The first morphological study of the Sloughi dog breed in Algeria

 

 

H.Y. Haddam^I; A. Mennani^II; O. Yilmaz^III; H. Abdelbari^I; I. Mkedder^IV; M.E.A. Benhamadi^I; W. Hamel^I; S.B.S. Gaouar^I

^IApplied Genetics in Agriculture, Ecology and Public Health Laboratory, Department of Biology, Faculty of Natural Sciences and Life and Earth Sciences and the Universe, University of Tlemcen, Algeria
^IIDepartment of Agronomic Sciences, Faculty of Sciences, University of M'sila, Algeria
^IIIFaculty of Agriculture, Department of Animal Sciences, Adnan Menderes University, Aydin, Turkey
^IVCenter for Research on Scientific and Technical Information Research (CRIST), Unit on the Mediation of Science, Tlemcen, Algeria

 

 

---

ABSTRACT

The present work aims to study morphometric profile, calculate body indices, and determine mathematical models that will be used for the estimation of adult live weights based on body measurements using multiple regression models in an endangered heritage breed, the Algerian Sloughi dog, raised in three different geographical regions of Algeria. The animal material of the study consisted of 105 Sloughi dogs, which included 30 individuals belonging to the eastern, 35 from the central, and 40 from the western regions of Algeria. Least squares means for height at withers and rump; head length and width; muzzle, ear, neck, body, tail lengths; muzzle, head, chest, waist, thigh, forearm, and wrist circumferences were used as body measurements. Least squares means for indices of massiveness, format, cephalic, body-tail, bone, muzzle-head, withers-rump, head-neck, ear-head length, and ear-head width were calculated using body measurements. Of the multiple regression models, the highest coefficients of determination (R²) were obtained from the models formed from height at withers, height at withers and thigh circumference or height at withers, thigh circumference and chest circumference together (R² = 0.70, R² = 0.79 and R² = 0.86 respectively). This study concluded that live weight of Algerian Sloughi dog breed could be estimated with a high accuracy using body measurements and statistical methods.

Keywords: Algeria, body indices, morphometric analysis, Sloughi, dog weight

---

 

 

Introduction

The Sloughi is a dog breed originating from North Africa, specifically the region encompassing Morocco, Algeria, Tunisia, and Libya. Developed for hunting purposes, this breed boasts a powerful and elegant appearance, with short, smooth hair; lop ears, and an athletic build. In addition to their striking aesthetic qualities, the Sloughi exhibits exceptional stamina, making them well suited for hunting a variety of prey, including hare, fox, jackal, gazelle, and big antelope species (www.akc.org). With a long-standing presence in North Africa, the Sloughi has been highly valued for its hunting prowess, endurance over long distances, and agility. In Algeria, the breed is regarded as a national heritage and is commonly found in the steppe areas of Wilayas, including Tebessa, Khenchela, Batna, Biskra, M'Sila, Djelfa, Tiaret, Laghouat (Nord), Saída, and Tlemcen. However, the breed's pure-blood status is threatened by pollution and genetic erosion, according to a statement from a prominent breeder (El-Hassani, 2022). The lack of research about the Sloughi in Algeria is a contributor to the breed's uncertain future. Generally, when defining the various characteristics of animals, researchers and experts rely on well-documented academic studies. However, the absence of such works regarding the Sloughi in Algeria has made it difficult to establish a clear understanding of the breed's history, characteristics, and potential avenues for preserving its lineage.

Generally, to define various characteristics of animals we use body measurements; the most important characteristics used by researchers and during breed selection are live weight and several body measurements. In order to preserve and protect breeds, two important parameters are usually targeted, live weight and growth characteristics. A breed is defined as a group of individuals sharing some transmissible and distinctive traits, which means that each breed holds certain traits that set it apart from other breeds (González et al., 2011). Another similar approach defines a breed as a group of individuals recognizable as being biologically different from others (Cavalli-Sforza, 2000). The methods used in the present study based on morphometric analysis are the same as those used in other dog breeds or other domestic species, such as sheep, goats, and camels (Traoré et al., 2008; González et al., 2011; Yilmaz et al., 2012; Meghelli et al., 2020; Urosevic et al., 2020a; Djaout et al., 2022).The aim of this work was to provide a morphometric profile and estimate body indices and live body weight of the Algerian Sloughi in order to provide a knowledge base for better breed identification and management of the genetic diversity of this important heritage and relevant biological resource. This study facilitates breeders to direct the selection of their dogs towards the desired activities without making crosses with other breeds.

 

Materials and Methods

The animal material of the study consisted of 105 Algerian Sloughis: 30 individuals from the eastern region, represented by Batna, Setif and M'sila; 40 individuals from the west, represented by Tlemcen, Sidi Bel Abbes, and Ain Temouchent; and 35 individuals from the central region, represented by the Wilaya of Laghouat (Figure 1). All of these animals were adults (> 36 months), purebred, fit, and well. All of these dogs used in the present study were neither in the reproductive period (concerning both males and females), in gestation, or lactating. In addition, all the individuals selected in this study were subjected to the same diet.

Sixteen body measurements represented by weight and ten calculated body indices were used in this study. Body measurements, including HW, HR, HL, HeW, ML, EL, NL, BL, and TL were taken using a measurement stick; all circumferences, including MC, HC, CC, WaistC, TC, FC, and WristC, were taken using a measuring tape; and live weights of dogs were measured using a digital scale (Figure 2). Figure 3 represents an example of the studied breed. The novelty of the work stems from all studied individuals that were measured directly on their owners' farms, taking into account the exact origin of these specimens and having information on at least three generations of their ancestors. Given the absence of a book of origins (studbook) in Algeria, we proceeded by eliminating any dogs of unknown origin. Body indices were calculated using mean values of the measurements in the formulas shown below using established methods used by local breeders and other studies on dogs (Drobnjak et al., 2010; Drobnjak et al., 2012) used by Oğrak et al. (2014).

The UNIVARIATE procedure of the SAS (1999) statistical package program was used to check normality of the data. The result of the analysis showed that the data for all the measured characteristics were normally distributed. Afterwards, the GLM procedure of same software was used to make variance analyses and to obtain least squares means for the investigated characteristics. Duncan's multiple comparison test was used to test for differences between significant means. The phe-notypic correlations between variables were also obtained using the PROC CORR procedures in SAS (1999).

Mathematical models used for analysis of variance is presented below:

Model used for weight and body measurement indices:

Model used for body measurements:

where Yijkl = observations for body measurements, weight, and index of massiveness (%); μ = overall mean of the trait; ai = fixed effect of gender (I = male and female); bj = fixed effect of region (j = east, west, and central); β1 = coefficient of regression of live weight; = mean weight; Xi = weight; eijk and eijkl = random errors with the assumption of N (0, σ²).

Estimation equations of live weights with multiple linear regression analysis using body measurements according to region and region χ sex groups were obtained by using stepwise multiple regression procedure in SAS (1999).

 

Results and Discussion

Descriptive statistics belonging to the body measurements, weight, and body indices (Table 1) show that the highest coefficients of variation among the studied individuals were for both waist circumference and weight with 11.94% and 11.39%, respectively; the lowest was for the withers-rump index at 2.49%. It is also remarkable that some individuals had a height at the withers greater than 80 cm, which are considered as giant type Sloughis, but unfortunately, this type is becoming increasingly rare. They were chosen during this work to be part of a new program for the preservation of the breed, with the agreement and the participation of the breeders. Height at withers is one of the most important traits in the selection of dogs, especially in the Sloughi. Local breeders of the mountain regions consider tall Sloughis as the best, and this is due to their utility for hunting wild boars and golden African wolves without any difficulty. In addition, these breeders and owners have noticed that these so-called giant Sloughis adapt better to mountainous regions compared to other Sloughis originating from plains and desert regions, where height at withers is less important.

Least squares means and standard errors for body measurements and weights belonging to Sloughi dog breed are presented in Appendix Table 1. It is understood that for HW, HR, HL, HeW, MC, HC, FC and weight, male traits were larger (P <0.001). For WaistC and TC, males were larger (P <0.01). For NL and BL, males were larger (P <0.05). ML, EL, TL, CC, and WristC measurements were similar between sexes (P >0.05).

Results obtained in this study showed that in the Algerian Sloughi, sexual dimorphism was evident in 11 measurements out of 16, without forgetting weight, where males were heavier than females. This phenomenon is common in dogs, according to the several dog breed standards of the AKC (American Kennel Club), where sexual dimorphism is evident in all dog breeds. In the Italian pointing dog (Braceo italiano), males are larger than females in HW, CC, BL, HR, and HL (Cecchi et al., 2015). For HW, HR, BL, CC, and HL in the Tarsus Çatalburun breed of Turkish hunting dog, males were larger (Ograk et al., 2014).

In Turkey, two studies on the Turkish Tazi have shown that there is a larger gain for males in HW, HR, CC, HL, HeW, and BL (Urosevic et al., 2020a), and for HW, HR, CC, BL, and weight (Yilmaz & Ertugrul, 2011). The third study was on the Akbash Turkish Shepherd, which shows sexual dimorphism and HW, HL, and back height in males are larger (Urosevic et al., 2020b). A study carried out in Brazil in the Labrador had also shown that the males have two larger measurements than females, HW and BL (Thuller et al., 2015). In the present study and the other citations mentioned above, HW, HR, CC, BL, HL, and HeW showed greater importance in morphometric differentiation between sexes. For live weight, males were always heavier than females. There was a difference in HW, HR, and FC (P <0.001) in western Sloughis with higher withers than central and eastern Sloughis; central Sloughis had greater HR and FC than the other regions (Appendix Table 1). Regional differences were apparent for NL, TL, CC, and WaistC (P <0.01), where western Sloughis had larger NL, TL, and CC than the two other regions; eastern Sloughis had a larger WaistC. ML, TC and WristC (P <0.05) were larger in central Sloughis than western and eastern dogs. HL, HeW, EL, BL, MC, HC, and weight were similar among regions (P >0.05).

The above differences could be due to the different environmental factors, such as climate, soil texture, and the shape of the land. Breeders report a difference between the Sloughi sub-populations and this is certainly related to the environment where they evolved, especially the use of Sloughi in hunting. The taller dogs are generally used against large game such as wild boar, the African golden wolf, or the jackal and the smaller dogs are used for hunting small game, such as the hare. In other words, Sloughis are selected according to criteria that meet the needs of breeders and users of this breed. In Turkish Tazi, when the study was carried out on three sub-populations of dogs belonging to three different regions of Turkey (southeast, central and eastern), the southeastern subpopulation had a longer body length than the two other sub-populations (Yilmaz & Ertugrul, 2011). This phenomenon is probably evident in other dog breeds.

Regressions between weight and all body measurements were statistically significant: between weight and HW, HR, EL, BL, TL, CC, TC, and FC (P <0.001); weight and HL, HeW, NL, MC, WristC, (P <0.01); and weight and ML, HC, and WaistC (P <0.05). Least squares means and standard errors for body indices are presented in Appendix Table 2. The results reported in Appendix Table 2 indicated differences between the sexes for IM (P <0.01), with females larger. Females had a larger FI (P <0.05) and female body length was similar to height at withers, such that females were more blockish than males. For all the remaining indices, differences were not apparent.

Body indices were influenced by the environment too. There were differences in IM and CI between dogs belonging to the three regions (P <0.001), where eastern Sloughis had larger indices, indicating that eastern dogs had a less athletic body and their head width represented more than 50% of their head length in comparison to the two other regions (53.32% vs. 51.86% and 51.31%, respectively) (Appendix Table 2). For BTI, WRI, and HNI, the differences were significant (P <0.05); western dogs' tail length represented 62.75% of their body length (62.75% versus 60.62% and 60.15% for central and eastern dogs, respectively). HW and HR were approximately equal in the central population compared to the western and the eastern populations, respectively, where HW was larger than HR (100.88% versus 101.81% and 102.43%, respectively). In the eastern population, HL represented 125.14% of NL, such that HL was 25.14% larger than NL; in the western population, HL was 18.38% larger than NL; and in the central population, HL was 22.54% larger than NL. FI, MHI, EHI, and EHWI were similar (P >0.05). In the Tarsus Çatalburun breed of Turkish hunting dog, females had larger indices than males for Index of Format (FI), index of massiveness (IM), and index of bone (BI). The reason could be due to the shorter height of shoulder in females (Ograk et al., 2014).

Phenotypic correlation coefficients between weight and body characteristics are presented in Appendix Table 3. A general evaluation shows that there were strongly positive phenotypic correlations between weight and body measurements in the study. Analysis of correlation coefficients between weight and body measurements according to region showed that the highest coefficients were obtained in the western region. It was also found that correlation coefficients between weight and all body measurements were strongest in western Sloughis for HW, HR, HL, HeW, NL, BL, TL, HC, CC, TC, and FC (P <0.001); EL, MC, and WristC (P <0.01); and ML and WaistC (P <0.05). Correlation coefficients between weight and HC and WaistC were not significant (P >0.05) in central Sloughis, and between weight and NL in eastern Sloughis.

In the present work, the prediction equations were specific to the studied breed; in German shepherds, Labrador retrievers, and Belgian Malinois, the results obtained were similar concerning the correlations between weight and body measurements (Dirlik, 2008). In Nigerian native dogs, the results were also similar (Emehelu et al., 2012). In their studies, the correlations between live weight and chest circumference were the highest. In Zagar dogs, live weight could be adequately predicted from body length, chest width, chest circumference, hind cannon bone circumference, and ear interval; in Zerdava dogs, live weight could be predicted from wither height, chest width, chest circumference, and head length; and in Çatalburun dogs, live weight could be predicted from wither height, body length, and chest circumference (Özkul et al., 2021). In Philippine native dogs, the correlation and regression analyses indicated that all body measurements had positive linear relationships with body weight regardless of sex, but the single best determinant of body weight for both sexes was thoracic girth, with a coefficient of determination (R²) of 0.468 (Valdez and Valencia, 2004). A study by the Gemlik Military Veterinary School indicated that ear length was the strongest predictor of live weight in German shepherd dogs (Elmaz et al., 2006). The use of the classification and regression tree method (CART) and multivariate adaptive regression splines (MARS) explained 68% and 91% of the variation in live weight of Turkish Tazi dogs with morphometric measures (Çelik & Yilmaz, 2018). Separate models were developed for each region for the estimation of weight from body measurements and coefficients of determination (R²) (Table 2).

Analysis of R² values indicated that the highest value was obtained from the second model of the central group (R² = 0.76); the lowest value was obtained from the model of the eastern group (R² = 0.46). HR with HeW could be used in the regression model. Separate models were developed for each region and sex for the estimation of weight from body measurements and coefficients of determination (R²) are presented in Table 3.

Analysis of R² values showed that the highest value was obtained from the third model in males from the central region and the lowest value was obtained from the model in males from the eastern region (R² = 086 versus R² = 0.37; Table 3). For the females, the highest R² was obtained in both the eastern and central females (R² = 0.65; Table 3). It was indicated that HW with TC and CC could be used in regression model for Sloughi males. HR or HeW could be used in regression models for females. The use of multiple regression models in other mammals, such as Karya sheep, showed that the highest coefficients of determination (R²) were obtained from the models including body length or body length and chest girth together (R² = 0.79, R² = 0.87), such that live weight of Karya sheep could be estimated with a high accuracy (Yilmaz et al., 2013). In dromedaries, live weight models with chest girth measurements had R² values of 0.74-0.99; chest girth was the most important measurement that could be used for estimate live weight (Meghelli et al., 2020).

It is remarkable that even the simplest calculations using morphometric measurements led to promising results in terms of accuracy in the prediction of live weight of Algerian Sloughi, especially HW, HR, CC, BL, TC, and HeW.

 

Conclusion

The models constructed with regression analysis used in this study facilitated the use of the most important body measurements that were likely to change between breeds, sexes, or geographical regions in the estimation of body weight. Results of the present study showed that there were statistically significant differences between sexes in the studied population of 105 individuals for the majority of body measurements, some body indices, and between the three different sub-populations for six of ten body indices and ten of sixteen body measurements. The western individuals were the tallest with good proportions between head measurements. Indices confirmed the statements of breeders, who use several historical methods to evaluate morphological profile, which is closely related to hunting potential. For the format index, central Sloughis had the best proportion; breeders use this parameter to evaluate the aesthetic of the Sloughi, which means that the proportion between wither and rump height should be equal to 100%. In Algeria, there is an important genetic potential and morphological diversity within the Sloughi population that are beneficial in a program of genetic improvement and the creation of a breed standard in order to preserve this endangered heritage breed. This type of study allows breeders to direct the selection of their dogs and direct each animal towards the most appropriate activity.

Compliance with ethical standards

All activities were performed according to Algerian laws and regulations. No ethical conflicts exist in the realization of this work.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Author contribution

HHY and GSBS developed and conceptualized the study; AM, BMEA, and WH provided field trips with HHY. Statistical analyses were made by OY and HHY. Writing, review, and editing were done by HA, MI, and HHY. All authors have read and agreed to the published version of the manuscript.

Data availability

Data will be made available on reasonable request.

 

Acknowledgement

We would like to thank T.T., L.M., A.K.H., A.B., and A.D. for their precious advice concerning the characteristics of Algerian Sloughi; the National Association for the Protection of Sloughi and Wildlife in Algeria (NAPSWL); and the Association of Life Team for Environment and Animals (A.L.T.E.A. 13) for their kind help during this study.

Funding

Self-funded: HYH

 

References

AKC Breeds Nomenclature, 2022. American Kennel Club website. Available at: https://images.akc.org/pdf/breeds/standards/Sloughi.pdf.         [ Links ]

Cavalli-Sforza, L., 2000. Genes, Pueblos y Lenguas. Barcelona, Ed. Crítica.         [ Links ]

Cecchi, F., Paci, G., Spaterna, A., Ciampolini, R., 2013. Genetic variability in Braceo italiano dog breed assessed by pedigree data. Ital J Anim Sci, 12:348-352.         [ Links ]

Celik, S., Yilmaz, O., 2018. Prediction of body weight of Turkish Tazi dogs using data mining techniques: Classification and regression tree (CART) and multivariate adaptive regression splines (MARS). J. Zoo. 50(2), 575-583.         [ Links ]

Dirlik, H., 2008. Gemlik askeri veteriner okulu ve eğitim merkezi komutanlığında yetiştirilen bazı köpek ırklarında vücut ölçüleri ve bu ölçüler arasındaki fenotipik korelasyonlar. Adnan Menderes Üniversitesi Sağlık Bilimleri Enstitüsü Yüksek Lisans Tezi.         [ Links ]

Djaout, A., Belharfi, F. Z., Kourichi, A., Larabi, N., Benidir, M., & Gaouar, S. B. S., 2022. Morphometric assess ment and physico-chemical description of the milk of Arbia goat breed in province of Tlemcen. Genet. Biodivers. J. 6(2), 143-158.         [ Links ]

Drobnjak, D., 2012. Turkish shepherd dog Kangal, Cepib Publisher; Zemun, Belgrade, Serbia.         [ Links ]

Drobnjak, D., Matic, V, Miliyevic D, 2010. Eksterijerpasaosnoveprocene, TipoStampa, Beogard, Serbia, 27-33.         [ Links ]

Elmaz, O., Aksoy, O.A., Dikmen, S. & Zonturlu, A., 2006. Growth performance, survival ratio, and body measurements until weaning age of German Shepherd dog. World Small Animal Veterinary Association, Proc. 31st World Congress, Prague, Czech Republic.         [ Links ]

Emehelu C. O., Eze J. I., Akune A., Chah K. F., 2012. Estimation of live body weight from body measurements in Nigerian local dogs. African Journals Online 30(2): 65-73.         [ Links ]

González, A., Luque, M., Rodero, E., González, C., Aguilera, R., Jiménez, J., Sepúlveda, N., Bravo, S. & Herrera, M., 2011. Use of morphometric variables for differentiating Spanish hound breeds. Int. J. Morphol. 29, 1248-1255.         [ Links ]

Kacimi El-Hassani, A., 2022. The causes of the disappearance of the Sloughi in Algeria. The Sloughi Review, 10, 05-10.         [ Links ]

Meghelli, I., Kaouadji, Z., Yilmaz, O., Cemal, I., Karaca, O., Gaouar, S.B.S., 2020. Morphometric characterization and estimating body weight of two Algerian camel breeds using morphometric measurements. Trop. An. Health Prod. 52, 2505-2512.         [ Links ]

Ograk, Y. Z., Yoldas, A., Urosevic, M., & Drobnjak, D., 2014. Some morphological traits of the Tarsus Çatalburun breed of Turkish hunting dog. Euras. J. Vet. Sci. 1, 25-29.         [ Links ]

Özkul B.Y., Doka P.C.K., Özen D., Özbaser F.T., Özarslan B., Atasoy F., 2021. Correlation between live weight and body measurements in certain dog breeds. S. Afr. J. Anim. Sci. 51(2), 151 -159.         [ Links ]

SAS, 1999. The SAS System SAS Institute Inc., Cary, NC, USA, Version 8.         [ Links ]

Thuller, M., Jangarelli, M., Couto, D., Araújo, A., 2015. Sexual dimorphism of Labrador retriever dogs by morphometry. Biosci. J. 31, 1475-1487.         [ Links ]

Traoré, A., Tamboura, H., Kaboré, A., Royo, L., Fernández, I., Álvarez, I., Sangaré, M., Bouchel, D., Poivey, J., Francois, D., Toguyeni, A., Sawadogo, L., Goyache, F., 2008. Multivariate analyses on morphological traits of goat in Burkina Faso. Archiv. Tierzucht, 6, 588-600.         [ Links ]

Urosevic, M., Drobnjak, D., Stojic, P., Ograk, Y. Z., 2020b. Morphometric Characterization of the Akbas (Akbash) Turkish Shepherd Dog. Turkish Journal of Agriculture - Food Science and Technology, 8(7): 1571-1576.         [ Links ]

Urosevic, M., Matejevic, M., Drobnjak, D., Ozkanal, U., 2020a. Use of Morphometric Variables for Differentiating Breed Variations in Turkish Tazi (Sighthound) Population. Pakistan J. Zool., 52:1765-1770.         [ Links ]

Valdez, C. A. and Valencia J. P. L., 2004. Weight prediction in medium to large sized adult Philippine native dogs using external body measurements. Philippine J. Vet. Anim. Sci. 30(1): 161-169.         [ Links ]

Yilmaz, O. and Ertugrul, M., 2011. Some morphological characteristics of Turkish Tazi (Sight Tazi). J. Anim. Pl. Sci., 21: 794-799.         [ Links ]

Yilmaz, O., Cemal, I., Karaca, O., 2013. Estimation of mature live weight using some body measurements in Karya sheep. Trop. Anim. Health Prod. 45: 397-403.         [ Links ]

 

 

Submitted 13 April 2023
Accepted 18 March 2024
Published 28 April 2024

 

 

# Corresponding author: hyh.genetics@gmail.com

 

 

Appendix Table 1 - Click to enlarge

 

 

Appendix Table 2 - Click to enlarge

 

 

Appendix Table 3 - Click to enlarge