SINET

Ethiopian Journal of Science

ISSN: 0379–2897 (Print)
2520–7997 (Online)

RESEARCH ARTICLES

Effect of phosphate solubilizing bio-inoculants

and vermicompost application on mineral
uptake and growth of coffee (Coffea arbica L.)
seedlings under greenhouse
condition …………………………….…………135-150
Reshid Abafita Abawari, Fassil Assefa and
Diriba Muleta

Evaluation of chickpea (Cicer arietinum L.)

genotypes for tolerance to Frost in controlled
environment…………….. ................................ 151-160 Indexed and abstracted by
Sintayehu Admas, Teklehaimanot American Mathematical Reviews
Haileselassie, Kassahun Tesfaye, Eleni Shiferaw BIOSIS
and K. Colton Flynn CAB International
Environment Abstracts

Contents continued on the outside back cover

SINET online:
http://www.ajol.info/

College of Natural and

Vol. 44 No. 2 computational Sciences
December 2021 Addis Ababa University
 SINET: ETHIOPIAN JOURNAL OF SCIENCE

EDITORIAL BOARD

Editor-in-Chief Emana Getu, Dept of Zoological Sciences, AAU

PO Box 31226, Addis Ababa, Ethiopia
Tel. 251–11–1239474, Fax: 251–11–1552350
E-mail: sinet@aau.edu.et

Associate Editors Araya Gebresilassie, Dept of Zoological Sciences, AAU

Ameha Atnafu, School of Earth Sciences, AAU
Dereje Teferi, Dept. of Computer Science, AAU
Shibru Temesgen, Dept of Statistics, AAU
Mesfin Redi, Dept of Chemistry, AAU
Belayneh Mesfin, Dept of Physics, AAU
Tigist Wondimu, Dept. of Plant Biology and Biodiversity Conservation,
AAU
Fitsum Tigu, Dept. of Microbial, Cellular and Molecular, AAU
Hunduma Legesse, Dept. of Mathematics, AAU
Aschenaki Taddese, Dept. of Sport Science, AAU
Yaregal Asabie, Dept. of Information Technology, AAU

EDITORIAL OFFICE

Editorial Assistant Yonas Yigezu

INTERNATIONAL ADVISORY BOARD

Dr Abebe Tessera, University of Port Elizabeth, South Africa

Prof. Aberra Mogessie, University of Graz, Austria
Dr R.D. Adams, International Seismic Centre, United Kingdom
Prof. Berhanu Abegaz Gashe, University of Botswana, Botswana
Prof. Berhanu Abegaz Molla, University of Botswana, Botswana
Prof. L.O. Björn, University of Lund, Sweden
Prof. B.S. Chandravanshi, Addis Ababa University, Ethiopia
Prof. Edemariam Tsega, Memorial University of NFD, Canada
Prof. Ermias Dagne, Addis Ababa University, Ethiopia
Prof. Fassil Nebebe, Concordia University, Canada
Prof. Hailemichael Bereda, University Paul Sabatier, France
Dr Hailu Gebre Mariam, EARO, Ethiopia
Dr Laike M. Asfaw, Addis Ababa University, Ethiopia
Prof. U. Lüttge, Darmstadt University of Technology, Germany
Prof. P. Mohr, National University of Ireland-Galway, Ireland
Prof. E. Oni, University of Ibadan, Nigeria
Prof. Shiferaw Berhanu, Temple University, USA
Prof. A. Taube, Uppsala University, Sweden
Prof. Theodros Solomon, Addis Ababa University, Ethiopia
Prof. G. Trenkler, University of Dortmund, Germany

© College of Natural and computational Sciences, Addis Ababa University, 2021

 SINET: Ethiopian Journal of Science
Volume 42, No. 2 (December 2021)

Contents

Plant biology and diversity management

Effect of phosphate solubilizing bio-inoculants and vermicompost application on mineral uptake and
growth of coffee (Coffea arbica L.) seedlings under greenhouse
condition……………….………………………………………………………………..…………………………135-150
Reshid Abafita Abawari, Fassil Assefa and Diriba Muleta

Evaluation of chickpea (Cicer arietinum L.) genotypes for tolerance to Frost in controlled
environment……………….…………………………………………………………..…………………………151-160
Sintayehu Admas, Teklehaimanot Haileselassie, Kassahun Tesfaye, Eleni Shiferaw and K. Colton Flynn

Molecular , cellular and microbial

Child nutritional status, mothers’ nutritional knowledge and practice and Household food security status
in Tehuledere Woreda, South Wollo, Ethiopia……………….……………………………………………………161-171
Ahmed Indris, Dilu Shaleka and Mogessie Ashenafi

Zoological science

Seasonal variation of non-volant small mammals in Gibe Sheleko national park, Southwestern
Ethiopia………………………………………………………………………………………………….………..172-181
Seyoum Kiros and Afework Bekele

Diurnal activity patterns, habitat use and foraging habits of Egyptian goose (Alopochena egyptiacus
Linnaeus, 1766) in the Boyo wetland, southern Ethiopia ………………………………………………….182-192
Mulugeta Kassa, M. Balakrishnan and Bezawork Afework

Survey and identification of termites (Insecta, Isoptera) using morphological and molecular methods
from eastern, central and western Ethiopia ……………………………………………………….……….193-204
Ashenafi Kassaye , Emana Getu , Mulatu Wakgari , Muluken Goftishu , Awol Seid Samantha J.
Montoya 3 and Gillian H. Gile

Chemistry

Determination of selected major and trace metals in lemongrass (Cymbopogon citratus) by microwave
plasma-atomic emission spectrometry…………………………………………………………………….205-214
Nitsuh Birhanu, Weldegebriel Yohannes and Bhagwan Singh Chandravanshi
284
SINET: Ethiop. J. Sci., 24(2), 2001 [Short communication]

Geology

Early hunters and herders of northern Ethiopia: The fauna from Danei Kawlos……..…………….215-222
Agazi Negash and Fiona Marshall

Information technology

An Integrated Knowledge-Based System for Early Detection of Eye Refractive Error Using Data
Mining……………………………………………………………………..…………………………….…….223-233
Teklay Birhane, Dr. Dereje Teferi and Tariku Mohammed

Sport science

Effects of plyometric and strength trainings on selected physical fitness variables in Ethiopia youth sport
academy female soccer players………………………………………………..……………………….…….234-241
Chalachew Lemecha1 and Aschenaki Taddese2,*

Guide to authors and acknowledgment

SINET:Ethiop. J. Sci., 44(2): 135–150, 2021 ISSN: 0379–2897 (PRINT)
© College of Natural and Computational Sciences, Addis Ababa University, 2021 eISSN: 2520–7997

Date received: February 03, 2021; Date revised: May 05, 2021; Date accepted: May 10, 2021
DOI: https://dx.doi.org/10.4314/sinet.v44i2.1

Effect of phosphate solubilizing bio-inoculants and vermicompost application on mineral

uptake and growth of coffee (Coffea arbica L.) seedlings under greenhouse condition

Reshid Abafita Abawari1, Fassil Assefa2 and Diriba Muleta3,*

1Department of Microbial, Cellular and Molecular Biology, Jimma Agricultural Research Centre, Jimma,
Ethiopia.
2Department of Microbial, Cellular and Molecular Biology, Addis Ababa University, Addis Ababa,

Ethiopia
3Institute of Biotechnology, Addis Ababa University, Addis Ababa, Ethiopia. E-mail:

diriba.muleta@aau.edu.et,

ABSTRACT: Arabica coffee (Coffea Arabica L.) is an economically important crop with the highest
export revenue in Ethiopia. This study was designed to evaluate the bacterial and fungal phosphate
solublization efficacy and to determine yield attributes of coffee seedlings under glasshouse conditions.
The study was conducted at Jimma Agricultural Research Center. The experiment was done with
completely randomized design (CRD) in three (3) replications. Three potent bacterial isolates viz.,
RCHVCB1, RScB1.19 and RMaB2.11 and three potent fungal isolates viz., RSCF1.19 RLVCF2 and RCHVCF2
were obtained from Jimma University, Veterinary Medicine, Microbiology Laboratory. The three
bacterial isolates were tested for antimicrobial resistance pattern and for their potential to serve as bio-
control agents. All the bacterial isolates showed 100% resistance to all the six antimicrobials tested. The
growth of pathogenic Fusarium xyloriodes was slightly inhibited by RSCF1.19. Single inoculation of
RSCF1.19+Phosphate fertilizer and dual inoculation of RSCF1.19 and RCHVCB1 with P fertilizer
significantly (p<0.05) increased plant height, root length, stem girth, leaf number, leaf area, fresh and
dry weights of coffee seedlings. However, all the treatments combined with vermicompost showed
suppressive characteristics with no seedlings emergency at all. RSCF1.19 and RCHVCB1 can be
recommended as bio-fertilizers after conducting necessary field trials in order to reduce the cost
required for chemical fertilizers.

Keywords/phrases: Bio-inoculants, coffee seedlings, dual inoculation, phyto-beneficial microbes,

single inoculation

production and productivity of coffee on

INTRODUCTION
Arabica coffee (Coffea arabica L.) grows in Ethiopia,
which is the place of its origin and hence its
production and consumption are closely
intertwined with Ethiopian history, culture and
economy (Alemayehu Teshome et al., 2008).
Arabica coffee is the most important Ethiopian
commodity and the principal source of revenue for
the agricultural sector and greatly requires
sustainable production with better product quality
to persist in the present world market competition.
The simplicity of production and low cost inputs
requirements are needed for coffee production,
particularly for farmers deprived of sufficient
resources. The use of liquid organic fertilizers as
important agricultural inputs to improve
_____________________
*Author to whom correspondence should be addressed.
sustainable basis has been reported by Ormeño-
Díaz et al.(2018). The use of biofertilizers in
agriculture has proven to be eco-friendly,
productive and accessible option to be utilized as
agricultural inputs (Sherazet al., 2010).
Accordingly, Tam et al.(2016) have isolated fungal
P-solubilizers from different plant rhizosphere and
carried out an investigation on their phosphate-
solubilizing capabilities. In addition, Satyaprakash
et al.(2017) have documented several strains of
bacteria isolated from coffee rhizosphere that were
capable of solublizing inorganic phosphate. The
discovery of phosphate-solubilizing bacteria and
fungi for solubilization of insoluble P compound in
Pikovskaya’s solid culture medium (Pikovskaya,
1948) opened the gate for today’s thorough
investigation to try out their application under
136
field conditions. Following this finding, several
strains of bacterial and fungal species have been
described and investigated in detail for their
inorganic phosphate-solubilizing capabilities (He
et al., 1997). Kunwar et al. (2018) have investigated
and reported that in vitro and greenhouse
experiments showed a significant improvement in
coffee seedlings treated with phosphate
solubilizing bacteria isolated from Coffea arabica
rhizosphere. The ability to solubilise P in a culture
medium is a potential activity but does not always
guarantee biofertilizer activity in the field.
Therefore, field experiments should be done with
the amendment of insoluble P source to test if a
potent bacterial and fungal isolates can enhance P
availability under field conditions and
consequently improve plant growth that indicates
their potential as biofertilizers. Fertile potting
media is fundamental components in soil fertility
management to establish healthy and vigorous
coffee seedlings in the nursery for better coffee
plantation in the future. To promote plant growth,
the soil must contain both macro and micro-
nutrients in the available forms to be easily taken
by roots. Among the bioelements, phosphorous is
a macro-nutrient which plays an important role in
plants in many physiological activities such as cell
division, photosynthesis, and development of
good root system and utilization of carbohydrate.
Although a plenty of phosphates are applied to
soil as chemical fertilizer, large portion of it
becomes rapidly immobilized and becomes
unavailable to plants in acidic and/or alkaline
soils. This fixed phosphate becomes mobilized by
the action of soil bacteria and fungi which are
termed as phosphate solubilizing microorganisms.
These phosphate solublizing microorganisms play
a pivotal role in solublizing phosphates and make
it to be available to the plants by allowing a
sustainable use of phosphate fertilizers (Sundara et
al., 2002). Coffee seedlings growth promotion can
be attributed to many mechanisms through which
rhizospheric bacteria and fungi contribute to
sustainable production. For sustainable crops
production, microbial phosphate solublization
efficacy and their application in agriculture are
getting greater attention due to their cost
effectiveness and eco-friendly use. Chemical
fertilizers pose health hazard and also are very
Reshid Abafita Abawari et al.
expensive as agricultural inputs. On the other
hand, application of phosphate solublizing
bacteria and fungi in the field condition has been
reported to increase crop yield. The current trends
of bio-fertilizers knowledge widely focused on
searching for better performing inoculants for
better yield gain in farming sector. Hence, it
necessitates a thorough research effort to obtain
rhizosphere competent inoculants of phosphate
solubilizing microbes (bacteria and fungi) for their
decisive use for on-farm application.
Phosphate-solubilizing bacterial isolates
(RCHVCB1, RScB1.19, RMaB2.11) and fungal isolates
(RSCF1.19, RCHVCF2, RLVCF2) which were isolated
and screened under in vitro conditions (Reshid
Abafita et al., 2021) were used for in vivo trials
under glasshouse condition prior to test in natural
nursery environment as bio-fertilizers. Hence, the
main objective of the present study was to test the
bacterial and fungal phosphate solublization
efficacy and their antibiotic susceptibility pattern
as well as to evaluate their suppressive effect
against coffee pathogenic Fusarium xyloriodes in
order to enhance coffee seedlings growth and
vigor.
MATERIALS AND METHODS
Location of study area
Glasshouse assay was carried out at Jimma
Agricultural Research Center (JARC) during
February to August of 2019. JARC is located at 363
km to the southwest of Addis Ababa. It is found at
7°40'47"N latitude and 36°49'47"E longitude. The
mean minimum and maximum temperature of
JARC are 26.2 and 11.3°C, respectively. The
elevation of the Center is 1,753 m above sea level
and it receives 1,529.5 mm average annual rainfall.
The total area of Jimma Zone is 18415 km2 and
located between latitudes 7018’N and 8056’N and
longitudes 35052’E and 37037’E. (Addis Tadsesse et
al., 2016).
Determination of antibiotic susceptibility
patterns of bacterial isolates
The microbial cultures viz., phosphate
solubilizing bacteria (PSB) were obtained from
Jimma University, College of Agriculture and
Veterinary Medicine, Microbiology Laboratory
(Reshid Abafita et al., 2021) and used to test their
SINET: Ethiop. J. Sci., 44(2), 2021
antibiotics susceptibility patterns. Susceptibility of
the isolates to antibiotics was performed by the
disc diffusion method as described by Bauer et al.
(1966) and Liasiet al. (2009) using commercially
available antibiotic discs (Oxoid). Briefly, the
purified bacterial colonies of the respective isolates
were inoculated into nutrient broth and incubated
at 36±2°C for 48 h. Sterile cotton swabs were
dipped into the bacterial broth suspension and
evenly spread on pre-dried surfaces of nutrient
agar plates. The inoculated plates were allowed to
dry before placing the diffusion discs containing
antibiotics. The plates were then incubated at 37°C
for 24 h. The commercial antibiotics used were
penicillin G (PG, 10 unit), Ceftazidime (Ce,10 μg),
Doxycycline (dxt, 30 μg), Erythromycine (E,15 μg),
Tetracycline (T,10 μg), and Vancomycin, (Van, 10
μg). After incubation of the plates, inhibition zone
diameters were measured by including the
diameters of the discs. The isolates were classified
as sensitive S (≥ 21 mm); intermediate, I (16-20
mm) or resistant, R (≤ 15 mm), respectively
according to Vlkováet al. (2006). For the purpose of
data analysis, the intermediates were considered
as sensitive (National Committee for Clinical
Laboratory Standards 2000; Rojo-Bezareset al.,
2006).
Co-culture test between phytobeneficial microbes
and pathogenic Fusarium xyloriodes
Co-culture assay was performed following
the method described by Santiago et al. (2017).
Each of the bacterial (RCHVCB1, RScB1.19, RMaB2.11)
and fungal (RSCF1.19, RCHVCF2 AND RLVCF2) isolates
were grown in nutrient agar medium at 30°C for at
least 3 days and then streaked perpendicularly on
freshly prepared nutrient agar medium; i.e., after
the first strain was allowed to grow at 30°C for 3
days. Thereafter, the second strain was streaked at
an angle of approximately 90° going outward from
the emerged colonies of the first strain and
incubated also at 30°C for 3 days. Finally,
photographic documentation of the agar plates
was obtained including those showing colony lines
and inhibition zones that appeared at the
intersection of the paired strains.
Inoculums preparation
The characterized and identified phosphate
solubilizing bacteria (PSB) and fungi (PSF) were
used to prepare the bio-inoculants (Reshid Abafita
137
et al., 2021). Inoculums of PSB were prepared in
Pikovskaya’s broth medium (Pikovskaya, 1948).
After multiplication of the selected elite isolates in
the PVK broth by incubating at 28±2°C under
shaking at 100 rpm for three days, the broth
culture was mixed with sterilized vermicompost
(VC) as carrier material. The viable count in the
inoculums was kept as 1x108 CFU/ml before mixing
with carrier material (VC) that was sterilized at
121°C and 15 psi pressure for an hour. Proper
water content of the sterile carrier material (VC)
was maintained and inoculated with broth cultures
of phosphate solublizing isolates (20 mL per 50 g
of VC) and was incubated at 28±2°C. For fungal
noculum preparation, phosphate solubilizing
fungal (PSF) isolates were mass cultured aseptically
in 90 mm diameter Petri plates each containing 15
mL of autoclaved PVK. The plates were incubated
at 28 ±2°C for 10 days. On the tenth day, spore
suspensions from the fungal isolates were
prepared by flooding the surface of the agar plates
with 10 mL sterile distilled water and the culture
surfaces were gently scraped using a sterile glass
rod to dislodge the spores. The spore suspension
was transferred separately to 500 mL flask
containing 400 mL sterile distilled water. Flasks
were shaken for 2 minutes to ensure that the
spores were properly mixed. The cultures were
filtered through Whatman No.42 filter paper into
sterile glass bottle. The spore suspension of 25 ml
(106 spores mL−1.) of fungal culture was used per
50 g of the sterilized carrier materials (VC) and
immediately stored at 4ºC until use. The mixed
and inoculated carrier material was evaluated for
plant growth promotion as bio-inoculants.
Test crop, potting media, pot volume and research
design
The test crop used was coffee variety 74110
which was coffee berry disease (CBD) tolerant, high
yield bearing and released variety and obtained
from Jimma Agricultural Research Center. This
variety is suitable for medium altitudes and
collected from Illu Aba Bora in western Ethiopia
(Bayetta Belachew et al., 2008). Endocarp
(parchments) was manually removed (Guimarães
et al., 2013) and the coffee seeds were surface
sterilized with 75% ethanol for 1 min followed by
1% sodium hypochlorite for 30 min with extensive
wash using sterile distilled water (Collavino et
al.,2010). The surface sterilized seeds of Coffea
138
arabica L. were placed in a sterile dish and mixed
with 4 ml of inoculants of bacterial and fungal
isolates and incubated for 6 hours at room
temperature (Mohamed and Almaroai, 2017).
Moreover, the seeds of Coffea arabica L. were also
inoculated with carrier based inoculums at a rate
of 15 g/100 g seeds (Mohamed and Almaroai,
2017) after moistening with 10 ml of sugar solution
(1 spoon table sugar per 10 ml water) and
thoroughly mixed with the seeds until the surfaces
were uniformly coated. The nursery experiments
were conducted from February to August of 2019
under controlled glasshouse condition at 28–32ºC
in plastic pots having 19 cm height with 21 cm top
and 18.5 cm bottom which were filled with 3.5 kg
of sand. Vermin-compost was applied at the rate of
20% of potting media per pot (Reshid Abafita et al.,
2014). The inorganic fertilizer treatment (4 g
DAP/pots) was mixed with sand before sowing
coffee seeds in the media. The fertilizer application
rates for the inorganic P treatments were
calculated from the published rates of inorganic
fertilizer recommendations for young coffee in the
field which was 1t ha-1 per year (Loga and Biscoe,
1987). Seedlings were thinned when they attained
two pairs of true leaves and one uniformly
growing seedling was left. The seedlings were
grown in a glasshouse until the emergence of
seven pairs of true leaves at a temperature of 28–
32ºC and 85% relative humidity. After completion
of trial, the plants were up-rooted and washed
thoroughly with water and several parameters
such as shoot and root length, leaf numbers, leaf
area, stem girth, fresh and dry weight of the whole
plant, NPK up-take of the leaves were measured
using standard procedures. The glasshouse assay
was done with completely randomized design
(CRD) in three (3) replications and designed with
Table 1.Effects of different antibiotics on the tested isolates.
Vancomycin Ceftazidime
Isolate code (Van) (Ce)
10 μg 10 μg
RCHVCB1 R R
RScB1.19 R R
RMaB2.11 R R
Legend
R=resistance
Reshid Abafita Abawari et al.
four groups. The first two groups were designed
with fourteen (T1 -T14,) and twenty (T1-T20)
treatments for both single and co- inoculation with
and without P fertilizer. The second two groups
were designed with nine (T1-T9) and twelve (T1-T12)
treatments for both single and co-inoculation with
vermicompost. The experiment was commenced
under protected glasshouse condition using sand
as the potting medium for the experiments.
Data Analysis
Data were collected in replicates and
analyzed using SAS Statistical Package Version
8.5.0) 2010, Origin Lab Corporation. To determine
the effects due to inoculation, Analysis of Variance
at the 0.05 level using SAS was done and means
were separated using Duncan multiple Range Test
at 0.05 level. Data obtained from the different
treatments were presented in the form of tables
using Microsoft Excel 2007.
RESULTS
Determination of antibiotic susceptibility
patterns of the isolates
Resistance to antibiotics is a threat phenomenon in
medical microbiology. However, in agriculture, it
is considered as advantageous for bio inoculants to
persist and well established in the soil by resisting
agro-chemicals such as pesticides, herbicides and
chemical fertilizers. Accordingly, all of the potent
isolates showed resistance to antimicrobials tested
(Table 1; Figure 1). The potent isolates selected for
bio-inoculants (RCHVCB1, RScB1.19, RMaB2.11) were
100% resistant to all the six tested antimicrobials
(Table 1).
Doxycycline Erythromycine penicilin G Tetracycline
(dxt) (E) (PG) (T)
30 μg 15 μg ,10 unit 10 μg
R R R R
R R R R
R R R R
SINET: Ethiop. J. Sci., 44(2), 2021
Resistance Sensitive
Figure 1. Pictorial presentation of antimicrobial susceptibility test results.
Co-culture test between Bacteria, fungi and
pathogenic Fusarium xyloriodes
In addition to phosphate solubilization,
biocontrol is one of the most desirable traits for
inoculants. To assess this we tested co-cultures of
both fungi and bacteria against pathogenic
Fusarium xyloriodes to see whether they could co-
exist or antagonistic to one another on agar plates
(Figure 2). The growth of Fusarium xyloriodes was
slightly inhibited by RSCF1.19 (Figure 2a).
However, a clear evidence of growth inhibition
was obtained when RSCF1.19 was streaked
perpendicular to RLVCF2. A sufficient inhibition
zone was created by RSCF1.19, which suppressed
the growth of RLVCF2 (Figure 2c), whereas the
growth of RSCF1.19 was inhibited by RCHVCF2
(Figure 2b). The growth of both RLVCF2 and
RCHVCF2 was not suppressed when both co-
cultured on the same plates (Figure 2d), which
showed the co-existence of both isolates on the
same plate with no trace of growth inhibition at
the center where the two isolates crossed each
other (Figure 2d). On the other hand, all the
bacterial and fungal isolates showed no growth
inhibition between each other (Figure 2e,f).
a. co-inoculation of RSCF1.19(F1) and Fusarium
xyloriodes ,b. co-inoculation of RSCF1.19(F1) and
RCHVCF2 (F2),c. co-inoculation of RSCF1.19(F1)and
RLVCF2(F3),d. co-inoculation of RCHVCF2 (F2) and
RLVCF2(F3), e. co-inoculation of Fusarium xyloriodes
and RCHVCB1(B1),f. co-inoculation of Fusarium
xyloriodes and RMaB2.11(B3).
139
Resistance Resistance
Effect of bacterial and fungal inoculation on
coffee seedlings
The individual application of phosphate
solubilizing bacteria (RCHVCB1, RScB1.19, RMaB2.1)
and fungi (RSCF1.19, RCHVCF2, RLVCF2) isolates had
effect on growth of coffee seedlings (Table 2).
Combined inoculation of bacteria and fungi in the
presence of inorganic chemical phosphate with the
treatments of RCHVCB1+P, RScB1.19+P, RMaB2.11+P,
RCHVCF2+P and RLVCF2+P, respectively showed
consistently increased growth parameters over the
un-inoculated coffee seedlings. The greenhouse
experiment results showed that inoculation with
phosphate solubilizing fungi (RSCF1.19+P)
significantly (p≤0.05) improved the growth
parameters such as shoot height, stem diameter,
leaf number, leaf area, fresh and dry weights
compared to treatments such as RCHVCB1+P,
RScB1.19+P, RMaB2.11+P, RCHVCF2+P, RLVCF2+P
and the controls (without bacterial and fungal
inoculums and without inorganic phosphate
fertilizer) (Table 2). The results showed that mean
values of number of leaves were not significantly
(p>0.05) different among all the treatments except
RSCF1.19+P. Generally, the effect of inoculation
was comparable or greater than the chemical
fertilizer and negative control. Isolate RCHVCF2+P
showed significant effect on shoot length and
fresh shoot biomass, dry shoot biomass but no
effect on leaves, leaf area, root fresh and dry
biomass as well as on stem girth of coffee seedling
(Table 2).
140 Reshid Abafita Abawari et al.

a b c

d e f
Figure 2. Co-culture test between pathogenic Fusariumxyloriodes, fungal and bacterial isolates .

.
The morphological growth performance of vermicompost showed suppressive characteristics
experimental plants was demonstrated in Figure 3. with no any seedlings emergency (Figure 3).
However, all the treatments combined with

-ve Control Bioinoculant +Ve control

Figure 3. Morphological growth performance of experimental plants, VC=Vermicompost,-Ve= Negative control ,+Ve=Positive
control

Effects of co-inoculation on the growth of coffee
seedlings
RSCF1.19 obtained from coffee rhizosphere
was identified as having the highest potential as
bio-inoculant phosphate solublizer. The dual
inoculation of RSCF1.19 and RCHVCB1 in the
presence of inorganic P source significantly
improved the growth of coffee seedlings over that
of the un-inoculated control and other tested
isolates in terms of plant shoot height, root height,
stem diameter, leaf number, leaf area, fresh
(p≤0.05) and dry weights (p≤0.05) (Table 3).
Accordingly, the increased levels of growth
parameters over the other treatments indicated
that the combined inoculation of fungi and
bacteria isolates (RSCF1.19 and RCHVCB1) enhanced
the growth of the plant by solublizing the applied
inorganic phosphate and posed better up-take by
plant roots. Moreover, combined inoculation of
RSCF1.19 and RMaB2.11 in the presence of inorganic
P significantly caused enhanced plant growth in
terms of root height, stem diameter, leaf number,
leaf area, fresh weight (p≤0.05) and dry weights
(p≤0.05) over treatments without inorganic
phosphate and both negative and positive controls
(Table 3).
 SINET: Ethiop. J. Sci., 44(2), 2021 141

Table 2.Growth response of coffee seedlings to sole PSF and PSB inoculants under glasshouse condition.

Treatments Shoot length/plant Root length/plant Shoot fresh Root fresh Shoot dry Root dry No of Stem Girth Leaf area
(cm) (cm) weight (g) weight (g) weight (g) weight (g) leaves (mm) (sq.cm)
T1 RCHVCB1 10.90cde 12.50cd 0.97c 0.14bc 0.48bc 0.08cd 8.00b 2.33bc 0.36i
T2 RScB1.19 16.00abc 12.00e 2.70bc 0.60ab 0.89bc 0.25ab 10.67ab 3.33ab 1.20d
T3 RMaB2.11 10.83cde 13.67ab 0.97c 0.14bc 0.36c 0.08d 8.00b 2.33bc 0.55gh
T4 RSCF1.19 13.83bcde 13.00bc 2.33bc 0.37abc 0.89bc 0.18abcd 9.00b 3.33ab 0.90e
T5 RCHVCF2 9.17de 12.17dde 1.03c 0.17bc 0.27c 0.09cd 8.00b 2.33bc 0.59gh
T6 RLVCF2 11.83bcde 12.67c 1.73bc 0.33abc 0.57bc 0.15bcd 9.33b 2.67abc 0.86f
T7 RCHVCB1+P 13.67bcde 12.33d 1.70bc 0.20bc 0.53bc 0.16bcd 10.00b 2.67abc 0.96de
T8 RScB1.19+P 10.67cde 13.50ab 1.23c 0.20bc 0.39c 0.09bcd 8.67b 2.00c 0.75fg
T9 RMaB2.11+P 12.67bcde 13.83a 1.53bc 0.20bc 0.39c 0.11bcd 10.00b 2.33bc 0.94de
T10 RSCF1.19 +P 21.17a 10.83def 5.13a 0.77a 1.71a 0.32a 13.00a 3.67a 2.78a
T11 RCHVCF2+P 16.53ab 11.83de 3.73ab 0.57abc 1.22ab 0.24abc 10.67ab 2.67abc 2.55b
T12 RLVCF2+P 14.50bcd 13.33abc 3.13abc 0.50abc 0.77bc 0.19abcd 10.00b 3.33ab 2.32c
T13 -ve control 8.67e 8.00f 0.97c 0.10c 0.39c 0.07d 8.00b 2.00c 0.52gh
T14 +ve control 9.00de 9.00ef 1.17c 0.30abc 0.41c 0.08cd 9.33b 3.00abc 0.86f
CV 26.05 11.10 65.13 85.68 64.76 63.54 17.10 23.55 79.55
Mean 12.82 12.02 2.02 0.33 0.66 0.15 9.48 2.71 1.19
T test (LSD) (0.05) 5.60 2.06 2.21 0.47 0.72 0.16 2.72 1.07 1.07

*Means followed by the same letter(s) in each column are not significantly different at α=5%, PSF=phosphate solubilizing fungi, PSB=phosphate solubilizing bacteria, *P=Chemical
phosphate, -Ve=negative control, +Ve=positive control

Table 3. Growth response of coffee seedlings to dual inoculation of PSF and PSB isolates under glasshouse condition.

Treatments Shoot Root Shoot fresh Root fresh Shoot dry Root dry No of Stem girth Leaf area
length/plant length/plant weight (g) weight (g) weight (g) weight (g) leaves (mm) (sq.cm)
(cm) (cm)
T1 RCHVCB1+ RSCF1.19 11.90cde 13.50ab 1.53cdef 0.21abcd 0.52bcde 0.13b 8.67cbd 3.00a 1.03abcd
T2 RCHVCB1+RCHVCF2 12.70cd 12.50abc 1.73bcdef 0.40ab 0.58bcde 0.16b 9.33abcd 3.00a 0.40d
T3 RCHVCB1+ RLVCF2 11.33cde 9.67bcd 1.73bcdef 0.33abcd 0.60bcde 0.15b 8.00cde 2.67ab 1.04abcd
T4 RScB1.19+ RSCF1.19 7.67f 13.33ab 0.67f 0.07d 0.25e 0.06b 7.33de 2.33ab 0.40d
T5 RScB1.19+ RCHVCF2 14.00b 10.17abcd 2.13bcdef 0.23abcd 0.63bcde 0.14b 10.67ab 2.33ab 1.21abcd
T6 RScB1.19+ RLVCF2 10.80cde 13.50ab 1.23cdef 0.17bcd 0.42cde 0.12b 8.67cbd 2.33ab 0.61cd
T7 RMaB2.11+ RSCF1.19 14.40bc 13.00abc 1.70bcdef 0.27abcd 0.58bcde 0.14b 10.33abc 2.67ab 0.86bcd
T8 RMaB2.11+RCHVCF2 13.27bcd 12.17abc 2.40abcde 0.33abcd 0.74abcde 0.15b 10.67ab 3.00a 1.99ab
T9 RMaB2.11+ RLVCF2 12.67cd 13.17ab 1.87bcdef 0.30abcd 0.51bcde 0.13b 8.67bcd 2.67ab 1.04abcd
T10 RCHVCB1+RSCF1.19 +P 16.50a 12.50abc 3.80a 0.47a 1.21a 0.19ab 11.33a 3.00a 2.13a
T11 RCHVCB1+ RCHVCF2+P 11.67cde 13.33ab 1.23cdef 0.14bcd 0.44cde 0.10b 8.67bcd 2.33ab 0.47d
T12 RCHVCB1+ RLVCF2+P 12.17cd 14.17a 1.47cdef 0.17bcd 0.48bcde 0.36a 10.00abc 2.33ab 0.94bcd
T13 RScB1.19+ RSCF1.19 +P 11.73cde 13.83a 1.13cdef 0.17bcd 0.37de 0.09b 9.33abcd 2.00b 0.46d
T14 RScB1.19+ RCHVCF2 +P 9.50de 10.17abcd 1.37cdef 0.17bcd 0.47cde 0.09b 10.00abc 2.00b 0.47d
T15 RScB1.19+ RLVCF2+P 13.83bcd 13.33ab 2.47abcd 0.33abcd 0.79abcd 0.15b 10.67ab 2.00b 1.41abcd
 142 Reshid Abafita Abawari et al.

T16 RMaB2.11+ RSCF1.19+P 14.83bc 13.17ab 3.23ab 0.47a 0.97ab 0.19ab 11.33a 3.00a 1.73abc
T17 RMaB2.11+ RCHVCF2+P 14.00b 12.83abc 2.67abc 0.37abc 0.87abc 0.15b 10.00abc 2.67ab 1.80ab
T18 RMaB2.11+ RLVCF2+P 11.17cde 11.67abcd 0.90ef 0.13bcd 0.30de 0.09b 6.00e 2.00b 1.24abcd
T19 -ve control 8.67de 8.00d 0.97def 0.10cd 0.39cde 0.07b 8.00cde 2.00b 0.52d
T20 +ve control 9.00de 9.00cd 1.17cdef 0.30abcd 0.41cde 0.08b 9.33abcd 3.00a 0.86bcd
CV 20.27 20.04 52.63 68.79 51.50 82.58 15.81 20.75 66.64
Mean 12.13 12.15 1.77 0.26 0.58 0.14 9.35 2.52 1.03
T test(LSD) (0.05) 2.03 4.02 1.54 0.29 0.49 0.19 2.44 0.86 1.14
*Means followed by the same letter(s) in each column are not significantly different at α=5%, *P=Chemical phosphate-Ve=negative control, +Ve=positive control

Table 4. Chemical properties of potting medium after taking data destructively at single inoculation treatments and nutrient uptake.

Treatments pH(1:2.5) OC (% ) Ex.Acidity CEC Available nutrients Nutrient uptake by plants

(meq/100g (meq/100g) AvailableP (ppm) TN (% ) K (ppm) AvailableP (ppm) TN (% ) K (ppm)
T1 RCHVCB1 6.50 1.65 0.30 20.70 10.37 0.02 50.00 78.90 0.30 1680.20
T2 RScB1.19 6.51 1.60 0.31 19.78 11.33 0.03 50.01 76.91 0.30 1580.10
T3 RMaB2.11 6.41 1.61 0.29 20.68 10.31 0.02 49.02 77.89 0.29 1650.30
T4 RSCF1.19 6.40 1.65 0.30 18.78 10.36 0.01 50.03 78.80 0.31 1681.29
T5 RCHVCF2 6.52 1.67 0.33 21.00 10.35 0.02 48.00 76.90 0.30 1580.20
T6 RLVCF2 6.50 1.64 0.30 19.70 11.30 0.02 50.01 75.70 0.31 1670.30
T7 RCHVCB1+P 5.30 1.93 0.38 18.14 53.50 0.04 75.05 124.04 0.32 3282.84
T8 RScB1.19+P 5.46 1.96 0.43 19.36 41.56 0.03 75.97 120.00 0.34 3182.85
T9 RMaB2.11+P 5.19 1.66 0.57 17.6 47.20 0.03 76.35 121.02 0.33 3282.80
T10 RSCF1.19 +P 5.10 1.86 0.38 17.58 88.64 0.04 75.78 214.20 0.33 3312.00
T11 RCHVCF2 +P 5.30 1.87 0.51 18.12 71.71 0.03 76.30 210.20 0.32 3212.01
T12 RLVCF2+P 5.28 1.53 0.41 18.54 56.01 0.05 73.42 212.20 0.31 3210.20
T13 -ve control 6.18 1.34 0.35 17.9 10.54 0.05 50.33 66.91 0.20 1270.10
T14 +ve control 5.18 1.64 0.46 17.48 34.82 0.04 53.76 121.02 0.25 3082.80
T15 VC 8.17 2.03 0.41 20.86 88.06 0.06 107.59
*p=Chemical phosphate, -Ve=negative control, +Ve=positive control, VC=Vermicompost, OC=organic carbon, EX=exchangeable,
SINET: Ethiop. J. Sci., 44(2), 2021
Similarly, the co-inoculations of isolates
RMaB2.11 and RCHVCF2 as well as RMaB2.11 and
RLVCF2 in the presence of inorganic phosphate
showed a better increase in plant growth
parameters such as plant shoot height, root
height, stem diameter, leaf number, leaf area,
fresh (p≤0.05) and dry weights (p≤0.05; Table3)
over the controls, but lower plant growth
parameters was obtained from both fungal and
bacterial inoculations in the absence of inorganic
phosphate indicating the lack of sufficient
inorganic phosphate to be solubilized in potted
medium.
Nutrient status of potting medium
After collecting the necessary data, the
potting medium was analyzed for the
physicochemical properties (pH, OC, Ex.
Acidity, CEC) and nutrient status (Table 4 and 5).
The potting medium reaction was near neutral
in the treatments containing bio-inoculants only
and in the negative control. But, there was a
slightly decreased pH value in the treatments
containing bio-inoculants and P fertilizer in both
singly and co-inoculated potting media. There
was no emergence of seedlings in the treatments
amended with vermicompost (T1-T9) and (T1-T12)
since the potting medium reaction was near
alkaline (Table 4 and 5; Figure 3).
Percent of organic carbon and the Cation
Exchange Capacity (CEC) of the potting soil was
the same in all the treatments by receiving bio-
inoculants amended with inorganic P fertilizer
as well as treatments without inorganic P
fertilizer (Table 4 & 5). However, available P
was to some extent increased in the treatments
containing bio-inoculants amended with
inorganic P fertilizer compared to treatments
with only bio-inoculants. But, percent organic
carbon and available P were high in treatments
inoculated with fungal bio-inoculants compared
to bacterial bio-inoculants in the presence of P
fertilizer. Similarly, available K was high in the
treatments containing bio-inoculants amended
with inorganic P fertilizer compared to
treatments with only bio-inoculants but
available K was higher in the treatments
containing co-inoculated bio-inoculants both
with and without P fertilizer amendments than
singly inoculated treatments (Table 4 and 5).
143
Nutrient uptake by coffee seedlings
The nutrient uptake of the coffee seedlings
grown in potting medium is presented in Tables
4 and 5. Although not significant, the highest P
and K-uptake by shoot of coffee seedlings
grown in potting medium was generally
observed with treatments that received bio-
inoculants and inorganic chemical phosphate
fertilizer compared to un-inoculated treatments
and negative control. Moreover, treatments that
received fungal bio-inoculants in the presence of
inorganic chemical fertilizer showed increased P
and K-uptake by coffee seedlings compared to
bacterial bio-inoculants amended with inorganic
chemical fertilizer treatments (Table 4).
Inoculations of bacteria and fungus alone or in
combination produced higher uptake of
available N, P and K compared to the control
(Table 5). Thus, the higher growth parameters
observed under fungal treatments can be
attributed to availability and uptake of balanced
and higher quantities of nutrients to coffee
seedlings through inorganic phosphate fertilizer
as well as bio-inoculants consortia compared to
treatments that received bio-inoculant without
inorganic phosphate fertilizer and negative
control.
Based on the correlation (r) analysis during
single inoculation of isolates, P-uptake was
greatly correlated with plant height (r=0.457*),
root length (r=0.529*), shoot fresh weight
(0.550*) and shoot dry weight (0.478*; Table 6).
Similarly, K uptake was correlated with root
length, shoot dry weight, number of leaves per
plant and stem girth (r=0.494*, 0.570**, 0.488*
and 0.502*, respectively) and was not
significantly (p>0.05) related with other growth
parameters (Table 6). Non-significant
relationships were observed with total N uptake
and all the growth parameters except in root
fresh weight and leaf area (r= 0.463* and 0.794**;
respectively). However, during dual inoculation,
only plant height and leaf area were greatly
correlated with P N uptake (r=0.448**, 0.682**
and 0.457**, 0.983**, respectively; Table 7) and
non-significant relationships were observed
with all other growth parameters.
144 Reshid Abafita Abawari et al.

Table 5. Chemical properties of potting medium after taking data destructively at co-inoculation treatments and nutrient uptake.

Treatments pH OC Ex.Acidity CEC Available nutrients Nutrient uptake by plants

(1:2.5) (%) (meq/100g) (meq/10 Available TN K (ppm) Available P TN K (ppm)
0g) (ppm) (% ) (ppm) (%)
T1 RCHVCB1+ RSCF1.19 6.50 1.67 0.24 18.10 12.33 0.03 70.94 78.80 0.30 1680.20
T2 RCHVCB1+ RCHVCF2 6.49 1.60 0.23 18.11 13.33 0.02 63.94 76.91 0.30 1580.10
T3 RCHVCB1+ RLVCF2 6.49 1.61 0.20 17.18 11.33 0.04 53.94 77.89 0.29 1650.30
T4 RScB1.19+ RSCF1.19 6.48 1.67 0.25 19.18 13.33 0.03 73.94 78.80 0.31 1681.29
T5 RScB1.19+ RCHVCF2 6.49 1.54 0.21 18.18 13.30 0.02 50.94 76.90 0.30 1580.20
T6 RScB1.19+ RLVCF2 6.49 1.67 0.20 17.18 13.13 0.02 71.94 75.70 0.31 1670.30
T7 RMaB2.11+ RSCF1.19 6.44 1.68 0.20 17.10 13.33 0.02 73.94 78.80 0.31 1680.30
T8 RMaB2.11+ RCHVCF2 6.51 1.60 0.22 18.12 11.33 0.03 70.94 77.90 0.26 1485.20
T9 RMaB2.11+ RLVCF2 6.49 1.62 0.23 17.18 13.33 0.02 71.94 76.90 0.30 1670.31
T10 RCHVCB1+ RSCF1.19 +P 5.40 1.73 0.51 19.38 73.53 0.03 100.67 120.81 0.39 3274.22
T11 RCHVCB1+ RCHVCF2+P 5.47 1.58 0.43 19.08 38.96 0.05 80.51 110.80 0.29 3174.22
T12 RCHVCB1+ RLVCF2+P 5.45 1.60 0.38 18.24 39.37 0.03 99.27 111.71 0.29 3274.22
T13 RScB1.19+ RSCF1.19 +P 5.36 1.69 0.46 18.08 66.96 0.05 101.85 122.53 0.47 3284.00
T14 RScB1.19+ RCHVCF2+P 5.41 1.59 0.38 19.20 59.76 0.02 95.12 123.50 0.47 3184.04
T15 RScB1.19+ RLVCF2+P 5.48 1.52 0.41 19.28 50.37 0.03 102.48 121.51 0.47 3244.01
T16 RMaB2.11+ RSCF1.19+P 5.21 1.71 0.40 18.56 59.77 0.04 94.22 112.20 0.31 3157.30
T17 RMaB2.11+ RCHVCF2+P 5.14 1.45 0.39 18.32 62.99 0.04 80.44 109.22 0.41 3057.31
T18 RMaB2.11+ RLVCF2+P 4.86 1.83 0.59 18.12 50.96 0.04 81.95 110.22 0.31 3237.30
T19 -ve control 6.18 1.34 0.35 17.9 10.54 0.05 50.33 69.91 0.20 1270.10
T20 +ve control 5.18 1.64 0.46 17.48 34.82 0.04 52.76 123.02 0.23 3082.80
*p=chemical phosphate, -Ve=negative, +Ve=positive

Table 6. Correlation coefficient for plant growth parameters and nutrient up take characters of coffee seedlings during single inoculation.

characters Shoot Root Shoot Root fresh Shoot dry Root dry No of Stem Girth Leaf area Available TN (% ) K
length/plant length/p fresh weight (g) weight (g) weight leaves (mm) (sq.cm) P (ppm) (ppm)
(cm) lant (cm) weight (g) (g)
Shoot length/plant(cm) 1
Root length/plant(cm) -0.149ns 1
Shootfresh weight(gm) -0.268ns 0.920** 1
Root fresh weight(gm) -0.196ns 0.967** 0.913** 1
Shoot dry weight(gm) -0.226ns 0.117ns 0.176ns 0.109ns 1
Root dry weight(gm) -0.011ns 0.881** 0.785** 0.810** 0.113ns 1
No of leaves -0.095ns 0.811** 0.701** 0.838** 0.057ns 0.745** 1
Stem Girth (mm) -0.142ns 0.862** 0.886** 0.863** 0.212ns 0.888** 0.721** 1
Leaf area(sq.cm) -0.040ns 0.337ns 0.430ns 0.281ns 0.638** 0.538* 0.187ns 0.281ns 1
AvailableP(ppm) 0.457* 0.529* 0.391ns 0.478* 0.215ns 0.550* 0.395ns 0.410ns 0.379ns 1
TN (% ) 0.183ns 0.127ns 0.163ns 0.021ns 0.463* 0.367ns 0.061ns 0.319ns 0.794** 0.400ns 1
K (ppm) -0.067ns 0.494* 0.441ns 0.570** -0.234ns 0.443ns 0.488* 0.502* 0.182ns 0.274ns 0.137ns 1
Legend: ns= not significant
 SINET: Ethiop. J. Sci., 44(2), 2021 145

Table 7. Correlation coefficient for plant growth parameters and nutrient up take characters of coffee seedlings during dual inoculation.

characters Shoot Root Shoot fresh Root fresh Shoot dry Root dry No of Stem Girth Leaf area Available TN (% ) K
length/plan length/plan wt (g) wt (g) wt (g) wt (g) leaves (mm) (sq.cm) P (ppm) (ppm)
t (cm) t (cm)
Shoot length/plant(cm) 1
Root length/plant(cm) -0.281ns 1
Shoot fresh wt(gm) -0.213ns 0.891** 1
Root fresh wt(gm) -0.233ns 0.966** 0.917** 1
Shoot dry weight(gm) -0.159ns 0.524** 0.410* 0.499** 1
Root dry weight(gm) -0.089ns 0.507** 0.574** 0.492** 0.318ns 1
No of leaves -0.340ns 0.812** 0.683** 0.792** 0.413* 0.229ns 1
Stem Girth (mm) -0.029ns 0.731** 0.723** 0.720** 0.447** 0.372* 0.565** 1
Leaf area(sq.cm) 0.406* -0.069ns 0.208ns 0.012ns -0.018ns 0.225ns -0.170ns 0.162ns 1
Available P(ppm) 0.448** 0.057ns 0.195ns 0.113ns 0.154ns 0.264ns -0.180ns 0.097ns 0.682** 1
TN (% ) 0.457** -0.074ns 0.191ns -0.010ns -0.012ns 0.184ns -0.191ns 0.196ns 0.983** 0.645** 1
Available K(ppm) 0.015ns 0.249ns 0.279ns 0.286ns -0.242ns 0.160ns 0.245ns 0.321ns 0.183ns -0.047ns 0.169ns 1

Legend: ns= not significant, ppm=parts per million, wt= weight

146
DISCUSSION
Determination of antibiotic susceptibility
patterns of bacterial isolates
Isolates with multiple antibiotic resistances have
greater advantage in establishing themselves as
biofertilizers in natural soil conditions as well as
any new ecological niche. Such isolates with high
level of intrinsic antibiotic resistance have their
own significance in establishing themselves in the
rhizosphere with greater capability when used as
bio-fertilizer in natural soil conditions. Antibiotic
sensitivity/resistance assay revealed that from the
all twelve isolates RCHVCB1, RScB1.19 and RM aB2.11
showed a very high level of resistance to all the 6
antibiotics and said to be advantageous for
establishing themselves in stressful environment.
Resistance to antibiotics is acquired by a change in
the genetic make- up of microorganisms which can
occur by either a genetic mutation or by transfer of
antibiotic resistant genes among microorganisms
(Spain and Alm, 2003). Similar investigations on
antibiotic resistance have been reported by Wani
and Irene (2014).
In-vitro co-culture test between bacteria, fungi
and pathogenic Fusariumxyloriodes
In addition to phosphate solubilization,
biocontrol is one of the most desirable traits for
inoculants. Therefore, in the present study, the co-
existence of isolates RSCF1.19 with RCHVCB1 was
confirmed by the absence of inhibition zones at the
intersection of the two colonies on the same plate
medium and this indicates the possibility for co-
colonization on the roots of coffee seedlings. These
results revealed that the combination of two
bacterial and two fungal phosphate solublizing
isolates could colonize coffee roots with inherent
ability to solubilize inorganic phosphate in order
to release plant growth-promoting hormones and
easily establish themselves in the eco-
physiologically stressed environments. In the
present study, synergy between the compatible
isolates (RCHVCB1 with RSCF1.19 +P) was evidenced
by the slightly increased growth parameters in the
coffee plant seedlings. Our results also in
agreement with the findings of Pandey et al.(2012)
who demonstrated that microbial diversity in soil
gives rise to a stable ecosystem through the
synergistic interactions of compatible microbes,
resulting in increased plant productivity.
Reshid Abafita Abawari et al.
Effect of bacterial and fungal inoculation on
coffee seedlings
A significantly increased plant growth in
terms of plant height, root length, stem diameter,
leaf number, leaf area, fresh weight and dry
weights with co-inoculation of RCHVCB1+ RSCF1.19
+P as well as single inoculation of fungus
(RSCF1.19+P) compared to both positive and
negative control. The higher growth parameters
observed under bioinoculants combined in the
presence of inorganic phosphate can be attributed
to the activity of phosphate solubilizing microbes
in rhizosphere that could release soluble P and also
through production of IAA, ACC deaminase,
siderophore, antibiotics and HCN compared to the
control (Zaidiet al.,2014). This revealed that the
potent microbes are not only phosphate
solubilizers but also promote plant growth
through the production of plant growth hormones
(Bottiniet al., 2004). Similarly, increase in biomass
due to treatment with phosphate solubilizing
bacteria has been reported in maize (Hameedaet
al., 2008). However, among inoculants, single
inoculation of fungal inoculums, RSCF1.19 in the
presence of inorganic P showed significantly
increased superior efficiency in promoting all the
measured plant growth parameters except in root
length compared to the bacterial inoculums. This
significant increase in growth parameters of the
plant is believed to be because of the fact that fungi
have a greater potential to solublize insoluble
phosphate compounds than bacteria and easily
establish in the soil (Nahas, 1996;
Mahadevamurthy et al.,2016). Next to fungal
inoculums, the three bacterial isolates showed
better plant growth parameters when combined
with P source than inoculants without P sources. A
better increase in plant growth in terms of plant
height, root length, stem diameter, leaf number,
leaf area, fresh weight and dry weights with
inoculation of Pseudomonus sp was also
documented by Mamta et al.(2010). Research
results documented by Prasad et al. (2014)
significantly increased coffee seedlings growth
when treated with Azosprillum sp, Pseudomonas
fluorescens , phosphate solubilizing bacteria (PSB)
and arbuscular mycorrhizal fungi (AMF).
Single application of chemical phosphate
without solublizers (+Ve control) did not
significantly improve plant growth parameters
(shoot and root dry weight) in coffee seedlings.
SINET: Ethiop. J. Sci., 44(2), 2021
The poor growth of seedlings observed in the
treatments inoculated with bio-inoculants but
without inorganic phosphate compared to the
seedlings under bio-inoculated treatments
combined with P sources that could be due to the
lack of adequate inorganic phosphate which is
essential to be solubilized and be available for
uptake by plants for the establishment of growth
parameters. In this study, vermicompost was used
as carrier material to enhance easy establishment
of bio-inoculants in the potting medium due to its
contribution to better growth of seedlings in all
combined treatments. This confirms that organic
matter is a predictable ingredient of potting
mixture when bioinoculants are used for raising
coffee seedlings even when the soil under
investigation in the potting mixture is deficient in
organic matter. However, all the treatments
amended with vermicompost showed suppressive
effect with no emerged seedlings. The suppressive
characteristics can be attributed to the high pH
value of the potting medium (vermicompost)
(pH>7.5) (Reshid Abafita et al., 2014).
Nutrient status of potting medium and uptake by
coffee seedlings
Bacterial and fungal phosphate solublizers,
which could solubilize insoluble phosphate
compounds by producing organic acids and
phosphatase enzymes improve P availability in
soils (Park et al., 2010) and stimulate growth due to
mineral uptake by plants. Consistently, our results
are in agreement with findings from other
researchers that indicate the importance of
selection and integration of the most efficient
bacterial and fungal P solublizers as bio-inoculants
in the presence of chemical phosphate fertilizer to
improve crop mineral nutrients in nutrient-
deficient soils.
Co-inoculation of bacterial and fungal
RSCF1.19 and RCHVCB1) isolates could promote
mineral uptake and growth of coffee seedlings.
Availability of P in the soil is crucial for facilitated
uptake and easy utilization of it by plant roots
(Vessey, 2003). Hence, higher available P due to
the addition of inorganic P-fertilizer and
solubilization with inoculated PSB and PSF might
cause an enhancement of P uptake and plant
growth. Generally, results from the present study
and others’ findings suggest that co-inoculation of
plant growth promoting microbes with other
147
different beneficial properties could be the future
trends of bio-fertilizers application for sustainable
crop production. It is likely that phosphate
solubilizing microbes (bacteria and fungi) might
have helped plant root development due to their
ability to produce phytohormones in the plant
rhizosphere to enhance absorption of water and
acquisition of nutrients such as phosphate by plant
roots (Bareaetal., 2005). From these results, we can
conclude that inoculation of coffee seeds with
efficient bio-inoculants significantly enhanced
plant growth in glasshouse experiments. In the
present study, the pronounced plant growth by
these isolates could be attributed to the production
of IAA, NH3, HCN, N-fixation and solubilization of
phosphate. These results are in concurrence with
the findings of many authors who reported
production of phytohormones and phosphate
solubilization by soil microbes (Dhurve et al.,2017).
The analytical data of nutrient status in the
potting medium clearly indicates more nutrient
availability in treatments containing bio-inoculants
and inorganic P fertilizer compared to the
treatments with only bio-inoculants. On the other
hand, a decrease in pH in treatments containing
bio-inoculants amended with inorganic P fertilizer
could be due to acid production by potent
microbes during P solubilization (Gaind, 2016).
Percent of organic carbon and the cation exchange
capacity (CEC) of the potting sand were the same in
all the treatments that received bio-inoculants in
the presence of inorganic P fertilizer as well as
treatments without inorganic P fertilizer. These
might be due to exclusion of organic amendments
from the treatments which may build organic
matter in the potting medium. Generally,
beneficial rhizospheric and nonrhizospheric
phosphate solubilizing microbes enhance growth
through synthesizing particular compounds for
plants or by facilitating the uptake of particular
nutrient from the soil or by preventing and
protecting the plants from pathogens (Yadav et al.,
2011).The results of our study revealed that the
added microbial inoculants have the advantage of
making nutrients available in balanced and
adequate quantities from the potting medium as
seen in the present experiments. PSF and PSB
inoculation increased total N and P concentration
in the tissues of coffee seedlings which has a
positive correlation with seedlings growth
parameters such as plant height, root length, stem
148
diameter, leaf number, leaf area, fresh weight and
dry weight. These increments were attributed to
the inherent bacterial and fungal growth-
promoting abilities through diverse mechanisms.
Nutrient uptake by coffee seedlings depends on
availability of nutrient. Some soil fungi and
bacteria are the most important phosphate
solublizers. Inoculation of these microbes helped
plant to take phosphorus compare to the control.
The increase in phosphorus uptake by inoculation
of microbes could be attributed to availability and
uptake of balanced and higher quantities of
phosphorous to coffee seedlings through inorganic
phosphate fertilizetion as well as consortia of bio-
inoculants compared to treatments received only
bio-inoculants without inorganic phosphate
fertilizer and negative control. Son etal. (2006) have
reported increased seed P content by phosphate
solubilizing microorganisms. Our results are in
agreement with reports of Jilani et al. (2007) that a
combination of 50% of recommended chemical
fertilizer and biofertilizer gave equal yield as 100%
of recommended chemical fertilizers. It is
concluded that bioavailability of precipitated
phosphorus is possible by fungus such as
Penicilium and bacteria such as Pseudomonqs spp.
Co-inoculation of both P-solubilizing fungi and
bacteria has positive effects on the growth and
nutrient status of the soil by providing growth
hormone and increasing the NPK uptake by
seedlings and thus provides healthy environment
for the next crop. The results in the present
investigation indicate the presence of a diverse
group of phosphate solubilizing microbes that
dwell in the rhizosphere of coffee plants and
vermicompost amendment. It is evident that
phosphate solubilizing microbes are widely
distributed and showed significant variations
among the microbes with respect to their
phosphate solublization efficacy. The use of TCP
along with the phosphate solubilizers, Penicillium
and Aspergillus spp. as well as Pseudomonas spp
and Bacillus spp. as biofertilizers could enhance the
phosphate solubilization in the soils. Bio-fertilizers
are eco-friendly, free from hazardous chemicals,
possess no detrimental health effects and are cost
effective. Therefore, the use of vermicompost and
indigenous coffee rhizospheric phosphate
solubilizing bacteria and fungi can be a reliable
alternative in low inputs sustainable agriculture.
These phosphate solublizers can be used in the
field as efficient and potential phosphatic
Reshid Abafita Abawari et al.
biofertilizers for the cultivation and growth of
coffee seedlings. Therefore, further field studies
are required to confirm the application of these
microorganisms to sustain maximum organic
coffee yields.
CONCLUSION
The results obtained from glasshouse experiments
demonstrated that addition of inorganic phosphate
to the soil and inoculation with native PSB and PSF
significantly increased plant growth by enhancing
nutrient uptake (N, P, and K) in coffee seedlings.
One of the fungal isolates, RSCF1.19 has shown
better characteristics of plant growth promotion
than the bacterial inoculums and thus has a greater
potential for use as biofertilizer in coffee
production. Moreover, co-inoculation of this
fungus with Pseudomonas (RCHVCB1+RSCF1.19)
showed statistically significant growth parameters
compared to the control. Similarly, the nutrient
status of potting medium and nutrient uptake by
coffee seedlings increased in the treatments that
received both fungal and bacterial inoculants
amended with inorganic P fertilizer. The results
suggested that phosphate solubilizing microbes
have better potential as bio-fertilizers under
greenhouse conditions and needs field trials for
their bio-inoculums production.
ACKNOWLEDGEMENTS
This work was supported by Ethiopian Institute of
Agricultural Research and Agricultural Growth Program
Component II (AGP II). We would like to thank also
Jimma University, School of Veterinary Medicine for
their provision of laboratory facilities.
REFERENCES
1. Addis Tadsesse, Endale Taye, Alemseged Yelma
and TeshomeMesfine(2016).Intercropping of
Arabica Coffee (Coffea arabica L.) with
Korerima (Aframomum korarima) in South
Western Ethiopia. World J. Biol. Med. Science 3
(1), 43-49
2. Alemayehu Teshome, Esayas Kebede and Kassu
Kebede (2008). Coffee Development and
Marketing Improvement Plan. In: Proceedings
of A National Work Shop Four Decades of
Coffee Research and Development in Ethiopia.
SINET: Ethiop. J. Sci., 44(2), 2021
14-17 August 2007, EIAR, Addis Ababa,
Ethiopia. pp. 375-381.
3. Barea, J., Pozo, M. J., Azcon, R.,Azcon-Aguilar,
C.(2005).Microbial cooperation in the
rhizosphereJ. Exp. Bot. 56: 1761 - 1778.
4. Bauer, AW., Kirby, WMM., Sherris, JC., Turck, M.
(1966). Antibiotic susceptibility testing bya
standardized single disk method. Am. J. Clin.
Pathol. 45:493-496.
5. Bayetta Belachew, Bahilu Atero, Fiqadu Tefera,
Ashenafi Ayano, Tadese Benti (2008). Genetic
diversity and Heterosis in Arabica Coffee
PP:50-58.in:Girma Adunya, Bayeta
Belachew,Tesfaye Shimbir,Endale Taye and
TayeKufa(eds)
6. Bottini, R., Cassán F., Piccoli, P. (2004). Gibberellin
production by bacteria and its involvement in
plant growth promotion and yield increase.
Appl.Microbiol. Biotechnol. 65:497-503.
7. Collavino, M.M., Sansberro, P.A., Mroginski, LA.,
Aguilar, O.M. ( 2010). Comparison of in vitro
solubilization activity of diverse phosphate-
solubilizing bacteria native to acid soil and
their ability to promote Phaseolus vulgaris
growth. Biol Fert Soil 46(7):727–738
8. Dhurve, N.G., Ingle, RW., Lad, RS., Madavi,
P.N.(2017).Characterization of phosphate
solubilizing bacteria isolated from paddy
rhizosphere of Vidarbha region. Int. J. of
Chemical Studies. 5(6): 24-30.
9. Gaind, S.(2016).Phosphate dissolving fungi:
Mechanism and application in alleviation of
salt stress in wheat, Microbiological Research 193:
94–102.
10. Guimarães, GC., Rosa, SDVF., Coelho, LFS.,
Veiga, AD., Clemente, A C S. (2013). Minimum
period to assess the potential of germination of
coffee seeds. J Seed Sci. 35(3):347-52.
11. Hameeda, B., Harini, G., Rupela, OP., Wani, SP.,
Reddy, G. (2008). Growth promotion of maize
by phosphate-solubilizing bacteria isolated
from composts and macrofauna. MicrobiolRes.
163(2):234-242.
12. He, Z.L., Wu, L., O’Donnell, N.G., Syers,
J.K.(1997). Seanonal responses in microbial
biomass carbon, phosphorus and sulphur in
soils under pasture. Biol. Fert. Soils.24: 421-428.
13. Jilani, G., Akram, A., Ali, R.M., Hafeez, FY.,
Shamsi, I.H., Chaudhry, A.N., Chaudhry, A.G.
(2007).Enhancing crop growth, nutrients
availability, economics and beneficial
rhizospheremicroflora through organic and
biofertilizers. Ann. Microbiol.57:177-183.
14. Kunwar, VS., Chimouriya, S., Lamichhane, J.,
Gauchan,Dh.P.(2018). Isolation and
149
Characterization of Phosphate Solubilizing
Bacteria from Rhizosphere of Coffee Plant and
Evaluating Their Effects on Growth and
Development of Coffee Seedlings. Biotechnol Ind
J.;14(5):173
15. Liasi, S.A., Azmi, T.I., Hassan, M.D., Shuhaimi, M.,
Rosfarizan, M., Ariff, A.B. (2009). Antimicrobial
activity and antibiotic sensitivity of three
isolates of lactic acid bacteria from fermented
fish product, Budu. Malaysian J. Microbiol.5:33-
37.
16. Logan, W.J.C.,Biscoe, J.(1987).Coffee Handbook
(Zimbabwe Coffee Growers Association.
Harare).
17. Mahadevamurthy, M.M., Channappa, Th.,
Sidappa, M.S., Raghupathi, M.K., Nagaraj, A.
(2016). Isolation of phosphate solubilizing fungi
from rhizosphere soil and its effect on seed
growth parameters of different crop plants. J.of
Appl. Biol. &Biotec. 4 (06): 022-026.
18. Mamta, PR., Pathaniad, V., Gulati, A., Singh,
B.,Bhanwra, R.K., Tewari, R. (2010) .Stimulatory
effect ofphosphate-solubilizing bacteria on
plant growth, stevioside andrebaudioside: A
contents of Stevia rebaudianaBertoni. Appl Soil
Ecol. 46:222-229.
19. Mohamed, M.H., Almaroai, A.Y. (2017). Effect of
Phosphate Solubilizing Bacteria on the Uptake
of Heavy Metals by Corn Plants in a Long-
Term Sewage Wastewater Treated Soil. Ijesd.
8(5): 366-371.
20. Nahas, E. (1996). Factors determining rock
phosphate solubilization by microorganisms
isolated from soil. World J. Microbiolology
Biotechnology. 12:567-572.
21. National Committee for Clinical Laboratory
Standards (2000). National Committee for
Clinical Laboratory Standards, Performance
Standards for Antimicrobial Disk
Susceptibility Tests, Approved Standard M2-
A7, 7th Ed., National Committee for Clinical
Laboratory Standards, Wayne, PA.
22. Ormeño Díaz M.A., Ovalle Silva,A.I., Rey, J.
C.(2018). Evaluation of the effect of liquid
organic fertilizers on the growth of seedlings of
coffee (Coffea arabica L.). Rev. Fac. Agron.
(LUZ). 2018, 35: 387-407. Octubre-Diciembre.
23. Pandey, P., Bisht, S., Sood, A., Aeron, A., Sharma,
G.D., Maheshwari, D.K .(2012). Consortium of
plant-growth-promoting bacteria: future
perspective in agriculture, p. 185–200. In
Maheshwari D.K. (ed.), Bacteria in
Agrobiology: Plant Probiotics. Springer, Berlin
Heidelberg.
150 Reshid Abafita Abawari et al.

24. Park, J., Bolan, N., Megharaj, M., Naidu, R . 32. Sheraz Mahdi, S., Hassan, G.I., Samoon, S.A.,
(2010). Isolation of Phosphate-Solubilizing Rather, HA.,Showkat, A., Dar Zehra, B. (2010).
Bacteria and their Effects Characterization of on Bio-fertilizers inorganic agriculture. Journal of
Lead Immobilization. Pedologist. 67-75. Phytology, 2(10): 42-54.
25. Pikovskaya, R. I.(1948). Mobilization of 33. Son, H.J., Park, G.T., Cha, M.S., Heo, M.S.(2006).
phosphorus insoil in connection with vital Solubilization of insoluble inorganic
activity of some microbial species Microbiology phosphates by a novel salt and pH-tolerant
17: 362-70. Pantoeaagglomerans R-42 isolated from soybean
26. Prasad, S.P., Hareesh, SB., Violet D’souza, M., rhizosphere. BioresourceTechnol. 97(2):204±210.
Manjunath, A., Jayarama, N. (2014). Evaluation https://doi.org/10.1016/j.biortech.2005.02.021
of Bio-inoculants Enriched Marginal Soils as PMID:16171676.
Potting Mixture in Coffee Nursery. J 34. Spain, A., Alm, E. (2003). Implications of microbial
BiofertilBiopestici 5: 148. doi:10.4172/2155- heavy metal tolerance in the environment.
6202.1000148 Reviews in Undergraduate Research. 2:1-6.
27. Reshid Abafita, Tesfaye Shimbir, Tesfu Kebede 35. Sundara, B., Natarajan, V., Hari, K
(2014). Effects of different rates of .(2002).Influence of phosphorus solubilizing
vermicompost as potting media on growth and bacteria on the changes in soil available
yield of tomato (Solanum lycopersicumL.) and phosphorus and sugar cane and sugar yields.
soil fertility enhancement Sky. J. Soil. Sci. Field Crops Research, 77:43-49.
Environ. Manage. 3(7): 073 – 077. 36. Tam, HTh.,Chau, D.Th. M., Diep, CNg.(2016).
28. Reshid Abafitai, Fasil Asefa, Diriba Muletae (2021). Isolation and Identification Phosphate-
Multi Traits of Phosphate Solublizing Bacterial Solubilizing Fungi from Ferralsols of tithonia
and Fungal Isolates and Evaluation of Their (TithoniaDiversifolia(Hamsl.) gray) in Daknong
Potential as Biofertilizer Agent for Coffee and Daklak Province(s), Vietnam.wjpps, Vol 5,
Production. International Journal of Applied Issue 9, 2016. www.wjpps.com
Agricultural Sciences. 7(1):1-15. doi: 37. Vessey, K.J . (2003). Plant growth promoting
10.11648/j.ijaas.20210701.11 rhizobacteria as biofertilizers. Plant Soil
29. Rojo-bezares, B., Saenz, Y., Poeta, P., Zarazaga, M., 255:571–586.
Ruizlarrea, F., Torres, C. (2006). Assessment of 38. Vlková, E., Rada, V., Popelářová, P., Trojanová, I.,
antibiotic susceptibility within lactic acid Killer, J. (2006). Antimicrobial susceptibility
bacteria strains isolated from wine. Int. J. Food ofbifidobacteria isolated from gastrointestinal
Microbiol. 111: 234–240. tract ofcalves. Livestock Science 105: 253-259.
30. Santiago, C.D., Yagi, S., Ijima, M., Nashimoto, T., 39. Wani, P.A., Irene, OI. (2014). Screening of microbes
Sawada, M., Ikeda, S., Asano, K., Orikasa, Y., for their metal, antibiotic resistance and plant
Ohwada, T.(2017). Bacterial Compatibility in growth promoting activity. Current Research in
Combined Inoculations Enhances the Growth Bacteriology. 7:22-31.
of Potato Seedlings. Microbes Environ. 32(1): 14-
40. Yadav, J., Verma, J.P., Tiwari, K.N. (2011). Plant
23.
growth promoting activities of fungi and their
31. Satyaprakash, M., Nikitha, T., Chaitanya, I., Reddi, effect on chickpea plant growth. Asian Journal of
EUB.(2017).Phosphate solubilisingrhizobacteria Biological Sciences. 4:291-299.
associated withCoffea Arabica L. in coffee
41. Zaidi, A., Khan, M.S., Ahmad, E. (2014).
plantation of northeastern ghats,
Microphos: Principles, Production and
visakhapatnam district,andhrapradesh,
Application Strategies. In Phosphate
india.Inte. J.of Current Research. 9(5): .49779-
Solubilizing Microorganisms. Springer,
49783.
Cham.1-30
SINET:Ethiop. J. Sci., 44(2): 151–160, 2021 ISSN: 0379–2897 (PRINT)
© College of Natural and Computational Sciences, Addis Ababa University, 2021 eISSN: 2520–7997

Date received: March 18, 2021; Date revised: June 06, 2021; Date accepted: July 23, 2021
DOI: https://dx.doi.org/10.4314/sinet.v44i2.2

Evaluation of chickpea (Cicer arietinum L.) genotypes for tolerance to Frost in

controlled environment

Sintayehu Admas1,2*, Teklehaimanot Haileselassie2, Kassahun Tesfaye2,3, Eleni Shiferaw1

and K. Colton Flynn4

1 Crop and Horticulture Biodiversity Directorate, Ethiopian Biodiversity Institute, PO Box

30726, Addis Ababa, Ethiopia. E-mail: sintayehu.admas@ebi.gov.et
2 College of Natural Sciences, Addis Ababa University, PO Box 3285, Addis Ababa, Ethiopia
3 Ethiopian Biotechnology Institute, PO Box 5954, Addis Ababa, Ethiopia
4 USDA-ARS, Grassland Soil and Water Research Laboratory, 808 East Blackl and Road,Temple,

TX 76502, USA

ABSTRACT: The study aimed to evaluate the frost tolerance variability of Ethiopian
chickpea (Cicer arietinum L.) germplasm under controlled environment using growth
chamber. A total of 72 genotypes were screened for frost tolerance using complete
randomized design with two replications. The analysis of variance result indicated that
there was a significant (P<0.01) difference amongst genotypes for plant height, number of
foliage, number of primary branch, growth rate, and fresh biomass weight. Based on plant
survival rate (SR), 31 (43.1%) genotypes scored above 0.8 values. Based on Freezing
tolerance rate (FTR), 37(51.4%) and 31(43.1%) genotypes were rated at a score of 1 to 3 in
freezing test 1 (T1) and freezing test 2 (T2), respectively. There was a strong negative
correlation between fresh biomass yields with SR (-0.75** for T1 and -0.71** for T2 at p<0.01),
while a strong positive correlation with FTR value (0.74** at p<0.01). Based on the combined
result of FTR and SR scores, 26 genotypes were found to be frost-tolerant genotypes at a
temperature level as low as -5oC at seedling stage. Based on our findings, Ethiopian
chickpea germplasm has a genetic potential for frost-tolerance traits for use in breeding
programs.

Key words/phrases: Chickpea, Freezing test, Frost screening, Frost tolerant

volume next to faba bean and haricot bean,

INTRODUCTION
Plant genetic resources play a significant role
in the variety development program by
serving as a reservoir for enormous genes
that confer tolerance to abiotic and biotic
stresses and potential sources of gene for
most important agronomic traits (Rao, 2004).
The maintenance of a wide array of genetic
pool for different crops is the main target for
gene bank manager. Recognizant of this, the
Ethiopian Biodiversity Institute (EBI) has been
collecting and maintaining plant genetic
resources in its genebank. And, extensive
genetic characterization and evaluation of the
germplasm for agronomic and quality traits
are required to make it more useful to
breeders and farmers (Castañeda-Álvarez et
al., 2016).
Chickpea (Cicer arietinum L.) is currently
the third largest food legume crop in Ethiopia
in terms of area covered and production
_____________________
*Author to whom correspondence should be addressed.
occupying roughly 239,786.13 ha of land
annually and producing 459,173,187 Kg with
an average productivity of 2025 kg/ha for
desi and 1682 kg/ha for kabuli type chickpea
(CSA, 2019). Currently, chickpea is introduced
to lowland areas using irrigation and also to
select areas of the Southern Nation and
Nationality People Region (Nigusie Girma et
al., 2017) contributing to the steady increase
of chickpea production. Furthermore,
chickpea production could be expanded into
highland locations (>2500masl) where frost is
a typical occurrence. The highland constitutes
2/3rd of the total cultivated land in Ethiopia
(Mulugeta Assefa et al., 2014) and chickpea
production can be extended to these areas if
frost-tolerant chickpea varieties are available
to the farmers.
Chickpea is a cold sensitive legume crops
and cold stress is the second most important
limiting factor in its production (Sassenrath et
152
al., 1990). Cold stress is classified as Chilling
injury (0 °C to 15 °C) and freezing/frost
injury (below -1.5 °C) (Croser et al., 2003; Jha
et al., 2014), both of which have and
overlapping effects on chickpea growth and
production (Croser et al., 2003; Jha et al., 2014).
Low field temperatures causes poor seed
germination, poor crop stand establishment,
chlorosis, wilting, necrosis of leaf tips,
reduced plant height and branches, full leaf
curling, and plant death (Croser et al., 2003;
Kumar et al., 2010). Moreover, frost stress
lowers leaf water status and chloroplast
membrane stability, resulting in the loss of
respiration and photosynthesis (Croser et al.,
2003; Yadav, 2010).
Frost-tolerance is one of the most important
pre-requisites to grow cool season legumes in
frost prone areas. The degree of frost
damages varies among genotypes due to the
differences in frost-tolerance capacity of the
genotypes. Evaluation of plant germplasm for
frost-tolerance variability is very crucial to
identify resistant genotypes. For example
genetic variation of frost-tolerance has been
reported for field pea seedlings (Bourion et
al., 2003), chickpea (Kanouni et al., 2009;
Nezami et al., 2012; Mir et al., 2019). However,
there was no documented information
regarding the potential of Ethiopian chickpea
genotypes for frost-tolerance variability.
Two types of chickpea frost screening
protocols; field screening under natural
condition and under controlled environment
using growth chamber have been used by
various breeders. The natural field screening
method is expensive and time consuming,
there is unpredictable frost severity and
irregular low temperature frequency
(Maqbool et al., 2010), the lowest temperature
is not controlled and there are large temporal
and spatial variations in the field (Nezami et
al., 2012). However, a controlled environment
(using freezing chamber) screening method
offers much more precise control of the
timing and intensity of frost treatment (Wu et
al., 2014). It is also inexpensive, quick and
highly reproducible (Nezami et al., 2012).
Therefore, the objective of the study was to
screen Ethiopian chickpea genotypes for
frost-tolerance variability in control
environment using growth chamber and
identify frost-tolerant genotypes to be used in
chickpea breeding program.
Sintayehu Admas et al.
MATERIAL AND METHODS
Plant materials
The study was conducted using 72 genotypes
constituting 51 Ethiopian chickpea genotypes,
13 genotypes from the International Center
for Agricultural Research in the Dry areas
(ICARDA), and 13 improved chickpea varieties
form Deber Zeit Agricultural Research
Centers. The genotypes were selected based
on the field performances of genotypes
showing poor to better reaction to frost
stresses at vegetative and grain filling period
(Sintayehu Admas et al., 2021).
Experimental design
Ten seeds of each genotype were planted in
pots with 20 cm internal diameter and 20 cm
depth. The pots were filled with
homogeneous soil mixture which was
prepared by mixing the sub-surface (0-30cm)
soil thoroughly collected from Debre Zeit
Research Center chickpea farm. The seedlings
were thinned to five plants per pot at four
leaf stage. A complete randomized design
with two replications was used. DAP (100
kg/ha) and other management practices were
applied. Each morphological and
physiological data were collected from all the
five individual plants.
A modified frost screening protocol using
freezing chamber designed by Nezami et al.
(2012); Zhu et al. (2014), and Mugabe et al.
(2019) were used. The plants were grown in
green house for two weeks at Ethiopian
Biodiversity Institute and moved to the
controlled environment chamber (Snijders
labs climate chambers) for five weeks.
Seedlings were subjected to a gradual low
temperature acclimation protocol for four
weeks. Acclimation started at seven days
with 7°C days/5°C nights, 11-h a photoperiod
(PP) followed by 5°C days/2°C nights, 10-h
PP for 7 days and then at 2°C days/0°C
nights, 9-h PP for 14 days with 250 mmol m2s1
levels of irradiance. Subsequently, the frost
treatment test took place for seven days
under 5°C days/-2°C night, 10-h PP for 3
days, 5°C days/-3°C nights, 10-h PP for 2 days
and finally 5°C days/-5°C nights, 10-h PP for
2 days. Finally, the pots were allowed to thaw
overnight at 4°C and the plants were moved
SINET: Ethiop. J. Sci., 44(2), 2021
back to the green house for one week for
scoring to conduct freezing test 1: to evaluate
the re-growth potential of both the foliage
and auxiliary buds of the genotypes and
freezing test 2: to test the viability of the
foliage using 1-9 scale as indicated below.
Data Collected
Plant Height (cm): Average canopy height
of five representative plants taken before and
after frost treatment.
Number of foliage leaf per plant: Average
number of foliage leaf per plant taken from
five representative plants before and after
frost treatment.
Number of primary branches per plant:
Average number of basal primary branches
per plant taken from five representative
plants before and after frost treatment.
Plant height growth rate: The ratio of the
difference between plant height before and
after frost stress to plant height after frost
stress.
Number of foliage leaf growth rate: The
ratio of the difference between number
foliage leaf per plant before and after frost
stress to number foliage leaf per plant.
Number of primary branches plant growth
rate: The ratio of the difference between
number of primary branches per plant before
and after frost stress to number of primary
branches per plant.
Fresh biomass yield (g/plant): Average
fresh weight of five plants.
Freezing tolerance rate (FTR): Visual
identification of viability of the foliage and
foliage and auxiliary buds re-growth scored
on 1-9 scale bases (Fiebelkorn, 2013 cited by
Mugabe et al., 2019), where, 1=Plant
completely green, 2= Plant with minimal
freezing damage, 3= Plant at least 75% green,
4= Plant between 50 and 75% green tissue, 5=
Plant 50% green, 6= Plant between 25 and
50% green tissue, 7= Plant 25% green, 8=
Plant almost dead but has minimal green, and
9= Plant completely dead.
Plant survival rate (SR): Calculated by
dividing the number of surviving plants after
the frost period by the number of emerged
plants after sowing was calculated
(Heidarvand et al., 2011).
153
Statistical Analysis
The performances of genotypes were tested
for significance by performing an analysis of
variance (ANOVA) in a complete randomized
design using R-4.1.0 statistical software
(Thomas et al., 2013). Treatment mean
comparison was performed for significance,
using a Fisher’s least significant difference
(LSD) test at 5% probability using R software.
Pearson correlation coefficients between
variable was estimated and tested for
significance using MINITAB release 14
statistical software (MINITAB, 1998).
RESULTS AND DISCUSSIONS
In Ethiopia, chickpea is commonly grown in
areas having vertisols soil type with an
altitude range of 1400 to 2300 meters above
sea level (Geletu Bejiga et al. 1996). There is
still an immense potential to introduce
chickpea as a new crops in highland area (>
2300 masl). In highland area, however, the
existence of frost stress limits crop
production. So to bring chickpea as an
alternative crop in this area; it requires the
improvement of chickpea towards the
capacity of chickpea for frost stress tolerance.
This requires an extensive germplasm
screening to identify frost tolerant genotypes.
Studying frost tolerance and breeding for
frost-tolerant chickpea varieties play a
fundamental role in increasing chickpea
production in frost prone areas. In the present
study, chickpea genotypes were evaluated
under controlled environment using growth
chamber. The experiment has shown the
response of 72 chickpea genotypes with
respect to frost stress tolerance, which
occurred during the two weeks old chickpea
seedlings under controlled conditions.
The analysis of variance raveled significant
differences at P  0.01 for frost-tolerance
traits variability amongst genotypes for plant
height, number of foliage, number of primary
branch, growth rate, and fresh biomass
weight (Table 1). This indicated that the
differences in the genotypes to the reaction of
frost damage were variable, which is an
indicator of the existence of variability
amongst genotypes for frost tolerances.
154 Sintayehu Admas et al.

Table 1. Analysis of variance and mean squares for tested traits grown under controlled environment at
Ethiopian Biodiversity Institute, 2021.

Sources of Degree of Mean Squares

Variation freedom
PLH PLH GR NF NF GR NPB GR FW
BF AF PLH BF AF NF AF NPB
Genotypes 71 8.0** 55.78** 0.04** 2.22** 45.43** 0.06** 31.55** 0.27** 26.27**
Error 72 2.12 7.48 0.01 1.45 3.95 0.01 2.58 0.02 2.94

PLHBF=Plant height before frost treatment in cm, PLHAF= Plant height after frost treatment in cm, GRPLH= Growth rate of
plant height during frost treatment, NFBF=Number of foliage before frost treatment, NFAF=Number of foliage after frost
treatment, GRNF=Growth of number of foliage during frost treatment, NPBAF=Number of primary branch after frost treatment,
GRNPB=Growth rate of number of primary branch during frost treatment, FW=Fresh weight in gm

Fisher’s least significant difference (LSD)
result indicated that the differences among
the means of the genotypes for a given trait
were significant (P<0.05). Wide mean ranges
were observed for all the collected traits in
the genotypes (Table 2). The LSD value and
range values confirmed the presence of a
variable response among genotypes for frost
stress. Similarly Mir et al. (2018) reported
presence of variability in chickpea germplasm
for frost stress.

Table 2. Mean value of quantitative traits of 72 chickpea genotypes grown under controlled
environment at Ethiopian Biodiversity Institute, 2021.

N Acc Type Source PLH PLH GR NF NF GR NP GRN FW SR FT1 FT

o s BF AF PLH BF AF NF B PB 2
AF
1 207674 Desi EBI 11.3 22.6 0.50 5.9 18.7 0.7 12.2 1.0 14.7 0.9 2 3
2 30350-B Desi EBI 10.6 21.1 0.49 5.4 14.8 0.6 7.4 1.0 10.5 1.0 2 3
3 41133-A Desi EBI 9.6 21.1 0.55 5.3 18.5 0.7 4.7 1.0 10.6 0.8 2 3
4 207173-B Desi EBI 13.1 25.3 0.48 5.4 17.5 0.7 10.4 1.0 13.6 1.0 1 1
5 207175-A Desi EBI 11.2 20.5 0.46 7.3 19.0 0.6 7.4 1.0 7.5 0.8 1 2
6 207766 Desi EBI 11.9 22.5 0.47 6.6 20.2 0.7 14.7 1.0 12.0 0.8 1 2
7 209026-B Desi EBI 8 17.2 0.53 5.8 11.1 0.5 3.0 1.0 13.2 1.0 1 1
8 227152-B Desi EBI 10.3 19.3 0.47 5.2 16.7 0.7 7.3 1.0 10.9 0.8 1 2
9 41301-B Desi EBI 8.1 16.4 0.50 5.2 13.7 0.6 6.8 1.0 11.1 1.0 1 2
10 207746 Desi EBI 7.5 14.5 0.48 5.5 8.0 0.3 1.5 1.0 10.7 1.0 2 3
11 Teketay Desi DZAR 8.3 16.1 0.48 6.2 17.5 0.7 11.2 1.0 11.1 1.0 1 1
C
12 Natoli Desi DZAR 5.0 9.7 0.48 5 8.6 0.4 0.8 0.5 10.2 1.0 2 3
C
13 Akaki Desi DZAR 9 19.9 0.55 5.7 13.5 0.6 4 1.0 11.6 1.0 1 1
C
14 9427 Kabu ICAR 9.1 16.6 0.46 6 12.3 0.5 5.8 1.0 9.2 1.0 1 1
li DA
15 69420 Kabu ICAR 7.1 17.2 0.59 5.5 9.9 0.5 0 0.0 11.9 1.0 1 1
li DA
16 Worku Desi DZAR 8.6 20.2 0.58 5.1 10.7 0.5 2.8 1.0 8.8 0.9 3 3
C
17 Mariye Desi DZAR 6.2 12.2 0.50 6.0 8.05 0.3 0 0.0 9.0 1.0 1 1
C
18 DBB Desi EBI 8.8 17.9 0.51 6.2 13.0 0.5 4.3 1.0 10.0 0.8 2 3
19 ENR Desi EBI 11.2 21.4 0.48 6.2 19.1 0.7 8.3 1.0 9.1 1.0 2 2
20 TEGR Desi EBI 9.9 20.8 0.52 6.2 18.1 0.7 10.0 1.0 9.1 1.0 2 3
21 30334-C Desi EBI 12.8 22.5 0.44 7.7 18.9 0.6 10.9 1.0 13.6 0.5 1 2
22 207648 Desi EBI 12.2 19.1 0.36 6.3 19.1 0.7 16.0 1.0 13 1.0 1 1
23 207728-A Desi EBI 10.6 18.0 0.41 5.6 14.0 0.6 5.2 1.0 12.1 1.0 2 2
24 208988-A Desi EBI 12.1 19.9 0.39 6.4 15.5 0.8 12.4 1.0 14.6 1.0 1 1
25 Kutaye Desi DZAR 8.3 15 0.45 5.4 14.5 0.6 8.9 1.0 9.3 1.0 1 1
C
26 Yelbie Kabu DZAR 8.4 13.8 0.39 5.9 8.9 0.3 5.6 1.0 9.8 1.0 2 3
li C
27 Teji Kabu DZAR 5.7 12.8 0.56 4.2 8.9 0.5 0.5 0.5 10.8 0.7 1 2
li C
28 Mastewal Desi DZAR 8.9 18.3 0.52 6.4 12.5 0.5 3.5 1.0 12.8 0.7 1 2
 SINET: Ethiop. J. Sci., 44(2), 2021 155

C
29 Shahso Kabu DZAR 8.1 19.2 0.58 5.5 15.0 0.7 6.9 1.0 9.4 0.6 1 1
li C
30 30300-A Desi EBI 12.3 24.2 0.49 6.5 13.5 0.5 7.3 1.0 13.5 0.9 4 5
31 30309-A Desi EBI 8.1 17.8 0.26 6.1 13.8 0.6 7.0 1.0 5.5 0.6 6 6
32 41075-C Desi EBI 11.9 20.1 0.41 6.1 12.0 0.5 11.4 1.0 9.4 1.0 4 4
33 41078-B Desi EBI 10.3 20.6 0.50 5.6 16.5 0.7 8.8 1.0 9.1 0.6 3 5
34 41094-C Desi EBI 7.3 14.0 0.47 5.1 15.. 0.7 6.8 1.0 7.9 0.8 4 5
35 41153-A Desi EBI 11.2 22.4 0.49 5.1 16.6 0.7 5.9 1.0 12.0 0.7 3 4
36 41282-B Desi EBI 11.4 20.1 0.44 6.0 16.9 0.6 8.2 1.0 13.8 0.7 3 3
37 41323-A Desi EBI 10.0 19.9 0.50 5.5 17.0 0.7 9.1 1.0 11.2 0.7 3 5
38 207167-A Desi EBI 11.8 20.9 0.43 6.9 18.9 0.6 6.5 1.0 10.3 0.8 4 4
39 207640 Desi EBI 10.4 14.2 0.27 5.1 12.7 0.6 3.8 1.0 10.3 0.7 5 6
40 207652 Desi EBI 9.5 17.4 0.46 4.5 16.0 0.7 6.7 1.0 9.2 0.9 4 5
41 207670 Desi EBI 11.4 21.4 0.42 8.0 18.9 0.6 13.5 1.0 12.1 0.7 5 6
42 209026-A Desi EBI 10.1 19 0.47 5.9 16.3 0.6 6.8 1.0 9.1 0.8 4 5
43 212477-A Desi EBI 11.0 17.7 0.38 5.8 12.4 0.5 3.9 1.0 12.5 0.6 1 2
44 241800-A Desi EBI 9.9 18.9 0.48 5.6 15.8 0.7 5.5 1.0 9.8 0.9 3 5
45 30339-A Desi EBI 10.1 19.3 0.48 6.3 13.1 0.5 5.8 1.0 10.1 1.0 4 5
46 Minjar Desi DZAR 8.8 17.5 0.50 4.3 12.4 0.7 4.3 1.0 11.2 0.6 3 4
C
47 140941 Kabu ICAR 10.3 16.6 0.38 6.2 14.2 0.6 9.3 1.0 11.7 0.7 4 4
li DA
48 141693 Kabu ICAR 8.6 18.7 0.54 6.5 8.2 0.2 0 0.0 1.5 0.5 8 8
li DA
49 125187 Kabu ICAR 8.3 14.4 0.71 5.1 10.4 0.5 2.9 1.0 3.7 0.5 8 8
li DA
50 Dubie Desi DZAR 8.1 15.7 0.48 3.9 17.3 0.9 4.2 1.0 10.8 0.6 6 6
C
51 140294 Kabu ICAR 9.1 16.9 0.47 4.5 7.1 0.4 0 0.0 8.2 0.6 3 5
li DA
52 132663 Kabu ICAR 6.6 14.1 0.53 5.3 11.0 0.5 3.6 1.0 5.8 0.8 4 4
li DA
53 Kasech Kabu DZAR 10.5 0 0 2.2 0 0 0 0.0 0 0.0 9 9
li C
54 16341-A Desi EBI 12.3 24.1 0.49 5.9 16.8 0.7 10.3 1.0 13.2 0.7 5 6
55 41081-A Desi EBI 9.2 16.9 0.46 5.1 11.7 0.6 6.1 1.0 5.2 0.5 7 7
56 207608 Desi EBI 8.3 17.4 0.52 3.9 10.1 0.6 0 0.0 5.1 0.4 8 9
57 207638 Desi EBI 9.3 19.8 0.53 4.4 14.4 0.8 6.8 1.0 10.4 0.4 7 7
58 207649-A Desi EBI 6.8 15 0.56 5.0 7.7 0.3 2.1 1.0 8.0 0.5 6 7
59 207668 Desi EBI 9.7 21.8 0.56 6.1 13.5 0.6 3.7 1.0 10.1 0.7 5 5
60 209008-A Desi EBI 9.2 18 0.49 5.8 16.6 0.7 9.4 1.0 11.4 0.6 5 5
61 209016-B Desi EBI 11.3 20.7 0.46 6.0 15.6 0.6 5.9 1.0 9.5 0.6 5 7
62 212688-C Desi EBI 10 23.2 0.57 5.4 14.4 0.6 4.7 1.0 10.4 0.7 5 6
63 215190-A Desi EBI 9.2 19.1 0.53 5.1 12.0 0.6 5.8 1.0 12 0.6 6 7
64 237054-B Desi EBI 6.4 16.9 0.62 4.9 11.5 0.6 2.8 1.0 6.9 0.4 7 7
65 Dimtu Desi DZAR 6.5 16.6 0.61 5.1 11.5 0.6 0.7 0.5 9.6 0.5 6 6
C
66 141720 Kabu ICAR 11.5 21.4 0.46 6.2 19.0 0.7 9.1 1.0 3.2 0.3 8 8
li DA
67 75095 Kabu ICAR 5.6 13.4 0.58 4.3 6.8 0.4 0.0 0.0 6.2 0.0 9 9
li DA
68 209026-A Desi EBI 10 18.9 0.47 5.1 13.0 0.6 4.0 1.0 4.8 0.4 7 8
69 10163 Kabu ICAR 3.7 10.4 0.65 3.0 5.6 0.5 0.0 0.0 1.5 0.4 8 8
li DA
70 8191 Kabu ICAR 8.2 0.0 0.0 1.7 0.0 0.0 0.0 0.0 0.0 0.0 9 9
li DA
71 139930 Kabu ICAR 6.1 0.0 0.0 4.4 0.0 0.0 0.0 0.0 0.0 0.0 9 9
li DA
72 73221 Kabu ICAR 8.2 0.0 0.0 3.9 0.0 0.0 0.0 0.0 0.0 0.0 9 9
li DA
Mean 9.3 17.3 0.5 5.4 13.1 0.5 5.6 0.8 9.2 - - -
±SE 1.5 2.8 0.1 1.2 2.0 0.1 1.6 0.2 1.9 - - -
LSD (5%) 2.9 5.5 0.2 2.4 3.9 0.2 3.0 0.3 3.8 - - -
Range 3.7- 0-25.3 0-0.71 1.7- 0- 0- 0- 0-1 0- 0- 1-9 1-9
13.1 8 20.2 0.8 16. 14.7 1.0

PLHBF=Plant height before frost treatment, PLHAF= Plant height after frost treatment, GRPLH= Growth rate of plant
height during frost treatment, NFBF=Number of foliage before frost treatment, NFAF=Number of foliage after frost
treatment, GRNF=Growth of number of foliage during frost treatment, NPBAF=Number of primary branch after frost
156 Sintayehu Admas et al.

treatment, GRNPB=Growth rate of number of primary branch during frost treatment, FW=Fresh weight, SR=Survival
rate, FT1 =Freezing test 1, and FT2=Freezing test 2. SE=pooled standard deviation, LSD=Least square difference,
EBI=Ethiopian Biodiversity Institute, DZAR=Deber Zeit Agricultural Research Centers, ICARDA=International Center
for Agricultural Research in the Dry areas

The reaction of genotypes to frost as

indicated by plant survival rate (SR) is given
in Table 3. Thirty one genotypes (43.1%) rated
more than 0.8 score, while the remaining
genotypes rated less than 0.8 score (56.9%).
Five genotypes scored 0 which means that
they were killed by frost stress since it had
poor reaction to frost stress. SR and FTR results
showed that differences in the levels of frost
tolerance between different genotypes were
highly variable. Based on FTR and SR scores
most genotypes were found to exhibit
moderate frost tolerance at a temperature of -
5°C. A plant survival score was used as an
index to describe genotypes tolerance to low
temperature (Heidarvand et al., 2011). The
susceptible four genotypes were killed at -
2°C, the remaining genotypes had shown
variable number of plant deaths which
indicates the different capacity of genotypes
for frost reactions.
Table 3. Plant survival rate (SR) of 72 chickpea
genotypes grown under controlled
environment at Ethiopian Biodiversity
Institute, 2021.
No SR Rating No of genotypes
1 0.8 31 (43.1 %)
2 0.6 to <0.8 22 (30.6 %)
3 0.4 to <0.6 9 (12.5 %)
4 0.2 to <0.4 5 (6.9 %)
5 <0.2 5 (6.9 %)
Total 72
The result of foliage and auxiliary buds
re-growth (Freezing test 1), and foliage
viability (Freezing test 2) is indicated in Table
4. The majority of the genotypes showed
recovery from frost damage. Four genotypes
did not recover because they were killed by
frost. The remaining genotypes recovered
with low to high rate (Fig. 1). The records of
leaf damage of the frost-susceptible
genotypes showed severely damaged
genotypes and all plant leaves died, while in
resistant genotypes the leaf damage were nil
to medium level (Fig. 1). Although the foliage
of these genotypes had injured foliage
following a frost, re-growth occurred from
auxiliary buds at the stem (Fig. 2). The scores
were done visually using freezing tolerance
rate (FTR). The FTR scores were taken at one
week after the end of the frost treatments.
Thirty seven (51.4%) and 31 (43.1%)
genotypes showed no or little leaf damage
due to frost injury for freezing test 1 and
freezing test 2 respectively. The remaining
genotypes, 35 (48.6%) and 41 (56.9%)
genotypes scored from 4 to 9 at freezing test 1
and freezing test 2, respectively. Theses
genotypes were moderately frost-tolerant to
highly frost-susceptible genotypes.

Table 4. Freezing tolerance rate (FTR) of 72 chickpea genotypes grown under controlled environment at
Ethiopian Biodiversity Institute, 2021.

No FTR Rating No. of genotypes (Freezing Test 1) No. of genotypes (Freezing Test 2)
1 1 19 (26.4%) 11 (15.3%)
2 2 10 (13.9%) 10 (13.9%)
3 3 8 (11.1%) 10 (13.9%)
Subtotal 37 (51.4%) 31 (43.1%)
4 4 9 (12.5%) 6 (8.3%)
5 5 7 (9.7%) 11 (15.3%)
6 6 5 (6.9%) 7 (9.7%)
7 7 4 (5.6%) 6 (8.3%)
8 8 5 (6.9%) 5 (6.9%)
9 9 5 (6.9%) 6 (8.3%)
Subtotal 35 (48.6%) 41 (56.9%)
Total 72 72
SINET: Ethiop. J. Sci., 44(2), 2021 157

Figure 1. The reactions of genotypes to frost injury [A (139930) and B (73221) highly frost-susceptible genotypes, C
(30339-A) moderately frost-tolerant genotypes, D (209026-A) and E (Teketay) frost-tolerant genotypes].

Figure 2. The variable potential of genotypes to the re-growth of the foliage and auxiliary buds one week after the
end frost treatments [A (973221), B (141693), and C (207649-A) genotypes did not show foliage and
auxiliary buds re-growth, D (Minjar) genotypes had shown foliage re-growth, E (Akaki) and F (41282-B)
genotypes had shown foliage and auxiliary buds re-growth. The arrow indicates the growing of primary
branches.

A rating scale of 1-9 has been used for
measuring frost stress injury during early
vegetative stage or seedling stage in earlier
studies (Singh et al., 1989). The score was
done by visual observation of the viability of
the foliage, and buds re-growth of foliage and
auxiliary. The susceptible genotypes were not
showed foliage and auxiliary bud re-growth
at all, in addition the percent of damage on
foliage and auxiliary buds were severe.
However, the frost-tolerant genotypes were
gave better reaction to frost stress; moderate
to high foliage and auxiliary bud re-growth
rate were observed. Freezing and/or chilling
range temperatures cause poor establishment,
reduced vigor resulting in stunted seedlings
and retarding plant growth and, in extreme
cases, may lead to plant death (Croser et al
2003; Maphosa et al., 2020). However, an
expected result was observed for 19
genotypes (Table 2, genotype listed from no
54 to 72) in which they showed better growth
rate of plant height, foliage leaf and primary
branch, while their reaction to frost stress
were poor with FTR score of 5 and above. This
happened because these genotypes had
performed well during a frost treatment of -
2°C and -3°C, however when frost treatment
temperature continued to drop at a level to -
5°C, then, these genotypes could not
withstand the frost stress and whole plant
death started which were manifested a week
after the end of frost treatment.
The phenotypic association of
agronomic and frost tolerance related traits
were analyzed based on the mean values of
the recorded traits of all genotypes and the
result is given in Table 5. SR and FTR scores
158 Sintayehu Admas et al.

were strongly correlated with plant height,
number of foliage, number of primary
branch, growth rate, and fresh biomass
weight at p<0.01. FTR score showed a negative
strong correlation between the traits
considered and SR score, while SR score
showed a positive strong correlation between
the recorded traits and FTR value. A strong
positive correlation between two traits means
that increasing one trait would be
accompanied by an increasing in the other
trait also. But, if the correlation is negative,
increasing one trait would result in the
reduction of the other. Such types of traits are
governed by a pleiotropic effect of genes or
linkage of genes controlling the inheritance of
two or more characters (Dabholkor, 1992).
Similar findings were reported by Mugabe et
al. (2019).

Table 5. Phenotypic Pearson’s correlation matrix for 11 traits of 72 chickpea grown under controlled
environment at Ethiopian Biodiversity Institute, 2021.

Traits PLH GR NF NF GR NPB GR FW SR Test 1 Test 2

AF PLH BF AF NF AF NPB
PLHBF 0.59** -0.12* 0.51** 0.62** 0.41** 0.71** 0.50** 0.44** 0.29** -0.26** -0.22**
PLHAF 0.65** 0.68** 0.84** 0.79** 0.60** 0.66** 0.69** 0.56** -0.45** -0.42**
GRPLH 0.31** 0.46** 0.59** 0.08ns 0.34** 0.41** 0.41** -0.31** -0.29**
NFBF 0.68** 0.42** 0.61** 0.55** 0.56** 0.61** -0.52** -0.51**
NFAF 0.89** 0.80** 0.75** 0.66** 0.57** -0.51** -0.48**
GRNF 0.65** 0.70** 0.62** 0.47** -0.4** -0.40**
NPBAF 0.67** 0.55** 0.47** -0.41** -0.39**
GRNPB 0.60** 0.54** -0.45** -0.43**
FW 0.74** -0.75** -0.71**
SR -0.87** -0.85**
Test 1 0.98**

PLHBF=Plant height before frost treatment, PLHAF= Plant height after frost treatment, GRPLH= Growth rate of plant
height during frost treatment, NFBF=Number of foliage before frost treatment, NFAF=Number of foliage after frost
treatment, GRNF=Growth of number of foliage during frost treatment, NPBAF=Number of primary branch after frost
treatment, GRNPB=Growth rate of number of primary branch during frost treatment, FW=Fresh weight, SR=Survival
rate, Test 2=Freezing test 2, and Test 1=Freezing test 1

As a conclusion, frost tolerant screening of wide range of chickpea growing

chickpea seedlings in controlled environment
using growth chamber has enabled the
identification of frost tolerant genotypes at an
early growth stage. The frost-tolerant
genotypes were selected based on SR and FTR
values (Freezing test 1 and freezing test 2).
Genotypes that were consistently rated as
frost-tolerant genotypes in both indices (SR
value of ≥ 0.8 and FTR score of 1 to 3) were
selected. Twenty six chickpea genotypes
(Table 2, genotypes listed from no. 1 to 26)
were identified as frost-tolerant genotypes
tested at seedling stage which can withstand
a temperature as low as -5°C. Based on these
findings, Ethiopian chickpea landraces have
genetic potential for frost resistance traits. It is
recommended that these genotypes can be
used for future frost-tolerant cultivar
development program through implementing
multi-locations and multi-year field trails to
test the frost tolerance adaptation to a more
environments.
ACKNOWLEDGEMENTS
Ethiopian Biodiversity Institute, Debre Zeit
Agricultural Research Center, and International
Center for Agricultural Research in the Dry Areas
for providing us chickpea genotypes for the study.
REFERENCES
1. Bourion, V., Lejeune-Henaut, I., Munier-Jolain,
N. and Salon, C. (2003). Cold acclimation
of winter and spring peas: carbon
partitioning as affected by light intensity.
Eur. J. Agron. 19:535-548.
2. Castañeda-Álvarez, N.P., Khoury, C.K.,
Achicanoy, H.A., Bernau, V., Dempewolf,
H., Eastwood, R.J., Guarino, L., Harker,
R.H., Jarvis, A., Maxted, N., Müller, J.V.,
Ramirez-Villegas, J. Sosa, C.C., Struik,
P.C., Vincent, H. and Toll, J. (2016).
SINET: Ethiop. J. Sci., 44(2), 2021
Global conservation priorities for crop
wild relatives. Nat. Plants. 2:16022.
3. Croser, J.S., Clarke, H.J., Siddique, K.H.M. and
Khan, T.N. (2003). Low-temperature
stress: implications for chickpea (Cicer
arietinum L.) improvement. CRC Crit Rev
Plant Sci. 22:185-219.
4. CSA (2019) Central Statistical Agency.
Agricultural Sample Survey Report on
Area and Production of Crops for Private
Peasant Holdings, Meher Season. Central
Statistical Agency, statistical bulletin 589,
Addis Ababa
5. Dabholkor, A.R. (1992). Elements of Biometrical
Genetics. Concept Publishing Company.
New Delhi.
6. Fiebelkorn, D. (2013). Characterization of
selected winter hardiness traits in pea
(Pisum sativum L.). MS thesis, North
Dakota State Univ., Fargo.
7. Geletu Bejiga, Million Eshete and Yadeta
Anbessa (1996). Improved cultivars and
production technology of chickpea in
Ethiopia. Research Bulletin No. 2.
DebreZeit Agricultural Research Center,
Alemaya University of Agriculture, Debre
Zeit, Ethiopia.
8. Heidarvand, L., Amri, R.M., Naghavi, M.R.,
Farayedi, Y., Sadeghzadeh, B. and
Alizadeh, K.H. (2011). Physiological and
morphological characteristics of chickpea
accessions under low temperature.
Russian J Plant Physi. 58:157-163.
9. Jha, U.C., Chaturvedi, S.K. Bohra, A., Basu, P.S.,
Khan, M.S., and Barh, D. (2014). Abiotic
stresses, constraints and improvement
strategies in chickpea. Plant Breed.
133:163-178.
10. Kanouni, H., Khalily, M. and Malhotra, R.S.
(2009). Assessment of Cold Tolerance of
Chickpea at Rainfed Highlands of Iran.
American-Eurasian J. Agric. & Environ. Sci.
5: 250-254.
11. Kumar, G.R., Sakthivel, K., Sundaram,R.M.,
Neeraja, C.N., Balachandran, S.M, Rani,
N.S., Viraktamath, B.C. and Madhav,
M.S. (2010). Allele mining in crops:
Prospects and potentials. Biotechnol. Adv.
28: 451-461.
12. Maphosa, L., Richards, M.F., Norton, S.L. and
Nguyen, G.N. (2020). Breeding for Abiotic
Stress Adaptation in Chickpea (Cicer
arietinum L.): A Comprehensive Review.
Crop Breed Genet Genom. 2: e200015.
13. Maqbool, A., Shafiq, S. and Lake, L. (2010).
Radiant frost tolerance in pulse crops-a
review. Euphytica 172:1-12.
14. MINITAB (1998). MINITAB users guide,
release 14 .MINITAB Inc
159
15. Mir, A.H., Bhat, M.A., Fayaz, H., Dar, S.A.,
Maqbool, S., Bhat, N.A., Thudi, M. and
Mir, R.R. (2019). Assessment of cold
tolerance in chickpea accessions in North-
Western Himalayas of Jammu and
Kashmir, India. J Pharmacogn Phytochem
8: 2268-2274.
16. Mugabe, D., Coyne, C.J., Piaskowski, J., Zheng,
P., Ma, Y., Landry, E., McGee, R., Main,
D., Vandemark, G., Zhang, H. and Abbo,
S. (2019). Quantitative trait loci for cold
tolerance in chickpea. Crop Sci. 59:573-582.
17. Mulugeta Assefa, Beneberu Shimelis, Punnuri,
S., Sripathi, R., Whitehead, W., and Singh,
B. (2014). Common Bean Germplasm
Diversity Study for Cold Tolerance in
Ehtiopia. Am. J. Plant Sci. 5:1842-1850.
18. Nezami, A., Bandara, M.S. and Gusta, L.V.
(2012). An evaluation of freezing
tolerance of winter chickpea (Cicer
arietinum L.) using controlled freeze
tests. Can. J. Plant Sci. 92:155-161.
19. Nigusie Girma, Asnake Fikre and Ojiewo, C.O.
(2017). The genotypic and phenotypic
basis of chickpea (Cicer arietinum L.)
cultivars for irrigation-based production
in Ethiopia. J. Agric. Sci. 9: 229-236.
20. Rao, N.K. (2004). Plant genetic resources:
Advancing conservation and use through
biotechnology. Afr. J. Biotechnol. 3:136-145.
21. Sassenrath, G. E., Ort, G. E. and Portis, A. R.
(1990). Impaired reductive activation of
stomatal biophosphatases in tomato
leaves following low temperature
exposure at high light. Arch. Biochem.
Biophy. 282:302-308.
22. Singh, K.B., Malhotra, R.S. and Saxena, M.C.
(1989). Chickpea evaluation for cold
tolerance under field conditions. Crop Sci.
29:282-285.
23. Sintayehu Admas, Teklehaimanot
Haileselassie, Kassahun Tesfaye, Eleni
Shiferaw and Flynn, K.C. (2021).
Evaluation of Ethiopian chickpea (Cicer
arietinum L.) genotypes for frost tolerance.
Acta Agric. Slov. 117:1-14.
24. Thomas, R., Vaughan, I. and Lello, J. (2013).
Data analysis with R statistical
software. A guidebook for scientists. Eco-
explore.
25. Wu, Y.F., Zhong, X.L., Hu, X., Ren, D.C., Lv,
G.H., Wei, C.Y. and Song, J.Q. (2014).
Frost affects grain yield components in
winter wheat. N. Z. J. Crop Hortic. Sci.
42:194-204.
26. Yadav, S,K. (2010). Cold stress tolerance
mechanisms in plants. A review. Agron
Sustain Dev. 30: 515-27.
27. Zhu, J., Pearce, S., Burke, A., See, D.R., Skinner,
D.Z., Dubcovsky, J. and Garland-
160 Sintayehu Admas et al.

Campbell, K. (2014). Copy number and frost tolerance in hexaploid wheat. Theor.
haplotype variation at the VRN-A1 and Appl. Genet. 127:1183-1197.
central FR-A2 loci are associated with
SINET:Ethiop. J. Sci., 44(2): 161–171, 2021 ISSN: 0379–2897 (PRINT)
© College of Natural and Computational Sciences, Addis Ababa University, 2021 eISSN: 2520–7997

Date received: July 07, 2021; Date revised: October 10, 2021; Date accepted: September 24, 2021
DOI: https://dx.doi.org/10.4314/sinet.v44i2.3

Child nutritional status, mothers’ nutritional knowledge and practice and Household food security
status in Tehuledere Woreda, South Wollo, Ethiopia

Ahmed Indris1, Dilu Shaleka2 and Mogessie Ashenafi1,*

1Center for Food Security Studies, 2Center for Gender Studies, College of Development Studies, Addis Ababa
University, Ethiopia. E-mail: mogessie.ashenafi@aau.edu.et

ABSTRACT: Child under nutrition due to household food insecurity remains critical issues in many
households in Ethiopia. Literature in nutrition knowledge and practice of mothers and the nutritional status of
their infants is scanty. This study aimed to assess the nutritional status of six to 23-month-oldchildren,
mothers‟ knowledge, attitude and practice of child nutrition and household food security status in a semi-
urban and a rural kebeles. A community based cross-sectional study was undertaken in semi-urban and rural
kebeles in Tehuledere Woreda (district) to assess household food security and nutritional status of six to 23-
month children. A total of 245 mother-child pairs were selected randomly from the two kebeles.
Anthropometric indices were used to determine the nutritional status of under-two children. Child age,
weight and height were measured and used to calculate weight-for-age, weight-for-height and height-for-age
Z-scores. Composite Index of Anthropometric Failure (CIAF) was calculated to determine total malnutrition.
Household food insecurity access scale (HFIAS) was used to assess food security status. Structured
questionnaires were used to collect data on mothers‟ knowledge, attitude and practice in child nutrition, food
diversity and child feeding. Data were statistically analyzed. Stunting was noted in 7.5% and 17% of under-
two children in the urban and rural kebeles, respectively. Similar levels of thinness (6%) were observed in both
kebeles. There were more under-weight children in the semi-urban (5.2%) than in the rural (3.6%) kebeles.
Chronic energy deficiency was noted in 20% and 15% of the children in semi-urban and rural kebeles,
respectively. CIAF was higher in children in the semi-urban kebele (48%) than in the rural kebele (31%). A small
proportion of study households were food secure (17.9%). The rest were either mildly (54.4%) or moderately
(27.8%) food insecure. Average knowledge of child nutrition among mothers in the semi-urban and rural
kebeles was very low (about 34% and 37%, respectively).The low anthropometric measurements of the
children in this study could be due to poor food diversity, insufficient food intake, and poor nutritional
knowledge and practice of mothers. Creating awareness in child feeding practices and diet diversity is
recommended.

Keywords: Food Security, Nutrition, Knowledge, Practice, Nutritional status, under-two children

status of mothers, infants and children has been a

INTRODUCTION
Undernourished infants and children are slow in their
physical and mental development and have increased
vulnerability to infections and early death. Nutrition
has increasingly been recognized as a basic pillar for
social and economic development. Malnutrition is a
major cause of morbidity and mortality in low-income
countries including Ethiopia (Wagnew et al., 2018).
The reduction of infant and young child malnutrition
is essential to achieve the sustainable development
goals. To this effect, WHO (2002) recommended best
feeding practices during the first year of life for
maximizing growth, development, and survival. To
meet their nutritional requirements, infants should
receive nutritionally adequate and safe solid or
semisolid complementary foods while breastfeeding
continues for up to two years of age or beyond food
insecurity and under nutrition remain critical issues in
poor households (Liu et al., 2016). Poor nutritional
_____________________
*Author to whom correspondence should be addressed.
consistent problem in Ethiopia. According to Ethiopia
Demographic and Health Survey (EDHS),
undernutrition is an underlying cause of 53% of infant
and child deaths (EPHI and ICF, 2021).
Although rates of stunting and underweight
have decreased over the past decades in Ethiopia, 37%
and 21% of infants and under-five children are still
stunted or underweight, respectively (EPHI and ICF,
2021).). Lack of dietary diversity and inappropriate
infant feeding practices contribute to the high rates of
under-nutrition in infants. Only half of infants are
exclusively breastfed with introduction of
complementary foods at the appropriate time, and
only 11% of young children are receiving a minimal
acceptable diet (EPHI and ICF, 2021).).Children in rural
areas are more stunted (40%) than those in urban
areas (26%), and great regional variations persist in
various regions in Ethiopia(EPHI and ICF, 2021).).Thus,
This study aimed at assessing it would be interesting
to assess child nutritional status and mother‟s KAP in
162 Ahmed Indris et al

child nutrition in a semi-urban and rural setting in literature and for logistic reasons. The two kebeles
Ethiopia. were selected based on the Woreda‟s report about the
The main objective of this study was, therefore, shift in agricultural products towards cash crops
to assess household food security status, mothers‟ (khat) and to study its effect on the household‟s food
knowledge, attitude and practice (KAP) of child security status and child nutritional status (TWOA,
nutrition and nutritional status of six to 23-month- 2016). Study households were selected randomly.
oldchildren in a semi-urban and a rural kebele in The representative sample for proportions was
Tehuledere Woreda, South Wollo Zone. This study determined as in Cochran (1963) using the equation
will help to come up with better intervention
methods, such as giving adequate training, to
improve the knowledge and practice of mothers in
child nutrition and, consequently the health status of
Where: n0 is the sample size, Z2 is the abscissa of the
their children,
normal curve that cuts off an area α at the tails (1 – α
equals the desired confidence level, e.g., 95%), e is the
desired level of precision (0.05), p is the estimated
MATERIALS AND METHODS
proportion of an attribute that is present in the
Description of the Study Area population, and q is 1-p. The value for Z is found in
statistical tables which contain the area under the
This study was carried out in a semi-urban and a rural
normal curve (Z statistics: 1.96)
kebele in Tehuledere Woreda (district), South Wollo
n0= (1.96 x 1.96 x 0.46 x 0.54) / (0.05 x 0.05) = 382.
Zone. The following description was based on
The sample size (n0) was adjusted using the following
information obtained from the woreda‟s agricultural
equation (Cochran, 1963).
office (TWOA, 2016). The worreda is located north of
Addis Ababa at a distance of 430 km. It covers an area
of 44,030 hectares and has a population of 143,445,
consisting of 53% males and 47% females. According
to the same source, the woreda is subdivided into 19
Where
rural and two small semi-urban kebeles (kebele is the
n0 – the initial sample size; n -adjusted sample size;
smallest administrative unit in Ethiopia).The semi-
and N -the population size.
urban and the rural kebeles considered in this study
n=382/1+ ((382-1)/625) =237
had a population of 7,103 and 4,571, respectively; the
total number of children in the woreda during the
Finally, with 5% non-response rate, the size is
study period (2019) was 72,494; and the number of six
249 households each with mother – child pair (6-23
to 23-month-oldchildren was 339 and 286 in the semi-
months). However, since more respondents than the
urban and rural study kebeles, respectively. The most
planned were found, 245 households (133 from the
important crops in the area are sorghum, teff and
semi-urban and 112 from rural kebele) with mother–
maize, whereas among the pulses, the main are
child pair were randomly selected for the study.
chickpea, grass pea and haricot bean, sometimes inter-
A community-based cross-sectional study design
cropped with sorghum and maize. Vegetables and
was used to collect data on household food insecurity
fruits are produced where farmers have access to
access (HFIA) and investigate food safety KAP of
small-scale irrigation, especially around Haik and
mothers in the study households in both kebeles.
Ardibo lakes. They also use rivers and water ponds for
Information, consisting of household socio-economic
the same purpose. A recent trend of the farming
and demographic parameters, such as age, sex,
system in the area is the increasing production of cash
educational level, monthly income, and occupation,
crops, such as khat, as source of income for
was also collected.
households.
Data collection
Study population:
Data on household socio-demographic
The study population was composed of
characteristics, household food security status, and
households withsix to 23-month-oldchildren, residing
mothers‟ KAP on child nutrition were collected during
in both rural and semi-urban kebeles. Sample size was
face-to-face interviews using structured
determined according to Cochran (1963). A total of
questionnaires.
245 households (133 from the semi-urban and 112
Household food insecurity status was assessed
from rural kebele) with mother–child pair were
by determining the household food insecurity access
randomly selected for the study.
scale (HFIAS)(Coates et al., 2007) as validated for
Ethiopia by Gebreyesus et al., 2015). A set of nine
Sampling and sample size determination
questions related to three different domains of food
The woreda was selected purposively due to insecurity access were used: (i) anxiety and
high prevalence of food insecurity reported in the uncertainty about the household food supply; (ii)
SINET: Ethiop. J. Sci., 44(2), 2021 163

insufficient quality in terms of variety and preferences Percentage of total knowledge/attitude/practice

of the type of food and (iii) insufficient food intake in
terms of reducing quantity of food. The nine
occurrence questions were asked along with the
frequency of occurrence within a recall period of four
weeks to determine whether a household was mildly,
moderately or severely food insecure. Households
which experienced none of the food insecurity
conditions, or rarely experienced anxiety, were
considered as food secure. Mildly food insecure
households faced anxiety and uncertainty more
frequently and sometimes compromised the quality of
food they ate. Moderately food insecure households
sacrificed more frequently the quality of food they ate,
and/or rarely or sometimes reduced the size of meals
or number of meals per day. Severely food insecure
households often cut back on meal size or number of
meals, and/or ran out of food, went to bed hungry, or
went a whole day and night without eating (Coates et
al., 2007). Food security status of households as food
secure, mildly food insecure and moderately food
insecure was determined according to (Coates et al.,
2007) using the formula:
Number of households with HFIA category
----------------------------------------------------------- X 100
Total number of households with a HFIA category
Mothers‟ KAP in child nutrition was assessed
with respect to child diet diversity and feeding
practice. Qualitative data, collected through
questionnaires, were converted to percentages and
used as indicators for level of knowledge, attitude and
practice on food safety. The assessment was based on
the FAO guidelines on food safety and nutrition
related KAP assessments (Macías and Glasauer 2014).
The knowledge, attitude and practice of the
population was calculated for each question by
dividing the total number of correct responses by the
number of respondents who answered the particular
question (Macías and Glasauer 2014). Respondents
who did not answer the question, or for whom
information was incomplete, were excluded.
among population =
KAP of mothers in child nutrition was classified
using the Bloom‟s cut-offs for KAP studies, namely,
>80% as good or positive; 60%-79% as moderate,
neutral or fair; and (<60%) as poor or negative as used
in Alrazeeni (2021).
Weight of the child was measured using a baby
hanging scale. Height was measured without shoes by
using either a horizontal length scale (for those who
could not stand erect) or standing height scale (for
infants above 13 months of age). Mid upper arm
circumference (MUAC) was measured using tape at the
midpoint between the shoulder and elbow of the
child. Information on age was obtained from a written
birth card or verbal information from the mother.
Anthropometric analysis was made by using
Emergency Nutritional Assessment Software. Height-
for-age (H/A), weight-for-height (W/H), and weight-
for-age (W/A) below -2SD, indicated stunting;
wasting (thinness); and underweight, respectively.
The data was analyzed using SPSS v22. Assessment
was also made in the two kebeles with regards to
knowledge, attitude and practice of mothers on child
nutrition.
Composite Index of Anthropometric Failure
(CIAF) was used to assign our study subjects into
„failure‟ and „no failure” groups, based on their
growth status (Bejarano et al., 2019, Destaw et al.,
2021). „No failure‟ groups attained normal growth
status by all the three conventional indices, namely
stunting, thinness, underweight/excess weight.
Children who failed to attain a normal growth in, at
least, one of the standard indices were categorized as
“Failure” (Table 1).
Thus, proportion of Anthropometric Failure is
the sum of the percentages of each failure groups
indicated in Table 8, except group A, and can be
calculated as in (Bejarano et al., 2019):
CIAF = (1 – A)%.

Table 1 CIAF categories for children of age group six to 23 months.

Categories Description Stunting Under- Thin- Over-

weight ness weight
A Without anthropometric failure No No No No
B Thinness only No No Yes No
C* Thinnessand Underweight No Yes Yes No
D Stunting, Thinness, and Underweight Yes Yes Yes No
E* Stunting and Underweight Yes Yes No No
F Stunting only Yes No No No
G Overweight No No No Yes
H Stunting and excess weight (overweight and obese) Yes No No Yes
Y* Underweight only No Yes No No

Adatpted from Bejarano et al. (2019)

164 Ahmed Indris et al

Data analysis their children and could properly answer questions

Descriptive analysis such as means, median and
frequencies were used to analysevariables of
household socio-demographic status. Inferential
statistics such as Pearson‟s chi-square testwas used to
see associations between categorical variables by cross
tabulation. Significance was determined at p=0.01 or
p=0.05. Student‟s t-test was used to compare means
between semi-urban and rural households,
Ethical consideration
As measurements were taken from human
subjects, research ethical issues were duly addressed.
Study conditions regarding anonymity,
confidentiality and willful participation were
explained to respondents and informed oral consent
was obtained from them.
RESULTS
In this study, only mothers with children six to 23-
months were considered as they were the closest to
regarding nutrition KAP at household level.
Socio-demographic information
The median age of mothers was 32 years and
over 90% of the mothers were young adults between
20 to 40 years. Over 90% were married and Muslim in
religion. More than half of the households had
between four to five children and over 10% had more
than six children. Only 13% of respondents from both
kebeles were illiterate while the majority (80%) of the
respondents could read and write or attended
primary or secondary school (Table 2).
The majority of the mothers in both kebeles were
farmers, although the proportion was higher (70%) in
the rural kebele (Table 2). About 57% of the semi-
urban and 75% of rural households had a monthly
income of <etb 2000 (USD 1=ETB 30 during the study
period).In general, the difference in socio-economic
and demographic parameters between households in
semi-urban and rural kebeles was not significant at
p=0.05. A significant association was, however,
observed between number of children and child
under-nutrition (p<0.1).

Table 2. Socio-economic and demographic status of the study population.

Semi-urban Rural
Variables Category n=133 n=112
No. (%) No. (%)
Age group 20-40 133 (100 %) 106 (94.5%)
(mother) 41-50 0 6 (5.5%)
Marital status Married 119 (89.5%) 103 (92%)
Divorced/Widowed 14 (10.5 %) 9 (8%)
Number of children <2 Three or less 25 (18.8%) 27 (24.1%)
years per household Four to six 94 (70.7%) 72 (64.3%)
>Six 14 (10.5%) 13 (11.6%)
Age of infant (months) 0-12 25 (18.8%) 34 (30.4%)
Above 12 108 (81.2%) 78 (69.6%)
Religion Muslim 124 (93.2%) 106 (94.6%)
Orthodox 9 (6.8%) 6 (5.4%)
Education (mothers) Illiterate 24 (18%) 8 (7.2%)
Read and write 36 (27.1%) 31 (27.7%)
Grades 1-8 62 (46.6 %) 66 (58.9%)
Grades 9-12 11 (8.3%) 6 (5.4%)
Occupation House wife 43 (32.3%) 28 (25%)
Farmer 78 (58.6%) 79(70.5%)
Merchant 12 (9.1%) 5 (4.5%)
Monthly household <2000 76 (57.1%) 84 (75%)
income (ETB) 2001-4000 41 (30.8%) 28 (25%)
4000 and above 16 (12.1%) 0

Household food security status
Findings of HFIAS indicated that a high
proportion of households (>75%) in both kebeles
rarely or sometimes experienced anxiety and
uncertainty or ate insufficient quality of food during
the four weeks prior to the interview. Although a
higher proportion of households in the rural kebele
(20%) had to eat insufficient quantity of food than
those in the semi-urban (5%) kebele, the difference
was, however not significant at p=0.05.(Table 3).
SINET: Ethiop. J. Sci., 44(2), 2021 165

Table 3. Mean values of food insecurity experience among rural (n=112) and semi-urban (n=133) households.

Household food insecurity experience Occurrence Frequency

Location No. % Rarely Sometimes Often
Anxiety and uncertainty Rural (n=112) 103 77.4 36.8% 40.6% 0
Semi-urban (n=133) 85 75.9 42% 33.9% 0
Reduced quality of food Rural (n=112) 100 75.2 36.1% 39.1% 0
Semi-urban (n=133) 85 75.9 41.1% 34.8% 0
Reduced quantity of food Rural (n=112) 7 5.3 5.3 0 0
Semi-urban (n=133) 23 20.5 13.4% 7.1% 0
Hunger Rural (n=112) 0 0 0 0 0
Semi-urban (n=133) 0 0 0 0 0
Rarely (1 or 2 times), sometimes (3 to 10 times), Often (more than 10 times)

Both semi-urbanand rural households belonged
to different categories of food security. The proportion
of mildly food insecure households in our study was
slightly higher in the semi-urban (57%) than in the
rural kebele (52%).Moderately food insecure
households in both kebeles had comparable
proportions (Table 4).There was, however, no severely
food insecure household in both kebeles. In general,
over 80% of the study population was in a state of
some degree of food insecurity.

Table 4. HFIA prevalence among households in the study kebeles.

Category Semi-urban (n=133) Rural (n=112) Total (n=245)

Food secure 17.8% 18% 17.9%
Mildly food insecure 56.5% 52.2% 54.4%
Moderately food insecure 25.7% 29.9% 27.8%
Severely food insecure 0 0 0

Knowledge, attitude and practice (KAP) of mothers on undernutrition

Knowledge: Average knowledge of child
nutrition among mothers in the semi-urban and rural
kebeles was very low (about 34% and 37%,
respectively) (Table 5).Although a higher proportion
of rural mothers (76%) had knowledge of how long
breastfeeding should continue than mothers in semi-
urban households (61%), there was no significant
difference in general knowledge of child nutrition
between mothers in semi-urban and rural households
at p=0.05. Knowledge of when to start complementary
feeding and the reasons for giving complementary
foods at six months was quite high and similar in both
kebeles (98%-100%).There was a wide gap between
semi-urban (100%) and rural (30%) households in
knowledge of appropriate consistency of meals. Most
rural households thought that complementary meals
should be watery in consistency because it was easier
for the child to swallow thinner meals. Knowledge of
the diverse foods that could be used to enrich starch-
based meals (e.g., porridge) was poor. Almost all
mothers in both kebeles (96%-100%) knew only about
legumes as enrichment foods to the starch-based
complementary child meals. All mothers in both
kebeles knew of over five different ways to encourage
children to eat.

Table 5. Mothers’ knowledge on child nutrition.

Correct response
Semi-urban (n=133) Rural (n=112)
1: Continued breastfeeding 81 (60.9%) 84 (75.5%)
2: Age of start of complementary foods 130 (97.7%) 112 (100%)
3: Reason for giving complementary foods at six months 133 (100%) 112 (100%)
4: Appropriate consistency of meals 133 (100%) 33 (29.5%)
5: Reason for appropriate consistency of meals 133 (100%) 33 (29.5%)
6: Dietary diversity and ways of enriching porridge by adding:
¨ Animal-source foods 0 0
¨ Pulses and nuts: 128 (96.2%) 112(100%)
¨ Vitamin-A-rich fruits and vegetables 0 0
¨ Green leafy vegetables 0 0
¨ Energy-rich foods 0 0
¨ Others 5 (3.8/%) 0
7: Responsive feeding: Number of ways to encourage young 5 (3.8/%) 6 (5.4%)
children to eat
Average Knowledge 33.9% 36.6%
166 Ahmed Indris et al

Attitude: Over 90% of mothers in both kebeles
were self-confident in preparing food for the child
and believed that it was good to give different types
of food to child each day, to feed child several times
each day and to continue breastfeeding beyond six
months (Table 6). Moreover, they believed that
continuing breastfeeding beyond six months was not
difficult (>90%). Although the majority of mothers in
both localities believed that it was good to feed a child
with different types of foods several times each day,
about 60% to 70%thought that this was difficult to
achieve. When asked why, they responded that it was
difficult because of low income of households.
Average positive attitude towards appropriate child
nutrition in the study kebeles ranged between 75%
and 80%.

Table 6. Mothers’ attitude towards child nutrition.

% Positive attitude
Perceived benefit or difficulty Semi-urban Rural
(n=133) (n=112)
Good to give different types of food to child each day 132(99.2%) 103 (92.0%)
Not difficult to give different types of food to child each day 33(24.8%) 49 (43.8%)
Goodto feed child several times each day 125 (94.0%) 86 (76.8%)
Not difficult to feed your child several times each day 58 (43.6%) 25 (22.3%)
Good to continue breastfeeding beyond six months 130(97.7%) 112 (100%)
Not difficult to continue breastfeeding beyond six months 130(97.7%) 103 (91.9%)
Self-confidence in preparing food for the child? 132 (99.2%) 109 (97.3%)
Support by health extension agents on child feeding 70 (52.6%) 61 (54.5%)
Average positive attitude 79.5% 74.9%

Practice: Over 98% of all mothers in both kebeles
fed breast milk to their children. Almost all mothers in
this study supplemented breast milk basically with
starchy foods. Complementary meals were dominated
by grains and legumes (>95%) followed by eggs (70-
80%) animal milk (60-80%) and flesh foods (around
30%) in both kebeles (Table 7).Consumption of food
groups such as vitamin A-rich fruits and vegetables
was very low (<7%). However, more mothers in semi-
urban (81%) than in rural kebele (56%) used other
fruits and vegetables to supplement their child‟s meal.
Average appropriate practice in child nutrition was
low (<45% in both kebeles)
A low proportion (<15%) of mothers in both
kebeles fed their child five times and less than 55%
only three times during the previous day. In general,
mothers in both study areas showed positive attitude
towards appropriate child nutrition (75%-805).

Table 7. Mothers’ practice in child nutrition.

Appropriate practice
Semi-urban (n=133) Rural (n=112)
1. Continued breastfeeding 130 (97.7%) 111 (99.1%)
2. Dietary diversity
Group 1: Foods made from grains, roots and tubers 132 (99.2%) 112 (100%)
Group 2: Foods made from Legumes and nuts 129 (97.0%) 101 (90.2%)
Group 3: Dairy products Infant formula(4x)
Milk, (powdered or fresh animal milk (3x) 80 (60.2%) 87 (77.7%)
Yogurt or drinking yogurt (1x) 2 (1.5%) 0
Cheese or other dairy products (1x) 6 (4.5%) 0
Group 4: Flesh foods 41 (30.8%) 39 (34.8%)
Group 5: Eggs 106 (79.7%) 81 (72.3%)
Group 6: Vitamin A, fruits and vegetables
Any dark green vegetables (kale) 2 (1.5%) 5 (4.5%)
Ripe mangoes, ripe papayas, musk melon 9 (6.8%) 0
Group 7: Other fruits and vegetables 107 (80.5%) 63 (56.3%)
Any oil, fats, or butter or foods made with these 15 (11.3%) 92 (82.1%)
Any sugary foods 0 0
Condiments for flavor 56 (42.1%) 54 (48.2%)
3. Minimum meal frequency
Number of times the child ate foods the previous day during 2x 2 (1.5%) 0
the day or at night? 3x 73 (54.9%) 57 (50.9%)
4x 48 (36.1%) 38 (33.9%)
5x 10 (7.5%) 17 (15.2%)
Average appropriate practice 41.3% 45.4%

Child Nutritional Status 23 months made up 44.1% of the under-two children

A total of 245six to 23-month-oldchildren from (Table 8). About 12% of all children were stunted with
both kebeles were considered in this study. those in the age group 12-17 months showing the
Distribution by sex was almost even (51% boys and highest proportion (18%).
49% girls). Children belonging to the age group 18 to
SINET: Ethiop. J. Sci., 44(2), 2021 167

Table 8 Nutritional status of six to 23-month-oldchildren in the study households.

Thinness W/H, <-2 z-score Stunting H/A, <2 z-score Underweight W/A, <-2 z-score MUAC
Age (Months <125 mm
Total
6-11 59 5 (10.1%) 7 (14%) 6 (10.1%) 18 (30.5%)
12-17 78 4 (5.1%) 14 (17.9%) 3 (3.8%) 15 (19.2%)
18-23 108 5 (4.6%) 9 (8.3%) 2 (1.9%) 11 (10.2%)
Total 245 15 (6.1%) 30 (12.2%) 11 (4.5%) 44 (18%)

About 4.5%of under-two children in this study
were underweight with higher proportion (11%) in
infants under six months. MUAC measurements
indicated that about 18% of the children suffered
from, at least, moderate under nutrition with the
highest proportion (56%) among children under six
months old.
A total of 39% of all children considered in this
study have an anthropometric failure (Table 9).
Children in the semi-urban kebele had lower values
(31%) of anthropometric failure when compared to
those in the rural kebele (48%). The difference,
however, was not significant (p>0.05). Most infants
suffered from only one anthropometric failure. A few
(about 5%) had combined failure of thinness and
underweight or stunting and excess weight.

Table 9. CIAF among six to 23-month children in the study area.

Category Semi-urban Rural Total

(n=133) (n=112)

A Without anthropometric failure(normal) 92 58 150

B Thinness only 8 7 15
C Thinness and Underweight 3 3 6
D Stunting, Thinness, and Underweight 0 0 0
E Stunting and Underweight 0 0 0
F Stunting only 11 19 30
G Excess weight (overweight and obese) 10 16 26
H Stunting and excess weight 2 5 7
I Underweight only 7 4 11
Total 133 112 245
Total anthropometric failure 41 54 95
CIAF 30.8% 48.2% 38.8%

of infants per household would put most households

DISCUSSION
In KAP studies, it is useful to consider socio-
demographic factors as they affect consumption rate
of diverse food groups and dietary diversity among
children aged 6–23 months(Baek and Chitekwe
(2019).The higher number of dependent children per
household, in our study, could expose families to, at
least, mild or moderate levels of food insecurity
(Miskir et al., 2017). The fact that only a small number
of respondents from both kebeles were illiterate
indicated that the communities, in both kebeles, were
appropriate for training interventions on nutrition by
health extension workers. According to Sultana and
Hasan (2020),in Bangladesh, child health and
nutrition depended on the mother‟s knowledge and
beliefs child nutritional practices.
A large proportion of our respondents from the
semi-urban and rural households had low monthly
income. Increased monthly income was significantly
associated with better child nutritional status (p<0.1).
The general low income combined with high number
in a marginal situation. According to Nankinga et
al.(2019), maternal income wasa significant
determinant of child nutritional status in Uganda.Oh
et al.(2019) also showed that family income had a
strong relationship with healthy child nutritional
practices in Senegal.
The proportion of food secure households in our
study was low. This proportion was much higher than
that reported from Boset woreda, East Shewa zone,
Ethiopia (Moroda et al.,2018),but lower than that
reported from Damot Gale woreda, Wolayta zone,
Ethiopia (Mota et al., 2019). In general, over 80% of the
study population was in a state of some degree of
food insecurity. Lower proportions of food insecurity
were reported from other parts of Ethiopia (Mota et
al., 2019; Berra, 2020). Considering the fact that the
survey was made immediately after harvest time,
when food in farming communities is usually
relatively abundant, the level of food insecurity was
telling of what could happen during the slim months
when amount of stored food in most households
gradually dwindles to nil.
168
It is crucial for mothers to know that a child‟s
body is nourished and its health influenced by
different types of foods. In this respect, knowledge of
what kinds of foods to choose for the child and how to
prepare them is useful. For mothers, adequate
knowledge of nutrition would help them to use
appropriate alternatives to fight against child
malnutrition using the means at hand. Moreover,
nutrition education improved the appropriate
complementary feeding knowledge and practice of
mothers (Muluye et al., 2020).Promotion of health and
nutrition education to mothers could enhance child
nutrition outcomes in Nigeria (Fadare et al.,. 2019).
Although the attitude of mothers towards giving
different types of food to the child each day was
considerably high, average appropriate practice in
child nutrition was poor. Consumption of diversified
diets and locally available nutrient-dense foods is
recommended to achieve adequate intake of macro
and micronutrients by infants and young children
(Kinabo et al., 2019).
The positive attitude shown by mothers in both
study areas towards appropriate child nutrition was,
unfortunately, not translated into practice mainly due
to low income of households. As nutritional
requirement of infants is continuously evolving,
infants should be fed with nutritionally adequate
complementary foods while breastfeeding continues
for up to two years of age or beyond(WHO, 2002).
Our observation of stunted children was lower
than the stunting observed in under-two children of
Orthodox Christian mothers in Tigray (Desalegn et al.,
2019). According to EPHI and ICF (2021),stunting of
children under five years in Ethiopia was high.
Stunting reflects failure to receive adequate nutrition
over a long period of time and is also affected by
recurrent and chronic illness. Height-for-age,
therefore, represents the long-term effects of
malnutrition in a population and is not sensitive to
recent, short-term changes in dietary intake EPHI and
ICF (2021). The proportion of children, in this study,
who suffered from thinness was much lower than that
reported among under-two children from elsewhere
in Ethiopia (Derso et al.,. 2017).
The proportion of under-two children in this
study who were underweight was higher than that
observed in children of similar age group from North
Ethiopia (Desalegn et al., 2019). The proportion of
moderate undernutrition among our under-two
children, as measured by MUAC was much lower than
that observed in northeast Ethiopia where a high
prevalence of acute malnutrition was observed among
under-two children (Gizaw et al.,. 2018).
CONCLUSION
The nutritional status of the children considered in
this study could be the consequence of a range of
factors such as poor food diversity, insufficient food
intake, and poor nutritional knowledge and practice.
Ahmed Indris et al
Child diet was not adequately diversified, thus,
leading to poor nutritional status of children. It is,
thus, recommended to give appropriate training to
mothers that can fill the gaps in child nutrition
knowledge and practice.
ACKNOWLEDGEMENTS
The paper is based on the M.Sc. study of AE. The Center of
Food Security Studies is acknowledged for facilitating the
study.
REFERENCES
1. Alrazeeni, D. 2021. Knowledge, attitude, and practice
toward COVID-19 among paramedics in Saudi
Arabia: Implications for educational program.
International Journal of Advanced and Applied
Sciences, 8: 71-77.
2. Baek, Y., Chitekwe, S. (2019) Sociodemographic factors
associated with inadequate food group
consumption and dietary diversity among infants
and young children in Nepal. PLoS ONE 14:
e0213610.
https://doi.org/10.1371/journal.pone.0213610
3. Bejarano, I., Oyhenart, E., Torres, M., Cesani, M., Garraza,
M., Navazo, B. (2019). Extended composite index
of anthropometric failure in Argentinean preschool
and school children. Public Health and Nutrition
22:3327-3335
4. Berra, W.(2020). Household Food Insecurity Predicts
Childhood Undernutrition: A Cross-Sectional
Study in West Oromia (Ethiopia). Journal of
Environment and Public Health. Volume 2020, Article
ID 5871980, ttps://doi.org/10.1155/2020
/5871980;2020.
5. Coates, J., Swindale, A., and Bilinsky, P.(2007).
Household Food Insecurity Access Scale (HFIAS)
for Measurement of Food Access: Indicator guide
(v.3). Food and Nutrition Technical Assistance
(FANTA) project.Washington DC.
6. Cochran,W.(1963). Cochran,. (1963). Sampling Techniques,
2nd Ed. John Wiley and Sons, Inc. NewYork.
7. Derso, T., Tariku, A., Biks, G.and Wassie, M.(2017).
Stunting, wasting and associated factors among
children aged 6-23months in Dabat health and
demographic surveillance system site: A
community based cross-sectional study in Ethiopia.
BMC Pediatrics 17: 96. doi: 10.1186/s12887-017-
0848-2
8. Desalegn, B., Lambert, C., Riedel, S., Negese, T., and
Biesalski, H.(2019). Feeding Practices and
Undernutrition in 6–23-Month-Old Children of
Orthodox Christian Mothers in Rural Tigray,
Ethiopia: Longitudinal Study. Nutrients.11:138;
https://doi.org/10.3390/nu11010138
9. Destaw, Z., Wencheko, E., Kidane, S., Challa, Y., Tiruneh,
M., Fite, M., Shaleka, D., and Ashenafi, M. 2021.
Use of modified composite index of
anthropometric failure and MUAC-for-age to
assess prevalence of malnutrition among school-
age children and adolescents involved in the
school feeding program in Addis Ababa, Ethiopia.
SINET: Ethiop. J. Sci., 44(2), 2021
BMC Nutrition (in press) https://doi.org/10.1186/
s40795-021-00471-x
10 EPHI and ICF, 2021. Ethiopia Mini Demographic and
Health Survey 2019: Final Report. Ethiopian Public
Health Institute (EPHI) [Ethiopia] and ICF. 2021.
Rockville, Maryland, USA: EPHI and ICF.
11. Fadare, O., Amare, M., Mavrotas, G., Akerele, D. and
Ogunniyi, A. (2019) Mother‟s nutrition-related
knowledge and child nutrition outcomes:
Empirical evidence from Nigeria. PLoS ONE14:
e0212775. https://doi.org/10.1371/journal.pone.
0212775
12. Gebreyesus, S., Lunde, T., Mariam, D. Woldehanna, T.,
and Lintjorn, B. (2015). Is the adapted Household
Food Insecurity Access Scale (HFIAS) developed
internationally to measure food insecurity valid in
urban and rural households of Ethiopia? BMC
Nutrition 1: 2. https://doi.org/10.1186/2055-0928-
1-2
13. Gizaw, Z., Woldu, W., and Bitew, B.(2018). Acute
malnutrition among children aged 6-59 months of
the nomadic population in Hadaleala district, Afar
region, northeast Ethiopia. Italian Journal of
Pediatrics44: 21. doi: 10.1186/s13052-018-0457-1.
14. Kinabo, J., Mamiro, P., Mwanri, A., Bundala, N., Kulwa,
K., Picado, J., Ntwenya, J., Nombo, A., Mzimbiri,
R., Ally, F., Salmini, A., Juma, A., Macha, E.,
Cheung, E., and Msuya, J. (2019). Adequacy of
macro and micronutrients in infants and young
children's diets in Zanzibar, Tanzania. African
Health Sciences, 19: 3063–3077. https://doi.org/b
10.4314/ahs.v19i4.28.
15. Liu, L., Oza, S., Hogan, D., Chu, Y., Perin, J., Zhu, J.,
Lawn, J., Cousens, S., Mathers, C, and Black, R.
(2016). Global, regional, and national causes of
child mortality: an updated systematic analysis for
2010 with time trends since 2000. Lancet; 388: 3027–
3035.
16. Macías, Y. and Glasauer, P. (2014).Guidelines for
assessing nutrition-related Knowledge, Attitudes
and Practices. Guidelines for assessing nutrition-
related Knowledge, Attitudes and Practices. Food
and Agriculture Organization of the United
Nations, Rome
17. Miskir, A., Godana, W., Girma, M., and G/Miskel,
F.(2017). Determinants of Acute Malnutrition
among Under-Five Children in Karat Town Public
Health Facilities, Southern Ethiopia: A Case
.
169
Control Study. Quality in Primary Care25: 242-252.
18. Moroda, G., Tolossa, D., and Semie, N. (2018). Food
insecurity of rural households in Boset district of
Ethiopia: A suite of indicators analysis. Agriculture
and Food Security. 7: 65
https://doi.org/10.1186/s40066-018-0217-x 7(1).
19. Mota, A. Lachore, S.and Handiso, Y.(2019). Assessment
of food insecurity and its determinants in the rural
households in Damot Gale Woreda, Wolaita zone,
southern Ethiopia. Agriculture and Food
Security8: 11 https://doi.org/10.1186/s40066-019-
0254-0.
20. Muluye, S., Lemma, T. and Diddan. T. (2020). Effects of
nutrition education on improving knowledge and
practice of complementary feeding of mothers with
6- to 23-month-old children in daycare centers in
Hawassa town, Southern Ethiopia: An institution-
based randomized control trial. Journal of Nutrition
and Metabolism
https://doi.org/10.1155/2020/6571583
21. Nankinga,O., Kwagala, B., Walakira, E. (2019) Maternal
employment and child nutritional status in
Uganda. PLoS ONE14: e0226720.
https://doi.org/10.1371/journal.pone.0226720
22. Oh, H., Kang, S., Cho, S., Ju, Y. and Faye, D. (2019)
Factors influencing nutritional practices among
mothers in Dakar, Senegal. PLoS ONE 14(2):
e0211787.
https://doi.org/10.1371/journal.pone.0211787
23. Sultana, P. and Hasan, K. (2020). Mother‟s Nutritional
Knowledge and Practice: A study on Slum Area of
Khulna City.European Journal of Medical and Health
Scienceshttp://dx.doi.org/10.24018/ejmed.2020.2.6
.582.
24. TWOA (2016). Tehuledere Woreda Office of Agriculture
(TWOA). Tehuledere, Ethiopia.
25. Wagnew, F., Tesgera, D., Mekonnen, M. and Alemu, A.
2018. Predictors of mortality among under-five
children with severe acute malnutrition, Northwest
Ethiopia: an institution based retrospective cohort
study. Archives of Public Health 76, 64.
https://doi.org/10.1186/s13690-018-0309-x
26. WHO (2002). Complementary feeding: report of the
global consultation, and summary of guiding
principles for complementary feeding of the
breastfed child. Geneva, Switzerland: World
Health Organization.
170 Ahmed Indris et al

Appendix I. Chi-square distribution of variables for nutritional status of children in semi-urban, Tehuledere Woreda.

Variable Category weight-for-age weight-for height Length/height for age MUAC (cm)
Under norma P wasting Norm P Stunting normal P Chronic energy normal p
weight l al deficiency
Sex of child Female 5(62.5%) 59 ns 4(50%) 60 Ns 6(50%) 58 ns 15(55.5%) 49 ns
Male 3(37.5%) 66 4(50%) 65 6(50%) 63 12(44.4%) 57
total 8 125 8 125 12 121 27 106
Household 500-1000 2(25%) 31 ns 4(50%) 29 ns 5(41.6%) 28 ns 15(55.5%) 20 .021
income 1001-1500 2(25%) 21 2(25%) 21 2(16.6%) 21 5(18.5%) 18 **
1501-2000 1(12.5%) 18 1(12.5%) 18 3(25%) 16 1(3.7%) 18
2001-2500 1(12.5%) 10 0 11 1(8.3%) 10 2(7.4%) 9
2501-3000 0 16 0 16 0 16 4(14.8%) 12
3001-4000 1(12.5%) 14 1(12.5%) 14 1(8.3%) 4 2(7.4%) 13
>4000 1(%12.5) 15 0 16 0 16 0 16
total 8 125 8 125 12 121 27 106
Education Illiterate 1(12.5%) 23 0 24 2(16.6%) 22 10(37%) 14
(mothers) Read &write 3(37.5%) 33 ns 3(37.5%) 33 ns 3(25%) 33 ns 6(22.2%) 30 ns
Grades 1-5 1(12.5%) 11 1(12.5%) 11 1(8.3%) 11 3(11.1%) 9
Grades 6-8 3(37.5%) 47 4(50%) 46 6(50%) 44 6(22.2%) 44
Grades 9-12 0 10 0 10 0 10 2(7.4%) 8
College/univ 0 1 0 1 0 1 0 1
Total 8 125 8 125 12 121
Number of Two 0 3 0 3 0 3 0 3
children Three 1(12.5%) 21 2(25%) 20 .ns 2(16.6%) 20 ns 5(18.5%) 17 ns
Four 3(37.5%) 32 ns 3(37.5%) 32 3(25%) 32 12(44.4%) 23
Five 2(25%) 37 2(25) 37 2(16.6%) 37 7(25.9%) 32
Six 2(25%) 18 1912.5%) 19 1(8.3%) 19 2(7.4%) 18
>six 0 14 0 14 0 14 1(3.7%) 13
Total 8 125 8 12 121 27
DD <4 food item 4(50% 4 .000* 1(12.5%) 7 Ns 4(33.3%) 4 .000* 2(7.4%) 6 ns
>= food item 4(50%) 121 7(87.5%) 118 8(66.7%) 117 25(92.6%) 100
Total 8 125 8
Chi-square significance: * = p < 0.01, ** =p < 0.05; n. s. not significant
SINET: Ethiop. J. Sci., 44(2), 2021 171

Appendix II. Chi-square distribution of variables for nutritional status of children in a rural kebele, Tehuledere Woreda.

Variable Category Nutritional status of children

weight-for-age weight-for-height Length/height for age/ MUAC (cm) Chronic energy deficiency (%)

Under nutrition normal P Wasting Normal P Stunting normal P Yes Normal p

Sex Female 1(33.3%) 57 ns 3(42.9%) 55 n.s 4(22.2%) 54 .006 9(52.9%) 49 n.s
Male 2(66.7%) 52 4(57.1%) 50 14(77.8%) 40 ** 8(47.1%) 46
Total 3 109 7 105 18 94 17 95
Household income 500-1000 3(100%) 38 n.s 4(57.1%) 32 n.s 5(27.8%) 36 n.s 9(52.9%) 32 n.s
1001-1500 0 26 2(28.6%) 24 5(27.8%) 21 4(23.5%) 22
1501-2000 0 18 1(14.3%) 12 5(27.85) 13 3(17.6%) 15
2001-2500 0 13 0 13 1(5.6%) 12 0 9
2501-3000 0 11 0 11 2(11.1%) 9 1(5.9%) 13
3001-4000 0 3 0 3 0 3 0 10
Total 3 109 7 105 18 94 17 95
Education (mothers) Illiterate 0 8 n.s 0 8 2(11.1%) 6 1(5.9%) 7 n.s
Read & write 1(33.3%) 30 1(14.3%) 30 n.s 5(27.8%) 26 n.s 7(41.2%) 24
Grades 1-5 1(33.3%) 27 2(28.6%) 27 5(27.8%) 23 5(29.4%) 23
Grades 6-8 1(33.3%) 37 4(57.1%) 37 5(27.8%) 33 3(17.6%) 35
Grades 9-12 0 6 0 6 1(5.6%) 5 1(5.9%) 5
Total 3 109 7 105 18 94 17 95
Number of children Two 0 9 0 9 n.s 1(5.6%) 8 4(23.5%) 5
Three 2(66.7%) 16 n.s 0 18 1(5.6%) 17 n.s 4(23.5%) 14 .026**
Four 0 30 4(57.1%) 26 7(38.9%) 23 1(5.9%) 29
Five 0 30 1(14.3%) 29 5(27.8%) 25 5(29.4%) 25
Six 0 12 0 12 2(11.1%) 10 0 12
>six 1(33.3%) 12 2(28.6%) 11 2(11.1%) 11 3(17.6%) 10
Total 3 109 7 105 18 94 17 95
DD <4 food items 0 10 n.s 4(57%). 4 .000* 4(5.6%) 4 .000* 9(23.5%) 5 n.s
>= 4food items 3(100%) 99 3(42.9%) 99 8(94.4%) 96 85(76.5%) 84
Total 3 109 7 103 12 100 17 89

Chi-square significance: * = p < 0.01, ** =p < 0.05; n. s. not significant

SINET:Ethiop. J. Sci., 44(2):172–181, 2021 ISSN: 0379–2897 (PRINT)
© College of Natural and Computational Sciences, Addis Ababa University, 2021 eISSN: 2520–7997

Date received: August 14, 2021; Date revised: September 05, 2021; date accepted: 09 September
DOI: https://dx.doi.org/10.4314/sinet.v44i2.4

Seasonal variation of non-volant small mammals in Gibe Sheleko national park, Southwestern
Ethiopia

Seyoum Kiros 1,* and Afework Bekele2

Addis Ababa, University, Addis Ababa, Ethiopia. E-mail: kirosseyoum@yahoo.com

ABSTRACT: The species composition and abundance of small mammals can vary within space
and time. The main objective of this study was to assess seasonal variation of non-volant small
mammals from randomly selected Acacia woodland, bushland, farmland, grassland, riverine forest
and wooded grassland habitats in Gibe Sheleko National Park, southwestern Ethiopia. Data were
collected using 49 Sharman live traps in 70 x 70 m sized square girds from December, 2018 to
August, 2020. Capture mark recapture technique was applied to estimate population size of the
existing small mammals and the data were analyzed using a chi-square test. A total of 1160
individual small mammals belonging to 10 species and 2 families were recorded. Three non-
captured species: Hystrix cristata, Xerus rutilus and Tachyoryctes splendens were also identified. There
was a significant (χ2= 31.12, df = 1, P < 0.05) difference in the total abundance of small mammals
between seasons. Of the total individuals captured, 675(58.19%) were trapped during the wet
season while 485(41.81%) individuals were during the dry season. Significant seasonal variation
was also observed in the total abundance of both sexes, i e. males (χ2= 11.99, df = 1, P < 0.05) and
females (χ2= 20.24, df = 1, P < 0.05). Among age groups, significant statistical seasonal variation was
shown in adults (χ2=15.14, df = 1, P < 0.05) and young (χ2=44.61, df = 1, P < 0.05) but not significant
in sub-adults (χ2=0.75, df = 1, P >0.05). The identified small mammals exhibited seasonal changes in
their abundance associated with changes in climatic and environmental conditions. However, a
long-term and annual based study is required to see the overall dynamics of existing small
mammals.

Key words/phrases: Abundance, age structure, Gibe Sheleko, seasonal variation, sex distribution,
small mammals

et al., 2015). Changes in species composition or

INTRODUCTION
Small mammals have high reproductive
potential and rapid turn-over rate to invade new
environment and wide range of habitats (Agerie
Addisu and Afework Bekele, 2015; Akpan et al.,
2015; Li et al., 2015; Ofori et al., 2015). The rapid
turn-over rate in small mammals may be
associated with their small body size, short
breeding period and the ability to consume
different food items (Akpan et al., 2015;
Adugnaw Admas and Mesele Yihune, 2016).
Inter and intra-specific small mammals have
different biotic potentials associated with
geographical location, habitat heterogeneity and
productivity, season and size of the female
animals as well (Happold, 2013).
Species composition and abundance of small
mammals vary within space and time. They
exhibit notable seasonal and inter-annual
differences in species composition and
abundance (Sintayehu Workeneh et al., 2011;
Agerie Addisu and Afework Bekele, 2015; Ofori
population size of small mammals in a particular
area is mainly determined by habitat
heterogeneity and productivity, climatic
variation, availability of natural resources,
natural predators, overgrazing, fire and the
extent of invasive exotic species (Sintayehu
Workeneh et al., 2011; Ofori et al., 2015).
Occasionally, deforestation, habitat
fragmentation and other anthropogenic activities
such as agricultural management techniques
have also a significant impact on the diversity
and population dynamics of small animals
(Gentili et al., 2014; Gezahegn Getachew et al.,
2016).
Population dynamics of small mammals is also
affected by the amount and pattern of rainfall in
space and time, as rainfall is the main driving
force of food availability which enhances the
breeding potential of small mammals (Ejigu
Alemayehu and Afework Bekele, 2013; Getachew
Bantihun and Afework Bekele, 2015; Ofori et al.,
2015). In the savanna ecosystem type, for

_____________________
*Author to whom correspondence should be addressed.
SINET: Ethiop. J. Sci., 44(2), 2021
instance, Mastomys natalensis commonly breeds
during the wet and early dry seasons of the year
due to the presence of sufficient food availability
and abundance (Happold, 2013). Overgrazing by
domestic animals may also affect the abundance
and species richens of herbivorous small
mammals of a given area, resulting in food
competition, habitat disturbance and exposure to
natural predators (Addishiwot Fekadu et al.,
2015; Gezahegn Getachew et al., 2016).
Protected areas are established throughout the
globe to conserve wildlife with special emphasis
on large mammals and birds from being extinct.
According to Caro (2001), in East Africa,
protected areas are established primarily to
conserve large mammals especially umbrella or
flagship species considering their tourism
revenue, but the importance and exploration of
small mammals are poorly known. The same is
true in some National Parks of Ethiopia
including Gibe Sheleko National Park.
Exploration on the species composition and
population dynamics of small mammal
communities is mandatory to enhance the
effectiveness of the existing conservation and
management strategies of the Park (Loeb et al.,
1999; Lavrenchenko and Afework Bekele, 2017).
In some developing nations including
Ethiopia, some rodents are considered as a major
challenge of human food security. Seasonal
rodent pest outbreaks can cause human
starvations due to considerable crop damage and
yield losses by rodents. Ecological investigation
on species composition and seasonal variation of
small mammals of a particular area is very
essential to identify economic importance of the
species of that area and to develop effective
rodent pest management strategies
(Lavrenchenko and Afework Bekele, 2017).
Understanding the species composition and
seasonal variation of small mammals from Gibe
Sheleko National Park deserves paramount
importance for the documentation of the existing
small mammals and to strengthen the
conservation efforts of other co-occurring
wildlife species. Therefore, this study was
conducted to investigate the seasonal variations
of small mammals in Gibe Sheleko National
Park.
173
MATERIALS AND METHODS
Study Area
Gibe Sheleko National Park (GSNP) is located in
Gurage zone, Southern Nations, Nationalities
and Peoples’ Region (SNNPR) at 178 and 18 km
away from Addis Ababa and Wolkite town,
respectively. It is geographically located at 8 °5´
00´´N to 8°16´00´´N latitude and 37°26´00´´E to
37°48´00’´E longitude (Fig. 1). The Park is
bordered by three districts of the zone namely
Abeshege, Cheha and Enemorener, and the Gibe
River along the western side bordering the
Oromia Regional State. The Park was established
in 2009 considering its conservation significance
for wildlife with special emphasis for birds. It
covers 360 km2 with variable altitudinal ranges
1050 to 1835 m asl (Alemneh Amare, 2015;
Kassahun Abie et al., 2019).
The Park possesses diverse topographic
features such as flat terrain, plateau, gorges,
relatively undulating and rocky steep slopes
(Hadis Tadele et al., 2020). The area is also
endowed with streams, rivers and hot springs
within and its surrounding. For instance, Wabe
and Nekem rivers flow inside the Park.
The study area is recognized by having
relatively hot weather condition. It receives a
rainfall between 960 and 1400 mm per annum
with average annual precipitation of 1163 mm.
The heaviest rainfall concentration is recorded
during the summer season (June to August) of
the year. Monthly average maximum and
minimum temperature of the area is about 29.1 °C
and 8.9°C, respectively, with average annual
temperature of 19.6°C. The maximum
temperature of the area is recorded during the
dry season (December to February) of the year.
The study area is mainly covered by acrisol,
nitisol and vertisol soil types derived from
Eocene–Palaeocene basaltic rock type (Solomon
Tadesse et al., 2003 as cited in Johansson et al.,
2021). The Park is covered by savannah
woodland vegetation type. Relatively flatlands of
the Park are covered by Acacia species such as
Acacia abyssinica, Acacia etbaica, Acacia nilotica,
Acacia polyacantha and Acacia seyal. Broad leaved
species such as Combretum molle, Combretum
collinum, Cussonia holstii and Ficus sycomorus
cover most of the escarpment area of the Park.
Dicrostachys cinerea, Fluegga virosa and Searsia
174
natalensis are the dominant and widely
distributed shrub species found in the study
area. Among the existing grass species,
Hyparrhenia dregeana, Hyparrhenia filipendula and
Bothriochloa insculpta are the most common and
widely distributed grasses. The surrounding area
of streams and rivers of the Park are mainly
covered by patches of evergreen woodland and
gallery forest vegetation types, respectively
(Johansson et al., 2021).
According to archives of Gibe Sheleko
National Park, the study area is endowed with
different species of vertebrates such as
mammals, reptiles, birds and fishes. Among
mammals, ungulates such as bohor reedbuck
(Redunca redunca), common bushbuck
(Tragelaphus scriptus) bush pig (Potamochoerus
Figure 1. Map of the study area and location of the sampling sites.
METHODS
During this study, six habitat types were
randomly selected based on the dominant
Seyoum Kiros and Afework Bekele
larvatus), hippopotamus (Hippopotamus
amphibious), warthog (Phacochoerus africanus) and
waterbuck (Kobus ellipsiprymnus), carnivores
such as civet (Civettistis civetta), honey badger
(Mellivora capensis), leopard (Panthera pardus) and
spotted hyaena (Crocuta crocuta), primates such
as colobus monkey (Colobus guereza), olive
baboon (Papio Anubis) and vervet monkey
(Chlorocbus pygerythrus) and, others like aardvark
(Orycteropus afer), porcupine (Hystirx cristata),
and rock hyrax (Procavia capensis) occur. Reptiles
such as the African rock python, Nile crocodile
and Nile monitor are also found within the Park.
According to Kassahun Abie et al. (2019) and
Hadis Tadele et al. (2020), GSNP supports more
than 110 bird species including endemic, near
endemic, rare, threatened and migratory species.
vegetation type, altitudinal zonation,
topographical feature and accessibility of the
area. The identified habitat types were: wooded
grassland (1634 m asl), bushland (1570 m asl),
SINET: Ethiop. J. Sci., 44(2), 2021
riverine forest (1490 m asl), farmland (1358 m
asl), Acacia woodland (1300 m asl) and grassland
(1100 m asl). Data collection was carried out for
two successive years having two dry and two
wet seasons from December, 2018 to August,
2020. Dry and wet season data collections were
conducted between December to February and
June to August, respectively. Data were collected
using Sherman live traps, naked eye or
binoculars aid visual surveys and using indirect
evidence such as presence of quills, burrows and
soil mounds for non-captured small mammals.
Capture mark recapture method was applied
to estimate the population size of the existing
small mammals (Afework Bekele and Leirs,
1997). In all habitats, 70 x 70 m sized permanent
square grid was randomly established. In each
gird, a total of 49 Sherman live traps at 10 m
intervals were placed for five consecutive days to
make sure maximum chances of capturing rare
and shy species. Trap stations were marked
using yellow coloured plastic tag. In each
trapping session, all traps were baited with a
mixture of peanut butter, crushed maize and
chickpea. To avoid small mammal mortality due
to harsh temperature during the dry season and
to minimize stealing from natural predators,
traps were covered by the existing grasses or
plant leaves. In each trapping session, traps were
checked and refreshed in the morning (7:00–9:00
a.m) and late afternoon (4:00–6.00 p.m.) per day.
In all trapping sessions, information such as
body weight, sex, approximate age structure and
reproductive condition of each live trapped
small mammal were assessed. Approximate age
structure was determined by considering the
pelage colour, body weight, and developmental
status of external body parts as well as
reproductive organs of the animal (Gezahegn
Getachew et al., 2016). Reproductive condition of
each live trapped small mammal was identified
following Afework Bekele and Leirs (1997),
Monadjem and Perrin (2003), and Tsegaye
Gadisa and Kitessa Hundera (2015). Thus,
sexually matured males are determined by
considering position of their testes either
abdominal or scrotal. Body weight, vaginal
orifice, abdominal and nipple sizes were used to
determine the reproductive condition of female
small mammals. Sex of young small mammals
was determined by considering the distance
between the anus and genital papilla, and the
presence of scrotal septum between the penis
and anus in males (Aplin et al., 2003). Each live
trapped small mammal was toe clipped and then
released into its natural habitat. Scientific names
175
of the live trapped small mammals were
identified using standard literature and
taxonomic keys (Yalden et al., 1976; Kingdon,
1997; Afework Bekele and Yalden, 2013;
Happold, 2013; Happold and Happold, 2013).
Data analyses
Trap success and relative abundance of small
mammals were calculated by the following
formulae:
Nm
TS  x100
Ntn
Where, TS= trap success, Nm = the number of
individuals trapped and Ntn = the number of
trap nights.
n
Ra  x100
Ns
Where, Ra = relative abundance, n = the total
number of captured individuals of a single
species and Ns = the total number of individuals
of the whole species.
Density was calculated as follows:
N
D
A
Where, D = density, N = the total number of
individuals per gird, A= area of the trapping
gird (4900 m2 = 0.49 ha) for each tapping season.
Chi-square test was applied to compare
seasonal variations in small mammal species
abundance, sex distribution and age structure.
All statistical data were analyzed using SPSS
version 24 computer software.
RESULTS
Seasonal variation
A total of 1160 live trapped small mammals
belonging to 10 species were recorded from the
six habitat types in Gibe Sheleko National Park.
Hystrix cristata, Xerus rutilus and Tachyoryctes
splendens were also documented through direct
observation and using indirect evidences. Of
these, 675(58.19%) individuals were trapped
during the wet season, while 485(41.81%) were
trapped during the dry season. There was a
significant (χ2= 31.12, df = 1, P < 0.05) difference
in the abundance of small mammals between
seasons. Of the total 2940 trap nights, overall trap
success was higher during the wet season
(22.96%) than the dry season (16.5%). Except
population of Mus tenellus and Rattus rattus,
more than 50% of the total populations of other
species were recorded during the wet season of
176 Seyoum Kiros and Afework Bekele

the study period (Fig. 2). However, significant
seasonal variation was observed only in the
abundance of three species; Arvicanthis niloticus
(χ2= 14.10, df = 1, P<0.05), Lemniscomys striatus
(χ2= 26.04, df = 1, P<0.05) and Stenocephalemys
albipes (χ2= 12. 52, df = 1, P <0.05).
Sex distribution and seasonal variation
In the present study, a total of 646(55.69%)
male and 514(44.31%) female small mammals
were recorded. There was a significant (χ2=
15.02, df =1, P < 0.05) difference in the
abundance of small mammals between sexes.
Males had higher (10.99%) capture rate than
females (8.74%) within 5880 trap nights. In
Grammomys dolichurus, nearly equal numbers of
male and female individuals were recorded. In
other species, the total captured males were
significantly higher than females. However,
except in Mastomys natalensis (χ2= 5.65, df =1, P <
0.05), no statistical significant difference was
observed in the abundance of sexes (Fig. 3).

250
Abundance

200 Dry
150 Wet
100
50
0

Species

Figure 2. Seasonal variation in the abundance of small mammal species.

250
200 Male
Abundance

Female
150
100
50
0

Species

Figure 3. Abundance of male and female small mammals.

SINET: Ethiop. J. Sci., 44(2), 2021 177

Seasonally, the overall abundance of male and

female small mammals varied. Out of the total Age structure and seasonal variation
646 male and 514 female small mammals, During the present study, a total of
367(56.81%) males and 308(59.92%) females were
785(67.67%), 262(22.59%) and 113(9.74%) adult,
recorded during the wet season with capture rate
of 12.48% and 10.48%, respectively in 2940 trap sub–adult and young mammals, respectively
nights. While the remaining 279(43.19%) males were recorded. A statistically significant (χ2=
and 206 (40.08%) females were registered during 644.24, df = 2, P < 0.05) difference was observed
the dry season of the study period with 9.49% in the total abundance of small mammals among
and 7.01% capturing rates, respectively (Table 1).
the three age groups. Overall, adults had high
Seasonal variation was observed in the total
abundance of both sexes, i e. males (χ2= 11.99, df (13.35%) capture rates, followed by sub-adults
= 1, P < 0.05) and females (χ2= 20.24, df = 1, P < (4.46%) and young (1.92%) in 5880 trap nights.
0.05). In some species, the ratio of male to female Seasonally, higher, 447(56.94%), 138(52.67%) and
showed some fluctuations within seasons. In 92(81.42%) adult, sub–adult and young small
Crocidura olivieri, 1:1 male to female ratio was
mammals, respectively were recorded during the
recorded during the wet season. In most species,
the numbers of females were higher during the wet season (Fig. 4). Statistical seasonal variation
wet season (Table 1). was shown in adults (χ2=15.14, df = 1, P < 0.05)
and young ones (χ2=44.61, df = 1, P < 0.05), but
Table 1. Seasonal variation and sex distribution of
not in sub–adults (χ2=0.75, df = 1, P >0.05).
small mammals.
During the wet season, the number of trapped
Dry season Wet season young ones increased by more than four times
Species Mal Femal Mal Femal Male:
e e e e Female
compared to the dry season (Fig. 4).
ratio
Arvicanthis 15 9 31 27 46:36
niloticus
Seasonal variation in density and trap success
Crocidura 5 4 7 7 12:11 The overall small mammal density and trap
olivieri
Grammomys 10 8 12 15 22:23 success of the study area was 98.64/ha and
dolichurus 19.73%, respectively. Small mammal density and
Lemniscomys 16 13 42 41 58:54
striatus trap success also varied among habitat types and
Mastomys 69 61 85 66 154:127 between seasons. Highest mean small mammal
awashensis
Mastomys 102 78 117 94 219:172 density was recorded in wooded grassland
natalensis (152.04/ha) followed by farmland (147.96/ha)
Mus tenellus 3 2 2 1 5:3
Myomys 24 11 22 19 46:30 during the wet season. While the lowest mean
fumatus
small mammal density was 50/ha recorded in
Rattus rattus 19 9 14 13 33:22
Stenocephalem 16 11 35 25 51:36 Acacia woodland during the dry season (Table 2).
ys albipes
Total 279 206 367 308 646:514
Overall, the highest (30.41%) and lowest (10%)
Relative 24.0 17.76 31.6 26.55 55.69:44.3 trap success was recorded in wooded grassland
abundance 5 4 1
(%)
and Acacia woodland habitats during the wet
Trap success 9.49 7.01 12.4 10.48 10.99:8.74 and dry seasons, respectively.
(%) 8
178 Seyoum Kiros and Afework Bekele

500
447
450
Dry
400
338 Wet
350
Abundance

300
250
200
124 138
150
92
100
50 21
0
Adult Sub-adult Young
Age group

Figure 4. Seasonal variation in age groups of small mammals.

Table 2. Seasonal variation in small mammal density and trap success in six habitats.

Season Total Density Trap Trap

Habitat captured ( ha-1) Nights success (%)
Acacia woodland Dry 49 50 490 10
Wet 86 87.76 490 17.55
Bushland Dry 75 76.53 490 15.31
Wet 89 90.82 490 18.16
Farmland Dry 134 136.74 490 27.35
Wet 145 147.96 490 29.59
Grassland Dry 76 77.55 490 15.51
Wet 107 109.18 490 21.84
Riverine forest Dry 61 62.25 490 12.45
Wet 99 101.02 490 20.20
Wooded grassland Dry 90 91.84 490 18.37
Wet 149 152.04 490 30.41

DISCUSSION Srinivasulu (2019). Capturing more individuals

Having a scientific knowledge about the seasonal
change of small mammals is very essential to
determine the small mammals’ life history,
patterns of population fluctuation and density
occurrences as well as the population regulatory
factors involved (Massawe et al., 2011; Fischer et
al., 2012; Ofori et al., 2015). It also helps to predict
the rodent pest outbreak season and to apply
appropriate rodent pest management strategies
in a particular area (Massawe et al., 2011).
In this study, higher number of individual
small mammals was recorded during the wet
than the dry seasons. This result goes in line with
Tilahun Chekol et al. (2012), Redwan Mohammed
et al. (2017), and Alembrhan Assefa and
during the wet season is associated with the
availability and quality of food and shelter. The
presence of diverse herbaceous plants during the
wet season may serve as a source of food and
shelter for non–volant small mammals (Li et al.,
2015). Microhabitat features of a given area such
as grass height and density, tree/shrub type and
density and other features may also determine
the species richness and abundance of small
mammals during the dry season (Delcros et al.,
2015).
In contrast, Demeke Datiko et al. (2007),
Tadesse Habtamu and Afework Bekele (2008),
Sintayehu Workeneh et al. (2011), Demeke Datiko
and Afework Bekele (2014), Delcros et al. (2015)
and Gezahegn Getachew et al. (2016) reported
SINET: Ethiop. J. Sci., 44(2), 2021
more number of individual small mammals were
documented during the dry season. This
variation may be related with the differences in
the degree of human induced disturbances,
overgrazing, fire, diversity and abundance of
natural predators and timing of the study period
(Tilahun Chekol et al., 2012; Alemrhan Assefa
and Srinivasulu, 2019).
According to Mulatu Osie et al. (2010) and
Kiros Welegerima et al. (2020), high individual
small mammals were recorded at the early dry
season due to the presence of sufficient food and
shelter following the rainfall patterns. However,
as the dry season progressed, the availability and
abundance of natural resources such as food and
water become scarce and hence, species richness
and abundance of small mammals decline as a
result of starvation and increasing mortality rate
(Happold and Happold, 1990).
In the study area, seasonal human induced
fire, deforestation, overgrazing and presence of
natural predators particularly snakes may affect
the species richness and abundance of small
mammals. According to Mulatu Osie et al. (2010),
due to the absence and/or reduction of plant
ground cover during the dry season, small
mammals may be exposed to natural predators
and food starvation. Burned areas during the dry
season may support few numbers of small
mammal species and individuals (Happold and
Happold, 1990).
In the present study, sex ratio was male biased
during both seasons. This is in line with Afework
Bekele (1996), Getachew Bantihun and Afework
Bekele (2015), Gezahegn Getachew et al. (2016)
and Shilereyo et al. (2020). Capturing of more
males may be related to larger home range
utilization behaviour of males to search for food
or receptive females. Lower number of trapped
females may also be associated with
reproductive costs of females such as in parental
care, pregnancy and lactation periods (Shilereyo
et al., 2020). In contrast, Tadesse Habtamu and
Afework Bekele (2008) and Alembrhan Assefa
and Srinivasulu (2019) reported that the number
of captured females was higher than the males,
this may be associated with the variations in
ecological conditions of the study areas such as
variations in natural resources. According to the
Trivers-Willard hypothesis, maternal diet status
of female small mammals plays an important
role in sex ratio bias, that is, females inhabited
nutritionally deficient areas such as caloric
content can produce more female offspring and
then, female skewed sex ratio may occur
179
(Rosenfeld and Roberts, 2004; Shilereyo et al.,
2020).
In this study, during both seasons, the
abundance of adult small mammals was higher
than the number of sub–adult and young small
mammals. This is in agreement with the studies
of Redwan Mohammed et al. (2017) and
Alembrhan Assefa and Srinivasulu (2019), where
capturing of more adult small mammals may be
associated with their body size making it suitable
for trappings or adults may have higher smelling
capacity to human induced food items such as
mixed peanut butter bait, as well as utilization of
relatively larger home ranges.
In this study, the abundance of young
increased during the wet season of the study
period. Reproductive potential of females may
be enhanced with the availability and quality of
food items. According to Afework Bekele and
Leirs (1997), Demeke Datiko et al. (2007), Tadesse
Habtamu and Afework Bekele (2008), Sintayehu
Workeneh et al. (2011), Tilahun Chekol et al.
(2012), Getachew Bantihun and Afework Bekele
(2015), Gezahegn Getachew et al.(2016), Redwan
Mohammed et al. (2017), Alemrhan Assefa and
Srinivasulu (2019), and Shilereyo et al. (2020),
increasing in the abundance of lactating,
pregnant and young of most small mammals
during the wet season could be assisted by the
availability of sufficient and nutritious food
items and sufficient vegetation cover. Capturing
low abundance of young during the dry season
may be due to their vulnerability towards
natural predators and harsh weather condition
as a result of reduction in ground cover (Redwan
Mohammed et al., 2017).
Overall, trap success was higher during the
wet season than the dry season. Among the six
habitat types, the highest and lowest trap success
and small mammal density were recorded in
wooded grassland and Acacia woodland habitats
during the wet and dry seasons, respectively.
According to Afework Bekele and Leirs (1997),
Mulatu Osie et al. (2010), Tilahun Chekol et al.
(2012), Dawd Yimer and Solomon Yirga, (2013),
and Redwan Mohammed et al. (2017), trap
success and density in a given area and time may
be attributed to several factors including habitat
heterogeneity and productivity, availability of
food and shelter, reproductive pattern, diversity
and abundance of natural predators and degree
of anthropogenic factors.
The study provided baseline information on
the small mammal species composition and their
temporal variations in Gibe Sheleko National
Park. From the present study, it is clear that the
180
small mammal species composition and
abundance varied spatially and temporally.
Among the identified species, Arvicanthis
niloticus, Lemniscomys striatus, Mastomys
awashensis, Mastomys natalensis, Rattus rattus
along with Hystrix cristata and Tachyoryctes
splendens were listed as the crop pests of the
study area. Hence, understanding the dynamics
of these species has a significant impact for
rodent pest management planning and
implementation. However, frequent ecological
assessments are needed to see the overall
seasonal variations of small mammals, as well as
the causal factors involved.
ACKNOWLEDGEMENTS
The authors greatly acknowledge Addis Ababa and
Wolkite Universities for their logistic and financial
support. Moreover, the authors wish to thank all staff
particularly field guides of Gibe Sheleko National Park
for their kind support.
REFERENCES
1. Addishiwot Fekadu, Afework Bekele and Demeke
Datiko (2015). Comparative study of species
composition, relative abundance and
distribution of rodents between exclosure
and control sites in the Web Valley of the
Bale Mountains National Park, Ethiopia.
Punjab Uni. J. Zool. 30:57–64.
2. Adugnaw Admas and Mesele Yihune (2016).
Species composition, relative abundance
and habitat association of rodents in
Yekoche Forest, East Gojjam, Ethiopia.
Int. J. Biodi. Cons. 8:216–223.
3. Afework Bekele (1996). Population dynamics of the
Ethiopian endemic rodent, Praomys albipes in
the Menagesha State Forest. J. Zool. Lond.
238:1–12.
4. Afework Bekele and Leirs, H. (1997). Population
ecology of rodents of maize fields and
grassland in central Ethiopia. Belg. J.
Zool. 127:39–48.
5. Afework Bekele and Yalden, D.W. (2013). The
Mammals of Ethiopia and Eritrea. Addis
Ababa University Press, Addis
Ababa.
6. Agerie Addisu and Afework Bekele (2015).
Population structure of rodents in Alage,
southern Ethiopia. J. Ecol. Nat. Envi.
7:7–13.
7. Akpan, A. U., Esenowo, I. K., Egwali, E. C. and
James, S. (2015). The checklist and abundance
of small mammals in Idu, Akwa Ibom State,
Nigeria. J. Appl. Sci. Envi. Mngt. 19:71–75.
Seyoum Kiros and Afework Bekele
8. Alembrhan Assefa and Srinivasulu, C. (2019).
Species composition and abundance of
rodents in Kafta Sheraro National Park,
Ethiopia: preliminary results. J. Threat. Taxa
11:13680–13689.
9. Alemneh Amare (2015). Conservation challenges of
Gibe Sheleko National Park, Southwestern
Ethiopia. Nat. Res. 6:286–289.
10. Aplin, K. P., Brown, P. R., Jacob, J., Krebs, C. J. and
Singleton, G. (2003). Field methods for
rodent studies in Asia and the Indo-Pacific.
ACIAR Monograph No. 100, Pp 223.
11. Caro, T. M. (2001). Species richness and abundance
of small mammals inside and outside
an African National Park. Biol. Cons. 98: 251–
257.
12. Dawd Yimer and Solomon Yirga (2013). Mammals
of the Mazie National Park, Southern
Nations, Nationalities and Peoples Regional
State, Ethiopia. SINET: Ethiop. J. Sci. 36:55–61.
13. Delcros, G., Taylor, P. J. and Schoeman, M. C.
(2015). Ecological correlates of small mammal
assemblage structure at different spatial
scales in the savannah biome of South Africa.
Mammalia 79:1–14.
14. Demeke Datiko and Afework Bekele (2014).
Habitat association and distribution of
rodents and insectivores in Chebera
Churchura National Park, Ethiopia. Trop.
Ecol. 55:221–229.
15. Demeke Datiko, Afework Bekele and Gurja Belay
(2007). Species composition, distribution and
habitat association of rodents from
Arbaminch forest and farmlands, Ethiopia.
Afr. J. Ecol. 45:651–65.
16. Ejigu Alemayehu and Afework Bekele (2013).
Species composition, distribution and habitat
association of rodents in bir farm
development and nearby natural habitat area,
Ethiopia. Int. J. Ecol. Envi. Sci. 39:195–202.
17. Fischer, C. M. G., Baldi, G., Codesido, M. and
Bilenca, D. (2012). Seasonal variations in
small mammal-landscape associations in
temperate agroecosystems: a study case in
Buenos Aires province, central Argentina.
Mammalia 76: 399–406.
18. Gentili, S., Sigura, M. and Bonesi, L. (2014).
Decreased small mammal species diversity
and increased population abundance along a
gradient of agricultural intensification.
Hystrix. Ital. J. Mammal. 25:39–44.
19. Getachew Bantihun and Afework Bekele (2015).
Diversity and habitat association of small
mammals in Aridtsy forest, Awi Zone,
Ethiopia. Zool. Res. 36: 88–94.
20. Gezahegn Getachew, Balakrishnan, M. and
Afework Bekele (2016). Species composition
habitat association of rodents in Yetere Forest
Central Ethiopia. Int. J. E. Sci. 42:193–200.
SINET: Ethiop. J. Sci., 44(2), 2021
21. Hadis Tadele, Afework Bekele, Solomon Wagaw
and Simeneh Admasu (2020). Assessment of
avifaunal assemblage and their distribution
pattern across different habitat types of Gibe
Sheleko National Park, South western
Ethiopia. Int. J. Biodi. Cons. 12:59–70.
22. Happold, D. C. D. and Happold, M. (1990). An
ecological study of small rodents in the
woodland savanna of Liwonde National
Park, Malawi. J. Zool., Lond. 221:219–235.
23. Happold, D. C. D. (2013). Mammals of Africa.
Volume III: Rodents, Hares and Rabbits.
Bloomsbury, London.
24. Happold, M. and Happold, D. C. D. (2013).
Mammals of Africa. Volume IV: Hedgehogs,
Shrews and Bats. Bloomsbury, London.
25. Johansson, M. U., Abebe, F. B., Nemomissa, S.,
Bekele, T. and Hylander, K. (2021). Ecosystem
restoration in fire managed savanna
woodlands: Effects on biodiversity, local
livelihoods and fire intensity. Ambio 50 :190–
202.
26. Kassahun Abie, Belete Tilahun, Abel Feyisa,
Tewodros Kumssa and Alemneh Amare
(2019). Bird species diversity and distribution
in case of protected area. Species 20:90–100.
27. Kingdon, J. (1997). The Kingdon Field Guide to
African Mammals. Academic Press, London.
28. Kiros Welegerima, Meheretu Yonas, Tsegazeabe H.
Haileselassie, Brhane Gebre, Dawit Kidane,
Massawe, A. W., Mbije, N. E. and Makundi,
R. H. (2020). Abundance and microhabitat
use of rodent species in crop fields and
bushland in Ethiopia. J. Vert. Biol. 69: 20054.
29. Lavrenchenko, L. A. and Afework Bekele (2017).
Diversity and conservation of Ethiopian
mammals: What have we learned in 30 years?
Ethio. J. Biol.Sci.16:1–20.
30. Li, B., Ran, J., Yue, B., Zhang, M. and Wu, Y. (2015).
Non-volant small mammals in landslides
caused by the Wenchuan Earthquake in a
fragmented forest of Sichuan, China. Pakis. J.
Zool. 47:535–544.
31. Loeb, S. C., Chapman, G. L. and Ridley, T. R.
(1999). Sampling small mammals in
Southeastern Forests: The importance of
trapping in trees. Proc. Annu. Conf. Southeast.
Assoc. Fish. Wildl. Agen. 53:415–424.
32. Massawe, A. W., Mulungu, L. S., Makundi, R. H.,
Dlamini, N., Eiseb, S. J., Kirsten, F., Mahlaba,
T., Malebane, P., Von Maltitz, E., Monadjem,
A., Taylor , P., Tutjavi, V. and Belmain, S. R.
(2011). Spatial and temporal population
dynamics of rodents in three geographically
different regions in Africa: Implication for
ecologically-based rodent management. Afri.
Zool. 46: 393–405.
181
33. Mesele Yihune and Afework Bekele (2012).
Distribution, diversity and abundance of
rodent community in Afro-alpine habitats of
the Seimien Mountains National Park,
Ethiopia. Int. J. Zool. Res. 8:137–149.
34. Monadjem, A. and Perrin, M. (2003). Population
fluctuations and community structure of
small mammals in a Swaziland grassland
over a three year period. Afr. Zool. 38:127–
137.
35. Mulatu Osie, Afework Bekele and Balakrishnan, M.
(2010). An ecological study on rodents of
natural vegetation and farm lands in Siltie,
Central Ethiopia. SINET: Ethiop. J. Sci. 33:59–
66.
36. Ofori, B.Y., Attuquayefio, D. K., Owusu, E. H.,
Musah, R. K., Quartey, J. K. and Ntiamoa-
Baidu, Y. (2015). Seasonal changes in small
mammal assemblage in Kogyae Strict Nature
Reserve, Ghana. Int. J. Biodi. Cons.7:238–244.
37. Redwan Mohammed, Afework Bekele and
Balakrishnan, M. (2017). Species composition
and pest status of rodents in Tendaho
Sugarcane Plantation, Afar Region, Ethiopia.
Mammal Study 42:31–38.
38. Rosenfeld, C. S. and Roberts, R. M. (2004). Maternal
diet and other factors affecting offspring sex
ratio: a review. Biol. Repro.71:1063–1070.
39. Shilereyo, M., Magige, F. J., Ogutu, J. O. and
Røskaft, E. (2020). Small mammal
community demography and reproductive
seasonality under protection, pastoralism and
agriculture in the Serengeti Ecosystem,
Tanzania. Int. Biodi. Cons. 12:253–269.
40. Sintayehu Workeneh, Afework Bekele and
Balakrishnan, M. (2011). Species diversity
and abundance of small mammals in
Nechisar National Park, Ethiopia. Afr. J. Ecol.
50:102–108.
41. Tadesse Habtamu and Afework Bekele (2008).
Habitat association of insectivores and
rodents of Alatish National Park,
northwestern Ethiopia. Trop. Ecol. 49:1–11.
42. Tilahun Chekol, Afework Bekele and Balakrishnan,
M. (2012). Population density, biomass and
habitat association of rodents and
insectivores in Pawe area, northwestern
Ethiopia. Trop. Ecol. 53:15–24.
43. Tsegaye Gadisa and Kitessa Hundera (2015). The
reproductive and feeding ecology of rodents
in Sekoru district, Southwest Ethiopia. J. Ecol.
Nat.Envi.7:188–195.
44. Yalden, D. W., Largen, M. J. and Kock, D. (1976).
Catalogue of the mammals of Ethiopia. Mon.
Zool. Italian. (Suppl) 8:1–118.
SINET:Ethiop. J. Sci., 44(2): 182–192, 2021 ISSN: 0379–2897 (PRINT)
© College of Natural and Computational Sciences, Addis Ababa University, 2021 eISSN: 2520–7997

Date recived: May 06, 2021; Date revised: June 22, 2021; Date accepted: June 23, 2021
DOI: https://dx.doi.org/10.4314/sinet.v44i2.5

Diurnal activity patterns, habitat use and foraging habits of Egyptian goose (Alopochena
egyptiacus Linnaeus, 1766) in the Boyo wetland, southern Ethiopia

Mulugeta Kassa1, M. Balakrishnan2 and Bezawork Afework1*

1
Addis Ababa University, College of Natural and Computational Sciences, P.O. Box 1176, Addis Ababa, Ethiopia.
E-mail: bezawork.afework@aau.edu.et
2
University of Kerala, Department of Zoology, India

ABSTRACT: Egyptian goose (Alopochena egyptiacus) is a resident bird species in Africa South of the
Sahara occurring throughout the entire Nile Valley. Despite the wide distribution, the available
information on its behavioral ecology is limited in Ethiopia. A study on the activity patterns, habitat use
and foraging habits of Egyptian goose was carried out in and around Boyo wetland, Ethiopia, during
the dry and wet seasons. Scan sampling method was used to study the activity patterns and habitat use
of Egyptian goose in grassland, mudflat and shallow water habitats of the wetland. The feeding
behavior of Egyptian goose was also observed in the surrounding farmland habitats using scan
sampling method. Generally, Egyptian geese spent most of their time resting (39.81%) followed by
foraging (32.64%). They spent 10.43% of their time in comfort movement preening or stretching. The
rest of their time was allocated for locomotion (6.63%), vigilance (5.75%), and social behavior (1.59%),
and other activities (2.86%). Most of the birds were engaged in foraging activity in the morning (07:00-
9:00 h) and afternoon (16:00 - 18:00 h) hours both during the wet and dry seasons. About 39% of
Egyptian geese were scanned in mudflat, 31.5% in grassland, and 30.05% in shallow water habitats
engaged in different activities. Most individuals used the grassland habitat for foraging during the dry
(59.5%) and wet (74%) seasons, while they used shallow water and mudflat habitats for resting both
during the wet and dry seasons seasons. The birds were observed foraging mainly grass during the
dry (93.62%) and wet (59.52%) seasons. The Egyptian geese show diurnal activity pattern with feeding
peaks in early morning and late afternoon hours as is observed in many other avian taxa. The Boyo
wetland is also as an important foraging ground for this species and other birds in the area. Further
ecological studies on the species and impact of human activities on the Boyo wetland should be
conducted for the conservation of the avifauna.

Keywords/phrases: Activity patterns, Boyo wetland, Egyptian geese, resident birds, scan sampling

tail feathers and rump are black, while the

INTRODUCTION
Egyptian goose Alopochena egyptiaca is a large (61-
75cm long) and distinctively plumaged bird in the
family Anatidae. These birds are monomorphic
but females are slighty smaller in size than males.
Egyptian goose is classified as “Least Concern”,
under the IUCN Red List criteria (BirdLife
International, 2018). This species is one of the birds
that had been raised in Egypt by the ancient
Egyptians for for its beautiful plumage used for
decoration in palaces and temples (Makram, 2018).
The adult plumage is predominantly grayish on
the head, neck, breast, under parts, flanks, and
back, with darker, chocolate brown tones around
eyes, nape, upper wing coverts, and with an
irregular blotch on the lower breast. The primaries,
secondaries are iridescent green and the upper
wing coverts are white except for a narrow black
bar extending across the front of the greater
secondary coverts (Johnsgard, 1978). Bill, legs, and
feet are pink (Mackworth-Praed and Grant, 1980;
Redman et al., 2009). The youngs on the other hand
are dull with a gray shade on their forewings.
Their crown and neck are darker and they have
yellowish legs and beak.
Egyptian goose is one of the non-native
waterfowl species in its native range in Africa,
particularly South of the sub-Sahara with a
population greater than 500,000 individuals
(Davies, 2005; Banks et al., 2008). In addition to its
native populations, there are successfully
established populations in Europe and are
considered one of the most rapidly spreading
SINET: Ethiop. J. Sci., 44(2), 2021
invasive species (Gyimesi and Lensink, 2012). In
North America, they occur in Florida, Texas and
California, among other regions (Callaghan and
Brooks, 2017). Egyptian goose show no regular
migration, but make irregular movements up to
1000 km in response to changes in water and
availability of food and presence of breeding
ground (Oatley and Prŷs-Jones, 1985).
Egyptian geese are primarily herbivores and
infrequently insectivores (Maclean,1988).
However, there are reports where Egyptian geese
cause damage to crops and agricultural lands,
resulting in extensive economic losses (Mangnall
and Crowe, 2002; Atkins, 2015) and their rapid
population growth has led to an increase in the
number of conflicts with people (Mangnall and
Crowe, 2002; Stephen, 2008). However, limited
information is available on the ecology and its
diurnal activity patterns (Callaghan and Brooks,
2016)
Bird’s activity study is significant in
understanding its life history, physical condition,
food availability, social structure, environmental
condition as well as ecological conditions (Asokans
et al., 2010; Aissaoui et al., 2011). Daily activity is
influenced by an individual’s need and its
interactions with organisms, both conspecific and
with other species, environmental factors, such as
ambient temperature, humidity, illumination and
precipitation and ecological factors such as group
size, habitat, food availability, and predation
(Lillywhite and Brischoux, 2012). In addition,
knowledge on habitat use and foraging ecology of
wetland birds has become fundamental in
providing an understanding of the ways in which
species in a habitat partition their resources
(Schulze et al., 2000).
Egyptian goose is mostly dependent on
wetland ecosystem (Brinson and Malvárez, 2002;
Mitsch and Hernandez, 2013). In Ethiopia,
Egyptian goose occurs across large areas; in or
near by open country wetlands, meadows and
grasslands of which Boyo wetland is one. Boyo
wetland is one of the Important Bird Areas in
Southern Ethiopia that supports different water
birds including the globally threatened species, the
wattled crane. Other bird species occurring in the
this wetland includes black crown crane (Balearica
pavonina) , squacco heron (Ardeola ralloides), cattle
egret (Bubulcus ibis), yellow-billed egret (Ardea
brachyrhyncha), yellow billed stork (Mycteria ibis),
glossy ibis (Plegadis falcinellus), sacred ibis
183
(Threskiornis aethiopicus), spur winged goose
(Plectropterus gambensis), and others (EWNHS, 1996).
Despite the different bird species it supports, the
wetland is threatened by expansion of range and
farmlands that cleared the tree and grass cover
causing major siltation and a decreas in the
wetland water depth. Settlement and introduction
of exotic tree species has further aggravated the
wetland deterioration. High disturbance of birds
by human and livestock is also reported (EWNHS,
1996).
Egyptian goose in its native range has a
population greater than 500,000 individuals and
increasing rapidly in numbers; i.e., more than 10%
increase per year on average. In addition,
populations in Europe are estimated at around
10.000 breeding pairs making them one of the most
rapidly spreading invasive species (Banks et al.,
2008). Their rapid population growth has led to an
increase in the number of conflicts with people and
human related activities, particularly within urban
and sub-urban landscapes (Mangnall and Crowe,
2002; Stephen, 2008)
Application and implementation of any
conservation measure towards a species or the
habitat it depends on a scientific study of the
ecological requirements of the target species and
its habitat (Dugan, 1990). Egyptian geese are the
dominant species in the study area being present
throughout the year in large numbers. Hence, this
study was aimed to investigating the activity
patterns, habitat use and feeding habits of
Egyptian goose in Boyo wetland, Southern
Ethiopia.
METHODS
Description of the study area
Boyo wetland is located in Southern Nations
Nationalities and Peoples Regional (SNNPR) State
of Ethiopia about 300 km away from Addis Ababa,
the capital city of Ethiopia. It is located in the
Central Rift Valley area of Ethiopia between
07º28'-07º32'N Latitudes and 38º00’-38º40'E
Longitudes (Fig. 1) (EWNHS, 1996). The Central Rift
Valley area of Ethiopia consist of a chain of lakes,
streams and wetlands with unique hydrological
and ecological characteristics. The landscapes and
ecosystems comprise extensive biodiversity-rich
wetlands, which support a wide variety of
endemic birds and other wildlife (Jansen et al.,
184
2007). Altitudinal ranges of the wetland vary from
1850 to 1900 m a.s.l. It is a swampy fresh water
wetland . Climate in Boyo wetland is generally
characterized by warm, wet summers (from June
to September) and dry, cold and windy winter
(December to March). The mean monthly rainfall
Figure 1. Map of Boyo wetland and surrounding Kebeles.
Data collection
Ecological studies on the diurnal activity
patterns, habitat use and foraging behavior of
Egyptian goose in Boyo wetland was carried out
during the dry (December 2017 to May 2018) and
wet (June to November 2018) seasons. Data were
Mulugeta Kassa et al.
of the area ranges from 9 mm in December to 195
mm in July. Maximum temperature of the area
reaches 27.4°C during the warmest month of April
and minimum in the coldest month of December
(13.6 ºC). As a result, the area is characterized as a
semi-arid climate (EWNHS, 1996).
collected for a total of 33 days, 15 days during the
wet season and 18 days during the dry season.
Activity pattern and habitat use data were
collected for 21 days and foraging habits for 12
days during the wet and dry seasons.Observation
was made under good weather conditions with the
aid of binoculars and/or naked eyes.
SINET: Ethiop. J. Sci., 44(2), 2021
Activity patterns
The diurnal activity patterns of Egyptian goose
were collected during both wet and dry seasons.
When flocks, pairs or individuals of Egyptian
geese were located, instantaneous scan sampling
was carried out (Altmann, 1974; Sutherland, 2004)
to collect the daily activity time budget of the
species. This method provides an overall estimate
of proportions of an individual’s engagement in
different behaviors (Webb et al., 2011) and
categorizing its major activities. Individuals were
scanned or observed for five minutes, during
which instantaneous behavioral observations were
recorded at 15minute intervals (Döpfner et al.,
2009; Chudzinska et al., 2013). The observations
were made from early morning to late evening
dividing the day into three time slots; morning
(7:00-9:00 hrs), mid-day (11:00-13:00 hrs), and
afternoon (16:00-18:00 hrs). Seven major
behavioral activities of the Egyptian geese were
distinguished: foraging, resting, comfort
movement, locomotion, vigilance, social behavior
and other activities. Foraging behavior refers to
searching for food while walking with lowered
head or head down, picking food items and
ingesting (Jónsson and Afton, 2006; Döpfneret al.,
2009; Webb et al., 2011). Resting behavior refers to
a goose pausing, sleeping or loafing. Locomotion
includes flying, swimming or walking while
raising the head, running, flight and flapping.
Comfort movements refer to cleaning or preening
as well as muscle stretching. Vigilance or alert
behavior refers to scanning or observing the
surroundings area by raising its head upward or
neck extended (Döpfner et al., 2009). Social
behavior refers to behaviors of aggression such as
chasing, pecking or biting and courtship.
Vocalizing, bathing and drinking were grouped as
other activities (Webb et al., 2011).Behavioral
activities were recorded whenever more than one
behavioral state occurred at the same time, the
more frequent one performed by the majority of
the individuals was taken (Edroma and Jumbe,
1983).
Habitat use
Habitat utilization of Egyptian geese was
recorded in the grassland, mudflat and shallow
water habitats of Boyo wetland using scan
sampling method during which the activities of the
birds within five minutes were recorded for each
185
habitat. During the scan sampling, the number of
birds in each habitat types and the activities they
were engaged were recorded to investigate how
the species utilize its surroundings to increase the
likelihood of its odds of survival.
Foraging habit
To collect data about the diet composition of
Egyptian geese, repeated observations were made.
Time spent on foraging was recorded using focal
sampling methods following Sutherland et al.
(2005). Individual bird was followed from a
distance of 5-10 m. All observations on diet studies
were carried out between the hours of 06:30-11:30
hrs in the morning, and 14:30-18:00 hrs in the
afternoon. These time periods were selected as it is
found to be active foraging times of the species
(Bibby et al., 1992). When the bird was located it
was first observed for 10 seconds without
recording any data. This time period minimized
the likelihood of recording only the conspicuous
behavior, and also ensured that the bird resumed
normal activity in the presence of the observer.
Observation began as soon as the focal bird began
foraging. When the focal bird stopped foraging or
lost from sight, another individual bird within the
flock was selected as the focal bird in order to
complete the observation period (De Melo and
Guiherme, 2016). To avoid re-sampling the same
bird, the observer moved 150 m from the location
before sampling of the next bird began (Munoz
and Colorado, 2012). During the observation, the
type of food items consumed in each minute was
recorded.
Data analysis
Data were analyzed by using SPSS version 20
computer software programme (SPSS inc, IL, USA)
and Microsoft Excel. Chi square test was used to
compare the different behavioral activities in the
three time blocks. Mann Whitney U-test was used
to compare data for wet and dry season activities
and foraged food items. Diurnal activities among
the different habitat types and seasons were
compared using two-way ANOVA.
RESULTS
Diurnal activity patterns
Egyptian geese were occurred in grassland
(31.5%), mudflat (38.45%) and shallow water
186
(30.05%) habitats. Overall, majority of the
individuals spent their time on resting (39.81%),
followed by foraging (32.64%), comfort movement
(10.63%), locomotion (6.65%), vigilance (5.78%),
social behavior (1.59%), and other activities
(2.89%).
During the dry season, Egyptian geese spent
most of its time foraging in the afternoon hours
(16:00 - 18:00 h) followed by morning (07:00-9:00
h) and mid-day hours (11:00-13:00). Resting in
Egyptian gees was the highest during mid-day
hours (11:00-13:00 h), and they were also seen
resting in the morning (07:00-9:00 h) (Fig.2a).
Time spent for foraging, comfort movement,
vigilance, locomotion activity, social and other
activities showed no significant statistically
difference with the time blocks of the day.
However, time spent for resting showed a
statistically significant difference with the time
blocks of the day (𝑥2=10.738, df=2, P=0.005).
During the wet season, foraging activity peaked
in the late afternoon (16:00 -18:00 h), followed by
morning (07:00-9:00 h), and mid-day hours (11:00 -
13:00 h). Resting was higher during the mid-day
(11:00 -13:00 h) and morning hours (07:00-9:00 h)
(Fig. 2b).
Time spent for foraging (𝑥2= 6.049, df=2, P= 0.049),
resting (𝑥2=14.102, df =2, P= 0.001) and other
behavioral activities (𝑥2= 9.556, df=2, P= 0.008)
showed a significant difference within the time
blocks of the day.
A comparison of the seven different diurnal
activities between dry and wet seasons at different
time blocks showed no statistically significant
difference (p > 0.05) for foraging, resting, comfort
movement and vigilance. However, a statistically
significant difference was observed in time spent
for locomotion(U= 351.5, df =1, P= 0.001), social
behavior (U= 350, df =1, P = 0.001) and other
activities (U= 308.5, df = 1, P = 0.00), showing that
relatively more time was spent during the dry
season than the wet season.
Habitat utilization
Egyptian geese spent most of their time
foraging in the grassland habitat followed by
shallow water during the dry season. They mainly
use the mudflats for resting followed by shallow
water habitat (Fig. 3a).
Mulugeta Kassa et al.
There was statstically significant difference in
habitat utilization for activities of foraging (F (2, 33)
= 8.729, p = 0.001), resting (F(2,33) = 7.318, p = 0.002)
and comfort movement (F (2,33) = 7.173, p = 0.003) of
Egyptian geese among the three habitat types.
However, there was no statistically significant
difference in activities of locomotion, vigilance,
social behavior, and other behavioral activities in
the three habitat types of the study area (p > 0.05).
During the wet season, most of the birds spent
their time foraging in the grassland habitats
followed by mudflats among the three habitats,
and they used the shallow water for resting
followed by mudflats habitat (Fig. 3b).
Habitat utilization of birds showed statistically
significant variation among the three habitats
during the wet season for foraging (F (2,33) =11.914,
p = 0.000), resting (F (2,33) = 8.271, p = 0.001),
comfort movement (F (2,33) = 3.344, p = 0.048) and
vigilance (F (2,33) = 7.274, p = 0.002). However,
there was no statistically significant difference in
time spent for locomotion, social behavior, and
other behavioral activities among the three habitat
types of the study area (p > 0.05).
Foraging habits
Egyptian geese in Boyo wetland were observed
foraging on a wide variety of food items including
grass, seeds, leaves, grain, crop seedlings, aquatic
rhizomes and tubers, and insects, invertebrates..
During the dry season, Egyptian geese consumed
grass (93.62%), followed by invertebrates (5.56%),
leaves (0.55%), and seeds (0.27%). During the wet
season, the highest percentage frequency of
Egyptian geese diet constituted grass (59.52%) and
invertebrates (40.5%) (Table 1). The percentage
frequency of foraging on grass, seeds, leaves and
invertebrates in the diet were not statistically
different between seasons (U = 6, df =1, P= 0.564;
U = 6, df =1, P= 0.317; U = 6, df =1, P= 0.317 and U
= 8, df =1, P= 1.000), respectively.
In addition, Egyptian geese were observed
foraging on harvested grains such as maize (Zea
mays), sorghum (Sorghum bicolor), and wheat
(Triticum aestivum) and also on unharvested teff
(Eragrostis tef) and soybean (Glycine max) which are
the major cultivated crops around Boyo wetlands.
SINET: Ethiop. J. Sci., 44(2), 2021 187

Figure 2. Percentage of time spent by Egyptian geese’s diurnal activity in the different time blocks during the dry(a) and wet
(b) seasons.
188 Mulugeta Kassa et al.

Figure 3. Percentage of time spent by Egyptian geese’s in different activities and habitat types during a) dry and b) wet seasons.
SINET: Ethiop. J. Sci., 44(2), 2021
Table 1. Percentage frequency of food items consumed
by Egyptian geese during the dry and wet
seasons.
Season Foraging food items
Grass Invertebrate Seeds Leaf
Dry 93.62% 5.56% 0.27% 0.55%
Wet 59.52% 40.5% 0% 0%
Mean 76.56% 23.03% 0.135% 0.275%
DISCUSSION
Most of the Egyptian geese were observed in
mudflat habitats. This habitat may provide a safe
ground for the birds from human disturbance.
They can also move to grassland habitat for
foraging with minimal energy expenditure to
move between habitats. Birds exhibit great
flexibility in adjusting time budget to maintain
their daily ecological requirements (Das et al., 2011)
that could be influenced by several factors such as
weather, season and habitat (Bull, 1997). The
overall diurnal activity patterns in Egyptian geese
showed that most of the birds were observed
resting, followed by foraging. Egyptian geese
spent maximum time forging early in the morning
and in the afternoon hours. However, Egyptian
gees are also reported foraging at night
(Johnsgard, 2010) and this might be the reason for
lower percentage of time spent for foraging
compared to resting during the day in this study.
In addition, human disturbance due to
agricultural activities in the Boyo wetlands might
hinder the birds from following their normal
diurnal activities and could have extended
foraging at night that requires further
investigation.
During the dry season, time spent by Egyptian
geese for foraging did not show any significant
difference among the different time blocks; they
mainly foraged in the afternoon hours (16:00 -
18:00 h) but they also foraged in the morning and
mid day hours. This might be due to the
limitation of the availability of food resources
during the dry season that requires the birds to
spend more time forging to fulfill their energy
requirements. In addition, dry season provides
favorable weather condition to forage
uninterrupted. Similar observation was recorded
in wetland birds such as African Jacana in Lake
Hawassa (Kidist Amha and Bezawork Afework,
2018). However, during the wet season time spent
189
foraging was the highest in the afternoon hours
than the other time blocks. This might be due to
the availability of enough forage and the bird’s
preference of safe foraging time with minimal
Total % human disturbance in the late afternoon hours.
100% The feeding peak at the end of the day might also
100% reflect the overnight energy requirement of birds
100% (Kelly, 1998). Insect preys may also be abundant
during the wet season compared to the dry season
that might have contributed to the relatively
shorter foraging duration by African jacana during
the wet season as insects fulfill protein
requirements of birds (Asokan et al., 2003).
Egyptian geese showed resting peak during
mid-day both during in the wet and dry seasons.
Many species of birds are known to exhibit
maximum time for feeding early in the morning
and in the late afternoon (Natarajan, 1991; Evers,
1994; Ramachandran, 1998; Rodway, 1998; Ali et
al., 2010, Asokanet al., 2010; Aissauoiet al., 2011)
with peak resting during the mid day, which is
common among birds to pass over the high
temperature. An increase in resting during midday
helps birds minimize heat load during high
environmental temperatures. Previous studies on
the activity patterns of birds have revealed similar
patterns of resting during the mid-day (Abrham
Megaze and Afework Bekele, 2013; Wlodarczyk,
2017).
Habitat utilization of Egyptian geese in Boyo
wetland showed that they used the grassland,
mudflat and shallow water habitats differently for
their activities. The grassland habitats was
predominantly used for foraging both during the
dry and wet seasons. As Egyptian geese are
primarily herbivores feeding on grasslands
(Tattan, 2004), this habitat may provide greater
food abundance compared to other habitats. On
the other hand, Egyptian geese were mainly
observed using mudflats and shallow water for
resting during the dry and wet seasons,
respectively.
In the Boyo wetlands, Egyptian geese
predominantly consumed grass both during the
wet and dry seasons . However, they consumed
invertebrates in high proportion during the wet
season. High feeding frequency on insects during
the wet season could be due to the growth of
different vegetation that support insects,
Moreover, the emergence of variety of insects
immediately following the rainfall during the dry
season ensuring insect abundance. Similar
190
observation was reported by Titan (2004) where
Egyptian geese are primarily herbivores, feeding
on grasslands at times far away from water bodies
feeding on aquatic vegetation but also animal prey
such as worms, insects, and frogs. In addition,
harvested grains, such as maize (Zea mays),
sorghum (Sorghum bicolor) and wheat (Triticum
aestivum) and also unharvested teff (Eragrostis tef)
and soybean (Glycine max) were foraged by the
birds from surrounding farmlands of Boyo
wetland. Similar observation was reported in
South Africa where these birds are considered
pests by farmers especially during the germination
periods of the months of April-July and harvesting
periods during October-December (Jancikova,
1996). Extensive economic losses caused by
Egyptian geese due to damage to crops and
agricultural lands were reported resulting in
human-wildlife conflict in South Africa. But their
ecologically important role in decreasing pest
populations around lakes or fields was also noted
(Mangnall and Crowe, 2002). Further investigation
is required on their ecological importance and the
extent of damage to agricultural products in the
farmlands surrounding the Boyo wetland.
CONCLUSION
Boyo wetlands support a wide variety of bird
species. The degradation of this wetland can have
an adverse effect on waterfowl of the area.
Egyptian goose is one of the wetland birds
commonly observed in and around Boyo wetland.
They actively forage in the grassland habitat both
during the wet and dry seasons with peak resting
during the mid day hours. Egyptian geese
primarily feed on grass but a wide variety of
insects were also consumed during the wet season
where they are abundantly available following the
rain. They were observed forging on seedlings of
crops grown nearby the wetland.
The ongoing degradation of Boyo wetland
caused by habitat destruction, heavy
sedimentation, wetland canalization, settlment and
overgrazing can aggravate the human-wildlife
conflict through the shortage of food available for
the Egyptian geese and other bird species in the
area. When natural food sources are insufficient to
meet their nutritional requirements, agricultural
lands on the edge of wetlands become the new
food sources for Egyptian geese and other birds
Mulugeta Kassa et al.
creating conflict. Hence, immediate conservation
intervention is required to maintain Boyo wetland,
an Important Bird Area for the conservation of the
Egyptian geese and their conspecifics.
ACKNOWLEDGEMENTS
We thank Addis Ababa University, Department of
Zoological Sciences and Hadiya Zone, ShashogoWoreda
for providing financial support to carry out this
research.
REFERENCES
1. Aberham Megaze and Afework Bekele (2013). Diet
preference and activity patterns of great white
pelicans (Pelecanus onocrotalus, Linnaeus, 1758)
at Lake Hwassa, Ethiopia. Ethiop. J. Biol. Sci. 12:
211–222.
2. Aissaoui, R., Tahar, A., Saheb, M., Guergueb, L. and
Houhamdi, M. (2011).Diurnal behaviour of
Ferruginous Duck (Aythyanyroca) wintering at
the El-Kala wetlands (northeast
Algeria).Bulletin de l’InstitutScientifique, Rabat,
section Sciences de la Vie 33: 67-75.
3. Ali, A. M. S., Asokan, S and Nithiyanandam, G. T.
(2010). Habitat-related density and activity
patterns of the white-breasted kingfisher
Halcyon smyrnensisin Cauvery Delta, Southern
India. Podoc. 5: 54–62.
4. Altmann, J. (1974). Observation study of behavior,
sampling methods.Behavior 49: 227-267.
5. Asokan, S., Ali, A. M. S., Manikannan, R and
Nithiyanandam, G. T. (2010). Population
densities and diurnal activity pattern of the
Indian Roller Coracias benghalensis (Aves:
Coraciiformes) in Nagapattinam District, Tamil
Nadu, India. J. Threaten. Ta. 2: 1185–1191.
6. Asokan, S., Thiyagesan, K. and Nagarajan, R. (2003).
Studies on Merops orientalis Latham 1801 with
special reference to its population in
Mayiladuthurai, Tamil Nadu. J. Env. Biol. 24:
477–482.
7. Atkins, A. (2015). An experimental assessment of the
efficacy of falconry to mitigate human-wildlife
conflict: Egyptian Geese (Alopochena egyptiaca)
at golf courses. PhD thesis, University of Cape
Town, Cape town. pp.7-13.
8. Banks, A. N., Wright, L., Maclean, I. M., Hann, C. and
Rehfisch, M. M. (2008). Review of the status of
introduced non-native water bird species in the
area of the African-Eurasian Water bird
Agreement: 2007 update. Norfolk, United
SINET: Ethiop. J. Sci., 44(2), 2021
Kingdom: British Trust for Ornithology. Pp.
489.
9. Bibby, C.J., Burgess, N.D. and Hill, D. (1992). Bird
census Techniques. Academic Press, London.
Pp.257
10. Bird Life International (2018). Egyptian goose
(Alopochena egyptiacus). The IUCN Red List of
Threatened species 2018: Downloaded from
http://www.birdlife.org on 27 January 2019.
11. Brinson, M. M. and Malvárez, A. I. (2002). Temperate
freshwater wetlands: types, status, and threats.
Environ. Conserv. 29: 115-133.
12. Bull, W. (1997). Daily variation in activity and flock
size of two parakeet species from Southeastern
Brazil. Willson Bullet. 109: 343–348.
13. Callaghan, C. T. and Brooks, D. M. (2016). Ecology,
behavior, and reproduction of invasive
Egyptian geese (Alopochena egyptiaca) in
Texas.Bull. Texas Ornithol. Soc.49: 37- 45.
14. Chudzinska, M., Madsen, J. and Nabe-Nielsen, J.
(2013). Diurnal variation in the behavior of the
Pink-footed goose (Anserbrachy rhynchus)
during the spring stopover in Trøndelag,
Norway. J. Ornithol. 154: 645-654.
15. Das, J., Deka, H. and Saikia, P. K. (2011). Diurnal
activity budgeting of large whistling teal
Dendrocygna bicolor (Vieillot, 1816)
(Anseriformes: Anatidae) in DeeporBeel
wetlands, Assam, India. J. of Threaten. Ta. 3:
2263–2267.
16. Davies, G. (2005). Egyptian goose (Alopochena
egyptiacus). Ducks, swans, and geese (J. Kear,
editor). Oxford University Press, Oxford, United
Kingdom.Pp. 401- 407.
17. De Melo, T and Guilherme, E. (2016). The foraging
behavior of the Large-headed Flatbill,
Ramphotrigonmegacephalum and the Duskey-
tailed Flatbill, Ramphotrigonfuscicauda (Aves:
Tyrannidae). Zoologia 33: e20160104.
18. Döpfner, M., Quillfeldt, P. and Bauer, H. G.
(2009).Changes in behavioral time allocation of
water birds in wing-molt at Lake
Constance.Water birds 32: 559-571.
19. Dugan, P. J. (Eds). (1990). Wetland conservation: A
review of current issues and required action. IUCN,
Gland, Switzerland.Pp. 94.
20. Evers, D. C. (1994). Activity budgets of a marked
common loon (Gavi aimmer) nestling
population, Hydrobiology 279-280: 415–-420.
21. EWNHS (1996). Important Bird Areas of Ethiopia: A
First Inventory. Ethiopian Wildlife and Natural
History Society. Addis Ababa. pp. 300.
22. Gyimesi, A. and Lensink, R. (2012). Egyptian goose
(Alopochena egyptiacus): an introduced species
191
spreading in and from the Netherlands.
Wildfowl 62: 128-145.
23. Jancikova, K. (1996). Crop and pasture damage by
Egyptian geese (Alopochena egyptiaca) in the
Overberg, Western Cape. Unpublished Nature
Conservation Report, Cape Technikon.
24. Jansen, H., Hengsdijk, H., Legesse, D., Ayenew, T.,
Hellegers, P. and Spliethoff, P. (2007). Land and
water resources assessment in the Ethiopian
Central Rift Valley.Alterra, Wageningen.
25. Johnsgard, P. A. (1978). Ducks, geese, and swans of the
World. University of Nebraska Press.pp.115-
116.
26. Johnsgard, P. A. (2010). Ducks, Geese, and Swans of the
World: Tribe Tadornini (Sheld geese and
Shelducks).University of Nebraska-
Lincoln.pp.115-116.
27. Jónsson, J. E. and Afton, A. D. (2006). Different time
and energy budgets of lesser snow geese in
rice-prairies and coastal marshes in southwest
Louisiana.Waterbirds 29: 451-458.
28. Kelley, K, Clark B, Brown, V. Sitzia, J. (2003). Good
practice in the conduct and reporting of survey
research. International Journal for Quality in
Health Care 15: 261-266.
29. Kidist Amha and Bezawork Afework (2018).Diurnal
activity patterns and feeding behaviourof
African jacana (Actophilornis africanus) in Lake
Hawassa, Ethiopia. SINET: Ethiop. J. Sci.,
41(2):53–60.
30. Lillywhite, H. B and Brischoux, F. (2012). Is it better
in the moonlight? Nocturnal activity of insular
cotton mouth snakes increases with lunar light
levels. J. Zool. 286: 194–199.
31. Maclean, L. (1988). Robert's birds of southern Africa.
New Holland Publishers, London.
32. Mangnall, M. J. and Crowe, T. M. (2002). Population
dynamics and the physical and financial
impacts to cereal crops of the Egyptian goose
(Alopochena egyptiacus) on the Agulhas Plain,
Western Cape, South Africa.Agri. Ecosys.
Environ. 90: 231-246.
33. Mackworth-Praed, C. W. and Grant, C. H. B. (1980).
African handbook of birds: Vol. 1. Longman
Group Limited, London.
34. Makram, A. (2018). Goose World. Poultry Production
Department, Faculty of Agriculture, Fayoum
University, Egypt.The 10th International Poultry
Conference-Proceeding from 26-29 November. pp.
1-12.
35. Mitsch, W. J. and Hernandez, M. E. (2013).
Landscape and climate change threats to
wetlands of North and Central America.
Aquatic Sci.75: 133-149.
192
36. Munoz, J and Colorado, G. (2012). Foraging ecology
of the Cerulean Warbler (setophaga cerulean) in
Andean agro-forestry ecosystems. Ornithlogia
Neotropical 23: 355-362.
37. Natarajan, V. (1991). Time budgeting by the southern
crow-pheasant (Centropussinensis paaaoti) at
point Calimere, Tamil Nadu. Bombay
Natur.Histor Soci 90: 92–95.
38. Oatley, T. O. and Prys-Jones, R. P. (1985). A
comparative analysis of movements of
southern African waterfowl (Anatidae), based
on ringing recoveries. S. Afri. J. Wildl. Resear. 16:
1-6.
39. Ramachandran, N. K. (1998). Activity patterns and
time budgets of the Pheasant-tailed
(Hydrophaslanus chirurgus) and Bronze winged
(Metopidius indicus) jacanas.Natur.Histor.Soci.
95: 234–245.
40. Redman, N., Stevenson, T. and Fanashawe, J. (2009).
Birds of the Horn of Africa.Princeton University
Press. Princeton and Oxford.pp. 496.
41. Rodway, M. S. (1998). Activity patterns, diet and
feeding efficiency of Harlequin Ducks
breeding in Northern Labrador. Canadi. J. Zool.
76: 902–909.
Mulugeta Kassa et al.
42. Stephen, V. (2008). A review of mitigation measures
and their effectiveness in reducing the nuisance
of geese on golf courses. Unpublished MSc thesis,
University of Cape Town.
43. Sutherland, W. J. (2004). Diet and Foraging
Behavior.In: Bird Ecology and Conservation; A
Handbook of Techniques, pp. 233-250 (Sutherland,
W. J., Newton, I. and Green, R. E., eds). Oxford
University Press, Oxford.
44. Tattan, A. 2004. "Alopochen aegyptiaca" (On-line),
Animal Diversity Web. Accessed June 22, 2021
at
https://animaldiversity.org/accounts/Alopoc
hen_aegyptiaca/
45. Webb, E. B., Smith, L. M., Vrtiska, M. P. and
Lagrange, T. G. (2011).Factors influencing
behavior of wetland birds in the Rainwater
Basin during spring migration.Waterbirds 34:
457-467.
46. Wlodarczyk, R. (2017). The daily activity pattern in
males and females of the Mute Swan (Cygnus
olor, Anseriformes) during different parts of the
breeding season.North-Western J. Z. 13: 85–93.
SINET:Ethiop. J. Sci., 44(2): 193–204, 2021 ISSN: 0379–2897 (PRINT)
© College of Natural and Computational Sciences, Addis Ababa University, 2021 eISSN: 2520–7997

Date received: July 20, 2021; Date revised: November 20, 2021; Date accepted: December 08, 2021
DOI: https://dx.doi.org/10.4314/sinet.v44i2.6

Survey and identification of termites (Insecta, Isoptera) using morphological and molecular
methods from eastern, central and western Ethiopia.

Ashenafi Kassaye 1,* , Emana Getu 2, Mulatu Wakgari 1, Muluken Goftishu 1, Awol Seid 1 Samantha J.
Montoya 3 and Gillian H. Gile 3

1School of Plant Sciences, Haramaya University, Haramaya, Ethiopia. E-mail: Ashenafik@gmail.com

2School of Life Sciences, Addis Ababa University
3School of Life Sciences Arizona State University, USA

ABSTRACT: The subfamily Macrotermitinae are the largest members among the Family Termitidae
which are the fungus growing sub-family and Odontotermes are the most abundant genus from the
subfamily. The taxonomy of termites is poorly described in Ethiopia. In the present study 168 termite
samples were collected from eight locations of Eastern, Western and Central Ethiopia. The collected
samples were identified based on morphological and molecular characteristics. Molecular identification
was done based on the DNA sequence of a portion of the mitochondrial 16S rRNA gene. A phylogenetic
analysis of the collected samples and the outgroup resulted in a consensus tree with four distinct
groups. Geographical distribution of the samples also supported the resulting clades. Odontotermes
were the most widely distributed termites from the collected samples. The genetic distance between the
sample showed that Odontotermes zambesiensis, Babile 33 is more distantly related with the rest of the
samples.

Key words /phrases: Ethiopia, Macrotermes, Microtermes, Odontotermes, Termitidae, 16s rRNA

reducing termite infestations (Fenetahun et al.,

INTRODUCTION
Termites, order Isoptera, inhabit a wide range of
ecosystems and can be found on all continents
except Antarctica. They play an important role as
one of the only animals that can break down
lignocellulose through their symbiotic
relationships with flagellates and fungi. Their
tendency to eat plant materials, fresh or decayed,
has marked them as a nuisance for homeowners
and farmers with 10% of described species being
labeled as pests (Wood, 1996).
Termite damage is first reported in Ethiopia in
the Kiltu Kara District, in Western Oromia Region
of Ethiopia (Abdullahi and Haile, 1986; Sanna,
1973). Although damage to crops has been
significant, few studies have been conducted to
analyze the diversity and distribution of termites
endemic to Ethiopia. There is pressure for proper
identification and control of termites, the former
practice of mound poisoning is discouraged as the
chemical compounds used can be quite harmful to
non-target organisms and current management by
physical destruction of the nests is ineffective at
2019; Logan et al., 1990; Weise et al., 1973). The type
of damage can vary based on the type of termites
causing the damage.
In Ethiopia Macrotermitinae termites are greatest
threat to agriculture (Cowie et al., 1990). These
termites are a subfamily of the Termitidae family,
which are sometimes called "higher" termites
based on their phylogenetic relationship to other
Isoptera families. This subfamily cultivates fungi to
facilitate the breakdown of lignocellulose and is
comprised of 12 genera with 373 living species and
two fossil species (Krishna et al., 2013). Of these
genera, Acanthotermes Jacobson (1905),
Pseudacanthotermes Sjöstedt (1924a),
Synacanthotermes Holmgren (1910), Allodontotermes
Silvestri (1912), and Protermes Holmgren (1910), are
unique to the Ethiopian (Afrotropical) region while
Macrotermes Holmgren (1909), Odontotermes
Holmgren (1910), Microtermes Wasmann (1902),
and Ancistrotermes Silvestri (1912) are found in
both Ethiopian and Oriental regions (Krishna et al.,
2013).
The genera Odontotermes comprises the majority
of the Macrotermitinae subfamily with close to 200

_____________________
*Author to whom correspondence should be addressed.
194 Ashenafi Kassaye et al.

species (Krishna et al., 2013). Macrotermes,
Odontotermes, and Pseudcanthotermes tend to grow
large nests that can run 50m long just below the
soil’s surface, this type of nesting can lead to ring
barking and root rot (Wood, 1991). In contrast,
Microtermes and Ancistrotermes create smaller more
diffuse nests that are deep underground and
tunnel upward to the surface; these termites can
tunnel through and up the roots weakening the
plants’ ability to uptake nutrients.In addition, the
type of crop can impact the type of damage caused
(Wood, 1991).
Termite infested rangelands can lead to
overgrazing of grass that can cause soil erosion
and there is an increase in reported diseases in
livestock that graze in termite infested rangelands
(Cowie and Wood, 1989; Fenetahun et al., 2019;
Wood, 1996). The practice of growing exotic trees,
such as Eucalyptus, in previously native woodland
areas in Ethiopia can also lead to accelerated soil
erosion and denudation as these trees are more
susceptible to termite damage (Cowie and Wood,
1989; Fenetahun et al., 2019; Wood, 1991).
The aim of this study is to survey species
diversity and distribution of termites in selected
agro-ecologies of Ethiopia. Highland, mid-
highland, and lowland regions across Western,
Central, and Eastern Ethiopia.
MATERIALS AND METHODS
Study areas
The study was conducted in areas rich in termite
diversity (Cowie et al., 1990) located in Eastern,
Central and Western Ethiopia. Survey was
conducted during the 2019/20 cropping season.
The survey sites were selected based on
representation of the different agro-ecologies viz.,
highlands (above 2000 m.a.s.l), mid-highlands
(1500-2000 m.a.s.l.) and lowlands (below 1500
m.a.s.l) as shown in Table 1.

Table 1.Sampling sites in Oromia state, Ethiopia.

Code Locality name Latitude Longitude Altitude (m) Habitat

2 Bako 9°05'57"N 37°06'00"E 1634 Grass
4 Bako 9°05'48"N 37°06'03"E 1629 Maize
5 Bako 9°05'30"N 37°06'14"E 1623 Maize
6 Bako 9°06'12"N 37°05'46"E 1704 Grazing land
8 Bako 9°05'47"N 37°06'60"E 2171 Grazing land
10 Ambo 8°58'24"N 37°57'04"E 2386 Grazing land
11 Ambo 8°58'35"N 37°57'55"E 2432 Grazing land
12 Ambo 8°58'35"N 37°58'08"E 2432 Grazing land
13 Holeta 9°03'41"N 38°30'31"E 2340 Forest
14 Holeta 9°03'41"N 38°30'33"E 2329 Forest
15 Holeta 9°03'38"N 38°30'38"E 2397 Forest
16 Holeta 9°03'38"N 38°30'40"E 2472 Forest
17 Bishoftu 8°38'17"N 39°04'52"E 1799 Acacia
24 Metehara 8°54'25"N 39°57'51"E 1262 Arid
27 Metehara 8°54'26"N 39°59'06"E 1105 Arid
28 Asebot 9°23'52"N 41°00'15"E 1626 Arid
30 Haramaya 9°24'53"N 42°01'58"E 1930 Mango
33 Babile 9°13'12"N 42°19'25"E 1328 Groundnut field

From each survey region, two to three districts
were selected for sampling based on their
accessibility and proximity to roads. From each
site, 10 soldiers and 10 worker termites were
randomly sampled for identification. Samples
were collected by opening termite nests, splitting
wood to expose specimens and excavating the soil
to a depth of 15-25cm using a shovel.
Microsites surveyed include: surface soil; deep
accumulations of leaf litter; decayed and dead
wood; and termite nests including subterranean,
epigeal, arboreal and mounds. Field samples were
collected from unhealthy looking groundnut crop,
other cultivated plants, and natural habitats.
Samples were collected in the morning and late
evening. Termites were extracted from the sample
soil by hand-sorting using a moistened camel's-
hair brush and preserved in 97% alcohol.
The collection localities were geo-referenced
(latitude, longitude and altitude) using GPS.
Voucher specimens were preserved in 85% ethanol
and deposited to the Arizona State University
SINET: Ethiop. J. Sci., 44(2), 2021
Hasbrouck Insect Collection under accession
numbers ASUHIC0095057 - ASUHIC0095074.
Morphological identification
Soldiers were used for identification to species
level using morphological characters. In this study,
head color, size, and shape, number of antennae
segments, mandible shape, pronotum shape, and
gula shape were considered as morphological
characters in the identification keys. Species
determination was made based on previous
descriptions of morphological identification using
taxonomic key by Krishna et al (2013) for
Macrotermitinae; key described by Bagine (1986)
was used to characterize the genus Odontotermes;
morphological characters described by Rambur
(1842) and Grassé (1937) were used for
Macrotermes; and descriptions used by Chhotani
(1997) was used for Microtermes. Samples
containing only workers were subjected to
molecular identification. Coordinates of each
collection location were plotted on the map using
QGIS 3.10 software.
Molecular characterization
Molecular characterization of the termites was
done using 16S rRNA gene sequences. Termite
samples were preserved in 97% ethanol for DNA
extraction and molecular characterization. DNA
was extracted from termite abdomens with the
GeneJET Genomic DNA Purification Kit
(ThermoFisher Scientific) according to the
manufacturer’s protocol. The mitochondrial 16S
rRNA gene was amplified by PCR using Econotaq
Plus Green PCR Master Mix (Lucigen) and primers
LR-N 5’-CGC CTG TTT ATC AAA AAC AT-3’ and LR-J 5’-
TTA CGC TGT TAT CCC TAA-3’ (Kambhampati and
Smith, 1995; Simon et al., 1994), under cycle
conditions of 3 min at 95 °C followed by 30 cycles
of 95 °C for 30s, 46 °C for 30s, and 72 °C for 60s.
PCR products were purified using NucleoSpin Gel
and PCR Clean-Up kit (Macherey-Nagel), and
sequenced directly on both strands on an Applied
Biosystems 3730 capillary sequencer. Termite
mt16S sequences generated in this study were
submitted to GenBank under accession numbers
MW324511-MW324528.
195
Phylogenetic analysis
New and previously published sequences of
Macrotermitinae 16S rRNA were assembled using
MEGA 10.1.8 (Kumar et al., 2018) software package
and aligned with Muscle using default settings.
The sequence alignment had a total of 50
nucleotide sequences in the final dataset. The
BLAST program
(http://www.ncbi.nlm.nih.gov/blast/) was used
to identify similarities between the sequences
obtained in this work and previously published
data. Sphaerotermes sphaerothorax Sjöstedt (1911a)
(Isoptera: Termitidae) (Gen Bank KP026279) and
Sphaerotermes sp. (Isoptera: Termitidae) (GenBank
KY238292.1) were used as the outgroup taxa.
The evolutionary history was inferred by using
the Maximum Likelihood method and Kimura 2-
parameter model (Kumar et al., 2018). Initial tree(s)
for the heuristic search were obtained
automatically by applying Neighbor-Join and
BioNJ algorithms to a matrix of pairwise distances
estimated using the Maximum Composite
Likelihood (MCL) approach, and then selecting the
topology with superior log likelihood value. The
tree is drawn to scale, with branch lengths
measured in the number of substitutions per site.
Evolutionary analyses were conducted in MEGA
10.1.8 (Kumar et al., 2018).
RESULTS AND DISCUSSION
Morphological identification
In the present study 168 samples were collected
where 44 samples had soldiers, 98 samples had
workers and 26 samples hadboth soldiers and
workers. Among the collected samples, seven
species of termites were identified representing
threegeneraof the family Termitidae viz.,
Odontotermes, Macrotermes and Microtermes (Figure
1). Among them the genus Odontotermes is
represented by four species viz., Odontotermes
montanus, Odontotermes stercorivorus, Odontotermes
tanganicus and Odontotermes zambesiensis; the genus
Macrotermes was represented by two species
Macrotermes jeanneli and Macrotermes subhyalinus;
and the genus Microtermes was represented by a
single species Microtermes obesi. Based on
morphological characters of the soldier caste, keys
were prepared for easy and accurate identification
of all the collected termite samples.
196 Ashenafi Kassaye et al.

Figure 1 Map of Ethiopia showing geographical distribution of Termitidae: Microtermes, Macrotrmes, and Odontotermes
included in the study. Dots represents collection sites.

Head yellowish to brownish yellow with relationships had strong quantitativesupport as

brownish tinge; Antennae with 13-14 segments; in indicated by bootstrap analysis. The highly
specimens with 14 articles, second longer than conserved 16S rRNA gene separated the collected
third; postmentum with scattered bristles and samples in to three genera Microtermes, Macrotrmes,
short hairs; pronotum slightly less hairy than head. and Odontotermes. O. stercorivorus and O. montanus
Head elongate oval, longer than broad, widest a share almost identical mitochondrial 16S rRNA
little below middle. Mandibles thin, delicate and gene sequence and they were identified using
weakly incurved apically Pronotum saddle morphologically
shaped, longer than half its width; anterior margin
distinctly notched in middle; posterior margin
very weakly depressed in middle; lateral lobes
prominent, broadly rounded (Ahmad, 1965;
Chhotani, 1997).

Molecular identifications
A phylogenetic analysis of Termitidae from eight
locations, 18 nucleotide sequences and two
Rhinotermitidae species (outgroups) having 447
base pair long fragment based on the DNA
sequence of a portion of the mitochondrial 16S
rRNA gene were performed (
Figure 2). Collections were made from West, Figure 2 PCR amplified products of 16S rRNA gene of various
Central and East Ethiopia, most of the inferred species of termites
SINET: Ethiop. J. Sci., 44(2), 2021 197

The sequences from O. stercorivorus and O. 14, Holeta 15, Holeta 16 and Haramaya 30
montanus were nearly identical, consistent with provided similar sequences to both O. montanus
previous studies, suggesting they should be and O. stercorivorus but morphological
considered one single species. Molecular identification of the samples revealed that all
characterization of samples including: Bako 8, belong to the same species O. montanus as shown
Ambo 10, Ambo 11, Ambo 12, Holeta 13, Holeta in Table 2.

Table 2 Measurements of soldier caste, n=20.

Mean ±SD (mm) Range (mm) Minimum (mm) Maximum (mm)

Total body length 10.32 ± 0.6 3.96 8.75 12.71
Head length with mandibles 5.42 ± 0.3 2.009 4.683 6.692
Length of mandible 1.41 ± 0.14 1.0226 0.9504 1.973
Head width 2.78 ± 0.09 0.547 2.643 3.19
Head height 4.01 ± 0.21 1.252 3.467 4.719

Morphological studies showed that O. montanus soldiers of O. stercorivorus and O. montanus

is larger than O. stercorivorus (Bagine, 1986). Table indicating the collected samples are O. montanus.
3 shows that comparative measurements of

Table 3. Comparative measurements of soldier in mm (Bagine, 1986).

O. montanus (Bagine, O. montanus (Harris, O. stercorivorus

1986) 1960) (Bagine, 1986)
head length 2.62 2.64 2.04
head width 1.96 1.96 1.35
left mandible length 1.58 1.46 1.06
pronotum width 1.36 0.82 1.01
hind tibia length 1.72 1.60 1.05
length of head with mandibles 4.20 4.00 3.10

The minimum divergence of 0.23 was seen
between M. subhyalinus and M. jeanneli species.
The maximum divergence of 0.255 was observed
between O. zambesiensis, Babile 33 and three rift
valley isolates M. jeanneli, Bishoftu 17, M. jeanneli,
Metehara 24 and M. jeanneli, Asebot 28. Samples
from west of the rift valley including Bako, Holleta
and Ambo are different from the central rift valley
samples: Bishoftu and Metehara but the Haramaya
species, O. montanus, Haramaya 30 is genetically
identical with the west samples. Outgroup taxa, R.
Arenicola and R. virginicus, have a value of more
than one indicating that they are completely
different groups (Table 4).
198 Ashenafi Kassaye et al.

Figure 3. Termitidae species from Ethiopia A. M. jeanneli. B.M. subhyalinus. C. M. obesi. D. O. montanus.E. O. tanganicus.F. O.
zambesiensis.

In the present study all the three genera,
Odontotermes, Macroterms and Microtermes, form a
monophyletic group and clustered together.
However, O. zambesiensis, Babile 33 collected from
Babile show the greatest diversity among the
Odontotermes species. The tree with the highest log
likelihood (-66926.94) is shown. The percentage of
trees in which the associated taxa clustered
together is shown next to the branches (Figure 3).
 SINET: Ethiop. J. Sci., 44(2), 2021 199

Table 4. Tajima-Nei pairwise genetic distances.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
1 Macrotermes subhyalinus, Bako 2
2 0.000 Macrotermes subhyalinus, Bako 5
3 0.023 0.023 Macrotermes jeanneli, Bishoftu 17
4 0.023 0.023 0.000 Macrotermes jeanneli, Metehara 24
5 0.023 0.023 0.000 0.000 Macrotermes jeanneli, Asebot 28
6 0.164 0.164 0.168 0.168 0.168 Microtermes obesi, Bako 4
7 0.164 0.164 0.168 0.168 0.168 0.000 Microtermes obesi, Bako 6
8 0.178 0.178 0.193 0.193 0.193 0.144 0.142 Odontotermes montanus, Bako 8
9 0.178 0.178 0.193 0.193 0.193 0.144 0.142 0.000 Odontotermes montanus, Holeta 14
10 0.177 0.177 0.192 0.192 0.192 0.153 0.148 0.006 0.006 Odontotermes montanus, Ambo 10
11 0.181 0.181 0.197 0.197 0.197 0.152 0.152 0.004 0.004 0.008 Odontotermes montanus, Ambo 11
12 0.177 0.177 0.192 0.192 0.192 0.148 0.145 0.003 0.003 0.008 0.000 Odontotermes montanus, Ambo 12
13 0.177 0.177 0.192 0.192 0.192 0.148 0.145 0.003 0.003 0.008 0.000 0.000 Odontotermes montanus, Holeta 13
14 0.177 0.177 0.192 0.192 0.192 0.148 0.145 0.003 0.003 0.008 0.000 0.000 0.000 Odontotermes montanus, Holeta 15
15 0.182 0.182 0.197 0.197 0.197 0.148 0.145 0.000 0.000 0.006 0.004 0.003 0.003 0.003 Odontotermes montanus, Holeta 16
16 0.182 0.182 0.197 0.197 0.197 0.148 0.145 0.000 0.000 0.006 0.004 0.003 0.003 0.003 0.000 Odontotermes montanus, Haramaya 30
17 0.192 0.192 0.208 0.208 0.208 0.168 0.168 0.039 0.039 0.043 0.036 0.035 0.035 0.035 0.039 0.039 Odontotermes tanganicus, Metehara 27
18 0.233 0.233 0.255 0.255 0.255 0.151 0.122 0.126 0.126 0.133 0.158 0.126 0.126 0.126 0.129 0.129 0.150 Odontotermes zambesiensis, Babile 33
19 1.046 1.046 1.083 1.083 1.083 1.083 0.988 1.127 1.127 1.098 1.096 1.119 1.119 1.119 1.132 1.092 1.038 1.024 Reticulitermes arenicola, AY168214.1
20 1.107 1.107 1.147 1.147 1.147 1.144 1.041 1.199 1.199 1.171 1.172 1.190 1.190 1.190 1.204 1.165 1.116 1.083 0.021 Reticulitermes virginicus, AY168224.1
200 Ashenafi Kassaye et al.

Figure 4. The maximum likelihood tree. The percentage of trees in which the associated taxa clustered together is shown next to
the branches.
SINET: Ethiop. J. Sci., 44(2), 2021
The mitochondrial 16S rRNA gene sequences
were obtained from each of the 18 Macrotermitinae
samples collected in this study. A phylogenetic
analysis of new and previously published
Macrotermitinae sequences with two
Sphaerotermitinae species as out groups was
performed. The tree with the highest log likelihood
(-66924.38) is shown in Figure 3.
Each of the three genera Microtermes,
Macrotermes, and Odontotermes formed a (Wanyonyi et al., 1984; Wood, 1991). Mitochondrial
moderately supported clade, with Odontotermesas
the deepest branch. New Odontotermes sequences
branched with previously published sequences
from the same species, except for Odontotermes
tanganicus. Instead, O. tanganicus sequences were
paraphyletic to Odontotermes amaniensis Sjöstedt
(Sjöstedt, 1924b), Odontotermes anceps Sjöstedt
(1911b), Odontotermes stercorivorus Sjöstedt
(Sjöstedt, 1907), and Odontotermes montanus.
Our new Macrotermes sequences did not branch
with previously published conspecifics, but
instead formed two supported sister clades within
a larger Macrotermes clade including Macrotermes
Figure 5. Distribution of termites in Ethiopia (Cowie et al., 1990).
201
natalensis and Macrotermes falcigerin addition to M.
jeanneli and M. subhyalinus. Similarly, our
Microtermes obesi sequence did not branch sister to
a previously published sequence though it did fall
within the same supported clade.
The phylogenetic analysis of Termitidae
provided molecular data for the Ethiopian termite
species which have been previously reported
based on morphological characters only
ribosomal RNA sequence has been used for
phylogenetic studies of termites (Darlington et al.,
2008; Eaton et al., 2016; Kambhampati et al., 1996;
Szalanski et al., 2006).
The result now provides evidence to the species
diversity of termites in east and west Ethiopia.
Three genera from the Termitidae family are
identified in this study including Microtermes,
Macrotermes, and Odontotermes. Cowie et al. (1990)
reported that 17 different termite genera from
different parts of the country (Figure 4) including
those identified in this study.
202
The results demonstrated two things. First, the
subfamily Macrotermitinae is the most common
termite groups in the survey area. Second, species
of Odontotermes are widely distributed in
Macrotermitinae at the studied locations. In East
Africa Odontotermes are abundant and diverse
groups of termites than any other termite genus
with thirty recorded species (Bagine, 1986).
This study also identified new termite species to
Ethiopia which includes O. tanganicus (Sjöstedt,
1924b), O. zambesiensis (Sjöstedt, 1914), M. obesi
(Holmgren, 1912) and M. jeanneli (Grassé, 1937).
The O. tanganicus and O. zambesiensis (Sjöstedt,
1914) are geographically distributed in Kenya
(Darlington et al., 1997; Davison et al., 2001;
Wanyonyi et al., 1984); Malawi (Munthali et al.,
1999); Tanzania (Mathur and Thapa, 1962;
Wanyonyi et al., 1984; Weidner, 1960). Whereas M.
obesi (Holmgren, 1912) is distributed in
Bangladesh, India, Pakistan and Sri Lanka
(Roonwal and Verma, 1977; Verma and Thakur,
1982); Thailand and Vietnam (Maiti, 1983) and
Bhutan, Cambodia, Myanmar (Verma, 1990). M.
jeanneli on the other hand is distributed in
Ethiopian region (Ghidini, 1955; Snyder, 1949),
Kenya (Darlington et al., 1997; Davison et al., 2001;
Snyder, 1949) and Sudan (Mathur and Thapa,
1962).
O. stercorivorus and O. montanus share almost
identical mitochondrial 16S rRNA gene sequence,
in agreement with previous studies (Davison et al.,
2001), though they differ morphologically
(Darlington et al., 2008). According to Bagine (1986)
O. montanus is morphologically larger than O.
stercorivorus as shown in Table 3. The findings in
this report confirms that the collected samples are
O. montanus where morphometric data of every
parameters greater than O. stercorivorus as
compared to Bagine (1986) as shown in Table 2.
This result ties well with previous studies of
Davison et al. (2001) that confirms O. stercorivorus
and O. montanus as they belong to a single genetic
population but having two distinct morphologies.
Genetic similarity between O. stercorivorus and O.
montanus was also observed between Kenyan
termite samples (Davison et al., 2001). These
samples are morphologically and behaviorally so
different but closely related groups and may
belong to a single genetic population (Darlington
et al., 2008).
Ashenafi Kassaye et al.
CONCLUSIONS
The paper concludes by arguing that termites are
important creatures as pests and in their ecological
role. Knowing the distribution and diversity of
termites will help to understand the kind of
species available in the country and will help to
identify the pest species easily. Importantly, our
results provide evidence that Odontotermes are the
most abundant termite species in the surveyed
areas.In addition, these findings provide
additional information about the new species
reported in this study for the first time in Ethiopia.
Future research on termite survey and
identification might extend to untouched parts of
the country. We believe that apart from using only
morphological identification, future research
should focus on molecular identification of
termites. In addition, further surveys in additional
geographic areas and climatic conditions in the
country might prove an important area for future
research.
ACKNOWLEDGEMENTS
We thank Haramaya University and Ministry of
Education of Ethiopia for their support. We extend our
thanks to anonymous reviewers and the editors for
useful suggestions that improved the quality of the
paper on the manuscript.
REFERENCES
1. Abdullahi, A. and Haile, A. (1986). Survey of termite
problems in Wellega Administrative Region.,
Committee of Ethiopian Entomologists Newsletter
5: 4–9.
2. Ahmad, M. (1965). Termites ( Isoptera ) Of Thailand,
Bulletin of The American Museum’of Natural
History, Vol. 131.
3. Bagine, R. K. N. (1986). Biosystematic Studies of the
Termite Genus Odontotermes (Holmgren) with
Special Reference to Kenya.
4. Chhotani, O. B. (1997). Isoptera (Termites), In The
Fauna of India and The Adjacent Countries,
Volume II., Calcutta: Zoological Survey of
India, p. xx + 801.
5. Cowie, R. H. and Wood, T. G. (1989). Damage to
crops, forestry and rangeland by fungus-
growing termites (Termitidae:
Macrotermitinae) in Ethiopia., Sociobiology 15:
139–153.
SINET: Ethiop. J. Sci., 44(2), 2021
6. Cowie, R. H., Wood, T. G., Barnett, E. A., Sands, W. A.
and Black, H. J. (1990). A checklist of the
termites of Ethiopia with a review of their
biology, distribution and pest status, African
Journal of Ecology 28(1): 21–33. Retrieved from
https://doi.org/10.1111/j.1365-
2028.1990.tb01133.x
7. Darlington, J. P. E. C., Benson, R. B., Cook, C. E. and
Walker, G. (2008). Resolving relationships in
some African fungus-growing termites
(Termitidae, Macrotermitinae) using molecular
phylogeny, morphology, and field parameters,
Insectes Sociaux 55(3): 256–265.
8. Darlington, J. P. E. C., Zimmerman, P. R., Greenberg,
J., Westberg, C. and Bakwin, P. (1997).
Production of metabolic gases by nests of the
termite Macrotermes jeanneli in Kenya, Journal of
Tropical Ecology 13(4): 491–510.
9. Davison, D., Darlington, J. P. E. C. and Cook, C. E.
(2001). Species-level systematics of some
Kenyan termites of the genus Odontotermes
(Termitidae, Macrotermitinae) using
mitochondrial DNA, morphology, and
behaviour, Insectes Sociaux 48(2): 138–143.
10. Eaton, T. D., Jones, S. C. and Jenkins, T. M. (2016).
Species Diversity of Puerto Rican Heterotermes
(Dictyoptera: Rhinotermitidae) Revealed by
Phylogenetic Analyses of Two Mitochondrial
Genes, Journal of Insect Science 16(1).
11. Fenetahun, Y., Xinwen, X. and Yongdong, W. (2019).
An Evaluation of the Effect of Termites on
Rangeland Degradation: The Case of Yabello,
Southern Ethiopia, Journal of Resources and
Ecology 10(5): 525–529. Retrieved from
https://doi.org/10.5814/j.issn.1674-
764x.2019.05.008
12. Ghidini, G. M. (1955). Missione biologica Sagan-Omo
diretta dal Prof. Edoardo Zavattari.
Termitidae., Rivista di Biologia Coloniale 15: 69–
82.
13. Grassé, P. P. (1937). Le Bellicositermes Jeanneli, n. sp.
constructeur de grandes termitières à cheminée,
In Bulletin de La Société Entomologique de France,
Vol. 42, pp. 71–73.
14. Harris, W. (1960). Further Records of East African
Termites-III, Further Records of East African
Termites—III*. Proceedings of the Royal
Entomological Society of London. Series B,
Taxonomy, 29(1–2): 17–21.
15. Holmgren, N. (1909). Termitenstudien. 1.
Anatomische Untersuchungen., In Kungliga
Svenska Vetenskaps-Akademiens Handlingar, Vol.
44, pp. 1–215 + 3 pls.
16. Holmgren, N. (1910). Das System der Termiten,
Zoologischer Anzeiger 35(10): 284–286.
17. Holmgren, N. (1912). Termitenstudien. 3. Systematik
203
der Termiten. Die Familie Metatermitidae.,
Kungliga Svenska Vetenskaps-Akademiens
Handlingar 48(4): 1–166 + 4.
18. Jacobson, G. (1905). Zur Kenntnis der Termiten
Russlands, Annuaire du Musée Zoologique de
l’Académie Impériale des Sciences de St.-
pétersbourg 9(1904): 57–107.
19. Kambhampati, S., Kjer, K. M. and Thorne, B. L.
(1996). Phylogenetic relationship among termite
families based on DNA sequence of
mitochondrial 16S ribosomal RNA gene, Insect
Molecular Biology 5(4): 229–238.
20. Kambhampati, S. and Smith, P. T. (1995). PCR
primers for the amplification of four insect
mitochondrial gene fragments., Insect molecular
biology 4(4): 233–236.
21. Krishna, K., Grimaldi, D., Krishna, V. and Engel, M.
(2013). Treatise on the Isoptera of the World,
Bulletin of the American Museum of Natural
History.
22. Kumar, S., Stecher, G., Li, M., Knyaz, C. and Tamura,
K. (2018). MEGA X: Molecular Evolutionary
Genetics Analysis across computing platforms.,
Molecular Biology and Evolution 35: 1547-1549.
23. Logan, J., Cowie, R. and Wood, T. (1990). Termite
(Isoptera) control in agriculture and forestry by
non-chemical methods: A review., Bulletin of
Entomological Research 80(3): 309–330.
24. Maiti, P. K. (1983). Termite fauna (Isoptera) of West
Bengal, India, their recognition, biology and
ecology, Records of the Zoological Survey of India,
Occasional Paper 42: i–iv + 1–152 + 4 pls.
25. Mathur, R. N. and Thapa, R. S. (1962). A revised
catalogue of Isoptera (white ants) of the
Entomological Reference Collection at the
Forest Research Institute, Dehra Dun., Indian
Forest Leaflet (Entomology) 167: 1–122.
26. Munthali, D. C., Logan, J. W. M., Wood, T. G. and
Nyirenda, G. K. . (1999). Termite distribution
and damage to crops on smallholder farms in
southern Malawi., Insect Science and Its
Application 19(1): 43–49.
27. Rambur, P. (1842). Histoire Naturelle Des Insectes.
Névroptères., Paris: Librairie Encyclopédique
De Roret.
28. Roonwal, M. L. and Verma, S. C. (1977). Re-survey of
the termite fauna of Rajasthan, India and its
zoogeography, Records of the Zoological Survey of
India 72(1–4): 425–480.
29. Sanna, E. (1973). Termites as Agricultural Pests in
Wes- tern Wellega., Appropriate Technology Unit,
CROA, Addis Ababa.
30. Silvestri, F. (1912). Termiti raccolte da S.A.R. la
Duchessa d’Aosta nelle regione dei grandi laghi
dell’Africa equa- toriale., Annuario del Museo
204
Zoologico della R. Università di Napoli (n.s.) 3(22–
23): 1–5.
31. Simon, C., Frati, F., Beckenbach, A., Crespi, B., Liu,
H. and Flook, P. (1994). Evolution, weighting,
and phylogenetic utility of mitochondrial gene
sequences and a compilation of conserved
polymerase chain reaction primers, Annals of
the Entomological Society of America 87(6): 651–
701.
32. Sjöstedt, Y. (1907). Scientific results of the Swedish
zoological expedition to Kilimandjaro, Meru
and the surrounding Massai steppes German
East Africa 1905–1906 under the direction of
prof. Dr. yngve Sjöstedt, Uppsala: Almqvist
and Wiksells,: Herausgegeben von der
Königliche Schwedischen Akademie der
Wissenschaften. 15. Corro- dentia. 1.
Termitidae. Vol. 3.
33. Sjöstedt, Y. (1911a). Zur Termitenfauna Kongos.
Entomologisk Tidskrift, 1911 32(3–4): 137–170.
34. Sjöstedt, Y. (1911b). Termitidae novae a Cl. Dom.
Maurice de Rothschild ex Aethiopia reportatae
quas descripsit., Entomologisk Tidskrift 32(3–4):
171–172.
35. Sjöstedt, Y. (1914). Termiten aus Zambesi, Rhodesia,
Nyassa und Süd-Nigeria., Arkiv för Zoologi
8(28): 1–9.
36. Sjöstedt, Y. (1924a). Neue Arten und Gattungen
Afrikanischer Termiten (vorläufige
Mitteilung)., Revue Zoologique Africaine
(Bruxelles) 12(2): 253–257.
37. Sjöstedt, Y. (1924b). Neue Afrikanische Termiten.,
Revue Zoologique Africaine (Bruxelles) 12(4): 490–
494.
38. Snyder, T. E. (1949). Catalog of termites (Isoptera) of
the world., Smithsonian Miscellaneous Collections
112(3953): 1–490.
39. Szalanski, A. L., Scheffrahn, R. H., Austin, J. W.,
Krecek, J. and Su, N.-Y. (2006). Molecular
Phylogeny and Biogeography of Heterotermes
(Isoptera: Rhinotermitidae) in the West Indies,
Ashenafi Kassaye et al.
Annals of the Entomological Society of America
97(3): 556–566.
40. Verma, S. C. (1990). Termites from Uttar Pradesh
Tarai Region (India) with a key for
identification of species., Indian Journal of
Forestry 12(4): 296–301.
41. Verma, S. C. and Thakur, R. K. (1982). Termites from
Madhya Pradesh, India, with new
distributional records (Insecta: Isoptera),
Records of the Zoological Survey of India 79: 311–
318.
42. Wanyonyi, K., Darlington, J. and Bagine, R. (1984).
Checklist of the species of termites (Isoptera)
recorded from East Africa, Journal of the East
Africa Natural History Society and National
Museum 181(10): 1–10.
43. Wasmann, E. (1902). Termiten, Termitophilen und
Myrmekophilen gesammelt auf Ceylon von Dr.
W. Horn 1899, mit anderm ostindischen
Material bearbeitet., Zoologische Jahrbücher,
Abteilung für Systematik, Ökologie und Geographie
der Tiere 17(1): 99–164 + 2 pls.
44. Weidner, H. (1960). Mission zoologique de
l’I.R.S.A.C. en Afrique orientale. (P. Basilewsky
et N. Leleup, 1957). IV. Isoptera., Annales du
Musée Royal du Congo Belge, Série in 8o, Sciences
Zoologiques 81: 30–39.
45. Weise, I. H., Basson, N. C. J., Basson, P. A., Naude, T.
W. and Maartens, B. P. (1973). The toxicology
and pathology of dieldrin and photodieldrin
poisoning in two antelope species,
Onderstepoort J. Vet. Res. 40: 31–40.
46. Wood, T. G. (1991). Termites in Ethopia: The
Environmental Impact of Their Damage and
Resultant Control Measures, Ambio 20(3–4):
136–138.
47. Wood, T. G. (1996). The agricultural importance of
termites in the tropics., Agric Zool Rev 7: 117–
155.
SINET:Ethiop. J. Sci., 44(2): 205–214, 2021 ISSN: 0379–2897 (PRINT)
© College of Natural and Computational Sciences, Addis Ababa University, 2021 eISSN: 2520–7997

Date received: December 12, 2020; Date revised: October 06, 2021; Date accepted: October 23, 2021
DOI: https://dx.doi.org/10.4314/sinet.v44i2.7

Determination of selected major and trace metals in lemongrass (Cymbopogon

citratus) by microwave plasma-atomic emission spectrometry

Nitsuh Birhanu, Weldegebriel Yohannes* and Bhagwan Singh Chandravanshi

Department of Chemistry, College of Natural and Computational Sciences, Addis Ababa

University, PO Box 1176, Addis Ababa, Ethiopia. E-mail: weldegebriel.yohannes@aau.edu.et

ABSTRACT: Cymbopogon citratus is very important and of great interest due to its commercially
valuable essential oils, its common use in food and beverage industries as well as in traditional folk
medicine. The objective of the present study was to determine the levels of selected major (K, Mg, Ca)
and trace (Mn, Fe, Cu, Zn, and Pb) metals in lemongrass (Cymbopogon citratus) samples collected from
Addis Ababa, Ankober, Finote Selam (Gojjam), and Wondogenet by microwave plasma-atomic
emission spectrometry. The optimized wet acid digestion method for the analysis of lemongrass was
found efficient for all the metals and it was evaluated through the recovery test and good percentage
recovery of 86.9 to 106% was obtained for all the metals identified. The metals K, Ca, Mg, Fe, Mn, Cu,
Zn and Pb were with a concentration of 743.8-1020, 123.1-129.3, 23.9-36.3, 10.35-22.3, 10.0-12.7, 1.48-2.53,
0.59-1.07, 0.13-0.20 mg/kg, respectively. The results of the Pearson correlation showed that there was a
weak and strong positive correlation between the concentrations of the metals analyzed. Statistical
analysis by using one-way ANOVA indicated that there were significant differences in the mean
concentrations of all the metals (except Ca and Mn) across each sampling points. The results also
showed that Cymbopogon citratus are beneficial sources of essential metals. The levels of the metals in
the analyzed samples were within the WHO maximum permissible limits and thus safe for human
consumption.

Keywords/phrases: Major and trace metals, Cymbopogon citratus, wet digestion, microwave plasma-
atomic emission spectroscopy

and America (Shah, 2011; Anal, 2014; Avoseh et al.

INTRODUCTION
Cymbopogon citratus popularly known as
lemongrass is one of the aromatic plants which are
used to make herbal medicine (Wilson et al., 2012;
Christopher et al., 2014; Singh, 2015). The lemon-
like odour of the plant could be ascribed to the
existence of a cyclic monoterpene (Manvitha and
Bidya, 2014).
The genus Cymbopogon belongs to the family of
Gramineae (syn. Poaceae), which are herbs known
around the world for their high yield of essential
oil. The herb is grown in a wide range of soil and
climatic conditions. It requires a hot and humid
climate with high rainfall and long sunlight (Aftab
et al., 2011; Thorat et al., 2017). The plant grows in
dense clumps up to two meters in diameter and
has leaves up to a height of one meter (Aftab et al.,
2011). It is widely distributed and cultivated in
tropical and subtropical countries of Asia, Africa,
_____________________
*Author to whom correspondence should be addressed.
2015).
Most of the population in Africa and other
developing countries rely on herbal medicine to
meet their primary health care needs for the
maintenance of good health. (Falkenberg, 2002;
Singh, 2015; Nkansah et al., 2016).
Cymbopogon citratus is of great interest due to its
commercially valuable essential oils, its common
use in food industries as well as in traditional
medicine (Manvitha and Bidya, 2014). The plant is
broadly used as flavoring agent in the food and
beverage industries. Moreover, its applications
have been reported in pharmaceutical, cosmetics,
soap, and detergent industries. Some studies
(Francisco et al., 2011; Ajayi et al., 2016; Ekpenyong
et al., 2015; Avoseh et al., 2015) have reported the
pharmacological activities of the plant such as
antibacterial, antifungal, anti-inflammatory,
anticancer, analgesic, antiseptic, etc. Thus, the
plant is used commonly as a remedy in the
treatment of illness from a toothache, swollen
206
gums, cough, rheumatism, malaria, bladder, and
digestive problems (Shah, 2011; Balakrishnan et al.,
2014; Ahmad and Viljoen, 2015; Babarinde et al.,
2016).
Human beings require minerals in varying
proportions for proper growth, health
maintenance, and general wellbeing. For example,
the major metal magnesium is essential to the
enzymatic function of all cellular life (Moomaw
and Maguire, 2008). Potassium is important for its
diuretic nature and helps in the proper function of
the brain as well as nerves and participates
actively in the maintenance of cardiac rhythm,
thereby preventing stroke (Silva et al., 2005; He and
MacGregor, 2008). Calcium is a multifunctional
nutrient essential to the body’s metabolism and the
formation of bones and teeth (Silva et al., 2005).
The trace metals are the components of the
structures of different active bio-compounds and
are necessary for the physiological and biological
function of the human body. For instance, zinc,
copper, and manganese are essential components
of many enzymes, also important for proper
growth, bone and connective tissues development,
elimination of free radicals, reproductive and the
central nervous system. Iron has various functions
in the human body including oxygen supply,
energy production, and immunity (Devi and
Sarma, 2013; Dghaim et al., 2015; Yohannes et al.,
2018). However, too low or excessive intake of
minerals beyond permissible limits in human diets
can result in adverse effects to human health as
they can cause various disorders of the body
function.
Some studies have been carried out on the levels
of major and trace metals of lemongrass by
different methods in different countries. For
instance, Anal (2014) has reported the metal
content in lemongrass in India by graphite
furnace-atomic absorption spectroscopy, Aftab et
al. (2011) studied the mineral content in
lemongrass in Pakistan by X-ray florescence
spectroscopy, Nkansah et al. (2016) has reported
metal content in lemongrass in Ghana by atomic
absorption spectrometry and Salami et al. (2007)
investigated mineral content in lemongrass in
Nigeria by atomic absorption spectrometry.
Recent studies have also been conducted on the
levels of essential and non-essential metals in some
spices and medicinal plants cultivated in Ethiopia
(Endalamaw and Chandravanshi, 2015; Dubale et
al., 2015; Hagos and Chandravanshi, 2016; Abdella
Nitsuh Birhanu et al.
et al., 2018; Admasu et al., 2018). In this study, MP-
AES was used because the technique is faster, cost-
efficient, sensitive, better linear dynamic range,
and has lower detection limit to sub ppb level than
conventional flame atomic absorption
spectroscopy (FAAS) (Vysetti et al., 2014;
Zamanian and Kashanaki, 2018; Ramos and
Lamorena, 2021). Unlike FAAS techniques, MP-AES
runs on nitrogen that can be extracted directly
from air instead of using hazardous and
flammable gases like nitrous oxide and acetylene
(Zamanian and Kashanaki, 2018).
Addis Ababa, Ankober, Gojjam and
Wondogenet are some of the areas in which
lemongrass is growing abundantly in Ethiopia.
However, the levels of the major and trace heavy
metals in the plant have not been studied yet in the
country. Lemongrass can help Ethiopian people to
balance the essential metals within the body when
it is used as folk medicine, flavoring agent in food
and beverage industries, herbal tea, etc. Therefore,
the main objective of the study was to detect the
presence and determine the levels of selected
major and trace metals, namely, K, Mg, Ca, Mn, Fe,
Cu, Zn, and Pb in lemongrass samples from the
different regions of the country by using simple,
sensitive and reliable method microwave plasma-
atomic emission spectroscopy (MP-AES). The
specific objectives of the study were: i) to develop
and optimize digestion procedure for the
determination of selected metals in lemongrass for
their subsequent determination by MP-AES, ii) to
compare the levels of the metals in lemongrass
from the sampling areas, iii) to compare the results
of the study with recommended values from WHO
and other reported values from the literature.
MATERIALS AND METHODS
Collection of samples
Lemongrass samples (1 kg from each site) were
collected by using a random sampling technique
from four different regions of Ethiopia (Addis
Ababa, Ankober, Gojjam, and Wondogenet). All
the samples collected from sampling areas were
stored in clean plastic bags. The selection of the
sample collection areas was based on more
accessibility for sampling. The geographical
locations (latitude, longitude, and elevation) of
sampling sites are given in Table 1.
SINET: Ethiop. J. Sci., 44(2), 2021
Chemicals, reagents, and standard solutions
For digestion of lemongrass samples, HNO3
(69.5%) (Sigma Aldrich Steinheim, Germany),
HClO4 (70%) (Sigma Aldrich Steinheim, Germany),
HCl (37%) (India) were used. Deionized water was
used for the preparation of standard solutions,
dilution and for cleaning purposes. 1000 mg/L
stock standard solutions of the metals (Sigma
Aldrich Steinheim, Germany) Ca, Cu, Zn, Mg, K,
Mn, Fe, and Pb were prepared for each element in
2% HNO3 and used for the preparation of
calibration curves for the determination of metals
in the samples.
Table 1. Geographical location, elevation of sampling
sites
Sample Latitude Longitude
Sites
Addis Ababa 9°01′29″ N 38°44′48″ E
Ankober 9° 35' 46N 39° 43' 57E
Gojjam 11°36′N 36°57′E 11°36′N 36°57′E
Wondogenet 7°1′N38°35′E 7°7°1′N38°35′E
Instrumentation and procedures
An electrical grinder was used for grinding
lemongrass leaves. Digital analytical balance
(Model 22ADAMT, Switzerland) with ±0.0001g
precision was used for weighing and plastic bags
for drying the samples at room temperature.
Kjeldahl apparatus (Gallenkamp, UK) was used to
digest the lemongrass samples, blank solution, and
spiked samples. A refrigerator was used to keep
the samples until analysis. Pipettes and
micropipettes (Dragonmed, Shanghai, China) were
used for measuring different amounts of acids and
standard solutions and for spiking of standard
solutions for the recovery test. 50 mL
volumetric flasks were used to dilute sample
solutions and prepare standard solutions. The
Agilent 4200 microwave plasma-atomic emission
spectrometer (MP-AES, USA) was used for the
analysis of the metals.
Lemongrass sample preparation for elemental
analysis
The lemongrass samples were washed first by tap
water and rinsed with distilled water to remove
earthy impurities. Then, lemongrass samples were
cut into small pieces by stainless scissors, allowed
to dry in air for ten days, chopped, and ground to
207
powder with an electrical grinder (Figure 1). Then
the powdered samples were sieved to 0.425 mm
mesh size and stored in dry and clean
polyethylene plastic bags until digestion. Finally,
0.5 g of the sample was taken from each sample for
digestion, and a solution for final metal
determination was prepared.
Optimization of digestion procedure
To select an optimum procedure for digestion,
parameters like, volume ratio of reagents,
digestion temperature, and digestion time were
optimized by varying one parameter at a time and
keeping the others constant. Parameters giving
clear and colorless solution at lower temperature,
at minimum reagent volume and digestion time
and maximum concentration of metals were
Altitude
(m) selected as an optimum procedure for digestion of
2405 lemongrass sample. Finally, the optimum
2896 procedure was chosen on the basis of these criteria
2072 requiring 3.0 mL of 69.5% HNO3, 2 mL of 37% HCl
1723
and 2.0 mL 70% HClO4 for complete digestion of
0.5 g lemongrass samples for a time of one and a
half hours at a temperature of 270 oC (Table 2).
Digestion of leaves of lemongrass samples
A 0.5 g of powdered lemongrass sample was
weighed and placed in a 250 mL round bottom
flask. Then, a total volume 7 mL of mixture of the
acids (3 mL HNO3, 2 mL HCl, and 2 mL of HClO4)
was added. The round bottom flask was fitted to a
reflux condenser and heated on a Kjeldahl
apparatus hot plate for one and half hours at a
temperature of 270 oC. The digested samples were
allowed to cool for 10 min without separating the
condenser and then further cooled to room
temperature for 10 min by separating the
condenser. The mixture was diluted with 20 mL of
deionized water and filtered with Whatman filter
paper No. 42 (Germany) into a 50 mL volumetric
flask. The round bottom flask was further rinsed
with 10 mL of deionized water and added to the
filtrate. Then, the flask was filled to the mark with
deionized water. For each sample the digestion
was done in triplicates. Triplicates of blank
samples were digested following the same
procedure as the samples. Finally, all the digests
were kept in the refrigerator until analysis using
MP-AES.
208 Nitsuh Birhanu et al.

a)
b)

Figure 1. Lemongrass plant a) and powdered lemongrass samples b).

Table 2. Digestion procedures for lemongrass samples.

Volume ratio of reagents Digestion temperature ( oC) Digestion time (h) Observation
3:1 (HNO3: HClO4) Yellow clear
3:2 (HNO3:HClO4) Yellow clear
2:1:1(HNO3:HCl:HClO4) 270 2 Light yellow clear
3:1:1(HNO3:HCl:HClO4) Light yellow clear
3:1:2(HNO3:HCl:HClO4) Very Light yellow clear
3:2:2(HNO3:HCl:HClO4) Colorless clear solution
180 Yellow clear solution
3:2:2(HNO3:HCl:HClO4) 210 2 Light yellow clear
240 Light yellow clear
270 Colorless clear solution
½ Yellow clear
3:2:2(HNO3:HCl:HClO4) 270 1 Light yellow clear
1:30 Colorless clear solution

Figure 2. Calibration curve for manganese determination.

SINET: Ethiop. J. Sci., 44(2), 2021 209

Calibration of the instrument and determination plate Kjeldahl apparatus by applying the
of the metals by MP-AES optimized digestion procedure used in sample
The 10 mg/L intermediate standard solutions of analysis. Then, the percentage recoveries of the
metals of interest were prepared from the standard analytes were calculated by using the equation:
stock solutions that contained 1000 mg/L. These (C M of spiked sample )− (C M of unspiked sample )
secondary standards were diluted with deionized %R= X 100
C M added for spiking
water to obtain working standards (5, 10, 15, 20
where CM is the concentration of metal and %R is
mg/L) of each metal, i.e., Ca, K, Mg, Mn, Fe, Cu,
the percent of the recovery. The results of the
Zn, and Pb. The Emission intensities of the
recovery analysis are presented in Table 3. The
working standard solutions were measured and
percentage recoveries lie within the range 86.9-106,
the calibration curves for each of the analyte
which is within the acceptable range for all the
metals were constructed. The calibration curve of
metals.
Mn has been shown as a representative example
(Figure 2). The concentrations of each metal in the
Table 3. Recovery test for lemongrass samples.
samples were determined by MP-AES after the
instrumental operating conditions were optimized Metal Conc. of Amount Conc. of % R ± SD
for the maximum signal intensity of the metal in spiked metal in
instrument. Triplicate determinations were carried unspiked (mg/L) spiked
out on each sample. The same analytical procedure sample sample
(mg/L) (mg/L)
was also employed for the determination of
K 5.100 2.5500 7.800 106 ± 0.5
elements in the digested blank solutions. Ca 0.647 0.3233 0.928 86.9 ± 0.7
Mg 0.181 0.0900 0.272 101 ± 0.1
Validation of the optimized procedure Mn 0.113 0.0570 0.167 94.7 ± 1.5
Fe 0.115 0.0580 0.174 102 ± 0.9
The results from a method validation procedure Cu 0.013 0.0063 0.019 95.2 ± 0.6
can be used to judge the quality, reliability, and Zn 0.00535 0.00268 0.008 98.9 ± 0.4
consistency of analytical results. Spiking Pb 0.00075 0.000375 0.00112 93.3 ± 0.5
experiments were performed to validate the
optimized procedure. For this purpose, standard
solutions of 1000 mg/L of the metals (Ca, K, Cu, RESULTS AND DISCUSSION
Fe, Mg, Mn, Zn and Pb) were used. For recovery
measurement, spiking was done for the metals Levels of metals in the leaves of lemongrass
whose concentrations were determined in samples
triplicates. Thus, 2550, 323, 6, 57, 56, 9, 0.13 and The MP-AES method was applied for the
0.02 µL of 1000 mg/L standard solution for K, Ca, determination of the levels of eight metals (Ca, K,
Cu, Mn, Fe, Mg, Zn and Pb, respectively, were Mg, Cu, Zn, Mn, Fe, and Pb) in lemongrass
spiked into round bottomed-flasks containing 0.5 g samples. Results obtained for each sample in terms
sample and 7 mL of acid mixture (3 mL HNO3, 2 of mg/kg with values for the mean, standard
mL HCl, and 2 mL of HClO4). Then, the round deviation, and ranges of metal concentration are
bottomed flasks containing the mixtures were shown in Table 4.
fitted with the condenser and digested
simultaneously with un-spiked samples on a hot

Table 4. Metal concentration in different lemongrass samples (mean ± SD mg/kg).

Metal Sample sites, conc. (mean ± SD) in mg/kg Range of metal conc.
Addis Ababa Ankober Gojjam Wondogenet (mg/kg)

K 913.5 ± 0.85 1020± 0.2 743.8 ± 0.5 879.3 ± 1.4 743.8–1020

Ca 125.1 ± 0.1 129.3 ± 0.5 123.1 ± 0.1 126.3 ± 0.9 123.1 –129.3
Mg 26.0 ± 1.8 36.3 ± 2.7 23.9 ± 1.7 33.0 ± 3.4 23.9 –36.3
Fe 18.5 ± 0.7 22.3 ± 1.2 18.5 ± 0.9 10.35 ± 1.4 10.35–22.3
Mn 12.6 ± 4.7 12.7 ± 4.3 10.0 ± 1.3 12.5 ± 2.8 10.0–12.7
Cu 2.35 ± 1.2 2.5 ± 0.8 1.48 ± 1.3 1.94 ± 1.9 1.48 –2.5
Zn 0.69 ± 0.5 1.07 ± 0.05 0.6 ± 0.1 0.59 ± 1.0 0.59–1.07
Pb 0.15 ± 0.3 0.20 ± 0.1 0.13 ± 0.4 0.17 ± 0.3 0.13–0.20
210
As indicated in Table 4, the metals Ca, K, Cu, Fe,
Mn. Mg, Zn and Pb are found in all the samples of
lemongrass. A relatively higher amount of K in the
sample from all areas could be due to its higher
natural abundance in the soil and the nature of the
lemongrass. This result is in good agreement with
the report from Godwin and his coworkers
(Godwin, 2002).
In comparison to other sampling areas, Ankober
is more exposed to human activity and the farms
may have been used for a long time which could
have a significant contribution to a gradual
accumulation of metals in lemongrass farms
through agricultural activities. The result showed
that the metal contents of lemongrass varied with
the geographical origin in which the lemongrass
grows. The types of soil, climatic condition,
humidity, the natural weathering of rocks,
agricultural activities could contribute to the
different concentrations of metals determined in
lemongrass.
Figure 3. The concentration of metals in lemongrass from different sample sites
Nitsuh Birhanu et al.
Concentration of metals in lemongrass collected
from different sample sites
The concentrations of metals in lemongrass
collected from different sample sites are given as
follows: In Addis Ababa, K is found in the highest
amount in the samples with concentration of 913.5
mg/kg followed by Ca (125.1 mg/kg), Mg (26.0
mg/kg), Fe (18.5 mg/kg), Mn (12.7 mg/kg), Cu
(2.35 mg/kg), Zn (0.69 mg/kg) and Pb (0.15
mg/kg). In Ankober area, K (1020 mg/kg), Ca
(129.5 mg/kg), Mg (36.3 mg/kg), Fe (22.3 mg/kg),
Mn (12.7 mg/kg), Cu (2.53 mg/kg), Zn (1.1
mg/kg) and Pb (0.20 mg/kg). In Gojjam, K (743.7
mg/kg), Ca (123.1 mg/kg), Mg (23.9 mg/kg), Fe
(18.5 mg/kg), Mn (10.0), Cu (1.48 mg/kg), Zn (0.6
mg/kg) and Pb (0.13 mg/kg) and in Wondogenet,
K (879.4 mg/kg), Ca (126.3 mg/kg), Mg (33.0
mg/kg), Fe (10.35 mg/kg), Mn (12.5), Cu (1.94
mg/kg), Zn (0.59 mg/kg) and Pb (0.17 mg/kg).
The comparison of metals in all sample sites is
indicated in Figure 3.
SINET: Ethiop. J. Sci., 44(2), 2021 211

In general, it can be summarized that the
lemongrass sample from the Ankober area
contains the highest amount of the metals studied
while the lowest concentration is in the Gojjam
sampling area (Figure 3). This could be attributed
to the fact that the sample taken from Gojjam is in
a very rural area with no industrialization and
thus fewer anthropogenic sources for the metals in
this sample site.
Analysis of variance (ANOVA)
In this study, the variation in sample means of
the analytes was tested whether they have
significant difference or not by using one way
ANOVA at 95% confidence level. Accordingly, there
is no significant difference among the sample
means for Ca and Mn. While there is significant
difference among the sample means for K, Mg, Fe,
and Cu, Zn, and Pb (Table 5). This may be due to
variations in factors other than the experimental
procedures. The source for this significant
difference between sample means could be
possibly attributed to the differences in soil
composition, usage of different fertilizers,
pesticides and industrial activities.
the correlation coefficient equal to zero (Bhan et al.,
2005).
In this study, to correlate the effect of one metal
concentration on the concentration of other metals,
Pearson correlation matrices for the samples were
applied for the metals determined in the samples
(Table 6). The results of the Pearson correlation
show positive correlations in the case of K, Ca. Mg,
Fe, Mn, Cu, Zn and Pb with each other. Besides,
the metal K with (Fe and Mn), Ca with (Mn and
Cu), Mg with (Fe, Mn with K, Ca, Cu, Zn, and Pb),
Fe with (K, Ca, Mg, Cu and Pd), Cu with (Ca, Fe,
Zn, and Pb), Zn with (Mn and Cu) and Pb with (Fe,
Mn, and Cu) were observed to have a positive
weak correlation.
The weak correlations indicate that one metal
affecting the other metal to a lesser extent and
strong correlations may arise from common
natural or anthropogenic sources as well as from
similarity in chemical properties. A good positive
correlation is observed between K and Ca and also
between Cu and Mn. In contrast, a weak
correlation is observed for Mn with K and Pb.
Table 6. Pearson correlation matrix for metals in leaves
of lemongrass samples.

Parameters K Ca Mg Fe Mn Cu Zn Pb
Table 5. Analysis of variance (ANOVA) between and K 1.00
Ca 0.75 1.00
within lemongrass samples at 95% confidence
Mg 0.51 0.69 1.00
level Fe 0.27 0.26 0.13 1.00
Metal K Ca Mg Fe Mn Cu Zn Pb Mn 0.04 0.27 0.52 0.63 1.00
Fcal 10.5 2.10 11.2 6.40 0.25 4.50 3984 5.30 Cu 0.59 0.35 0.55 0.43 0.76 1.00
Fcrit 5.27 3.29 5.10 5.64 0.35 1.23 3816 4.29 Zn 0.66 0.61 0.55 0.70 0.14 0.42 1.00
Pb 0.66 0.71 0.67 0.31 0.028 0.34 0.65 1.00

Comparison of the concentrations of the metals in

Pearson correlation of metals within lemongrass
samples
A Pearson correlation is a number between -1
and 1 that indicates the extent to which two
variables are linearly related. If the correlation
coefficient approaches to positive one there is a
strong positive relationship among the two
variables. If the correlation coefficient is -1, the two
variables have a strong negative relationship. If
there is no relationship between the two variables
lemongrass sample obtained in the present study
with literature and acceptable range of medicinal
plant by WHO
The findings of the present study were
compared to that of other authors from different
countries and an acceptable range of medicinal
plant by WHO. The comparative levels of metals in
the studied samples with the levels reported in the
literature are indicated in Table 7.
212 Nitsuh Birhanu et al.

Table 7. Comparison of metal concentrations of lemongrass samples of this study with reported literature and an
acceptable level in medicinal plants by WHO.

Metal Conc. of metals in lemongrass sample (mg/kg) Conc. of metals

in medicinal
plant (mg/kg).
Ethiopia Pakistan (XRF) Indian (AAS) WHO
K 743.8–1020 723 - -
Ca 123.1–129.3 65 5.2-6.0 -
Mg 23.9–36.3 60 0.76-0.79 2000
Fe 10.35–22.3 8.7 1.5-1.98 261-1239
Mn 10.0–12.7 5.2 0.19-0.27 200
Cu 1.48–2.5 0.266 0.05-0.07 20-150
Zn 0.59–1.07 2.23 0.11-0.12 50
Pb 0.13–0.20 - 0.14 10
Reference This study (Aftab et al., (Anal et al., (WHO,1998;
2011) 2014) Nkuba and
Mohammed, 2017)

As shown in Table 7, the concentrations of all the
above listed metals are greater in Ethiopia when
compared to India and in good agreement with the
results from Pakistan. The variations observed
might be due to differences in soil fertility, climatic
conditions, usage of fertilizers, etc.
The level of most of the metals found in the
lemongrass samples collected from four selected
areas (Addis Ababa, Ankober, Gojam, and
Wondogenet) were found comparable to the
maximum tolerable limits recommended by WHO.
CONCLUSION
metals. The levels of metals in the samples were
found to be comparable with the results reported
from different countries and within the permissible
limits of WHO.
ACKNOWLEDGEMENTS
The authors would like to acknowledge the Department
of Chemistry, Addis Ababa University, Ethiopia for
providing laboratory facilities. Nitsuh Birhanu is also
grateful to the graduate female scholarship program of
Addis Ababa University for sponsoring her study.

The determination of the level of the metals K, Ca, REFERENCES

Mg, Fe, Mn, Cu, Zn, and Pb in Cymbopogon citratus
samples was carried out by microwave plasma-
atomic emission spectrometry. The result was
evaluated through the recovery test and good
percentage recovery of 86.9 to 106% was obtained
for all the metals identified. The concentration of
the metals in the samples can be expressed in the
following order: K > Ca > Mg > Fe > Mn > Cu > Zn
> Pb.
Statistical analysis by using one-way ANOVA at a
95% confidence level indicated that there were no
significant differences among the sample mean
concentrations for the metals Ca and Mn while
there were significant differences in the mean
concentration of K, Mg, Fe, Cu, Zn, and Pb in the
samples. This could be attributed to the differences
in soil composition, usage of different fertilizers,
pesticides, and industrial activities. The results of
the Pearson correlation showed that there was a
weak and strong positive correlation between the
1. Aftab, K., Ali, M. D., Aijaz, P., Beena, N., Gulzar, H. J.,
Sheikh, K, Sofi, Q. and Tahir A. S. (2011).
Determination of different trace and essential
element in lemongrass samples by x-ray
florescence spectroscopy technique.
International Food Research Journal 1 (8):265-270.
2. Ahmad, A. and Viljoen, A. (2015). The in vitro
antimicrobial activity of Cymbopogon essential
oil (lemon grass) and its interaction with silver
ions. Phytomedicine 22(6):657-665.
3. Ahmed Abdella, Chandravanshi, B. S. and
Weldegebriel Yohannes (2018). Levels of
selected metals in coriander (Coriandrum
sativum L.) leaves cultivated in four different
areas of Ethiopia. Chemistry International
4(3):189-197.
4. Ajayi, E. O., Sadimenko, A. P. and Afolayan, A. J.
(2016). GC–MS evaluation of Cymbopogon
citratus (DC) Stapf oil obtained using modified
hydrodistillation and microwave extraction
methods. Food Chemistry 209:262-266.
SINET: Ethiop. J. Sci., 44(2), 2021
5. Amare Aregahegn, Chandravanshi, B. S. and Kidane
Fanta Gebremariam (2015). Levels of major and
trace metals in the leaves and infusions of
Croton Macrostachyus. Bulletin of the Chemical
Society of Ethiopia 29(1):11-26.
6. Anal, J. M. (2014). Trace and essential elements
analysis in Cymbopogon citratus (DC.) stapf
samples by graphite furnace-atomic absorption
spectroscopy and its health concern. Journal of
Toxicology 2014:1-5.
7. Avoseh, O., Oyedeji, O., Rungqu, P., Nkeh-Chungag,
B. and Oyedeji, A. (2015). Cymbopogon species;
ethnopharmacology, phytochemistry and the
pharmacological importance. Molecules
20(5):7438-7453.
8. Babarinde, A., Ogundipe, K., Sangosanya, K. T.,
Akintola, B. D. and Hassan, A. O. E. (2016).
Comparative study on the biosorption of Pb(II),
Cd(II) and Zn(II) using Lemon grass
(Cymbopogon citratus): kinetics, isotherms and
thermodynamics. Chemistry International
2(2):89-102.
9. Balakrishnan, B., Paramasivam, S. and Arulkumar, A.
(2014). Evaluation of the lemongrass plant
(Cymbopogon citratus) extracted in different
solvents for antioxidant and antibacterial
activity against human pathogens. Asian Pacific
Journal of Tropical Disease 4(1):S134-S139.
10. Bleasel, N., Tate, B. and Rademaker, M. (2002).
Allergic contact dermatitis following exposure
to essential oils. Australasian Journal of
Dermatology 43(3):211-213.
11. Devi, R. K. and Sarma, H. N. (2013). Profile of trace
elements in selected medicinal plants of North
East India. IOSR Journal of Applied Physics 4:47-
51.
12. Dghaim, R., Al Khatib, S., Rasool, H., and Ali Khan,
M. (2015). Determination of heavy metals
concentration in traditional herbs commonly
consumed in the United Arab Emirates. Journal
of Environmental and Public Health 2015:1-6.
13. Ekpenyong, C. E., Akpan, E. and Nyoh, A. (2015).
Ethnopharmacology, phytochemistry, and
biological activities of Cymbopogon citratus (DC.)
Stapf extracts. Chinese Journal of Natural
Medicines 13(5):321-337.
14. Ekpenyong, C. E., Akpan, E. E. and Daniel, N. E.
(2014). Phytochemical constituents, therapeutic
applications and toxicological profile of
Cymbopogon citratus Stapf (DC) leaf extract.
Journal of Pharmcognosy and Photochemistry 2
(1):133-141.
15. Falkenberg, T., Sawyer, J. and Zhang, X. (2002).
WHO traditional medicine strategy 2002–2005.
WHO, Geneva, Switzerland.
213
16. Feleke Demissie Endalamaw and Chandravanshi, B.
S. (2015). Levels of major and trace elements in
fennel (Foeniculum vulgari Mill.) fruits
cultivated in Ethiopia. SpringerPlus 4(1):1-10.
17. Francisco, V.; Figueirinha, A.; Neves, B. M.;
GarcíaRodríguez, C.; Lopes, M. C.; Cruz, M. T.
and Batista T. (2011). Cymbopogon citratus as
source of new and safe anti-inflammatory
drugs: bio-guided assay using
lipopolysaccharide-stimulated macrophages.
Journal of Ethno pharmacology 13 (3):818-827.
18. He, F. J. and MacGregor, G. A. (2008). Beneficial
effects of potassium on human health.
Physiologia Plantarum 133(4):725-735.
19. Manvitha, K. and Bidya, B. (2014) Review on
pharmacological activity of Cymbopogon
citratus. International Journal of Herbal Medicine
1(6):5-7.
20. Mebrahtu Hagos and Chandravanshi, B. S. (2016).
Levels of essential and toxic metals in
fenugreek seeds (Trigonella Foenum-Graecum L.)
cultivated in different parts of Ethiopia.
Brazilian Journal of Food Technology 19: 1-12.
21. Moomaw, A. S., Maguire, M. E. (2008). The unique
nature of Mg2+ channels. Physiology 23(5):275-
285.
22. Nkansah, M. A., Hayford, S. T., Borquaye, L. S.,
Ephraim, J. H. (2016). Heavy metal contents of
some medicinal herbs from Kumasi, Ghana.
Cogent Environmental Science 2(1):1-8.
23. Nkuba, L.L. and Mohammed, N. K. (2017). Heavy
metals and essential elements in selected
medicinal plants commonly used for medicine
in Tanzania. Chemical Science International
Journal 19(2):1-11.
24. Ramos, N. C. and Lamorena, R. B. (2021). Detection
of copper, cadmium, manganese, lead, and zinc
content in milled rice using microwave plasma
atomic emission spectroscopy. Philippine Journal
of Science, 150(4):765-776.
25. Salami, S. J., Akande, E. A. and Zachariah, D. M.
(2007). Level of heavy metals in soils and lemon
grass in Jos, Bukuru and environs, Nigeria.
Global Journal of Pure and Applied Sciences
13(2):193-196.
26. Shah, G.; Shri, R.; Panchal, V.; Sharma, N.; Singh, B.
and Mann, A. S. (2011). Scientific basis for the
therapeutic use of Cymbopogon citratus, Stapf
(Lemon grass). Journal of Advanced
Pharmaceutical Technology and Research 2 (2):3 –
8.
27. Silva, F. D., Santos, R. H. S., Andrade, N. J. D.,
Barbosa, L. C. A., Casali, V. W. D., Lima, R. R.
D. and Passarinho, R. V. D. M. (2005). Basil
conservation affected by cropping season,
214
harvest time and storage period. Pesquisa
Agropecuária Brasileira 40(4):323-328.
28. Singh, R. (2015). Medicinal plants review. Journal of
Plant Sciences, 3 (1):50-55.
29. Solomon Admasu, Chandravanshi, B. S. and
Weldegebriel Yohannes (2018). Levels of
selected metals in the leaves of Ruta chalepensis
L. (Rue) collected from four different areas of
Ethiopia. Bulletin of the Chemical Society of
Ethiopia 32(2):185-197.
30. Thorat, P. P.; Sawate, A. R.; Pati, B. M. and
Kshirsagar, R. B. (2017). Effect of lemongrass
powder on proximate and phytochemical
content of herbal cookies. Journal of
Pharmacognosy and Photochemistry 6(6):155-159.
31. Vysetti, B., Vummiti, D., Roy, P., Taylor, C., Kamala,
C. T., Satyanarayanan, M. and Abburi, K.
(2014). Analysis of geochemical samples by
Nitsuh Birhanu et al.
Microwave Plasma-AES. Atomic Spectroscopy,
35(2):65-78.
32. Weldegebriel Yohannes, Chandravanshi, B. S. and
Ghirma Moges (2018). Assessment of trace
metals and physicochemical parameters of
commercially available honey in Ethiopia.
Chemistry International 4(2):91-101.
33. Wilson, K.; Asiyanbi, H. T.; Hammed, A. M. and
Simsek, S. (2012). Biological properties of
lemongrass. Journal International Food Research
21(2):455-462.
34. World Health Organization (1998). Quality Control
Methods for Medicinal Plants Materials. WHO,
Geneva, Switzerland.
35. Zamanian, F. and Kashanaki, R. (2018). Microwave
Plasma Atomic emission Spectroscopy (MP-
AES). Iranian Journal of Laboratory Knowledge 6:7-
14.
SINET:Ethiop. J. Sci., 44(2): 215–222, 2021 ISSN: 0379–2897 (PRINT)
© College of Natural and Computational Sciences, Addis Ababa University, 2021 eISSN: 2520–7997

Date received: March 03, 2021; Date revised: August 08, 2021; Date accepted: September 19, 2021
DOI: https://dx.doi.org/10.4314/sinet.v44i2.8

Early hunters and herders of northern Ethiopia: The fauna from Danei Kawlos

Agazi Negash 1, * and Fiona Marshall 2

1 Paleoanthropology and Paleoenvironment Program, Addis Ababa University, Ethiopia. E-mail:

agazi.negash@aau.edu.et
2 Department of Anthropology, Washington University in St Louis, St Louis, USA

ABSTRACT: It is generally agreed that Ethiopia is one of the world’s primary centers of
prehistoric plant domestication. It is also known that domestic fauna (cattle and domestic
caprines) were brought in from outside. Unfortunately, very few Holocene archaeological
sequences have been excavated in the Horn. Even fewer sites have yielded domestic fauna
dating to > 3000 years ago. The excavations at the site Danei Kawlos in northern Ethiopia
provide new Holocene archaeological sequences for Northern Ethiopia and document the
presence of cattle, sheep and goat with a direct date of 3358 ± 47 BP on a Bos molar. We discuss
here the zooarchaeological data from the site.

Key words /phrases: Danei Kawlos; domestication, fauna, Tigrai, Ethiopia

Arabian Peninsula (Lesur et al., 2014a). This

INTRODUCTION
It is generally agreed that Ethiopia is one of the
world’s primary center of prehistoric plant
domestication (e.g., Vavilov, 1951). Indigenous
Ethiopian plant domesticates include, among
others, teff (Eragrostis tef), noog (Guizotia
abyssinica),finger millet (Eleucine coracana),
enset(Ensete ventricosum), chat (Catha edulis) and
coffee (Coffea arabica). These play an important
role in contemporary Ethiopia’s food
producing systems (Westphal, 1975), and
without doubt, these must have played an
important role in prehistory.
Domestic cattle and sheep and goats have
also played an important role in pastoral and
farming systems over the long term, but were
brought to the region from northeastern Africa
or Arabia. However, we know very little about
the processes involved, and the relative timing
of the appearance of domestic livestock. It has
generally been assumed that cattle herders
moved into Eritrea and Ethiopia from northern
Sudan, where they first appeared 5000 years
ago (see review in Edwards [2007] and Lesur et
al., [2014a]). However given the short distance
from the coast of Africa to Yemen, it is also
possible that cattle and sheep and goats were
introduced to the Horn of Africa from the
_____________________
*Author to whom correspondence should be addressed.
could have occurred before 5000 BP.
Unfortunately, very few Holocene
archaeological sequences have been excavated
in the Horn (Fig.1). Even fewer sites have
yielded domestic fauna dating to > 3000 years
ago. The excavations at Danei Kawlos near
Abiy Adi in Tigrai provide new Holocene
archaeological sequences for Northern
Ethiopia. We report here on zooarchaeological
data from the site.
Brief Background
Danei Kawlos is a rock shelter located in
northern Ethiopia near the modern town of
Abiy Adi, a few km from the rock art bearing
rock shelters of Mihdar Abur and Tselim
Ba'ati. It measures 14 x 8 meters. Five 1 x 1
meter units were excavated, 3 to bedrock, circa
145cm below surface. The substrate was sandy
throughout the sequence. Five Cultural
Stratigraphic Units were discerned. The site
was dug in 5cm pits and dry sieved though
2mm mesh. This allowed excellent recovery of
small faunal specimens.
Given the sandy nature of the substrate and
the general issue of bioturbation in rock shelter
deposits, obtaining direct dates was a priority
of our research. At Danei Kawlos it was
possible to obtain a direct apatite date on a Bos
216 Agazi Negash and Fiona Marshall

lower third premolar from 130-135 cm below fits with this at 3380 ± 160 (GX-27745). An AMS
surface in stratigraphic Layer 5. This date is date on a small charcoal sample came out at
3358 ± 47 (A-0214). A conventional charcoal 1,103 ± 41 and is almost certainly intrusive.
date on this layer (95-100 cm below surface)

Figure 1:-Map showing modern towns and archaeological sites mentioned in the text and represented here in numbers
(1=Dabo Zellelew; 2= Mihdar Abur; 3=Tselim Ba'ati; 4=Danei Kawlos; 5=Ba'ati Ataro; 6=Gobedra; 7=Quiha;
8=Mezbir; 9=Lalibela/Natchebiet; 10=Lake Beseka; 11=Koka/Kumali; 12=Mochena Borago;
13=Mota/Tuwatey; 14=Mome Gongolo 1 and 2; 15=Yabello; 16=Laga Oda; 17=Goda Buticha).

Specimens were initially sorted by body part

METHODS
Following methods developed for the study of
African faunas (Gifford et al., 1980; Marshall
1990), all identifiable specimens were
individually bagged, labeled, sorted, and
identified. All such specimens were given
zooarchaeological catalog numbers. Minimally
identifiable specimens, such as rib shaft
fragments, cranial, axial, or long bone shaft
fragments were left bagged in groups by
provenience category under one catalog
number. The same procedure was followed for
non-identifiable specimens.
and subsequently by taxon. All identifications
were made by reference to comparative
material housed in the National Museum in
Addis Ababa or, for those specimens that were
not available, by reference to identification
manuals such as Walker (1985) and Schmid
(1972). Levels of identification used vary from
class, Mammalia, orders such as Artiodactyla
or Carnivora, genera such as bovid or suid to
more specific tribal or species level
distinctions. In some cases, as is common in
African zooarchaeology, size classes were also
used. In general, because of lack of
comparative material, identifications were
cautious.
SINET: Ethiop. J. Sci., 44(2), 2021 217

We recorded the presence of cut marks, RESULTS

gnaw marks, burning, root action or
weathering for individual specimens. Detailed A total of 1697 faunal specimens were studied
study of the location of these marks was not (Table 1; Figs 2 and 3) with cattle and caprines
undertaken however. Measurements were present in all stratigraphic layers. The fauna is
taken on all complete specimens and on all well-preserved but highly fragmentary. The
complete long bone ends, following the bulk of the specimens occur in the lower two
recommendations of von den Driesch (1976). stratigraphic layers (layers 4 and 5). This fauna
All specimens were counted. Number of is associated with obsidian and chert lithics
identifiable specimen or NISP is the unit of with a high proportion of scrapers and backed
quantification used here. The sample sizes for pieces. Ceramics in Layer 4 and 5 were
any one unit were not large enough to warrant burnished and slipped. Most of the decorated
calculation of minimum numbers of sherds, with closely spaced, parallel or zigzag
individuals (MNI). line motifs, occurred in level 5.

Table 1. Species Representation at Kawlos.

well
Porcupine

CSU Total
cf. Rodent

identified
Rodents
Caprine

Mollusc
Primate
Human
Reptile
cf. Fish
Hyrax
Cattle

Hare

Suid

Not
Stratigraphic layer 1 3 1 1 5 76 86
Stratigraphic layer 2 2 1 1 1 4 181 190
Stratigraphic layer 3 1 1 3 8 1 124 138
Stratigraphic layer 4 2 1 4 1 37 3 1 295 344
Stratigraphic layer 5 4 1 8 4 4 1 2 116 27 1 1 10 760 939
Total 12 5 14 5 4 1 4 2 170 31 1 1 11 1436 1697

Nine hundred and thirty nine faunal primate as well as one fish specimen were
specimens were studied from Layer 5, many found. A small number of mollusks (10/179 or
were identifiable to size class, but only 179 or 5.6%) are also present.
19% of the assemblage were identifiable to Three hundred and forty-four specimens
tribe or genus (Table 1; Figs. 2 and 3). This were studied from Layer 4, 49 or 14.2% of
sample is dominated by rodents (143/179 or which were identifiable beyond size class. Just
79.9%), including porcupine some with cut as in unit five, rodents dominate (40/49 or
marks. Hyrax are also present in small but about 81.6%), hyrax (4/49 or 8.1%) and hare
significant numbers (8/179 or 4.5%). Even (1/49 or 2.1%) are present, as are cattle (2/49
smaller numbers of hare (4 or 2.2%) and wild or 4%) and caprines (1/49 or 2.1%).
suid (4 or 2.2%) are present. Domestic cattle (4 The samples from Stratigraphic Layers 3, 2,
or 2.2%) and possibly caprines (1 or 0.5%) are and 1 are small, but the taxonomic
represented. One human specimen and one representation is very similar.
218 Agazi Negash and Fiona Marshall

Figure 2:- Selected Species Representation at Danei Kawlos (numbers at the bottom of each column represent cultural
stratigraphic layers).

100%

90%

80%

70%

60%
Size 4
50% Size 2
Size 1/2
40%

30%

20%

10%

0%
1 2 3 4 5

Figure 3:- Size Class Representation at Danei Kawlos (numbers at the bottom of each column represent cultural
stratigraphic layers).
SINET: Ethiop. J. Sci., 44(2), 2021
DISCUSSION
The faunas from Danei Kawlos are
important as the first sizable excavated and
systematically analyzed faunal assemblages
from long Holocene sequences in northern
Ethiopia and provided the earliest direct date
for the presence of cattle in the region. These
findings have interesting implications for the
timing and paleoclimatic context of the
appearance of early herding in northern
Ethiopia.
The inhabitants of Danei Kawlos relied quite
heavily on small wild animals. The lack of
large wild mammals was unexpected and
differs in patterns of the use of wild fauna
from sites dating to similar periods in
Southern Ethiopia at Yabello (Girma 2001), in
Djibouti (Lesur et al., 2014a), in the Sudan
(Fattovich et al., 1984) and in Kenya (Marshall
2000). This may relate to the intensity of land
use in the Abiy Adi area, however, little is
known about this at this time.
Danei Kawlos is situated very close to the
sites of Mihdar Abur, Tselim Ba'ati and Dabo
Zellelew. Of these Dabo Zellelew, an open air
site, yielded lithic artifacts similar to the lowest
layers of Danei Kawlos, but dates for this site
are lacking. The sites of Mihdar Abur and
Tselim Ba'ati are rock shelters but have rocky
surfaces and cannot be excavated. However,
they do contain rock art that mostly represents
the earliest stage of rock art tradition of the
Horn (Agazi 1997a, b; 2001) that is
hypothesized to date to the mid-Holocene
(Brandt and Carder 1987).
It is, therefore, possible that use of the
landscape was already quite intensive by the
time that the site was occupied and that large
mammals were scarce for this reason. This line
of reasoning fits with the presence of domestic
stock at the site and the increasing dependence
on small mammals. The presence of a wide
spectrum of small wild animals is intriguing
and it could also be that large wild animals
were scarce due to aridity. There is nothing in
the sediments or artifacts to suggest high
rainfall, and the fauna would be compatible
with fairly arid conditions. In the absence of
information on paleoclimate, or settlement
patterns, it is difficult to fully understand the
significance of the reliance on small wild
219
mammals. Paleoclimatic reconstructions for
Ethiopia by Gasse and others (Gasse 1977;
Gasse and Street 1978), based on diatoms,
isotopes, pollen, and levels of Rift Valley lakes
show gradually increasing aridity from circa
6000BP, with many oscillations. A particularly
arid phase occurs circa 3,500BP. The dates from
Stratigraphic Unit 5, Danei Kawlos suggest
occupation around 3,300BP, close to this arid
phase.
The presence of domestic cattle from the
base of the sequence at Danei Kawlos is also
intriguing. The direct date of 3358± 47 BP on a
Bos lower third premolar from 130-135cm
below surface in Stratigraphic Layer 5,
together with the charcoal date for this level of
3380 ± 160 BP provide one of the first secure
contexts for the presence of early domestic
stock in the Horn of Africa. These data provide
a setting against which to view older dates for
early domestic stock in the region.
The dates from Danei Kawlos are early and
especially secure due to direct dating but fit
with the estimated range of dates for domestic
cattle or sheep and goat or chicken from other
sites with sequences that range between 4000-
3000 BP. Also located in the northern Ethiopian
highlands (Fig. 1), Gobedra stratum 2A,
preserves a probable domestic cow, with a
bone collagen date of a little younger than 3000
BP (Phillipson 1977). At the site of Mezbir,
domestic chicken bones have been directly AMS
dated to 819-755 BC or around 3000 BP
(Woldekiros and D'Andrea 2017). At Lake
Beseka, in the Rift Valley in central Ethiopia,
cattle are thought to occur at FEJX 3 c. 3500 BP.
There is no direct date on these remains but
they are associated with aridity and low lake
levels characteristic of the period (Brandt
1982). Similar dates for cattle are found at the
site of Laga Oda in eastern Ethiopia (Clark and
Prince 1978). At Yabello, in southern Ethiopia,
cattle remains have been found in levels
thought to date to 4100 BP. Stratigraphic
interpretation was complex at this site,
however, and direct dates are lacking (Girma
2001).
Other Ethiopian archaeological sites did not
yield domestic animals or have very late dates
for domestic faunas. The absence of domestic
fauna in the Holocene layers of Mochena
220
Borago is striking, as noted by Lesur (Lesur et
al., 2007) as well as Goda Buticha (Zelalem et
al., 2014), and Mome Gongolo 1 and 2 (Lesur et
al., 2014b). Similarly, there is no report of the
recovery of fauna in the contemporaneous
layers of Porc Epic (Clark and Williamson
1984) in eastern Ethiopia or Haroruna
(Bachechi 2005) in south central Ethiopia. At
Lalibela and Natchebiet caves in northwestern
Ethiopia legumes and chickpeas in association
with sheep or goat and possible domesticated
cattle are dated to no earlier than 2500 BP
(Dombrowski, 1970, 1971). Recent
investigations at the sites of Koka and Kumali
in Keffa have yielded domestic fauna dating to
around 2000 BP (Hildebrand et al., 2010).
Contemporaneous domestic animals have
been recovered also from Akirsa (Poisbaud
2002). The domestic cattle and sheep or goat at
Mota cave in the Gamo highlands date to only
around 1700-1600 years ago (Arthur et al.,
2019). These recent dates for domestic stock
revealed by recent research in much of
Ethiopia contrast with the presence of rock art
containing mainly depictions of cattle (see
Agazi [1997c, 2018, 2020]) in the region that is
hypothesized to date to at least the mid-
Holocene (Brandt and Carder, 1987).
Archaeological evidence in neighboring
Sudan provides indirect but important
information bearing on the introduction of
cattle and the Neolithic of northern Ethiopia.
Fieldwork in eastern Sudan has yielded
evidence for the "herding" of cattle and sheep
and goats by the 6th-5th millennium BP
(Fattovich, 1988), followed by the possible
addition of sorghum cultivation at Gash
Group sites by ca. 5000 to 3500 BP (Fattovich et
al., 1984; Sadr 1991). Contact between eastern
Sudan and the highlands of northern Ethiopia
is inferred from the ceramics and obsidian
artifacts recovered from Kassala Phase sites in
Eastern Sudan dating from the middle of the
6th to the end of the 4th millennium BP.
Indeed, Fattovich and his colleagues go on to
state that "by the 2nd millennium BC there was
considerable contact between the eastern area
of the Atbai Tradition and Ethiopia (Fattovich
et al., 1984:185).
A Sudanese origin for early cattle and sheep
and goats in northern Ethiopia moving with
people or via exchange and trade contacts
Agazi Negash and Fiona Marshall
from the Khartoum Nile and sites like Kadero
and then via the sites of the Butana phase of
the southern Atbai, therefore, seems probable.
This hypothesis would be supported by dates
for the earliest cattle or sheep and goat in
Ethiopian sites falling within the last 4500
years. But if the dates for the earliest domestic
stock date to 5000 BP or older, then we would
have to consider an Arabian or Near-Eastern
origin for domestic stock in the Horn of Africa.
It is ultimately likely that both eastern and
western routes and diverse small scale social
processes played a role in introduction of cattle
and sheep and goat to northern Ethiopia but
more fine-grained data and direct dates will be
needed to parse relationships between pastoral
influences, trade and exchange and the reach
of ancient states on specific places and periods.
CONCLUSION
Taken together the fauna from sites in the
Horn of Africa now suggests a relatively late
appearance of cattle, in the fourth millennium
BP, and possibly slow introduction of domestic
stock to Horn of Africa. Disease challenges for
cattle outside of their wild range posed by the
epizootical threats of wild herbivores and the
wildlife-livestock disease interface may have
slowed the spread of cattle in the highland
Horn just as suggested for much of Africa by
Gifford-Gonzalez (2000, 2017). The elevational
extremes and woodlands of northern Ethiopia
may also have resulted in a slow spread of
early herders to the Horn (Marshall and
Hildebrand 2002). Throughout Africa, early
herders moved preferentially into high
productivity grasslands below 3500 masl
(Marshall and Hildebrand 2002; Marshall et al.,
2018). Moreover, ecological stress related to
climate may have compounded herder choices
as livestock appears in the Horn of Africa
4000-3500 BP during a particularly arid period
when grazing, though better than the
lowlands, would have been difficult. It is also
possible that in the south, resilient hunter-
gatherer societies who were geographically
isolated from the Nile and Red Sea networks of
trade and exchange provided social resistance
to the spread of herding ways of life.
SINET: Ethiop. J. Sci., 44(2), 2021
Dates ranging between 4000 and 3500 BP are
consistent with a Sudanese origin from the
Khartoum Nile and sites like Kadero, via the
sites of the Butana phase of the southern Atbai.
There may have been repeated small pulses of
movement of cattle and people into the Horn
during the 4thmillennium BP, perhaps in
response to drought conditions in the
lowlands. There is currently no evidence for a
South Arabian origin for the cattle of the Horn.
More well preserved and carefully studied
archaeological sequences, as well as direct
dates on early domestic stock, are needed to
test the hypotheses of late and African origins
for domestic cattle in the Horn.
ACKNOWLEDGEMENTS
Thanks are due to the Authority for Research and
Conservation of Cultural Heritage, the government
body that oversees archaeological research in the
country, for permits. Funding for research was
provided by the Wenner Gren and L. S. B. Leakey
Foundations. We also thank the anonymous
reviewers for helpful comments.
REFERENCES
1. Agazi Negash (1997a). Preliminary results of an
archaeological reconnaissance in Tigray,
Northern Ethiopia. Nyame Akuma 47:27-32.
2. Agazi Negash (1997b). Temben’s place in the
Neolithic of northern Ethiopia. In K. Fukui
et al. (Eds.), Ethiopia in broader perspective:
Papers of the 13th international conference of
Ethiopian studies (Vol. 1, pp. 389–398).
Kyoto: Shokado Book Sellers.
3. Agazi Negash (1997c). Ethiopia and the Horn:
rock art. In O. Vogel (ed). The encyclopedia
of precolonial Africa: archaeology, history,
languages, cultures, and environments,
pp.357-361. Altamira: Walnut Creek:
Altamira.
4. Agazi Negash (2001). The Holocene prehistoric
archaeology of the Temben region, northern
Ethiopia. PhD dissertation, University of
Florida.
5. Agazi Negash (2018). Regional variation of the
rock art of Ethiopia: a geological
perspective. African Archaeological Review
35(3):407-416.
6. Agazi Negash (2020). Enset (Ensete ventricosum)
and the Archaeology of Southwestern
221
Ethiopia. African Archaeological Review
37:627-637.
7. Arthur, J, Curtis, M., Arthur, K., Coltotri, M.,
Pierucciini, P., Lesur, J., Fuller, D., Lucas,
L., Conyers, L., Stock, J., Stretton, S., and
Tykot, R. (2019). The Transition from
Hunting–Gathering to Food Production in
the Gamo Highlands of Southern Ethiopia.
African Archaeological Review 36:5-65.
8. Bachechi, Luca. (2005). Preliminary report on the
2002 excavation of Harurona cave deposit.
Atti del Museo di Storia Naturale della
Maremma, supplement 21:67-78.
9. Brandt, S. A. (1982). A Late Quaternary cultural
environmental sequence for Lake Besaka,
Southern Afar, Ethiopia. PhD Dissertation.
Berkeley: University of California.
10. Brandt, S. A.,& Carder, N. (1987). Pastoral rock
art in the Horn of Africa: Making sense of
udder chaos. World Archaeology 19:194–213.
11. Clark, J. D. and Prince, G. R. (1978). Use-wear on
Later Stone Age Microliths from Laga Oda,
Hararghi, Ethiopia and Possible Functional
Interpretations. Azania, 13:101-110.
12. Clark, J. D., and Williamson, K. D. (1984). A
Middle Stone Age occupation site at Porc-
Epic Cave, Dire Dawa (east-central
Ethiopia). The African Archaeological Review,
2:37–71.
13. Dombrowski, J. (1970). Preliminary Report on
Excavations in Lalibela and Natchabiet
Caves, Begemeder, Annales d'Ethiopie 8:21-
29.
14. Dombrowski, J. (1971). Excavations in Ethiopia:
Lalibela and Natchabiet Caves, Begemder
Province. PhD Dissertation, Boston
University.
15. Edwards, D. (2007). The archaeology of Sudan
and Nubia. Annual Review of Anthropology
36:211–28.
16. Fattovich, R. (1988). Remarks on the Late
Prehistory and Early History of Northern
Ethiopia. In Taddese Beyene (ed).
Proceedings of the Eighth International
Conference of Ethiopian Studies. Addis
Ababa, Institute of Ethiopian Studies, pp.
85-105.
17. Fattovich, R., Marks, A. E., and Mohamed-Ali,
A. (1984). The Archaeology of the Eastern
Sahel, Sudan: Preliminary Results. African
Archaeological Review 2:173-188.
18. Gasse, F. (1977). Evolution of Lake Abhe
(Ethiopia and TFAI) from 70,000 BP.
Nature, 265:42-45.
222
19. Gasse, F. and Street, A. F. 1978. Late Quaternary
Lake-Level Fluctuations and Environments
of the Northern Rift Valley and Afar
Region (Ethiopia and Djibouti).
Palaeogeogeography, Palaeoclimatoloqy, and
Palaeoecology 24:279-325.
20. Gifford-Gonzalez, D. (2017). “Animal disease
challenges” fifteen years later: The
hypothesis in light of new data. Quaternary
International 436:283-293.
21. Gifford-Gonzalez, D. (2000). Animal disease
challenges to the emergence of pastoralism
in sub-Saharan Africa. African
Archaeological Review 18:96-139.
22. Gifford, D., Isaac, G., and Nelson, C. (1980).
Evidence for Predation and Pastoralism at
Prolonged Drift: a Pastoral Neolithic Site in
Kenya. Azania: Archaeological Research in
Africa 15:57-108.
23. Girma Hundie (2001). The emergence of prehistoric
pastoralism in southern Ethiopia. PhD
dissertation, University of Florida,
Gainesville, Florida.
24. Hildebrand, E. A., Brandt, S. A. and Lesur-
Gebremariam, J. (2010). The Holocene of
southwest Ethiopia: New insights from the
Kafa archaeological project. African
Archaeological Review 27(4):255–289.
25. Lesur, J., Vigne, J-D, and Gutherz, X. (2007).
Exploitation of wild mammals in South-
west Ethiopia during the Holocene (4000
BC–500 AD): the finds from Moche Borago
shelter (Wolayta). Environmental
Archaeology 12(2):139-159.
26. Lesur, J., Hildebrand,. E., Abawa, G., and
Gutherz, X. (2014a). The advent of herding
in the Horn of Africa: New data from
Ethiopia, Djibouti and Somaliland.
Quaternary International 343:148-158.
27. Lesur, J., Gutherz, X., Jallot, L., Diaz, A., and
Arthur, J. (2014b). Late Holocene societies
from Southern Ethiopia: Results from
archaeological project in Nechisar National
Park. Annales D'Ethiopie 29:265-268.
28. Marshall, F. (1990). Origins of specialized
pastoral production in East Africa.
American Anthropologist 92(4):873-894.
29. Marshall, F., (2000). The origins and spread of
domestic animals in East Africa. In: Blench,
M., MacDonald, K.C. (Eds.), The Origins
and Development of African Livestock:
Archaeology, Genetics, Linguistics and
Agazi Negash and Fiona Marshall
Ethnography. UCL Press, London, pp. 191-
221.
30. Marshall, F., & Hildebrand, E. (2002). Cattle
before crops: The beginnings of food
production in Africa. Journal of World
Prehistory 16, 99–143.
31. Marshall, F., Reid, R, Goldstein, S., Stotozum,
M., Wreschnig, A., Hu, L., Kiura, P.,
Shahack-Gross, R., and Ambrose, S. (2018).
Ancient herders enriched and restructured
African grasslands. Nature https://
doi.org/10.1038/s41586-018-0456-9.
32. Phillipson, D. W. (1977). The Excavation of
Gobedra Rock Shelter, Axum: An Early
Occurrence of Cultivated Finger Millet in
Northern Ethiopia. Azania 12:53-82.
33. Poisblaud, B. (2002). Une figurine de bovine
dans la site d’Akirsa. Annales d’Éthiopie
18:191–198.
34. Sadr, K. (1991). The Development of Nomadism in
Ancient Northeast Africa. Philadelphia:
University of Pennsylvania Press.
35. Schmid, E. (1972). Atlas of Animal Bones.
Amsterdam and New York: Elsevier.
36. Vavilov, N. (1951). The Origin, Variation,
Immunity, and Breeding of Cultivated
Plants. Chronica Botanica 13:1-364.
37. Von den Driesch, A. 1976. A Guide to the
Measurement of Animal Bones from
Archaeological Sites. Peabody Museum of
Archaeology and Ethnology Monographs.
38. Walker, R. 1985. A guide to post-cranial bones of
East African animals. Norwich: Hylochoerus
Press.
39. Westphal, E. (1975). Agricultural Systems of
Ethiopia. Center for agricultural publishing
and documentation, Wageningen.
40. Woldekiros, H. and D'Andrea. C. (2017). Early
evidence for domestic chickens (Gallus
gallus domesticus) in the Horn of Africa.
International Journal of Osteoarchaeology
27:329-341.
41. Zelalem Assefa, Pleurdeau, D., Duquesnoy, F.,
Hovers, E., Pearson, O., Asfawossen Asrat,
Constantinos T/Tsion, Yin Man Lam.
(2014). Survey and explorations of caves in
southeastern Ethiopia: Middle Stone Age
and Later Stone Age archaeology and
Holocene rock art. Quaternary International
343:117-135.
SINET:Ethiop. J. Sci., 44(2): 223–233, 2021 ISSN: 0379–2897 (PRINT)
© College of Natural and Computational Sciences, Addis Ababa University, 2021 eISSN: 2520–7997

Date received: April 26, 2021; Date revised: September 03, 2021; Date accepted: October 23, 2021
DOI: https://dx.doi.org/10.4314/sinet.v44i2.9

An integrated knowledge-based system for early detection of eye refractive error

using data mining

Teklay Birhane1,*, Dr. Dereje Teferi2 and Tariku Mohammed3

1Departmentof Information Science, Mekelle University, Mekelle, Ethiopia.

*E-mail corresponding author: teklaybirhane12@yahoo.com
2School of Information Science, Addis Ababa University, Addis Ababa, Ethiopia. E-mail:

dereje.teferi@aau.edu.et
3Department of Information Science, Haramaya University, Haramaya, Ethiopia. E-mail:

tarejimma@gmail.com

ABSTRACT: Refractive error is one of optical defect in the human visual system.
Refractive error is a very common disease these days in all populations and in all age
groups. Uncorrected and undetected refractive error contributes to visual impairment,
blindness and places a considerable burden on a person in the world. The long use of
technological devices such as smart phones also poses a new burden on the human eye. The
intensity and brightness of these digital devices open a new door for high prevalence of eye
refractive errors. Early medical diagnosis of the disease may help in avoiding complications
and blindness. Data mining algorithms can be applied to help in ophthalmology and
detection of an eye disease at an early stage. So mining the ophthalmology data in efficient
manner is a critical issue. This research work deals with development of an integrated
knowledge-based system that helps to detect eye refractive error early and provides
appropriate advice for the patients. In this study, the hybrid knowledge discovery process
model of data mining that was developed for academic research is used. About 9000
ophthalmology data from selected eye health centers are used to build the model. The
sample data was preprocessed for missing values, outliers, and noise. Then the model is built
using decision tree (J48 and REPTree) and rule induction (JRip and PART) algorithms. The
PART algorithm has registered better predictive performance with accuracy of 60% and
96.45% for subjective and objective based model evaluation, respectively as compared to J48,
REPTree, and JRip. Finally, the knowledge discovered with this algorithm is further used to
build the knowledge-based systems. The Java programing language is used to integrate data
mining results to knowledge-based system. The performance of the proposed system is
evaluated by preparing test cases. Overall, the knowledge based system resulted in 89.2%
accuracy. Finally the study concludes that discovering knowledge using data mining
techniques could be used as a functional eye refractive error detection system.

Keywords/Phrases: Data Mining Techniques, Eye Refractive Error, Knowledge-Based System

are particularly vulnerable groups for this

INTRODUCTION
Refractive error is one of optical defect in the
visual system, which causes induced blurred
vision and blindness. Refractive error
nowadays is one of the frequent reasons
behind the visual impairment and blindness
in the world (Fageeri et al., 2017). Refractive
error places a considerable burden and
creates a negative impact on individuals and
society. People living in low-income countries
are majority of those affected by refractive
error with minimal access to eye care
(Varadarajan et al., 2018). School-age children
disease (Jafer Kedir and Abonesh Girma,
2014). Where uncorrected and undetected,
refractive error have an adverse effect on
learning capability and educational potential,
as well as high economic cost to the family
and government (Nebiyat Kassa et al., 2014).
It causes a serious blurred vision and
blindness. The poor vision and inability to
read material on the chalkboard due to
refractive error can greatly affect a child‟s
participation in education system and other
social interactions (Sheeladevi et al., 2019).
Moreover, children often think that their
problem is common to all and may not speak

_____________________
*Author to whom correspondence should be addressed.
224
about it to their family or teachers. Due to
this, many children are forced to leave their
school in Ethiopia (Tibebu Kassa and Getu
Degu, 2000). This problem is common in
Ethiopia since the country has relatively poor
health coverage and eye health care system.
The lack of skilled workers in eye
healthcare contributes to poor diagnosis and
weak treatment of patient having eye
refractive error conditions in Ethiopia, mainly
due to lack of domain experts and eye health
care facilities (Nebiyat Kassa et al., 2014).
There are novel technologies such as artificial
intelligence (AI) that offer a low-cost method
of screening and diagnosing eye disease in
the developing world (Varadarajan et al.,
2018). Therefore, the use of AI is important in
order to solve such problems. Artificial
Intelligence (AI) has shown promising results
in the diagnosis and interpretation of medical
datasets to make intelligent decisions with the
design and development of intelligent
algorithms (Puaschunder, 2020).
As sub part of AI, data mining techniques
enable to extract useful knowledge from large
data sources using various analysis
techniques that integrate traditional methods
for analyzing data with complex algorithms
(Cheng et al., 2018).
Therefore, by considering the importance of
early medical diagnosis of a disease, data
mining techniques can be applied to help in
ophthalmology and detection of eye disease
at an early stage. Early identification of vision
problems and eye refractive errors, including
high refractive errors like hyperopia, myopia
and astigmatism is critical for optimal
treatment (Fageeri et al., 2017).
Data mining extract accurate information
from large amount of data generated by
medical centers which cannot typically
extract by medical experts (Codreanu et al,
2011). Such information can be used as key
input knowledge in developing effective and
efficient knowledge-based system (KBS)
(Kedir Eyasu et al., 2020). KBS is part of
Artificial Intelligence which contains the
knowledge and analytical skills of human
experts in a specific problem domain (Blondet
et al., 2019). It is a good solution to reduce
human and medical errors, functioning in a
specific domain to offer wise decisions with
better reasoning ability (Tripathi, 2011).
Fageeri et al. (2017) and Varadarajan et al.
(2018) developed predictive model for
Teklay Birhane et al.
diagnosis of eye refractive error using data
mining algorithm techniques such as decision
tree, support vector machine, naïve Bayes and
deep learning neural network. Although,
there are some developed models that
perform well, they lack the provision of
advice for the patients and optometrists about
what course of action to be taken in
accordance to the detected eye refractive error
condition. The integration of eye refractive
error detection with a KBS system will help
ophthalmologists and optometrists to provide
a better service to patients. To this end, there
is a need to integrate data mining results with
a reasoning system to design an intelligent
and self-learning knowledge based system.
This research aimed to develop an
integrated knowledge-based system that
automatically acquires knowledge from
ophthalmology dataset by applying data
mining techniques. The researchers
integrated the data mining model with
knowledge base systems to detect eye
refractive disease and provide advice for
optometrists and ophthalmologists. The
knowledge-based system has a learning
capability to update the knowledge base as
the size of data increases and new rules are
generated.
With the aforementioned in mind, this
study explores and addresses the following
research questions.
1. What are the relevant attributes for the eye
refractive error detection model?
2. How do we develop an integrated
prototype model for early detection of eye
refractive error?
3. To what extent does the integrated
prototype model works in detection process?
RESEARCH METHODOLOGY
Research Design
A Hybrid Knowledge Discovery Process
(HKDP) model is followed in this research. It
was developed based on the more industry
based CRISP-DM model by adopting it to
academically inspired research (Cios et al.,
2019). This model provides a more general
and research-oriented description of the
steps.
The CRISP-DM model has only three major
feedback sources, while the hybrid model has
more detailed feedback sources and a
SINET: Ethiop. J. Sci., 44(2), 2021 225

modified final step for the application of the reduction, and data transformation. Before
discovered knowledge. Subsequently, the feeding the data to Data Mining, the quality
hybrid knowledge discovery modeling of data has to be ensured. Well-accepted
process is chosen for this study. As multidimensional data quality measures such
mentioned, the HKDP model is a six step as accuracy, completeness, consistency,
process namely: understanding the business timeliness, and interpretability are used.
problem, understanding the data, data
preparation, data mining, evaluation of the Table 1: List of raw data variables in the initial
discovered knowledge, and use of discovered dataset
knowledge.
S. Attribu Value Description
No te Type
Problem Understanding Name
Domain problem understanding is assessed 1. uid Nominal Patient ID number
and investigated in depth with domain 2. age Numeric Patient age
3. od_sph Numeric The force required
experts for constructing eye refractive error from the lens of the
predictive model to improve the early right eye to correct
detection process of eye diseases.Ten-domain focus values ranging
experts are selected from Fitsum Birhan between 0 to +/- 20
4. od_cyl Numeric A vision defect, which
Specialized Eye Center and St. Louis Eye collects X-hyphen-to-
Clinics in Mekelle, Ethiopia to define the right eye on many
business problem and determine the data- points, so-called
mining goal. In addition, observations and astigmatism. values
ranging between 0 to
review of articles is done to understand the +/- 25
problem area in depth. 5. od_axis Numeric This is linked to CYL
Based on the suggestion from domain and tells where the
experts, variables age, sex, both parts of right place of
astigmatism.Values
and left eye (spherical, cylindrical and axial) ranging between 0 to
refractive error measure values are chosen to 180 degrees.
detect and identify refractive error condition. 6. os_sph Numeric Left eye sphere, which
indicates the power of
the lens prescribed to
Data Understanding correct the Left eye
In this study, real life medical dataset is vision.
7. os_cyl Numeric Left eye cylinder
employed. The eye diseases dataset was
8 os_axis Numeric Left eye axis
collected from the two Eye Clinics. Data from 9 Class Nominal Class of diseases:
2018 and 2019 is considered for the study. Hyperopic
Initially, the dataset collected in paper Astigmatism, Myopia,
based table format is converted to MS-Excel Myopic Astigmatism,
Hyperopia, Mixed
format by selecting the attributes chosen to be Astigmatism or
important for the problem domain. The Normal
dataset has 9000 instances and 9 attributes
with 508 missing values. Table 1 briefly Attribute Selection
describes the variables that are being The importance of attributes in eye
considered in the initial dataset.
refractive error predictive modeling is
checked by evaluating on all training data.
Data Preparation and Preprocessing
Attributes selection has passed through all
Data mining uses cleaned and transformed
possible combinations of attributes in the data
data, searches the data by using different
to find which subset of attributes works best
techniques and algorithms, and then outputs
for refractive error prediction. Accordingly,
patterns and relationships to the evaluation
attributes selected using best first techniques
step of the knowledge discovery from data
in WEKA are od_sph, od_cyl, od_axis, os_sph,
(Li et al., 2011 and Mihaela, 2006).
os_cyl, and os_axis. From the eight (8)
According to Han and Kamber (2012), the
attributes stated in table 1 six (6) of them are
data processing task of data mining includes
used in the study the remaining two
data cleaning, data integration, data
attributes uid and age has no effect in this
226
study because the detection process of eye
refractive error using refraction based eye
examination does not depends on patient id
or patient age. Thus, before preceding to the
experiments, the accuracy of the model in
WEKA with the selected attributes is
evaluated.
Weka data mining tool was used for
knowledge extraction and data analysis. To
map, the knowledge acquired from the data
mining classifiers model with the knowledge-
based system, Java Net Beans IDE 8.2 with JDK
8 is employed. In addition, SWi-prolog
integrated with JPL is applied to develop and
implement the system.
EXPERIMENTATION AND RESULTS
The next step in the hybrid knowledge
discovery process model, after preparation
and preprocessing of the data, is model
development followed by model evaluation.
A dataset with 9000 records is used for
training and testing the predictive model
constructed in this study.
Four data mining classification algorithms
are experimented to discover knowledge
from the data for predicting refractive error
and to build the analytical model for
diagnosis and treatment of refractive error.
Classification is a data mining technique,
based on supervised machine learning, which
maps data into predefined groups or classes
(Sharma et al., 2013). It is the process of
building a model of classes from a set of
records that contain class labels (Liu and Wu,
2012). It often describes these classes by
looking at the characteristics of data already
known to belong to the classes (Sharma et al.,
2013).
The four classification algorithms used in
this study are J48 pruned, REPTree, PART, and
JRip. J48 and REPTree are tree-based
classifiers in WEKA whereas JRip and PART are
rule-based classifiers. The four algorithms are
selected because they use decision tree or rule
based structure. These kind of algorithms are
known to perform well for numeric data
types.
Hence the aim of this study is constructing
of rule based KBS and because of their capable
of generating powerful rules the four
classifiers algorithms are selected. A rule-
based classification extracts a set of rules that
Teklay Birhane et al.
show relationships between attributes of the
dataset and the class label ( Kedir Eyasu et al.,
2020). The dataset is split into training and
test using 10-fold Cross-validation. Thus, for
each classifier, 10 experiments are conducted
and the average evaluation result is reported.
Experiment 1- J48 classifier
A decision tree produces a sequence of
rules that can be used to classify the given
data of attributes together with its classes
(Milovic, 2012 and Brázdil et al., 2018). It is
mainly used in classification and prediction.
It is a simple and powerful way of
representing knowledge (Milovic, 2012).The
strengths of J48 are simple to understand and
visualize, requires little data preparation, and
can handle both numerical and categorical
data. Drawn backs of J48 are poor results on
very small datasets and overfitting can easily
occur.
The experiment shows that, the model
developed using J48 classifier has a tree of 63
nodes with 32 leaves. The algorithm has
correctly classified 8702 instances and only
298 instances are classified incorrectly taking
0.47 seconds to build the model. In Table 2
and all the following experiments, A = Mixed
Astigmatism, B = Hyperopia, C = Normal, D
= Myopic Astigmatism, E = Myopia, F =
Hyperopic Astigmatisms
Table 2. Confusion Matrix for J48 classification
algorithm.
Confusion Matrix
A B C D E F
A 1396 70 22 6 1 5
B 0 1432 57 1 6 4
C 7 54 1397 27 9 6
D 1 8 0 1490 0 1
E 2 4 4 3 1487 0
F 0 0 0 0 0 1500
Experiment 2- REPTree classier
The second experiment is based on
REPTree algorithm. REPTree creates multiple
trees in different iterations using the
regression tree logic. After that it selects the
best one from all generated trees and that will
be considered as the representative. Basically
REPTree is a fast decision tree learner that
builds a decision/regression tree using
information gain/variance and prunes
predictions made by the tree using the mean
square error measure. This algorithm has
SINET: Ethiop. J. Sci., 44(2), 2021 227

correctly classified 8703 instances out of 9000. the one, which performed better, is selected
That is, it has incorrectly classified 297 as prime choice for the knowledge acquisition
instances taking 0.13 seconds to build the step. Both objective and subjective based
model. In addition, the tree generated has 53 model evaluation methods are used to select
nodes. the best model for development of the KBS.
Objective based model evaluation
Table 3. Confusion Matrix for REPTree measurement is generally based upon the
classification algorithm. statistical and inherent structure of the mined
patterns.
Confusion Matrix The accuracy, sensitivity (TPR), precision,
A B C D E F
recall, and F-measure of all the four
A 1390 70 22 11 2 5
B 0 1489 0 2 5 4 classifiers, obtained from the experiment, are
C 10 97 1350 27 10 6 shown in Table 6.
D 4 8 0 1481 6 1
E 0 4 0 3 1493 0 Table 5. Confusion Matrix for JRip classification
F 0 0 0 0 0 1500
algorithm.

Experiment 3-PART classier Confusion Matrix

PART is a rule-based classifier that uses a set A B C D E F
of IF-THEN rules for classification. It is an A 1384 72 25 12 2 5
expression of IF condition THEN conclusion. B 16 1442 30 5 1 6
PART, rule induction algorithm generated 25 C 7 88 1363 25 9 8
rules by involving all the attributes of the D 0 6 2 1490 1 1
E 3 5 5 4 1480 3
dataset. The algorithm registered prediction
F 0 0 0 0 0 1500
accuracy of 96.47% in which 8682 instances
out of 9000 are correctly classified. The
algorithm has incorrectly classified only 318
Table 6. Performance of Classifiers.
instances and took 0.64 seconds to build.

Table 4. Confusion Matrix for PART

Objective Classification algorithms
classification algorithm.
model J48 REPTree JRip PART
evaluations pruned
Confusion Matrix Correctly 96.69% 96.7 % 96.21% 96.47%
A B C D E F Classified
A 1410 57 23 4 1 5 Incorrectly 3.31% 3.3 % 3.79% 3.53%
B 30 1409 50 4 3 4 Classified
C 7 68 1390 27 2 6 TP Rate 96.7% 96.7% 96.2% 96.5%
D 1 7 1 1490 0 1 Precision 96.7% 96.9% 96.3% 96.5%
E 2 3 9 3 1483 0 Recall 96.7% 96.7% 96.2% 96.5%
F 0 0 0 0 0 1500 F-Measure 96.7% 96.7% 96.2% 96.5%
Time taken 0.47 0.13 10.64 0.67
Experiment 4 – JRip classifier in seconds

The other rule induction algorithm selected

As can be seen from Table 6, REPTree
for this study is JRip. JRip is a rule-based
classification algorithm performs better in all
classifier algorithm that uses a set of IF-THEN
objective based model evaluation methods.
rules for classification process. JRip correctly
However, over 3.3% of the test data is still
classified 8659 instances from 9000. The
incorrectly classified. The rule acquired from
numbers of incorrectly classified instances are
the classifier algorithms will be used for
341. The algorithm has generated 15 rules. See
constructing the knowledge base and hence
Table 5.
these errors need to be minimized as much as
possible. To this end, subjective model
Model evaluation and selection evaluation method is used to identify and
exclude the incorrect rules from the
All the selected classifier algorithms
knowledge base.
generated rules from the dataset. The
Discussion was held with domain experts
performance of algorithms is compared, and
about the significance and correctness of the
228
rules generated by the four algorithm. The
rules provided by the models can be easily
understood by human experts without any
difficulty. The rules of the selected four
algorithms are distributed to ten purposively
selected domain experts and their evaluation
is collected as shown in Table 7.
Table 7.Subjective model evaluation results.
Classification algorithms
Subjective model J48 REP JRip
evaluation pruned Tree
Domain Expert 10% 10% 20%
Knowledge
No of rules 32 53 15
Rule Extraction
From the four models that are built, the
model developed with PART rule induction
classifier is selected for developing the
knowledge base because it registered better
performance (average of objective and
subjective model evaluation method). Some
of the interesting rules generated by PART
model are presented below.
RULE 1: IF os_cyl<= -0.2 AND os_sph<=
0.454007 AND od_cyl> 0 AND os_cyl>
-0.75 THEN Eye refractive error
disorder = Hyperopic Astigmatism
(72.0).
RULE 2: IF os_cyl<= -0.2 AND os_sph<=
0.454007 AND od_sph<= 0 AND
od_cyl<= -0.25 THEN Eye refractive
error disorder = Myopic Astigmatism
(741.0).
RULE 3: IF os_sph<= -0.25 AND os_cyl> -0.2
AND od_cyl> -0.25 AND od_sph<= -
1.75 THEN Eye refractive error
disorder = Myopia (1479.0).
RULE 4: IF os_cyl<= -0.2 AND os_sph<= 0 AND
od_sph<= -2 THEN Eye refractive
Teklay Birhane et al.
error disorder = Myopic Astigmatism
(550.0/2.0).
RULE 5: IF os_cyl<= -0.2 AND od_cyl<= -0.25
AND os_sph> 0.5 AND od_sph> 0.75
THEN Eye refractive error disorder =
Mixed Astigmatism (1333.0).
RULE 6: IF os_sph> 0 AND os_cyl<= 0.25 AND
od_cyl<= 0.25 AND od_sph> 0 AND
os_cyl> -0.2 THEN Eye refractive error
disorder = Hyperopia (1089.0/15.0).
RULE 7: IF os_sph<= 0 AND os_axis<= 0 AND
PART
od_sph<= 0 AND od_cyl> -0.25 AND
60% os_sph> -0.75 THEN Eye refractive
error disorder = Normal (1169.0).
25
Mapping Predictive Model to Knowledge
Based Systems
Java programming language is used to
integrate the data-mining model to the
knowledge-based systems. An integrator
module that translates a Java program from a
text file to a Prolog representation is
implemented.
The knowledge discovered using data
mining techniques (PART classification
algorithm) is used in building the knowledge-
based systems that predict eye refractive error
conditions. Then, the knowledge-based
systems are used for early detection of
uncorrected refractive error by inferring from
the inference engine of the knowledge-based
system. Moreover, the developed knowledge
based systems have a self-learning capability
to generate new rules and update its
knowledge base automatically.
Architecture of the Prototype System
Architecture is a blueprint showing how
the components of the integrated knowledge-
based system interacts and interrelates.
Figure 1 illustrates the architecture of the
proposed system.
SINET: Ethiop. J. Sci., 44(2), 2021
Figure 1. General Architecture of the prototype KBS.
Integrator Module
The results of the data mining are the basis
for the development of the knowledge base
system through the integrator application.
The facts generated from the Data mining can
be represented as a rule in knowledge base.
The PART model is integrated with
knowledge-based system automatically for
designing intelligent early detection of eye
refractive error as depicted in Figure 2.
Figure 2. Conceptual Design of the Integration Process.
Overall, the following six steps are used to
develop the KBS. Each step is discussed as
follows:
1. Dataset Preparation: - The ophthalmology
dataset is collected from two-eye clinic
centers manual database/record named
229
Fitsum Birhan Specialized Eye Center and St.
Louis Eye Clinic.
2. Rule mining: - at this stage, rules are
extracted from the ophthalmology dataset by
applying classifier algorithm using Weka.
PART classifier algorithm is selected as the
best predictive model.
3. Rule Tokenization: - at this stage, removal
of undesirable characters and replacing some
tokens with other equivalent tokens is
performed.
For example, comparison operator „<=‟ (less
than or equal to) in PART rules is replaced by
its PROLOG equivalent „=<‟. The character
“:”in PART is replaced by “:-” in equivalent
character which means IF in PROLOG.
Moreover, the conjunction operator „AND‟ is
replaced by its PROLOG equivalent „,‟ used for
joining two conditions.
4. Rules and facts Parsing - this is the process
of analyzing a string of symbols, either in
natural language or in computer languages,
according to the rules of a formal grammar.
In this context, parsing is analyzing the
components of the generated PART rules. A
given PART rule is composed of: (condition)
230 Teklay Birhane et al.

implication (conclusion). If condition is diagnosis of eye refractive error and the

evaluated true then conclusion is executed. Prolog‟s built-in inference mechanism is
5. Rule Reversing: - the rule reverse is used forward chaining, a goal derived forward
to exchange the place of the Left Hand Side inference mechanism that tries to prove or
(LHS) and the Right Hand Side (RHS) of the disprove the goal is used.
rule. This is done because PROLOG uses the D. User Interface: - It is the means of by
reversed rule format. That means PROLOG which the user and a computer system
starts with conclusion and goes for the interact. The acceptability of a knowledge-
conditions that made the conclusion. PART based system depends on the quality of the
algorithm rule is reversed from the format IF user interface. The user interface is used as
… THEN to the format THEN… IF PROLOG the means of interaction between the user and
understandable format. For example: the knowledge based system.
Rule: attribute1= value1 AND attribute2 = value2
AND attribute n = value n THEN (conclusion).
Reversed rule format: (conclusion):- attribute1=
value1, attribute2 = value2, attribute n = value n.
Finally, period (.) is placed at the end of all
reversed rules to tell PROLOG end of
statement.
6. Rule Normalization: - normalization is
done to change all tokens in reversed rule
into lower case to be understood by PROLOG.

Knowledge based system

A typical knowledge based system consists Figure 3. User Interface of the Developed KBS
of the following subsystems: knowledge
representation, knowledge base, Inference E. Explanation facility: The KBS system can
engine, explanation facility and user interface. describe “what”, a request to repeat for
A. Knowledge representation: - The most clarification before it reaches on its
commonly used methods of knowledge conclusions. This ability is usually important
representation are production rule, frame, because the type of problems to which a KBS
and network. Knowledge captured from is needed requires an explanation of the
experts and other sources must be organized result presented to the end-users. It has also
in such a way that a computer program can the ability of justifying “why” a certain
access this knowledge whenever needed and problem is being questioned in order to
draw conclusions. In KBS, production rules reason out what it means and how it benefits.
are used and represented in the form of rules The system included “what” and “why”
by rule-based representation technique, since explanation facilities in problem solving.
it permits the relationships that make up the Moreover, the explanation facilities included
knowledge base to be broken down into in the system are easily understandable by
manageable unit. the end-users.
B. Knowledge Base: - is a set of rules or
encoded knowledge about detection process
of eye refractive error disorders. The
knowledge collected from domain experts,
document analysis and data mining is stored
in the knowledge base as set of rules using a
rule-based knowledge representation
method. The rules are stored as refraction
based library.
C. Inference Engine: - An inference engine
is the brain of the Knowledge Based System,
which directs the system to derive a
conclusion by looking for possible solutions Figure 4. User interface of the explanation facility.
from the knowledge base and recommending
the best possible one. Since the objective of
the proposed Knowledge Based System is
SINET: Ethiop. J. Sci., 44(2), 2021 231

System Testing
Before deployment of the system to actual
use, it has to be evaluated by domain experts
to assure whether the system meets its
objectives or not. To evaluate the
performance of the KBS, test cases are
prepared and given to the system. The
outputs of the system are compared against
domain experts‟ judgment.
Figure 5. User interface for detected eye disorder result.

Table 8. KBS accuracy test.

Prototype System Response/output

Diagnoses Normal Myopia Hyperopia Mixed- Hyperopic- Myopic- Total
Type astigmatism astigmatism astigmatism
Normal 7 0 0 0 0 0 7
Myopia 3 5 0 0 0 0 8
Domain Hyperopia 1 0 10 0 0 0 11
Experts Mixed- 0 0 0 4 0 0 4
suggestion astigmatism
Hyperopic- 0 0 0 0 3 0 3
astigmatism
Myopic- 0 0 0 0 0 4 4
astigmatism
Total 11 5 10 4 3 4 37
knowledge base of the KBS can be taken as a
The system accuracy test, Table 8, shows strong feature of the developed system.
that test cases evaluation by the KBS and The integration process began by taking
domain experts‟ judgment. The column side initial samples of ophthalmology datasets.
shows the result of KBS and the row side The collected datasets was preprocessed and
shows suggestion of domain experts. made suitable for mining steps. Since the
The entry under column Normal indicates system focuses on refraction-based eye
that out of eleven instances seven of the examination process, automatic knowledge
instances are correctly identified as Normal, acquisition mechanism is applied in this
three of the instances are incorrectly study to make intelligent decision-making.
identified as Myopia and one instance is PART data mining classifier is employed for
incorrectly identified as hyperopia. The automatic knowledge acquisition. PART
entries in all the rest columns show that the performed best among the selected four-
system has correctly identified the instance. classifier algorithms namely J48, REPTree and
Overall, the system has correctly detected and JRip with an accuracy of 60% and 96.45% for
diagnosed 33 test instances out of 37 to their subjective and objective based model
correct class. This means the system has evaluation respectively. Automatic integrator
89.2% detection accuracy. application is built to map the datamining
model with knowledge-based system and
construct knowledge base. Following the
DISCUSSION successful integration of induced knowledge
with the knowledge-based system, an
Eye disease diagnosis is one of real-world integrated KBS for detection of eye refractive
medical problem. Detection of eye disease in error conditions is built. System performance
its early stages can prevent complications and testing is undertaken to make sure that the
blindness. KBS works well. Testing the system registered
In this study, the possibility of integrating a promising result of 89.2% accuracy.
data mining models with knowledge-based In this study, promising result is achieved
system is realized and explored. The use of in integrating data mining induced patterns
data mining techniques to build the with knowledge-based system for detecting
232
eye refractive error conditions and providing
appropriate advice for ophthalmologists.
This study provides automated medical
diagnosis and treatment of eye disorders and
diseases. The knowledge-based system
improves decision-making process on eye
refractive error detection. The system
provides advice about detected refractive
error conditions to the ophthalmologist as
well as which action to take in accordance to
the detected eye refractive error condition.
The system can also become an expert
knowledge-sharing tool to be used by other
medical personnel who are not specialists in
diagnosis of eye refractive errors, especially
for hospitals that do not have an
ophthalmologist. Moreover, this research
helps to increase awareness on the negative
impacts of eye refractive errors and helps to
detect them early for better treatment.
In addition, this study will motivate future
researchers to work on developing
knowledge-based system using data mining
techniques in other fields of studies especially
in areas where there is shortage of domain
experts.
The developed integrated KBS supports
specific types of eye refractive error
conditions. However, each type of eye
refractive error has sub categories. It is
possible to increase the number of eye
diseases in future work using similar studies.
The following are recommended as potential
research area for future study.
 The need for additional types of eye
refractive disorders solution for users who
have multiple eye disorder diagnosis.
 The use of image datasets to improve
the performance of the model in detection
of eye refractive error conditions.
 Developing hybrid knowledge based
system, which is capable of employing
rule based reasoning, and case based
reasoning with integrated data mining.
ACKNOWLEDGMENTS
The authors acknowledge Mekelle University for
allowing this research as well as Haramaya
University for writing support letter to the eye
hospitals where the study was conducted.
Moreover, the authors are grateful to the
management of the eye hospitals considered in this
study for providing the dataset.
Teklay Birhane et al.
REFERENCES
1. Sheeladevi S, Seelam B, Nukella PB, Borah RR,
Ali R and Keay L. (2019). Prevalence of
refractive errors, uncorrected refractive
error, and presbyopia in adults in India:
A systematic review. Indian Journal of
Ophthalmology, 67:583-92. https://
doi.org/10.4103/ijo.IJO
2. Blondet, G., Duigou, J. Le, and Boudaoud, N.
(2019). A knowledge-based system for
numerical design of experiments
processes in mechanical engineering.
Expert Systems with Applications,
122(6):289–302. https://doi.org/10.1016/
j.eswa.2019.01.013
3. Brázdil, T., Chatterjee, K., Křetínský, J., and
Toman, V. (2018). Strategy representation
by decision trees in reactive synthesis.
Journal of Artificial Intelligence and Lecture
Notes in Bioinformatics, 385–407.
https://doi.org/10.1007/978-3-319-
89960-2
4. Cheng, Y., Chen, K., Sun, H., Zhang, Y., & Tao,
F. (2018). Data and knowledge mining
with big data towards smart production.
Journal of Industrial Information Integration,
9(8):1–13. https://doi.org/10.1016/
j.jii.2017.08.001
5. Kedir Eyasu,Worku Jimma andTakele Tadesse
(2020). Developing a Prototype
Knowledge-Based System for Diagnosis
and Treatment of Diabetes Using Data
Mining Techniques. Ethiopian Journal of
Health Sciences, 30(1):115–124. https://
doi.org/10.4314/ejhs.v30i1.15
6. Fageeri, S. O., Ahmed, S. M. M., 1. Almubarak, S.
A., and Mu‟Azu, A. A. (2017). Eye
refractive error classification using
machine learning techniques. Journal of
International Conference on Communication,
Control, Computing and Electronics
Engineering. https://doi.org/10.1109/
ICCCCEE.2017.7867660
7. Cios, K.J., Pedrycz, W., Swiniarski, R.W. and
Kurgan, L.A. (2019). Data Mining A
Knowledge Discovery Approach.
International Journal of Chemical Information
and Modeling, 53(9): 1689-1699.
8. Han, J., Kamber, and Pei, J. (2012). Data mining
Concepts and Techniques Preface and
Introduction. Morgan Kaufmann, 3rd
edtion,1-740.
9. Nebiyat Kassa, Alemayehu Woldeyes and
Chalachew Misganaw (2014). Prevalence
and Factors Associated With Refractive
Error Among Primary School Children in
Addis Ababa, Ethiopia. International
Journal of Medical and Health Sciences
Research, 1(9):92–104.
10. Jafer Kedir and Abonesh Girma. (2014).
Prevalence of refractive error and visual
SINET: Ethiop. J. Sci., 44(2), 2021
impairment among rural school-age
children of Goro District, Gurage Zone,
Ethiopia. Ethiopian Journal of Health
Sciences, 24(4): 353–358. https://doi.org/
10.4314/ejhs.v24i4.11
11. Codreanu, D.E., Parpandel, D.E. and Popa, I.
(2011) Data Mining : Extracting
Knowledge from Data. Nicolae Titulescu
University Publishing.
12. Mihaela, O. (2006). On the Use of Data-Mining
Techniques in Knowledge-Based Systems.
1–4.
13, Li, J.-S., Yu, H.-Y., and Zhang, X.-G. (2011).
Data Mining in Hospital Information
System. New Fundamental Technologies
in Data Mining. https://doi.org/
10.5772/14052.
14. Liu, Q., & Wu, Y. (2012). Encyclopedia of the
Sciences of Learning. Encyclopedia of the
Sciences of Learning. https://doi.org/
10.1007/978-1-4419-1428-6
15. Milovic, B. (2012). Prediction and decision
making in Health Care using Data
Mining. International Journal of Public
Health Science (IJPHS), 1(2): 69–76.
https://doi.org/10.11591/ijphs.v1i2.1380
16. Puaschunder, J. M. (2020). The Potential for
Artificial Intelligence in Healthcare.
Electronic Journal of Social Science Research
Network (SSRN), 6(2):94–98. https://
233
doi.org/10.2139/ssrn.3525037
17. Sharma, S., Agrawal, J., Agarwal, S., & Sharma,
S. (2013). Machine learning techniques for
data mining: A survey. Journal of
International Conference on Computational
Intelligence and Computing Research, I.
https://doi.org/10.1109/ICCIC.2013.672
4149
18. Tibebu Kassa, G. D. A. (2000). prevalance of
refractive error in preschool and school
childern of Debark and Kola Diba
towns,Northwest Ethiopia. Ethiopian
Journal of Health Development, 17(2): 117–
124.
19. Tripathi, K. P. (2011). A Review on Knowledge-
based Expert System : Concept and
Architecture. Journal of Artificial
Intelligence Techniques - Novel Approaches &
Practical Applications, 4(4):19–23.
20. Varadarajan, A. V., Poplin, R., Blumer, K.,
Angermueller, C., Ledsam, J., Chopra, R.,
Keane, P. A., Corrado, G. S., Peng, L., and
Webster, D. R. (2018). Deep learning for
predicting refractive error from retinal
fundus images. International Journal of
Investigative Ophthalmology and Visual
Science, 59(7):2861–2868. https://doi.org/
10.1167/iovs.18-23887
SINET:Ethiop. J. Sci., 44(2): 234–241, 2021 ISSN: 0379–2897 (PRINT)
© College of Natural and Computational Sciences, Addis Ababa University, 2021 eISSN: 2520–7997

Date received: June 14, 2021; date revised: November 13, 2021; date accepted: November 15, 2021
DOI: https://dx.doi.org/10.4314/sinet.v44i2.10

Effects of plyometric and strength trainings on selected physical fitness variables in

Ethiopia youth sport academy female soccer players

Chalachew Lemecha1 and Aschenaki Taddese2,*

1 EthiopiaYouth Sport Academy, E-mail:chalewlemecha@gmail.com

2 Department of Sport Science, College of Natural and Computational Sciences, Addis Ababa
University, Ethiopia. E-mail: aschenakitaddese1@gmail.com

ABSTRACT: Plyometric is an exercise in which muscles exert force in short intervals of time.
Back then, it was only employed in track and field events and was known as “jump training.” The
purpose of this study was to investigate the effects of plyometric and strength trainings on selected
physical fitness variables in Ethiopia Youth Sport Academy female soccer players. The study used
quasi-experimental design. Random sampling technique was used to group the participants into
plyometric and strength training groups. Both groups performed selected exercises for consecutive
twelve weeks, implemented two days per week and two hours in each session. Paired sample t-test
and independent t-test were conducted to analyze the change scores from pre and posttest for all
subjects and from separated groups. The paired sample t-test results revealed that players in the
plyometric group significantly improved their agility (p<0.05) and explosive power performance
(p<0.05) from baseline to the end of the intervention. Whereas, female soccer players in the strength
group only improved their explosive power (p<0.05) following twelve-week of strength training
intervention. As the independent sample t-test result identified no significant performance
difference in agility (p>0.05) and explosive power (p>0.05) performances between the groups. In
general, following twelve-week of continuous plyometric and strength trainings the study
identified that both training interventions improved the agility and explosive power ability of
female soccer players and no significant performance differences had been identified between the
two groups.

Keywords/phrases: agility, explosive power, plyometric, strength

kicking, tackling, jumping, turning, sprinting,

INTRODUCTION
Soccer is one of the most popular ball games
which is widely played and followed by a high
number of supporters all over the world at a
professional level (Acar et al., 2008). In modern
soccer, physical fitness is one of the most
important factors in indicating successfor soccer
players. Out of physical fitness variables,
explosive power and agility are the performance
indicators for soccer players during trainings and
matches. Plyometric is a training technique used
by athletes in all types of sports to increase
strength and explosiveness. It consists of
physical exercises in which muscles exert
maximum force at short intervals of time to
increase dynamic performances. Out of those
performances, agility is one of the basic
performance indicators in soccer which can be
developed by plyometric training.
According to Bangsbo et al. (2006), during a
90-minute soccer match, professional soccer
players make numerous explosive bursts, such as
_____________________
*Author to whom correspondence should be addressed.
and changing pace. Plyometric exercise consists
of a rapid stretching of a muscle (eccentric
action) immediately followed by a concentric
(shortening action) of the same muscle (Baechle,
1994). According to different scholars,
plyometric trainings in soccer have a significant
effect on players’ power, especially explosive
power of players in training and game situations
(Chu, 1998).
Hoff (2004) suggests that strength training
using high loads, few repetitions and maximal
mobilization of force in the concentric mode
have proved to be effective in the development
of strength and related parameters. Strength
(resistance) training is a training technique that
improves muscle strength or size. But Herman, et
al. (2008), studied that strength training alone
does not alter knee and hip kinematics in female
recreational athletes and that further research is
needed to determine the effects of strength
training in combination with other intervention
methods on lower extremity biomechanics and
performance improvement. Those two studies
SINET: Ethiop. J. Sci., 44(2), 2021
brought different results, making them
controversial. Nowadays, many coaches,
researchers and players are looking for ways to
improve the performance of players in soccer.
For instance, Ethiopia Youth Sport Academy has
been delivering trainings for talented soccer
players from all over Ethiopia for the last five
years. However, the effectiveness of plyometric
and strength trainings with reference to soccer
related physical fitness variables hasn’t been
investigated yet. Therefore, the main aims of this
study were to answer for the following
questions1. Does twelve-week of plyometric
training has an effect on the agility of Ethiopia
Youth Sport Academy female soccer players? 2.
Does twelve-week of plyometric training has an
effect on the explosive power of Ethiopia Youth
Sport Academy female soccer players? 3. Does
twelve-week of strength training has an effect on
the agility of Ethiopia Youth Sport Academy
female soccer players? 4. Does twelve-week of
strength training has an effect on the explosive
power of Ethiopia Youth Sport Academy female
soccer players? 5. Which type of training does
significantly bring change on the explosive
power and agility of female soccer players?
MATERIALS AND METHODS
This study used quasi-experimental research
design. The sample population of the study was
20.Using the baseline fitness tests, players were
assigned into two equal groups (plyometric
group= 10 players and strength group =10
players). Participants in the plyometric and
strength training groups were performing their
task immediately after the warming-up.
The interventions or treatments were
performed for twelve weeks, two days per week
and two hours per session. Participants
continued their regular soccer training routines
that were identical for every participant, except
the two days of plyometric and strength training
program.
The experiment consisted of two test
sessions (pre and post exercise test) and two
interventions (plyometric and strength). Pre-tests
were performed three days before the beginning
of training and it contained field tests (Illinois
agility test and counter movement jump test) to
evaluate agility and power respectively. Then,
those programs were followed by post-tests,
three days after the last training session. All
testing sessions began with a dynamic stretching
235
of all muscles with warming up and at the end,
cooling down.
Test Procedures
The vertical jump test and Illinois agility
test were used to know the explosive power and
agility level of players respectively. Before the
tests, the subjects had performed warming–up
exercises for 15 minutes along with dynamic
stretching. Then, the researcher demonstrated
how they could perform the test. During the
Illinois agility test, an assistant score recorder
commanded them to “Go!” simultaneously
starting his stop watch which went on until the
athletes finished the obstacle. In each trial the
time to cover each Illinois agility test was
recorded in seconds. The players were given two
chances and took full recover between the first
and second trials. As for the vertical jump test,
the players’ standing height with extended hand
was measured and recorded. Then, the athletes
jumped vertically as high as they could and
touched the wall with their fingertips. In the end,
the final result was recorded as the difference
between the first standing height and the
jumping height. The vertical jump tests were
taken at baseline (pre-test) and end of the twelve
weeks (post-test).
Dependent and Independent Variables
In this study, plyometric and strength
trainings were used as independent variables of
the study. On the other hand, the dependent
variables were agility and explosive power
performances/results of the participants
respectively.
Validity and Reliability of Test Instrument
To ensure the validity and reliability of the
tests, the study used standardized and
internationally accepted testing tools, procedures
and protocols. Valid tests were applied to both
dependent variables of the experiment (explosive
power and agility). Those were
countermovement jump test and Illinois agility
test.
Inclusion criteria were: Free from any
physical health problems, willing to stay in the
academy for more than three months, player in
236 Chalachew Lemecha and Aschenaki Taddese

the under 17 years old football team, willing to RESULTS

eat, sleep and train in the academy during the
experiment weeks and do not take any Table 1: Demographic Characteristic of the Players.

medication.
Description N Descriptive Statistics
Mean SD
Data Analysis Age (Years) 20 15.9 1.07
Training experience (years) 20 2.7 .73
The data gathered through pre-test and BMI (kg/m2) 20 18.9 .67
post-tests were coded and arranged for
analysis.The pre and post-test results were As shown in the table, the participants were
presented as mean (SD). The coded and arranged 20 female EYSA soccer players. The average age of
data were analyzed by using paired t-test to see participants was 15.9 years (SD = 1.07). The age of
performance changes in both groups from participants ranged from 14 to 18 years (M =
baseline (pre-test) to post-test and independent 15.90, SD = 1.07). The average training age of the
t-test to identify performance differences players was 2.7 years which indicated that the
between the groups. The researchers also used athletes started their regular training after
SPSS version 20.0 software to describe, compare, entering the academy. The above table also
summarize and analyze the changes in the shows that the average body mass index of the
dependent variables. The level of significance players was 18.9.
was set at p ≤ 0.05%.

Table 2. Descriptive Statistics of the pre- and post-test results of the dependent variable (Illinois Agility test).

Variables Experimental Group N Mean SD

Illinois Agility Pre- test (sec.) Plyometric training group 10 18.4 0.87
Strength training group 10 18.5 0.52
Illinois Agility Post-test (Sec.) plyometric training group 10 17.8 0.78
Strength training group 10 18.0 0.40

Test results in table 2 reveals the pre- and
post-test assessment results of the two groups’
Illinois agility tests. As indicated in the table, the
mean (SD) values of the pre-test results of the
Illinois agility testsresults of players in the
plyometric and strength groups were 18.4(0.87)
Sec. and 18.5(0.52) Sec. respectively. The post-
test mean (SD) values of 17.8(0.78) and 18.0(0.40)
cm were scored by players in the plyometric and
strength training groups at the end of the
12thweek.
The data in the above table (Table 3) reveals
the descriptive analysis results ofthe pre- and
post-test CMJ assessment results of the two
groups. As indicated in the table, the mean (SD)
values of the pre-test results of the CMJ of players
in the plyometric and strength groups were
31.40(9.10) cm and 25.7(8.07) cm respectively.
The post-test mean (SD) values of 35.20(7.11) and
35.80(7.23) cm were scored by players in the
plyometric and strength training groups after 12
weeks of study weeks.

Table 3. Descriptive Statistics of the pre and post-test Table 4. Paired sample t-test results of the Pre
results of the dependent variables (Counter to post-test dependent variables
Movement Jump test). (Plyometric on agility).

Test Experimental N Mean SD Variables Groups

groups Sig.
Pre-test score on Plyometric training 10 31.40 9.10 Pre- to post-test Illinois Plyometric training
CMJ(cm) group Agility Test(sec.) group 0.000
Strength training 10 25.70 8.07 Strength training 0.113
group group
Post-test score on Plyometric training 10 35.20 7.11
CMJ(cm) group
Strength training 10 35.80 7.23 To see the effects of 12 weeks of plyometric
group training on the agility performance of the female
soccer players, paired sample t-test was
conducted. Based on the result of the analysis
SINET: Ethiop. J. Sci., 44(2), 2021
(table 4), statistically, significant change (p<0.05)
had been identified among players in the
plyometric training group. In the contrary, the
plyometric group no meaningful change had
been reported for players who didn’t receive
plyometric training (strength group).
Table 5: Paired sample t-test results of the Pre- to
post-test dependent variables (strength on
explosive power).
Variables Groups
Pre to post-test Counter Plyometric
Movement Jump Test (cm) training group
Strength training
group
Table 6: Independent sample t-test results of the post-test differences in the Illinois agility and counter
movement jump tests.
Test variables
Between groups Post-test performance difference in
Illinois agility test
Between groups Post-test performances difference in
counter movement jump test
The study applied two types of interventions -
plyometric and strength trainings. After 12
weeks of the training interventions, the study
aimed at identifying the performance differences
between the groups in their agility and explosive
power abilities. Based on the results from
independent sample t-test, the study reported no
significant difference between the two study
groups due to the designed plyometric (p=0.483)
nor the strength (p=0.854) training interventions.
Although the analysis results revealed no
significant performance differences between the
two study groups, the paired sample t-test
analysis results had reported positive changes in
the pre to post- training interventions in agility
performance in the plyometric group (p=0.000);
in the explosive power performance, both in the
plyometric (p=0.036) and strength training
groups (p=0.02), even though no performance
(pre- to post-test) change was reported in the
strength training group (p=0.113).
DISCUSSION
Some previous studies agreed with this finding.
Thomas et al. (2009) study the effect of six weeks
237
During the 12 weeks of the study time, the
female soccer players received plyometric
(plyometric group) and strength (strength group)
trainings. At baseline (pre-test) and at the end of
the strength training interventions, the effect of
the training on the explosive power ability of the
female soccer players was tested and the results
were analyzed using paired sample t-test. The
data in table 5 reveals in comparison to the pre-
test players in the plyometric group (who didn’t
Sig.
receive strength training) and strength group
(who received strength trainings) significantly
0.036 improved their explosive power (p<0.05).
0.02
Pre-test Post-test
Mean Difference Sig. Mean Difference
Sig.
0.165 0.019 0.202 0.483
-5.700 0.156 0.600 0.854
of plyometric training and reported increases in
vertical jump height and change in agility. The
study concluded that both depth jump and
CMJ plyometric were worthwhile training
activities for improving power and agility in
young male soccer players.In this study, the
participants were female and the result was a
similar improvement on the athletes’ agility and
explosive power performance.
Taking a look at other literatures, Rubley et
al. (2011), studied the effect of plyometric
training on power and kicking distance in
adolescent female soccer players. From the
study, the plyometric group had a significantly
higher vertical jump after fourteen weeks. In
addition, Adams and O’Shea, (1992), indicated
that both strength and plyometric trainings were
necessary for improving hip and thigh power
production for vertical jumping ability. Based on
these findings, the researcher observed increased
vertical jump and agility.
Herman et al. (2008) studied that strength
training alone does not alter knee and hip
kinematics in female recreational athletes.
Further research is needed to determine the
effect of strength training in combination with
238
other intervention methods on lower extremity
biomechanics and performance improvement.
According to Mirzaei and Norasteh, (2014),
effects of six weeks of depth jump vs.
countermovement jump training on sand on
muscle soreness and performance, there was a
significant effect of jumping trainings or
plyometric trainings on the agility level of
players. This study backed up one of the findings
of the current study: plyometric trainings
improve the agility level of the athletes.
However, the training environment was different
and it may have affected the results.
Also, according to Chu (1998), plyometric is
a training technique used by athletes in all types
of sports to increase strength and explosiveness.
Different studies (Parsons et al.,1998; Yap et al.,
2000; Miller et al., 2001; Young et al., 2001; Craig,
2004)also supported the discussed tables, i.e.
plyometric trainings have a significant change in
the agility of soccer players. In this study, the
researchersalso showed plyometric trainings
bringing significant change on the agility
performance of the players within the allotted
time.
CONCLUSION
Following twelve-week of continuous plyometric
and strength trainings, the study identified that
both training interventions improved the agility
and explosive power ability of female soccer
players and no significant performance
differences had been identified between the two
groups due to the selected two interventions.
REFERENCES
1. Aagaard, P., Simonsen, E. B., Trolle, M., Bangsbo,
J., & Klausen, K. (1994). Effects of different
strength training regimes on moment and
power generation during dynamic knee
extensions. European journal of applied
physiology and occupational physiology, 69(5),
382-386.
2. Acar, M. F., Yapicioglu, B., Arikan, N., Yalcin, S.,
Ates, N., & Ergun, M. (2008). Analysis of
goals scored in the 2006 World Cup.
In Science and football VI (pp. 261-268).
3. Adams, K., O’Shea, J. P., O’Shea, K. L.,
&Climstein, M. (1992).The effect of six weeks
of squat, plyometric and squat-plyometric
training on power production. Journal of
applied sport science research, 6(1), 36-41.
Chalachew Lemecha and Aschenaki Taddese
4. Ball, J. R., Rich, G. Q., & Wallis, E. L. (1964).
Effects of isometric training on vertical
jumping. Research Quarterly. American
Association for Health, Physical Education and
Recreation, 35(3), 231-235.
5. Bangsbo, J., Mohr, M., &Krustrup, P. (2006).
Physical and metabolic demands of training
and match-play in the elite football
player. Journal of sports sciences, 24(07), 665-
674.
6. Beato, M., Bianchi, M., Coratella, G., Merlini, M.,
& Drust, B. (2018). Effects of plyometric and
directional training on speed and jump
performance in elite youth soccer players. The
Journal of Strength & Conditioning
Research, 32(2), 289-296.
7. Baechle, T. R. (1994). Strength training and
conditioning. Human Kinetics: Champaign,
IL.
8. Bloomfield, J., Polman, R., O'Donoghue, P., &
McNaughton, L. (2007). Effective speed and
agility conditioning methodology for random
intermittent dynamic type sports. Journal of
Strength and Conditioning Research, 21(4), 1093.
9. Brughelli, M., Cronin, J., Levin, G., & Chaouachi,
A. (2008). Understanding change of direction
ability in sport. Sports medicine, 38(12), 1045-
1063.
10. Chu, D. A. (1998). Jumping into plyometric.
Human Kinetics.
11. Cissik, J. M. (2002). Basic principles of strength
training and conditioning. NSCA’s
Performance Training Journal, 1(4), 7-11.
12. Davis, B., Bull, R., Roscoe, J., Roscoe, D., & Saiz,
M. (2000). Physical education and the study
of sport (p. 388). London: Mosby.
13. Davies, P. (2005). Total Soccer Fitness. The
Complete Guide To Soccer Conditioning.
14. Getchell, B. (1979). Physical fitness: A way of life.
15. Hamou, M. O., & Zaoual, M. (2017). Physical
Determinants of Success of Algerian Football
Players of U20 Category. Revue Académique
des Études Socialeset Humaines, (17), 25.
16. Handzel, T. M. (2003). Core training for improved
performance. NSCA’s Performance Training
Journal, 2(6), 26-30.
17. Herman, D. C., Weinhold, P. S., Guskiewicz, K.
M., Garrett, W. E., Yu, B., & Padua, D. A.
(2008). The effects of strength training on the
lower extremity biomechanics of female
recreational athletes during a stop-jump
task. The American journal of sports
medicine, 36(4), 733-740. Medicine, 34(3), 165-
180.
18. Jerry R. Ball, George Q. Rich & Earl L.
Wallis (1964) Effects of Isometric Training on
Vertical Jumping Hoff, J., & Helgerud, J.
(2004). Endurance and strength training for
soccer players. Sports, 35(3), 231-235.
SINET: Ethiop. J. Sci., 44(2), 2021
19. Jovanovic, M., Sporis, G., Omrcen, D., &
Fiorentini, F. (2011).Effects of speed, agility,
quickness training method on power
performance in elite soccer players. The
Journal of Strength & Conditioning
Research, 25(5), 1285-1292.
20. Jullien, H., Bisch, C., Largouët, N., Manouvrier,
C., Carling, C. J., & Amiard, V. (2008). Does a
short period of lower limb strength training
improve performance in field-based tests of
running and agility in young professional
soccer players?. The Journal of Strength &
Conditioning Research, 22(2), 404-411.
21. Klavora, P. (2000). Vertical-jump tests: A critical
review. Strength and Conditioning
Journal, 22(5), 70-75.
22. Koutedakis, Y. (1995). Seasonal variation in fitness
parameters in competitive athletes. Sports
Medicine, 19(6), 373-392.
23. Lambert, M. I., Viljoen, W., Bosch, A., Pearce, A.
J., & Sayers, M. (2008).General principles of
training. The olympic textbook of medicine in
sport, 1-48.
24. Laursen, P. B., & Jenkins, D. G. (2002).The
scientific basis for high-intensity interval
training. Sports medicine, 32(1), 53-73.
25. MacDougall, J. D., Wenger, H. A., & Green, H.
(1991).The purpose of physiological
testing. Physiological testing of the high-
performance athlete, 1-5.
26. Meylan, C., & Malatesta, D. (2009). Effects of in-
season plyometric training within soccer
practice on explosive actions of young
players. The Journal of Strength & Conditioning
Research, 23(9), 2605-2613.
27. Miller, M. G., Herniman, J. J., Ricard, M. D.,
Cheatham, C. C., & Michael, T. J. (2006).The
effects of a 6-week plyometric training
program on agility. Journal of sports science &
medicine, 5(3), 459.
28. Mirzaei, b., Norasteh, a. a., de Villarreal, e. s.,
&Asadi, a. (2014). Effects of six weeks of
depth jump vs. countermovement jump
training on sand on muscle soreness and
performance. Kinesiology, 46(1).
29. Reilly, T. (2006). The science of training–Soccer: A
scientific approach to developing strength,
speed and endurance. Routledge.
30. Reilly, T., & Williams, A. M. (Eds.). (2003). Science
and soccer. Routledge.
239
31. Roth, S. M., Ivey, F. M., Martel, G. F., Lemmer, J.
T., Hurlbut, D. E., Siegel, E. L., & Wernick, D.
M. (2001). Muscle size responses to strength
training in young and older men and
women. Journal of the American Geriatrics
Society, 49(11), 1428-1433.
32. Rubley, M. D., Haase, A. C., Holcomb, W. R.,
Girouard, T. J., & Tandy, R. D. (2011). The
effect of plyometric training on power and
kicking distance in female adolescent soccer
players. The Journal of Strength & Conditioning
Research, 25(1), 129-134.
33. Sander, A., Wirth, K., & Schmidtbleicher, D.
(2014). Long-term strength training effects on
change-of-direction sprint performance. The
Journal of Strength & Conditioning
Research, 28(1), 223-231.
34. Sheppard, J. M., & Young, W. B. (2006). Agility
literature review: Classifications, training and
testing. Journal of sports sciences, 24(9), 919-
932.
35. Sood, H. (2013). Physical and performance
correlates of agility in tennis players.
36. Stojanović, E., Ristić, V., McMaster, D. T.,
&Milanović, Z. (2017). Effect of plyometric
training on vertical jump performance in
female athletes: a systematic review and
meta-analysis. Sports Medicine, 47(5), 975-986.
37. Stølen, T., Chamari, K., Castagna, C., &Wisløff, U.
(2005).Physiology of soccer. Sports
medicine, 35(6), 501-536.
38. Taube, W., Leukel, C., &Gollhofer, A. (2012). How
neurons make us jump: the neural control of
stretch-shortening cycle movements. Exercise
and sport sciences reviews, 40(2), 106-115.
39. Thomas, K., French, D., & Hayes, P. R. (2009).The
effect of two plyometric training techniques
on muscular power and agility in youth
soccer players. The Journal of Strength &
Conditioning Research, 23(1), 332-335.
40. Tudor O. Bompa, G. Gregory Haff , Periodization:
theory and methodology of training.--5th ed.
41. Twist, C., &Eston, R. (2005). The effects of
exercise-induced muscle damage on maximal
intensity intermittent exercise
performance. European journal of applied
physiology, 94(5-6), 652-658.
42. Winett, R. A., &Carpinelli, R. N. (2001). Potential
health-related benefits of resistance training.
Preventive medicine, 33(5), 503-513.
240 Chalachew Lemecha and Aschenaki Taddese

APPENDEXS

Training protocols/ interventions / for plyometric and strength groups/ for twelve week

Table 1:- Plyometric trainings protocol for first, second and third weeks.

No. Plyometric Trainings Reps Set Active Rest b/n Set Intensity Duration
1 Box step front jump /40''/ 6-8 3 4' Medium
2 Calf exercise /40''/ 6-8 x 3 3 4' Medium 1:20'
3 Plyo push ups 6-8 3 4' Medium include warming up and cooling down
4 Cross leg movement 6-8 3 4' Medium
5 Tuck jump 6-8 3 4' Medium
6 Box step side jump 6-8 3 4' Medium
7 Burpees 6-8 3 4' Medium

Table 2:- Plyometric trainings protocol for fourth, fifth and sixthweeks.

No. Plyometric Trainings Reps Set Active Rest Intensity Duration

b/n Set
1 Box step front jump 6-8 3 3' Medium/high
2 Calf exercise 6-8 x 3 3 3' Medium/high 1:20'
3 Plyo push ups 6-8 3 3' Medium/high include
4 Cross leg movement 6-8 3 3' Medium/high warming up
5 Tuck jump 6-8 3 3' Medium/high and cooling
6 Box step side jump 6-8 3 3' Medium/high down
7 Burpees 6-8 3 3' Medium/high

Table 3:- Plyometric trainings protocol for seventh, eighth and ninth weeks.

No. Plyometric trainings Reps Set Active Rest Intensity Duration

b/n Set
1 Box step front jump 6-8 3 2' high
2 Calf exercise 6-8 x 3 3 2' high 1:20'
3 Plyo push ups 6-8 3 2' high include warming up and
4 Cross leg movement 6-8 3 2' high cooling down
5 Tuck jump 6-8 3 2' high
6 Box step side jump 6-8 3 2' high
7 Burpees 6-8 3 2' high

Table 4:- Plyometric trainings protocol for ten, eleven and twelfth weeks.

No. Plyometric Trainings Reps Set Active Rest b/n Set Intensity Duration
1 Box step front jump 6-8 4 2' high
2 Calf exercise 6-8 x 3 4 2' high 1:20'
3 Plyo push ups 6-8 4 2' high include warming up and cooling down
4 Cross leg movement 6-8 4 2' high
5 Tuck jump 6-8 4 2' high
6 Box step side jump 6-8 4 2' high
7 Burpees 6-8 4 2' high

Table 5:- Strength trainings protocol for first, second and third weeks.

No. Strength trainings Reps Set Active Rest b/n Intensity Duration
Set
1 Squat 8 3 4' medium 1:20'
2 Bar bell good morning 8 3 4' medium include warming up
3 Lunch 8 3 4' medium and cooling down
4 Crunches 10 3 4' medium
5 Calf exercise 10 3 4' medium
SINET: Ethiop. J. Sci., 44(2), 2021 241

Table 6:- Strength training protocol for fourth, fifth and sixth weeks.

No. Strength Trainings Reps Set Active Rest b/n Set Intensity Duration
1 Squat 8 3 3' Medium/high
2 Bar bell good morning 8 3 3' Medium/high 1:20'
3 Lunch 8 3 3' Medium/high include warming up and cooling down
4 Crunches 10 3 3' Medium/high
5 Calf exercise 10 3 3' Medium/high

Table 7:- Strength training protocol seventh, eighth and ninth weeks.

No. Strength Trainings Reps Set Active Rest b/n Set Intensity Duration
1 Squat 10 3 2' high
2 Bar bell good morning 10 3 2' high 1:20'
3 Lunch 10 3 2' high include
4 Crunches 12 3 2' high warming up
5 Calf exercise 12 3 2' high and cooling
down

Table 8:- Strength training protocol ten, eleven and twelfth weeks.

No. Strength Trainings Reps Set Active Rest b/n Set Intensity Duration
1 Squat 10 4 2' high
2 Bar bell good morning 10 4 2' high 1:20'
3 Lunch 10 4 2' high include
4 Crunches 12 4 2' high warming up
5 Calf exercise 12 4 2' high and cooling
down
 SINET: Ethiop. J. Sci., 33(1):73–74, 2010

GUIDE TO AUTHORS

Research manuscripts in science and technology and related disciplines are considered for publication.
Manuscripts must be in English. Submission of a manuscript to the JOURNAL must be accompanied by a
letter stating that no similar paper, other than an abstract or an oral presentation, has been or will be
submitted for publication elsewhere. Manuscripts should be typed with wide margins (minimum 30 mm)
on good quality A4 (210 x 297 mm) paper using double spacing throughout. Two hard copies and a soft
copy in MS Word should be submitted. E-mail attachment is also possible. Each page of the typescript
must be numbered.

The manuscript must be addressed to: The Editor-in-Chief, PO Box 31226, Addis Ababa, Ethiopia. All
copies should be carefully checked and any corrections clearly indicated.

The first page should contain the full title of the manuscript, the name(s) of the author(s) including
address(es) and the institution(s) in which the research was carried out. If more than one author are
involved, the author to whom all correspondence will be addressed should be indicated by an asterisk.
The first page should also contain the ABSTRACT. The ABSTRACT must not be more than 250 words. Key
words (up to five, and in alphabetical order) should be given next to the ABSTRACT in a separate line.

A desirable, though not strictly prescribed plan, for the organization of a research paper is to start with an
INTRODUCTION giving a description of the problem and its relation to other works in the same field. The
objective(s) of the investigation should also be stated in this section. Abbreviations (e.g., BAP, 6-
Benzylaminopurine; GA3, gibberellic acid 3; etc.) should be put under the INTRODUCTION in a separate
paragraph. The remaining sections can then cover MATERIALS AND METHODS, RESULTS AND DISCUSSION AND
CONCLUSION. Then follows ACKNOWLEDGEMENTS, if any, and the last item would be a list of REFERENCES.

References in the text should have the following form:

- Darwin and Morgan (1993) or, if more than two authors, Andersson et al. (1993).
- (Hartmann and Kester, 1975; Andersson et al., 1993; Darwin and Morgan, 1993) —
chronologically.

Ethiopian names should be in direct order, that is, the full first name followed by the father's name; e.g.,
Solomon Kassa and not Kassa, S.
Listing of references is alphabetical, and should have the form:

- Kalb, J.E. (1978). Miocene to Pleistocene deposits in the Afar depression, Ethiopia. SINET:
Ethiop. J. Sci. 1:87–98.
- Hartmann, H.T. and Kester, D.E. (1975). Plant Propagation: Principles and Practices.
Prentice-Hall, Inc., Englewood Cliffs, New Jersey, pp. 53–66.

To cite some pages in a publication prepared by one or more editors, an acceptable format is:

- Fraser, D.A.S. (1971). Events, information processing, and the structural model (with
discussion). In: Foundations of Statistical Inference, pp. 32–35, (Godame, V.P. and Sprott,
D.A., eds). Holt, Rinehart and Winston, Toronto.

Tables, however small, should bear arabic numerals and be referred to in the text by their numbers, e.g.
`Table 4'. Each table must be typed on a separate sheet and should be placed at the end of the manuscript.
The approximate position of each table should be indicated in the text. Footnotes in tables should be
indicated by superscript letters beginning with `a' in each table. Descriptive material not designated as a
footnote may be placed under a table as a NOTE. Vertical lines should not be used to separate columns;
instead extra space should be left between the columns. Ditto marks should be avoided.

All illustrations should be given separately, not stuck on pages and not folded. They should be numbered
as figures in sequence with arabic numerals. Each figure should have a descriptive legend and this should
be given on a separate sheet after the manuscript. Clear, glossy black and white photographs (100 x 70
mm) can be submitted to the Journal; they should be clearly numbered on the back in pencil.

Only International System of Units (SI) are acceptable. Symbols and nomenclatures should conform to
international recommendations with respect to specific fields of specialization.
 COLUMNS
of
SINET: Ethiopian Journal of Science

Research articles
Report of original research in any branch of science will be considered for
publication in this section.

Review articles
Covering recent advances in an area in which an author has been actively engaged
will be published under this column.

Short communications
News of interest to researchers including progress report on ongoing researches,
records of observations — short comments, correction and reinterpretations of
articles previously published in SINET, etc. Manuscripts submitted for this column
should be prepared following the preceding instructions, and should not be more
than six pages and should not contain more than two figures and/or two tables.

Feature articles
These are articles that may be published in SINET feature columns covering a
range of topics in science education, and science and society.

Science education: Articles related to science education in schools and

colleges, new or simple methods of scientific experiments for science educators,
discussion of science curricula textbooks, teaching aids, etc.

Science and society: Articles related to applications of science to solving daily

problems of human; the scientific method as applied to solving humans'
problems; the role of science experiments in the "Third World", etc., are
considered in this column.

Book review
A critical evaluation of recently published books in basic sciences, science
education, technology and development will be published under this column.
 SINET: ETHIOPIAN JOURNAL OF SCIENCE

SINET is a peer-reviewed, bi-annual journal of science published by the College of Natural y

of Sciences (Formerly Faculty of Science), Addis Ababa University, Ethiopia. The Journal is
designed for an international readership both within Africa and overseas. Since its inception in
1978, SINET has been publishing original research articles, review articles, short
communications and feature articles in basic and applied sciences.

Objectives

 To serve as a communication medium among professionals in science and related fields.

 To publish original scientific works relevant to the solution of Ethiopian as well as global
problems.

 To encourage Ethiopian researchers to align their investigations in the direction of solving

the basic needs of the country.

 To promote the effective teaching of science in schools by publishing newer, better and
more efficient ways of carrying out experiments.

 To serve as a forum for discussing the applications of basic science principles to

development and for the identification of problems of development and their possible
solutions.

All correspondence about back issues, single copies, change of address and claims for missing
issues should be sent to:

The Editorial Manager

SINET Editorial Office
PO Box 31226
Addis Ababa
Ethiopia

Fax: (+251–11) 155 23 50

Tel: (+251–11) 123 26 68
(+251–11) 123 94 74

E-mail: sinet@aau.edu.et
 SINET: Ethiopian Journal of Science
Volume 42, No. 2 (December 2021)

Contents continued from outside front cover

Molecular , cellular and microbial

Child nutritional status, mothers’ nutritional knowledge and practice and Household food security status
in Tehuledere Woreda, South Wollo, Ethiopia……………….……………………………………………………161-171
Ahmed Indris, Dilu Shaleka and Mogessie Ashenafi

Zoological science

Seasonal variation of non-volant small mammals in Gibe Sheleko national park, Southwestern
Ethiopia………………………………………………………………………………………………….………..172-181
Seyoum Kiros and Afework Bekele

Diurnal activity patterns, habitat use and foraging habits of Egyptian goose (Alopochena egyptiacus
Linnaeus, 1766) in the Boyo wetland, southern Ethiopia ………………………………………………….182-192
Mulugeta Kassa, M. Balakrishnan and Bezawork Afework

Survey and identification of termites (Insecta, Isoptera) using morphological and molecular methods
from eastern, central and western Ethiopia ……………………………………………………….……….193-204
Ashenafi Kassaye , Emana Getu , Mulatu Wakgari , Muluken Goftishu , Awol Seid Samantha J.
Montoya 3 and Gillian H. Gile

Chemistry

Determination of selected major and trace metals in lemongrass (Cymbopogon citratus) by microwave
plasma-atomic emission spectrometry…………………………………………………………………….205-214
Nitsuh Birhanu, Weldegebriel Yohannes and Bhagwan Singh Chandravanshi

Geology

Early hunters and herders of northern Ethiopia: The fauna from Danei Kawlos……..…………….215-222
Agazi Negash and Fiona Marshall

Information technology

An Integrated Knowledge-Based System for Early Detection of Eye Refractive Error Using Data
Mining……………………………………………………………………..…………………………….…….223-233
Teklay Birhane, Dr. Dereje Teferi and Tariku Mohammed

Sport science
Effects of plyometric and strength trainings on selected physical fitness variables in Ethiopia youth sport
academy female soccer players………………………………………………..……………………….…….234-241
Chalachew Lemecha1 and Aschenaki Taddese2,*

© College of Natural and computational Sciences, Addis Ababa University