ORIGINAL ARTICLE Genetic Resources (2023), 4 (7), 56–67 DOI: 10.46265/genresj.IXPJ9541 https://www.genresj.org ISSN: 2708-3764 Phenotypic diversity between and within Harar and Ogaden cattle breeds in eastern Ethiopia: The first step for conservation Amine Mustefa *,a, Tesfalem Aseged a, Seble Sinkie a, Fasil Getachew b, Tesfu Fekensa a, Manaye Misganaw a and Abebe Hailu a a Ethiopian Biodiversity Institute, Addis Ababa, Ethiopia b International Livestock Research Institute, Addis Ababa, Ethiopia Abstract: Fourteen morphological and eight morphometric traits on a total of 456 adult cattle (354 cows and 102 oxen) from 5 purposively selected districts were recorded to characterize the cattle populations in eastern Ethiopia. Frequency, general linear model and multivariate analysis procedures of Statistical Analysis Software (SAS 9.0) were used to analyze the data. The morphological similarities among the cattle populations from Tulo, Jarso and Fedis districts confirmed their uniqueness and belonging to Harar cattle. Morphological similarities were also observed between the cattle populations from Jigjiga and Kebri Beyah districts, which categorize them as Ogaden cattle. The Harar cattle were characterized by their forward-oriented, widely spaced, and medium-length horns while the Ogaden cattle possessed upright-oriented, narrowly spaced, short horns. The majority of the Ogaden cattle had grey body colour while the Harar cattle had multiple body colours with red combined with white and black observed frequently. Most of the morphometric measurement values were higher for the oxen, while the cows had longer horns. Moreover, Ogaden cattle had higher morphometric parameters than Harar cattle. According to the multivariate analysis results, the cattle populations in the study area were separated into two breeds – Harar and Ogaden. However, these results only showed phenotypic differences, which might not necessarily be due to genetic differences. Therefore, further molecular characterization is recommended to understand their level of relationships which will help to design appropriate conservation and breeding programmes. Keywords: Breeds, Characterization, Indigenous, Morphology, Morphometric, Multivariate Citation: Mustefa, A., Aseged, T., Sinkie, S., Getachew, F., Fekensa, T., Misganaw, M., Hailu, A. (2023). Phenotypic diversity between and within Harar and Ogaden cattle breeds in eastern Ethiopia: The first step for conservation. Genetic Resources 4 (7), 56–67. doi: 10.46265/genresj.IXPJ9541. © Copyright 2023 the Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Introduction The cattle population size of Ethiopia, 70.3 million head (Central Statistical Agency, 2021), is the largest in Africa and more than double the cattle population registered by Chad (32.2 million) which is the second country with high cattle population (Statista, 2020). Cattle genetic resources, the major contributor to the livestock sector, serve as sources of milk, meat, draught power, hide, manure, nutrient recycling and foreign exchanges for Ethiopia (Central Statistical Agency, ∗Corresponding author: Amine Mustefa (aminemustefa32@gmail.com) 2021). Due to the very important role cattle genetic resources play in the economy of the country, various diversity and genetic improvement studies have been made so far to ensure their sustainable utilization. Diversity studies in animal genetic resources are important to better understand the breed and design appropriate breeding programmes for current and future research and development works (FAO, 2012). Variation within and among breeds is among the key inputs in genetic improvement and conservation programmes; the more diverse the population the more likely it leads to bring genetic improvement (Falconer, 1989). Identification, phenotypic and genetic characteriza- tion, and advanced performance evaluation can help Received: 27.01.2023 Accepted: 09.06.2023 Published online: 04.07.2023 https://www.genresj.org https://www.dx.doi.org/10.46265/genresj.IXPJ9541 https://www.genresj.org https://www.dx.doi.org/10.46265/genresj.IXPJ9541 mailto:aminemustefa32@gmail.com Genetic Resources (2023), 4 (7), 56–67 Characterization of cattle in eastern Ethiopia 57 us understand the diversity of a given population. The first diversity assessment of Ethiopian cattle genetic resources was made by the International Livestock Cen- tre for Africa (ILCA), now the International Livestock Research Institute (ILRI), in 1992 (Rege, 1999; Rege and Tawa, 1999). It sought to determine the status and compile information on the characteristics of cattle. It does not constitute the level of assessment required to make decisions on use and conservation, as the informa- tion was incomplete due to its very broad brush survey. However, it provides a basis for additional, targeted sur- veys (Rege, 1999; Rege and Tawa, 1999; Mustefa et al, 2021). According to Ethiopia’s focal point for genetic resources, the Ethiopian Biodiversity Institute, Ethiopia possesses 28 indigenous cattle breeds; Abergelle, Anuak, Adwa, Afar, Ambo, Arado, Arsi, Bale, Begait, Begaria, Borena, Fogera, Gofa, Gojam Highland Zebu, Guraghe, Hammer, Harar, Horro, Irob, Jem-Jem, Jijiga, Kereyu, Medenes, Mursi, Ogaden, Raya, Sheko, and Smada (Rege and Tawa, 1999; EBI, 2016). Among the listed cattle breeds, some (Adwa, Ambo, Bale, Guraghe, Hammer, Harar, Jem-Jem, Jijiga, and Smada) do not have a detailed description of their physical appearance. Therefore, it is important to consider filling the missing information to capture a country-wide picture. According to Getachew et al (2014), Ogaden cattle are thought to be a variety of the Borena cattle found in the Ogaden area of the Somali region of Ethiopia and bordering East Hararghe. They have a well-developed hump, large dewlap and short horns. They are mainly kept for milk production but are good beef animals. The Ogaden cattle were also reported to be distributed up to Jigjiga and parts of East Hararghe (Rege and Tawa, 1999; Mengesha, 2019). On the other hand, according to the reports of Rege and Tawa (1999), the Harar cattle are found in the East and West Hararghe plateau. They have short, thick horns and well-developed dewlaps. The common coat colours are black, roan and red and are used primarily for draught. These two cattle breeds are found adjacent to one another. To quantify the level of relationship between them as well as to know the available cattle diversity in that area, phenotypic characterization is required. Moreover, phenotypic characterization is a crucial first step for in situ conservation. The current study aims to phenotypically characterize these two cattle breeds and to quantify the level of relationship between them. Materials and methods Study areas This study was conducted in five districts of three zones in two regions: Tulo district of West Hararghe zone, Jarso and Fedis districts of East Hararghe zone of Oromia region and Jigjiga and Kebri Beyah districts of Fafen zone of Somali region (Figure 1). The weather conditions, altitude, ethnicity, and human and cattle population sizes of the sampled districts are presented in Table 1. Site selection According to Rege and Tawa (1999), Harar cattle were found in East and West Hararghe zones of the Oromia region. Similarly, Getachew et al (2014) reported the distribution of Ogaden cattle to be the Somali region and adjacent areas of eastern Oromia. Therefore, the current study took representative samples from the two breeds of cattle. Accordingly, for the Harar cattle, Tulo district was randomly selected from West Hararghe zone while Jarso and Fedis districts were randomly selected from East Hararghe zone. On the other hand, Jigjiga and Kebri Beyah districts were selected randomly from Fafen zone to represent Ogaden cattle. The kebeles (sampling sites) within each district were categorized into highland and lowland areas. One kebele from the highland and one kebele from the lowland areas were selected randomly. Households were also randomly selected from each kebeles (sampling sites) and 2–3 adult animals (four years old and above) were sampled randomly from each household. Data collection Data collection procedures were adapted from the FAO guidelines for the Phenotypic Characterization of Animal Genetic Resources (FAO, 2012). Sampled cattle were carefully handled by trained labourers and their owners. To avoid measurement error, the cattle were made to stand properly on flat grounds with parallel legs. Four researchers were involved in the data collection: two for the quantitative and two for the qualitative data recording. Measurements were carried out by the same researcher throughout the study to minimize subjectivity errors. Data recording was carried out early in the morning before the animals were fed and watered. Textile measurement tape in a centimetre unit was used to record quantitative data. Eight morphometric/quantitative (Table 2) and fourteen morphological/qualitative traits (Table 3, Figure 2) were recorded on a total of 456 adult cattle (354 females and 102 males). Data analysis Univariate analysis Data entry and management were performed using Microsoft Excel© worksheet (Microsoft Office 2016). UNIVARIATE procedure of Statistical Analysis Software (SAS) 9.0 (SAS Institute, 2002) was used to test the normality of the morphometric data. Data on morphological/qualitative traits were subjected to chi- square (χ2) tests of the frequency (FREQ) procedure of SAS 9.0 software (SAS Institute, 2002) . Qualitative data analysis was performed using the following model by fitting sex, breed and location as class variables. Yijk = µ + Si + Bj + Lk + eijk where Yijk is an observation, µ is the overall mean, Si is the fixed effect of ith sex, Bj is 58 Mustefa et al Genetic Resources (2023), 4 (7), 56–67 Figure 1. Map of the study areas the fixed effect of jth breed, Lk is the fixed effect of kth location, and eijk is the random error attributed to the nth observation. On the other hand, morphometric data analysis was performed separately for males and females by fitting location and breed as fixed variables. Morphometric data were analyzed using the general linear model (GLM) procedure of SAS 9.0 software, with adjusted Tukey-Kramer test to separate the least square means (LSM). Morphometric data analysis was performed using the following model: Yi = µ + Li + ei where Yi is an observation, µ is the overall mean, Li is the fixed effect of ith location, and ei is the random error attributed to the nth observation. Yij = µ+ Bi + Lj + eij where Yij is an observation, µ is the overall mean, Bi is the fixed effect of ith breed, Lj is the fixed effect of jth location, and eij is the random error attributed to the nth observation. Multivariate analysis Quantitative measurements that better discriminate the cattle populations from different locations were identified using the forward selection method of the stepwise discriminant function analysis (STEPDISC) procedure of SAS 9.0. The discriminant function analysis (DISCRIM) procedure of SAS 9.0. was also used to assign observations to locations and evaluate probabilities of misclassifications. A linear combination Table 1. Weather and population-related information of the selected districts (City Population, 2007; Keskes et al, 2013; EHZLFDO, 2018; WHZLFDO, 2018; Ayana, 2019; UNHCR, 2020; Kebede and Utta, 2021; Abdi and Shiferaw, 2022) Agroecology Tulo Jarso Fedis Jigjiga Kebri Beyah Human population projection 2022 215,337 165,712 161,214 417,688 242,880 Area (km2) 430.6 515.1 720.8 2,859 4,839 Temperature(oC) 17–32 12–25 14–28 16–29 16–27 Rainfall (mm) 600–900 400–900 516.3 300–500 582.4 Altitude (m.a.s.l.) 1,600–2,400 1,500–3,060 1,702 500–1,600 1,530 Cattle population 131,643 76,873 - - - Ethnicity Oromo Oromo Oromo Somali Somali Genetic Resources (2023), 4 (7), 56–67 Characterization of cattle in eastern Ethiopia 59 Table 2. Description of the collected morphometric traits. Adapted from FAO (2012). Measurement was performed using a centimetre (cm) unit. No. Morphometric traits Definitions 1 Body length Horizontal length from the point of the shoulder to the pin bone 2 Heart girth Measurement around the animal right behind its front legs 3 Height at withers Height from the bottom of the front foot to the highest point of the withers 4 Pelvic width Horizontal distance between the extreme lateral points of the hook bone (tuber coxae) of the pelvis 5 Muzzle circumference Circumference of the mouth a little above the nostrils and around the point where the dewlap meets the chin 6 Ear length Length of the back side of the ear from the root to the tip 7 Horn length The longest distance from the root of the horn to its tip along the outer curvature 8 Cannon bone length Distance from the lateral tuberculum of the os metacarpale IV to the fetlock joint of morphometric measurements that provide maximal separations between locations was performed using the canonical discriminant function analysis (CANDISC) procedure of SAS 9.0. The scored canonical variables were used to plot pairs of canonical variables to get a visual interpretation of location differences. Pairwise squared Mahalanobis distances between locations were computed as: D2 (i|j) = (xi − xj) ′ cov−1 (xi − xj). Where D2 (i|j) is the distances between locations i and j, cov−1 is the inverse of the covariance matrix of measured variables, xi and xj are the means of variables in the ith and jth populations. Results Morphological traits The qualitative characteristics of both sexes (male and female) and both breeds (Harar and Ogaden) along with their chi-square values and levels of significance are presented in Table 3. Accordingly, sex and breed significantly (p < 0.05) affected the qualitative characteristics of the cattle populations. The majority of the males had straight-shaped lateral and upright horn orientation, shaded body colour pattern, large hump size located at thoracic position, and large dewlap width. On the other hand, the majority of the females had curved and forward-oriented horn, uniform body colour pattern, small hump size located at cervico- thoracic position, and medium dewlap width. Moreover, the studied cattle breeds showed signifi- cantly different qualitative characteristics. The majority of Harar cattle had wide, curved and forward-oriented horn. They also possessed a sloppy rump with a small to medium hump located at the cervivo-thoracic posi- tion (Figure 3A). On the other hand, the Ogaden cattle possessed both narrow and wide horn spacing, straight and curved horn shapes, as well as flat and sloppy back profiles at an equivalent ratio. They also possessed an upright-oriented horn and a small hump placed at tho- racic position. The majority of the Ogaden cattle also had a long tail with no naval flap (Figure 3B). The qualitative characteristics of the cattle popula- tions from the different locations along with their chi- square values and levels of significance are presented in Table 4. All the studied qualitative characteristics of the cattle populations were significantly (p < 0.05) affected by location differences. Accordingly, almost all the cattle populations from Tulo and Jarso districts pos- sessed wide horn spacing, while this was comparably narrow in cattle from Jigjiga and Kebri Beyah districts. The horn shape of the majority of the cattle popula- tions was curved while it was straight in cattle from Kebri Beyah district. Upright horn orientation was domi- nantly observed in cattle from Jigjiga and Kebri Beyah districts while it was oriented forward in cattle from Tulo and Jarso districts. The hump of the cattle popula- tions from Jigjiga and Kebri Beyah districts was located at the thoracic position while the hump of most of the cattle populations from the other locations was located at the cervico-thoracic position. Flat-rump profile was observed in the majority of the cattle populations from Kebri Beyah district while all the cattle from Tulo, Jarso and Fedis districts had a sloppy rump. The majority of the cattle populations possessed a uniform body colour pattern, erected small hump, and straight back profile. The body colour of the cattle populations disaggre- gated into male and female is presented in Figure 2A. Higher proportion of grey, red, white, and red + white body colours were observed in females than the males. On the other hand, black body colour and its mix with other colours (i.e. white + black and red + black) were observed frequently in the males. Body colour disaggregated by cattle breeds (Harar and Ogaden) is presented in Figure 2B and representa- tive examples shown in Figure 3. Grey body colour was observed predominantly in Ogaden cattle while several body colours were observed frequently in Harar cattle. The body colour of the cattle populations from different locations is presented in Figure 2C. Location affected the body colour of the cattle populations significantly (p < 0.0001). Grey body colour was dominantly observed in Jigjiga and Kebri Beyah districts while cattle in Tulo and Jarso districts were predominantly red and the cattle population from Fedis district had white body colour followed by grey. 60 Mustefa et al Genetic Resources (2023), 4 (7), 56–67 Figure 2. A: Effect of sex on body colour (chi-square value 93.2, p < 0.0001); B: Effect of breed on body colour (chi-square value 161.5, p < 0.0001); C: Effect of location on body colour (chi-square value 275.8, p < 0.0001). Genetic Resources (2023), 4 (7), 56–67 Characterization of cattle in eastern Ethiopia 61 Table 3. Percentages of qualitative characteristics of cattle populations by sex and breed. N, number of animals sampled; *, p < 0.05; **, p < 0.01; ***, p < 0.0001; NS: not significant. Qualitative traits Sex Breed Male Female X2 value P Harar Ogaden X2 value P N 102 354 293 163 Horn spacing Narrow 38.2 22.0 10.9 ** 11.6 50.9 84.9 *** Wide 61.8 78.0 88.4 49.1 Horn shape Straight 65.3 24.0 61.9 *** 26.6 45.4 16.6 *** Curved 34.3 76.0 73.4 54.6 Horn orientation Lateral 42.1 12.2 56.5 *** 25.2 7.4 126.3 *** Upright 36.3 33.6 15.7 67.5 Forward 19.6 44.9 50.2 19.6 Dropping 2.0 9.3 8.9 5.5 Colour pattern Uniform 26.5 71.7 78.3 ** 55.0 73.6 20.3 ** Spotty 2.0 3.7 5.1 0 Pied 9.8 4.8 6.1 5.5 Shaded 61.7 19.8 33.8 20.9 Hump shape Erect 86.3 100 50.1 *** 95.2 100 8.0 ** Dropping 13.7 0.0 4.8 0.0 Hump size Small 12.7 78.0 232.7 *** 55.6 77.3 50.7 *** Medium 35.3 21.5 35.2 5.5 Large 52.0 0.5 9.2 17.2 Hump position Thoracic 85.3 44.1 54.1 *** 30.0 95.1 178.1 *** Cervico-thoracic 14.7 55.9 70.0 4.9 Back profile Straight 96.1 88.1 5.5 * 89.1 91.4 0.6 NS Curved 3.9 11.9 10.9 8.6 Rump profile Flat 29.4 12.7 16.1 *** 0.0 46.0 161.4 *** Sloppy 70.6 87.3 100 54.0 Tail length Short 6.9 7.9 5.3 NS 10.2 3.1 31.4 *** Medium 24.5 35.9 40.3 20.9 Long 68.6 56.2 49.5 76.0 Naval flap width Absent - 41.0 NA NA 30.0 61.3 44.4 *** Small - 42.4 45.7 36.3 Medium - 12.4 17.8 2.4 Large - 4.2 6.5 0.0 Perpetual sheath Absent 7.8 - NA NA 0.0 20.5 50.1 *** Small 27.5 - 11.1 53.9 Medium 62.7 - 88.9 20.5 Large 2.0 - 0.0 5.1 Dewlap width Small 2.0 22.9 65.7 *** 2.4 46.6 143.6 *** Medium 43.1 59.6 63.1 43.0 Large 54.9 17.5 34.5 104 Morphometric measurements Least square means with their respective standard errors (LSM ± SE) and pairwise comparison of the morphometric measurements of both cattle breeds disaggregated by sex are presented in Table 5. Overall, males of each breed and location had higher size- related linear body measurements than their female counterparts. Both the Ogaden male and female cattle had higher measurements for most of the morphometric parameters (i.e. heart girth, height at withers, pelvic width, muzzle circumference, and canon bone length). Some traits (horn, ear and body length) of the two breeds were sex dependent. Within the females, Harar cows had longer horns while in Ogaden cattle, males possessed longer horns. Similarly, Harar cows’ body length was significantly higher than their counterparts from Ogaden while males’ body length did not differ significantly. Moreover, Ogaden oxen’s ear length was significantly higher than their counterparts from Harar while the value was not significantly different between the cows. 62 Mustefa et al Genetic Resources (2023), 4 (7), 56–67 Table 4. Percentages of qualitative characteristics of cattle populations from different locations. N, number of animals sampled; *, p < 0.05; **, p < 0.01; ***, p < 0.0001; NS, not significant. Qualitative traits Tulo Jarso Fedis Jigjiga Kebri Beyah X2 value P N 95 103 95 97 66 Horn spacing Narrow 2.1 5.8 27.4 46.4 57.6 106.1 *** Wide 97.9 94.2 72.6 53.6 42.4 Horn shape Straight 21.0 20.0 39.0 38.1 56.1 31.9 *** Curved 79.0 80.0 61.0 61.9 43.9 Horn orientation Lateral 25.3 20.4 30.5 10.3 3.0 152.5 *** Upright 10.5 13.6 23.2 62.9 74.2 Forward 61.0 56.3 32.6 20.6 18.2 Dropping 3.2 9.7 13.7 6.2 4.6 Colour pattern Uniform 50.5 46.6 68.4 74.2 72.7 33.8 ** Spotty 5.3 5.8 4.2 0 0 Pied 8.4 6.8 3.2 7.2 3.0 Shaded 35.8 40.8 24.2 15.6 24.3 Hump shape Erect 89.5 97.1 99.0 100 100 24.2 *** Dropping 10.5 2.9 1.0 0 0 Hump size Small 56.9 66.0 43.1 82.5 69.7 74.5 *** Medium 34.7 22.3 49.5 4.1 7.6 Large 8.4 11.7 7.4 13.4 22.7 Hump position Thoracic 27.4 26.2 36.8 91.8 100 181.8 *** Cervico-thoracic 72.6 73.8 63.2 8.2 0 Back profile Straight 93.7 97.1 75.8 86.6 98.5 34.7 *** Curved 6.3 2.9 24.2 13.4 1.5 Rump profile Flat 0 0 0 38.1 57.6 172.1 *** Sloppy 100 100 100 61.9 42.4 Tail length Short 12.6 4.9 13.7 3.1 3.0 40.4 *** Medium 45.3 39.8 35.8 19.6 22.7 Long 42.1 55.3 50.5 77.3 74.3 Naval flap width Absent 23.4 32.4 34.2 70.5 45.6 62.3 *** Small 57.1 43.3 36.7 28.2 50.0 Medium 16.9 16.2 20.2 1.3 4.4 Large 2.6 8.1 8.9 0 0 Perpetual sheath Absent 0 0 0 31.6 10.0 70.0 *** Small 33.3 3.5 0 42.1 65.0 Medium 66.7 96.5 100 15.8 25.0 Large 0 0 0 10.5 0 Dewlap width Small 2.1 2.9 2.1 57.7 30.3 207.8 *** Medium 76.8 70.9 41.1 29.9 62.1 Large 21.1 26.2 56.8 12.4 7.6 Least square means with their respective standard errors (LSM ± SE) and pairwise comparison of the morphometric measurements disaggregated by the five locations for both sexes are presented in Table 6. Most of the morphometric measurements were significantly affected by the location of the cattle populations. Significantly higher heart girth, pelvic width, muzzle circumference and cannon bone length values were observed for the populations from Jigjiga and Kebri Beyah districts while the horn length of Tulo and Jarso cows was significantly higher than the others. The shortest horn was registered in oxen from Fedis district. Multivariate analysis for discrimination of cattle populations Cannon bone length, horn length and pelvic width were the three most important morphometric variables used in discriminating the cattle populations from different locations (Table 7). These results were confirmed by Wilk’s lambda test where the selected variables made a highly significant (p < 0.0001) contribution in discriminating the cattle populations (Table 7). Results of a location-wise discriminant function analysis (Table 8) show the overall classification of individual animals into their location (population). The Genetic Resources (2023), 4 (7), 56–67 Characterization of cattle in eastern Ethiopia 63 Table 5. Least square means (LSM ± SE) and pairwise comparisons of the morphometric measurements of both cattle breeds under both sexes. N, number of animals sampled; BL, Body length; HG, Heart girth; HW, Height at withers; PW, Pelvic width; MC, Muzzle circumference; EL, Ear length; HL, Horn length; CBL, Cannon bone length. *, p < 0.05; **, p < 0.01; ***, p < 0.0001; NS, not significant. Traits Females Males Harar Ogaden P Harar Ogaden P N 230 124 63 39 BL 106.1 ± 0.38 104.2 ± 0.52 ** 108.6 ± 0.91 110.6 ± 1.16 NS HG 140.0 ± 0.53 149.1 ± 0.72 *** 143.8 ± 1.17 165.6 ± 1.49 *** HW 112.3 ± 0.30 113.6 ± 0.40 * 115.2 ± 0.73 120.9 ± 0.93 *** PW 35.3 ± 0.16 38.5 ± 0.22 *** 33.5 ± 0.38 40.6 ± 0.48 *** MC 38.6 ± 0.13 40.0 ± 0.17 *** 40.1 ± 0.30 44.6 ± 0.38 *** EL 17.9 ± 0.11 17.8 ± 0.16 NS 18.0 ± 0.19 17.4 ± 0.24 * HL 20.8 ± 0.54 17.2 ± 0.73 *** 13.1 ± 0.72 15.9 ± 0.92 * CBL 20.6 ± 0.12 27.7 ± 0.16 *** 21.0 ± 0.22 27.6 ± 0.28 *** Table 6. Least square means (LSM ± SE) in centimetre units and pairwise comparisons of the morphometric measurements of the cattle populations from different locations by sex. N, number of animals sampled; BL, Body length; HG, Heart girth; HW, Height at withers; PW, Pelvic width; MC, Muzzle circumference; EL, Ear length; HL, Horn length; CBL, Cannon bone length. *, p < 0.05; **, p < 0.01; ***, p < 0.0001; NS, not significant. Traits Location P Tulo Jarso Fedis Jigjiga Kebri Beyah Females N 77 74 79 78 46 BL 106.7 ± 0.66a 106.1 ± 0.67ab 105.5 ± 0.65ab 104.0 ± 0.65b 104.5 ± 0.85ab * HG 142.8 ± 0.90b 139.0 ± 0.92c 138.3 ± 0.89c 147.9 ± 0.89a 151.2 ± 1.16a *** HW 112.9 ± 0.51ab 112.4 ± 0.52ab 111.6 ± 0.50b 113.4 ± 0.51ab 113.9 ± 0.66a * PW 35.5 ± 0.27c 35.6 ± 0.28c 34.7 ± 0.27c 37.7 ± 0.69b 39.7 ± 0.35a *** MC 38.9 ± 0.22b 38.49 ± 0.22b 38.3 ± 0.22b 39.9 ± 0.22a 40.1 ± 0.28a *** EL 18.3 ± 0.19a 17.5 ± 0.20bc 18.0 ± 0.19ab 18.1 ± 0.19ab 17.1 ± 0.25c ** HL 25.0 ± 0.86a 21.9 ± 0.88a 15.8 ± 0.85b 17.8 ± 0.86b 16.0 ± 1.12b *** CBL 21.1 ± 0.20c 20.7 ± 0.20cd 20.1 ± 0.20d 28.1 ± 0.20a 27.0 ± 0.26b *** Males N 18 29 16 19 20 BL 107.9 ± 1.69 110.4 ± 1.33 106.1 ± 1.79 109.16 ± 1.64 111.95 NS HG 146.9 ± 2.10b 145.2 ± 1.66bc 138.0 ± 2.23c 163.6 ± 2.05a 167.5 ± 1.99a *** HW 115.2 ± 1.34b 116.9 ± 1.05ab 112.3 ± 1.42b 120.5 ± 1.3a 121.3 ± 1.27a *** PW 33.6 ± 0.70b 34.1 ± 0.55b 32.3 ± 0.74b 40.0 ± 0.68a 41.2 ± 0.66a *** MC 40.8 ± 0.55b 40.5 ± 0.43b 38.8 ± 0.58b 44.6 ± 0.53a 44.6 ± 0.52a *** EL 18.6 ± 0.35a 17.8 ± 0.28ab 17.8 ± 0.37ab 17.8 ± 0.34ab 16.9 ± 0.33b * HL 16.1 ± 1.19a 14.6 ± 0.93a 6.9 ± 1.26b 16.2 ± 1.15a 15.7 ± 1.12a *** CBL 21.7 ± 0.39c 20.8 ± 0.30c 20.4 ± 0.41c 28.6 ± 0.38a 26.7 ± 0.37b *** Table 7. Order of traits used in discriminating the cattle populations from different locations. Step Variables entered Partial R-Square F value Pr > F Wilks’ Lambda Pr < Lambda 1 Cannon bone length 0.7927 431.02 < 0.0001 0.2073 < 0.0001 2 Horn length 0.1445 19.00 < 0.0001 0.1773 < 0.0001 3 Pelvic width 0.1495 19.73 < 0.0001 0.1508 < 0.0001 4 Body length 0.1080 13.56 < 0.0001 0.1345 < 0.0001 5 Ear length 0.0884 10.83 < 0.0001 0.1226 < 0.0001 6 Heart girth 0.0436 5.08 0.0005 0.1173 < 0.0001 7 Height at withers 0.0362 4.18 0.0025 0.1130 < 0.0001 8 Muzzle circumference 0.0159 1.79 0.1290 0.1112 < 0.0001 64 Mustefa et al Genetic Resources (2023), 4 (7), 56–67 Figure 3. A: Representative Harar cattle bull; B: Representa- tive Ogaden cattle cows. Photos: Amine Mustefa overall analysis shows medium (61%) classification of individuals into their corresponding location with an error rate of 39%. Furthermore, the highest (69.7%) classification of individuals into their location was observed in Kebri Beyah district while the lowest (45.6%) classifications were recorded in Jarso district. Results of a breed-wise discriminant function analysis (Table 9) show the overall classification of individual animals into their breed. The overall analysis shows high (99%) classification of individuals into their corresponding breed with a small error rate of 1%. Furthermore, a higher (99.39%) classification of individuals into their breed was observed in Ogaden cattle than Harar cattle (98.63%). Location-wise pairwise squared Mahalanobis dis- tances are presented in Table 10. The distances were highly significant (p < 0.0001). The shortest distance (0.77) was obtained between Tulo and Jarso popula- tions, while Fedis and Jigjiga populations were most distantly related (27.34). The breed-wise analysis also showed long Mahalanobis distance between the two breeds (22.15, p < 0.0001). Location wise, multivariate statistics outputs showed the significance of Can 1 due to its high eigenvalue (5.3) and proportion (93%) to discriminate the cattle popu- lations from the different locations. Can 2–Can 4 com- bined had only 7% proportion in discriminating the cat- tle populations with significantly low eigenvalue (0.37 combined). Similarly, Can 1 significantly discriminated the breeds during the breed-wise analysis with high eigenvalue (5.1) and 100% proportion. Therefore, the outputs of Can 1 are indicative and significant. Location- wise and breed-wise plot of the first two canonical vari- Figure 4. Location-wise (A) and breed-wise (B) plot of canonical discriminant analysis. Districts in A) are indicated by numbers: 1, Tulo; 2, Jarso; 3, Fedis; 4, Jigjiga; 5, Kebri Beyah. Breeds in B) are indicated by Ha, Harar and Og, Ogaden. ables to discriminate the cattle populations is presented in Figure 4. The cattle populations from Tulo, Jarso and Fedis districts were inseparable and in the same group; similarly, the populations from Jigjiga and Kebri Beyah districts were also inseparable and placed in the same group. However, those two groups were clearly sepa- rated from each other. Discussion Qualitative morphological traits can help to easily dif- ferentiate breeds. The observed qualitative characteris- tic similarities among the cattle populations from Tulo, Jarso and Fedis districts support the presence of a unique cattle breed (the Harar cattle) in East and West Hararghe zones of Oromia region. This is also backed by the report of Rege and Tawa (1999), which stated the East and West Hararghe zones of Oromia region as the breeding tract of Harar cattle. Similarly, morphological similarities shared between the cattle populations from Jigjiga and Kebri Beyah districts and their variation from the previ- ous group were also reported by Getachew et al (2014) who characterized them as Ogaden cattle. These quali- tative results confirm the presence of two cattle breeds (Harar and Ogaden) in the eastern part of the coun- try. In line with different publications including Mustefa et al (2021) on Raya cattle and Terefe et al (2015) on Mursi cattle, the results of the current study also con- firm the existence of within-breed variations besides the between-breed differences. The Ogaden cattle possess a relatively unique body colour compared to the Harar cattle, which has multiple body colours. The presence of multicolour (red, red + Genetic Resources (2023), 4 (7), 56–67 Characterization of cattle in eastern Ethiopia 65 Table 8. Number (and percent) of observations classified into locations. From district Tulo Jarso Fedis Jigjiga Kebri Beyah Total Tulo 54 (56.84) 27 (28.42) 12 (12.63) 1 (1.05) 1 (1.05) 95 (100) Jarso 28 (27.18) 47 (45.63) 27 (26.21) 0 1 (0.97) 103 (100) Fedis 15 (15.79) 17 (17.89) 63 (66.32) 0 0 95 (100) Jigjiga 0 0 1 (1.03) 64 (65.98) 32 (32.99) 97 (100) Kebri Beyah 0 0 0 20 (30.30) 46 (69.70) 66 (100) Error rate 0.4316 0.5437 0.3368 0.3402 0.3030 0.3911 Priors 0.2 0.2 0.2 0.2 0.2 Table 9. Number (and percent) of observations classified bybreed. From breed Harar Ogaden Total Harar 289 (98.63) 4 (1.37) 293 (100) Ogaden 1 (0.61) 162 (99.39) 163 (100) Error rate 0.0137 0.0061 0.0099 Priors 0.5 0.5 Table 10. Pairwise squared Mahalanobis distances between locations. ***, p < 0.0001 From District Tulo Jarso Fedis Jigjiga Kebri Beyah Tulo 0 - - - - Jarso 0.77 *** 0 - - - Fedis 2.09 *** 0.87 *** 0 - - Jigjiga 22.17 *** 24.13 *** 27.34 *** 0 - Kebri Beyah 20.23 *** 21.20 *** 24.47 *** 1.80 *** 0 white, red + black, white + black, white, and grey) cattle in Harar might be due to the relatively highland- dominated areas of its distribution, especially the Tulo and Jarso districts. The frequently observed grey body in Ogaden cattle was due to the agropastoralists’ preferences and selection of criteria for that particular coat colour (Getachew et al, 2014). Getachew et al (2014) also related the uniformly patterned grey body colour with their adaptation mechanism to the arid and semi-arid agroecologies of the Ogaden rangelands. This was in line with the report of Titto et al (2016), who reported animals with light coat colouring absorb less heat than those with darker coats. Alongside the most observable qualitative character- istics, morphometric measurements can also produce more reliable information in characterizing and differ- entiating cattle breeds. The above grouping made by the qualitative characteristics of the cattle populations was also supported by the results of the quantitative measurements. Significantly higher values of heart girth, pelvic width, muzzle circumference and cannon bone length with shorter horns were observed in the Ogaden cattle (Jigjiga and Kebri Beyah districts) than in the Harar cattle. Like for the qualitative results, within-breed varia- tions were also observed in the morphometric measure- ments. Within Ogaden cattle, values of heart girth and pelvic width measurements were significantly higher for the population from Kebri Beyah district than Jigjiga dis- trict. These results made the population from Jigjiga dis- trict relatively closer to the Harar cattle. This might be due to the presence of mid- and high-altitude areas of Jigjiga district compared to the Kebri Beyah district, as well as the closeness in ground distance of the Jigjiga district to the distribution areas of Harar cattle. In comparison to other Ethiopian indigenous cattle breeds, the morphometric measurements of both Harar and Ogaden breeds were found to be significantly lower than some lowland cattle breed like Begait cattle (Mulugeta, 2015). Similarly, body length, height at withers, ear length and horn length measurements of both breeds from the current study were lower than those of Raya cattle, while the reverse was true for heart girth and pelvic width measurements (Mustefa et al, 2021). Compared to the adjacent Afar cattle, the Ogaden cattle had higher values for height at withers and hearth girth while the Afar cattle had a longer body than both Harar and Ogaden cattle breeds (Tadesse et al, 2008). The observed higher size-related linear body mea- surements of the males in each breed and location fol- low Rensch’s rule (Rensch, 1950), which states that males are usually larger than females. Such differ- ences between males and females might be due to the testosterone secreted in males which causes the growth of muscle mass and skeletal development (Baneh and Hafezian, 2009). Estrogen secreted in females has a lim- ited effect on growth (Chriha and Ghadri, 2001; Baneh and Hafezian, 2009). The current results were compa- rable with the reports of Mustefa et al (2021) on Raya cattle, Terefe et al (2015) on Mursi cattle, and Genzebu 66 Mustefa et al Genetic Resources (2023), 4 (7), 56–67 et al (2012) on Arado cattle. Some size-linked morpho- logical parameters (i.e. hump size and dewlap width) were also larger for males than females, as these traits are associated with the overall size of the cattle. The morphometric variables, which discriminated the cattle populations, were ranked according to their importance. The inclusion of horn length within the top-three discriminatory variables is comparable with the reports of Mustefa et al (2021), who also classified it among the top-three variables to discriminate Raya cattle from other highland cattle breeds. The high error rate of the discriminant function analysis among the different districts shows a lack of uniqueness within each location. On the contrary, some similarities were shared among locations. The cattle populations from Tulo, Jarso and Fedis districts shared similarities justifying their belonging to the same group (the Harar cattle group). Similarly, the cattle populations from Jigjiga and Kebri Beyah districts shared some similarities, which support their categorization into the same group (the Ogaden cattle group). This confirms the idea of previous studies which state the presence of Harar and Ogaden cattle breeds in the eastern part of Ethiopia (Rege and Tawa, 1999; Getachew et al, 2014; Mengesha, 2019). This grouping was also supported by the morphometric and morphological results of the current study. The pairwise squared distance results between locations confirmed the already known differences between Harar and Ogaden cattle breeds, supporting the morphological, morphometric and multivariate results. However, these distances showed only the relative size differences between each population. Such differences might not necessarily be due to genetic differences (Zechner et al, 2001; Mustefa et al, 2021, 2022). Therefore, further diversity studies using molecular techniques are recommended to understand the level of genetic diversity within and between each breed. In conclusion, two cattle breeds in Eastern Ethiopia listed under the FAO Domestic Animal Diversity Information System (DAD-IS), were phenotypically characterized to obtain and quantify the within- and among-breed diversity. Strong within-breed similarities and large between-breed differences (distance) were observed. Thus, the current study confirmed the presence of two cattle breeds (the Harar cattle and the Ogaden cattle) in Eastern Ethiopia. Besides breed differentiation, this study will be used to design conservation and genetic improvement programmes for each breed. Data availability The data that has been used is confidential. Acknowledgments The authors are highly grateful to the Ethiopian Biodiversity Institute (EBI) for funding the work. Our special appreciation also goes to the farmers for providing their animals to this work for free. We also take this opportunity to appreciate the animal science experts at zonal, district and kebele level for their endless help during the data collection. A special word also goes to our friend and work partner Mr. Tadesse Hunduma for mapping the study area. Conflict of interest statement The authors declare that they have no conflict of interest. Author contributions All authors contributed to the study conception and design. Material preparation and data collection were performed by Amine Mustefa, Tesfalem Aseged, Seble Sinkie, Fasil Getachew, Tesfu Fekensa, and Manaye Misganaw. Amine Mustefa performed the data analysis and wrote the first draft of the manuscript. All authors commented on previous versions of the manuscript, and read and approved the final manuscript. References Abdi, D. Y. and Shiferaw, T. (2022). Determinants of the adoption of rainwater harvesting technologies in Kebri-Beyah District. International Journal of Water Resources and Environmental Engineering 14(3), 55– 65. doi: https://doi.org/10.5897/IJWREE2022.1031 Ayana, D. D. (2019). On Farm Phenotypic Characteriza- tion, Breeding, Husbandry and Fattening Practices of Harar Cattle in Hararghe Highland. MSc Thesis, Hara- maya University. Haramaya, Ethiopia. Baneh, H. and Hafezian, S. H. (2009). Effect of environmental factor on growth traits in Ghezel sheep. African Journal of Bio-technology 8, 2903–2907. url: https://www.ajol.info/index.php/ ajb/article/view/60943 Central Statistical Agency (2021). Ethiopian agricultural sample survey 2020/2021 (2014 E. C.). In Report on Livestock and Livestock Characteristics, volume II, Central Statistical Agency, Addis Ababa, Ethiopia. Chriha, A. and Ghadri, G. (2001). Caprine in the Arab world. 2nd edition. Department of Livestock Production. (Fateh University: Libby Conservation of Biodiversity and Environments in the Arab Countries), 478p. City Population (2007). Population statistics for coun- tries, administrative divisions, cities, urban areas and agglomerations, population projection 2022 accord- ing to the Ethiopian Statistics Service population cen- sus on 2007. url: https://www.citypopulation.de/. EBI (2016). Ethiopian National Strategy and Plan of Action for conservation and utiliza- tion of Animal Genetic Resources (Addis Ababa, Ethiopia: Ethiopian Biodiversity Institute). url: https://www.ebi.gov.et/wpcontent/uploads/2013/ 01/Final%20ENSAP final submitted%20(1).pdf. https://doi.org/10.5897/IJWREE2022.1031 https://www.ajol.info/index.php/ajb/article/view/60943. https://www.ajol.info/index.php/ajb/article/view/60943. https://www.citypopulation.de/ https://www.ebi.gov.et/wpcontent/uploads/2013/01/Final\%20ENSAP_final_submitted\%20(1).pdf https://www.ebi.gov.et/wpcontent/uploads/2013/01/Final\%20ENSAP_final_submitted\%20(1).pdf Genetic Resources (2023), 4 (7), 56–67 Characterization of cattle in eastern Ethiopia 67 EHZLFDO (2018). East Hararghe Zone Livestock and Fisheries Development Office Annual Report (Harar, Ethiopia). Falconer, D. S. (1989). Introduction to Quantita- tive Genetics (Longman Scientific and Technical), 3rd edition, 438p. doi: https://doi.org/10.1017/ S0016672300028573 FAO (2012). Phenotypic characterization of animal genetic resources. FAO Animal Production and Health Guidelines n.11, (Rome: FAO). url: www.fao.org/ docrep/015/i2686e/i2686e00.pdf. Genzebu, D., Hailemariam, M., and Belihu, K. (2012). Morphometric characteristics and livestock keeper perceptions of “Arado” cattle breed in Northern Tigray, Ethiopia. Livestock Research for Rural Development 24(6). url: https://www.lrrd.org/lrrd24/ 1/hail24006.htm. Getachew, F., Abegaz, S., Misganaw, M., and Fekansa, T. (2014). On-farm phenotypic characteriza- tion of Ogaden cattle populations of Jigjiga zone, south eastern Ethiopia. Ethiopian Jour- nal of Animal Production 14, 66–83. url: https:// www.researchgate.net/publication/325011260 On- farm phenotypic characterization of Ogaden cattle populations of Jigjiga zone southeastern Ethiopia. Kebede, M. B. and Utta, H. Z. (2021). Effects of Applying Blended Mineral NPS and Nitrogen Fertilizers on Growth, Yield Components, and Yield of Maize (Zea mays L.) in Fedis District. Eastern Ethiopia. East African Journal of Sciences 15(2), 167–174. doi: https: //doi.org/10.20372/eajs.v15i2.1688 Keskes, S., Ibrahim, M., Tessema, T. S., Tamir, B., Regassa, F., Kassa, T., and Dawo, F. (2013). Pro- duction systems and reproductive performances of Camelus dromedarius in Somali regional state, east- ern Ethiopia. Journal of Agriculture and Environment for International Development 107(2), 243–266. doi: https://doi.org/10.12895/jaeid.20132.166 Mengesha, A. M. (2019). On-farm Phenotypic Charac- terization of Ogaden Cattle Breed in Gode District of Somali Regional State in Ethiopia. International Jour- nal of Science and Business 3(3), 172–183. Mulugeta, F. G. (2015). Production system and phenotypic characterization of Begait cattle, and effects of supplementation with concentrate feeds on milk yield and composition of Begait cows in Humera ranch Western Tigray Ethiopia. Ph.D. thesis, Addis Ababa University, Ethiopia. Mustefa, A., Belayhun, T., Melak, A., Hayelom, M., Tadesse, D., Hailu, A., Assefa, and A (2021). Phenotypic characterization of Raya cattle in northern Ethiopia. Tropical Animal Health and Production 53, 48–48. doi: https://doi.org/10.1007/s11250- 020-02486-1 Mustefa, A., Engdawork, A., Sinke, S., and Hailu, A. (2022). Phenotypic characterization of Gesha horses in south western Ethiopia. Genetic Resources 3(5), 36– 50. doi: https://doi.org/10.46265/genresj.KPIL8781 Rege, J. E. O. (1999). The state of African cattle genetic resources I. Classification framework and identification of threatened and extinct breeds. Animal Genetic Resources Information 25, 1–25. doi: https://doi.org/10.1017/S1014233900003448 Rege, J. E. O. and Tawa, C. L. (1999). The state of African cattle genetic resources II. Geographical distribution, characteristics and uses of present- day breeds and strains. Animal Genetic Resources Information 26, 1–25. doi: https://doi.org/10.1017/ S1014233900001152 Rensch, B. (1950). Die Abhangigkeit der relative sexual differenz von der korpergrosse. Bonner Zoologische Beitrage 1, 58–69. url: https://www. biodiversitylibrary.org/partpdf/119381. SAS Institute (2002). Statistical Analysis Software (SAS). url: https://www.sas.com/enus/home.html. Statista (2020). Countries with the largest cattle population in Africa. url: https://www.statista. com/statistics/1290046/cattle-population-in- africa-by-country#:∼:text=Countries%20with %20the%20largest%20cattle%20population %20in%20Africa%202020&text=That%20same %20year%2C%20Africa%20had. Tadesse, D., Ayalew, W., and Hegde, B. P. (2008). On- farm Phenotypic Characterization of Cattle Genetic Resources in South and North Wollo Zones of Amhara Region. Ethiopian Journal of Animal Production 8(1), 22–38. url: https://hdl.handle.net/10568/34135. Terefe, E., Dessie, T., Haile, A., Mulatu, W., and Mwai, O. (2015). On-farm phenotypic characterization of Mursi cattle in its production environment in South Omo Zone, Southwest Ethiopia. Animal Genetic Resources 57, 15–24. doi: https://doi.org/10.1017/ S2078633615000132 Titto, C. G., Verissimo, C. J., Pereira, A. M. F., Geraldo, A. M., Katiki, L. M., and Titto, E. A. L. (2016). Thermoregulatory Response in Hair Sheep and Shorn Wool Sheep. Small Ruminant Research 144, 341–345. doi: https://doi.org/10.1016/j.smallrumres.2016.10. 015 UNHCR (2020). Kebri Beyah settlement profile, Somali region, Ethiopia. url: https://data2.unhcr.org/ en/documents/details/83312. WHZLFDO (2018). West Hararghe Zone Livestock and Fisheries Development Office Annual Report (Harar, Ethiopia). Zechner, P., Zohman, F., Solkner, J., Bodo, I., Habe, F., Marti, E., and Brem, G. (2001). Morphological description of the Lipizzan horse population. Livestock Production Science 69, 163–177. doi: https://doi.org/ 10.1016/S0301-6226(00)00254-2 https://doi.org/10.1017/S0016672300028573 https://doi.org/10.1017/S0016672300028573 www.fao.org/docrep/015/i2686e/i2686e00.pdf www.fao.org/docrep/015/i2686e/i2686e00.pdf https://www.lrrd.org/lrrd24/1/hail24006.htm https://www.lrrd.org/lrrd24/1/hail24006.htm https://www.researchgate.net/publication/325011260_On-farm_phenotypic_characterization_of_Ogaden_cattle_populations_of_Jigjiga_zone_southeastern_Ethiopia https://www.researchgate.net/publication/325011260_On-farm_phenotypic_characterization_of_Ogaden_cattle_populations_of_Jigjiga_zone_southeastern_Ethiopia https://www.researchgate.net/publication/325011260_On-farm_phenotypic_characterization_of_Ogaden_cattle_populations_of_Jigjiga_zone_southeastern_Ethiopia https://www.researchgate.net/publication/325011260_On-farm_phenotypic_characterization_of_Ogaden_cattle_populations_of_Jigjiga_zone_southeastern_Ethiopia https://doi.org/10.20372/eajs.v15i2.1688 https://doi.org/10.20372/eajs.v15i2.1688 https://doi.org/10.12895/jaeid.20132.166 https://doi.org/10.1007/s11250-020-02486-1 https://doi.org/10.1007/s11250-020-02486-1 https://doi.org/10.46265/genresj.KPIL8781 https://doi.org/10.1017/S1014233900003448 https://doi.org/10.1017/S1014233900001152 https://doi.org/10.1017/S1014233900001152 https://www.biodiversitylibrary.org/partpdf/119381 https://www.biodiversitylibrary.org/partpdf/119381 https://www.sas.com/enus/home.html https://www.statista.com/statistics/1290046/cattle-population-in-africa-by-country/\#:~ :text=Countries\%20with\%20the\%20largest\%20cattle\%20population\%20in\%20Africa\%202020&text=That\%20same\%20year\%2C\%20Africa\%20had https://www.statista.com/statistics/1290046/cattle-population-in-africa-by-country/\#:~ :text=Countries\%20with\%20the\%20largest\%20cattle\%20population\%20in\%20Africa\%202020&text=That\%20same\%20year\%2C\%20Africa\%20had https://www.statista.com/statistics/1290046/cattle-population-in-africa-by-country/\#:~ :text=Countries\%20with\%20the\%20largest\%20cattle\%20population\%20in\%20Africa\%202020&text=That\%20same\%20year\%2C\%20Africa\%20had https://www.statista.com/statistics/1290046/cattle-population-in-africa-by-country/\#:~ :text=Countries\%20with\%20the\%20largest\%20cattle\%20population\%20in\%20Africa\%202020&text=That\%20same\%20year\%2C\%20Africa\%20had https://www.statista.com/statistics/1290046/cattle-population-in-africa-by-country/\#:~ :text=Countries\%20with\%20the\%20largest\%20cattle\%20population\%20in\%20Africa\%202020&text=That\%20same\%20year\%2C\%20Africa\%20had https://www.statista.com/statistics/1290046/cattle-population-in-africa-by-country/\#:~ :text=Countries\%20with\%20the\%20largest\%20cattle\%20population\%20in\%20Africa\%202020&text=That\%20same\%20year\%2C\%20Africa\%20had https://hdl.handle.net/10568/34135 https://doi.org/10.1017/S2078633615000132 https://doi.org/10.1017/S2078633615000132 https://doi.org/10.1016/j.smallrumres.2016.10.015 https://doi.org/10.1016/j.smallrumres.2016.10.015 https://data2.unhcr.org/en/documents/details/83312 https://data2.unhcr.org/en/documents/details/83312 https://doi.org/10.1016/S0301-6226(00)00254-2 https://doi.org/10.1016/S0301-6226(00)00254-2 Introduction Materials and methods Study areas Site selection Data collection Data analysis Univariate analysis Multivariate analysis Results Morphological traits Morphometric measurements Multivariate analysis for discrimination of cattle populations Discussion Data availability Acknowledgments Conflict of interest statement Author contributions