By

Hideto Furuya

August, 2001


Chairperson: K.H. Quesenberry
Major Department: Agronomy


Seedling establishment is a critical aspect associated with the profitability of annual

forage production in Florida. The objective of this research was to evaluate the effect of

chromosome doubling on seed production and seedling characteristics of red clover

(Trifolium pratense L.), a cool-season forage legume adapted to winter-spring conditions

in north Florida. Autotetraploid (4x) populations were produced by nitrous oxide

treatment within the diploid (2x) cv. 'Cherokee.' Four clones of 2x and 4x seedlings from

each of eight crosses were grown in the field to increase seed with the 2x and 4x

populations physically isolated. Seeds were harvested from individual clones, and seed

number and weight of each genotype were determined at each ploidy level. The mean seed

number per plant for the 2x and 4x populations was 806 and 26, respectively. Mean seed

weight of the 4x was approximately 1.5 times that of the 2x. There was a ploidy level x









cross interaction effect, indicating that superior seed-producing genotypes at the 2x level

were different from those at the 4x level. Thereafter, seeds from all genotypes within each

ploidy level were composite into two populations. The 2x and 4x seeds were germinated

in growth pouches at 12, 20, and 280C in the dark to compare seedling growth of 2x and

4x populations. The seedling characteristics measured were total length, hypocotyl length,

root length, crown diameter, hook diameter, and middle diameter of the seedling between

the hook and crown. The characteristics were measured at four dates in each temperature.

The 4x had greater means for all seedling characters at any temperature. Especially, the 4x

hypocotyl mean was longer than the 2x. The increases in diameter may indicate greater

emerging forces of the 4x. The effect of seed weight on seedling growth at both ploidy

levels was tested at a constant temperature of 200C using regression analysis. Seed weight

was shown to influence all the growth characters measured. The influence of seed weight

was generally the same regardless of ploidy level. Thus, the greater seedling response

means of 4x were primarily due to heavier individual seed weights. The effect of various

chemical additions and combinations (Hoagland solution, indole-3-acetic acid, kinetin, and

sucrose) in the germination medium and presoaking treatments was studied to determine

their effect on seedling growth at both ploidy levels. Among the chemicals tested, only

sucrose had a positive effect on root development. These experiments indicated that the 4x

seeds were heavier and the 4x seedlings were longer in length and thicker in diameter. This

would suggest that, under field conditions, the use of 4x seeds would probably result in

superior establishment. However, the 4x seed fertility in the first generation was low

limiting practical use unless greater fertility can be obtained in successive generations.















CHAPTER 1
INTRODUCTION

General

Approximately 11.5 million acres of Florida are occupied by grasslands, 43.5% of

which are grazed forest lands, 30.4% planted in pasture and 26.1% in native range. The

importance of the grasslands is shown by the fact that these grasslands account for 95% of

beef cattle nutrition, 60% for goats, 40% for horses, and 10-15% for dairy. It is estimated

that the forage contributes approximately $380 million a year to animal livestock value and

the value of hay in Florida (L.E. Sollenberger, personal communication, 2001). As new

species and cultivars were introduced, the forage system has changed from native ranges

to planted pasture, and the livestock number doubled between 1940 and 1980 (Mislevy et

al., 1999).

Forage crops are primarily used for pasture, but also for stored feed. Grazing is

more convenient and costs less than half the amount of stored feed, which has additional

costs associated with hay and silage harvesting and feeding. One goal of forage production

in a grazing program is to extend the growing season of the crop. Since the price of stored

feed is increasing, producers should maximize the production of grazed forage. Depending

on the size of farm, producers should consider various combinations of forage

management and production (Ball et al., 1996).









2

Influence of Climate in Southeastern USA

Compared to central USA, southern USA is warmer and has reasonably distributed

rainfall, but occasional droughts restrict the growth of plants. There are frequent

thunderstorms during summer. The humid atmosphere prevents rapid loss of heat, and

there is little air movement in this region, resulting in conditions favorable for disease

development. The frequent rainfall during summer makes haymaking difficult, and quick

drying grasses are favored in summer production. Winter rainfall is caused by weather

frontal activity. Sudden polar air movements drop temperature occasionally, causing frost

or cold damage. Severe cold winter temperatures are a major factor determining the

survival and adaptation of warm-season perennial or tropical grasses. The general forage

scheme in the southern USA is warm-season perennial grasses supplemented by cool-

season annual grasses and legumes (Ball et al., 1996).

However, there is a critical shortage of feed from late fall through early spring in

north and northwest Florida, and in winter in south Florida. This shortage is primarily due

to the unpredictable weather patterns described above. Good cool-season forage

establishment and production can be readily accomplished during years with fall seasons

having moderate to cool temperature and adequate rainfall, but in years when fall seasons

are hot and dry, inadequate establishment of the cool-season forages can result.

Occasional and yearly variable frosts may also impair the establishment and production of

cool-season annual forages (Mislevy et al., 1999).

There are numerous solutions to the forage deficit problem of subtropical Florida.

Extending the growing season of warm-season forages, later in fall and earlier in spring,











can help limit the shortage of feed during winter time. Some new warm-season grass

cultivars such as 'Floralta' limpograss [Hemarthria altissima (Poir.) Stapf and C.E.

Hubb.] (Quesenberry et al., 1987) are able to produce forage under short days and cool

temperature. Winter production of small grains, cool-season legumes, and use of fall

produced peanut hay, are also options. Production and purchase of hay or deferring

grazing can increase the animal feed (Ball et al., 1996; Mislevy et al., 1999).

Various attempts have been made to reduce the cool-season forage production

problems by evaluation of different perennial forage species, but most perennial temperate

species do not tolerate hot summer weather and diseases. Use of corn (Zea mays L.) and

forage sorghum (Sorghum bicolor L. Moench) for silage production are possible

alternatives, but their production may be limited by the drought during late spring and

early summer. Increasing fuel costs discourage hay import as well as use of expensive

nitrogen fertilizer for grasses. There are also water use restrictions created by increasing

urban demand that has priority over irrigation for hay production. The high cost of

irrigation does not justify its return (Mislevy et al., 1999). Therefore, needs exist for the

development and use of more persistent cool season legumes to provide feed during the

winter season.

Red Clover

The plant used in this research is red clover (Trifolium pretense L.), which is a

self-incompatible, cross-pollinated species. It is a cool-season, short-lived perennial

legume that is believed to have originated in southeastern Europe and Asia Minor. This

species belongs to the genus Trifolium, which has approximately 230 other species.









4

Clovers per se have not only been used as legume crops, but also used as charms, religious

symbols, good luck symbols, emblems, food, medicine, and decoration (Taylor, 1985).

They certainly have some intrinsic value to humans. James Whitcomb Riley wrote a poem

of clover:

...And so I love clover it seems like apart
Of the sacerdest sorrows and joys of my hart;
And wherever it blossoms, oh, thare let me bow
And thank the good God as I'm thankin' Him now;
And I pray to Him still fer the stren 'th when I die,
To go out in the clover and tell it good-bye,
And lovin 'ly nestle my face in its bloom
While my soul slips away on a breth ofpurfume. (Manlove, 1982)

Forage Uses

Compared to grasses, clovers in general have higher nutritive value such as more

crude protein, digestibility, minerals, and vitamins. When mixed with grasses such as tall

fescue (Festuca arundinacea Schreb), they can reduce endophyte toxicity (Lacefield and

Ball, 2000). The animals that consume red clover are primarily beef, dairy, and sheep.

However, red clover can also cause antiquality problems such as bloat, isoflavones, and

slaframine if not managed properly (Taylor and Quesenberry, 1996).

Mode of Pollination

Because of the self-incompatible system, red clover depends on bees for cross

pollination. Honeybees (Apis mellifera L.) and bumblebees (Bombus spp.) are known to

be the principal pollinators. The latter are more effective in the pollination, but they are

not present in a large number in seed production fields. Five honeybee colonies ha-1

generally provide good seed production, but the long corolla tube of the flower reduces









5

the effectiveness especially in the tetraploid form (Rincker and Rampton, 1985). For seed

production, the planting pattern should be carefully designed, otherwise the pollinators

may pollinate other attracting plants nearby (Smith et al., 1985).

Nitrogen Fixation

The true clovers form mutual relationships with Rhizobium spp. In this symbiosis,

the host plant forms nodules on roots, which are filled with Rhizobium bacteroids. These

bacteroids obtain energy and nutrients for growth from the host plants, and, in turn, fix or

reduce atmospheric nitrogen (N2) into an organic form that is available to the plants.

Environmental effects such as temperature and water stress also influence the correct

specific interaction (Leonard and Dodson, 1933). In general, the host range of Rhizobium

is limited to legume plants, even though within the legumes, there exists host-species and

Rhizobium strain specificity. In the evolution of the symbiotic relationship with the host,

Rhizobium has lost the ability to fix N2 independently (Dilworth and Parker, 1969). The

process ofN2 fixation has been considered the second most important biochemical

reaction on earth after photosynthesis. As a result of N2 fixation, legumes depend less on

soil N as well as N fertilizers. The pasture and pulse legumes provide about 85% of N2

fixed in agricultural soils. The legume plants are a major source of human and animal

proteins, and are very important world crops (Vance and Johnson, 1981).

Red Clover in Florida

Red clover and other cool-season forage legumes are not well adapted to Florida

or the southeastern USA in general. This is due to drought, heat, soil acidity, low soil

fertility, pests, and other production requirements (Ball et al., 1996). Even though red









6

clover is a short-lived perennial plant in temperate regions, it grows primarily as an annual

plant in Florida (Chambliss and Quesenberry, 2000). Therefore, it is desirable to obtain

more adapted cultivars of this species in this region.

'Cherokee' red clover was developed at the University of Florida and released in

1991 (Quesenberry et al., 1993). Prior to that time, the red clover cultivars used in Florida

were developed in the Midwest USA, and had winter dormancy; therefore, they had slow

growth in spring (Chambliss and Quesenberry, 2000). Cherokee was developed through

the use of recurrent selection for early spring growth, vigorous early flowering, regrowth

vigor, early dry matter yield, and adaptation to the southeastern USA. Field selection in an

area with high root knot nematode (RKN) (Meloidogyne spp.) population also resulted in

increased resistance to RKN (Quesenberry et al., 1993). Comparisons of dry matter yields

of Cherokee with other cool-season legumes as well as other red clover cultivars in spring

indicated that it generally has superior first harvest and total seasonal yields (Chambliss

and Quesenberry, 2000).

Red clover, in general, can be grazed, made into green chop, or used as hay. It is

an upright bunch type clover without stolons or rhizomes. It has leafy stems, which arise

from a thick crown. The root system of this plant is a taproot with numerous adventitious

branches. Soil pH should range between 6 and 6.5. A moderate amount of water is

required for this plant, but red clover will not tolerate flooding conditions for a extended

period of time. Higher soil organic matter or clay content is desirable to keep the soil

moist (Chambliss and Quesenberry, 2000).

As a result of the limited winter growing season in Florida as well as its general











sandy soils and high temperatures, water shortages can be a problem for red clover

production. In addition, scarce rainfall in the fall seeding time can also be a critical

problem in the establishment of cool-season forage crops. Thus, the cool-season plants

need to have rapid seedling growth. Especially root elongation is helpful for effective

water uptake from the water stored deeper in the soil.

The size of seeds has an effect on early growth and development. This is because

heavier or larger seeds generally have more nutrition that is necessary for the development

of seedling growth (Black, 1959). This effect would be more pronounced in small-seeded

crops, and the seeds are generally sown at shallow depths. At this level, however, the

seeds are subjected to quickly drying soil conditions, compared with deeper levels.

Therefore, it is desirable to have heavier seed weights in small-seeded crops. Moon (1993)

reported increased seed weight of red clover following five cycles of half-sib family and

four cycles of mass selection methods. The former resulted in a 22.5% increase in seed

weight, while the latter resulted in a 9.2% increase. However, these changes did not lead

to differences in dry matter yield of 4-wk-old seedlings.

Polyploidy

Polyploidy is a multiple set of basic genomes. The change in ploidy levels indicates

a change of the number of basic genomes (x) from diploid (2x) to tetraploid (4x) or higher.

The number of chromosomes in a gamete is defined as n; thus, normal sporophytic tissue

is 2n. Diploid red clover has 14 somatic chromosomes (2n=2x=14), and tetraploid red

clover has 28 chromosomes (2n=4x=28). Polyploids can be allopolyploid or

autopolyploid. The allopolyploid is one that has two or more different basic genomes,











resulting from hybridization of genetically distant parents followed by chromosome

doubling. On the other hand, an autopolyploid has multiple copies of the same basic

genome because of chromosomes doubling within the same species (Schulz-Schaeffer,

1980).

In general, autotetraploids have greater vegetative volume, larger seed weight, and

adaptability to wider ecological region, but lower reproductive fertility than their diploid

counterparts. At the cellular level, the cytoplasm and nucleus of polyploids are larger than

those of normal plant. Chemical composition of the plants may also change. As the ploidy

level increases, the concentration of some chemical contents may increase (Swaminathan,

1970; Poehlman and Sleper, 1995; Li, 1976; Schulz-Schaeffer, 1980). In addition,

Stebbins (1947) noted that tetraploids often flower and fruit later than diploids. Levan

(1942) suggested that the slower growth of polyploids was caused by a slower mitotic

rhythm.

Generally, autopolyploids have reduced reproductive fertility compared to their

diploid counterparts. Stebbins (1947) listed three causes of reduced fertility: first,

"irregular chromosomal distribution caused by unequal separation of multivalents [such as

trivalents and quadrivalents]," second, "irregular distribution caused by meiotic

abnormalities of a physiological nature, presumably controlled genetically," and finally,

"genetic physiological sterility of an unexplained nature, but not associated with meiotic

irregularity" (Swaminathan, 1970; Schulz-Schaeffer, 1980).

Polyploidy is considered important in the evolutionary trends of plants. Around

30% to 50% of all angiosperms are estimated to be polyploids, 70% of grasses, and 23%











of legumes (Poehlman and Sleper, 1995). Moore et al. (1998) mentioned that polyploids

also tolerate colder temperatures, so the polyploid species are found in higher latitudes.

They can survive in the harsh environment where selection pressure is more intense. This

characteristic allows them to develop into permanent populations. Moore et al. (1998)

stated that allopolyploid was an immediate cause of speciation; that is, creation of a new

species. Despite the fact that autopolyploids produce fewer seeds, autopolyploidy can also

be a force in speciation.

Swaminathan (1970) summarized the ways polyploids contribute to evolutionary

processes among species. The genetic redundancy buffers small amounts of mutations as

well as adaptation without interfering in reproduction rate, and allows wider hybridization,

thus, increasing genetic variability. It would also influence gene actions and interaction.

More frequent bivalent pairings in later generations help keep the new ploidy level species

in existence, while maintaining higher chromosome numbers.

Followed by diploidization, the polyploidy can even promote greater number of

chromosomes in plants (Swaminathan, 1970). One species of fern Ophioglossum

reticulatum has as many as 1260 somatic chromosomes, but it behaves as a diploid

(Abraham and Ninan, 1954). Polyploidy is rare in gymnosperms and some woody

angiosperms, but their chromosome number is high. This may suggest they may have gone

through a polyploidization process. However, the recent evolutionary trend was not

favorable for them. The "changes in physiological and developmental rhythm often arising

from polyploidy seem to be of negative selective value" (Swaminathan, 1970).









10

However, the evolutionary trend is not always from lower to higher ploidy level. It

can be reversed (Hougas and Peloquin, 1958). Doubling chromosome numbers does not

always lead to large size and vigorous growth. Since the increase in size seems to have a

limitation after a certain increase in ploidy levels, there might be an optimal ploidy level in

each species. Welsh (1981) points out that there exists "a delicate balance in most plants

for numbers of chromosomes within each cell." Thus, it is necessary to test and evaluate

the effects of altering chromosome numbers of each species.

As for the effect of doubling chromosomes on the legume and Rhizobium trifolii

symbiosis, Weir (1961) compared the infection of three strains of Rhizobium on

autotetraploid and diploid red clover. He found that the nodule production on diploid

plants was 6 d faster than that on tetraploids. In both ploidy levels, 100% nodulation was

achieved 7 wk after the inoculation, but the number of nodules on diploids was greater

than that of tetraploids. The author indicated that the difference in the total number of

nodules between diploids and tetraploids was due to the faster rate of nodulation on

diploids between 22 and 30 d after planting. Also, the size of nodules on tetraploids was

larger than that of diploids. Wipf and Cooper (1940) found "disomatic cells" at the site of

infection in the normal diploid plants. Thus, the disomatic cells in tetraploid plants could

be octaploid. In one particular case, differences between nodulation on the diploid and

autotetraploid were pronounced (Weir, 1961).

Methods of Chromosome Doubling

The methods of chromosome doubling include: decapitation followed by callus

formation with or without indoleacetic acid, twin seedlings, cold and heat shocks,











colchicine, chloral hydrate, ether, chloroform, acenaphthene, phenylurethane, ethyl-

mercury-chloride, sulfanilamide, nitrous oxide (NzO), and others chemicals (Allard, 1966;

Briggs and Knowles, 1967). The formation of polyploids may also be genetically

controlled. If a plant has a homozygous recessive gene for asynapsis, it may produce

diploid (2n) gametes, instead of haploids, leading to autotetraploids (Allard, 1966; Briggs

and Knowles, 1967). In nature, the most frequent method of formation of polyploids was

speculated to be the union of unreduced gametes (Swaminathan, 1970).

The use of N20 for chromosome doubling results in higher percentage of

tetraploidy in Trifolium compared to the use of colchicine. The use of colchicine resulted

in 9% chromosome doubling (Neubauer and Thomas, 1966). However, the use of N20

increased doubling up to 100% (Berthaut, 1968). Previously, Taylor et al. (1976)

produced tetraploid red clover with N20. N20 was applied 24 h after pollination and

withdrawn after the subsequent 24 h. They obtained tetraploid plants, ranging from 50 to

100% among the crosses tested. Their results indicated the superiority of N20 use in

tetraploid induction on red clover.

Research Justification and Objectives

There are three justifications for this research. First, the economic importance of

red clover is primarily the vegetative part for forage, so reduced fertility associated with

tetraploidy is less of a concern than with grain crops. Second, the basic chromosome

number of red clover is low. The use of polyploid breeding in species with low

chromosome numbers is generally more successful (Poehlman and Sleper, 1995). In fact,

16% of the species in the genus Trifolium are polyploid (Cleveland, 1985). Taylor et al.











(1979) concluded that Trifolium species with higher chromosome numbers are cross

pollinators and perennial. This may suggest that red clover can possibly function as a

polyploid species in the genus. Finally, red clover is cross fertile (Levan, 1942; Bingefors

and Ellerstrom, 1964). This may lead to greater hybrid vigor at the multiple loci of

tetraploids than diploids.

The objective of this research was to examine the effects of chromosome doubling

in red clover seedling development to determine if the autotetraploid seedlings have the

potential for better establishment and consequently leading to the increased winter forage

production.














CHAPTER 2
THE EFFECT OF PLOIDY LEVEL ON SEED NUMBER AND WEIGHT

Introduction

Autotetraploid plants generally have thicker leaves and stems, increased winter

hardiness, slower growth, and increased seed weight but reduced seed number compared

to their diploid counterparts. The reduced fertility and lower seed numbers may be related

to pairing problems and unbalanced chromosome numbers in the gametes (Schulz-

Scharffer, 1980). The reduced seed numbers could be also caused by the changes in

morphological characteristics of the autotetraploid plants. Pollinating insects may not be

capable of obtaining pollen because of the longer corolla tubes. However, the increases in

seed volume and weight due to the gigantism from chromosome doubling, may overcome

the problem of low seed production (Swaminathan, 1970) or it could be possible to find

favorable environments for seed production of the tetraploids (Bingefors and Ellerstrom,

1964).

Literature Review

Bingefors and Ellerstrom (1964) conducted a cytological experiment to produce

tetraploid red clover (Trifolium pratense L.) using a colchicine solution applied to the

seeds of a local diploid red clover cultivar, Ultuna. This diploid cultivar and resulting

tetraploid cultivar, Ulva, were compared for their seed production in different

environments for 8 yr. Their results indicated that the tetraploids produced fewer seeds











than diploids. They suggested possible causes. Ulva (4x) produced a lower number of

flower heads per unit area; 25-30% lower than Ultuna (2x). Lower seed set could be

attributed to Ulva (4x) having about 0.74 mm longer corolla tube than Ultuna (2x),

resulting in poorer pollination by bees. Ulva (4x) had a narrower environmental range such

as a shorter flowering period, which may have contributed lower seed production. Valle et

al. (1960) also indicated that environmental factors, such as rainfall pattern, temperature,

relative humidity, and appropriate pollinators, can influence seed production. As for seed

weight, the tetraploid plants produced heavier seeds. The average weight of 1,000

tetraploid seeds was 2.75 g, whereas that of diploids was 1.78 g (Bingefors and

Ellerstrom, 1964).

In a case of rye (Secale cereale L.), Pfahler et al. (1987) found in their 4-yr study

that tetraploid cultivars had lower grain yield (1,080 kg ha-1) than that of comparative

diploids (1,850kg ha-1), lower seed filling per spikelet (62.4%) compared to the diploids

(74.3%), but higher mean seed weight (21.1 mg seed-') than diploids (18.8 mg seed-'). In

all cases, they found highly significant cultivar differences, suggesting the selection of more

fertile diploid lines for the induction of chromosome doubling could lead to more fertile

tetraploid lines.

Selection to improve autotetraploid fertility may be possible. Gilles and Randolph

(1951) selected for improved fertility in tetraploid corn (Zea mays L.) for 10 cycles. They

found lower incidences of multivalents in the last generation and more frequent incidences

of bivalents. Swaminathan and Sulbha (1959) also obtained a similar correlation in

Brassica campestris var. Toria after 19 generations. This tendency of bivalent pairing in









15

autotetraploids is called diploidization (Schulz-Schaeffer, 1980) or preferential pairing in a

case of complete intervarietal chromosome pairing, which could lead to higher and more

stable heterosis (Briggs and Knowles, 1967).

The objective of this experiment was to compare the effects of doubling the

chromosome number of red clover on seed number and mean seed weight in the generation

immediately after doubling. The hypothesis is that seed number will be lower for these

autotetraploids due to cytogenetic and morphological factors. However, the seed weight

would probably higher for the tetraploids than the diploids.

Materials and Methods

Production of Tetraploid Plants

All experiments were conducted using plants (FRU1, FRU2, FRU4, FRU5, FRU6,

FRU7, A155, and J041) representing advanced selection cycles of the diploid red clover

cv. 'Cherokee.' The first six were high yielding lines derived from field selection. The last

two were clones resistant to root knot nematodes (Meloidogyne spp).

These clones were crossed without emasculation, and after 24 h they were placed

into a nitrous oxide (N20) tank and maintained at 0.62 MPa for the next 24 h. N20 is

known to disturb the spindle fiber formation during the first embryonic division of the

zygote. Each chromosome is not pulled to the opposite poles, and remains in the middle of

the cell. In the next cell cycle when N20 is removed, normal mitosis should occur except

the cell contains double the chromosome number. From this N20 treatment, some

tetraploid and some diploid plants were obtained from the same cross. Plants produced

from the above treatment were classified as diploid or tetraploid based on size and









16

morphology of dry pollen (Taylor et al., 1976). Full-sib diploid and tetraploid plants from

the same cross on the same plant were used to establish corresponding diploid and

tetraploid populations.

Field Study

A diploid and a tetraploid plant resulting from each of eight crosses were

evaluated. Four ramets of each of the eight diploid and tetraploid plants were produced by

crown bud cuttings. The apical meristems were cut and treated with Rhizopon@ and

placed into a wet vermiculite tray to induce root formation. Each rooted ramet was

transferred to a pot to stabilize growth in the soil. Then, both tetraploid and diploid plants

were transplanted into the field at the University of Florida in May 2000; the different

ploidy groups were separated by approximately 2 km to avoid interpollination between

ploidy levels. Each field had four blocks, and the ramets were randomly distributed in each

of the four blocks. Naturally-occurring bees were allowed to cross-pollinate the plants

within and among the four blocks. The flower heads were harvested from an individual

ramet from each replication as they matured between June and August 2000.

Seeds were threshed by hand rubbing, and severely shriveled seeds were discarded.

After cleaning the seeds, the seed number was counted. The total seed weight of each

individual plant was measured. The mean seed weight was calculated by the total seed

weight divided by seed number.

The experimental design was a randomized block with four replications. As a result

of the heterogeneity of variance, seed number was transformed to base 10 logarithms.

Data were analyzed by analysis of variance as a 2 x 8 factorial. The minimum differences









17

for significance among means were compared using Duncan's Multiple Range Test for the

maximum number of means to be compared (Harter, 1960).

Results

Seed Number

There were effects of ploidy level (p<0.001) and the interaction between ploidy

level and cross (p<0.05), but no effect of cross (p>0.05) (Table 2-1). Overall crosses, the

mean numbers of seeds plant-' among diploid plants was 806, whereas that of tetraploid

plants was 26 (Table 2-2). Therefore, the tetraploid lines produced only about 3% as much

seed as the diploids. As an interaction effect was detected, cross 4 x A155 had lower seed

number than all the other crosses within the diploid level. On the other hand, A155 x 2 and

1 x A155 had the lowest seed number produced within the tetraploid level (Table 2-2).

Mean Seed Weight

There were significant effects (p<0.001) of cross, ploidy level, and the interaction

on mean seed weight (Table 2-1). The mean seed weight was higher for tetraploids than

diploids in all crosses (Table 2-3). The weight increases associated with the ploidy level

increase were not uniform. The cross with the greatest mean seed weight was 5 x A155 at

both ploidy levels, but the lowest mean seed weight was from 4 x A155 at the diploid level

and 4 x J041 at the tetraploid level (Table 2-3).

Discussion and Conclusions

From the results, the seed number of tetraploids was substantially lower than that

of the diploids and was dependent upon the genotype of the cross. However, it should be

remembered that the plant materials tested in this experiment were the first generation









18

after the N20 treatment, and that no selection had been practiced for tetraploid fertility or

balanced chromosome pairing. Additional generations of selection for fertility or

diploidization would improve seed set. Maximum fertility of autopolyploids was generally

attained after several generations of interpollination after doubling (Gilles and Randolph,

1951; Swaminathan and Sulbha, 1959; Briggs and Knowles, 1967).

Although the induction of autotetraploidy resulted in reduced fertility expressed as

lower seed number, some crosses appeared to be superior, suggesting a genetic component

for fertility. Also, good seed-producing diploid crosses did not necessarily produce high

fertility tetraploids. Thus, identification of high seed-producing diploid lines would not

necessarily result in high seed-producing tetraploids. Evaluation of tetraploid fertility from

other genotypes is needed in future research.

Overall, tetraploid seeds were heavier than diploid seeds. This could possibly be

the result of greater availability of photosynthates per seed throughout the growing season

for the tetraploids, compared to the need of the diploid seeds to divide assimilate among

many seeds. However, there was no correlation between seed number and seed weight.

So, the increase in the seed weight probably was due to the increased cell volume

associated with autotetraploidy. Even though the fertility of tetraploids will probably never

exceed that of diploids, the increased mean seed weight might overcome this problem by

increasing the total seed yield per unit area, as reported by Valle et al. (1960).

The mean seed weights at both ploidy levels obtained in this experiment were

considerably less than those reported by other authors (Bingefors and Ellerstrom, 1964;

Anderson, 1971). This was probably because this red clover seed was produced in the









19

summer season in Florida, which is generally adverse for optimal seed production. The

plants should have been planted as early as in February rather than in May to optimize seed

production in this subtropical region. This change would have probably produced more

vigorous growth, resulting in greater seed weight and numbers. The performance and

adaptability of the tetraploids especially for seed production should be evaluated at

different locations and years to determine the extent of the genotype effect.











Table 2-1. Mean squares and significance levels from the analyses of variance of seed
number plant-' (transformed) and seed weight.

Source of variation df Seed number Seed weight

Treatment 15 3.85*** 0.61***
Ploidy level (PL) 1 44.38*** 7.11***
Cross (CR) 7 0.76 0.18***
PL x CR 7 1.16* 0.11***
Error 39 0.42 0.02

*, *** F value significance at the 5 and 0.1% levels, respectively.











Table 2-2. Mean seed number plant-' from each cross at each ploidy level. The log 10
transformed mean is given after each mean in parentheses for statistical comparisons.

Cross Ploidy level Cross
means
2x 4x

A155 x 2 621 (2.70)t 3 (0.35) 312(1.52)
A155 x 6 1658 (3.19) 51 (1.65) 855 (2.42)
1 x A155 1062 (2.95) 2 (0.39) 532 (1.67)
4x A155 41 (1.67) 99(1.47) 70(1.57)
5x A155 871 (2.28) 2 (0.78) 437 (1.53)
4 x J041 1453 (3.06) 8 (0.89) 731 (1.98)
6x J041 471 (2.53) 6(0.66) 239(1.60)
J041 x 7 273 (2.33) 33 (1.20) 153 (1.77)

Ploidy level mean 806 (2.59) 26 (0.92)

t Minimum differences for significance among the means in the parenthesis were: ploidy
level means= F value significant at the 0.1% level; cross means= F value not significant
and; any combination of ploidy level and cross means= 1.12 and 1.48 at the 5 and 1%
levels, respectively.











Table 2-3. Mean seed weight from each cross at each ploidy level.

Cross Ploidy level Cross
mean
2x 4x
---------------------mg------------------ -----------------

A155 x 2 1.53 ? 2.17 1.85
A155 x 6 1.54 2.42 1.98
1 x A155 1.57 2.16 1.86
4xA155 1.22 2.23 1.72
5 x A155 1.74 2.60 2.17
4x J041 1.53 1.88 1.71
6x J041 1.59 2.09 1.84
J041 x 7 1.54 2.04 1.79

Ploidy level mean 1.53 2.20

tMinimum differences for significance among the means were: ploidy level means= F
value significant at the 0.1% level; cross means= 0.17 and 0.22 mg at the 5 and 1% levels,
respectively and; any combination of ploidy level and cross means= 0.25 and 0.33 mg at
the 5 and 1% levels, respectively.














CHAPTER 3
THE EFFECT OF TEMPERATURE ON DIPLOID AND TETRAPLOID RED
CLOVER SEEDLING GROWTH

Introduction

To improve seedling establishment, it is desirable to have rapid seedling and, more

importantly, root growth so that the root can remain in contact with moist soil while sown

at the soil surface. At the same time, greater shoot growth can enhance competition with

weeds, and thus maximizing growth of the developing seedling by photosynthesis. During

the period of establishment with grasses or seeded in a pure stand, the plants are subject to

intra- or inter-species competition. These plants compete for light, water, and soil

minerals. Those that intercept the light first would have an advantage in receiving light and

producing chemical energy for further growth of both shoots and roots (Kendall and

Stringer, 1985).

Literature Review

Seedling Characteristics of Clovers

Seedling growth begins with germination of the seed. Germination is defined as the

imbibition of water, resulting in "the rupture of the testa by an extruding radicle" (Black,

1959). However, many plants have mechanisms to prevent germination as a means to

ensure that proper environmental conditions exist for proper development after

germination. Seed coat hardness is one of them. Nevertheless, it can be broken by cold and











subsequent alternating temperature, microbial action, abrasive materials, sulfuric acid,

radio frequency electrical treatment, and other factors (Kendall and Stringer, 1985).

The type of germination and seedling development in red clover is called epigeal.

In this type, the radicle emerges from the seed, the first root absorbs water and nutrients

are transported to the hypocotyl. The hypocotyl elongates toward the soil surface with

cotyledons suspended by the hook (Moore et al., 1998). In the dark, seedling development

is entirely dependent on the cotyledons for energy. However, on exposure to light, the

cotyledons are capable of photosynthesis (Kendall and Stringer, 1985).

Seedling growth is divided into three phases: heterotrophic, transitional, and

autotrophic. During the heterotrophic phase, a plant is dependent on nutrient reserves in

the endosperm and cotyledons for energy, since it is not capable of photosynthesis. This

phase begins from germination and ends at the emergence of cotyledons from the soil.

Planting depth is important in this phase because the plant has to expend energy to

elongate the hypocotyl. During the transitional phase, the embryo obtains energy from

both the cotyledon and newly-emerging leaves. Finally, the autotrophic phase starts when

the plant becomes independent; that is, it can produce its own energy from photosynthesis.

Through these phases, seed weight, temperature, and seeding depth, are related to

competitiveness (Kendall and Stringer, 1985).

Temperature Effect on Seedling Growth

In diploid subterranean clover (Trifolium subterraneum L.), the effects of

temperature (7, 14, 21, and 28C) and depth of sowing (1.3, 2.5, 3.8, and 5.1 cm) on

seedling growth were examined by Black (1955), using pots. He found no differences in











total plant weight among seeds sown at any depth at the same temperature on any day

before emergence. The proportions of cotyledon, hypocotyl, and root, were the same

regardless of the sowing depth. However, there were differences in days of emergence

because of the depth at each temperature. At higher temperatures, the seedlings were able

to emerge from the soil quicker than at lower temperature, even though they were sown

more deeply. The seeds planted at 7C never emerged.

It was found that the emerged cotyledon weights decreased as the depth of sowing

increased. The transfer rate of nutrient from cotyledons was calculated from the loss of

cotyledon weight and the gains in root or hypocotyl weight. The stored energy in the

cotyledons is used for both respiration and growth of hypocotyl and root, but the

respiration cost was ignored because he assumed it had a relatively small effect. It was

found that regardless of the depth of seed sowing, 21 C germination temperature had the

greatest remaining cotyledon weights at emergence. That is, 21 C was the optimum

temperature for subterranean clover seedling growth (Black, 1955).

Temperature and depth effects were both important as well as the interaction effect

for seedling emergence. The interaction was due to the lack of parallelism of the response

to depth for temperatures of 14 and 21 C. The rate of decrease in cotyledon weight was

independent of the depth of sowing. Thus, there were no differences on any day between

plants sown at different depths in the relative portions of cotyledon and hypocotyl (Black,

1955).

Sullivan and Pfahler (1986) studied temperature effects (12, 20, and 28C) on

seedling growth in five diploid and corresponding tetraploid rye (Secale cereale L.)











populations. They found highly significant effects of temperature, ploidy, and genetic

background on shoot and root growth. There was also a temperature x genetic

background interaction, and temperature x genetic background x ploidy interaction effects

on shoot length. There was a temperature x ploidy interaction on root length. At both

ploidy levels, an increase in temperature resulted in increases in shoot growth and primary

root growth. The tetraploids had shoot growth of 12.0, 20.0, and 28.1 mm d-', whereas

those of diploids were 11.1, 18.4, and 27.0 mm d-1, at 12, 20, and 28C, respectively. The

primary root growth was always greater for the tetraploids than diploids by 16.7 vs. 15.0

mm d-1 at 12C, 26.9 vs. 25.8 mm d-1 at 20C, and 39.7 vs. 37.4 mm d-1 at 28C,

respectively. On the other hand, there were highly significant effects of ploidy,

temperature, and temperature x genetic background interaction on primary root cross-

sectional area. Increasing temperature resulted in a decrease in primary root cross-

sectional area at both ploidy levels, 0.12 mm2 for diploid and 0.17 mm2 for tetraploid at

12C, 0.10 and 0.15 mm2 at 20C, and 0.09 and 0.13 mm2 at 28C, respectively. Since

autotetraploids have a heavier seed weight, they should have advantages in seedling

competition compared to diploids (Swaminathan, 1970).

In this experiment, the seedling development of each ploidy level (2n=2x=14 and

2n=4x=28) of red clover (Trifolium pratense L.) was measured over a range of

temperatures. The hypothesis was that the increase in ploidy level increased rate and

length of seedling growth. Also, as temperature increases, the rate of growth would

increase within the ploidy level, since an increase in temperature would stimulate

enzymatic or respiratory activities.











Materials and Methods

Because of the shortage of tetraploid seeds, the diploid and tetraploid seeds of

various crosses (Chapter 2) were bulked together at each ploidy level for this and

subsequent experiments. The same crosses were used if at all possible but a number of

tetraploid seeds were added from crosses in which no diploid seeds were available. The

crosses and estimated seed number are shown in Table 3-1. As a result, there were two

populations tested: one diploid (2x) and the other tetraploid (4x). To increase the

germination percentage, both diploid and tetraploid seeds were scarified by immersion in

95% sulfuric acid solution for 5 min with thorough stirring. After the acid was removed,

the seeds were rinsed 5 times with distilled water.

Plastic pouches (Northrup King Co., Minneapolis, MN, Seed Pack Growth Pouch,

U. S. Patent 3241264) which control moisture level uniformly in and among the pouches,

were used in this experiment to measure seedling development. To enhance germination

uniformity, all seeds were soaked in distilled water for 24 h at 50C. Thereafter, the seeds

were germinated and grown in darkness. Each pouch contained 20 seeds. There were two

replications for each date. The pouches were inserted vertically into the slits of growth

cans. The replicated pouches were separated from each other within the same can. The

cans were stored in a temperature chamber at each constant temperature throughout the

experiment. The whole experiment was repeated twice.

According to the Association of Official Seed Analysts (1995), testing for

laboratory germination for red clover is conducted at a temperature of 20C. However, to

evaluate the effect of different temperatures on growth, the temperature ranges of 12, 20,









28

and 28C were chosen based on the preliminary information (data not shown). Preliminary

experiments were also conducted to determine the number of days required for seedling

growth under dark conditions at the above germination temperatures. The seedling

characteristics measured were total seedling length (TL), hypocotyl length (HL), root

length (RL), crown diameter (CD), hook diameter (HD), and middle point diameter

between crown and hook (MD). The measurements were made to the nearest mm.

Diameter measurements were made in mm converted to [im, using a microprojector at a

magnification of 50X. The dates of measurements were on 2, 4, 6, and 8 d at 280C, 3, 6,

9, and 12 d at 200C, and 7, 14, 21, and 28 d at 120C after germination initiation or after

the seeds were placed in the pouches. Day effect was included to examine the growth rate

over time.

The experimental units were dates of measurement and ploidy levels at each

temperature. The sampling units were the seeds at each temperature. From each pouch,

ten seedlings were measured. The experimental design was 2 x 4 factorial at each

temperature. The data were analyzed by analysis of variance. The minimum differences for

significance among means were compared using Duncan's Multiple Range Test for the

maximum number of means to be compared (Harter, 1960).

Results

Evaluation at 28C

Length measurement

There were effects of ploidy level on TL and HL (p<0.001) and on RL (p<0.05)

(Table 3-2). The means of the tetraploids were always greater than those of the diploids









29

for any character (Table 3-3). Day effects were also observed for all characters (p<0.001)

(Table 3-2). That is, differences in the means among the days were detected. However,

days 6 and 8 were shown to be not significantly different for any character. The elongation

of TL and HL appeared to stop after about day 6. In addition, day 4 was not significantly

different from day 6 or day 8 for RL. The root elongation appeared to stop after about day

4 (Table 3-3).

The ploidy x day interaction effect was not significant except for HL (p<0.05)

(Table 3-2). Initially, the tetraploids had greater TL means than diploids at day 2, and the

differences between ploidy levels were maintained throughout the rest of the period (Table

3-3). Therefore, their growth rates were parallel. As for HL, the mean differences between

the diploids and tetraploids widened from 4, 5, 8, to 9 mm, at days 2, 4, 6, and 8,

respectively (Table 3-3). Therefore, the interaction effect was from the progressive

increase of HL by days.

Diameter measurement

There were effects of ploidy levels for CD, MD, and HD (p<0.001) (Table 3-4).

The tetraploids always had greater means than those of diploids (Table 3-5). The day

effect was only found for HD (p<0.001), but not for CD or MD. The interaction effect

was only found for MD, but not on CD and HD (Table 3-4).

Evaluation at 20C

Length measurement

There were significant effects of ploidy level on TL, HL, and RL (p<0.001) (Table

3-6). The tetraploids always had greater means than those of diploids (Table 3-7). Day











effects were also found for all the responses (p<0.001) (Table 3-6). The day effects

indicated that the mean lengths of TL, HL, and RL, were different among the given days.

For all characters, day 12 had the greatest means, but they were not significantly different

from those of day 9 (Table 3-7).

The effect of interaction between ploidy level and day was only expressed for TL

(p<0.05) (Table 3-6). Although the TL means increased as days increased, those of the

diploids stopped increasing at day 9, whereas the means of tetraploids continued to

increase (Table 3-4). Therefore, the differences in TL means between ploidy levels at each

day increased.

Diameter measurement

There were effects of ploidy level on CD, MD, and HD (p<0.001) (Table 3-8).

The tetraploids had larger means than those of diploids (Table 3-9). Unlike at 28C, the

day effect was expressed in MD (p<0.01), but not in CD and HD (Table 3-8). In general,

the means of MD decreased as days increased. The MD of earlier days were significantly

greater than those of later days (p<0.05) (Table 3-9). No interaction effects were

observed (Table 3-8). At both ploidy levels, the means of CD and HD did not change

significantly across days (Table 3-9).

Evaluation at 12C

Length measurement

There were effects of ploidy level on TL, HL, and RL (p<0.001) (Table 3-10). The

tetraploids always had greater means than those of diploids (Table 3-11). Day effects were

also found for all the responses (p<0.001) (Table 3-10). The day effects indicated that the











mean lengths of TL, HL, and RL, were different among the given days. The means

increased as the days progressed until day 21 for all responses, but the means of day 28

were not different from that of day 21 for TL and RL. In HL, the mean continued to

increase until the end of the measurement period (Table 3-11).

The interaction effect between ploidy level and day was only expressed for HL

(p<0.05), but not for TL and RL (Table 3-10). In HL, the increase of tetraploids between

day 7 and day 14 were significantly greater from that of diploids (Table 3-11).

Diameter measurement

There were effects of ploidy level on CD, MD, and HD (p<0.001) (Table 3-12).

The tetraploids had the greater means than those of diploids (Table 3-13). Similar to the

results at 20C, the day effect was expressed for MD (p<0.001), but not in CD and HD

(Table 3-12). The means of MD decreased as the days progressed (Table 3-13). Also

similar to the results at 20C, no interaction effects were expressed in any of the characters

(Table 3-12).

Discussion and Conclusions

The tetraploid seedlings always had greater length and diameter than those of

diploids at all germination temperatures. Also, as temperature increased, the rate of

seedling growth increased at both ploidy levels. These results agreed with Black (1955)

who found that as temperature increased, emergence was improved within the range of

temperatures. Even though the growing environment in this experiment was different from

his, when the diploid and tetraploid seeds were grown within this temperature range, the









32

tetraploid materials are expected to show greater seedling vigor and emergence than that

of diploids.

The ploidy level effects were pronounced in all response variables measured. The

tetraploid seedlings had longer hypocotyls and longer roots than the diploids. However, it

was observed during the experiment that at lower temperatures, the seedlings tended to

achieve longer hypocotyl than those at higher temperatures within the range of

temperature, though the reason is unknown. This may be because of the assimilate priority

of the cotyledon reserve to the hypocotyl during the period of the allocation or because of

negative gravitropic growth.

Even though significant differences in diameter measurement were found, the

differences between the ploidy levels were generally greater than those among days. The

day effect on HD at 280C (Table 3-4) seemed to be random in nature. However, the day

effects on MD (Table 3-8 and 3-14) suggest the elongation of red clover seedlings is due

to cell elongation rather than cell division. The relative role of cell division and elongation

in hypocotyl length was discussed with the conclusion that cell elongation was the major

factor (Jones and Moll, 1983). The hypocotyl elongation of lettuce (Lactuca sativa L.),

which also has the epigeal germination pattern, was shown to be caused by cell elongation.

Therefore, a major part of the increase in hypocotyl length in the tetraploids was probably

the result of the larger cell size of the tetraploids compared to that of diploids.

In general, when a seedling must emerge through a compacted soil, a plant

experiences a thickening of the seedling or root. This retards the growth, but it helps the

seedling push up through the compacted soil (Mayer and Poljakoff-Mayber, 1989). The









33

tetraploid seedlings were shown to have both of the positive characteristics, faster growth

and larger diameter for greater emergence ability. The latter characteristic should be

especially important by the increased HD, since the hook area of the hypocotyl protects

the cotyledon as well as meristematic region during the penetration toward the soil

surface.

The diploid and tetraploid red clover populations used in this study were

genetically similar, and this experiment showed the clear ploidy differences in the

expression of the seedling growth. Overall, the lack of interaction indicated a parallel

growth of diploid and tetraploid seedlings. The initially-greater means of the tetraploids

may imply a more rapid germination of tetraploid seeds compared to that of the diploids.

Variations in length at later evaluation dates were found probably because of the

heterozygosity of red clover, even though there was less variation in the initial period.











Table 3-1. The crosses and the estimated seed numbers composite to develop the diploid
(2x) and tetraploid (4x) populations used in this and subsequent studies.

Cross 2x 4x

155x1 32
155x2 2419 13
155x6 18545 997
1x 155 7583 7
2x 155 -52
3 x 155 -36
4x155 144 456
5x155 3108 6
6x 155 -983
7x 155 -233
4xJ041 5203 32
6xJ041 3412 48
J041x7 971 537
J054 x 3 420
MI119x4 197











Table 3-2. Mean squares and significance levels from the analyses of variance of total
seedling length (TL), hypocotyl length (HL), and root length (RL) at 28C.

Source of variance df Length character

TL HL RL

Treatment 7 8975*** 5857*** 350***
Ploidy level (PL) 1 5208*** 3699*** 129*
Day(D) 3 19022*** 12326*** 724***
PLxD 3 183 108* 49
Error 312 70 34 32

*, *** F value significant at the 5 and 0.1% levels, respectively.











Table 3-3. Means of total seedling length (TL), hypocotyl length (HL), and root length
(RL) at each ploidy level (PL) at 28C on each of four days after germination initiation.

Day after germination initiation
Character PL 2 4 6 8 PL mean

------------------------mm-------- -------------------
TLt 2x 43 68 73 73 64
4x 49 73 84 84 72

Day mean 46 70 78 78


HL 2x 19 38 43 43 36
4x 23 43 51 52 43

Day mean 21 41 47 47


RL 2x 24 30 29 31 28
4x 25 29 33 31 30

Day mean 25 29 31 31

tMinimum differences for significance among the means were: ploidy level means = F
value significant at the 0.1% level; day means = 3 and 4 mm at the 5 and 1% levels,
respectively and; any combination of ploidy level and day means = 4 and 6 mm at the 5
and 1% levels, respectively.
Minimum differences for significance among the means were: ploidy level means = F
value significant at the 0.1% level; day means = 2 and 3 mm at the 5 and 1% levels,
respectively and; any combination of ploidy level and day means = 3 and 4 mm at the 5
and 1% levels, respectively.
Minimum differences for significance among the means were: ploidy level means = F
value significant at the 5% level; day means F value =2 and 3 mm at the 5 and 1% levels,
respectively and; any combination of ploidy level and day means = 3 and 4 mm at the 5
and 1% levels, respectively.









37

Table 3-4. Mean squares and significance levels from the analyses of variance of crown
diameter (CD), middle diameter (MD), and hook diameter (HD) at 28C.

Source of variance df Diameter character
CD MD HD

Treatment 7 191966*** 200351*** 174962***
Ploidy level (PL) 1 1292861*** 1357205*** 1128125***
Day(D) 3 14825 6182 25088***
PLxD 3 2141 8902* 7115
Error 312 5853 3020 2936

*, *** F value significant at the 5 and 0.1% levels, respectively.












Table 3-5. Means of crown diameter (CD), middle diameter (MD), and hook diameter
(HD) at each ploidy level (PL) at 28 C on each of four days after germination initiation.

Day after germination initiation
Character PL 2 4 6 8 PL mean
----------------------------m------------------ --------
CDt 2x 760 781 787 776 776
4x 889 901 928 894 903

Day mean 825 841 857 835


MD\ 2x 665 663 640 651 655
4x 796 778 800 767 785

Day mean 730 721 720 709


HD 2x 537 573 566 567 561
4x 656 678 712 673 679

Day mean 590 625 639 620

tMinimum differences for significance among the means were: ploidy level means=F value
significant at the 0.1% level; day means = F value not significant and; any combination of
ploidy level and day means = 39 and 51 plm at the 5 and 1% levels, respectively.
Minimum differences for significance among the means were: ploidy level means=F value
significant at the 0.1% level; day means = F value not significant and; any combination of
ploidy level and day means = 28 and 37 plm at the 5 and 1 % levels, respectively.
Minimum differences for significance among the means were: ploidy level means=F value
significant at the 0.1% level; day means = 18 and 24 plm at the 5 and 1% levels,
respectively and; any combination of ploidy level and day means = 28 and 36 plm at the 5
and 1% levels, respectively.











Table 3-6. Mean squares and significance levels from the analyses of variance of total
seedling length (TL), hypocotyl length (HL), and root length (RL) at 20C.

Source of variance df Length character
TL HL RL

Treatment 7 7393*** 5213*** 260***
Ploidy level (PL) 1 3768*** 1144*** 760***
Day(D) 3 15810*** 11722*** 316***
PLxD 3 184* 60 38
Error 312 60 27 26

*, *** F value significant at the 5 and 0.1% levels, respectively.












Table 3-7. Means of total seedling length (TL), hypocotyl length (HL), and root length
(RL) at each ploidy level (PL) at 20C on each of four days after germination initiation.

Day after germination initiation
Character PL 3 6 9 12 PL mean
---------------- mm----------- ------------
TLt 2x 44 64 71 71 63
4x 47 72 78 81 70

Day mean 46 68 75 76


HL 2x 22 40 46 46 39
4x 24 45 50 51 43

Day mean 23 43 48 49


RL 2x 22 23 25 25 24
4x 23 27 28 30 27

Day mean 23 25 27 27

tMinimum differences for significance among the means were: ploidy level means = F
value significant at the 0.1% level; day means = 3 and 4 mm at the 5 and 1% levels,
respectively and; any combination of ploidy level and day means = 4 and 5 mm at the 5
and 1% levels, respectively.
Minimum differences for significance among the means were: ploidy level means = F
value significant at the 0.1% level; day means = 2 and 3 mm at the 5 and 1% levels,
respectively and; any combination of ploidy level and day means = 3 and 4 mm at the 5
and 1% levels, respectively.
Minimum differences for significance among the means were: ploidy level means = F
value significant at the 5% level; day means F value =2 and 3 mm at the 5 and 1% levels,
respectively and; any combination of ploidy level and day means = 3 and 4 mm at the 5
and 1% levels, respectively.









41

Table 3-8. Mean squares and significance levels from the analyses of variance of crown
diameter (CD), middle diameter (MD), and hook diameter (HD) at 20C.

Source of variance df Diameter character
CD MD HD

Treatment 7 193500*** 247368*** 166092***
Ploidy level (PL) 1 1333861*** 1682000*** 1139031***
Day(D) 3 5998 12083** 6255
PL x D 3 881 4443 1615
Error 312 6051 2927 3123

**, *** F value significant at the 1 and 0.1% levels, respectively.












Table 3-9. Means of crown diameter (CD), middle diameter (MD), and hook diameter
(HD) at each ploidy level (PL) at 20C on each of four days after germination initiation.

Day after germination initiation
Character PL 3 6 9 12 PL mean
-----------------------------------------------------
CDt 2x 782 765 761 782 772
4x 912 898 895 901 901

Day mean 847 831 828 841


MD\ 2x 685 669 646 653 663
4x 808 823 802 799 808

Day mean 746 746 724 726


HD 2x 583 577 568 556 571
4x 708 704 690 698 700

Day mean 645 640 629 627

tMinimum differences for significance among the means were: ploidy level means=F value
significant at the 0.1% level; day means = F value not significant and; any combination of
ploidy level and day means = 40 and 52 plm at the 5 and 1% levels, respectively.
Minimum differences for significance among the means were: ploidy level means=F value
significant at the 0.1% level; day means = 18 and 24 plm at the 5 and 1% levels,
respectively and; any combination of ploidy level and day means = 28 and 36 plm at the 5
and 1% levels, respectively.
Minimum differences for significance among the means were: ploidy level means=F value
significant at the 0.1% level; day means = F value not significant and; any combination of
ploidy level and day means = 29 and 37 pLm at the 5 and 1% levels, respectively.











Table 3-10. Mean squares and significance levels from the analyses of variance of total
seedling length (TL), hypocotyl length (HL), and root length (RL) at 12C.

Source of variance df Length character
TL HL RL

Treatment 7 7499*** 5849*** 149***
Ploidy level (PL) 1 4183*** 1975*** 410***
Day(D) 3 15947*** 12854*** 184***
PLxD 3 156 136* 28
Error 312 62 46 15

*, *** F value significant at the 5 and 0.1% levels, respectively.












Table 3-11. Means of total seedling length (TL), hypocotyl length (HL), and root length
(RL) at each ploidy level (PL) at 12C on each of four days after germination initiation.

Day after germination initiation
Character PL 7 14 21 28 PL mean
---------------- mm----------- ------------
TLt 2x 41 60 67 68 59
4x 44 67 75 78 66

Day mean 42 63 71 73


HL 2x 26 42 48 51 42
4x 27 48 55 58 47

Day mean 26 45 51 54


RL 2x 15 18 19 17 17
4x 18 19 21 21 19

Day mean 16 18 20 19

tMinimum differences for significance among the means were: ploidy level means = F
value significant at the 0.1% level; day means = 3 and 4 mm at the 5 and 1% levels,
respectively and; any combination of ploidy level and day means = 4 and 5 mm at the 5
and 1% levels, respectively.
Minimum differences for significance among the means were: ploidy level means = F
value significant at the 0.1% level; day means = 2 and 3 mm at the 5 and 1% levels,
respectively and; any combination of ploidy level and day means = 3 and 5 mm at the 5
and 1% levels, respectively.
Minimum differences for significance among the means were: ploidy level means = F
value significant at the 5% level; day means F value =1 and 2 mm at the 5 and 1% levels,
respectively and; any combination of ploidy level and day means = 2 and 3 mm at the 5
and 1% levels, respectively.









45

Table 3-12. Mean squares and significance levels from the analyses of variance of crown
diameter (CD), middle diameter (MD), and hook diameter (HD) at 12C.

Source of variance df Diameter character
CD MD HD

Treatment 7 182617*** 262203*** 175773***
Ploidy level (PL) 1 1247501*** 1690711*** 1212781***
Day(D) 3 2165 42581*** 3551
PLxD 3 8108 5655 2325
Error 312 7618 3819 4531

*, *** F value significant at the 5 and 0.1% levels, respectively.












Table 3-13 Means of crown diameter (CD), middle diameter (MD), and hook diameter
(HD) at each ploidy level (PL) at 12C on each of four days after germination initiation.

Day after germination initiation
Character PL 7 14 21 28 PL mean
----------------------------m-------------------------
CDt 2x 775 753 785 762 768
4x 896 895 883 900 893

Day mean 835 824 834 831


MD\ 2x 677 636 639 612 641
4x 814 792 766 774 786

Day mean 745 714 702 692


HD 2x 557 547 563 549 554
4x 677 665 679 688 677

Day mean 617 606 621 619

tMinimum differences for significance among the means were: ploidy level means=F value
significant at the 0.1% level; day means = F value not significant and; any combination of
ploidy level and day means = 45 and 58 plm at the 5 and 1% levels, respectively.
Minimum differences for significance among the means were: ploidy level means=F value
significant at the 0.1% level; day means = 21 and 27 plm at the 5 and 1% levels,
respectively and; any combination of ploidy level and day means = 32 and 41 plm at the 5
and 1% levels, respectively.
Minimum differences for significance among the means were: ploidy level means=F value
significant at the 0.1% level; day means = F value not significant and; any combination of
ploidy level and day means = 35 and 45 pLm at the 5 and 1% levels, respectively.














CHAPTER 4
THE EFFECT OF SEED WEIGHT ON SEEDLING GROWTH OF DIPLOID AND
TETRAPLOID RED CLOVER

Introduction

The subtropical climate of the southeastern USA, as well as unpredictable patterns

of autumn weather, often cause a problem in establishment of forage grasses and legumes.

Drought conditions of this area are a primary restriction of seedling growth and

establishment (Mislevy et al, 1999; Ball et al. 1996). Seed weight is known to influence

seedling growth as well as subsequent early vegetative growth. Black (1959) summarized

some agronomic characteristics of legume seeds. There seems to be a positive relationship

between seed weight and seedling growth. Large-seeded species or heavier seeds had

greater emergence as well as they were able to emerge from deeper planting depths under

adequate moisture availability.

Literature Review

Seed weight is correlated with the ability of the seed to push the cotyledons

through the soil surface, thus, allowing normal seedling development to progress. Williams

(1956) measured the strength of emerging hypocotyls of subterranean clover (Trifolium

subterraneum L.), crimson clover (T. incarnatum L.), rose clover (T. Hirtum All.), and

alfalfa (Medicago sativa L.). Among the species, he found that the large-seeded

subterranean clover had the highest force, and that small-seeded alfalfa was lowest. The









48

correlation coefficient found was 0.99. He also found a correlation (r = 0.84) between the

emergence force and weight of hydrolyzable carbohydrates in the seeds.

Black (1957) showed the effect of seed weight on early seedling weight, using

three strains of subterranean clover grown in pots. He found a positive relationship

between seed or embryo weight and seedling growth, independent of strains. One strain of

this species had greater early dry matter production than the others, but it was found that

the strain had the greater seed weight on average. The same weight of seeds from different

strains had similar dry weight production at the early stages. The only physiological

differences found were the number and area of the leaves. The heavier seed strain resulted

in larger but fewer leaves. The total leaf area and the relative growth rate remained similar

among the strains.

Black (1956) examined the effect of seed weight on emergence from different

depths, 1.3, 3.2, and 5.1 cm in subterranean clover. When seeds of the same weight were

used, depth of planting had no effect on the weight of seedlings. The weight of cotyledons

of emerging seedlings was shown to decrease, as the length of hypocotyls increased.

However, the lower weight subterranean clover seeds had delayed emergence from the

soil at the 3.2 cm depth, compared to other larger size seeds, and this size seeds failed to

emerge from a 5.1 cm depth. The effects of seed weight and planting depth were

significant, and no interaction effect was found. The maximum hypocotyl elongation was

also measured at each seed weight, by placing the seedling in the dark after emergence.

The smallest seeds (mean of 3.0 mg) had the shortest elongation of 37 mm, and the mean

seed weights of 5.0 and 8.0 mg had 52 and 67 mm elongation, respectively.









49

Black (1956) also examined the effect of seed size on initial leaf area during 30 d.

There was a positive relationship between them, but no interaction effect. The growth rate

was concluded to be the same regardless of seed weight. As for planting depth, there was

no effect on leaf production. It was shown that the cotyledon areas were independent of

cotyledon weight. That is, the cotyledon weights decreased as hypocotyls grew, but the

cotyledon areas remained identical to the initial areas. Since the cotyledons function as a

storage as well as the first photosynthetic tissue, there is an advantage to having a larger

cotyledon area for initial photosynthate production. Therefore, for the species that have

epigeal germination, it is advantageous to having larger seed weight or size to emerge

from deeper soils in addition to having better early vegetative growth.

Anderson (1971) compared the seedling growth of diploid and tetraploid red

clover (T. pratense L.) growing in pots at the first harvest (32 d from sowing). The lowest

mean temperature was 90C (fluctuating 2 to 170C), and the highest mean temperature was

210C (13 to 270C) during the experimental period. There was a highly significant ploidy

effect. The tetraploid seeds (mean = 3.4 mg seed-') had more dry matter production than

diploids (mean = 2.1 mg seed-') at the first harvest (25 mg vs 16 mg, respectively). At the

second harvest (39 d from sowing), tetraploids had dry matter of 77 mg compared to 48

mg for the diploids. The relative shoot growth rate was not significantly different. He also

observed longer petiole length of tetraploids than diploids, but a smaller leaf number for

the tetraploids.

Anderson (1971) also separated two sizes of seed populations at each ploidy level,

though the smaller tetraploid mean was heavier (2.71 mg) than larger diploid population









50

(2.01 mg). Within each ploidy level, the heavier population produced more dry matter at

the first (28 days from sowing) and second (38 days) harvests. Yet, the relative shoot

growth rate was not different. The relationship between seed weight and seedling growth

between ploidy levels seemed to follow that described by Black (1959) at the normal

diploid level.

Since the weight of red clover seeds was increased from doubling chromosome

number, the tetraploids should have greater seedling growth than that of diploids. If the

relationship between seed weight and seedling growth was verified, there would be greater

potential to sow the heavier seeds, or tetraploid seeds, deeper to avoid desiccation at or

closer to the soil surface as well as to remain in contact with a receding soil moisture

front, thus increasing water use efficiency. The objective of this research was to evaluate

the effect of seed weight at each ploidy level on seedling growth. The hypothesis was that

an increase in seedling growth occurs as the seed weight increases regardless of ploidy

levels.

Methods and Materials

The seedling growth procedure was described in Chapter 3, but in this study the

seeds were not presoaked. Four hundred seeds at each ploidy level were individually

weighed to 0.01 mg, maintaining the identities of seeds with their corresponding weight

throughout the experiment. The temperature in the growth chamber was maintained at

20C. The growth measurements at 4, 7, 10, and 13 d were recorded. The seedling

characteristics measured included total seedling length (TL), hypocotyl length (HL), root

length (RL), crown diameter (CD), hook diameter (HD), and middle point diameter











between crown and hook (MD). At each day of measurement, the seedlings from 100

weighted seeds at both ploidy levels were measured. The seed weight distributions within

the ploidy level were not significantly different among days. During the experiment,

obviously abnormal seedlings were discarded. A regression of seed weight against all the

response variables described above at various dates was conducted to determine whether

slope (b-value) at each ploidy level was different from zero. The slope differences between

ploidy levels were compared using Student's t-test. Each response variable at both ploidy

levels on each day was compared using the F-test.

Results

The total seed weight distributions at both ploidy levels are shown in Table 4-1.

The mean tetraploid seed weight (2.28 mg) was greater than that of diploids (1.54 mg) by

approximately 1.5 times. The seed weight distributions were overlapping. The seed weight

distributions at both ploidy levels were normal as shown by Shapiro and Francia (1972)

(Table 4-1). Therefore, the seed weight distributions of the samples in any population

were statistically equal.

The linear regression equations were shown in Table 4-2 for length characters and

in Table 4-3 for diameter characters. All the b-values were positive and significantly

different from zero. Therefore, an increase in seed weight would lead to an increase in the

values of any character at both ploidy levels. The differences in b-values between ploidy

levels at the same dates for each character were also tested using the t-test, shown in the

same tables. Overall, there were few significant differences in b-values between the ploidy

levels with occasional exceptions. When differences were detected, the values of diploids











tended to be greater than those of tetraploids. For the length characters, no additional

growth was obtained after 10 d (Table 4-2).

Discussions and Conclusion

From the results, all the responses were found to be related to seed weight

regardless of the ploidy levels. Increasing seed weights led to an increase in the values of

all the characteristics at both ploidy levels. This would explain the greater mean growth

measurements of tetraploids than diploids, since the mean seed weight of the former was

greater than the latter. Therefore, the hypothesis that an increase in seed weight within

each ploidy level results in the increase in seedling growth, was supported by these results.

Selection to increase population seed weight would probably improve seedling growth at

both ploidy levels.

The finding of no significant differences in b-values between ploidy levels indicated

that each increase in seed weight led to the same increase in growth. However, the mean

seed weight or the weight distributions were not the same, the mean tetraploid seedlings

had greater means values of any characteristics measured. This result also supported the

advantage of the heavier seeds in small-seeded species, at least in the early growth or

seedling stage.

The b-values of diploids were occasionally greater than those of tetraploids.

However, since the weight distribution or mean weight of tetraploids were larger than

those of the diploids, the differences in seed weight offset the expected higher rate of

increase of diploids.









53

The advantage of heavier seeds within each ploidy levels was also supported in this

study. This experiment indicated that the induction of autotetraploidy resulted in heavier

seeds compared to the diploid. It would be more efficient than selection for increasing

seed weight at a diploid level. Since tetraploids are generally associated with a greater cell

volume, the length and diameter of the tetraploid seedlings possibly could elongate to a

greater extent, and the tetraploid seedlings had greater length and diameter than the

diploids. Thus, the heavier or tetraploid seeds should result in better seedling growth and

establishment of red clover.











Table 4-1. The seed weight mean, standard error (SE), standard deviation (SD), maximum
and minimum values, normality test, and total sample number of diploid (2x) and tetraploid
(4x) populations.

2x 4x

Seed weight mean (mg)t 1.54 2.28
SE 0.01 0.02
SD 0.23 0.29
Maximum (mg) 2.15 3.05
Minimum (mg) 0.83 1.49
Normality test (W values)$ 0.9949 0.9965
Total sample number 363 381

t The seed weight means of diploids and tetraploids were significantly different at 1%
level.
1 The W values were not statistically significant indicating normal distribution in both
populations (Shapiro and Francia, 1972).












Table 4-2. Linear regression analyses [L (mm) = a (mm) + b (mm mg-') SW (mg), where
seed weight (SW) is the independent variable] between SW and the various characters at
each date and ploidy level. The data reported are the mean, regression equation, and
standard error (SE) of the b-value for total seedling length (TL), hypocotyl length (HL)
and root length (RL) at each ploidy level (PL) on each day.
Character Date PLt Mean j Regression equation SE of b t-value
--mm--


53*
56
70***
78
76***
82
75***
83

36
36
52**
55
56***
60
56***
60

17***
20
18***
23
20*
22
19***
23


27 + 16 SW***
18 + 16 SW***
34 + 24 SW***
28 + 22 SW***
35 + 26 SW***
51 + 13 SW***
38 + 24 SW***
44 + 17 SW***

20 + 10 SW**
15 + 9 SW***
29 + 15 SW***
29 + 11 SW***
28 + 18 SW***
38 + 9 SW***
33 + 15 SW***
36 + 11 SW***

8+ 6 SW***
3 + 7 SW***
5+ 9 SW***
-1 +10 SW***
7 + 8 SW***
13 + 4 SW*
5+ 9 SW***
9+ 6 SW***


0.09

0.52

3.79***

2.01*


0.53

1.27

3.11**

1.61


0.92

0.82

2.21*

1.31


tFor all characters at each date at each ploidy level, the number of pairs measured (n)
were: date 4=89 and 93, date 7=91 and 95, date 10=92 and 97, and date 13=91 and 96 for
diploid and tetraploid, respectively.
Ploidy means at each day and for each character were tested using a F-test.
*, **, *** F-value significant at 5, 1, and 0. % levels, respectively.
*, **, *** a- and b-values significantly different from zero at 5, 1, and 0.1% levels,
respectively.
T The differences between the b-values of each ploidy level at each day and for each
character were tested using a t-test.
*, **, *** t-value significantly different at 5, 1, and 0.1% levels, respectively.


4 2x
4x
7 2x
4x
10 2x
4x
13 2x
4x

4 2x
4x
7 2x
4x
10 2x
4x
13 2x
4x

4 2x
4x
7 2x
4x
10 2x
4x
13 2x
4x











Table 4-3. Linear regression analyses [L (ptm) = a (ptm) + b ([tm mg-') SW (mg), where
seed weight (SW) is the independent variable] between SW and the various characters at
each date and ploidy level. The data reported are the mean, regression equation, and
standard error (SE) of the b-value for crown diameter (CD), middle diameter (MD), and
hook diameter (HD) at each ploidy level (PL) on each day.
Character Date PLt Mean j Regression equation SE of b t-value
--[m--
CD 4 2x 876*** CD = 644 +148 SW*** +33 2.40*
4x 1000 CD = 832 + 73 SW* + 30
7 2x 862*** CD = 673 +124 SW** +44 0.07
4x 989 CD = 713 +121 SW*** +31
10 2x 862*** CD = 543 +208 SW*** +35 1.95
4x 974 CD = 634 +148 SW*** +27
13 2x 867*** CD = 685 +118 SW*** +32 0.12
4x 981 CD = 706 +122 SW*** +31

MD 4 2x 700*** MD = 563 + 88 SW*** +19 1.66
4x 858 MD = 732 + 55 SW* +21
7 2x 662*** MD = 505 +103 SW*** +20 0.78
4x 817 MD = 615 + 88 SW*** +19
10 2x 653*** MD = 502 + 99 SW*** +22 1.82
4x 807 MD = 494 +137 SW*** +20
13 2x 657*** MD = 453 +133 SW*** + 18 2.23*
4x 815 MD = 612 + 90 SW*** +20

HD 4 2x 545*** HD = 418 + 81 SW*** +21 1.15
4x 686 HD = 448 +104 SW*** + 18
7 2x 532*** HD = 391 + 92 SW*** + 23 0.69
4x 651 HD = 473 + 78 SW*** +20
10 2x 524*** HD = 407 + 76 SW*** +20 2.72**
4x 658 HD = 356 +132 SW*** +21
13 2x 522*** HD = 391 + 85 SW*** + 18 0.35
4x 655 HD = 479 + 78 SW*** +21
tFor all characters at each date at each ploidy level, the number of pairs measured (n)
were: date 4=89 and 93, date 7=91 and 95, date 10=92 and 97, and date 13=91 and 96 for
diploid and tetraploid, respectively.
8 Ploidy means at each day and for each character were tested using a F-test.
*** F-value significant at 0.1% level.
*, **, *** a- and b-values significantly different from zero at 5, 1, and 0.1% levels,
respectively.
T The differences between the b-values of each ploidy level at each day and for each
character were tested using a t-test.
*, ** t-value significantly different at 5 and 1% levels, respectively.














CHAPTER 5
THE EFFECTS OF VARIOUS CHEMICALS ON DIPLOID-TETRAPLOID RED
CLOVER SEEDLING GROWTH

Introduction

Tetraploid seedlings have been shown to grow faster and more vigorously than

diploids; apparently, the heavier mean seed weight of tetraploids was primarily responsible

for the increased growth (Chapter 3 and 4). Approximately 90% of ungerminated seed

weight in dicotyledons is the cotyledon, the primary energy source during germination

before photosynthesis is initiated in the seedling (Black, 1955). Heavier seed weight can

supply more energy for seedling growth.

Various chemicals are reported to influence seedling growth during germination in

a number of species (McWilliam et al., 1970; Copeland, 1976; Stevenson and Cleland,

1981). Auxin is known to cause cell elongation by increasing cell wall extensitivity,

hydraulic conductivity, water potential gradient across the plasma membrane, and turgor

pressure (Cleland, 1987). Cytokinin is known to increase amylolytic activity of cotyledons

(Bewley and Black, 1985), enhance cellular division and expansion of cotyledons, and

increase sugar content (Moore et al., 1998). The application of sucrose can also increase

seedling growth because the stored energy in the cotyledons is metabolized into highly

mobile sucrose and transported to newly-forming tissues (Copeland, 1976). Other sugars

and ions are known to be other osmoregulators maintaining the turgor pressure in the cells

of some species (Stevenson and Cleland, 1981). Hoagland solution which contains all the











elements essential for plant development, seemed to increase the seedling growth

(Hoagland and Arnon, 1938). Much of the research with these compounds has been with

excised seedling components: that is, hypocotyls and cotyledons. Literature on the

application of these chemicals to intact seedlings is limited. However, these chemicals may

influence seedling growth when seeds are associated with chromosome doubling or

increasing seed weight. Some chemical treatments may have a positive influence on diploid

seedling growth to make it comparable to that of tetraploids. Even though there are many

other chemicals that may influence the seedling growth, only the influence of chemicals

described above were tested due to the limited amount of time and the number of seeds.

The objective of this study was to test the effects of the chemicals added to the

germination medium to determine the seedling responses in both diploid and tetraploid red

clover (Trifolium pratense L.). In addition, the effect of presoaking seeds in various

chemicals and then drying of the seeds at both ploidy levels before placing on germination

media was tested to determine the practical uses of presoaking the seeds, or to determine

possible chemical treatments that can be applied to seeds and then dried before sowing. In

this way, desirable chemical treatments could be used to improve seedling establishment

under field conditions.

Materials and Methods

The diploid (2x) and tetraploid (4x) populations and their genotype composition

were described previously (Table 3-1). Six chemical treatments were tested: 1) no

chemical added (control); 2) 2 ppm indole-3-acetic acid (IAA) (C10HNOzNa, FW 197.2);

3) 3 ppm kinetin (K) (6-Benzylaminopurine- C12HlNs5HC1, FW 261.7); 4) 2 ppm IAA +











3 ppm K; 5) 2% sucrose (S) (C12H2201, FW 342.3); and 6) 2 ppm IAA + 3 ppm K + 2%

S. These six treatments were added to either deionized water or half the recommended

concentration of Hoagland's No. 2 Basal Salt Mixture (Sigma H-2395) (Hoagland and

Amon, 1938). The concentrations were determined either from preliminary studies, in the

case of Hoagland solution, and/or information obtained from the literature showing a

positive influence in the species studied (McWilliam et al., 1970; Copeland, 1976;

Stevenson and Cleland, 1981).

The seeds at each ploidy level were soaked in each of 12 treatments, six chemical

treatments and each of two Hoagland solution treatments (- and +; indicates no addition

of Hoagland solution and + indicates the addition in Tables 5-2, 5-3, 5-4, 5-5, and 5-6),

for 48 h at 5C. After the soaking treatment at 5C, one half of the seeds were

immediately germinated in growth pouches at 20C in the dark with the same solution in

which they had been soaked. This group is referred to as 'media' treatment in Tables 5-2,

5-3, 5-4, 5-5, and 5-6. Each growth pouch with the corresponding 25 mL Hoagland

solution and chemical solution, contained 20 seeds. The osmotic potentials of the various

germination solutions were not measured. The other half of the seeds were dried for 24 h

at room temperature. These seeds were then soaked in deionized water for 48 h at 5C

and germinated in deionized water at 20C in the dark. This second group is referred to as

presoakedd' treatment in Tables 5-2, 5-3, 5-4, 5-5, and 5-6. This seed was then germinated

at 20C in deionized water in the dark. There were two replications (growth pouches) for

each treatment. The pouches were inserted vertically into the slits of growth cans. The

replicated pouches were separated from each other within the same can. Measurements









60

were made at 10 d after the seeds were placed in the pouches. The whole experiment was

repeated twice.

The experimental units were the means of each ploidy level, chemical, Hoagland

solution, and presoaking treatment. Thus, there were 96 experimental units. The sampling

units were the eight randomly selected normal seedlings. The seedling characteristics

measured were hypocotyl length (HL), root length (RL), crown diameter (CD), hook

diameter (HD), and middle point diameter between crown and hook (MD). The

measurements were made to the nearest mm for length characters. Diameter

measurements were made in mm converted to rim, using a microprojector at a

magnification of 50X. The treatments were all factor combinations of two ploidy levels x

two soaking treatments x two Hoagland solutions x six chemicals arranged in factorial

design. The minimum differences for significance presented in the tables were obtained

using Duncan's Multiple Range Test for the maximum number of means to be compared

(Harter, 1960).

Results

The analyses of variance for all seedling characteristics are shown in Table 5-1.

Generally, all the main effects were significant for most characters. The ploidy level effect

was always found for all characteristics, and the means of tetraploids were almost always

greater than those of diploids.

Hypocotyl Length

For the media treatment, the use of Hoagland solution slightly increased HL in

control and in combination with IAA and sucrose at the diploid level (Table 5-2). The









61

increases were comparable to the means of tetraploids without Hoagland solution. On the

other hand, tetraploids did not respond to the Hoagland solution effect except in

combination with sucrose. Applications of IAA and sucrose alone did not lead to an

increase in HL. The applications of K, IAA + K, and IAA + K + S resulted in considerable

decreases in HL at both ploidy levels. There were no significant effects of Hoagland

solution in these cases.

For the presoaking treatment, there were generally decreases in elongation of HL

in control, IAA, and sucrose treatment at both ploidy levels. There were generally no

significant differences in K, IAA + K and IAA + K + S treatment. The Hoagland solution

effect was not present in the presoaking treatment.

Root Length

There were generally no increases from Hoagland treatment at either ploidy levels

in either media or presoaking treatment (Table 5-3). The additions of IAA, K, IAA + K,

and IAA + K + S resulted in significant decreases in RL in both the media and presoaked

treatments at both ploidy levels. Only sucrose addition produced an increase in RL in the

media treatment at both ploidy levels. However, a significant interaction between sucrose

and Hoagland solution was observed. The combination of Hoagland solution and sucrose

offset the increase of sucrose alone. In the presoaked treatment, the addition of sucrose

had no effect over the control.

The presoaking treatment resulted in an increase in RL of the control, compared to

non-presoaking at both ploidy levels. After presoaking, the effect of sucrose application on

RL growth was not significantly different from control, but was lower before presoaking.











This treatment increased RL in the IAA and IAA + K + S applications as well as

combination with Hoagland solution. The RL in the IAA application after presoaking was

comparable to that of control (Table 5-3).

Crown Diameter

The Hoagland solution generally had no effect on CD (Table 5-4). Applications of

IAA, K, IAA + K, and IAA + K + S resulted in increases in CD at both ploidy levels, but

there was no effect of sucrose. Comparing the presoaking treatment with media treatment,

presoaking with IAA, K, IAA + K, and IAA + K + S resulted in significant decreases in

CD. However, the CD from K, IAA + K, and IAA + K + S applications were still greater

than that of control without Hoagland solution (Table 5-4).

Middle Diameter

The Hoagland solution generally had no effect on MD at both ploidy levels (Table

5-5). Applications of K, IAA + K, and IAA + K + S resulted in significant increases, but

IAA and sucrose had no effect. The presoaking treatment generally had no effect, but it

decreased MD when K was applied. The MD from K, IAA + K, and IAA + K + S

applications were greater than that of the control after presoaking (Table 5-5).

Hook Diameter

There were generally no Hoagland solution and presoaking treatment effects on

HD at either ploidy level (Table 5-6). The applications of K, IAA + K and IAA + K + S

resulted in greater increases in HD.











Discussion and Conclusions

From these results, greater HL was obtained from addition of Hoagland solution

and from the Hoagland solution in combination with IAA and sucrose at the diploid level.

Significant RL growth increase at both ploidy levels was observed from sucrose

application only. IAA actually resulted in decreases in RL. The use of K and combinations

of K with other chemicals resulted in decreases in length characters. The measurement was

made only at one day, when the maximum growth was achieved at each ploidy level. Thus,

the effects of these chemicals on growth rate within this growing period were not

determined in this experiment.

From this experiment, it would be reasonable to establish the general relationship

between seedling length and diameter. The negative relationship between HL and diameter

found in this research supports the statement by Stuart et al. (1977) that the hypocotyl

elongation was due to cell elongation not cell division in lettuce (Lactuca sativa L.), which

also has the epigeal germination pattern as red clover. Since the cell volume is fixed,

elongation probably results in decreased diameter. This may be the case in red clover.

The reductions from IAA application on RL and from K on both HL and RL and

their associated combinations would indicate sensitivity of red clover to these

concentrations. Cleland (1987) also pointed out the contradictory evidences of the effects

of growth regulators because the tests were conducted using either intact plants or

isolated growth regulator-responsive tissues. Since these growth regulators are involved in

cell growth and/or elongation and already present, the additions of these growth regulators

might be redundant. Obviously, the negative effects indicated that addition of the growth











regulators was inhibitory. Further research to determine the optimal concentrations of

various growth regulators and chemicals should be pursued in future programs.

The previous experiment (Chapter 4) indicated that the increase in seedling growth

of tetraploids was because of heavier seed weights. The tetraploids were expected to have

larger cotyledons: therefore, more stored energy in seeds. Therefore, the addition of

sucrose was expected to increase seedling growth. From a different point of view, Bewley

and Black (1985) noted the feedback mechanisms of end products so that the addition of

any nutrient would not lead to an increase in seedling growth. Nevertheless, it was

surprising to find the increase in seedling growth because of sucrose application was found

only in RL, but not HL at both ploidy levels. A practical use of sucrose in seedling

establishment should be explored further, and the measurement at various dates should be

made in the future research.

On the other hand, Hoagland solution application showed a slight increase in HL

elongation at the diploid level. Hoagland solution contains all the essential chemical

elements necessary for plant growth (Hoagland and Arnon, 1938). Therefore, it was not

known which elements or combinations had the primary impact on increasing HL. In an

experiment by McWilliam et al. (1970), seedling growth of subterranean clover (Trifolium

subterraneum L.) and Australian phalaris (Phalaris tuberosa L.) was examined using the

Hoagland solution. Their results were similar to this experiment. In their longitudinal

study, the seedlings increased RL also, but the proportional growth of the shoot was

greater than the root at later stages of measurement.









65

One of the important findings of this experiment was the interaction of sucrose and

Hoagland solution. The effect of sucrose on RL growth was offset by Hoagland at both

ploidy levels. The reason is unknown, but the result showed the increases in both HL and

RL were not obtained, when sucrose and Hoagland solution were applied together. The

applications of both solutions would not be recommended.

Despite the fact that the sucrose application increased the RL growth of both

diploid and tetraploid red clover, Hoagland solution application only increased HL in the

diploid seedlings. This may indicate that the tetraploids attained maximum hypocotyl

extension. Further increases in seed weight may not lead to an increase in HL (Black,

1959).

The presoaking treatment resulted in reduction of HL and slight increase in RL.

Copeland (1976) listed some advantages and disadvantage of presoaking. The uptake of

solutions was evidenced by the dramatic reductions of seedling length from IAA or K.

Nevertheless, in the normal conditions, the leaching from the seeds would have caused the

reduction in growth. Therefore, this practice would not be favored.












Table 5-1. Mean squares and significance levels from the analyses of variance of hypocotyl
length (HL), root length (RL), crown diameter (CD), middle diameter (MD), and hook
diameter (HD).

Source of variance df HL RL CD MD HD


Treatment
Ploidy level (PL)
Presoaking (PS)
Hoagland (H)
Chemicals (C)
PL x PS
PLxH
PSxH
PLx C
PSx C
HxC
PL x PS x H
PL x PSxC
PL x H x C
PSxHxC
PL x PS x H x C


Error


14S


47 2457*** 2772***
1 5061*** 2638***
1 4524*** 9836***
1 455*** 1058***
5 18219***18836***
1 63 44
1 179* 6
1 124 1242***
5 90* 162***
5 2399*** 2696**
5 108*** 809***
1 29 25
5 40 40
5 33 13
5 89* 504***
5 33 25
88 35 24


1096134*** 380533***
4108538*** 7095938***
13615024*** 149626***


37604
4813718***
11704
28017
153600***
54658***
1648469**
49124***
43350
44957**
28382*
23135
40630**
11423


36038*
1975830***
4134
10004
69876***
6498
85172***
10190
51
4289
2540
9959
9402
6178


374831***
7284771***
13
14138
1948727**
27846*
24544
45719*
14314
49199***
3118
2763
11294
1993
9004
5804
7172


*, **, and *** F value significant at the 5, 1, and 0.1 % levels, respectively.












Table 5-2. Means of hypocotyl length (HL) in various chemical treatments at both ploidy
levels and between media and presoaked treatments 10 days after germination initiation.
HS = Hoagland solution. (-) indicates no HS added. (+) indicates HS added.

Media treatment Presoaked treatment
Treatment HS 2x 4x 2x 4x
--------------mm-------------
0 47t 51 38 44
+ 52 52 39 41
IAA 45 48 39 41
+ 49 51 40 44
K 26 31 30 33
+ 26 30 30 34
IAA + K 27 29 29 32
+ 28 27 28 32
S 46 52 38 45
+ 50 55 41 43
IAA+K+S 28 33 27 33
+ 29 33 30 35

t Minimum differences for significance among the means were 3 and 4 mm at the 5 and
1% levels, respectively.












Table 5-3. Means of root length (RL) in various chemical treatments at both ploidy levels
and between media and presoaked treatments 10 days after germination initiation. HS =
Hoagland solution. (-) indicates no HS added. (+) indicates HS added.

Media treatment Presoaked treatment
Treatment HS 2x 4x 2x 4x
--------------mm-------------
0 23t 26 26 29
+ 22 26 26 30
IAA 13 14 27 29
+ 9 12 28 30
K 11 13 12 13
+ 13 13 12 13
IAA + K 5 7 12 13
+ 5 5 12 13
S 35 42 27 30
+ 19 26 26 29
IAA+K+S 6 11 12 13
+ 6 7 12 15

t Minimum differences for significance among the means were 3 and 4 mm at the 5 and
1% levels, respectively.












Table 5-4. Means of crown diameter (CD) in various chemical treatments at both ploidy
levels and between media and presoaked treatments 10 days after germination initiation.
HS = Hoagland solution. (-) indicates no HS added. (+) indicates HS added.

Media treatment Presoaked treatment
Treatment HS 2x 4x 2x 4x
---------------- ------------------------
0 816t 904 779 912
+ 800 904 773 877
IAA 1151 1288 794 873
+ 1068 1354 741 853
K 1017 1101 943 1011
+ 1042 1076 915 1021
IAA + K 1224 1329 925 1042
+ 1274 1388 964 1063
S 757 880 763 863
+ 821 910 749 852
IAA+K+S 1219 1221 941 1044
+ 1244 1385 953 1005

t Minimum differences for significance among the means were 66 and 67 [m at the 5 and
1% levels, respectively.












Table 5-5. Means of middle diameter (MD) in various chemical treatments at both ploidy
levels and between media and presoaked treatments 10 days after germination initiation.
HS = Hoagland solution. (-) indicates no HS added. (+) indicates HS added.

Media treatment Presoaked treatment
Treatment HS 2x 4x 2x 4x
-------------------m-- ------------
0 686t 831 685 816
+ 696 853 689 832
IAA 684 842 697 807
+ 701 859 661 812
K 871 1030 829 947
+ 890 1020 796 932
IAA + K 859 1002 814 963
+ 880 1026 851 985
S 649 771 680 802
+ 678 814 667 810
IAA+K+S 819 896 844 982
+ 829 971 834 951


t Minimum differences for significance among the means were 48 and 49 [m at the 5 and
1% levels, respectively.












Table 5-6. Means of hook diameter (HD) in various chemical treatments at both ploidy
levels and between media and presoaked treatments 10 days after germination initiation.
HS = Hoagland solution. (-) indicates no HS added. (+) indicates HS added.

Media treatment Presoaked treatment
Treatment HS 2x 4x 2x 4x
----------------m------------------
0 591t 705 601 738
+ 573 719 618 731
IAA 611 756 606 719
+ 592 751 573 721
K 718 875 732 884
+ 728 874 687 871
IAA + K 795 913 749 898
+ 776 923 726 896
S 564 658 600 737
+ 580 686 586 718
IAA+K+S 756 838 746 902
+ 750 885 709 872

t Minimum differences for significance among the means were 52 and 53 [im at the 5 and
1% levels, respectively.














CHAPTER 6
SUMMARY AND CONCLUSIONS

Eight crosses of red clover (Trifolium pratense L.) cv. 'Cherokee' were exposed

to a N20 atmosphere to induce autotetraploidy in the zygotes 24 h after pollination.

Comparisons of the seed number and weight of the diploid and the autotetraploid plants

obtained in each cross were tested in the field during the 2000 summer season. Over all

crosses, the fertility (seed number) of the first generation of tetraploid plants was only 3%

that of diploid plants. Crosses were found to be a significant factor controlling seed

number or fertility within each ploidy level. Therefore, the induction of autotetraploidy in

diverse crosses would be important to find more fertile lines at this ploidy level in future

research. Since the seed number or fertility of the diploid and tetraploid plants was tested

in the summer season which is not optimum for seed production in red clover and, more

importantly, the plants represented the first generation after autotetraploidy induction,

maximum seed yields and numbers were not expected. However, in successive

generations of interpollination among the tetraploids, fertility should improve through the

process of diploidization which improves pairing regularity and gamete fertility. Also, the

tetraploid plants may require slightly different pollen vectors. Because the tetraploid

flowers have longer corolla tubes, bumblebees (Bombus spp.) rather than honeybees (Apis

mellifera L.) may be more effective pollinators of the tetraploid plants.

Over all crosses, the mean seed weight of the diploids and tetraploids was 1.53 and

2.20 mg, respectively. A highly significant ploidy level x cross interaction was found,

72











indicating that the tetraploid seed weight in the crosses was not associated with diploid

seed weight. The autotetraploids have been shown to have greater mean seed weights and

larger vegetative parts than their diploid ancestors presumably because of the increased

cell volume associated with chromosome doubling.

Because of the shortage of tetraploid seeds, the diploid and tetraploid seeds were

bulked within each ploidy level. Therefore, in additional studies, one diploid and one

tetraploid population were used. The procedure involved seedling growth in pouches to

compare seedling characteristics at four dates at three constant germination temperatures

(12, 20, and 28C) in the dark. The response variables measured were total length

(hypocotyl length + root length), hypocotyl length, root length, crown diameter, hook

diameter, and middle point diameter of the seedling between hook and crown. The

tetraploid seedlings had greater length and diameter than those of diploids at all

temperatures evaluated. The growth pattern over dates indicated that the tetraploids were

longer at the beginning of germination and this difference was maintained over dates. The

greater diameter of tetraploids may indicate a greater emerging ability of the tetraploids in

general. As the temperature increased, the days to achieve the maximum length were

reduced. The critical temperature at a time of sowing should range between 20 and 28C.

The effect of seed weight on seedling characteristics within each ploidy level was

evaluated using regression analysis. Increasing seed weight resulted in increases in the

values of all the response variables measured. The comparisons of b-values between

diploids and tetraploids were made for each response variable at each day of measurement.

Generally, no differences were detected between the b-values of the two ploidy levels,











indicating that seed weight had the same effect on the responses at both ploidy levels.

Since tetraploids are generally associated with a greater cell volume, the greater length and

diameter of the tetraploid seedlings possibly conferred on them the ability to elongate to a

greater extent than the diploids. Thus, the use of heavier or tetraploid seeds should result

in better seedling growth and establishment of red clover.

The additions of Hoagland solution, indole-3-acetic acid (IAA), kinetin, sucrose,

and some combinations in the germination media as well as a presoaking treatment were

made to determine the effect on seedling characteristics at both ploidy levels. The addition

of sucrose had no effect on hypocotyl length, but greatly increased root length at both

ploidy levels. Hoagland solution addition slightly increased hypocotyl length of the diploid

seedling, but not the tetraploids. IAA, kinetin, and their combinations resulted in reduction

of root and/or hypocotyl length in both the diploid and tetraploid seedlings. The

presoaking treatment was not favored in this research due to decreased hypocotyl length.

These aspects should be further explored to determine if certain growth regulators and /or

chemicals might improve germination characteristics and enhance seedling establishment.

This research primarily compared the seedling stage of diploid and tetraploid red

clover, and the tetraploid seedlings were shown to have greater and faster elongation than

those of diploids. However, greater seedling growth is not necessarily related to the final

yield (Black, 1959). Therefore, other important traits in red clover production such as

forage quality, pest resistant and so forth, need to be evaluated in the future before

releasing a tetraploid cultivar.















REFERENCES


Abraham, A., and C.A. Ninan. 1954. The chromosomes of Ophioglossum reticulatum.
Curr. Sci. 23: 213-214.

Allard, R.W. 1966. Principles of plant breeding. 3rd ed. John Wiley & Sons, Inc., New
York.

Anderson, L.B. 1971. A study of some seedling characters and the effects of competition
on seedlings in diploid and tetraploid red clover (Trifolium pratense L.). N.Z. J.
Agric. Res. 14: 563-571.

Association of Official Seed Analysts. 1995. Rules for testing seeds. J. Seed Technol.
16:39.

Ball, D.M., C.S. Hoveland, and G.D. Lacefield. 1996. Southern Forages. 2nd ed. The
Potash & Phosphate Institute, Atlanta.

Berthaut, J. 1968. L'emploi du protoxyde d'ozote dans la creation de varieties
autotetraploids chez le trefle violet (Trifolium pratense L.). (In French, with
English abstract.) Ann. Amelior. Plantes. 18: 381-390.

Bewley, J.D., and M. Black. 1985. Seeds: physiology of development and germination.
Plenum Press, New York.

Bingefors, S., and S. Ellerstrom. 1964. Polyploidy breeding in red clover. The tetraploid
variety Svalof s Ulva compared with some diploid and tetraploid varieties. Z.
Pflanzenzuecht. 51: 315-334.

Black, J.N. 1955. The influence of depth of sowing and temperature on pre-emergence
weight changes in subterranean clover (Trifolium subterraneum L.). Aust. J. Agric.
Res. 6: 203-211.

Black, J.N. 1956. The influence of seed size and depth of sowing on pre-emergence and
early vegetative growth of subterranean clover (Trifolium subterraneum L.). Aust.
J. Agric. Res. 7: 98-109.











Black, J.N. 1957. The early vegetative growth of three strains of subterranean clover
(Trifolium subterraneum L.) in relation to size of seed. Aust. J. Agric. Res. 8: 1-
14.

Black, J.N. 1959. Seed size in herbage legumes. Herb. Abstr. 29: 235-241.

Briggs, F.N., and P.F. Knowles. 1967. Introduction to plant breeding. Reinhold Publishing
Corporation, New York.

Chambliss, C.G., and K.H. Quesenberry. 2000. Cherokee red clover. SS-AGR-40.
Agronomy Department. Florida Cooperative Extension Service. Institute of Food
and Agrocultural Sciences. Univ. of Florida. http://edis.ifas.ufl.edu/BODYAA190.

Cleland, R.E. 1987. Auxin and cell elongation. p. 132-148. In P.J. Davies (ed.). Plant
hormone and their role in plant growth and development. Kluwer Academic
Publishers. Dordrecht, Netherlands.

Cleveland, R.W. 1985. Reproductive cycle and cytogenetics. p. 71-110. In N.L. Taylor
(ed.) Clover science and technology. Agron. Monogr. 25. ASA, CSSA, and SSSA,
Madison, WI.

Copeland, L.O. 1976. Principles of seed science and technology. Burgess Publishing
Company. Minneapolis, Minnesota.

Dilworth, M.J., and C.A. Parker. 1969. Development of the nitrogen-fixing system in
legume. J. Theor. Biol. 25: 208-218.

Gilles, A., and L.F. Randolph. 1951. Reduction of quadrivalent frequency in autotetraploid
maize during a period of 10 years. Am. J. Bot. 38: 12-17.

Harter, H.L. 1960. Critical values of Duncan's multiple range test. Biometrics. 16: 671-
685.

Hoagland, D.R., and D.I. Amon. 1938. The water-culture method for growing plants
without soil. Calif. Agric. Expt Sta. Circ. 347.

Hougas, R.W., and S.J. Peloquin. 1958. The potential of potato haploids in breeding and
genetic research. Am. Potato J. 35: 701-707.

Jones, R.L., and C. Moll. 1983. Gibberellin-induced growth in excised lettuce Hypocotyls.
p. 95-128. In A. Crozier (ed.). The biochemistry and physiology of gibberellins.
Vol. 2. Praeger Publishers. Westport, Connecticut.











Kendall, W.A., and W.C. Stringer. 1985. Physiological aspects of clover. p. 111-159 In
N.L. Taylor (ed.) Clover science and technology. Agron. Monogr. 25. ASA,
CSSA, and SSSA, Madison, WI.

Lacefield, G.D., and D.M. Ball. 2000. Ten great reasons for growing clover. p. 1-4. In G.
Lacefield et al. (ed.) Proc. Natl. Clover Symp. 6 Aug. 2000. Oregon Clover
Commission and National Association of County Agricultural Agents.

Leonard, L.T., and W.R. Dodson. 1933. The effects of non-beneficial nodule bacteria of
Austrian winter pea. J. Agric. Res. 46: 649-664.

Levan, A. 1942. Plant breeding by induction of polyploidy and some results in clover.
Hereditas. 28: 245-246.

Li, C.C. 1976. First course in population genetics. Boxwood Press, Pacific Grove, CA.

Manlove, D.C. 1982. The Best of James Whitcomb Riley. Indiana Univ. Press,
Bloomington.

Mayer, A.M., and A. Poljakoff-Mayber. 1989. The germination of seeds. 4th ed.
Pergamon Press, Oxford.

McWilliam, J.R., R.J. Clements, and P.M. Dowling. 1970. Some factors influencing the
germination and early seedling development of pasture plants. Aust. J. Agric. Res.
21. 19-32.

Mislevy, P., K. Quesenberry, J. Atkins, C. Chambliss, T. Hewitt, R. Kalmbacher, M.
Kistler, and L. Sollenberger. 1999. Florida FIRST Base Papers-Forage Crops. p.
261-271. In J. Smith et al. (ed.) Florida FIRST Base Papers. Univ. of Florida.
Institute of Food and Agricultural Sciences.

Moon, D.E. 1993. Breeding red clover for increased seed weight. M.S. thesis. Univ. of
Florida, Gainesville.

Moore, R., W.D. Clark, and D.S. Vodopich. 1998. Botany. 2nd ed. WCB/McGraw-Hill,
New York.

Neubauer, G., and H.L. Thomas. 1966. Effects of various colchicine pH levels of seed
treatment on polyploid cells and other cytological variations in root tips of red
clover. Crop Sci. 6: 209-210.

Pfahler, P.L., R.D. Barnett, and H.H. Luke. 1987. Diploid- tetraploid comparisons in rye.
IV. Grain production. Crop Sci. 27: 431-435.














Poehlman, J.M.,and D.A. Sleper. 1995. Breeding field crops. 4th ed. Iowa State Univ.
Press, Ames.

Quesenberry, K.H., W.R. Ocumpaugh, O.C. Ruelke, L.S. Dunavin, and P. Mislevy. 1987.
Registration of'Floralta' limpograss. Crop Sci. 27: 1087.

Quesenberry, K.H., G.M. Prine, O.C. Ruelke, L.S. Dunavin, and P. Mislevy. 1993.
Registration of'Cherokee' red clover. Crop Sci. 33: 208-209.

Rincker, C.M., and H.H. Rampton. 1985. Seed Production. p. 417-443. In N.L. Taylor
(ed.) Clover science and technology. Agron. Monogr. 25. ASA, CSSA, and SSSA,
Madison, WI.

Schulz-Scharffer, J. 1980. Cytogenetics-Plants, Animals, and Humans. Springer-Verlag,
New York.

Shapiro, S.S., and R.S. Francia. 1972. An approximate analysis of variance test for
normality. J. Am. Stat. Assoc. 67: 215-216.

Smith, R.R., N.L. Taylor, and S.R. Bowley. 1985. Red clover. p. 457-470. In N. L.
Taylor (ed.) Clover science and technology. Agron. Monogr. 25. ASA, CSSA, and
SSSA, Madison, WI.

Stebbins, G.L,Jr. 1947. Types of polyploids: their classification and significance. Adv.
Genet. 1: 403-429.

Stevenson, T.T., and R.E. Cleland. 1981. Osmoregulation in the Avena coleoptile in
relation to auxin and growth. Plant Physiol. 67: 649-753.

Stuart, D.A., D.J. Durnam, and R.L. Jones. 1977. Cell elongation and cell division in
elongating lettuce hypocotyl sections. Planta. 135: 249-255.

Sullivan, B.P., and P.L. Pfahler. 1986. Diploid-tetraploid comparisons in rye. III.
Temperature effects on seedling development. Crop Sci. 26: 795-799.

Swaminathan, M.S. 1970. The significance of polyploidy in the origin of species and
species groups. p. 87-96. In O.H. Frankel and E. Bennett (ed.) Genetic resources
in plants-their exploration and conservation. International Biological Programme
Handbook No. 11. Blackwell Scientific, Oxford.











Swaminathan, M.S., and K. Sulbha. 1959. Multivalent frequency and seed fertility in raw
and evolved tetraploids ofBrassica campestris var. Toria. Z. fur Vererb. 90: 385-
392.

Taylor, N.L. (ed.) 1985. Clover science and technology. Agron. Monogr. 25. ASA, CSSA,
and SSSA, Madison, WI.

Taylor, N.L., M.K. Anderson, K.H. Quesenberry, and L. Watson. 1976. Doubling the
chromosome number of Trifolium species using nitrous oxide. Crop Sci. 23: 1191-
1194.

Taylor, N.L., and K.H. Quesenberry. 1996. Red clover science. Kluwer Academic
Publishers, Dordrecht, The Netherlands.

Taylor, N.L., K.H. Quesenberry, and M.K. Anderson. 1979. Genetic system relationships
in Trifolium. Econ. Bot. 33: 431-441.

Valle, O., M. Salminen, and E. Huokuna. 1960. Pollination and seed setting in tetraploid
red clover in Finland II. Suomen Maataloustieteellisen Seuran Julkaisuja. 97: 2-63.

Vance, C.P., and L.E.B. Johnson. 1981. Nodulation: a plant disease perspective. Plant Dis.
65: 118-124.

Weir, J.B. 1961. A Comparison of the nodulation of diploid and tetraploid varieties of red
clover inoculated with different rhizobial strains. Plant Soil. 14: 85-89.

Welsh, J.R. 1981. Fundamentals of plant genetics and breeding. John Wiley & Sons, New
York.

Williams, W.A. 1956. Evaluation of the emergence force exerted by seedlings of small
seeded legumes using probit analysis. Agron. J. 48: 273-274.

Wipf, L., and D.C. Cooper. 1940. Somatic doubling of chromosomes and nodular
infection in certain Leguminosae. Ann. Bot. 27: 821-824.














BIOGRAPHICAL SKETCH

Hideto Furuya, the second son of Toshihiko and Tomoe Furuya, was born on

December 13, 1977, in Konosu city in Saitama prefecture, which is very close to Tokyo in

Japan. After graduating from Josai Kawagoe High School in 1996, he went to Georgetown

College, Georgetown, KY, for one year (1996-1997). During his brief attendance at

Georgetown, he received the Outstanding Freshman Mathematics Student Award. From

1997-1999, he attended the University of Kentucky, majoring in agronomy with minors in

agricultural economics and sociology graduating in December 1999 with a B.S. degree

(Cum Laude). He was awarded the George Roberts Memorial Scholarship, and became a

member of Gamma Sigma Delta, University of Kentucky Chapter. Graduate study in the

Agronomy Department at the University of Florida was begun in January 2000. A research

assistantship associated with the small grain breeding and genetics was granted Spring

Term 2001. The Master of Science degree with an agronomy major (plant breeding and

genetics specialty) and a horticultural sciences minor will be awarded in August 2001.

Since 1997, he was actively involved in Aikido. At this time, he is at the second

Kyu level. On Thursday, 4 January 2001, his photograph appeared in the local & state

section of the Gainesville Sun (local daily newspaper) showing him diligently practicing

this difficult martial art.




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs