GENOTYPE X ENVIRONMENT INTERACTION AND YIELD

STABILITY OF EARLY MATURING SORGHUM [Sorghum bicolor (L.)

Moench] GENOTYPES IN DRY LOWLAND AREAS OF ETHIOPIA

M.Sc. THESIS

ZIGALE SEMAHEGN YITAYEH

NOVEMBER, 2018
JIMMA UNIVERSITY

1
Genotype x Environment Interaction and Yield Stability of Early Maturing
Sorghum [Sorghum bicolor (L.) Moench] Genotypes in Dry Lowland Areas
of Ethiopia

A Thesis Submitted to the School of Graduate Studies

Jimma University College of Agriculture and Veterinary Medicine

In Partial Fulfillment of the Requirements for the Degree of

MASTER OF SCIENCE IN AGRICULTURE (PLANT BREEDING)

By

Zigale Semahegn Yitayeh

Major Advisor: Prof. Kassahun Bantte

Co- Advisor: Taye Tadesse (PhD.)

November, 2018
Jimma, Ethiopia

I
 DEDICATION

I dedicate this thesis manuscript to my father Ato Semahegn Yitayeh and My mother W/ro
Emiye Ayalew, who were committed to grow me up in the process of education..

II
 STATEMENT OF THE AUTHOR

First, I declare that this thesis is my work and that all sources of materials used for this thesis
have been duly acknowledged. This thesis has been submitted in partial fulfillment of the
requirements for M.Sc. degree at the Jimma University and is deposited at the University
Library to be made available to borrowers under rules of the Library. I solemnly declare that
this thesis is not submitted to any other institution anywhere for the award of any academic
degree, diploma, or certificate. Brief quotations from this thesis are allowable without special
permission provided that accurate acknowledgement of source is made. Requests for
permission extended quotation from or 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 interests of
scholarship. In all other instances, however, permission must be obtained from the author.

Name: Zigale Semahegn Signature: ……………………

Place: Jimma University, Jimma
Date of Submission: …………

III
 LIST OF ABBREVIATIONS AND ACRONYMS

AMMI Additive Main Effects and Multiplicative Interaction

ANOVA Analysis of Variance
ASV AMMI Stability Value
CV Coefficient of Variation
DAP Diammonium Phosphate
DF Degrees of Freedom
EIAR Ethiopian Institute of Agricultural Research
ESIP Ethiopian Sorghum Improvement Program
GxE Genotype x Environment
GGE Genotype and Genotype by Environment
IBPGR Institute for Biodiversity of Plant Genetic Resources
ICRISAT Institute of Crop Research Institute for the Semi Arid Tropics
IDRC International Development Research Centre
IPCA Interaction Principal Component Axis
JLR Joint Genotypear Regression
LMM Linear Mixed Model
MARC Melkassa Agricultural Research Center
m.a.s.l. Meter above sea level
MET Multi-environment Trials
MS Mean Square
PCA Principal Component Analysis
PGRC/E Plant Genetic Resources Centre/Ethiopia
RCBD Randomized Complete Block Design
REML Residual Maximum Likelihood
SV Source of Variation

IV
 BIOGRAPHICAL SKETCH

The author was born on June 14, 1989 in Bure, West Gojjam of Ethiopia. He attended
elementary and junior Secondary school at Bure Ras Bit Woded Mengesha Atikem School
from 1997 to 2004. He pursued high school and preparatory School at Bure Shikudad from
2005 to 2008. After the completion of his high school education, he joined the then Hawassa
University in 2009 and graduated with a Bachelor of Science degree (B.Sc.) in Horticulture
on July, 2011. Soon after graduation, he was employed by Burea of Agriculture as
Horticulture expert in Womberma woreda Agricultural office from October 2011 to April
2014 and from May 2014 to September 2016 as assistance researcher I in Ethiopian Institute
of Agricultural Research. In the year 2016, he joined the School of Graduate Studies at
Jimma University to pursue his graduate study in Plant Breeding.

V
 ACKNOWLEDGEMENTS

First and foremost I bless the almighty God for keeping me healthy and strong enough to
successfully undertake both the course and field research works as well as compiling this
manuscript. I would like to express my deep gratitude to my major advisor Prof. Kassahun
Banttee and my co-advisor Dr. Taye Tadesse for their guidance, support and encouragement
throughout the entire period of this thesis work. I would also like to thank to all staff members
of Sheraro Agricultural Research Centre, Kobo Agricultural Research Sub Centre, Shoarobit
Agricultural Research Sub Centre, Mieso Agricultural Research Sub Center and Fedis
Agriculture Research Centre for their help in planting, collecting data while I was on course
work. Many people have contributed to the completion of this thesis research and it would
probably take a whole chapter to name them all. Thank you all!!!

I am highly indebted to the Ethiopian Institute of Agricultural Research for giving me the
chance to pursue my study. I am also highly indebted to my beloved family and extend my
gratitude to my father Ato Semahegn Yitayeh, My mother W/ro Emiye Ayalew and My Wife
W/ro Wude Amsalu for their moral.

VI
 TABLE OF CONTENTS

CONTENTS PAGES

DEDICATION.................................................................................................... II
STATEMENT OF THE AUTHOR................................................................. III
LIST OF ABBREVIATIONS AND ACRONYMS ........................................ IV
BIOGRAPHICAL SKETCH ............................................................................. V
ACKNOWLEDGEMENTS ............................................................................. VI
TABLE OF CONTENTS ................................................................................VII
LIST OF TABLES ............................................................................................ IX
LIST OF FIGURES ............................................................................................X
LIST OF TABLES IN APPENDICES ............................................................ XI
LIST OF FIGURES IN APPENDICES .........................................................XII
ABSTRACT .................................................................................................... XIII
1. INTRODUCTION ...........................................................................................1
2. LITERATURE REVIEW ...............................................................................5
2.1. Sorghum Research in Ethiopia ........................................................................................ 5
2.2. Genotype by Environment Interaction ............................................................................ 5
2.3. Causes of Genotype X Environment Interaction ............................................................. 9
2.4. Concept of Stability ......................................................................................................... 9
2.5. Methods of Statistical Analysis of G x E Interaction .................................................... 11
2.5.1. Conventional analysis of variance .......................................................................... 11

2.5.2. Spatial and linear mixed model .............................................................................. 12

2.5.3. Estimation of stability parameters .......................................................................... 14

2.5.3.1. Regression coefficient (bi) and deviation mean square (S2di) ..................................... 14

2.5.3.2. Ecovalence (Wi) ............................................................................................................ 15

2.5.4. Multivariate analysis methods ................................................................................ 16

2.5.4.1. Additive main effects and multiplicative interaction (AMMI) ...................................... 16

VII
 2.5.4.2. GGE Biplot Model ......................................................................................................... 17

3. MATERIALS AND METHODS ..................................................................18

3.1 Description of the Study Area ........................................................................................ 18
3.2. Genetic Materials .......................................................................................................... 18
3.3. Experimental Design ..................................................................................................... 22
3.4. Trial Management ......................................................................................................... 22
3.5. Data Collection and Sampling Techniques ................................................................... 22
3.6. Data Analysis ................................................................................................................ 23
3.6.1. Analysis of data for each location and combined over location ............................. 23

3.7. Stability Analysis .......................................................................................................... 24

3.7.1. Eberhart and Russell’s joint regression model ....................................................... 24

3.7.2. The additive main effects and multiplicative interaction (AMMI) method............ 25

3.7.3. GGE Biplot Method ................................................................................................ 25

3.7.4. AMMI’s stability value (ASV) ............................................................................... 26

4. RESULTS AND DISCUSSION ....................................................................27

4.1. Data analysis of variance ............................................................................................... 27
4.1.1 Single Location data analysis .................................................................................. 27

4.1.2. Combined analysis of data ...................................................................................... 30

4.2. General and relative performance of the genotype........................................................ 38

4.3. Stability analysis ............................................................................................................ 46
4.3.1 Eberhart and Russel's Regression Model ................................................................. 46

4.3.2. AMMI Model .......................................................................................................... 50

4.3.4. AMMI stability value.............................................................................................. 57

5. SUMMERY AND CONCLUSION ..............................................................62

REFERENCES ...................................................................................................67
APPENDIX .........................................................................................................76

VIII
 LIST OF TABLES

Tables Pages

Table 1: Agro-ecological features of the experimental locations. ........................................... 18

Table 2: Description of sorghum genotypes tested at five locations during 2017 main cropping
season ........................................................................................................................ 19
Table 3: Variance component estimation by REML and spatial model of early maturing
sorghum genotypes for grain yield, phenological and other major traits tested at Erer,
Kobo, Mieso, Sheraro and Shewa Robit during 2017 main cropping season. .......... 29
Table 4: Variance component estimation by REML of early maturing sorghum genotypes for
grain yield, phenological and other major traits tested at five locations during 2017
main cropping season. ............................................................................................... 31
Table 5: Means for grain yield, phenolgical and other major traits of sixty early maturing
sorghum genotypes at Erer, Kobo, Mieso, Sheraro and Shewarobit during 2017 main
cropping season. ........................................................................................................ 40
Table 6: Genotype by environment interaction for mean grain yield of 60 early maturing
sorghum genotypes ....................................................Error! Bookmark not defined.
Table 7: Means for grain yield, phenological and other major traits of the top 15 and the
bottom 15 early maturing advanced sorghum genotypes tested across locations
during 2017 main cropping season. .......................................................................... 45
Table 8: Genotype x Environment Interaction analysis of variance by Eberhart and Russel's
Model of early maturing sorghum genotypes on mean grain yield (kg/ha) tested at
five sites. ................................................................................................................... 47
Table 9: Eberhart and Russell’s (1966) stability parameters of early maturing sorghum
genotypes tested at five locations.............................................................................. 49
Table 10: Genotype x Environment Interaction analysis of variance by AMMI for grain yield
(kg/ha) of early maturing advanced sorghum genotypes tested at five locations
during 2017 main cropping season. .......................................................................... 52
Table 11: IPCA1 and IPCA 2 scores; and ASV for the sixty early maturing sorghum
genotypes sorted on ASV evaluated at five locations during 2017 main cropping
season. ....................................................................................................................... 58

IX
 LIST OF FIGURES

Figures Pages

Figure 1: Heat map of spatial trend within a field of sixty early maturing sorghum genotypes
on plot grain yield performance tested at Erer during 2017 cropping season. .......... 27
Figure 2: AMMI 1 biplot for grain yield of 60 early maturing sorghum genotypes evaluated
across five locations in Ethiopia during 2017 main cropping season. .................... 54
Figure 3: AMMI 2 biplot for grain yield (kg/ha) showing the interaction of IPCA2 against
IPCA1 scores of 60 early maturing sorghum genotypes in five environments......... 55
Figure 4: The ‘which-won-where’ feature of the biplot. ......................................................... 59
Figure 5: The relationship among testers and mega environments. ......................................... 60
Figure 6: GGE-biplot showing a comparison of five testing environments with in ideal
environment for grain yield (kg/ha). ......................................................................... 61
Figure 7: GGE-biplot showing a comparison of all genotypes with in ideal genotypes for grain
yield (kg/ha) .............................................................................................................. 62

X
 LIST OF TABLES IN APPENDICES

Tables Pages

Appendix Table 1: Variance component estimation by REML of early maturing sorghum

genotypes for grain yield, phenological and other traits tested at Erer, Kobo, Mieso,
Sheraro and Shewa Robit during 2017 main cropping season.................................. 76
Appendix Table 2: Variance component estimation by REM of early maturing sorghum
genotypes for grain yield, phonological and other complex traits tested at five
locations during 2017 main cropping season. ........................................................... 77
Appendix Table 3: Predicted mean grain yield (kg/ha) of sixty early maturing sorghum
genotypes by spatial model tested at Erer, Kobo, Mieso, Sheraro and Shewa Robit
during 2017 main cropping season. .......................................................................... 78
Appendix Table 4: Means for grain yield, phenological and other complex traits of sixty early
maturing advanced sorghum genotypes by mixed model tested at Erer during 2017
main cropping season. ............................................................................................... 80
Appendix Table 5: Means for grain yield, phonological and other complex traits of sixty
early maturing advanced sorghum genotypes by mixed model tested at Kobo during
2017 main cropping season. ...................................................................................... 83
Appendix Table 6: Means for grain yield, phonological and other complex traits of sixty
early maturing advanced sorghum genotypes by mixed model tested at Mieso during
2017 main cropping season. ...................................................................................... 86
Appendix Table 7: Means for grain yield, phonological and other complex traits of sixty early
maturing advanced sorghum genotypes by mixed model tested at Sheraro during
2017 main cropping season. ...................................................................................... 89
Appendix Table 8: Means for grain yield, phonological and other complex traits of sixty early
maturing advanced sorghum genotypes by mixed model tested at Shaorobit during
2017 main cropping season. ...................................................................................... 92
Appendix Table 9: Means for grain yield, phenolgical and other complex traits of sixty early
maturing advanced sorghum genotypes at Erer, Kobo, Mieso, Sheraro and Shewa
Robit during 2017 main cropping season.................................................................. 95
Appendix Table 10: Means for grain yield, phonological and other traits of sixty early
maturing advanced sorghum genotypes tested at five sites during 2017 main
cropping season. ........................................................................................................ 96
Appendix Table 11: The IPCA 1 and IPCA 2 scores for the five sites, sorted on environmental
mean yield, used in the study. ................................................................................... 99

Appendix Table 12: Total monthly rainfall (mm) and mean monthly temperature (°C) of the
four tested locations during 2014 main cropping season. ........................................................ 99

XI
 LIST OF FIGURES IN APPENDICES

Figure Page

Appendix Figure 1: Mean grain yield graphical display of genotype by environment

interaction................................................................................................................ 100
Appendix Figure 2: Graphical display of Correlation between locations with heat map (MET
analysis) on grain yield performance of early maturing sorghum genotypes. ........ 101
Appendis Figure 3: Spatial trend and spatially independent residuals from the spatial model
for grain yield (kg/ha) in each location plotted against row and column positions. 101

XII
 ABSTRACT

The yield performance of crop varieties is highly affected by genotype x environment

interaction, which is the major concern to plant breeders while developing improved
varieties. In Ethiopia, high yielding, early maturing and stable varieties that wisthstand
drought in the dry lowland areas are limited. In view of this, the yield performance of 60
early maturing advanced sorghum genotypes were tested at five locations with the objectives
of estimating the magnitude of GxE for grain yield and other traits and to identify high yielder
and stable genotypes across locations. The experiment was conducted using Randomized
Complete Block Design with Row Column arrangement and three replications. Grain yield,
phenological and other traits were recorded. The Residual Maximum Likelihood combined
analysis of variance across locations showed very highly significant (P<0.0001) difference
for grain yield among locations and significant (P<0.05) among genotypes and highly
significant (P<0.0001) among interactions for most of the traits studied. Significant variation
for locations and genotype indicates variation in the performance of genotypes for grain
yield, phonological and other traits in different environments. On the other hand, significant
GxE interaction showed inconsistency in the performance of sorghum genotypes across
locations. Based on Eberhart and Russell’stability analysis, genotypes 05MW6026,
14MWLSDT7207, 14MWLSDT7332, 12MW6444, 12MW6302 and 14MWLSDT7042 were the
most stable and high yielders. The AMMI stability value showed that genotypes
14MWLSDT7207, 05MW6026 and 14MWLSDT7234 were the high yielders and stable, while
the GGE stability value showed that genotypes 14MWLSDT7207, 05MW6026,
14MWLSDT7209 and 12MW6440 were the best yielders and stable ones. Results of ASV
parameter also showed six most widely stable and high yielder genotypes i.e.,
13MWF6#6037, 05MW6026, 14MWLSDT7207, 14MWLSDT7234,14MWLSDT7042 and
12MW6440. Genotypes 05MW6026 and 14MWLSDT7207, were selected with four stability
parameters as a high yielder and stable, and therefore, are the promising ones. Generally,
this study showed the importance of testing early maturing sorghum genotypes for their yield
and stability across diverse dry lowland areas of Ethiopia.

Keywords: Early maturing, Sorghum, AMMI, GGE, Stability, Joint regression

XIII
 1. INTRODUCTION

Sorghum [Sorghum bicolor (L.) Moench] is a C4 tropical crop (Smith and Frederiksen, 2000).
It is a monocot crop, which belongs to the family Gramnieae (Harlan and De Wet, 1972).
Sorghum is naturally self-pollinated short day plant with the degree of spontaneous cross
pollination, in some cases, reaching up to 30%, depending on panicle types. It has 2n = 2x =
20 chromosomes and it has relatively small size of 730 Mbp (Paterson et al., 2009). The
genus sorghum is very diverse and all cultivated sorghums belong to Sorghum bicolor
species, which is divided, based on morphology, into five races: bicolor, caudatum, guinea,
durra, and kafir (Harlan and De Wet, 1972). Sorghum is an indigenous crop to Ethiopia where
it is grown in a wider area of adaptation ranging from hot, dry lowland, intermediate to the
highland environments.

The world sorghum production is estimated to be 62.3 million tons from 42 million hectares
of land (USDA, 2017). In Africa, sorghum production is 29.14 million tons from 26.03
hectares of land (USDA, 2017). In Ethiopia, it ranks third in area coverage, after maize and
teff (CSA, 2017). In Ethiopia, sorghum is contributing 16.4% of the total annual cereal grain
production (CSA, 2017). In Ethiopia, the area covered with sorghum is 1.9 million ha, and its
total production is 4.8 million tons of grain (CSA, 2017). It is the dominant crop in the dry
lowlands, which accounts for 66% of the total cultivated areas of the country and the national
average productivity of sorghum in Ethiopia is 2.5 tons/ha (CSA, 2017). However, research
has shown that there is a potential to increase sorghum productivity from 3 to 6 tones/ha using
improved varieties and production practices (Asfaw et al., 2005).

In the arid and semi-arid tropics of Africa and Asia, sorghum is primarily grown as a food
grain crop while in the developed world the majority of the grain produced is used for
animal feed (Rakshit et al., 2014). Sorghum grain is preferred next to tef, a small cereal grain
crop, used as injera (Fantaye and Hintsa, 2017). Although there is variability in the grain
quality depending on the end use product, larger seed size, white and light red types of
sorghum grains are predominantly preferred for the preparation of injera. The grain is also
used for the preparation of local beverages. In addition, the stover is equally valued as the
grain, which is used for animal feed, fuel wood and construction purposes (Taye et al., 2016).
1
In Ethiopia, the current rate of yield increment in sorghum is not adequate. Although a
number of biotic (diseases, insects, striga etc.) and abiotic factors (drought, poor soil fertility,
soil salinity etc. ) contributed to the lower grain yield increment, drought is considered as one
of the major abiotic factors impeding sorghum productivity in Ethiopia. Drought is occurring
due to delay in onset, dry spell after sowing, drought during critical crop stage ( flowering and
grain filling stage) and too early cessation of rain. This limitation is reduced by developing
sorghum genotypes which are more adapted to moisture stress areas (Fantaye and Hintsa,
2017). Over eighty percent of sorghum in Ethiopia is produced under high to moderate
drought stress condition (Fantaye and Hintsa, 2017). In Ethiopia, drought occurred in any
stage of sorghum development with the high frequency of terminal stress in dry lowland
areas. Genetic improvement is considered as an integral part of overcoming the challenges of
drought in the dry lowlands of the country. A major challenges of sorghum production in the
dry lowland parts of the country is lack of early maturing improved varieties that can escape
drought or tolerant to stresses, such as drought and lack of stable varieties across
environments (Fantaye and Hintsa, 2017). Development of an early maturing and/or varieties
that can withstand terminal stress through introgression of stay-green trait are the two major
strategies being implanted in the sorghum breeding in Ethiopia.

In countries such as Ethiopia, where environmental differences are very diverse and
unpredictable (Dugan and Labuschahne, 2002) and led to significant G x E interaction even
within a small geographic area, making genotype improvement and recommendation is more
problematic. Under such conditions, testing of genotypic performance at various locations
delivers valuable information to determine their adaptation and stability (Crossa, 1990). As a
result, it is not only average genotype performance that is important in selection of superior
genotypes, but also, the magnitude of the interactions (Gauch and Zobel, 1997). As a result,
several methods of measuring and describing genotypic response across environments have
been developed and utilized (Luthra et al, 1974).

Various studies have been conducted to analyze the effect of GxE interaction in sorghum
varieties. Asfaw (2007) studied yield stability in sorghum and tested 14 hybrids and one open
pollinated variety and he recommended four genotypes for the drought stressed sorghum
growing areas of Ethiopia. Abiy and Firew (2016) studied genotype x environment interaction

2
and stability of early maturing sorghum genotypes in Ethiopia and identified three genotypes
suitable for lowland areas of Ethiopia. Kinde et al. (2016) evaluated sorghum varieties and
environments for yield performance and stability and recommended two varieties for moisture
stress area of West Hararghe. Sintayehu and Kassahun (2017) studied genotype-by-
environment interaction and yield stability in sorghum genotypes and identified three
genotypes for the North Shewa sorghum growing areas. Similarly, Fantaye and Hintsa (2017)
studied 12 sorghum hybrids in the moisture stress conditions of Abergelle District, Northern
Ethiopia and identified three hybrid genotypes for moisture stressed areas of Abergelle. These
authors reported the existence of high interaction of genotypes with environments. However,
the previous authors conducted their trials with previously released and very limited
genotypes.

Yield stability is one of the impediments facing plant breeders in developing widely adapted
varieties with superior yield (Asfaw, 2007). Several research institutions have been working
and able to identify sorghum varieties for yield that resist or tolerate harsh environmental
conditions and produce consistently better grain yield. In line with this, the national and
regional sorghum improvement programs have released a number of stable and early maturing
sorghum genotypes for the moisture stress lowland areas of Ethiopia (EIAR, 2014).
Information on nature and extent of genotype by environment interaction of sorghum
genotypes is important to identify superior and stable variety in drought areas of Ethiopia.
However, there is no information on genotype by environment interaction of early maturing
sorghum genotypes developed by the natonal program. Therefore, this study was conducted
with the following objectives

3
General Objectives
 To assess the performance of early maturing sorghum genotypes for yield and stability
across locations.

Specific Objectives
 To estimate the magnitude of GxE interactions and performance of early maturing
sorghum genotypes for yield and yield related traits.
 To identify high yielder and stable genotypes across locations

4
 2. LITERATURE REVIEW

2.1. Sorghum Research in Ethiopia

The Ethiopian Sorghum Research was commenced in 1957 by Haramaya University, the then
College of Agriculture with the subsequent initiation of Ethiopian Sorghum Improvement
Program (ESIP) with the fund from International Development Research Centre (IDRC), and
Canada (Firew Mekbib, 2005). Sorghum breeding activities were conducted in the diverse
ecological parts of the country by support of different national and international organizations
(Firew Mekbib, 2005). The collection, evaluation, characterisation and conservation were one
of the primary sorghum activities. Closer to 8000 indigenous collections were done (PGRC/E,
1986).

Currently different types of crossing programs with designed product concept were
undertaken by the national sorghum improvement program to solve sorghum production
problems related to production and productivity. Research has also been conducted to develop
technologies in the areas of variety development, management and cropping system,
protection, food science, socio economics and research extension with the major products in
the variety development (EIAR, 2014). On the past five decades more than 50 improved
sorghum varieties with different desirable features were released for the four major agro-
ecologies of Ethiopia (dry lowlands, humid lowlands, intermediate and highland elevation
areas) (MARC unpublished document).

2.2. Genotype by Environment Interaction

Living organisms are identified neither by their genes nor by their environment; they are the
result of the interaction of genes and environment (Suzuki et al., 1981). A genotype x
environment interaction is a change in the relative performance of two or more genotypes
tested in two or more environments. Interactions may involve change in rank order for
genotypes between environments and change in the absolute and relative magnitude of the
genetic, environmental and phenotypic variances between environments (Bowman, 1972). In
general, genotype by environment interactions happen when two or more genotypes perform
differently in several environments, and are selected as differential genotypic sensitivities to

5
environments (Sarah and Maria, 2011) and GXEs are also considered as a challenge to crop
improvement in a target region (Kang, 1998).

The genotype by environment interaction and yield stability have been challenges to the
breeders and biometricians for many years. A significant genotype by environment interaction
is used to diminish the genotype means across environments for choosing and advancing high
yielder genotypes to the next stage of selection (Pham and Kang, 1988, Asfaw, 2007). A non-
significant genotype by environment interaction simplified the selection because the ‘best’
genotype in one environment would also be the ‘best’ genotype for all target environments
(Basford and Cooper, 1998, Asfaw, 2007).

The phenotype of an individual is determined by the effects of its genotypes, the environment,
and the interaction between the genotype of the individual and the environment (Yan and
Tinker; 2005, Abiy and Firew, 2016). The genotype by environment interaction results in
non-stable performances between the genotypes across environments. Thus, significant GXE
results from the changes in the degree of differences between genotypes in diverse
environments or changes in the relative ranking of the performance of the genotypes
(Falconer, 1952) and (Fernandez, 1991). According to Baker (1990) and Cornelius et al.
(1996) GXEs have been categorized in to crossover and non-crossover interactions.

The change in the response of genotypes to different environments is called a crossover

interaction when genotype ranks change from one environment to another. A main
characteristic of crossover interaction is intersecting genotypes in a graphical illustration. If
the genotypes do not intersect, there is no crossover interaction (Kang, 1998). Non-crossover
(quantitative) interactions indicate changes in extent of genotype performance, but rank order
of genotypes across different environments remains constant.

Genotype x environment interaction is one of the main challenge in the selection of broad
adaptation and stable genotypes in most breeding programs. Numerous studies have shown
that both environmental and genetic factors are the cause for the interaction, but in some
studies the large difference of genotypes or environments has been the real cause of the
interaction (Hagos and Fetien, 2011; Mahnaz et al., 2013; Sewagegne et al., 2013).

6
Domitruk et al (2001) showed that the analysis of variance technique is a valuable tool for
estimating the presence and magnitude of GXE. In multi-environment trial, the combined
analysis of variance is vital for estimating variance components related to different sources of
variation, including genotypes, environment and GXE. In multi-environment trial,
environment describes 80 % or greater of the total yield variation (Yan and Hunt, 2002; Abiy
and Firew, 2016). The environmental factors that are causal to the variances in mean grain
yield across various environments and years may include soil types, sowing dates,
temperature and amount of rainfall during the crop cycle (Dagnachew et al., 2014; Abiy and
Firew, 2016).

Different authors have evaluated their experiment on various crops and as they have stated in
a multi environment trial for yield, the total variation of the contribution of environmental
sum square takes the largest share Abiy and Firew (2016); Kinde et al. (2016); Asfaw (2007,
2008), and Vangge et al. (2014) on sorghum genotypes;Hagos and Fetien (2011) on sesame;
Sewagegne et al. (2013) on rice; Muez et al. (2014) on malt barely; Akcuraet et al. (2006) on
durum wheat; Shrestha et al. (2012) on maize; Dagnachew et al. (2014) on triticale.

The effects of genotype and environment on sorghum grain were investigated using 15
sorghum genotypes, which were tested during three years (2003–2005) at three different
locations (Melkasa, Kobo, and Mieso) to evaluate the effect of GXE on sorghum yield
performance in the drought stressed parts of Ethiopia (Asfaw, 2007). This study showed that
the effect of environments and genotypes were different and the performance of genotypes in
various environments was different. The contribution of genotypes, environments and GXE
were 5.9 %, 73.8 % and 20.3 % of the total sum of squares, respectively. The magnitude of
the GXE sum of squares was 3.41 times larger than the genotypes, indicating, that there were
significant differences in genotype response across environments.

Different authors such as Kinde et al. (2016) studied on the effect of genotype and
environment on sorghum yield to improve the yield potential and stability of sorghum
genotypse. Genotype and environment effect were investigated using 5 sorghum varieties
were grown during 2013 at three different locations (Mechara, Mieso and Hawi Gudina) to
test the effect of GxE on sorghum yield performance. The contribution of genotype,

7
environment and GxE were 4.53%, 86.89% and 2.28% of the total sum of squares,
respectively. The environment where the genotype tested was very diverse, while the
genotypes were not diverse due to small total of square of variety. The magnitude of the GxE
sum of squares was 2 times smaller than that of genotypes. Genotype and environment effect
were evaluated using 25 sorghum genotypes were grown during 2014 at four different
locations (Erer, Kobo, Mieso and Shewa Robit) to test the effect of GxE on sorghum yield
performance (Abiy and Firew, 2016). The contribution of genotypes, environments and GxE
were 6.86 %, 74.19 % and 18.98 % of the total sum of squares. This study showed that the
effect of environments and genotypes were different and the performance of genotypes in
various environments was different. The magnitude of GXE sum of squares was 2.78 times
larger than that of genotypes.

The strategies for crop development for drought resistance include identification and selection
of traits that contribute to the well performance of the crop under drought conditions. This
allows selective addition of the traits that contribute to drought resistance or tolerance for a
specific target environment (Blum, 1983; Rosenow et al., 1983). Taye et al. (2008) reported
that grain yield is a function of linear additive effects of various yield components, it has
always been important to pay close attention to these traits. The relative contribution of the
traits may differ also depending upon the environment and crop management conditions
where the crop is grown.

In terms of yield, Ludlow and Muchow (1990) reported that late flowering varieties tend to
yield higher than early flowering ones. Early flowering varieties mostly produce fewer
assimilating organs (i.e., leaf area) which results in less production of assimilates but good to
escape late season drought stress (Assar et al., 2009). Bakheit (1990) studied 22 sorghum
genotypes for drought stress and omitting one irrigation during stages of before flowering
period, kernel filling period, and physiological maturity period and reported that moisture
stress was reduced grain yield, plant height and panicle weight during flowering stage while
1000 seed weight was during grain filling period.

Drought stress during grain development also shortens the grain filling period, which reduces
the final grain size and if the stress is extreme the grain will be small, and shrunken which

8
finally reduce yield (Younesi and Moradi, 2009). Water deficient during grain development
frequently interrupts development and results in small grain size (Cruz-Aguado et al., 2000).
The reduction in grain size is due primarily to a shortening of the grain filling period rather
than an inhibition of grain growth rate (Vieira et al., 1992).

2.3. Causes of Genotype X Environment Interaction

Genotype x environment interaction is one of the main challenges in the selection of wide
adaptation in most breeding programs. The phenotype effect of an individual is determined by
the combined effect of the environment, the genotype and the interaction. Several studies have
revealed that both environmental and genetic factors are the cause for the interaction, but in
some studies have shown that the large difference of genotypes or environments has been the
real cause of the interaction (Asfaw, 2007; Hagos and Fetien, 2011; Mahnaz et al.,
2013; Sewagegne et al.2013; Abiy and Firew, 2016; Kinde, 2016).

Environmental effect is the greatest (Asfaw, 2007; Hagos and Fetien, 2011; Mahnaz et al.,
2013; Sewagegne et al., 2013; Abiy and Firew, 2016; Kinde, 2016) but it is challenging to
selection. In Ethiopia, the relationship between selection environments and target production
environment had been a major problem because many of the selected activities performed by
the conventional approach are in on-stations which are good production environments
(Temesgen et al., 2015). Many statistical approaches consider all of the phenotypic variation
(i.e., means across environments), which may be misleading. Genotype by environment
interaction is not merely a problem, it is also an opportunity" (Simmonds, 1991). The varietal
stability could be challenged not only due to the change in the test environment but also due
to change in growing season per environment (Dagnachew et al., 2014). Specific adaptations
can make the difference between a good variety and a superior variety. Some environmental
variations are predictable (soil type, soil fertility, plant density) and others also may be
unpredictable (rainfall, temperature, humidity etc.,) (Allard and Bradshaw, 1964).

2.4. Concept of Stability

Stability refers to the adaptation or suitability of genotypes to diverse sets of environments

and used to select stable genotypes (MOORTHY et al., 2012). The term stability is used to
9
characterize a genotype, which shows a relatively consistent yield, independent of change in
environmental conditions. Due to this idea, genotypes with a minimal variance for yield
across different environments are considered to be stable (Sabaghnia et al., 2006; Abiy and
Firew., 2016). Yield stability affected by genotype X environments interaction. The cause of
differences between genotypes in their yield stability is the wide occurrence of GXE.

A variety or genotype is considered to be more adaptive or stable if it has a high mean yield
but a low degree of fluctuation in yielding ability when grown over diverse environments.
Two basic phenotypic stability concepts are distinguished as the biological or static concept,
and the agronomic or dynamic concept. The biological concept of stability refers to the
constant performance of a genotype over a wide range of environments. According to Becker
and Leon (1988), in biological stability, a genotype possesses unchanged performance
regardless of variation of the environments, thus, implying that its variance among
environments is zero. On the other hand, dynamic stability, also termed as agronomical
concept of stability, implies that a stable genotype should always give high yield expected at
the level of productivity of the respective environments, which means that a variety with GXE
as small as possible (Becker, 1981, cited by Abiy and Firew, 2016).

Becker and Leon (1988) stated that all stability procedures based on quantifying GXE effects
belong to the dynamic stability concept. This includes the procedures for partitioning the
GXE of Wricke’s (1962) ecovalence and Shukla’s (1972) stability of variance, procedures
using the regression approach such as proposed by Finlay and Wilkinson (1963), Eberhart and
Russell (1966) and Perkins and Jinks (1968), as well as non-parametric stability statistics.

Lin et al. (1986) identified three concepts of stability: type one concept a stable genotype
possesses an unchanged performance regardless of any variation of the environmental
conditions. Parameters used for this type of stability are coefficient of variability used by
Francis and Kannenburg (1978) for each genotype as a stability parameter and the genotypic
variances across environments. Becker and Leon (1988) called this stability a biological
concept of stability. Type two stability concepts select a stable genotype, if a genotype has no
deviations from the general response to environments and thus permits a predictable response
to environments. A regression coefficient by Finlay and Wilkinson (1963) and Shukla (1972)
10
stability variance can be used to measure this type of stability. Becker and Leon (1988) called
this stability agronomic concept of stability. Type three stability concept refers to a genotype
that has a small mean deviation. Therefore a genotype is considered to be stable if the residual
mean square from the regression model on the environmental index is small. Breeding for
broad adaptability requires a different interpretation and approach to the stability analysis
procedure than breeding for specific adaptability. According to Becker and Leon (1988) this
is part of the agronomic stability concept. Methods to describe this type of stability are the
methods of Eberhart and Russell (1966) and Perkins and Jinks (1968).

2.5. Methods of Statistical Analysis of G x E Interaction

The statistical analysis of G x E interaction is important in the analysis of experiments in

plant breeding and crop production (Kang, 1996, Zerihun T., 2011). Different statistical
methods have been used for the estimation and partitioning of G x E interactions. The
most common methods are the conventional analysis of variance, stability analysis,
multivariate methods and qualitative methods (Freeman, 1990; Zerihun, 2011).

2.5.1. Conventional analysis of variance

In an experiment the yield of genotypes (G) is measured in environments (E) and each with
replicates (R), the classic model for analyzing the total yield variation contained in GXE
observations is the analysis of variance (Fisher R.A, 1925; Alberts, 2004). The residual mean
square within environment measures the error in estimating the genotype. After removing
the replicate effect when combining the data, the GXE observations are partitioned into
two sources: (a) additive main effect for genotypes and environments and (b) non-additive
effects due to GXE interaction. The analysis of variance of the combined data
expresses the observed (Yij) mean yield of the ith genotype at the jth environment as;

Yij = µ + Gi + Ej + (GXE)ij + eijk

Where, µ is the general mean; Gi, Ej and GXEij represent the effect of the genotype,
environment, and G x E interaction, respectively; and eij is the average of the random errors
associated with the kth plot that receives the ith genotype in the jth environment.

11
The non-additivity interaction (GXEij) implies that the expected value of the ith genotype
in the jth environment (Yij) depends not only on the levels of G and E separately but also on
the particular combination of levels of G and E (Crossa, 1990, Zerihun, 2011). The major
problem in ANOVA is that the error variances over environments should be homogenous to
test for genotypic differences.

2.5.2. Spatial and linear mixed model

In evaluation of sorghum trials in Mali, (Leiser et al., 2012, Sarker and Singh, 2015)
concluded that spatial models add further value to experiment over the classical complete
block design model. Plant breeders evaluate much larger number of genotypes, e.g. 70
genotypes of sorghum in trials in Mali evaluated by Leiser et al. (2012). In such situations, a
row–column design with most suitably identified spatial model would be expected to result in
even further higher efficiency, heritability and genetic gain over the conventional complete
block designs.

There is a need to identify sources of variation to fit an appropriate model to data set. This is
due to spatial analysis taking into account variations that result from plot or row and column
location in field experiment data. Spatial variation modeling is particularly important at the
early stages of selection when there are many varieties grown in few locations with few
replicates. When modeling spatial variation, the error variation is broken up. Local trends are
modeled using a covariance structure while global and extraneous variation is modeled using
design factors and functions of the co-ordinates of the plots. The model for spatial analysis is
the same as the linear mixed model and is proposed by Gilmour et al. (1997).

y = Xτ+ Zu + e

Where y is the observed outcomes, τ is the fixed effects (over all mean and genotype effects)
and u is the random effects (replication, row and column effects). X is the design matrix
associated with the fixed effects, Z is the design matrix associated with the random effects
and e is the residual.

12
The linear mixed model is an expansion of the classical linear model and so its basic
properties may be applied to the analysis of mixed models. The difference of the mixed
model, compared to the linear model presented is that the mixed model assumes that some
effects are random variable such as replicate, row and column etc., termed as random effects.
In the linear model only the residual terms are considered as random.

The mixed model is essential in the analysis of plant genotype selection data. Growing areas
for crops differ greatly in location and season; therefore it is important to consider that not all
genotypes prosper to the same level in different environments (GXE). The advantage of the
linear mixed model approach is that it can handle unbalanced data, and can easily be extended
to include complex covariance structures between trials and genotypes. Genotype effects
made be considered as fixed or random, but GXE interactions are always regarded as random.
Regarding genotype effects as random allows for more reliable predictions of the genotypes
performance and addresses issues of selection bias. The linear mixed model and spatial model
(Gilmour et al. 1997) are similar.

Multi-environment trial; a trial carried out at a series of different locations, perhaps one or
over several years. Multi-environment trials are the typical experimental design for comparing
multiple test new genotypes of a crop against existing genotype or standard for beneficial
properties such as increased yield and drought tolerant. This is done to make
recommendations on which varieties are to be identified and selected for a consistent
performance of optimum yield in a diverse range of environments. Thus genotype by
environment interaction is a crucial factor in selection programs and hence why MET's are the
common design adopted when making variety recommendations.

There are several stages of testing and selection in the breeding of a new variety for improved
yield. Selections are based on analysis of yield data, with minimal possible error to ensure
accuracy in future grain yield predictions for new varieties. Multi-environment trial analysis is
consisted of two stages; initially the mean yield for genotypes are estimated for the individual
locations, followed by a combined overall analysis of genotype performance across all trials
(Cullis et al, 1998). Both cases can be analyzed by linear mixed models, usually fitted using
residual maximum likelihood (REML) estimation. The advanced Multi-environment trial
13
analysis methods involve spatial modeling of the MET data. These models include genotype
(G) and environment (E) effects as well as GXE. The effects G and GXE are regarded as
random and assumed independent with constant variance. This constant variance assumption
indicates that all environments have the same genetic variance which is unlikely to be true
(Cullis. et al., 2006). A general mixed model for MET analysis is defined as:

y = Xτ + Zgug +Zpup +e

Where, Y is the combined individual plot yields across trials, τ is the fixed effects including
an overall mean for each of the sites as well as other trial-specic fixed effects, with
corresponding design matrix X, ug are the variety effects in each environment with associated
design matrix Zg, up is the extraneous trend effects in each environment (for example, rows
and columns) with corresponding design matrix Zp, e is the residual errors across all trials

2.5.3. Estimation of stability parameters

Stability analysis provides a summary of response of genotypes to environmental changes. To

identify stable genotypes, the G x E interaction must be divided into stability statistics that are
assigned to each genotype evaluated across environments. Stability indices have allowed
researchers to identify widely adapted genotypes for use in breeding (Yayeh and Bosland,
2000).

2.5.3.1. Regression coefficient (bi) and deviation mean square (S2di)

Joint regression analysis or joint linear regression is the main type of stability analysis and it
is the regression of the genotypes means on environmental index (Freeman, 1973). Joint
regression analysis provides testing whether the genotypes linear responses to changes in
environments. Joint regression analysis was first proposed by Yates and Cochran (1938) and
then widely used and reviewed by various authors (Alberts, 2004).

Eberhart and Russell (1966) method involves the use of joint linear regression where the yield
of each genotype is regressed on the environmental mean yield. In this model the SS due to
environments and GXE are partitioned into environments (genotypear), GXE (linear) and

14
deviations from regression (pooled deviation over all the genotypes). The genotype
regressions term was tested for significance using an F-ratio by taking the deviations from
regressions mean square as the error term. The deviations from regressions mean square were
tested for significance using the error term for overall GXE in the ANOVA.

Simple linear regression provides a model for genotypic stability and is the most
widely used statistical technique in plant breeding (Ramagosa and Fox, 1993). This model
is also called the Finlay and Wilkinson (1963) approach. During regression analysis the G x E
interaction is partitioned into two: i) a component due to linear regression (bi) of the ith
genotype on the environment mean and ii) a deviation (dij):

(GE)ij = biEj + S2di and thus,

Yij = µ + Gi + Ej + (biEj + S2di) + eij

Different authors used different bi values to define genotype stability. Finlay and Wilkinson
(1963) defined a genotype with bi = 0 as stable (static concept) and Eberhart and
Russell (1966) defined a genotype with bi = 1 as stable (dynamic concept). Becker and
Leon (1988) suggested that ecovalence rather be used, since it combines bi and S2di as a
stability parameter. Many scientists consider bi as a response parameter and S2di as a stability
parameter.

Many Authors such as Abiy (2016) on sorghum, Parmar D.J et al (2016) on rice ; P. O. Kurt
Polat (2016) on bread wheat; Dagnachew Lule (2014) on Triticale, conducted their trial
experiment on yield stability and identified four, five, three, three stable and high yielder
genotypes using Eberhart and Russel joint regression model, respectively.

2.5.3.2. Ecovalence (Wi)

Wricke (1964) defined the concept of ecovalence as the contribution of each genotype to the
G x E interaction sum of squares. The ecovalence (Wi) or stability of the ith genotype is its
interaction with the environments squared and summed across environments. For this reason,
genotypes with a low Wi value have smaller deviations from the mean across environments
and are thus more stable. It is express as:
15
Where, Yij is the mean performance of genotype i in the jth environment and are
the genotype and environment mean deviations respectively, and Y is the overall mean.

According to Becker and Leon (1988), ecovalence measures the contribution of a genotype to
the G x E interaction; a genotype with zero ecovalence is regarded as stable. Genotypes with,
an average, small residues are preferred because they show variability that is more
predictable. The ecovalence, however, strongly depends on the environments included in the
test and may lead to different ranking orders of genotypes.

2.5.4. Multivariate analysis methods

According to Crossa (1990), multivariate analysis has three main purposes: a) to eliminate
noise from the data pattern (i.e., to distinguish systematic from non-systematic variation); b)
to summarize the data; and c) to reveal a structure in the data. In contrast with
classic statistical methods, the function of multivariate analysis is to elucidate the internal
structure of the data from which hypotheses can be generated and later tested by
statistical methods (Gauch, 1982a; Gauch, 1982b).

2.5.4.1. Additive main effects and multiplicative interaction (AMMI)

AMMI model has been used for statistical analysis of yield trials, and the best as
compared to other statistical analysis such as ANOVA, principal component analysis and
linear regression analysis (Gauch and Zobel, 1988, Zerihun T., 2011) because it combines
the analysis of variance for the genotype and environment main effects with principal
components analysis of the G x E interaction and several graphically represented interactions
for principal component axis (IPCA) (Crossa et al., 1990). AMMI is also used for predicting
genotypic yields in specific environments (Annicchiarico, 1997) and summarizing the pattern
and relationship of genotypes, environment and their interaction (Gauch and Zobel, 1988).

16
2.5.4.2. GGE Biplot Model

GGE biplot provides breeders with a complete and visual estimation of all features of the data
by forming a biplot that simultaneously represents both mean performance and stability,
optimized environments for specific genotypes and identifies mega-environments. GGE
analysis partitions G + GE into principal components through singular value decomposition of
environmentally centered yield data (Yan, 2001).

Different authors have been conducted multi-environment trials or GXE on different crops
using GGE biplot such as Abay and Bjornstad (2009); Dehghani et al. (2006); Tesfaye et al.
(2008) and Gasura et al. (2015) in their finding reported the existence of a good testing
environment for selecting widely adaptable and high yielding cultivar.

17
 3. MATERIALS AND METHODS

3.1 Description of the Study Area

The field experiment was conducted during the 2017 main cropping season at five locations
(Sheraro, Kobo, Mieso, Shaorobit and Erer), representing the dry lowland areas of Ethiopia
located in the altitude range of 1297 - 1513 m.a.s.l, where sorghum is widely grown. The
detailed agro-ecological features of the locations are presented in Table 1.
Table 1: Agro-ecological features of the experimental locations.

Altitude Soil Rainfall Minimum Maximum

Location Longitude Latitude
in m.a.s.l type in mm To To
Mieso 39o21΄E 8o30΄N 1470 vertisol 571.9 16.0 31.5
Erer 42o 15’E 9o 10’ 1297 vertisol 510.0 15.5 31.1
Shoarobit 39o 93’E 10o 35’ N 1500 vertisol 713.0 17.7 32.6
Kobo 39o38’ E 12o09’N 1513 vertisol 677.8 14.8 32.1
Sheraro 38o9’ E 14o6’ N 1179 vertisol 615.0 20.4 33.7

Source: National Metrology data of 2017 main cropping season, m.a.s.l = meters above sea
level, To = Temperature.

3.2. Genetic Materials

Planting materials used for the experiment comprised of fifty nine early maturing advanced
sorghum genotypes, which were advanced from pedigree breeding at Melkassa Agricultural
Research Center and one sorghum variety melkam (released from Melkassa Agricultural
Research Center for low moisture stress areas of the Ethiopian lowlands in 2009) as a
standard check. Originally, the parental genotypes were developed from germplasm collection
and characterization and were crossed for grain yield and early maturity at Melkassa
Agricultural Research Center, then genotypes were evaluated, and advanced to F6 stage
through pedigree selection at Erer, Mieso, Shorobit, Kobo and Sheraro by the Ethiopian
national sorghum breeding program. The selected genotypes were advanced to preliminary
yield trial (the F7 stage) and tested for one year across five locations, then the best and
superior genotypes with early type were selected and advanced to early maturing sorghum
national variety trial in five locations, where represent the dry low land areas of Ethiopia
(MARC unpublished Sorghum field book and manual).

18
 Table 2: Description of sorghum genotypes tested at five locations during 2017 main cropping season

Entry#. Genotype Pedigree Seed source Status

1 Melkam WSV 387 2016MS Breeder Seed Inc. Standard check
2 14MWLSDT7060 Macia/76T1#23 2017MW Seed increase#2 Advanced genotype
3 12MW6251 WSV 387/76T1#23 2017MW Seed increase#3 Advanced genotype
4 14MWLSDT7410 ICSR24010/B-35 2017MW Seed increase#4 Advanced genotype
5 12MW6302 WSV 387/76T1#23 2017MW Seed increase#5 Advanced genotype
6 14MWLSDT7322 SDSL2690-2/76T1#23 2017MW Seed increase#6 Advanced genotype
7 14MWLSDT7395 MR812/76T1#23 2017MW Seed increase#7 Advanced genotype
8 14MWLSDT7400 WSV387/76T1#23 2017MW Seed increase#8 Advanced genotype
9 14MWLSDT7310 Teshale/B-35 2017MW Seed increase#9 Advanced genotype
10 13MWF6#6077 ICSR 24010/Brihan 2017MW Seed increase#10 Advanced genotype
11 14MWLSDT7325 SDSL2690-2/76T1#23 2017MW Seed increase#11 Advanced genotype
12 2005MI5069 M36121/P9402 2017MW Seed increase#12 Advanced genotype
13 14MWLSDT7196 WSV387/76T1#23 2017MW Seed increase#13 Advanced genotype
14 14MWLSDT7311 Teshale/B-35 2017MW Seed increase#14 Advanced genotype
15 14MWLSDT7157 WSV387/E-36-1 2017MW Seed increase#15 Advanced genotype
16 14MWLSDT7193 WSV387/E-36-1 2017MW Seed increase#16 Advanced genotype
17 14MWLSDT7332 SDSL2690-2/76T1#23 2017MW Seed increase#17 Advanced genotype
18 14MWLSDT7115 ICSR24010/B_35 2017MW Seed increase#18 Advanced genotype
19 14MWLSDT7176 WSV387/E-36-1 2017MW Seed increase#19 Advanced genotype

19
20 14MWLSDT7209 WSV387/76T1#23 2017MW Seed increase#20 Advanced genotype
21 12MW6440 LocalBulk(White)/SRN-39/76T1#23 2017MW Seed increase#21 Advanced genotype
22 14MWLSDT7201 WSV387/76T1#23 2017MW Seed increase#22 Advanced genotype
23 12MW6146 WSV 387/E-36-1 2017MW Seed increase#23 Advanced genotype
24 14MWLSDT7364 2005MI5060/B-35 2017MW Seed increase#24 Advanced genotype
25 Pipline 2 (Teshale/E-36-1)BC3F3 2017MW Seed increase#25 Advanced genotype
26 14MWLSDT7413 WSV387/E-36-1 2017MW Seed increase#26 Advanced genotype
27 13MWF6#6037 2001 MS 7007/SRN-39 2017MW Seed increase#27 Advanced genotype
28 14MWLSDT7207 WSV387/76T1#23 2017MW Seed increase#28 Advanced genotype
29 14MWLSDT7040 WSV387/76T1#23 2017MW Seed increase#29 Advanced genotype
30 14MWLSDT7036 WSV387/76T1#23 2017MW Seed increase#30 Advanced genotype
31 14MWLSDT7324 SDSL2690-2/76T1#23 2017MW Seed increase#31 Advanced genotype
32 12MW6243 WSV 387/76T1#23 2017MW Seed increase#32 Advanced genotype
33 12MW6420 LocalBulk(White)/SRN-39/E36-1 2017MW Seed increase#33 Advanced genotype
34 14MWLSDT7238 Macia/E-36-1 2017MW Seed increase#34 Advanced genotype
35 12MW6444 LocalBulk(White)/SRN-39/76T1#23 2017MW Seed increase#35 Advanced genotype
36 14MWLSDT7402 WSV387/76T1#23 2017MW Seed increase#36 Advanced genotype
37 14MWLSDT7234 Macia/E-36-1 2017MW Seed increase#37 Advanced genotype
38 12MW6471 IESV92084/E36-1 2017MW Seed increase#38 Advanced genotype
39 14MWLSDT7042 WSV387/76T1#23 2017MW Seed increase#39 Advanced genotype
40 14MWLSDT7033 WSV387/76T1#23 2017MW Seed increase#40 Advanced genotype

20
41 14MWLSDT7241 Macia/E-36-1 2017MW Seed increase#41 Advanced genotype
42 14MWLSDT7191 WSV387/E-36-1 2017MW Seed increase#42 Advanced genotype
43 2005MI5093 PGRCE22880/P9403 2017MW Seed increase#43 Advanced genotype
44 2401 (S35/B35)/S35 2017MW Seed increase#44 Advanced genotype
45 2004MW6197 SDSL-2690-2/SAR-39 2017MW Seed increase#45 Advanced genotype
46 2005MI5064 WSV387/P9403 2017MW Seed increase#46 Advanced genotype
47 2523 (ICSV111/B35)/ICSV111 2017MW Seed increase#47 Advanced genotype
48 04MW 6043 WSV387/Dabar 2017MW Seed increase#48 Advanced genotype
49 2005MI5057 WSV387/P9401 2017MW Seed increase#49 Advanced genotype
50 04MW 6079 SDSL2690-2/Dabar 2017MW Seed increase#50 Advanced genotype
51 14MWLSDT7202 WSV387/76T123 2017MW Seed increase#51 Advanced genotype
52 14MWLSDT7291 Macia/76T123 2017MW Seed increase#52 Advanced genotype
53 2001MS7036 PGRCE222878/ICSV708 2017MW Seed increase#53 Advanced genotype
54 90MW5319 85LPYT-224/(148/Framida) 2017MW Seed increase#57 Advanced genotype
55 99MW4047 ((148/E-35-1)-4/CS3541derive5-4-2-1)/P9401 2017MW Seed increase#55 Advanced genotype
56 05MW6026 M36121/P9401 2017MW Seed increase#56 Advanced genotype
57 14MWLSDT7421 Macia/76T123 2017MW Seed increase#58 Advanced genotype
58 2003MW6053 ICSV112BF/SRN-39 2017MW Seed increase#59 Advanced genotype
59 2294 (S35/B35)/S35 2017MW Seed increase#60 Advanced genotype
60 2003MW6038 ICSV111/SRN-39 2017MW Seed increase#61 Advanced genotype

21
3.3. Experimental Design

The trial was conducted using RCBD with row column arrangement and three replications.
The experimental plot consisted of 2 rows, each 5m in length with 0.75m between row
spacing and 0.15m between plants. The total area of each plot had a size of 7.5m2.

3.4. Trial Management

As per the recommendation for sorghum prduction in the lowland areas of Ethiopia, Di-
ammonium phosphate (DAP) and urea were applied at the rate of 100 kg/ha and 50 kg/ha,
respectively. Di-ammonium phosphate was applied by incorporating into the soil during
planting of the seeds and Urea was applied as side dressing at knee height stage (35 days )
after planting of the seed. Thinning was done after three weeks of planting to maintain the
space between plants and to balance the plant density. Other crop management practices were
applied following the recommended practices.

3.5. Data Collection and Sampling Techniques

Data were collected from the two rows for plot based data and five randomly sampled plants
for Plant based data following the descriptors for sorghum (IBPGR/ICRISAT, 1993). The
details of the data collection were as follow:
Days to emergency: the number of days from planting to when 50% of plants started
emergency.
Days to flowering: the number of days from emergence to when 50% of plants started
flowering.
Days to maturity: the number of days from planting to the date when 95% of the
plants matured physiologically.
Plant height (cm): Plant height was measured from five randomly sampled main plants from
the two rows at 75 % physiological maturity. The average height from the five plants was then
recorded for the plot.
Panicle length (cm): the average length of five randomly selected plants from the base of the
panicle to the tip was measured.

22
Panicle width: the average width of individual panicle as measured panicle width
measurement in the widest diameter of the panicle on five randomly selected plants per plot at
maturity.
Panicle weight (g): the total weight of panicles per plot after harvest was measured.
Number of panicle: the total number of panicles per plot during harvest was counted.
Panicle yield (g): the yield was obtained by threshing the total number of panicles per plot
during harvest.
Grain filling rate (kg/ha/day): it is the ratio of grain yield (kg/ha) to grain filling period and
calculated as follows:
Grain filling rate (kg/ha/days) = Grain yield (kg/ha)
Grain filling period (days)
Thousand seed weight (TSW): Weight of 1000 seeds in gram that was measured from each
plot after the moisture level adjusted to 10%.
Drought score: tolerance to moisture deficit and scored as 1-5, 1= more tolerant, 2= mild
drought tolerant, 3= moderate tolerant, 4= severe drought and 5= extreme drought.
Staygreen: level of greenness scored 1-5 after maturity, 1 indicates completely green normal
size leaves (no leaf death), 2 = 25% of the leaves died, 3 =26 to 50% of the leaves died, 4 =51
to 75% are dead, 5 =76 to 100% of the leaves and stem are dead (complete plant death).
Panicle exersion: Score (1=well exserted more than 10cm between ligule of flag leaf to head
base, 2=2-10cm exsertion, 3=less than 2cm but ligule below the head base, 4=peduncle
recurved but head is below the ligule and clearly exposed splitting the leaf sheath, 5=head
covered by leaf sheath).

3.6. Data Analysis

R software using spatial analysis and mixed model (Culllis et al. 1998, Gilmour et al. 1997)
were used to analyze all the collected data from individual location and combined data over
locations.

3.6.1. Analysis of data for each location and combined over location

Using the raw data collected on fifteen characters of sixty genotypes, which were grown at
five locations, analysis of data using spatial and linear mixed and model of RCBD with row

23
column arrangement was computed. Before pooling the data over locations, Bartlett’s test of
homogeneity of variance was adopted for most of the traits to determine the validity of the
overall mixed analysis of the data of combined locations. This analysis revealed the
homogeneity of error variance. Therefore, overall mixed analysis was done to determine the
effects of the genotypes, locations and their first order interactions using mixed model.
Genotypes were assumed to be fixed and environment effects random. Least significance
difference was used to determine the significance of differences among the genotype means
for each character. The model for spatial analysis is the same as the linear mixed model and is
proposed by Gilmour et al. (1997).

y = Xτ+ Zu + e

Where y is the observed outcomes, τ is the fixed effects (over all mean and genotype) and u is
the random effects (replication, row and column effects). X is the design matrix associated
with the fixed effects, Z is the design matrix associated with the random effects and e is the
residual.

3.7. Stability Analysis

The following four analysis of the stability models were performed for grain yield (kg/ha)
using R software.

3.7.1. Eberhart and Russell’s joint regression model

Eberhart and Russell (1966) method involves the use of joint linear regression where the yield
of each genotype is regressed on the environmental index. The behavior of the genotype was
determined by the model: = + + δ using R Software.

Where: Y = the mean performance of the ith genotype in the jth environment, μ = the grand
mean of the ith genotype over all the environments, β = the regression coefficient which
measures the response of the ith genotype on environmental index, I = the environmental
index obtained by the difference between the mean of each environment and the grand mean
and = the deviation from regression of ithvariety in the jth environment.

24
The pooled deviations mean square was tested against the pooled error mean square by the F-
test to evaluate the significance of the differences among the deviations of genotypes being
evaluated from their expected performances. Therefore, to test the validity of the hypothesis
that whether there is significant difference among the 60 genotypes with respect to their mean
grain yields or not and whether there is significant difference among the regression coefficient
or not, genotypes mean square and regression mean square were tested against the pooled
deviation using the F-test.

3.7.2. The additive main effects and multiplicative interaction (AMMI) method

Additive main effects and multiplicative interaction (AMMI) model was performed for the
mean data of grain yield (ton/ha) from each location using R software. The AMMI model
equation is given as:

Y = μ+α +ß + λ γ δ +θ +ε

Where:Y = the mean yield of genotype i in environment j, μ = the grand mean, α = the
deviation of the genotype mean from the grand mean, β = the deviation of the environment
mean from the grand mean, λ = the singular value for the IPCA n, N = the number of PCA
axis retained in the model, γ = the PCA score of a genotype for PCA axis n,δ = the
environmental PCA score for PCA axis n, θ = the AMMI residual and E = the residuals.

The degrees of freedom (DF) for the IPCA axis were calculated based on the following
method (Zobel et al., 1988). DF = G + E – 1 – 2n; Where: G = the number of genotypes, E =
the number of environments and n = the nth axis of IPCA.

3.7.3. GGE Biplot Method

Genotype by environment interaction analysis was done by GGE biplot, which uses singular
value decomposition (SVD) to decompose GGE into two or more principal components. Each
principal component consisted of a set of genotype scores multiplied by a set of environment
scores, to generate a two-dimensional biplot (Ding et al., 2007). In GGE biplots genotype plus
genotype × environment (G + GE) interaction was studied together and to achieve this G +
GE effect is separated out from the observed mean from Equation (1) (by omitting random

25
error and block effect) and eventually model becomes as ij j i ij Y μ β =α +αβ (2) The GGE
(G + GE) effect was partitioned into multiplicative terms using SVD. The model based on
singular value decomposition (SVD) of first two principal components (Ding et al., 2007) is:

Yij – μ – βj = λiξi1η1j + λ2ξi2η2i + ξij

Where λ1 and λ2 are the singular values of the first and second largest principal components,
PC1 and PC2, respectively; ξ1 and ξ2 are the eigenvectors of genotype I for PC1 and PC2,
respectively, and η1 and η2 are the eigenvectors of environment j for PC1 and PC2,
respectively.

3.7.4. AMMI’s stability value (ASV)

In order to compute and rank genotypes according to their yield stability, the additive main
effect and multiplicative interaction effect stability value (ASV) was proposed by Purchase
(1997). It was calculated using Microsoft excel (2007) by employing the following formula:

[ ( )]
= +( 2 ) , Where: ASV = AMMI’s
stability value, IPCA1= interaction principal component analysis one, and IPCA 2=
interaction principal component analysis II.

26
 4. RESULTS AND DISCUSSION

4.1. Data analysis of variance

4.1.1 Single Location data analysis

The heat map/plot map revealed the spatial trends within a field, which is a source of error
that could potentially bias the estimation of genotype effects (Figure 1). Figure 1 showed
areas of high, average and low yield and that neighboring plots tend to be more similar than
plots a greater distance apart. In other words, plots located close to each other are more likely
to have similar yields than distance apart (Gilmour et al., 1997). This correlation between
plots can be used to adjust for trend reducing error. Spatial models [ar1 (row), id (col), id
(row)] are used to adjust spatial variability within a field to each location (Gleeson and Cullis,
1987; Gilmour et al., 1997). These models were fitted to the dataset, while the model ar1
(column) was not fitted to dataset that is not significantly different in the column direction due
to this reason it was rejected.

Areas of
average

yield

Areas of

Areas high yield

of low
yield

Figure 1: Heat map of spatial trend within a field of sixty early maturing sorghum genotypes
on plot grain yield performance tested at Erer during 2017 cropping season.

27
The separate REML analysis of the six characters (days of flowering, days of maturity, plant
height, grain filling period, grain filling rate, grain yield, plant height, panicle weight and
thousand seed weight) for sixty early maturing sorghum genotypes is presented in Table 3.
The result of spatial analysis which is the variance component for grain yield at each site
revealed highly significant difference between genotypes (Table 3). This showed that at each
location there were genetic diversity among genotypes for grain yield.

Genotypes were significantly different for days to flowering at each location (Table 3). This
shows the observed numbers of days that genotypes spent to flower were statistically different
at all locations. Genotypes were significant for days to maturity at Sheraro, Shaorobit, and
Mieso (Table 3). This finding is in contrast to Abiy and Firew (2016) who indicated that
genotypes were not significantly different for days to maturity at Shaorobit and Mieso,
indicating the number of days that genotypes took to mature were similar. However,
genotypes were not-significantly different for days to maturity at Erer and Kobo; show no
genotypic effect for days to maturity.

Genotypes were significantly different at Shaorobit, Erer, Kobo and Mieso but non-significant
for grain filling period at Sheraro. This showed that genotypes took different period to fill
their grain at all locations except at Sheraro. Genotypes were significantly different for plant
height, grain filling rate and 1000 seed weight at all locations. This indicates the performance
of genotypes for plant height and grain filling rate are different in all locations.

28
Table 3: Variance component estimation by REML and spatial model of early maturing sorghum genotypes for grain yield,
phenological and other major traits tested at Erer, Kobo, Mieso, Sheraro and Shewa Robit during 2017 main cropping season.

Traits
Variance
Site DF DTF DTM GFP GFR GY PHT TSW
Component
Genotype 59 14*** 0.81** 9.92*** 501.9*** 520310* 1117*** 10.4***
SR Replicate 2 0.56ns 0.15ns 1.16ns 14.7ns 0.0042* 0.0001ns 0.001ns
Error 118 5.87 1.18 6.39 96.34 41394.7 37.3 2.8
Genotype 59 10.4*** 0.55ns 15.4** 476*** 449593* 411*** 9.1***
ER Replicate 2 0.001ns 2.14ns 2.68ns 5.62ns 0.00127* 0.002ns 0.1ns
Error 118 10.5 18.9 26.8 77.89 12562.1 31.2 3.1
Genotype 59 5.8** 2.16* 3* 114.5*** 130520* 482*** 13.39***
MS Replicate 2 1.19ns 1.75ns 0.0001ns 0.11ns 0.0015* 0.001ns 0.006ns
Error 118 9.32 9.41 7.14 14.41 14606.5 21.7 2.12
Genotype 59 18.1*** 2.13ns 15.8** 209*** 688377* 732*** 13.09***
KB Replicate 2 0.43ns 0.001ns 0.84ns 1.01ns 941.948ns 0.001ns 0.04ns
Error 118 12.8 12.1 20.1 32.64 55189.6 51 2.83
Genotype 59 12.35*** 6.52*** 6.33ns 317.4*** 244754* 735*** 11.54***
SH Replicate 2 0.0003ns 0.0001ns 0.001ns 0.000012ns 0.0006* 0.55ns 0.0023ns
Error 118 1.78 6.83 35.4 179.9 57574.5 68 2.1

Where REML = Residual Maximum Likelihood, DF = Degree of Freedom, DTF = Days to flowering, DTM = Days to maturity, GY
= Grain yield, PHT = Plant height, GFP = Grain filling period, GFR = Grain filling rate, TSW = 1000 seed weight, SR = Shaorobit,
ER = Erer, MS = Mieso, KB = Kobo, SH = Sheraro, *** = vary highly significant (P≤ 0.0001), ** = highly significant (P≤ 0.001), * =
significant (P<0.05) and ns = insignificant (P>0.05).

29
4.1.2. Combined analysis of data

The combined variance component estimation by REML of the six characters (days of
flowering, days of maturity, plant height, grain filling period, grain filling rate, grain yield,
plant height, panicle weight and thousand seed weight) for sixty early maturing sorghum
genotypes is presented in Table 4.

The result showed that there were highly significant (P<0.0001) differences among locations,
and differences among the genotypes were significant (P<0.05) for grain yield. This indicates
that the high diversity of the growing conditions in the five locations and the small variability
in the genotypes for grain yield performance. Significant effect of location on yield of
sorghum varieties was reported by Asfaw (2007); Maposa et al. (2010); Almeida et al.
(2014); Abiy and Firew (2016); Lyle et al.(2016). Furthermore, the GxE was also very highly
significant (p≤ 0.0001), showing inconsistencies in the performance of sorghum genotypes
across locations which shows difference in the response of sorghum genotypes at different
environments. This outcome is in agreement with the findings of Kenga et al. (2003); Asfaw
(2007); Almeida et al. (2014); Abiy and Firew (2016);.

The result of the combined variance components showed that the total variation in yield was
attributed to location (86.4 %), genotype (1.9 %) and GxE (10.9 %) effects (Table 4). The
largest proportion of the variance showed on grain yield performance was due to locations
than other variances. Therefore, high percentage of the location component of variation is an
indication that environment is the major factor that affect grain yield performance of sorghum
genotypes in the dry lowlands areas of Ethiopia. Similar results of large location effects were
also reported by Akcuraet et al. (2006); Asfaw (2007, 2008); Hagos and Fetien (2011);
Shrestha et al. (2012); Sewagegne et al. (2013); Muez et al. (2014); Dagnachew et al. (2014);
Vangge et al. (2014); Abiy and Firew (2016); Kinde et al. (2016);. The variance component
due to GxE was 5.74 times higher than that of the genotypes. The highest magnitude of the
interaction as compared to the genotype component revealed that the grain yield performance
of sorghum genotypes across locations was different (Asfaw, 2007; Abiy and Firew, 2016),
indicating inconsistencies in the performances of sorghum genotypes across environments. A
significant GxE may be either a non-cross-over or cross-over type (Baker, 1990; Cornelius et

30
 al., 1996). In the current study, the interaction was of cross-over type as the ranking of
genotypes for grain yield changed at every location.

Significant differences among genotypes were found for days of flowering (P≤ 0.0001), grain
filling period (P≤0.0001), grain filling rate (P<0.05), plant height (P≤0.0001), date of
maturity(P<0.05) and 1000 seed weight (P≤ 0.0001). These indicates the presence of the
effect of genetic differences for these traits. Differences among locations were non-significant
(P>0.05) for these traits except grain yield, indicating the absence of variation among
locations for these traits which are independent of location effect in contrast to Asfaw (2007);
Abiy and Firew (2016); Kinde (2016) who indicated that the large presence of variation
among locations had high effect for these traits. Genotype by environment were significant
for days of flowering (P≤ 0.001), grain filling period (P≤ 0.001), days to maturity (P≤ 0.001),
grain filling rate (P≤ 0.0001), plant height (P≤ 0.0001) and 1000 seed weight (P≤ 0.0001)
(Table 4). The highest magnitude of the interaction revealed inconsistencies of the
performance of sorghum genotypes for these traits across locations were different.
Table 4: Variance component estimation by REML of early maturing sorghum genotypes for
grain yield, phenological and other major traits tested at five locations during 2017
main cropping season.

Estimate
Error
Genotype Replicate/Site Site Genotype:Site
Traits (DF = 590)
(DF = 59) (DF = 10) (DF = 4) (DF = 236)
GY 61140* 76.11ns 2743000*** 346000*** 25870
DTF 10.48*** 0.08ns 23.15ns 1.49** 8.41
DTM 0.55* 0.54ns 53.95ns 1.8** 9.95
GFP 6.61*** 0.46ns 73.55ns 3.31** 19.62
GFR 65.51* 2.07ns 2098.39ns 256.85*** 82.54
PHT 491.94*** 0.01ns 1138.56ns 203.7*** 41.91
GY 61140* 76.11ns 2743000*** 346000*** 25870
TSW 4.25*** 0.02ns 7.03ns 7.24*** 2.61
Where REML = Restricted Maximum Likelihood, DTF = Days to flowering, DTM = Days to
maturity, GY = Grain yield, PHT = Plant height, GFP = Grain filling period, GFR = Grain
filling rate, TSW = 1000 seed weight, *** = vary highly significant (P≤ 0.0001), ** = highly significant (P≤ 0.001),
* = significant (P<0.05) and ns = insignificant (P>0.05).

31
4.2. General and relative performance of the genotype

Means for grain yield, phenolgical and other major traits of sixty early maturing advanced
sorghum genotypes at Erer, Kobo, Mieso, Sheraro and Shewa Robit is presented in Table 5.
Shaorobit and Sheraro were the highest yielding environments with mean values of 5174.0
kg/ha and 3706.7 kg/ha, respectively, indicating that the two environments are suitable for
sorghum production, whereas Mieso, Erer and Kobo were the poorest yielding environments
with mean grain yields of 1099.0 kg/ha, 1508.5 kg/ha and 2832.1 kg/ha, respectively (Table
5), revealing that these environments were not favorable for sorghum production. Therefore,
Mieso and Erer were found to be the lowest yielding environment as compared to other
environments. This is mainly attributed to the high drought observed during the study period
(2017).

Flowering days of genotypes at Erer (70.1days) and Sheraro (69.9 days) were statistically
similar. At these two locations the required mean flowering days of genotypes were early than
mean days of flowering at Mieso (77.7 days), Shaorobit (72.3 days) and Kobo (80.7 days).
Compared to the overall locations flowering date of genotypes was late at Kobo (80.7 days).
This finding is similar to Abiy (2016) pointed out the flowering date of genotypes at Kobo
was late. Compared to the overall locations maturity date, it was only at Kobo and Mieso, had
above the mean of the five locations maturity date (109.9 days). At Sheraro and Erer, the
mean maturity days were lower than the grand mean (Table 5). The mean maturity days of
locations were different except Mieso and Shaorobit. At Sheraro (100.6 days), the required
mean of maturity days of genotypes were early than mean days of maturity of genotypes at
Erer (104.9 days), Mieso (112.1 days), Shaorobit (112.4 days) and Kobo (119.62 days).
Compared to the overall locations maturity date, it was only at Kobo, Mieso and Shaorobit,
had above the mean of the five locations maturity date (109.9 days. At Sheraro and Erer, the
mean maturity days were lower than the grand mean (Table 5).

Average grain filling periods of genotypes at Erer (34.8 days) and Mieso (34.5 days) were
similar. At these two locations, grain filling period of genotypes was short as compared to the
other locations, except at Sheraro (31.1 days). At Sheraro, the average period of genotypes to
fill their grain was 31.1 days, this makes the grain filling period of genotypes at Sheraro

38
faster than Erer, Mieso, Shorobit and Kobo. This variation might be due to the differences in
the amount of rain fall and temperature, (Table 5).

The average grain filling rates of genotypes were 45.8 kg/day/ha at Errer, 53.6 kg/day/ha at
Kobo, 32.1 kg/day/ha at Mieso, 121.5 kg/day/ha at Sheraro and 130.8 at Shaorobit (Table 5).
The variation among means of grain filling rate of genotypes in each location was diverse.
The grand mean grain filling rate of locations was 76.5 kg/day/ha, Shewarobit and Sheraro
were the two locations that had faster grain filling rate than the rest three locations. At
Shewarobit, genotypes filled their grains at a faster rate than the genotypes in the other
locations. At Mieso, grain filling rate was the poorest of all the locations.

The mean plant height of all the genotypes at the tested locations was 185.5 cm. Average
height of genotypes at Errer (149.8 cm), Kobo (189.0 cm), Mieso (151.9 cm), Sheraro (216.5
cm) and Shewa Robit (220.1 cm) were different. The highest mean plant height of the
genotypes was observed at Shaorobit, it is higher than the rest locations (Table 5). The lowest
mean were observed at Erer and Mieso. This shows that drought is highly affected the
performance of genotypes with plant height.

The mean 1000 seed weight of all the genotypes at the tested locations was 30.1 g. Average
1000 seed weight of genotypes at Kobo (31.6 g), Mieso (33.3 g), Sheraro (29.3 g) and
Shewarobit (30 g) were statistically similar. The lowest 1000 seed weight was observed at
Erer (26.2 g), shows drought is highly affected the performance of genotypes with seed
weight.

39
Table 5: Means for grain yield, phenolgical and other major traits of sixty early maturing
sorghum genotypes at Erer, Kobo, Mieso, Sheraro and Shewarobit during 2017 main
cropping season.

Traits
Location
DTF GFP DTM GFR GY PHT TSW
Erer 70.1 d
34.8c
104.9 c
45.8b 1508.5d 149.8 c
26.2b
Kobo 80.7a 52.9a 119.6a 53.6b 2832.1c 189.0b 31.6a
Mieso 77.7b 34.5c 112.1b 32.1c 1099.0e 151.9c 33.3a
Sheraro 69.9d 31.1d 100.6d 121.5a 3706.7b 216.5a 29.3a
Shaorobit 72.3c 40.1b 112.4b 130.8a 5174.0a 220.1a 30a
Mean 74.1 38.68 109.9 76.5 2863.9 185.5 30.1
LSD(0.05) 1.9 2.5 1.2 11 238.74 13.5 2.6
CV (%) 3.8 11.3 2.8 11.7 5.6 3.5 5.4

Where DTF = Date of flowering, DTM = Date of maturing, GFP = Grain filling period, GFR
= Grain filling rate, GY = Grain Yield, PHT = Plant height, TSW = 1000 seed weight, LSD =
Least significance difference and CV = coefficient of variation.
Genotype by environment interaction for mean grain yield of sixty early maturing sorghum
genotypes is presented in Table 6. In table 6 existence of cross-over GXE interaction is
evident in which case a significant change in rank occurs from one environment to another. In
the present investigation, the interaction is of cross-over type as the ranking of genotypes
changed at every environment. Table 6 indicates relative performance of each genotype at
each environment. Now we can see that, similarity and difference of overall performance of
the genotypes considering the interaction of genotype by environment. For example,
14MWLSDT7238 has high performance in overall mean; is ranked 3rd in Erer, 30th in Kobo,
1st in Mieso, 12th in Sheraro and 1st in Shoarobit. Visibly one can detect that from the table,
the mean yield of genotypes shows interactions across the test locations with the possible
existence of crossover interaction (Table 6). In general, ranking of genotypes changed from
one environment to another, indicates that, a remarkable GXE and require further
investigation to understand the patterns of interactions.

40
Table 6: Genotype by environment interaction for mean grain yield of 60 early maturing
sorghum genotypes.

Genotype Genotype Name ER KB MS SH SR Genotype

code mean
1 Melkam 1358 2851 1048 3661 4857 2755
2 14MWLSDT7060 1341 3193 1189 4300 5506 3105.8
3 12MW6251 1475 3398 1331 4763 6238 3441
4 14MWLSDT7410 1309 2746 983 3433 4527 2599.6
5 12MW6302 1622 2684 1080 3489 5145 2804
6 14MWLSDT7322 2184 2655 1289 3741 6413 3256.4
7 14MWLSDT7395 1024 3088 1019 3928 4546 2721
8 14MWLSDT7400 1145 2854 965 3549 4359 2574.4
9 14MWLSDT7310 1828 2519 1089 3288 5297 2804.2
10 13MWF6#6077 589 2742 697 3030 2816 1974.8
11 14MWLSDT7325 1926 2252 1013 2835 4991 2603.4
12 2005MI5069 1486 3018 1170 4047 5495 3043.2
13 14MWLSDT7196 1657 2896 1185 3909 5654 3060.2
14 14MWLSDT7311 1957 2442 1107 3212 5446 2832.8
15 14MWLSDT7157 1691 2799 1157 3745 5539 2986.2
16 14MWLSDT7193 1557 2856 1129 3778 5337 2931.4
17 14MWLSDT7332 1388 2965 1109 3893 5158 2902.6
18 14MWLSDT7115 2018 2282 1062 2941 5268 2714.2
19 14MWLSDT7176 1240 3342 1214 4527 5568 3178.2
20 14MWLSDT7209 2113 2146 1040 2736 5219 2650.8
21 12MW6440 1773 2613 1108 3435 5356 2857
22 14MWLSDT7201 1814 3099 1335 4381 6434 3412.6
23 12MW6146 1101 2897 966 3606 4342 2582.4
24 14MWLSDT7364 1460 3122 1205 4229 5644 3132
25 Pipline 2 1587 3235 1304 4514 6173 3362.6
26 14MWLSDT7413 1938 2537 1140 3381 5592 2917.6
27 13MWF6#6037 1115 3401 1190 4570 5392 3133.6
28 14MWLSDT7207 1435 2896 1097 3787 5129 2868.8
29 14MWLSDT7040 1478 3029 1172 4064 5500 3048.6
30 14MWLSDT7036 1426 2803 1053 3605 4918 2761
31 14MWLSDT7324 1693 2691 1111 3540 5325 2872
32 12MW6243 1269 2621 913 3176 4183 2432.4
33 12MW6420 835 3153 972 3947 4230 2627.4
34 14MWLSDT7238 2135 2857 1357 4098 6706 3430.6
35 12MW6444 1800 2700 1157 3616 5596 2973.8
36 14MWLSDT7402 1340 2720 984 3401 4549 2598.8

41
 37 14MWLSDT7234 1436 3283 1266 4522 5913 3284
38 12MW6471 1994 2079 964 2543 4801 2476.2
39 14MWLSDT7042 1600 2868 1150 3824 5461 2980.6
40 14MWLSDT7033 1500 2980 1160 3983 5453 3015.2
41 14MWLSDT7241 1617 2924 1182 3940 5616 3055.8
42 14MWLSDT7191 1360 2399 853 2802 3947 2272.2
43 2005MI5093 1689 2721 1122 3594 5377 2900.6
44 2401 1515 2957 1156 3948 5444 3004
45 2004MW6197 1194 3146 1111 4130 5065 2929.2
46 2005MI5064 2084 2528 1194 3445 5921 3034.4
47 2523 733 2774 768 3172 3224 2134.2
48 04MW 6043 2356 2158 1141 2891 5815 2872.2
49 2005MI5057 1592 2922 1171 3924 5555 3032.8
50 04MW 6079 964 3351 1109 4393 4935 2950.4
51 14MWLSDT7202 1426 2868 1082 3730 5051 2831.4
52 14MWLSDT7291 1710 2749 1143 3660 5484 2949.2
53 2001MS7036 1273 2840 1010 3593 4634 2670
54 90MW5319 1273 3205 1167 4285 5370 3060
55 99MW4047 1414 2897 1089 3777 5080 2851.4
56 05MW6026 1244 3009 1071 3897 4906 2825.4
57 14MWLSDT7421 1430 2941 1115 3869 5207 2912.4
58 2003MW6053 988 2862 906 3478 4003 2447.4
59 2294 1685 2746 1131 3640 5418 2924
60 2003MW6038 1325 2620 935 3204 4313 2479.4
Environment Mean 1509 2832 1099 3707 5174 2864.2
LSD (0.05) 181.72 378.18 193.59 123.45 330.68 504.87
CV (0.05) 7.45 8.26 10.89 2.06 3.95 5.6

Where ER = Erer, KB = Kobo, MS = Mieso, SH Sheraro, SR = Shoarobit.

The mean for grain yield, phenological and other major traits of the top fifteen early maturing
genotypes for combined location is presented in Table 7. The mean grain yield over all the
locations and genotypes was 2863.9 kg/ha; with genotype mean grain yield ranging from
2024.6 kg/ha to 3533.1 kg/ha, indicating wide difference in yield potential across locations
(Table 7). The lowest yielding genotype is 14MWLSDT7191 (2024.6 kg/ha) and is mainly
attributed to the high moisture stress occurred, during the study period, which is similar to the
finding of Menezes et al. (2015); Tardin et al. (2013). All the top fifteen genotypes were
statistically similar to each other. However, they were significantly different with all the

42
genotypes at this bottom (Table 7).The grain yield performance of the standard check was
statistically different to the first six top high yielder genotypes but not with the grand mean
(2863.9 kg/ha) across locations (Table 7), indicating the grain yield performance of the
standard check was not among the first six top yielder genotypes in similar to Abiy and Firew
(2016).

Genotypes with mean grain yield, phenological and other major traits found in the top fifteen
genotypes that vary in rank are better than the rest genotypes, which is high yielder, early
flowering and maturing, short grain filling period, high grain filling rate, tall in plant height
and high thousand seed weight traits are more preferable and vice versa (Table 7). From the
top fifteen genotypes 13MWF6#6037, Pipline 2, 14MWLSDT7176 and 90MW5319 were
early in days to flowering than other genotypes thus making it more adaptable in the moisture
stress conditions of the locations while the remaining eleven genotypes and the check were
late flowering.

Earlness indicates the response of genotypes that escape from moisture stress condition. The
earlness traits (days to flowering, grain filling period and days to physiological maturity)
enables them to flower, grain fill and mature early. That is why days to flowering and
maturity are of the most important attributes that need to be considered in selecting genotypes
for drought affected areas. In this study, number of days to flowering ranged from 64.6 to
79.8 days. The five bottom genotypes (14MWLSDT7325, 2001MS7036, 2523,
14MWLSDT7115 and 14MWLSDT7191) were statistically significant different with the
check. However, the three genotypes (14MWLSDT7325, 14MWLSDT7191 and
14MWLSDT7115) were late flowering and low yielding in contrast with Ludlow and
Muchow (1990) reported late flowering varieties tend to offer yield higher than early
flowering ones. Early flowering genotypes consists of adapting the crop cycle to water
availability and evaporative demand, usually by reducing its duration, thereby reducing the
total demand for water and withstand terminal stress (Tardieu, 2013). Flowering time is the
most critical factor to optimize adaptation, hence grain yield, in environments differing in
water availability and distribution during the growing season (Richards, 2006).

43
Mean days to maturity ranged from 107.5 to 113.6 with a mean value of 109.9 days (Table 7).
The top genotypes were statistically similar with the check and the all bottom genotypes
except genotype 14MWLSDT7191. Genotype 14MWLSDT7191 was late maturing and low
yielding in contrast with Ludlow and Muchow (1990) pointed out that late maturing varieties
tend to yield higher than early maturing ones.

Mean grain filling period ranged from 34.9 to 47.8 days with a mean value of 38.7 days
(Table 7). The top three genotypes (14MWLSDT7176, 04MW 6043 and 90MW5319) were
significantly different from the check and these genotypes spent long time to fill their grain,
indicating high yielder due to the amount of solar energy available for grain filling increase.
Similarly, the bottom three genotypes were statistically significant from the check while the
remaining genotypes were statistically similar from the check.

Mean grain filling rate ranged from 54.1 to 101.4 kg/day/ha with a mean value of 76.8
kg/day/ha (Table 7). The top genotypes were statistically similar to the check. Most of the top
genotypes were statistically significant different from the bottom genotypes. This indicates
high yielder genotypes had the faster grain filling rate and vice verca. However, the four
bottom genotypes were statistically significant different from the check.

Plant height is a complex trait, it is being affected by environmental conditions and

management practices. (Butler et al. 2005 and Al-Temimi et al. 2013) reports indicated that
plant height is directly linked to the productive potential of plant in terms of grain yield since
it represents a good storage organ for photosynthetic metabolites. Therefore, significant
reduction in plant height was noticed due to water stress; however, tolerant cultivars attained
more plant height.

Plant height plays a major role in the acceptance of varieties by users (farmers) in the study
areas, with preference being given to tall plants, which can serve a dual purpose as food and
feed. In plant height, there were a significant difference and statistical different between
genotypes and check. Mean plant height ranged from 125 cm to 238.2 cm with average of
185.5 cm (Table 7). Most of the top fifteen genotypes were statistically significant different
from the check, indicating the wide difference in plant height across locations. In addition,
most of the bottom genotypes were stsistically significant different from the check in similar

44
to Sintayehu (2017) reported that top of the genotypes were statistically significant different
from the standard check. The check is small in stature as compared to most of the top and
bottom genotypes.

Mean thousand seed weight ranged from 21.4 g to 34 g with average of 30.1 g (Table 7). All
the top genotypes except 12MW6302 were statistically similar with the check. Most of the
bottom genotypes were statistically similar with the check. Genotype 14MWLSDT7176 was
the highest 1000 seed weight as compared to the other top genotypes except four genotypes
(Pipline 2, 14MWLSDT7201, 12MW6251, 14MWLSDT7042 and 2005MI5057). Genotypes
that had the maximum 1000 seed weight indicated post flowering drought tolerance. Hence,
traits associated with post-flowering drought tolerance include improved longer grain filling
period, stay-green and seed weight (Borrell et al., 2000b; Burke et al., 2010; Van Oosterom et
al., 1996). Drought stress during seed development shortens the seed filling period (Younesi
and Moradi, 2009) which results earlier maturation of the seeds (Meckel et al., 1984).

Table 7: Means for grain yield, phenological and other major traits of the top 15 and the
bottom 15 early maturing advanced sorghum genotypes tested across locations during
2017 main cropping season.

Genotype Name GY DTF GFP DTM GFR PHT TSW

14MWLSDT7238 3533.1 75.7 37.1 110 98.4 193.1 28.3
14MWLSDT7234 3528.5 74.9 36.8 108.9 101.4 184.6 31
14MWLSDT7196 3494.6 73.7 38.2 109.1 90.9 176.7 30
13MWF6#6037 3465.1 71.9 38.8 107.9 88.7 165.9 29.2
Pipline 2 3462.7 71.9 38.8 107.9 96.9 204.9 30.5
14MWLSDT7176 3453.1 70.6 42.3 110.1 84.7 187.9 34
14MWLSDT7201 3343.6 74.3 37 108.5 93.8 190 31.7
12MW6251 3284.6 75.1 38.1 110.4 90.2 194.7 30.5
04MW 6043 3256.9 74.6 44.2 111.3 81 200.4 27.8
14MWLSDT7332 3249.2 76.2 36.9 110.3 90.4 180 29.2
14MWLSDT7042 3200.9 74.3 38.4 109.9 85 194.2 33.8
2005MI5057 3190.6 74.4 38.5 110.1 87 192.4 33.9
14MWLSDT7241 3174.8 76.4 36.3 109.9 89 197.7 29.6
90MW5319 3169.9 65.4 45.1 107.7 71.6 168.5 30.2
12MW6302 3099.9 76.4 35.3 108.9 89 146.8 26.2

2004MW6197 2594.4 74.7 37.7 109.6 70.6 176.5 32.2

99MW4047 2589.2 73.1 38.9 109.2 66.9 146.8 30

45
 14MWLSDT7310 2579.1 76.2 38.5 111.9 69.4 218.4 27.5
14MWLSDT7402 2540.4 74.5 38.5 110.2 69.3 238.2 31.9
14MWLSDT7410 2435 76.6 36.5 110.3 69.3 192.2 27.3
14MWLSDT7325 2408.7 79.1 34.9 111.1 70.1 212 30.2
2003MW6038 2397 73 39.3 109.5 64 191.2 32.1
2001MS7036 2382.1 69.7 36.9 109.9 75.5 194.8 33.1
12MW6243 2380.7 75.3 36.5 109 65.8 125 25.9
2523 2380 67.1 44.3 108.6 54.1 160.2 30
2003MW6053 2368.3 70.8 41.3 109.3 58.4 171 30
14MWLSDT7115 2354.5 78.2 35.9 111.3 72.3 204.7 28
14MWLSDT7395 2343.4 69 44.9 111.1 54.6 160.3 29.1
13MWF6#6077 2086.8 76.7 35.9 109.9 60.6 129.5 21.4
14MWLSDT7191 2024.6 79.8 36.5 113.5 57.5 174.3 30.3
Check 2885.1 75.1 36.9 109.1 80.7 159.3 30.9
Maximum 3533.1 79.8 47.8 113.6 101.4 238.2 34
Minimum 2024.6 64.6 34.9 107.5 54.1 125 21.4
Mean 2863.9 74.1 38.7 109.9 76.8 185.5 30.1
LSD at 5% 504.9 2.6 3.9 2.8 21 18.4 3.5
CV (%) 5.6 3.8 11.3 2.8 11.7 3.5 5.4

Where DTF = Date of flowering, DTM = Date of maturing, GFP = Grain filling period, GFR
= Grain filling rate, PHT = Plant height, GY = Grain yield and TSW = 1000 seed weight.
4.3. Stability analysis

4.3.1. Eberhart and Russel's Regression Model

The analysis of Variance by Eberhart and Russel's joint regression model of early maturing
advanced sorghum genotypes on mean grain yield (kg/ha) tested at five locations is presented
in Table 8. The result of Eberhart and Russell’s ANOVA showed highly significant (P≤
0.001) difference among the genotypes for grain yield, indicating differences in the
performance of genotypes, which may attribute to the genetic make up of the materials. The
GXE (linear) interaction was not significant (P>0.05). Thus, the GXE interaction was non-
linear type and showed the non-existence of genetic differences among genotypes for their
response to various locations, which is in agreement with earlier findings of Kenga et al.
(2003), Wedajo (2014), and Fekadu et al. (2009), Abiy (2016). Pooled deviations were very
highly significant against pooled error.

46
Pooled deviation (non-linear) in the joint regression analysis was very highly significant (p ≤
0.0001) difference (Table 8). The significance of the mean squares due to pooled deviations
from regressions revealed the performance of some genotypes were not stable across various
locations. This finding is similar to Abiy (2016). In addition, only 21 % of the GXE sum of
squares accounted by regression sum of square, and the remaining 79 % was accounted for
the SS of the regression deviation. This indicated that the largest proportion of the interaction
component of variation was explained by the deviation from regression. Hence, according to
Khan et al. (1988) and Ashraf et al. (2001), such differences in stability were due to deviation
from linear regression only. This means the variation in the yield performance of genotypes
are entirely unpredictable in nature.

Table 8: Genotype x Environment Interaction analysis of variance by Eberhart and Russel's

Model of early maturing sorghum genotypes on mean grain yield (kg/ha) tested at five
sites.

Sources of variation Sum Squares Mean Squares

Df
Total 299 782604108 2617405
Genotypes 59 38961387 660362**
Env + (Gen x Env) 240 743642721 3098511ns
Env (linear) 1 659944480 659944480ns
Gen xEnv (linear) 59 17443148 295647ns
Pooled deviation 180 66255093 368084***
Pooled error 600 5086911 8478

*** = very highly significant (P<0.0001), ** = highly significant (P<0.001), * = significant

(P<0.05) and ns = non-significant (P>0.05).
The stability parameters of Eberhart and Russell’s (1966) model for the yield of early
maturing sorghum genotypes tested at five locations is presented in table 9. According to this
model the genotype’s performance is described in terms of three parameters, mean yield,
regression coefficient and the deviation from the regression. Thus, a stable genotype is one
with high mean yield, bi=1, and S2di is small value.

47
The bi estimate of genotypes 14MWLSDT7410 (0.85), 12MW6420 (0.86), 12MW6444
(0.87), 14MWLSDT7332 (0.83), 04MW 6043 (0.84), Melkam (0.98), 12MW6302 (0.89),
14MWLSDT7400 (0.91), 14MWLSDT7325 (0.88), 14MWLSDT7115 (0.94), 12MW6146
(0.91), 12MW6243 (0.89), 12MW6444 (0.87), 14MWLSDT7042 (0.97) and 2005MI5064
(0.95), 2003MW6053 (0.94), 05MW6026, 14MWLSDT7207 are relatively near to unity. The
average yield performance of genotype 14MWLSDT7410, 12MW6420, 14MWLSDT7325,
12MW6146, 14MWLSDT7400, 12MW6243, and 2003MW6053 were below average (Table
9). Therefore, considering their above average mean grain yield, bi value closest to unity and
the S2di value is small, genotype Melkam, 05MW6026, 14MWLSDT7207, 2005MI5064,
12MW6302, 14MWLSDT7332, 12MW6444, 14MWLSDT7209 and 14MWLSDT7042 were
stable genotypes based on Eberhart and Russell’s model. However, the check (Melkam) and
14MWLSDT7209 genotypes were low yielder as compared to these selected stable
genotypes. Thus, genotypes 05MW6026, 14MWLSDT7207, 12MW6444, 12MW6302,
14MWLSDT7332 and 14MWLSDT7042 were the most stable and high yielder genotypes. In
contrary, the S2di value of genotype 14MWLSDT7060, 12MW6251, 14MWLSDT7310,
14MWLSDT7157,14MWLSDT7193,12MW6440, 14MWLSDT7201, 14MWLSDT7364,
Pipline2, 14MWLSDT7040, 14MWLSDT7036, 14MWLSDT7324, 14MWLSDT7234,
14MWLSDT7238, 14MWLSDT7291, 14MWLSDT7402, 14MWLSDT7033,
14MWLSDT7241, 14MWLSDT7191, 2003MW6038, 2005MI5093, and 2004MW6197 had
significantly deviation from regression (Table 9).

Wachira et al. (2002) grouped genotypes adaptability to specific environments based on their
estimate of bi as adaptable to high and low yielding environments. The bi values above one
define genotypes with higher sensitivity to environmental change (below average stability)
and are suitable to high yielding environments, whereas bi below one delivers a measurement
of more resistance to environmental change (above average stability), and are adaptable to
low yielding environments. Based on this idea the present study shows (Table 9), genotype
14MWLSDT7060, 12MW6251, 14MWLSDT7322, 14MWLSDT7196, 14MWLSDT7193,
14MWLSDT7176, 12MW6440, and 14MWLSDT7201, 14MWLSDT7364, Pipline 2,
14MWLSDT7413,13MWF6#6037, 14MWLSDT7040, 14MWLSDT7238, 14MWLSDT7234,
14MWLSDT7033, 14MWLSDT7241, 14MWLSDT7291, 90MW5319 had bi value of greater
than one and above mean yield performance in similar to Abiy and Firew (2016) reported that

48
eight sorghum genotypes had bi value of greater than one. Therefore, these genotypes were
highly contributed to the GXE and were suitable for conducive environments. In contrary,
genotype 2003MW6038, 14MWLSDT7410, 13MWF6#6077, 14MWLSDT7324,
14MWLSDT7325, 12MW6243, 14MWLSDT7402, 12MW6471, 14MWLSDT7191, 2523,
04MW6043, and 2003MW6053 had b_ivalue less than one and these genotypes were
contributed less to the GXE. Hence, these genotypes are suitable for harsh environments
(Table 9).

Table 9: Eberhart and Russell’s (1966) stability parameters of early maturing sorghum
genotypes tested at five locations.

Entry# Genotype Name bi S2di GY Rank

1 Melkam 0.975 240745 2885.09 28
2 14MWLSDT7060 1.1904 76071.9985* 3013.5 22
3 12MW6251 1.3746 52648.5568** 3284.55 8
4 14MWLSDT7410 0.8537 202975 2434.99 50
5 12MW6302 0.8854 249180 3099.92 15
6 14MWLSDT7322 1.2583 1036075 2937.86 26
7 14MWLSDT7395 1.0267 429131 2343.43 58
8 14MWLSDT7400 0.9089 640670 2690.96 40
9 14MWLSDT7310 1.0765 174810.832* 2579.09 48
10 13MWF6#6077 0.5125 404144 2086.77 59
11 14MWLSDT7325 0.8765 513248 2408.68 51
12 2005MI5069 1.0592 604144 2617.49 45
13 14MWLSDT7196 1.0095 860163 3494.64 3
14 14MWLSDT7311 1.1796 432721 2652.33 21
15 14MWLSDT7157 0.9478 161901.518* 3028.49 44
16 14MWLSDT7193 1.1664 211478.473* 2882.85 30
17 14MWLSDT7332 0.8319 145175 3249.17 10
18 14MWLSDT7115 0.9398 1438882 2354.53 57
19 14MWLSDT7176 1.1377 813739 3453.1 6
20 14MWLSDT7209 0.8449 360775 2884.52 29
21 12MW6440 1.0625 112372.72* 2868.93 31
22 14MWLSDT7201 1.2498 262459.621* 3343.65 7
23 12MW6146 0.9069 364183 2670.21 43
24 14MWLSDT7364 1.1225 81966.2491* 3034.68 20
25 Pipline 2 1.1647 127102.647* 3462.67 5
26 14MWLSDT7413 1.176 461902 2950.72 24
27 13MWF6#6037 1.0772 709836 3465.1 4

49
 28 14MWLSDT7207 0.812 179922 3050.93 19
29 14MWLSDT7040 1.1211 48725.9222* 3080.19 16
30 14MWLSDT7036 1.0547 28800.1309** 2685.27 41
31 14MWLSDT7324 0.9601 109177.745* 2847.04 32
32 12MW6243 0.89 212369 2380.68 54
33 12MW6420 0.8594 317889 2673.53 42
34 14MWLSDT7238 1.2476 130011.226* 3533.08 1
35 12MW6444 0.8681 536943 3052.16 18
36 14MWLSDT7402 0.869 19555.8851* 2540.45 49
37 14MWLSDT7234 1.0279 32651.5533* 3528.52 2
38 12MW6471 0.6455 1641709 2818.12 35
39 14MWLSDT7042 0.9733 190083 3200.93 11
40 14MWLSDT7033 1.12 186283.566* 3067.98 17
41 14MWLSDT7241 1.0565 23459.6288** 3174.82 13
42 14MWLSDT7191 0.8137 49118.5803* 2024.65 60
43 2005MI5093 1.1112 73272.0194* 2830.55 34
44 2401 1.0597 406132 2830.79 33
45 2004MW6197 1.1859 158557.494* 2594.45 46
46 2005MI5064 0.9451 1057996 2964.41 23
47 2523 0.6313 1131878 2380.03 55
48 04MW 6043 0.8369 1245803 3256.91 9
49 2005MI5057 0.9476 74065.6386* 3190.58 12
50 04MW 6079 1.0097 358134 2799 38
51 14MWLSDT7202 1.067 202160.958* 2803.42 37
52 14MWLSDT7291 1.0136 95315.8035* 2908.82 27
53 2001MS7036 1.0197 172813.299* 2382.12 53
54 90MW5319 1.1087 511415 3169.89 14
55 99MW4047 1.1251 30825.8423** 2589.24 47
56 05MW6026 0.8115 133843 2947.87 25
57 14MWLSDT7421 1.0224 118487.126* 2803.77 36
58 2003MW6053 0.9368 502830 2368.26 56
59 2294 1.2299 392500 2783.55 39
60 2003MW6038 0.8041 26489.915* 2397.01 52

Where bi = Coefficent of regression, S2di = deviation from regression, GY Mean Grain yield,
* = Significant (P<0.05) and ** = highly significant (P<0.001).

4.3.2. AMMI Model

The combined AMMI model ANOVA of the sixty early maturing advanced sorghum
genotypes over five sites for grain yield (kg/ha) is presented in Table 10. The ANOVA

50
indicated highly significant differences (p<0.001) for treatments (environments, genotypes
and GXE). The total variation explained (%) was 99.1 % for treatment and 0.9 % for error.
The larger contribution of the treatment than the error reveals the reliability of this multi-
location experiment (Table 10). The treatment variation was largely due to among
environments variation, genotype and GXE accounted 84.36 %, 4.86 % and 10.78% for the
treatment variation, respectively. As stated earlier, the great percentage of the location is an
indication that the major factor that affect yield performance of sorghum in drought
areas of Ethiopia is the environment. Various authors also reported similar results for other
sorghum genotypes tested at different locations (Asfaw (2007, 2008); Hagos and Fetien
(2011); Mahnaz et al. (2013); Sewagegne et al. (2013); Abiy and Firew (2016); Kinde (2016).
In the AMMI ANOVA, the GXE was further partitioned by PCA. The Gollob F-test used to
measure significant of the GXE components. The number of PCA axis to be retained is
determines by testing the mean square of each axis with the estimate of residual through the
F-statistics. The result of ANOVA showed that the first two IPCA are significant at 0.001
probability level, this outcome suggests the addition of the first two interactions PCA axes in
the model. Hence, the best fit AMMI model for this multi-location yield trial data was
AMMI-2 (Table 10).

In particular, the first IPCA captured 49.8 % of the total interaction sum of squares while the
second IPCA explained 26.65 % of the interaction sum of squares. Gauch and Zobel (1996)
and Yan et al. (2002) also proposed that the most accurate model for AMMI can be predicted
by using the first two IPCAs. In the present study the first two IPCAs accounted for a total of
76.45 % of the interaction with 122 of the corresponding degrees of freedoms. This indicates
that the GXE of the sixty sorghum genotypes with five locations was sufficiently predicted by
the first two principal components axes and therefore, most evidence may well to graphically
display in AMMI1 and AMMI2 biplot.

51
 Table 10: Genotype x Environment Interaction analysis of variance by AMMI for grain yield
(kg/ha) of early maturing advanced sorghum genotypes tested at five locations during
2017 main cropping season.

Source DF SS MS %Total %Treatment %GXE %Cumulative

Total 899 2376230888 2643193
Treatments 299 2353974003 7872823** 99.1
Genotypes 59 114373528 1938534** 4.86
Environment 4 1985932733 496483183** 84.36
Replications 10 460154 46015ns
Interactions 236 253667743 1074863** 10.78
IPCA 1 62 126333038 2037630** 49.8 49.8
IPCA 2 60 67611000 1126850** 26.65 76.45
Residuals 114 59723705 523892** 23.54 100
Pooled Error 590 21796731 36944 0.9

DF = degree of freedom, SS =sum of squares, MS = mean of squares, ** = highly significant

(P<0.001) and ns = non significant.

The AMMI 1 was used to analyse biplot graph (Fig. 2) using individual environments and
mean grain yield performances of sorghum genotypes. As indicated in Figure 2, each
environment and genotype main effect was plotted along the abscissa against their respective
IPCA1 score as ordinate. The AMMI 1 biplot, showing main effects means of grain yield on
the abscissa and IPCA 1 values as the ordinates, genotypes or environments that appear on a
vertical line have similar means and those that appear on a horizontal line have similar
interaction (similar IPCA 1 value) patterns (Crossa et al., 1990) but they have not similar
means. According to these authors, genotypes or environments with large IPCA 1 scores
(either positive or negative) have high interactions, whereas genotypes or environments with
IPCA 1 scores near zero have small interactions. Similarly, in the biplot (Fig. 2) three of the
five environments had below average main effects and were unsuitable to the performance of
half of the total genotypes as compared to the rest two environments. The remaining two
environments (Shaorobit and Sheraro) had the highest main effects and were favorable to the
performance of almost half of the genotypes. Similar outcomes have reported by Das et al.
(2010) and Kulsum et al. (2013). On the contrary, environments Mieso, Erer and Kobo were

52
the most unsuitable environments because they had below average main effects. In general
Shaorobit showed higher main effect values whereas Erer and Kobo showed below average
main effect values. There were high interaction at Kobo due to high IPCA 1 score value and
very small interaction (small IPCA 1 score value) at Mieso. Inconsistency in interaction at
Kobo present difficulty in producing variety recommendation for that target location. Entries
21, 55, 57, 30, 17, 39, 7, 37, 56, 28, 36, 60, 45, 20 and environments Mieso and Sheraro were
least interactive due to low IPCA 1 score value. Similar findings and interpretation have been
made by Adugna (2007); Anandan et al. (2009) and Islam et al. (2014). Entries 21, 17, 39, 30,
56, 28, 37 and 55 placed closed to the biplot origin and were the most stable but 55 and 30
had below grand mean. Similar result was reported by Anandan et al. (2009); Crossa, (1990)
and Kempton R A, (1984). Entries 39, 17, 37 21, 56 and 28 had above average main effect
and small interaction which makes them the most stable genotypes. On the contrary, entries
19, 27, 13 and 22 had similar and large main effects but 19 showed larger interaction due to
high IPCA 1 score and more unstable genotype (Fig. 2).

53
Figure 1: AMMI 1 biplot for grain yield of 60 early maturing sorghum genotypes evaluated
across five locations in Ethiopia during 2017 main cropping season.

AMMI 2 biplot presents the pattern of the first two IPCA of the interaction effects and helps
in visual interpretation of the G x E interaction patterns and identify genotypes or locations
that show small and large interaction effects. In AMMI 2 biplot, environments fell into three
sections (Fig 3). Among the environments 17MS (Mieso) had very short spoke and 17SH
(Sheraro) had short spokes. They do not exert strong interaction but the environments 17ER
(Erer), 17KB (Kobo) and 17SR (Shoarobit) had long spokes and hence indicate the most
enfluencing environments. In AMMI 2 biplot, the entries, 38, 47, 19, 3 and 6 are the best or
poorest genotypes in some or all environments because they are farthest from the origin where
as the best genotype is 6 with respect to the best enhancing environment 17SR and the poor
entry is 47 due to its value below average value. On the other hand, the entries 37, 56, 28, 30,
25, 21, 39, 57, 55, 29, 41, 52, 36, 31 and 17 were close to the origin and therefore were

54
 less/non-sensitive to environmental interaction. However, entries 57, 55 and 31 were low
yielding due to below average grain yield (Fig 3). Hence, entries 28, 56 and 37 were the most
yielding and stable due to very close to the origin. Similar outcomes were reported by
Sintayehu (2017); Dagnachew et al. (2014); Anandan et al. (2009); Crossa, (1990) and
Kempton R A, (1984).

IPCA 2-
26.65%

IPCA 1-49.82%

Figure 2: AMMI 2 biplot for grain yield (kg/ha) showing the interaction of IPCA2 against
IPCA1 scores of 60 early maturing sorghum genotypes in five environments.

Where 1 = Melkam , 2 = 14MWLSDT7060, 3 = 12MW6251, 4 = 14MWLSDT7410, 5 =

12MW6302, 6 = 14MWLSDT7322, 7 = 14MWLSDT7395, 8 = 14MWLSDT7400, 9 =
14MWLSDT7310, 10 = 13MWF6#6077, 11 = 14MWLSDT7325, 12 = 2005MI5069, 13 =
14MWLSDT7196,14 = 14MWLSDT7311,15 = 14MWLSDT7157, 16 = 14MWLSDT7193,
17 = 14MWLSDT7332,18 = 14MWLSDT7115,19 = 14MWLSDT7176,20 =
14MWLSDT7209,21 = 12MW6440,22 = 14MWLSDT7201,23 = 12MW6146,24 =
14MWLSDT7364, 25 = Pipeline 2, 26 = 14MWLSDT7413, 27 = 13MWF6#6037, 28 =
14MWLSDT7207,29 = 14MWLSDT7040, 30 = 14MWLSDT7036, 31 = 14MWLSDT7324,
32 = 12MW6243, 33 = 12MW6420,34 = 14MWLSDT7238,35= 12MW6444,36 =
14MWLSDT7402,37 = 14MWLSDT7234, 38 = 12MW6471, 39 = 14MWLSDT7042,40 =

55
14MWLSDT7033, 41 = 14MWLSDT7241,42 = 14MWLSDT7191, 43 = 2005MI5093, 44 =
2401, 45 = 2004MW6197, 46 = 2005MI5064,47 = 2523,48 = 04MW 6043,49 = 2005MI5057,
50 = 04MW 6079, 51 = 14MWLSDT7202, 52 = 14MWLSDT7291, 53 = 2001MS7036, 54 =
90MW5319,55 = 99MW4047,56 = 05MW6026,57 = 14MWLSDT7421, 58 = 2003MW6053,
59 = 2294, 60 = 2003MW6038 and 17ER = Erer, 17MS = Mieso, 17KB = Kobo, 17SR =
Shaorobit, 17SH = Sheraro.

4.3.3. GGE biplot model

Stability analysis of the tested genotypes (represented by entry number) based on their IPCA
scores using the GGE biplot of analysis is presented in Fig 4 and Fig 5. The polygon of lines
in Fig. 4 is formed by connecting vertex genotypes, by connecting straight lines and rest of
genotypes (represented by entry number) fall inside the polygon. The vertex entries were 19,
13, 34, 6, 18, 42, 10 and 47 (Fig 4). These entries are either the best or poorest genotypes in
some or all environments because they are farthest from the origin (Yan and Kang, 2003).

In the present study, the GGE biplot graphic analysis of the sixty early maturing sorghum
genotypes tested at five locations revealed that the first two principal components explained
71.30% of the total variance (Fig 4). This GGE biplot is used to facilitate visual analysis of
the genotype by location data. Genotypes close to the origin are not sensitive (unfavorable) to
the environments and those distant from the origin are sensitive (favorable) to environments
and have large interactions. Accordingly, statistically stable genotypes and locations were
located near to the biplot origin, with scores practically zero for the two interaction axis
(IPCA1 and IPCA2). The entires, which lie near the origin and practically stable were entries
28, 56, 57, 20 and 21and had wide adaptability. However, entry 57 was below average yield.
Hence, entries 28, 56, 20 and 21 were high yielding and stable across the five locations. On
the other hand entries 19, 13, 34, 6, 18, 42, 10, 47, 27, 22, 46 and 8 were located far away
from the origin, which were more sensitive to environment changes and are considered as
specifically adapted genotyypes in agreement with the finding of Abay and Bjornstad (2009);
Dehghani et al. (2006); Tesfaye et al. (2008) and Gasura et al. (2015) reported high yielder
and stable genotype as well as low yielding and poorly stable ones.

The graphic analysis of the first principal component (IPCA1) represents variety productivity,
and the second principal component (IPCA2) variety stability (Yan et al., 2000). Therefore,

56
the GGE biplot revealed that the ideal genotype must have a high IPCA1 value (high
productivity) and an IPCA2 value near to zero (more stable). Thus based on the graphic
interpretation, the present study revealed that entry 34 had the largest IPCA1 score, and hence
had the highest average grain yield. The IPCA2 value of this genotype, however, was the
largest indicating its specific adaptation. In contrast, 57 yielded poorly at all sites, but was
relatively stable, as indicated by its small IPCA1 scores (low yielding) and relatively small
IPCA2 scores (stable). Genotypes that had IPCA1 scores >0 were identified as higher yielding
and those that had IPCA1 scores <0 were identified as lower yielding. Accordingly, the
average yield of 42, 10, 18 and 47 were below average, and the lower yielding genotypes due
to highly unstable (large absolute IPCA2 scores). In contrast, entries 19, 27, 13, 34, 22 and 48
had positive IPCA1 scores and were identified as high yielder genotypes.

Figure 5 revealed the relationship among genotypes in biplot analysis, hence; the biplot
revealed that 19, 27 and 13 were similar, because the angle between them was smaller. In
contrast, genotype 19 and 46 were dissimilar, as their angle was larger. In addition, the
positions of the genotypes in opposing quadrants on the Cartesian plan also revealed their
dissimilar genetic performance, as observed for 19 and 46, 19 and 6, 27 and 46, 27 and 6, 13
and 46, 13 and 6, 34 and 10, 34 and 42. Moreover, the relationship among testers was also
graphically described in the GGE biplot (Fig 5). The cosines of the angle between the testers
normally estimated the correlation coefficient between them. Hence, the pair of testers, which
were positively correlated had an angle between their vectors less than 90o (17ER and 17SR,
17ER and 17MS, 17MS and 17SH, 17SR and 17MS, 17SR and 17SH, 17SH and 17KB);
while the angle between vectors of tester 17ER and 17KB was approximately 90o, and were
not correlated (Fig 5).

With respect to the contribution of testing locations to the GXE, location mieso (17MS) had
least contribution as it lied closest to the origin, but locations Shaorobit (17SR) and Kobo
(17KB) were made the highest contribution. The biplot in this trail also indicated that entry
13, 27, 19 and 40 were performed above average in location Kobo (17KB); while 34, 37,
22,25,3,48,24,49,17,21,41,39 and 2 in 17SR (Shoarobit), 17MS (Mieso), 17ER (Erer) and
17SH (Sheraro). Meanwhile, discriminating ability was an important measure of a test
environment. Another equally important measure of a test environment was its

57
representativeness of the target environment. An ideal environment should be highly
differentiating of the genotypes and at the same time representative of the target environment
(Yan et al.2000 and Yan etal.2006). Hence in the current experiment, locations 17SR
(Shaorobit) and 17KB (Kobo) were most discriminating of the entries as indicated by the
longer distance between their markers and the origin. However, due to their relatively large
IPCA2 score, genotypic differences observed at both locations did not exactly reflect the
genotypes in average yield over all sites. On the other hand, location Mieso (17MS) was not
actually the most discriminating as distance of its vector was the smallest, but varietal
difference at this site was highly consistent with those averaged over all sites, because it had
the lowest IPCA2 scores and IPCA1 scores. In agreement with this finding, Abay and
Bjornstad (2009); Dehghani et al. (2006); Tesfaye et al. (2008) and Gasura et al. (2015)
reported the existence of a good testing environment for selecting widely adaptable and high
yielding cultivar.

To clearly display the ‘which-won-where’ pattern and sensitivity degree between the
genotype and environment, polygon view of a GGE biplot based on the IPCA1 and IPCA2
values was displayed in Fig 5. The perpendicular genotypes in the biplot have divided the
biplot in to 5 sectors in which each location fell in either of the sectors. Yan and Kang (2003)
explained that the polygon view of a biplot was the best way to visualize the interaction
patterns between genotypes and environments, and to effectively interpret a biplot. In this
study, the ‘which won where’ feature of the biplot identified wining genotypes; i.e., entry 13
for instance was the winning/corner genotype in locations 17SH (Sheraro) and 17KB (Kobo);
entry 25 in locations 17MS (Mieso) and 17SH (Sheraro); entry 22 in locations 17ER (Erer)
and 17SR (Sheraro). Similarly, entry 34 was the vertex/winning genotype in location 17SR
and 17MS. According to the findings of Yan and Tinker (2006), the vertex genotypes were
the most responsive genotypes, as they have the longest distance from the origin in their
direction. On the other hand, the result also showed some genotypes, which fall in sectors
where there were no locations at all; these genotypes are poorly adapted to five of the testing
locations (18, 46, 35,38, 11, 9, 31, 52, 36, 4, 42, 10, 32, 58, 47, 45, 50, 44, 24, 7 and 33).

Yan et al. (2000) defined mega environments as a cluster of locations or environments that
constantly share the same best variety. Hence, the result of this experiment identified two
58
Sorghum growing mega-environments. The first mega-environment contained locations of
(Erer) 17ER, (Shoarobit) 17SR, (Mieso) 17MS and (Sheraro) 17SH with winning entry 34;
while the second mega-environment on the other hand, contained location (Kobo) 17KB with
winning entry 19.

Figure 3: The ‘which-won-where’ feature of the biplot.

59
IPCA 2 -
33.01 %

IPCA 1- 38.29 %

Figure 4: The relationship among testers and mega environments.

Where 1 = Melkam , 2 = 14MWLSDT7060, 3 = 12MW6251, 4 = 14MWLSDT7410, 5 =

12MW6302, 6 = 14MWLSDT7322, 7 = 14MWLSDT7395, 8 = 14MWLSDT7400, 9 =
14MWLSDT7310, 10 = 13MWF6#6077, 11 = 14MWLSDT7325, 12 = 2005MI5069, 13 =
14MWLSDT7196,14 = 14MWLSDT7311,15 = 14MWLSDT7157, 16 = 14MWLSDT7193,
17 = 14MWLSDT7332,18 = 14MWLSDT7115,19 = 14MWLSDT7176,20 =
14MWLSDT7209,21 = 12MW6440,22 = 14MWLSDT7201,23 = 12MW6146,24 =
14MWLSDT7364, 25 = Pipeline 2, 26 = 14MWLSDT7413, 27 = 13MWF6#6037, 28 =
14MWLSDT7207,29 = 14MWLSDT7040, 30 = 14MWLSDT7036, 31 = 14MWLSDT7324,

60
32 = 12MW6243, 33 = 12MW6420,34 = 14MWLSDT7238,35= 12MW6444,36 =
14MWLSDT7402,37 = 14MWLSDT7234, 38 = 12MW6471, 39 = 14MWLSDT7042,40 =
14MWLSDT7033, 41 = 14MWLSDT7241,42 = 14MWLSDT7191, 43 = 2005MI5093, 44 =
2401, 45 = 2004MW6197, 46 = 2005MI5064,47 = 2523,48 = 04MW 6043,49 = 2005MI5057,
50 = 04MW 6079, 51 = 14MWLSDT7202, 52 = 14MWLSDT7291, 53 = 2001MS7036, 54 =
90MW5319,55 = 99MW4047,56 = 05MW6026,57 = 14MWLSDT7421, 58 = 2003MW6053,
59 = 2294, 60 = 2003MW6038 and 17ER = Erer, 17MS = Mieso, 17KB = Kobo, 17SR =
Shaorobit,17SH=Sheraro.

Comparison Biplot of five test environments: The average environments coordinate (AEC) is
a line that pass through the average environment (represented by small circle) and biplot
origin. A test environment that has a small angle with the AEC is more representative of other
test environments (Yan et al., 2000 and Yan et al., 2006). Thus, 17SR (Shoarobit) was more
representative testing location (Figure 6). In agreement with this finding Gasura et al., (2015)
reported the existence of a good testing environment for high yielding cultivar.

Figure 5: GGE-biplot showing a comparison of five testing environments with in ideal

environment for grain yield (kg/ha).

61
An ideal genotype should have both high mean grain yield performance across environments.
It is one which is near or at the centre of the concentric circle, and is also a genotype to be on
average environmental coordinate (AEC) on positive direction and has vector length equal to
the longest vector of the genotype and indicated by an arrow pointed to it (Yan et al., 2006
and Kaya et al., 2006). The Biplot indicated that genotype 34 is the most ideal genotypes,
where as 37, 25, 3, 22, 41, 17 and 39 were nearest to the ideal genotype (the center of
concentric circle). Therefore, these genotypes are more required and ideal genotypes than
other tested genotypes. In line with this finding (Gasura et al., 2015) found the presence of
that ideal genotype.

Figure 6: GGE-biplot showing a comparison of all genotypes with in ideal genotypes for grain
yield (kg/ha).

62
4.3.4. AMMI stability value

In additive main effect and multiplicative interaction stability analysis (ASV) method, a genotype with lowest ASV score is the most
stable across diverse environments and the higher the ASV value (either negative or positive) the more specifically adapted a genotype
is to specific environments (Purchase, 1997). Table 11 shows ASV for each genotype and the ranks of the genotypes according to
their ASV values. The results revealed that from the tested early maturing sorghum genotypes fifty one of them had ASV of below
one.AccordinglyMelkam,14MWLSDT7060,12MW6251,14MWLSDT7410,12MW6302,14MWLSDT7395,14MWLSDT7310,13MW
F6#6077,14MWLSDT7325,2005MI5069,14MWLSDT7196, 14MWLSDT7311, 14MWLSDT7, 14MWLSDT7193, 14MWLSDT733
2,14MWLSDT7209,12MW6440,14MWLSDT7201,14MWLSDT7364,Pipline2,14MWLSDT7413,13MWF6#6037,14MWLSDT7207,
14MWLSDT7040,14MWLSDT7036,14MWLSDT7324,12MW6243,12MW6420,14MWLSDT7238,12MW6444,14MWLSDT7402,1
4MWLSDT7234,14MWLSDT7042,14MWLSDT7033,14MWLSDT7241,14MWLSDT7191,2005MI5093,2401,2004MW6197,2005
MI5057,04MW6079,14MWLSDT7202,14MWLSDT7291,2001MS7036,90MW5319,99MW4047,05MW6026,14MWLSDT7421,200
3MW6053,2294,2003MW6038, were relatively widely stable (Table 11).

In contrary, due to their large ASV genotype 14MWLSDT7322, 14MWLSDT7400, 14MWLSDT7115, 14MWLSDT7176,
12MW6146, 12MW6471, 2005MI5064, 2523 and 04MW 6043 were the most unstable genotypes (Table 11). The mean yield of
genotypes is also considered for selection of genotypes as a high yielder and stable genotypes. Among the selected widely stable early
maturing genotypes the mean yield of twenty seven genotypes are above the grand mean. Therefore, based on mean yield and ASV,
genotypeMelkam,14MWLSDT7060, 12MW6251, 12MW6302, 14MWLSDT7196, 14MWLSDT7157, 14MWLSDT7193,14MWLSD
T7332,14MWLSDT7209, 12MW6440, 14MWLSDT7201, 14MWLSDT7364, Pipline 2, 14MWLSDT7413, 13MWF6#6037, 14MWL
SDT7207, 14MWLSDT7040, 14MWLSDT7238,12MW6444,14MWLSDT7234,14MWLSDT7042,14MWLSDT7033,14MWLSDT72
41,2005MI50714MWLSDT7291, 90MW5319 and 05MW6026 are relatively high yielder and widely stable genotypes while the

57
remaining twenty four genotypes are below the grand mean due to this they are low yielder and not selected as stable. The six most
stable and high yielde early maturing sorghum genotypes obtained using this model were genotype
13MWF6#6037, 05MW6026, 14MWLSDT7207 ,14MWLSDT7234 , 14MWLSDT7042 and 12MW6440 (Table 11).
Table 11: IPCA1 and IPCA 2 scores; and ASV for the sixty early maturing sorghum genotypes sorted on ASV evaluated at five
locations during 2017 main cropping season.

Entry# Genotype Name GY (kg/ha) GY Rank IPCA 1 Score IPCA 2 Score ASV ASV Rank
27 13MWF6#6037 3465.1 4 0.5295 0.0861 0.0073 1
30 14MWLSDT7036 2685.27 41 0.0298 0.0648 0.0599 2
55 99MW4047 2589.24 47 -0.011 -0.2288 0.0735 3
56 05MW6026 2947.87 25 -0.049 -0.0007 0.091 4
28 14MWLSDT7207 3050.93 19 -0.0487 -0.0233 0.0916 5
37 14MWLSDT7234 3528.52 2 -0.054 0.0217 0.1009 6
39 14MWLSDT7042 3200.93 11 -0.023 0.274 0.1178 7
21 12MW6440 2868.93 31 0.0109 0.1171 0.1375 8
57 14MWLSDT7421 2803.77 36 -0.052 -0.2189 0.1459 9
4 14MWLSDT7410 2434.99 50 -0.164 -0.0592 0.155 10
2 14MWLSDT7060 3013.5 22 0.0727 -0.1466 0.1573 11
36 14MWLSDT7402 2540.45 49 -0.076 0.157 0.1666 12
41 14MWLSDT7241 3174.82 13 0.0891 -0.0697 0.1713 13
60 2003MW6038 2397.01 52 -0.083 0.1359 0.1746 14
17 14MWLSDT7332 3249.17 10 0.0151 0.3894 0.1798 15

58
7 14MWLSDT7395 2343.43 58 -0.043 -0.361 0.2102 16
29 14MWLSDT7040 3080.19 16 0.1148 -0.0436 0.2164 17
45 2004MW6197 2594.45 46 0.0523 -0.3767 0.2397 18
43 2005MI5093 2830.55 34 0.12 -0.1736 0.2543 19
25 Pipline 2 3462.67 5 0.0653 -0.3671 0.2568 20
20 14MWLSDT7209 2884.52 29 -0.067 0.3706 0.2627 21
24 14MWLSDT7364 3034.68 20 -0.113 -0.2562 0.2775 22
53 2001MS7036 2382.12 53 0.1122 -0.2763 0.286 23
50 04MW 6079 2799 38 -0.152 -0.0832 0.2909 24
34 14MWLSDT7238 3533.08 1 -0.138 -0.2093 0.3015 25
5 12MW6302 3099.92 15 0.0807 0.3983 0.3095 26
42 14MWLSDT7191 2024.65 60 -0.148 0.2156 0.3234 27
49 2005MI5057 3190.58 12 -0.178 0.1166 0.3458 28
52 14MWLSDT7291 2908.82 27 -0.195 -0.0825 0.3709 29
3 12MW6251 3284.55 8 0.0742 -0.501 0.3896 30
44 2401 2830.79 33 -0.127 -0.4071 0.4023 31
33 12MW6420 2673.53 42 0.1862 0.259 0.415 32
31 14MWLSDT7324 2847.04 32 -0.224 0.0018 0.4186 33
15 14MWLSDT7157 3028.49 44 -0.247 -0.0463 0.4633 34
9 14MWLSDT7310 2579.09 48 -0.277 -0.0263 0.5187 35
1 Melkam 2885.09 28 0.2653 0.2019 0.5365 36

59
51 14MWLSDT7202 2803.42 37 0.2858 -0.1505 0.5566 37
40 14MWLSDT7033 3067.98 17 0.2825 -0.1764 0.5591 38
32 12MW6243 2380.68 54 0.3084 0.0863 0.5836 39
16 14MWLSDT7193 2882.85 30 0.305 -0.1885 0.6055 40
26 14MWLSDT7413 2950.72 24 0.3381 -0.1374 0.6506 41
22 14MWLSDT7201 3343.65 7 -0.328 -0.2135 0.6588 42
59 2294 2783.55 39 0.3492 -0.3608 0.7827 43
11 14MWLSDT7325 2408.68 51 -0.44 0.006 0.8222 44
14 14MWLSDT7311 2652.33 21 -0.442 -0.016 0.8268 45
58 2003MW6053 2368.26 56 0.4762 0.0405 0.8915 46
54 90MW5319 3169.89 14 0.451 -0.263 0.9119 47
12 2005MI5069 2617.49 45 -0.451 -0.3249 0.9487 48
35 12MW6444 3052.16 18 -0.502 0.1086 0.9501 49
10 13MWF6#6077 2086.77 59 0.2831 0.6555 0.9586 50
13 14MWLSDT7196 3494.64 3 0.5006 0.2322 0.9892 51
8 14MWLSDT7400 2690.96 40 0.5376 0.0629 1.0084 52
23 12MW6146 2670.21 43 0.3848 0.2066 1.1322 53
19 14MWLSDT7176 3453.1 6 0.6155 -0.164 1.177 54
48 04MW 6043 3256.91 9 -0.402 0.6738 1.2054 55
46 2005MI5064 2964.41 23 -0.674 -0.0712 1.2652 56
6 14MWLSDT7322 2937.86 26 -0.684 -0.2423 1.3361 57

60
 47 2523 2380.03 55 0.7144 0.3957 1.4914 58
18 14MWLSDT7115 2354.53 57 -0.813 -0.012 1.5191 59
38 12MW6471 2818.12 35 -0.443 1 1.8276 60
Mean 2863.9

Where GY = Grain yield by AMMI, ASV = AMMI Stability value, IPCA 1 Score = Interaction prinicipal component analysis Score 1
and IPAC 2 Score = Interaction prinicipal component analysis Score 2.

61
 5. SUMMERY AND CONCLUSION

Sorghum (Sorghum bicolor) is one of the major important cereal crops which the lives of
millions of people depend and grows in areas where other major cereals marginally grown.
However, drought becomes the major problem for sorghum production in Ethiopia especially
in drought prone areas of the country which needs the development of drought tolerant or
resistance varieties. The yield performance of crop varieties is also highly affected by
genotype x environment interaction which is the major concern to plant breeders while
developing improved varieties. In view of this, the yield performance of 60 early maturing
advanced sorghum genotypes were tested for further confirmation with the objectives of
estimating the magnitude of GXE for grain yield and other traits and to identify high yielder
and stable genotypes across locations.

The experiment was conducted at five locations during 2017 main cropping season. The
experiment was carried out using RCBD with Row Column arrangement and three
replications. Recommended fertilizer application and agronomic practices were followed at
each location. Data on grain yield, phenological and other traits were collected from all
locations. The data collected were subjected to spatial analysis and linear mixed model for
individual and across locations and stability analysis using various models.

In spatial analysis, plots that are located together in the field are more likely to have similar
yield due to shared micro-environment. The spatial analysis for each location revealed that the
genotypes were significantly different for grain yield. This showed that there were genetic
difference among genotypes for grain yield at each location. The REML combined analysis of
variance across locations showed very highly significant (P<0.0001) difference among
locations, significant (P<0.05) among genotypes and highly significant (P<0.0001) among
interactions for most of the traits studied. Significant component of variation for locations and
genotype indicated variation in the performance of genotypes for grain yield, phenological
and other traits in different environments. On the other hand, significant GXE interaction
showed inconsistency in the performance of sorghum genotypes across locations, that is, the

62
relative performances of genotypes were significantly affected by the varying environmental
conditions. The partitioning of the combined analysis of variance component for grain yield
indicated that environments, genotypes and G x E interaction contributed 86.4 %, 1.9 % and
10.9 %, respectively, for the total variance. Larger proportion of environment variance
indicated the larger effect of environment on yield performance of sorghum genotypes in the
test environment than other variances components.

The mean yield of genotypes across location was 2863.9 kg/ha; with genotype means grain
yield ranged from 2024.6 kg/ha to 3533.1kg/ha, indicating wide difference in yield potential
across locations. The genotype with the lowest mean grain yield was 14MWLSDT7191
(2024.6 kg/ha). The performance of the standard check (Melkam = 2885.1kg/ha) was
statistically different to the top fifteen high yielding genotypes but higher than the grand mean
(2863.95 kg/ha). Shaorobit and Sheraro were the highest yielding environments with mean
values of 5174.0 kg/ha and 3706.7 kg/ha, respectively, indicate the two environments are
suitable for sorghum production, whereas Mieso, Erer and Kobo were the poorest yielding
environments with mean grain yields of 1099.0 kg/ha, 1508.5 kg/ha and 2832.1 kg/ha,
respectively.

Eberhart and Russell’s joint regression model, AMMI, GGE biplot model and ASV models
were used for grain yield to identify superior and relatively stable genotypes across location.
Eberhart and Russell’s joint regression ANOVA showed that the performance of genotypes
for grain yield was statistically different. The GXE (linear) was not significant and the pooled
deviation was significant. The interaction sum square was accounted largely by the pooled
deviation (79 %) and only 21 % by the GXE (linear).

The combined AMMI ANOVA showed significant differences among genotypes and the
presence of interaction effect. In this study including the IPCA residual both the first two
IPCAs were significant. For the total variation the treatment variation accounted about 99.1 %
%, and for the treatment variation was attributed to genotype variation 4.86 %, location
variation 84.36 %% and interaction 10.78%. In addition 76.46 % of the interaction effect was
explained by the first two IPCAs.

63
Based on Eberhart and Russell’s stability analysis, considering their above average mean
grain yield, bi value closest to unity and the S2di value is small, genotype Melkam,
05MW6026, 14MWLSDT7207, 2005MI5064, 12MW6302, 14MWLSDT7332, 12MW6444,
and 14MWLSDT7042 were stable genotypes. However, the check (Melkam) genotype was
low yielder as compared to these selected stable genotypes. Thus, genotypes 05MW6026,
14MWLSDT7207, 14MWLSDT7332, 12MW6444, 12MW6302 and 14MWLSDT7042
were the most stable and high yielder genotypes. Genotype 14MWLSDT7060, 12MW6251,
14MWLSDT7322, 14MWLSDT7196, 14MWLSDT7193, 14MWLSDT7176, 12MW6440,
and 14MWLSDT7201, 14MWLSDT7364, Pipline 2, 14MWLSDT7413,13MWF6#6037,
14MWLSDT7040, 14MWLSDT7238, 14MWLSDT7234, 14MWLSDT7033,
14MWLSDT7241, 14MWLSDT7291, 90MW5319 had value of greater than one and
above mean yield performance. Therefore, these genotypes were highly contributed to the
GXE and were suitable for conducive environments. In contrary, genotype 2003MW6038,
14MWLSDT7410, 13MWF6#6077, 14MWLSDT7324, 14MWLSDT7325, 12MW6243,
14MWLSDT7402, 12MW6471, 14MWLSDT7191, 2523, and 2003MW6053 had value less
than one and these genotypes were contributed less to the GXE. Hence, these genotypes are
suitable for harsh environments.

AMMI 2 biplot helps in visual interpretation of the G x E interaction patterns and identify
genotypes or locations that show small and large interaction effects. Accordingly,
environments Erer, Kobo and Shaorobit were the most enfluencing environments due to large
interaction while Mieso and Sheraro had less enfluencing environments. In AMMI 2 biplot,
the genotypes, 12MW6471, 2523, 14MWLSDT7176, 12MW6251, 14MWLSDT7322 had
large interactions due to large IPCA 1 and IPCA 2, and are the best or poorest genotypes in
some or all environments because they are farthest from the origin where as the best genotype
is 14MWLSDT7322 with respect to the best enhancing environment Shaorobit and the poor
genotype is 2523 due to its value below average value. On the other hand, the genotypes
14MWLSDT7234, 05MW6026, 14MWLSDT7207, 14MWLSDT7036, Pipline 2, 12MW6440
,14MWLSDT7042,14MWLSDT7421,99MW4047,14MWLSDT7040,14MWLSDT7241,
14MWLSDT7291, 14MWLSDT7402, 14MWLSDT7324 and 14MWLSDT7332 were close
to the origin and therefore were less/non-sensitive to environmental interaction. However,

64
genotypes 14MWLSDT7421, 99MW4047 and 14MWLSDT7324 were below average yield.
Genotypes 14MWLSDT7207, 05MW6026 and 14MWLSDT7234 were the most yielding and
stable due to very close to the origin.

GGE biplot is the best method as compared to AMMI biplot to show graphically the ‘which-
won where’ and the relationship between testers and mega environments. Therefore, based on
the overall mean grain yield and stability of this multi-location trial, genotype
14MWLSDT7238 could be recommende to locations Erer, Shaorobit, Mieso and Sheraro
areas that share similar characteristics. Likewise, genotype 14MWLSDT7176 could be
recommended to location Kobo. Genotypes 14MWLSDT7207, 05MW6026,
14MWLSDT7209 and 12MW6440 fortunately, could be grown in all testing locations due to
the high yielding and stable genotypes. However, genotype 14MWLSDT7421 was the lower
yielding and stable genotype.

Additive Main Effects and Multiplicative Interaction stability value (ASV) was one of the
stability models to identify the stable genotype for this study. Accordingly, The six most
stable and high yielder early maturing sorghum genotypes on this model were genotype
13MWF6#6037, 05MW6026, 14MWLSDT7207, 14MWLSDT7234, 14MWLSDT7042 and
12MW6440.

The results of genotype x environment interaction and stability analysis methods indicated
that, Eberhart and Russell’s stability analysis, AMMI 2 biplot, GGE biplot and ASV analysis
identified two early maturing sorghum genotypes 05MW6026 and 14MWLSDT7207 that had
a high mean performance and high stability for grain yield. Therefore, genotype 05MW6026
and 14MWLSDT7207 can be recommended as a candidate for releasing over a wide range of
locations of the lowland Ethiopia.

In general, results from this study gave valuable information for future studies related to
allocation of early maturing sorghum genotypes to different growing conditions in the dry
lowlands areas of Ethiopia. The sorghum growing dry lowland areas of Ethiopian were
various and contributed largely to the changes of genotypes yield performance over locations.
Therefore, further study on the GXE effects and stability of early maturing sorghum
genotypes is needed in multi locations for a number of years and location to determine the

65
interaction effect of genotypes and select stable genotypes. Since the current study was
conducted only for one year, the work should be repeated at least for some more years to give
sound conclusions and reliable recommendations.

66
 REFERENCES

Abay Fetien and A. Bjornstad, 2009. Specific adaptation of barley varieties in different
locations in Ethiopia. Euphytica, 167:161-195.

Abiy Legesse and Firew Mekbib., 2016. Genotype X Environment Interaction and Stability of
Early Maturing Sorghum [Sorghum bicolor (L.) Moench] Genotypes in Ethiopia. M.Sc.
Thesis, Alemaya University of Agriculture, Ethiopia:2016-01-04T05:19:25Z.

Abush Tesfaye, V P. Gupt and Ketema belete, 2001. Genotype X Environment Interaction in
Land Race Derived Advanced Lines of Sorghum (Sorghum bicolor L. Moench). M.Sc.
Thesis.

Akcura, M., Kaya, Y., Taner, S. and Ayranci, R., 2006. Parametric stability analyses for grain
yield of durum wheat. Plant Soil Environ, 52(6): 254–261.

Ali A., Qamar, I.A, Arshad, M., and Sheikh, S., 2001. Correlation of economically important
traits in sorghum bicolor varieties. Ongenotype J of Biolog. Science. 1 (5):330 331.

Alberts, M.J.A., 2004. Comparison of statistical methods to describe genotype x environment

interaction and yield stability in multi-location maize trials. M.Sc. Thesis, University of the
Free State. Bloemfontein, South Africa.

Allard, R.W. and Bradshaw, A.D., 1964. Implications of Genotype by environment

interactions in applied plant breeding. Crop science, 4:503-508.

Almeida, F.J.E., Tardin, F.D., Daher, R.F., Barbe, T.C., Paula, C.M., Cardoso, M.J. and
Godinho, V.P.C. 2014. Stability and adaptability of grain sorghum hybrids in the off-season.
Genetic and Molecular Research, DOI http://dx.doi.org/10.4238/.

Al-Temimi, H.N., Al-Shahwany, A.W. and Alsaadawi, I.S., 2013. Screening of bread wheat
cultivars (Triticum aestivum L.) to water deficit stress under field conditions. Journal of
Science, 54(3): 577-584.

Annicchiarico, P., 1997. Joint regression Vs. AMMI analysis of genotype-environment

interactions for cereals in Italy. Euphytica, 94: 53-62.

Anandan, A, R Eswaran, T Sabesan and M Prakash, 2009. Additive main effects and
multiplicative interactions analysis of yield performances in rice genotypes under coastal
sagenotype environments. Advances in Biological Research, 3:43-48.

Asfaw Adugnua, 2007. Assessment of Yield Stability in Sorghum, African Crop Science
Journal, vol.15, No, 2 pp 83-92.

Assar, A. H., R. Uptmoor, A.A. Abdelmula , C. Wagner , M. Salih, A. M. Ali, F. Ordon and
W. Friedt, 2009. Assessment of sorghum genetic resources for genetic diversity and drought

67
tolerance using molecular markers and agro-morphological traits. U. of K. J. Agric. Sci.
17(1):1-22.

Baker, R.J., 1990. Crossover genotype by environmental interaction in spring wheat. pp. 42–
51. In: Kang, M.S. (Ed.) Genotype by Environment Interaction and Plant Breeding. Louisiana
State University Agricultural Center, Baton Rouge, Louisiana.

Bakheit, B. R., 1990. Variability and correlations in grain sorghum genotypes (Sorghum
bicolor [L.] Moench) under drought conditions at different stages of growth. Journal of
Agronomy and Crop Science, 165(5): 355–360.

Basford, K. E. and Cooper, M., 1998. Genotype x environment interactions and some
considerations of their implications for wheat breeding in Australia. Australian Journal of
Agricultural Research, 49:153-174.

Becker, H. C. and Léon, L., 1988. Stability analysis in plant breeding. Plant Breeding,
101:123.

Blum, A., 1983. Genetic and physiological relationship in plant breeding for drought
resistance. Agr. Water Mgmt. 7:195-205.

Borrell A, Hammer G, van Oosterom E (2001). Stay-green. A consequence of the balance

between supply and demand for nitrogen during grain filling? Ann. Appl. Biol. 138:91-95.

Borrell, A.K., G.L. Hammer and R.G. Henzell, 2000b. Does maintaining green leaf area in
Sorghum improve yield under drought? II. Dry matter production and yield. Crop
Science, 40: 1037-1048.

Bowman. J. C., 1972. Genotype X Environment Interactions (1), Department of Agriculture,

University of Reading, Great Britain, Ann. Genet. Sel. anim., I972,4(I), II7-I23.

Burgueno, J., A. Cadena, J. Crossa, M. Banziger, A.R. Gilmour, B. Cullis., 2000. User's
Guide for Spatial Analysis of Field Variety Trials Using ASREML. Mexico, D.F.: CIMMYT,
ISBN: 970-648-060-9.

Burke, J.J., Franks, C.D., Burow, G. and Xin, Z., 2010. Selection system for the stay-green
drought tolerance trait in sorghum germplasm. Journal of Agronomy, 102(4): 1118-1122.

Butler, J. D., Byrne, P. F., Mohammadi, V., Chapman P. L., and Haley, S. D., 2005.
Agronomic performance of Rht Alleles in a spring wheat population across a range of
moisture levels. Crop Science, 45:939- 947.

Burton, G.W. and E.H. DeVane, 1953. Estimation of heritability in tall Festuca (Festuca
arudindcea) from replicated clonal material. Agronomy J., 45: 78-481.

Ceccarelli, S. and S. Grando, 2007. growing Areas of Tigray Region, northern

Ethiopia. Decentralized-participatory plant breeding:. An example of demand driven
research. Euphytica, 155(3): 349-360.

68
Central Statistical Agency of Ethiopia (CSA), 2017. Agricultural Sample Survey 2016/2017.
Stat Bull, 584.

Crossa, J., Gauch, H.G. and Zobel, R.W., 1990. Additive main effects and multiplicative
interaction analysis of two international maize cultivar trials. Crop Science, 30: 493 500.

Cruz- Aguado, J., R. Rodes, I. Perez and M. Dorado, 2000. Morphological characteristics and
yield components associated with accumulation and loss of dry mass in the internode of
wheat. Field Crop Research, 66: 129-139.

Cullis B. R, Gogel B. J, Verbyla A. P, Thompson R., 1998. Spatial analysis of multi-

environment early generation trials, Biometrics, vol.54, pp.1-18.

Cullis B.R, Smith A.B, Coombes N.E, 2006. 'On the Design of Early GenerationVariety
Trials With Correlated Data', Journal of Agriculture, Biological and Environmental Statistics,
vol.11, no.4, pp.381-393.

Dagnachew Lule, Masresha Fetene, Santie de Villiers and Kassahun Tesfaye, 2014. Additive
Main Effects and Multiplicative Interactions(AMMI) and genotype by environment
interaction (GGE) biplot analyses aid selection of high yielding and adapted finger millet
varieties. J. Appl. Biosci., 76:6291–6303.

Das, S, R C Misra, M C Patnaik and S R Das., 2010. G × E interaction, adaptability and yield
stability of midearly rice genotypes. Indian J of Agric. Res., 44:104-111.

Dehghani H., A. Ebadi, and A. Yousefi, 2006. Biplot Analysis of Genotype by Environment
Interaction for Barley Yield in Iran. Agron. J. 98:388–393.

Dillon SL, Shapter FM, Henry RJ, Cordeiro G, Izquierdo L, Lee LS, 2007. Domestication to
Crop Improvement: Genetic Resources for Sorghum and Saccharum Andropogoneae. Ann
Bot, 100:975-989.

Ding, M., Tier, B. and Yan, W., 2007. Application of GGE Biplot Analysis to Evaluate
Genotype, Environment and GxE Interaction on P. radiata: A Case Study.
http://proceedings.com.au/afgc/slides%20(PDF)/thurs%201600%20ding.pdf.

Domitruk, D.R., Duggah, B.L. and Fowler, D.B. 2001. Genotype by environment interaction
of no-till winter wheat in Western Canada. Canadian Journal of Plant Science, 81:7-16.

Dugan, W. and Labuschagne, M.T. , 2002. G x E interaction and phenotypic stability analysis
of linseed in Ethiopia. Plant Breed., 121: 66-71.

Eberhart, S.A. and Russell, W.A. 1966. Stability parameters for comparing varieties. Crop
Science, 6: 36-40.

Ethiopian Institute of Agricultural Research (EIAR), 2014. Ethiopian strategy for sorghum.
Counry strategy document.

69
Ezeaku IE, Mohammed SG, 2006. Character association and path analysis in grain sorghum.
Afr. J. Biotechnol., 5(14): 1337-1340.

Fantaye Belay and Hintsa Meresa, 2017. Performance evaluation of sorghum [Sorghum
bicolor (L.) Moench] hybrids in the moisture stress conditions of Abergelle District, Northern
Ethiopia. Academic journal, Vol. 8(4), pp. 26-32, April 2017:DOI: 10.5897/JCO2016.0168.

Finlay,G.J., Bullock, P.R., Sapirstein, H.D., Naeem, H.A., Hussain, A., Angadi, S.V., and
DePauw, R.M., 2007. Genotypic and environmental variation in grain, flour, dough and
bread-making characteristics of western Canadian spring wheat. Can. J. Plant Sci., 87: 679–
690.

Finlay, K.W. and Wilkinson, G.N., 1963. The analysis of adaptation in a plant breeding
program. Australian Journal of Agricultural Research, 14: 742-754.

Fisher R.A., 1925. Statistical methods for research works. Oliver and Boyd, London.

Francis, T.R. and Kannenburg, L.W., 1978. Yield stability studies in short season maize. A
descriptive method for grouping genotypes. Canadian Journal of Plant Sciences,
58:1029-1034.

Freeman, G.H., 1973. Statistical methods for the analysis of genotype-environment

interactions. Heredity, 31: 339-354.

Freeman, G.H., 1990. Modern statistical methods for analysing genotype x environment
interactions. In: Kang, M.S.(ed.).Genotype-by-environment interaction and plant breeding.
Louisiana State University Agricultural Center, Baton Rouge, La. pp.118 125.

Gasura, E., Setimela, P. S. and Souta, C. M., 2015. Evaluation of the performance of sorghum
genotypes using GGE biplot. Canadian. Journal of Plant Sci. 95(6): 1205–1214.

Gauch, H.G. 1982a. Noise reduction by eigenvector ordinations. Ecology, 63: 1634-1649.

Gauch, H.G., 1982b. Multivariate analysis in community ecology. 1sted, Cambridge

University press, London and New York, ISBN-10: 0521282403.

Gauch, H.G. 1988. Model selection and validation for yield trials with interaction. Biometrics,
44: 705-715.

Gauch, H.G. and Zobel, R.W. 1997. Identifying mega-environments and targeting genotypes.
Crop Science, 37(2): 311-326.

Gebeyehu Geremew, Asfaw Adugna, Taye Tadesse, Tesefaye Tesso, Ketema Belete,
Hailemichael H, 2004. Development of sorghum varieties and hybrids for dry land areas of
Ethiopia. Uga J Agri Sci., 9:594-605.

70
Gebrekidan H, 2003. Grain Yield Response of Sorghum [Sorghum bicolor (L.) Moench] to
Tied Ridges and Planting Methods on Entisols and Vertisols of Alemaya Area, Eastern
Ethiopian Highlands.

Gleeson, A. C., and Cullis, B. R., 1987. Residual Maximum Likelihood (REML) Estimation
of a Neighbour Model for Field Experiments, Biometrics, 43, 277-288.

Gilmour, A. R., Cullis, B. R., and Verbyla, A. P., 1997. Accounting For Natural And
Extraneous Variation In The Analysis Of Field Experiments,”Journal of Agricultural,
Biological, and Environmental Statistics, 2, 269–273.

Gul, I., Saruhan, V., Basbag, M., 2005. Determination of yield and yield components and
relationship among the components of grain sorghum cultivars grown as main crop. Asian
Journal of Plant Sciences, 4: 613-618.

Hagos T. and Fetien A., 2011. Additive Main Effects and Multiplicative Interactions analysis
of yield performance of sesame genotypes across environments in Northern Ethiopia. Journal
of the drylands, 4(1): 259-266.

Harlan J, De Wet J., 1972. A simplified classification of cultivated sorghum. Crop Science,
12:172-176.

IBPGR and ICRISAT, 1993. Descriptors for sorghum [sorghum bicolor(L.)Moench].

International Board for plant genetic Resources, Rome, Italy; International Crops Research
Institute for the Semi -Arid Tropics, Patancheru, India.

Frew Mekbib, 2005. The contribution of farmers’ breeders in meeting food security: the case
of sorghum (Sorghum bicolor (L.) Moench) in Ethiopia. Journal.

Kang, M.S., 1996. Using genotype-by-environment interaction for crop cultivar development.
Advances in Agronomy, 62: 199-252.

Kang, M.S., 1998. Using genotype by environment interaction for crop cultivar development.
Advances in Agronomy, 62: 199-252.

Kassahun Amare, Habtamu Zeleke and Geremew Bultosa, 2015. Variability for yield, yield
related traits and association among traits of sorghum (Sorghum Bicolor (L.) Moench)
varieties in Wollo, Ethiopia. M.Sc. Thesis. Haramaya University, College of Agriculture,
Department of Plant Sciences, Haramaya, Ethiopia, Vol7(5),pp125-133, May 2015: DOI:
10.5897/JPBCS2014.0469.

Kempton, R A., 1984. The use of biplots in interpreting variety by environment interactions. J
of Agric. Sci., 103:123135.

Kenga, R., Alabi, S. and Gupta, O., 2003. Yield stability of sorghum hybrids and parental
genotypes. African Crop Science Journal, 11(2): 65-73.

71
Kinde L., Gebeyehu C., Abubeker T., Dadi G., Shanene H., Abdela U., 2016. Evaluation of
Sorghum (Sorghum bicolor (L) Moench) Varieties and Environments for Yield Performance
and Stability, Journal of Biology, Agriculture and Healthcare, Vol.6, No.21, 2016.

Kulsum, M U, M Jamil Hasan, A Akter, H Rahman and P Biswas., 2013. Genotype-

environment interaction and stability analysis in hybrid rice: an application of additive main
effects and multiplicative interaction. Bangladesh J. Bot., 42(1): 73-81.

Leiser, W.L., Rattunde, H.F., Piepho, H.-P., Parzies, H.K., 2012. Getting the most out of
sorghum low-input field trials in West Africa using spatial adjustment. J. Agron.Crop
Sci. 198, 349–359.

Lin, C.S., Binns, M.R. and Lefkovitch, L.P., 1986. Stability analysis: where do we stand
?.Crop Science, 26: 894-899.

Ludlow, M.M. and Muchow, R.C.,1990. A critical evaluation of the traits for improving crop
yield in water limited environments. Advance Agronomy, 43: 107-153.

Lyle F. Friesen, Anita L. Brûlé-Babel, Gary H. Crow, and Patricia A. Rothenburger., 2016. Mixed
model and stability analysis of spring wheat genotype yield evaluation data from Manitoba,
Canada: Can. J. Plant Sci., 96: 305–320 (2016):dx.doi.org/10.1139/cjps-2015-0252.

Mahalanobis, P.C., 1936. On the generalized distance in statistics. Pro. Nat. Sci India, B 2:
49-55.

Mahnaz, R., Ezatollah, F. and Mohammad, M. J., 2013. Additive Main Effect and
Multiplicative Interaction Analysis of phenotypic stability in chickpea genotypes over stress
and non-stress environments. International Journal of Agriculture and Crop Sciences,
5(36):253-260.

Maposa, D., Mudimu, E. and Ngwenya, O. A., 2010. Multivariate analysis of variance
(MANOVA) of the performance of sorghum genotypes in different agro-ecological regions of
Zimbabwe. African Journal of Agricultural Research, 5(3): 196-203.

McBee GG., 1984. Relation of senescence, nonsenescence, and kernel maturity to

carbohydrate metabolism in sorghum. In: Mughogho LK (Ed.) Sorghum root and stalk
diseases, a critical review. Proceedings of the consultative group discussion of research needs
and strategies for control of sorghum root and stalk diseases. Bellagio, pp. 119-129.

Meckel, L., D.B. Egli, R.E. Phillips, D. Radcliiffe and J.E. Leggett, 1984. Effect of moisture
stress on seed growth in soybeans. Agronomy Journal, 76: 647-650.

MOORTHY, S. M., N.K. DAS and K.MANDAL, 2012. Genotype x Environment interaction
and stability analysis in bivoltine silkworm genotypes of Bombyx moriL. IJBPAS, 1(10):
1443-1449.

72
Muez Mehari, Sentayehu Alamerew, Berhane Lakew, Haddis Yirga and Mizan Tesfay. 2014.
Parametric stability analysis of malt barley genotypes for grain yield in Tigray, Ethiopia.
World Journal of Agricultural Sciences, 10 (5): 210-215.

Parmar D.J., G.N. Motaka, J.S. Patel and S.G. Patel, 2016. Study on different stability
procedures for yield of rice genotypes (Oryza sativa L.): Department of Agricultural
Statistics, B.A. College of Agriculture, AAU, Anand- 388110. International Journal of
Science, Environment, ISSN 2278-3687 (O) and Technology, Vol. 5, No 3, 2016, 1503 –
1514

Paterson, A.H., Bowers, J.E., Bruggmann, R., Dubchak, I., Grimwood, J., Gundlach, H.,
Haberer, G., Hellsten, U., Mitros, T., Poliakov, A. and Schmutz, J., 2009.The Sorghum
bicolor genome and the diversification of grasses. Nature, 457(7229), pp.551-556.

Pham, H. N. and Kang, M.S. 1988.Interrelationships among and repeatability of several

stability statistics estimated from international maize trials. Crop Science,28: 925-928.

P. O. Kurt Polat, E. A. Cifci, and K. Yagdı, 2016. Stability Performance of Bread Wheat
(Triticum aestivum L.) Genotypes: J. Agr. Sci. Tech. (2016), Vol. 18: 553-560.

Plant Genetic Resources Centre/Ethiopia, 1986. Ten years of collection, conservation and
utilization, 1976-86. PGRC/E, Addis Ababa, Ethiopia.

Purchase, J.L,. 1997. Parametric analysis to describe genotype by environment interaction and
stability in winter wheat. PhD. thesis. Department of Agronomy, Faculty of Agriculture,
University of the Orange Free State, Bloemfonten, South Africa.

Rakshit S, Hariprasanna K, Gomashe S, Ganapathy KN, Das IK, Ramana OV, Dhandapani A,
Patil JV, 2014. Changes in Area, Yield Gains, and Yield Stability of Sorghum in Major
Sorghum Producing Countries, 1970 to 2009. Crop Sci., 54:1571–1584.

Richards R. A., 2006. Physiological traits used in the breeding of new cultivars for water-
scarce environments. Agric. Water Manage, 80: 197–211.

Sabaghnia, N., Dehghani, H. and Sabaghpour, S.H. 2006. Non-parametric methods for
interpreting genotype by environment interaction of lentil genotypes. Crop Science Journal,
46: 1100-1106.

Samonte, S.O.P.B., L.T. Wilson, A.M. McClung and J.C. Medley, 2005. Targeting Cultivars
on to Rice Growing Environments Using AMMI and SREG GGE Biplot Analyses. Crop
Science 45:2414-2424.

Sarah Brumlop and Maria R. Finckh, 2011. Applications and potentials of marker assisted
selection (MAS) in plant breeding.

Sewagegne Tariku, Taddesse Lakew, Mulugeta Bitew and Mitiku Asfaw. 2013. Genotype by
environment interaction and grain yield stability analysis of rice (Oryza sativa L.) genotypes
evaluated in north western Ethiopia. Net Journal of Agricultural Science, 1(1): 10-16.

73
Shrestha, S.P., Asch, F., Dusserre, J., Ramanantsoanirine, A. and Brueck, H. 2012. Climate
effects on yield components as affected by genotypic responses to variable environmental
conditions in upland rice systems at different altitudes. Field Crops Research, 134:216-228.

Shukla, G.K. 1972. Some statistical aspects of partitioning genotype-environmental

components of variability. Heredity, 29: 237-24.

Sintayehu Admas and Kassahun Tesfaye, 2017. Genotype-by-environment interaction and

yield stability analysis in sorghum [Sorghum bicolor (L.) Moench] genotypes in North
Shewa, Ethiopia. Ethiopian Biodiversity Institute, P.O. Box 30726, Addis Ababa, Ethiopia:
DOI: 10.1515/ausae-2017-0008.

Smith CW, Frederiksen RA, 2000. History of cultivar development in the United States: From
“memories of A.B. Maunder – Sorghum breeder”. In: Smith CW, Fredriksen RA (eds)
Sorghum: Origin, history, technology and production, pp 191-223.

Suzuki, D., Griffiths, A. and Lewontin, R., 1981. An introduction to genetic analysis. W. H.
Freeman and Company. San Francisco, USA.

Tardieu, F., 2013. Plant response to environmental conditions: assessing potential production,
water demand, and negative effects of water deficit. Frontiers in physiology, 4:17.

Taye T. Mindaye , Emma S. Mace , Ian D. Godwin , David R. Jordan, 2016. Heterosis in
locally adapted sorghum genotypes and potential of hybrids for increased productivity in
contrasting environments in Ethiopia. The University of Queensland, Queensland, Alliance
for Agriculture and Food Innovation, Hermitage Research Facility. 604 Yangan Rd, Warwick,
QLD, 4370, Australia. Journal of crop science: https://doi.org/10.1016/j.cj.2016.06.020

Taye T., T. Tesfaye., and G. Ejeta, 2008. Combining ability of introduced sorghum parental
lines for major morpho-agronomic traits. Journal of SAT Agricultural Research.

Taylor, J. R. N., 2003. Overview importance of sorghum in Africa [Ongenotype].Available

from: http: WWw.sciencedirect/science.

Tesfaye Letta, M. G. D’Egidio and Mohammed Abinasa. 2008. Analysis of multi-

environment yield trials in durum wheat based on GGE-biplot. Journal of Food, Agriculture &
Environment Vol.6 (2): 217 – 221.

Thomas H, Rogers LJ., 1990. Turning over an old leaf. Univ. Wales Rev. Sci. Tech. 6:29-38.

Tolk JA, Howell TA, Miller FR, 2013. Yield component analysis of grain sorghum grown
under water stress. Field Crop Res., 145:44-51.

USDA, 2017. World Agricultural Production U.S. Department of Agriculture Foreign

Agricultural Service / Office of Global Analysis International Production Assessment
Division (IPAD): Ag Box 1051, Room 4630, South Building Washington, DC 20250-1051.

74
Vange, T., Ango I. and Nache A.V., 2014. Stability analysis of six improved sorghum
genotypes across four environments in the Southern Guinea Savanna of Nigeria.
Internationals Journal of Advances in Agricultural Science and Technology, 2 (2): 01-14.

Van Oosteriom, E.J., R. Jayachandram, and F.R. Bidininger. 1996. Diallel analysis of the stay
green traits and its components in sorghum. Crop Sci. 36:549-555.

Vieira, R.D., D.M. Tekrony and D.B. Egli, 1992. Effect of drought stress and defoliation in
the field on soybean seed germination and vigor. Crop Science, 32: 471-475.

Wolfe DW, Henderson DW, Hsiao TC, Alvino A., 1988. Interactive water and nitrogen
effects on senescence of maize. II. Photosynthetic decline and longevity of individual leaves.
Agron. J. 80:865-870.

Wricke, G., 1964. Zur berechnung der ıkovalenz bei sommerweizen und hafer. Z.
Pflanzenzuchtg, 52: 127-138.

Yan, W., 2001. GGE biplot: a windows application for graphical analysis of Multi-
Environment trial data and other types of two way data. Agronomy Journal 93: 1111-1118.

Yan, W. and Tinker, N.A. 2005. An integrated biplot analysis system for displaying,
interpreting and exploring genotype by environment interactions. Crop Science, 45: 1004–16.

Yates, F. and Cochran, W.G., 1938. The analysis of group of experiments. Journal of
Agricultural Science, 28: 556-580.

Yayeh, Z and P.W. Bosland. 2000. Evaluation of genotype, environment and genotype by
environment interaction for capsaicinoids in Capsicumannum L. Euphytica, 111: 185-190.

Younesi, O., and Moradi, A., 2009. The effect of water limitation in the field on sorghum seed
germination and vigor. Australian Journal of Basic and Applied Sciences, 3(2): 1156-1159,
INSInet Publication.
Zerihun Tadesse, Dagne wagery and Sintayehu Alamrew, 2011. Genotype X Environment
Interaction Yield stability analysis of Maize (Zea Mays L) in Ethiopia: M.Sc thesis.

75
 APPENDIX

Appendix Table 1: Variance component estimation by REML of early maturing sorghum genotypes for grain yield, phenological and
other traits tested at Erer, Kobo, Mieso, Sheraro and Shewa Robit during 2017 main cropping season.

Traits
Site VarComponent DTE PN PL PW PE DS STG PWt
Genotype 0.08ns 29.81*** 3.54*** 0.8*** 1.56*** 0.68*** 0.56*** 336386***
SR Replicate 0ns 0.02ns 0.07ns 0.1ns 0ns 0.01ns 0.02ns 766ns
Error 0.79 1.58 1.41 0.8 0.16 0.36 0.3 41246
Genotype 0.06ns 59.77*** 2.63*** 0.3*** 0.57*** 0.15*** 0.32*** 303386***
ER Replicate 0.01ns 0ns 0ns 0.1ns 0ns 0ns 0ns 108.8ns
Error 0.46 2.21 0.59 0.3 0.06 0.06 0.1 13009
Genotype 0.12* 42.35*** 4.31*** 0.1* 0.44*** 0.18** 0.2** 200479***
MS Replicate 0.1ns 0.16ns 0ns 0ns 0ns 0ns 0ns 0ns
Error 0.51 4.54 2.51 0.2 0.18 0.31 0.3 35254
Genotype 0.01ns 22.65*** 3.94*** 0.1ns 0.72*** 0.27*** 0.4*** 506450***
KB Replicate 0.01ns 0ns 0ns 0ns 0ns 0ns 0ns 0ns
Error 0.19 11.82 1.04 0.4 0.24 0.23 0.2 24507
Genotype 0.03* 37.34*** 1.91* 1* 0.16*** 0.55*** 0.93*** 175877***
SH Replicate 0.02ns 0.04ns 0.03ns 0.9ns 0.01ns 0ns 0ns 5284ns
Error 0.12 7.18 3.79 3.1 0.11 0.21 0.2 48397

Where REML = Residual Maximum Likelihood, DTE = Days to Emergency, PN = Panicle number, PL = Panicle length, PW =
Panicle width, PE = Panicle exersion, DS = Drought score, STG = Staygreen, PWt = Panicle weight, SR = Shaorobit, ER = Erer, MS
= Mieso, KB = Kobo, SH = Sheraro, *** = vary highly significant (P≤ 0.0001), ** = highly significant (P≤ 0.001), * = significant
(P<0.05) and ns = insignificant (P>0.05).

76
Appendix Table 2: Variance component estimation by REM of early maturing sorghum
genotypes for grain yield, phonological and other complex traits tested at five locations during
2017 main cropping season.

Estimate
error
Traits Gen Rep/Site Site G:S
GY 61140* 76.11ns 2743000*** 346000*** 25870
DTE 0.04** 0.02** 1.42ns 0.01ns 0.43
DTF 10.48*** 0.08ns 23.15ns 1.49** 8.41
DTM 0.55* 0.54ns 53.95ns 1.8** 9.95
GFP 6.61*** 0.46ns 73.55ns 3.31** 19.62
GFR 65.51* 2.07ns 2098.39ns 256.85*** 82.54
PHT 491.94*** 0ns 1138.56ns 203.7*** 41.91
PWt 42903.5* 574.9ns 1849005ns 261291.8*** 33091
PN 7.3* 0.03ns 176.72ns 31.05*** 5.48
PL 2.56*** 0.01ns 4.31ns 0.7*** 1.88
PW 0.28*** 0.1ns 1.31ns 0.12* 1.06
PE 0.52*** 0.00003ns 0.47ns 0.17*** 0.15
DS 0.1* 0.0003ns 0.58ns 0.27*** 0.241
STG 0.29*** 0.002ns 0.24ns 0.19*** 0.21
TSW 4.25*** 0ns 7.03ns 7.24*** 2.61

Where REML = Residual Maximum Likelihood, DTE = Days to Emergency, DTF = Days to
flowering, DTM = Days to maturity, GFP = Grain filling period, GFR = Grain filling rate,
PHT = Plant heght, PN = Panicle number, PL = Panicle length, PW = Panicle width, PE =
Panicle exersion, DS = Drought score, STG = Staygreen, PWt = Panicle weight, TSW =
Thousand weight, *** = vary highly significant (P≤ 0.0001), ** = highly significant (P≤
0.001), * = significant (P<0.05) and ns = insignificant (P>0.05).

77
Appendix Table 3: Predicted mean grain yield (kg/ha) of sixty early maturing sorghum
genotypes by spatial model tested at Erer, Kobo, Mieso, Sheraro and Shewa Robit during
2017 main cropping season.

Expperimental Locations
Genotype
Genotype Name ER KB MS SH SR
code
1 Melkam 1358 2851 1048 3661 4857
2 14MWLSDT7060 1341 3193 1189 4300 5506
3 12MW6251 1475 3398 1331 4763 6238
4 14MWLSDT7410 1309 2746 983 3433 4527
5 12MW6302 1622 2684 1080 3489 5145
6 14MWLSDT7322 2184 2655 1289 3741 6413
7 14MWLSDT7395 1024 3088 1019 3928 4546
8 14MWLSDT7400 1145 2854 965 3549 4359
9 14MWLSDT7310 1828 2519 1089 3288 5297
10 13MWF6#6077 589 2742 697 3030 2816
11 14MWLSDT7325 1926 2252 1013 2835 4991
12 2005MI5069 1486 3018 1170 4047 5495
13 14MWLSDT7196 1657 2896 1185 3909 5654
14 14MWLSDT7311 1957 2442 1107 3212 5446
15 14MWLSDT7157 1691 2799 1157 3745 5539
16 14MWLSDT7193 1557 2856 1129 3778 5337
17 14MWLSDT7332 1388 2965 1109 3893 5158
18 14MWLSDT7115 2018 2282 1062 2941 5268
19 14MWLSDT7176 1240 3342 1214 4527 5568
20 14MWLSDT7209 2113 2146 1040 2736 5219
21 12MW6440 1773 2613 1108 3435 5356
22 14MWLSDT7201 1814 3099 1335 4381 6434
23 12MW6146 1101 2897 966 3606 4342
24 14MWLSDT7364 1460 3122 1205 4229 5644
25 Pipline 2 1587 3235 1304 4514 6173
26 14MWLSDT7413 1938 2537 1140 3381 5592
27 13MWF6#6037 1115 3401 1190 4570 5392
28 14MWLSDT7207 1435 2896 1097 3787 5129
29 14MWLSDT7040 1478 3029 1172 4064 5500
30 14MWLSDT7036 1426 2803 1053 3605 4918
31 14MWLSDT7324 1693 2691 1111 3540 5325
32 12MW6243 1269 2621 913 3176 4183
33 12MW6420 835 3153 972 3947 4230
34 14MWLSDT7238 2135 2857 1357 4098 6706

78
 35 12MW6444 1800 2700 1157 3616 5596
36 14MWLSDT7402 1340 2720 984 3401 4549
37 14MWLSDT7234 1436 3283 1266 4522 5913
38 12MW6471 1994 2079 964 2543 4801
39 14MWLSDT7042 1600 2868 1150 3824 5461
40 14MWLSDT7033 1500 2980 1160 3983 5453
41 14MWLSDT7241 1617 2924 1182 3940 5616
42 14MWLSDT7191 1360 2399 853 2802 3947
43 2005MI5093 1689 2721 1122 3594 5377
44 2401 1515 2957 1156 3948 5444
45 2004MW6197 1194 3146 1111 4130 5065
46 2005MI5064 2084 2528 1194 3445 5921
47 2523 733 2774 768 3172 3224
48 04MW 6043 2356 2158 1141 2891 5815
49 2005MI5057 1592 2922 1171 3924 5555
50 04MW 6079 964 3351 1109 4393 4935
51 14MWLSDT7202 1426 2868 1082 3730 5051
52 14MWLSDT7291 1710 2749 1143 3660 5484
53 2001MS7036 1273 2840 1010 3593 4634
54 90MW5319 1273 3205 1167 4285 5370
55 99MW4047 1414 2897 1089 3777 5080
56 05MW6026 1244 3009 1071 3897 4906
57 14MWLSDT7421 1430 2941 1115 3869 5207
58 2003MW6053 988 2862 906 3478 4003
59 2294 1685 2746 1131 3640 5418
60 2003MW6038 1325 2620 935 3204 4313
Mean 1509 2832 1099 3707 5174

Where ER = Erer, KB = Kobo, MS = Mieso, SH = Sheraro, SR =

Shaorobit.

79
Appendix Table 4: Means for grain yield, phenological and other complex traits of sixty early maturing advanced sorghum genotypes
by mixed model tested at Erer during 2017 main cropping season.

Entry# Genotype Name DTE DTF GFP DTM GFR PHT PWt TSW PN PL PW PE DS STG
1 Melkam 7.0 71.7 32.0 103.7 49.3 128.5 1937.1 29.3 33.7 27.8 15.0 2.9 2.8 3.9
2 14MWLSDT7060 6.7 71.0 33.0 104.0 36.0 125.0 1286.9 25.3 44.0 20.9 15.1 2.4 2.5 3.0
3 12MW6251 7.3 72.3 34.7 107.0 35.9 147.4 1769.9 25.7 38.3 23.5 15.1 3.2 2.8 3.2
4 14MWLSDT7410 6.7 72.0 31.3 103.3 36.2 139.3 1536.1 24.3 38.0 28.2 15.5 3.4 3.5 3.0
5 12MW6302 6.7 73.0 28.0 101.0 81.7 125.1 1943.3 24.6 42.0 24.9 14.6 1.9 3.1 4.2
6 14MWLSDT7322 6.7 73.0 33.0 106.0 66.4 175.5 1873.7 23.3 34.7 21.2 14.1 1.9 2.9 2.9
7 14MWLSDT7395 6.7 63.3 43.0 106.3 11.8 125.0 430.2 26.7 35.0 24.4 14.6 1.0 2.7 2.2
8 14MWLSDT7400 6.0 67.7 39.7 107.3 17.1 144.1 653.5 26.5 26.3 23.9 15.4 3.1 3.7 4.3
9 14MWLSDT7310 7.3 72.0 33.7 105.7 46.6 157.8 2226.5 22.2 32.7 22.6 15.5 1.7 3.0 3.4
10 13MWF6#6077 7.0 74.0 26.0 100.0 54.1 107.7 1357.9 19.1 40.3 24.5 14.1 2.1 2.4 2.9
11 14MWLSDT7325 6.7 74.7 30.7 105.3 47.5 144.1 1145.0 25.0 26.0 21.5 14.5 1.9 2.8 3.3
12 2005MI5069 8.0 72.3 31.7 104.0 48.6 163.1 1597.1 29.9 34.3 21.4 14.5 3.3 2.8 3.7
13 14MWLSDT7196 7.3 69.3 35.7 105.0 52.8 140.3 1760.5 27.6 31.3 22.3 15.3 3.8 3.6 4.0
14 14MWLSDT7311 6.7 71.7 33.7 105.3 54.4 192.5 1692.4 25.2 37.3 22.3 14.6 1.3 3.2 3.3
15 14MWLSDT7157 6.7 71.7 31.3 103.0 59.3 154.9 1883.4 28.1 38.7 21.8 14.8 2.0 2.5 3.0
16 14MWLSDT7193 6.0 62.3 45.3 107.7 19.0 135.7 975.5 26.6 53.0 21.0 15.3 3.1 3.4 4.2
17 14MWLSDT7332 7.0 72.3 29.7 102.0 85.7 179.0 2281.7 20.7 36.0 19.3 14.8 3.8 3.0 3.6
18 14MWLSDT7115 7.0 73.0 30.3 103.3 65.0 178.8 1876.0 23.3 38.3 24.0 14.3 3.3 2.9 2.3
19 14MWLSDT7176 7.3 66.3 42.7 109.0 25.2 142.7 1119.0 31.7 31.7 24.2 15.7 1.9 3.0 3.5
20 14MWLSDT7209 7.7 75.3 32.3 107.7 62.6 176.6 2101.1 26.0 35.0 21.5 14.3 1.6 2.1 1.9
21 12MW6440 7.3 70.7 30.7 101.3 56.2 150.9 1975.8 22.4 36.3 23.1 15.0 3.8 3.5 4.0
22 14MWLSDT7201 7.3 70.3 31.3 101.7 73.7 151.0 1312.4 25.7 34.0 22.0 15.3 2.2 2.8 3.3
23 12MW6146 7.3 71.3 38.0 109.3 31.7 110.3 1009.4 25.3 33.3 25.7 15.3 2.1 2.4 3.5
24 14MWLSDT7364 6.3 72.0 31.0 103.0 48.7 167.3 2309.2 25.4 50.7 22.8 15.2 1.9 3.0 3.4
25 Pipline 2 6.3 68.7 35.7 104.3 40.3 166.0 1601.3 23.4 36.0 21.7 16.3 2.1 3.2 3.7

80
26 14MWLSDT7413 7.7 73.7 30.3 104.0 28.3 137.0 1203.8 24.9 30.0 22.7 15.0 2.7 3.4 3.2
27 13MWF6#6037 7.3 67.7 31.0 98.7 55.4 153.5 1755.3 24.7 42.0 21.2 17.0 2.6 3.0 3.9
28 14MWLSDT7207 5.7 71.7 27.3 99.0 56.4 152.1 1769.8 31.0 37.3 22.9 15.2 1.8 1.9 2.1
29 14MWLSDT7040 6.7 70.0 35.3 105.3 46.6 150.3 1326.3 29.7 29.3 22.1 15.6 2.2 2.8 4.1
30 14MWLSDT7036 7.3 72.0 35.3 107.3 45.5 141.3 1159.3 30.0 25.7 22.3 15.1 3.0 3.3 4.0
31 14MWLSDT7324 7.0 73.0 33.3 106.3 52.7 187.0 1495.3 20.0 35.3 21.4 13.7 1.8 2.5 3.5
32 12MW6243 7.0 70.3 35.7 106.0 19.2 115.0 1563.7 23.5 42.0 24.0 14.7 2.8 2.5 3.8
33 12MW6420 7.0 75.0 29.3 104.3 56.5 123.2 1755.5 19.5 43.0 20.7 15.0 4.2 2.6 2.8
34 14MWLSDT7238 7.0 72.0 33.0 105.0 62.1 160.7 2182.1 24.6 42.0 24.7 14.5 1.9 2.3 2.3
35 12MW6444 6.3 72.3 33.7 106.0 72.6 169.3 2192.5 24.0 43.3 22.1 14.9 2.2 2.7 2.9
36 14MWLSDT7402 6.7 71.0 31.7 102.7 48.9 185.6 1524.1 27.1 27.0 23.0 14.2 1.6 2.6 2.9
37 14MWLSDT7234 6.3 72.0 29.7 101.7 81.9 156.3 2179.7 25.1 50.7 23.4 14.9 1.8 2.1 2.3
38 12MW6471 6.7 75.7 32.7 108.3 111.3 187.8 3063.5 29.5 47.0 21.0 14.4 1.8 2.8 2.5
39 14MWLSDT7042 7.0 71.3 33.0 104.3 72.9 159.0 2188.1 30.9 27.7 23.0 14.9 2.7 2.8 3.9
40 14MWLSDT7033 7.3 68.3 39.0 107.3 27.2 126.6 1042.2 22.7 25.3 23.0 15.5 2.2 2.7 3.9
41 14MWLSDT7241 7.0 73.0 35.3 108.3 44.5 162.7 1791.2 30.9 41.7 24.1 15.1 1.9 2.2 2.5
42 14MWLSDT7191 6.0 76.3 31.0 107.3 38.0 143.3 1204.8 25.2 27.0 20.9 14.7 4.1 2.9 3.1
43 2005MI5093 7.0 72.0 34.0 106.0 29.6 138.0 1138.3 26.2 39.0 20.4 14.6 1.8 3.2 3.8
44 2401 6.7 62.3 47.7 110.0 19.0 136.3 986.4 26.3 48.0 22.1 14.3 1.5 2.9 2.7
45 2004MW6197 7.0 70.0 35.0 105.0 20.7 167.9 1080.0 26.3 56.0 21.8 14.3 2.1 3.4 3.3
46 2005MI5064 7.3 73.3 30.7 104.0 73.2 150.8 2500.8 30.0 50.0 21.6 14.2 1.6 2.6 3.2
47 2523 7.0 62.7 42.3 105.0 14.2 136.4 674.7 28.4 48.0 22.7 14.0 3.6 2.9 3.1
48 04MW 6043 7.7 62.0 41.7 103.7 92.6 176.2 2434.5 25.0 38.0 24.4 14.6 2.9 3.8 3.5
49 2005MI5057 6.3 71.7 35.3 107.0 63.9 163.0 2122.8 32.3 46.3 24.1 14.5 1.9 2.7 2.9
50 04MW 6079 7.3 72.0 32.3 104.3 52.5 139.3 2160.2 24.7 41.7 22.2 13.9 3.0 2.7 3.2
51 14MWLSDT7202 7.3 68.7 35.3 104.0 23.4 148.3 1016.2 22.3 45.7 20.5 14.4 2.3 2.3 2.7
52 14MWLSDT7291 6.7 70.3 36.3 106.7 44.8 177.4 1765.4 28.2 39.3 22.5 14.2 1.4 2.8 2.6
53 2001MS7036 6.7 63.0 48.3 111.3 8.9 155.5 876.5 30.2 44.0 25.3 14.1 1.3 3.0 3.0

81
 54 90MW5319 6.3 62.3 41.0 103.3 19.4 129.0 873.3 24.0 53.0 22.6 15.4 1.7 2.3 2.8
55 99MW4047 6.7 67.7 39.0 106.7 24.6 120.4 1088.9 27.8 43.0 22.7 14.1 2.5 3.3 3.5
56 05MW6026 6.7 65.7 38.3 104.0 40.8 144.3 1676.6 27.3 53.0 22.3 13.8 1.8 2.2 3.1
57 14MWLSDT7421 6.7 68.7 31.3 100.0 36.4 154.9 1626.8 32.5 38.3 21.5 13.8 1.9 2.6 2.7
58 2003MW6053 7.3 68.0 37.7 105.7 13.0 149.1 587.9 26.4 44.3 25.7 14.8 2.9 2.4 2.8
59 2294 6.7 62.0 44.7 106.7 8.3 113.0 523.1 31.0 26.7 22.1 14.8 1.4 2.7 2.4
60 2003MW6038 6.7 68.7 35.3 104.0 36.6 147.7 1328.5 25.4 40.0 23.3 14.9 2.8 2.7 3.2
Mean 6.9 70.1 34.8 104.9 45.8 149.8 1558.5 26.2 38.8 22.8 14.8 2.4 2.8 3.2

Where, DTE = Days to Emergency, DTF = Days to flowering, DTM = Days to maturity, GFP = Grain filling period, GFR = Grain
filling rate, PHT = Plant heght, PN = Panicle number, PL = Panicle length, PW = Panicle width, PE = Panicle exersion, DS = Drought
score, STG = Staygreen, PWt = Panicle weight, TSW = Thousand weight.

82
Appendix Table 5: Means for grain yield, phonological and other complex traits of sixty early maturing advanced sorghum genotypes
by mixed model tested at Kobo during 2017 main cropping season.

Entry# Genotype Name DTE DTF GFP DTM GFR PHT PWt TSW PN PL PW PE DS STG
1 Melkam 7.7 80.3 51.0 117.3 70.0 149.4 3638.2 32.4 39.3 22.2 13.4 4.0 3.0 4.3
2 14MWLSDT7060 8.7 83.0 51.3 120.3 60.5 178.4 3681.0 32.8 45.3 18.7 13.9 2.3 3.7 3.7
3 12MW6251 8.0 81.7 53.7 121.3 60.2 190.8 3568.0 34.7 44.3 19.6 13.7 4.3 3.7 3.7
4 14MWLSDT7410 8.0 86.7 50.3 123.0 36.5 204.5 2872.1 26.3 43.0 25.4 14.0 4.0 4.3 3.3
5 12MW6302 7.7 81.3 49.0 116.3 73.2 138.3 3427.4 24.0 51.0 23.2 13.6 2.7 3.3 5.0
6 14MWLSDT7322 8.0 86.3 48.3 120.7 33.4 231.0 2146.1 35.7 45.7 18.5 13.9 2.7 4.3 3.3
7 14MWLSDT7395 8.0 78.0 58.3 122.3 30.2 163.8 2256.4 31.5 42.7 20.7 13.8 1.0 3.3 2.3
8 14MWLSDT7400 8.0 80.7 51.3 118.0 72.9 180.1 4218.3 31.7 48.0 20.4 14.3 3.7 3.3 5.0
9 14MWLSDT7310 8.0 82.7 51.7 120.3 40.1 219.6 2408.1 31.0 51.3 19.3 13.9 2.7 4.0 4.7
10 13MWF6#6077 7.7 81.0 46.7 113.7 57.1 131.3 3031.1 19.2 51.3 21.2 14.3 3.3 3.7 4.0
11 14MWLSDT7325 8.3 89.3 43.3 118.7 33.4 225.6 1722.9 31.9 40.7 18.2 13.5 2.3 4.0 4.3
12 2005MI5069 7.3 86.7 46.0 118.7 30.6 178.5 1907.1 34.3 26.7 18.4 14.4 3.7 4.3 4.3
13 14MWLSDT7196 8.3 80.0 54.3 120.3 90.4 176.1 4854.8 33.3 47.3 19.7 14.6 4.0 3.0 4.0
14 14MWLSDT7311 8.0 83.0 53.3 122.3 33.2 229.8 2332.9 29.0 47.0 17.1 13.6 1.3 4.3 3.0
15 14MWLSDT7157 8.0 84.0 50.3 120.3 48.1 201.8 2953.8 34.6 42.7 16.3 13.4 3.0 3.3 4.0
16 14MWLSDT7193 7.7 72.0 57.0 115.0 61.5 150.9 3534.1 38.9 43.3 15.8 13.9 3.3 3.3 5.0
17 14MWLSDT7332 8.0 84.0 53.0 123.0 65.2 171.7 4198.8 34.7 46.0 14.0 13.6 4.3 3.0 3.3
18 14MWLSDT7115 8.0 91.7 51.0 128.7 12.7 215.4 1833.8 23.2 40.7 21.7 13.6 4.7 5.0 2.0
19 14MWLSDT7176 7.7 76.0 59.3 121.3 78.4 191.7 3481.1 37.7 45.3 20.1 14.4 3.0 3.7 4.3
20 14MWLSDT7209 8.0 79.3 54.3 119.7 57.8 186.1 3402.3 31.1 41.3 18.9 14.1 1.7 3.7 2.7
21 12MW6440 8.0 79.7 52.0 117.7 59.9 185.0 3399.0 31.4 48.7 19.6 13.8 4.3 4.3 4.3
22 14MWLSDT7201 8.3 80.7 53.0 119.7 50.5 181.8 3079.7 35.6 39.0 19.9 13.5 2.7 4.3 4.3
23 12MW6146 8.3 80.3 51.3 117.7 67.9 150.7 3692.0 27.1 44.3 23.0 14.8 3.0 3.3 4.3
24 14MWLSDT7364 7.7 81.7 55.3 123.0 47.3 193.2 2919.9 32.1 44.0 20.2 13.9 2.3 3.7 4.3
25 Pipline 2 8.0 76.0 55.7 117.7 60.5 207.8 3644.1 33.7 49.0 17.8 14.8 2.0 3.7 4.7

83
26 14MWLSDT7413 8.0 79.0 53.7 118.7 70.9 200.5 3898.1 29.9 46.7 17.6 13.2 3.3 3.7 4.3
27 13MWF6#6037 8.0 77.3 54.7 118.0 84.6 160.8 4804.8 34.8 45.7 17.9 15.2 3.0 2.3 3.3
28 14MWLSDT7207 7.7 79.7 54.7 120.3 49.7 184.5 2859.9 30.7 42.7 18.9 14.2 2.3 4.0 3.7
29 14MWLSDT7040 7.7 77.3 53.7 117.0 62.8 187.6 3681.5 33.0 41.0 20.1 14.1 2.7 3.3 4.7
30 14MWLSDT7036 8.0 80.7 51.0 117.7 54.6 181.9 3344.3 34.8 41.3 18.8 14.3 2.7 3.3 4.3
31 14MWLSDT7324 8.0 83.0 46.0 115.0 51.0 243.9 2656.6 31.0 50.0 18.8 13.9 2.3 3.7 4.7
32 12MW6243 8.7 84.0 48.0 118.0 63.3 139.4 3890.2 28.8 47.3 21.5 14.2 3.7 3.0 5.0
33 12MW6420 7.7 84.3 50.0 120.3 59.7 142.2 3634.1 22.6 50.3 17.8 14.4 4.7 4.3 3.7
34 14MWLSDT7238 8.0 83.7 50.7 120.3 66.8 190.8 3777.8 29.5 49.7 22.9 14.4 2.7 3.3 2.7
35 12MW6444 8.0 82.3 52.0 120.3 38.8 206.1 2749.0 32.6 48.7 18.8 13.9 3.3 4.3 4.0
36 14MWLSDT7402 8.0 80.7 55.3 122.0 42.3 258.1 3068.6 33.7 47.0 18.0 13.8 1.3 3.7 3.3
37 14MWLSDT7234 8.0 81.7 54.3 122.0 62.8 186.2 3481.5 28.6 45.0 20.0 13.9 2.3 3.0 3.3
38 12MW6471 8.0 85.7 50.3 122.0 52.9 211.5 3347.3 34.8 42.0 19.7 13.8 2.3 2.7 3.3
39 14MWLSDT7042 8.0 79.0 53.0 118.0 64.1 196.4 4368.7 35.0 50.7 19.9 14.3 2.7 3.3 4.3
40 14MWLSDT7033 8.0 79.7 57.3 123.0 62.6 188.2 4079.3 35.6 49.0 21.5 14.5 3.3 3.0 4.3
41 14MWLSDT7241 8.3 86.0 45.7 117.7 74.5 210.7 3693.9 28.4 48.3 20.0 13.7 2.7 3.3 3.7
42 14MWLSDT7191 8.0 86.7 48.7 121.3 34.9 170.4 2519.6 29.9 45.0 18.6 14.5 5.0 4.3 4.0
43 2005MI5093 8.3 82.0 50.7 118.7 60.5 215.9 3351.9 31.3 51.7 16.1 14.1 2.3 3.7 4.7
44 2401 7.7 70.7 61.3 118.0 35.3 180.2 2959.8 32.7 50.3 17.9 13.2 2.0 3.7 3.7
45 2004MW6197 8.3 85.0 48.7 119.7 48.4 179.7 2096.4 32.8 45.0 19.6 14.3 2.7 4.7 3.7
46 2005MI5064 8.0 87.3 44.7 118.0 34.0 185.1 1718.3 31.4 26.7 19.1 13.3 1.7 4.7 4.0
47 2523 8.0 70.7 62.0 118.7 61.0 155.5 3831.4 32.0 49.0 20.0 14.2 3.7 3.0 2.7
48 04MW 6043 8.7 82.3 49.3 117.7 63.3 190.1 4055.8 28.0 37.3 20.6 14.2 3.7 3.7 5.0
49 2005MI5057 8.3 82.0 52.3 120.3 55.4 211.4 3305.8 36.7 43.0 21.7 13.7 2.0 3.7 3.7
50 04MW 6079 7.7 83.7 50.0 119.7 43.9 189.6 2718.5 25.7 42.3 19.3 14.0 2.7 4.0 4.0
51 14MWLSDT7202 8.0 74.0 62.3 122.3 52.6 194.1 3389.3 35.6 45.3 15.0 14.1 2.7 3.0 4.0
52 14MWLSDT7291 8.0 80.0 56.0 122.0 43.7 249.0 3251.3 34.0 46.7 18.5 13.4 1.3 5.0 2.7
53 2001MS7036 8.0 80.7 55.0 121.7 41.8 211.6 2514.4 27.4 48.0 22.3 13.3 2.0 5.0 4.3

84
54 90MW5319 8.0 69.3 63.7 119.0 61.2 167.7 3728.7 32.8 44.7 17.7 13.8 2.3 3.0 4.3
55 99MW4047 8.0 82.3 48.7 117.0 48.7 147.1 2817.5 33.3 38.0 18.2 13.6 3.0 3.7 3.7
56 05MW6026 8.0 77.7 52.3 116.0 49.2 195.2 2560.2 34.0 48.0 17.9 13.5 2.3 3.7 3.7
57 14MWLSDT7421 8.0 78.0 54.7 118.7 45.2 213.0 2766.4 30.1 46.3 17.6 13.1 3.0 4.0 4.0
58 2003MW6053 8.3 73.3 59.3 118.7 54.3 182.7 3470.7 32.0 33.3 22.8 14.0 3.0 3.0 3.7
59 2294 8.0 68.3 67.0 121.3 51.0 175.4 3308.2 32.3 44.3 18.9 14.3 2.0 3.7 3.0
60 2003MW6038 7.7 77.3 58.0 121.3 36.3 206.7 2489.8 34.7 39.7 18.8 14.2 3.7 4.3 4.0
Mean 8.0 80.7 52.9 119.6 53.6 189.0 3206.5 31.6 44.6 19.4 14.0 2.9 3.7 3.9

85
Appendix Table 6: Means for grain yield, phonological and other complex traits of sixty early maturing advanced sorghum genotypes
by mixed model tested at Mieso during 2017 main cropping season.

Entry# Genotype Name DTE DTF GFP DTM GFR PHT PWt 1000GW PN PL PW PE DS STG
1 Melkam 6.7 78.7 36.0 114.7 21.7 126.3 920.3 34.7 13.3 32.8 14.8 2.7 3.3 3.3
2 14MWLSDT7060 5.7 76.0 34.0 110.0 18.7 129.2 1034.0 34.8 17.0 22.4 13.4 2.0 3.0 3.0
3 12MW6251 6.7 75.7 34.7 110.3 26.8 148.7 1444.7 35.3 24.3 24.8 14.0 2.7 3.3 2.7
4 14MWLSDT7410 6.3 79.0 32.7 111.7 40.5 159.2 1671.1 31.7 25.7 31.8 14.1 3.0 2.3 2.3
5 12MW6302 6.7 80.3 33.3 113.7 31.2 120.5 1106.8 30.9 24.7 25.1 13.2 1.7 3.7 3.7
6 14MWLSDT7322 6.3 80.0 31.3 111.3 22.6 180.3 1491.0 30.0 23.0 21.4 13.4 1.7 3.0 3.0
7 14MWLSDT7395 6.3 75.0 39.7 114.7 25.4 138.9 1355.5 30.0 22.7 25.8 13.6 1.0 2.3 2.7
8 14MWLSDT7400 5.7 77.0 34.0 111.0 34.1 164.4 1550.9 28.7 37.0 24.7 13.1 2.0 3.3 3.7
9 14MWLSDT7310 6.3 80.7 36.7 117.3 17.0 170.5 967.6 28.9 31.3 22.3 13.2 1.3 3.3 3.3
10 13MWF6#6077 6.0 79.3 38.0 117.3 17.2 111.8 1646.5 29.7 23.7 28.1 13.6 2.0 2.3 2.3
11 14MWLSDT7325 5.7 81.0 33.3 114.3 39.6 178.4 1835.2 31.7 31.3 22.6 13.8 1.3 2.3 3.0
12 2005MI5069 7.3 78.0 35.0 113.0 18.0 139.8 1271.0 29.4 16.3 22.6 13.3 2.7 3.0 3.3
13 14MWLSDT7196 6.7 79.0 31.3 110.3 36.8 148.2 2059.2 28.7 32.0 23.3 13.2 3.3 3.3 3.7
14 14MWLSDT7311 5.7 79.3 33.7 113.0 29.8 200.6 1648.7 32.0 38.0 24.0 13.3 1.0 3.7 3.7
15 14MWLSDT7157 5.3 76.7 33.3 110.0 49.1 168.7 1789.8 33.8 34.3 25.8 13.7 2.0 3.3 2.7
16 14MWLSDT7193 5.7 69.7 37.7 107.3 23.9 135.3 1726.4 33.0 20.7 23.0 13.4 2.3 3.3 3.3
17 14MWLSDT7332 6.0 79.0 32.7 111.7 39.9 138.8 2046.9 35.3 32.3 21.1 14.0 3.0 2.7 3.0
18 14MWLSDT7115 6.3 78.7 35.7 114.3 26.5 165.1 1734.3 42.0 23.0 27.7 13.6 2.7 2.7 2.3
19 14MWLSDT7176 6.3 76.0 34.0 110.0 42.7 143.8 2111.2 39.7 26.7 26.7 14.0 1.7 3.3 3.3
20 14MWLSDT7209 6.0 79.0 34.0 113.0 38.3 165.1 2219.2 32.3 32.0 23.3 13.4 1.3 2.7 2.7
21 12MW6440 7.0 80.0 31.0 111.0 22.6 142.9 1452.0 33.7 16.7 25.2 13.3 3.3 3.3 4.0
22 14MWLSDT7201 6.7 78.7 32.0 110.7 30.9 147.3 2211.6 39.7 27.3 22.4 13.3 1.7 3.0 3.7
23 12MW6146 7.0 80.3 31.0 111.3 24.5 131.8 988.2 39.9 22.0 26.4 13.4 2.0 2.7 3.3
24 14MWLSDT7364 5.3 77.3 34.7 112.0 35.5 144.2 1570.2 41.0 29.7 23.9 14.0 1.3 3.0 2.7
25 Pipline 2 6.0 75.7 34.3 110.0 53.5 165.9 1968.6 35.3 34.3 24.0 14.2 2.0 2.7 3.0

86
26 14MWLSDT7413 6.7 77.3 34.3 111.7 29.9 127.1 1947.0 30.1 36.7 23.6 13.1 2.7 3.3 3.0
27 13MWF6#6037 7.3 77.7 33.3 111.0 32.6 138.6 1324.4 29.2 22.7 23.6 14.4 2.0 3.3 3.7
28 14MWLSDT7207 5.9 76.0 33.7 109.7 63.8 195.9 2503.2 31.7 34.3 26.0 13.6 1.3 1.3 1.7
29 14MWLSDT7040 6.0 77.7 34.0 111.7 25.7 142.6 1343.1 29.8 25.7 22.9 13.4 1.7 3.3 4.0
30 14MWLSDT7036 6.3 81.3 31.7 113.0 20.4 171.0 1124.6 32.1 16.0 23.7 13.2 2.3 3.7 3.7
31 14MWLSDT7324 6.0 81.3 32.3 113.7 40.3 201.3 2019.9 30.3 36.0 24.8 13.4 1.3 2.3 3.3
32 12MW6243 6.7 80.0 30.3 110.3 29.4 96.0 2027.6 28.7 27.7 26.2 13.6 2.7 3.7 3.3
33 12MW6420 6.7 83.0 34.3 117.3 19.0 123.1 1093.3 30.4 22.0 22.2 13.2 4.0 3.3 2.7
34 14MWLSDT7238 6.0 77.3 34.7 112.0 40.8 162.0 2080.9 33.3 31.0 28.0 13.9 1.7 3.0 2.3
35 12MW6444 5.7 79.0 35.7 114.7 46.5 152.0 2539.0 32.0 32.7 24.2 13.2 2.0 2.7 3.0
36 14MWLSDT7402 6.7 78.0 36.0 114.0 25.3 174.1 1744.2 36.0 27.7 24.3 13.1 1.0 2.7 3.3
37 14MWLSDT7234 5.3 75.3 34.3 109.7 44.4 162.0 1153.5 43.1 37.3 24.4 13.4 1.7 2.7 2.3
38 12MW6471 6.0 82.0 35.0 117.0 15.5 152.6 1290.2 39.7 18.7 22.0 13.6 1.3 3.0 2.7
39 14MWLSDT7042 6.3 77.3 35.3 112.7 28.0 160.0 1405.5 38.0 22.3 24.8 13.6 2.0 3.3 3.7
40 14MWLSDT7033 6.0 77.3 34.7 112.0 35.7 179.5 2240.5 33.7 32.0 24.8 13.3 2.0 2.7 3.3
41 14MWLSDT7241 6.3 75.3 34.0 109.3 40.9 156.0 1899.8 29.7 33.0 27.6 13.6 1.7 2.3 2.7
42 14MWLSDT7191 5.0 82.0 35.3 117.3 14.7 140.7 1223.8 32.1 18.0 22.6 13.6 3.3 3.3 3.3
43 2005MI5093 6.7 78.3 32.0 110.3 33.8 155.8 1966.1 31.7 22.3 22.1 13.0 1.3 3.7 3.3
44 2401 6.0 74.7 37.3 112.0 45.3 148.2 2477.8 39.2 32.3 23.2 13.8 1.0 1.7 1.7
45 2004MW6197 6.3 76.7 34.0 110.7 20.0 135.6 1425.1 36.0 21.7 25.0 13.4 2.0 3.7 4.0
46 2005MI5064 6.0 75.3 35.3 110.7 49.4 160.5 1773.3 31.8 31.3 22.7 13.4 1.0 2.3 2.7
47 2523 6.3 75.7 36.0 111.7 33.2 150.8 1156.8 34.0 33.3 23.2 13.1 3.0 3.7 4.0
48 04MW 6043 6.7 79.3 34.0 113.3 30.9 152.6 2068.0 29.8 22.3 26.1 13.3 2.3 3.3 3.7
49 2005MI5057 5.0 73.3 36.7 110.0 38.0 151.5 2020.7 28.0 31.0 26.7 13.7 1.7 2.7 3.3
50 04MW 6079 6.7 82.7 33.3 116.0 25.2 147.6 1342.5 30.8 16.0 24.8 14.2 2.7 2.3 2.7
51 14MWLSDT7202 6.0 77.3 35.7 113.0 31.4 172.5 1506.5 32.3 27.7 23.7 13.3 2.0 3.3 3.3
52 14MWLSDT7291 6.3 76.7 33.3 110.0 39.8 190.3 2229.6 31.4 37.0 25.0 13.3 1.0 2.7 3.0
53 2001MS7036 6.0 77.3 35.7 113.0 25.7 158.1 1169.0 33.3 27.7 27.8 13.1 1.0 2.7 3.3

87
 54 90MW5319 5.0 69.0 39.3 108.3 38.3 135.1 1217.4 30.3 37.0 23.8 13.4 1.0 2.3 2.3
55 99MW4047 6.7 78.7 32.7 111.3 22.0 103.3 1521.4 32.6 22.3 26.0 13.2 2.0 3.3 3.3
56 05MW6026 5.7 76.7 32.7 109.3 58.0 191.2 2883.7 36.0 37.0 23.8 13.0 1.3 2.0 2.7
57 14MWLSDT7421 6.0 77.7 34.0 111.7 40.6 147.0 1769.4 37.6 33.3 22.1 13.1 1.7 2.3 2.7
58 2003MW6053 7.0 75.3 35.3 110.7 18.7 133.6 972.4 29.4 20.7 26.1 13.3 2.0 2.7 2.7
59 2294 5.7 67.3 44.3 111.7 22.5 134.7 995.2 36.1 27.3 21.9 13.1 1.0 3.3 3.0
60 2003MW6038 6.7 80.3 32.7 113.0 33.7 147.9 1329.1 35.7 24.0 25.9 13.2 2.0 2.3 3.3
Mean 6.2 77.7 34.5 112.1 32.1 151.9 1643.4 33.3 27.3 24.6 13.5 1.9 2.9 3.1

Where, DTE = Days to Emergency, DTF = Days to flowering, DTM = Days to maturity, GFP = Grain filling period, GFR = Grain
filling rate, PHT = Plant heght, PN = Panicle number, PL = Panicle length, PW = Panicle width, PE = Panicle exersion, DS = Drought
score, STG = Staygreen, PWt = Panicle weight, TSW = Thousand weight.

88
Appendix Table 7: Means for grain yield, phonological and other complex traits of sixty early maturing advanced sorghum genotypes
by mixed model tested at Sheraro during 2017 main cropping season.

Entry# Genotype Name DTE DTF GFP DTM GFR PHT PWt 1000GW PN PL PW PE DS STG
1 Melkam 5.0 70.0 27.7 97.7 133.0 204.3 4038.3 27.3 48.3 27.3 15.3 2.0 1.3 4.0
2 14MWLSDT7060 5.0 69.3 32.0 101.3 134.9 198.3 4521.7 26.3 68.0 21.0 15.1 1.0 1.0 2.7
3 12MW6251 5.0 70.0 30.7 100.7 156.5 239.5 4265.6 27.7 58.0 26.2 15.8 1.7 1.7 3.7
4 14MWLSDT7410 5.0 70.0 30.7 100.7 113.6 218.7 4007.8 26.8 72.0 26.4 16.2 1.3 2.7 2.7
5 12MW6302 5.0 70.0 30.0 100.0 117.5 186.7 3792.8 25.5 55.7 25.9 14.7 1.0 1.7 4.0
6 14MWLSDT7322 5.0 71.7 31.3 103.0 119.6 248.4 3982.8 33.5 64.7 24.2 13.1 1.0 1.0 2.0
7 14MWLSDT7395 5.0 65.0 36.0 101.0 109.7 185.6 3661.7 29.6 63.7 25.2 14.7 1.0 1.3 1.3
8 14MWLSDT7400 5.0 67.7 28.3 96.0 125.5 202.4 4192.0 24.3 52.7 26.7 16.7 1.0 3.7 4.0
9 14MWLSDT7310 5.0 70.3 34.3 104.7 95.7 272.8 3551.7 26.0 70.3 25.9 16.9 1.0 1.3 1.7
10 13MWF6#6077 5.0 73.0 28.7 101.7 108.3 153.2 3387.2 20.7 56.3 24.0 14.0 1.0 1.3 2.7
11 14MWLSDT7325 5.0 75.3 28.3 103.7 99.8 258.2 2906.1 31.3 54.3 23.0 14.4 1.0 1.0 2.3
12 2005MI5069 5.0 71.0 30.3 101.3 136.0 236.1 4182.1 34.8 50.3 22.9 14.7 1.3 1.0 3.7
13 14MWLSDT7196 4.7 69.0 29.7 98.7 132.3 207.7 4087.0 31.3 46.0 22.6 14.9 2.3 3.0 4.0
14 14MWLSDT7311 4.0 72.0 30.0 102.0 107.2 229.8 3906.1 25.8 68.3 26.2 15.3 1.0 1.0 2.7
15 14MWLSDT7157 4.7 71.7 32.0 103.7 117.3 240.9 4239.6 29.5 69.3 22.9 14.9 1.0 1.0 2.3
16 14MWLSDT7193 5.0 62.3 35.7 98.0 106.1 179.1 3626.1 33.8 51.0 23.9 16.2 1.3 3.3 4.0
17 14MWLSDT7332 4.7 72.0 30.0 102.0 131.4 216.9 4029.3 27.6 56.0 24.3 15.8 2.0 2.7 4.0
18 14MWLSDT7115 4.7 72.0 26.0 98.0 113.3 233.7 3497.7 26.5 57.0 22.0 13.6 1.0 1.3 1.7
19 14MWLSDT7176 4.7 68.7 29.0 97.7 156.5 228.2 4348.4 32.7 57.0 26.1 16.9 1.0 2.7 3.0
20 14MWLSDT7209 5.0 75.3 26.0 101.3 105.3 229.4 3108.0 26.0 50.3 22.2 13.8 1.0 1.0 1.0
21 12MW6440 4.7 69.7 30.3 100.0 113.4 180.6 3634.5 28.3 49.3 24.0 15.8 2.0 2.3 3.7
22 14MWLSDT7201 4.7 68.3 30.3 98.7 144.8 226.5 4448.9 30.3 56.7 23.4 16.7 1.7 1.0 2.0
23 12MW6146 5.0 71.3 30.3 101.7 121.7 210.8 4377.2 22.2 51.3 28.6 17.1 1.3 1.7 4.0
24 14MWLSDT7364 4.3 69.7 31.3 101.0 136.7 208.3 4291.7 32.5 53.7 23.8 16.2 1.0 1.0 3.0
25 Pipline 2 5.0 71.3 24.7 96.0 188.4 242.4 4666.1 28.9 61.7 22.4 18.0 1.0 2.7 4.0

89
26 14MWLSDT7413 4.7 74.3 27.7 102.0 121.7 206.2 3664.1 27.8 59.3 27.2 16.4 2.0 3.3 2.7
27 13MWF6#6037 5.0 68.0 33.0 101.0 142.7 187.9 4298.7 30.0 59.0 21.4 19.6 1.3 1.7 4.0
28 14MWLSDT7207 4.7 72.0 33.7 105.7 112.5 214.6 3797.3 25.2 59.0 24.2 16.4 1.0 1.0 1.3
29 14MWLSDT7040 5.0 69.3 33.7 103.0 120.9 186.3 4222.0 30.0 54.7 22.2 16.7 1.0 1.3 3.7
30 14MWLSDT7036 4.7 70.3 29.7 100.0 121.6 222.1 3785.2 33.0 48.7 25.7 16.2 1.3 1.7 2.7
31 14MWLSDT7324 4.7 71.7 29.3 101.0 120.9 192.8 3465.9 28.3 62.3 22.8 12.7 1.0 1.0 2.7
32 12MW6243 4.7 70.3 28.3 98.7 112.0 135.0 3602.3 21.1 61.7 23.8 15.3 2.0 1.3 4.0
33 12MW6420 4.7 75.7 31.3 107.0 127.3 198.3 3948.4 28.5 62.7 22.4 15.8 3.0 1.3 1.7
34 14MWLSDT7238 5.0 72.7 28.3 101.0 147.0 230.8 4075.9 26.0 65.0 23.8 13.3 1.0 1.3 1.7
35 12MW6444 4.7 72.3 29.3 101.7 123.9 240.2 3765.0 28.0 63.7 23.0 15.1 1.0 1.0 1.7
36 14MWLSDT7402 5.0 70.0 30.0 100.0 114.4 289.5 3530.0 29.7 55.3 26.6 14.4 1.0 1.0 1.7
37 14MWLSDT7234 4.7 70.7 28.3 99.0 160.4 194.7 4455.0 27.0 63.7 25.7 17.3 1.0 1.7 1.7
38 12MW6471 5.0 78.0 29.3 107.3 86.8 243.8 2410.7 32.0 60.3 20.6 14.4 1.0 2.7 1.3
39 14MWLSDT7042 5.0 69.3 32.0 101.3 119.7 228.3 3788.3 31.8 61.7 25.7 16.1 1.3 1.0 3.3
40 14MWLSDT7033 5.0 69.7 28.7 98.3 139.6 225.2 4116.0 32.2 56.3 23.8 16.7 1.3 1.0 3.3
41 14MWLSDT7241 5.0 71.3 29.7 101.0 132.9 221.1 3887.8 26.3 69.3 24.4 16.0 1.0 1.3 1.0
42 14MWLSDT7191 5.0 77.0 30.0 107.0 93.7 205.2 3056.8 32.7 53.3 22.2 14.9 2.3 1.3 2.3
43 2005MI5093 4.3 72.0 31.0 103.0 116.2 251.5 3650.0 31.3 58.7 22.2 16.0 1.0 2.0 3.7
44 2401 4.7 62.3 33.7 96.0 117.4 240.6 3799.4 33.5 59.0 25.3 15.3 1.0 3.3 2.0
45 2004MW6197 5.0 68.7 30.7 99.3 135.9 206.6 4210.0 32.8 45.3 22.8 15.0 1.3 1.7 2.3
46 2005MI5064 4.3 70.7 34.0 104.7 101.9 217.7 4350.0 34.3 55.0 23.0 14.3 1.0 1.3 3.0
47 2523 4.3 63.3 33.0 96.3 96.1 192.2 3323.3 27.7 55.7 25.0 14.0 2.0 1.7 2.7
48 04MW 6043 5.0 75.0 58.0 108.0 65.1 235.3 3140.2 27.0 51.7 26.6 14.7 1.3 3.7 1.0
49 2005MI5057 4.3 72.0 29.7 101.7 133.6 200.0 3801.1 35.5 56.3 23.8 14.3 1.0 1.0 1.0
50 04MW 6079 5.0 72.0 31.0 103.0 143.3 226.7 4491.7 27.8 59.0 22.3 11.3 1.7 1.0 1.7
51 14MWLSDT7202 5.0 67.0 29.7 96.7 125.5 228.9 3685.6 25.2 58.3 23.1 14.7 1.0 1.0 1.0
52 14MWLSDT7291 4.0 69.7 29.0 98.7 126.3 253.1 3750.0 29.3 62.7 25.8 14.5 1.0 1.0 1.7
53 2001MS7036 5.0 64.0 34.0 98.0 105.8 205.8 3416.7 26.2 45.0 25.4 14.6 1.0 1.3 2.3

90
 54 90MW5319 4.7 61.7 34.3 96.0 124.7 199.6 4426.7 32.0 58.7 25.6 17.3 1.0 1.0 2.0
55 99MW4047 5.0 67.0 32.7 99.7 115.7 208.9 4021.1 30.3 62.3 24.2 14.0 1.0 3.0 3.3
56 05MW6026 5.0 63.3 33.0 96.3 118.3 205.6 3728.3 34.2 57.0 24.3 12.3 1.0 1.3 2.7
57 14MWLSDT7421 4.7 69.0 32.0 101.0 123.8 256.7 4274.3 32.7 63.0 23.8 13.7 1.0 1.7 1.7
58 2003MW6053 5.0 67.3 32.3 99.7 108.2 180.2 3701.7 29.0 55.0 27.4 15.6 1.0 1.0 1.3
59 2294 4.7 61.7 35.3 97.0 102.5 188.0 3883.9 35.5 58.3 26.4 15.0 1.0 1.0 1.0
60 2003MW6038 4.7 68.3 29.3 97.7 109.5 230.8 3546.7 32.7 50.3 25.4 15.3 1.0 1.3 2.3
Mean 4.8 69.9 31.1 100.6 121.5 216.5 3863.7 29.3 57.8 24.3 15.3 1.3 1.6 2.5

Where, DTE = Days to Emergency, DTF = Days to flowering, DTM = Days to maturity, GFP = Grain filling period, GFR = Grain
filling rate, PHT = Plant heght, PN = Panicle number, PL = Panicle length, PW = Panicle width, PE = Panicle exersion, DS = Drought
score, STG = Staygreen, PWt = Panicle weight, TSW = Thousand weight.

91
Appendix Table 8: Means for grain yield, phonological and other complex traits of sixty early maturing advanced sorghum genotypes
by mixed model tested at Shaorobit during 2017 main cropping season.

Entry# Genotype Name DTE DTF GFP DTM GFR PHT PWt 1000GW PN PL PW PE DS STG
1 Melkam 7.0 74.7 37.7 112.3 129.4 188.0 4473.3 30.9 51.0 28.8 16.4 2.7 3.3 4.0
2 14MWLSDT7060 7.3 74.3 39.0 113.3 142.3 193.9 4649.7 24.0 45.7 21.6 17.9 4.0 2.3 2.7
3 12MW6251 7.3 76.0 36.7 112.7 171.7 247.4 5420.0 29.0 62.0 23.5 16.9 3.7 2.7 2.7
4 14MWLSDT7410 7.0 75.3 37.7 113.0 119.6 239.1 4540.0 27.1 56.3 29.0 17.8 5.0 4.7 3.7
5 12MW6302 5.7 77.3 36.3 113.7 141.4 163.2 4545.0 26.0 59.7 25.2 17.1 2.0 4.3 4.0
6 14MWLSDT7322 6.3 76.3 37.3 113.7 173.6 281.3 4535.0 35.0 57.0 20.6 16.0 2.0 3.3 3.3
7 14MWLSDT7395 6.7 63.7 47.3 111.0 96.0 188.5 4170.0 27.6 58.7 25.8 16.4 1.0 3.7 2.3
8 14MWLSDT7400 6.0 69.0 42.0 111.0 104.1 197.2 4320.0 27.6 53.0 23.9 17.4 4.7 4.7 4.3
9 14MWLSDT7310 8.0 75.3 36.3 111.7 147.5 271.3 4906.7 29.3 62.0 22.9 17.8 1.7 3.7 3.7
10 13MWF6#6077 7.7 76.3 40.3 116.7 66.6 143.8 3430.0 18.3 59.0 24.7 14.6 2.0 2.7 2.7
11 14MWLSDT7325 7.3 75.0 38.7 113.7 130.0 253.8 4393.3 31.2 59.3 22.4 16.1 2.3 4.0 3.7
12 2005MI5069 7.7 75.0 38.7 113.7 142.4 220.7 5035.0 34.6 54.7 21.9 15.6 5.0 3.3 3.7
13 14MWLSDT7196 7.3 71.0 40.0 111.0 142.0 211.3 5158.3 29.0 57.7 23.7 18.5 5.0 5.0 4.3
14 14MWLSDT7311 7.7 73.0 39.3 112.3 139.3 274.3 4757.8 27.8 68.0 21.9 16.1 1.7 4.3 3.7
15 14MWLSDT7157 7.0 73.0 39.0 112.0 142.5 246.4 4390.0 32.4 63.3 22.3 17.1 2.0 2.7 3.0
16 14MWLSDT7193 6.3 64.0 47.3 111.3 114.4 180.3 4762.6 35.3 60.7 21.3 17.6 4.7 3.7 4.3
17 14MWLSDT7332 7.3 73.7 39.0 112.7 129.7 193.8 5025.4 27.8 62.7 17.6 15.9 5.0 3.7 4.0
18 14MWLSDT7115 7.3 75.7 36.7 112.3 144.2 230.8 5010.0 25.0 56.7 24.5 16.4 4.3 3.0 3.3
19 14MWLSDT7176 7.3 66.0 46.3 112.3 120.6 233.4 5412.2 28.4 61.0 23.7 17.5 1.7 3.0 3.3
20 14MWLSDT7209 8.3 76.7 35.7 112.3 148.1 232.8 4476.7 30.9 61.3 21.4 15.8 2.0 1.3 1.3
21 12MW6440 7.3 73.3 38.7 112.0 139.7 222.1 4798.3 27.8 51.0 23.7 17.1 5.0 4.0 3.7
22 14MWLSDT7201 7.0 73.7 38.3 112.0 169.2 243.5 5608.9 27.5 64.3 22.4 17.6 2.3 3.0 3.3
23 12MW6146 7.3 73.3 39.7 113.0 109.0 165.1 3582.2 26.1 41.3 24.7 15.9 2.0 2.0 2.3
24 14MWLSDT7364 7.0 76.0 35.3 111.3 160.1 225.0 5063.3 27.9 64.3 23.5 16.6 2.3 4.3 3.7
25 Pipline 2 6.3 67.7 43.7 111.3 142.0 242.6 5593.3 31.2 61.3 22.7 18.3 3.3 3.7 3.7

92
26 14MWLSDT7413 8.3 74.3 38.7 113.0 148.6 209.9 5446.4 28.7 59.3 22.4 17.2 2.7 3.0 2.7
27 13MWF6#6037 7.0 69.0 42.0 111.0 128.3 188.6 4937.4 27.4 64.3 22.0 18.6 3.3 5.0 4.7
28 14MWLSDT7207 7.0 73.3 38.0 111.3 133.3 220.9 5029.8 26.6 61.3 22.6 16.5 2.0 1.7 1.7
29 14MWLSDT7040 6.3 72.0 40.7 112.7 135.9 225.2 4865.8 33.7 52.0 23.3 18.1 2.7 3.3 4.0
30 14MWLSDT7036 7.3 75.0 36.7 111.7 135.3 227.9 4913.2 35.9 61.7 21.2 16.8 5.0 4.7 5.0
31 14MWLSDT7324 7.7 76.3 36.0 112.3 147.9 264.6 4330.9 32.1 58.7 19.3 14.9 2.0 3.3 3.3
32 12MW6243 7.0 72.0 40.0 112.0 105.2 139.7 3322.1 27.5 52.7 24.4 15.5 2.7 2.3 2.7
33 12MW6420 7.0 75.3 40.0 115.3 103.7 148.9 3973.3 26.9 59.7 20.6 16.6 5.0 2.0 3.0
34 14MWLSDT7238 7.3 73.0 38.7 111.7 175.3 221.4 5575.6 27.9 68.7 23.9 16.5 2.0 2.0 2.7
35 12MW6444 6.7 73.3 38.0 111.3 146.0 236.3 5048.9 29.0 62.7 22.5 17.5 2.0 3.0 3.0
36 14MWLSDT7402 6.3 73.0 39.3 112.3 115.4 284.0 4465.0 33.0 57.0 23.1 15.4 2.7 3.3 3.3
37 14MWLSDT7234 7.3 75.0 37.3 112.3 157.3 223.8 5246.7 31.1 61.7 23.5 15.1 2.0 1.3 1.3
38 12MW6471 7.3 74.3 39.0 113.3 121.2 243.7 4647.2 32.2 65.0 21.5 15.9 2.0 3.0 2.7
39 14MWLSDT7042 7.7 74.7 38.7 113.3 140.3 227.5 4251.9 33.4 51.3 21.8 15.7 4.0 3.7 4.3
40 14MWLSDT7033 8.0 68.3 43.7 112.0 125.9 222.0 5696.3 31.7 57.3 21.9 17.6 2.0 4.3 4.3
41 14MWLSDT7241 6.7 76.3 37.0 113.3 152.3 238.2 5083.3 33.0 61.0 24.3 17.0 2.0 2.0 2.3
42 14MWLSDT7191 6.3 77.0 37.3 114.3 106.0 212.0 3949.1 31.9 61.7 20.2 15.7 5.0 3.0 2.7
43 2005MI5093 7.0 73.7 38.3 112.0 143.6 251.8 5116.7 27.5 66.3 21.3 15.2 2.0 4.0 3.7
44 2401 6.7 64.0 47.7 111.7 114.5 214.6 4419.3 33.4 66.0 22.1 14.9 1.3 3.0 3.3
45 2004MW6197 7.0 73.0 40.3 113.3 127.8 192.6 4422.2 32.9 47.0 19.9 14.5 2.3 4.0 3.3
46 2005MI5064 8.3 77.3 36.3 113.7 163.2 237.5 4912.8 29.6 61.0 21.7 15.7 2.0 2.7 2.7
47 2523 6.7 63.3 48.0 111.3 65.9 166.0 3283.3 28.1 54.3 22.5 14.6 5.0 3.7 3.0
48 04MW 6043 7.7 75.7 38.0 113.7 153.0 247.8 5260.0 29.3 55.3 24.3 16.2 3.7 4.7 4.7
49 2005MI5057 7.0 73.0 38.3 111.3 143.9 236.4 4873.3 37.1 60.0 24.3 16.3 2.7 3.3 3.7
50 04MW 6079 7.3 73.7 38.7 112.3 125.9 221.9 4460.0 28.9 57.7 22.3 16.1 4.7 4.0 4.3
51 14MWLSDT7202 7.7 72.3 40.0 112.3 127.9 243.6 4450.0 29.5 63.0 20.2 15.4 3.3 2.3 2.3
52 14MWLSDT7291 6.7 72.7 39.0 111.7 142.6 258.2 5047.3 32.3 64.3 20.9 15.7 2.0 2.3 3.0
53 2001MS7036 6.3 63.3 48.3 111.7 97.0 242.8 4120.0 28.8 63.7 25.9 15.2 1.0 2.7 2.3

93
 54 90MW5319 7.0 64.7 47.0 111.7 114.6 211.3 5220.0 32.0 62.0 23.1 17.1 1.7 3.0 2.3
55 99MW4047 6.3 69.7 41.7 111.3 123.2 154.4 3358.9 26.1 52.3 22.3 15.8 3.3 3.3 3.7
56 05MW6026 6.7 69.3 42.7 112.0 114.7 219.1 4187.4 29.9 61.3 23.3 16.5 2.0 2.0 3.3
57 14MWLSDT7421 7.0 70.0 41.3 111.3 126.9 257.0 4520.0 36.3 59.0 22.6 15.5 1.7 3.0 2.3
58 2003MW6053 7.3 70.0 41.7 111.7 97.8 209.4 3693.5 33.1 53.0 26.4 16.3 4.7 3.0 3.3
59 2294 7.3 63.7 47.7 111.3 116.3 192.5 4630.0 32.4 66.3 21.2 16.7 1.3 2.7 2.7
60 2003MW6038 6.0 70.3 41.3 111.7 103.7 222.7 3936.7 31.9 54.0 23.2 16.9 3.7 3.0 3.0
Mean 7.1 72.3 40.1 112.4 130.8 220.1 4645.9 30.0 58.9 22.8 16.4 2.9 3.2 3.3

Where, DTE = Days to Emergency, DTF = Days to flowering, DTM = Days to maturity, GFP = Grain filling period, GFR = Grain
filling rate, PHT = Plant heght, PN = Panicle number, PL = Panicle length, PW = Panicle width, PE = Panicle exersion, DS = Drought
score, STG = Staygreen, PWt = Panicle weight, TSW = Thousand weight.

94
Appendix Table 9: Means for grain yield, phenolgical and other complex traits of sixty early maturing advanced sorghum genotypes at
Erer, Kobo, Mieso, Sheraro and Shewa Robit during 2017 main cropping season.

Traits
Locations DTE DTF GFP DTM GFR PHT GY PWt 1000GW PN PL PW PE DS STG
Erer 6.9 70 35 105 45.8 150 1508 1559 26.1836 39 23 15 2 3 3.2
Kobo 8 81 53 120 53.6 189 2831 3207 31.6309 45 19 14 3 4 3.9
Mieso 6.2 78 34 112 32.1 152 1099 1643 33.3028 27 25 13 2 3 3.1
Sheraro 4.8 70 31 101 122 216 3707 3864 29.2672 58 24 15 1 2 2.5
Shaorobit 7.1 72 40 112 131 220 5174 4646 29.9823 59 23 16 3 3 3.3

Where, DTE = Days to Emergency, DTF = Days to flowering, DTM = Days to maturity, GFP = Grain filling period, GFR = Grain
filling rate, PHT = Plant heght, PN = Panicle number, PL = Panicle length, PW = Panicle width, PE = Panicle exersion, DS = Drought
score, STG = Staygreen, PWt = Panicle weight, TSW = Thousand weight.

95
Appendix Table 10: Means for grain yield, phonological and other traits of sixty early maturing advanced sorghum genotypes tested at
five sites during 2017 main cropping season.

Entry# Genotype Name DTE DTF GFP DTM GFR PHT GY PWt 1000GW PN PL PW PE DS STG
1 Melkam 6.7 75.1 36.9 109.1 80.7 159.3 2885.1 3001.5 30.9 37.1 27.8 15.0 2.9 2.8 3.9
2 14MWLSDT7060 6.7 74.7 37.9 109.8 78.5 165.0 3013.5 3034.6 28.7 44.0 20.9 15.1 2.3 2.5 3.0
3 12MW6251 6.9 75.1 38.1 110.4 90.2 194.7 3284.6 3293.6 30.5 45.4 23.5 15.1 3.1 2.8 3.2
4 14MWLSDT7410 6.6 76.6 36.5 110.3 69.3 192.2 2435.0 2925.4 27.3 47.0 28.2 15.5 3.4 3.5 3.0
5 12MW6302 6.3 76.4 35.3 108.9 89.0 146.8 3099.9 2963.1 26.2 46.6 24.9 14.6 1.8 3.2 4.2
6 14MWLSDT7322 6.5 77.5 36.3 110.9 83.1 223.3 2937.9 2805.7 31.5 45.0 21.2 14.1 1.8 2.9 2.9
7 14MWLSDT7395 6.5 69.0 44.9 111.1 54.6 160.3 2343.4 2374.8 29.1 44.5 24.4 14.6 1.0 2.7 2.2
8 14MWLSDT7400 6.1 72.4 39.1 108.7 70.7 177.6 2691.0 2987.0 27.8 43.4 23.9 15.4 2.9 3.7 4.3
9 14MWLSDT7310 6.9 76.2 38.5 111.9 69.4 218.4 2579.1 2812.1 27.5 49.5 22.6 15.5 1.7 3.1 3.3
10 13MWF6#6077 6.7 76.7 35.9 109.9 60.6 129.5 2086.8 2570.5 21.4 46.1 24.5 14.1 2.1 2.5 2.9
11 14MWLSDT7325 6.6 79.1 34.9 111.1 70.1 212.0 2408.7 2400.5 30.2 42.3 21.5 14.5 1.8 2.8 3.3
12 2005MI5069 7.1 76.6 36.3 110.1 75.1 187.6 2617.5 2798.4 32.6 36.5 21.4 14.5 3.2 2.9 3.8
13 14MWLSDT7196 6.9 73.7 38.2 109.1 90.9 176.7 3494.6 3584.0 30.0 42.9 22.3 15.3 3.7 3.6 4.0
14 14MWLSDT7311 6.4 75.8 38.0 111.0 72.8 225.4 2652.3 2867.6 28.0 51.7 22.3 14.6 1.3 3.3 3.3
15 14MWLSDT7157 6.3 75.4 37.2 109.8 83.3 202.5 3028.5 3051.3 31.7 49.7 21.8 14.8 2.0 2.6 3.0
16 14MWLSDT7193 6.1 66.1 44.6 107.9 65.0 156.2 2882.9 2924.9 33.5 45.7 21.0 15.3 3.0 3.4 4.2
17 14MWLSDT7332 6.6 76.2 36.9 110.3 90.4 180.0 3249.2 3516.4 29.2 46.6 19.3 14.8 3.6 3.0 3.6
18 14MWLSDT7115 6.7 78.2 35.9 111.3 72.3 204.7 2354.5 2790.4 28.0 43.1 24.0 14.3 3.2 3.0 2.3
19 14MWLSDT7176 6.7 70.6 42.3 110.1 84.7 187.9 3453.1 3294.4 34.0 44.3 24.2 15.7 1.8 3.1 3.5
20 14MWLSDT7209 7.0 77.1 36.5 110.8 82.4 198.0 2884.5 3061.5 29.3 44.0 21.5 14.3 1.5 2.2 1.9
21 12MW6440 6.9 74.7 36.5 108.4 78.4 176.3 2868.9 3051.9 28.7 40.4 23.1 15.0 3.7 3.5 3.9
22 14MWLSDT7201 6.8 74.3 37.0 108.5 93.8 190.0 3343.6 3332.3 31.7 44.3 22.0 15.3 2.1 2.8 3.3
23 12MW6146 7.0 75.3 38.1 110.6 71.0 153.7 2670.2 2729.8 28.1 38.5 25.7 15.3 2.1 2.4 3.5
24 14MWLSDT7364 6.1 75.3 37.5 110.1 85.6 187.6 3034.7 3230.9 31.8 48.5 22.8 15.2 1.8 3.0 3.4
25 Pipline 2 6.3 71.9 38.8 107.9 96.9 204.9 3462.7 3494.7 30.5 48.5 21.7 16.3 2.1 3.2 3.8

96
26 14MWLSDT7413 7.1 75.7 36.9 109.9 79.9 176.1 2950.7 3231.9 28.3 46.4 22.7 15.0 2.7 3.4 3.2
27 13MWF6#6037 6.9 71.9 38.8 107.9 88.7 165.9 3465.1 3424.1 29.2 46.7 21.2 17.0 2.4 3.1 3.9
28 14MWLSDT7207 6.2 74.5 37.5 109.2 83.1 193.6 3050.9 3192.0 29.0 46.9 22.9 15.2 1.7 2.0 2.1
29 14MWLSDT7040 6.3 73.3 39.5 109.9 78.4 178.4 3080.2 3087.8 31.2 40.5 22.1 15.6 2.0 2.8 4.1
30 14MWLSDT7036 6.7 75.9 36.9 109.9 75.5 188.9 2685.3 2865.3 33.1 38.7 22.3 15.1 2.9 3.3 3.9
31 14MWLSDT7324 6.7 77.1 35.4 109.7 82.5 217.9 2847.0 2793.7 28.4 48.5 21.4 13.7 1.7 2.6 3.5
32 12MW6243 6.8 75.3 36.5 109.0 65.8 125.0 2380.7 2881.2 25.9 46.3 24.0 14.7 2.8 2.6 3.8
33 12MW6420 6.6 78.7 37.0 112.9 73.2 147.1 2673.5 2880.9 25.6 47.5 20.7 15.0 4.2 2.7 2.8
34 14MWLSDT7238 6.7 75.7 37.1 110.0 98.4 193.1 3533.1 3538.5 28.3 51.3 24.7 14.5 1.8 2.4 2.3
35 12MW6444 6.3 75.9 37.7 110.8 85.6 200.8 3052.2 3258.9 29.1 50.2 22.1 14.9 2.1 2.7 2.9
36 14MWLSDT7402 6.5 74.5 38.5 110.2 69.3 238.2 2540.4 2866.4 31.9 42.8 23.0 14.2 1.5 2.7 2.9
37 14MWLSDT7234 6.3 74.9 36.8 108.9 101.4 184.6 3528.5 3303.3 31.0 51.7 23.4 14.9 1.8 2.2 2.2
38 12MW6471 6.6 79.1 37.3 113.6 77.6 207.9 2818.1 2951.8 33.6 46.6 21.0 14.4 1.7 2.8 2.5
39 14MWLSDT7042 6.8 74.3 38.4 109.9 85.0 194.2 3200.9 3200.5 33.8 42.7 23.0 14.9 2.5 2.8 3.9
40 14MWLSDT7033 6.9 72.7 40.7 110.5 78.2 188.3 3068.0 3434.8 31.2 44.0 23.0 15.5 2.2 2.7 3.8
41 14MWLSDT7241 6.7 76.4 36.3 109.9 89.0 197.7 3174.8 3271.2 29.6 50.7 24.1 15.1 1.8 2.2 2.4
42 14MWLSDT7191 6.1 79.8 36.5 113.5 57.5 174.3 2024.6 2390.8 30.3 41.0 20.9 14.7 4.0 3.0 3.1
43 2005MI5093 6.7 75.6 37.2 110.0 76.7 202.6 2830.6 3044.6 29.6 47.6 20.4 14.6 1.7 3.3 3.8
44 2401 6.3 66.8 45.5 109.5 66.3 184.0 2830.8 2928.6 33.0 51.1 22.1 14.3 1.4 2.9 2.7
45 2004MW6197 6.7 74.7 37.7 109.6 70.6 176.5 2594.4 2646.7 32.2 43.0 21.8 14.3 2.1 3.5 3.3
46 2005MI5064 6.8 76.8 36.2 110.2 84.3 190.3 2964.4 3051.1 31.4 44.8 21.6 14.2 1.4 2.7 3.1
47 2523 6.5 67.1 44.3 108.6 54.1 160.2 2380.0 2453.9 30.0 48.1 22.7 14.0 3.4 3.0 3.1
48 04MW 6043 7.1 74.6 44.2 111.3 81.0 200.4 3256.9 3391.7 27.8 40.9 24.4 14.6 2.8 3.8 3.6
49 2005MI5057 6.2 74.4 38.5 110.1 87.0 192.4 3190.6 3224.7 33.9 47.3 24.1 14.5 1.9 2.7 2.9
50 04MW 6079 6.8 76.8 37.1 111.1 78.2 185.0 2799.0 3034.6 27.6 43.3 22.2 13.9 2.9 2.8 3.2
51 14MWLSDT7202 6.8 71.9 40.6 109.7 72.2 197.5 2803.4 2809.5 29.0 48.0 20.5 14.4 2.3 2.4 2.7
52 14MWLSDT7291 6.3 73.9 38.7 109.8 79.4 225.6 2908.8 3208.7 31.0 50.0 22.5 14.2 1.4 2.8 2.6
53 2001MS7036 6.4 69.7 44.3 111.1 55.9 194.8 2382.1 2419.3 29.2 45.7 25.3 14.1 1.3 2.9 3.1

97
 54 90MW5319 6.2 65.4 45.1 107.7 71.6 168.5 3169.9 3093.2 30.2 51.1 22.6 15.4 1.5 2.3 2.8
55 99MW4047 6.5 73.1 38.9 109.2 66.9 146.8 2589.2 2561.6 30.0 43.6 22.7 14.1 2.4 3.3 3.5
56 05MW6026 6.4 70.5 39.8 107.5 76.2 191.1 2947.9 3007.2 32.3 51.3 22.3 13.8 1.7 2.2 3.1
57 14MWLSDT7421 6.5 72.7 38.7 108.5 74.6 205.7 2803.8 2991.4 33.8 48.0 21.5 13.8 1.8 2.7 2.7
58 2003MW6053 7.0 70.8 41.3 109.3 58.4 171.0 2368.3 2485.2 30.0 41.3 25.7 14.8 2.7 2.4 2.8
59 2294 6.5 64.6 47.8 109.6 60.1 160.7 2783.6 2668.1 33.5 44.6 22.1 14.8 1.4 2.7 2.4
60 2003MW6038 6.3 73.0 39.3 109.5 64.0 191.2 2397.0 2526.1 32.1 41.6 23.3 14.9 2.6 2.7 3.2
Mean 6.6 74.1 38.7 109.9 76.8 185.5 2863.9 2983.6 30.1 45.5 22.8 14.8 2.3 2.9 3.2

Where, DTE = Days to Emergency, DTF = Days to flowering, DTM = Days to maturity, GFP = Grain filling period, GFR = Grain
filling rate, PHT = Plant heght, PN = Panicle number, PL = Panicle length, PW = Panicle width, PE = Panicle exersion, DS = Drought
score, STG = Staygreen, PWt = Panicle weight, TSW = Thousand weight.

98
Appendix Table 11: The IPCA 1 and IPCA 2 scores for the five sites, sorted on environmental mean yield, used in the study.

Location Environment Mean IPCA - 1 IPCA - 2

Errer 1508.48 -0.6346 -0.07441

Kobo 2831.281 1 -0.28507

Mieso 1099.278 -0.07839 0.656635

Sheraro 3706.683 0.152699 0.267812

Shaorobit 5174.024 -0.43971 -0.56497

Where IPCA - 1 = Interaction Principal Component Analysis Score 1 and IPCA – 2 = Interaction Principal Component Analysis Score
2.
Appendix Table 12: Total monthly rainfall (mm) and mean monthly temperature (°C) of the four tested locations during 2014 main
cropping season.

Total Rain fall (mm) Mean Temperature (°C)

Month Errer Kobo Mieso Shaorobit Sheraro
Errer Kobo Mieso Shaorobit Sheraro
Min. Max. Min. Max Min Max Min Max. Min Max.
June 95 4.5 13.9 14.9 60.9 16.9 32.3 13.7 31.8 16.7 35 19.2 36.2 21.7 36.2
July 115 154 154.2 244.9 140 16.8 31.3 16.1 33.9 18.8 33.2 19.3 34.3 20.7 32.4
Aug. 110 256 90.1 188.1 253.9 17 30.8 17.1 34 18 31.2 18.6 31.8 20.3 29.7
Sept. 90 160 158.7 98.2 152.5 16.7 31.5 14.5 30.9 16.7 30.6 18 32 19.6 32
Oct. 70 97 147.5 141.9 6.5 13.8 32.1 13.4 30.7 14 29 16.1 30.8 21.2 35.8
Nov. 30 6.3 7.5 25 1.2 11.9 28.6 14.2 31.2 11.8 29.8 14.7 30.3 19.1 36.3

99
 SR

SH

KB

ER

MS

1Appendix Figure 1: Mean grain yield graphical display of genotype by environment interaction.
Where G1 = Melkam , G2 = 14MWLSDT7060, G3 = 12MW6251, G4 = 14MWLSDT7410, G5 =
12MW6302, G6 = 14MWLSDT7322, G7 = 14MWLSDT7395, G8 = 14MWLSDT7400, G9 =
14MWLSDT7310, G10 = 13MWF6#6077, G11 = 14MWLSDT7325, G12 = 2005MI5069, G13 =
14MWLSDT7196, G14 = 14MWLSDT7311, G15 = 14MWLSDT7157, G16 = 14MWLSDT7193,
G17 = 14MWLSDT7332, G18 = 14MWLSDT7115, G19 = 14MWLSDT7176, G20 =
14MWLSDT7209, G21 = 12MW6440, G22 = 14MWLSDT7201, G23 = 12MW6146, G24 =
14MWLSDT7364, G25 = Pipeline 2, G26 = 14MWLSDT7413, G27 = 13MWF6#6037, G28 =
14MWLSDT7207, G29 = 14MWLSDT7040, G30 = 14MWLSDT7036, G31 = 14MWLSDT7324,
G32 = 12MW6243, G33 = 12MW6420, G34 = 14MWLSDT7238, G35= 12MW6444, G36 =
14MWLSDT7402, G37 = 14MWLSDT7234, G38 = 12MW6471, G39 = 14MWLSDT7042, G40 =
14MWLSDT7033, G41 = 14MWLSDT7241, G42 = 14MWLSDT7191, G43 = 2005MI5093, G44 =
2401, G45 = 2004MW6197, G46 = 2005MI5064, G47 = 2523, G48 = 04MW 6043, G49 =
2005MI5057, G50 = 04MW 6079, G51 = 14MWLSDT7202, G52 = 14MWLSDT7291, G53 =
2001MS7036, G54 = 90MW5319, G55 = 99MW4047, G56 = 05MW6026, G57 = 14MWLSDT7421,
G58 = 2003MW6053, G59 = 2294, G60 = 2003MW6038, GY = mean of grain yield and 17ER = Erer,
17MS = Mieso, 17KB = Kobo, 17SR = Shaorobit, 17SH = Sheraro

100
2Appendix Figure 2: Graphical display of Correlation between locations with heat map (MET analysis) on grain yield
performance of early maturing sorghum genotypes.

101
3Appendis Figure 3: Spatial trend and spatially independent residuals from the spatial model for grain yield (kg/ha) in
each location plotted against row and column positions.

101