Unlocking the potential of local rabbit population: morphological insights for sustainable rabbit farming in Burkina Faso’s challenging environments

Abstract

This study explores the potential of rabbits as a sustainable solution for poverty alleviation and food security in Burkina Faso, a country facing socio-economic and environmental challenges. In the context of limited resources and employment opportunities, rabbits offer advantages such as small size, short generation interval and high reproductive capacity. The research focuses on characterising the morphology of Burkina Faso’s local rabbit population, comparing it with two Italian breeds: a local medium-growing breed called Leprino di Viterbo (LV) and a fast-growing hybrid (a crossbred New Zealand × California, NZC). Utilising path analysis and principal component analysis, the study identifies key morphometric traits crucial for breeding programs and examines the influence of specific traits on body weight (BW) gain and heat stress resilience through ‘Transpiration Indexes.’ Findings highlight the importance of variables like chest girth (CG), abdominal girth (AG), rump width (RW) and nose to shoulders length (NSL) in influencing body weight (BW) and reveal significant size and shape differences among the breeds. The study suggests potential climate adaptation in Burkina Faso’s rabbit population, providing practical insights for breeding programs in resource-constrained regions. This research not only advances scientific understanding but also provides practical insights for breeding programs in resource-constrained regions, emphasising the importance of certain morphological traits for heat dissipation.

Keywords:

• Small livestock
• rabbit production
• morphological traits
• poverty alleviation

Introduction

Persistent challenges of poverty and hunger, particularly in sub-Saharan Africa, accentuate the cycle of destitution due to limited access to resources, education and employment opportunities. Small livestock, notably rabbits, have emerged as a promising yet unexplored alternative for poverty alleviation in sub-Saharan Africa and developing countries in general (Oseni and Lukefahr 2014).

Burkina Faso, sited in West Africa, faces numerous socio-economic and environmental obstacles hindering progress towards sustainable development.

In the face of challenges like climate change and population growth, rabbits have become a favoured meat protein source, especially in ‘focusing on ending hunger, achieving food security, improving nutrition, and promoting sustainable agriculture’ (UN General Assembly 2015; Mutsami and Karl 2020). Unlike larger livestock species, such as cattle or goats, rabbits require minimal space and financial investment and so offers possibilities to rural and urban poor people to benefit from high and increasing demands for animal products (Chriki et al. 2020; Wongnaa et al. 2023). Furthermore, rabbits can be raised on a diet that does not require grains, crucial in a geopolitical context with rising grain prices and growing demand (Mutsami and Karl 2020). Understanding the potential of Burkina Faso local rabbit populations (BLF) is crucial, particularly in contest of heat stress, which significantly impacts rabbit growth (Zeferino et al. 2013; Farghly et al. 2020).

In West Africa, NGOs are actively promoting small livestock husbandry as a powerful tool against food insecurity, providing training and support to local communities (Hermans et al. 2023).

The current paper presents the results of a study carried out in the frame of a Cooperation project, aimed at characterising the morphology of a local rabbit population. To produce valuable comparative data, the same morphometric measurements were also collected from two distinct Italian rabbit populations: the first one is a local medium-growing breed called Leprino di Viterbo (LV), and the last, a fast-growing hybrid (a crossbred New Zealand × California, NZC) (ANCI, https://www.anci-aia.it/), mainly used in intensive breeding systems. The use of body measurements together with live body weight (BW) is of great importance in defining performance in livestock, as well as the correlation between body measurements and performance traits (Cam et al. 2010). Furthermore, to assess the breeding potential of the Burkina Faso rabbit populations, a path analysis was conducted on the measures, to identify key morphometric traits that can be prioritised in future breeding plans. These data were also used to perform a PCA analysis in order to compare morphometrics features with the aim of identifying key morphometric differences in terms of shape and proportion of some traits, for example, ears length (EL) and tail length (TL) as evidence of ‘shape-shifting’ – (changes in appendage size) in response to heat stress (Ryding et al. 2021). Beyond scientific inquiry, the study emphasises the strategic role of small livestock like rabbits in Burkina Faso, offering a potential solution to food insecurity and income generation for rural populations. The results aim to underscore the importance of science-driven interventions in unlocking the full potential of local animal resources in adaptation to harsh environments.

Materials and methods

The project ‘Lapin’

The present study is based on the Lapin Project, titled ‘Development of the rabbit production chain and family agroecology: initiatives to fight food insecurity and rural exodus in Burkina Faso’, and it took place from 1 March 2021 to 28 February 2022. The project, led by Tamat, an Italian NGO (tamat.org), aimed to face a variety of challenges, among which food insecurity, children malnutrition, irregular migration and the empowerment of women, through small livestock production. To achieve these objectives, technical training on rabbit farming methods was provided to 40 small-scale farming families in the Loumbila municipality, just outside of the capital Ouagadougou. These families received essential equipment and raw materials for rabbit farming, and breeding stock was purchased and enhanced to improve genetic traits.

Climate and environment of Burkina Faso

The study was carried on a sample of local population of rabbit (BFL) available in Burkina. The sampling was performed in the village of Tanlargho in the municipality of Loumbila, among the familiar breeding units created by the project Lapin. The breeding conditions are characterised by wood cages, one per family. The climate is hot ‘semi-arid’ because it has a prolonged dry season with high temperatures and relatively low humidity, and a wet season; however, due to climate change, many alterations have been occurred. Burkina Faso grapples with noticeable climate shifts, impacting traditionally more favourable regions like the east and southwest with rising temperatures and intermittent droughts. Geographically situated with a dry tropical climate, Burkina Faso experiences distinct wet and dry seasons. Its three climatic zones face varying average annual rainfalls, with the Sahelian north receiving less than 600 mm, the central north-Sudanian zone between 600 and 900 mm, and the southern Sudanian zone exceeding 900 mm. Climate change poses threats of reduced rainfall, water scarcity and lower agricultural yields, with an anticipated 10% decline in rainfall and a 1.4–1.6 °C temperature rise by 2050, heightening drought and fire risks (UNDP, https://www.undp.org/).

Animal sampling and body measurements

A total of 100 BFL mature rabbits from the project Lapin with a minimum age of 5 months were sampled; 28 mature rabbits (of a minimum of 5 months) of NZC and 20 mature rabbits (of a minimum of 5 months) of LV.

The measure collection of the commercial Italian strain NZC took place in the Experimental Section of the Department of Agricultural, Food and Environmental Sciences of the University of Perugia (Vestricciano, Perugia, Italy). NZC is the result of a selection program conducted by ANCI (https://www.anci-aia.it/) to obtain a F1 of meat rabbits to share with Italian rabbit breeders.

LV is a rabbit population showing a medium-sized animal; it adapts well to breeding in warrens, on straw-covered ground, and with mobile arches, and is considered a native breed originating from Lazio, the central region of Italy. The measure collection of the local Italian strain LV took place in a breeding farm that is member of the ‘Consorzio del Coniglio Verde Leprino di Viterbo’ (coniglioleprinoviterbo.it), whose aim is to improve and preserve this local population.

The following described morphological traits were measured using a measuring tape (cm) for the exception of live weight (kg) and recorded (Awuor et al. 2018; Rotimi 2021; Eshimutu et al. 2023; Isaac and Oriaku 2023). Abdominal girth (AG): circumference of the abdomen; body length (BL): distance between the shoulders (the first thoracic vertebrae) and the base of the tail. Chest girth (CG): circumference of the chest just behind the forelimbs. Ear length (EL): distance from ear base to the end of the ear. Foot length (FL): distance from the ankle joint to the tip of the foot. Nose to shoulder length (NSL): distance from the point of the nose to the shoulder (fore limb). Tail length: distance from base of the tail to end of tail tip. Rump width (RW): distance between the most posterior point of pin bones. Thigh circumference (TC): circumference of the thigh. Live BW of the animals. All measurements were carried out by the same person.

Data processing and statistical analysis

All the statistical analyses were performed in R software (R Core Team 2021). Descriptive statistics have been performed for all the quantitative body measurements and BW. The means, minimum, maximum and standard deviation (SD) were estimated for each measurement separately for each breed. Spearman’s correlation coefficient was performed to assess the degree and the direction of the relationship among the variables for BLF.

To identify the variables to apply in structure equation modelling (SEM) for path analysis, a multiple regression analysis was first performed: $$Y=\alpha +b_1*X_1+b_2*X_2+b_3*X_3+b_4*X_4+b_5*X_5+b_6*X_6+b_7*X_7+b_8*X_8+b_9*X_9$$ where * is the multiplication, Y is the dependent variable (BW), α is the intercept, b₁–b₉ is the coefficient of regression, X₁–X₉ is the independent variables (biometric traits = AG; BL; CG; EL; FL; NSL; TL; RW; TC).

The path analysis is part of structural equation modelling technique, that allows for the simultaneous assessment of both direct and indirect relationships among variables within a hypothesised model. This model permits the capturing of the intricate biological connections between BW and morphometric characteristics and avoid to account for causality and multicollinearity among variables (Norris et al. 2015; Adenaike et al. 2023; Li et al. 2023). Path analysis was performed with R software (R Core Team 2021) through ‘LAVAAN’ library (Rosseel 2012). The SEM equation applied to the analysis is as follows: $$\text{BW}=\alpha +{\beta }_1\mathrm{}*\text{AG}+{\beta }_2\mathrm{}*\text{CG}+{\beta }_3*\text{RW}+{\beta }_4\mathrm{}*\text{NSL}$$ $$\text{AG}=\gamma *\text{CG}$$ $$\text{RW}=\delta *\text{AG}$$ $$\text{RW}=\epsilon *\text{NSL}$$ $$\mathit{\text{bc}}=c*b$$ $$\mathit{\text{bd}}=b*d$$ $$\mathit{\text{cd}}=c*d$$ where * is the multiplication; BW is the endogenous variable; AG, CG, RW and NSL are the exogenous variables; α is the intercept; β₁, β₂, β₃ and β₄ are the regression coefficients representing the direct effects of AG, CG, RW, NSL on BW; γ, δ and ε are the partial regression coefficients indicating the relationships between AG, RW with CG, AG, NSL; bc, bd, and cd are the indirect effects of CG on BW mediated by AG (bc), AG on BW mediated by RW (bd) and CG on BW mediated by both AG and RW (cd).

The goodness of fit of the path model was assessed using several fit indices suggested in literature (Schermelleh-Engel et al. 2003), including the comparative fit index (CFI), Tucker–Lewis index (TLI), root mean square error of approximation (RMSEA) and standardised root mean square residual (SRMR) (Schermelleh-Engel et al. 2003). The model was estimated using the maximum-likelihood (ML) estimation method, and the significance of each path coefficient (PC) was determined using standard z-test.

Finally, a PCA analysis was performed on all the measurements recorded for the three rabbit populations through library ‘FactoMineR’ (Lê et al. 2008) in R to understand deeply the morphology of BFL (Akounda et al. 2023; Tyasi and Tada 2023). PCA helps in reducing the dimensionality of the data, making easier to visualise and comprehend the underlying patterns in shape and size variations among individuals or species. PCA aids in identifying the most important variables or components contributing to morphological differences.

Furthermore, following the ‘Shape-Shifting’ hypothesis (Ryding et al. 2021), morphometric indexes called in this study ‘Transpiration Index’ (TI) were calculated in relation to measurements suggesting the shape of the animal (BL and CG) and the appendixes (EL and TL) associated to the heat dissipation (Conley and Porter 1985).

An ANOVA and a Tukey test were performed to validate the differences observable in shape between the three rabbits’ population considered through R software.

Results

Descriptive statistics regarding body live weight and body measurements of rabbits’ population are shown in Table 1.

Table 1. Descriptive statistics of body weight and body measurements of rabbits.

BLF	Mean	SD	Min	Max
EL, cm	10.41	0.79	8.5	12
TL, cm	6.86	0.9	4	9
NSL, cm	15.71	0.93	13	18.5
BL, cm	28.16	3.51	13	35
RW, cm	9.85	0.87	8	13
FL, cm	9.6	0.68	7	11
TC, cm	16.32	1.96	13	22
AG, cm	29.07	2.8	22	36.5
CG, cm	25.18	1.87	21	30
BW, kg	1.81	0.33	1.15	3
LV	Mean	SD	Min	Max
EL, cm	14.53	0.75	13	15.5
TL, cm	9.13	1.09	7	11.5
NSL, cm	18.32	0.61	17	19
BL, cm	40.74	2.65	37	49
RW, cm	11.76	1.48	9	15
FL, cm	12.68	0.48	12	13.5
TC, cm	23.32	1.79	19	26
AG, cm	38.92	2.64	34	43
CG, cm	32.97	2.32	28.5	38
BW, kg	4.7	0.48	3.81	5.68
NZC	Mean	SD	Min	Max
EL, cm	13.53	0.73	11.5	15
TL, cm	9.95	1.05	8	12
NSL, cm	17.3	0.73	16	19
BL, cm	41.65	2.67	34	45.5
RW, cm	14.12	1.74	10.5	17
FL, cm	12.25	0.95	10.5	14
TC, cm	22	2.12	18	26
AG, cm	35.58	2.28	32	42.2
CG, cm	32.4	1.39	30	36
BW, kg	4.11	0.23	3.75	4.46

EL: ear length; TL: tail length; NSL: nose to shoulders length; BL: body length; RW: rump width; FL: foot length; TC: thigh circumference; AG: abdominal girth; CG: chest girth; BW: live body weight; SD: standard deviation; BLF= Burkina Faso Local rabbit; LV= Leprino di Viterbo; NZC= New Zealand x California cross (Italian Commercial strain).

The differences reveal that BFL rabbits exhibit consistently smaller values in terms of BW and several key morphometric measures when compared to NZC and LV counterparts, as predictable. LV rabbits exhibited the longest ear mean 14.53 cm, while NZC rabbits showed the longest tail, mean 9.95 cm. Moreover, NZC showed the highest values for BL and RW and LV for AG and CG. Spearman’s correlations between all the measured traits of BFL and BW are presented in Table 2.

Table 2. Spearman’s correlations between the observed traits of Burkina Faso local rabbit populations (BFL) rabbits.

	TL	NSL	BL	RW	FL	TC	AG	CG	BW
EL	0.37**	0.29*	0.19	0.19	0.24*	0.169	0.06	0.11	0.25*
TL		0.13	0.04	0.25*	0.21*	−0.07	0.03	0.00	0.16
NSL			0.38**	0.33**	0.14	0.29*	0.22*	0.41**	0.43**
BL				0.28*	0.27*	0.33**	0.18	0.19*	0.33**
RW					0.36**	0.17	0.41**	0.39*	0.60**
FL						0.24	0.34**	0.37**	0.45**
TC							0.40	0.48**	0.46**
AG								0.80**	0.76**
CG									0.77**

EL: ear length; NSL: nose to shoulders length; TL: tail length; BL: body length; RW: rump width; FL: foot length; TC: thigh circumference; AG: abdominal girth; CG: chest girth; BW: live body weight.

* p < .05.

** p < .001.

There was a significant association between BW and most biometric traits such as AG and CG with the higher correlations (respectively, 0.76 and 0.77). BW also shows a moderately strong positive correlation with RW (0.60), suggesting that animals with wider rumps tend to have higher BWs.

Multiple regression analysis demonstrated a robust model (adjusted R-squared value of 0.751) indicating that 75.1% of the variability BW can be explained by RW, AG and CG, identified as influential factors in predicting BW.

RW and AG showed strong positive associations with BW (t = 3.928, p < .001 for BW-RW; t = 4.116, p < .001 for BW-AG). CG also influences BW consistently, even if at a slower significance (t = 2.686, p < .05). Conversely, several predictor variables, TL, BL and FL, do not show statistically significant associations with live BW.

Another regression model was applied to AG, to also validate the connection of this variable with the others. The NSL variable (t = −2.829, p = .005) was added to the path analysis with the intention of disclosing the variability explained by AG.

The variance inflation factor (VIF) was also consulted to avoid multicollinearity among predictor variables; neither of the models exhibited any issues with multicollinearity (data not shown). Accordingly, with the multivariate regression results, the only variables included in the path analysis were RW, AG, CG and NSL. The covariance among the error terms of these variables was also added to the model as a common variance is expected due to the high-level relationship between CG–AG and RW–AG.

The model fit statistics, including CFI and TLI values of 0.991 and 0.960, respectively, an RMSEA of 0.108, and an SRMR of 0.031, indicate that the structural equation model fits the data well; comparison with a baseline model using a Chi-square test suggests that the used model is a significantly better fit, and the R² values for BW, AG and RW are 0.753, 0.647 and 0.248, respectively.

The PC reveals significant standardised direct and positive effects of AG, CG and RW on BW with respective values of 0.046 (z = 4.427, p < .001), 0.052 (z = 3.336, p = .001) and 0.108 (z = 4.920, p < .001).

Additionally, the model reveals significant direct effects, such as PC = 1.202 (z = 13.534, p < .001) for AG on CG, PC = 0.111 (z = 3.825, p < .001) for RW on AG, and PC = 0.239 (z = 2.746, p = .006) for RW on NSL.

The indirect effects computed within the model highlight the substantial impact of CG on rabbit BW through AG (PC = 0.133, z = 3.681, p < .001) and through NSL (PC = 0.287, z = 2.691, p < .01), as well as the influence of AG on BW through NSL (PC = 0.027, z = 2.702, p < .01).

In the investigation of rabbit populations across three distinct groups, a principal component analysis (PCA) was conducted to gain insights into the underlying morphological differences among these populations and highlight the characteristics of BFL. The analysis revealed that the first two principal components (PC1 and PC2) collectively accounted for a substantial 84.4% of the total variability in the dataset (eigenvalues PC1 = 7.806; eigenvalues PC2 = 0.632). Figure 1 represents the contribution of each variable to explain the variability inside the analysis.

Figure 1. PCA of the body measurements of rabbits and their correlations with the first two axes.

Figure 1 visually demonstrates the grouping of positively correlated variables, with the distance from the origin indicating representation quality and variables farther from the origin better represented. To PC1 explanation, contribute EL, BL, FL, CG and BW (10.67%; 10.64%; 10.77%; 11.01%; 12.14%). To PC2 explanation, contribute TL, RW and AG (30.69%; 35.44%; 13.12%). To visually represent the distribution of variability among individuals, a PCA graph was generated (Supplementary Figure 1).

PC1 exhibited a highly significant result (p < .001), indicating substantial differences in size, and PC2 also showed significant differences (p < .001) in shape among the rabbit breeds.

Post hoc Tukey’s HSD tests confirmed significant differences in PC2 scores between populations LV and NZC compared to BFL (<.001), as well as a significant difference between populations NZC and LV.

The analysis of variance, with a significance level of p < .05, confirmed the validity of the computed TIs (TL/CG and EL/BL ratio) among the three rabbit breeds (LV, BFL and NZC).

According to Tukey’s multiple comparisons test, the TL_CG_Ratio difference between LV and BFL was not statistically significant (p = .90). However, the TL_CG_Ratio difference between NZC and BFL was significant (p = .002). Regarding EL_BL_Ratio, there was no significant difference between NZC and LV (p = .32) or LV and BFL (p = .45). Yet, the EL_BL_Ratio difference between NZC and BFL was highly significant (p = .006), highlighting distinct characteristics between these two populations.

Discussion

As expected, BFL rabbits exhibited the lowest average values for most of the measured morphometric parameters, indicating a relatively smaller size compared to the other breeds. These differences suggest that BFL rabbits are characterised by small size, while LV rabbits exhibit larger body sizes and distinctive ear sizes and NZC rabbits have elongated bodies.

BFL shape is reasonably similar to other studies conducted on local rabbits’ breeds. BFL shows similar measurements for BL, CG, EL and BW to local rabbit populations studied in Western Algeria, which showed averages, respectively: 28.80 cm, 26.11 cm, 10.59 cm and 1.97 kg (Mogharbi et al. 2021). BFL rabbits show also similarities with body measurements of domesticated Kenyan rabbit population especially to TC, AG, and CG (16.92 cm, 28.42 cm and 25.45 cm), and slower BL than the three native Egyptian rabbit population (BL = 32.8; 32.5; 35.2) but similar average value for BW (1.66 kg) (Abdel-Kafy et al. 2018). BLF’s BW is also similar to the body live weight observed in a Nigerian population of adult rabbits; the average of BW for the two sexes in the study was 1.88 kg (Ajayi and Oseni 2012). Small body sizes are also confirmed in other African breeds such as Sudan rabbit population that showed smaller BL and NSL (24.58 cm; 10.35 cm), similar CG and AG (26.28 cm; 28.71 cm) and lightly smaller BW (1.71 kg) (Hassan et al. 2012).

Correlation’s results also offer insight into the relation between variables in analysed rabbit’s populations. The increase in the BW in BLF rabbits was positively correlated to the biometric traits taken; similar results were obtained in other studies. In 430 Kenyan domesticated rabbits (Awuor et al. 2018), EL, BL and TC got higher correlations with BW, respectively, r = +0.56, r = +0.67 and r = +0.55, while NSL, RW, AG and CG got slower values than BLF, respectively, r = +0.19, r = +0.37, r = +0.50 and r = +0.66. In 160 Nigerian cross-bred rabbits, higher correlation BW for BL was obtained than in BLF, with r = +0.63, and a BW for AG correlation equal to 0.76 (Rotimi 2021).

In 79 local rabbits in Sudan, higher correlations were observed between BW-EL and BW-TL, respectively (r = +0.35 and r = +0.26) but lower correlations were observed between BW and BL (r = +0.33), NSL (r = +0.23) and AG (r = +0.61) (Hassan et al. 2012).

Path analysis results also revealed similar insight in relationship between variables, like in other studies. Awuor et al. (2018) obtained results like those obtained in this research; the same predictor variables AG, CG, RW and NSL were investigated, yielding similar findings in terms of their direct effects on BW by AG, CG and RW. All these traits had significant positive direct effects on BW, with PCs comparable to those found in BLF. In BLF, NSL's direct effect remained above all smaller but statistically significant and positive (PC = 0.048), while in female Kenyan rabbits it was higher but negative (PC = −0,265) (Awuor et al. 2018). In comparison with a female Kenyan population, BLF rabbit shows notably higher and significant PC value regarding the direct effect of RW on BW (PC = 0.108 and PC = –0.001 ns, respectively), while lower PC was estimated for the direct effect of CG on BW (respectively PC = 0.052 for BLF and PC = 0.368 for the Kenyan population) (Awuor et al. 2018). For the PC between BW and AG, only in BLF there was a significant direct effect (PC = 0.046) while in the Kenyan population, this relationship was similar but not significant (PC = 0.061 ns) (Awuor et al. 2018). In the male Kenyan population, the results were similar, with the exception of higher PC for CG on BW (PC = 0.458) and a negative and still not significant PC between BW and AG (PC = −0.082 ns) (Awuor et al. 2018).

Many similarities lie in the magnitude and significance of the indirect effects for the indirect pathways, particularly CG's impact on BW through AG (PC = 0.133) was lower in BLF than in male and in female Kenyan rabbits (PC = 0.368; PC = 0.288); the indirect effect of CG on BW through NSL in BFL was higher and stronger than the Kenyan results (PC = 0.187 female rabbits; PC = 0.284 male rabbits). AG's indirect influence on BW through NSL in BLF was PC = 0.027, while the Kenyan’s result was non-significant (PC = 0.041 ns). In contrast, NSL's direct effect on BW was still significant and positive in BLF, while it was negative but significant in Kenyan rabbits (PC = –0.255) (Awuor et al. 2018).

In another study conducted on Saibei rabbit population (Wu Z-f et al. 2008), a path analysis reveals similar insight in direct effect between body variables; the study confirmed the strong and positive direct effect of CG on BW (PC = +0.73) resulting even higher than the one observed for BLF.

These findings confirmed the importance of choosing variables to understand their effect on BW: CG, AG, RW and NSL are important variables to evaluate if the scope is to enhance the growing performance of rabbits. Especially in contest where the tools affordable to make this evaluation are very limited, increasing the activity of collecting morphometric measures could be a valuable strategy to enhance the potential of local populations.

These results underscore the applicability of path analysis in guiding breeding decisions and optimising rabbit growth and weight management efforts.

The PCA results demonstrate significant differences in both size and shape among the BFL, NZC and LV rabbit populations. The communality values obtained in this study were high, implying that the variance of each variable was well represented in the extracted two components and hence the adequacy of the PCA procedure. Comparing our results with those of other researchers provides a broader perspective on the morphological differences within rabbit populations. The variables contributing most to these differences are EL, BL, FL, CG and BW for size and TL, RW and AG for shape. For instance, Yakubu and Ayoade (2009) extracted two PCA to explain the morphology of 103 New Zealand White × Chinchilla crossbred rabbits: PC1 explained 74.98% of the variance and PC2 contributed to 15.29% (Yakubu and Ayoade 2009). Their study emphasised the correlation between PC1 and BL and CG, emphasising the bond of PC1 on defining size and suggesting a different set of variables influencing shape-related difference (Yakubu and Ayoade 2009). Additionally, Udeh (2013) identified in a similar analysis a single dominant principal component that accounted for a substantial portion of the total variance, approximately 75%, and they reported that PC1 was highly correlated with BL, CG and TC (Udeh 2013).

According to bibliography, in rabbits and mice, ears and tails, respectively, are linked to heat exchange, consequently animals may increase the size of their appendages relative to body size, thus altering overall body shape, in order to adapt to increasing temperatures (Williams and Moore 1989; Ryding et al. 2021). The results obtained in the study for morphometric measures could suggest that in BLF population, adaptation to climate conditions is still in action, since there are not huge differences with the morphological traits related to heat dissipation with the other two populations. Moreover, it cannot be ignored that heat stress exposes animal’s growth and health mainly through induction of oxidative stress and inflammation (Khalid et al. 2020), for this reason it cannot be attributed with certain that this population, with its smaller size, is already showing morphological adaptation to the environment. Further studies on BFL population morphology could enhance their potential for coping with heat stress and thermoregulation.

The results of the analysis of variance and the Tukey tests conducted to validate the TIs, represented by TL_CG_Ratio and EL_BL_Ratio, among three distinct rabbit populations (LV, BFL and NZC), have yielded intriguing and statistically significant results (p < .05). These findings shed light on the variations in ear-to-BL ratios and tail-to-CG ratios within these breeds, providing valuable insights for rabbit breeding and adaptation strategies. Regarding the TL_CG_Ratio among the breeds can be made the following observation: the negligible difference in TL_CG_Ratio between LV and BFL (p = .90) suggests that these two breeds exhibit similar ratio, indicating that in this case, in proportional terms, there are not differences for this population relative to the tail-to-CG ratios. This funding lets us assume that in BFL and LV population, though they are reared in different climate contests, they have similar tale length-CG proportion, probably due to a higher responsive morphology to environmental condition of an outdoor breeding system. Further studies are required to validate this statement. As already suggested, according to the literature, the escalating temperatures associated with climate change might favour the development of larger appendages, which enhance effective heat dissipation and effectively shape the body of the animal (Ryding et al. 2021). On the contrary, the substantial statistical significance observed in TL_CG_Ratio between NZC and BFL (p = .002) highlights a clear differentiation between the proportion of these two breeds, confirming a different growing potential and a different shape. Furthermore, the significant difference in TL_CG_Ratio between NZC and LV suggests that also LV and NZC breeds have different ratios that could be attributed to different shape. These findings, as preliminary results, emphasise the need for careful consideration of both the goals of a possible genetic improvement of BLF population: growing rates and heat resistance. In a marginal contest characterised by adverse environment, adaptation to heat stress is a priority to include in a sustainable breeding program.

Nonetheless, these results are not confirmed by the EL_BL_Ratio, which reveal distinct patterns among the breeds. The non-significant differences between NZC and LV (p = .32) and between LV and BFL (p = .45) suggest a lack of substantial variation in ear and BL proportions between these breed pairs or suggest the need to analyse a bigger sample size. However, the highly significant difference between NZC and BFL (p = .006) underscores the notable dissimilarity in this ratio, confirming that the BLF population needs further studies to apply a breeding program that could also enhance their growing potential.

Therefore, for further implication, the TL_CG_Ratio and EL_BL_Ratio could be an easy tool to apply for guiding the selection not only on growing potential, but also to morphological traits that could enhance rabbit’s resistance to heat stress. Analysing morphometric ratios such as the ear-to-BL and tail-to-BL, is crucial to comprehensively address the adaptation of rabbit populations to heat stress and to better choose the animals with best morphology. These ratios offer a unique perspective on how specific body parts can evolve relative to overall body size, potentially facilitating more efficient heat dissipation. By focusing on these ratios, it is possible to identify and select new traits, such as ‘Transpiration Indexes’ and so animals with higher values, indicating larger appendages relative to their body size. Moreover, incorporating these ratios alongside traditional traits related to live BW can enhance breeding programs, allowing for a holistic approach to addressing climate challenges and ensuring the sustainability of rabbit populations in varying conditions.

Conclusions

The present study sheds light on the morphological diversity and characteristics of the BFL local rabbit population in comparison to NZC and LV rabbits. As expected, BFL rabbits displayed smaller body sizes across most morphometric parameters, underscoring their potential of adaptability to their specific breeding conditions and the possible influence of environmental adaptations, particularly in coping with heat stress.

The comparisons made with other rabbit populations from different part of Africa, reveal both similarities and variations in morphometric parameters, emphasising distinctions in BL and weight. Furthermore, correlations and path analysis unveil critical relationships between morphometric traits (CG, AG, RW and NSL) and rabbit BW, emphasising the importance of these variables in growth potential. Especially in regions with limited resources for evaluating growth potential, measurement of CG, AG, RW emerges as key tool for consideration in improving rabbit growth rates.

Attention should be given to variations in EL and TL to understand the impact of breeding conditions on size and morphology, especially in coping with heat stress.

Transpiration indexes highlight potential differences between outdoor-reared breeds (BFL and LV) and hybrids for intensive breeding (NZC), warranting further investigation. The findings have implications for sustainable rabbit population management and utilisation in Burkina Faso and beyond, offering insight for future research and conservation efforts related to rabbit population and their potential adaptation to local climate conditions.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data that support the findings of this study are available from the corresponding author, [S.G.], upon reasonable request.

Additional information

Funding

The project was funded by the Presidency of the Council of Ministers, with financial support from the 8x1000 IRPEF Fund donated by citizens to direct state management for the year 2018.