AGROMORPHOLOGICAL CHARACTERIZATION AND EVALUATION

OF OKRA [Abelmoschus esculentus (L.) Moench] GENOTYPES AT

MELKASSA, CENTERAL ETHIOPIA

MSc THESIS

NESRU TEMAM

AUGUST 2019

HARAMAYA UNIVERSITY, HARAMAYA

AGROMORPHOLOGICAL CHARACTERIZATION AND EVALUATION
OF OKRA [Abelmoschus esculentus (L.) Moench] GENOTYPES AT
MELKASSA, CENTERAL ETHIOPIA

A Thesis Submitted to School of Plant Sciences,

Postgraduate Program Directorate
HARAMAYA UNIVERSITY

In Partial Fulfillment of the Requirements for the Degree of

MASTER OF SCIENCE IN HORTICULTURE

Nesru Temam

August 2019

Haramaya University, Haramaya

 HARAMAYA UNIVERSITY
POSTGRADUATE PROGRAM DIRECTORATE

We hereby certify that we have read and evaluated this Thesis entitled: “Agro morphological
Characterization and Evaluation of Okra [Abelmoschus esculentus (L.) Moench] Genotypes at
Melkassa, Centeral Ethiopia’’ prepared, under our guidance, by Nesru Temam. We recommended
that to be submitted as fulfilling the thesis requirement.

Wassu Mohamed (PhD) _______________ ______________

Major Advisor Signature Date

Shimelis Aklilu (PhD) ______________ ____________

Co-Advisor Signature Date

As a member of the Board of Examiners of the MSc Thesis Open Defense Examination, I certify
that I have read and evaluated the Thesis prepared by Nesru Temam and examined the candidate.
I recommended that the Thesis be accepted as it fulfills the Thesis requirement for the Degree of
Master of Science in Horticulture.

__________________ ______________ _____________

Chair Person Signature Date

_________________ _______________ _____________

Internal Examiner Signature Date

________________ ________________ ____________

External examiner Signature Date

Final approval and acceptance of the Thesis is contingent upon the submission of its final copy to
the council of Graduate Study (CGS) through the candidate’s department or school graduate
committee (DGC or SGC).

i
 DEDICATION
I dedicated this thesis to my beloved father Temam Hajifeto and my mother Zeynabu Haji
Mohammed for their immeasurable support, inspiration, guidance and love as well as contributed
for the success of my life.

ii
 STATEMENT OF THE AUTHOR

By my signature below, I declare and affirm that this Thesis is my own work. I have followed all
ethical and technical principles of scholarship in preparation, data collection, data analysis and
compilation of this Thesis. Any scholar matter that is included in the Thesis has been given
recognition through citation.

This MSc Thesis is submitted in partial fulfillment of the requirements for MSc degree in
horticulture at Haramaya University. The Thesis is deposited at the Haramaya University Library
and made available to borrowers under the rules of the Library. I solemnly declare that this MSc
Thesis work is not submitted to any other institution anywhere for the award of academic degree,
diploma or certificate.

Brief quotations from this Thesis are allowable without special permission provided that accurate
acknowledgement of source is made. Request for permission for extended quotation or for
reproduction of this manuscript in whole or in part may be granted by the head of the major
department or the dean of the school of Graduate studies when in his or her judgment the proposed
use of the material is in the interest of scholarship. In all other instances, however, permission must
be obtained from the author.

Name: Nesru Temam Signature_______________

Date: ________________

Schools: School of Plant Science

iii
 BIOGRAPHICAL SKETCH

The author was born in September 12, 1988 G.C in Gassera woreda, Bale zone, Oromia, Ethiopia.
He attended Elementary School Education at Baneba and Anesa elementary and High School
Education at Gassera high school, Bale zone. After completion of his High School Education, he
joined Alage ATVET College in September 2005 and completed his Diploma in Plant science in
July 2007. Soon after graduation, he employed at Raitu woreda as plant science expert and he
served from November 2007 to August 2012 and then he joined Mada Walabu University and
completed his Bsc education at 2015 in plant science and again back to Raitu working for 1 year.
Then he joined Ethiopian Biodiversity Institute as assistant researcher-I in February 2016 and
served until he joined the School of Graduate Studies at Haramaya University in pursue his MSc
degree.

iv
 ACKNOWLEDGEMENTS

First, I would like to thank the almighty God for giving me the strength and substance to complete
this thesis work; as well as for giving me opportunity to express my heartfelt gratitude to people
whose moral support and kind cooperation encouraged me during the course of study. I would like
to express my deepest gratitude and respect to my major advisor, Dr. Wassu Mohammed, for his
valuable contributions, guidance, patience, encouragement and constructive suggestions towards
the completion of this thesis. My appreciation also goes to my co-advisor, Dr. Shimelis Aklilu, for
his valuable comments during research work and preparation of Thesis write up.

A special thanks to my brother Adem Temam for direct encouragements and financial support.
Words fail me to express my gratitude from the very bottom of my heart to my wife Mandarina
Umer for being a part of my life and her beautiful patience during my career, Melkasa agricultural
research center for their provision of land as well as other necessary input and materials during
field work. I would like to thank Dajane Dida and Ayelech Woldatsadik staff member of
biodiversity for laboratory work mainly mucilage extraction. I would also like to thank Ethiopian
Biodiversity Institute for access to experimental material and financial support. My appreciation
and thanks goes to my colleagues and friends Wubshet Teshome, Kiflu Tarekegn, Temam Arebo
and Shibru Tilahun for their tough encouragement. Finally, I would like to acknowledge Haramaya
University, School of Plant Sciences, for giving me the chance to study my MSc Degree.

v
 ACRONYMS AND ABBREVIATIONS

ANOVA Analysis of Variance

DMRT Duncan’s Multiple Range

EBI Ethiopian Biodiversity Institute

EDHA Ethiopian Horticultural Development Association

FAO Food and Agriculture Organization

FAOSTAT Food and Agriculture Organization Statistics

IBC Institute of Biodiversity Conservation

IBPGR International Board for Plant Genetic Resources

IPGRI International Plant Genetic Resource Institute

MARC Malkasa Agricultural Research Center

Masl Meter above Sea Level

MoANR Ministry of Agriculture and natural Resource

PGRC Plant Genetic Resource Conservation

SAS Statistical Software

USDA United States Department of Agriculture

vi
 TABLE OF CONTENTS

STATEMENT OF THE AUTHOR iii

BIOGRAPHICAL SKETCH iv

ACKNOWLEDGEMENTS v

ACRONYMS AND ABBREVIATIONS vi

LIST OF TABLES ix

LIST OF TABLES IN APPENDIX x

LIST OF FIGURE xi

LIST OF FIGURE IN APPENDIX xii

ABSTRACT xiii

1. INTRODUCTION 1

2. LITERATURE REVIEW 4
2.1. Taxonomy, Origin, and Distribution of Okra 4
2.2. Biology of Okra 5
2.3. Growth and Ecological Requirement of Okra 7
2.4. Germplasm Management and Breeding of Okra Plant 8
2.5. Nutritional Composition and Economic Importance of Okra 9
2.6. Quantitative and Qualitative Traits of okra 11
2.6.1. Quantitative Traits of Okra Genotypes 12
2.6.2. Qualitative Traits of Okra Genotypes 13
2.7. Estimates of Genetic Variability Components in Okra 15
2.7.1. The Genotypic and Phenotypic Coefficient of Variation in Okra 15
2.7.2. Heritability and Genetic Advance 16
2.8. Association of Traits in Okra 18
2.8.1. Phenotypic and Genotypic Correlations 18
2.8.2. Path Coefficient Analysis 21
2.9. Genetic Divergence 22

3. MATERIALS AND METHODS 24

3.1. Description of Study Area 24
3.2. Experimental Materials and Design 24
vii
 3.3. Experimental Procedure 25
3.4. Data Collection 26
3.4.1. Crop Phenology and Growth Traits 26
3.4.2. Yield and Yield Components 28
3.4.3. Tender Fruit Quality Related Traits 29
3.4.4. Qualitative Traits 30
3.5. Data Analysis 31
3.5.1. Analysis of Variance and Descriptive Statistics 31
3.5.2. Variability Components 32
3.5.3. The Phenotypic and Genotypic Correlation Coefficient 35
3.5.4. Path Coefficient Analysis 36
3.5.5. Genetic Distance and Clustering 37
3.5.6. Principal Component Analysis 37

4. RESULT AND DISCUSSION 38

4.1. Analysis of Variance and Genotypes Performance of Okra 38
4.1.1. Crop Phenology and Growth Traits 38
4.1.2. Fruit Yield and Yield Components of Okra Genotypes 41
4.1.3. Fruit Quality Related Traits of Okra Genotypes 44
4.2. Qualitative Traits of Okra Genotypes 45
4.2.1. Plant Habit and Stem Related Traits 45
4.2.2. Leaf and Flower Characteristics of Okra 46
4.3. Estimation of Variability Components 50
4.3.1. Genotypic and Phenotypic Variations 50
4.3.2. Heritability and Genetic Advance 51
4.4. Association of Traits 54
4.4.1. Genotypic and Phenotypic Correlation of Fruit Yield with Other Traits 54
4.4.2. Genotypic and Phenotypic Correlations among other Characters 55
4.5. Path Coefficient Analysis 62
4.5.1. Genotypic Path Analysis 62
4.5.2. Phenotypic Path Analysis 64
4.6. Clustering and Genetic Distances of Okra Genotypes 65
4.6.1. Clustering of Genotypes 65
4.6.2. Genetic Distances of Genotypes 67
4.6.3. Cluster Mean Analysis 71
4.7. Principal Component Analysis 74

5. SUMMARY AND CONCLUSION 79

6. REFERENCES 82

7. APPENDICES 98

viii
 LIST OF TABLES

Table Page

1. List and description of okra genotypes 25

2. Schematic representation of analysis of variance (ANOVA) table of lattice design 32
3. Mean squares from analysis of variance for 14 phenology and growth traits
of 36 okra genotypes evaluated at Melkassa during 2018/19 40
4. Mean squares from analysis of variance for 11 tender fruit yield and yield
components as well as 2 quality related traits of 36 okra genotypes evaluated
at Melkassa during 2018/19 43
5. Estimates of range, mean and variability components of 36 okra genotypes
evaluated at Melkassa in 2018/19 53
6. Genotypic (above diagonal) and phenotypic (below diagonal) correlation
coefficients among 24 quantitative traits 60
7. Genotypic path analysis of 36 okra genotypes of 24 quantitative data those had
significant correlation with pod yield. 63
8. Phenotypic path analysis result of seven quantitative traits and had significant
correlation with pod yield 36 okra genotypes 64
9. Number of genotypes grouped in 13 clusters, genotype code, and place of
collection 36 okra genotypes evaluated at Melkassa in 2018/19 67
10. Range, mean, standard deviation and Coefficient of variation Euclidean
distances of 36 okra genotypes estimated from 24 quantitative traits
evaluated at MARC in year 2018/19 70
11. Mean values of 13 clusters for 24 quantitative traits of 36 okra genotypes
evaluated at Melkassa in year 2018/19 73
12. The principal component values of four principal component from 24 quantitative
traits for 36 okra genotypes evaluated at Melkassa in 2018/19 76

ix
 LIST OF TABLES IN APPENDIX

Appendix Table Page

1. Mean values of 36 okra genotypes for crop phenology and growth traits
evaluated at Melkassa in year 2018/19 97
2. Mean values of 36 okra genotypes for pod yield and related traits evaluated
at Melkassa in year 2018/19 99
3. The 13 Qualitative Traits of 36 Okra Genotypes Evaluated at Melkassa in
Year 2018/19 103
4. Description and score of qualitative traits of okra 104
5. Euclidean distances of 36 okra genotypes based on 24 quantitative traits
evaluated at Melkassa in 2018/19 106

x
 LIST OF FIGURE

Figure Page
1. Number of okra genotypes as distributed into categories of plant habit (a),
stem color (b) evaluated at Melkassa 2018/19 46
2. Number of okra genotypes as distributed into categories of leaf color (a),
leaf petiole color (b), leaf pubescence (c), flower color (d) and shape of leaf (e)
evaluated at Melkassa 2018/19 47
3. Number of okra genotypes as distributed into categories of Immature fruit
color (a), Position of fruit on stem (b), Fruit pubescence (c), Nature of
fruit base (d), Fruit shape (e) and Shape of seed (f) evaluated at Melkassa 2018/19 49
4. Dendrogram shows the dissimilarity of 36 okra genotypes based on 24
quantitative traits by using UPGMA 66
5. Biplot (axes PC1, PC2, PC3 AND PC4) OF 24 quantitative traits of 36 okra
genotypes evaluated at Melkassa 2018/19 78

xi
 LIST OF FIGURE IN APPENDIX

1. Fruit Shape Score (IBPGR, 1991). 105

xii
 AGROMORPHOLOGICAL CHARACTERIZATION AND EVALUATION OF OKRA
[Abelmoschus esculentus (L.) Moench] GENOTYPES AT MELKASSA, CENTERAL
ETHIOPIA

ABSTRACT

Okra [Abelmoschus esculentus (L). Moench] is one of indigenous and endemic genetic resource of
Ethiopia as well as Benishangul Gumuz. However, only few studies were carried out to assess its
diversity and performance throughout the country and specifically no research conducted to assess
diversity of okra within regional state. Therefore, this study was conducted with the objective to
assessing the genetic divergence and estimate the genetic variability components in okra genotypes
collected from regional state, and to determine the association of traits and the direct and indirect
effects of traits on fruit yield. A total of 36 genotypes of which 33 okra genotypes were collected
from different areas of Benishangul Gumuz Regional State, 3(three) checks, of 2 introduced and
1(one) released were evaluated for 27 quantitative and 13 qualitative traits at MARC in 2018/19
using simple lattice design (6 x 6). The results from analysis of variance revealed the presence of
significant variation for all quantitative traits except 3 traits. Moreover, the variation of genotypes
for fruit yield per hectare ranged from 9.44 to 32.88ton/ha-1 with mean 19.59ton/ha-1 for genotypes
29620 and 29618, respectively. Most of genotypes had high mean performance as compared to
checks for majority of traits. The majority of genotypes had almost desirable qualitative traits such
as color of fruit, pubescence and general growth habits. The genotypic (GCV) and phenotypic
(PCV) coefficient of variation were in range of 6.3 to 54.19% and 9.19 to 55.51%, respectively.
Heritability in broad sense (H2) and genetic advance as percent of mean (GAM) had ranged from
43.46 to 97.34% and 9.16 to 109.14%, respectively. The variability components (GCV, PCV, H2
and GAM) were high for all traits except few traits like days to 90% maturity, hundred seed weight
low in PCV and GCV. Number of ridge and hundred seed weight showed moderate H2 whereas
days to 90% maturity, number of ridge and percentage of mucilage content had moderate GAM
and low for hundred seed weight. It is found that number of primary branches, fruit diameter,
number of tender fruit per plant, seed yield, and number of harvests are correlated to fruit yield at
genotypic and phenotypic levels, whereas average fruit weight was correlated at phenotypic level.
Traits like number of primary branches, seed yield and average fruit weight had positive and
significantly high direct effect on fruit yield. Therefore, they should be considered during selection
for fruit yield improvement. The genetic distance of genotypes ranged from 2.83 to 12.24 with
mean, standard deviation and coefficient of variation 6.73, 1.63 and 24.18(%), respectively. All the
36 genotypes clustered in 13 distinct cluster consisting 11 (30.56%) in cluster I to seven cluster in
which genotypes are solitary clustered. Among cluster, cluster VII had the highest fruit yield by
leading other seven cluster greater than overall cluster means. Principal component analysis
revealed four principal components (PC1 to PC4) with eigenvalues ranged from 1.83 to 7.58 which
accounted for a total of 71.34% cumulative contributions. The result observed in these study was
the presence of a wide genetic variation among genotypes collected from Benishangul Gumuz
Regional State.
Keywords: Correlation, Direct Effect, Genetic Distance, Genetic variability, Principal Component

xiii
 1. INTRODUCTION

Okra [Abelmoschus esculentus (L.) Moench] belongs to family Malvaceae and it is very important
vegetable crop grown in tropical and sub-tropical parts of the world (Kisher et al., 2016). Okra is
proposed to be originated in Tropical Africa and it is native to North Eastern Africa in the area of
Ethiopia and Sudan from where it extensively spread to Asia, America, Southern Europe and other
countries (Oyelade et al., 2003; Santos et al., 2012). It is self-pollinated, mainly propagated by
seeds with a duration of 3 to 4 months (Muhammad et al., 2013; Osawaru et al., 2014).

Cultivated okra constitutes a major economic crop in Asia, West and Central Africa and in America
as a result of its vital importance as a component of various recipes in many cuisines and
preparations. It has a considerable area under cultivation in Africa and Asia in particular because
of its contribution to the human diet by supplying fats, proteins, carbohydrates, minerals, and
vitamins. Its mucilage is suitable for medicinal and various industrial applications (Lamont, 1999;
FAO, 2004; Saifullah and Rabbani, 2009; Haruna et al., 2016). The unripe green finger-like seed
capsule of okra, usually called “pod” are processed and consumed as stews and salads, soups,
sliced, boiled and fried vegetables (Akanbi et al., 2010; Daniela et al., 2012). The fruits contain
effortlessly digestible fiber, fat-free contents and low calories (Kumar and Sreeparvathy, 2010;
Reddy et al., 2013). Okra typically differs from most other common vegetables in having high
mucilage content (Jideani and Bello, 2009). The seed is used as a coffee additive or substitute
(Moekchantuk and Kumar, 2004).

There is no complete data on production area and productivity of okra in Ethiopia, although it is a
traditional crop in southwestern, western and northwestern Ethiopia (Miheretu et al., 2014a).The
crop is cultivated from landraces over the years in the country (Tesfa and Yosef, 2016). Recently
in 2016, the first improved variety (Bamia Humera) has been recommended for cultivation
(MoANR, 2016). Ethiopia is not benefiting from okra either for food and nutrition security as well
as foreign currency income generation as an export commodity. However, the few studies
conducted indicated the country is rich for the diversity of the crop (PGRC, 1995; Miheretu et al.,
2014a and b; Muluken et al., 2015, 2016; Tesfa and Yosef, 2016; Wassu et al., 2017) which is a
good opportunity to improve the crop through characterization and evaluation the available
germplasm in the country.
 2

The collection of desirable plant germplasm relies on the proven accession features and genetic
divergence, which are essential in genetic resources utilization (Olaoye et al., 2009; AdeOluwa and
Kehinde, 2011). Genetic diversity denotes the variability in different crop species, and its links
with accessions identification which is important in genebank curators (Osekitar and Akinyele,
2008; Bello et al., 2011). In Ethiopia, little attention has been given to okra research, conservation,
and development and it was recommended the need for immediate action to conserve okra
germplasm (Tesfaye, 2007). In this regard, morphological characterization of plants has been
recommended as the first step to be adopted prior to far-reaching molecular research and
biochemical analyses (Akash et al., 2013).

Improvement in plant breeding scheme leans on high genetic differences in the population and the
magnitude of inheritance of favorable attributes (Olawuyi et al., 2015). The variability obtained in
a population is apportioned into non-heritable and heritable parameters utilizing genetically related
components including heritability, the genotypic coefficient of variation and genetic advance,
which are the core for selection (Muluken et al., 2016; Seth et al., 2016). Progress and gain from
the selection in any breeding program depend upon the magnitude of useful variability present in
the population and the degree to which the desired traits are heritable. Yield is a complex
quantitative traits and controlled by several genes that interact with the environment. It is the
product of various factors known as yield components. Therefore, the efficiency of selection in any
breeding program mainly depends upon the knowledge of the association of the characters. The
knowledge of associations between yield and its related traits could appreciably enhance the
efficiency of the crop improvement through the utilization of the appropriate selection indices
(Adekoya et al., 2014; Aminu et al., 2016).

A wide variability recorded for pod yield among the genotypes of different geographic locations
has been reported in Ethiopia (Mihretu et al., 2014a and b; Muluken et al., 2015 and 2016; Tesfa
and Yosef, 2016; Wassu et al., 2017). The morphological variation observed in these studies is an
indicative of differences in the genetic makeup of germplasm. Although few of these studies
reported that some of the traits are highly heritable and could be used in the selection process during
improvement programs, further studies are required to determine the association of traits in okra.

The recently conducted research tried to characterize and reported the presence of diversity in okra
collection in Ethiopia (Miheretu et al., 2014a and b; Muluken et al., 2015 and 2016; Tesfa and
 3

Yosef, 2016; Wassu et al., 2017). However, these studies did not focus to assess the genetic
variability among okra genotypes in each major okra growing region in Ethiopia. Okra is
traditionally cultivated from landraces and low exchange of genotypes among farmers of
southwestern and western parts of Ethiopia (Mihretu et al., 2014a and b; Tesfa and Yosef, 2016).
In such situation, it is expected that the crop population in different regions is diverse with each
other. This is because genetic differentiation and thereby the presence of population structure is the
function of low dispersal rates between local populations (Slatkin, 1987). Pluess and Stocklin
(2004) indicated spatial isolation of populations adapted to different environments may result in an
increase in gene diversity between populations. Geographic distances and environment differences
are the two major causes of genetic diversity among plant populations (Slatkin, 1987; Nosil et al.,
2009). Even though, Benishangul Gumuz Regional State one of the potential producers of okra
among the major producers and many genotypes collected from regional state no one can try to
investigate the genetic variability as well as their performance for yield and related traits. Moreover,
among current experimental materials 24 genotypes are new collections. Therefore, research
focusing on the assessment of diversity among okra genotypes collected from Benishangul Gumuz
Regional State is important to generate additional information and fill the gap of insufficient
information generated on okra germplasm in the region. The information generated from such study
also helps to design appropriate okra breeding and germplasm conservation strategies in the region.
Therefore, this research is initiated and intended to attain the following objectives

Objectives

o To assess the genetic divergence and estimate the genetic variability components in okra
genotypes collected from Benishangul Gumuz Regional State, and
o To determine the association of traits and the direct and indirect effects of traits on fruit
yield.
 4

2. LITERATURE REVIEW

2.1. Taxonomy, Origin, and Distribution of Okra

Okra (Abelmoschus esculentus L.) belongs to the family Malvaceae, genus Abelmoschus. It is native
to tropical Africa where it serves as a staple vegetable crop (Kisher et al., 2016). The genus
Abelmoschus may be easily distinguished from the closely related Hibiscus by its 5-toothed calyx,
which splits longitudinally along a single suture at flowering. After flowering the whole bloom,
consisting of calyx, petals and staminal column, which are adnate at the base, drops together,
leaving the ovary and the epicalyx segments behind. The Genus Abelmoschus is now considered to
consist of nine or 10 species of which four are cultivated (Tripathi et al., 2011).

The naming of this vegetable crop is very diverse which is in English ladies finger and there are
similar names in Arabic and other languages known from the early 13th century onwards. In
French, okra is called gombo. In several vernacular languages in many Anglophone African
countries, the name okro is used rather than okra. The name derere is used for okra in Zimbabwe.
In Ethiopia, it has various names such as Kenkase for round pod type and Bamia for the longer type
by Rutana, Andeha by Gumzigna, Bamia in Oromiffa/Amharic (Schippers, 2000; Habtamu and
Negussie, 2014; Habtamu et al., 2015).

Many Author’s including Vavilov (1951); Santos et al. (2012); Muhammad et al. (2013); Sathish
and Eswar, (2013); Osawaru et al. (2014) reported that even there is a taxonomic puzzle, Crops
which have a part-Asian and a part-African origin, the common okra is proposed to be originated
in tropical Africa and it is native to North Eastern Africa in the area of Ethiopia and Sudan from
where it extensively spread to Asia, America, Southern Europe and other countries. Tesfaye (2007)
have indicated that okra might have been domesticated in Ethiopia and confirmed that the existence
of Abelmoschus ficulneus, the close wild relatives (ancestor) of the cultivated species (Abelmoschus
esculentus L.) in western Ethiopia. It is an annual fruit vegetable crop grown in tropical and sub-
tropical regions as well as in warmer temperate regions of the Mediterranean region (Dhankhar and
Mishra, 2009).
 5

2.2. Biology of Okra

Okra is diverse an annual, hardy, erect, and high-yielding plant, which varies in size, pod shape,
pigmentation, the degree of branching, a period of maturity, and plant height (Purquerio et al.,
2010; Muluken et al. 2016; Fozia, 2018). Okra plant is robust, erect, and variable in branching and
varying from 0.5 to 4.0 meters in height and stem is pigmented with green or reddish tinges color
(Tripath et al., 2011). Muluken et al. (2016) reported okra plant shows densely branched base
(DBB) characters were 92% in frequency, densely branched over and non-branched growth habit
4% for each among evaluated genotypes. In addition other authors Tesfa and Yosef (2016), Oppong
et al. (2011) and Wassu et al. (2017) reported that the growth habit and degree of branching of okra
germplasm are highly variable. About 59% of genotypes were erect type while 24.7 and 16.4 % of
accession characterized with medium growth and procumbent type of growth alignment,
respectively. Stem of okra was 32, 24, and 44% green, green with red patches and red/purple color,
respectively.
The seeds are dicotyledonous and they vary in shape, which is either round, kidney or spherical
with epigeal germination (Hamon et al., 1991and Ariyo, 1993). Almost relative result reported by
Oppeng et al. (2011) that among genotypes 24, 40 and 36% of shows round, spherical (oval) and
kidney(reniform) seed shape, respectively.

About 35-60 days after emergence the plant begins to flower and the flowers are axillary and
solitary, borne on a peduncle 2.0 – 2.5 cm long within the leaf axil (Valeriana, 2011). The monoic
flowers of okra are self-compatible (Martin, 1983; Hamon et al., 1989). The flower usually remains
open for a day. It is mostly self-fertilized; however, insects such as honeybees and bumblebees can
cross-pollinate. Okra is self-compatible, and passive self-pollination can take place in its
hermaphrodite flowers (Al-Ghzawi et al., 2003). Okra is often self-pollinated but cross-pollination
is possible. Okra had two distinct flower colors. Flower color for all accessions was red color at
both sides (Muluken et al., 2016). Wassu et al. (2017) reported three (12%) and 22 (88%) okra
genotypes had flowers with red color inside only and red color at both sides, respectively.

Okra genotypes shows 48% heart shapes, 28% broadly ovate shape, 16% star shaped and 8%
palmately lobed with serrated margin of leaf shape (Muluken et al., 2016). The leaves are dark
green in color (Kumar et al., 2010). Wassu et al. (2017) among 25 okra genotypes 60% shows
green with red vein and 40% had totally green leaf colors. Tesfa and Yosef (2016) reported three
 6

distinct leaf colors viz. 26.5% green with red veins, 60.3% totally green and 13.2% red colors. In
addition they also reported that leaf pubescence as 67.1% slight pubescence, 19.2% glabrous and
13.7% had conspicuous leaf pubescence among 50 tested genotypes.

Okra genotypes had high variability for fruit yield and reported that 72 and 28% genotypes had
green and yellow green immature fruit color, respectively. In addition they was reported position
of fruit on the main stem as erect 68% and intermediate 32% among genotypes studied (Muluken
et al., 2016). Tesfa and Yosef (2016) were also reported as 56.1% yellowish green immature fruit
color, 25.8% green, 12.1% green with red, 3% dark green and 3% dark red immature fruit colors
and they also identify fruit pubescence 29.6% downy, 35.2% slight rough and 35.2% prickly.
Another authors from Nigeria Adeoluwa and Kehinde (2011) reported that 60% erect and 40%
horizontal fruit position on main stem and 56% smooth and 44% rough fruit pubescence. They also
reported three distinct fruit color of green (42.85%), purple (48.57%) and green yellow (8.57%).
Oppeng et al. (2011) states fruit shape of okra under study was 1, 3, 4 and 15 respectively according
to fruit shape descriptors.

Several researchers have reported on the occurrence of heterosis in considerable quantities for fruit
yield and its various components (Weerasekara et al., 2007; Jindal et al., 2009). The nomenclature
of (Abelmoschus sp.) with varied chromosome numbers of both cultivated and wild species of
Abelmoschus genus as distinctly reported involved most species of the genus. The most frequently
observed number of chromosome is 2n = 66-144 for Abelmoschus esculentus L. cultivated type
(Hamon and Yapo, 1986). Although Dutta and Naug (1968) reported the chromosome numbers
2n=72, 108, 120, 132 and 144 are in regular series of polyploids with n=12. In general okra plant
had high genetic variability as to genetic diversity is primarily a function of sexual recombination.
During meiosis, homologous chromosomes undergo crossing over which results in appearance of
several new recombination. Different phenomena affect the genetic diversity in plants.
Evolutionary forces like selection, mutation, migration and genetic drift are the basis of crop
genetic diversity. These might be the cause of continuous changes in allelic frequency in the
population and influence the genetic diversity of the crop plant (Bhandari et al., 2017).
 7

2.3. Growth and Ecological Requirement of Okra

Okra is mainly propagated by seeds and has a duration of 90-100 days. It is generally an annual
plant. Its stem is robust, erect, and variable in branching and varying from 0.5 to 4.0 meters in
height. Leaves are alternate and usually palmately five-lobed, whereas the flower is axillary and
solitary. Okra plants are characterized by indeterminate growth. Flowering is continuous but highly
dependent upon biotic and abiotic stress. The plant usually bears its first flower one to two months
after sowing. The fruit is a capsule and grows quickly after flowering. The greatest increase in fruit
length, height and diameter occur during 4th to 6th day after pollination. It is at this stage that fruit
is most often plucked for consumption. The okra pods are harvested when immature and high in
mucilage, but before becoming highly fibrous. Generally, the fiber production in the fruit starts
from 6th day onwards of fruit formation and a sudden increase in fiber content from the 9th day is
observed (Nath, 1976). Okra plants continue to flower and to fruit for an indefinite time, depending
upon the variety, the season and soil moisture, and fertility. In fact, the regular harvesting stimulates
continued fruiting, so much that it may be necessary to harvest every day in climates where growth
is especially vigorous (Tripathi et al., 2011).

Okra is tolerant to a wide range of climatic condition (Akanbi et al., 2010). It requires a long, warm
and humid growing environment. It can be successfully grown in hot humid areas (Tripathi et al.,
2011). The crop grows well in hot weather, especially in the regions with warm nights (>20 °C)
(Anwar et al., 2011). It is sensitive to frost at its all growth stage and injured below 10 °C (Tripathi
et al., 2011). Okra needs an average optimum temperature of 20 °C to 35°C (Voss and Bell, 2007,
Sorapong, 2012, Tripathi.et al., 2011,) with 15 oC and 42 o C minimum and maximum temperature
respectively (Voss and Bell, 2007).

Okra grows best on well-drained sandy loam soils at altitudes up to 550-650 m a.s.l. Poorly drained
soils may result in drowning (low oxygen) of the plants. Okra prefers slightly acidic soils with a
pH between 5.8 and 6.5. On clay soils. Okra is very sensitive to soils with hardpan and soil
compaction can severely restrict plant growth. Okra is a hot weather crop. The optimum soil
temperature for growth is 24 to 32 °C, while the minimum soil temperature is 18 °C. Damping-off
and seed decay are likely at soil temperatures below 21 °C. Planting dates may vary with favorable
soil temperature (The Department of Agriculture, Forestry, and Fisheries of South Africa, 2012).
 8

2.4. Germplasm Management and Breeding of Okra Plant

Okra is minor and under‐utilized crops are often cultivated by farmers following traditional farming
practices. These crops are adapted to extreme edaphic and climatic conditions. However, due to
lack of genetic improvement, minor and under‐utilized crops give a lower yield. Hence, these crops
are not widely grown by farmers. Processing difficulties, economic and cultural changes have also
affected the wider use of minor and under‐utilized crops (IBC, 2012). According to information
obtained from Ethiopian Biodiversity Institute data base system up to year 2018 only 156 accession
of okra plant was conserved in gene bank (EBI data base, 2018).

Okra (Abelmoschus esculentus) has also recently emerged as important export vegetables (EIA,
2012). Ethiopia has favorable climate and edaphic conditions for the production of tropical, sub-
tropical and temperate vegetables in the lowlands (<1500 meters above sea level), Midlands (1500-
2200), and highlands (>2200), respectively (EHDA, 2011, 2012).

For a successful breeding program, the evaluation of germplasm is pre-requisite on which the future
line of action is based. The value of germplasm collection depends not only on the number of
genotypes it possesses, but also up on the genetic variability present in those genotypes for yield
and yield components. Yield and yield components, in general, are polygenic in nature and are
subjected to different degrees of non-heritable variation (Lush, 1940).

There is genetic diversity among Cultivated Ethiopian okra (Mihretu et al., 2014a and b; Muluken
et al., 2016). Genetic diversity is a broad term encompassing all the variability occurring among
different genotypes with respect to total genetic make-up of genotypes related to single species or
between species (Bhandari et al., 2017). Mating system of crop plants also affect genetic diversity.
Inbreeding reduces while out breeding increases genetic diversity. Genetic drift can lead to loss of
rare alleles thereby reduces genetic diversity. Gene flow is the phenomenon in which individuals
migrate from one population to another population and the migrants are able to breed successfully
with the members of the recipient population. Gene flow depends not only on migration, but also
on the ability of the migrants’ alleles to be passed to subsequent generations (Brooker, 2012). Gene
flow within population increases the genetic diversity as new alleles are introduced (Osawaru et
al., 2015). Genetic diversity is importance for selecting parents in combination breeding of different
autogamous crops to obtain transgressive segregants (Pradip et al., 2010).
 9

Genetic distance or diversity of genotypes is considered as a good start in plant breeding to improve
crops either by means of hybridization or direct selection of genotypes for their desirable traits.
The traits which had more contribution to total genetic divergence are under the control of additive
gene action and will offer a good scope of improvement through selection breeding (Muluken et
al., 2015). The breeder should adopt suitable breeding methodology to utilize both additive and
non-additive gene effects simultaneously, since varietal and hybrid development will go a long way
in the breeding programs especially in case of okra (PhaniKrishna et al., 2015). Weerasekar (2006),
and Shujaat et al.(2014) suggested that genetic variations is an important feature to get the
genotypes inherited the traits together to meet the diversified goals of plant breeding including
higher and quality yield, resistance to diseases, early maturing and wider adaptations. The high
yielding varieties in okra has been developed by exploiting the genetic diversity available in the
crop (Muluken et al., 2016).

In breeding programs of okra, the traits that need to be given emphasis include medium tall to tall
plants, short internodes, low position of first flowering and fruiting node, high number of fruiting
nodes and early maturity for enhanced productivity, medium long to short green, smooth (downy),
five ridged, angular, strait fruits with blunt tip for enhanced fruit quality and appearance and
tolerate to biotic stresses for stable and sustainable production (Thirupathi et al., 2012a).

2.5. Nutritional Composition and Economic Importance of Okra

FAO/WHO recommends 400 gram of fruits and vegetables per day or alternatively five servings a
day; at least two servings of fruits and three servings of vegetables (WHO, 2003). Moreover,
experts recommend consumption of fresh, frozen, dried, or canned fruits and vegetables of a variety
of colors and kinds (Deckelbaum, 1999; USDA, 2000).

In recent years, awareness of the nutritional and health benefits of vegetables in Ethiopia has been
increasing due to public health advocacy on the role of vegetables in human nutrition and health
through its provision of antioxidants such as vitamin A, C and E that are important in neutralizing
free radicals (oxidants) known to cause cancer, cataracts, heart disease, hypertension, stroke and
diabetes (Demissie et al., 2009; Getachew and Mohammed, 2012). Increasing consumption of
vegetables helps to fight hidden hunger, malnutrition. Vegetables are also used as a source of raw
 10

material for the local processing industry and are produced for exports making a significant
contribution to the national economy (Shimelis, 2000 and Baredo, 2013).

Okra mucilage has the potential for use as food, none food production, and medicine application.
Mucilage from okra is used in medicine as protective food additive against irritating and
inflammatory gastric diseases and also used to stabilize blood sugar by regulating the rate at which
sugar is absorbed from the intestinal tract (Gemede et al., 2014; Saurabh et al., 2015). The fruit
contents comprises of 9.7 % carbohydrate, 86.1 % water, 1.0 % fiber, 0.8 %, 0.2 % fat and 2.2 %
protein (Saifullah and Rabbani, 2009). Furthermore, the unripe pods are very rich sources of
potassium, vitamins, calcium, and composition of okra leaves per 100 g edible portion is: water
81.50 g, energy 235.00 kJ (56.00 kcal), protein 4.40 g, fat 0.60 g, carbohydrate 11.30 g, fibre 2.10
g, Ca 532.00 mg, P 70.00 mg, Fe 0.70 mg, ascorbic acid 59.00 mg, β-carotene 385.00 μg, thiamin
0.25 mg, riboflavin 2.80 mg, niacin 0.20 mg (Gopalan et al., 2007; Varmudy, 2011). The seeds can
be dried and the dried seeds are a nutritious material that can be used to prepare vegetable curds or
roasted and ground to be used as coffee additive or substitute (Moekchantuk and Kumar, 2004).

Okra fruit is an important vegetable in the tropics and subtropics (Jideani and Bello, 2009). The
fruits of okra has different shape, length, color and other attributes and mucilage which is thick
slimy polysaccharides and is used to thicken soups and stews (Ahiakpa et al., 2014; Biswal et al.,
2014). Mucilage is a plant hydrocolloid which is a polymer of a monosaccharide or mixed
monosaccharide (Deogade et al., 2012). It is a common constituent of okra (Abelmoschus
esculentus) plant (Kaewmanee et al., 2014).

Okra typically differs from most other common vegetables in having high mucilage content
(Jideani and Bello, 2009). The food applications include a whipping agent for reconstituted egg
whites, an additive in the formulation of flour-based adhesives, an additive for clarifying sugarcane
juice. It is also used to modify the food quality in terms of food stability, texture, and appearance
properties by acting as an emulsifier, thickener, gelling agent, or texture modifier (Noorlaila et al.,
2015).

Habtamu et al. (2018) were reported the mean mucilage content (2.43 g/100 g) of Ethiopian okra
is higher than the value reported by Adetuyi and Dada (2014) for Nigerian okra (1.4 g/100 g). Okra
mucilage also contributes to improved functionality especially water-binding, emulsifying, and
 11

foaming properties of food products (Jideani and Bello, 2009). Mucilage from okra contains
significant levels of protein, carbohydrate, neutral sugars, minerals, and other complex
polysaccharides (Ahiakpa et al., 2014). It is medically proven to be linked to anticancer,
antimicrobial, hypoglycemic, anti-ulcer activities (Ansari et al., 2005), as well as its ability to bind
cholesterol and bile acid carrying toxins by filtering the liver (Shui and Peng, 2004). The chemical
compositions, molecular structures, monosaccharide sequences, glycoside linkages configuration,
and position in the backbone and side chains are some of the factors that can affect the functional
properties of natural plant mucilage (Mirhosseini and Amid, 2012).

Okra seeds contain about 20 to 40% oil and also protein (Sorapong, 2012). Okra seed oil yield is
comparable to most oilseed crops except oil palm and soybean (Sanjeet et al., 2010). Moreover,
okra seed oil has a potential hypocholesterolemic effect. The potential for wide cultivation of okra
for edible oil as well as for cake is very high (Sanjeet et al., 2010). Okra seed flour could also be
used to fortify cereal flour (Adelakun et al., 2009). For example, supplementing maize with okra
meal increases protein, ash, oil and fiber content (Akingbala et al., 2003). Okra seed flour has been
used to supplement corn flour for a very long time in countries like Egypt to make the better quality
dough (Sanjeet et al., 2010).

2.6. Quantitative and Qualitative Traits of okra

Morphological characters usually vary with environments and evaluation of traits requires growing
the plants to full maturity prior to identification. The high degree of wide morphological variation
exhibited among accessions of okra which is measured qualitatively and quantitatively (Tesfa and
Yosef, 2016). The significant variations among okra genotypes may be due to either environmental
factors or variations in the genetic potential of genotypes (Shujaat et al., 2014). Muluken et al.
(2016) reported the presence of genetic distance among Ethiopian okra collections. The result
indicated the presence of wide genetic diversity among tested okra genotypes which suggested the
importance of a further study on the genetic diversity of okra genotypes in the country which allows
for the identification of promising accessions for okra breeding in Ethiopia.
 12

2.6.1. Quantitative Traits of Okra Genotypes

Phenology and growth traits recorded includes days to 50% emergence, days to first flowering,
days to 50% flowering, days to 90% maturity, number (frequency) of harvest, plant height (cm),
stem diameter (mm), number of primary branches per stem, number of internodes, internodes
length (cm), leaf length (cm), leaf width (cm), number of epicalyxes and fruit peduncle length (cm).
Fruit and yield traits were recorded from fruit length (cm), fruit diameter (mm), average fruit weight
(g), number of tender fruits per plant, number of ridges on fruit, yield per plot (kg), yield per hectare
(t/ha), number of seeds per pod and hundred seed weight (g). Okra fruit quality related like dry
matter and mucilage contents are also be studied (IPGRI, 1991).

The yield of okra is a complex quantitative trait, which is conditioned by the interaction of various
genotypic and environmental components of growth and physiological processes throughout the
life cycle (Johnson et al., 1955; Adeniji and Peter, 2005). Mihretu et al. (2014b) reported
quantitative traits showed highly significant differences (P< 0.01) for all traits. The variation in
traits studied was two times of the minimum value for a number of seed per pod, three-fold for a
number of pod per plants, fivefold for fruit diameter and six-fold for fruit per plot.

According to Tesfa and Yosef (2016) many quantitative traits are an important measure of yield
and they are a very high variability in almost all Ethiopian okra accessions. Some of these are first
fruit producing node (3 to 25), plant height (64 to 258 cm), fruit length (51 to 193mm), fruit
diameter (16 to 43mm), number of seed per pod (21.3 to 136.7), total fruit per plant (313.3 to
6698.7g) and average fruit weight (6.47 to 61.67g) in their mean values. A wide range of flowering
periods among the accessions and indeterminate growth trait of okra plant resulted in varying
maturity periods even on the same plant making it difficult for harvesting and practically unfeasible
for mechanization. However, these type of cultivars are appropriate for home gardening for
continuous harvest. A number of pods per plant, days to flowering and plant height are some of the
most variable quantitative characters of okra (Singh and Singh, 1977). Traits such as days to 50%
flowering, stem diameter, number of seed per pod, number of pod per plants, internodes length,
internodes number and plant height contributed to the variation due to genetic diversity for use in
improvement program of okra (Mihretu et al., 2014). A number of fruits per plant, fruit weight,
and total fruit production often show high variability as compared to other quantitative traits
(Ahiakpa et al., 2013).
 13

Muluken et al. (2016) reported the mean square of most of the traits studied a high significant (P<
0.01) genotypic differences to days to pod formation, days first flowering, days to 50% flowering,
days to maturity, plant height, stem diameter, number of branch, leaf length, leaf width, number of
epicalyx, fruit length, fruit diameter, fruit weight, fruit ridge, fruit yield, number of mature pod,
fresh weight of mature pod and dry weight of mature pod. On the other hand, genotypes exhibited
significant (P<0.05) differences for days to 50% emergency, number of fruit per plant, number of
seeds per pod and hundred seed weight. However, non-significant differences among genotypes
were observed for internode length, a number of epicalyx per flower and peduncle length
characters. The analysis of variance result indicated the presence of substantial variability among
Ethiopian okra collections for most of the characters studied.

Aminu et al. (2016) reported that plant height at harvest varied from 1.11 to 1.49 m with a mean of
1.25 m, the range observed for days to 50% flowering was 33.00–50.33 days with the overall mean
of 46.82 days which revealed that dissimilar okra cultivars take a significantly different number of
days to flower and that cultivars were morphologically different from one another in flower-bearing
habits. Branches per plant were phenotypically different. The mean fresh pod length and fresh pod
diameter ranged between 11.49 and 14.88 cm and between 1.22 and 1.84 cm respectively. The
highest fresh weight per pod was (16.88 g) and the lowest (13.14 g).

2.6.2. Qualitative Traits of Okra Genotypes

Qualitative descriptors for okra are grouped into three main types; color, shape and other features.
The descriptors tend to be highly subjective (Hammon and Van Stolen, 1989). Muluken et al.
(2016) reported that plant habit, flower color, leaves color, leave petiole color, pod color, stem
color, leave shape, fruit position, fruit shape, and fruit pubescence are 10 most widely used for okra
genotypic evaluation.

The okra genotypes collected from different regions and countries may distributed into varied
categories of plant habit, stem color and plant alignment and proportion of genotypes in each
category may varied either due to inherent characteristics differences and environmental
differences where the genotypes grow or selection of different genotypes by human being.
Evolutionary forces like selection, mutation, migration and genetic drift are the basis of crop
 14

genetic diversity (Bhandari et al., 2017). There may be also differences of data registration because
the descriptors tend to be highly subjective (Hammon and Van-Stolen, 1989).

Stem of okra plant is robust, erect, and variable in branching and varying from 0.5 to 4.0 meters in
height and stem is pigmented with green or reddish tinges color (Tripath et al., 2011). Muluken et
al. (2016) reported densely branched base (DBB) characters were 92% in frequency, densely
branched over and non-branched growth habit 4% for each. Oppong et al. (2011), Tesfa and Yosef
(2016) and Wassu et al. (2017) reported that the growth habit and degree of branching of okra
germplasm are highly variable. About 59% of accessions are erect type while 24.7 and 16.4 % of
accession characterized with medium growth and procumbent type of growth alignment,
respectively. Stem of okra was 32, 24, and 44% shows green, green with red patches and red/purple
color, respectively. In addition they reported that most okra genotypes have densely branched at
base characters followed by densely branched overall and branched at apex (Wassu et al., 2017).

Okra had great variations of leaf related traits and different scholars reported the presence of
variability for leaf shape, color and others. Wassu et al. (2017) reported that among 25 okra
genotypes 60% shows green with red vein and 40% had totally green leaf colors while Tesfa and
Yosef (2016) reported three distinct leaf colors viz. 26.5% green with red veins, 60.3% totally green
and 13.2% red colors. They also reported that leaf pubescence as 67.1% slight pubescence, 19.2%
glabrous and 13.7% had conspicuous leaf pubescence among 50 tested genotypes. Muluken et al.
(2016) reported that genotypes shows 48% heart shapes, 28% broadly ovate shape, 16% star shaped
and 8% palmately lobed with serrated margin of leaf shape. Hammon and Van Stolen (1989)
reported 3 colors of the leaf as light green, green and deep green.

Wassu et al. (2017) also reported 22 (88%) and three (12%) okra genotypes had flowers with red
color at both sides and red color inside only, respectively. Muluken et al. (2016) reported that all
okra accessions showed uniform red at both side flower colors. Oppong-Sekyere et al. (2011) also
reported that purple, green and green with red veins petiole color representing 48%, 44%, and 8%,
respectively. They also identified the position of pod on main axis as 60% intermediate or unique
orthotropy axis and 20% erect position, 12% horizontally while only 4% hanging down.

Fruit showed seven different types of shape from short and triangular to long straight or long curve.
It's fruits with shape scores of 1, 2, 3, 4, 12, 14 and 15, according to the descriptor (IPGRI, 1991).
 15

Fruits with characteristics such as smooth, spineless, slender with green (light or dark) skin are
very desirable (Sinnadurai, 1992; Duzyaman, 2005). Adeoluwa and Kehinde (2013) reported that
three distinct fruit color 42.85% green, 48.57% purple and 8.57% green-yellow fruit color. Fruits
position on the main stem of the accessions showed 68% fruits positioned erect and 32% fruits
positioned intermediate. Fruit pubescence showed two variations among the okra accessions. Fruit
pubescence scored as smooth and rough among okra accessions 56% of all accessions bored smooth
fruit and 44% of accessions bored rough fruit (Muluken et al., 2016).

Anteneh (2017) observed the ranges of 26.45 to 54.03% and 4 to 25.33% for dry matter content
and mucilage contents of fruits of 25 okra genotypes evaluated at Dire Dawa, respectively. As
reported by Habtamu et al. (2017) the mean mucilage content (2.43 g/100 g) of Ethiopian okra is
higher than the value of Nigerian okra (1.4 g/100 g).

2.7. Estimates of Genetic Variability Components in Okra

2.7.1. The Genotypic and Phenotypic Coefficient of Variation in Okra

The hereditary parameters genotypic (GCV) and phenotypic coefficient of variations (PCV) can be
categorized as low (<10%), moderate (10-20%) and high (> 20%) (Sivasubramanian and
Madhavamenon, 1973). The large magnitude of PCV and GCV with a small difference between
the two heredity parameters indicated a smaller amount of environmental influence on the
phenotypic expression. Several researchers reported the consistent differences of okra cultivars due
to cultivars and environmental interactions (Thirupathi et al., 2012; AdeOluwa and Kehinde, 2013;
Ehab et al., 2013 and Adekoya et al., 2014; Aminu et al., 2016). The wide variability recorded for
fruit yield among the genotypes shows that there is a copious opportunity for selection and
improvement. This variability is possibly heritable and could be used in the selection processes
during improvement programs (Muluken et al., 2016).

Adeoluwa and Kehinde, (2011) reported low PCV and GCV for 100 seed weight and numbers of
ridges per fruit. Ehab et al. (2013) and Aminu et al. (2016) reported that number of pods per plant,
fruit length, plant height, the number of primary branches, fresh pod length, fresh pod yield per
plant and fresh seed per pod exhibited high values more than 20 % for both PCV and GCV with a
considerably low degree of variation between them. Muluken et al. (2016) also reported that the
GCV and PCV ranged between 3.08 ‒ 26.3% and 5.5 – 34.14% for number of epicalyx and number
 16

of pod, respectively. Sibsankar et al. (2012) states low PCV and GCV values for days to first
flowering and numbers of ridge per fruit.

The slight range of difference and hindered possibility for selection of these characters.
Furthermore, the least GCV and PCV estimate of characters implied higher impacts environmental
conditions on these characters. Therefore selection based on phenotypic basis will not be useful
for the genetic progress of the crop (Sankara and Pinaki, 2012; Thirupathi et al., 2012a; Ehab et
al., 2013 and Kishor et al., 2016). The number of pods per plant, fruit length, plant height, the
number of primary branches, fresh pod length, fresh pod yield per plant and fresh seed per pod
exhibited high values more than 20 % for both PCV and GCV with a considerably low degree of
variation between the two (Ehab et al., 2013 and Aminu et al., 2016). Many researchers noticed
that high magnitude of GCV and PCV inferred a low environmental manifestation effects on the
characters which probably increase greater improvement prospects through selection scheme
(Swati et al., 2014; Kishor et al., 2016 and Muluken et al., 2016). Thus, selections of favorable
characters by utilizing high phenotypic and genotypic estimates could be exploited in enhancing
the characters during the breeding program. The progress from selection does not only depend on
the knowledge of interrelationships of yield and yield contributing characters but it is also useful
in the efficient selection of characters (Aminu et al., 2016).

2.7.2. Heritability and Genetic Advance

In worry of improvement, the breeder thought as a matter of fact about crossing the best lines with
a hypothesis that it is possible to accumulate in a same genotype favorable genes present in different
genotypes. However, it is known that before we turn our attention to breeding methods in self-
pollinated crop species, we have to consider the degree to which a trait is transferred from
generation to generation. Heritability has been widely used to assess the degree to which a character
is transmitted from one parent to the offspring. More often in biological research, broad and narrow
sense heritabilities are used to evaluate the proportion of heredity and environment in the
expression of a character (Adeniji, 2007).

The heritability estimates provide a background genetic information on the inheritance of traits
from parents to progenies as well as the amount of genetic progress that can be made in selection
 17

of a trait or traits of interest (Sabesan et al., 2009). Genetic advance under selection is helpful in
detecting the actual genetic gain expected (Ogunniyan and Olakojo, 2014).

Heritability is the only component which is transmitted to the next generation, calculated as the
ratio of genetic variance to phenotypic variance. The heritability is a good index of the degree of
transmission of the characters from parents to their offspring (PhaniKrishna et al., 2015) and the
consistency in the performance of progeny in succeeding generations depends mainly on the
magnitude of the heritable portion of variation. Heritability indicates the possibility and extent to
which improvement can be brought about through selection and it may be of broad and narrow
senses (Robinson, 1966).

Heritability estimates reported were only based on broad sense and hence the total genetic variance
may include dominance and epistatic components which are not available for selection. High
heritability coupled with high genetic advance as percent of the mean was more valuable in
predicting the effect of selection because heritability is a single numerical expression on the ratio
of the two variances which may not lead to success if the selection is based on heritability estimates
alone (Johnson et al., 1955). High heritability along with moderate genetic advance in days to pod
formation, stem diameter, leaf length, fruit diameter, and fruit ridges while moderate values for
both heritability and genetic advance for days to first flowering, leaf width, 100 seed weight and
peduncle length (Muluken et al., 2016). The traits mentioned above could be exploited through the
manifestation of dominance and epistatic components through heterosis (Yassin and Anbu, 1997).

Célestin et al. (2012) reported a wide range of heritability values for most characters ranging from
0.17 to 0.63 in which fruit diameter (0.63) had the highest broad sense heritability followed by fruit
length (0.61) and seeds number per pod (0.50). Concerning the narrow-sense heritability, the
highest value was recorded by the fruit length character (0.55), fruit diameter (0.54) and the seeds
number per pod character (0.47), respectively. The other parameters had low values and the lowest
of these values were presented by the fruit number per plant (0.08).

The evaluations of the two heritabilities being raised but differing in intensity for all considered
characters, which indicates the presence of the additive and non-additive genetic variance (Esmail
et al., 1999). An elevated heritability mentions the dominant role of the additive effects of the genes
in the expression of the aforesaid characters (Yiguo, 2005). In other words, the values raised of the
 18

heritability of the characters suggest that the selection based on the phenotypic performance could
be efficient. The high values of the heritability indicate the higher prospects in the future programs
of okra improvement for the characters such as diameter of the fruit, length of the fruit and number
of seeds per pod. It is important to apply some backcross to concentrate these characters in
genotypes because a lot of traits appear to be controlled by the genes with additive effects.
Knowledge of heritability involved in the expression of traits and the existence of a correlation
between traits would facilitate the further improvement of okra (Subrata et al., 2004; Célestin et
al., 2012).

2.8. Association of Traits in Okra

The yield of a crop plant is the result of the interaction of a number of inter-related characters (Ren
et al., 1995). The correlation coefficients among the quantitative traits in the accessions of okra
selection for a single character may increase the value of the trait which is positively correlated and
declines the values for negatively correlated traits (Ahiakpa et al., 2014). Correlation of yield and
other desirable characters with another character should also be studied because sometimes the
selection on the basis of the component character of yield may not be effective due to low
heritability. Therefore, it is necessary to make a selection on the basis of character other than the
yield contributing characters. The progress from selection does not only depend on the knowledge
of interrelationships of yield and yield-contributing characters but it is also useful in the efficient
selection of characters (Mehta et al., 2006).

All the fruit traits like fruit length, fruit width, fruit weight, total number of fruits per plant and
number of fruits per plant showed significant positive association with fruit yield per plant and also
among themselves except fruit width (Bello et al., 2006; Mehta et al., 2006; Reddy, 2013). Positive
association of fruit length with fruit weight and total yield per plant was reported by Dakahe et al.
(2007) and Jaiprakashnarayan and Mulge (2004), respectively.

2.8.1. Phenotypic and Genotypic Correlations

The correlation between the different characters shows the existence of both positive meaningful
and significant and negative meaningful and non-meaningful values. A positive and meaningful
correlation translates the existence of a high relation between the parameters put in the inscription.
 19

On the other hand, the existence of a negative meaningful correlation between two parameters
suggests that all growth of one induced a reduction of the other. Therefore, the increase in the length
of the fruit is followed by a reduction of the diameter of this fruit. In the same way, a precocious
genotype (of which the day of 50% of flowering is reached quickly) presents long fruits. The traits
that exercise negative and non-meaningful correlation one on the other can be difficult to select
and even should not be taken in account in a program of varietals improvement (Celestin et al.,
2012; Henry and Krishna, 1990). Selection for a single character may increase the trait values of
positively correlated characters and decline the values for negatively correlated traits (Ahiakpa et
al., 2014; Tesfa and Yosef, 2016).

Phenotypic correlation (rp) measures the extent to which the two observed characters are linearly
related while genotypic correlation (rg) measures the extent to which degree the same genes or
closely linked genes cause co-variation (simultaneous variations) in two different traits (Singh and
Chaudhary, 1977; Falconer et al., 1996; Mehta et al., 2006).

Muluken et al. (2016) reported fruit yield t ha-1 had positive and highly significant (P<0.01)
genotypic correlation with days to pod formation, number of branches per plant, leaf length, and
number of tender fruit per plant. Mihretu et al. (2014) also reported that average fruit weight, fruit
diameter, seed yields, hundred seed weight and a number of tender fruit per plant positively and
significantly correlated at (P<0.01) to fresh pod yield genotypic level. Simon et al. (2013) reported
pod yield had a positive and highly significant genotypic correlation with seed yield (0.75).
Oppong-Sekyere (2012) reported 100 seed weight (0.282) were positive and significant (P< 0.05).
While number of total fruits per plant revealed significantly high and positive correlation.

Célestin et al. (2012) reported a highly significant and positive correlations between fruit length
and fruit peduncle length, seeds number per pod and plant height, seeds number per pod and fruit
length, fruit diameter and 50(%) flowering day, fruit diameter and plant height, fruit diameter and
seeds number per pod. Highly significant and positive correlations were also reported between stem
diameter and plant height, stem diameter and seeds number per pod, stem diameter, and fruit
diameter. Significant and positive correlations were exhibited by plant height and 50 % flowering
day, fruit peduncle length and plant height, seeds number per pod and fruit peduncle length, stem
diameter and 50 % a flowering day, 100 seeds weight and seeds number per pod. Only one highly
 20

significant and negative correlation was exhibited by fruit length and 50 % a flowering day.
Significant and negative correlation exists between fruit diameter and fruit length.

Simon et al., (2013) and Chattopadhyay et al. (2011) reported a highly significant positive
correlation between the mean seed yield and pod weight; both genotypic and phenotypic. Pod yield
was highly significant and positively correlated with pod length and pod weight phenotypically,
indicating that selection based on these traits could significantly improve total yield. According to
the report, the number of pod per plant and pod yield influenced the total yield of okra. Due to
variations in plant height, length of fruit and other morphological structure of different lines the
individual fruit weight of different lines were varied from each other (Chandra et al., 1996). Hazra
and Basu (2000) reported that genotypes differed significantly for individual fruit weight.

Tesfa and Yosef (2016) reported there was a positive correlation between first fruit-producing node
(FFPN) with maximum plant height, a number of seeds per fruit with mature fruit length and
commercial fruit, fruit yield with fruit numbers. Rath et al. (1991) reported that the improvement
of okra through phenotypic selection would be effective for characters like plant height, number of
branches per plant, number of nodes per the main stem, fruit length, yield per plant, and seeds per
fruit.

The phenotypic and genotypic correlation coefficients of the okra traits showed a highly significant
and positive correlation of the number of pods yield per plant with plant height (r = 0.61), days to
50(%) flowering (r = 0.72), fresh weight per pod (r = 0.44), and pod length (0.43) this indicated
that increasing these attributes could invariably increase pod yield (Aminu et al., 2016). Rashwan
(2011) reported pod yield per plant was positively associated with the number of branches per plant,
number of pods per plant, and pod length. He also stated traits that are phenotypically associated
but not genotypically correlated will be relevant for selection because selection is sometimes due
to the phenotypical performance of traits. The significant relationship between plant height and pod
yield in flowering also revealed that yield could be improved via direct selection of plant height
during flowering, as a single trait could be practicable during crop improvement. Days to 50%
flowering were positive and highly significantly associated with plant height (r = 0.99) and the
number of pods per plant (r = 0.45) and fresh weight per pod (r = 0.55). The days to 50% flowering
that positively correlated with plant height morphologically indicated that when the internodes’
formation ceases at floral initiation the early maturing okra cultivars usually have short status.
 21

Subrata et al. (2004) reported that days to the first flower and fruit yield per hectare varied
significantly among the genotypes. Significant positive correlations among yield and yield
contributing traits that would increase yield are suitable in crop improvement programs (Adeoluwa
and Kehinde, 2011).

Days to pod maturity had positive and significant correlation with days to 50% flowering, a number
of branches as well as negative and significantly correlated to fruit length (Muluken et al., 2016).
Akinyele and oseikita (2006), Saitwal et al. (2011), Reddy et al. (2013) reported that plant height
had positive and significant phenotypic and genotypic correlation with stem diameter, number of
branches per plant, number of internodes per plant, fruit diameter, number of fruit per plant and
number of seeds per pod. Increasing internode number might increase the number of tender fruit
per pod this is because a fruit of okra is usually borne in the leaf axil at nodes of the stem (Aminu
et al., 2016).

Reddy et al. (2013), Mihretu et al. (2014b), Muluken et al. (2016), Aminu et al. (2016), and Kisher
et al. (2016) reported that seed yield is correlated to number of tender fruit per plant, fruit length,
number of seed per pod, pod yield, number of primary branches and hundred seed weight.

2.8.2. Path Coefficient Analysis

The correlation coefficients were classified into indirect and direct effects utilizing the path
coefficient analysis as suggested by Dewey et al. (1959). Path coefficient analysis was developed
to study the relationship between two characters through their direct and by the way of the indirect
influence of the other characters (Simon et al., 2013). Path coefficient analyses revealed that fruit
weight, the total number of fruits per plant and number of marketable fruits per plant had a strong
influence on marketable pod yield per plant and are the main determiners of marketable pod yield
per plant (Reddy et al., 2013). Kaul et al. (1978) observed that seed yield per plant followed by a
number of primary branches per plant had a high direct effect on pod yield per plant. Kumar and
Reddy (1982) observed that number of fruits per plant had the highest direct effect on seed yield
followed by plant height, a number of primary branches and days to flowering. Ariyo et al. (1987)
reported that edible pod weight had the highest positive direct effect on fruit yield with its largest
indirect effect through reduction on fruit width.
 22

Reddy et al. (2013) reported that number of branches per plant, internodal length, fruit length and
total number of fruits per plant had a negligible positive direct effect at phenotypic level, while
plant height, days to 50% flowering; first fruiting node and fruit width had a negligible negative
direct effect on marketable yield per plant. Number of marketable fruits per plant and total yield
per plant had a high positive direct effect on marketable yield per plant. Fruit weight, the total
number of fruits per plant and number of marketable fruits per plant had a high positive direct
effect, while total yield per plant had a high negative direct effect on marketable pod yield per
plant.

Rambabu et al. (2019) reported number of plant branch, number of tender fruit per plant, seed yield
had high positive direct effect on pod yield and even other with low and negligible direct effect had
moderate and positive high indirect effects via other traits. Prasath et al. (2017) reported that
number of tender fruit per plant, number of the fruit harvest, hundred seed weight and seed per pod
had a direct positive effect on pod yield. Aminu et al. (2016) reported the number of pods per plant
had the greatest direct influence on pod yield, followed by fresh weight per pod, which had a
positive genotypic association with pod yield. Trait having positive and high indirect effect on pod
yield and high association at genotypic level suggests the relevance of considering this trait
indirectly during selection. But direct selection of this trait might be not effective (Subrata et al.,
2004; Adeniji et al., 2007). Number of pods per plant had the greatest direct influence on pod yield
(p = 6.63), followed by fresh weight per pod (p = 6.13), which had a positive genotypic association
with pod yield (Aminu et al., 2016).

2.9. Genetic Divergence

Multivariate genetic clustering method is a very useful techniques to assess the genetic diversity
among different strains/varieties or entries of a species and to provide most reliable information
about the real genetic distance of the genotypes (Singh and Pawar, 2005). Euclidean or straight line
measure of distance is the most commonly used statistics for estimating genetic distance between
individuals (genotypes or populations) by morphological data. Mohammadi and Prasama (2003)
have described in different measures of genetic distance in detail.

Fozia (2018) reported that the genetic distance of all possible pairs of 24 okra genotypes that ranged
from 1.96 to 11.36 with the mean, standard deviation and coefficient of variation of 5.85, 1.97 and
 23

33.75%, respectively (Table 9).The highest genetic distances (Euclidean distance) was computed
between the released okra variety, 23793 (Bamya-Humera) and 245162 (11.36) followed by
between Humera 01 and 245162 (11.28) and between Humera 01 and 240207 (10.69). Conversely,
the lowest genetic distances (Euclidean distance) was estimated between exotic genotypes.
Generally, 43 (15.58%) pair of genotypes had genetic distances <3.88, 110 (39.86%) pair of
genotypes had genetic distances between 3.89-5.85, 74 (26.81%) pairs of genotypes had genetic
distance between the interval 5.86 to 7.83 while 40 (14.49%) pair of genotypes had genetic
distances between 7.84 to 9.79 and 9(3.26%) pair of genotypes had genetic distance >9.79. She
also reported that genetic distances among the introduced varieties were lower than genetic
distances among genotypes collected from Ethiopia ranged from 1.96 to 10.01.

Muluken et al. (2016) reported that the genetic distances for all possible pairs of 25 okra accessions
ranged from 3.89 to 18.24 with mean and standard deviation of 9.57 and 3.44, respectively. The
most distant pairs of accessions were Indian okra variety and Benishangul (Asossa) okra accession.
The more close pairs of accessions were obtained from Benishangul (Metekel) and Gambella okra
collection respectively. They also states that Ethiopia okra collections exhibited wide genetic
distances in the range between 5.16 and 11.14 while exotic varieties had 11.95 to 13.78. This
suggested that the higher chance of improving the crop production through collection,
characterization, evaluation and selection or hybridization of okra genotypes from different regions
of Ethiopian other than introducing okra varieties from other countries. Mihretu et al. (2014a) also
reported the existence of considerable genetic distance among Gambella okra collections.

Wassu et al. (2017) reported the Euclidean distances of 300 pair of 25 genotypes evaluated at Dire
Dawa and reported the genetic distance ranged from 3.1 to 12.6 with 7, 2.2 and 27.85% mean,
standard deviation and coefficient of variation, respectively. Mihretu et al. (2014a) also reported
very high genetic distance among local collections.
 24

3. MATERIALS AND METHODS

3.1. Description of Study Area

The study was conducted at Melkassa Agriculture Research center (MARC), Ethiopia in 2018/19
main cropping season. Malkasa is located 8024’N latitude and 39021’E by having a distance of
around 112 K.M from Addis Abeba on the Eastern direction at an altitude of 1550 m.a.s.l. The area
is characterized by low and erratic rainfall with a mean annual rainfall of 763 mm with peaks in
July and August. The dominant soil type of the center is andosol of volcanic origin with pH that
ranges from 7 to 8.2. The mean annual temperature is 21.20C with a minimum of 140C and a
maximum of 28.40C (MARC, 2019 http://www.eiar.gov.et/marc).

3.2. Experimental Materials and Design

A total of 36 genotypes were evaluated of which 33 okra genotypes were collected from different
areas of Benishangul Gumuz Regional State of Ethiopian by Ethiopian Biodiversity Institute and 2
(two) of the varieties were introduced from India and now registered as commercial variety in
Ethiopia by one company and 1 (one) variety is released from Humera research center (Table 1).
The okra genotypes were collected at different altitudes ranging from 661 to 1518 m.a.s.l. The three
registered varieties will be used as the standard checks. Genotypes was evaluated on the field in 6
x 6 simple lattice designs. Each plot had 0.8 m x 5.4 m (4.32 m2) consisting of one row and a total
of 12 plants per row or per plot. The spacing between plant, plots and adjacent replications were
0.45, 0.8 and 2m, respectively. Three seeds were sown and thinned to one plant per hill when plants
reached 4-5 leaves stage.
 25

Table 1. List and description of okra genotypes

Accession Collection Altitude Accession Collection Altitude

No No
code woreda (m.a.s.l) code woreda (m.a.s.l)
1 29408 Guba 752 19 242433A Assosa 1480
2 29409 Menge 1100 20 242449A Assosa 1395
3 29410 Menge 1100 21 242445A Assosa 725
4 29411 Kurmuk 737 22 29051 Tango 1289
5 29412 Kurmuk 735 23 29413 Assosa 1358
6 29622 Assosa 661 24 29615 Assosa 1309
7 29414 Kurmuk 1346 25 29625 Kurmuk 1160
8 29415 Balojiganfoy 1192 26 29617 Assosa 1383
9 29416 Balojiganfoy 1195 27 29624 Bambasi 1459
10 29418 Guba 929 28 29620 Assosa 1518
11 29616 Assosa 1338 29 29621 Assosa 687
12 29618 Assosa 1491 30 29623 Assosa 691
13 29052 Assosa 1142 31 29619 Assosa 1497
14 T240204 Mandur 1200 32 242451A Assosa 1105
15 29417 Menge 1358 33 T242444 Menge 1105
16 T242443 Menge 935 34 SOH701 Standard check
17 240207A Dibate 1400 35 SOH714 Standard check
18 240209A Mandur 1050 36 BamiaHumera Standard check
Source: EBI 2018.

3.3. Experimental Procedure

Land preparation: The land was deeply ploughed and cross ploughed with the help of tractor and
power tiller for the experiment. After tractor plough, the land was further prepared by hand
breaking up the soil clods, levelled and pulverized for bringing a good tilth which is necessary for
proper plant growth and root development. The ridges was made according to the proper planting
spacing by ridger mounted tractor.

Planting: Available planting materials (36 genotypes) were planted in early August 2018 as soon
as the start of the rainy season and when the soil is becoming moist to support seed emergence.

Cultural practices: Weeding, cultivation, watering, and earthing-up was done at the appropriate
time to facilitate root growth, to control weed infestation and control waterlogging. Integrated
weeding and hoeing were done two times in order to improve soil structure and reduce competition
of weeds and earthingup were done as required to prevent exposure of roots to direct sunlight and
 26

to give anchorage due soil washed by irrigation water during irrigating the field. Supplementary
irrigation was applied depending upon the rainfall condition. Other cultural practices were applied
as per the usual practices used.

Harvesting: When the plants reached commercial maturity i.e. when tender fruits are observed,
ten plants from each row or plot, leaving the plants growing at both ends of each row to avoid edge
effects was harvested to estimate tender fruit yield and other yield-related traits.

3.4. Data Collection

International Plant Genetic Resources Institute (IPGRI, 1991) descriptor list for okra species were
used to record data on quantitative and qualitative traits. Quantitative traits were recorded from 10
plants per row leaving the two plants grown at both ends of the row as border plants and the two
border plants were used for mature pod and seed traits measurement. Five randomly selected tender
fruits from each harvest in each plot were used to record tender fruit characters.

3.4.1. Crop Phenology and Growth Traits

Days to 50% emergence: This refers to the time required in days for the shoot to emerge from the
seeds above the soil. It was recorded by counting the number of days from planting to the
emergence of 50% of plants seedling in each plot.

Days to first flowering: The number of days was taken from the date of sowing to onset of the
first flower appears on the plant in each plot.

Days to 50% flowering: This refers to the time required in days for the okra plants to flower. It
was recorded by counting the number of days from planting to the flowering of 50% of plants in
each plot.

Days to 90% maturity: Refers to the time required by the plant to reach the stage when 50% of
the plants in each plot start to senesce. Days to maturity was recorded when 90% of the plants in
each plot were ready for harvest as indicated by senescence of pods and leaves. The average number
of days from sowing to the date of the first harvest of 10 sample plants of the rows was recorded.
 27

Number of harvests: It was registered for each genotype by counting the number of tender fruits
harvested from each plot.

Plant height (cm): The height of the plants from the ground surface to the tip of the main stem
was measured in centimeter at the final harvest and the plant height of 10 randomly selected plants
were averaged to record the mean plant height in centimeter for statistical analysis.

Stem diameter (cm): The diameter of the stem at the basal region was measured at the final harvest
using a standard graduated scale Vernier caliper. The diameter of 10 randomly selected plants were
averaged to record the mean stem diameter in centimeter.

A number of primary branches per stem: The total number of primary branches per plant was
counted at final picking an average of 10 plants was calculated.

A number of internodes: The total number of internodes per plant was counted at final picking an
average of 10 plants were calculated.

Internodes length (cm): The length of the internodes between the fifth and sixth node was
measured at the time of maturity before the first tender fruit harvest.

Leaf length (cm): The length of 15 leaves on the main stem from each plot was sampled randomly
from the ninth and eleventh node when the plants come into flowering. As the leaves from 7th node
onwards are representative of the shape and the size of the variety. Leaf was measured from the
attachment of the base of the leaf petiole to the tip of the leaf using a ruler.

Leaf width (cm): The width of 15 leaves on the main stem from each plot were sampled randomly
from the ninth and eleventh node when the plants came into flowering. Leaves were measured from
the widest part of the leaf.

Number of epicalyxes: The number of epicalyxes flowers of the five samples per plant at the
flowering stage from each plot were counted and recorded and finally their mean used for
computing of data.

Peduncle length (cm): Pedicel length of the five fruits per plot prior to picking were measured at
the fully matured stage and described as 1) - from 1-3 cm, 2) - more than 3 cm
 28

3.4.2. Yield and Yield Components

Fruits were harvested every three days and the number and weight of all tender fruits were recorded
in each harvest. Data from each harvest were summed at the end of the growing season to estimate
the number of tender fruits/plant and tender fruits weight/plant. Average tender fruits weights were
calculated by dividing the total weight of tender fruits from all harvest to the total number of tender
fruits harvested. Five to seven randomly chosen tender fruits from each harvest in each plot and
totally not less than fifty tender fruits from each plot were used to record average tender fruits
diameter, length, and pedicel length.

From two plants at both ends of each row/plot, all mature pods those produced between the 6th and
20th nodes were harvested at the end of the growing season to collect data on mature pod length,
seed number/pod and 100 seed weight. Data collection procedures and traits description of yield
and yield components are listed below.

Fruit length (cm): Fruit length was measured as the distance from the fruit cap scar at the base to
the tip end of the pod (USDA, 1997). The length of five tender fruits per plot in each harvest was
measured from the base of the calyx to the tip of the fruit. The average was calculated by dividing
the sum of all tender fruits length by the total number of fruits measured.

Fruit diameter (mm): The diameter of the fruits in the peduncle insertion zone was measured
using a standard graduated scale Vernier caliper. The five tender fruits per plot which fruit length
was measured the average was calculated like that of the fruit length.

Average fruit weight (g): Each of five tender fruits per plot that were used to measure fruit length
and width was weighed using sensitive balance and the average weight of tender fruit was
calculated and recorded accordingly.

A number of tender fruits per plant: Fruits of ten plants in each plot at each harvest was counted
and summed at the end of the harvest and the average number of tender fruits per plant was
calculated for statistical analysis.

Number of ridges on fruit: The number of ridges was counted and the average was also calculated
from five tender fruits per plot at each harvest that was used to measure fruit length and width
which described as 1) - 5-edged, 2) - more than 5-edged
 29

Yield per plot (kg): Weight of tender fruits from each plot in each harvest were recorded and
summed to record yield per plot.

Yield per hectare (t/ha): It was estimated from the 10 plants tender fruit yield in each plot.

Number of seeds per pod: Ten fully matured and dried pods were collected randomly from the
two plants in each plot as indicated above and seeds were extracted, counted and the average
number of seeds per pod was computed to be considered for statistical analysis.

Hundred seed weight (g): Seeds extracted from matured pods as indicated above was kept in the
open air under the sun and the dried100 seeds were randomly counted and weighed to estimate 100
seeds weight.

Seed yield per plant (g): Seeds were extracted from the dry pods of two border plants in each plot,
kept in the open air under the sun for five days, then at drying room for 15 days, the weight of seeds
were measured and the average weight of seed per plant adjusted to 10% moisture content and
registered accordingly.

Seed yield per hectare (kg): Seed yield per hectare was calculated from the mean seeds yield of
the plant in a plot.

3.4.3. Tender Fruit Quality Related Traits

Dry matter content of tender fruit (%): Five tender fruits from three harvests were randomly
taken from each plot, weighed at harvest, sliced and dried in an oven at 750C until a constant weight
was attained and the dry matter in percent was calculated according to Williams and Woodbury
(1968) as follows.

𝑤𝑒𝑖𝑔ℎ𝑡 𝑜𝑓 𝑠𝑎𝑚𝑝𝑙𝑒 𝑎𝑓𝑡𝑒𝑟 𝑑𝑟𝑦𝑖𝑛𝑔(𝑔)

Dry matter(%) =
𝑖𝑛𝑖𝑡𝑖𝑎𝑙 𝑤𝑒𝑖𝑔ℎ𝑡 𝑜𝑓 𝑠𝑎𝑚𝑝𝑙𝑒(𝑔)

Estimation of mucilage content of okra fruits: A known weight of okra fruits harvested from
experimental plots were initially dried under shade for 24 hours and further dried in an oven at
500C until constant weight obtained. The dried fruits were powdered in grinder and mucilage
extraction was done in two steps (Uzma et al, 2013). Step 1: Powdered fruit (200 g) keep in 500
ml distilled water for 6 hours and then heated with continuous stirring at 600C for 2 hours. The
 30

concentrated solution filters through a muslin cloth and allows cooling. Step 2: Treat the extract
with an equal volume of acetone (100 %) for 30 minutes and filter through a muslin cloth. The
residue which has a high fraction of mucilage was dried in an oven at 400C then cool to room
temperature, weight and finally, the percentage of mucilage in fruit samples were calculated.

3.4.4. Qualitative Traits

The qualitative traits was recorded on a plot basis according to the International Plant Genetic
Resources Institute (IPGRI, 1991) descriptor list for okra species as follows.

Plant habit: This was identified how the plants in each plot branched and described as: (1) Densely
branched at the apex (DBA), (2) Densely Branched Base (DBB) (3) Densely branched all over
(DBO).

Flower color: Red coloration of petals base was assessed at both side and described as (1) red color
inside only or (2) red color on both sides.

Leaf color: This were assessed from leaves lamina and ribs and described as (1) totally green and
(2) green with the red vein.

Leaf petiole color: This was assessed from petioles color at both sides and described as (1) Green,
(2) Red above but green below and (3) Red on both sides.

Immature fruit color: The main color of the pods were observed at harvesting stage and described
as (1) Yellow-Green, (2) Green, (3) Green with red patches, (4) Red, (5) Dark-red and (6) other.

Stem color: This was assessed from stems color of plants at first harvest stage and described as:
(1) Green (2) Green with a red patch and (3) Red or Purple. The color chart was used for all color
identification of pod, stem, and leaf (http://w3schools.comhtml/html color full asp).

Shape of leaf: This was assessed from leaves of plants that was produced up to the first harvest
and described as (1) oval undulate (2) heart-shaped (3) broadly ovate (4) star-shaped (palmately
lobed) (5) palmately triangular lobes (6) palmately lobed with dentate margins (7) palmately lobed
with serrated margins and (8) linear-oblong or triangular lobes.
 31

The shape of seed: It is recorded after harvesting the matured crop and it described as 1) round, 2)
reniform, and 3) other specified

Position of fruits on the main stem: The position of fruits on the main stem was observed and it
was described in five distinct variations as (1) erect (2) intermediate (3) horizontal (4) slightly
falling and (5) totally falling.

Fruit shape: It is assessed from fruits at harvest stage and described as with shape scores of 1, 2,
3, 4, 12, 14 and 15, according to the descriptor (IBPGR, 1991).

Fruit pubescence: this was observed at harvesting stage and described as 1) smooth and 2) rough.

Leaf pubescence: It is observed at fully grown leaf stages and described as 1) glabrous, 2) slight
and 3) conspicuous.

Nature of fruit base: 1) ringed or protrude, 2) ring less or flat, and 3) sunken.

3.5. Data Analysis

3.5.1. Analysis of Variance and Descriptive Statistics

The quantitative data were subjected to analysis of variance (ANOVA) and computed with SAS
statistical software (9.0) (SAS, 2004). The quantitative data were collected in simple lattice
(partially balanced or incomplete block) design and analysis of variance was computed considering
the general linear model as follows.

Yijk= μ+ ri+ Gj+ b(r)ik+ + eijk;

Where;
Yijk= Yield
μ = grand mean
ri= effect of the ith replication
Gj= effect of the jth genotype
b (r)ik= effect of the ith block in kth replication
eijk = random error effects
 32

Table 2. Schematic representation of analysis of variance (ANOVA) table of lattice design

Source of DF Sum of Mean square Computed F Expected mean

variation squares (SS) squares

Replication r-1 SSR MSR 2 + kb2 + vr2

Treatment k2 - 1 SST (unadj.) MST (unadj.) 𝑘

σ2 (𝑘+1) mσt2
(unadj.)

Blocks within r(k-1) SSB (adj.) MSB (adj.) 2 + 2t+ kb2

replication (adj.)

Intrablock error (k-1)(rk-k-1) SSE MSE 𝑎2

RCB Error (t-1) (r-1) SSe MSe 2e
Total rk2 - 1 SSTO

r = Number of replications. k2 = Number of treatments, k = Number of plots in a block, SS = Sum

square, MS = Mean square, 2 = Variance, t = Number of genotypes, MSe = Mean squares for error
and 2e= Error variance.

Relative efficiency of square lattice to randomized complete block design (RCBD) was computed
as:
Mean square error in square lattice design
𝑅𝑒𝑙𝑎𝑡𝑖𝑣𝑒 e𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑐𝑦 = 𝑥100
Mean square error in RCBD

The mean performance of genotypes comparison was done following the significance of mean
squares using Duncans Multiple Range (DMRT) at P<0.05. The traits that exhibited significant
mean squares in general ANOVA were further subjected to genetic analyses. Based on mean square
expectations from the analysis of variance, phenotypic and genotypic coefficients of variations,
heritability and genetic advance was calculated as follows.

3.5.2. Variability Components

The phenotypic and genotypic variability of each quantitative trait was estimated as phenotypic
and genotypic variances and coefficients of variation. The genotypic variance was estimated
 33

according to the method suggested by Burton and De vane (1953) considering mean square
expectations from the analysis of variance as follows:

Msg−Mse
𝜎2 𝑔 = ( )
r

Where;

σ2g= genotypic variance

Msg= genotype/treatment mean square

Mse= error mean square

Phenotypic Variance (𝜎 2 𝑝) = 𝜎 2 𝑔 + 𝜎 2 𝑒

Where, 𝜎 2 𝑝 = phenotypic variance

𝜎 2 𝑔 = genotypic variance

𝜎 2 𝑒 = Environmental variance

√𝜎2𝑔
𝐺𝐶𝑉 = [ ] 𝑥100
X

√𝜎2𝑝
𝑃𝐶𝑉 = [ ] 𝑥100
X

Where:𝐺𝐶𝑉 = Genotypic coefficient of variation

𝑃𝐶𝑉 = Phenotypic coefficient of variation

𝐗 = Population mean of the character being evaluated

PCV and GCV values was categorized as low, moderate, and high values as indicated by
Sivasubramaniah and Menon (1973) as follow.

0 - 10% = Low

10 – 20 = Moderate

> 20 = High
 34

Heritability and genetic advance

Broad-sense heritability values were estimated using the formula adopted by Falconer and Mackay
(1996) as follows:

H2 = (σ2g/σ2p) x 100

Where, H2 = heritability in a broad sense

σ2p = phenotypic variance

σ2g = Genotypic variance

The heritability percentage was categorized as low, moderate and high as suggested by Robinson
et al. (1955).

0 - 30% = Low

30 – 60 = Moderate

> 60 = High

Expected genetic advance under selection (GA)

Genetic advance in the absolute unit (GA) and percent of the mean (GAM), assuming selection of
superior 5% of the genotypes were estimated in accordance with the methods illustrated by Johnson
et al., (1955) as:

𝐺𝐴 = 𝐾 𝑥 𝑆𝐷𝑝 𝑥 𝐻 2

Where 𝐺𝐴 = Genetic advance

𝑆𝐷𝑝 = Phenotypic standard deviation on mean basis

𝐻 2 = Heritability in the broad sense

𝐾 = the standardized selection differential at 5% selection intensity (K = 2.063)
Genetic advance as percent of the mean

Genetic advance as percent of the mean was estimated as follows:

 35

𝐺𝐴
𝐺𝐴𝑀 = ( ) 𝑥100
𝑋

Where, 𝐺𝐴𝑀 = Genetic advance as percent of mean

𝐺𝐴 = Genetic advance
𝑋 = Populations mean character to be evaluated
The GA as a percent of the mean was categorized as low, moderate and high as suggested by
Johnson et al., (1955) as follows.
0 - 10% = Low
10 – 20 = Moderate
>20 = High

3.5.3. The Phenotypic and Genotypic Correlation Coefficient

Phenotypic (rp) and genotypic (rg) correlations between two traits were estimated using the formula
suggested by Johnson et al., (1955) and Singh and Chaudhury (1985).

covpxy
𝑟𝑝𝑥𝑦 =
√σ2 px . σ2 𝑝𝑦

Where, 𝑟𝑝𝑥𝑦 = phenotypic correlation coefficient between character x and y

covpxy = phenotypic covariance between character x and y

σ2 px= phenotypic variance for character x

σ2 𝑝𝑦 = phenotypic variance for character y

𝑐𝑜𝑣𝑔𝑥𝑦
𝑟𝑔𝑥𝑦 =
√ 𝜎 2 𝑔𝑥 . σ2 𝑔𝑦

Where: 𝑟𝑔𝑥𝑦 = genotypic correlation coefficient between character x and y

𝑐𝑜𝑣𝑔𝑥𝑦 = genotypic covariance between character x and y

σ2 𝑔𝑥 = genotypic variance for character x

σ2 𝑔𝑦 = genotypic variance for character y

 36

The coefficient of correlation at the phenotypic level was tested for significance by comparing the
values of correlation coefficient with tabulated r-value at g-2 degree of freedom, where ‘g’ is a
number of genotypes. However, the coefficient of correlations at the genotypic level was tested for
the significance using the formula described by Robertson (1959).

(𝑟𝑔𝑥𝑦)
t=
𝑆𝐸𝑟𝑔𝑥𝑦

The calculated‘t’ value was compared with the tabulated’t’’ value at g-2 degree of freedom at 5%
level of significance. Where, g= number of genotypes, 𝑟𝑔𝑥𝑦 =genotypic correlation coefficient
and, 𝑆𝐸𝑟𝑔𝑥𝑦 = standard error of genotypic correlation coefficient between character x and y
which was calculated as:

(1 − r 2 )2
𝑺𝑬𝒓𝒈𝒙𝒚 = √
2𝐻 2 𝑥 . 𝐻 2 𝑦

Where: 𝑺𝑬𝒓𝒈𝒙𝒚 =standard error of genotypic correlation coefficient between character x and y,
𝐻 2 𝑥 =Heritability value of character x and

𝐻 2 𝑦 = heritability value of character y.

3.5.4. Path Coefficient Analysis

Based on genotypic and phenotypic correlations, path coefficient analysis which refers to the
estimation of direct and indirect effects of the fruit yield attributing characters (independent
character) on fruit yield (dependent character) was calculated based on the method used by Dewey
and Lu, (1959) as follows:

𝑟𝑖𝑗 = 𝑝𝑖𝑗 + Σ𝑟𝑖𝑘𝑝𝑘𝑗

Where, 𝑟𝑖𝑗 = mutual association between the independent character (i) and dependent character (j)
as measured by the genotypic and phenotypic correlation coefficients. 𝑝𝑖𝑗 = direct effects of the
independent character (i) on the dependent variable (j) as measured by the genotypic path
coefficients, and Σ𝑟𝑖𝑘𝑝𝑘𝑗 = Summation of components of indirect effects of a given independent
character (i) on a given dependent character (j) via all other independent characters (k).
 37

The residual effect, which determines how best the causal factors account for the variability of the
dependent factor yield, was computed using the formula;

1 = 𝑝2 𝑅 + Σ𝑝 𝑖𝑗 𝑟𝑖𝑗

Where, 𝑝2 𝑅 is the residual effect and Σ𝑝 𝑖𝑗 𝑟𝑖𝑗 is the product of direct effect of any variable and
its correlation coefficient with yield

3.5.5. Genetic Distance and Clustering

Euclidean distance (ED) was computed from all data collected for okra accessions after
standardization (subtracting the mean value and dividing it by the standard deviation) as:

𝐸𝐷𝑗𝑘 = √∑𝑛𝑖=1(𝑋𝑖𝑗 − 𝑋𝑖𝑘)2 (Sneath and Sokal, 1973),

Where 𝐸𝐷𝑗𝑘 = distance between accessions j and k; 𝑋𝑖𝑗 and 𝑋𝑖𝑘 = phenotype traits values of the
ith character for genotypes j and k, respectively; and n = number of phenotype traits used to
calculate the distance. The distance matrix from phenotype traits were used to construct
dendrograms based on the Unweighted Pair-group Method with Arithmetic Means (UPGMA). The
results of cluster analysis were presented in the form of dendrograms. In addition, mean ED was
calculated for each accession by averaging of a particular genotype to the other 35 genotypes. The
calculated average distance (ED) was used to estimate which genotype(s) is closest or distant to
others.

3.5.6. Principal Component Analysis

Principal component analysis (PCA) was computed to find out the characters, which accounted
more to the total variation. The data was standardized to mean zero and variance of one before
computing principal component analysis. The principal component based on a correlation matrix
was calculated using XLSTAT software (2014).
 38

4. RESULT AND DISCUSSION

4.1. Analysis of Variance and Genotypes Performance of Okra

4.1.1. Crop Phenology and Growth Traits

The results of analysis of variances for 14 phenology and growth traits of 36 okra genotypes are
presented in Table 3. Mean squares of phenology and growth traits of 36 okra genotypes were
significant except nonsignificant mean squares were observed for leaf length, leaf width and
number of epicalyxes. These results indicated the presence of significant variations among
genotypes for most phenology and growth traits that may give a good opportunity for breeders to
identify best performing genotypes for traits of interest to improve in breeding programs. Tesfa and
Yosef (2016) evaluated 50 okra genotypes at Melkassa Agricultural Research Center and Muluken
et al. (2016) evaluated 25 okra accessions at Werer Agricultural Research Center and reported the
presence of significant differences among genotypes for phenology and growth traits. Mihratu et
al. (2014) also reported significant differences among 25 accessions for days to pod formation,
days first flowering, days to 50% flowering, days to maturity, plant height, stem diameter, number
of branch, fruit length, fruit diameter, fruit weight, fruit yield, number of mature pod, and fresh
weight of mature pod.

The mean values of genotypes for crop phenology and growth traits are presented in Appendix
Table 1. The genotypes variations for days to seedling emergence, days to first flowering, and days
to 50% flowering ranged from 7.5 to 11, 44.5 to 71 and 48.5 to 77.5, respectively. The genotypes
showed variations for days to 90% maturity and number of harvest in the range between 75.5 and
104.5 days and 5.4 and 8, respectively. The seeds of 29621 and 29620 showed significantly delayed
emergence compared to check varieties (SOH701, SOH714, and Bamia Humera) and to other
genotypes. The accessions, 29617, 29418 had significantly delayed in days to first flowering and
50% flowering as compared to check varieties and to other accessions. Accessions, 29052, 29416
and 29617 were late for days to 90% maturity while 240204 and SOH714 were early maturing.
The accessions, 29414, 29418, 29618 29052, 29615 and 29616 had long duration of fruit harvest
while 29411 and 29417 had short duration of fruit harvest.
 39

The 33 accessions collected from Benishangul Gumuz Regional State had overall mean values
greater than mean values of the three check varieties for all phenology and growth traits except
internode length. A total of 28 (84.85%) and 19 (57.58%) of accessions collected from different
Woredas of Benishangul Gumuz Regional State, western Ethiopia had mean performance greater
than mean values of the three check varieties for days to seedling emergence and duration of fruit
harvest, respectively. All except one accession collected from Benishangul Gumuz Regional State
had delayed days to first flowering and days to 50% flowering and all except two accessions had
delayed maturity than the three check varieties (SOH701, SOH714, and Bamia Humera) (Appendix
Table 1). Muluken et al. (2016) also reported delayed seed emergence, flowering and maturity of
okra genotypes collected from Ethiopia than the two commercial varieties (SOH701) and
(SOH714). Muluken et al. (2016); Anteneh (2017) and Fozia (2018) also observed delayed seedling
emergence, days to first flowering, days to 50% flowering and maturity of okra genotypes collected
from Ethiopia than introduced SOH701 and SOH714 commercial varieties.
 40

Table 3.Mean squares from analysis of variance for 14 phenology and growth traits of 36 okra
genotypes evaluated at Melkassa during 2018/19

Blocks/ Genotype Error

Rep(Adj.) (Unadj.) RCBD Intra CV
Traits Rep (1) (10) (35) (35) Block (25) (%)
Days to 50% emergence 0.22 0.42 2.41** 0.39 0.38 7.04
Days to first flower 9.39 14.91 97.94** 10.47 8.7 5.45
Days to 50% flower 20.06 16.34 115.39** 13.63 12.54 5.98
Days to 90% maturity 25.68 2.11 137.04** 3.82 4.51 2.43
Number of harvest 0.61 0.6 1.501** 0.33 0.22 6.82
Stem diameter(mm) 3.83 1.47 22.24** 1.58 1.63 6.12
Plant height (cm) 43.71 53.86 2155.28** 61.59 64.69 6.19
Number of primary branch 0.09 1.31 12.28** 0.58 0.29 12.02
Number of internode 2.72 2.96 73.27** 2.69 2.59 6.21
Internode length (cm) 0.25 0.45 7.79** 0.35 0.3 11.35
Leaf length(cm) 1 9.71 9.24ns 6.95 5.84 13.86
Leaf width (cm) 0.13 17.06 17.15ns 13.87 12.59 15.29
Number of epicalyxes 0.01 1 1.06ns 1.03 1.04 10.31
Peduncle length (cm) 0.05 0.04 0.4366** 0.04 0.04 9.8
ns, * and **= nonsignificant, significant at P<0.05 and P<0.01, respectively. Rep= replication, Blocks/Rep (adj.) = blocks in
replication adjusted, Unadj. = unadjusted, RCBD= randomized complete block design, and CV (%) = coefficient of variation
in percent. Number in parenthesis represents degree of freedom for the respective source of variation.

The variation among okra genotypes was in the range between 14.5 and 30.5 cm, 73.05 and 194.5
cm, and 1.3 to 12.7 for stem diameter, plant height and number of branches, whereas it was ranged
from 13.5 to 39cm and 2.5 to 10.25 cm for number of internodes, and internode length, respectively.
The Genotypes (29615 and 29414), (29416 and T240204), (29418 and 29409) shows significant
difference for plant branches, number of internode and internode length whereas genotypes (29408
and 29417), (29622 and 29417), and (T242444 and 29620) had significantly lower number of
branches, number of internode and internode length as compared to checks and other Benishangul
Gumuz collections, respectively.

Among the 33 accessions collected from Benishangul Gumuz Regional State, 93.94, 87.88 and
81.82% had mean values greater than mean values of check varieties for days to 90% maturity,
stem diameter and plant height, respectively. In addition, 39.39, 90.91 and 69.70% accessions had
a mean values greater than mean values of check varieties for internode length, leaf length and
number of epicalyxes, respectively. In addition, 96.97, 84.85 and 66.67% of genotypes had greater
 41

performance than the overall mean values of check varieties for leaf width, peduncle length as well
as number of primary branches and number of internodes each, respectively. The results showed
that most of collections from Benishangul Gumuz Regional state had more robust/vigorous growth
characteristics than the commercial varieties. Muluken et al. (2016) also observed robust growth
of 23 okra genotypes collected from different regions of Ethiopia than the two commercial varieties
(SOH701) and (SOH714). He also reported that higher mean performances of 23 okra genotypes
collected from Ethiopia for growth traits (plant height, stem diameter, number of primary branches,
number of internode and number of epicalyxes) than the two introduced commercial varieties.
Anteneh (2017) reported the presence of wide variation among Ethiopian collections and also
higher performance of genotypes than commercial varieties.
Researchers from different countries in the world also reported the presence of significant
differences among okra genotypes for phenology and growth traits. Aminu et al. (2016) reported a
mean value of 46.82, 1.25m, 2.92, for days to 50% flowering, plant height and number of primary
branches, respectively among okra evaluated at Nigeria. Chandra et al. (2014) also reported 47.33
to 54.67, 1.47 to 3.53, 102.53 to 155.67cm, and 53.33 to 63.33 ranges for days to first flowering,
number of primary branches, average plant height, and days to maturity, respectively from India.
Shivaramegowda et al.(2016) evaluated 36 okra genotypes at southern India and reported mean
values of 41, 2.08, and 96.73 cm for days to first flowering, a number of primary branches and
plant height respectively. Davinder et al. (2018) also reported the presence of high genetic
variability for agro-morphological traits of 30 okra accessions from India.

4.1.2. Fruit Yield and Yield Components of Okra Genotypes

The results of analysis of variance (ANOVA) of 36 Okra genotypes for 11 fruit yield and yield
components traits showed significant difference at (P<0.01) except for hundred seed weight and
number of fruit ridges traits at (P<0.05) level of significance (Table 4). Tesfa and Yosef (2016)
observed significant differences among 50 okra genotypes collected from Ethiopia for number of
fruits per plant, mean fruit weight and total fruit production. Similar results was reported by
Muluken et al. (2016) and Mihretu et al. (2014) also reported the presence of significant difference
among okra genotypes collected from Ethiopia for fruit length, fruit diameter, fruit weight, fruit
ridge, fruit yield, number of mature pod, fresh weight of mature pod and dry weight of mature pod,
number of seed per pod and hundred seed weight.
 42

The mean values of tested genotypes are presented in Appendix Table (1). They exhibited variation
in ranges 5.5 cm to 22.3 cm, 21.6 mm to 35.75 mm, 17.04 g to 67.23 g, 9.7 to 55.2, and 5 to 8.3
with the mean values of 11.83 cm, 27.79 mm, 37.19(g), 21.72, and 5.54 for fruit length, fruit
diameter, average fruit weight, number of tender fruit per plant and number of fruit ridge,
respectively. On another hand, the mean values of 704.82g, 19.58 ton/ha-1, for fuit yield per plant,
fruit yield per hectare, with a ranged from 339.71 to 1183.6 g and 9.44 to 32.88 tons, respectively.
The genotypes had wide range of variation 58.5 to 111.9, 5.08 to 7.16g, 33.48 to 198.04g, and
929.95 to 5501.01 kg/ha-1 and with mean values 79.56, 5.93g, 99.7g, and 2769.34kg/ha-1 for
number of seed per plant, hundred seeds weight, seed yield per plant and seed yield per hectare,
respectively. The Genotypes 29408, T240204, 29417, 29618 and 29409 showed significantly
higher for fruit length, fruit diameter, average fruit weight, number of tender fruit per plant and
number of fruit ridge whereas genotypes 29624, 29625, 29618, 29411 and 29619 shows
significantly lower as compared to three check varieties and other Benishangul Gumuz collections,
respectively. In addition genotypes 29618 and 29620 showed significantly higher and lower in fruit
yield performance, respectively. Genotypes 29409, 242449A, and 29618 shows the highest
performance for number of seed per pod, weight of hundred seed and seed yield per hectare-1
whereas genotypes 29052, 29412 and 29417 had the lowest performance as compared to checks
and to the rest of genotypes collected from regional states, respectively. Muluken et al. (2016),
Mihretu et al. (2014a) and (2014b) and Tesfa and Yosef (2016) reported a greater degree of
variability in all Ethiopian okra collection and higher in performance for fruit and fruit yield
components than introduced commercial varieties.

Similar results are reported from other countries, Chandra et al. (2014) reported that five accession
are evaluated at India and shows variation for a number of fruits per plant that ranged from 9.17 to
20.22, fruit length from 7.03 to 15.08 cm and fruit width from 1.29 to 1.69 cm. Average fruit weight
varied from 7.79 to 14.69 g, a number of fruit ridge 4.67 – 7, fruit yield per plant 108.4 to 291.97
and fruit yield per hectare 7.23 to 19.46 t ha-1. Aminu et al. (2016) have also reported from
experiment carried out at Nigeria on 10 okra genotypes the mean values of 13.08cm, 1.54cm,
15.25(g) and 523.75(g) for, number of pod per plant, fruit length, fruit diameter, fresh pod weight
and pod yield per plant respectively. Oppong-Sekyere et al. (2011) reported 25 accessions of okra
collected and evaluated at Ghana all the okra accessions showed relatively wide ranges of variations
for all morphological characters observed. Shivaramegowda et al. (2016) reported from similar
 43

study carried in India 16.1, 14.26(g), 13.09(g), 231.27(g) of mean values for a number of fruits per
plant, fruit weight, fruit length, and fruit yield per plant respectively. A number of fruits per plant,
fruit weight, and total fruit production often show high variability as compared to other quantitative
traits (Ahiakpa et al., 2013).

Among the 33 accessions collected from Benishangul Gumuz Regional State 15 (45.45%), 18
(54.55%), 18 (54.55), 21 (63.64%), and 18 (54.55) of genotypes had a mean greater than the mean
values of checks varieties for fruit length, fruit diameter, average fruit weight, number of tender
fruit per plant, and number of ridges, respectively. The mean performance of 25 (75.76%) and 21
(63.64%) okra genotypes showed higher than the mean performance of three checks for fruit and
seed yields, respectively. Sixteen (48.48%) and 12 (36.36%), of genotypes showed higher mean
values than the mean values of three commercial varieties for number of seed per pod and weight
of hundred seed, respectively.

Table 4. Mean squares from analysis of variance for 11 tender fruit yield and yield components as
well as 2 quality related traits of 36 okra genotypes evaluated at Melkassa during 2018/19

Blocks/ Genotype Error

Rep(Adj.) (Unadj.) RCBD Intra CV
Traits Rep (1) (10) (35) (35) Block (25) (%)
Fruit length (cm) 0.01 0.76 37.27** 0.84 0.87 7.87
Fruit diameter (mm) 2.33 0.34 28.93** 0.38 0.39 2.25
Average fruit weight (g) 4.85 6.11 428.44** 7.68 8.31 7.75
Number of fruits per plant 89.76 36.49 187.39** 29.63 26.88 23.87
Number of fruit ridge 2.57 0.87 1.28* 0.61 0.5 10.86
Number of seed per pod (g) 217.01 59.63 448.82** 79.34 87.22 11.74
Hundred seed weight (g) 0.55 0.19 0.44* 0.16 0.15 6.6
Seed yield per plant (g) 7.17 1102.37 3002.77** 580.19 371.31 19.33
Seed yield per hectare (kg) 5535.68 850596 2316952** 447674 286506 19.33
Fruit yield per plant (g) 13452 45120 139861** 18965 8503.08 13.08
Fruit yield per hectare (ton) 10.38 34.81 107.92** 14.63 6.56 13.08
Dry matter content of fruits (%) 2.65 1.24 14.5109** 1.08 1.01 3.86
Mucilage content fruits (%) 0.02 1.12 41.525** 1.64 1.84 9.96
ns, * and **= nonsignificant, significant at P<0.05 and P<0.01, respectively. Rep= replication,
Blocks/Rep (adj.) = blocks in replication adjusted, Unadj. = unadjusted, RCBD= randomized complete
block design, and CV (%) = coefficient of variation in percent. Number in parenthesis represents degree
of freedom for the respective source of variation.
 44

The result indicated that the genotypes collected from Benishangul Gumuz region had higher mean
values than the mean values of checks varieties viz. Bamia Humera, SOH701 and SOH714 for all
tender fruit yield and yield components except for fruit length and weight of hundred seed.

They showed double values for pod and seed yield as compared to checks. This showed that the
higher degree of variability and chance of developing varieties for tender fruit yield and yield
components through collection and selection of okra genotypes with the desired traits.

The scope of improvement in any crop depends upon the magnitude of genetic variability present
in the available germplasm. Greater the variability in the available germplasm higher would be the
chances of selecting superior genotypes (Chandra et al., 2014).

4.1.3. Fruit Quality Related Traits of Okra Genotypes

The genotypes showed highly significant (P<0.01) differences for percentage dry matter contents
of tender fruits and fruits percentage of mucilage content (Table 4). The tender fruits of genotypes
had dry matter content in the range between 22.035 and 34.55% with overall mean values of
26.08%, whereas mucilage content of fruits was in the range between 6.93 and 23.01% with overall
mean values of 13.63% (Appendix Table 2 and Table 4). Anteneh (2017) observed the ranges from
26.45 to 54.03% and 4 to 25.33% for dry matter content and mucilage contents of fruits of 25 okra
genotypes evaluated at Dire Dawa, respectively. Gemede et al. (2018), reported the mean mucilage
content (2.43 g/100g) of Ethiopian okra is higher than the value of Nigerian okra (1.4g/100g).

Genotypes having lowest and highest dry matter and mucilage contents are 29617, 29624 and
29408, 29619 respectively. Genotypes 29624 and 29619 are significantly different in contrast to
checks by having the highest dry matter and mucilage contents respectively which is 23 (69.7%)
and 24 (72.73%) percentage of dry matter contents and percentage of mucilage contents compared
to the mean value of three checks, respectively. This result indicates the possibility of improving
the mucilage and dry matter contents of okra from Benishangul Gumuz collections as it’s the major
determinants of okra nutritional qualities. Okra mucilage also contributes to improved
functionality, especially water-binding, emulsifying, and foaming properties of food products
(Jideani and Bello, 2009). Mucilage from okra contains significant levels of protein, carbohydrate,
neutral sugars, minerals, and other complex polysaccharides (Ahiakpa et al., 2014).
 45

4.2. Qualitative Traits of Okra Genotypes

4.2.1. Plant Habit and Stem Related Traits

A total of 18 (50%), 14 (38.89%) and 4 (11.11%) of genotypes had a plant habit of densely branched
base, densely branched all over and densely branched at the apex, respectively. All genotypes under
study showed erect growth habits and large proportion of 26 (72.22%) okra genotypes had green
with a red patch stem color while 7 (19.44%) and 3 (8.33%) genotypes had red or purple and green
stem colors, respectively (Figure 1). This indicated that the okra genotypes were distributed into
three categories of each trait though 50 to 72.22% genotypes were grouped in one of the category
of each trait. The okra genotypes collected from different regions and countries may be distributed
into varied categories of plant habit, stem color and plant alignment and proportion of genotypes
in each category may varied either due to inherent characteristics differences and environmental
differences where the genotypes are grow or selection of genotypes differences by human being.
Evolutionary forces like selection, mutation, migration and genetic drift are the basis of crop
genetic diversity (Bhandari et al., 2017). There may be also differences of data registration because
the descriptors tend to be highly subjective (Hammon and Van Stolen, 1989).

Similar with Tripath et al. (2011) reported stem of okra plant is robust, erect, and variable in
branching and varying from 0.5 to 4.0 meters in height and stem is pigmented with green or reddish
tinges color. Muluken et al. (2016) reported densely branched base (DBB) characters were 92%
in frequency, densely branched over and non-branched growth habit 4% for each. In addition other
authors Oppong-sekyere et al. (2011), Tesfa and Yosef (2016) and Wassu et al. (2017) reported
that the growth habit and degree of branching of okra germplasm are highly variable. About 59%
of accessions are erect type while 24.7 and 16.4 % of accession characterized with medium growth
and procumbent type of growth alignment, respectively. Stem of okra was 32, 24, and 44% shows
green, green with red patches and red/purple color, respectively. In addition they reported that most
okra genotypes have densely branched at base characters followed by densely branched overall and
branched at apex.
 46

Figure 1. Number of okra genotypes as distributed into categories of plant habit (a), stem color (b)
evaluated at Melkassa 2018/19.

4.2.2. Leaf and Flower Characteristics of Okra

The tested genotypes had great variations for leaf related qualitative traits viz. leaf color, leaf petiole
color, leaf pubescence and shape of leaf (Figure 2). Out 36 genotypes 22 (61.11%) and 14 (38.89%)
genotypes showed green with red veins and totally green leaf color, respectively. The leaf petiole
of 36 genotypes were observed as red above but green below 27(75%) and red on both side 9(25%)
among evaluated genotypes. The leaf pubescence traits observed among 36 genotypes were
31(86.11%) slight pubescence, 3(8.33%) glabrous and 2(5.56%) are conspicuous. The study also
revealed five distinct leaf shape 11(30.56) heart shaped, 10(27.78%) broadly ovate, 10(27.78%)
palmately lobed with serrated margin, 3(8.33%) star shaped and 2(5.56%) with palmately triangular
lobes. Similar results reported by Wassu et al. (2017) among 25 okra genotypes 60% showed green
with red vein and 40% had totally green leaf colors. Tesfa and Yosef (2016) reported three distinct
leaf colors viz. 26.5% green with red veins, 60.3% totally green and 13.2% red colors. In addition
they also reported that leaf pubescence as 67.1% slight pubescence, 19.2% glabrous and 13.7% had
conspicuous leaf pubescence among 50 tested genotypes. Muluken et al. (2016) reported that
genotypes shows 48% heart shapes, 28% broadly ovate shape, 16% star shaped and 8% palmately
lobed with serrated margin of leaf shape.

All okra genotypes shows flower color with red color on both side except one genotypes which
shows red color inside only. Similar result was reported by Muluken et al. (2016) where all Okra
genotypes showed uniform red at both side flower colors. Wassu et al. (2017) also reported, 22
 47

(88%) and three (12%) okra genotypes had flowers with red color at both sides and red color inside
only, respectively.

Figure 2. Number of okra genotypes as distributed into categories of leaf color (a), leaf petiole color
(b), leaf pubescence (c), flower color (d) and shape of leaf (e) evaluated at Melkassa 2018/19.
 48

4.2.3. Fruit Related Trait and Seed Shape of Okra Genotypes

The current research reveals the presence of variation among tested genotypes for immature fruit
color, position of fruit on main stem, fruit shape, fruit pubescence and nature of fruit base (Figure
3). The immature fruit color of 20 (55.56%) of genotypes shows green with red patch, 15 (41.67%)
green and 1(2.78%) had yellow green color. All genotypes had slight fruit pubescence except one
genotype which had irritating character at harvest unless we used gloves during harvest. In addition
to this 25 (69.44%), 10 (27.78%) and 1(2.78%) genotypes shows erect, intermediate and horizontal
fruit position on main stem, respectively. The genotypes had three different fruit base natures; of
which 25 (69.44%), 6(16.67%) and 5(13.89%) are ringed protrude, ring less (flat) and sunken fruit
base, respectively. Fruit shape showed the greatest diversity among the okra accessions; from shape
score 1, 2, 3, and 4 as 41.67, 13.89, 16.67 and 27.78%, respectively according to fruit shape score
(Appendix figure. 1). This result shows the presence of variation among genotypes studied which
may be due to distribution of genotypes and environmental difference that indicates the possibility
of improving fruit qualitative traits through selection and/or crossing of genotypes from regional
state.

Muluken et al. (2016) observed that 72% and 28% genotypes had green and yellow green immature
fruit color, respectively, and 68 and 32% genotypes had erect and intermediate position of fruit on
the main stem, respectively. Tesfa and Yosef (2016) also reported 56.1% yellowish green immature
fruit color, 25.8% green, 12.1% green with red, 3% dark green and 3% dark red immature fruit
colors. They also identify fruit pubescence 29.6% downy, 35.2% slight rough and 35.2% prickly.
Adeoluwa and Kehinde (2013) from Nigeria reported that 60% erect and 40% horizontal fruit
position on main stem and 56% smooth and 44% rough fruit pubescence. They also reported three
distinct fruit color of green (42.85%), purple (48.57%) and green yellow (8.57%).

The genotypes showed three seed shape which is 4 (11.11%) reniform (kidney) shape and 16(44.44)
round and spherical shaped each. Oppeng et al. (2011) reported that among genotypes 24, 40 and
36% of showed round, spherical (oval) and kidney (reniform) seed shape, respectively.
 49

Figure 3. Number of okra genotypes as distributed into categories of Immature fruit color (a),
Position of fruit on stem (b), Fruit pubescence (c), Nature of fruit base (d), Fruit shape (e) and
Shape of seed (f) evaluated at Melkassa 2018/19.
 50

4.3. Estimation of Variability Components

4.3.1. Genotypic and Phenotypic Variations

The estimated genotypic (GCV) and phenotypic (PCV) coefficient of variations for 24 quantitative
traits of okra genotypes are presented in (Table 5). The genotypic and phenotypic coefficients of
variation ranged from 6.3 to 54.19% and 9.19 to 55.51%, respectively. Among evaluated traits,
days to 90% maturity, and weight of hundred seed showed lower both for PCV and GCV, while
number of fruit ridge and percentage of dry matter content had less than 10% GCV. Number of the
branch, followed by a number of tender fruit per plant, internode length, seed yield, average fruit
weight, pod yield, fruit length, percentage of mucilage content, plant height, internode length, and
peduncle length had higher GCV and PCV.

According to Sivasubramanian and Madhavamenon (1973) PCV and GCV can be categorized as
low (<10%), moderate (10-20%) and high (> 20%). Based on this delineation PCV and GCV for
days to 50% emergence, days to first flowering, days to 50% flowering, stem diameter, fruit
diameter, and number (frequency) of fruit harvest had moderate values (10-20%). The difference
between the values of PCV and GCV was low (<5%) for all of the traits except 6.41% for number
of tender fruit per plant. This result is consistent with the finding of Adeoluwa and Kehinde (2011)
who reported low PCV and GCV for 100 seed weight and number of ridges per fruit and Ehab et
al. (2013), and Aminu et al. (2016) reported that number of pods per plant, fruit length, plant height,
the number of primary branches, fresh pod length, fresh pod yield per plant and fresh seed per pod
exhibited high values of more than 20 % for both PCV and GCV with a considerably low degree
of variation between them. Sibsankar et al. (2012) also estimated low PCV and GCV values for
days to first flowering and numbers of ridge per fruit. In contrast, Muluken et al. (2016) reported
the highest GCV and PCV are observed for number of matured pod per plant and number of primary
branches 34.14 and 27.19, respectively.

The large magnitude of PCV and GCV with a small difference between the two heredity parameters
indicated a smaller amount of environmental influence on the phenotypic expression; several
researchers reported the consistent differences of okra cultivars due to cultivars and environmental
interactions (Thirupathi et al., 2012; AdeOluwa and Kehinde, 2013; Ehab et al., 2013; Adekoya et
al., 2014). Therefore, high phenotypic and genotypic coefficients of variation is an indication of
 51

the less influence of environmental factors in the expression of traits and the higher chance to
improve the traits through selection breeding.

4.3.2. Heritability and Genetic Advance

Estimates of heritability in a broad sense (H2) and genetic advance as percent of the mean (GAM)
for 24 quantitative traits of okra genotypes are presented in (Table 5). The heritability values ranged
from 43.46% to 97.34 % for a number of fruit ridge to fruit diameter respectively. As suggested by
Johnson et al. (1955), heritability values are categorized as low (<30%), moderate (30-60%) and
high (>60%). Based on this classification, all traits had high heritability except a number of ridge
and hundred seed weight which had moderate heritability. The high values of the heritability
indicate that in the future programs of okra improvement for the characters such as diameter of the
fruit, length of the fruit and number of seeds per pod. It is important to apply some backcross to
concentrate these characters in genotypes because a lot of traits appear to be controlled by the genes
with additive effects. Knowledge of heritability involved in the expression of traits and the
existence of a correlation between traits which would facilitate the further improvement of okra
(Célestin et al., 2012; Subrata et al., 2004).

The heritability is a good index of the degree of transmission of the characters from parents to their
offspring (PhaniKrishna et al., 2015). The consistency in the performance of progeny in succeeding
generations depends mainly on the magnitude of the heritable portion of the variation. Heritability
indicates the possibility and extent to which improvement can be brought about through selection
and it may be of broad and narrow senses (Robinson, 1966).

The genotypes showed genetic advance as percent of mean in the range between 9.16 and 109.14
% for the trait of hundred seed weight to a number of branches, respectively. According to Johnson
et al. (1955a), the range of genetic advance as percent mean are classified as low (<10%), moderate
(10-20%) and high (>20%). Accordingly, all the traits had high GAM at 5% selection intensity
except moderate for days to 90% maturity, number of fruit ridge and percentage of dry matter
content whereas low for hundred seed weight. Similar results reported by Muluken et al. (2016) in which
genetic advance was observed for number of branches, fresh weight of mature pod, number of
mature pod per plant, fruit length, fruit weight, plant height, fruit yield ha-1, dry weight of mature
pod, number of fruit per plant, days to 50% flowering, days to maturity and number of seeds per
 52

pod but disagree for high heritability in number of ridge per pod. High genetic advance (198.15)
was reported for the number of primary branches (Mihretu et al., 2014b)

High heritability coupled with high genetic advance as percent of the mean was more valuable in
predicting the effect of selection because heritability is a single numerical expression on the ratio
of the two variances which may not lead to success if the selection is based on heritability estimates
alone (Johnson et al., 1955). In this regard high heritability coupled with high genetic advance for
all traits except moderate genetic advance for days to 90% maturity and percentage of dry matter
content but, moderate heritability for a number of ridge and hundred seed weight coupled with
moderate to low in genetic advances respectively. Therefore, this will pave ways for the success of
selection of genotypes for the improvement program.

The heritability estimates provide a background genetic information on the inheritance of traits
from parents to progenies as well as the amount of genetic progress that can be made in selection
of a trait of interest (Sabesan and Saravanan, 2009). While genetic advance under selection is
helpful in detecting the actual genetic gain expected (Ogunniyan and Olakojo, 2014).
Understanding and knowledge of the magnitude of genetic variability among the okra genotypes
for the traits evaluated are very important and a pre-requisite for selection of good parental lines to
make a significant improvement in okra breeding (Gerrano, 2017).
Table 5. Estimates of range, mean and variability components of 36 okra genotypes evaluated at Melkassa in 2018/19

Variables Range Mean SE 𝜎2g 𝜎2p GCV PCV H2 (%) GA GAM%

Eme 7 - 11 8.78 0.50 1.02 1.40 11.48 13.46 72.65 1.77 20.18
Dflo 44 - 71 54.17 2.21 44.62 53.32 12.33 13.48 83.68 12.61 23.27
Flo 47.5 - 77.5 59.25 2.59 51.42 63.97 12.10 13.50 80.39 13.26 22.39
Mat 74.5 - 104.5 87.51 1.38 66.27 70.77 9.30 9.61 93.63 16.25 18.57
SD 14.5 - 30.5 20.87 0.89 10.31 11.94 15.38 16.56 86.33 6.15 29.49
PH 73.05 - 194.5 129.88 5.55 1045.30 1109.98 24.89 25.65 94.17 64.73 49.84
PB 1.3 - 12.7 4.52 0.42 5.99 6.29 54.19 55.51 95.31 4.93 109.14
NI 13.5 - 39 25.92 1.16 35.34 37.93 22.94 23.76 93.17 11.84 45.68
IL 2.5 - 10.25 4.86 0.41 3.74 4.05 39.82 41.41 92.48 3.84 79.00
PL 1.53 - 3.45 2.15 0.15 0.20 0.24 20.57 22.79 81.49 0.82 38.31
FL 5.5 - 22.3 11.83 0.65 18.20 19.07 36.07 36.92 95.45 8.60 72.70
FD 21.6 - 35.75 27.79 0.43 14.27 14.66 13.59 13.78 97.34 7.69 27.66
AFW 17.04 - 67.23 37.19 1.96 210.06 218.38 38.97 39.73 96.19 29.33 78.85
NTFPP 9.7 - 55.2 21.72 3.80 80.25 107.14 41.24 47.65 74.91 16.00 73.64
NR 5 - 8.3 6.54 0.53 0.39 0.89 9.52 14.44 43.46 0.85 12.95
YPP 339.71 - 1183.60 704.82 72.37 65678.96 74182.04 36.36 38.64 88.54 497.48 70.58
yhaton 9.44 - 32.88 19.58 2.01 50.68 57.24 36.36 38.64 88.54 13.82 70.58
NSPP 58.5 - 111.9 79.56 6.30 180.80 268.02 16.90 20.58 67.46 22.78 28.64
SDW 5.08 - 7.16 5.94 0.30 0.14 0.30 6.39 9.19 48.33 0.54 9.16
syppg 33.48 - 198.04 99.70 14.86 1315.73 1687.04 36.38 41.20 77.99 66.09 66.29
syph 929.95 - 5501.01 2769.34 412.80 1015223 1301729 36.38 41.20 77.99 1835.69 66.29
Nha 5.4 - 8 6.93 0.31 0.64 0.86 11.54 13.41 74.14 1.42 20.51
DMC (%) 22.035 - 34.55 26.08 0.73 6.75 7.76 9.96 10.68 86.93 5.00 19.16
PMC (%) 6.93 - 23.01 13.63 0.90 19.84 21.68 32.69 34.17 91.49 8.79 64.50
Eme= Days to 50% emergence, Dflo= Days to first flower, Flo= Days to 50% flowering, Mat= Days to 90% maturity, SD=Stem diameter (mm), PH= Plant height (cm), PB= Number
of primary branch, NI= Number of internode, IL= Internode length (cm), PL= Peduncle length (cm), FL= Fruit length (cm), FD= Fruit diameter (mm), AFW=
Average fruit weight (g), NTFPP= Number of tender fruit per plant, NR= Number of fruit ridge, YPP= Pod yield per plant (g), Yhaton= Pod yield per hectare (tons),
NSPP= Number of seed per pod, SDW= Hundred seed weight, SYPP= Seed yield per pod (g), Syph= Seed yield per hectare (Kg), Nha= Number of harvest, DMC=
Dry matter content (%), PMC= Mucilage content (%).
 54

4.4. Association of Traits

4.4.1. Genotypic and Phenotypic Correlation of Fruit Yield with Other Traits

The results of genotypic and phenotypic correlation coefficients is presented in (Table 6). The
results showed that number of primary branches, fruit diameter, average fruit weight, number of
tender fruit per plant, seed yield and number of harvest had positive and significant genotypic
correlation with fruit yield per hectare at (P<0.01). Whereas, fruit diameter and weight of hundred
seed had significant correlation at (P<0.05). However, all traits had positive and significant
phenotypic correlation with fruit yield per hectare at (P<0.01) level of significance. The presence
of significant and positive correlation of number of primary branches, number of tender fruit per
plant, seed yield, number of harvests, hundred seed weight and fruit diameter with fruit yield per
hectare indicated that the importance of these traits in a selection program to identify the genotypes
with high fruit yield. The signficant and positive correlation coefficients among the quantitative
traits in the accessions of okra indicate that the selection of one of genotypes for the correlated
traits may increase the value of the other traits while the negative correlations between traits
showed the in selection of genotypes for one trait may result the selection of genotypes for the
lower mean value of the other negatively correlated traits.

The genotypic correlation coefficents were higher in magnitude than the corresponding phenotypic
correlations coefficient in growth traits and yield components that had positive and significant
genotypic correlation with fruit yield. This is in agreement with other authors reports that estimates
of genotypic correlation coefficients were in most cases higher than their corresponding phenotypic
correlation coefficients among agronomic traits of okra genotypes (Akinyele and Osekita, 2006;
Bello et al., 2006; Mehta et al., 2006; Rashwan, 2011; Somashekhar et al., 2011; Reddy et al.,
2013 and Muluken et al., 2016). Muluken et al. (2016) reported that fruit yield ha-1 had positive
and highly significant (P<0.01) genotypic correlation with days to pod formation, number of
branches per plant and number of tender fruit per plant in 25 okra genotypes evaluated at Werrer.
Mihretu et al. (2014) also reported positive and significant genotypic corrleations of fruit yield
with average fruit weight, fruit diameter, seed yields, hundred seed weight and a number of tender
fruit per plant. Simon et al. (2013) also reported that genotypic correlation between fruit yield and
seed yield was positive and highly significant (P<0.01). Positive and significant correlation of fruit
yield with hundred seed weight and number of total fruits per plant was reported by Oppong-
 55

Sekyere (2012). Rashwan (2011) reported positive and significant correlation of fruit yield per
plant with the number of branches per plant, number of fruits per plant, and fruit length.

Correlated characters are of prime importance for breeders because of genetic causes of correlations
through pleiotropic action, linkage or developmental interactions of genes and changes brought
about by a natural or artificial selection (Singh, 1993; Falconer et al., 1996; Sharma, 1998). The
presence of strong correlation of traits enables to perform indirect selection for a quantitative trait,
usually hard to be selected by another directly correlated trait of higher genetic gain or easy visual
selection. Besides, it is also able to access how a trait can interfere on another (Bárbaro et al., 2007;
Cruz et al., 2012).

The genotypic and phenotypic correlation coefficient between different trait pairs being similar in
sign and nature indicates the corresponding of heritable factors to phenotypic expression (Akinyele
and Osekita, 2006; Bello et al., 2006; Mehta et al., 2006; Rashwan, 2011 and Somashekhar et al.,
2011). Phenotypic correlation measures the extent to which the two observed characters are linearly
related while genotypic correlation measures the extent to which degree the same genes or closely
linked genes cause co-variation or simultaneous variations in two different characters (Singh and
Chaudhary, 1977; Falconer et al., 1996; Sharma, 1998; Mehta et al., 2006). The genotypic
correlation between two traits determines the genetic breeding as it involves an association of
heritable nature whereas, the phenotypic value is lessened by the significant interaction of
environment (Kumar and Reddy, 2016). Kumar and Reddy (2016) suggested that considerable
number of growth traits and yield components had positive and significant genotypic and
phenotypic correlations with tender fruit yield indicating the correspondence of genetic factors and
phenotypic expressions and the strong association between those characters genetically.

4.4.2. Genotypic and Phenotypic Correlations among other Characters

It was observed positive and significant genotypic and phenotypic associations among phenology
(days to 50% emergence, days to first flower, days to 50% flowering, days to 90% maturity and
number of harvest) of okra genotypes except days to 50% emergence and number of harvest had
positive nonsignificant genotypic and phenotypic correlation. Days to first flower and days to 50%
flowering with number of primary branches and peduncle length showed positive and significant
genotypic and phenotypic correlations, and days to first flowering and days to 50% flowering with
 56

number of primary branches showed positive and significant correlation at genotypic and
phenotypic levels.

In agreement with Fozia (2018) that phenological traits (days to emergence, days to first flowering,
days to50% flowering and days to maturity) showed significant correlation among them both at
genotypic and phenotypic levels except the genotypic correlation between days to emergence and
days to maturity was positive but non-significant. She also reported the genotypic correlation of
phenological traits with growth traits of okra was non-significant except the genotypic correlation
of days to emergence with number of internodes per plant and days to maturity with stem diameter
showed negative and positive significant genotypic association, respectively.

Crop phenological traits (days to 50% emergence, days to first flowering, days to 50% flowering
and maturity) had nonsignificant correlation with growth traits (plant height, stem diameter,
number of inter node and inter node length) at genotypic level. However, primary branches had
positive and significant correlation with first flowering, days to 50% flowering and maturity both
at genotypic and phenotypic level. Days to first flowering with number of tender fruits per plant
and seed yield per hectare had positive significant correlation at both genotypic and phenotypic
level. Positive and significant genotypic and phenotypic correlation were observed between days
to 50% flowering and number of tender fruits per plant. Days to 50% flowering with seed yield
per hectare and days to 90% maturity with number of tender fruits per plant showed positive and
significant correlation at phenotypic level.

Mihretu et al. (2014) reported positive and significant association among phenological traits of
okra. They also stated that days to 50% flowering and days to maturity with average fruit weight,
fruit diameter and seed per pod had negative correlations at genotypic and phenotypic levels.
Muluken et al. (2016) also observed positive and significant correlations of days to pod maturity
with days to 50% flowering and number of branches. Samim et al. (2018) also reported that days
to first flowering with days to 50% emergence and number of branches had positive and signficant
correlation in okra genotypes.

All phenology parameters of studied okra genotypes had negative and significant genotypic and
phenotypic correlation coefficients with fruit length (Table 6). The negative correlation of these
phenological traits with fruit length suggested that selection for earliness has negative effect on
 57

fruit length. Negative and significant correlations of traits with one another both at phenotypic and
genotypic levels will make difficult the simultaneous selection and improvement of traits (Akinyele
and Osekita, 2006). Muluken et al. (2016) observed negative and significant correlation between
days to pod maturity and fruit length.

The presence of strong genotypic and phenotypic correlation among phenology parameters of okra
genotypes suggested that the selection of genotypes based on one phenology traits resulted in the
selection of the genotypes for the other four traits. The positive and significant correlations between
most of phenology parameters with growth traits indicated that as the crop had delayed flowering
and maturity, did the higher mean values of genotypes for growth traits. However, only one or two
phenology parameters showed positive and significant correlations with only two yield components
(number of tender fruits per plant and seed yield per hectare) suggested that the late or early
flowering and maturity had negligible effects on yield components. Besides fruit length with all the
phenology parameters had negative and significant correlation coefficients suggested simultaneous
selection of genotypes for phenology parameters and fruit length was hardly possible since
selection for one vary the other trait in opposite direction.

Stem diameter had positive and significant correlation with pant height and number of internode
both at genotypic and phenotypic level. Plant height also had positive and significant correlation
with number of internode and internod length both at genotypic and phenotypic level. Number of
primary branches had positive and significant correlation with number of internode both at
genotypic and phenotypic level. However, number of primary branch had positive and
nonsignificant correlation with stem diameter and plant height both at genotypic and phenotypic
level.

Similar result was reported by Fozia (2018) Plant height with stem diameter showed positive and
significant association at phenotypic level. Number of branches and number of internodes per plant
had positive and significant correlations with plant height; stem diameter at genotypic and
phenotypic levels. Number of branches with number of internodes per plant at both levels had
positive and significant correlation coefficient. Kumar and Reddy (2016) reported the presence of
positive and significant phenotypic correlation among plant height, number of branches per plant
and internodes length.
 58

Stem diameter and plant height with fruit diameter, number of primary branches and number of
internode with number of tender fruit per plant and seed yield per hectare, and internode length
with average fruit weight and number of fruit ridges showed positive and significant genotypic and
phenotypic correlations. Fruit length and diameter with average fruit weight, number of fruit ridges
and number of seeds per pod, and average fruit weight with number of fruit ridges and number of
seeds per pod showed positive and significant correlation at genotypic and phenotypic levels.
Number of tender fruits per plant with seed yield per hectare and pod mucilage content while
number of fruit ridges and number of seeds per pod had strong genotypic and phenotypic
correlation. The genotypic correlation between hundred seed weight and seed yield per hectare was
positive and significant.

The genotypic and phenotypic correlations within and between growth traits and yield components
suggested that selection of genotypes for robust plant growth had a positive effect to increase the
mean performance of genotypes for yield components. Increasing internode number might increase
the number of tender fruit per pod this is because a pod of okra is usually borne in the leaf axil at
nodes of the stem (Aminu et al., 2016). Increase in primary branches results in increases of the
number of tender fruits per plant which is principal component of fruit yield. Kisher et al., (2016)
reported that selection of genotype with more primary branches leads to correlated selection for
more number of fruits per plant. Mihretu et al. (2014b) and Tesfa and Yosef (2016) reported fruit
length was positive and highly correlated to a number of seed per pod.

Akinyele and oseikita (2006), Saitwal et al. (2011), Reddy et al. (2013) and Muluken et al. (2016),
reported that plant height had positive and significant phenotypic and genotypic correlation with
stem diameter, number of branches per plant, number of internodes per plant, fruit diameter,
number of fruit per plant and number of seeds per pod. Kumar and Reddy (2016) reported growth
traits had positive and significant phenotypic correlation with number of fruits per plant, fruit length
and fruit weight.

The present study showed that negative and significant correlations were observed between number
of primary branches with fruit length, number of inter node with fruit length and fruit ridge,
internode length with fruit mucilage content, peduncle length with number of tender fruit per plant,
fruit yield per hectare, seed yield per hectare, number of harvest and mucilage content at phenotypic
 59

level. Negative and significant correlations of traits with one another make difficult the
simultaneous selection and improve traits (Akinyele and Osekita, 2006).

Average fruit weight with hundred seed weight, and hundred seed weight with seed yield per
hectare and pod mucilage content had positive and significant phenotypic correlations. Fruit
mucilage content showed negative and significant genotypic and phenotypic correlations with fruit
diameter, average fruit weight and number of fruit ridge, and it had also negative and significant
phenotypic correlations with fruit length and fruit dry matter content. The genotypic and phenotypic
correlation between number of tender fruits per plant and number of seed per pod was negative and
significant and the genotypic correlation between number of tender fruit per plant and number of
fruit ridges was negative and significant. The simultaneous selection of genotypes for those growth
traits and yield components showed either positive phenotypic correlations alone or negative
significant correlation at phenotypic and/or genotypic level may be hardly possible since traits had
not strong genetically correlations. Reddy et al. (2013), Aminu et al. (2016), Kisher et al. (2016)
and Muluken et al. (2016) reported that seed yield is correlated to number tender fruit per plant,
fruit length, number of seed per pod, fruit yield and hundred seed weight.

It was noted that traits that exhibited significant genotypic correlation coefficients showed also
significant phenotypic correlation. Moreover, the higher significant correlation magnitude were
observed for genotypic level than the cross ponding phenotypic correlation level which indicated
traits were genetically controlled over environments. Therefore, traits that are positive and
significantly correlated to each other at the genotypic level had to be directly advanced to selection
progress whereas those shows low correlation at phenotypic level and none correlated traits
suggested independence association that would possible to select independently in question for
diverse directions. This is in agreement with other authors reports that estimates of genotypic
correlation coefficients were in most cases higher than their corresponding phenotypic correlation
coefficients among agronomic traits of okra genotypes (Somashekhar et al., 2011; Reddy et al.,
2013). Akinyele and Osekita (2006) reported that negative and significant correlations both at
phenotypic and genotypic correlation with one another will be difficult to select for characterization
of desirable traits to those with the negative association. However, nonsignificant correlation will
be disregarded in selection for crop improvement.
 60

Table 6. Genotypic (above diagonal) and phenotypic (below diagonal) correlation coefficients among 24 quantitative traits

Traits Emer Dflo Flo Mat SD PH PB NI IL PL FL

ns ns ns ns ns ns
Emer 0.555** 0.543** 0.458** -0.007 0.176 0.226 0.192 -0.083 0.103 -0.34*
Dflo 0.489** 0.994** 0.890** 0.211ns 0.179ns 0.431** 0.261ns -0.080ns 0.333* -0.45**
ns ns ns ns
Flo 0.474** 0.989** 0.894** 0.224 0.199 0.403* 0.268 -0.073 0.337* -0.44**
ns ns ns ns ns
Mat 0.392** 0.852** 0.852** 0.294 0.219 0.381* 0.261 -0.063 0.317 -0.43**
ns ns ns ns ns ns
SD -0.012 0.190 0.199 0.281* 0.447** 0.136 0.503** -0.025 -0.205 0.067ns
PH 0.156ns 0.184ns 0.204ns 0.210ns 0.428** 0.163ns 0.484** 0.578** -0.022ns 0.079ns
PB 0.199ns 0.402** 0.369** 0.367** 0.131ns 0.160ns 0.395* -0.082ns -0.196ns -0.34*
NI 0.201ns 0.249* 0.255* 0.251* 0.469** 0.466** 0.372** -0.22ns -0.242ns -0.307ns
ns ns ns ns ns ns ns
IL -0.076 -0.057 -0.054 -0.052 -0.020 0.561** -0.083 -0.221 0.255ns 0.292ns
PL 0.064ns 0.307** 0.315** 0.312** -0.166ns -0.012ns -0.191ns -0.220ns 0.231ns -0.016ns
FL -0.298* -0.412** -0.405** -0.424** 0.066ns 0.085ns -0.322** -0.303** 0.277* -0.014ns
FD 0.183ns 0.156ns 0.178ns 0.214ns 0.452** 0.431** 0.035ns 0.199ns 0.275* 0.037ns 0.200ns
AFW -0.112ns -0.218ns -0.212ns -0.172ns 0.285* 0.274* -0.162ns -0.119ns 0.347** 0.089ns 0.828**
ns
NTFPP 0.195 0.367** 0.331** 0.285* 0.007ns 0.005ns 0.837** 0.397** -0.241ns -0.32** -0.49**
NR -0.215ns -0.253ns -0.247* -0.270* 0.101ns 0.034ns -0.152ns -0.358** 0.241* 0.029ns 0.515**
ns
yhaton 0.123 0.189ns 0.151ns 0.113ns 0.207ns 0.147ns 0.797** 0.208ns 0.012ns -0.24ns 0.108ns
NSPP -0.134ns -0.165ns -0.176ns -0.204ns 0.086ns -0.004ns -0.107ns -0.099ns 0.102ns -0.004ns 0.446**
SDW 0.134ns -0.111ns -0.131ns -0.095ns 0.040ns 0.047ns 0.159ns -0.008ns 0.147ns 0.015ns 0.116ns
ns ns ns ns
Syph 0.204 0.299* 0.260* 0.216 0.095 0.016 0.832** 0.372** -0.183ns -0.31** -0.26*
Nha 0.220ns 0.453** 0.417** 0.369** 0.169ns 0.152ns 0.757** 0.503** -0.135ns -0.25* -0.45**
ns ns ns ns ns ns ns ns
PDM -0.128 0.179 0.191 0.247* 0.137 0.045 -0.064 0.089 0.006 0.50** -0.063ns
PMC 0.175ns 0.085ns 0.072ns 0.035ns -0.017ns -0.048ns 0.224ns 0.229ns -0.252* -0.24* -0.30*
ns, *and **, nonsignificant, significant at P<0.05 and P<0.01, respectively. Eme= Days to 50% emergence, Dflo= Days to first flower, Flo= Days to
50% flowering, Mat= Days to 90% maturity, SD=Stem diameter (mm), PH= Plant height (cm), PB= Number of primary branch, NI= Number of internode, IL=
Internode length (cm), PL= Peduncle length (cm), FL= Fruit length (cm), FD= Fruit diameter (mm), AFW= Average fruit weight (g), NTFPP= Number of tender
fruit per plant, NR= Number of fruit ridge, YPP= Pod yield per plant (g), Yhaton= Pod yield per hectare (tons), NSPP= Number of seed per pod, SDW= Hundred
seed weight, Syph= Seed yield per hectare (Kg), Nha= Number of harvest, DMC= Dry matter content (%), PMC= Mucilage content (%).
 61

Table 5. Continued.

Traits FD AFW NTFPP NR yhaton NSPP SDW syph Nha PDM PMC
Emer 0.209ns -0.131ns 0.226ns -0.257ns 0.149ns -0.173ns 0.185ns 0.228ns 0.292ns -0.121ns 0.182ns
Dflo 0.167ns -0.238ns 0.405* -0.300ns 0.201ns -0.218ns -0.121ns 0.333* 0.501** 0.218ns 0.085ns
Flo 0.184ns -0.229ns 0.377* -0.296ns 0.172ns -0.219ns -0.124ns 0.305ns 0.470** 0.242ns 0.073ns
ns ns ns ns ns ns ns ns
Mat 0.217 -0.170 0.324 -0.309 0.128 -0.244 -0.124 0.241 0.420* 0.262ns 0.033ns
SD 0.473** 0.302ns 0.017ns 0.128ns 0.228ns 0.116ns 0.020ns 0.115ns 0.214ns 0.144ns -0.021ns
ns ns ns ns ns ns ns ns
PH 0.441** 0.280 -0.004 0.044 0.156 -0.002 0.069 0.015 0.174 0.058ns -0.049ns
PB 0.037ns -0.171ns 0.891** -0.257ns 0.800** -0.124ns 0.168ns 0.877** 0.797** -0.064ns 0.225ns
NI 0.202ns -0.123ns 0.443** -0.44** 0.243ns -0.110ns 0.036ns 0.432** 0.586** 0.111ns 0.238ns
ns
IL 0.290 0.364* -0.280ns 0.3375* 0.014ns 0.118ns 0.152ns -0.223ns -0.179ns 0.007ns -0.256ns
PL 0.033ns 0.093ns -0.312ns -0.009ns -0.242ns -0.010ns 0.039ns -0.310ns -0.242ns 0.552** -0.244ns
ns
FL 0.205 0.830** -0.54** 0.637** 0.104ns 0.490** 0.146ns -0.291ns -0.51** -0.073ns -0.308ns
FD 0.608** -0.179ns 0.469** 0.329* 0.340* 0.254ns 0.081ns 0.130ns 0.044ns -0.35*
AFW 0.599** -0.47** 0.676** 0.322ns 0.600** 0.287ns -0.109ns -0.325ns 0.086ns -0.35*
NTFPP -0.177ns -0.43** -0.51** 0.610** -0.36* 0.029ns 0.845** 0.837** -0.107ns 0.364*
NR 0.363** 0.554** -0.39** 0.119ns 0.646** 0.152ns -0.201ns -0.44** -0.001ns -0.48**
yhaton 0.302** 0.311** 0.599** 0.175ns 0.212ns 0.373* 0.820** 0.646** -0.010ns 0.131ns
NSPP 0.310** 0.549** -0.33** 0.50** 0.186ns 0.264ns 0.130ns -0.221ns 0.153ns -0.156ns
SDW 0.201ns 0.240* -0.014ns 0.128ns 0.311** 0.271* 0.339* 0.165ns 0.180ns 0.031ns
ns ns ns ns
Syph 0.062 -0.101 0.834** -0.123 0.789** 0.153 0.328** 0.820** 0.024ns 0.307ns
Nha 0.110ns -0.29* 0.791** -0.31** 0.636** -0.207ns 0.182ns 0.782** -0.002ns 0.335*
PDM 0.037ns 0.083ns -0.085ns 0.006ns -0.004ns 0.145ns 0.125ns 0.025ns -0.011ns -0.254ns
PMC -0.34** -0.35** 0.348** -0.37** 0.131ns -0.125ns 0.028ns 0.30* 0.302** -0.25*
ns, *and **, nonsignificant, significant at P<0.05 and P<0.01, respectively. Eme= Days to 50% emergence, Dflo= Days to first flower, Flo= Days to
50% flowering, Mat= Days to 90% maturity, SD=Stem diameter (mm), PH= Plant height (cm), PB= Number of primary branch, NI= Number of internode, IL=
Internode length (cm), PL= Peduncle length (cm), FL= Fruit length (cm), FD= Fruit diameter (mm), AFW= Average fruit weight (g), NTFPP= Number of tender
fruit per plant, NR= Number of fruit ridge, YPP= Pod yield per plant (g), Yhaton= Pod yield per hectare (tons), NSPP= Number of seed per pod, SDW= Hundred
seed weight, Syph= Seed yield per hectare (Kg), Nha= Number of harvest, DMC= Dry matter content (%), PMC= Mucilage content (%).
 62

4.5. Path Coefficient Analysis

4.5.1. Genotypic Path Analysis

In this study, the genotypic and phenotypic path analyses conducted that pod yield as dependent
variable and the other traits as independent variables. The results of genotypic path analysis area
presented in (Table 7). This study result revealed that seed yield per plant/or ha-1, number of
primary branches, number (frequency) of harvest, number of tender fruit per plant had a positive
and highly significant genotypic correlation with fruit yield of okra plant. Weight of hundred seed
and fruit diameter had a positive and significant genotypic correlation with okra pod yield and had
also had a direct effect on pod yield of okra plant.

The mere estimation of correlation coefficients does not give an idea about the real contribution of
an independent character to the yield. Path coefficient is a tool, which provides an effective measure
of the direct and indirect cause of association and depicts the importance of each factor involved in
contributing towards yield (Patil et al., 2017 and Simon et al., 2013). The correlation coefficients
were classified into indirect and direct effects utilizing the path coefficient analysis as suggested
by (Dewey et al., 1959). In order to obtain such developmental relations, the cause and effect
relationship between pod yield and yield components were studied in okra through path coefficient
analysis, the results of which are discussed below.

According to the rate set by Lenka and Mishra (1973), the direct and indirect effects into negligible
(0.00-0.09), low (0.10-0.19), moderate (0.20-0.29), high (0.30-1.00) and very high (>1.00). Based
on these rates the genotypic path analysis resulted that number of primary branches (0.706) and
seed yield per plant (0.621) had a positive high direct effect on okra fruit yield. But a number of
tender fruit per plant (-0.445) had a negative high direct effect on okra fruit yield and the rest traits,
fruit diameter (0.177), hundred seed weight (0.025), and number of harvests (-0.08) shows positive
low and negligible and negative negligible direct effect on okra fruit yield, respectively.

Number of a primary branch and seed yield per plant or/ha-1 had revealed a highly significant
correlation with fruit yield per hectare and they had also high positive direct effect on fruit yield
but the number of primary branches had negative high indirect effect on fruit yield via a number of
tender fruit per plant. Therefore, it is important to consider this trait directly while we are selecting
for improvement programs.
 63

Number of tender fruit per plant shows positive and highly significant genotypic correlation with
pod yield as well as a negative and high direct effect on pod yield of okra. Even though it had a
negative direct effect on pod yield; it had a high positive indirect effect on pod yield via a number
of primary branches per plant and seed yield per hectare. Selecting this character were possible
indirectly via a number of branch and seed yield per hectare. Number (frequency) of harvest had a
positive and highly significant genotypic correlation with pod yield and negative and negligible
direct effect on pod yield at genotypic level but this trait had a positive and high indirect effect on
pod yield via number primary of branches and seed yield per hectare.

Almost similar results reported by Rambabu et al. (2019) number of plant branch, number of tender
fruit per plant, seed yield was a high positive direct effect on pod yield and even other with low
and negligible direct effect had moderate and positive high indirect effects. He also summarizes
that traits with a positive direct effect on fruit yield per plant favors yield improvement through
selection and indirect selection based on plant height will be effective in yield improvement.
Prasath et al. (2017) reported that number of tender fruit per plant, number of the fruit harvest,
hundred seed weight and seed per pod had a direct positive effect on pod yield. Aminu et al. (2016)
reported the number of pods per plant had the greatest direct influence on pod yield, followed by
fresh weight per pod, which had a positive genotypic association with pod yield. Since it had
positive and high indirect effect on pod yield and high association at genotypic level; considering
this trait indirectly during selection will be better. But direct selection of this trait might be not
effective.

Table 7. Genotypic path analysis of 36 okra genotypes of six quantitative data those had
significant correlation with pod yield.

Traits PB FD NTFPP SDW Syph Nha gp

PB 0.706 0.006 -0.397 0.004 0.544 -0.064 0.800**
FD 0.026 0.177 0.080 0.006 0.050 -0.010 0.329*
NTFPP 0.629 -0.032 -0.445 0.001 0.525 -0.067 0.610**
SDW 0.119 0.045 -0.013 0.025 0.210 -0.013 0.373*
syph 0.619 0.014 -0.376 0.009 0.621 -0.066 0.820**
Nha 0.563 0.023 -0.373 0.004 0.509 -0.080 0.646**
Residual 0.182
PB= Nunmer of primary branches, FD= fruit diameter, NTFPP= Number of tender fruit per pod, SDW=
hundred seed weight, Syph= Seed yield per hectare, and Nha= number of harvest.
 64

4.5.2. Phenotypic Path Analysis

The current study result showed a positive and highly significant correlation of all traits at
phenotypic correlation levels as shown in (Table 8). Eventhough, it shows positive and highly
significant phenotypic correlation with fruit yield, it is mandatory to understand their direct and
indirect effect as well as their implications on fruit yield. According to partitioning range the
phenotypic path coefficient analysis resulted for average fruit weight (0.526) and number of
primary branches (0.498) had high and positive direct effect on okra fruit yield. Whereas, seed
yield per plant/or hectare (0.241) and number (frequency) of harvest (0.195) had positive moderate
direct effect on fruit yield. Eventhough, all these traits show highly significant phenotypic
correlation with fruit yield; fruit diameter (-0.059) negative, a number of tender fruit per plant
(0.044) and hundred seed weight (0.004) had a positive negligible direct effect on fruit yield.
Therefore, looking at the indirect effect of those traits could be effective. Among traits number of
tender fruit per plant had a positive and high indirect effect on fruit yield via a number of primary
branches, moderate indirect effect via seed yield per hectare and number (frequency) of harvest.

Similar result had been reported by Reddy et al. (2013) fruit weight and total number of fruits per
plant had a direct strong influence on fruit yield per plant and are the main determiners of fruit
yield per plant. Kaul et al. (1978) reported that seed yield per plant and number of primary branches
per plant had a maximum direct effect on fruit yield per plant. Average fruit weight had positive
and high direct effects on fruit yield, indicating their importance as reliable selection criteria for
the improvement of yield in okra (Subrata et al., 2004, Adeniji et al., 2007).

Table 8. Phenotypic path analysis result of seven quantitative traits and had significant correlation
with pod yield 36 okra genotypes

Traits PB FD AFW NTFPP SDW syph Nha rp

PB 0.498 -0.002 -0.085 0.037 0.001 0.200 0.147 0.797**
FD 0.017 -0.059 0.315 -0.008 0.001 0.015 0.021 0.302**
AFW -0.080 -0.035 0.526 -0.019 0.001 -0.024 -0.057 0.311**
NTFPP 0.417 0.010 -0.228 0.044 0.000 0.201 0.154 0.599**
SDW 0.079 -0.012 0.126 -0.001 0.004 0.079 0.035 0.311**
syph 0.414 -0.004 -0.053 0.037 0.001 0.241 0.152 0.789**
Nha 0.377 -0.007 -0.153 0.035 0.001 0.188 0.195 0.636**
Residual 0.116
PB= Nunmer of primary branches, FD= fruit diameter, AFW= average fruit weight, NTFPP= Number of tender
fruit per pod, SDW= hundred seed weight, Syph= Seed yield per hectare, and Nha= number of harvest.
 65

4.6. Clustering and Genetic Distances of Okra Genotypes

4.6.1. Clustering of Genotypes

The Euclidean distance matrix of 630 pair of genotypes estimated from 24 phenotypic traits was
used to construct dendrograms based on the Unweighted Pair-group methods with Arithmetic
Means (UPGMA). Accordingly, all 33 local collections and 3 checks are grouped into 13 distinct
clusters (Figure 1). The highest number of genotypes were grouped in the first Cluster contained
11 genotypes (30.56%) which included 3 checks, 1 from Guba, 2 from Menge, 3, from Kurmuk
and 2 from Assosa woredas followed by Cluster VIII which consisted of 6 (16.67%) genotypes and
all genotypes were collected from Assosa woreda. Cluster IX consisted of 4 (11.11%) genotypes
of which 2 each from Mandur and Assosa woredas, while Cluster X and III had 3 and 2 genotypes,
respectively. The other clusters (IV, V, VII, XI, XII, and XIII) were all solitary and each consisted
genotype (Figure 4 and Table 9).

Wassu et al. (2017) grouped 25 okra genotypes into seven major clusters in which the three clusters
(Cluster II, III and V) were solitary consisted of one genotype and each cluster had distinct
characters. Tesfa and Yosef (2016) from Melkasa have grouped 50 okra accession collected from
four major okra growing areas of the country into IV clusters. Muluken et al. (2016) reported 25
okra accessions grouped into ten major clusters and four clusters were solitary that each cluster
consists of one accession. They also reported that other six clusters consisted of more than one up
to the maximum 10 accessions. Mihretu et al. (2014) were also able to grouped 25 okra genotypes
collected from two regions into five major clusters. Davinder et al. (2018) from India reported 30
okra genotypes are clustered into six groups based on 10 quantitative characters. Clustering is a
multivariate technique that can conveniently show the pattern of genetic relationships or proximity
among accessions (Afifi and Clark, 1990). Each group is homogeneous with respect to certain
characteristics and each group should be different from other groups with respect to some
characteristics (Anderson, 1989).
 66

Figure 4. Dendrogram shows the dissimilarity of 36 okra genotypes based on 24 quantitative traits
by using UPGMA.
 67

Table 9. Number of genotypes grouped in 13 clusters, genotype code, and place of collection 36
okra genotypes evaluated at Melkassa in 2018/19

Number of Collection
Cluster No genotypes Genotypes Woredas
29408 Guba
29409 and 29417 Menge
29411, 29625 and 29412 Kurmuk
I 11 (30.56%)
29622 and 29413 Assosa
SOH701 and SOH714 India
Bamia Humera Humera
II 1 (2.78%) 29410 Menge
29414 Kurmuk
2 (5.56)
III 240207A Dibate
IV 1 (2.78%) 29415 Balojiganfoy
V 1 (2.78%) 29416 Balojiganfoy
29418 Guba
VI 3 (8.33%)
29616 and 29615 Assosa
VII 1 (2.78%) 29618 Assosa

VIII 6 (16.67%) 29052, 242445A, 29620,

29621, 29623, and 29619 Assosa
T240204 and 240209A Mandur
IX 4 (11.11%)
242433A and 242451A Assosa
T242443 and T242444 Menge
X 3 (8.33%)
29051 Tango
XI 1 (2.78%) 242449A Assosa
XII 1 (2.78%) 29617 Assosa
XIII 1(2.78%) 29624 Bambasi

4.6.2. Genetic Distances of Genotypes

The genetic distances of 630 pair of okra genotypes are presented in (Appendix Table 5). The mean
genetic distance of 36 okra genotypes was calculated to generate information about the most distant
and closest genotypes (Table 10). The genetic distance for all possible pairs of 36 genotypes ranged
from 2.83 to 12.24 with the mean, standard deviation, and coefficient of variation of 6.73, 1.63 and
24.18, respectively. The highest genetic distances (Euclidean distance) was computed between
accession 29618 and 29417 (12.24) followed by genotype [29618 to 29408 (12.04), 29409 (11.92)
and 29411 (11.48)], respectively. Whereas, the lowest was computed between genotypes 29625
and 29412 (2.83) followed by SOH714 and Bamia Humera (2.88), 242433A and 240209A (2.87),
 68

29620 and 29623 (3.051), respectively. The largest proportion 312 (49.52%) of pair of genotypes
had Euclidean distances of <6.73 (overall mean ED), small percentage 25(3.97%) pair of genotypes
had Euclidean distances of >9.56 and the remaining 293(46.51%) pair of genotypes had Euclidean
distances in between 6.73 to 9.56 (Table 10). The result suggested that the presence of a
considerable number of distant okra genotypes to others that could be used in crossing program to
combine the desirable traits of the genotype.

The genotype 29618 which collected from Assosa woreda on 1419 m.a.s.l. and two genotypes
29416 and 29415 collected from Balojiganfoy Woreda on 1195 and 1192 m.a.s.l. had higher genetic
distance among evaluated genotypes as well as checks (Table 10). Since most genotypes had
greater genetic distance than released variety and introduced varieties, there is a higher chance of
improving fruit yield and fruit related traits through selection and/or crossing of okra genotypes
from the regional state. Wassu et al. (2017) reported that Euclidean distances of 300 pair of 25
genotypes evaluated at Dire Dawa and reported the genetic distance ranged from 3.1 to 12.6 with
7, 2.2 and 27.85% mean, standard deviation and coefficient of variation, respectively. They also
reported the presence of high genetic distance between Ethiopian accessions. Muluken et al. (2015)
reported that Ethiopian okra collections exhibited wide genetic distances in the range between 5.16
and 11.14. Mihretu et al. (2014b) were also reported very high genetic distance among local
collections.

The genotype 29618 had the highest mean Euclidean distance of 8.94 followed by 29416 (7.93),
T240204 (7.63) and 29415 (7.62). Whereas genotypes 242433A (5.47) followed by genotypes
242445A (5.61) and 240209A (5.66) and 29413 (5.73) had the lowest Euclidean distance compared
all other genotypes (Table 10). The genotypes with high genetic distances between them have the
potential to produce heterotic hybrids through crossing made among genotypes. Among tested 33
local collections, 20 (60.6%) genotypes had mean genetic distances greater than the overall mean
genetic distance of genotype (Table 10). The results indicated that the genotypes were highest
distant to others or/and had genetic distance above the average to other genotypes and 13 (39.4) of
the new collection had less than the overall mean Euclidean distance. Whereas, the released variety
Bamia-Humera (6.05), and the two other introduced checks viz. SOH701 (5.98) and SOH714 (6.37)
had a distance lower than the mean Euclidean distances.
 69

This results revealed the presence of diverse okra genotypes with a wide range of genetic distances
which enables the researchers to improve the okra tender fruit yield and other desirable traits either
through direct selection or crossing of okra genotypes having different desirable traits. Availability
of genetically broad based variation for yield and its component traits is a prerequisite for the
development of new cultivars of okra. Okra breeders all over the world have been utilizing the
available genetic resources to modify the varieties (Reddy et al., 2012). Maximum genetic
recombination is expected from the hybridization of the parents selected from divergent
combinations (Mihretu et al., 2014).
 70

Table 10. Range, mean, standard deviation and Coefficient of variation Euclidean distances of 36
okra genotypes estimated from 24 quantitative traits evaluated at MARC in year 2018/19

No Genotype Minimum Maximum Mean SD Cv (%)

1 29408 4.66 12.03 7.07 1.79 25.25
2 29409 4.03 11.92 7.06 1.84 26.14
3 29410 4.89 10.14 6.89 0.94 13.62
4 29411 3.92 11.48 6.74 1.93 28.66
5 29412 2.83 10.83 6.52 1.92 29.39
6 29622 3.65 11.29 6.79 1.89 27.88
7 29414 3.83 8.93 6.71 1.38 20.63
8 29415 5.49 10.14 7.62 0.98 12.80
9 29416 4.97 9.78 7.93 1.26 15.85
10 29418 3.59 9.36 7.06 1.43 20.27
11 29616 3.13 8.99 6.31 1.52 24.15
12 29618 4.35 12.24 8.94 1.99 22.29
13 29052 4.06 9.62 6.85 1.52 22.15
14 T240204 4.47 9.66 7.63 1.41 18.53
15 29417 4.13 12.24 7.44 1.72 23.10
16 T242443 3.84 9.06 7.02 1.29 18.45
17 240207A 3.83 9.07 6.77 1.26 18.67
18 240209A 2.87 8.81 5.66 1.10 19.41
19 242433A 2.87 8.15 5.47 0.99 18.07
20 242449A 5.04 9.51 6.99 1.20 17.15
21 242445A 3.99 8.05 5.61 1.25 22.24
22 29051 3.81 9.13 5.95 1.09 18.31
23 29413 3.65 9.90 5.73 1.57 27.43
24 29615 3.13 9.90 7.11 1.65 23.21
25 29625 2.83 10.52 6.66 1.78 26.76
26 29617 4.54 9.24 7.45 1.27 17.10
27 29624 4.24 9.78 7.30 1.15 15.80
28 29620 3.05 9.63 6.70 1.46 21.76
29 29621 3.71 9.24 6.41 1.40 21.84
30 29623 3.05 8.89 6.06 1.36 22.51
31 29619 3.72 9.58 6.74 1.44 21.44
32 242451A 3.74 9.26 6.87 1.27 18.49
33 T242444 3.81 8.72 6.00 1.39 23.21
34 SOH701 3.36 9.91 5.98 1.57 26.26
35 SOH714 2.88 10.16 6.37 1.90 29.78
36 Bamia Humera 2.88 8.92 6.05 1.57 25.93
Overall 2.83 12.24 6.73 1.63 24.19
 71

4.6.3. Cluster Mean Analysis

The mean values of the 13 clusters for 24 quantitative traits were presented in (Table 11). The
unique features of cluster for plant phenology and growth traits are as follows. Among the overall
mean of thirteen clusters, Cluster I had the lowest mean values for days to first flowering, days to
90% maturity and a number of primary branches. Days to 50% emergence is early in cluster II and
III while late in cluster VIII, XI and XII. Cluster III had the lowest stem diameter from others. A
number of internode and internode length are poor traits of cluster IV and VI, respectively. Whereas
cluster XII shows the highest plant height but too late for flowering and maturity as compared to
other clusters. Cluster II, IV, V and VII had the highest mean values for stem diameter, internode
length, number of internode and number of primary branches, respectively among other clusters.
The shortest plant height was observed for cluster X. In short among thirteen okra clusters; cluster
I, II, III, VII, IX, and X contain almost early maturing genotypes. This indicates it is better to
develop early maturing varieties through further selection and/or crossing from these clusters
accompanied by further evaluation. But in contrary to this clusters V, XII, and XIII are too late as
compared to the overall mean values.

The clusters having distinguishing character in pod yield and related traits among other clusters
and also in comparison with the overall mean values of quantitative traits. Therefore, genotypes in
cluster IV, X, I and IX had longer fruit length and the highest fruit diameter were observed in
cluster V followed by cluster II, IX, and XI. Cluster IV, V, IX, X, and XI are distinguished by
higher average fruit weight. Cluster VII had the highest number of tender fruit per plant, pod yield,
seed yield and also mucilage contents and the lowest average fruit weight as compared to other
clusters. Cluster (III and IV), (IV, VI, III, and X), (V, VI, and XI), and (XI, X and VIII) had also
better in number of tender fruit per plant, pod yield, seed yield and percentage of mucilage contents
than the overall mean values of clusters. Whereas number of seed per pod is higher in cluster V
followed by IX, IV, and X respectively. Percentage of dry matter content is higher in cluster XIII
followed by IV and V respectively. On the other hand cluster VII was characterized by the lowest
average fruit weight, number of seed per pod and percentage dry matter contents. Whereas, seed
and pod yield, number of harvest as well as number of tender fruit per plant are lowest in cluster I.

Wassu et al. (2017) grouped 25 genotypes under seven clusters each clusters having distinguishing
characters. Muluken et al. (2015) were reported that 25 okra accession were clustered into ten
 72

distinct groups based on their quantitative and qualitative similarity of genotypes. Therefore, each
cluster had unique features among other clusters. Mihretu et al. (2014a) were also reported that
different clusters had distinguishing traits from others. Davinder et al. (2018) reported 30 okra
genotypes grouped into six clusters in which all clusters had their distinguishing traits. Oppong-
Sekyere et al. (2011) from Ghana reported 25 accessions grouped into 4 distinct clusters in which
each cluster had uniformity in their qualitative and quantitative features
 73

Table 11. Mean values of 13 clusters for 24 quantitative traits of 36 okra genotypes evaluated at Melkassa in year 2018/19

Cluster
Traits I II III IV V VI VII VIII IX X XI XII XIII Means
Eme 7.86 7.50 8.25 8.50 8.0 8.83 9.50 9.75 9.38 8.83 10.50 10.50 9.50 8.99
Dflo 48.00 54.50 48.25 61.0 60.5 64.50 58.50 59.42 51.63 49.67 54.0 71.0 60.50 57.04
Flo 52.73 60.0 52.0 66.0 67.0 70.33 63.50 65.08 56.25 54.0 59.50 77.50 67.0 62.38
Mat 80.86 87.50 81.0 92.0 102.5 95.83 87.50 94.67 85.25 81.50 89.50 103.0 96.0 90.55
SD 19.41 30.50 17.75 21.0 28.6 21.67 18.50 20.25 22.25 21.17 24.0 22.50 18.2 21.98
PH 117.02 178.0 137.50 155.5 148.5 104.35 115.50 126.53 164.21 88.30 186.50 194.50 141.0 142.88
PB 2.24 3.70 7.50 6.30 6.25 7.38 12.70 4.0 5.3 4.03 4.95 5.0 4.7 5.70
NI 21.23 29.00 29.75 18.50 39.0 26.67 30.50 28.33 27.75 23.83 32.50 31.50 26.50 28.08
IL 5.18 7.40 4.75 10.25 3.4 3.02 3.70 3.73 6.08 3.03 5.85 6.60 6.40 5.34
PL 2.13 2.10 1.75 3.30 2.3 1.93 1.90 2.41 1.93 2.04 1.53 1.88 3.45 2.20
FL 14.26 9.87 10.77 18.80 13.25 8.18 5.77 6.85 14.17 15.79 12.15 12.45 5.50 11.37
FD 26.04 35.15 23.08 26.34 35.2 27.20 24.10 26.89 33.35 27.79 29.39 28.95 29.10 28.66
AFW 39.09 39.66 23.36 60.32 55.9 30.85 17.04 23.14 55.46 47.0 46.12 25.36 23.66 37.46
NTFPP 13.27 20.40 37.25 17.0 24.4 33.63 55.20 22.70 18.65 19.73 26.80 24.90 22.0 25.84
NR 6.96 6.90 5.75 7.0 7.2 6.20 5.70 5.52 7.30 6.93 6.0 6.30 6.50 6.48
YPP 473.89 670.23 977.77 1016 869.66 993.69 1183.6 502.65 957.41 872.13 924.80 541.91 556.10 810.76
Yhaton 13.16 18.62 27.16 28.22 24.16 27.60 32.88 13.96 26.59 24.23 25.69 15.05 15.45 22.52
NSPP 82.07 63.20 64.55 94.0 106.0 81.87 60.50 66.10 95.18 92.07 80.70 65.20 63.50 78.07
SDW 5.75 5.28 5.90 6.84 5.96 5.55 6.02 5.92 6.19 6.31 7.16 5.24 6.24 6.03
SYPP 63.31 65.46 139.93 109.23 154.08 152.26 198.04 87.51 109.31 114.69 150.53 84.79 87.46 116.66
Syph 1758.7 1818.4 3887.1 3034.1 4280.1 4229.5 5501 2431 3036.5 3185.8 4181.3 2355.3 2429.3 3240.6
Nha 5.94 7.30 7.95 6.80 7.50 8.0 8.0 7.13 7.03 6.97 7.60 7.70 7.40 7.33
DMC 25.85 24.56 25.58 30.20 32.48 26.99 22.56 25.87 24.59 24.82 27.88 22.04 34.55 26.77
PMC 11.15 11.02 13.84 12.01 7.17 13.89 20.01 16.43 11.59 19.91 19.93 16.01 7.85 13.91
Eme= Days to 50% emergence, Dflo= Days to first flower, Flo= Days to 50% flowering, Mat= Days to 90% maturity, SD=Stem diameter (mm),
PH= Plant height (cm), PB= Number of primary branch, NI= Number of internode, IL= Internode length (cm), PL= Peduncle length (cm), FL= Fruit
length (cm), FD= Fruit diameter (mm), AFW= Average fruit weight (g), NTFPP= Number of tender fruit per plant, NR= Number of fruit ridge,
YPP= Pod yield per plant (g), Yhaton= Pod yield per hectare (tons), NSPP= Number of seed per pod, SDW= Hundred seed weight, SYPP= Seed
yield per pod (g), Syph= Seed yield per hectare (Kg), Nha= Number of harvest, DMC= Dry matter content (%), PMC= Mucilage content
 74

4.7. Principal Component Analysis

The principal component analysis (PCA) of 24 quantitative traits are presented in (Table 12). The
results was also included the factor score of each trait among the 36 okra genotypes, Eigenvalues
and the percentage of contribution to the total variability accounted for 4 principal components.
Principal component analysis (PCA) was computed to find out the traits which accounted more to
the total variation (Chahal and Gosal, 2002).

This principal component analysis resulted in four principal components (PC1 to PC4) with
eigenvalues ranged from 7.582 to 1.832. The four principal components accounted for the varied
percentage of total variance of 31.591%, 18.397%, 13.754%, and 7.596% for PC1, PC2, PC3, and
PC4, respectively. These four components accounted for a total of 71.34% cumulative
contributions. In PCA principles, if >50% of the variations catches with the PCs in which each
contribution is having >10% contribution and Eigenvalue >1, it is acceptable (Table 12). Therefore,
since the total variation of PC1 to PC4 > 50%, the other could be ignored. There was no guideline
to determine the significance of eigenvectors (Duzyaman, 2005). The higher coefficients for traits
substantiated the relatedness of that trait with respective PC axis (Broschat, 1979).

The total contribution of the four principal component axis of this study result was higher and
similar with the results reported by other authors. Osawuru et al. (2014) reported the variation
observed up to five principal component axis ranges from 6.90 to 22.97% for PC1 to PC5,
respectively. They also reported the cumulative variation of five PCA was accounted 70.2% of
variation. Mihretu et al. (2014b) reported six principal components for 20 traits of 25 okra
genotypes in which eigenvalues were 10.65, 3.04, 2.41, 1.7, 1.62, and 1.32 which accounted 83%
of the cumulative variation. Muluken et al. (2016) reported the first three principal components
PC1, PC2 and PC3 with values of 32.4%, 16.7%, and 8.2%, respectively and contributed more to
the total of 57.3% variation. Asare and Asare-bediako (2016) also reported eigenvalues of four PCs
3.42, 1.34, 1.11 and 1.06 from PC1 to PC4 respectively that accounted for 77% of cumulative
variations. Davinder et al. (2018) reported four principal components for ten quantitative traits and
the eigenvalues of each component was 3.414, 3.215, 1.239 and 0.915 from PC1 to PC4,
respectively which accounted for 87.84% of cumulative variation. Ahiakpa (2012) reported that
principal component axis contributed 64.32% of cumulative variation.
 75

According to Chahal and Gosal (2002), traits with the largest absolute values closer to unity with
in the first principal component influence the clustering more than those with lower absolute values
closer to zero. Therefore, in the present study, the differentiation of the traits was due the
cumulative effect of number of traits rather to the small contribution of each trait
(± 0.006 to 0.923).

The traits that highly contributed to differentiation were in principal component (PC-I) was number
(frequency) of fruit harvest, number of tender fruit per plant, number of primary branches, seed
yield per plant, pod yield, days to first flowering, days to 50% flowering, number of internode, and
days to 90% maturity had relatively high contribution. For PC-2 average fruit weight, number of
fruit ridge, number of seed per pod, fruit length, fruit diameter, and fruit yield had relatively higher
contributions. In third principal component days to 90% maturity, days to 50% flowering, days to
first flowering, fruit peduncle length and fruit diameter are relatively major contributors (Table 12).
 76

Table 12. The principal component values of four principal component from 24 quantitative traits
for 36 okra genotypes evaluated at Melkassa in 2018/19

No Trait Eigenvectors
PC1 PC2 PC3 PC4
1 Days to 50% emergence 0.469 -0.164 0.370 0.015
2 Days to first flower 0.634 -0.231 0.641 -0.169
3 Days to 50% flowering 0.611 -0.233 0.667 -0.151
4 Days to 90% maturity 0.559 -0.223 0.687 -0.086
5 Stem diameter (mm) 0.232 0.348 0.351 0.549
6 Plant height (cm) 0.170 0.323 0.469 0.597
7 Number of primary branch 0.891 0.191 -0.147 -0.098
8 Number of internode 0.578 0.023 0.100 0.565
9 Internode length (cm) -0.247 0.363 0.337 0.090
10 Peduncle length (cm) -0.190 -0.160 0.627 -0.540
11 Fruit length (cm) -0.541 0.652 -0.052 -0.012
12 Fruit diameter (mm) 0.083 0.639 0.491 0.213
13 Average fruit weight (g) -0.317 0.847 0.224 0.006
14 Number of tender fruit per plant 0.902 -0.123 -0.301 -0.034
15 Number of fruit ridge -0.470 0.701 0.091 -0.147
16 Pod yield per plant (g) 0.682 0.638 -0.182 -0.148
17 Pod yield per hectare (tons) 0.682 0.638 -0.182 -0.148
18 Number of seed per pod -0.218 0.675 -0.016 -0.215
19 Hundred seed weight (g) 0.157 0.476 -0.072 -0.251
20 Seed yield per pod (g) 0.876 0.296 -0.281 -0.173
21 Seed yield per hectare (Kg) 0.876 0.296 -0.281 -0.173
22 Number of harvest 0.923 0.031 -0.074 0.076
23 Dry matter content (%) 0.025 0.046 0.445 -0.399
24 Mucilage content (%) 0.380 -0.280 -0.360 0.180
Eigenvalue 7.582 4.415 3.301 1.823
Difference 3.167 1.114 1.478 0.503
Contribution to Variability (%) 31.591 18.397 13.754 7.596
Cumulative contribution % 31.591 49.988 63.742 71.338
PC = Principal Component.

Mihretu et al. (2014b) reported that days to first flowering, number of seed per pod, number of
tender fruit per plant, internode length and number of internode were major contributors of diversity
in okra plant and they confirms the presence of genetic diversity for further improvement programs.
Pradip et al. (2010) reported plant height, number of internode and fruit ridges as a major
contributor of okra diversity. In contrary Muluken et al. (2015) reported the maximum contributors
on component axis were leaf width followed by days to 50% flowering, pod yield per plant, hundred
 77

seed weight, stem diameter, internode length, fruit length, plant height, peduncle length and number
of tender fruit per plant. Relative reports by Ahiakpa et al. (2013) reported that the first four PC-
axis contributed for 82.97% of the variations in okra was number of branches per plant, days to
50% flowering, intermodal length, number of fruits per plant and fruit yield contributed to the
variation in PC1. Stem diameter, days to 50% flowering, internodal length, fruit weight, and fruit
yield accounted for the variations observed in PC2. Plant height, stem diameter, fruit diameter and
fruits per plant contributed to the variations in PC3. Stem diameter and fruit diameter contributed
to variations in PC4. These variations may suggest the existence of genetic diversity in okra that
can be an input to improve the crop.

Davinder et al. (2018) reported an assessment of relative contribution of ten characters towards
total genetic divergence revealed that branches per plant had contributed highest (41.61%) followed
by fruit yield per plant (31.49%), fruits per plant (15.17%) and stem diameter (1.61). He also
suggests characters with the highest genotypic variability should be considered while selecting
parents for hybridization programs.

In general, this result implied that traits such as number of harvests, number of tender fruit per
plant, number of primary branches, seed yield, pod yield, days to flowering, number of internodes,
maturity date, average fruit weight, fruit length, and fruit diameter which associated with PC1 and
PC2 are implicated for being responsible for the phenotypic divergence observed in the cultivars
and can be used for cultivar discrimination for improvement program. The traits which contributed
much in each PC1 to PC4 also presented in figure for visual observation (Figure 5).
 78

Figure 5. Biplot (axes PC1, PC2, PC3 AND PC4) OF 24 quantitative traits of 36 okra genotypes
evaluated at Melkassa 2018/19.
 79

5. SUMMARY AND CONCLUSION

It is believed that Ethiopia is both a center of origin and diversity for okra (Abelmoschus esculentus)
which is mainly known for its edible pod around south western and western parts of Ethiopia. It is
grown traditionally around Benishangul Gumuz Regional State mainly for its edible pod which had
cultural value. Even though, it is important vegetable throughout the world it’s negligible for our
country as well as regional state in terms of export and domestic use. This may arised due to lack
of appropriate agricultural technology for okra plant i.e. improved varieties and agricultural
packages. Therefore, it’s one of unexploited and underutilized genetic resource in Ethiopia.
Therefore, this study was conducted with the objective of assessing the genetic divergence and
estimate the genetic variability components in okra genotypes collected from Regional State, and
determine the association of traits and the direct and indirect effects of traits on fruit yield. The
field experiment was conducted in simple lattice design at MARC in year 2018/19.

This research identified the presence of significant difference among genotypes at (P<0.01) and
(P<0.05) level of significance for all quantitative traits except leaf length, leaf width and number
of epicalyxes. The two major yield in okra plant which is pod and seed yield ranged from 9.44 to
32.88ton ha-1 with mean yield of 19.58ton ha-1 and 929.95 to 5501.01Kg ha-1 with a mean of
2769.34Kg ha-1, respectively. Among local collection many genotypes had better performance in
terms of pod and seed yield as compared to checks Viz. Bamia Humera, SOH701 and SOH714.
This indicated that the genotypes collected from regional state had high variability both in terms of
fresh pod yield as well as seed yield. Therefore, the result of this study is an indicator for the
presence of higher chance to develop okra varieties for both pod and seed yield through
selection/crossing of okra genotype collected from Benishangul Gumuz regional state.

The okra genotypes collected from regional state had wide variation for leaf, fruit, flower and seed,
shape, color and other quality factors like pubescence which determinant factor in preference of
consumer.

The genotypic (GCV) and phenotypic (PCV) coefficient of variation of 24 pod and pod yield
related traits of 36 okra genotypes under study revealed within the range of 6.3 to 54.19 and 9.19
to 55.51%, respectively. The heritability in broad sense (H2) and genetic advance as percent of
mean (GAM) had a ranges 43.46 to 97.34% and 9.16 to 109.14%, respectively. The result of
 80

variability components indicates that number of primary branches, followed by a number of tender
fruit per plant, internode length, seed yield, average fruit weight, pod yield, fruit length, percentage
of mucilage content, plant height, internode length, and peduncle length had higher GCV and PCV;
while days to 90% maturity and hundred seed weight had the low GCV and PCV. In addition this
traits had shown high heritability and genetic advance as percent of mean except moderate for
number of ridge in both and hundred seed weight and percentage of dry matter content,
respectively. But low GAM for hundred seed weight. This result confirmed that traits were highly
heritable and variation observed was due to genetic contributions.

The association of traits like genotypic and phenotypic correlation coefficient between fruit yield
and related traits were positive and significantly correlated with number of primary branches,
number of tender fruit per plant, seed yield, and number of harvests both at genotypic and
phenotypic correlations whereas average fruit weight was only correlated at phenotypic levels.
Hundred seed weight and fruit diameter are positive and significantly correlated to fruit yield at
(P<0.01) phenotypic correlation and significantly correlated at (P<0.05) at genotypic correlations.
All traits with positive and significant genotypic correlation levels were higher in magnitude than
the corresponding phenotypic correlations coefficient.

Number of primary branches had positive and highly direct effect on fruit yield at both genotypic
and phenotypic path whereas, seed yield per plant/or ha-1 and average fruit weight had positive
and significantly high direct effect at genotypic and phenotypic level, respectively. Number of
tender fruit per plant and (frequency) of harvest had high negative and negative negligible direct
effect genotypically, in respective sequence, however they had positive and high indirect effects
via number of branches and seed yield per hectare. Consequently this traits should be considered
indirectly during selection. In addition to this number of harvest and fruit diameter had negative
and negligible direct effect on pod yield genotypically and phenotypically, respectively. Even
though they are negligeble they had contributed for yield via other traits. Therefore all those traits
listed above had significant importance during selection for pod yield improvements.

The genetic distance of 36 okra genotypes ranged from 2.83 to 12.24 with the mean, standard
deviation, and coefficient of variation of 6.73, 1.63 and 24.18, respectively. The largest proportion
312 (49.52%) of pair of genotypes had Euclidean distances of <6.73 (overall mean ED), small
percentage 3.97% (25) Pair of genotypes had Euclidean distances of >9.56 and the remaining 293
 81

(46.51%) pair of genotypes had Euclidean distances between 6.73 to 9.56. Among genotypes under
study 20 (55.56%) had mean genetic distance of >6.73 (overall mean distances of genotypes) and
16 (44.44%) including all the three checks had a mean genetic distance of <6.73 (overall mean
distances of genotypes). The Euclidean distance matrix of 630 pair of genotypes estimated from 24
quantitative traits was used to construct dendrograms and accordingly, 33 local genotypes and 3
checks are grouped into 13 distinct clusters. The highest number of genotypes are grouped in the
first cluster and contains of 11 genotypes (30.56%) followed by cluster VIII which consisted of 6
(16.67%) genotypes and cluster IX accommodates 4 (11.11%). Cluster X and III consisted of 3 and
2 genotypes, respectively. The rest clusters IV, V, VII, XI, XII, and XIII are all solitary consisted
of each one genotype. The 13 cluster varied for varying number of traits such as the eight of clusters
(III, IV, V, VI, VII, XI, X and XI) had higher fruit yield performance >22.52 t ha -1 (overall mean
performance of clusters for fruit yield) as compared to all other clusters. The four principal
components (PC1 to PC4) accounted for a total of 71.34% cumulative contributions to total
variations, in which PC1 and PC2 had larger contribution of 31.591 and 18.397%, respectively,
while PC3 and PC4 contributed 13.754 and 7.596%, respectively.

The study revealed the presence of wide variations among okra genotypes for all agro morphology
traits except for three traits. The high heritability and genetic advance as percent of mean were
estimated for quantitative traits. The genotypes also exhibited genetic divergence grouped to 13
distinct clusters and had variations for varying traits. This suggested the higher chance of
developing varieties either through selection and/or hybridization of okra genotypes for
Benishangul Gumuz regional state. It is recommended to conduct similar experiment over seasons
and locations since this research was conducted for one season and at one location and it is better
if molecular characterization is to be followed the current agro morphological characterization and
evaluation work.
 82

6. REFERENCES

Adekoya, M.A., Ariyo, O.J., Kehinde, O.B., Adegbite, A.E. 2014. Correlation and path
analyses of seed yield in okra (Abelmoschus esculentus (L.) Moench) grown under
different cropping seasons. Pertanika Journal of Tropical Agricultural and Science, 37(1):
39–49.
Adelakun, O.E., Oyelade, O.J., Ade-Omowaye, B.I.O., Adeyemi, I.A., Van de Venter, M. and
Koekemoer, T.C. 2009. Influence of pre-treatment on yield chemical and antioxidant
properties of a Nigerian okra seed (Abelmoschus esculentus Moench) flour. Food and
chemical toxicology, 47(3): 657-661.
Adeniji O.T. 2007. Heritability and Number of Genes Governing Pod Yield in West African Okra
(Abelmoschus caillei (A.chev) Stevens. Agricultural Journal, 2 (4): 483-486.
Adeniji, O.T., Kehinde, O.B., Ajala, M.O. Adebisi, M.A. 2007. Genetic studies on seed yield of
West African okra (Abelmoschus caillei) (A chev.) Stevels). Journal of Tropical
Agriculture, 45(1–2): 36–41.
Adeniji, O.T. and Peter, J.M. 2005.Stepwise regression analysis of pod and seed yield characters
in segregating F2 population of West African okra (Abelmoschus caillei). Proceedings of
30th Conference, Genetics Society of Nigeria, pp..250-258.
AdeOluwa, O.O. and O.B. Kehinde. 2011. Genetic variability studies in West African Okra
(Abelmoschus caillei). Agriculture and Biological Journal of North America, 2(10): 1326–
1335.
Adeoluwa, O.O. and Kehinde, O.B. 2013. Genetic variability studies in West African okra
(Abelmoschus caillei). Agriculture and Biology Journal of North America.; 2(10):1326-
1335.
Adetuyi, F.O., and Dada, I.B,O. 2014. Nutritional, phytoconstituent and antioxidant potential of
mucilage extract of Okra (Abelmoschus esculentus), water leaf (Talinum triangulare) and
Jews mallow (Corchorus olitorius). International Food Research Journal, 21(6): 98–107.
Afifi, A.A. and Clark, V. 1990. Computer aided multivariate analysis. Van Nostrand
Reinhold.NewYork, USA, USA, Pp 505.
Ahiakpa, J.K. 2012. Characterization of twenty-nine (29) accessions of okra [Abelmoschus Spp (L.)
Moench] in Ghana.Master of Philosophy, University of Ghana, Ghana.
 83

Ahiakpa, J.K., Kaledzi, P.D., Adi, E.B., Peprah, S. and Dapaah, H.K. 2013. Genetic diversity,
correlation and path analyses of okra [Abelmoschus spp. (L.) Moench] germplasm
collected in Ghana. International Journal of Development and Sustainability, 2(2):1396-
1415.
Ahiakpa, J.K., Amoatey, H.M., Amenorpe, G., Apatey, J., Ayeh, E.A., and Agbemavor, W.S.K.
2014. Mucilage Content of 21 accessions of Okra (Abelmoschus spp L.). Scientia
Agriculture, 2(2): 96–101.
Akanbi, W.B., A.O. Togun, J.A., Adediran, and E.A.O. Ilupeju. 2010. Growth, dry matter and fruit
yield components of okra under organic and inorganic sources of nutrients. American-
Eurasian Journal of Sustainable Agriculture, 4(1): 1–13.
Akash, M.W., S.M. Shiyab and M.I. Saleh. 2013. Yield and AFLP analyses of inter–landrace
variability in okra ( Abelmoschus esculentus L.). Life Science Journal, 10(2): 2771–2779.
Akingbala, JO., Akinwande, BA., and Uzo-Peters, PI. 2003. Effects of color and flavor change on
the acceptability of ogi supplemented with okra seed meals. Plant Foods Human Nutrition,
58:1-9.
Akinyele B.O, Oseikita OS. 2006. Correlation and path coefficient analyses of seed yield attribute
in okra (Abelmoschus esculentus (L.) Moench). African Journal of Biotechnology, 5: 1330-
1336.
Aminu, D., Bello, O.B., Gambo, B.A., Azeez, A.H., Agbolade, O.J., Iliyasu, A. and Abdulhamid,
U.A., 2016. Varietal performance and correlation of okra pod yield and yield
components. Acta Universitatis Sapientiae, Agriculture, and Environment, 8(1), pp.112-
125.
Anderson, T.W. 1989. An introduction to multivariate statistical analysis. John Wileyand Son,
New York. PP 675.
Ansari, N.M., Houlihan, L., Hussain, B., and Pieroni, A. 2005. Antioxidant activity of five
vegetables traditionally consumed by South-Asian migrants in Bradford, Yorkshire, UK.
Phytotherapy Research, 19(10): 907– 911.
Anteneh Bekele. 2017. Characterization and Evaluation of Indigenous and Exotic Okra
[Abelmoschus esculentus (L.) Moench] Genotypes for Yield and Yield Related Traits at
Dire Dawa, Eastern Ethiopia. MSc Thesis, Haramaya University, Ethiopia.
 84

Anwar F., Umer R., Zahid M., Tahira I., Tufail H. and Sherazi. 2011. Inter-varietal variation in the
composition of okra [Hibiscus esculentus (L).] seed oil. Pakistan Journal of Botany.
43(1):271-280.
Ariyo, O. J. 1993. Genetic diversity in West African Okra (Abelmoschus caillei (A.Chev.) Stevels)
– Multivariate analysis of morphological and agronomic characteristics. Genetic Resources
and Crop Evolution. 40:125-132.
Ariyo, O.J., Aken‟Ova, M.E., Fatokun, C.A. 1987. Plant character correlation and path analysis of
pod yield in okra (Abelmoschus esculentus). Euphytical, 36: 677-686.
Asare, A.T., Asare-Bediako, E., Agyarko, F., Taah, K. and Osei, E.O., 2016. Phenotypic traits
detect genetic variability in Okra (Abelmoschus esculentus L. Moench). African Journal of
Agricultural Research, 11(33), pp.3169-3177.
Baredo Yishak. 2013: Gamo Gofa Zone Diagnosis and Planning Document. Livestock and
irrigation value chains for Ethiopian Smallholders (LIVES) project research report,
Addis Ababa, Ethiopia.
Basu, A.N., 1995. Bemisia tabaci (Gen.) Crop pest and principal whitefly vector of plant
viruses. West view
Bello, D., Sajo, A.A., Chubado, D. and Jellason, J.J. 2006. Variability and correlation studies in
okra (Abelmoschus esculentus (L.) Moench). Journal of Sustainable Development in
Agriculture and Environment, 2(1): 120-126.
Bello, O.B., M.S. Afolabi, S.A. Ige, S.Y. Abdulmalik, M.A. Azeez, J. Mahmood. 2011. Yield
response of diallelic crossed maize (Zea mays L.) cultivars to varying nitrogen regimes in
Nigeria. Journal of BioSciences, 19: 43–52.
Bendale, V.W, Kadam, S.R, Bhave, S.G, Mehta, J.L. and Pethe, U.B. 2003. Genetic variability and
correlation studies in okra. Orissa J. Hort., 31(2): 1-4.
Bhandari, H.R., Bhanu, A.N., Srivastava K., Singh M.N. and Shreya. 2017. Assessments of genetic
diversity in crop plants-An overview. Advance plant.
Biswal, B., Karna, N., and Patel, R. 2014. Okra mucilage act as a potential binder for the
preparation of tablet formulation. Der Pharmacia Lettre, 6(3): 31–39.
Board, J.L. and Sutcliffe, C.M., 1988. The weekend effect in UK stock market returns. Journal of
business finance & accounting, 15(2): 199-213.
 85

Brooker, R.J. 2012. Concepts of genetics. Population genetics. (1st edition). McGraw-Hill
Companies, America .Pp 4-627.
Broschat, T.K. 1979. Principal component analysis in horticultural research. Horticultural Science,
14: 114-117.
Burton GW, De Vane EH. 1953. Estimating heritability in tall fescue (Fistvea arundiancea) from
the replicated clonal material. Agriculture Journal, 45: 284-291.
Célestin Dabandata, Bell Joseph Martin, Nsimi Mva Armand, Amougou Akoa, Ngalle Hermine
Bille, and Ndoumbe Nkeng Michel. 2012. Heritability And Correlations Study In Okra
[Abelmoschus esculentus (L.) Moench]. Continental Journal of Agronomy, 6(1): 24 – 29.
Chahal, GS and Gosal SS. 2002. Principles and Procedures of Plant Breeding Biotechnology and
Conventional Approaches. Narosa Publishing House, New Delhi. Pp 604.
Chandra, D.E.O., Singh, K.P., Panda, P.K. and Deo, C. 1996. Genetic variability, correlation and
path analysis in okra. Environment and Ecology, 14 (2): 315 – 319.
Chandra, S., Bhardwaj, M.L., Kumar, R., Kumar, D., Kumar, S., Gautam, N., Dogra, B. and
Sharma, S., 2014. Estimation of parameters of variability for different quantitative traits in
okra, [Abelmoschus esculentus (L) Moench]. International Journal of Farm Sciences, 4(3):
33-41.
Chattopadhyay A, Dutta S, Chatterjee S. 2011. Seed yield and quality of okra as influenced by
sowing dates. African Journal of Biotechnology 10: 5461-5467.
Dakahe, K., Patil, H.E. and Patil, S.D. 2007. Genetic variability and correlation studies in okra
(Abelmoschus esculentus (L) Moench.). Asian Journal of Horticulture, 2(1): 201-203.
Daniela, F.O., M. Alicia, R.C. Alicia, H.Rodolfo and Z.V. Sonia. 2012. Quality attributes of okra
(Abelmoschus esculentus (L.) Moench) pods as affected by cultivar and fruit size. Journal
of Food Research, 1(4), 224–235.
Das, S,A, Chattopadhyay, S.B. Chattopadhyay, S.Dutta and P.Hazra. 2013. Breeding okra for
higher productivity and yellow vein mosaic tolerance. International Journal of Vegetable
Science, 19: 58–77.
Datta, P.C. and Naug, A. 1968. A few strains of Abelmoschus esculentus (L) Moench. Their
karyological in relation to phylogeny and organ development. Beitr. Biology of Pflanzen,
45: 113-126.
 86

Dattijo Aminu, Omolaran Bashir Bello, Babagana Abba Gambo, Alafe Hakeem Azeez, Oludare
James Agbolade, Ali Iliyasu, and Usman Abdulrahman Abdulhamid. 2016.Varietal
performance and correlation of okra pod yield and yield components. Acta Universitatis
Sapientiae Agriculture and Environment. 8 (16): 112−125.
Davinder Singh, B.S. Dudi, S.K. Dhankhar and Rajkumar. 2018. Genetic Diversity Analysis of
Okra Genotypes Using Morphological Markers. International Journal of Current
Microbiology and Applied Sciences, 7 (1): 2319-7706
Deblock, C. and Haji, R. 2008. Commission on Growth and Development, the Growth Report.
Strategies for Sustained Growth and Inclusive Development, Washington, the
International Bank for Reconstruction and Development World Bank. 2008. CEIM:
Montréal, mai.
Deckelbaum R. 1999. Summary of a scientific conference on preventive nutrition: pediatrics to
geriatrics. Circulation, 100(4):450-456.
Demissie Tsegaye., Ahmed Ali and Dilnesaw Zerfu. 2009. Availability and consumption of fruits
and vegetables in nine regions of Ethiopia with special emphasis to vitamin A deficiency.
In: Ethiopian Journal of Health Development 23 (3): 216-222.
Deogade, U.M., Deshmukh, V.N., and Sakarkar, D.M. 2012. Natural gums and mucilage are in
NDDS: Applications and recent approaches. International Journal of PharmTech
Research, 4(2): 799–814.
Department of Agriculture, Forestry and Fisheries Republic of South Africa, 2012. Okra (Hibiscus
esculentus).
Dewey, D.R., Lu, K.H. 1959. A correlation and path coefficient analysis of components of crested
wheatgrass seed production. Agronomy Journal of Horticulture, 57(4): 342–346.
Dhankhar, B.S., and Mishra, J. P. 2009. Origin, history, and distribution. In B. S. Dhankhar and R.
Singh (Eds.), Global production, processing, and crop improvement, Okra handbook, New
York: 3–12.
Dicko, A., Natta, A.K., Biaou, H.S.S. and Akossou, A., 2019. Assessing Morphological Traits
Variation and Fruit Production of Lophira lanceolata (Ochnaceae) in Benin. American
Journal of Plant Sciences, 10(6): 1048-1060.
 87

Datta,. P.C and Naug, A. 1968. A few strains of Abelmoschus esculentus (L) Moench. Their
karyological in relation to phylogeny and organ development, Beitr, Biol, Pflanzen.45. pp
113-126.
Duzyaman, E. 2005. Phenotypic diversity within a collection of distinct okra (Abelmoschus
esculentus L.) cultivars derived from Turkish landraces. Genetics Research and Crop
Evolution, 52: 1019-1030.
EBI (Ethiopian Biodiversity Institute), 2018. Conserved Genetic Resource Data Base Information.
Ehab. AA. I., Mohamed.Y.A. and Ali M.M. 2013. Genetic behavior of families selected from some
local okra [Abelmoschus esculentus (L.) Moench] populations in Egypt. Plant Breeding
Biotechnology, 1(4): 396-405.
EHDA (Ethiopian Horticulture Development Agency) 2011/2012: Exporting Fruit and
Vegetables from Ethiopia: Assessment of development potentials and investment options
in the export-oriented fruit and vegetable sector. Addis Ababa, Ethiopia.
Esmail, R.M., Hendawy, F.A., Rady, M.S. and Hamid, A.M. 1999. Genetic studies on yield and
yield component in one inter and two intraspecific crosses of cotton. Egyptian Journal of
Agronomy, 21: 37-51.
Ethiopian Investment Commission. 2012: Investment opportunity profile for production of fruits
and vegetables in Ethiopia. Ethiopian Privatization Agency, Addis Ababa, Ethiopia.
Falconer DS, Mackay TFC, 1996. An Introduction to quantitative genetics. Ed, 4.Printice Hall
London. 1996;464.
FAOSTAT, 2004. (http://fao.org).
FAOSTAT. 2010. Economic and Social Department. The Statistics Division. Major Food and
Agricultural Commodities and Producers. Retrieved from
http://faostat.fao.org/site/339/default.aspx
Fozia Yimam. 2018. Genetic diversity and association of seed yield and related traits of Okra
[Abelmoschus esculentus (L.) Moench] in Ethiopia. MSc thesis, Haramaya University,
Haramaya, Ethiopia.
Gemede Habtamu Fekade, Haki Gulilat Dessie, Beyene Fikadu, Rakshit SK, Woldegiorgis
Ashagrei. 2018. Indigenous Ethiopian okra (Abelmoschus esculentus) mucilage: A novel
ingredient with functional and antioxidant properties. Food Science and Nutrition, 200:1–
9. https://doi.org/10.1002/fsn3.596
 88

Gerrano, A. S. 2017. Preliminary study on variability and heritability estimates of micro nutrient
composition in the immature fruits of okra (Abelmoschus esculentus L. ) genotypes in
South Africa. Cogent Food & Agriculture, 9(1): 14.
Getachew Tabor and Mohammed Yesuf, 2012. Mapping the current knowledge of carrot
cultivation in Ethiopia. Research Report Submitted to Carrot Aid (August), Denmark.
Downloadable at http://www.carrotaid.org/reports/taboryesuf2012.pdf, accessed on April
19, 2014.
Gopalan C, Sastri SBV, Balasubramanian S. 2007: Nutritive value of Indian foods, National
Institute of Nutrition (NIN), ICMR, India.
Habtamu Fekede Gemede, Nigussei Ratta., Gulelat Dessie Haki., Ashagrei Z Woldegiorgis, and
Fekadu Beyene. 2015. Nutritional quality and health benefits of Okra (Abelmoschus
esculentus): A review. Journal of Food Processing and Technology, 6: 458.
Habtamu Fekede Gemede, and Negussie Ratta. 2014. Antinutritional factors in plant foods:
Potential health benefits and adverse effects. International Journal of Nutrition and Food
Sciences, 3: 284–289.
Habtamu Fekede Gemede, Gulelat Dessie Haki., Fekadu Beyene., Ashagrie Z Woldegiorgis., and
Rakshit, S. K. 2016. Proximate, mineral, and antinutrient compositions of indigenous Okra
(Abelmoschus esculentus) pod accessions: Implications for mineral bioavailability. Food
Science and Nutrition, 4(2): 223–233.
Hammer, K. and Mbewe, D. N. 1994. The Role of Traditional Knowledge in Germplasm
Collecting. In: A. Putter (ed). Safeguarding the Genetic Basis of Africa's Traditional
Crops. CTA, the Netherland/IPGRI/Rome: 147-155.
Hammon S, and Van Stolen DH. 1989. Characterization and evaluation of okra. The Use of Plant
Genetic Resource. 173-174.
Hamon, S. and Nairot, M. 1991. Some proposed procedures for obtaining a core collection using
quantitative plant characterization. International Workshop on okra genetic resources held
at NBPGR. International Crop Network Series, 5: 89 – 94.
Hamon, S. and Yapo, A. 1986. Peturbation induced within the genus Abelmoschus by the discovery
of a second edible okra species in West Africa. In: Maesen, L.J.G. Van Der (Ed.), First
International Symposium on Taxonomy of Cultivated Plants. Acta Horticulture. 182:133-
144.
 89

Haruna, S., Aliyu, B. S., and Bala, A. 2016. Plant gum exudates (Karau) and mucilages, their
biological sources, properties, uses and potential applications: A review. Bayero Journal
of Pure and Applied Sciences, 9(2): 159–165.
Hazra P, Basu D. 2000. Genetic variability, correlation and path analysis in okra. Annals of
Agriculture Research, 21(3): 452-453.
Henry, A. and Krishna, G.V. 1990. Correlation and path coefficient analysis in pigeon pea. Madras
Agriculture Journal, 77 (9-12): 443-446.
Institute of Biodiversity Conservation of Ethiopia, 2012. Ethiopia: Third Country Report on The
State of Plant Genetic Resources For Food and Agriculture. Institute of Biodiversity
Conservation (IBC) Addis Ababa, October 2012.
Ijoyah, M. O. and D. M. Dzer. 2012. Yield Performance of okra (Abelmoschus esculentus L.
Moench) and maize (Zea mays L.) as affected by time of planting maize in Makurdi,
Nigeria. Agriculture and Biological Journal of North America, 1(6): 1328– 1332.
IPGRI. 1991. Okra Descriptor list. International Crop Network Series 5. International Board for
Plant Genetic Resources (IBPGR), Rome, Italy.
Jaiprakashnarayan, R.P. and Mulge, R. 2004. Correlation and path analysis in okra (Abelmoschus
esculentus (L.) Moench). Indian Journal of Horticulture, 61(3): 232- 235.
Jideani, V. A., and Bello, B. M. 2009. Functional properties of okra protein products containing
different levels of mucilage. Journal of Food, Agriculture & Environment, 7(2): 252–255.
Jindal, S. K., Arora, D., and Ghai, T. R. 2009. "Heterobeltiosis and combining ability for earliness
in okra (Abelmoschus esculentus (L.) Moench)", Crop Improvement, 36 (2): 59-66
Johnson, H.W., H. F. Robinson and R. E. Comstock. 1955. Estimation of the genetic and
environmental variability of soybean. Agronomy Journal, 47: 314-318.
Lush, J.L. 1940. Intersire correlations and regression of offspring on dams as a method of
estimating heritability of traits Proceedings of American Society of Animal Breeding 33:
293-301.
Kaewmanee, T., Bagnasco, L., Benjakul, S., Lanteri, S., Morelli, C. F., Speranza, G., & Cosulich,
M. E. 2014. Characterisation of mucilages extracted from seven Italian cultivars of flax.
Food Chemistry, 148: 60–69.
 90

Kaul, T. G., and Peter, K. V. 1978. Correlation and path coefficient analysis of components of
earliness, pod yield and seed yield in okra. Indian Journal of Agricultural Science, 48: 459-
463.
Kisher, D. S., K. Arya, K. J. Yogeesh, K. Y. Vinod and K. Hee-Jong. 2016. Genotypic variation
among okra (Abelmoschus esculentus (L.) Moench) germplasms in South Indian Plant
Breeding and Biotechnology, 4(2): 234–241.
Kumar, P. S. and S. Sreeparvathy 2010. Studies on heterosis in okra (Abelmoschus esculentus (L.)
Moench). Electronic Journal of Plant Breeding, 1(6): 1431–1433.
Kumar, S. and Reddy, M.T., 2016. Correlation and path coefficient analysis for yield and its
components in okra (Abelmoschus esculentus (L.) Moench). Advances in Agricultural
Science, 4(1), pp.72-83.
Kumar, S., Dagnoko, S., Haougui, A., Ratnadass, A., Pasternak, D., and Kouame, C. 2010. Okra
(Abelmoschus spp.) in West and Central Africa: Potential and progress on its improvement.
African Journal of Agricultural Research, 5(25): 3590–3598.
Lamont, W. J. 1999. Okra-A versatile vegetable crop. Horticultural Technology, 9: 179–184.
Lenka, D. and Mishra, B. 1973.Path coefficient analysis of yield in rice varieties. Indian Journal
of Agricultural Sciences, 43: 376-379.
Martin, F. W. Rhodes AM, Ortiz M, and Diaz F. 1983. Variation in okra. Euphytica 30:697-705.
Maundu, P. M. 1995. Methodology for collecting and sharing indigenous knowledge: a case study.
Indigenous knowledge and Development Monitor 3(2): 3-5.
Melkassa Agricultural Research Center. 2019. Home page datas. http://www.eiar.gov.et/marc
Mehta, D.R., Dhaduk, L.K. and Patel, K.D. 2006. Genetic variability, correlation and path analysis
studies in okra (Abelmoschus esculentus (L.) Moench). Agricultural Science Digest, 26(1):
15-18.
Mihretu Yonas, Weyessa Garedew and Adugna Debela, 2014a. Multivariate Analysis among okra
[Abelmoschus esculentus (L.)Moench] collection in South Western Ethiopia. Journal of
Biological science, 9(2): 43-50.
Mihretu Yonas, Weyessa Garedew and Adugna Debela, 2014b.Variability and association of
quantitative characters among okra [Abelmoschus esculentus (L.)Moench] collection in
South Western Ethiopia. Journal of Biological science, 14: 336-342.
 91

Mirhosseini, H., and Amid, B. T. 2012. A review study on chemical composition and molecular
structure of newly plant gum exudates and seed gums. Food Research International, 46(1):
387–398.
MoANR (Ministry of Agriculture and Natural Resources). 2016. Crop variety registers Issue No.
19, p.211. Addis Abeba, Ethiopia.
Moekchantuk, T. and Kumar, P., 2004. Export okra production in Thailand. Inter-country
programme for vegetable IPM in South and SE Asia phase II Food and Agriculture
Organization of the United Nations, Bangkok, Thailand, 56.
Mohammadi, S.A. and Prasana, B.M. 2003.Analysis of genetic diversity in crop plant salient.
Statistical tools and considerations. Crop Science 43(4):1235-1248.
Muhammad, R. S., A. Muhammad, Z. Khurram, M. J. Muhammad, A. Saeed, I. Qumer and N.
Aamir. 2013. Growth, yield and seed production of okra as influenced by different growth
regulators. Pakistan Journal of Agricultural Science, 50(3): 387–392.
Muluken Demelie, Wassu Mohammed, and Endale Gebre, 2015. Genetic Diversity of Ethiopian
Okra Collections through Multivariate Analysis at Werer, Rift Valley of Ethiopia:, The
International Journal Of Science and Technolodge, 3(8): 186–193.
Muluken Demelie, Wassu Mohammed and Endale Gebre. 2016. Variability, Heritability and
Genetic Advance in Ethiopian Okra [Abelmoschus esculentus (L.) Monech] Collections for
Tender Fruit Yield and Other Agro-morphological Traits’, Journal of Applied Life Sciences
International,4(1): 1–12.
Nath, P. 1976. Vegetables for the Tropical Region. ICAR. New Delhi.
Noorlaila, A., Siti Aziah, A., Asmeda, R., and Norizzah, A. R. 2015. Emulsifying properties of
extracted Okra (Abelmoschus esculentus L.) mucilage of different maturity index and its
application in coconut milk emulsion. International Food Research Journal, 22(2): 122–
131.
Nosil, P., Harmon, L.J. and Seehausen, O., 2009. Ecological explanations for (incomplete)
speciation. Trends in ecology and evolution, 24(3):145-156.
Ogunniyan, D. J., and Olakojo, S. A. 2014. Genetic variation, heritability, genetic advance and
agronomic character association of yellow elite inbred lines of maize (Zea mays L.).
Nigerian. Journal of Genetics, 28(2): 24–28
 92

Olaoye, G., O. B. Bello, L. S. Olayiwola and A. Y. Abubakar. 2009. Analyses of moisture deficit
grain yield loss in drought tolerant maize (Zea mays L.) germplasm accessions and its
relationship with field performance. African Journal of Biotechnology, 8(14): 3229‒3238.
Olawuyi, O.J., O. B. Bello, C., V. Ntube, and A. O. Akanmu. 2015. Progress from selection of
some maize cultivars’ response to drought in the derived savanna of Nigeria. Agrivita,
37(1): 8‒17.
Oppong-Sekyere D, Akromah R, Nyamah EY, Brenya E, and Yeboah S, 2011. Characterization of
okra (Abelmoschus spp. L.) germplasm based on morphological characters in Ghana.
Journal of Plant Breeding and Crop Science; 3(13):367-378.
Oppong-Sekyere, D., Akromah, R., Nyamah, E.Y., Brenya, E. and Yeboah, S., 2012. Evaluation
of some okra (Abelmoschus spp L.) germplasm in Ghana. African Journal of Plant
Science, 6(5), pp.166-178.
Osawaru, M. E.; Ogwu, M. C., and Omologbe, J. 2014. Characterization of three Okra
[Ablemoschus esculentus (L.)] accessions using morphology and SDS-PAGE for the basis
of conservation. Egyptian Academic Journal of Biological Sciences H. Botany, 5(1): 55 –
65.
Osawarus, M.E., Ogwu, M.C. and Aiwansoba, R.O .2015.Hierachical approaches to the analysis
of genetic diversity in plants: A systematic Overview. University of Mauritius .Research
Journal, 21:1-33.
Osekita, O. S. and B. O. Akinyele. 2008. Genetic analysis of quantitative traits in ten cultivars of
okra – Abelmoschus esculentus (L.) Moench. Asian Journal of Plant Sciences, 7: 510–513.
Patel, M., Ganga, M., Jawaharlal, M. and Jeyakumar, P., 2017. Assessment of Qualitative,
Quantitative and Visual Flower Quality Parameters of Certain Commercial Jasmine
Varieties during Peak Flowering Season. Int. J. Curr. Microbiol. App. Sci, 6(11), pp.3246-
3251.
PGRC. 1995, Ethiopia: Country report to the FAO international technical conference on plant
genetic resources (Leipzig), 1996. Plant Genetic Resources Center, Addis Ababa, Ethiopia.
PhaniKrishna M., Hameedunnisa Begum, A. Manohar Rao1 and N. Sunil Kumar. 2015. Estimation
of Heritability and Genetic Advance in Okra [Abelmoschus esculentus (L.) Moench.]. Plant
Archives, 15 (1): 489-491.
 93

Pluess, A.R. and Stöcklin, J., 2004. Population genetic diversity of the clonal plant Geum reptans
(Rosaceae) in the Swiss Alps. American Journal of Botany, 91(12): 2013-2021.
Prasath G., Reddy K.R., and Pidigam Saidaiah. 2017. Correlation and Path Coefficient Analysis of
Fruits Yield and Yield Attributes in Okra (Abelmoschus esculentus (L.) Moench).
International Journal of Current Microbiology and Applied Sciences ISSN: 2319-7706
Volume 6 (3): 463-472
Prakash, M., T.B. Alloli, D. Satish, C.P. Mansur, Venkateshalu, S. Raghavendra and Hadimani,
H.P. 2017.Genetic diversity analysis in okra [Abelmoschus esculentum (L.) Moench]
Genotypes for Productivity and Quality Traits. International Journal of Current
Microbiology and Applied science 6(10):2951-2956.
Pradip, K., Akotkar, D.K. De and Pal, A.K. 2010. Genetic variability and diversity in okra
[Abelmoschus esculentus (L).Moench]. Electronic Journal of Plant Breeding, 1(4): 393-
398.
Purquerio, L. F. V., Lago, A. A., Passos, F. A. 2010. Germination and hard seedness of seeds. in
okra elite lines. Horticulture Brasil Brasília, 28(2): 23–32.
Rambabu, B., D.P. Waskar and Khandare, V.S. 2019. Correlation and Path Coefficient Analysis of
Fruits Yield and Yield Attributes in Okra [Abelmoschus esculentus (L.) Moench].
International Journal of Current Microbiology and Applied Science. 8(04): 764-774.
Rashwan, A.M.A., 2011. Study of genotypic and phenotypic correlation for some agro-economic
traits in okra (Abelmoschus esculentus (L.) Moench). Asian Journal of Crop Science, 3(2):
85-91.
Rath, U. K. Sahu, G. S. and Mishra, R. S. 1991. Studies on genetic variability in okra. Orissa
journal of Agriculture Research, 4(1-2): 48-51.
Reddy, M. T., K. H. Babu, M. Ganesh, H. Begum, R. S. K. Reddy and J. D. Babu. 2013.
Exploitation of hybrid vigor for yield and its components in okra [Abelmoschus esculentus
(L.) Moench]. American Journal of Agricultural Science and Technology, 1: 1–17.
Reddy T.M., Haribhau, K., Ganesh, M., Chandrasekhar R.K., and Begum, H. 2012. Genetic
divergence analysis of indigenous and exotic collections of Okra [Abelmoschus esculentus
(L.) Moench]. Journal of Agricultural Technology, 8(2): 611-623.
 94

Ren, J., J. R. McFerson, R. Li, Kresovich and W.F. Lamboy, 1995. Identities and relationships
among Chinese vegetable Brassicas as determined by random amplified polymorphic
DNA markers. Journal of the American Society for Horticultural Science 120: 548-555.
Robinson H.F. 1966. Quantitative genetics in relation to breeding on the centennial of Mendelism.
Indian Journal of Genetics., 26: 171-187.
Ruel M.T, Minot N, Smith L. 2005. Patterns and determinants of fruit and vegetable consumption
in Sub-Saharan Africa: A multi-county comparison. Int. Food Policy Res Inst Washington
DC, USA.
Sabesan, T., Suresh, R., and Saravanan, K. 2009. Genetic variability and correlation for yield and
grain quality characters of rice the grown in the coastal saline low land of Tamil Nadu.
Electronic Journal of Plant Breeding, 1: 56–59.
Saifullah M and M. G. Rabbani. 2009. Evaluation and characterization of Okra (Abelmoschus
esculentus L. Moench.) cultivars. SAARC Journal of Agriculture, 7(1): 92–99.
Samim S., Sonia S., Akhilesh S., Anuradha V. and Amandeep K. 2018. Morphological
Characterization of Okra [Abelmoschus esculentus (L.) Moench]. International Journal of
Current Microbiology and Applied Sciences ISSN: 2319-7706 Volume 7(10). Department
of Vegetable Science and Floriculture CSK Himachal Pradesh Krishi Vishvavidyalaya,
Palampur (HP)-176 062, India
Sanjeet K., Sokona D., Adamou H., Alain R., Dov P., and Christophe, K. 2010. Okra (Abelmoschus
spp.) in West and Central Africa: Potential and progress on its improvement. African
Journal of Agricultural Research, 5(25): 3590-3598.
Sankara, R.K. and Pinaki, A. 2012. Performance of okra [Abelmoschus esculentus (L.) Moench]
cultivars under summer and rainy environments. International Journal of Advanced Life
Sciences 2: 17-26.
Santos BM, Dittmar PJ, Olson SM, Webb SE, Zhang S. 2012. Okra production in Florida.
University of Florida IFAS Extension. ; 163-171.
SAS Institute. 2008. SAS software. SAS Institute INC., Cary. NC. USA.
Sathish, D., and Eswar, A. 2013. A review on [Abelmoschus esculentus (L.) (Okra)]. International
Research Journal of Pharmaceutical and Applied Sciences, 3: 129–132.
 95

Saurabh, D. B., Ekta, K., Sunita, S. 2015. A natural polymer as a pharmaceutical excipient and
their application in the different pharmaceutical formulation. A review world journal of
Pharmaceutical Research.
Sharma, J.R., 1998. Statistical and Biometrical Techniques in Plant Breeding. New Age
International (P) Limited Publishers, New Delhi. Pp 432.
Shimelis Aklilu. 2000: Research achievement on variety development and seed production of
vegetable crops in Ethiopia. In: Chadha, M.L., E.C. Altoveros, R. Nono-Womdim and
H. Mndiga (eds.): AVRDC Africa Regional Program 2000. Varietal evaluation and
seed production of vegetable crops. Proceea dings of a workshop held at AVRDC Africa.
Regional Program, Arusha, Tanzania from 29 September to 05 October 1997.
AVRDC ARP Publication No. 2000-2. Asian Vegetable Research and Development
Center – Africa Regional Program, Arusha, Tanzania: 6-11.
Shui, G., and Peng, L. L. 2004. An improved method for the analysis of major antioxidants of
(Hibiscus esculentus L.) Journal of Chromatography A, 1048(1): 17–24.
Shivaramegowda, K.D., Krishnan, A., Jayaramu, Y.K., Kumar, V. and Koh, H.J. 2016. Genotypic
variation among okra (Abelmoschus esculentus (L.) Moench) germplasms in South
India. Plant Breeding and Biotechnology, 4(2): 234-241.
Shujaat, A., Azhar, H. S., Rehmani, G., Habib, A., Hasnian, N. and Sikandar, K. S. 2014. Morpho-
agronomic characterization of okra [Abelmoschus esculentus (L.) Moench]. World Applied
Sciences Journal, 31 (3): 336-340.
Sibsankar. D., Arup, C., Sankhendu, B. C., Subrata, D. and Pranab, H. 2012. Genetic parameters
and path analysis of yield and its components in okra at different sowing dates in the
Gangetic plains of eastern India. African Journal of Biotechnology, 11(95): 16132-16141.
Simon, S. Y; Gashua, I. B, and Musa, I. 2013. Genetic variability and Trait Correlation Studies in
Okra (Abelmoschus esculentus (L.) Moench). Agriculture and Biology Journal of North
America, 4(5): 532-538.
Singh, B.D. 1993. Plant Breeding Principles and Methods. Kalyani Publishers. Ludhiana. New
Delhi.
Singh, R.K. and Chaudhary. 1977. Biometrical methods in quantitative genetic analysis. Kalyani
publishers, New Delhi-Ludhiana, India.
 96

Singh, K.B., and Singh, H.N. 1977. Path coefficient analysis for yield in okra. Indian Journal of
Agricultural Sciences, 49: 244-246.
Singh, S. and Pawar, I.S.2005. Theory and Application of Biometrical Genetics. CBS Publishers,
India.
Sivasubramaniah S, Meron M. 1973. Heterosis and inbreeding depression in rice. Madras
Agriculture Journal, 60:1139-1144.
Sneath, P.H.A., and Sokal, R.R. 1973. Numerical Taxonomy: the principles and practice of
numerical classification. Fransisco USA W.F Freeman, Pp. 573.
Somashekhar, G., Mohankumar, H.D. and Salimath, P.M. 2011. Genetic analysis of association
studies in segregating population of okra. Karnataka Journal of Agricultural Science.
24(4), 432-435.
Sorapong Benchasr. 2012. Okra (Abelmoschus esculentus (L.) Moench) as a Valuable Vegetable
of the World. Ratarstvo povrtarstvo, 49:105-112.
Subrata, S., Hazra, P. and Chattopadhyay, A. 2004. Genetic variability, correlation and path
analysis in okra (Abelmoschus esculentus L. Moench). Horticulture Journal, 17(1): 59-66.
Sugri I, Abdulai M. S, Larbi A, Hoeschle-Zeledon I, Kusi F. and Agyare R. Y. 2015. Participatory
variety selection of okra (Abelmoschus esculentus L.) genotypes for adaptation to the semi-
arid agro-ecology of Northern Ghana. African Journal of Plant Science, 9(12): 466-475.
Swati, B. Reena, N., Meenakshi, R. and Jain PK.2014. Genetic variability in okra [Abelmoschus
esculentus (L.) Moench]. An International Quarterly Journal of Environmental Sciences,
6:153-156.
Tesfa Binalfew and Yosef Alemu. 2016. Characterization of Okra (Abelmoschus esculentus (L.)
Moench) Germplasms Collected from Western Ethiopia. International Journal of
Research in Agriculture and Forestry, 3(2):11-17.
Tesfaye Awas. 2007. Plant Diversity in western Ethiopia: Ecology, Ethnobotany, and
Conservation, Dissertation presented for the degree of Doctoral Philosophy, Department
of Biology, Faculty of Mathematics and Natural Sciences, University of Oslo, Norway.
Thirupathi, R. M., B. K. Hari, M. Ganesh, R. K. Chandrasekhar, H. Begum, R. B. Purushothama
and G. Narshimulu. 2012. Genetic variability analysis for the selection of elite cultivars
based on pod yield and quality from the germplasm of okra [Abelmoschus esculentus (L.)
Moench].Journal of Agricultural Technology, 8, 639-655.
 97

Tripathi, K.K., Govila, O.P., Ranjini, W. and Vibha, A. 2011. Biology of okra [Abelmoschus
esculentus (L).Moench].serious of crop specific biology document. Ministry of
Environment and forests government of India and department of biotechnology ministry
of science and technology government of India; 22.
USDA, 2000. Nutrition and your health: dietary guidelines for Americans. Home and Garden Bull.
United State of America, Department of Agriculture, Washington, D.C.
Uzma Farooq, Rishabha Malviya and Pramod Kumar Sharma. 2013. Extraction and
Characterization of Okra Mucilage as Pharmaceutical Excipient. Academic Journal of
Plant Sciences, 6 (4): 168-172.
Valeriana, P. J. 2011.Training Resource Text on Crop Development, Major Economic Practices,
Disease and Insect Ecology, Insect Pests, Natural Enemies and Diseases of Okra.
Varmudy V. 2011: Marking survey needs to boost okra exports. Department of economics,
Vivekananda College, Puttur, Karnataka, India.
Vavilov, N.I. 1951. The origin, variation, immunity and breeding of cultivated plants. Chron.
Botany. 13: 1949-1950.
Voss, R. and Bell, M. 2007. Okra .World Vegetables, 2nd ed. Aspen Publications. Pp 843.
Wassu Mohammed., Anteneh Bekele, and Kumar, V. 2017. Characterization and Evaluation of
Okra [Abelmoschus esculentus (L.) Moench] Collections in Eastern Ethiopia. Edited and
Compiled by, p.211- 238. Proceedings of 34th Annual review workshop, 6-8 April 2017.
Haramaya University, Ethiopia.
Weerasekara, D., Jagadeesha, R. C., Wali, M. C., Salimath, P. M., Hosamani, R. M. and
allappanavar, K. 2007. Heterosis for yield and yield components in okra. Vegetable
Science. 34(1): 106-107.
World Health Organization, 2003. Diet, nutrition and the prevention of chronic diseases. Report of
a joint WHO/FAO expert consultation. (V. 916). World Health Organization, Geneva;
Switzerland.
Yasin, G. M., and S. Anbu. 1997. Variability Studies in Bhendi. South Indian Horticulture, 45: 13-
15.
Yiguo, L. 2005. Diallel and stability analysis of kenaf (Hibiscus cannabinus L.) in South Africa.
Master of Science in Agriculture, University of the Free State Bloemfontein, South Africa.
 98

7. APPENDICES
 97

Appendix Table 1. Mean values of 36 okra genotypes for crop phenology and growth traits evaluated at Melkassa in year 2018/19.

Genotype Emer Dflo Flo Mat SD PH PB NI IL LL LW Nep

29408 7.5ef 47.5k-o 53i-l 77lmn 24cde 128.7f-i 1.3r 26.5fg 2.65m 22.75a 27.5ab 9.4abc
29409 7.5ef 49j-o 54.5i-l 81.5j-m 18ijk 148de 1.75qr 16.5lm 8.3b 14.5cde 20.5a-d 11.5a
29410 7.5ef 54.5e-k 60e-l 87.5f-i 30.5a 178ab 3.7k-n 29def 7.4bc 17.5a-e 24.5a-d 10abc
29411 8.5c-f 49j-o 53.5i-l 84hij 20.5f-j 126f-k 2.7n-q 19jkl 7.6bc 15.5b-e 17.5cd 11.3ab
29412 8.5c-f 47.5k-o 53i-l 78k-n 18.5ijk 138.5d-j 2.05pqr 22hij 5.15e-h 14.5cde 19.5a-d 10.7abc
29622 7.5ef 50.5j-o 55h-l 88fgh 17.5jk 78.8o 2.7n-q 13.5m 3.3klm 15.5b-e 21.5a-d 10.3abc
29414 9b-e 52g-m 55.5h-l 85.5f-i 14.5l 136e-h 8.5b 26fg 4.7f-j 16b-e 25a-d 9abc
29415 8.5c-f 61cde 66c-f 92def 21e-i 155.5cd 6.3ef 18.5jkl 10.25a 20.5ab 26abc 10.2abc
29416 8def 60.5c-f 67b-e 102.5a 28.6ab 148.5de 6.25efg 39a 3.4j-m 18.5a-e 24.5a-d 8.8bc
29418 8.5c-f 68.5ab 75ab 97.5bc 22d-h 117.5h-m 7.8bc 26.5fg 3.3klm 18.5a-e 26.5ab 10.7abc
29616 9.5bcd 61cde 66c-f 90fg 22.5d-g 122.5g-l 7.6bcd 27efg 3lm 18a-e 25.5a-d 9.6abc
29618 9.5bcd 58.5c-h 63.5c-h 87.5f-i 18.5ijk 115.5i-m 12.7a 30.5cde 3.7i-m 17a-e 24a-d 9.5abc
29052 9.5bcd 61cde 67b-e 104.5a 25cd 172bc 8.2b 32.5bcd 4.5g-k 19.5a-d 23.5a-d 9.35abc
T240204 8def 46mno 51.5i-l 75.5n 20.5f-j 184.75ab 5.7e-i 35.5b 7.4bc 20.5ab 28a 11.5a
29417 8.5c-f 51.5h-n 57g-k 89.5fg 18ijk 127f-j 1.6qr 16.5lm 7.3bc 18.5a-e 22.5a-d 10.2abc
T242443 9.5bcd 51i-o 55h-l 81.5j-m 22d-h 75.2o 5.7e-i 17.5lk 3.2klm 18.5a-e 22a-d 11ab
240207A 7.5ef 44.5no 48.5kl 76.5mn 21e-i 139d-g 6.5de 33.5bc 4.8f-i 18.5a-e 22a-d 9.8abc
240209A 10abc 53.5f-l 58f-j 85f-i 20f-j 136e-h 4.2j-m 19jkl 7.5bc 16.5b-e 25a-d 9.7abc
242433A 9cde 53.5f-l 57.5f-j 91ef 22.5d-g 149.8de 5.2f-j 23.5ghi 3.7g-k 19a-e 27.5ab 10.05abc
242449A 10.5ab 54e-l 59.5e-j 89.5fg 24cde 186.5ab 4.95h-k 32.5bcd 5.85def 18.5a-e 22.5a-d 10abc
242445A 8def 55.5d-j 60d-i 91.5def 22d-h 107.9j-n 4.3j-m 27efg 3.65i-m 17.5a-e 24.5a-d 9.35abc
29051 8.5c-f 52g-m 55h-l 81j-m 22.5d-g 116.5i-m 2.7n-q 33.5bc 3.4j-m 20abc 24.5a-d 10.1abc
29413 8.5c-f 52g-m 55.5h-l 83ijk 21e-i 89.45no 2.3o-r 22hij 3.25klm 19a-e 25.5a-d 9.5abc
 98

Appendix Table 1 Continued.

Genotypes Eme DFlo FlO Mat SD PH PB NI IL LL LW Nep

29615 8.5c-f 64bc 70abc 100ab 20.5f-j 73.05o 6.75cde 26.5fg 2.75m 20.5ab 25a-d 10abc
29625 7.5ef 46mno 52i-l 78.5k-n 23def 136.75d-g 1.8pqr 27efg 4.1h-l 15.5b-e 20.5a-d 9.6abc
29617 10.5ab 71a 77.5a 103a 22.5d-g 194.5a 5g-j 31.5cd 6.6cd 18a-e 26.5ab 10.7abc
29624 9.5bcd 60.5c-f 67b-e 96bcd 18.2ijk 141d-g 4.7i-l 26.5fg 6.4cde 19a-e 25.5a-d 9.1abc
29620 11a 60.5c-f 67b-e 95cde 21e-i 103.5lmn 2.75n-q 27efg 2.6m 14de 19bcd 9.4abc
29621 11a 59c-g 64.5c-g 92.5def 19.5g-k 117.75h-m 3.1m-p 32bcd 2.65m 15.5b-e 21a-d 9.8abc
29623 10abc 62.5b-d 68.5b-d 88.5fgh 16.5kl 115.5i-m 2.85n-q 25.5fgh 3.4j-m 16b-e 23.5a-d 10abc
29619 9cde 58c-i 63.5c-h 96bcd 17.5jk 142.5def 2.8n-q 26fg 5.6d-g 17b-e 24.5a-d 8.4c
242451A 10.5ab 53.5f-l 58f-j 89.5fg 26bc 186.3ab 6.1e-h 33bc 5.7d-g 16.5b-e 23.5a-d 9.6abc
T242444 8.5c-f 46mno 52i-l 82jkl 19h-k 73.2o 3.7k-n 20.5ijk 2.5m 16b-e 20.5a-d 9.6abc
Mean 8.89 55.00 60.20 88.50 21.16 132.17 4.67 26.14 4.84 17.66 23.62 9.93
Stv 1.07 6.69 7.20 7.92 3.33 33.36 2.53 6.22 2.03 2.04 2.60 0.74
CV 11.99 12.17 11.95 8.94 15.75 25.24 54.06 23.79 41.96 11.53 11.00 7.42
Range- Min 7.50 44.50 48.50 75.50 14.50 73.05 1.30 13.50 2.50 14.00 17.50 8.40
Max 11.00 71.00 77.50 104.50 30.50 194.50 12.70 39.00 10.25 22.75 28.00 11.50
Checks
SOH701 8def 47l-o 51jkl 78.5k-n 17.5jk 107.5k-n 2.5n-r 19.5jkl 6.4cde 17.5a-e 22a-d 10abc
SOH714 7f 44o 48l 74.5n 17.5jk 106lmn 2.3o-r 24.5gh 3.5i-m 13.5e 17d 8.9bc
Ba-Humera 7.5ef 44o 47.5l 77lmn 18ijk 100.5mn 3.6l-o 26.5fg 5.4d-g 14de 17d 9.6abc
Mean 7.50 45.00 48.83 76.67 17.67 104.67 2.80 23.50 5.10 15.00 18.67 9.50
Stv 0.50 1.73 1.89 2.02 0.29 3.69 0.70 3.61 1.47 2.18 2.89 0.56
Cv 6.67 3.85 3.88 2.64 1.63 3.52 25.00 15.34 28.88 14.53 15.46 5.86
Range(min) 7.00 44.00 47.50 74.50 17.50 100.50 2.30 19.50 3.50 13.50 17.00 8.90
Max 8.00 47.00 51.00 78.50 18.00 107.50 3.60 26.50 6.40 17.50 22.00 10.00
Mean values with similar letter(s) had not significant differences in each row, SD =Standard deviation, CV (%) = Coefficient of variation
in percent and LSD (5%) = Least significant difference at P<0.05.
 99

Appendix Table 2. Mean values of 36 okra genotypes for pod yield and related traits evaluated at Melkassa in year 2018/19.

Genotype PL FL FD AFW NTFPP NR YPP yhaton

29408 1.9h-k 22.3a 26.8ef 58.09c 10.3j 7.2a-f 568.75i-m 15.80i-m
29409 2.15e-i 14.12d-g 30.01bc 45.55efg 10.9j 8.3a 455.78j-n 12.66j-n
29410 2.1e-j 9.87l-o 35.15a 39.66g-j 20.4e-j 6.9a-f 670.23g-j 18.62g-j
29411 2.15e-i 14.75cde 25.55fgh 42.49f-i 9.7j 7.7abc 396.52mn 11.01mn
29412 2f-k 11.02i-n 22.04j 25.27l-p 12.9hij 6.8a-g 345.56mn 9.6mn
29622 2.5c-f 16.01cd 24.18hi 41.65f-j 11.8ij 7.4a-e 501.47i-n 13.93i-n
29414 1.9h-k 10.91j-n 24i 26.15l-o 38.8b 6c-h 1056.79a-d 29.36a-d
29415 3.3ab 18.8b 26.34f 60.32bc 17f-j 7a-f 1016a-e 28.22a-e
29416 2.3d-h 13.25e-i 35.2a 55.9cd 24.4c-i 7.2a-f 869.66c-g 24.16c-g
29418 2.3d-h 10.41k-n 31.1b 38.59hij 28.39b-f 7.1a-f 1173.17a 32.59a
29616 1.9h-k 8.23opq 27ef 35.11jk 34.4bcd 5.8e-h 954.4b-e 26.51b-e
29618 1.9h-k 5.77r 24.1i 17.04q 55.2a 5.7e-h 1183.60a 32.88a
29052 2.15e-i 6.71qr 26.15fg 22.96m-q 30.79b-e 5.1gh 685.82f-i 19.05f-i
T240204 1.9h-k 16.67c 35.75a 65.77ab 18.5e-j 7.4a-e 1082.10abc 30.06abc
29417 2.8c 20.87a 30.34bc 67.23a 10.4j 6.7a-h 632.37h-l 17.57h-l
T242443 1.73ijk 14.45def 29.3cd 43.96e-h 24.3c-i 7.8ab 1156.25ab 32.12ab
240207A 1.6jk 10.63k-n 22.15j 20.56n-q 35.7bc 5.5fgh 898.75c-f 24.97c-f
240209A 2f-k 13.12e-j 31.45b 44.66e-h 17.2f-j 7a-f 803.16e-h 22.31e-h
242433A 1.9h-k 12.89e-j 31.3b 50.21de 20.3e-j 7.6a-d 866.66c-g 24.07c-g
242449A 1.53k 12.15g-k 29.39cd 46.12efg 26.8b-g 6c-h 924.8cde 25.69cde
242445A 1.9h-k 6.9pqr 26fg 20.33opq 27.6b-f 5.8e-h 633.65h-l 17.6h-l
29051 2.45c-g 16.04cd 25.58fgh 50.41de 16.5f-j 6.4b-h 659.68g-k 18.34g-k
29413 2.6cde 9.25mno 25.95fg 30.33lk 14.1hij 6.8a-g 488.96i-n 13.58i-n
29615 1.6jk 5.89r 23.5i 18.87q 38.1b 5.7e-h 853.52d-g 23.71d-g
29625 1.75ijk 11.86h-l 21.6j 26.89l-o 13.9hij 6.3b-h 359.09mn 9.97mn
29617 1.88h-k 12.45f-k 28.95cd 25.36l-p 24.9c-h 6.3b-h 541.91i-n 15.05i-n
 100

Appendix Table 2. Continued

Genotypes PL FL FD AFW NTFPP NR YPP Yhaton

a r cd l-q d-j b-h i-n
29624 3.45 5.5 29.1 23.66 22 6.5 556.1 15.45i-n
29620 2.7cd 6.19qr 28de 21.77m-q 16.4f-j 5.5fgh 339.71n 9.44n
29621 1.93h-k 6.78qr 29.3cd 24.57l-p 21.6e-j 5.8e-h 436.25k-n 12.12k-n
29623 2.9bc 8.9nop 25.8fg 22.84m-q 22.2d-j 5.9d-h 485.84i-n 13.5i-n
29619 2.9bc 5.6r 26.1fg 26.39l-o 17.6f-j 5h 434.61lmn 12.07lmn
242451A 1.9h-k 13.1d-h 34.9a 61.2abc 18.6e-j 7.2a-f 1077.74abc 29.94abc
T242444 1.95g-k 16.87c 28.5d 46.64ef 18.4e-j 6.6a-h 800.46e-h 22.23e-h
Mean 2.18 11.79 27.90 37.77 22.12 6.55 724.52 20.13
Stv 0.48 4.51 3.85 15.12 10.01 0.82 267.24 7.42
Cv(%) 21.99 38.23 13.79 40.03 45.26 12.60 36.89 36.89
Range-
Min 1.53 5.50 21.60 17.04 9.70 5.00 339.71 9.44
Max 3.45 22.30 35.75 67.23 55.20 8.30 1183.60 32.88
h-k i-m b ijk f-j a-f i-n
SOH701 1.8 11.24 31 36.44 16.5 7.1 554.94 15.42i-n
SOH714 1.83h-k 12.53fk 24.8ghi 28.57lm 16f-j 6.1b-h 424.97lmn 11.8lmn
Ba-Humera 1.95g-k 12.9e-j 24.2hi 27.46lmn 19.42e-j 6.2b-h 484.38i-n 13.46i-n
Mean 1.86 12.22 26.67 30.82 17.31 6.47 488.10 13.56
Stv 0.08 0.87 3.76 4.89 1.85 0.55 65.06 1.81
Cv (%) 4.32 7.15 14.12 15.88 10.67 8.52 13.33 13.33
Range-
Min 1.80 11.24 24.20 27.46 16.00 6.10 424.97 11.80
Max 1.95 12.90 31.00 36.44 19.42 7.10 554.94 15.42
Mean values with similar letter(s) had not significant differences in each row, SD =Standard deviation, CV (%) = Coefficient of variation
in percent and LSD (5%) = Least significant difference at P<0.05.
 101

Appendix Table 2. Continued

Genotypes NSPP SDW syppg syph Nha PDM PMC

29408 77.1e-n 6.04b-h 47.72lmn 1325.48lmn 5.5gh 27.94bcd 6.93l
29409 111.9a 6.06b-g 75.61i-n 2100.33i-n 5.5gh 26.02d-i 9.87h-l
29410 63.2k-n 5.28fgh 65.46i-n 1818.35i-n 7.3a-e 24.56f-l 11.02g-j
29411 68.5i-n 5.86c-h 40.83mn 1134.27mn 5.4h 22.12m 8.24jkl
29412 85.5b-k 5.08h 57.05l-n 1584.65l-n 5.5gh 27.2cde 14.58def
29622 98.5a-e 5.44e-h 61.73j-n 1714.86j-n 5.5gh 25.98d-i 9.39i-l
29414 67j-n 5.86c-h 154.45bc 4290.26bc 8a 24.9e-k 13.1e-h
29415 94a-g 6.84ab 109.23c-i 3034.11c-i 6.8b-f 30.2b 12.01f-i
29416 106abc 5.96b-h 154.08bc 4280.08bc 7.5a-d 32.48a 7.17kl
29418 90a-i 5.68d-h 147.17b-e 4088.03b-e 8a 28.48bc 14.16d-g
29616 83d-m 5.44e-h 157.06b 4362.66b 8a 25.3e-j 13.81efg
29618 60.5mn 6.02b-h 198.04a 5501.01a 8a 22.56klm 20.01ab
29052 58.5n 6.12b-g 110c-i 3055.68c-i 8a 25.13e-j 17.2bcd
T240204 106.5ab 6.68abc 131.37b-g 3649.14b-g 7.3a-e 23.97h-m 13.01e-h
29417 62.5lmn 5.2gh 33.48n 929.95n 5.4h 25.51d-j 8jkl
T242443 89b-j 6.68abc 143.88b-f 3996.68b-f 7.6a-d 23.7i-m 20.07ab
240207A 62.1lmn 5.94b-h 125.42b-h 3483.91b-h 7.9ab 26.25c-h 14.57def
240209A 95a-g 6.1b-g 98.01f-k 2722.41f-k 6.7c-f 23.74i-m 12.13f-i
242433A 83.7c-l 6b-h 102e-k 2833.46e-k 6.9a-f 23.95h-m 13.13e-h
242449A 80.7d-n 7.16a 150.53bcd 4181.3bcd 7.6a-d 27.88bcd 19.93ab
242445A 65k-n 5.82c-h 101.55e-k 2820.96e-k 7.5a-d 26.53c-g 18.06bc
29051 101.2a-d 5.88b-h 99.82f-k 2772.66f-k 6.7c-f 26.5c-g 21.41a
29413 83d-m 6.1b-g 71.25i-n 1979.12i-n 6.2e-h 28.55bc 11.47f-i
29615 72.6g-n 5.52d-h 152.56bc 4237.75bc 8a 27.2cde 13.7efg
29625 78e-n 5.44e-h 58.61j-n 1628.08j-n 5.9fgh 26.93c-f 21.8a
29617 65.2k-n 5.24gh 84.79g-m 2355.25g-m 7.7abc 22.04m 16.01cde
29624 63.5k-n 6.24a-f 87.46g-m 2429.33g-m 7.4a-d 34.55a 7.85jkl
29620 73.5f-n 6.4a-e 76.99i-n 2138.53i-n 6.2e-h 26.34c-h 10.29h-k
29621 65k-n 5.6d-h 78.71h-n 2186.4h-n 7.1a-e 22.22lm 17.02bcd
29623 70.5h-n 5.94b-h 93.7g-l 2602.68g-l 7.3a-e 27.11cde 13.02e-h
29619 64.1k-n 5.64d-h 64.12i-n 1781.2i-n 6.7c-f 27.92bcd 23.01a
242451A 95.5a-f 5.98b-h 105.87d-j 2940.78d-j 7.2a-e 26.68c-g 8.1jkl
T242444 86b-k 6.36a-e 100.38f-k 2788.28f-k 6.6c-g 24.25g-m 18.25bc
Mean 79.58 5.93 101.18 2810.53 6.94 26.20 13.89
Stv 15.38 0.48 39.93 1109.28 0.90 2.77 4.61
Cv (%) 19.32 8.13 39.47 39.47 12.98 10.55 33.20
Range (min) 58.50 5.08 33.48 929.95 5.40 22.04 6.93
Max 111.90 7.16 198.04 5501.01 8.00 34.55 23.01
 102

Appendix Table 2. Continued

Genotypes NSPP SDW syppg syph Nha PDM PMC

e-n a-d h-m h-m c-f d-j
SOH701 76 6.42 81.41 2261.39 6.7 25.6 8jkl
SOH714 69.3i-n 5.91b-h 66.47i-n 1846.45i-n 6.5d-h 25.26e-j 10.12h-l
Ba- Humera 92.5a-h 5.72d-h 102.27e-k 2840.91e-k 7.2a-e 23.27j-m 14.23d-g
Mean 79.27 6.02 83.38 2316.25 6.80 24.71 10.78
Stv 11.94 0.36 17.98 499.50 0.36 1.26 3.17
Cv (%) 15.06 6.05 21.56 21.56 5.30 5.09 29.37
Range (min) 69.30 5.72 66.47 1846.45 6.50 23.27 8.00
Max 92.50 6.42 102.27 2840.91 7.20 25.60 14.23
Mean values with similar letter(s) had not significant differences in each row, SD =Standard
deviation, CV (%) = Coefficient of variation in percent and LSD (5%) = Least significant difference
at P<0.05.
 103

Appendix Table 3. The 13 Qualitative Traits of 36 Okra Genotypes Evaluated at Melkassa in

Year 2018/19.

No ACCESSION PH FCR LCR LPCR IFCR STC SL SS SF PFMS FP LP NFB

1 29408 1 1 1 2 2 1 3 2 1 1 2 1 1
2 29409 2 2 2 3 3 2 7 1 1 1 2 2 1
3 29410 2 2 2 3 2 2 3 1 1 1 2 1 1
4 29411 2 2 2 3 3 2 4 2 4 1 2 2 1
5 29412 1 2 2 2 2 2 4 1 4 1 2 2 1
6 29622 1 2 1 2 1 1 2 2 1 2 1 2 1
7 29414 2 2 1 2 2 2 7 3 4 1 2 2 1
8 29415 2 2 1 2 2 2 4 2 1 1 2 2 1
9 29416 3 2 1 2 2 2 2 2 2 2 2 2 3
10 29418 3 2 1 2 2 1 2 1 2 2 2 2 3
11 29616 2 2 2 3 3 3 2 2 2 1 2 3 3
12 29618 2 2 2 3 3 3 2 3 3 2 2 2 2
13 29052 3 2 2 3 3 2 7 1 1 1 2 3 1
14 T240204 2 2 2 3 3 3 3 2 1 1 2 1 1
15 29417 2 2 2 2 2 2 3 1 4 2 2 2 1
16 T242443 2 2 2 2 3 2 3 1 1 1 2 2 1
17 240207A 2 2 1 2 2 2 5 1 4 1 2 2 1
18 240209A 3 2 1 2 2 2 3 2 4 1 2 2 1
19 242433A 3 2 2 2 3 2 7 3 2 1 2 2 3
20 242449A 2 2 2 2 3 3 3 1 1 1 2 2 1
21 242445A 3 2 2 3 3 2 7 3 3 1 2 2 2
22 29051 2 2 1 2 3 2 7 1 4 2 2 2 1
23 29413 2 2 2 2 3 2 2 1 4 2 2 2 1
24 29615 3 2 2 3 3 3 2 1 3 1 2 2 2
25 29625 3 2 2 2 3 2 7 1 1 1 2 2 1
26 29617 3 2 2 2 2 2 3 2 1 1 2 2 1
27 29624 3 2 2 2 2 2 3 2 3 3 2 2 2
28 29620 3 2 2 2 3 2 2 1 3 2 2 2 2
29 29621 3 2 2 2 3 3 7 2 3 1 2 2 2
30 29623 3 2 2 2 3 2 2 2 4 2 2 2 1
31 29619 3 2 2 2 3 3 2 1 2 2 2 2 3
32 242451A 2 2 1 2 3 2 2 2 1 1 2 2 1
33 T242444 2 2 1 2 3 2 3 2 1 1 2 2 1
34 SOH701 2 2 1 2 2 2 7 1 4 1 2 2 1
35 SOH714 1 2 1 2 2 2 7 2 1 1 2 2 1
36 B/Humera 2 2 1 2 2 2 5 2 1 1 2 2 1
PH= plant habit, FCL= Flower color, LCR= Leaf color, LPCR= Leaf petiole color, IFCR= Immature fruit
color, STC= Stem color, SL= Shape of leaf, SS= shape of seed, SF= Shape of fruit, PFMS= Position of fruit
on main stem, FP= Fruit pubescence, LP= Leaf pubescence and NFB= Nature of fruit base.
 104

Appendix Table 4 . Description and score of qualitative traits of okra

No Traits Scores description No Traits Score Description
1 Plant habit 1 Densely branched at 8 Immature 1 Yellow-Green
the apex (DBA) fruit color
2 Densely Branched 2 Green
Base (DBB)
3 Densely Branched 3 Green with red
Overall (DBO) patches
2 Flower color 1 red color inside 4 Red
only
2 red color on both 5 Dark- red
sides
3 Leaf color 1 Totally green 6 other specify
2 Green with the red 9 Position of 1 Erect
vein fruits on the
main stem
4 Leaf Petiole 1 Green 2 Intermediate
color
2 Red above but 3 Horizontal
green below
3 Red on both sides 4 Slightly Falling
5 Shape of leaf 1 Oval undulate 5 Totally Falling
2 Heart-shaped 10 Fruit 1 Smooth
Pubescence
3 Broadly ovate 2 Rough
4 star-shaped 11 Nature of 1 Ringed or
(palmately lobed) fruit base Protrude
5 palmately 2 Ringless or Flat
triangular lobes
6 palmately lobed 3 Sunken
with dentate
margins
7 palmately lobed 12 Shape of 1 Round
with serrated Seed
margins
8 Linear-oblong or 2 Spherical
Triangular lobes (Oval)
6 Leaf 1 Glabrous 3 Kidney
pubescence (Reniform)
2 Slight
3 Conspicuous
7 Stem Color 1 Green
2 Green with red
patches
3 Red or Purple
 105

Appendix Figure 1. Fruit Shape Score (IBPGR, 1991).

 106

Appendix Table 5. Euclidean distances of 36 okra genotypes based on 24 quantitative traits evaluated at Melkassa in 2018/19

29409 29410 29411 29412 29622 29414 29415 29416 29418 29616 29618 29052 T240204 29417 T242443 240207A 240209A 242433A
29408 5.465 6.589 4.690 5.350 4.863 8.851 7.889 8.304 9.085 8.664 12.035 9.222 7.391 5.112 7.823 7.658 6.102 5.524
29409 6.935 4.115 4.962 4.389 8.672 6.691 8.748 8.836 8.794 11.915 9.619 6.693 5.564 7.625 8.592 4.409 5.441
29410 6.107 6.816 7.722 8.049 7.859 6.908 7.259 6.820 10.141 6.320 7.092 6.671 8.139 7.004 5.621 4.893
29411 4.170 4.290 8.351 7.609 9.781 9.361 8.609 11.477 8.570 8.064 4.133 7.959 7.904 4.795 5.409
29412 4.130 7.700 8.556 9.640 9.034 7.788 10.826 8.185 8.764 5.826 8.281 6.814 5.653 6.056
29622 8.109 7.590 9.110 8.339 8.007 11.293 9.175 8.736 4.853 7.418 8.186 5.494 5.664
29414 7.732 8.293 5.763 3.852 4.349 5.769 7.209 8.928 5.720 3.831 5.889 5.541
29415 7.276 6.619 7.698 10.140 8.074 6.808 7.003 7.314 8.656 5.485 6.189
29416 4.972 6.217 9.685 7.245 6.999 9.564 8.014 8.477 7.280 5.988
29418 3.591 6.680 5.887 7.591 9.149 5.970 7.420 6.208 5.141
29616 4.897 4.448 7.521 8.995 5.624 5.258 5.618 4.793
29618 6.432 9.658 12.242 7.391 6.300 8.805 8.145
29052 8.786 9.299 7.894 5.951 6.782 5.961
T240204 8.490 6.180 7.171 5.219 5.055
29417 8.830 9.070 5.612 6.131
T242443 6.635 4.910 4.351
240207A 6.650 6.053
240209A 2.870
242433A
 107

Appendix Table 5. Continued

Bamia
242449A 242445A 29051 29413 29615 29625 29617 29624 29620 29621 29623 29619 242451A T242444 SOH701 SOH714 Humera
29408 8.375 7.073 5.275 4.811 9.559 5.353 9.210 8.506 7.507 7.817 7.439 8.186 6.945 5.260 5.007 4.656 5.893
29409 8.483 7.482 6.084 5.096 9.729 6.237 8.947 7.964 7.798 8.164 7.344 7.824 7.181 5.917 4.032 5.742 5.476
29410 7.276 5.658 6.478 6.367 8.112 6.484 6.059 7.365 7.360 6.515 7.130 6.944 5.467 7.292 6.090 6.838 6.780
29411 8.763 6.637 6.295 4.596 9.298 5.280 7.862 7.861 6.636 6.679 6.581 6.925 7.577 5.748 3.920 4.555 5.193
29412 8.627 5.518 5.191 3.822 8.262 2.833 7.912 7.448 6.315 6.028 5.710 5.540 8.171 5.758 4.840 3.713 4.100
29622 9.475 6.460 5.555 3.650 8.324 5.489 8.992 7.755 6.707 7.267 6.134 6.986 8.270 4.797 4.764 4.629 4.934
29414 5.943 4.938 6.706 7.051 4.799 7.896 7.501 7.500 7.745 6.609 6.017 7.279 7.030 5.948 6.521 6.834 5.764
29415 6.842 7.871 6.821 7.041 8.866 9.018 8.484 6.433 8.223 9.244 7.196 7.889 6.432 7.233 7.128 8.801 8.133
29416 6.623 7.376 6.906 7.811 7.411 9.485 8.146 7.383 8.278 8.598 8.017 9.144 5.022 8.199 8.454 9.520 8.842
29418 6.382 5.758 6.745 7.296 4.595 9.111 6.618 6.816 7.558 7.175 6.252 7.741 5.897 7.035 7.948 9.011 8.112
29616 5.495 4.181 5.934 6.638 3.129 7.729 5.973 7.153 6.581 5.474 5.390 6.842 5.901 6.155 7.159 7.567 6.412
29618 7.366 6.928 9.130 9.903 5.389 10.516 8.795 9.779 9.633 8.325 8.351 9.375 9.255 8.716 9.907 10.163 8.920
29052 5.399 4.064 6.862 7.119 4.951 7.575 4.543 6.370 5.720 4.906 5.361 5.390 6.896 7.619 7.886 8.020 7.442
T240204 5.692 8.049 6.119 7.983 9.611 8.648 9.238 9.386 9.606 8.918 8.889 9.579 4.471 6.306 6.371 7.969 6.885
29417 9.508 7.691 6.781 5.824 9.903 6.956 8.327 7.756 7.533 7.705 6.984 7.066 7.586 6.429 5.629 6.226 6.912
T242443 5.759 6.193 5.812 6.614 6.921 8.011 8.954 9.059 8.226 7.518 7.417 8.762 6.371 3.842 6.119 7.490 6.392
240207A 6.046 4.270 5.925 6.510 5.927 6.039 8.233 8.062 7.965 6.818 6.772 7.185 7.377 5.896 6.147 5.367 4.670
240209A 5.525 5.468 4.930 4.825 7.219 6.418 6.459 6.815 6.022 5.674 5.480 6.480 4.443 4.268 3.738 5.862 4.941
242433A 5.165 4.758 4.485 4.858 6.480 6.177 6.195 6.985 6.217 5.430 5.677 6.694 3.735 3.974 4.454 5.839 5.282
 108

Appendix Table 5. Continued

242445A 29051 29413 29615 29625 29617 29624 29620 29621 29623 29619 242451A T242444 SOH701 SOH714 Bamia Humera

242449A
5.908 5.944 7.431 7.085 7.950 7.112 7.695 7.345 6.817 6.990 7.527 5.036 6.414 7.071 8.239 7.358
242445A
4.841 4.323 3.990 4.800 5.933 5.991 4.979 3.998 4.035 4.258 7.061 4.975 5.278 4.920 4.557
29051
4.319 7.215 4.425 7.502 7.454 6.264 5.778 5.485 5.800 6.062 3.808 5.626 5.300 4.233
29413
6.971 4.387 7.884 5.460 4.262 5.394 4.059 5.102 7.126 4.280 3.719 3.719 4.286
29615
8.120 6.733 7.248 6.722 5.955 5.569 6.920 8.164 7.068 7.810 7.862 7.075
29625
8.021 8.207 6.752 6.036 6.334 5.453 8.369 5.465 5.550 3.816 4.133
29617
7.371 6.370 4.723 5.728 6.104 7.007 8.521 8.238 8.753 8.167
29624
5.135 6.836 4.242 5.262 7.694 8.129 6.814 7.736 7.930
29620
3.713 3.051 4.630 7.713 6.577 6.222 6.456 6.750
29621
3.785 4.553 7.229 6.112 6.280 6.177 5.956
29623
3.723 7.532 5.954 5.757 5.863 5.778
29619
8.490 6.833 6.710 6.334 6.288
242451A
6.664 6.651 8.225 7.550
T242444
4.126 4.509 3.998
SOH701
3.362 3.849
SOH714
2.882