INHERITANCE OF OKRA LEAF TYPE, GOSSYPOL GLANDS

AND TRICHOMES IN COTTON ( Gossypium hirsutum L.)

BY

NAUSHERWAN NOBEL NAWAB

M.Sc. (Hons.) Agriculture (Plant Breeding & Genetics)

A thesis submitted in partial fulfilment of the requirements for the degree of

DOCTOR OF PHILOSOPHY

IN

PLANT BREEDING AND GENETICS

FACULTY OF AGRICULTURE

UNIVERSITY OF AGRICULTURE, FAISALABAD PAKISTAN

2010

The Controller of Examinations University of Agriculture Faisalabad.

We, the supervisory committee, certify that the contents and form of thesis

submitted by Mr. Nausherwan Nobel Nawab, Reg. No. 95-ag-1416 have been found

satisfactory and recommend that it be processed for evaluation by external

examiner(s) for the award of degree.

Supervisory committee:

Chairman _________________________________ (Prof. Dr. Iftikhar Ahmad Khan)

Member _________________________________ (Prof. Dr. Asif Ali)

Member _________________________________

(Prof. Dr. Muhammad Amjad Aulukh)

CONTENTS

CHAPTER TITLE PAGE

LIST OF TABLES i

LIST OF FIGURES iii

LIST OF APPENDICES iv

1 INTRODUCTION 1

2 REVIEW OF LITERATURE 8

2.1 Impact of insect/pest damage to the cotton crop 8

2.2 Insect resistance in cotton 9

2.3 Genetics of insect resistance in cotton 15

2.4 Inheritance of insect resistance traits in cotton 18

2.5 Measurement of insect resistance traits in cotton 21

2.6 Effect of insect resistant traits on yield and other morphological traits 26

2.7 Effect of insect resistant traits on fibre quality traits 28

2.8 Assessment of inheritance studies for yield, fibre and other

morphological traits.

29

2.9 Assessment of heterosis, heritability, genetic advance, correlations and

inbreeding depression for yield, fibre and other morphological traits

35

3 MATERIALS AND METHODS 41

3.1 Development of plant material for genetic studies 41

3.1.1 Generation developed in glasshouse 42

3.1.2 Generation development in field 42

3.1.3 Field sowing and planting geometry 43

CHAPTER TITLE PAGE

3.1.4 Field evaluation at maturity 43

3.2 Fibre quality characteristics 45

3.3 Morphological characters affecting insect resistance. 46

3.4 Biochemical characters affecting insect resistance. 51

3.5 Statistical Analyses 56

3.5.1 Genetic basis of variation in genotypic responses for traits under study 56

3.5.2 Estimation of narrow sense heritability. 58

3.5.3 Genetic advance 58

3.5.4 Heterosis and inbreeding depression 58

3.5.5 Correlations 59

3.5.6 Chi-square analysis 59

4 RESULTS 60

4.1 Genetic basis of variation for morphological, fibre and insect resistant

traits

60

4.1.1 Analysis of variance for morphological traits 60

4.1.2 Analysis of variance for fibre related traits 71

4.1.3 Analysis of variance for insect related traits 79

4.2 Generation mean analysis for various plant traits 85

4.3 Generation variance analysis for various plant traits 92

4.4 Inheritance studies pertaining to insect resistant traits 92

4.5 Estimation of heritability and genetic advance for various plant traits 99

4.6 Estimation of heterosis and inbreeding depression for various plant traits 102

4.7 Correlations 104

CHAPTER TITLE PAGE

5 DISCUSSION 114

5.1 Genetic basis of variation for morphological, fibre and insect resistant

traits

114

5.2 Generation mean analysis for various plant traits 115

5.3 Generation variance analysis for various plant traits 120

5.4 Inheritance studies pertaining to insect resistant traits 121

5.5 Estimation of heritability and genetic advance for various plant traits 125

5.6 Estimation of heterosis and inbreeding depression for various plant traits 127

5.7 Correlations 128

6 SUMMARY 135

LITERATURE CITED 140

ACKNOWLEDGEMENTS

I am grateful to the Almighty God, Who blessed me with health, wisdom, knowledge,

thoughts and opportunity to make some contribution in the form of present effort. The

research work embodied in this manuscript was accomplished under the able guidance and

affectionate supervision of Prof. Dr. Iftikhar Ahmad Khan, Dean Faculty of Agriculture,

University of Agriculture, Faisalabad. I will remember his moral encouragement and

valuable advice throughout the course of this study.

I owe my obligation to Dr. Asif Ali, Professor, Department of Plant Breeding and

Genetics, University of Agriculture, Faisalabad for his skillful guidance, positive criticism,

and keen interest throughout my research programme. I am also obliged to Dr. Muhammad

Amjad Aulukh, Associate Professor, Institute of Horticultural Sciences, University of

Agriculture, Faisalabad for his behaviour and encouragement, in spite of his busy

assignments.

I am thankful to Dr. Abdus Salam Khan, Professor and Chairman, Department of

Plant Breeding and Genetics, University of Agriculture, Faisalabad for his generous

cooperation and guidance.

I am extremely obliged to all teachers who taught me with a great zeal and spirit

especially Dr. Zulfiqar Ali, Assistant Professor, Department of Plant Breeding and Genetics,

University of Agriculture, Faisalabad for their encouragement, support and guidance at every

stage of studies and research. I also acknowledge with thanks the services rendered by the

field and laboratory staff. I pay my best gratitude to Mr. Qamar Shakil, Mr. Akhtar Saeed,

Mr. Mumtaz –Ul- Hassan and Mr. Kashif Nadeem for their timely support in thesis writing.

In the end, I pay cordial obligations to my father and dear sister, who prayed for my

success and performed my responsibilities at home during the course of the studies.

Nausherwan Nobel Nawab

LIST OF TABLES

Table Description Page

3.1 Distinctive morphological features of the upland cotton accessions assessed for

the traits under study

41

3.2 Scheme of crossing 42

3.3 Coefficients of genetic effects for the weighted least squares analysis of

generation means (Mather and Jinks, 1982)

57

3.4 Coefficients of genetic variance components for the weighted least square

analysis of generation variances (Mather and Jinks, 1982)

57

4.1 Generation means and variances for plant height in three single crosses 61

4.2 Generation means and variances for number of monopodial branches in three

single crosses

61

4.3 Generation means and variances for number of sympodial branches in three

single crosses

65

4.4 Generation means and variances for number of bolls in three single crosses 65

4.5 Generation means and variances for seed cotton yield in three single crosses 68

4.6 Generation means and variances for boll weight in three single crosses 68

4.7 Generation means and variances for lint percentage in three single crosses 72

4.8 Generation means and variances for fibre length in three single crosses 72

4.9 Generation means and variances for fibre strength in three single crosses 76

4.10 Generation means and variances for fibre elongation in three single crosses 76

4.11 Generation means and variances for fibre uniformity ratio in three single

crosses

80

4.12 Generation means and variances for fibre fineness in three single crosses 80

4.13 Generation means and variances for number of trichomes in three single crosses 83

4.14 Generation means and variances for gossypol content in two single crosses 86

4.15 Generation means and variances for total gossypol % in two single crosses 86

4.16 Components of generation means parameters, mean (m), additive [d],

dominance [h], additive × additive [i], additive × dominance [j] and dominance

× dominance [l] for various plant traits in different crosses.

89

i

4.17 Components of Variance components, D, H, E following weighted analysis of

components of variance for various cotton traits.

93

4.18 Chi-Squared values and probabilities of goodness of fit of segregation ratios of

F2 and backcross generations in a study of inheritance of okra leaf type trait

97

4.19 Chi-Squared values and probabilities of goodness of fit of segregation ratios of

F2 and backcross generations in a study of inheritance of leaf trichomes trait

98

4.20 Chi-Squared values and probabilities of goodness of fit of segregation

ratios of F2 and backcross generations in a study of inheritance of gossypol

glanding trait on cotton bolls in HRVO-1 × HG- 142

100

4.21 Chi-Squared values and probabilities of goodness of fit of segregation ratios of

F2 and backcross generations in a study of inheritance of gossypol glanding trait

on cotton bolls in HRVO-1 × Acala 63-74

100

4.22

Estimates of heterosis and inbreeding depression for various plant traits in

different crosses.

103

4.23 Genotypic (upper value) and phenotypic (lower value) correlations among

insect resistant and fibre related traits in HRVO-1 × FH 1000

106

4.24 Genotypic (upper value) and phenotypic (lower value) correlations among

insect resistant and fibre related traits in HRVO-1 × CIM 446

107

4.25 Genotypic (upper value) and phenotypic (lower value) correlations among

insect resistant and fibre related traits in HRVO-1 × Acala 63-74

108

4.26 Genotypic (upper value) and phenotypic (lower value) correlations among

insect resistant and morphological and yield related traits in HRVO-1 ×

FH 1000

110

4.27 Genotypic (upper value) and phenotypic (lower value) correlations among

insect resistant and morphological and yield related traits in HRVO-1 ×

CIM 446

111

4.28 Genotypic (upper value) and phenotypic (lower value) correlations among

insect resistant and morphological and yield related traits in HRVO-1 × Acala

63-74

112

ii

LIST OF FIGURES

Figure Description Page

3.1 Variable classes in leaf types 47

3.2 Variable classes in leaf trichomes 49

3.3 Variable classes of boll gossypol glands 52

4.1 Frequency distribution in F2 generation for plant height in three crosses 62

4.2 Frequency distribution in F2 generation for No. of monopodial branches

per plant in three crosses

63

4.3 Frequency distribution in F2 generation for No. of sympodial branches per

plant in three crosses

66

4.4 Frequency distribution in F2 generation for number of bolls per plant in

three crosses

67

4.5 Frequency distribution in F2 generation for seed cotton yield in three

crosses

69

4.6 Frequency distribution in F2 generation for boll weight in three crosses 70

4.7 Frequency distribution in F2 generation for lint % in three crosses 73

4.8 Frequency distribution in F2 generation for fibre length in three crosses 74

4.9 Frequency distribution in F2 generation for fibre strength in three crosses 77

4.10 Frequency distribution in F2 generation for fibre elongation in three

crosses

78

4.11 Frequency distribution in F2 generation for fibre uniformity ratio in three

crosses

81

4.12 Frequency distribution in F2 generation for fibre fineness in three crosses 82

4.13 Frequency distribution in F2 generation for number of trichomes in three

crosses

84

4.14 Frequency distribution in F2 generation for gossypol content in two

crosses

87

4.15 Frequency distribution in F2 generation for total gossypol in two crosses 88

4.16 Segregation in F2 generation for leaf shape in three crosses 96

iii

LIST OF APPENDICES

Appendix Description Page

I Preliminary assessment of germplasm for okra leaf type, gossypol

glands and trichomes

164

II Analysis of variance for plant height for six generations in 3 crosses 165

III Analysis of variance for number of monopodial branches for six

generations in 3 crosses

166

IV Analysis of variance for number of sympodial branches for six

generations in 3 crosses

167

V Analysis of variance for number of bolls for six generations in 3 crosses 168

VI Analysis of variance for seed cotton yield for six generations in 3

crosses

169

VII Analysis of variance for boll weight for six generations in 3 crosses 170

VIII Analysis of variance for lint percentage for six generations in 3 crosses 171

IX Analysis of variance for fibre length for six generations in 3 crosses 172

X Analysis of variance for fibre strength for six generations in 3 crosses 173

XI Analysis of variance for fibre elongation for six generations in 3 crosses 174

XII Analysis of variance for fibre uniformity ratio for six generations in 3

crosses

175

XIII Analysis of variance for fibre fineness for six generations in 3 crosses 176

XIV Analysis of variance for number of trichomes for six generations in 3

crosses

177

XV Analysis of variance for gossypol content for six generations in 3

crosses

178

XVI Analysis of variance for total gossypol for six generations in 3 crosses 179

XVII

Computation of the standard aliquots for the development of standard

curve in HRVO-1 × Acala 63-74 (Normal × glandless)

180

XVIII Computation of the standard aliquots for the development of standard

curve in HRVO-1 × HG-142 (Normal × High glanding)

180

iv

CHAPTER 1

INTRODUCTION

Cotton refers to those species of the genus Gossypium which bear spinnable seed coat fibres.

The history of cotton growing is as old as the history of mankind. There are compelling

archeological evidences which advocate this statement. The cotton fabrics found during the

excavations in the Indus River Valley of Pakistan, provide the evidence of its usage dating

back to about 3000 B.C (Gulati and Turner, 1928). Cotton crop is anciently grown in

Pakistan since fifth million B.C (Ahmad and Ali, 1993). Chowdhuri and Buth (1971) found

cotton remains in Egyptian Nubia which was estimated to be 4500 years old, and concluded

that cotton was cultivated for its seed, which was used as cattle feed, rather than as a fibre

crop. It is evident from the DNA sequencing studies on the species that the genus of

Gossypium might have emerged 10-12 million years ago, although the geographic origin of

the genus has not been identified (Wendel and Albert, 1992). These discoveries provide

evidence for an ancient culture of cotton in both the old and the new world. However, the

earliest written record of its textile use is in the ancient digest ascribed to Manu, written in

800 B.C, which indicates that the Hindus had long known cotton both as a plant and as a

textile. Later references by such writers as Pliny and Marco Polo indicate that India was the

centre of the cotton industry in the old world until well into the Christian era. When

Columbus visited West Indies in 1492, during his discovery journey to the United States of

America, he found that cotton was abundantly being grown. The source of the cotton seed

stocks used in the early American colonies is unclear. Records indicate that seed of both

Asiatic and G. hirsutum types was obtained early from the West Indies. The Asiatic types

soon disappeared, and by 1784, when the first cotton was grown for export the green-seeded

cottons (G. hirsutum) introduced from the Caribbean and Central America predominated

(Niles, 1980a).

By the 1840s, India was not capable of fulfilling the increasing demands of huge

quantities of cotton fibres needed by mechanized British textile industries; moreover,

shipping bulky, low-price cotton from India to Britain was time-consuming and expensive.

These factors coupled with the stronger fibres of American cotton encouraged the British

1

traders to purchase cotton from slave plantations in the United States and the Caribbean.

Therefore, British industrialists preferred cotton from the southern United States, over that of

Indian (Niles, 1980a).

Cotton is grown in tropical and sub-tropical regions of more than 80 countries world

over. About two-thirds of the world cotton is grown between latitudes 300 and 370 North,

which includes China, the former Soviet Union (primarily Uzbekistan) and the United States.

Small quantities of cotton come from 400 North, where Bulgaria, Russia, China and Korea

are located (Bell and Gillham, 1989). The remaining quantities are mostly produced in

countries located at the latitudes 300 North to 300 South, such as Greece, India and Pakistan.

(Bell and Gillham, 1989). The leading cotton producing countries are China, USA, India and

Pakistan (FAO, 2000; ICAC, 2007a). The cotton industry of Pakistan is the economic

backbone, which provides employment to millions of farm and factory workers. Cotton

accounts for 7.5 percent of the value added in agriculture and about 1.6 percent to GDP

(GOP, 2007-08 a). The area and production under cotton crop during 2007-08 was 3054

thousand hectares and 11.6 million bales respectively. This area and production during the

fiscal year 2007-08 was about 0.6 percent and 9.3 percent less than the last year’s (2006-07)

record (GOP, 2007-08 b). World cotton production had also declined by 3 percent than in

2006-07 (GOP, 2007-08 c).

Since 1984, the government of Pakistan has gradually abolished restrictions on the

establishment of mills and has significantly reduced import tariffs on textile machinery.

These reforms liberalized Pakistan’s cotton industry. The numbers of cotton mills in 1997

were almost double the number of installations in 1984. Correspondingly, the output of

cotton and cotton-products increased on average by 11% annually between 1984 and 1995

(FAO, 2000). The rapid expansion in textile industry, local consumption and consequently

cotton production elevated the country to the fifth largest cotton producer in the world.

Moreover, Pakistan has aggressively expanded both its market share of cotton yarn, fabric

and clothing in the global market, and has become a significant exporter since 1985. Cotton

is an occupation of 1.5 million farming families and is the chief source of livelihood for

several millions of labour in cities and towns as well. In cotton growing areas, sale of cotton

produce may account as much as 40% of cash income of rural households. It provides raw

material to 503 textile mills, 1135 ginning factories and 5000 oil expellers (GOP, 2005-06).

2

In addition to providing nearly 44 percent of the world's fibre and supplies 10 percent of the

world's edible oil (Fertilizer Statistics, 1999-2000). It not only meets the needs of fibre of the

local industry but also provides food in the form of edible oil and feed in the form of seed

cake. The cotton accounts for about 85% of the edible oil production, which is used, mixed

with other edible oils, in cooking and making margarine, and low grade cotton seed oil is

used in making soap manufactures and lubricants. Residual seed cake is a valuable protein

concentrate used for livestock feed. Besides, fibre and oil, the second-cut linters are used in

the chemical industry with other compounds to produce cellulose derivatives such as

cellulose acetate, nitrocellulose and a wide range of other compounds (Gregory et al., 1999).

About a century ago the indigenous short staple Desi cotton species, G. arboretum, was

mainly planted in the Indo-Pak sub-continent (Bell and Gillham, 1989). Since their

introduction around 1884, the Upland varieties belonging to G. hirsutum species, developed

rapidly and now comprise approximately 95% or more of Pakistan’s cotton production (Bell

& Gillham, 1989). Cotton is mainly produced in two provinces of Pakistan, Punjab and

Sindh, which jointly account for more than 99% of the total production. However, for the last

few years, some climatic and biotic factors such as unexpected heavy rainfall, drought and

insect/pest infestation led to cotton production instability.

According to a report (Introduction & economic importance to cotton, 2001) yield

trends can be divided into five different phases: (www.ccri.gov.pk).

1950s: constant yields: In the 1950s, yields remained almost constant for the entire period,

from 1949-50 to 1959-60, at around 200 kilogram per hectare.

1960s: steady growth: the first gush of growth took place in the 1960s, when yield trend

remained constant from 200 to 300 kilograms per hectare in 1970-71, and to 361 kilograms

in 1971-72.

1970s: The first cotton crisis: An intensive and importunate attack of the

Heliothis armigera devastated the crop during the 1970s, resulting in wide fluctuations in

yields between 377 and 233 kilograms. The figures of 1971-72 were re-attained only in

1979-82.

1980s: Rapid growth: In 1980s there was a dramatic increase in yields, from 364 kilograms

per hectare in 1982-83 to 769 kilograms in 1991-92. This was actually an era when there was

the major pesticide use expansion.

3

1990s: The second cotton crisis: Another crisis arose in 1990s after the peak in cotton

production achieved in 1991-92 followed by severe and persistent pest attack of whitefly,

resulting in the cotton leaf curl virus. Consequently, yield graph dropped dramatically from

769 to between 500 and 600 kilograms per hectare.

The seed cotton yields in Pakistan increased dramatically during the 1980s with the

introduction of new varieties with the national average of 687 pounds per acre reached in

1991-92. However, average yields were reduced to about 510 pounds (231 Kg) in 2000-01

due to increasing disease and difficulties in controlling the insect infestation. Pakistan has

made some initial progress in the development of varieties tolerant to the leaf curl virus, but

the newly developed varieties have intrinsically lower ginning ratios than the achieved

targets of yields in the late 1980s (World cotton market condition, 2001).

The decline in the cotton production over many years may be attributed to various

factors and among those the vulnerability of the crop to insect attack, holds a chief position

in cotton production (Ahmad, 2001). The incidence of insect pests considerably reduces both

yield and quality of cotton production (Arshad et al., 2001). There are about more than 1326

species of insects which have been reported in commercial cotton fields worldwide but only

small proportions are pests (Matthews and Tunstall, 1994). In Pakistan, about 150 different

species of insect and mite pests have been found attacking and reducing the cotton yield and

quantity (Attique and Rashid, 1983). In Pakistan, the chief insect/pest of cotton that pose

threat to the crop are aphid, jassid, thrips, whitefly, Heliothis bollworm, spotted bollworm,

pink bollworm and most recently the new emerging insect/ pest is mango mealy bug. The

mealy bug host records are reported to extend 76 families and over 200 genera, with some

preference for Fabaceae, Malvaceae and Moraceae (Mani, 1989 and Garland, 1998).

For minimizing the losses to cotton production in Pakistan, caused by the insect pests,

the use of synthetic insecticides has become necessary for the last many years for pest

control. This has increased the cost of cotton production in the country, which is exceeding to

40 % of the total cost of growing cotton. The latest studies by International Cotton Advisory

Committee (ICAC) studies have shown that the cost of production ranges from less than 50

cents to over US$ 2.5/kg lint (ICAC, 2006). Such fluctuations show that the cost of

4

production can be minimised and it is a great challenge for researchers to do so. Farmers are

ready to accept the current yield level if the cost of production can be reduced.

During 1970’s and 80’s the use of insecticides increased tremendously in almost all

cotton producing countries of the world including Pakistan. In Pakistan, during 2007-08 a

total of 28 thousand tonnes of insecticides was imported which measures the foreign

exchange of worth 6330 million Pak. rupees (GOP, 2007-08 d). Chemical pesticides affect

drastically human health as well as biological diversity and quality of surface and

underground water. Some pesticides leave persistent residues in soil, groundwater and the

food chain, thus exposing the human population to slow and cumulative poisoning. Cotton

field workers in India and Pakistan are most vulnerable because of lack of awareness of

pesticide impact, lack of strict implementation of safety measures, lack of readily available

running water and exposure to pesticide-contaminated water for drinking or cleaning (Bell &

Gillham, 1989). Pesticides also affect wildlife, domestic animals and biological diversity.

Given the prevailing agronomic wisdom of the times, cotton farmers in the last half-century

sought to transform the ecological system to eliminate insects altogether. Extensive uses of

pesticides have also caused damage to soil quality and fertility (Dinham 1993; Edwards

1993; Murray 1994). Traces of pesticides were also found in soil samples. Murray (1994)

reported that the figures for Central America cotton producers were as high as 18 kilograms

per hectare and as many as 20 to 30 sprays per season, as compared with between 8-13

sprays per season in Pakistan. The indiscriminant and incessant use of synthetic insecticides

is not only developing resistance in the insect/pests but also posing great threat to the

ecosystem in the form of pollution (Reynolds, 1970; Van Dinther, 1972; Renou et al., 2001).

Therefore, while focusing on attaining the national target of increased yield; insect pests have

also to be kept to the minimum.

Genetic resistance in the form of resistant varieties is an affective means of

minimizing yield losses caused by insect pests but also leads to the reduction in the use of

insecticides (Vaden Bosch, 1972; Van Dinther, 1972, Maxwell et al., 1972 and

Bhatti et al., 1976). The emerging trading scenarios under WTO demands cotton production

free from insecticides. Transgenic crops that are genetically modified to produce insecticidal

proteins from the common bacterium Bacillus thuringiensis (Bt) can be effective in

controlling pests thereby reducing reliance on insecticide sprays. Pink bollworm

5

(Pectinophora gossypiella) has been effectively controlled by Bt cotton. It has no resistance

against the sucking insects (Tabashnik, 2003). Instead, transgenic crops can be used in

harmony with other plant protection measures as part of integrated pest management. The use

of transgenic crops however, can greatly reduce the reliance on the hazardous insecticides, as

were achieved in Arizona cotton. Nonetheless, pink bollworm and other insects will

eventually evolve resistance, so any particular transgenic crop variety is not a permanent

solution to pest problems (Tabashnik, 2003).

Nature has provided cotton with traits like okra leaf type, gossypol glands and

trichomes which confer non-preference to the insect pest infestation. Non-preference refers to

various features of host plant which make the host undesirable or unattractive whereas,

antibiosis refers to the adverse effects of the host plant on the development and reproduction

of insect pests which feed on resistant plants. In some cases antibiosis leads to even death of

the insects. Resistant plants retard the growth and rate of reproduction of insect pests.

Although growers spend much of their time and resources in protecting their cotton crop

against attack by insect pests and diseases, the plant itself is already well equipped against

these invaders. Of the particular note are the terpenoid aldehydes, such as gossypol, found in

the oil rich gossypol glands all over the plant. These glands are characteristic of cotton and

its wild relatives and are full of oil that is rich in different terpenoid chemicals. According to

the latest studies, it has been found to be an anti-cancer drug, which has broadened its worth

from medical science (Sotelo et al., 2005).

Conventional breeding has not lost its significance even in the modern era of

biotechnology and genetic engineering. The manipulation and transference of the genes

controlling these traits into suitable cultivars will be of a significant impact for providing

resistance to at least major insect pests in Pakistan’s environmental conditions. It has been

reported that okra leaf in cotton confers resistance to the white fly which is a serious pest and

vector for cotton leaf curl virus. Heavy pubescence confers resistance against pink bollworm,

thrips and jassids while, gossypol glands confer resistance to Heliothis armigera, jassid, and

aphids (Niles, 1980 a). The genes for these traits have higher effect in terms of magnitude

and are reported to be governed by oligogenes (Niles, 1980; Endrizzi et al., 1984).

6

Yield is the ultimate goal of any breeding programme, but improvement of fibre

attributes also holds a chief position. Demand of the finer quality cotton ensues from the

demand of finer cloth on one side and modernization in the ginning, spinning and weaving

industry on the other. The introduction of ISO-9000 standards has further intensified need of

producing and processing better quality cotton for domestic as well as international market.

Breeders have made enormous efforts over decades, to satisfy the fibre quality requirements

of the industry. Fibre and its quality parameters make cotton an industrial plant and therefore,

cotton is grown for its end users, i.e., textile industry. Changes in the textile industry

prioritize relative importance of fibre quality characters. Fibre colour, length, uniformity,

strength, fineness, and maturity are the primary determinants of quality traits in cotton. The

fibre consists of long, fine and convoluted hairs called ‘lint’, which can be detached easily

from the seed. The value and quality of cotton variety depends on the fineness of the fibre as

well as its length. The longer and finer fibres produce thinner and lighter textiles without

knots or uneven surfaces.

Cotton breeders in the country have made enormous efforts over decades, to improve

seed cotton yield and to satisfy the fibre quality requirements of the industry. The enhanced

standards, however, demand more concrete efforts for the synthesis of physiologically

efficient cotton cultivars, which could produce internationally comparable fibre yield and

quality with reduced plant protection coverage. The present studies were carried out to study

the inheritance of these traits in the segregating populations along with their parents, F1 and

backcrosses in order to assess the genetic effects, heritability, heterosis, inbreeding

depression and correlation among the traits of significance. The information reported herein,

would be useful for continued genetic improvement for the development of insect resistance

lines/cultivars with improved yield and quality attributes.

7

CHAPTER 2

REVIEW OF LITERATURE

2.1. Impact of insect/pest damage to the cotton crop

Cotton is a pest loving plant and due to this habit it has become a problematic crop for the

farmers. The sustainability of cotton production worldwide has been affected due to the

piercing, sucking insect pests and bollworms which are a serious threat to the cotton crop.

Matthews and Tunstall (1994) reported more than 1326 species of insects in the commercial

cotton fields worldwide, out of these only small proportions are pests. Younus et al. (1980)

reported the damage caused to the cotton crop by 96 insects. Among these whitefly, jassids,

aphids and mites hold fundamental importance from their damage perspective in Pakistan. Of

the 30 pests of cultivated G. hirsutum been studied, the most important were caterpillars of

Helicoverpa armigera and Helicoverpa punctigera and aphids (Schepers, 1989), whitefly,

jassids, bugs and the spider mite Tetranychus urticae (Shaw, 2000; Pyke and Brown, 2000).

Aston and Winfield (1972) listed out 46 groups of insects known to occur in cotton

throughout the world; among these 46 groups, 42 were classified as economically important

in one or more of the cotton-producing nations. The bollworm/budworm complex is a

primary insect pest problem with larvae attacking squares and bolls causing significant yield

losses if left uncontrolled. Several bollworm control tactics consisting of different pyrethroid

insecticides, applied at different rates, and using different spray application schedules, were

compared over several years for bollworm efficacy, boll damage and cotton lint yields

(Herbert, 2000). Cotton aphid (Aphis gossypii) is the main insect pest of cotton. Honeydew

produced by the aphid can contaminate cotton lint, reducing its quality (Schepers, 1989). It is

an important pest affecting the profitability of cotton production. Cotton whitefly (Bemisia

tabaci) is also a serious pest of primary importance for fibre (Dittrich et al., 1986)

horticultural and ornamental crops worldwide. It could cause extensive damage through

direct feeding, honeydew production and as a viral vector. "Stickiness" in cotton, a major

problem affecting throughout in cotton gins and spinning mills is thought to be caused by the

deposition of sugars by insects, principally aphid and whitefly, on the open boll (Barton et

al., 2005). Most recently, there is emerging threat to many plant species including cotton

crop by mealy bug (Maconellicoccus hirsutus). It had been reported by Kairo et al. (2000) to

8

be native to southern Asia, and had spread to other continents of the world like Africa, and

more recently North America and Caribbean and is still spreading globe wise. The growing

points of the infested plant of cotton with cotton mealy bug become stunted and swollen and

it mainly depends upon the susceptibility of each host species. In highly susceptible plants,

even brief probing of unexpanded leaves causes severe crumpling of the leaves, and heavy

infestation can cause defoliation and even death of the plant. From the dead plant tissues, the

mealybugs migrate to healthy tissue, so the colonies migrate from shoot tips to twigs to

branches and finally down the trunk affecting the whole plant structure. The mealybugs are

readily visible, though sometimes hidden in the swollen growth. Each adult female lays 150–

600 eggs over a period of about one week, and these hatch in 6–9 days (Bartlett, 1978 and

Mani, 1989). Infestations of M. hirsutus are often associated with attendant ants (Ghose,

1970 and Mani, 1989). In its native range, M. hirsutus has been recorded causing economic

damage to many crops. In India, losses have been reported for cotton by Dhawan et al.

(1980) and Muralidharan and Badaya (2000).

Plant protection products have proven to be of limited effectiveness against

M. hirsutus because of its habit of hiding in crevices, and the waxy covering of its body

(Williams, 1996). Most granular insecticides sprayed were proved to be ineffective against

M. hirsutus (Mani, 1989). Systemic insecticides were only used to control heavy infestations.

Inorganic oil emulsion sprays gave good control of M. hirsutus on guava. Any insecticide

used against M. hirsutus should be carefully selected to avoid injury to its natural enemies.

IPM using both coccinellid beetle predators and insecticides (dichlorvos and chlorpyrifos)

had been achieved on grapevine (Mani, 1989). Biological control studies in the release of

natural enemies were proved very successful. Cryptolaemus montrouzieri had been used

successfully to reduce large populations of M. hirsutus in India (Mani and Krishnamoorthy,

2001) and the Caribbean (Kairo et al., 2000).

2.2. Insect resistance in cotton

Insect pests constitute a major factor in production in all over the cotton growing areas of the

world. In recent times, insect control has been totally based on the use of chemical

insecticides. Little emphasis was placed on the plant genetic resistance as a means of

suppressing insect pests. Painter (1951) noted that there were no efforts made to develop

genetic resistance to cotton insect pests. In the present scenario, by failing to cope with all

9

other possible means for insect control, research has been accelerated on the host plant

resistance. In cotton the insect resistance is associated with various morphological traits

(Jayaraj and Murgesan, 1988; Jenkins, 1989 and Watson, 1989) and biochemical traits

evaluated and reported by Singh and Agarwal (1988) and Hedin and McCarty (1990).

Of the morphological traits, okra leaf trait is characterized by deeply cleft and

narrowly lobed leaves with less surface area per leaf than normal leaf of cotton. This type of

leaf is supposed to be non-preferred by the insect pests. Okra and super okra leaf were

proposed as modified leaf types having no direct effect in suppressing insects, except

possibly whiteflies. However, such leaves provided better penetration and coverage of

insecticides (Maxwel, 1977). The level of resistance to pink bollworm was reported to be

increased by transfering okra leaf into resistant background (Wilson, 1987). Effects of okra

leaf types on pink bollworm damage and agronomic properties of cotton were studied by

Wilson and George (1982). They found that okra leaf trait appeared to have value as a pink

bollworm resistant trait as well as improved agronomic performance. Similar types of studies

were conducted on yield, earliness and improved fibre characteristics with pink bollworm

resistance by Wilson (1989). Lines sustained less seed damage caused by pink bollworm with

equal in lint yield and earliness. According to Wilson et al. (1991) the okra leaf isolines had

76 % as much damage as that of normal leaf strains. Moreover, there was 41 % reduction in

the insecticide usage against the Pink bollworm attack in the genotypes with okra leaf trait. In

another study on different cotton varieties, regarding the sucking insect pest population,

Bhatangar and Sharma (1991) found that okra leaf varieties were less infected to whitefly

(Kalifa and Gameel, 1983), jassid and thrips attack as compared to the check. Morphological

leaf surface features of cotton were studied in two types of genotypes i.e; normal leaf

genotypes versus okra leaf genotypes. Compared to normal leaf types, okra leaf accessions

had resistance for Bemisia argentifolli colonization (Chu et al., 2000 a). Gossypium thurberi

Todaro is a wild cotton species native to Mexico and parts of the southwestern USA. It was

found to be resistant against silverleaf whitefly, Bemisia argentifolii, an important pest of

cotton in many regions of the world. Naturally developing field infestations of silverleaf

whitefly in plots of G. thurberi were significantly lesser than in the plots of the commercial

cotton cultivars DP 5415, Siokra L23, and Stoneville 474. Two important traits of smooth

and okra-leaf in Gossypium thurberi have been associated with lower levels of whitefly

susceptibility however, the levels of resistance observed in G. thurberi were significantly

10

greater than in the cotton cultivar DP 5415, which is a smooth-leaf cotton, and Siokra L23

which, like G. thurberi, has both smooth- and okra-leaf traits. This gives an understanding

that the high level of resistance in G. thurberi is due to factors above and beyond smooth and

okra-leaf. Siokra L23, was less susceptible to whitefly than other cotton cultivars, but it

developed whitefly populations over 30 times more than on G. thurberi. The difference in

whitefly population development between G. thurberi and the other two cotton cultivars was

even more striking, up to a 475-fold difference. In contrast to the clear results on naturally

developing field infestations, experiments comparing nymphal survival among G. thurberi

and commercial cotton cultivars did not detect antibiosis, and both choice and no-choice

oviposition experiments did not detect antixenosis. Thus, the mechanisms of resistance in

G. thurberi remain unknown (Walker and Natwick 2005). The whitefly, Bemisia tabaci

(Gennadius) (Homoptera: Aleyrodidae), is usually considered to have originated from the

Indian sub-continent, although little information has so far been gathered on the molecular

diversity of populations present in this region. Three distinct genotypes were indicated by

molecular diversity studies, apparently indigenous to India, which are also present in China,

Malaysia, Nepal, Pakistan, and Thailand. These genotypes have coexistence with the B

biotype, which was first reported in 1999 in India and spread rapidly to the other states in

south India. The B biotype was more common than the indigenous B. tabaci. This is

reminiscent of the situation in the Americas during the early 1990s, where the B biotype

replaced existing biotypes and caused unprecedented losses to agriculture (Rekha et al.,

2005). Okra and super okra leaf were proposed as modified leaf types having no direct effect

in suppressing insects, except possibly whiteflies. However, such leaves provided better

penetration and coverage of insecticides (Maxwel, 1977). In the study of eight United States

Deltapine genotypes, six Australian cotton cultivars and breeding lines the relationships

between cotton leaf morphology and whitefly population densities showed that okra-leaf

cultivars and lines were colonized with fewer whitefly adults, eggs and nymphs as compared

to the normal-leaf cultivars (Chu et al., 1999). In a comparison of smooth-leaf okra and

normal-leaf upland cotton (Gossypium hirsutum L.) strains and cultivars for susceptibility to

colonization by Bemisia tabaci (Gennadius) biotype B. Okra-leaf strains and cultivars, as a

group, had lower numbers of adults, eggs, and nymphs compared with normal-leaf strains

and cultivars indicated the potential of okra-leaf genetic trait for reducing colonization by

B. tabaci. Results also suggested that okra-leaf shape may provide less favorable micro-

11

environmental conditions for the habitat of B. tabaci because of more open canopy as

evidenced by higher leaf perimeter to leaf area ratio (Chu et al., 2002). In another study, it

was found that both normal leaf type and okra leaf shape were susceptible to silver whiteflies

in cotton in relationship to the hairiness background (Chu et al., 2000 b). In a screening study

conducted by Kular and Butter (1999) against the whitefly, Bemisia tabaci conducted on 51

cotton genotypes in a screenhouse with artificially infested Bemisia tabaci adults. Cotton

cultivars with narrow leaves (okra leaf type) were found tolerant to the pest. Bollworms also

pose a great threat to the cotton crop by damaging the leaves and most especially the cotton

bolls. Pink bollworm can be controlled by the trait okra leaf in cotton cultivars. Resitance to

other two types of bollworm is also essential. In a study on the comparison of the oviposition

of H. armigera and H. punctigera on four cotton cultivars including okra leaf and normal leaf

cultivars. In the field, both Heliothis spp., had highest oviposition on okra leaf plants as

compared to the normal plant types (Hassan et al., 1990). More number of eggs was found on

mature leaves as that of young leaves. Syed et al. (1996) investigated the relative resistance

of twenty cotton varieties and observed the highest and lowest thrips population on super

okra and riode okra. Arif et al. (2006) studied the role of some morpho-physical plant

characters of various cotton genotypes including the genotype, HRVO-1 in developing

resistance against thrips. Their results revealed that the genotype HRVO-1 was found

resistant to thrips infestation. Soomoro et al. (1998) observed that okra leaf line CRIS-151

was resistant to boll rot. Boll formation and boll opening was earlier in this okra leaf trait

line.

Another important insect resistant trait is the hairiness/trichomes. There may be one

state of hairiness and the other one of glabrousness on the basis of the densities of the

trichomes. Pubescence phenotypes are described as smooth (no trichomes), hirsute (moderate

pubescence) or pilose (dense pubescence). Profusely state of hairiness is termed as pilose or

velvet hairiness. Most modern cultivars of cotton are smooth (glabrous). The role of

trichomes in plant defence was evaluated by Levin (1973). According to him trichomes occur

in a multitude of forms and sizes. Although they have been used widely for taxonomic

purposes, their adaptive significance has been all but ignored by the evolutionist and

ecologist. It is clear that trichomes play a role in plant defence, especially with regard to

phytophagous insects. In numerous species there is a negative correlation between trichome

density and insect feeding and oviposition responses, and the nutrition of larvae. Specialized

12

hooked trichomes may impale adults or larvae as well. Trichome may also complement the

chemical defence of a plant by possessing glands which exude terpenes, phenolics, alkaloids

or other substances which are olfactory or gustatory repellents. In essence, glandular

trichomes afford an outer line of chemical defence by advertising the presence of "noxious"

compounds. In some groups of plants, protection against large mammals is achieved by the

presence of stinging trichomes. Intraspecific variation for trichome type and density is known

in many species, and often is clinal in accordance with ecographic parameters. The presence

of such correlations does not imply that differences in predator pressure are the causal

factors, although this may indeed be the case. Hairiness has been reported to have a

resistance against the sucking insect pests of cotton. The primary source of resistance in

G. hirsutum is the presence of trichomes (Lee, 1985). The degree of hair or trichome density

on the leaves of Gossypium species and cultivars is related to varying degrees of

resistance/susceptibility to sucking pests, like whiteflies (Meagher et al., 1997), aphids, and

jassids (Jenkins, 1989 and Watson, 1989), or to the boll weevil (reviewed in Thomson and

Lee, 1980 and Percy and Kohel, 1999). The degree of jassid resistance had definite

correlation with the pilosity of the plant. The more tufted types were less prone to jassid

attack. On the relative importance of the characteristics of hairiness studied, length of hair

seemed to be of prime importance, closely followed by density of hair on lamina whereas,

hair on the midrib did not seem to play any resistance to pest. Length of hairs with hair

density on the lamina was considered to be the best selection index in breeding resistance to

jassid attack (Sikka et al., 1966).

However, hairiness leads to high incidence of Heliothis and whitefly (Niles, 1980 and

Butler et al., 1991). Singh et al. (2001) reported negative correlation of the T. chilonis

parasitization with number of trichomes. A study of the relationship of the silver leaf

whitefly, the number of eggs, nymphs and adults was conducted by Chu et al. (2000 a) on

normal leaf with higher number of trichomes were found similar to that of the smooth and

okra leaf except for the leaf with okra shape coupled with high trichome density. In addition

to the trichomes, the physiomorphic plant characters like number of gossypol glands, hair

density and length of hair had resistance against sucking insect pests. They also reported that

whitefly adult population had positive correlation with hair density on leaf lamina and midrib

and there was positive correlation with gossypol glands on leaf vein and midrib. Adult and

nymph of Jassids were negatively correlated with hair density on leaf lamina, midrib and

13

vein. Thrips population was also negatively correlated with hair density on leaf lamina and

midrib while a positive correlation was reported with gossypol glands on leaf lamina, midrib

and vein. Similar type of research findings were also found by Raza et al. (2000),

Bashir et al. (2001), Gulzar et al. (2005) and Arif et al. (2006). On the other hand, glabrous

trait provided significant resistance to bollworms and significantly reduced oviposition of

Heliothis in comparison with the normal genotypes with hirsute character (Lukefahr et al.,

1971 and Wilson and Wilson, 1976). Similarly in another study, Simmons and Gurr (2005)

concluded that the pesticidal losses on cultivated tomato can be reduced by the incorporation

of trichome based host plant resistance. For trichome-based host-plant resistance to be

utilized as a pest management tool, trichomes of wild species need to be introgressed into the

cultivated tomato. Hybrids between the cultivated tomato and the wild species Lycopersicon

hirsutum f. glabratum, Lycopersicon pennellii and Lycopersicon cheesmanii f. minor have

been produced and useful levels of resistance to Acarina, Diptera and Hemiptera pests have

been exhibited, although these effects may be tempered by effects on natural enemies.

In cotton, a high level of gossypol, flavanols, silica and low sugar contents were

reported to have some role in insect resistance (Singh and Agarwal, 1988 and Hedin and

McCarty, 1990). Wilson and Smith (1976) proposed that gossypol glands constituted

gossypol, a phenolic compound which acted as an insecticide, repellant and growth retardant.

Duhoon et al. (1981) and Ilango and Uthamasamy (1989) reported that high gossypol content

had deleterious effects on bollworm/spotted bollworms. The relationship of gossypol gland

density with bollworm incidence was studied by Mohan et al. (1995) which revealed lowest

incidence of bollworm in three genotypes with highest gossypol gland density on the ovary.

Density of glands had an influence on Heliothis larval growth. The number of pigment

glands per cm2 of leaf tissue was negatively correlated with larval weight after five days of

feeding, came from the research findings of Bryson (1983). It was reported by

Jenkins et al. (1966) that glandless cottons were more susceptible to bollworms than glanded

cottons. Another study by Mohan et al. (1994) revealed that gossypol gland count on the

cotyledonary leaves was significantly and positively associated with free gossypol content in

leaf and seed.

14

2.3. Genetics of insect resistance in cotton

Resistance to insects may be governed by oligogenes or polygenes. Oligogenic resistance is

governed by one or few major genes, whereas polygenic resistance is under the control of

several minor genes. Various morphological characters such as hairiness, gossypol glands,

okra leaf etc are associated with insect resistance. The genetics of these mentioned insect

resistance traits have been reported to be governed by oligogenes (Endrizzi et al., 1984).

Okra leaf is a deeply lobed leaf shape that is a monogenic trait governed by the L0 gene

which is incompletely dominant to normal l0. A more extreme leaf shape, termed super okra

is produced by the Ls allele at the L0 locus. Ls is incompletely dominant. Expressions of the

L0 and Ls alleles are modified by genetic backgrounds. The hybrid of normal × okra is

intermediate between the two phenotypic extremes, which indicate towards the incomplete

pattern of inheritance (Niles, 1980 and Hammond, 1941). Whereas, according to the previous

studies by Andries et al. (1969) okra leaf type trait belongs to an allelic series having a

minimum of five members: L0 (okra), Ls (super okra), Le (Sea Island), Lu (sub okra) and l

(normal).

It is important to have an idea of developmental biology of okra leaf type. Okra (L2

O)

is a semidominant mutation of cotton (Gossypium barbadense) that alters leaf shape by

increasing the length of lobes and decreasing lamina expansion. Chimeras containing L2O and

wild-type tissue were generated using Semigamy (Se), a mutation that blocks syngamy

during fertilization and produces haploid maternal/paternal chimeral progeny at low

frequency. In sectorial chimeras, changes in leaf morphology coincide with the boundary

between mutant and wild-type tissues, suggesting that L2O does not regulate a laterally

diffusible factor within the leaf. However, in mericlinal or periclinal chimeras, the presence

of L2O in tissue derived from any of the three histogenic layers (L1, L2, or L3) of the shoot

apical meristem produced leaves with a partial mutant phenotype. The presence of L2O in the

epidermis (an L1 derivative), or in the subepidermal mesophyll of the leaf (L2 derivatives)

reduced the growth of the lamina and thus increased the depth of leaf lobes. The presence of

L2O in the middle mesophyll of the lamina and the vasculature of major lateral veins

(L3 derivatives) had no local effect on the expansion of the lamina, but significantly increased

lobe length. These results demonstrate that L2O is active in every tissue layer of the leaf

(Dolan and Poethig, 1998).

15

According to Niles (1980), the nature of biochemical resistance against the insect

pests ascribed to “high gossypol”. Generally, increasing gland density in cotton plant

appeared to result in increasing concentration of the toxic compounds. Gland density is

regulated by genes at six loci designated as gl1 through gl6. At each locus, the alleles that

increase gland formations are identified Gl; those that reduce gland density are designated gl.

The principal determinants of gland density are Gl1, Gl2 and Gl3 alleles. Gl1 is responsible for

gland formation only in stems, petioles and carpel walls, whereas the Gl2 and Gl3 affects

gland formation in cotyledons and leaves, as well as those organs affected by Gl1. In other

words it can be said that Gl2 and Gl3 mask the effect of Gl1. According to Lee (1962), Gl2

and Gl3 are the major loci regulating gland production. The effects of Gl2 and Gl3 in flower

parts are largely additive with some epistatic interactions. Studies by Wilson and Lee (1971)

showed that seedling damage was least and number of larvae were lowest on plants of

genotypes Gl2 Gl2 Gl3 Gl3, intermediate on Gl2 Gl2 gl3 gl3 and gl2 gl2 Gl3 Gl3, and highest on

gl2 gl2 gl3 gl3 ( Lee, 1971 and Niles, 1980).

Trichomes may be unicellular or multicellular outgrowths from the epidermis of

leaves, shoots and roots. Leaf trichomes in Arabidopsis are unicellular epidermal hairs with a

branched morphology. They undergo successive endo-reduplication rounds early during cell

morphogenesis. Mutations affecting trichome nuclear DNA content, such as triptychon or

glabra3, alter trichome branching. The trichomes of these mutants presented an increased

DNA content, although to a variable extent suggesting a developmental program controlling

DNA increases via the counting of endo-reduplication rounds (Perazza et al., 1999).

Trichomes can be generally divided into either non-glandular or glandular forms. Non-

glandular trichomes are typically simple hairs found on the aerial surfaces of many plants

while the glandular trichomes display much greater diversity as these are capable of

producing toxic chemicals. The presence of glandular trichomes may protect alfalfa

(Medicago sativa L.) against certain stem, leaf, and fruit-eating insect pests. The dominance

genetic variance was greater than the additive genetic variance. The average degree of

dominance exceeded a value of `1' indicating that erect glandular trichome density may be

influenced by digenic epistasis, and/or repulsion phase linkage disequilibrium (Garcia et al.,

2004).

The trichome cover of a plant surface is collectively called pubescence. An increase

in the plant hairiness above the normal degree is governed by two major genes and a complex

16

of modifier genes. The major gene is designated as H1 for sparse hairing. A second major

gene, H2 controls the finely dense pubescence in an upland mutant designated as ‘Pilose’.

The H2 gene has a pleiotropic effect on fibre length, resulting in a short fibre length, too short

to be of commercial use. In the F1 populations, both H1 and H2 show incomplete dominance

(Niles, 1980). An other study by Saunders (1965) regarding genetics of hairiness being

transferred from Gossypium raimondii to G. hirsutum proved to be successful in transferring

the gene, H6 for hairiness from the wild diploid Gossypium raimondii to G. hirsutum race

punctatum. The usefulness of the hairiness gene for jassid resistance is considered to be of

less importance than its value as a marker of D5 genome chromosome segment introduced

into cultivated tetraploid cottons. The presence in the D genome of a hairiness gene similar in

its effect as H1 of the genome, suggests the possibility that these are homologous genes.

Prior to 1985, a series of major genes (H1, H2, H6, Sm2, Sm1-smooth stem, smooth

leaf, Sm3) and modifier genes (H3-stem, H4-lower leaf surface, H5-length) of diverse origins

(G. hirsutum, G. barbadense, G. raimondii, G. tomentosum, G. armourianum) influencing

pubescence had been identified (Endrizzi et al., 1984). Knight (1952) identified two genes

being important for hairiness in cotton. The gene H1 was reported for the tetraploid New

World and also for the diploid Old World cottons. The gene H2 was extracted from the

Hawaiian tetraploid spps. He viewed that these two genes were not allelic, but the studies of

Simpson (1947) revealed allelic nature. Because of the presumable allelic relationships

between some of these loci, there became the need for the revision of genetics of the

hairiness-smoothness system. Genes affecting plant trichome density and pattern were

grouped into five major loci, namely t1 to t5. Corresponding allelic series were also renamed,

T1, and as in the example of t1 (Lee, 1985). The t1 locus is known to be part of cytological

group IV on chromosome 6 (Percy and Kohel, 1999), as originally described by Knight

(1952). Based on quantitative measures of young and mature leaves, Wright et al. (1999)

mapped four QTLs. The t1 locus on chromosome 6 imparts dense leaf pubescence. Another

QTL located on chromosome 25, is homoelogous to chromosome 6 defines the t2 locus.

Significant phenotypic variation in leaf pubescence were found to be associated with two

additional QTLs, QLP(1) and QLP(2). These may represent the t3, t4, or t5 loci. QTLs are

specific in action for particular developmental stage for example, QLP(1) reduced hairiness

only in young leaves while QLP(2) increased hairiness in mature leaves. A single locus

associated with variation in trichome density on the stem did not correspond to the

17

genes/QTLs affecting leaf trichomes, suggesting that these traits may largely be controlled by

different genes. A widely used qualitative classification system for scoring trichome density

(DTL) detected only the chromosome 6 locus and was apparently not sensitive enough to

detect alleles such as t2 having smaller phenotypic effects (Wright et al., 1999).

Two key genes regulating the initiation of trichome development, GLABROUS1

(GL1) and TRANSPARENT TESTA GLABRA (TTG) were found in Arabidopsis (Larkin et

al., 1994). GL1 is a member of the myb gene family. The maize R gene, which can

functionally complement the Arabidopsis ttg mutation, encodes a basic helix-loop-helix

protein. Copies of the GL1 and R genes were used to test hypotheses about the roles of GL1

and TTG in trichome development. The results support that, TTG and GL1 cooperate at the

same point in the trichome developmental pathway. Furthermore, the constitutive expression

of both GL1 and R in the same plant caused trichomes to develop on all shoot epidermal

surfaces. Results were also obtained indicating that TTG plays an additional role in inhibiting

neighboring cells from becoming trichomes.

2.4. Inheritance of insect resistance traits in cotton

The development of leaves by the allometric method of monogenic differences in leaf shape

represents the genetically analyzed multiple allelic series of genes for leaf shape of a

tetraploid American cotton, G. hirsutum, and of a diploid Asiatic cotton, G. arboreum

(Hammond, 1941). The action of three allelic genes for leaf shape was compared in the Acala

variety of Upland cotton, viz., normal (broad), okra (narrow), and superokra. In the

development of a single leaf, the mutant gene for okra leaf produced a comparatively deeper

sinus by delaying the appearance of relatively narrower lateral lobes in the young

primordium as compared to the gene for normal leaf. The leaf of the hybrid of normal × okra

is intermediate. The mutant gene for okra leaf produced a deeper sinus and a narrower lobe in

the series of successive leaves that appear in plant development, as compared with the gene

for normal leaf. The superokra leaf exhibits the same leaf width as that of okra leaf. During

the latter part of the developmental period, it attains the narrower lobe. In G. arboreum, the

allelic genes, laciniate, intermediate broad, recessive broad, and mutant broad were studied in

the same genetic background explaining the shape differences between broad and narrow

leaves due to the differences in relative numbers of cells in the length and width planes.

Genes for leaf shape in both species affect leaf length as well as shape. The cell number is

18

greatly increased in the narrow-type leaves (G. arboretum) and broad leaves of G. hirsutum.

In addition to increase in cell number in G. hirsutum, an increase in cell size as measured by

stomatal length was also noticed. In the series of successive leaves that appear in plant

development, okra and laciniate genes act to produce a longer leaf than that of the broad

types by increasing leaf length at node 1 and by accelerating leaf length increase from node

to node up the stem. In G. hirsutum, absolute petiole length is not affected by the genes for

leaf shape, whereas, in G. arboreum, the genes for leaf shape affect absolute petiole length in

a sequence which does not correspond to their sequence in regard to other phenotypic effects.

Leaf shape and hairiness in cotton are monogenically controlled. The gene for profuse

hairiness (Pilose) and narrow okra leaf is controlled by H2 and L0 respectively

(Endrizzi et al., 1984). The studies of Simpson (1947) indicated that the pilose trait is

controlled by a single gene, which gives incomplete dominance in F1 generation and

segregated into the ratio of 1:2:1 of pilose, intermediate and smooth classes in the F2

generation. Based upon the early genetic studies, the genes, H1 and H2 are independently

inherited. According to the studies of Muttuthamby et al. (1969) the genetical control of

pubescence on leaves is by two pairs of genes i.e; HP1 and HA

2. HP1 seems to induce hair of

sufficient length and density and is completely dominant to hP1. HA

2 allele seems to induce

hairiness but to a smaller degree. It acts additively to HP1 giving profusely hairy plants. There

was another gene EA discovered which displays an epistatic effect on HA2 gene. This gene has

only a minor effect on the HP1. Apart from this the presence of intensifying or modifying

genes affects the density and length of hair resulting in deviations even in the individual

groups.

The expressivity of the two genes (H2 and L0) for pilose hairing and okra leaf type

were studied by Rahman and Khan (1998) in F1 and F2 generations in different genetic

backgrounds, by involving a strain (HR-Velvet Okra) with other broad leafed and semi/

sparsely haired varieties. In F1 generation both the pilose and okra leaf traits were partially

dominant. The F2 generation segregated into four classes of hairiness and four classes of leaf

shape and fitted into the theoretical 1:2:1 ratio of partial dominance. The two homozygous

extremes for both the traits were easily distinguishable. However, the phenotypic expression

in heterozygous condition was affected by the genetic background, i.e; modifying gene

effects. Hairiness as compared to leaf shape was more influenced by the minor modifier gene

19

effects. A significant level of linkage existed between H2 and L0 genes indicating higher

number of velvet-okra combinations in the advance segregating generations (Rahman and

Khan, 1998). From the studies on aneuploids of Gossypium hirsutum L. by Endrizzi and

Ramsay (1983) showed that the H1, H2 and Sm2 genes, all are located on chromosome 6, and

that the H2 and Sm2 are located in the long arm of the chromosome. The H2 gene was

mapped 4 units from the centromere. Based on the latter data, three genes may either be

closely linked or alleles. In addition to the cytogenetic approach, F2 population of three

crosses H1 × H2, H1 × Sm2 and Sm2 × H2 revealed that the three genes segregated as alleles.

Therefore, in view of the segregation of alleles a revised nomenclature for the smooth-

hairiness genetical system was devised (Lee, 1985).

A study on the interaction of two loci that affect trichome density in upland cotton

was studied by Kloth (1995). Genetic interaction between two dominant, non- allelic loci that

impart extreme phenotypes for hair density were investigated and revealed that the gene T1

imparted dense pubescence on leaves and stems, and places hairs on the capsule. The gene T2

arm reduced hairs to the margins of leaves (glabrous plant type). An inheritance model based

on the interaction between T1 and T2 arm was devised. This model predicted the frequencies

of the phenotypes for leaf trichome density in the progeny from self-pollinating T1t1T2armt2

plants to be 3 glabrous, 3 normal pubescent and 10 densely pubescent. The F2 and BC1

population and F2 derived F3 lines were used to test the model. No significant deviations

from the expected ratios were found and all predictions were met. Therefore, T1 was epistatic

on T2arm when phenotypic classes were limited to the presence of trichomes on the leaves or

when T1 was homozygous and T2arm is heterozygous. In all other situations, T2 arm is

epistatic on T1.

The inheritance of gossypol was studied by Lee (1973) in two strains of cultivated

Gossypium barbadense L. The normal alleles, Gl2 and Gl3 are “native” to G. barbadense,

whereas the mutant alleles, gl2 and gl3 were introduced from Gossypium hirsutum L. through

backcrossing. Additive effects accounted for more than 90% of the total genetic variance for

seed gossypol level. Epistatic effects, though small, were frequently significant. In

G. barbadense Gl2 and Gl3 were associated with the production of similar amounts of

gossypol, whereas previous trials with cultivated varieties of G. hirsutum showed that Gl2

was more than twice as expressive as Gl3. The greater average productivity of seed gossypol

20

in cultivated G. barbadense, as compared with, was attributed to greater activity at the Gl3

locus in the former species. In G. hirsutum, Kohel (1987) found additive effects were greater

in the crosses involving glandless lines than in the crosses involving glanded lines. Similarly,

the inheritance of high glanding trait in the high glanding cultivars was investigated by

Calhoun, 1997 by developing the crosses among high glanding, normal glanding and

glandless genotypes and with the isolines of the high glanding breeding. The isoline XG-15

(gl2Gl3) expressed high glanding phenotype, suggesting that high glanding was conferred by

a special Gl3 allele derived from XG-15. Crosses of high glanding and normal glanding

parents resulted in the high glanding plants in the F1 whereas a 3 HG:1 NG ratio was

observed in F2 and ratios of 1 HG: 2 segregating: 1 NG amongst F2:3 progeny.

The effects of genes at two independent loci, identifiable by their gland-producing

pattern, on level of seed gossypol were studied by Lee et al. (1967). Application of the

analyses to the data on gossypol level showed no maternal effects, highly significant

dominance and epistatic effects but which when combined accounted for only 6% of the

genetic variance, and one locus, Gl2, to contribute about three times as much additive

variance as the other locus, Gl3.

2.5. Measurement of insect resistance traits in cotton

There are some traits for which the quantitative method of measurement can not be applied.

Instead, of this these can be measured on phenotypic basis. The visual rating system is an

efficient way for classifying such traits. Similar is the case with leaf shape. There had been

reported visual based rating of leaf shape by various scientists (Hammond, 1941; Simpson,

1947; Rahman and Khan, 1998 and Frelichowski et al., 2005).

In the case with the hairiness/trichomes, it can be measured by two methods. There is

a qualitative grading system (Lee, 1968 and Kloth, 1995) based upon the distribution of

trichomes on the leaves and leaf veins. Leaf trichomes are confined to the margins and to the

epidermal surfaces. The classification system uses five grades: 1 is the absence of trichomes

at all stages and 5 is the presence of trichomes on the petiole and at all stages of vein

branches (Lee, 1968; Wright et al., 1999; Bourland et al., 2003; Stiller et al., 2004 and

Lacape and Nguyen, 2005). Hairiness index was proposed by Rayburn (1986) with three

classes labelled as smooth, moderately hairy and hairy. There is a quantitative means of

21

measurement of leaf trichome density on leaf surfaces. Trichome counts were concurrently

made on the underside of a young and mature leaf from each plant (Muttuthamby et al.,

1969). Two counts were made on each leaf, one on the right and the other on the left of the

mid-vein. All trichomes inside a 6 mm ring (28.27 mm2) were counted under a wide-field

microscope. The density of stem trichomes was scored in the upper 5 cm of each plant, on a

scale 1 (glabrous) to 5 (highly pubescent) (Wright et al., 1999 and Lacape and Nguyen

2005). In another study by Wanjura et al. (1976) trichome counts were made on the petiole,

leaf margin near tip, the midvein and the blade on both sides of the midvein near the bottom

of the leaf.

Trichome numbers were counted with help of an index card of 0.65 cm diameter hole

(0.33 cm-2). All the trichome counts were made within the specific unit area with the aid of a

stereo-microscope (Bourland et al., 2003). Trichomes in 0.36 cm-2 grid area on five different

places were counted by Bryson et al. (1983) on both the abaxial and adaxial surfaces and

averaged. On account of the covering trichome counts made by Bryson et al. (1983) there

was a significant difference in count of the trichomes between the adaxial and the abaxial

leaf surfaces. The frequency and density of pubescence per leaf was greater on the abaxial

surface (1842 cm-2 to 9 cm-2) than that of the adaxial surface (1435 cm-2 to 18 cm-2). The

trichome count studies of Simpson (1947) showed an average of 12.0 hairs per square

millimeter on the upper surface of the leaves and 18.8 on the lower. In the studies conducted

by Smith (1964) evaluated trichomes on cotton genotype and recorded the average number of

trichomes on leaf blades ranging from 2 to 205 trichomes cm-2. On the basis of this he

defined a cultivar Deltapine as smooth leaf with 5 trichomes cm-2. Differences in density of

trichomes between leaf pubescence ratings were found and evaluated by Bourland et al.,

2003. Regardless of the expected pubescence rating of the cultivar, leaves in the top of the

plant always exhibited the highest leaf pubescence ratings, and leaves from the bottom of the

canopy tended to have lowest pubescence ratings. Thus it can be inferred that pubescence

ratings and trichome density were highest for upper canopy leaves. Trichome number

became less as leaves enlarged, then tended to abscise as leaves aged. In the light of the

above justifications, Smith (1964) inferred that the leaf pubescence should be rated using the

youngest, fully expanded main stem leaves.

22

Gossypols are the pigment glands distributed on the plant body covering the stem,

leaf, bract, calyx and carpel walls. These pigment glands are visible from both leaf surfaces.

Gossypol glands provide resistance against insect pests. Studies in quantitative inheritance

are generally conducted as the analysis of the effects of groups of genes acting in concert to

produce the character under consideration. It is thus of some interest when the number loci

genes involved in the production of a character can be known through the use of a method

which allows for discrimination among units of expression by qualitative assays, yet have the

genes express themselves in some other way, the nature of which is quantitative. The

character, expression of pigment gland size and number in the plant body and seeds of

various species of Gossypium, fits into this category (Lee et al., 1967).

The studies of Calhoun (1997) were based on the counting of the gossypol glands on

the half grown flower buds under a stereo microscope at 10X magnification. Gossypol glands

in 0.36 cm2 grid area on five different places were counted with the aid of a grid ocular

micrometer at 35 X by Bryson et al. (1983) on both the abaxial and adaxial surfaces and

averaged. To count glands, a disk was removed from the middle of each leaf approximately

0.5 cm away from the midrib using a corkborer (0.125 cm2). Prior to removal, the area from

which the disk was taken was rubbed with a finger to remove a hazy film which made the

counting of the glands easier. Pigment glands were then counted with the help of a dissection

microscope (Agarwal and Karban, 2000).

In the studies relating to bollworm incidence, Mohan et al. (1994) evaluated the

genotypes for gossypol glands per unit area on leaf, bract, calyx and ovary surfaces.

Gossypol glands were spherical on leaf, bract and ovary surfaces, elliptical on stem and

stigma surfaces; and both oval and spherical on calyx surfaces. Moreover, it was found that

the elliptical glands were largest and the spherical were smallest. On account of the gossypol

counts made by Bryson et al. (1983) there was no significant differences in count of the

gossypol glands on between the adaxial and the abaxial leaf surfaces. Gland numbers on

glanded cottons ranged from 1935 cm-2 to 805 cm-2. This difference in the glanded strains

was attributed to the gland frequency, as well as their presence or absence and can be

manipulated genetically. Calhoun (1997), found that the gossypol gland counts on the sepals

of the true breeding high glanding plant varied from 22 to 79 glands on upper portion of the

sepals and overlapped with the plants whose progeny segregated for high glanding

23

(5 to 55 glands on upper portion of sepals). In seeds the total number of glands per section in

mm2 was estimated (Benbouza et al., 2002) after the removal of the teguments and was

assessed with the help of light and florescent microscope. Mohan et al. (1995) reocorded data

on the number of gossypol glands per mm2 on the adaxial leaf surface, number of gossypol

glands per mm2 seed surface and free gossypol content (%) of the leaf and seed. Gossypol

gland number of cotyledonary leaves was significantly and positively associated with leaf

free gossypol content and seed gossypol gland number. Leaf gossypol content was

significantly and positively associated with number of gossypol glands and free gossypol

content in the seed. Number of seed gossypol glands was highly significantly associated with

seed free gossypol content.

The most frequently used analytical procedures used for quantification include

spectrophotometry and HPLC (Abou-Donia et al., 1981; Stipanovic et al., 1988; Hron et al.,

1990 and Tchatchueng et al., 1992). The spectrophotometric method of quantification was

applied on the decorticated, seed lot dried, over CaC1, (ca. 6%). The kernels were ground to

fine meal and returned to coldstorage. After all the seed lots had been processed, the samples

were extracted and assayed for total gossypol according to the spectrophotometric method

(Smith, 1958 and Lee, 1973).

Because of the sensitivity and repeatability, HPLC is the method of choice for the

measurements of low gossypol concentrations (Abou-Donia et al., 1981). It is however,

tedious to apply chemical measurements on a large number of seed samples. In a study by

Benbouza et al. (2002), a significant correlation was found between the % gossypol content

determined on single seeds by HPLC and number of gossypol glands per section area. This

new technique is rapid and accurate and is particularly valuable in breeding programs to

screen the progeny of cotton genotypes showing a high degree of segregation in the gossypol

content of their seeds. The accuracy and reproducibility of the HPLC method is evident from

a comparative study by Cai et al. (2004), on the gossypol content of various cotton varieties

was conducted through an optimized high-performance liquid chromatography (HPLC) on a

C-18 column with menthanol 5% acetic acid aqueous solution, 90:10 (V/V), as mobile phase,

at a flow rate of 8 mL/min and UV detection at 254 nm. The method was shown to be highly

reproducible, with precision and accuracy, as relative standard deviation and relative mean

error, less than 10 %. Absolute recoveries were greater than 94 %.

24

Sotelo et al. (2005) reported gossypol content in leaves and seeds of 10 Malvaceae

species by HPLC. No gossypol content was noticed in the leaves and seeds of Malvavicus

arborues II Cav Schltdl and Hibicus sabdariffa L. Whereas, in species like Hampea

integerrina Schltdl, Hibicus clypeatus L. Shesh Tendal and Pavonia schideana Stend. J.

gossypol contents were limited to seed only with the concentration of 1180.0, 4.37 and 3.33

mg per 100 g of dried sample respectively. Out of 100 g dried sample in each of the species,

Anoda cristata L. Schltdl, Hibicus rosa-sinensis L. Shesh Tendal and Malvavicus arboreus

Cav Schltdl; gossypol content was noticed in seeds as 27.24, 2.05 and 4.47 mg, respectively.

Whereas, the leaves of the same spp., exhibited 3.52, 1.87 and 0.75 mg, respectively.

However, in Gossypium hirsutum L., the gossypol content in leaves (847.00 mg/100 g) and

seeds (297 mg/100 g) was reported. In another study regarding variation in 10 seed

characteristics in the species groups, common okra (Abelmoschus esculentus), with edible

pods, and in related species by Martin and Rhodes (1983) found that gossypol or gossypol-

like compounds were lower in common okras than in related species. Varieties particularly

low in toxic substances were identified.

Another simple, fast and cost effective method for isolation, identification and

quantification of gossypol, using packed micro-tips columns in combination with HPLC was

performed on different parts of the cotton plant comprising of seeds, stems and leaves by

Meyer et al. (2004). The minimum detection limit of gossypol was determined to be 10 ng

(absolute gossypol). Absolute recovery was greater than 94 %.

The newly developed competitive direct enzyme-linked immunosorbent assay

(cdELISA) technique developed by Wang et al. (2005) could also be a valuable and feasible

alternative for determination of “free” gossypol, in the condition especially when the

available sample is limited with relatively low gossypol concentration. The detection limit

for gossypol was 0.005 mu g/mL. A good correlation between the cdELISA method and the

AOCS official method for “free” gossypol analysis of cotton seed meals was also established.

In the present studies gossypol estimation was carried out using spectrophometric

method for quantifying gossypols by keeping in view its access, validity and reproducibility

of the results as it also became evident from the studies of Vlessidis et al. (2004). Moreover,

in a comparative study using three methods of quantification i.e; Spectrophometric, TLC and

25

HPLC, Fayek and Anwer (2007) concluded that there was no significant difference between

each of the three methods. Therefore, any of the methods could be used depending upon ease

and accessibility.

2.6. Effect of insect resistant traits on yield and other agronomic traits

Yield and other agronomic traits are important for any breeding programme. The main

emphasis of any breeding programme is on these two important principles. By incorporation

of the insect resistance traits, the effect on yield and other agronomic traits is discussed

below.

In a study of the effects of okra leaf, nectariless, frego bracts and glabrous leaf on

yield traits were studied by Thomson et al. (1987) under two insecticide spray regimes

(heavy & light) in field conditions. Under the heavy spray regime, there were few consistent

differences between mutant genotypes and normal except for low yields associated with the

glabrous leaf, nectariless, frego bract genotype and the okra leaf genotype. Under light spray

regime, the genotypes with okra leaf gene or nectariless genes were associated with higher

yields than the normal and the genotypes with frego bracts or glabrous leaf genes were

associated with lower yields. Positive epistatic interactions occurred in the okra leaf gene in

the glabrous, normal bract backgrounds under heavy spraying, and with all backgrounds

under light spraying, and for glabrous leaf in both the okra normal and okra frego

backgrounds under heavy spraying. Pronounced negative epistatic interactions occurred in

only light spray regime, including glabrous leaf gene in all backgrounds and frego bract in all

backgrounds except the normal leaf, normal hair. Rahman et al. (2005) also observed high

yield in okra leaf accessions and also concluded that the genes for high yield in HRVO-1 and

HR-107NH were different from those controlling high yield in HR109-RT. In addition to

increased yield, (Wells and Meredith, 1986 and Meredith and Wells, 1986) associated with

okra leaf shape. Andries et al. (1969) and Pettigrew (2003) found a significant reduction in

the incidence of boll rot with increase in earliness in comparison with the normal leaf shape.

According to Andries et al. (1969) okra leaf trait reduced the leaf area per plant to such an

extent that it might be suspected of causing a reduction in yield. Open plant canopy in okra

leaf plots allowed better air exchange and more sunlight to penetrate to the lower plant zones.

These factors might have increased the photosynthetic efficiency, increasing the yields. But

the findings of Monks et al. (1999) concluded that normal leaf types yielded 17% more than

26

the okra leaf isolines. Similar results of reduced yields from okra leaf cottons were obtained

by other researchers (Andries et al., 1971 and Wilson, 1986).

Studies of Wilson and George (1982) pointed that okra leaf appeared to have value

over frego bract and smooth leaf not only in pink bollworm resistance but also in

improvement in the agronomic performance. Five leaf shapes (normal, semi-okra, sub-okra,

okra and super-okra were evaluated in a study on the leaf shape traits by Jones et al. (1988).

Okra and sub-okra leaf shapes were found superior in different genetic backgrounds in terms

of lint yield. Super-okra accessions were more affected by different environments than the

other leaf types. Moreover, super-okra leaf accessions increased earliness in different genetic

backgrounds.

Okra leaf was supposed to be associated with high yield and earliness but the findings

of Wilson (1989) suggested that all nectariless, nectariless-okra, nectariless-semi-smooth leaf

and nectariless-semi-smooth-okra lines were equal in lint yield and earliness. In contrary,

when Wilson et al. (1991) compared, nectariless-okra line with nectaried-normal line the

former yielded more lint and was significantly earlier than the later. Sometimes, combining

all useful characters into a single genotype does not always result in a single ideotype. A

study conducted by Meredith et al. (1996) involving the traits like sub-okra, semi-smooth

leaf, and nectariless; were backcrossed into DES 119 from MD65-HS to evaluate the effects

of these three traits and their interactions. No significant total yield response due to any trait

was detected; however, sub-okra leaf types produced significantly higher yields than the

normal. In another study Meredith and Wells (1987) found that the sub okra leaf cotton

averaged significantly higher yields (3%) than the normal leaf cotton.

In the study regarding the adaptability potentials McCarty et al. (1983) suggested

high adaptability potentials for nectariless cottons than for the other traits glandless, high

gossypol, okra leaf and frego bract. The genetic potential for improvement in agronomic

traits exists in the population with the okra leaf morphology (Ulloa, 2006).

Similarly, variability of different growth contributing parameters of some okra

(Abelmoschus esculentus L.) accessions and their interrelation effects on yield with leaf

shape were noticed by Alam and Hossain (2008). Two accessions having okra leaf

27

morphology gave a maximum green pod yields of 10.49 and 10.57 t/ha. Fresh green pod

yield, plant height, nodes per plant, leaf length, leaf breadth and length of petiole had a

positive correlation with okra leaf morphology than the other leaf morphologies of palmatifid

and palmatipartite.

Hairiness is imparted due to two alleles. The effect of pilosity of these two alleles on

agronomic traits was studied by Lee (1984). No, significant inferences were inferred for the

trichome count per cm, lint percentage and boll weight. In order to develop potato cultivars

with insect resistance based on the glandular trichomes, Kalazich and Plaisted (1991) found

associations between agronomic and trichome characters with both insect resistance and

acceptable agronomic characteristics on Solanum berthaultii. A strong association between

the presence of B trichome droplets and undesirable agronomic characteristics was

established in backcrossed generations. The backcross plants bearing droplets produced

significantly lower yields, fewer tubers, later maturing plants, and poorer foliage and tuber

appearance than their sibs without the droplets. In intercrosses, no associations were found.

Good MEBA scores were seldom found in backcrossed plants without B droplets. The

associations observed are speculated to be due to linkage or structural genomic

differentiation between the genomes of the species involved in these populations.

In a study of gossypol gland density with boll worm incidence and yield, Mohan et al.

(1994) found by evaluating genetically diverse genotypes that those genotypes having

highest gossypol gland density on their ovary had lowest incidence of bollworms, thus

highest seed cotton yield was achieved.

2.7. Effect of insect resistant traits on fibre quality traits

In cotton the fibre quality traits hold a special position from the industrial point of view. The

physical parameters required for good spinning performance of cotton fibre challenge

breeders. With the objective of insect resistance in the form of incorporation of insect

resistant traits the effect imparted on the fibre quality attributes is discussed below.

In the study on the effects of okra leaf on fibre quality traits by Andries et al. (1969)

it was concluded that Okra leaf shape had no effect on fibre length, fibre strength and fibre

uniformity but caused reduction in the fibre elongation. Same research findings were

28

conceived by Thomson et al. (1987). Normal leaf types had longer fibres and lower fibre

elongation than the sub-okra leaf types (Meredith et al. 1996).

Wilson et al. (1991) in a study, compared nectariless-okra line with nectaried-normal

line the former yielded more lint and was significantly earlier, but the fibre properties were

inferior as that of the later. In a study of potassium level to okra leaf-type isolines, Pettigrew

(2003) observed that though that low K had only minor effects on fibre quality. In the study

of dryland cultivation of the okra leaf-type cultivars the fibre lengths were reduced to 4%

than the normal irrigated cultivation of the same cultivars (Stiller et al., 2004).

The pilosity of cotton plant in relation to insect control has been the subject of a

number of investigations. The pilose condition is associated with decreased fibre length and

increased micronaire (Simpson, 1947 and Lee, 1964, 1984). But Kloth (1993) discovered a

pilose like plant with unexpectedly low micronaire value among the homozygous pilose

plants.

Fibre length, fibre strength, length uniformity and micronaire values demonstrated no

significant differences between the glandless and glanded isolines of any cultivar. The lack of

significant differences in most of the cultivar backgrounds between glandless and glanded

near isogenic lines showed no effect of the glandless/glanded gene on the fibre quality traits

(Yuan et al., 2000). Moreover, gossypol content had hardly any adverse effects on fibre

quality traits (Phogat et al., 2000).

2.8. Assessment of inheritance studies for yield, fibre and other morphological traits

Inheritance studies are meant for the assessment of the nature of the genetic effects in the

breeding material. On the basis of the knowledge of the genetic analysis, it becomes evident,

the direction of the genes for devising a suitable direction in terms of the best breeding

procedure towards the crop improvement. For qualitatively controlled traits, the inheritance

studies are somewhat easy to conduct, as the effect of the individual genes is more

pronounced. In case of quantitative inheritance, the inheritance pattern is complex as the

individual gene effect is controlled by minor genes.

For quantitatively inherited traits, mainly diallel and generation mean analyses are

commonly used. Generation mean analysis is a powerful technique for the assessment of the

29

gene effects. The mode of inheritance of stomatal conductance in crosses of six

G. barbadense parents varying in origin, degree of agronomic development and stomatal

conductance were studied, through generation means analysis. Inheritance of stomatal

conductance varied in complexity from a simple additive-dominance model to models

displaying digenic epistatic interactions in the crosses. Significant additive mean effects for

stomatal conductance were detected in all crosses. The interpretation of results shows that the

mode of inheritance for stomatal conductance is multigenic. Recouping higher stomatal

conductance levels from genetically wider crosses appears feasible and could proceed at a

moderate rate. Fixing higher levels of stomatal conductance in populations from crosses of

elite germplasm may be more difficult because of the presence of dominant mean effects and

digenic epistatic interactions (Percy et al., 1996).

Similarly generation mean analysis technique was also used for agronomically

important traits in oats. Corbit, an agronomically important oat cultivar, was crossed with the

highly regenerable, but agronomically undesirable line, GP-1. Callus was induced from

mature seeds of each parent (P1 and P2); F1, F2 and their reciprocals; and backcross (BC1 and

BC2) generations. The number of somatic embryos was recorded before transfer to

regeneration medium and the number of plants regenerated was recorded. Gene effects, using

generation mean analysis, were computed when GP-1 was the maternal parent (Set 1) and

when Corbit was the maternal parent (Set 2). From this study it was concluded that selection

for callus weight and plant number would be expected to produce only small gains per cycle

because of the substantial negative dxd and dominance effects and these two traits might not

be improved simultaneously when selection is practiced for one of them. However, two

important characters-callus fresh weight and plant number were positively correlated when

GP-1 was chosen as the maternal parent. Therefore, backcross strategies for improvement

were recommendable in the direction of the cross as the highly regenerable plant

characteristics observed were considerably influenced by maternal inheritance (Rowena

et al., 2002).

The nature of gene action and of maternal influence governing cottonseed oil

attributes were determined by Dani and Kohel (1989) with four lines, two each with high and

low seed-oil percentage. For this purpose, P1, P2, F0, F1, F2 and alternative sets of BC1 and

BC2 generations were analyzed in six cross-combinations and their reciprocals. Marginal

30

extents of heterosis for seed-oil percentage were noticeable in F1, with inbreeding depression

in F2. Data from reciprocal backcrosses provided evidence in favour of maternal rather than

cytoplasmic effects of seed-oil development. Relatively higher extents of heterosis, sizeable

inbreeding depression and reciprocally unequal F2 averages were characteristic of the seed

index trait, which often showed a reversal of effects from F1 to F2. Reverse reciprocal

backcrosses exhibited some differences, including greater resemblance between the types,

(A/B)A and (B/A)A, in addition to variable dose effects in seed index. Thus, the differences

between F1 seed index values were not due to cytoplasmic influence. Positive heterotic

effects for seed-oil index, especially among the backcrosses, ranged between 16.08 % and

47.29 % over mid parent averages. Genetic component estimates from analysis of similar sets

of crosses differing only in reciprocal backcrosses, and also from sets of reciprocal crosses

between any two parental combinations, were inconsistent. Scaling tests detected presence of

epistasis within and between a majority of cross-combinations. Despite reciprocal

differences, additive gene effects for seed-oil percentage were significant in 7 out of 24

crosses, representing high × low, low × high and low × low seed-oil parents. Those were

however, accompanied by significant dominance effects of higher order. In crosses involving

low seed-oil percentage parents SA1060 and SA229, all six components were detected

significant, with opposite effects of dominance, dominance × dominance and epistatic

components. Significant additive components were also detected for seed index and seed-oil

index in 7 and 5 out of 24 crosses, respectively. In the inheritance of seed index and seed-oil

index, dominance effects were more important. Epistatic components of additive × additive

and to a lesser extent those of dominant x dominant were found significant.

The morphological and yield and yield related traits play an important role in a

breeding programme (Kumar and Raveendran, 1999). Four generations (F1, F2, B1, and B2 in

which B1 represents a backcross with the parent 1 and B2 represents a backcross with the

parent 2) and parents were tested in upland cotton. Generation mean analysis was used to

estimate the type of gene action determining yield and yield components. It was concluded

that additive, dominance and epistatic gene effects were responsible for the inheritance of lint

yield, boll number per plant and lint percentage whereas only dominance effects were

involved in the inheritance of seed cotton weight per boll and 100-seed weight

(Mert et al., 2003). Pathak and Singh (1970) concluded additive gene effects for boll weight

and lint percentage. In addition to boll weight, Gad et al., 1974 detected significant additive

31

effects for yield, number of bolls per plant lint percentage and fibre strength. Whereas,

Bertini et al. (2001) also studied the gene action through generation mean analysis and

observed dominance effects for number of bolls in addition to the boll weight and fibre yield.

But the studies of Kalsey and Vithal (1980) proved almost an equal magnitude for yield and

the importance of dominance variance for plant height, number of bolls per plant and fibre

length. Kalsy and Garg (1988) showed additive, dominance and epistatic type of gene action

for the inheritance of seed cotton yield and boll weight. The involvement of epistasis for seed

cotton yield and number of bolls per plant was noticed by Gill and Kalsy (1981) and

Randhawa et al. (1986) while partitioning the components of generation means for yield

some yield related traits. Similarly, epistasis was also observed for the traits related to fibre

quality except for fibre length, Singh et al. (1983) by evaluating the crosses involving six

generations in generation mean analysis. In another study, genic effects for yield of seed

cotton and number of sympodial branches per plant were estimated from two Upland cotton

crosses following generation mean analysis by Iqbal and Nadeem (2003) from six

populations (P1, P2, F1, F2, BC1 and BC2). The generation mean analysis advocated the

presence of additive gene action in crosses i.e., S-12 × S-14, S-12 × Albacala (69)11,

LRA-5166 × S-12 and LRA-5166 × S-14 for number of sympodial branches per plant. The

scaling test revealed involvement of epistasis in all the crosses, except for S-14 × LRA-5166

for yield of seed cotton per plant. The rest of all the crosses were predominately under non

additive genetic control except for S-14 × LRA 5166 for yield of seed cotton plant, hence

delayed selection would be fruitful in these crosses. Liu et al. (2000) by using generation

mean analysis assessed the genetic effects of agronomic characters in intraspecific hybrids of

transgenic × non-transgenic upland cotton. The studies pointed out that plant height and boll

number, 2.5 % span length, fibre strength and micronaire were controlled by epistatic gene

action. While seed cotton yield and lint yield were controlled by additive, dominance and

epistatic gene action, with dominance being an important element. Dhillon and Singh (1980)

inferred from his studies on generation mean/variance analysis, that expression of different

components of variation were much influenced by environmental effects. Prakash (1982)

while analyzing parentals, BC1, BC2 and F2 generations of the crosses in Gossypium

arboreum revealed the importance of both additive and non-additive components of genetic

variance for yield and yield related traits.

32

Bollworm attack affect yield. Genetics of yield and its component in cotton under

artificial bollworm infestation was studied by Murthy (1998). The study revealed the

predominance of dominance gene action, although both additive and dominance gene effects

were involved in the control of the majority of yield components, as well as bollworm

damage. The parents for the traits seed index, lint index and 2.5 % span length, and boll and

locule damage. Overdominance for all characters was observed except for 2.5 % span length

which exhibited partial dominance. Epistasis was observed in the case of number of bolls and

seed cotton yield per plant. All traits studied were controlled by 1-2 groups of dominant

genes, except for seed cotton yield. Studies of Ramalingam and Sivasamy (2002) indicated

the importance of additive and non- additive gene effects in the expression of seed cotton

yield. However, the estimates of genetic components of variance indicated the importance of

additive, additive × dominance and dominance × dominance type of non- allelic interaction

for seed cotton yield. Additive × additive type of interaction had been reported for lint

percentage and boll weight (McCarty et al., 2004 b). In contrary, the findings of

Khan et al. (1999) and Ahmad et al. (2001) showed absence of epistasis for any trait.

However, additive and non- additive type of gene action had been reported for

morphological, economic and yield traits (Kumaresan et al., 1999 and Deshpande and Baig,

2003). Plant height, number of bolls per plant, boll weight (Khan et al., 1999 and Ahmad

et al., 2001) and seed cotton yield (Ahmad et al., 2001) showed additive type of gene action.

Plant height and number of sympodial branches per plant were also under the control of

additive type of gene action (Neelima et al., 2004). Stoilova and Taofik (1998) and

Hassan et al. (1999) showed that boll weight was controlled by additive effects. Lint

percentage and seed cotton yield (Sayal and Sulemani, 1996 and Liu et al., 2000) was

controlled by additive type of gene action (Subhan et al., 2002). Plant height and fibre

fineness were found being controlled through non- additive type of gene action (Islam et al.,

2001).

Findings of Hassan et al. (1999) in order to study the nature of the gene action,

ascertain non- additive genetic effects for boll number, seed cotton yield, plant height, lint

percentage and staple length. Findings of Subramanian et al. (2002) and Neelima

et al. (2004) pin pointed non- additive type of gene action for number of bolls per plant, boll

weight, seed cotton yield, whereas, Subramanian et al. (2002) also observed non-additive

33

type of gene action for lint percentage, plant height, number of sympodia. Number of

monopodia per plant also showed non- additive type of gene action (Neelima et al., 2004).

A number of studies related to fibre traits in upland cotton have been reported in the

literature which directs the attention of the plant breeders for the fibre improvement. Studies

of Nadeem and Azhar (2005) suggested the traits like fibre length and fibre strength, studies

of Pathak (1975) also suggested additive type of gene action for fibre strength and

Mukhtar et al. (2000) suggested fibre fineness, fibre elongation and uniformity ratio under

additive type of gene action with no non-allelic interactions. Pathak (1975) used generation

mean analysis and found incomplete dominance for long fibre over short fibre and over

dominance for fibre fineness. In a study conducted by Liu et al. (2000) 2.5 % span length,

fibre strength and micronaire were controlled by epistatic gene action. Hendawy et al. (1999)

observed additive × additive type of gene interaction for fibre traits, however, fibre length,

uniformity ratio, fibre strength and fibre fineness displayed additive type of gene action,

though there was a high magnitude and significant dominance effects were also observed.

McCarty et al. (2004 b) also observed additive × additive interaction for fibre strength.

Pavasia et al. (1999) investigated that lint percentage, 2.5 % span length and fibre fineness

was controlled by additive type of gene action. Liu and Han (1998) and Nistor and Nistor

(1999) observed that fibre length exhibited additive type of gene action. Sayal et al. (1996),

Hassan et al. (1999) and Nistor and Nistor (1999) observed staple length, Islam et al. (2001)

observed fibre fineness, Mukhtar et al. (2000) observed fibre strength, Haq and Azhar (2005)

observed fibre strength and fibre fineness and Liu and Han (1998) observed fibre uniformity

ratio, fibre elongation and fibre fineness being controlled by non-additive (dominance) type

of gene action.

Genetic effects for fibre properties in cotton by Gamble’s six-parameter model in the

analysis of generation means indicated that although dominance and dominance × dominance

genetic effects were more important in the fibre length, additive genetic effects were also

important. An over dominance type of gene action for fibre fineness while, additive type of

gene action governed for fibre strength (Pathak, 1975). Similarly, inheritance of fibre quality

traits in upland cotton having non-preference traits for insect pests were studied in 8 × 8

diallel cross and were evaluated in F1 and F2 generations. Additive and dominance

components were studied. The values of H1 and H2 for all characters were higher than those

34

of D which indicated the predominance of non-additive genetic variance to control the most

of these characters under study. H2 component was estimated to be smaller than H1,

indicating unequal portion of positive and negative allelic frequencies. The positive F-value

indicated gene asymmetry, i.e; there were more dominant than recessive alleles in the parents

for these characters. It was further confirmed by relative portion of dominant and recessive

alleles, which were more than 1 for all these traits, proving dominant alleles were in excess

than the recessive alleles. The dominance effect estimated by the heritability estimates was

not due to heterogeneity of the loci in these traits. The environmental component of variation

remained non-significant and indicated insignificant role in the phenotypic expression of

these traits. Recurrent selection was suggested for improvement of these traits (Murtaza

et al., 2004).

Significant genetic variation was found for fibre length, uniformity ratio, fibre

strength, fibre elongation and fibre fineness. Dominance genetic variance was greater than

additive genetic variance for all of the fibre traits (May and Green, 1994 and Tang

et al., 1996). The significant additive × environment variance components in the studies of

Tang et al. (1996) indicated a lack of useful additive genetic variability for fibre traits. This

suggested that selections for pure lines within the F2 populations would have limited success

in improving fibre traits.

2.9. Assessment of heterosis, heritability, genetic advance, correlations and inbreeding

depression for yield, fibre and other morphological traits

(i) Heterosis

In general heterosis refers to the increase of F1 fitness and vigour over the parents. Heterosis

and hybrid vigour are synonymously used terminologies. In general, cross-pollinated species

show heterosis, particularly when inbreds are used as parents. Heterosis in the form of hybrid

or synthetic varieties is the commercial utilization of heterosis. Various scientists’

contributions are available in literature regarding heterosis for yield, fibre and other plant

attributes in cotton. Some of them are described as below:

Wu et al. (2002) formulated his heterotic results on yield related attributes.

Significant results of heterosis over better parent and standard for seed cotton yield were

inferred from the studies of Soomro et al. (2000), Banumathy et al. (2001), Qian et al.

35

(2001), Zhang et al. (2003). Soomro et al. (2000), from their results of high magnitude of

heterosis pinpoint towards heavier bolls and high lint percentage. Potdukhe (2001) in

addition to seed cotton yield, high estimates of heterosis over the better parent were found for

number of monopodia, number of sympodia, bolls per plant lint percentage % and plant

height. It was inferred that this high magnitude of heterosis over better parent was due to

number of sympodia and bolls per plant. While Hassan et al. (1999) showed that superiority

in yield may be attributed to number of bolls per plant than boll weight. Hassan et al. (1999)

and Baloch (2003) showed increased heterotic effects for seed cotton yield and number of

bolls per plant.

Kamaresan et al. (1999) studied the traits bolls per plant, seed cotton yield and boll

weight and found significant heterosis for these traits, while heterosis of all types were found

significant come from the studies by Rajan et al. (2000) for plant height, number of

sympodial branches and number of bolls per plant. Malek and Shamsuddin (1999),

Kaynak et al. (2000) and Kowsalya et al. (2000) observed positive heterosis for seed cotton

yield over mid parent. Interspecific and intraspecific hybridization studies by Ravindranath

et al. (2000) and Manimaran and Raveendran (2002) revealed that there is more pace of

heterosis for interspecfic hybrids than intraspecific hybrids for the traits like number of bolls

per plant, boll weight and lint percentage.

Positive heterosis over mid parent for lint percentage and staple length and fibre

strength were observed whereas, negative effect were inferred for fibre fineness

(Arshad et al., 2001), whereas, positive heterotic effects for fibre yield traits were observed

(Bertini et al., 2001 and Desphande and Baig, 2004). Positive and significant heterosis for

fibre quality traits with the absence of epistasis was observed by Feki and Gelil (2001) in

Egyptian cotton. Baloch (2003) observed high heterotic effect for fibre length while

Manimaran and Raveendran (2002) observed highest heterotic effects for fibre length and

fibre strength.

(ii) Heritability and Genetic advance

Both heritability and genetic advance are important selection parameters. In crop

improvement, only the genetic component of variation is important since this component is

transmitted to the next generation. The extent of contribution of genotype to the phenotypic

36

variation for a trait in a population is ordinarily expressed as the ratio of the genetic variance

to the total variance. Estimates of heritability serve as a useful guide to the plant breeder.

Proportion of variation either genotypic (broad sense) or additive (narrow sense) is helpful to

the plant breeder in one or the other way. High heritability for any trait makes the selection

easier than that of a character with low heritability which makes selection difficult or

virtually impractical due to masking effect of the environment on genotypic effects.

Improvement in the mean genotypic value of selected plants over the parental population is

known as genetic advance. It is actually the measure of genetic gain under selection.

Estimates of genetic advance help in understanding the type of gene action involved in the

expression of various polygenic characters. High values of genetic advance are indicative of

additive gene action and low values are indicative of non- additive gene action.

Pandey and Singh (2002) drew high heritability and genetic advance estimates for

seed cotton yield, boll weight and plant height. Kumari and Chamundeswari (2005) inferred

high estimates of heritability and genetic advance for seed cotton yield. High heritability and

low genetic advance was reported for number of monopodia. For number of bolls per plant,

high estimates of heritability and moderate genetic advance was noticed, while the other

traits like plant height, boll weight and sympodial branches exhibited low heritability coupled

with low genetic advance. According to these findings by Kumari and Chamundeswari

(2005) while studying heritability and genetic advance on various attributes of cotton,

suggested that high heritability and genetic advance were under control of additive type of

gene action. A high heritability with low genetic advance was an indication towards non-

additive type of gene action. High estimate of heritability with moderate genetic advance

showed the involvement of additive and non-additive gene actions whereas; for traits with

low heritability and genetic advance revealed that these characters were controlled by

polygenes.

For fibre quality traits like fibre length, fibre strength, lint percentage %, uniformity

ratio %, fibre elongation and fibre fineness, heritability estimates meant a lot. Nistor and

Nistor (1999) reported high estimates of narrow sense heritability for fibre length whereas,

Pandey and Singh (2002) not only reported high estimates of heritability for fibre length but

also for lint percentage %. Yuan et al. (2002) in a study on fibre attributes like fibre strength,

fibre length, fibre elongation and fibre fineness, found high estimates of broad sense

37

heritability. However, there was some involvement of environmental interaction. High

heritability with low environmental it was observed for fibre length and fibre strength

whereas, low heritability in broad sense was estimated for fibre length and uniformity ratio.

High environmental interaction was also observed for fibre elongation and fibre fineness.

Feki and Gelil (2001) and Feki et al. (1998) also estimated narrow sense heritability for fibre

quality attributes. Adl et al. (2001) estimated high estimates of broad sense heritability than

narrow sense heritability. Ulloa (2006) observed high heritability estimates in an okra-leaf

population of cotton and suggested for some practical applications for simultaneous

improvement of multiple fibre traits.

(iii) Correlations

Correlation estimates are necessary for any breeding programme. This actually gives the

exact relationship of one character and its resultant effect on to the other character. Hence,

improvement in one trait may be helpful in subsequent improvement in the related trait.

Wu et al. (2003) suggested that selecting and breeding new cultivar of upland cotton with

high yield, good quality and resistance to diseases and insects is possible if the correlations

between those important characters are well coordinated. The genotypic correlations (Tyagi,

1987) were found generally higher than that of the phenotypic correlations for yield

parameters studied in F2 and F3 generations of upland cotton. For morphological and yield

traits, Arshad et al. (1993), Echekwu (2001) and Naveed et al. (2004 a) observed significant

association of plant height and number of bolls per plant with seed cotton yield. Azhar et al.

(1984) concluded that number of bolls per plant, boll weight and lint percentage are

positively and significantly correlated with seed cotton yield. In the study of Singh et al.,

1968 and Konoplya et al., 1979, it was found that the plant height and number of bolls were

positively and significantly correlated with each other at both levels.

Sultan et al. (1999) drew their inferences for other morphological traits with fibre

yield and found that number of sympodia per plant, boll number per plant, boll weight and

lint percentage showed highly significant positive correlation with fibre yield at genotypic

and phenotypic levels. There was an antagonistic association between boll number and boll

weight. Boll weight was positively correlated with seed cotton yield (Azhar et al., 1984).

Echekwu (2001) also observed a significant and positive correlation of plant height with seed

cotton yield, number of sympodial and monopodial branches. Whereas, a significant negative

38

correlation was obtained between plant height, lint percentage, number of monopodial and

sympodial branches.

Auld et al. (2000), Herring et al. (2004), Avgoulas et al. (2005) and Ulloa (2006)

observed positive correlation between fibre length and strength. But Naveed et al. (2004 b)

observed negative association between fibre fineness and fibre length at both levels and

significant only at phenotypic level, however, seed cotton yield was positively and

significantly associated with fibre fineness and fibre strength. This positive and significant

correlation between seed cotton yield and fibre length and strength was also noticed by

Echekwu (2001), who not only observed a significant and positive correlation of seed cotton

yield with fibre length and micronaire index but also a significant negative correlations

between plant height and lint percentage, fibre strength and micronaire index.

Fibre fineness and staple length both have been reported to influence lint percentage

(Ulloa and Meredith Jr., 2002). Ulloa (2006) observed a positive correlation of fibre fineness

with uniformity ratio. lint percentage % was found to be negatively associated with fibre

length but was positively correlated with fibre strength, similarly negative association of

fibre length with micronaire means positive association of fibre length with fibre fineness

were recorded by Singh et al. (2002) and Badr and Aziz (2000). The findings of Ulloa (2006)

gave negative correlation of lint percentage % with fibre strength.

(iv) Inbreeding depression

It refers to decrease in fitness and vigour due to inbreeding. The information of heterosis and

inbreeding depression together provide the information about the type of gene action,

involved in the expression of various quantitative traits. A high heterosis followed by

inbreeding depression is the indication towards non- additive type of gene action. If the genes

behave same in both F1 and F2, then there is the involvement of additive genes. About 50%

inbreeding depression expressed by F2 hybrids in the study by Baloch (2002). Significant or

highly significant inbreeding depression ratios were observed for seed cotton yield, boll

weight and plant height (Qian et al., 2001). In the study of Yadav and Yadava (1987) while

investigating the nature of gene action for yield, number of bolls per plant and boll weight

found a low level inbreeding depression for these traits.

39

The current scenario of pesticide usage and its import, which is an additional burden

on the farming community and a matter of concern for the policy makers to deal with.

Therefore, it is desired to exploit the natural traits like okra leaf, trichomes and gossypol

glands; conferring non-preference against the insect pest infestation. The literature supports

the existence of their non-preference nature against the insect pest population. An extensive

review of the genetics of these insect non-preference traits along with the traits of agronomic

importance have been reviewed in this section with the possible aspects of their mutual

associations. The present research is proposed on the following hypothetical lines:

• Transference and incorporation of the genes conferring insect non-preference traits in

different genetic backgrounds of better cotton genotypes.

• Genetic analysis of the traits related to the insect resistance and those related to

agronomic importance through the powerful technique of generation mean analysis

for the study of the gene action.

• Study of the inheritance pattern involved in the expression of the insect non-

preference traits.

• Study of the effect of these insect non-preference traits on the quality of cotton fibre,

yield and other agronomic traits.

40

CHAPTER 3

MATERIALS AND METHODS

Present studies pertaining to the inheritance of insect non-preference (insect resistant) plant

traits like okra leaf type, gossypol glands and trichomes along with other agronomic and fibre

related characteristics were conducted in the Department of Plant Breeding and Genetics,

University of Agriculture, Faisalabad (Pakistan). Seed of the 31 cotton accessions was

collected from Cotton Research Institute (CRI), Ayub Agricultural Research Institute,

Faisalabad, Central Cotton Research Institute (CCRI), Multan and the Department of Plant

Breeding and Genetics, University of Agriculture, Faisalabad. The collected germplasm was

assessed for the traits i.e; okra leaf type, normal leaf, high gossypol glands, normal gossypol

glands, glandless types, pilose hairiness and low/sparse hairiness. List of germplasm used for

preliminary assessment for the above mentioned traits is given in Appendix-I.

3.1. Development of plant material for genetic studies

Based upon the assessment made on the above mentioned traits, five genotypes were

selected. The particulars of the selected genotypes are described in Table 3.1.

Table 3.1: Distinctive morphological features of the upland cotton accessions assessed

for the traits under study

S. No.

Variety/ Accession

Parentage Distinct Features Origin

1 Acala 63-74 - Broad leaf, Glandless Exotic

2 CIM 446 CP 15/2 × S 12 Normal leaf, Normal gossypol glanding and

High fibre quality.

CCRI, Multan Pakistan

3 FH 1000 S 12 × CIM 448 Normal leaf, Normal gossypol glanding,

High yielding.

CRI, Faisalabad Pakistan

4 HRVO-1 B-557/2/Gambo Okra/Rajhans/3/Rajhans

Okra leaf, pilose hairiness

CRI, Faisalabad Pakistan

5 HG- 142 - Broad leaf, High Gossypol glands

Exotic

41

3.1.1 Generation developed in glasshouse

The selected germplasm was selfed for four generations by growing twice a year, in a

glasshouse and field during 2003 to 2004, to ensure homozygosity for the traits under study.

The selected parents with contrasting traits (Table 3.1) were planted in 30 cm × 30 cm

earthen pots, containing a mixture of equivalent proportion of sand, soil and farmyard

manure, during November, 2004 in a glasshouse. Temperature in the glasshouse was

maintained at 30 ± 20C during the day and 25 ± 20C at night by using built in steam heaters.

The plants were exposed to natural sunlight supplemented with artificial lighting, for a

photoperiod of 16 hours (ICAC, 2007b). Seedlings were thinned to one plant per pot after

two weeks of planting and after 14 days, 0.25 g of urea (46% N) was added to each pot and

plants were watered daily. Crosses were attempted in four groups to obtain F0 seed during

February through March, 2005. The emasculation of the floral buds was done in the evening

and pollination was carried out the next day morning. In order to avoid any chances of

foreign pollen contamination, the emasculated buds were covered with a soda straw tube and

the flower from the pollen parent was covered with a butter paper bag. The selfed seed of the

parents was obtained by covering their floral buds with butter paper bags. The crossing

scheme is given in Table 3.2.

Table 3.2: Scheme of crossing

S. No. CROSS TRAIT CONSIDERED

1 HRVO-1 × FH-1000 Okra leaf, pilose hairiness × Normal/broad leaf, glabrous

2 HRVO-1 × CIM-446 Okra leaf, pilose hairiness × Normal/broad leaf, glabrous

3 HRVO-1 × Acala 63-74 Okra leaf, pilose hairiness, normal glanding × Normal/broad leaf, glabrous, glandless

4 HRVO-1 × HG-142 Normal glanding × High glanding

3.1.2 Generation development in field

The F1 and their parents were planted in the research area of the Department of Plant

Breeding and Genetics, University of Agriculture, Faisalabad during the normal crop season

of 2005-06. Seed for the F0, F2, BC1 and BC2 generations was produced for each of the four

combinations through manual selfing and crossing. The F1 plants of each cross were divided

42

in three groups for developing BC1, BC2 and F2 for each combination. The fresh F0 seed was

developed through selfing.

3.1.3 Field sowing and planting geometry

All the six generations of the four crosses (Table 3.2) were sown during the normal crop

season of the year 2006-07. The experimental field was fertilized with N-P-K at the rate of

100-75-00 Kg/ha. Irrigation both by canal and turbine water was applied to the experimental

material with the interval of 7-10 days. The experiment in the field was laid out in a

Randomized Complete Block Design with three replications of each of the six generations of

the three crosses. A single plot per replication was assigned to each of the parents and their

respective F1, while, four plots per replication were assigned to each of the backcrosses and

eight plots per replication were assigned to raise the F2 population of each cross. The length

of the plot was maintained at 4.5 m, accommodating approximately 15 plants spaced 30 cm

apart. The distance between the rows was 75 cm. All other agronomic and cultural practices

were kept uniform to minimize the experimental error. Recommended doses of sprays of

pesticide were also applied to the experimental material from time to time in order to prevent

possible insect damage.

3.1.4 Field evaluation at maturity

For the parents and F1, data were recorded on 10 randomly selected competitive plants in

each replication for each trait except for number of trichomes/unit area and quantification of

gossypols for which five competitive plants were randomly selected. For F2 and backcross

generations, the data for all the traits were recorded from 50 and 30 randomly selected

competitive plants respectively in each replication. The data on the individual plants in each

generation were recorded at appropriate times for the following parameters.

3.1.4.1. Plant height (cm)

When the apical bud of main stem ceased to grow, final height of plants was measured in

centimeters from the first cotyledonary node to the apical bud. The average plant height of all

the generations in each replication was computed.

43

3.1.4.2. Number of monopodial branches

At maturity the monopodial branches on each plant were counted from all the guarded

selected plants in each of the generations and average number of monopodia was counted.

3.1.4.3. Number of sympodial branches

The number of sympodia on individual plants were counted and recorded. Average number

of sympodial branches was worked out for all the genotypes.

3.1.4.4. Number of bolls

The effective bolls picked at the time of picking were counted on individual plant basis. The

total count of bolls was obtained summing the bolls picked in the two pickings. Average

number of bolls per plant was calculated in all the generations.

3.1.4.5. Boll weight (g)

Average weight per boll was obtained by dividing whole produce of seed cotton yield per

plant by its respective number of bolls. The mean boll weight of a generation in each

replication was calculated for the purpose of data analysis.

3.1.4.6. Seed cotton yield (g)

Seed cotton was hand picked from the mature bolls in two pickings. The total plant produce

of seed cotton was weighed by using electrical balance, and the average seed cotton yield of

each generation was calculated in each replication.

3.1.4.7. Lint percentage

Clean and dry samples of seed cotton were ginned separately with a single roller electric

ginning machine. The lint obtained from each sample was weighed, and ginning percentage

was calculated using the following formula. Mean lint percentage of each generation was

calculated for the purpose of data analysis.

100× sample a in cotton seedof Weight

sample a in lintof Weight= percentage Lint

44

3.2. Fibre quality characteristics

Fibre quality characteristics like fibre length, fibre fineness, fibre strength, fibre elongation

percentage and fibre uniformity ratio of each plant of a generation were measured using Spin

lab high volume instrument (HVI-900-A), M/S Zellweger Uster Switzerland, available in the

Department of Fibre Technology, University of Agriculture, Faisalabad. HVI-900-A is a

computerized high volume instrument, which provides a comprehensive profile of fibre

quality characteristics. A minimum of 10 g sample of lint from each of the guarded plants in

each generation was pre-conditioned to moisture applicability for at least four to five hours

prior to testing in the HVI. HVI measures fibre quality characteristics according to the

international trading standard. Means of these fibre characters were obtained for genetic

analysis.

3.2.1. Fibre length (mm)

Fibre length was measured using HVI-900-A, on the basis of the fibrograph. The samples

were prepared at fibro sampler in the form of fibre comb and the fibrograph-910 brushed the

sample fibres automatically by vacuum action, and optical density of the sample was

displayed on the screen. The mean fibre length was derived according to the ASTM Standard

(1977a) through the procedure laid down by Hunter (1991). The average fibre length was

calculated in mm for further analysis.

3.2.2. Fibre fineness (micronaire)

Fibre fineness of micronaire value was measured on fibrofine-920 device of HVI-900- A.

When lid of the chamber containing a sample is closed, the sample is compressed to a fix and

known volume. The sample was weighed on an electric balance before placing in the test

chamber. This balance transmitted the mass through the control processor. This mass was

accepted if the weight was between 8.5 and 11.5 g of cotton. The measured values of mass

and pressure calculated in fineness value according to ASTM standard (1977c). Thus the

readings were recorded and averaged for statistical analysis.

3.2.3. Fibre strength (g/tex)

Fibre bundle strength was determined by Pressley strength tester using the flat bundle

method as specified by ASTM standard (1977b).

45

3.2.4. Fibre elongation (%)

The fibre elongation was also measured on HVI-900-A. When a sample was moved into an

optical sensor, where the test for length and strength were performed, the percentage increase

in the length before fibre breakage was measured (Anonymous, 1992a, 1992b).

3.2.5. Fibre uniformity ratio (%)

The sample was moved into an optical sensor of HVI-900-A and the reading of optical

density of the sample was displayed on the screen. The uniformity ratio of fibre was

measured according to ASTM standard (1977a). Recorded readings were then averaged for

statistically analysis.

3.3. Morphological characters affecting insect resistance

3.3.1. Okra leaf

Okra leaf trait is characterized by deeply cleft and narrowly lobed leaves with less surface

area per leaf than normal leaf of cotton (Fig.3.1b).

3.3.1.1. Rating system for leaf shape

In order to classify leaf shape, a qualitative system of classification including a visual rating

of leaf shape was used. Leaf shape was categorized into, broad normal leaf (grade-I), sub-

okra (grade-II) and narrow okra (grade-III) following Rahman and Khan (1998). Data for leaf

classification were recorded in accordance with the given categories.

3.3.2. Trichomes

Trichomes represent the presence of small hairs on the cotton plant (Fig. 2). The trichome

density on leaves was estimated following two criteria proposed by Bourland et al. (2003).

46

Fig. 1. Variable classes in leaf types

(a) Normal leaf

(b) Okra leaf

47

(c) Sub-okra leaf

(d) Semi-okra leaf

48

Fig. 2. Variable classes in leaf trichomes

(a) Pilose hairiness

(b) Normal hairiness

49

(c) Intermediate class of hairiness

50

3.3.2.1. Qualitative grading system for trichomes

Three leaves at random each from upper, middle and lower portion of the selected plants

were used to assess for the trichome density rating. A rating system of trichomes on the

abaxial surface of leaf, using a scale of 1 for sparsely (non) hairiness, 2 for moderate number

of trichomes, 3 for (pilose) hairiness was carried out.

3.3.2.2. Quantitative measure of leaf trichomes

The same leaves mentioned above for the study of qualitative grading were used to assess for

the quantitative measure of trichomes on the abaxial leaf surface (Muttuthamby et al., 1969).

Observations pertaining to the number of trichomes were recorded with the help of an index

card within an area of 0.1 cm2 laid over the abaxial side of each leaf from three different

positions and averaged. The resultant mean values for number of trichomes from three

different portions of the plant were worked out as the final reading for average number of

trichomes per unit area. Trichomes in the 0.1 cm2 area were counted with the aid of high

magnifying power microscope (Olympus Z61). Each bunch of stellate trichomes was counted

as a separate trichome.

3.4. Biochemical characters affecting insect resistance

3.4.1. Gossypol glands

Gossypol glands are dot like toxic glands found on all parts of the plant. Fig. 3 shows

variable classes of gossypol glands present on the unopened bolls. Total gossypol defines

gossypol and gossypol derivatives, both free and bound, which are capable of reacting with

3- amino-1-propanol in dimethylformamide solution to form a diaminopropanol complex,

which then reacts with aniline to form dianilinogossypol. The gossypol glands present on the

surface of the unopened bolls were quantified on a spectrophotometer according to the

protocol (A.O.C.S., Official Method, 1989) as described in the following sections. For

gossypol studies two crosses (HRVO-1 × HG- 142 and HRVO-1 × Acala 63-74) comprising

of a high glanding (HG- 142) and glandless (Acala 63-74) parent.

51

(a) High glanding Gl2Gl2Gl3Gl3

(b) Normal glanding Gl2Gl2gl3gl3

(c) Intermediate high glanding Gl2Gl2Gl3gl3

(d) Glandless gl2gl2gl3gl3

(e) Intermediate glandless Gl2gl2gl3gl3

Fig. 3. Variable classes of boll gossypol glands

52

3.4.2.1. Apparatus/glassware/plasticware

1. Water bath, for operation at 95-100 0C with clamps for supporting volumetric

flasks and test tube stands.

2. Spectrophotometer, for operation at 440 nm 1 cm light-path cells.

3. Volumetric flasks 25 and 50 mL.

4. Graduated cylinders of 1, 10 and 100 mL.

5. Beakers of 25, 50, 100 and 500 mL.

6. Test tubes of 15 mL capacity with test tube stand.

7. Mouth pipette of 10 mL.

8. Micro-pipettes of 1 and 5 mL.

9. Medium retention, 11 cm diameter, Whatman No. 2, S and S 597, filter papers.

10. Water cooled distillation apparatus.

3.4.2.2. Preparation of Reagents

The laboratory grade Isopropyl alcohol (2-propanol), n-hexane (boiling range 68-

690C), dimethylformamide, 3-amino-1-propanol, glacial acetic acid and aniline were

purchased from SIGMA suppliers and reagents were prepared as follows:

1. The aniline was redistilled over zinc dust, using a water cooled condenser.

Redistillation was carried out with a water cooled condenser apparatus, set in a way that from

one end water comes and leaves the apparatus from the other end making the vapours

condensed. First and last 10 % of distillate was discarded to avoid any impurities.

2. Isopropyl alcohol-hexane mixture was prepared by mixing 60 volumes of

isopropyl alcohol with 40 volumes of n-hexane.

3. Complexing reagent was prepared with 2mL of 3-amino-1-propanol and 10 mL

glacial acetic acid made to 100 mL volume with dimethyl formamide.

4. Standard gossypol acetic acid solution was prepared by dissolving 24 mg of

gossypol acetic acid (powder) in the complexing reagent and volume was made to 50 mL

with the complexing reagent. Thus the solution contained 0.48 mg gossypol acetic acid per

mL. The mg gossypol acetic acid used was multiplied with 0.8962 to obtain mg of gossypol

(A.O.C.S, 1989).

53

3.4.2.3. Preparation of standard curve of gossypol acetic acid

1. From the standard gossypol acetic acid solution prepared (Step-4 of Preparation

of reagents), the aliquots of 1, 2, 4, 6, 8 and 10 mL were taken and a final volume of 10 mL

was made with the complexing reagent. Pure complexing reagent (10 mL) of pure

complexing reagent was used as blank.

2. The separate flasks containing a total volume of 10 mL made for each of the

aliquots and blank solution were heated in a water bath (95-1000C) for 30 minutes, cooled to

room temperature, and finally diluted to a total volume of 50 mL with isopropyl alcohol-

hexane solution and mixed well. These aliquots of standard gossypol acetic acid and blank

were stored as stock solutions in the refrigerator.

3. 2 mL volume of each of these aliquots of the standard and blank were pipetted in

duplicate into separate volumetric flasks of volume 25 mL.

4. One set of the standard aliquots and the reagent blank were diluted to make the

final volume of 25 mL with the isopropyl alcohol-hexane solution and reserved as reference

solutions for absorbance measurements.

5. 2 mL aniline was added to the other set of standard aliquots and the blank, heated

in a water bath (95-1000C) for 30 minutes, cooled to room temperature, finally diluted up to

the volume of 25 mL with the isopropyl alcohol-hexane solution and mixed well. Allowed to

cool down for 1 h at room temperature before determining absorbance.

6. Optical density (OD) of reagent blank and the standard aliquots was determined

on a spectrophotometer at 440 nm wave length. The OD value of reagent blank was

subtracted from the OD value of each standard to obtain the corrected value.

Corrected absorbance = OD of each standard – OD of reagent blank

7. Calibration factor was determined by dividing mg gossypol in standards by

corrected OD of each standard to obtain calibration factors. Average of the factors were

determined for each of the standards, and used to calculate mg gossypol in sample aliquots.

Factor = mg gossypol in standard

Corrected OD

54

3.4.2.4. Sample preparation

Sample weight and aliquot used for aniline reaction depends on expected total gossypol

content. Ideally, the analytical sample should contain 0.5-5.0 mg of gossypol, and the aliquot

for the aniline reaction about 0.1 mg gossypol. Before the sample preparation the unopened

cotton boll was washed with water. The outer surface of the bolls containing the gossypol

glands was peeled off and weighed on a digital balance. About 1 g sample thus obtained was

crushed in a mortal and pestle using one drop of glacial acetic acid and one drop of 70 %

aqueous acetone. A small piece of aluminum foil was used for weighing and transferring

sample to flask.

3.4.2.5. Protocol of sampling for quantification of gossypols

1. The crushed sample was transferred into a test tube and 1 mL of the complexing

reagent was added.

2. 1mL of complexing reagent was used as reagent blank.

3. Sample and blank were heated in a water bath (95-1000C) for 30 minutes, cooled to

room temperature and diluted to 4 mL volume with isopropyl alcohol-hexane mixture and

shook well.

4. Sample extract was filtered through 11 cm medium retention paper into test tube,

discarding first 1 mL of filtrate.

5. Two mL duplicate aliquots of sample and blank were taken into test tubes.

6. One set of the sample and blank aliquots was diluted to 10.5 mL volume with the

isopropyl alcohol-hexane mixture and reserved as reference solutions for absorbance

measurement.

7. One mL aniline was added to the other set of sample and blank aliquots, heated in

water bath (95-1000C) for 30 minutes, cooled to room temperature, diluted with volume of

9.5 mL of isopropyl alcohol-hexane solution and mixed well. Allowed to stand for 1 h at

room temperature before determining absorbance.

8. Optical density (OD) of the reagent blank reacted with aniline was determined

using blank aliquot without aniline as reference solution. The OD value of reagent blank was

subtracted from the OD value of each standard to obtain the corrected absorbance.

55

9. OD the sample aliquots reacted with aniline was determined using diluted sample

aliquot without aniline as reference solution. The OD value of reagent blank was subtracted

from the OD value of the sample aliquot reacted with aniline to obtain corrected absorbance.

Corrected absorbance = OD of sample aliquot – OD of reagent blank

(Aniline treated) (Aniline treated)

10. From the corrected absorbance mg gossypol in sample aliquot were determined

by multiplying OD by either the mean calibration factor, or reference to calibration graph.

Total gossypol % was calculated by the formula (A.O.C.S., Official Method, 1989).

Total gossypol % = 5 × G W × V

Where,

G = mg gossypol in sample aliquot.

W = weight of sample in grams.

V = volume of sample aliquot used for analysis.

3.5. Statistical Analyses

The data regarding all traits measured at plant maturity were analyzed using analysis of

variance technique (Steel and Torrie, 1980) using MSTATC version 1.5.

3.5.1. Genetic basis of variation for traits under study

The genetic bases of variation for the measured traits were studied by analyzing the genetic

data on the six generations (P1, P2, F1, BC1, BC2, and F2) of the three crosses. The generation

mean analysis (Mather and Jinks, 1982) was performed using a Computer programme written

by Dr. H.S. Pooni, School of Biological Sciences, University of Birmingham, U.K. Means

and variances of the parents, BC1, BC2, F1 and F2 generations used in the analysis were

calculated from individual plant data pooled over replications. The coefficients of genetic

components of generation mean are presented in Table 3.3. A weighted least square analysis

(Mather and Jinks, 1982) was performed on the generation means commencing with the

simplest model using parameter m only. Further models of increasing complexity (md, mdh,

etc.) were fitted if chi-square value was significant. The best fit model was one which had

56

significant estimates of all parameters along with non-significant chi-square value. For each

trait the higher value parent was taken as P1 in the model fitting.

Table. 3.3: Coefficients of genetic effects for the weighted least squares analysis of

generation means (Mather and Jinks, 1982)

Generations Components of genetic effects

m [d] [h] [i] [j] [l]

P1 1 1.0 0.0 1.0 0.0 0.0

P2 1 -1.0 0.0 1.0 0.0 0.0

F1 1 0.0 1.0 0.0 0.0 1.0

F2 1 0.0 0.5 0.0 0.0 0.25

BC1 1 0.5 0.5 0.25 0.25 0.25

BC2 1 -0.5 0.5 0.25 -0.25 0.25

A weighted least square analysis of variance was also performed. The coefficients of

genetic components i.e; additive (D), dominance (H), additive × dominance (F) and

environmental variation (E) are presented in Table 3.4. The model fittings E, (D and E),

(D, H and E), (D, F and E) and (D, H, F and E) were tried. The best fit model was selected,

when estimates of chi-square were not significant with all other significant parameters.

Table.3.4: Coefficients of genetic variance components for the weighted least square

analysis of generation variances (Mather and Jinks, 1982)

Generations Components of variation

D H F E

P1 0.00 0.00 0.00 1

P2 0.00 0.00 0.00 1

F1 0.00 0.00 0.00 1

F2 0.50 0.25 0.00 1

BC1 0.25 0.25 -0.50 1

BC2 0.25 0.25 0.50 1

57

3.5.2. Estimation of narrow sense heritability

Estimation of narrow sense heritability (h2N) in F2 (Warner, 1952) and F infinity generations

(Mather and Jinks, 1982) from the components of variance from the best fit model of the

weighted least squares analysis by using the formulae:

a) h2 N (F2) = 0.5D/ VF2

b) h2 (F∞ ) = D/D+E

3.5.3. Genetic advance

Expected genetic advance in the next generation was computed by the following formula

(Falconer and Mackay, 1996).

G.A = K. ∧σp .h2

Where,

G.A = Genetic advance

K = Selection differential, being 2.06 at 5 % selection intensity ∧σp = Standard deviation of phenotypic variance of the population under selection

h2 = heritability estimates in fraction of the trait under study.

3.5.4. Heterosis and inbreeding depression

Magnitude of heterosis in F1 (HF1) and inbreeding depression in F2 was estimated using the

formulae of Miller and Marani (1963)

HF1 = F1 – MP

Where, MP

F1 = mean value of F1

MP = mid-parent mean value and

Inbreeding depression = F1- F2

Where, F1 F1 = mean value of F1

F2 = mean value of F2

58

The significance of heterosis and inbreeding depression was tested by calculating critical

difference (cd) by the formula,

cd = S.E × t

Where,

S.E. is standard error of difference of varietal means and is equal to (2EMS/r)1/2

EMS is the error mean square

r is number of replications

3.5.5. Correlations

Phenotypic and genotypic correlation coefficients between pairs of plant traits were also

determined using the F2 data. Phenotypic correlation coefficients were calculated following

Dewey and Lu (1959) using Minitab computer programme. The genetic correlations (rg)

between two characters X and Y were calculated by the following formula (Falconer, 1981).

rg = COVg (X, Y)

√Vg (X). Vg (Y)

Where,

COVg (X, Y) = COV (X, Y) F2 – COV (X, Y) E

COV (X, Y) E = (1/4) [COV (X, Y) P1 + COV (X, Y) P2 + 2COV (X, Y) F1]

COVg (X, Y), COV (X, Y) E, COV (X, Y) P1, COV (X, Y) P2, COV (X, Y) F1 and

COV (X, Y) F2 are covariances of X and Y associated with genetic effects,

non-genetic variances of X and Y respectively.

3.5.6. Chi-square analysis

The segregating ratios of plants in F2 and back crosses for the traits, okra leaf type, gossypol

glands and trichomes were tested for their fitness to a theoretical ratio through chi-square test

(Harris, 1912).

59

CHAPTER 4

RESULTS

4.1. Genetic basis of variation for morphological, fibre and insect resistant traits

4.1.1. Analysis of variance for morphological traits

Ordinary analysis of variance was applied separately for each cross to determine the

significance of the generation effects on plant height. There were significant differences

(P< 0.05) among generation means of the six generations of each of the three crosses

(Appendix II). Generation means and variances for plant height for three crosses are shown

in Table 4.1 which reflect significant differences between the parents HRVO-1 and CIM 446,

while non-significant differences were observed between the parents HRVO-1, FH 1000 and

Acala 63-74. A higher level of magnitude of variances was observed in F2 and backcrosses

than that of parental and an F1 generation in all the three crosses. The frequency distributions

for plant height in F2 generations of three crosses are presented in Fig. 4.1.

Analysis of variance for number of monopodial branches was applied for each cross

to determine the significance of the generation effects. There were significant differences

(P< 0.05) among generation means of the six generations of each of the three crosses shown

in Appendix III. Table 4.2 shows the generation means and variances for number of

monopodial branches for three crosses which reflects significant differences between the

parents HRVO-1, CIM 446 and Acala 63-74. Non-significant differences were observed

between the parents HRVO-1 and FH 1000. The higher level of magnitude of variances was

observed in F2 and backcrosses than that of parental and F1 generations in the crosses,

HRVO-1 × FH 1000 and HRVO-1 × CIM 446, while in the cross HRVO-1 × Acala 63-74 the

magnitude of variance in F2 was found higher than parental and F1 generations. The

magnitude of variance for backcrosses for this cross was almost at par with that of parental

and F1 generations. In F2 the frequency distributions for number of monopodial branches in

three crosses are presented in Fig. 4.2.

60

Table 4.1: Generation means and variances for plant height in three single crosses

Generation HRVO-1 × FH-1000

Generation HRVO-1 × CIM-446

Generation HRVO-1 × Acala 63-74

Mean Variance Mean Variance Mean Variance P1 (HRVO-1) 116.06 1.098 P1 (HRVO-1) 115.07 6.064 P1 (HRVO-1) 115.433 1.736

P2 (FH-1000) 117.00 1.586 P2 (CIM-446) 121.60 5.765 P2 (Acala 63-74) 115.833 1.626

F1 117.66 1.436 F1 124.97 5.136 F1 116.900 1.989

F2 114.99 13.939 F2 119.07 41.397 F2 108.540 8.040

BC1 115.7 6.279 BC1 118.93 27.793 BC1 115.080 4.005

BC2 116.77 5.186 BC2 122.58 22.735 BC2 116.350 3.130

LSD (0.05) 1.71 LSD (0.05) 1.12 LSD (0.05) 2.53

Table 4.2: Generation means and variances for number of monopodial branches in three single crosses

Generation HRVO-1 × FH-1000

Generation HRVO-1 × CIM-446

Generation HRVO-1 × Acala 63-74

Mean Variance Mean Variance Mean Variance P1 (HRVO-1) 1.267 0.547 P1 (HRVO-1) 1.466 0.632 P1 (HRVO-1) 1.700 0.480

P2 (FH-1000) 1.000 0.544 P2 (CIM-446) 1.366 0.642 P2 (Acala 63-74) 3.033 0.454

F1 1.330 0.506 F1 1.533 0.651 F1 1.633 0.475

F2 1.640 2.535 F2 1.500 0.977 F2 1.640 0.809

BC1 1.400 0.955 BC1 1.489 0.769 BC1 1.566 0.420

BC2 0.988 1.483 BC2 1.433 0.697 BC2 1.955 0.452

LSD (0.05) 0.32 LSD (0.05) 0.057 LSD (0.05) 0.34

61

Plant height (cm)

Plant height (cm)

Plant height (cm)

62

No. of monopodial branches/plant

No. of monopodial branches/plant

No. of monopodial branches/plant

63

Significance of the generation effects in case of number of sympodial branches was

determined by the ordinary analysis of variance for each cross. The three crosses exhibited

significant differences (P< 0.05) among generation means of the six generations in each of

the three crosses (Appendix IV). Generation means and variances for number of sympodial

branches for three crosses are shown in Table 4.3 which reflected significant differences

between the parents HRVO-1, FH 1000, CIM 446 and Acala 63-74. The F1 mean was

statistically higher than those of the other generations in the crosses HRVO-1 × FH 1000 and

HRVO-1 × CIM 446. The variances in F2 and backcrosses were mostly higher in magnitude

than that of parental and F1 generations in all the three crosses for this trait. The frequency

distributions for number of sympodial branches in F2 generations of three crosses are

presented in Fig. 4.3.

There were significant differences (P< 0.05) among generation means of the six

generations of each of the three crosses for the trait number of bolls (Appendix V).

Significant differences between the parents HRVO-1, FH 1000, CIM 446 and Acala 63-74

were observed (Table 4.4). The F1 mean was higher than that of the mean values of the other

generations in the three crosses. The variances in F2 and backcrosses were higher in

magnitude than that of parental and F1 generations in all the three crosses for this trait. The

frequency distributions for number of bolls in F2 generations of three crosses are presented in

Fig. 4.4.

The frequency distributions for seed cotton yield in F2 generations of three crosses are

presented in Fig. 4.5. Analysis of variance was applied separately for each cross to determine

the significance of the generation effect for seed cotton yield. Significant differences

(P< 0.05) among generation means of the six generations of each of the three crosses were

observed (Appendix VI). Generation means and variances for seed cotton yield for three

crosses are shown in Table 4.5 which reflected significant differences between the parents

HRVO-1, FH 1000, CIM 446, and Acala 63-74. The F1 mean was statistically higher than

that of the mean values of the other generations in all the three crosses. The magnitude of

variances in F2 and backcrosses was higher than parental and F1 generations in all the three

crosses.

64

Table 4.3: Generation means and variances for number of sympodial branches in three single crosses

Generation HRVO-1 × FH-1000

Generation HRVO-1 × CIM-446

Generation HRVO-1 × Acala 63-74

Mean Variance Mean Variance Mean Variance P1 (HRVO-1) 9.966 0.654 P1 (HRVO-1) 10.00 0.827 P1 (HRVO-1) 10.26 1.582

P2 (FH-1000) 17.733 0.616 P2 (CIM-446) 16.13 0.878 P2 (Acala 63-74) 15.60 1.731

F1 19.000 0.633 F1 18.20 0.717 F1 16.20 1.424

F2 15.860 11.020 F2 15.58 13.103 F2 14.65 4.659

BC1 14.580 3.840 BC1 14.43 6.607 BC1 13.26 2.782

BC2 18.500 3.720 BC2 17.34 4.655 BC2 16.19 1.959

LSD (0.05) 0.69 LSD (0.05) 0.25 LSD (0.05) 0.40

Table 4.4: Generation means and variances for number of bolls in three single crosses

Generation HRVO-1 × FH-1000

Generation HRVO-1 × CIM-446

Generation HRVO-1 × Acala 63-74

Mean Variance Mean Variance Mean Variance P1 (HRVO-1) 15.30 0.63 P1 (HRVO-1) 15.50 1.086 P1 (HRVO-1) 15.37 1.309

P2 (FH-1000) 23.36 0.70 P2 (CIM-446) 18.46 1.402 P2 (Acala 63-74) 18.27 1.443

F1 24.03 0.65 F1 22.06 1.754 F1 19.87 1.912

F2 19.10 11.66 F2 18.25 9.921 F2 17.63 7.253

BC1 19.30 5.75 BC1 18.67 6.335 BC1 17.47 4.948

BC2 22.31 4.57 BC2 19.77 4.893 BC2 18.53 2.847

LSD (0.05) 0.89 LSD (0.05) 0.38 LSD (0.05) 0.49

65

No. of sympodial branches/plant

No. of sympodial branches/plant

No. of sympodial branches/plant

66

No. of bolls/plant

No. of bolls/plant

No. of bolls/plant

67

Table 4.5: Generation means and variances for seed cotton yield in cotton single crosses

Generation HRVO-1 × FH-1000

Generation HRVO-1 × CIM-446

Generation HRVO-1 × Acala 63-74

Mean Variance Mean Variance Mean Variance P1 (HRVO-1) 54.04 0.439 P1 (HRVO-1) 52.87 1.676 P1 (HRVO-1) 52.91 4.377

P2 (FH-1000) 101.04 0.442 P2 (CIM-446) 78.02 1.596 P2 (Acala 63-74) 87.13 3.904

F1 111.96 0.458 F1 96.21 1.753 F1 92.00 4.777

F2 83.23 78.889 F2 81.00 8.176 F2 73.45 76.359

BC1 90.13 25.00 BC1 74.50 5.490 BC1 68.26 38.367

BC2 104.49 27.199 BC2 86.80 4.015 BC2 86.49 50.246

LSD (0.05) 0.94 LSD (0.05) 0.60 LSD (0.05) 2.84

Table 4.6: Generation means and variances for boll weight in three single crosses

Generation HRVO-1 × FH-1000

Generation HRVO-1 × CIM-446

Generation HRVO-1 × Acala 63-74

Mean Variance Mean Variance Mean Variance P1 (HRVO-1) 3.46 0.086 P1 (HRVO-1) 3.45 0.015 P1 (HRVO-1) 3.41 0.020

P2 (FH-1000) 4.37 0.115 P2 (CIM-446) 4.22 0.014 P2 (Acala 63-74) 4.77 0.021

F1 4.59 0.100 F1 4.40 0.018 F1 4.63 0.021

F2 4.04 0.133 F2 4.02 1.211 F2 4.17 0.192

BC1 3.99 0.105 BC1 3.95 0.441 BC1 3.92 0.087

BC2 4.35 0.097 BC2 4.27 0.629 BC2 4.66 0.077

LSD (0.05) 0.099 LSD (0.05) 0.06 LSD (0.05) 0.08

68

Seed cotton yield (g)

Seed cotton yield (g)

Seed cotton yield (g)

69

Boll weight (g)

Boll weight (g) Boll weight (g)

Boll weight (g) Boll weight (g)

70

The significance of generation effects for boll weight in each of the three crosses was

determined by the ordinary analysis of variance (Appendix VII). There were significant

differences (P< 0.05) among generation means of the six generations of each of the three

crosses. Generation means and variances for boll weight for three crosses are shown in

Table 4.6 which reflected significant differences between the parents HRVO-1, FH 1000,

CIM 446 and Acala 63-74. The F1 mean was statistically higher than that of the mean values

of the other generations in the crosses HRVO-1 × FH 1000 and HRVO-1 × CIM 446. The

Variances in F2 and backcrosses were greater in magnitude than that of the parental and F1

generations in the three crosses. The frequency distributions for boll weight in F2 generations

of three crosses are presented in Fig. 4.6.

4.1.2. Analysis of variance for fibre related traits

Significant differences (P< 0.05) among generation means of the six generations in each of

the three crosses for lint percentage were recorded (Appendix VIII). Generation means and

variances for lint percentage for three crosses are shown in Table 4.7. There existed

significant differences between the parents HRVO-1, FH 1000, CIM 446 and Acala 63-74.

The F1 mean was statistically higher than the mean values of the other generations in the

crosses HRVO-1 × FH 1000 and HRVO-1 × CIM 446. The backcrosses and F2 secured

higher magnitude of variances than in the parental and F1 generations in the three crosses.

The variance of BC1 in the cross HRVO-1 × FH 1000 was found higher in magnitude than

that of the variance of F2. The frequency distributions for lint percentage in F2 generations of

three crosses are presented in Fig. 4.7.

In case of fibre length and fibre strength a separate analysis of variance was applied

on the six generations of each cross for this trait to determine the significance of the

generation effects. Among the generation means of the six generations of each of the three

crosses (Appendices IX and X), significant differences (P< 0.05) were found. From the

Tables 4.8 and 4.9, the generation means and variances for fibre length and fibre strength in

two crosses reflected significant differences between the parents HRVO-1, FH 1000 and

CIM 446. While in HRVO-1 and Acala 63-74 the differences for these two traits was

71

Table 4.7: Generation means and variances for lint percentage in three single crosses

Generation HRVO-1 × FH-1000

Generation HRVO-1 × CIM-446

Generation HRVO-1 × Acala 63-74

Mean Variance Mean Variance Mean Variance P1 (HRVO-1) 33.09 4.38 P1 (HRVO-1) 32.95 1.14 P1 (HRVO-1) 32.89 1.41

P2 (FH-1000) 38.65 3.68 P2 (CIM-446) 35.95 1.14 P2 (Acala 63-74) 36.52 1.24

F1 40.94 2.46 F1 36.40 1.43 F1 35.23 1.36

F2 37.77 5.46 F2 34.47 5.36 F2 34.98 2.97

BC1 37.00 6.93 BC1 34.12 2.61 BC1 34.07 3.19

BC2 38.24 2.68 BC2 36.05 2.15 BC2 35.53 1.94

LSD (0.05) 0.98 LSD (0.05) 0.14 LSD (0.05) 0.53

Table 4.8: Generation means and variances for fibre length in three single crosses

Generation HRVO-1 × FH-1000

Generation HRVO-1 × CIM-446

Generation HRVO-1 × Acala 63-74

Mean Variance Mean Variance Mean Variance P1 (HRVO-1) 23.34 1.065 P1 (HRVO-1) 23.33 0.136 P1 (HRVO-1) 22.99 0.368

P2 (FH-1000) 25.92 1.039 P2 (CIM-446) 26.99 0.141 P2 (Acala 63-74) 23.10 0.448

F1 25.82 1.087 F1 26.65 0.138 F1 23.40 0.448

F2 25.26 2.387 F2 25.10 4.512 F2 22.96 2.295

BC1 24.84 2.749 BC1 24.85 2.141 BC1 23.07 1.129

BC2 25.61 1.892 BC2 26.74 2.095 BC2 23.34 1.149

LSD (0.05) 0.18 LSD (0.05) 0.23 LSD (0.05) 0.19

72

Lint (% age)

Lint (% age)

Lint (% age)

73

Fibre length (mm)

Fibre length (mm)

Fibre length (mm)

74

non-significant. The magnitude of variances observed in F2 and backcrosses was higher than

in the parental and F1 generations in the three crosses for fibre length and fibre strength. For

fibre length, the variance of BC1 was the highest in the cross HRVO-1 × FH 1000 among the

variances recorded for other five generations of the same cross. The frequency distributions

for fibre length and fibre strength in F2 generations of three crosses are presented in Fig. 4.8

and 4.9.

Analysis of variance for the trait fibre elongation for the six generations of three

crosses showed significant (P< 0.05) differences (Appendix XI). Generation means and

variances for fibre elongation in three crosses reflect significant differences between the

parents HRVO-1, FH 1000, CIM 446 and Acala 63-74, and are shown in Table 4.10. The F1

mean was statistically higher than the mean values of the other generations of the crosses

HRVO-1 × FH 1000 and HRVO-1 × CIM 446. A higher magnitude of variances was

observed in the F2 and backcrosses than in parental or F1 generations in all the three crosses.

The frequency distributions for fibre elongation in F2 generations of three crosses are

presented in Fig. 4.10.

Significant differences (P< 0.05) among the six generations were found in all the three

crosses for fibre uniformity ratio (Appendix XII). Generation means and variances for fibre

uniformity ratio in three crosses are shown in Table 4.11. Significant differences were

observed between the parents HRVO-1, FH 1000, CIM 446 and Acala 63-74. In F1 of three

crosses, the mean of F1 of the cross HRVO-1 × Acala 63-74 was significantly higher than the

mean values of the other generations. While in the other two crosses it remained lesser than

the P2 but higher than the common parent HRVO-1. F2 and backcrosses secured a higher

magnitude of variances than the parental and F1 generations in all the three crosses. The

frequency distributions for fibre uniformity ratio in F2 generations of the three crosses are

presented in Fig. 4.11.

75

Table 4.9: Generation means and variances for fibre strength in three single crosses

Generation HRVO-1 × FH-1000

Generation HRVO-1 × CIM-446

Generation HRVO-1 × Acala 63-74

Mean Variance Mean Variance Mean Variance P1 (HRVO-1) 22.70 0.562 P1 (HRVO-1) 22.99 0.544 P1 (HRVO-1) 22.85 0.385

P2 (FH-1000) 25.00 0.592 P2 (CIM-446) 25.96 0.560 P2 (Acala 63-74) 23.08 0.404

F1 24.80 0.582 F1 25.64 0.550 F1 23.62 0.398

F2 24.12 1.636 F2 24.68 2.068 F2 23.31 2.236

BC1 23.92 1.420 BC1 24.60 1.613 BC1 23.26 1.502

BC2 24.95 1.437 BC2 24.27 1.164 BC2 23.35 1.120

LSD (0.05) 0.46 LSD (0.05) 0.67 LSD (0.05) 0.27

Table 4.10: Generation means and variances for fibre elongation in three single crosses

Generation HRVO-1 × FH-1000

Generation HRVO-1 × CIM-446

Generation HRVO-1 × Acala 63-74

Mean Variance Mean Variance Mean Variance P1 (HRVO-1) 5.22 0.005 P1 (HRVO-1) 5.31 0.0308 P1 (HRVO-1) 5.36 0.027

P2 (FH-1000) 5.54 0.006 P2 (CIM-446) 6.00 0.030 P2 (Acala 63-74) 4.48 0.028

F1 5.67 0.005 F1 6.23 0.032 F1 4.84 0.022

F2 5.49 0.164 F2 5.91 0.125 F2 4.75 0.184

BC1 5.50 0.048 BC1 5.66 0.087 BC1 4.99 0.073

BC2 5.58 0.064 BC2 6.11 0.060 BC2 4.64 0.062

LSD (0.05) 0.057 LSD (0.05) 0.08 LSD (0.05) 0.06

76

Fibre strength (g/tex)

Fibre strength (g/tex)

Fibre strength (g/tex)

77

Fibre elongation (%)

Fibre elongation (%)

Fibre elongation (%)

78

For fibre fineness analysis of variance was performed separately for the six generations of the

three crosses (Appendix XIII). There were significant differences (P< 0.05) among the six

generations. Generation means and variances for fibre fineness in three crosses are shown in

Table 4.12, which reflected significant differences between the parents HRVO-1, FH 1000,

CIM 446 and Acala 63-74 in the three crosses. In cross HRVO-1 × FH 1000, F1 mean values

were lower than in both the parents. A higher magnitude of variances than in parental and F1

generations of all the three crosses was observed in the F2 and backcrosses. The frequency

distributions for fibre fineness in F2 generations of three crosses are presented in Fig. 4.12.

4.1.3. Analysis of variance for insect related traits

In case of number of trichomes, the analysis of variance was performed separately for the six

generations of the three crosses (Appendix XIV). Significant differences (P< 0.05) among the

P1, P2, F1, F2, BC1 and BC2 generations were found of all the three crosses. Generation means

and variances for number of trichomes in three crosses are shown in Table 4.13. Significant

differences of the means for number of trichomes were found between the parents HRVO-1,

FH 1000, CIM 446 and Acala 63-74. The F1 mean in three crosses was higher than the mean

values of the other generations. A higher magnitude of variances in F2 and backcrosses of all

the three crosses was observed as compared to parental and F1 generations. The variances in

for almost all the three crosses were higher than their respective backcrosses. The frequency

distributions for number of trichomes in F2 generations of three crosses are presented in Fig.

4.13.

For gossypol content two crosses HRVO-1 × Acala 63-74 and HRVO-1 × HG-142

were studied. Significant differences (P< 0.05) among the P1, P2, F1, F2, BC1 and BC2

generations were found in the two crosses (Appendix XV). Generation means and variances

for gossypol content in two crosses are shown in Table 4.14. Table 4.14 showing generation

means for gossypol content revealed significant differences between the parental means of

HRVO-1, Acala 63-74 and HG-142. In the mean comparison of F1 in these two crosses, the

mean of F1 of the cross HRVO-1 × HG- 142 was closer to the mid parent value. In the

segregating generations of F2 and backcrosses a higher magnitude of variances were

observed than in the parental and F1 of these two crosses. The variances in F2 of the two

79

Table 4.11: Generation means and variances for fibre uniformity ratio in three single crosses

Generation HRVO-1 × FH-1000

Generation HRVO-1 × CIM-446

Generation HRVO-1 × Acala 63-74

Mean Variance Mean Variance Mean Variance P1 (HRVO-1) 42.13 0.156 P1 (HRVO-1) 42.78 0.232 P1 (HRVO-1) 42.54 0.226

P2 (FH-1000) 52.46 0.149 P2 (CIM-446) 45.22 0.212 P2 (Acala 63-74) 42.11 0.237

F1 44.05 0.159 F1 44.97 0.243 F1 43.10 0.226

F2 44.80 16.163 F2 44.12 2.205 F2 41.97 1.350

BC1 43.23 6.364 BC1 43.65 1.477 BC1 42.62 0.622

BC2 46.52 7.47 BC2 44.50 2.045 BC2 42.28 0.488

LSD (0.05) 0.37 LSD (0.05) 0.37 LSD (0.05) 0.22

Table 4.12: Generation means and variances for fibre fineness in three single crosses

Generation HRVO-1 × FH-1000

Generation HRVO-1 × CIM-446

Generation HRVO-1 × Acala 63-74

Mean Variance Mean Variance Mean Variance P1 (HRVO-1) 5.08 0.046 P1 (HRVO-1) 5.21 0.183 P1 (HRVO-1) 5.30 0.014

P2 (FH-1000) 4.67 0.048 P2 (CIM-446) 4.49 0.188 P2 (Acala 63-74) 4.70 0.014

F1 4.49 0.049 F1 4.30 0.189 F1 4.77 0.014

F2 4.39 0.189 F2 4.61 0.445 F2 4.70 0.147

BC1 4.65 0.164 BC1 4.72 0.345 BC1 4.92 0.071

BC2 4.45 0.148 BC2 4.29 0.235 BC2 4.65 0.052

LSD (0.05) 0.09 LSD (0.05) 0.06 LSD (0.05) 0.06

80

Fibre uniformity ratio (%)

Fibre uniformity ratio (%)

Fibre uniformity ratio (%)

81

Fibre fineness (Micronaire)

Fibre fineness (Micronaire)

Fibre fineness (Micronaire)

82

Table 4.13: Generation means and variances for number of trichomes in three single crosses

Generation HRVO-1 × FH-1000

Generation HRVO-1 × CIM-446

Generation HRVO-1 × Acala 63-74

Mean Variance Mean Variance Mean Variance P1 (HRVO-1) 240.20 12.84 P1 (HRVO-1) 240.20 4.84 P1 (HRVO-1) 239.80 4.16

P2 (FH-1000) 40.76 10.09 P2 (CIM-446) 31.73 4.43 P2 (Acala 63-74) 43.76 4.51

F1 103.70 14.58 F1 60.73 4.67 F1 198.40 4.23

F2 122.73 5218.67 F2 102.52 7108.62 F2 173.20 633.71

BC1 158.79 4259.46 BC1 169.57 2907.23 BC1 217.93 384.37

BC2 77.00 4140.61 BC2 47.64 2617.77 BC2 108.16 413.62

LSD (0.05) 5.06 LSD (0.05) 2.09 LSD (0.05) 13.13

83

No. of trichomes

No. of trichomes

No. of trichomes

84

crosses were found higher than the variances of their respective backcrosses. Frequency

distributions for gossypol content in F2 generations of two crosses are presented in Fig. 4.14.

Similarly for total gossypol the same two crosses were studied as were studied for the

trait gossypol content. The analysis of variance for total gossypol in the six generations of the

two crosses of HRVO-1 × Acala 63-74 and HRVO-1 × HG- 142 is given in Appendix XVI.

Significant differences (P< 0.05) among the P1, P2, F1, F2, BC1 and BC2 generations were

found in the two crosses. Generation means and variances for total gossypol in two crosses

are shown in Table 4.15, which revealed significant differences between the parents

HRVO-1, Acala 63-74 and HG-142. In the mean comparison of F1 of the two crosses, the

mean of F1 in the cross HRVO-1 × HG-142 was almost equal to the mid parent value. A

higher magnitude of variances in the segregating generations of F2 and backcrosses was

observed than that of non-segregating generations of the parental and F1 in both the crosses.

The variances of F2 of the two crosses, as shown in the Table 4.15, were almost higher than

their respective backcross variances. Frequency distributions for total gossypol in F2

generations of the two crosses are presented in Fig. 4.15.

4.2. Generation mean analysis for various plant traits

The estimates were made using the cotton plant data on various traits, in order to see the best

fitness of the genetic model for components of generation means for the traits. The estimates

of the best fit model for generation mean parameters, mean (m), additive [d], dominance [h],

additive × additive [i], additive × dominance [j] and dominance × dominance [l] for various

plant traits in the three crosses along with the chi- squared (χ2) are given in Table 4.16.

As is evident from the Table 4.16 the estimates recorded for plant height revealed that

four-parameter model (m, d, h and i) was adequate for the crosses HRVO-1 × FH 1000 and

HRVO-1 × CIM 446 whereas, in the cross HRVO-1 × Acala 63-74 four-parameter model

(m, h, i and l) was found adequate. From the estimates for number of monopodial branches

none of the genetic effects and interaction component appeared to be involved in the

expression of the trait in the cross HRVO-1 × CIM 446. In the other two crosses i.e;

85

Table 4.14: Generation means and variances for gossypol content in two single crosses

Generation HRVO-1 × HG-142

Generation HRVO-1 × Acala 63-74

Mean Variance Mean Variance P1 (HRVO-1) 0.60 0.0045 P1 (HRVO-1) 0.59 0.0010

P2 (HG-142) 1.14 0.0054 P2 (Acala 63-74) 0.040 0.0015

F1 0.88 0.0048 F1 0.140 0.0014

F2 0.88 0.0405 F2 0.200 0.0447

BC1 0.74 0.0201 BC1 0.373 0.0229

BC2 1.03 0.0179 BC2 0.11 0.0127

LSD (0.05) 0.018 LSD (0.05) 0.057

Table 4.15: Generation means and variances for total gossypol % in two single crosses

Generation HRVO-1 × HG-142

Generation HRVO-1 × Acala 63-74

Mean Variance Mean Variance P1 (HRVO-1) 0.240 0.0007 P1 (HRVO-1) 0.233 0.0017

P2 (HG-142) 0.455 0.0008 P2 (Acala 63-74) 0.020 0.0015

F1 0.351 0.0008 F1 0.050 0.0011

F2 0.351 0.0065 F2 0.081 0.0059

BC1 0.295 0.0032 BC1 0.149 0.0038

BC2 0.411 0.0029 BC2 0.041 0.0030

LSD (0.05) 0.018 LSD (0.05) 0.018

86

Gossypol content (mg/1g)

Gossypol content (mg/1g)

87

Total gossypol (%)

Total gossypol (%)

88

Table 4.16: Components of generation means parameters, mean (m), additive [d], dominance [h], additive × additive [i], additive × dominance

[ j ] and dominance × dominance [l] for various plant traits in different crosses. Traits Crosses m [d] [h] [i] [j] [l] χχχχ2 (d.f) Prob.

Plant height HRVO-1 × FH 1000

HRVO-1 × CIM 446

HRVO-1 × Acala 63-74

112.61 ± 0.55

113.62 ± 1.02

86.96 ± 1.09

0.56 ± 0.13

3.33 ± 0.28

-

5.08 ± 0.69

11.42 ± 1.27

56.38 ± 2.56

3.95 ± 0.58

4.75 ± 1.08

28.67 ± 1.08

-

-

-

-

-

-26.45 ±1.56

3.22

0.989

1.42

2

2

1

0.25-0.10

0.75-0.50

0.25-0.10

Number of monopodial branches/ plant

HRVO-1 × FH 1000

HRVO-1 × CIM 446

HRVO-1 × Acala 63-74

1.38 ± 0.071

1.47 ± 0.043

1.60 ± 0.047

0.19 ± 0.082

-

0.54 ± 0.065

-

-

-

-0.29 ± 0.12

-

0.74 ± 0.10

-

-

-

-

-

-

7.74

1.043

5.07

3

5

3

0.10-0.05

0.97-0.95

0.25-0.10

Number of sympodial branches/ plant

HRVO-1 × FH 1000

HRVO-1 × CIM 446

HRVO-1 × Acala 63-74

13.85 ± 0.09

13.10 ± 0.11

13.00 ± 0.14

3.88 ± 0.96

3.05 ± 0.11

2.77 ± 0.13

5.14 ± 0.17

5.15 ± 0.19

3.32 ± 0.26

-

-

-

-

-

-

-

-

-

4.96

1.77

1.92

3

3

3

0.25-0.10

0.75-0.50

0.75-0.50

Number of bolls/plant

HRVO-1 × FH 1000

HRVO-1 × CIM 446

HRVO-1 × Acala 63-74

14.39 ± 0.47

14.75 ± 0.52

15.38 ± 0.48

4.03 ± 0.10

1.43 ± 0.12

1.36 ± 0.13

9.93 ± 0.55

7.39 ± 0.69

4.47 ± 0.66

4.95 ± 0.49

2.26 ± 0.54

1.42 ± 0.51

-1.95 ±0.70

-

-

-

-

-

2.40

2.89

1.44

1

2

2

0.25-0.10

0.25-0.10

0.50-0.25

Seed cotton yield

HRVO-1 × FH 1000

HRVO-1 × CIM 446

HRVO-1 × Acala 63-74

68.49 ± 1.06

65.43 ± 0.15

55.0 ± 1.27

23.50 ± 0.087

12.50 ± 0.15

17.20 ± 0.25

43.46 ± 1.09

30.74 ± 0.28

36.81 ± 1.45

9.03 ± 1.07

-

14.81 ± 1.28

-

-

-

-

-

-

0.26

3.31

1.65

2

3

2

0.75-0.50

0.50-0.25

0.50-0.25

Boll weight HRVO-1 × FH 1000

HRVO-1 × CIM 446

HRVO-1 × Acala 63-74

3.49 ± 0.08

3.83 ± 0.015

3.75 ± 0.066

0.44 ± 0.03

0.38 ± 0.015

0.69 ± 0.017

1.08 ± 0.122

0.56 ± 0.026

0.88 ± 0.083

0.42 ± 0.09

-

0.34 ± 0.07

-

-

-

-

-

-

0.36

1.50

2.85

2

3

2

0.90-0.75

0.75-0.50

0.25-0.10

Lint percentage

HRVO-1 × FH 1000

HRVO-1 × CIM 446

HRVO-1 × Acala 63-74

36.34 ± 0.15

32.63 ± 0.41

34.86 ± 0.07

2.74 ± 0.25

1.59 ± 0.12

1.69 ± 0.12

-

3.82 ± 0.57

-

-

1.83 ± 0.44

-

-2.86 ±0.79

-

-

4.78 ± 0.36

-

-

5.57

3.06

6.87

2

2

4

0.10-0.05

0.25-0.10

0.25-0.10

89

Table 4.16. Continued

TRAITS CROSSES m [d] [h] [i] [j] [l] χχχχ2 (d.f) Prob.

Fibre length HRVO-1 × FH 1000

HRVO-1 × CIM 446

HRVO-1 × Acala 63-74

24.63 ± 0.11

23.95 ± 0.28

23.02 ± 0.046

1.16 ± 011

1.83 ± 0.046

-

1.18 ± 0.22

2.70 ± 0.30

0.34 ± 0.08

-

1.21 ± 0.28

-

-

-

-

-

-

-

3.98

4.58

8.18

2

2

4

0.25-0.10

0.25-0.10

0.10-0.05

Fibre strength HRVO-1 × FH 1000

HRVO-1 × CIM 446

HRVO-1 × Acala 63-74

23.85 ± 0.09

24.40 ± 0.086

22.96 ± 0.073

1.12 ± 0.092

1.47 ± 0.083

0.66 ± 0.14

0.933 ± 0.17

1.10 ± 0.16

-

-

-

-

-

-

-

-

-

-

5.95

7.22

6.21

3

3

4

0.25-0.10

0.10-0.05

0.25-0.10

Fibre elongation

HRVO-1 × FH 1000

HRVO-1 × CIM 446

HRVO-1 × Acala 63-74

5.38 ± 0.009

5.64 ± 0.020

4.66 ± 0.064

0.15 ± 0.009

0.36 ± 0.019

0.42 ± 0.018

0.29 ± 0.016

0.54 ± 0.038

0.17 ± 0.082

-

-

0.25 ± 0.07

-

-

-

-

-

-

5.63

7.27

3.69

3

3

2

0.25-0.10

0.10-0.05

0.25-0.10

Fibre uniformity ratio

HRVO-1 × FH 1000

HRVO-1 × CIM 446

HRVO-1 × Acala 63-74

44.07 ± 0.069

42.83 ± 0.32

40.98 ± 0.18

5.16 ± 0.051

1.18 ± 0.058

0.30 ± 0.054

-

2.13 ± 0.37

2.15 ± 0.25

3.22 ± 0.086

1.16 ± 0.33

1.31 ± 0.20

-3.76 ± 0.78

-

-

-

-

-

3.49

4.86

2.42

2

2

2

0.25-0.10

0.10-0.05

0.50-0.25

Fibre fineness HRVO-1 × FH 1000

HRVO-1 × CIM 446

HRVO-1 × Acala 63-74

4.28 ± 0.078

4.83 ± 0.048

4.62 ± 0.056

0.20 ± 0.025

0.38 ± 0.045

0.29 ± 0.014

0.21 ± 0.10

-0.56 ± 0.09

0.15 ± 0.07

0.59 ± 0.084

-

0.38 ± 0.06

-

-

-

-

-

-

0.27

3.54

0.61

2

3

2

0.90-0.75

0.50-0.25

0.75-0.50

Number of trichomes

HRVO-1 × FH 1000

HRVO-1 × CIM 446

HRVO-1 × Acala 63-74

140.43 ± 0.61

135.96 ± 0.27

141.81 ± 0.37

99.70 ± 0.61

104.22 ± 0.27

97.99 ± 0.37

-36.78 ± 1.15

-75.22 ± 0.48

56.62 ± 0.65

-

-

-

-

-

-

-

-

-

0.33

0.81

1.61

3

3

3

0.97-0.95

0.90-0.75

0.75-0.50

Gossypol content (mg)

HRVO-1 × Acala 63-74

HRVO-1 × HG-142

0.32 ± 0.006

0.88 ± 0.006

0.27 ± 0.006

0.28 ± 0.010

-0.17 ± 0.01

-

-

-

-

-

-

-

3.32

1.50

3

4

0.50-0.25

0.90-0.75

Total gossypol (%)

HRVO-1 × Acala 63-74

HRVO-1 × HG-142

0.13 ± 0.006

0.35 ± 0.002

0.11 ± 0.005

0.11 ± 0.004

-0.07 ± 0.011

-

-

-

-

-

-

-

3.84

1.06

3

4

0.50-0.25

0.95-0.90

90

HRVO-1 × FH 1000 and HRVO-1 × Acala 63-74 a similar trend of adequacy in the form of

three-parameter model (m, d and i) was observed for the number of monopodial branches. A

three-parameter model (m, d and h) was best fit from the observed to expect estimated

generation means for number of sympodial branches in all the three crosses under study.

In case of number of bolls, four-parameter model involving m, d, h and i was

adequate for the crosses, HRVO-1 × CIM 446 and HRVO-1 × Acala 63-74. A five-parameter

model m, d, h, i and j showed the best fitness of the observed to the expected generation

means for number of bolls in the cross of HRVO-1 and FH 1000. A similar pattern of

adequacy of four-parameter model m, d, h and i for the trait of seed cotton yield was

observed in the crosses HRVO-1 × FH 1000 and HRVO-1 × Acala 63-74. However, a simple

additive-dominance model (three-parameter model m, d and h) was best fit from the observed

to the expected estimated generation means for seed cotton yield in the cross of HRVO-1 and

CIM 446. Four- parameter model (m, d, h and i) showed best fitness of the observed to the

expected generation means for boll weight in the crosses HRVO-1 × FH 1000 and HRVO-1

× Acala 63-74 whereas, m, d and h showed its fitness for the same trait in the cross HRVO-1

and CIM 446.

In case of lint percentage, four-parameter model (m, d, j and l) was fit in the cross

HRVO-1 × FH 1000 whereas, m, d, h and i was found fit in the cross HRVO-1 × CIM 446.

However, in the cross HRVO-1 × Acala 63-74 the simplest model of two parameters i.e; m

and d was found to be operative. The Chi- square analysis in case of fibre length indicated a

three-parameter model (m, d and h) to be fit in the cross HRVO-1 × FH 1000, a four-

parameter model (m, d, h and i) in the cross HRVO-1 × CIM 446 and two-parameter model

i.e; m and h in case of HRVO-1 × Acala 63-74. Similarly, for fibre strength the same three-

parameter model (m, d and h) was operative in the crosses HRVO-1 × FH 1000 and

HRVO-1 × CIM 446 whereas, in case of HRVO-1 × Acala 63-74, two parameter model

i.e; m and d observed to be fit.

In fibre elongation, three-parameter model (m, d and h) was found adequate in the

crosses HRVO-1 × FH 1000 and HRVO-1 × CIM 446. In the cross HRVO-1 × Acala 63-74,

91

four-parameter model (m, d, h and i) showed its adequacy from the observed to the expected

generation means in the cross HRVO-1 × Acala 63-74. In case of fibre uniformity ratio, a

four-parameter model (m, d, h and i) was adequate in the crosses HRVO-1 × CIM 446 and

HRVO-1 × Acala 63-74 whereas, a four-parameter model (m, d, i and j) showed its adequacy

in the cross HRVO-1 × FH 1000. The four-parameter model (m, d, h and i) for fibre fineness

was found adequate in the crosses HRVO-1 × FH 1000 and HRVO-1 × Acala 63-74 while a

three-parameter model (m, d and h) proved to be best fit in the cross HRVO-1 × CIM 446.

For the number of trichomes, a similar pattern of genetic effects was observed in all

the three crosses under study. The three-parameter model (m, d and h) observed to be best fit

in terms of the observed to the expected generation means, showing adequacy for this model

in all the three crosses. The best fit model for gossypol content observed in the cross

HRVO-1 × Acala 63-74 was three-parameter model with m, d and h effects whereas, in

HRVO-1 × HG-142 a simplest model comprising of two-parameters (m and d) showed its

adequacy for the observed to the expected mean values. In case of total gossypol in the

crosses i.e; HRVO-1 × Acala 63-74 and HRVO-1 × HG-142 a similar trend of genetic effects

was followed as were observed for the gossypol content in these two crosses.

4.3. Generation variance analysis for various plant traits

The estimates of the components of genetic variances and Chi-square (χ2) values for various

plant traits are consolidated in Table 4.17. In the generation variance analyses, a model

incorporating the additive (D) and environmental (E) components was sufficient to explain

the variation in cotton crosses for all the traits except for lint percentage in cross, HRVO-1 ×

FH 1000, where the additive (D), additive × dominance (F) and environmental (E) model

appeared to show its best fitness. The Chi- squared (χ2) values for the estimated genetic

components were detected as non-significant for all the traits in the crosses.

4.4. Inheritance studies pertaining to insect resistant traits

The inheritance studies were conducted for three important insect resistant traits, namely,

okra leaf type, trichomes and gossypol glands in Gossypium hirsutum L, on the basis of the

F2 data and test/back cross populations. Chi-square test was employed to test the differences

of the observed vs the expected segregating phenotypic ratios.

92

Table 4.17: Variance components, D, H, E following weighted analysis of components of variance for various traits Traits Crosses D H F E χχχχ2 (d.f) H2

n.s

∞ G.A

Plant height HRVO-1 × FH 1000

HRVO-1 × CIM 446

HRVO-1 × Acala 63-74

21.45 ± 2.06

74.69 ± 7.57

10.70 ± 1.37

-

-

-

-

-

-

1.33 ± 0.19

5.70 ± 0.84

1.58 ± 0.23

5.37

1.40

5.43

4

4

4

0.77

0.90

0.67

0.94 5.92

0.93 11.96

0.87 3.89

Number of

monopodial

branches/plant

HRVO-1 × FH 1000

HRVO-1 × CIM 446

HRVO-1 × Acala 63-74

3.51 ± 0.44

0.67 ± 0.28

0.58 ± 0.19

-

-

-

-

-

-

0.51 ± 0.07

0.61 ± 0.08

0.40 ± 0.05

5.79

0.81

7.08

4

4

4

0.69

0.34

0.36

0.87 2.27

0.53 0.70

0.59 0.66

Number of

sympodial

branches/plant

HRVO-1 × FH 1000

HRVO-1 × CIM 446

HRVO-1 × Acala 63-74

16.62 ± 1.46

21.90 ± 1.92

5.30 ± 0.94

-

-

-

-

-

-

0.62 ± 0.09

0.79 ± 0.11

1.46 ± 0.21

8.08

4.44

5.90

4

4

4

0.75

0.84

0.57

0.96 5.16

0.96 6.23

0.78 2.53

Number of

bolls/plant

HRVO-1 × FH 1000

HRVO-1 × CIM 446

HRVO-1 × Acala 63-74

19.96 ± 1.73

16.93 ± 1.76

10.59 ± 1.33

-

-

-

-

-

-

0.65 ± 0.09

1.41 ± 0.20

1.52 ± 0.22

2.42

3.18

7.73

4

4

4

0.86

0.85

0.73

0.97 6.02

0.92 5.54

0.87 4.05

Seed cotton

yield

HRVO-1 × FH 1000

HRVO-1 × CIM 446

HRVO-1 × Acala 63-74

127.47 ± 10.03

12.72 ± 1.54

131.65 ± 11.44

-

-

-

-

-

-

0.44 ± 0.06

1.66 ± 0.24

4.31 ± 0.64

7.38

2.21

2.52

4

4

4

0.81

0.78

0.86

0.99 14.78

0.88 4.58

0.97 15.52

Boll weight HRVO-1 × FH 1000

HRVO-1 × CIM 446

HRVO-1 × Acala 63-74

0.064 ± 0.04

2.22 ± 0.17

0.30 ± 0.029

-

-

-

-

-

-

0.093 ± 0.01

0.015 ± 0.002

0.020 ± 0.003

1.86

3.75

3.18

4

4

4

0.24

0.92

0.78

0.40 0.18

0.99 2.08

0.93 0.71

Lint percentage HRVO-1 × FH 1000

HRVO-1 × CIM 446

HRVO-1 × Acala 63-74

4.0 ± 1.67

7.46 ± 0.98

3.65 ± 0.81

-

-

-

3.96 ± 0.87

-

3.61 ± 0.49

1.19 ± 0.17

1.42 ± 0.20

2.40

2.29

8.07

3

4

4

0.37

0.70

0.61

0.52 1.76

0.86 3.32

0.72 2.18

Fibre length HRVO-1 × FH 1000

HRVO-1 × CIM 446

3.20 ± 0.68

7.97 ± 0.65

-

-

-

-

1.18 ± 0.17

0.14 ± 0.02

8.78

0.027

4

4

0.67

0.88

0.73 2.14

0.98 3.87

HRVO-1 × Acala 63-74 4.13 ± 0.36 - - 0.14 ± 0.02 0.24 4 0.90 0.97 2.81

93

Table 4.17: Continued

TRAITS CROSSES D H F E χχχχ2 (d.f) h2 n.s ∞ G.A

Fibre strength HRVO-1 × FH 1000

HRVO-1 × CIM 446

HRVO-1 × Acala 63-74

2.50 ± 0.42

3.15 ± 0.43

3.67 ± 0.41

-

-

-

-

-

-

0.62 ± 0.09

0.56 ± 0.08

0.39 ± 0.06

3.32

2.68

1.92

4

4

4

0.76

0.76

0.82

0.80 2.01

0.85 2.26

0.90 2.53

Fibre

elongation

HRVO-1 × FH 1000

HRVO-1 × CIM 446

HRVO-1 × Acala 63-74

0.26 ± 0.021

0.18 ± 0.024

0.25 ± 0.027

-

-

-

-

-

-

0.005 ± 0.0007

0.030 ± 0.004

0.024 ± 0.003

8.47

3.05

9.34

4

4

4

0.79

0.72

0.68

0.98 0.66

0.86 0.52

0.91 0.60

Fibre

uniformity ratio

HRVO-1 × FH 1000

HRVO-1 × CIM 446

HRVO-1 × Acala 63-74

29.34 ± 2.32

7.02 ± 0.60

1.85 ± 0.22

-

-

-

-

-

-

0.15 ± 0.023

0.22 ± 0.03

0.22 ± 0.32

1.65

4.29

6.57

4

4

4

0.91

0.84

0.69

0.99 7.52

0.95 3.53

0.89 1.64

Fibre fineness HRVO-1 × FH 1000

HRVO-1 × CIM 446

HRVO-1 × Acala 63-74

0.34 ± 0.04

0.49 ± 0.10

0.23 ± 0.02

-

-

-

-

-

-

0.05 ± 0.007

0.18 ± 0.025

0.014 ± 0.002

3.93

3.31

4.71

4

4

4

0.90

0.56

0.78

0.87 0.81

0.73 0.76

0.94 0.62

Number of

trichomes

HRVO-1 × FH 1000

HRVO-1 × CIM 446

HRVO-1 × Acala 63-74

9507.85 ±743.23

12476.466 ± 972.4

1706.78 ± 133.93

-

-

-

-

-

-

12.49 ± 2.63

4.64 ± 0.97

4.29 ± 0.90

1.78

2.87

1.22

4

4

4

0.91

0.88

0.91

0.99 135.56

0.99 152.42

0.99 57.53

Gossypol

content (mg)

HRVO-1 × Acala 63-74

HRVO-1 × HG-142

0.075 ± 0.006

0.064 ± 0.006

-

-

-

-

0.00 0.001 ± 0.0002

0.0047 ± 0.0009

9.14

1.79

4

4

0.83

0.79

0.99 0.36

0.93 0.33

Total gossypol

(%)

HRVO-1 × Acala 63-74

HRVO-1 × HG-142

0.0087 ± 0.001

0.010 ± 0.001

-

-

-

-

0.001 ± 0.0003

0.00075 ± 0.00015

1.92

1.78

4

4

0.67

0.77

0.90 0.11

0.93 0.13

94

Okra leaf type

Inheritance of okra leaf type was studied in three crosses. Fig. 4.16 shows the segregating

pattern for leaf type in F2 population of the three crosses. It is easy to explain from the Fig.

3.1 that the leaf type segregated into three major different types of shapes. Almost an equal

number of plants exhibited okra and normal leaf types, while a large number of plants

exhibited intermediate leaf type (sub-okra) in the F2 generation.

Chi-Squared values and probabilities of goodness of fit of segregation ratios of F2 and

backcross generations in a study of inheritance of okra leaf type trait in three crosses is

shown in Table 4.18. Non- significant chi- squared values were observed for the segregating

ratios in F2 and backcross generations of the three crosses. Observations of 1 normal : 2 sub-

okra :1 okra, leaf types were observed in the F2 populations of the three crosses. In the

backcrosses with parent-I, ratios of 1 okra : 1 sub-okra, leaf types were obtained. Similarly,

in the backcrosses with parent-II, ratios of 1 normal : 1 sub-okra, leaf types were observed.

Trichomes/Hairiness

The inheritance pattern for hairiness/trichomes was studied in the three cross combinations. It

is evident in all the three cross combinations, that leaf trichomes segregated into three

distinct classes (Fig. 3.2). Chi-Squared values and probabilities of goodness of fit of

segregation ratios of F2 and backcross generations in a study of inheritance of leaf trichomes

trait in three cross combinations are shown in Table 4.19. Non-significant chi- squared values

were observed for the segregating ratios in F2 and backcross generations of the three crosses.

Observations of 1 sparse hairiness : 2 intermediate class of hairiness : 1 pilose hairiness on

leaves were observed in the F2 populations of the three crosses. In the backcrosses with

parent-I, ratios of 1 pilose : 1 intermediate, leaf hairiness were obtained. Similarly, in the

backcrosses with parent-II, ratios of 1 sparse : 1 intermediate, leaf hairiness were observed.

Gossypol glands

The inheritance of the insect resistant trait i.e; gossypol glands was studied in two cross

combinations. One cross of normal glanding with glandless (HRVO-1 × Acala 63-74),

95

96

Table 4.18: Chi-Squared values and probabilities of goodness of fit of segregation ratios of F2 and backcross generations in a study of

inheritance of okra leaf type trait

Cross Generation Expected Ratios

Observed value Expected value χχχχ2

value Probability Normal

leaf Sub- okra

Okra leaf type

Normal Leaf

Sub- okra

Okra leaf type

HRVO-1 × FH 1000

F2 1: 2:1 43 76 31 37.5 75 37.5 1.95 0.25-0.10

BC1 1:1 - 42 48 - 45 45 0.40 0.75-0.50

BC2 1:1 38 52 - 45 45 - 1.88 0.25-0.10

HRVO-1 × CIM 446

F2 1: 2:1 30 80 40 37.5 75 37.5 2.00 0.50-0.25

BC1 1:1 - 50 40 - 45 45 1.11 0.50-0.25

BC2 1:1 39 51 - 45 45 - 1.60 0.25-0.10

HRVO-1 × Acala 63-74

F2 1: 2:1 39 69 42 37.5 75 37.5 1.08 0.75-0.50

BC1 1:1 - 44 46 - 45 45 0.04 0.90-0.25

BC2 1:1 49 41 - 45 45 - 0.71 0.50-0.25

97

Table 4.19: Chi-Squared values and probabilities of goodness of fit of segregation ratios of F2 and backcross generations in a study of inheritance of leaf trichomes trait

Cross Generation Expected Ratios

Observed value Expected value χχχχ2

value Probability Sparse

Hairy

Medium Hairy

Pilose Velvet

Sparse Hairy

Medium Hairy

Pilose Velvet

HRVO-1 × FH 1000

F2 1: 2:1 34 79 37 37.5 75 37.5 0.55 0.90-0.75

BC1 1:1 - 53 37 - 45 45 2.84 0.10-0.05

BC2 1:1 36 54 - 45 45 - 3.60 0.10-0.05

HRVO-1 × CIM 446

F2 1: 2:1 41 69 40 37.5 75 37.5 0.97 0.75-0.50

BC1 1:1 - 37 53 - 45 45 2.84 0.10-0.05

BC2 1:1 48 42 - 45 45 - 0.40 0.75-0.50

HRVO-1 × Acala 63-74

F2 1: 2:1 35 74 41 37.5 75 37.5 0.50 0.90-0.75

BC1 1:1 - 47 43 - 45 45 0.18 0.75-0.50

BC2 1:1 52 38 - 45 45 - 2.18 0.25-0.10

98

Similarly, in the cross HRVO-1 × HG-142 (normal glanding × high glanding) three classes of

high, intermediate and normal glanding types were observed from the F2 data.

Non-significant chi- squared values were observed for the segregating ratios in F2 and

backcross generations of the two crosses. Chi-Squared values and probabilities of goodnessof

fit of segregation ratios of F2 and backcross generations in a study of inheritance of gossypol

glanding trait on unopened cotton bolls are described in Tables 4.20 and 4.21. Observations

of 1 glandless : 2 intermediate glanding : 1 normal glanding, were observed in the F2

populations of the cross, HRVO-1 × Acala 63-74. In the backcross with parent-I, ratios of 1

intermediate glandless : 1 normal glanding whereas, in the backcross with parent-II, ratios of

1 intermediate glandless : 1 glandless were obtained.

In the cross HRVO-1 × HG-142 again a ratio of 1 high glanded : 2 intermediate

glanded : 1 normal glanded obtained in F2. While, in the backcross with parent-I, ratios of 1

intermediate glanded : 1 normally glanded and in the backcross with parent-II, ratios of 1

intermediate glanded : 1 high glanded genotypes were assessed.

4.5. Estimation of heritability and genetic advance for various plant traits

The estimates of narrow sense heritability and expected genetic advances in F2 and estimates

of heritability in F infinity generations are presented in Table 4.17.

In general a high magnitude of narrow sense heritabilities was noticed in all the three

crosses for the trait plant height. The range of the estimates was from 0.67 for HRVO-1

× Acala 63-74 to 0.90 for the cross HRVO-1 × CIM 446. The infinity generation heritability

estimates were found higher than the estimates of narrow sense heritability in F2 generation

for the same trait. In case of number of monopodial branches the narrow sense heritability

estimates were found inconsistent. For the cross HRVO-1 × FH 1000 a high estimate of

narrow sense heritability (0.69) and low estimates (0.34 and 0.36) for HRVO-1 × CIM 446

and HRVO-1 × Acala 63-74 were noticed respectively. The infinity generation heritability

estimates were found higher than their respective estimates of narrow sense heritability in F2

generation. A moderate to high estimate of narrow sense heritability were recorded for

number of sympodial branches. Moderate narrow sense heritability estimates of 0.57 in the

99

Table 4.20: Chi-Squared values and probabilities of goodness of fit of segregation ratios of F2 and backcross generations in a study of

inheritance of gossypol glanding trait on cotton bolls in HRVO-1 × HG- 142

Cross Generation Expected Ratios

Observed value Expected value χχχχ2 value

Probability

Normal glanding

Intermediate glanding

High glanding

Normal glanding

Intermediate glanding

High glanding

HRVO-1 × HG 142

F2 1:2:1 40 71 39 37.5 75 37.5 0.44 0.90-0.75

BC1 1:1 40 50 - 45 45 - 1.11 0.50-0.25

BC2 1:1 - 42 48 - 45 45 0.40 0.75-0.50

Table 4.21: Chi-Squared values and probabilities of goodness of fit of segregation ratios of F2 and backcross generations in a study of

inheritance of gossypol glanding trait on cotton bolls in HRVO-1 × Acala 63-74

Cross Generation Expected Ratios

Observed value Expected value χχχχ2 value

Probability

Normal glanding

Intermediate glandless

Glandless Normal glanding

Intermediate glandless

Glandless

HRVO-1 × Acala 63-74

F2 1:2:1 48 68 34 37.5 75 37.5 3.91 0.25-0.10

BC1 1:1 46 44 - 45 45 - 0.044 0.90-0.75

BC2 1:1 - 49 41 - 45 45 0.18 0.75-0.50

100

The narrow sense heritability estimates in F2 generation of the three crosses for number of

bolls ranged from 0.73 to 0.86. High estimates of narrow sense heritability for seed cotton

yield in three crosses ranging from 0.78 to 0.86 were noticed. In both the traits there was a

consistent increase in the heritability values for infinity generations. The estimates of

heritability in infinity generation for these crosses also showed the same pattern of increase.

High values of narrow sense heritability (0.78 and 0.92) estimates in F2 were observed for

boll weight in two crosses i.e; HRVO-1 × Acala 63-74 and HRVO-1 × CIM 446 respectively,

while lowest value of heritability in narrow sense (0.24) was obtained in the cross HRVO-1

× FH 1000. It is also apparent from the Table 4.17 that the infinity generation heritability

estimates were consistently higher than those in F2 generation. A low to high narrow sense

heritability estimates ranging from 0.37 for HRVO-1 × FH 1000 to 0.70 for HRVO-1

× CIM 446 were observed for the trait lint percentage. The same trend was observed for fibre

length where the estimates of narrow sense heritability in F2 generation ranged from 0.67 to

0.90. The infinity generation heritability estimates were consistently higher than those in F2

generation. In the case of fibre strength, high estimates of narrow sense heritability were

observed in the F2 generation of all the three crosses. From the estimates of the infinity

generation heritability a consistent higher trend was noticed in the values than those in F2

generation.

For fibre elongation and fibre uniformity ratio an increased trend was observed in the

narrow sense heritability estimates in F2, ranging from 0.68 to 0.79 and 0.69 to 0.91

respectively. The infinity generation, heritability estimates were consistently higher than

those in F2 generation. In case for fibre fineness moderate to high estimates of narrow sense

heritability were obtained in F2 generation of the three crosses. High estimates of narrow

sense heritability were observed for HRVO-1 × FH 1000 (0.90) and HRVO-1 × Acala 63-74

(0.78), while a moderate estimate for h2 (ns) (0.56) for fibre fineness in the cross HRVO-1

× CIM 446 was observed. The infinity generation (F) heritability estimates were consistently

higher than those recorded in F2 generation.

For number of trichomes, very high estimates of narrow sense heritability in F2

generation were observed in all the three crosses ranging from 0.88 to 0.91. Almost the same

pattern of heritability estimates in infinity generation was obtained for number of trichomes

101

in all the three crosses. For the two traits i.e; gossypol content and total gossypol, the narrow

sense heritability estimates in F2 generation of two crosses viz, HRVO-1 × Acala 63-74 and

HRVO-1 × HG-142 followed the same pattern as were observed for number of trichomes.

The h2 (ns) estimates ranged from 0.79 to 0.83 for gossypol content in HRVO-1 × HG-142

and HRVO-1 × Acala 63-74, while 0.67 to 0.77 for total gossypol in HRVO-1 × Acala 63-74

and HRVO-1 × HG-142 respectively were recorded. From the estimates of the infinity

generation heritability a consistent higher trend was noticed in the values than those noticed

for the narrow sense heritability estimates in F2 generation.

Based upon the estimates of narrow sense heritability, the extent of genetic advance is

shown in the Table 4.17. High estimates of genetic advance i.e; 135.56, 152.42 and 57.53

were obtained for number of trichomes in the three crosses i.e; HRVO-1 × FH 1000,

HRVO-1 × CIM 446 and HRVO-1 × Acala 63-74 respectively. For seed cotton yield high

estimates of genetic advance i.e; 14.78 and 15.52 were observed in crosses HRVO-1

× FH 1000 and HRVO-1 × Acala 63-74 respectively. For plant height, in the cross, HRVO-1

× CIM 446 high estimates of genetic advance (11.96) were recorded. In fibre uniformity ratio

higher estimate of genetic advance (7.52) was recorded for HRVO-1 × FH 1000. Moderate

estimates of genetic advance (5.92) were recorded for plant height in HRVO-1 × FH 1000.

Similarly, in the crosses HRVO-1 × FH 1000 and HRVO-1 × CIM 446, moderate estimates

of 5.16 and 6.23 respectively were recorded for number of sympodial branches and 6.02 and

5.54 respectively for number of bolls. For all other traits the estimates of genetic advance

remained less which ranged from 0.13 to 4.65.

4.6. Estimation of heterosis and inbreeding depression for various plant traits

The estimates of heterosis and inbreeding depression for various plant traits are given in

Table 4.22. Significant and positive heterosis for boll weight (17.07 %) was observed in

HRVO-1 × FH 1000 while in the other two crosses, HRVO-1 × CIM 446 and HRVO-1

× Acala 63-74 positive and highly significant heterotic estimates of 14.58 % and 13.20 %

respectively, were observed. For number of monopodial branches, positive and highly

significant heterosis with the value of 8.26 % in the cross HRVO-1 × CIM 446. Whereas,

significant but negative heterosis (-30.81 %) was reported for the number of monopodial

branches in the cross HRVO-1 × Acala 63-74. For fibre fineness significant but negative

102

Table 4.22: Estimates of heterosis and inbreeding depression for various plant traits in different crosses

Traits HRVO × FH-1000 HRVO × CIM-446 HRVO × Acala 63-74 Het (%) I.B.D (%) Het (%) I.B.D (%) Het (%) I.B.D (%)

Plant height 0.97 2.27 5.88 4.72 1.10 7.15

No. of monopodial branches/plant

17.39 -23.31 8.26** -8.28** -30.81* -0.43

No.of sympodial branches/plant

37.18 16.53 39.36 14.36 25.29 9.57

No. of bolls/plant 24.31 20.52 29.95 17.29 18.07 11.21

Seed cotton yield 44.40 25.66 47.01 15.81 31.39 20.16

Boll weight 17.07* 11.92* 14.58** 8.64* 13.20** 9.91*

Lint percentage 14.13 7.74 5.66 5.31 1.53 0.71

Fibre length 4.83 2.13 5.92 5.82 1.52 1.88

Fibre strength 4.11 2.90 4.78 3.74 2.87 1.30

Fibre elongation 5.39 3.09 9.91 4.67 -1.63 1.86

Fibre uniformity ratio -6.86 -1.71 2.20 1.89 1.83 2.62

Fibre fineness -7.86* 2.25 -11.34** -7.21* -4.60 1.47*

No. of trichomes -26.18 -18.35** -55.33 -68.81 39.93 12.70

Traits HRVO × Acala 63-74 HRVO × HG-142

Het (%) I.B.D (%) Het (%) I.B.D (%)

Gossypol content -55.56** -42.86** 1.15 15.91*

Total gossypol -60.32** -62.00** 0.29 0.00

*, ** = Significant at 0.05 and 0.01 probability levels, respectively.

103

value of -7.86 % was observed in HRVO-1 × FH 1000 while highly significant and negative

estimate of –11.34 % was observed in the cross HRVO-1 × CIM 446 where fibre fineness

was measured in micronaire. For gossypol content and total gossypol though negative but

highly significant heterotic values i.e; -55.56 % and –60.32 % were recorded in HRVO-1

× Acala 63-74. A positive and significant inbreeding depression i.e; 11.92 %, 8.64 % and

9.91 % was recorded for boll weight in the crosses HRVO-1 × FH 1000, HRVO-1

× CIM 446 and HRVO-1 × Acala 63-74 respectively. Significant and positive inbreeding

depression with the value of 1.47 % was recorded for fibre fineness in the cross HRVO-1

× Acala 63-74 but significantly negative inbreeding depression value in HRVO-1 × CIM 446

was recorded in the same trait. Positive and significant inbreeding depression 15.91 % for

gossypol content in the cross HRVO-1 × HG-142. Highly significant though negative values

of inbreeding depression were recorded for number of monopodial branches (-8.28 %) in

HRVO-1 × CIM 446, for number of trichomes (-18.35 %) in HRVO-1 × FH 1000 and for

gossypol content and total gossypol in the cross HRVO-1 × Acala 63-74 as – 42.86 % and

– 62.00 % respectively.

4.7. Correlations

The phenotypic and genotypic correlations among the agronomic, fibre and insect related

traits were studied making use of the data of the F2 populations in three crosses. Correlation

at both the phenotypic and genotypic levels for the three crosses are given in the Tables 4.23

to 4.28.

In general, the magnitude of the genetic correlations was higher than that of the

correlations at the phenotypic level in all the three crosses.

4.7.1 Correlation among insect resistant and fibre traits

(a) Leaf type

The correlation of leaf type with all other traits related to fibre (fibre length, fibre strength,

fibre uniformity ratio, fibre fineness and lint percentage) shown in Table 4.23, 4.24 and 4.25

showed non-significant associations except for fibre elongation where significant correlation

was recorded in all the three crosses. In the cross HRVO-1 × Acala 63-74 there existed

positive and significant association of leaf type with fibre elongation both at the phenotypic

104

and genotypic levels. In rest of the two crosses, there was negative correlation between leaf

type and fibre elongation.

(b) Trichomes/Hairiness

In all of the three crosses as shown in Table 4.23, 4.24 and 4.25 there existed positive and

significant correlation of number of trichomes with lint percentage. For fibre length and fibre

strength there was significant and negative correlation with number of trichomes. A

significant and positive correlation of number of trichomes with micronaire (negative

correlation of number of trichomes with fibre fineness) was observed in all of the three

crosses. In case of fibre uniformity ratio and fibre elongation, a negative association with

number of trichomes was observed in crosses, HRVO-1 × FH 1000 and HRVO-1 × CIM 446

respectively, whereas, in the cross HRVO-1 × Acala 63-74 there existed a positive but

non-significant association of number of trichomes with fibre elongation and fibre uniformity

ratio.

(c) Gossypol content and total gossypol

Non-significant correlation of gossypol content with all of the fibre traits in the single cross

of HRVO-1 with Acala 63-74 was recorded (Table 4.25).

4.7.2 Correlation among fibre quality traits

Correlation among fibre traits in the three crosses is shown in Tables 4.23 to 4.25. Lint

percentage was significantly and negatively associated to fibre length in the crosses of

HRVO-1 × FH 1000 and HRVO-1 × CIM 446 respectively. In contrary, a positive and

significant correlation of lint percentage with fibre strength, fibre elongation and fibre

uniformity ratio was recorded in these two crosses. In the cross of HRVO-1 and Acala 63-74

a significant and negative correlation existed between lint percentage and fibre elongation. In

all the three crosses, a positive (negative correlation with fibre fineness) and significant

correlation between lint percentage and micronaire was observed. Fibre length was positively

and significantly correlated with fibre strength in all the three crosses. Positive and

105

Table 4.23: Genotypic (upper value) and phenotypic (lower value) correlations among insect resistant and fibre related traits in HRVO-1 × FH 1000

Traits L% FL FS FE U% FF T Lt

Lint percentage (L %)

-0.94* -0.92**

0.92* 0.90**

0.97* 0.93**

0.42* 0.42

0.83 0.82**

0.81 0.81**

-0.58 -0.56

Fibre length

(FL)

0.99* 0.97**

0.95* 0.91**

0.68* 0.67**

-0.83 -0.82**

-0.96 -0.96**

-0.48 -0.47

Fibre strength (FS)

0.88* 0.85**

0.70* 0.68**

-0.76 -0.74**

-0.98 -0.96**

-0.37 -0.39

Fibre elongation (FE)

0.43 0.42

-0.96 -0.94**

-0.83 -0.81**

-0.70 -0.68**

Fibre uniformity ratio (U%)

-0.23 -0.23

-0.83 -0.83**

-0.55 -0.58

Fibre fineness (FF)

0.69* 0.68

-0.60 -0.65

L% = Lint percentage, FL = Fibre length, FS = Fibre strength, FE = Fibre elongation, U% = Fibre uniformity ratio, FF = Fibre fineness, T = No. of trichomes, Lt = Leaf type

*, ** = Significant at 0.05 and 0.01 probability levels, respectively.

106

Table 4.24: Genotypic (upper value) and phenotypic (lower value) correlations among insect resistant and fibre related traits in HRVO-1 × CIM 446

Traits L% FL FS FE U% FF T Lt

Lint percentage (L %)

-0.99* -0.98**

1.00* 0.97**

0.95* 0.93**

0.94* 0.92**

0.94 0.93**

0.92 0.91**

-0.53 -0.55

Fibre length (FL)

1.01* 0.99**

0.92* 0.91**

0.96* 0.95**

-0.91 -0.90**

-0.95 -0.95**

-0.38 -0.45

Fibre strength (FS)

0.95* 0.92**

0.97* 0.96**

-0.94 -0.92**

-0.98 -0.96**

-0.37 -0.39

Fibre elongation (FE)

0.92* 0.91**

-0.96 -0.95**

-0.95 -0.94**

-0.76 -0.75**

Fibre uniformity ratio (U%)

-0.86 -0.85**

-0.97 -0.96**

-0.60 -0.61

Fibre fineness (FF)

0.90* 0.89

-0.70 -0.71

L% = Lint percentage, FL = Fibre length, FS = Fibre strength, FE = Fibre elongation, U% = Fibre uniformity ratio, FF = Fibre fineness, T = No. of trichomes, Lt = Leaf type

*, ** = Significant at 0.05 and 0.01 probability levels, respectively.

107

Table 4.25: Genotypic (upper value) and phenotypic (lower value) correlations among insect resistant and fibre related traits in HRVO-1 × Acala 63-74 Traits L

% FL FS FE U% FF T Lt G mg Tg%

Lint percentage (L %)

0.05 0.04

0.44 0.43

-0.99 -0.98**

-0.34 -0.31

0.56 0.55*

0.88 0.88**

-0.42 -0.48

0.34 0.36

0.33 0.35

Fibre length (FL)

0.57 0.56*

0.75* 0.77**

0.71 0.63**

-0.86 -0.79**

-0.75* -0.72*

-0.04 -0.03

-0.26 -0.24

-0.26 -0.24

Fibre strength (FS)

-0.41 -0.39

-0.13 -0.10

-0.54 -0.50*

-0.90 -0.80*

-0.30 -0.28

-0.60 -0.56

-0.62 -0.59

Fibre elongation (FE)

0.31 0.29

-0.54 -0.54*

0.41 0.42

0.99* 0.98**

0.54 0.56

0.53 0.56

Fibre uniformity ratio (U%)

-0.91* 0.88**

0.36 0.35

0.40 0.39

0.46 0.45

0.45 0.43

Fibre fineness (FF)

0.43* 0.43

-0.60 -0.65

0.34 0.36

0.35 0.36

L% = Lint percentage, FL = Fibre length, FS = Fibre strength, FE = Fibre elongation, U% = Fibre uniformity ratio, FF = Fibre fineness, T = No. of trichomes, Lt = Leaf type, G mg = Gossypol content, Tg% = Total gossypol *, ** = Significant at 0.05 and 0.01 probability levels, respectively.

108

Significant correlation of fibre length with fibre elongation and fibre uniformity ratio was

recorded in all the three crosses. Fibre length was significantly and negatively correlated with

micronaire (fibre length significantly and positively correlated to fibre fineness) in all of the

three crosses under study. Fibre strength in all the three crosses was significantly and

negatively correlated with micronaire value (a positive correlation of fibre strength with fibre

fineness). For two crosses, HRVO-1 × FH 1000 and HRVO-1 × CIM 446, fibre strength was

significantly and positively correlated with fibre elongation and fibre uniformity ratio.

In case of fibre elongation, there was a significant and negative correlation with

micronaire (means a positive correlation with fibre fineness) in all of the three crosses. While

a positive and significant correlation of fibre elongation with fibre uniformity ratio in

HRVO-1 × CIM 446 was observed. Fibre uniformity ratio was significantly and negatively

correlated with micronaire (means a positive correlation with fibre fineness) in the crosses

HRVO-1 × CIM 446 and HRVO-1 × Acala 63-74.

4.7.3 Correlation among other morphological and yield related traits

In case of plant height, significant and positive correlation, (Table 4.26 to Table 4.28) in the

crosses HRVO-1 × FH 1000 and HRVO-1 × CIM 446, existed with number of sympodial

branches, number of bolls, seed cotton yield and boll weight. In the third cross, HRVO-1 ×

Acala 63-74, same trend of correlation of plant height with other traits was noticed except for

boll weight which was non-significant. A significant and negative correlation of number of

monopodial branches with seed cotton yield and boll weight was noticed only in the cross

HRVO-1 × Acala 63-74. Number of sympodial branches in all of the three crosses was

significantly and positively associated with number of bolls, seed cotton yield and boll

weight. In case of number of bolls a positive as well as significant correlation was observed

with seed cotton yield and boll weight in all the three crosses. Seed cotton yield in all the

three crosses was positively and significantly correlated with boll weight.

4.7.4 Correlation among morphological, yield and insect resistant traits

Correlation among morphological, yield and insect resistant traits are depicted in the Tables

4.26, 4.27 and 4.28.

109

Table 4.26: Genotypic (upper value) and phenotypic (lower value) correlations among insect resistant and morphological and yield related traits in HRVO-1 × FH 1000

Traits PH NMB NSB NBP SCY BWt T Lt

Plant height (PH)

-0.35 -0.34

0.48* 0.45

0.63* 0.61**

0.61* 0.59**

0.49* 0.48*

-0.35 -0.34

-0.39 -0.38

No. of monopodial branches/plant (NMB)

0.18 0.18

0.03 0.04

0.09 0.09

0.22 0.21

0.16 0.15

0.39* 0.36

No. of sympodial branches/plant (NSB)

0.98* 0.97**

0.98 0.97**

1.00* 0.99**

-0.18 -0.18

0.77 -0.77**

No. of bolls/plant (NBP)

0.99* 0.99**

0.98* 0.98**

-0.29 -0.28

0.81 0.81**

Seed cotton yield (SCY)

0.98* 0.94**

0.27 0.26

0.78 0.78**

Boll weight (BWt.)

0.17 -0.18

0.76 0.76**

PH = Plant height, NMB = No. of monopodial branches/plant, NSB = No. of sympodial branches per plant, NBP = No. of bolls/plant, SCY = Seed cotton yield, BWt = Boll weight, T = No. of trichomes, Lt = Leaf type *, ** = Significant at 0.05 and 0.01 probability levels, respectively.

110

Table 4.27: Genotypic (upper value) and phenotypic (lower value) correlations among insect resistant and morphological and yield related traits in HRVO-1 × CIM 446

Traits PH NMB NSB NBP SCY BWt T Lt

Plant height (PH)

0.05 0.05

0.92* 0.92**

0.94* 0.93**

0.97* 0.96**

0.95* 0.95**

-0.88 -0.87**

-0.71 -0.70**

No. of monopodial branches/plant (NMB)

0.37 0.35

0.26 0.24

0.25 0.24

0.24 0.24

-0.07 -0.06

0.17* 0.17

No. of sympodial branches/plant (NSB)

0.92* 0.91**

0.96* 0.96*

0.99* 0.99**

0.67 0.68

0.67 0.67**

No. of bolls/plant (NBP)

0.99* 0.98**

0.91* 0.90**

-0.59 -0.58

0.64 0.74*

Seed cotton yield (SCY)

0.96* 0.95**

0.48 0.48

0.56 0.55*

Boll weight (BWt)

0.41 0.41

0.74 0.74**

PH = Plant height, NMB = No. of monopodial branches/plant, NSB = No. of sympodial branches per plant, NBP = No. of bolls/plant, SCY = Seed cotton yield, BWt = Boll weight, T = No. of trichomes, Lt = Leaf type *, ** = Significant at 0.05 and 0.01 probability levels, respectively.

111

Table 4.28: Genotypic (upper value) and phenotypic (lower value) correlations among insect resistant and morphological and yield related traits in HRVO-1 × Acala 63-74

Traits PH NMB NSB NBP SCY BWt T G mg Tg% Lt

Plant height (PH)

0.22 0.19

0.19* 0.18

0.32* 0.30

0.33* 0.32

0.27 0.27

-0.12 -0.11

-0.08 -0.07

-0.10 -0.09

-0.13 -0.13

No. of monopodial branches/plant (NMB)

0.29 0.28

0.29 0.15

-0.41* -0.40

-0.55 -0.54*

-0.29 -0.28

-0.55 -0.54

-0.52 -0.51

-0.78 -0.76**

No. of sympodial branches/plant (NSB)

0.94* 0.93**

0.95* 0.94**

0.94* 0.93**

-0.40 -0.40

-0.23 -0.23

-0.34 -0.34

0.80 0.79**

No. of bolls/plant (NBP)

0.95* 0.96**

0.89* 0.88**

-0.41 -0.41

0.25 0.24

0.27 0.26

0.65 0.65**

Seed cotton yield (SCY)

0.98* 0.98**

-0.35 -0.35

0.95 0.94**

0.96 0.95**

0.83 0.83**

Boll weight (BWt)

-0.48 -0.48

-0.98 -0.98**

-0.98 -0.98**

0.92 0.93**

PH = Plant height, NMB = No. of monopodial branches/plant, NSB = No. of sympodial branches per plant, NBP = No. of bolls/plant, SCY = Seed cotton yield, BWt = Boll weight, T = No. of trichomes, Lt = Leaf type, G mg = Gossypol content, Tg% = Total gossypol *, ** = Significant at 0.05 and 0.01 probability levels, respectively.

112

Plant height was recorded to be in negative and significant association with the leaf

type in HRVO-1 × CIM 446. Number of monopodial branches had a significant and positive

correlation with the leaf type in the crosses HRVO-1 × FH 1000 and HRVO-1 × CIM 446

whereas, a negative but significant correlation for the same trait with leaf type in the cross of

HRVO-1 × Acala 63-74. A significant and positive correlation of number of sympodial

branches, number of bolls, seed cotton yield and boll weight with leaf type was recorded in

all the three crosses.

The number of trichomes were negatively but significantly correlated with plant

height in the cross HRVO-1 × CIM 446. Non-significant correlations of number of trichomes

with number of monopodial branches, number of sympodial branches, number of bolls, seed

cotton yield and boll weight were recorded in all of the three crosses studied.

For gossypol content and total gossypol, non-significant correlation with plant height,

number of monopodial branches, number of sympodial branches and number of bolls was

recorded in the cross HRVO-1 × Acala 63-74. Significant and positive correlation of

gossypol content and total gossypol with seed cotton yield was observed. In case of boll

weight, significant and negative association with gossypol content and total gossypol was

recorded in the cross HRVO-1 × Acala 63-74.

113

CHAPTER 5

DISCUSSION

In order to conduct the inheritance studies on three important insect resistant traits i.e; okra

leaf type, gossypol glands and trichomes, a germplasm comprising of 31 entries from

different sources was collected and evaluated on the phenotypic basis for the presence of

okra leaf type, gossypol glands and trichomes (Appendix I). On the basis of the assessment

for the above mentioned three traits, altogether five entries were selected to study the

inheritance pattern of these traits alongwith other morphological and fibre related

characteristics.

5.1. Genetic basis of variation for morphological, yield, fibre and insect resistant traits

Analysis of variance (Steel and Torrie, 1980) revealed significant variation for the characters

under study in all the generations of all the crosses (Appendices II- XVI). The generation

mean comparison based on LSD (0.05) values also indicated significant variation for all the

morphological, fibre and insect related traits (Tables 4.1- 4.15). The frequency distribution of

various agronomic and fibre traits in F2 populations was observed to be normal (Fig. 4.1 -

4.12). The distribution showed continuous variation indicating the quantitative nature of

these traits. In all the traits some F2 plants excelled their parents indicating transgressive

segregation.

In case of the insect resistant traits like number of trichomes (Fig. 4.13), gossypol

content (Fig. 4.14) and total gossypol (Fig. 4.15), the segregation pattern in F2 indicated the

discontinuous variation, which confirmed the qualitative nature of the traits

(Endrizzi et al., 1984). For number of trichomes, almost an equal number of plants showed

pilose hairiness and normal/sparse hairiness, while a large number of plants exhibited

intermediate hairiness in the F2 generation which indicated incomplete dominance for

trichomes (Knight, 1952; Niles, 1980). It was noticible from the Fig. 4.13 (a) that a very

small proportion of plants fell in another intermediately resembling hairiness category. This

phenotypic expression of the intermediate hairy state in heterozygous condition was probably

114

affected by the genetic background of the parents indicating, modifying gene effects

(Rahman and Khan, 1998). Three classes were observed genotypically for gossypol content

and total gossypol in F2 (Fig. 4.14 and 4.15) which indicated incomplete dominance

(Calhoun, 1997). The significant differences (P< 0.05) found in the results for all these traits

provided the basis for further studies in relation to the gene action by using more

sophisticated techniques.

5.2. Generation mean analysis for various plant traits

The knowledge about the nature and magnitude of genetic effects prevailing in the breeding

material is necessary to decide the kind of breeding procedure to be followed. In case of

quantitative characters, the inheritance pattern is complex as the individual gene effect is

minor. For quantitatively inherited traits, mainly diallel and generation mean analyses are

commonly used. The adoption of a biometric method, which could provide information on

the inheritance pattern for various plant traits in cotton become important. Generation mean

analysis falls into first order statistics which gives the meaning that the assessment for

various attributes are on mean basis. The calculation of first order statistics is simple and

estimates are more robust and reliable due to their precision. It is therefore, a powerful

technique for the assessment of the gene effects. The generation mean analysis makes use of

the data obtained from the segregating and non-segregating generations and provides

information on the genetic effects i.e; additive and non-additive (Mather and Jinks, 1982).

This technique of genetic analysis had been used previously by Percy et al. (1996) for

stomatal conductance, Rowena et al. (2002) for agronomically important traits in oats, Dani

and Kohel (1989) for the nature of gene action in cottonseed oil attributes, Kumar and

Raveendran (1999) for the morphological and yield and yield related traits in cotton, Bertini

et al. (2001) for the study of gene action in yield and fibre related traits in cotton, Liu et al.

(2000) for agronomic, yield and fibre attributes and Iqbal and Nadeem (2003) for the study of

the genic effects for yield of seed cotton and number of sympodial branches in Upland cotton

crosses following generation mean analysis.

In quantitatively inherited traits, gene action is described as additive, dominance and

epistatic effects. Additive effect is defined as the average effect of genes; dominance as the

interaction of allelic genes and epistatsis as the interaction of non-allelic genes that influence

115

a particular trait. Generation mean analysis of the data showed additive and non-additive

types of gene effects (Table 4.16). The genetic analysis for various plant traits revealed that

mostly additive component [d] was involved in the inheritance of various plant traits but it

was noticed that the dominance component [h] was greater than the corresponding additive

component [d] whenever, it was present for most of the plant traits except for the traits fibre

strength, number of trichomes, gossypol content and total gossypol in the crosses. The

additive × additive [i] in most of the plant traits where it was present found greater than the

additive [d] component except for seed cotton yield, boll weight and fibre elongation. The

additive × dominance [j] and dominance × dominance [l] component, whenever, it was

present also found greater than the additive [d] component. The low magnitude of the

additive component [d] than that of the non-additive components indicated involvement of

additive and non-additive genes in the inheritance of various plant traits. The additive

× additive [i] interaction suggested the fixation of additive alleles in the later segregating

generations, as suggested by Singh and Narayanan (2000), where ever it was present. These

alleles in the later generations can be fixed by making simple selections. The positive sign for

additive × additive [i] interaction, as evident from Table 4.16, was for almost every trait

where ever it was present. The positive sign was the indication towards the effect of

favourable or increasing alleles and vice versa but this was reverse in case of number of

monopodial branches for the cross HRVO-1 × FH 1000 which indicated that the number of

monopodial branches will be reduced in the later generations. For fibre strength, number of

trichomes, gossypol content and total gossypol the additive component [d] was found greater

than the dominace [h] component which indicated the preponderance of the additive effects

over the dominance effect. The absence of epistatic interactions where ever noticed in the

computation of the genetic effects in plant traits, showed simple inheritance and selection

could be done to make improvement in those traits right from the early generations. In

contrary, in the presence of epistasis selection is delayed to the later generations. For number

of monopodial branches, no genetic effect was observed in the cross HRVO-1 × CIM 446,

only mean value was fit. This indicated that the existing variability for this trait was due to

environmental influence. The non-significant χ2 values for various traits revealed that the

models were best fit of the observed to estimated values (Mather and Jinks, 1982).

116

In the present investigation, for plant height, the involvement of additive and

non-additive effects in first two crosses (Table 4.16) were in accordance to the findings of

Kalsey and Vithal (1980), Randhawa et al. (1986), Tyagi (1988) and Kumaresan et al.

(1999). However, Khan et al. (1999) and Neelima et al. (2004) observed additive effects

while the findings of Hassan et al. (1999), Islam et al. (2001) and Subramanian et al. (2005)

recorded the involvement of dominance and epistatic effects in the control of plant height as

has been observed in the cross HRVO-1 × Acala 63-74 in the present studies. For number of

monopodial branches in crosses HRVO-1 × FH 1000 and HRVO-1 × Acala 63-74 additive

and non-additive components were observed. These findings were similar to the findings of

Neelima et al. (2004). In case of number of sympodial branches in all the three crosses

additive and dominance effects were observed. However, only additive effects were observed

for this trait by Neelima et al. (2004) while dominance and epistatic effects were noticed by

Subramanian et al. (2005).

For the number of bolls a substantially high amount of positive dominance [h] effects

in the direction of the higher parent were observed in all the three crosses. Additive and non-

additive components of generation means were evident in the cross HRVO-1 × FH 1000,

accompanied by negative dominance × dominance [l], thus corroborated the findings of

Pathak and Singh (1970), Gad et al. (1974), Bhatade and Bhale (1983) and Silvia and Alves

(1983). However, additive effects were noticed for number of bolls by Gad et al. (1974),

Gill and Kalsey (1981), Silvia and Alves (1983) and Rehman et al. (1988). Simple additive-

dominance model was best fit for seed cotton yield with no interaction in HRVO-1

× CIM 446. These findings of additive-dominance effects for seed cotton yield were in

confirmity to the findings of Kalsey and Vithal (1980), Parkash (1982) and Kalsey and Garg

(1988). In other two crosses, additive and non-additive components of generation means

were observed. The results obtainted in these two crosses corroborated the findings of

Mert et al. (2003), Ramalingam and Sivasamy (2002) and Liu et al. (2000). The same trend

for boll weight in all the three crosses was observed as had been observed for seed cotton

yield. Additive-dominance model alongwith a non-allelic interaction (additive × additive [i])

accounted for boll weight in two crosses HRVO-1 × FH 1000 and HRVO-1 × Acala 63-74

(Silvia and Alves, 1983; Kaseem et al., 1984; Kalsey and Garg, 1988 and

117

Kumaresan et al., 1999). While Khan et al. (1999) and Ahmad et al. (2001) observed

additive effects for boll weight. Dominance effects were observed for boll weight by Neelima

et al. (2004) and Subramanian et al. (2005).

In the case of lint percentage, additive and non-additive types of gene action

(Mert et al., 2003) were recorded in first two crosses as evident from the Table 4.16, while

only additive component showed its fitness in the cross HRVO-1 × Acala 63-74

(Bertini et al., 2001 and Subhan et al., 2002). In the study of fibre length, simplest model m,

additive [d] and dominance [h] was fit in the cross HRVO-1 × FH 1000, which were in line

with the findings of Nistor and Nistor (1999) and Babar and Khan (1999). However, in the

cross HRVO-1 × Acala 63-74 dominance effects were recorded as were observed by

Murthy (1998). The findings of Pathak (1975) indicated dominance and dominance

× dominanace [l] effects controlling fibre length. Murtaza et al. (2004) also reported epistatic

effects in the control of fibre length. In the cross of HRVO-1 and CIM 446, additive [d],

dominance [h] and additive × additive [i] interaction were involved in the control of the fibre

length, which were in accordance to the findings of Sayal et al. (1996) and to some extent

with the findings of Singh and Yadavendra (2002) who in addition to these components also

reported additive × dominance [ j ] component. In fibre strength three- parameter model was

best fit in HRVO-1 × FH 1000 and HRVO-1 × CIM 446. These results obtained from these

two crosses were in agreement to the findings of Murtaza et al. (2004). In another study

Pathak (1975) and Hendaway et al. (1999) reported additive, dominance as well as additive

× additive interactions. However, in the cross HRVO-1 × Acala 63-74 only additive effects

were noticed which were in agreement to the findings of Nadeem and Azhar (2005).

However, Meredith and Bridge (1972) concluded additive effects for fibre strength, fineness

and lint yield.

In first two crosses for fibre elongation simplest additive-dominance model was best

fit while in the third cross (HRVO-1 and Acala 63-74) additive-dominance model along with

an additive × additive [i] interaction was best fit. Only additive effects were observed for

fibre elongation by Mukhtar et al. (2000). In the case of fibre uniformity ratio additive,

dominance as well as additive × additive interaction were observed in crosses HRVO-1

× CIM 446 and HRVO-1 × Acala 63-74, while in the cross HRVO-1 × FH 1000, dominance

118

alongwith additive × additive [i] and additive × dominance [ j ] were observed. The findings

of Mukhtar et al. (2000) reported the presence additive effects only. In case of fibre fineness,

additive, dominance as well as additive additive interaction (Lin and Zhao, 1988) were

recorded in HRVO-1 × FH 1000 and HRVO-1 × Acala 63-74. Whereas, in the cross

HRVO-1 × CIM 446 simplest additive-dominance model was adequate. The negative sign of

dominance [h] is an indication of the direction of dominance effects towards the lower

parent. Pathak (1975), Rehman et al. (1993), Babar and Khan (1999) and

Mukhtar et al. (2000) also observed similar results, however, other scientists like

Innes et al. (1975) and Pavasia et al. (1999) reported additive type of gene action governing

the inheritance of this character.

In the present studies trichome counts were made and subjected to generation mean

analysis, in order to investigate the type of gene action involved. A view of the Table 4.16

showing the components of generation means for these traits indicate that in all the three

crosses, m, d and h components were found. The magnitude of the additive [d] component

was higher than the respective dominance [h] component. Moreover, in the study of the trait

number of trichomes in all the three crosses, the dominance [h] component, though lesser

than additive component, but of high magnitude was recorded. The negative sign of the

dominance [h] effects in the crosses HRVO-1 × FH 1000 and HRVO-1 × CIM 446 indicated

the direction of the intermediate class towards the parents (FH 1000 and CIM 446)

contributing less number of trichomes (Fig. 4.13 a and 4.13 b). In contrary, the cross

HRVO-1 × Acala 63-74 had positive value of the dominance [h] effect which indicated the

direction of the intermediate class of trichomes towards the high trichomes contributing

parent (HRVO-1) indicating incomplete dominance (Fig. 4.13 c). For the number of

trichomes incomplete pattern of inheritance had already been reported by Simpson (1947),

Knight (1952), Niles (1980) and Rahman and Khan (1998 ). The basis of their findings were

purely visual and phenotypic. So the qualitative studies of the above scientists for the gene

action in the trait of number of trichomes supported the quantitatively driven findings in the

present studies.

In the case of gossypol content and total gossypol traits studied in two crosses i.e;

HRVO-1 × Acala 63-74 and HRVO-1 × HG-142, the inheritance pattern was investigated

119

only on quantitative basis. It has been reported in the literature by Niles (1980 a) that for the

gossypol glands Gl2 and Gl3 act as two principal determinants which acting additively. In the

present studies, for gossypol content as well as for total gossypol, m, additive [d] and

dominance [h] components were observed for HRVO-1 × Acala 63-74. However, it was

noticible that the magnitude of the dominance [h] component was very negligible than the

magnitude of the respective additive [d] component in the two crosses. The findings of Kohel

(1987) proved that the additive effects were greater in crosses involving glandless lines than

in crosses involving glanded lines. This finding corroborates to the present findings in the

cross HRVO-1 × Acala 63-74. The negative sign for the dominance [h] component in both

the traits was the indication of this component towards the lower parent considered on the

basis of the parental differences in gossypol content and total gossypol. In the study of these

two traits in the cross HRVO-1 × HG-142, m and additive [d] components were principally

observed. These quantitatively derived results for the inheritance pattern of glanding trait

were similar to those of Lee et al. (1968), who proposed the glandulosity of embryos were

largely additive and Lee (1973) who showed that additive effects accounted for more than

90% of the total genetic variance for seed gossypol level.

5.3. Generation variance analysis for various plant traits

A model incorporating additive (D) and environmental (E) components of varaince (Table

4.17) showed adequacy for most of the traits in three crosses except for lint percentage in the

cross HRVO-1 × FH 1000, where additive (D), additive × dominance (F) and environmental

(E) showed adequacy. The same results were achieved in a study by Singh and Sandhu

(1979) for lint percentage. A satisfactory fit of the model incorporating only additive (D) and

environmental (E) in the generation variance analysis suggested the preponderance of

additive genetic variation in the control of all the traits except for lint percentage in the cross

HRVO-1 × FH 1000, where there was the involvement of an additive × dominance (F)

interaction. All the genetic parameters were found significant in view from their respective

values of standard error. The significance of the environmental component (E) in all the traits

indicated the role of environment in the control of these traits. In critical view of the Table

4.17 it is noted that the values of additive (D) componenet of generation variance was found

higher than that of the respective environmental component of variance for all the traits in all

120

the crosses, which suggested the importance of additive gene effects for the expression of a

character suggesting, no need of further progeny testing. The selection product could be

utilized as a variety, pure line or strain to be improved within a population. In other words,

intra-population selection methods will be effective in accumulating favourable alleles

whereas, in the case of lint percentage in the cross HRVO-1 × FH 1000, the involvement of

additive × dominance interaction suggested delayed selection after progeny testing for the

trait rather than early selection. The non-significant χ2 values for various traits in the

generation variance analysis revealed that the models were best fit of the observed to the

estimated values (Mather and Jinks, 1982). Additive genetic variance was held responsible

for the genetic variation for various cotton plant traits as reported by Bary et al. (1975),

Singh and Singh (1981), Singh (1982), Randhawa et al. (1986), Shory et al. (1986), Yadav

and Yadava (1987) and Kohel (1987). Other scientists like Pathak and Singh (1970), El-

Fawal et al. (1974), Kalsey and Vithal (1980), Prakash (1982), Gupta (1987) and Ji and Zhu

(1988) stressed for both additive and non-additive variances for various plant traits.

5.4. Inheritance studies pertaining to insect resistant traits

The present studies were conducted to ascertain the inheritance pattern for the insect

resistance traits in the crosses. The traits, okra leaf type, gossypol glands and trichomes were

selected on the basis of their relative importance and resistance to the insect pest population

in cotton plants. These traits were reported to be oligogenically controlled

(Endrizzi et al., 1984). In view of their qualitative mode of inheritance, Chi-square test was

employed to test the differences of the observed vs the expected segregating phenotypic

ratios (Harris, 1912).

Okra leaf is a deeply lobed leaf shape that is a monogenic trait governed by the L0

gene which is incompletely dominant to normal l0 (Hammond, 1941; Niles, 1980). Fig. 4.16

showing the segregating pattern in F2 for leaf shape into three different types or shapes

suggested incomplete dominance in the three crosses. The two homozygous extremes for leaf

type: okra (L0L0) and broad/normal (l0l0) were easily distinguishable. In the present studies,

the parents in the three crosses involving okra leaf and normal leaf plants were hybridized to

obtain sub-okra (L0l0) progeny in F1 showing incomplete dominance. The segregation in the

backcrosses with parent-I and parent-II also fit to the theoretical ratio of 1:1 further

121

confirmed the incomplete pattern of inheritance.The segregation of the leaf shape in F2

generation into three classes: okra leaf, normal and the intermediate leaf shape (sub-okra) and

fitting into the theoretical 1:2:1 a monohybrid ratio of incomplete dominance in the present

studies corroborated the findings of Green (1953) and Rahman and Khan (1998). The

non-significant χ2 in F2 for leaf shape in all the crosses fit well against the theoretical ratio.

However, the phenotypic expression of sub-okra leaf shape in heterozygous condition was

affected by the genetic background of the parents i.e; modifying gene effects (Rahman and

Khan, 1998).

The trichome cover of a plant surface is collectively called as pubescence. From the

Table 4.19 the phenotypic classes were developed on the basis of the trichome count per 0.1

cm-2, from the abaxial leaf surface in three crosses involving pilose hairy and normal/sparse

hairy plants. The intermediate (H2h2) progeny in F1 was due to incomplete dominance. The

two homozygous extremes for trichomes/hairiness: pilose hairy (H2H2) and sparse/normal

hairy (h2h2) were easily distinguishable (Fig. 4.13). The F2 data regarding number of

trichomes were categorized into three main classes: pilose hairy (H2H2), sparse/normal hairy

(h2h2) and intermediate hairy (H2h2) (Simpson, 1947). The major gene is designated as H1 for

sparse hairing. A second major gene, H2 controls the finely dense pubescence in an upland

mutant designated as ‘Pilose’. In the F1 populations, both H1 and H2 show incomplete

dominance (Knight, 1952; Niles, 1980). But Lee (1985) reported the inheritance of leaves

and identified five loci different loci from t1 to t5 denoting the trichomes. According to this

revision in trichome genetics, t1 locus was assigned to previously using H2. Kloth (1995)

proposed that the gene T1 imparted dense pubescence on leaves and stems, and places hairs

on the capsule. The gene T2arm reduced hairs to the margins of leaves (glabrous plant type).

The expressivity of H2 in different genetic backgrounds was studied by Simpson (1947) and

Rahman and Khan (1998), further supported the authenticity of the present studies on

trichome inheritance.The segregation in the backcrosses with parent-I and parent-II also fit to

the theoretical ratio of 1:1 which further confirmed the incomplete dominance pattern of

inheritance. The non-significant χ2 in F2 for trichomes in all the crosses fit well against the

theoretical monohybrid ratio of 1:2:1. This ratio is in proximity to the understanding made

from the Fig. 4.13.

122

Generally increasing gland density in cotton plant results in increasing concentration

of the toxic compounds. The principal determinants of gland density are Gl1, Gl2 and Gl3

alleles. Gl1 is responsible for gland formation only in stem, petioles and carpel walls,

whereas the Gl2 and Gl3 affect gland formation in cotyledons and leaves, as well as the

organs affected by Gl1. In other words it can be said that Gl2 and Gl3 mask the effect of Gl1

(Niles 1980 a). Gossypol glands are also one of the important insect resistant traits studied

during the present studies. Two crosses, normal glanding × glandless and normal glanding

× high glanding were made to obtain the F1 progeny which showed incomplete dominance.

The F1 of normal glanding × glandless (HRVO-1 × Acala 63-74) cross was intermediate

glandless while the F1 of normal glanding × high glanding (HRVO-1 × HG-142) cross was

intermediate high glanding. The same results were achieved by Calhoun, 1997 in the F1

generation of the upland cotton. The Fig. 4.14 and 4.15 showing the segregating pattern in F2

for glanding trait into three types which depicts the genotypic ratio (1 : 2 : 1) in the two

cotton crosses.

In the glandless parent (Acala 63-74) and F1, gossypol content of 0.04 mg/1g and

0.140 mg/1g (Table 4.14) was recorded respectively. Similarly, the total gossypol (%) in the

parent (Acala 63-74) and F1 was recorded as 2% and 5%, respectively as explained from the

Table 4.15. The mean gossypol yields from the studies of Lee (1973), who while crossing a

direct normal glanding parent (Gl2Gl2gl3gl3) to four glandless parents (gl2gl2gl3gl3), yielded

the gossypol level ranging from 0.068 mg to 0.320 mg in F1 and in the reciprocal

arrangement with four normal glanding parents, the gossypol level ranged from 0.064 mg to

0.253 mg in F1. In the cross of glandless with four glandless parents, the gossypol level

ranged from 0.004 mg to 0.014 mg in F1. He termed the gossypol yields ranging from 0.004

mg to 0.320 mg as glandless. Mansour et al. (2004), examined the gossypol content in

cultivars in relation to the bollworm infestation. The range of gossypol content determined

was 20-25 mg/100gram (0.20-0.25 mg/1gram), which was considered low in relation to the

non-significant association with bollworm incidence.The F2 data regarding the glanding trait

inheritance in HRVO-1 × Acala 63-74 cross was categorized into three main classes, normal

glanding (Gl2Gl2gl3gl3): intermediate glandless (Gl2gl2gl3gl3): glandless (gl2gl2gl3gl3), on the

basis of the quantification for glanding trait made through the spectrophotometric method by

123

A.O.C.S. (Reapproved 1989). The significant differences found between the parents and their

F1 justified the distinctness of three classes (Tables 4.14 – 4.15). But the studies of Calhoun

(1997) categorized two main classes of glandless and normal glanding in the F2 of the cross

of normal glanding and glandless. His studies were based on visual classification. The

appearance of intermediate glandlessness in F1 of the cross HRVO-1 × Acala 63-74 is an

indication that the single dose of Gl2 was inadequate for gland production on bolls

(Calhoun, 1997).

In the cross HRVO-1 × HG-142 (normal glanding × high glanding), three genotypic

classes (Fig. 4.14 and Fig. 4.15) were observed in F2. The data regarding the glanding trait

inheritance in HRVO-1 × HG-142 cross was categorized into three main classes high

glanding, intermediate high glanding and normal glanding (Gl2Gl2gl3gl3) plants on the basis

of the quantification for glanding trait made through the spectrophotometric method by

A.O.C.S. (Reapproved 1989). The significant differences between the parents and their F1

justified the distinctness of the three classes (Tables 4.14 – 4.15). But the studies of Calhoun

(1997) which were based on visual classification, categorized two main classes of high

glanding and normal glanding in F2 of the cross of normal glanding and high glanding

parents. The non-significant χ2 in F2 for the inheritance of glanding trait in these two crosses

fit well against the theoretical expected ratio of 1:2:1. This pattern of segregation in these two

crosses was further confirmed from the non-significant χ2 values obtained in the backcrosses

with parent-I & II which meant that the observed values fit well against the theoretical

expected ratios of 1:1. Observations of 1 high glanding: 2 intermediate high glanding: 1

normal glanding in F2 from the Table 4.20 in the cross HRVO-1 × HG-142 indicated

incomplete dominance.

According to Lee (1962), Gl2 and Gl3 are the major loci regulating gland production.

Calhoun (1997) explained that high glanding was controlled primarily by a single locus but

the expression was affected by interaction with recessive (gl) alleles at Gl2 or Gl3 loci or

both. The segregation pattern in F2 according to the findings of Calhoun (1997) helped in

further genotyping. The two homozygous extremes were: normal glanding (Gl2Gl2gl3gl3) and

glandless (gl2gl2gl3gl3) in the cross HRVO-1 × Acala 63-74 and normal glanding

(Gl2Gl2gl3gl3) and high glanding (Gl2Gl2Gl3Gl3) in the cross HRVO-1 × HG-142 were easily

124

distinguishable. Studies by Wilson and Lee (1971), Lee (1971) and Niles (1980 a) showed

that seedling damage was least and number of larvae were lowest on plants of genotypes

Gl2Gl2Gl3Gl3, intermediate on Gl2Gl2gl3gl3 and gl2gl2Gl3Gl3, and highest on gl2gl2gl3gl3).

Though these studies were conducted on the basis of seedling damage, yet these findings

confirmed the behaviour of the genotypes proposed in terms of the quantity of the gossypol

glands.

From the data reported here, it is clear that the glanding traits (high glanding and

glandless) are simply inherited traits and in the crosses between normal glanding and high

glanding and between normal glanding and glandless, can be selected in much the same way

as any other incompletely dominant allele behaves. However, the environment as well as

minor genes affect the degree of expression of the glanding trait (White et al., 1982).

5.5. Estimation of heritability and genetic advance for various plant traits

Partitioning of the phenotypic variance into its genotypic and environmental components is

not enough to look into deeply the properties of the breeding material. The genotypic

variance needs further partitioning into additive and dominance variances. Environmental

variance which is due to non-genetic causes is also involved in the final expression of a trait.

According to Falconer and Mackay (1996) the estimates of heritability are subject to

considerable environmental variation, and considerable caution is necessary in their

interpretation and use. A general adequate fitness of the two parameter model (D) and (E) to

the generation variances reflects that additive variance appeared to account for largest

proportion of the total genetic variance for all the traits (Table 4.17). From the estimates of

various components of variance, heritability in narrow sense, heritability in infinity

generation and genetic advance was worked out.

Overall, high to low estimates of narrow sense heritability were observed for various

traits. The highest estimates of narrow sense heritability for most of the traits in different

crosses indicated that a large proportion of the genetic variance was composed of the additive

genetic component and selection for improvement of such characters would be rewarding.

Infinity generation heritability was consistently higher than the narrow sense heritability

estimates. Low estimates of narrow sense heritability as noticed in some traits showed the

125

preponderance of non-additive gene action and could be utilized in heterosis breeding. The

environmental variance estimated in the traits with moderate to low heritability clearly

showed its role. For plant height, number of bolls, seed cotton yield, boll weight and lint

percentage high estimates of narrow sense heritability were observed by Singh and Singh

(1981), Singh (1982), Gupta (1987), Pandey and Singh (2002) and Nadeem and Azhar

(2005). Findings of these researchers supported the present investigations. Very high

estimates of heritability in narrow sense were noticed in the number of trichomes, gossypol

content and total gossypol in different crosses. These high estimates pointed towards their

additive mode of inheritance. Fibre length, fibre strength and fibre fineness are important

traits from fibre quality point of view. The present studies showed normally medium to high

narrow sense heritability estimates for these traits which supported the findings of

Desphande et al. (1984), Vyahalkar et al. (1984), Nadarajan and Rangasamy (1990), Pandey

and Singh (2002) Nadeem and Azhar (2004), Nadeem and Azhar (2005) and Ulloa (2006).

High estimates of genetic advance as well as high narrow sense heritability for

number of trichomes in all the three crosses; for seed cotton yield in the crosses HRVO-1

× FH 1000 and HRVO-1 × Acala 63-74 (Kumari and Chamundeswari, 2005), in the cross,

HRVO-1 × CIM 446 for plant height and in fibre uniformity ratio in HRVO-1 × FH 1000,

were most likely due to additive gene effects and selection may be effective to carry on the

breeding improvement. Moderate estimates of genetic advance with high narrow sense

heritability estimates as noted for plant height in the cross HRVO-1 × FH 1000 and for

number of sympodial branches and number of bolls (Kumari and Chamundeswari, 2005) in

the crosses HRVO-1 × FH 1000 and HRVO-1 × CIM 446 indicated the preponderance of

both additive and non-additive type of gene effects. However, it is not necessary that a

character showing high heritability will exhibit high genetic advance (Johnson et al., 1955).

In most of the traits high narrow sense heritability estimates coupled with low genetic

advance were recorded which indicated that the high heritability might be due to non-

additive gene effects and the favouable environmental influence rather than genotype.

Selection for such traits may not be rewarding. For boll weight in the cross HRVO-1

× FH 1000 and in all the three crosses for number of monopodial branches, low narrow sense

heritability with low genetic advance indicated that these traits are highly influenced by the

126

environment variance as the Table 4.17 clearly shows high values for the environmental

component of variance.

5.6. Estimation of heterosis and inbreeding depression for various plant traits

Heterosis in F1 and inbreeding depression in F2 considered together are helpful in giving the

idea about the genetic control of a character. These two phenomena are important in cotton

although less pronounced and less consistent in occurrence than in cross pollinated species of

plants. Significant and positive heterosis and inbreeding depression recorded for the trait boll

weight in the different crosses indicated the presence of non-additive gene effects and

suggested the full potential of heterosis for boll weight could be exploited in the first

generation hybrids (Singh and Narayanan, 2000). These results supported the findings of

Hassan et al. (1999), Kamaresan et al. (1999), Bertini et al. (2001), Qian et al. (2001) and

Nadeem and Azhar (2004). Monopodial branches are non-fruiting branches, so plants with

reduced number of monopodial branches are considered to be desirable. For number of

monopodial branches, the highly significant positive heterosis and highly significant negative

inbreeding depression in HRVO-1 × CIM 446 which meant a highly significant negative

heterosis and highly significant positive inbreeding depression indicated that the parents are

not best combiners for this trait. A significant though negative heterosis with non-significant

inbreeding depression in the cross HRVO-1 × Acala 63-74, for the same trait was noted

which meant a positive heterosis suggested the retentive nature in advance generations.

While selecting for fibre fineness one has to be careful as fineness is expressed in micronaire

value. Higher the micronaire value the coarser the fibre and vice-versa. For fibre fineness in

the cross HRVO-1 × FH 1000 a significant but negative heterotic effect (Arshad et al., 2001)

with non-significant inbreeding depression indicated that the parents in these crosses were

good combiners and heterosis in F1 was of retentive nature and could be utilized in later

generations. In the cross of HRVO-1 and CIM 446 for the same trait a negative and highly

significant heterotic estimate with significantly negative inbreeding depression was an

indicative that the parents in this cross were good combiners and heterosis could be best

utilize in F1 generation only whereas, in HRVO-1 × Acala 63-74 a significant and positive

inbreeding depression suggested the possibility of selection of better hybrids in F1. The

findings of Kamaresan et al. (1999), Rajan et al. (2000), Manimaran and Raveendran (2002)

127

and Nadeem and Azhar (2004) showed significant heterosis for plant height, number of bolls,

boll weight and seed cotton yield. High estimates of heterosis for seed cotton yield were

found by Hassan et al. (1999), Kaynak et al. (2000), Kowsalya et al. (2000),

Banumathy et al. (2001), Zhang et al. (2003) and Desphande and Baig (2004).

For number of trichomes a negative and highly significant inbreeding depression in

HRVO-1 × FH 1000 suggested the possibility of transgressive segregants in the F2 generation

and selection of better F2 hybrids would be fruitful in delayed selection for this insect non-

preference trait. For gossypol content and total gossypol in HRVO-1 × Acala 63-74, a highly

significant though negative heterosis with negative inbreeding depression suggested the

possibility of transgressive segregants in the F2 generation and selection of better hybrids

would be fruitful in delayed selection.

For fibre length, fibre strength, fibre elongation and lint percentage high and

significant heterotic effects were studied by Soomro et al. (2000), Arshad et al. (2001),

Bertini et al. (2001), Qian et al. (2001), Feki and Gelil (2001), Manimaran and Raveendran

(2002), Baloch (2003) and Zhang et al. (2003).

5.7. Correlations

Correlation coefficient is a statistical measure which is used to find out the degree and

direction of the relationship between two or more variables. A positive value gives the

indication of the same direction of the two variables in question and vice-versa. Whereas, the

negative value shows movement of the variables in opposite directions. Correlation in plant

breeding is a useful tool of indirect selection of the secondary trait with the improvement in

the primary trait. The objective of the present study was to find correlation among insect

resistant, fibre, other morphological and yield related traits in cotton.

The fibre fineness is recorded in micronaire value. This means that higher the

magnitude of micronaire value, lesser will be the fineness of the fibre and vice- versa. In case

of leaf type, the data generated for the expression of this trait was on visual ranking system.

The maximum value was assigned to the okra leaf, while the minimum to the normal leaf

morphologies. Positive correlation value for leaf type, indicated the okra leaf type while the

128

negative correlation value indicated the normal leaf morphology. The correlation among

various plant traits are shown in Table 4.23 to 4.28. In general the magnitude of the genetic

correlations was higher than that of the correlations at the phenotypic level in all the three

crosses. This indicated genetic correlation between the two characters. Similar type of high

magnitude for genetic correlations was observed by Dhanda et al. (1984) and Tyagi (1987).

The higher values of phenotypic correlation coefficients than the genotypic correlation

coefficients where ever present, indicated that the correlation between the two characters was

not only due to genes but environment also played its role in the expression of the character.

Significant correlation between the two characters where ever present gave dependency of

the two characters, while non-significant correlation where ever present indicated the

independent nature of the two characters under study as suggested by Singh and Narayanan

(2000).

5.7.1 Correlation among insect resistant and fibre traits

(a) Leaf type

In the present study the non-significant correlation of leaf type with all fibre traits (fibre

length, fibre strength, fibre uniformity ratio, fibre fineness and lint percentage) except for

fibre elongation revealed no correlation of okra leaf type with the expression of these traits as

shown in Table 4.23. The present results are in agreement to the findings of Andries et al.

(1969), Wells and Meredith Jr. (1986), Thomson et al. (1987) and Percy (2001). In all the

three crosses, the significant values showed correlation between fibre elongation and leaf

type. The positive and significant correlation of leaf type with fibre elongation in the cross

HRVO-1 × Acala 63-74, meant that okra leaf type would result in enhanced fibre elongation.

In the other two crosses the negative correlation gave the understanding of negative

correlation between these two traits which meant that the decrease in the fibre elongation was

correlated with the okra leaf morphology. These results of negative correlation between the

two traits were in agreement to the findings of Andries et al. (1969), Thomson et al. (1987)

and Meredith et al. (1996).

(b) Trichomes/Hairiness

The positive correlation of number of trichomes with lint percentage in all of three crosses

under study showed that the selection for pilose hairiness would be helpful in enhancing the

129

lint percentage. The negative correlation of fibre length, fibre strength, fibre fineness, fibre

elongation and fibre uniformity ratio with number of trichomes in the crosses under study

corroborated to the findings of Simpson, 1947; Lee, 1964 and Lee 1984. But Kloth (1993) in

his study discovered a pilose like plant with unexpectedly low micronaire value (high fibre

fineness) among the homozygous pilose plants.

(c) Gossypol content and total gossypol

The non-significant correlations obtained in the cross HRVO-1 × Acala 63-74 indicated no

effect of gossypol content or total gossypol on the fibre quality attributes. The results

obtained herein got a strong support from the findings of Phogat et al. (2000) and Yuan et al.

(2000).

5.7.2 Correlation among fibre quality traits

Negative and significant correlation between fibre length and lint percentage in the crosses

HRVO-1 × FH 1000 and HRVO-1 × CIM 446, indicated that improvement in one trait led to

a proportionate decrease in the other trait. Similar results were obtained in the studies by

Ulloa and Meredith Jr. (2002). Positive correlation of lint percentage with fibre strength in

the crosses HRVO-1 × FH 1000 and HRVO-1 × CIM 446, indicated that improvement in one

trait led to a simultaneous improving in the other trait and vice- versa. These results got the

support from the findings of Badr and Aziz (2000) and Singh et al. (2002). Positive

correlation of lint percentage with micronaire meant a negative correlation between lint

percentage and fibre fineness in the three crosses as observed by Ulloa and Meredith Jr. (200-

2). Positive and significant correlation of fibre fineness with fibre uniformity ratio in the

crosses, HRVO-1 × CIM 446 and HRVO-1 × Acala 63-74 were in agreement to the findings

of Ulloa (2006). Negative correlation of fibre length with micronaire meant positive

correlation of fibre length with fibre fineness in all the three crosses corroborated the findings

of Badr and Aziz (2000) and Singh et al. (2002). Whereas, a positive correlation between

fibre length and fibre strength in different crosses got the support from Badr and Aziz (2000),

Singh et al. (2002), Ying and Jun (2004), Herring et al. (2004) and Ulloa (2006).

130

5.7.3 Correlation among other morphological and yield related traits

Positive and significant correlation of plant height with number of sympodial branches,

number of bolls, seed cotton yield and boll weight in the crosses HRVO-1 × FH 1000 and

HRVO-1 × CIM 446 indicated that increase in plant height results in the increase in the other

traits. Similar type of results were inferred from the findings of Arshad et al. (1993), Naveed

et al. (2004 a) and Echekwu (2001). Negative correlation of monopodial branches with

number of bolls and boll weight suggested that an increase in monopodia reduced these two

yield attributing traits in the cross HRVO-1 × Acala 63-74. The positive and significant

correlation of number of sympodial branches with seed cotton yield, number of bolls and boll

weight in all the three crosses suggested that an increase in number of sympodial branches

produce more number of bolls with an average increase in the boll weight, resulting

ultimately increase seed cotton yield and vice-versa. Positive correlation between number of

bolls with seed cotton yield and boll weight in all the three crosses was observed. Similarly,

positive correlation of seed cotton yield with boll weight in all the crosses indicated that seed

cotton yield (Azhar et al., 1984) is highly dependent upon number of sympodial branches,

number of bolls and boll weight. So, these traits could be selected independently as selection

criteria in breeding programme with the ultimate goal of high yield (Singh et al., 1968;

Konoplya et al., 1979).

5.7.4 Correlation among other morphological, yield and insect resistant traits

In cotton the insect resistance is associated with various morphological (Jayaraj and

Murgesan, 1988; Jenkins, 1989 and Watson, 1989) and biochemical traits evaluated and

reported by Singh and Agarwal (1988); Hedin and McCarty (1990). Of these traits, okra leaf

type, trichomes and gossypol glands hold a special position in insect pest resistance

perspective. Okra leaf trait appeared to have pink bollworm resistance as it is evident from

the findings of various researchers, Wilson and George (1982), Wilson (1987) and

Wilson et al. (1991). The findings of some other scientists like Kalifa and Gameel (1983) and

Bhatangar and Sharma (1991) indicate that the okra leaf genotypes were found to be less

infected by whitefly, thrips, aphids and jassids. Number of sympodial branches, number of

bolls, boll weight and seed cotton yield are important yield determining traits in cotton.

These traits had a positive correlation with leaf type which means that okra leaf type is

131

positively correlated to these yield related traits and thus results in an increased yield. These

types of results have been well documented in the literature (Wells and Meredith, 1986;

Meredith and Wells, 1986; Thomson et al., 1987; Jones et al., 1988; Wilson, 1989;

Rahman et al., 2005). In addition to increased yield, associated with okra leaf shape,

Andries et al. (1969) found a significant reduction in the incidence of boll rot with increase

in earliness in comparison with the normal leaf shape (Pettigrew, 2003). The reason for this

increased yield may be attributed to the reduced leaf area per plant in the okra leaf with open

plant canopy in okra leaf plots which allowed better air flow and more sunlight to penetrate

to the lower plant zones. These factors may increase the photosynthetic efficiency, resulting

in increased yields. Ulloa (2006) has also advocated the genetic potential for improvement in

agronomic traits in the populations with the okra leaf morphology.

Non-significant correlations among the agronomic/yield related traits and number of

trichomes/hairiness revealed that hairiness or number of trichomes had no effect on the

expression of these agronomic/yield realted traits. These results match with the findings of

Lee (1984).

Similarly in case of gossypol content and total gossypol, the non-significant

correlation of these traits with morphological/yield related traits like plant height, number of

monopodial branches, number of sympodial branches and number of bolls indicated

independence of these traits in the cross HRVO-1 × Acala 63-74. Boll weight was negatively

correlated with gossypol content and total gossypol which reveals that an increase in boll

weight is related to a decrease in gossypol content or total gossypol. This in another way can

be interpreted that glandlessness is associated positively with bollweight (Soomro, 2000).

The present results apparently pointed towards significant and positive correlation of seed

cotton yield with gossypol content and total gossypol. An increase in the gossypol content or

total gossypol results into increase seed cotton yield and vice-versa. Althought, the literature

does not support this observation. I could not happen to see any supporting literature to the

present observations. However, the increased yield could be due to high number of gossypol

glands present resulting into a decreased insect/pest activity as stated by Wilson and Lee

(1971), Lee (1971), Niles (1980) and Mohan et al. (1994).

132

Conclusions:

The study was aimed to reveal the inheritance pattern in different genetic backgrounds for

okra leaf type, gossypol glands and trichomes. Incorporation of the insect non-preference is

one aspect of the proposed study, while the other was focused on the improvement of yield

and fibre quality traits. The transference of the genes conferring insect non-preference traits

is easy and under the control of dominant genes.

Non-significant correlation of okra leaf type was observed with fibre length, fibre

strength, fibre uniformity ratio, fibre fineness and lint percentage. Whereas, fibre elongation,

number of sympodial branches, number of bolls, boll weight and seed cotton yield revealed

significant correlation with this trait, which clearly shows that introduction of the gene for

okra leaf morphology would not alter the fibre traits but would add to the insect resistance

coupled with improvement in the agronomic traits.

Positive correlation of number of trichomes with lint percentage in all of three crosses

was observed, which gives the insight for increase in the lint percentage. However, the

negative correlation of fibre length, fibre strength, fibre fineness, fibre elongation and fibre

uniformity ratio with number of trichomes, shows the negative impact on these fibre quality

attributes. At the same time, there is a possibility to explore for the better combinant in a

large segregating generation. However, morphological/yield related traits had a non-

significant correlation with trichomes/hairiness, which reveals that the introduction of

hairiness would not hamper the agronomic traits of economic importance.

Non-significant correlations of gossypol content and total gossypol, with the fibre and

morphological/yield related traits like plant height, number of monopodial branches, number

of sympodial branches and number of bolls which shows independency of these traits with

gossypol content and total gossypol. However, boll weight was negatively correlated with

gossypol content and total gossypol, which further suggests exploring for the better

combination in a large segregating population. Significant and positive association of seed

cotton yield with gossypol content and total gossypol gives the justification that an increase

in the gossypol content or total gossypol is the result of the increased seed cotton yield and

133

vice-versa. However, the increased yield could be due to high number of gossypol glands

present resulting into a decreased insect/pest activity.

The high yielding genotypes with good fibre quality attributes like FH 1000 and CIM

446 were used in the present research. The results obtained so far are encouraging, which

indicated that the traits like okra leaf type, trichomes and gossypol glands can be easily

introduced into any of the commercial lines/varieties through conventional breeding. The

single plants selected in the segregating generations can be further progressed through

pedigree/bulk methods for varietal development. Large population is required in F3 or in F4 to

search for the better combinants keeping in view of the agronomic and insect resistant traits.

On the same side the F1 developed in the four cross combinations could also be progressed

through convergent/multiple crossing.

134

CHAPTER 6

SUMMARY

The present studies were undertaken in the Department of Plant Breeding and

Genetics, University of Agriculture, Faisalabad (Pakistan), in order to study the inheritance

of okra leaf type, gossypol glands and trichomes along with other morphological and fibre

related characters. Thirty-one varieties/genotypes were collected from different sources and

assessed for okra leaf type, gossypol glands and trichomes/hairiness. Five parents were

selected, including one parent (HRVO-1) with okra leaf type and pilose hairiness (trichomes)

traits common to make four cross combinations. The F1, F2 and backcrosses both with the

parents I and II were developed and studied along with the parents. These basic six

generations in each of the four crosses were evaluated in a randomized complete block

design with three replications. Means and variances of each cross combination were

calculated and analyzed separately to estimate heterosis, inbreeding depression, genetic

advance and narrow sense heritability for morphological, fibre and insect resistant traits. The

nature and magnitude of genetic effects involved in the inheritance of these traits were

determined. The nature and extent of correlation between various morphological, fibre and

insect resistant traits were also determined in the F2 generation of each cross.

Ordinary analysis of variance was applied separately for each cross to determine the

significance of the generation effect on various morphological, fibre and insect resistant

traits. The knowledge about the nature and magnitude of genetic effects prevailing in the

breeding material is necessary to decide the kind of breeding procedure to be followed.

Generation mean analysis indicated all three kinds of gene effects (additive, dominance and

their interactions) involved in the inheritance of different traits. In plant height four-

parameter model (m, d, h and i) was adequate for the crosses HRVO-1 × FH 1000 and

HRVO-1 × CIM 446 whereas, in the cross HRVO-1 × Acala 63-74 four-parameter model

(m, h, i and l) was found adequate. None of the genetic effects and interaction component

appeared to be involved in the expression of number of monopodial branches in the cross

HRVO-1 × CIM 446, while in the other two crosses i.e; HRVO-1 × FH 1000 and

135

HRVO-1 × Acala 63-74 a similar trend of adequacy in the form of three-parameter model

(m, d and i) was observed for the same trait. A three-parameter model (m, d and h) was best

fit from the observed to the expected estimated generation means for number of sympodial

branches in all the three crosses under study. In case of number of bolls, four-parameter

model involving m, d, h and i was adequate for the crosses, HRVO-1 × CIM 446 and

HRVO-1 × Acala 63-74 while a five-parameter model m, d, h, i and j showed the best fitness

in the cross of HRVO-1 and FH 1000. Adequacy of four-parameter model m, d, h and i was

observed for seed cotton yield in the crosses HRVO-1 × FH 1000 and HRVO-1

× Acala 63-74. However, a simple additive-dominance model (three-parameter model m, d

and h) was best for seed cotton yield in the cross of HRVO-1 and CIM 446. For boll weight,

four-parameter model (m, d, h and i) showed best fitness for crosses HRVO-1 × FH 1000 and

HRVO-1 × Acala 63-74 whereas, m, d and h showed its fitness for the same trait in the cross

HRVO-1 and CIM 446. Four-parameter model (m, d, j and l) was fit in the cross HRVO-1

× FH 1000 whereas, m, d, h and i was found fit in the cross HRVO-1 × CIM 446 in case of

lint percentage. However, in the cross HRVO-1 × Acala 63-74 the simplest model of two

parameters i.e; m and d was found to be operative for the same trait. Fibre length indicated

best fitness of three-parameter model (m, d and h) in the cross HRVO-1 × FH 1000, while a

four-parameter model (m, d, h and i) in the cross HRVO-1 × CIM 446 and two-parameter

model i.e; m and h in case of HRVO-1 × Acala 63-74. Similarly, for fibre strength the same

three-parameter model (m, d and h) was operative in the crosses HRVO-1 × FH 1000 and

HRVO-1 × CIM 446 whereas, in case of HRVO-1 × Acala 63-74, two parameter model i.e;

m and d observed to be fit. In fibre elongation, three-parameter model (m, d and h) was found

adequate in the crosses HRVO-1 × FH 1000 and HRVO-1 × CIM 446 whereas, in the cross

HRVO-1 × Acala 63-74, four-parameter model (m, d, h and i) showed its adequacy in the

cross HRVO-1 × Acala 63-74. A four-parameter model (m, d, h and i) was adequate for fibre

uniformity ratio, in the crosses HRVO-1 × CIM 446 and HRVO-1 × Acala 63-74 whereas, a

four-parameter model (m, d, i and j) showed its adequacy in the cross HRVO-1 × FH 1000.

The four-parameter model (m, d, h and i) for fibre fineness was found adequate in the crosses

HRVO-1 × FH 1000 and HRVO-1 × Acala 63-74 while a three-parameter model (m, d and h)

proved to be best fit in the cross HRVO-1 × CIM 446. For the number of trichomes, a similar

pattern of genetic effects was observed in all the three crosses under study. The three-

136

parameter model (m, d and h) observed to be best fit in terms of the observed to the expected

generation means, showing adequacy for this model in all the three crosses. The best fit

model for gossypol content observed in the cross HRVO-1 × Acala 63-74 was three-

parameter model with m, d and h effects whereas, in HRVO-1 × HG-142 a simplest model

comprising of two-parameters (m and d) showed its adequacy for the observed to the

expected mean values. In case of total gossypol a similar trend of genetic effects was

followed as were observed for the gossypol content.

In the generation variance analyses, a model incorporating the additive (D) and

environmental (E) components was sufficient to explain the variation in cotton crosses for all

the traits except for lint percentage in cross, HRVO-1 × FH 1000, where the additive (D),

additive × dominance (F) and environmental (E) model appeared to show its best fitness.

The inheritance studies of three important insect resistant traits, namely, okra leaf

type, trichomes and gossypol glands indicated the ratio of 1 normal : 2 sub-okra :1 okra leaf

types in the F2 populations of the three crosses. In the backcrosses with parent-I, ratios of

1 okra : 1 sub-okra leaf types were obtained. Similarly, in the backcrosses with parent-II,

ratios of 1 normal : 1 sub-okra leaf types were observed. Trichomes/hairiness segregated into

three classes in F2. Observations of 1 sparse hairiness : 2 intermediate class of hairiness : 1

pilose hairiness on leaves were observed in the F2 populations of the three crosses. In the

backcrosses with parent-I, ratios of 1 pilose : 1 intermediate leaf hairiness were obtained.

Similarly, in the backcrosses with parent-II, ratios of 1 sparse : 1 intermediate leaf hairiness

were observed. The inheritance of gossypol glands was studied in two cross combinations i.e;

HRVO-1 × Acala 63-74 and HRVO-1 × HG-142. Observations of 1:2:1 in F2 and 1:1 in the

backcross/testcross were observed in both the crosses.

In general, a high magnitude of narrow sense heritability was noticed in all the three

crosses for the trait plant height. In case of number of monopodial branches the heritability

estimates were found inconsistent. Moderate to high estimates of heritability were recorded

for number of sympodial branches. Moderate heritability estimates of 0.57 in the cross

HRVO-1 × Acala 63-74, while high estimates of 0.75 in HRVO-1 × FH 1000 and 0.84 in the

cross HRVO-1 × CIM 446. The narrow sense heritability estimates in three crosses for

137

number of bolls ranged from 0.73 to 0.86. High estimates were noticed for seed cotton yield

in three crosses ranging from 0.78 to 0.86. High values of heritability (0.78 and 0.92)

estimates for boll weight in two crosses i.e; HRVO-1 × Acala 63-74 and HRVO-1

× CIM 446 respectively, while lowest value of 0.24 for the same trait was obtained in the

cross HRVO-1 × FH 1000. A low heritability (0.37) for lint percentage in HRVO-1 × FH

1000 while high estimates (0.70) for the same trait in HRVO-1 × CIM 446 were observed.

The same trend was observed for fibre length where the estimates of heritability ranged from

0.67 to 0.90. In the case of fibre strength, high estimates were observed in all the three

crosses. For fibre elongation and fibre uniformity ratio an increased trend was observed in

the estimates ranging from 0.68 to 0.79 and 0.69 to 0.91 respectively. In case of fibre

fineness moderate to high estimates were obtained in the three crosses. For number of

trichomes, very high were observed in all the three crosses ranging from 0.88 to 0.91. For the

two traits i.e; gossypol content and total gossypol, the heritability estimates of two crosses

viz, HRVO-1 × Acala 63-74 and HRVO-1 × HG-142 followed the same pattern as were

observed for number of trichomes. The infinity generation (F) heritability estimates were

consistently higher than those recorded in F2 generation for most of the traits under study.

High estimates of genetic advance were obtained for number of trichomes in the three

crosses i.e; HRVO-1 × FH 1000, HRVO-1 × CIM 446 and HRVO-1 × Acala 63-74. For seed

cotton yield high estimates of genetic advance were observed in crosses HRVO-1 × FH 1000

and HRVO-1 × Acala 63-74. For plant height, only in the cross, HRVO-1 × CIM 446 high

estimates were recorded. Similarly, in fibre uniformity ratio genetic advance was higher for

HRVO-1 × FH 1000. Moderate estimates of genetic advance were recorded for plant height

in HRVO-1 × FH 1000. Similarly, in the crosses HRVO-1 × FH 1000 and HRVO-1

× CIM 446, moderate estimates were recorded for number of sympodial branches and

number of bolls. For all other traits the estimates of genetic advance remained less which

ranged from 0.13 to 4.65.

Significant and positive heterosis and inbreeding depression recorded for the trait boll

weight in different crosses. For fibre fineness a significant but negative heterotic effects with

non-significant inbreeding depression values were recorded in all the three crosses.

138

No correlation of okra leaf type was observed with fibre length, fibre strength, fibre

uniformity ratio, fibre fineness and lint percentage whereas, fibre elongation, number of

sympodial branches, number of bolls, boll weight and seed cotton yield revealed significant

correlation with this trait which is a clear cut evidence that introduction of the gene for okra

leaf morphology will not alter the fibre traits but will add to the insect resistance coupled

with improvement in the agronomic traits. Positive correlation of number of trichomes with

lint percentage in all of three crosses was observed whereas; a negative correlation of fibre

length, fibre strength, fibre fineness, fibre elongation and fibre uniformity ratio with number

of trichomes in all the crosses was recorded. Morphological/yield related traits had a non-

significant correlation with trichomes/hairiness. For gossypol content and total gossypol,

there existed non-significant correlations with the fibre and morphological/yield related traits

like plant height, number of monopodial branches, number of sympodial branches and

number of bolls except for boll weight which was negatively correlated in HRVO-1 × Acala

63-74. Negative and significant correlation between fibre length and lint percentage in the

crosses HRVO-1 × FH 1000 and HRVO-1 × CIM 446 and positive correlation of lint

percentage with fibre strength in the crosses HRVO-1 × FH 1000 and HRVO-1 × CIM 446

were observed. Similarly, lint percentage was observed to be positively correlated with

micronaire in the three crosses. Fibre length was observed to be negatively correlated with

micronaire in all the three crosses.

Numbers of sympodial branches were significantly and positively correlated with

seed cotton yield, number of bolls and boll weight in all the three crosses. There also existed

a positive correlation between number of bolls, seed cotton yield and boll weight in all the

three crosses. Similarly, in all the crosses a positive correlation of seed cotton yield with boll

weight was observed.

139

140

LITERATURE CITED .

A. O. C. S. Official Method Ba 8-78 (Reapproved 1989).

Abou-Donia, S. A., S. A. Lasker and M. B. Abou-Donia. 1981. High-performance liquid

chromatography analysis of gossypol. J. Chromatogr. 206: 606–610.

Adl, A. M., Z. M. Diasty, A. A. Awad, A. M. Zeina and A. M. A. Bary. 2001. Inheritance of

quantitative traits of Egyptian cotton (G. barbadense L.). A-Yield and yield

component traits. Egyptian J. Agric. Res. 79(2): 625-646.

Agrawal, A. A. and R. Karban. 2000. Specificity of constitutive and induced resistance: pigment

glands influence mites and caterpillars on cotton plants. Entomologia Experimentalis

et Applicata 96: 39–49.

Ahmad, I., A. Ali, M. Zubair and I. A. Khan. 2001. Mode of gene action controlling seed cotton

yield and various components in Gossypium hirsutum L. Pak. J. Agric. Sci. 38: 3-4.

Ahmad, M. 2001. Agricultural productivity growth differential in Punjab, Pakistan: A district-level

analysis. Pakistan Development Review. 40(1): 1-25.

Ahmad, Z. and M. Ali. 1993. Study of cotton production prospects for next decade. The World

Bank, Washington, U.S.A.

Alam, A. K. M. A. and M. M. Hossain. 2008. Variability of different growth contributing parameters

of some okra (Abelmoschus esculentus L.) accessions and their interrelation effects on

yield. J.Agric. Rural Dev. 6(1&2): 25-35.

Andries, J. A., A. G. Douglas and R. C. Albritton. 1971. Performance of normal, okra, and superokra

leaf types in three row widths. p. 59-60. In J. M.Brown (ed.) Proc. Beltwide Cotton

Prod. Res. Conf., Atlanta, GA. 11-12 Jan., Natl. Cotton Council of Am., Memphis,

TN.

Andries, J. A., J. E. Jones, L. W. Sloane and J. G. Marshall. 1969. Effects of okra leaf shape on boll

rot, yield and other important characters of upland cotton, Gossypium hirsutum L. Crop

Sci. 9: 705-710.

Anonymous, 1992 a. Operational manual of HVI 4000 prepared by MCI, Peyer, USA.

Anonymous, 1992b. HVI test results and their interpretation, Peyer 04441. 08.92 (Switzerland).

141

Arif, M. J., M. D. Gogi and G. Ahmad. 2006. Arab. J. Pl. Prot. 24(1): 58-60.

Arshad, M., M. Hanif, N. Ilani and S. M. Shah. 1993. Correlation studies on some commercial

cotton varieties of Gossypium hirsutum L. Sarhad J. Agri. 9(1): 49-53.

Arshad, M., N. Illahi, M. Rashid and Z. U. Qamar. 2001. Estimation of useful heterosis in

interspecific F1 cotton hybrids. J.Agric.Res. 4(1): 65-67.

ASTM Committee, 1977a. Standard test for length and length uniformity of fibrograph

measurement. ASTM Designation D 1474-77. Amer. Soc. For Testing Materials,

Philadelphia. 34th Ed. Part 33.

ASTM Committee, 1977b. Tentative method of test for strength of cotton fibre. ASTM Designation

D 1445-75. Amer. Soc. For Testing Materials, Philadelphia. 34th Ed. Part 33.

ASTM Committee, 1977c. Standard method of test for micronaire reading of cotton fibres. ASTM

Designation D 1440-44. Amer. Soc. For Testing Materials, Philadelphia. 34th Ed. Part

33.

Aston, J. S. and B. Winfield, 1972. Insect section introduction. In: Breeding plants resistant to

insects. F.G. Maxwell and P. R. Jennings., John Wiley and Sons, USA, pp: 337-370.

Attique, M. R. and A. Rashid. 1983. Efficacy of pyrethroid pesticides for control of cotton pests.

J.Agric.Res. 4(1): 65-67.

Auld, D. L., R. G. Cantrell, E. Bechere, E. F. Hequet, P. Dugger and D. Ritcher. 2000. Inheritance of

fibre length in two mutants of cotton. Proceedings Beltwide Cotton Conferences, San

Antonio, USA. 4-8 Jan. 2000. 1: 518-520.

Avgoulas, C., L. Bouza, A. Koutrou, S. Papadopoulou, S. Kosmas, E. Makridou, P. Papastylianou

and D. Bilalis. 2005. Evaluation of five most commonly grown cultivars (Gossypium

hirsutum L.) under Mediterranean conditions: productivity and fibre quality. J. Agron.

Crop. Sci. 191(1): 1-9.

Azhar, F. M., A. R. Abid and M. A. Khan. 1984. Association of yield with various economic

characters in Gossypium hirsutum L. The Pak. Cottons 28(2): 127-134.

Azhar, F. M., M. Naveed and A. Ali. 2004. Correlation analysis of seed cotton yield with fibre

characteristics in Gossypium hirsutum L. International J. Agri. Biol. 6(4): 656-658.

Babar, M. and I. A. Khan. 1999. Genetic analysis of some agronomic and fibre characters in upland

cotton (Gossypium hirsutum L.). Pak. J. Biol. Sci. 2(4): 1484- 1487.

142

Badr, S. S. M. and M. Aziz. 2000. Comparative study of fibre properties and yield of five new

Egyptian cotton cultivars. J. Agric. Res. 78(1): 279-291.

Bahtnagar, P. and P. D. Sharma, 1991. Comparative incidence of sucking insect pests on different

isogenic lines of cotton variety H-777. J. Insect Sci. 4(2): 170-171.

Baloch, M. J. 2002. Relative performance of F1 and F2 intrahirsutum hybrids for some quantitative

traits in upland cotton. Pak. J. Scientific and Industrial Res. 45(6): 407-411.

Baloch, M. J. 2003. Genetic distance and its relation to the performance of intrahirsutum F1 hybrids.

Pak. J. Scientific and Industrial Res. 46(1): 63-67.

Banumathy, S. and S. Patil. 2001. Heterosis in cotton. Journal of Soils and Crops. 11(1): 28-32.

Bartlett, B. R .1978. Pseudococcidae. In: Introduced Parasites and Predators of Arthropod Pests and

Weeds: a World Review (Ed. Clausen CP), Pp. 137–170. Agriculture Handbook No.

480. USDA, Washington (US).

Barton, F. E., J. D. Bargeron, G. R. Gamble, D. L. McAlister and E. Hequet. 2005. Analysis of

Sticky Cotton by Near-Infrared Spectroscopy. Applied Spectroscopy. 59(11): 1388-

1392.

Bashir, M. H., M. Afzal, M. A. Sabri and A. B. M. Raza. 2001. Relationship between sucking insect

pests and physico-morphic plant characters towards resistance/ susceptibility in some

new genotypes of cotton. Pak.Entomol. 23(1-2): 75-78.

Bell, T. M. and F. E. M. Gillham. 1989. The World of Cotton. In: M.Sc. Thesis, Faculty of the Univ.

Georgia, Athens, Georgia.

Benbouza, H., G. Lognay, R. Palm, J. P. Baudoin and G. Mergeai. 2002. Development of a visual

method to quantify the gossypol content in cotton seeds. Crop Sci. 42: 1937-1942.

Bertini, C. H. C - de- M., F. P-da- Silva, R- de- P. Nunes and J. H. R- dos- Santos. 2001. Gene

action, heterosis and inbreeding depression of yield characters in mutant lines of

upland cotton. Pesquisa-Agropecuaria- Brasileira. 36(7): 941-948.

Bhatade, S. S. and N. L. Bhale. 1983. Estimation of gene effects for yield characters in Gossypium

arboreum L. Ind. J. Agric. Sci. 53(6): 423-427.

Bhatti, M. A., M. Saeed, N. Chatta and S. Iqbal. 1976. Host plant resistance and importance of insect

population suppression in cotton crop. Proc. Cotton. Prod. Seminar, ESSO, Pak.

Fertilizer Co. Ltd. pp: 132-142.

143

Bourland, F. M., J. M. Hornbeck, A. B. McFall, and S. D. Calhoun. 2003. A rating system for leaf

pubescence of cotton. J. of Cotton Sci. 7: 8-15.

Bryson, C. T., J. C. McCarty, Jr., J. N. Jenkins and W. L. Parrott. 1983. Frequency of pigment

glands and capitate and covering trichomes in nascent leaves of selected cottons.

Crop Sci. 23: 369-371.

Butler, G. D., Jr., F. D. Wilson and G. Fishler. 1991. Cotton leaf trichomes and populations of

Empoasca lybica (Homoptera: Cicadellidae) and Bemisia tabaci (Homoptera:

Aleyrodidae). Plant Prot. 10: 461-464.

Cai, Y. F., H. Zhang, Y. Zeng, J. C. Mo, j. K. Bao, C. Miao, J. Bai, F. Yan and F. Chen. 2004. An

optimized gossypol high-performance liquid chromatography assay and its

application in evaluation of different gland genotypes of cotton. J. Bio.Sci. 29(1): 67-

71.

Calhoun, D. S. 1997. Inheritance of high glanding, an insect resistance trait in cotton. Crop.Sci. 37:

1181-1186.

Chowdhuri, K. A. and G. M. Buth. 1971. Cotton seeds from the Neolithic in Egyptian Nubia and

origin of old world cotton. In Breeding plants resistant to insects. F.G. Maxwell and

P. R. Jennings., John Wiley and Sons, USA, pp: 337-370.

Chu, C. C., E. T. Natwick, T. J. Henneberry, P. Dugger and R. Richter. 2000 b. Susceptibility of

normal leaf and okra leaf shape cottons to silver leaf whiteflies and relationships to

trichome densities. Proceedings Beltwide Cotton Conferences, San Antonio, USA. 2:

1157-1158.

Chu, C., A. C. Cohen, E. T. Natwick, G. S. Simmons and T. J. Henneberry (1999). Bemisia tabaci

(Hemiptera: Aleyrodidae) biotype B colonization and leaf morphology relationships

in upland cotton cultivars. Australian Journal of Entomology 38 (2): 127–131.

Chu, C., E. T. Natwick, T. J. Henneberry (2002). Bemisia tabaci (Homoptera: Aleyrodidae) biotype

B colonization on okra- and normal-leaf upland cotton strains and cultivars. J. Econ.

Entomol. 95(4): 733-738.

Chu, S. S., T. P. Freeman, J. S. Buckner, E. T. Natwick, T. J. Henneberry and D. R. Nelson. 2000 a.

Silverleaf whitefly colonization and trichome density relationships on upland cotton

cultivars. Southwestern- Entomologist. 25(4): 237-242.

144

Dani, R. G. and R. J. Kohel. 1989. Maternal effects and generation mean analysis of seed-oil

content in cotton (Gossypium hirsutum L.). TAG. 77(4): 569-575.

Deshpande, L. A., G. R. Vyahalkar, A. S. Asingkar and S. S. Mane. 1984. Nature of gene action for

yield and fibre traits in upland cotton. Ind. J. Agric. Sci. 54(2): 97-99.

Deshpande, L. A., K. S. Baig. 2003. Combining ability analysis for yield, economic and

morphological traits in American cotton (Gossypium hirsutum L.). J. Res. ANGRAU.

31(3): 28-34.

Deshpande, L. A., K. S. Baig. 2004. Heterosis for yield, economic and morphological traits in

American cotton (Gossypium hirsutum L.). Nat. J. Pl. Improvement. 6(2): 136-138.

Dewey, D. R. and K. H. Lu. 1959. A correlation and path-coefficient analysis of components of

crested wheat grass seed production. Agron. J. 51: 515:518.

Dhanda, S. S., A. A. Tyagi and D. S. Jatasra. 1984. Character association among quantitative and

quality attributes of upland cotton. Ind. J. Agric. Sci. 54(1): 24-29.

Dhawan, A. K., J. Singh and A. S. Sidhu .1980. Maconellicoccus sp. attacking arboreum cotton in

Punjab. Science and Culture. 46: 258.

Dhillon, S. S. and T. H. Singh. 1982. Genetic control of some quantitative characters in upland

cotton (Gossypium hirsutum L.). J. Agri. Sci., U.K. 94(3): 539-543.

Dinham, B. 1993. The Pesticide Hazard: A Global Health and Environmental Audit, London: The

Pesticides Trust.

Dittrich, V., S. O. Hassan and G. H. Ernst. 1986. Development of a new primary pest of cotton in the

Sudan: Bemisia tabaci, the whitefly. Agriculture, Ecosystems and Environment. 17(1-

2): 137-142.

Dolan, L. and R. S. Poethig. 1998. The okra leaf shape mutation in cotton is active in all cell layers

of the leaf. Am. J. Bot. 85(3): 322-327.

Duhoon, S. S., M. Singh and A. K. Basu. 1981. Preferential behaviour of the spotted bollworm and

two varieties of Desi cotton (G. arboreum L.). In Cotton Breeding, IInd edition, 2004

eds. Phundun Singh., Kalyani Publishers, New Delhi. pp: 136-146.

Echekwu, C. A. 2001. Correlations and correlated responses in upland cotton

(Gossypium hirsutum L ). Tropicultura. 19(4): 210-213.

145

Edwards, C. A. 1993. The Impact of Pesticides on the Environment, in D. Pimental and H. Lehman,

eds. The Pesticide Question: Environment, Economics, and Ethics, New York:

Chapman and Hall.

El- Fawal, M. A., A. M. Gad., A. A. A. Bary and A. A. El- Kishen. 1974. Study on gene action in an

interspecific cross of cotton. II. Estimation of genetic variance and heritability.

Egyptian J. Genet. Cytol. 3(2): 236-345.

Endrizzi, J. E., E. L. Turcotte and R. J. Kohel. 1984. Qualitative genetics, cytology and cytogenetics.

In Cotton Breeding, IInd edition, 2004 eds. Phundun Singh., Kalyani Publishers, New

Delhi. pp: 136-146.

Falconer, D.S., 1981. Introduction to Quantitative Genetics, IInd edition. Longman Inc. New York.

Falconer, D. S. and T. F. C. Mackay. 1996. Introduction to Quantitative Genetics. Chapman and

Hall, London. U.K.

Fayek, Y. N. and T. E. Anwer. 2007. Development and validation of spectrophotometric, TLC and

HPLC methods for the determination of lamotrigine in presence of its impurity.

Chemical and pharmaceutical bulletin. 55: 541-545.

Feki, T. A. and M. A. A. Gelil. 2001. Diallel analysis of cotton hybrids (G. barbadense L.).

II—Genetic behaviour of fibre to seed attachment force, fuzz index and fibre quality

index. Egyptian J. Agric. Res. 79(2): 607-623.

Feki, T. A., M. A. A. Gelil and A. A. Zeina. 1998. Diallel analysis of cotton hybrids (G. barbadense

L.). I- Earliness characters of F1’s and F2’s. Egyptian J. Agric. Res. 76(1): 219-230.

Fertilizer Statistics. 1999-2000. Fertilizer Association of India New Delhi. 3: 38-39.

Food and Agriculture Organization of United Nations. 2000. Pakistan Cotton Production and Trade

Development and Prospects. In M.Sc. Thesis, Faculty of the Univ. Georgia., Athens,

Georgia.

Frelichowski, J., M. Ulloa, C. A. Shafter, A. Percival, J. Stewart, A. R. Fayetteville, R. Cantrell and

N. C. Cary. 2005. Germplasm evaluation of cotton accessions from the U. S. cotton

germplasm collection, USDA-ARS (Gossypium hirsutum L. Landraces of Mexico).

Beltwide cotton conferences, New Orleans, Lousiana. January, 4-7, 2005.

146

Gad, A. M., M. A. El- Fawal, M. B. Bashir and A. A. El- Kishen. 1974. Studies on gene action in an

interspecific cross of cotton. I. Manifestation of types of gene effects. Egyptian J.

Genet. Cytol. 3(1): 117-124.

Garcia, J. N., I. M. Ray, J. A. Henning and L. W. Murray. 2004. Quantitative genetic analysis of

erect glandular trichome density in diploid alfalfa. Euphytica. 111(1): 61-65.

Garland, J. A. 1998. Pest Risk Assessment of the pink mealybug Maconellicoccus hirsutus (Green),

with particular reference to Canadian greenhouses. PRA. 96–21. Canadian Food

Inspection Agency, Ottawa (CA).

Ghose, S. K. 1970. Predators, parasites and attending ants of the mealybug, Maconellicoccus

hirsutus. Plant Protection Bulletin, India. 22: 22–30.

Gill, M. S. and H. S. Kalsy. 1981. Genetic analysis in four crosses of upland cotton. Crop Imp. 8(2):

95-99.

GOP, 2005-06. Economic Survey of Pakistan, Finance Division, Economic Advisor’s Wing,

Government of Pakistan, Islamabad.

GOP, 2007-08 a. Pakistan Economic Survey, Economic Adviser’s Wing, Finance Division.

Government of Pakistan, Islamabad.

GOP, 2007-08 b. Agricultural Statistics of Pakistan, Economic Wing, Ministry of Food and Agri.

Government of Pakistan, Islamabad.

GOP, 2007-08 c. Pakistan Economic Survey, Economic Adviser’s Wing, Finance Division.

Government of Pakistan, Islamabad.

GOP, 2007-08 d. Federal Statistics Bureau, Ministry of Food, Agri. and Livestock, Government of

Pakistan, Islamabad.

Green, J. M. 1953. Sub-okra; a new leaf type in upland cotton. J. Hered. 44: 229- 232.

Gregory, S. R., E. Hernandez and B. R. Savoy .1999. Cotton seed processing. In WC Smith, JT.

Gulati, A. M. and A. N. Turner. 1928. A note on the early history of cotton. In Breeding plants

resistant to insects. F.G. Maxwell and P. R. Jennings., John Wiley and Sons, USA,

pp: 337-370.

Gupta, S. P. 1987. Genetics of seed index, lint index, earliness and plant height in upland cotton. Ind.

J. Agric. Sci. 57(6): 427-428.

147

Hammond, D. 1941.The Expression of Genes for Leaf Shape in Gossypium hirsutum L. and

Gossypium arboreum L. II. The Expression of Genes for Leaf Shape in Gossypium

hirsutum L. American Journal of Botany, Vol. 28(2): 138-150.

Haq, I. and F. M. Azhar. 2005. Genetic analysis of some fibre quality characters in Gossypium

hirsutum L. Int. J. Agri. Bio. 7(2): 266-268.

Harland, S. C. 1939. The genetics of cotton. London: Jonathan Cape.

Harris, J. A. 1912. A simple test of goodness of fit of Mendelian ratios. Am. Nat., 46: 741-745.

Hassan, G., G. Mahmood, N. Khan and A. Razzaq. 1999. Combining ability and heterobeltiotic

estimates in a diallel cross of cotton Gossypium hirsutum L. Sarhad J. Agri. 15(6):

563-568.

Hassan, S. T. S., L. T. Wilson and P. R. B. Blood. 1990. Oviposition by Heliothis armigera and H.

Punctigera (Lepidoptera: Noctuidae) on okra leaf and smooth leaf cottons.

Env.Entomol. 19: 710-716.

Hedin, P. A. and J. C. McCarty. 1990. Possible roles of cotton bud sugars and terpenoids in

oviposition by the boll weevil. In Cotton Breeding, IInd edition, 2004 eds. Phundun

Singh., Kalyani Publishers, New Delhi. pp: 136-146.

Hendaway, F. A., M. S. Rady, A. M. Hamid and R. M. Esmail. 1999. Inheritance of fibre traits in

some cotton crosses. Egyptian J. Agron. 21: 15-36.

Herbert, D. A. 2000. Bollworm management in Virginia cotton: history, strategies and successes.

Proceedings Cotton Conferences. 2: 1101-1104.

Herring, A. D., D. L. Auld, M. D. Ethridge, E. F. Hequet, E. Bechere, C. J. Green and R. G. Cantrell.

2004. Inheritance of fibre quality and lint yield in a chemically mutated population of

cotton. Euphytica. 136(3): 333-339.

Hron R. J, M. S. Kuk, and G. Abraham. 1990. Determination of free and total gossypol by high

performance liquid chromatography. J. Am. Oil. Chem. 67: 182–187.

Hunter, L. 1991. HVI Technology, its use and misuses, ICAC Recorder 9(3).

IISD, and WWF .1997. The Cotton Industry: Towards an Environmentally Sustainable Commodity

Chain, Report Prepared for the Workshop on Cross-National Environmental Problem-

Solving, School of International and Public Affairs, Columbia University.

148

Ilango, K. and S. Uthamasamy. 1989. Biochemical and physical bases of resistance to bollworm

complex in cotton varieties. In Cotton Breeding, IInd edition, 2004 eds. Phundun

Singh., Kalyani Publishers, New Delhi. pp: 136-146.

Innes, N. L., R. H. Wimble and H. E. Gridley. 1975. Estimates of genetic parameters for lint quality

in upland cotton (Gossypium hirsutum L.). Theor. Appl. Genet. 46(5): 249-256.

International Cotton Advisory Committee. 2006. New Frontiers in Cotton Production. Washington,

DC. USA.

International Cotton Advisory Committee. 2007 a. Cotton: Review of the World situation.

Washington, DC. USA. 60(4): 3-18.

International Cotton Advisory Committee. 2007 b. Update on Cotton Production Research. The

ICAC Recorder. Washington, DC. USA. 25(3): 3-19.

Introduction & economic importance to cotton. 2001. Retrieved on 10.10.2007, from the website:

www.ccri.gov.pk

Iqbal, M. Z. and M. A. Nadeem. 2003. Generation mean analysis for seed cotton yield and number

of sympodial branches per plant in cotton (Gossypium hirsutum L.). Asian J. Pl. Sci.

2(4): 395-399.

Islam, Z., H. A. Sadaqat and F. M. Azhar. 2001. Evaluation of some cotton strains for their

combining ability in important genetic traits. Int. J. Agri. Biol. 3(4): 409- 410.

Jayaraj, S. and M. Sellammal. 1988. Studies on bollworm resistance in cotton. Bollworm Seminar,

Nagpur, India. In Cotton Breeding, IInd edition, 2004 eds. Phundun Singh., Kalyani

Publishers, New Delhi. pp: 136-146.

Jenkins, J. N., M. G. Maxwell and H. N Lafever. 1966. The comparative preference of insects for

glanded and glandless cottons. J.Eco. Ento. 59(2): 352-356.

Jenkins, J. N. 1989. State of the art in host plant resistance in cotton. In Cotton Breeding, IInd edition,

2004 eds. Phundun Singh., Kalyani Publishers, New Delhi.pp: 136-146.

Ji, D. F. and J. Zhu. 1988. Genetic analysis of oil and amino acid content in shelled seed of upland

cotton hybrids. Acta Agronomica Sinica. 14(1): 1-6.

Johnson, H. W., H. F. Robinson and R.E. Comstock. 1955. Estimates of genetic and environmental

variability in soybean. Agron. J. 47 : 314-318.

149

Jones, J. E., J. I. Dickson and R. G. Novick. 1988. Another look at effects of leaf shape traits on

agronomic performance of upland cotton. Proc.Beltwide Cotton Prod. Res. Conf. P.

94.

Kairo, M. T. K., G. V. Pollard, D. D. Peterkin and V. F. Lopez. 2000. Biological control of the

hibiscus mealybug, Maconellicoccus hirsutus in the Caribbean. Integrated Pest

Management Reviews. 5: 241–254.

Kalazich, J. C. and R. L. Plaisted. 1991. Association between trichome characters and agronomic

traits in Solanum tuberosum (L.) X.S. berthaultii (hawkes) hybrids. Am. J. Potato

Res. 68(12): 833-847.

Kalsy, H. S. and B. M. Vithal. 1980. Inheritance of some qualitative charaters in upland cotton

(Gossypium hirsutum L.). J. Res. Punjab Agric. Univ., India, 17(1): 1-7.

Kalsy, H. S. and H. R. Garg. 1988. Analysis of generation means for metric traits in upland cotton

(Gossypium hirsutum L.). Ind. J. Agric. Sci. 58(5): 397-399.

Kaseem, E. S., M. A. Khalifa, M. A. El-Morshidy and F. G. Younis. 1984. Genetical analysis of

some agronomic characters in cotton II. Yield and its components. Agric. Res.

Review, 59(9): 62-82.

Kaynak, M. A., A. Unay, I. Ozkan and I. Basal. 2000. Determination of the heterotic effects and

phenotypic correlation of earliness, and agronomical and quality characteristics in

cotton (Gossypium hirsutum L.). Turkish J. Agri. Forestry. 24(1): 105-111.

Khalifa, H. and O. I. Gameel, 1983. Breeding cotton cultivars resistant to whitefly, Bemisia tabaci.

In: Rev. Appl. Entomol., (A), 1984. 72(1): 930.

Khan, M. A. J., M. A. Khan and T. E. Lodhi. 1999. Genetic study of yield and yield related

components in cotton. (Gossypium hirsutum L.). J. A. P. S. 9(1): 73-75.

Kloth, R. H. 1993. New evidence relating the pilose allele and micronaire reading in cotton. Crop

Sci. 33: 683-687.

Kloth, R. H. 1995. Interaction of two loci that affect trichome density in upland cotton. J. Heredity.

86(1): 78-80.

Knight R. L. 1952. The genetics of jassid resistance in cotton. I. The genes H1 and H2. J. Genet. 51:

46–66.

150

Kohel, R. J. 1987. Seed oil content of glanded and glandless cottons. J. Amer. Oil Chem. Soc. 64(9):

1337-1340.

Konoplaya, S. P., V. N. Fursov and A. Odemamedov. 1979. Heritability estimates and genetic

correlations in respect of some characters in first generation hybrids of Gossypium

hirsutum L. Turkmenistan SSR XylmLar Akademijasynyn Habarlary, Biologik

YlymLaryn No. 2: 34-40.

Kowsalya, R., T. S. Raveendran and T. Pushpalatha. 2000. Heterosis and combining ability analysis

for seed quality traits in Gossypium hirsutum L. Madras Agric. J. 86(1/3): 127-134.

Kular, J. S. and N. S. Butter, 1999. Influence of some morphological traits of cotton genotypes on

resistance to whitefly, Bemisia tabaci. J. Insect. Sci., 12(1): 81-83.

Kumar, P. R. and T. S. Raveendran. 1999. Genetic analysis of yield and quality traits in upland

cotton (Gossypium hirsutum L.) through generation mean analysis. Indian J. Agric.

Sci. 69(7): 538-540.

Kumaresan, D., J. Ganesan, P. Azhaguvel and P. S. Kumar. 1999. Heterosis for some important

quantitative traits in cotton (Gossypium hirsutum L.). Tropical Agric. Res. 11: 149-

153.

Kumaresan, D., P. Senthilkumar and J. Ganesan. 1999. Combining ability studies for quantitative

traits in cotton (Gossypium hirsutum L.). Crop Res. Hisar. 18(3): 430-432.

Kumari, S. R. and N. Chamundeswari. 2005. Studies on genetic variability, heritability and genetic

advance in cotton (Gossypium hirsutum L.). Research on crops. 6(1): 98-99.

Lacape, J. M. and T. B. Nguyen. 2005. Mapping Quantitative Trait Loci Associated with Leaf and

Stem Pubescence in Cotton. Journal of Heredity 2005 96(4): 441-444.

Larkin, J. C., D. G. Oppenheimer, A. M. Lloyd, E. T. Paparozzi and M. David. 1994. Roles of the

GLABROUS1 and TRANSPARENT TESTA GLABRA genes in Arabidopsis

trichome development. The Plant Cell. 6: 1065-1076.

Lee, J. A. 1973. The inheritance of gossypol level in gossypium II: inheritance of seed gossypol in

two strains of cultivated gossypium barbadense L. Genetics. 75 : 259-264.

151

Lee, J. A. 1968. Genetical studies concerning the distribution of trichomes on the leaves of

Gossypium hirsutum L. Genetics. 60 : 567-575.

Lee, J. A. 1985. Revision of the genetics of the hairiness smoothness system of Gossypium. J. Hered.

76: 123–126.

Lee, J. A. 1962. Genetical studies concerning the distribution of pigment glands in the cotyledons

and leaves of upland cotton. Genetics 47: 131-142.

Lee, J. A. 1964. Effects of the pilose allele, H2, on a long staple upland cotton. Crop Sci. 4: 442-443.

Lee, J. A. 1984. Effects of two pilosity alleles on agronomic and fibre traits in upland cotton. Crop

Sci. 24: 127-129.

Lee, J. A. 1971. Some problems in breeding smooth- leaved cottons. Crop Sci. 11: 448-450.

Lee, J. A., C. C. Cockerham and F. H. Smith. 1967. The inheritance of gossypol level in Gossypium.

I. Additive, dominance, epistatic, and maternal effects associated with seed gossypol

in two varieties of Gossypium hirsutum L. Genetics 59: 295-298.

Levin, D. A. 1973. The Role of Trichomes in Plant Defence. The Quarterly Review of Biology.

48(1): 3-15.

Lin, Y. and L.Y. Zhao. 1988. Estimation of genetic effects on the main fibre quality characteristics

in upland cotton. Acta Genetica Sinica, 15(6): 401-408.

Liu, H. T., X. M. Guo and J. Y. Xia. 2000. Genetic effects of agronomic characters in intraspecific

hybrids of transgenic cotton x non-transgenic upland cotton (Gossypium hirsutum L.).

Acta-Gossypii-Sinica. 12(3): 118-121.

Liu, Y. X. and X. M. Han. 1998. Research on the combining ability and inheritance of 12 economic

characters in upland cotton. China-Cottons. 25(3): 9-11.

Lukefahr, M. J., J. E. Houghtaling and H. M. Graham. 1971. Suppression of Heliothis populations

with glabrous cotton strains. In Cotton Breeding, IInd edition, 2004 eds. Phundun

Singh., Kalyani Publishers, New Delhi. pp: 136-146.

Malek, M. A. and A. K. M. Shamsuddin. 1999. Combining ability and heterosis for yield and yield

contributing characters in upland cotton (Gossypium hirsutum L.). Annals of

Bangladesh Agri. 8(1): 49-56.

152

Mani, M. 1989. A review of the pink mealybug – Maconellicoccus hirsutus. Insect Science and its

Application. 10: 157–167.

Mani, M. and A. Krishnamoorthy. 2001. Suppression of Maconellicoccus hirsutus on guava. Insect

Environment. 6: 152.

Manimaran, R. and T. S. Raveendran. 2002. Heterosis in intra and interspecific cotton hybrids.

Advances Pl. Sci. 15(1): 255-259.

Mansour, M. H., N. M. Zohdy, El- Gengaihi and E. M. Amr. 2004. The status of gossypol and tannin

contents in some Egyptian cotton cultivars in correlation to bollworm infestation.

Bulletin of the National Res. Centre. 29(1): 121-128 (Cab abstract No. 2004316267).

Martin, F. W. and A. M. Rhodes. 1983. Seed characteristics of okra and related Abelmoschus

species. Chemistry and Materials Science. 33(1): 41-49.

Mather, K. and J. L. Jinks. 1982. Biometrical Genetics, third edition, Champman and Hall Ltd.

London.

Matthews, G. A. and J. P. Tunstall. 1994. Insect pests of cotton. C.A.B. International, Wallingford,

England.

Maxwell, F. G. 1977. Plant resistance to cotton insects. E.S.A. Bulletin. Vol. 23(3): 199-203.

Maxwell, F. G., N. J. Johnie and W. L. Parrott. 1972. Resistance of plant to insects. Adv. Agron. 24:

187-265.

May, O. C. and C. C. Green. 1994. Genetic variation for fibre properties in elite Pee Dee cotton

populations. Crop Sci. 34(3): 684-690.

McCarty, J. C., J. N. Jenkins and J. X. Wu. 2004 b. Primitive accession derived germplasm by

cultivar crosses as sources for cotton improvement: II. Genetic effects and genotypic

values. Crop Sci. 44(4): 1231-1235.

McCarty, J. C., W. R. Meredith, J. N. Jenkins, W. L. Parrott and J. C. Bailey. 1983 a. Genotype ×

environment interaction of cottons varying in insect resistance. Crop Sci. 23: 970-

973.

Meagher, R. L., C. W. Smith, and W. J Smith. 1997. Preference of Gossypium genotypes to Bemisia

argentifolii (Homoptera: Aleyrodidae). J. Econ. Entomol 90(4): 1046–1052.

153

Meredith, W. R. and R. Wells 1986. Normal vs. okra leaf yield interactions in cotton. I -Performance

of near-isogenic lines from bulk populations. Crop Sci 26: 219-222.

Meredith, W. R., Jr., and R. R. Bridge. 1972. Heterosis and gene action in cotton (Gossypium

hirsutum L.). Crop Sci. 12(3): 304-310

Meredith, W. R., W. T. Pettigrew and J. J. Heitholt. 1996. Sub- Okra, semi- smoothness and

nectariless effect on cotton lint yield. Crop. Sci. 36: 22-25.

Meredith, W. R. Jr. and R. Wells. 1987. Sub-okra leaf influence on cotton yield. Crop Sci. 27: 47-48.

Mert, M., O. Gencer, Y. Akscan, and K. Boyac. 2003. Inheritance of yield and yield components in

cotton (Gossypium hirsutum L.). Turkish J. Field Crops. 8(2): 62-67.

Meyer, R., S. Vorster and I. A. Dubery. 2004. Identification and quantification of gossypol in cotton

by using packed micro-tips columns in combination with HPLC. Analytical and

Bioanalytical Chemistry. 380(4): 719-724.

Miller, P. A. and A. Marani. 1963. Heterosis and combining ability in diallel crosses of upland

cotton. Crop Sci. 3(5): 441-445.

Mohan, P., P. Singh, A. B. Dongre, and S. S. Narayanan. 1995. Gossypol gland density and free

gossypol content in seed and cotyledonary leaf of upland cotton (Gossypium

hirsutum). Indian. J. Agric. Sci. 65(1): 66-68.

Mohan, P., P. Singh, S. S. Narayanan and R. Ratan. 1994. Relation of gossypol gland density with

bollworm incidence and yield in tree cotton. Indian. J. Agric. Sci. 64(10): 691-696.

Monks, C. D., M. G. Patterson, D. P. Delaney, and M. D. Pegues. 1999. The effects of row width

and leaf shape on cotton growth and yield. Alabama Agricultural Experiment Station

Luther Waters, Director Auburn University Auburn, Alabama Bulletin 638: 3-9

Mukhtar, M. S., T. M. Khan and A. S. Khan. 2000. Gene action study in some fibre traits in cotton

(Gossypium hirsutum L.). Pak. J. Biol. Sci. 3(10): 1609- 1611.

Muralidharan, C. M. and S. N. Badaya .2000. Mealy bug (Maconellicoccus hirsutus) outbreak on

herbaceum cotton (Gossypium herbaceum) in Wagad cotton belt of Kachch. Indian J.

Agri. Sc. 70: 705–706.

Murray, D. L. 1994. Cultivating Crisis: The Human Cost of Pesticides in Latin America, Austin:

University of Texas Press.

154

Murtaza, N., G. M. Ali and H. Rahman. 2004. Inheritance of fibre quality characters in upland cotton

genotypes having non-preference traits for insect pests. J. Food, Agri. &

Environment. 3(1): 179-183.

Murthy, J. S. V. S. 1998. Genetics of yield and its components in cotton under artificial bollworm

infestation. Annals of Agric. Res. 19(3): 294-298.

Muttuthamby, S., M. Aslam and M. A. Khan. Inheritance of leaf hairiness in Gossypium hirsutum L.

cotton and its relationship with jassid resistance. Euphytica 18 : 435-439.

Nadarajan, N. and S. R. S. Rangasamy. 1990. Combining ability and variability studies in

Gossypium hirsutum L. Ind. Society for cotton Improve. J. 15(1): 16-19.

Nadeem, K. and F. M. Azhar. 2004. Genetic analysis of seed yield and its components in Gossypium

hirsutum L. Int. J. Agri. Bio. 6(5): 865-868.

Nadeem, K. and F. M. Azhar. 2005. Genetic analysis of fibre length and strength of Gossypium

hirsutum L. Int. J. Agri. Bio. 7(2): 263-265.

Naveed, M., F. M. Azhar and A. Ali. 2004 a. Estimates of heritabilities and correlations among seed

cotton yield and its components in Gossypium hirsutum L. Int. J. Agri. Bio. 6(4): 712-

714.

Naveed, M., F. M. Azhar and A. Ali. 2004 b. Correlation analysis of seed cotton yield with fibre

characteristics in Gossypium hirsutum L. Int. J. Agri. Bio. 6(4): 656-658.

Neelima, S., V. C. Reddy and A. N. Reddy. 2004. Combining ability studies for yield and yield

components in American cotton. Annals of Agri. Bio. Res. 9(1): 1-6.

Niles, G. A. 1980 a. Breeding cotton for resistance to insects. In Breeding plants resistant to insects.

F.G. Maxwell and P. R. Jennings., John Wiley and Sons, USA, pp: 337-370.

Niles, G. A. 1980. Breeding cotton for resistance to insects. In Cotton Breeding, IInd edition, 2004

eds. Phundun Singh., Kalyani Publishers, New Delhi. pp: 136-146.

Nistor, T. and G. Nistor. 1999. Inheritance of fibre length in cotton. Analele-Institutului-de-

Cercetari-si-Plante-Tehnice,-Fundulea. 66: 13-23.

Painter, R. H., 1951. Insect resistance in crop plants. In: Breeding plants resistant to insects. F.G.

Maxwell and P. R. Jennings., John Wiley and Sons, USA, pp: 337-370.

155

Pandey, S. K., P. Singh. 2002. Genetic variability in upland cotton (Gossypium hirsutum L.).

Progressive Agri. 2(2): 178-179.

Parkash, O. 1982. Mode of inheritance of yield and fibre contents in cotton (Gossypium

arboreum L.). Thesis Abst. 8(4): 389- 390.

Pathak, R. S. 1975. Gene effects for fibre properties in upland cotton (Gossypium hirsutum L.).

Theor. Appl. Genet. 46(3): 129- 133.

Pathak, R. S. and R. B. Singh. 1970. Genetics of yield characters in upland cotton. Ind. J. Genet. Pl.

Br. 30(3): 679-689.

Pavasia, M. J., P. T. Shukla and U. G. Patel. 1999. Combining ability analysis over environments for

fibre characters in upland cotton. Indian J. Genet. Pl. Br. 59(1): 77-81.

Perazza, D., M. Herzog, M. Hülskamp, S. Brown, A. M. Dorne and J. M. Bonneville. 1999.

Trichome cell growth in Arabidopsis thaliana can be derepressed by mutations in at

least five genes. Genetics. 152: 461-476.

Percy R. G. and Kohel R. J. 1999. Qualitative genetics In: Cotton. Origin, history, technology, and

production. (Smith CW and Cothren JT, eds). New York: Wiley, 319–360.

Percy, R. G., Z. Lu, J. W. Radin, E. L. Turcotte and E. Zeiger (1996). Inheritance of stomatal

conductance in cotton (Gossypium barbadense). Physiologia Plantarum 96(3): 389–

394.

Percy, R. G. 2001. Registration of PS-6ne, PS-6L, PS-6neL, P62ne, P62L and P62neL, Extra- long

staple cotton germplasm. Crop Sci. 41: 602-603.

Pettigrew, W. T. 2003. Relationships between insufficient potassium and crop maturity in cotton.

Agron. J. 95: 1323-1329.

Phogat, D. S., D. P. Singh and J. S. Hooda. 2000. Correlation studies on bollworm incidence,

gossypol, seed cotton yield and fibre characters in American cotton

(Gossypium hirsutum L.). Annals of Biology, Ludhiana. 16(1): 89-93.

Potdukhe, N. R. 2001. Heterosis and identification of superior crosses in upland cotton. Crop

Improvement. 28(2): 187-190.

Pyke, B. A. and E. H. Brown .2000. The cotton pest and beneficial guide.

156

Qian, D. S., N. Y. Xu, J. C. Di, X. G. Zhang and J. G. Liu. 2001. Heterosis and F2 inbreeding

depression of hybrid cotton “Suza 16”. Jiangsu J. Agric. Sci. 17(1): 7-12.

Rahman, H. and W. S. Khan. 1998. Expressivity of H2 gene of hairiness and L0 gene of leaf shape of

cotton under different genetic backgrounds. Pak. J. Bot. 30(1): 95-100.

Rahman, H., A. Bibi and M. Latif. 2005. Genetic variability in okra-leaf upland cottons for

agronomic and fibre traits. J. Appl. Genet. 46(2): 149-155.

Rajan, G. A. S., G. Kandasamy, C. Vanniarajan and G. Rajaravindran. 2000. Study of heterosis in

American cotton. Indian J. Agric. Res. 34(3): 156-159.

Ramalingam, A. and N. Sivasamy. 2002. Genetics and order effects of seed cotton yield in upland

cotton (Gossypium hirsutum L.)- triallel analysis. Indian J. Genet. Pl. Br. 62(4): 359-

360.

Randhawa, L. S., G. S. Chahal and T. H. Singh. 1986. Role of epistasis in the inheritance of yield

and its components in upland cotton. Ind. J. Agric. Sci. 56(7): 494- 496.

Ravindranath, K. S. B. Mohiddin, R. R. Reddy and D. Balakondappa. 2000. Promising desi cotton

hybrid for Rayalaseema region. J. Res. ANGRAU. 29(4): 106-108.

Rayburn, S. T., Jr. 1986. A leaf hairiness index for nine cotton cultivars. p. 434. In Proc. Beltwide

Cotton Prod. Res. Conf., Las Vegas, NV. 4-9 Jan. 1986. Natl. Cotton Counc. Am.,

Memphis, T. N.

Raza, A. B. M., M. Afzal, M. Sarfraz, M. H. Bashir, M. D. Gogi and M. S. Sarwar. 2000. Physico-

morphic plant characters in relation to resistance against sucking insect pests in some

new cotton genotypes. Pak.Entomol.. 22(1-2): 73-77.

Rehman, A., M. A. Khan and I. Hassan. 1993. A diallel analysis of varietal differences for some

ginning and fibre traits in Gossypium hirsutum L. crosses. Pak. J. Agric. Res. 31(3):

257-266.

Rehman, S., M. A. Khan, M. A. Khan and M. A. Khan. 1988. Genetic analysis of yield and yield

components in various crosses of American upland cotton. Sarhad J. Agric. 4(4): 495-

514.

157

Rekha, A. R., M. N. Maruthi, V. Muniyappa and J. Colvin (2005). Occurrence of three genotypic

clusters of Bemisia tabaci and the rapid spread of the B biotype in south

India.Entomologia Experimentalis et Applicata 117 (3): 221–233.

Renou, A., J. Pages, D. Dessauw, and J. Charusri. 2001. First steps to improve cotton crop

management in Thailand. Mededelingen (Rijksuniversiteit te Gent. Fakulteit van de

Landbouwkundige en Toegepaste Biologische Wetenschappen). 66(2a): 425-435.

Reynolds, H. T. I., 1970. Current pest control problems on cotton in the south- western United

States. M.Sc. Thesis, Dept. of Agric. Entomol., Univ. Agri., Faisalabad.

Rowena, Y. K., M. S. Kang, A. E. Zipf, D. E. Wesenberg and G. C. Sharma. 2002. Genetic analysis

of plant regenerability in oats (Avena spp. L.) Journal of new seeds: Innovations in

production, biotechnology, quality, and marketing. 4(4): 31-40.

Saunders, J. H. 1965. Genetics of hairiness transferred from Gossypium raimondii to G. hirsutum.

Euphytica 14: 276-282.

Sayal, O. U. and M. Z. Sulemani. 1996. Comparison of gene action controlling the qualitative traits

in some early maturing cultivars of American cotton (Gossypium hirsutum L.).

Sarhad J. Agri. 12(6): 653-659.

Schepers, A. 1989. Chemical control. In: World Crop Pests: Aphids, Their Biology, Natural Enemies

and Control, Vol. C. (eds AK Mink and P Harrewijn). Pp. 89–122.

Shaw, A. J. 2000. Cotton pest Management guide 2000/2001. Cotton Technical Specialist, NSW

Agriculture.

Sikka, S. M., V. M. Sahni and D. K. Butani. 1966. Studies on jassid resistance in relation to hairiness

of cotton leaves. Euphytica 15: 383-388.

Silivia, F. P. D. A. and J. F. Alves. 1983. Estimation of epistatic, additive and dominance variation

in cotton (Gossypium hirsutum L.) race latifolium Hutch. Revista Brasileria de

Genetica (Brazil). 6(3): 491-503.

Simpson, D. M. 1947. Fuzzy leaf in cotton and its association with short lint. J. Herd. 38: 153-156.

Simmons, A. T. and G. M. Gurr. 2005. Trichomes of Lycopersicon species and their hybrids: effects

on pests and natural enemies. 7(4): 265-276.

158

Singh, D. and J. P. Yadavendra. 2002. Gene actions for ginning outturn and staple length in cotton.

Annals Agri. Res. 23(4): 558- 562.

Singh, J. P., J. Singh, K. S. Brar and J. Singh. 2001. Efficiency of Trichogramma chilonis Ishii on

different varieties of cotton. Insect Environment. 7(1): 15-16.

Singh, P. and H. G. Singh. 1981. Gene action, heritability and genetic advance in upland cotton. Ind.

J. Agric. Sci. 51(4): 209-213.

Singh, P. and S. S. Narayanan. 2000. Biometrical Techniques in Plant Breeding. IInded. Kalyani

Publ. Ludhiana, New Delhi, India.

Singh, R. and R. A. Agarwal. 1988. Role of chemical components of resistant and susceptible

genotypes of cotton and okra in ovipositional preference of leaf hoppers. In Cotton

Breeding, IInd edition, 2004 eds. Phundun Singh., Kalyani Publishers, New Delhi. pp:

136-146.

Singh, R. B., M. P. Gupta, B. R. Mor and D. K. Hain. 1968. Variability and correlation studies on

yield and quality characters in hirsutum cotton. Ind. J. Genet. 28: 216-222.

Singh, R. P., V. P. Singh, M. Singh and C. B. Lal. 2002. Selection of non-fuzzy (naked seed) Punjab

American Cotton (Gossypium hirsutum L) lines with high ginning out turn

(lint percentage). Annals Agri. Res. 23(2): 191-196.

Singh, S. 1982. Genetics of yield, its components and quality characters in upland cotton

(Gossypium hirsutum L.). Thesis Abst. Haryana Agri. Univ., Hissar, India. 8(2): 186-

187.

Singh, T. H. and B. H. Sandhu. 1979. Estimates of gene effects in Gossypium hirsutum L. Ind. J.

Agric. Sci. 49(5): 325-329.

Singh, T. H., M. A. Quader and G. S. Chahal. 1983. Estimates of gene effects for some quantitative

characters in upland cotton (Gossypium hirsutum L.). Cotton et Fibres Trop. 38(4):

319-322.

Smith, A. L. 1964. Leaf trichomes of upland cotton varieties. Crop Sci. 4: 348-349.

Smith, F. H. 1958. Spectrophotometric determination of total gossypol in cottonseed meats. J. Am.

Oil Chem. Soc. 35: 261-265.

Soomro, A. R. 2000. Assessment of useful heterosis in glandless (Gossypium hirsutum ) cotton

strains through their performance in hybrid combination. Pak. J. Bot. 32(1): 65-68.

159

Soomro, A. R., A. M. Memon, G. H. Nachnani and A. H. Baloch. 1998. Effect of okra leaf shape on

boll rot, earliness and yield of upland cotton (Gossypium hirsutum L.). Pak. J.

Scientific and Industrial Res. (Pakistan). 41(1): 36-38.

Sotelo, A., H. Villavicencio, I. Montalvo and M. T.Gonzalez. 2005. Gossypol content on leaves and

seeds from some wild malvaceae species. African J.Traditional, Complementary and

Alternative Medicines. 2(1): 4-12.

Steel, R. G. D. and J. H. Torrie. 1980. Principles and Procedures of Statistics, A Biometrical

Approach. McGraw Hill Book Co., New York, USA.

Stiller, W. N., P. E. Reid and G. A. Constable. 2004. Maturity and leaf shape as traits influencing

cotton cultivar adaptation to dryland conditions. Agron. J. 96: 656-664.

Stipanovic, R. D., D. W. Altman, D. L. Begin, G. A. Greenblatt, and J. H. Benedict. 1988. Terpenoid

aldehydes in upland cottons: analysis by aniline and HPLC methods. J. Agric. Food.

Chem. 36: 509–515.

Stoilova, A. and S. Taofik. 1998. Inheritance and heterosis related to yield and its components in

interspecific hybrids of cotton (Gossypium hirsutum × G. barbadense). Rasteniev dni-

Nauki. 35(2): 101-104.

Subhan, M., M. Qasim, D. A. Riaz and H. Niazi. 2002. Genetics of yield and its components in

upland cotton (Gossypium hirsutum L.). Asian J. Pl. Sci. 1(3): 250-253.

Subramanian, A., R. Ravikesavan, K. Iyanar, K. Thangaraj and P. Vindhyavarman. 2005.

Combining ability analysis in upland cotton. Plant Archives. 5(1): 23-28.

Sultan, M. K. B. N. Mitra, R. Choudhury, M. M. Kamruzzaman and M. A. Matin. 1999. Correlation

and path analysis in upland cotton (Gossypium hirsutum L.). Bangladesh J. of

Scientific and Industrial Res. 34(1): 55-58.

Syed, T. S., G. H. Abro, R. D. Khuro and M. H. Dhauroo. 1996. Relative resistance of cotton

varieties against sucking insect pests. Abstract No. PM-17.2nd International Congress

of Entomol. Sci., March 19-21, Pak. Entomol. Soc., PARC, Islamabad, pp.52.

Tabashnik, B., T. Dennehy, Y. Carrière, Y. Liu, S. Meyer, A. Patin, M. Sims and C. Ellers-kirk.

2003. Resistance Management: Slowing Pest Adaptation to Transgenic Crops. Acta

Agriculturae Scandinavica, Plant Soil Science. 53 (1): 51 - 56

160

Tang, B., J. N. Jenkins, C. E. Watson, J. C. McCarty and R. H. Creech. 1996. Evaluation of genetic

variances, heritabilities and correlations for yield and fibre traits among cotton hybrid

populations. Euphytica. 91(3): 315-322.

Tchatchueng, J. B., R. Kamga, G. J. Kayem and R. A. Oderinde. 1992. Dosage

spectrophotométrique du gossypol par complexation avec les ions molybdiques en

milieu acide. Cot. Fib. Trop. 47: 121–129.

Terry, P. T. World cotton market conditions. 2001. International Cotton Advisory Committee. 1629

K, Street NW, suit 702, Washington., D. C. 20006 U.S.A.

Thomson L. J. and J. A. Lee. 1980. Insect resistance in cotton: a review and prospectus for Australia.

J. Aust. Inst. Agric. Sci. 1980: 75–86.

Thomson, N. J., P. E. Reid and E. R.Williams. 1987. Effects of the okra leaf, nectariless, frego bract

and glabrous conditions on yield and quality of cotton lines. Euphytica 36: 545-553.

Tyagi, A. P. 1987. Correlation studies on yield and fibre traits in upland cotton

(Gossypium hirsutum L.). Ind. J. Agric. Res. 22(2): 75-80.

Tyagi, A. P. 1988. Genetic architecture of yield and its components in upland cotton

(Gossypium hirsutum L.). Ind. J. Agric. Res. 22(2): 75-80.

Ulloa, M. 2006. Heritability and correlations of agronomic and fibre traits in an okra-leaf upland

cotton population. Crop Sci. 46: 1508-1514.

Ulloa, M. and W. R. Meredith, Jr., 2002. Assessment of agronomic and fibre traits in an upland

cotton population from F2 to 3 to F2 to 6 generations. P 212-216. In Proc. Beltwide

Cotton Conf. Atlanta Georgia 8-12 Jan. 2002. Natl. Cott. Amer. Am., Memphis, T.N.

Vaden Bosch, R., 1972. The cost of poisons. Environment. 14(7): 18-22.

Van Dinther, J. B. M., 1972. Insect control and new approaches of world crops. M.Sc.Thesis, Dept.

Agric. Entomol. Univ. Agri., Faisalabad.

Vlessidis, A. G., M. E. Kotti and N. P. Evmiridis. 2004. A study for the validation of

spectrophotometric methods for detection, and of digestion methods using a flow

injection mainfold, for the determination of total phosphorous in wastewaters. J.

Analytical Chem. 59(1): 77-85.

161

Vyahalkar, G. R., N. L. Bhale and L. A. Deshpande. 1984. Inheritance of fibre traits in Gossypium

arboreum L. Ind. J. Agric. Sci. 54(9): 702-704.

Walker, G. P., E. T. Natwick (2006). Resistance to silverleaf whitefly, Bemisia argentifolii

(Hem., Aleyrodidae), in Gossypium thurberi, a wild cotton species Journal of Applied

Entomology 130 (8): 429–436.

Wang, J., X. Wang, F. Chen, P. J. Wan, G. Q. He and Z. G. Li. 2005. Development of competitive

direct ELISA for gossypol analysis. J. Agric. Food Chem. 53(14): 5513-5517.

Wanjura, D. F., R. V. Baker and L. L. Ray. 1976. Leaf and bract trichome density and boll type

influence on cotton lint grade index. Crop Sci. 16: 588-591.

Warner, J. N. 1952. A method for estimating heritability. Agron. J. 44: 427-430.

Watson, J. S. 1989. Recent progress in breeding for insect resistance in three cotton. In Cotton

Breeding, IInd edition, 2004 eds. Phundun Singh., Kalyani Publishers, New Delhi. pp:

136-146.

Wells, R. and W. R. Meredith. 1986. Normal vs. okra leaf yield interactions in cotton. II. Analysis of

vegetative and reproductive growth. Crop Sci. 26: 223-228.

Wendel, J. F. and V. A. Albert. 1992. Phylogenetics of cotton genus (Gossypium). Character state

weighted parsimony analysis of chloroplast–DNA restriction site date and its

systematic and biogeographic implications. In M.Sc. Thesis, Faculty of the Univ.

Georgia., Athens, Georgia.

White, W. H., J. N. Jenkins, W. L. Parrot, J. C. McCarty. Jr., D. H. Collum and P. A. Hendin. 1982.

Strain and within-season variability of various allelochemicals within a diverse group

of cottons. Crop Sci. 22: 1235-1238.

Williams, D. J. 1996. A brief account of the hibiscus mealybug Maconellicoccus hirsutus, a pest of

agriculture and horticulture, with descriptions of two related species from southern

Asia. Bulletin of Entomological Research. 86: 617–628.

Wilson, F. D. 1987. Pink bollworm resistant lint yield and earliness of cotton isolines in a resistant

genetic background. Crop Sci., 27: 957-60. In Cotton Breeding, IInd edition, 2004 eds.

Phundun Singh., Kalyani Publishers, New Delhi. pp: 136-146.

162

Wilson, C. D. 1986. Pink bollworm resistance, lint yield, and lint yield components of okra-leaf

cotton in different genetic backgrounds. Crop Sci. 26: 1164-1167.

Wilson, F. D. 1989. Yield, earliness and fibre properties of cotton carrying combined traits for pink

bollworm resistance. Crop Sci. 29: 7-12.

Wilson, F. D. and B. W. George. 1982. Effects of okra leaf, frego-bract, and smoothleaf mutants on

pink bollworm damage and agronomic properties of cotton. Crop Sci. 22: 798-801.

Wilson, F. D. and J. A. Lee. 1971. Genetic relationship between tobacco budworm feeding response

and gland number in cotton seedlings. Crop Sci. 11: 419-421.

Wilson, F. D. and J. N. Smith. 1976. Some genetic relationship between gland density and gossypol

content in G. hirsutum. Crop Sci. 16: 830–832.

Wilson, F. D., H. M. Flint, L. A. Bariola and C. C. Chu. 1991. Reduction in insecticide use

associated with cotton resistant to pink bollworm. Crop. Sci. 31: 363-366.

Wilson, R. L. and F. D. Wilson. 1976. Some genetic relationship between gland density and

gossypol content in G. hirsutum. Crop Sci. 16: 830–832.

Wright, R. J., P. M. Thaxton, K. M. El-Zik, and A. H. Paterson, 1999. Molecular mapping of genes

affecting pubescence of cotton. J . Hered. 90: 215–219.

Wu, Y. T., T. Z. Zhang, X. F. Zhu, and G. M. Wang. 2002. Relationship between F1, F2 hybrid yield,

heterosis and genetic distance measured by molecular markers and parent

performance in cotton. Agric. Sci. in China. 1(5): 498-507.

Wu, Z. B., P. Chen, Z. Li. Zhao and H. B. Xie. 2003. Studies on correlation of some important

characters of new lines in Gossypium hirsutum L. Acta- Agriculturae Shanghai.

19(2): 20-23.

Ying, Z. W. and M. Y. Jun. 2004. Analysis of genetic correlation between boll shape traits and fibre

quality traits in upland cotton. Acta Agronomica Sinica. 30(8): 816-820.

Younus, M., M. Yousaf and G. Jilani. 1980. Insect and spider mite pests of cotton in Pakistan. M.Sc.

Thesis, Dept. of Agri. Entomol., Univ. Agri., Faisalabad.

Yuan, Y. L., T. Z. Zhang, W. Z. Guo and J. J. Pan. 2002. Genetic stability of fibre qualities in upland

cotton. Cotton Sci. 14(2): 67-70.

163

Yuan, Y. L., Y. H. Chen, M. Tang, S. R. Jing, S. L. Liu, J. J. Pan, R. J. Kohel and T. Z. Zhang.

2000. Effects of the dominant glandless gene Gl2 e on agronomic and fibre characters

of Upland cotton. Plant Breeding. 118: 59-64.

Zhang, Z. S., X. B. Li, Y. H. Xiao, M. Luo, D. J. Liu, S. L. Huang and F. X. Zhang. 2003.

Combining ability and heterosis between high strength lines and transgenic Bt

(Bacillus thuringiensis) bollworm resistant lines in upland cotton

(Gossypium hirsutum L.). Agric. Sci. in China. 2(1): 13-18.

164

APPENDIX - I: Preliminary assessment of germplasm for okra leaf type, gossypol glands and trichomes

Variety/Accession Leaf type Hairiness Gossypol glands

Acala 63-74 Normal Less hairy Glandless

Acala 63-75 Normal Less hairy Low glanding

Bar F/8 Normal Medium hairy Normal glanding

Brant 2-41 Normal Less hairy Medium glanding

Brown Okra Okra Medium to high hairy Medium glanding

CIM 446 Normal Sparse hairy Medium glanding

Cocker Normal Low to medium hairy Low to medium hairy

Cocker- 304 Normal Less hairy Medium glanding

Ding Dong Normal Less hairy Medium glanding

DP-15-26 Normal Less hairy Medium glanding

DP-15-A Normal Medium hairy Medium glanding

DP-45 Broad/normal Medium hairy Medium glanding

DP-65 Normal Less hairy Medium glanding

F-280 Normal Sparse hairy Glandless

FH-1000 Normal Less hairy Medium glanding

HA-106 Normal Less hairy Medium glanding

HG-142 Normal/broad Less hairy High glanding

HGT- 469 Normal Less hairy Medium glanding

HR-107 NH Okra Sparse hairy Low glanding

HRVO-1 Okra Pilose Normal glanding

LR-4-B Normal Less hairy Low to medium glanding

Mex- 4 Normal Less hairy Medium glanding

Mex-10 Normal Less hairy Medium glanding

NIAB- Karishma Normal Less hairy Medium glanding

OKR-BWP Okra Medium hairy Medium glanding

Rebah-55 Normal Medium hairy Dense glanding

Seakland Normal Less hairy Medium glanding

UA-31-4 Normal Less hairy Dense glanding

V-16 Normal Less hairy Medium glanding

V-16 Normal Less hairy Low to medium glanding

Xli Normal Less hairy High glanding

165

APPENDIX - II

Analysis of variance for plant height for six generations in 3 crosses

Analysis of variance for plant height for six generations in HRVO-1 × FH 1000

Source d.f S.S M.S F-ratio Prob

Replication 2 3.95 1.97 1.61 0.25

Genotypes 5 114.37 22.87 18.63 0.0001

Error 10 12.27 1.23

Analysis of variance for plant height for six generations in HRVO-1 × CIM 446

Source d.f S.S M.S F-ratio Prob

Replication 2 2.38 1.19 3.15 0.086

Genotypes 5 178.22 35.64 94.22 0.0000

Error 10 3.78 0.37

Analysis of variance for plant height for six generations in HRVO-1 × Acala 63-74

Source d.f S.S M.S F-ratio Prob

Replication 2 9.13 4.56 2.47 0.13

Genotypes 5 135.43 27.08 14.68 0.0002

Error 10 18.44 1.84

166

APPENDIX - III

Analysis of variance for number of monopodial branches for six generations in

3 crosses

Analysis of variance for number of monopodial branches for six generations in HRVO-1 × FH 1000

Source d.f S.S M.S F-ratio Prob

Replication 2 0.17 0.08 3.20 0.084

Genotypes 5 0.95 0.19 6.98 0.004

Error 10 0.27 0.02

Analysis of variance for number of monopodial branches for six generations in HRVO-1 × CIM 446

Source d.f S.S M.S F-ratio Prob

Replication 2 0.01 0.005 3.47 0.071

Genotypes 5 0.14 0.027 19.60 0.0001

Error 10 0.014 0.001

Analysis of variance for number of monopodial branches for six generations in HRVO-1 × Acala 63-74

Source d.f S.S M.S F-ratio Prob

Replication 2 0.098 0.049 1.34 0.3028

Genotypes 5 4.730 0.946 26.15 0.0000

Error 10 0.362 0.036

167

APPENDIX - IV

Analysis of variance for number of sympodial branches for six generations in

3 crosses

Analysis of variance for number of sympodial branches for six generations in HRVO-1 × FH 1000

Source d.f S.S M.S F-ratio Prob

Replication 2 0.142 0.071 0.234

Genotypes 5 136.35 27.27 89.951 0.000

Error 10 3.032 0.303

Analysis of variance for number of sympodial branches for six generations in HRVO-1 × CIM 446

Source d.f S.S M.S F-ratio Prob

Replication 2 0.371 0.186 1.27 0.3212

Genotypes 5 125.313 25.063 172.13 0.0000

Error 10 1.456 0.146

Analysis of variance for number of sympodial branches for six generations in HRVO-1 × Acala 63-74

Source d.f S.S M.S F-ratio Prob

Replication 2 0.004 0.002 0.026

Genotypes 5 74.917 14.98 177.36 0.0000

Error 10 0.845 0.084

168

APPENDIX - V

Analysis of variance for number of bolls for six generations in 3 crosses

Analysis of variance for number of bolls for six generations in HRVO-1 × FH 1000

Source d.f S.S M.S F-ratio Prob

Replication 2 2.129 1.065 3.95 0.054

Genotypes 5 154.557 30.911 114.94 0.000

Error 10 2.689 0.269

Analysis of variance for number of bolls for six generations in HRVO-1 × CIM 446

Source d.f S.S M.S F-ratio Prob

Replication 2 0.354 0.177 0.96

Genotypes 5 65.825 13.165 71.80 0.0000

Error 10 1.833 0.183

Analysis of variance for number of bolls for six generations in HRVO-1 × Acala 63-74

Source d.f S.S M.S F-ratio Prob

Replication 2 0.073 0.037 0.504

Genotypes 5 33.168 6.634 91.455 0.0000

Error 10 0.725 0.073

169

APPENDIX - VI

Analysis of variance for seed cotton yield for six generations in 3 crosses

Analysis of variance for seed cotton yield for six generations in HRVO-1 × FH 1000

Source d.f S.S M.S F-ratio Prob

Replication 2 50.907 25.453 3.808 0.0591

Genotypes 5 6552.764 1310.553 195.767 0.0000

Error 10 66.944 6.694

Analysis of variance for seed cotton yield for six generations in HRVO-1 × CIM 446

Source d.f S.S M.S F-ratio Prob

Replication 2 2.262 1.131 0.586

Genotypes 5 2984.823 589.765 305.77 0.0000

Error 10 19.288 1.929

Analysis of variance for seed cotton yield for six generations in HRVO-1 × Acala 63-74

Source d.f S.S M.S F-ratio Prob

Replication 2 1.285 0.643 0.263

Genotypes 5 3258.645 651.729 267.197 0.0000

Error 10 24.391 2.439

170

APPENDIX - VII

Analysis of variance for boll weight for six generations in 3 crosses

Analysis of variance for boll weight for six generations in HRVO-1 × FH 1000

Source d.f S.S M.S F-ratio Prob

Replication 2 0.003 0.002 0.52

Genotypes 5 2.391 0.478 163.09 0.0000

Error 10 0.029 0.003

Analysis of variance for boll weight for six generations in HRVO-1 × CIM 446

Source d.f S.S M.S F-ratio Prob

Replication 2 0.002 0.001 0.580

Genotypes 5 1.643 0.329 181.57 0.0000

Error 10 0.018 0.002

Analysis of variance for boll weight for six generations in HRVO-1 × Acala 63-74

Source d.f S.S M.S F-ratio Prob

Replication 2 0.004 0.002 1.23 0.3322

Genotypes 5 4.218 0.844 468.04 0.0000

Error 10 0.018 0.002

171

APPENDIX - VIII

Analysis of variance for lint percentage for six generations in 3 crosses

Analysis of variance for lint percentage for six generations in HRVO-1 × FH 1000

Source d.f S.S M.S F-ratio Prob

Replication 2 0.454 0.227 0.778

Genotypes 5 100.366 20.073 68.85 0.0000

Error 10 2.915 0.292

Analysis of variance for lint percentage for six generations in HRVO-1 × CIM 446

Source d.f S.S M.S F-ratio Prob

Replication 2 0.468 0.234 4.085 0.0505

Genotypes 5 21.696 4.339 75.720 0.0000

Error 10 0.573 0.057

Analysis of variance for lint percentage for six generations in HRVO-1 × Acala 63-74

Source d.f S.S M.S F-ratio Prob

Replication 2 0.480 4.556 2.96 0.0975

Genotypes 5 22.782 0.081 56.27 0.0000

Error 10 0.810

172

APPENDIX - IX

Analysis of variance for fibre length for six generations in 3 crosses

Analysis of variance for fibre length for six generations in HRVO-1 × FH 1000

Source d.f S.S M.S F-ratio Prob

Replication 2 0.105 0.053 1.52 0.263

Genotypes 5 15.500 3.100 90.20 0.0000

Error 10 0.344 0.034

Analysis of variance for fibre length for six generations in HRVO-1 × CIM 446

Source d.f S.S M.S F-ratio Prob

Replication 2 0.155 0.077 3.183 0.0852

Genotypes 5 29.724 5.945 244.41 0.0000

Error 10 0.243 0.024

Analysis of variance for fibre length for six generations in HRVO-1 × Acala 63-74

Source d.f S.S M.S F-ratio Prob

Replication 2 0.052 0.026 0.62

Genotypes 5 1.183 0.237 5.59 0.0103

Error 10 0.423 0.042

173

APPENDIX - X

Analysis of variance for fibre strength for six generations in 3 crosses

Analysis of variance for fibre strength for six generations in HRVO-1 × FH 1000

Source d.f S.S M.S F-ratio Prob

Replication 2 0.240 0.120 1.85 0.2074

Genotypes 5 11.614 2.323 35.84 0.0000

Error 10 0.648 0.065

Analysis of variance for fibre strength for six generations in HRVO-1 × CIM 446

Source d.f S.S M.S F-ratio Prob

Replication 2 0.351 0.175 1.33 0.3085

Genotypes 5 18.456 3.691 27.91 0.0000

Error 10 1.322 0.132

Analysis of variance for fibre strength for six generations in HRVO-1 × Acala 63-74

Source d.f S.S M.S F-ratio Prob

Replication 2 0.006 0.003 0.11

Genotypes 5 1.063 0.213 7.42 0.0038

Error 10 0.286 0.029

174

APPENDIX - XI

Analysis of variance for fibre elongation for six generations in 3 crosses

Analysis of variance for fibre elongation for six generations in HRVO-1 × FH 1000

Source d.f S.S M.S F-ratio Prob

Replication 2 0.006 0.003 1.26 0.3243

Genotypes 5 0.234 0.047 19.22 0.0001

Error 10 0.024 0.002

Analysis of variance for fibre elongation for six generations in HRVO-1 × CIM 446

Source d.f S.S M.S F-ratio Prob

Replication 2 0.005 0.0025 0.49

Genotypes 5 1.692 0.338 69.07 0.0000

Error 10 0.049 0.005

Analysis of variance for fibre elongation for six generations in HRVO-1 × Acala 63-74

Source d.f S.S M.S F-ratio Prob

Replication 2 0.006 0.003 2.35 0.1454

Genotypes 5 1.386 0.277 203.79 0.0000

Error 10 0.014 0.001

175

APPENDIX - XII

Analysis of variance for fibre uniformity ratio for six generations in 3 crosses

Analysis of variance for fibre uniformity ratio for six generations in HRVO-1 × FH 1000

Source d.f S.S M.S F-ratio Prob

Replication 2 0.013 0.007 0.060

Genotypes 5 198.162 39.632 367.53 0.0000

Error 10 1.078 0.108

Analysis of variance for fibre uniformity ratio for six generations in HRVO-1 × CIM 446

Source d.f S.S M.S F-ratio Prob

Replication 2 0.235 0.118 2.75 0.1117

Genotypes 5 12.147 2.429 56.82 0.0000

Error 10 0.427 0.043

Analysis of variance for fibre uniformity ratio for six generations in HRVO-1 × Acala 63-74

Source d.f S.S M.S F-ratio Prob

Replication 2 0.015 0.007 0.26

Genotypes 5 2.655 0.531 18.74 0.0001

Error 10 0.283 0.028

176

APPENDIX - XIII

Analysis of variance for fibre fineness for six generations in 3 crosses

Analysis of variance for fibre fineness for six generations in HRVO-1 × FH 1000

Source d.f S.S M.S F-ratio Prob

Replication 2 0.008 0.004 2.019 0.1834

Genotypes 5 0.910 0.182 86.72 0.0000

Error 10 0.021 0.002

Analysis of variance for fibre fineness for six generations in HRVO-1 × CIM 446

Source d.f S.S M.S F-ratio Prob

Replication 2 0.014 0.007 3.13 0.0882

Genotypes 5 1.729 0.346 158.46 0.0000

Error 10 0.022 0.002

Analysis of variance for fibre fineness for six generations in HRVO-1 × Acala 63-74

Source d.f S.S M.S F-ratio Prob

Replication 2 0.003 0.002 0.762

Genotypes 5 1.961 0.392 175.16 0.0000

Error 10 0.022 0.002

177

APPENDIX - XIV

Analysis of variance for number of trichomes for six generations in 3 crosses

Analysis of variance for number of trichomes for six generations in HRVO-1 × FH 1000

Source d.f S.S M.S F-ratio Prob

Replication 2 16.439 8.220 1.062 0.3818

Genotypes 5 72788.532 14557.706 1880.82 0.0000

Error 10 77.401 7.740

Analysis of variance for number of trichomes for six generations in HRVO-1 × CIM 446

Source d.f S.S M.S F-ratio Prob

Replication 2 5.523 2.762 2.082 0.1755

Genotypes 5 98963.315 19792.663 14919.20 0.0000

Error 10 13.267 1.327

Analysis of variance for number of trichomes for six generations in HRVO-1 × Acala 63-74

Source d.f S.S M.S F-ratio Prob

Replication 2 299.734 149.867 2.88 0.1031

Genotypes 5 82486.009 16497.202 316.68 0.0000

Error 10 520.943 52.094

178

APPENDIX - XV

Analysis of variance for gossypol content for six generations in 2crosses

Analysis of variance for gossypol content for six generations in HRVO-1 × Acala 63-74

Source d.f S.S M.S F-ratio Prob

Replication 2 0.000 0.000 0.66

Genotypes 5 0.629 0.126 1257.83 0.0000

Error 10 0.001 0.000

Analysis of variance for gossypol content for six generations in HRVO-1 × HG-142

Source d.f S.S M.S F-ratio Prob

Replication 2 0.000 0.000 1.212 0.3379

Genotypes 5 0.546 0.109 559.35 0.0000

Error 10 0.002 0.000

179

APPENDIX - XVI

Analysis of variance for total gossypol for six generations in 2 crosses

Analysis of variance for total gossypol for six generations in HRVO-1 × Acala 63-74

Source d.f S.S M.S F-ratio Prob

Replication 2 0.000 0.000 1.266 0.3234

Genotypes 5 0.098 0.020 811.65 0.0000

Error 10 0.000 0.000

Analysis of variance for total gossypol for six generations in HRVO-1 × HG-142

Source d.f S.S M.S F-ratio Prob

Replication 2 0.000 0.000 1.212 0.3378

Genotypes 5 0.091 0.018 817.39 0.0000

Error 10 0.000 0.000

180

APPENDIX - XVII

Computation of the standard aliquots for the development of standard curve in HRVO-1 × Acala 63-74 (Normal × glandless)

Concentration of

stock solution Gossypol in

gossypol acetic acid in standard solutions (mg)

(mg) of gossypol = gossypol acetic acid

(mg) × 0.8962

OD (Optical density)

reading (A)

Corrected Absorbance =

(A-B)

Calibration factor = gossypol(mg)/

corrected absorbance

1mL 0.048 0.043 0.132 0.13 0.3308

2mL 0.096 0.086 0.188 0.19 0.4600

4mL 0.192 0.172 0.239 0.237 0.7257

6mL 0.280 0.251 0.292 0.29 0.8655

8mL 0.380 0.341 0.363 0.361 0.9446

10mL 0.480 0.430 0.418 0.42 1.0300

Blank OD reading = Zero Blank OD reading (A) – (Blank Aniline treated “B”) = 0.002

Mean 0.7261

APPENDIX - XVIII

Computation of the standard aliquots for the development of standard curve in HRVO-1 × HG-142 (Normal × High glanding)

Concentration of

stock solution Gossypol in

gossypol acetic acid in standard solutions (mg)

(mg) of gossypol = gossypol acetic acid

(mg) × 0.8962

OD (Optical density) reading

(A)

Corrected Absorbance =

(A - B)

Calibration factor = gossypol(mg)/

corrected absorbance

1mL 0.048 0.043 0.107 0.087 0.4943

2mL 0.096 0.086 0.17 0.15 0.5733

4mL 0.192 0.172 0.24 0.22 0.7818

6mL 0.280 0.251 0.31 0.29 0.8655

8mL 0.380 0.341 0.4 0.38 0.8974

10mL 0.480 0.430 0.49 0.47 0.9149 Blank OD reading = Zero Blank OD reading (A) – (Blank Aniline treated “B”) = 0.02

Mean 0.7545